Shaman-Queens and Sacral Princesses

Home , Age of the Gods, Amatsu Shrine, Aoi Matsuri, Emperor Keitai, Emperor Suinin, Empress Genmei

SHAMAN-QUEENS AND SACRAL PRINCESSES

A Re-Examination of the “Golden Age” Narrative of Female Sacral Power in Ancient Japan

Natalie Louise McKay

ORCID: 0000-0002-2620-0372

Doctor of Philosophy – Arts

October 2020

Asia Institute, Faculty of Arts, University of Melbourne

School of Japanese History, Graduate School of Letters, University of Kyoto

Submitted in partial fulfilment of the requirements of the degree of Doctor of Philosophy – Arts

Abstract

Among scholarship of ancient Japanese women’s history, a traditional narrative has formed in which the ancient period emerges as a “golden age” of female power, an age in which women – through magico-religious performance – occupied positions of authority that were later usurped by the coming of patriarchal value systems, surviving only in traces. Recent decades have seen revisions to this “golden age” narrative, proposing more complex models of how this ancient “power” may have been configured, and how it may have been eroded by changing gender norms. However, many core assumptions from the early “golden age” narrative persist throughout modern scholarship, such as the existence of contrapuntal male-female rule as a standard format of power across the archipelago, or the interpretation of later female positions of (sacral or monarchic) power as direct legacies of a more empowered age.

This research thus performs a critical re-evaluation of the “golden age” narrative in its various incarnations, examining its arguments against images of women and female sacral performance from a variety of surviving contemporary texts. It also aims to bring to light new perspectives on the discourse, such as the overlooked consideration of regional diversity in ancient

Japan, and the multiplicity encompassed within the concept of “power”. Through this work, it can be seen that, while modern scholarship has developed to encompass greater nuance and diversity, several threads of the “golden age” narrative persist into the modern day, contributing to the oversimplification and homogenisation of what in actuality appear as a complex range of relationships between women, religion, and power throughout ancient Japan. By identifying and challenging these points of homogenisation, we may come to a more nuanced and comprehensive understanding of women’s experiences in ancient Japan.

This is to certify that the thesis comprises only my original work except where indicated in the preface; due acknowledgement has been made in the text to all other material used; this thesis is

100,219 words in length, inclusive of footnotes, but exclusive of tables, maps, appendices, and bibliography.

……………………………………………………………….

I would like to acknowledge the help and support given to me by my supervisors, Dr. Claire Maree

and Dr. Ikuko Nakane, in the writing of this thesis. I would also like to acknowledge the assistance provided by the input and feedback of Dr. Etsuko Toyoda, Dr. Anne McLaren, Prof. Shinji Yoshikawa,

and Mr. Yūta Watanabe.

I would also like to acknowledge the Traditional Custodians of the land on which I live and work, the

Wurundjeri and Boon wurrung peoples of the Kulin nation, and pay my respects to Elders past and

present.

This research was funded through an Australian Postgraduate Award scholarship, carrying on into a

Research Training Program scholarship, as well as a Monbukagakushō Scholarship.

All translations are the author’s own unless otherwise specified.

Table of Contents

CHAPTER 1: A RE-EXAMINATION OF THE “GOLDEN AGE” NARRATIVE ...... 8 1.1. Rationale ...... 9 1.2. Research Questions ...... 12 1.3. Structure of Thesis ...... 13 1.4. A Few Notes on Sources...... 17 1.5. Notes on Language...... 19

CHAPTER 2: AN INTRODUCTION TO THE “GOLDEN AGE” NARRATIVE ...... 25 2.1. Japanese Women’s History and the “Golden Age” Narrative ...... 28 2.1.1. The Beginnings of the “Golden Age” Narrative ...... 28 2.1.2. The “Golden Age” Flourishing: Takamure and Yanagita ...... 34 2.1.3. The “Golden Age” in Modern Scholarship ...... 39 2.2. Methodology ...... 44 2.2.1. History and Hermeneutics ...... 44 2.2.2. Reading Texts in Context ...... 46 2.2.3. Reclaiming the Humanity of Historical Women ...... 52 2.2.4. Ryukyu as Model for Ancient Japan ...... 55 2.2.5. Defining Power ...... 57

CHAPTER 3: WOMEN IN THE “GOLDEN AGE” ...... 63 3.1. Reading Prehistoric Society ...... 64 3.1.1. Theories of Prehistoric Matrilineality ...... 64 3.1.2. Goddesses and Women ...... 73 3.2. Women and the Spiritual World ...... 78 3.2.1. The Miko: Female Shamanism in Ancient Japan ...... 79 3.2.2. The Gendering of Sacral Power ...... 89 3.2.3. The Female Body as Source of Pollution ...... 95

CHAPTER 4: FEMALE RULERSHIP IN THE “GOLDEN AGE” ...... 104 4.1. Himiko, Queen of Wa ...... 106 4.1.1. An Introduction to Himiko ...... 106

4.1.2. Situating Himiko in History...... 114 4.1.3. Himiko and Tensions of Power ...... 121 4.2. Jingū, Empress-Regent ...... 126 4.2.1. An Introduction to Jingū ...... 126 4.2.2. Situating Jingū in History...... 131 4.2.3. Power and Gender in the Jingū Myth ...... 135 4.3. The Himehiko System ...... 143 4.3.1. An Introduction to the “Himehiko System” within the “Golden Age” Narrative ...... 143 4.3.2. Himehiko and Formats of Power in Kyushu ...... 147 4.3.3. Himehiko and Formats of Power in the Eastern Provinces ...... 155 4.3.4. Himehiko and Formats of Power in the Kinai ...... 158

CHAPTER 5: WOMEN IN THE “SILVER AGE” ...... 167 5.1. Centralisation and Gender: The Impact of Yamato Control ...... 169 5.1.1. The Process of Yamato Centralisation ...... 169 5.1.2. Confucianism, Patriarchy, and the End of the “Golden Age” ...... 174 5.2. Priestesses in the “Silver Age” ...... 185 5.2.1. Ritsuryō and Religious Centralisation ...... 185 5.2.2. Priestesses and Power within Hegemonic Systems ...... 187 5.2.3. Priestesses and Power: Beyond the Bureaucracy ...... 191 5.3. Women, Religion, and Imperial Dominion: Tensions of Power within the Ise Saigū and Kamo Saiin ...... 196 5.3.1. The Ise Saigū and Kamo Saiin: Historical Background...... 196 5.3.2. Hime in Decline: Saigū as Sacral Partner...... 202 5.3.3. Saigū and Saiin as Mechanisms of Regional Control ...... 212 5.3.4. Tensions of Power within the Saiō ...... 221 5.4. The Buddhist Nun as Hime ...... 227 5.4.1. Nuns and Monks as Replication of Hime and Hiko ...... 228 5.4.2. Nuns and Power: A Brief Note ...... 233

CHAPTER 6: EMPRESSES IN THE “GOLDEN AGE” NARRATIVE...... 240 6.1. Empresses as Hime: Traditions of Female Rule ...... 243 6.2. Gendering the Tennō ...... 249 6.2.1. Empresses and Femininity ...... 249

6.2.2. Empresses and Masculinity ...... 260 6.3. Empresses, Gender, and Kami-Worship ...... 266 6.3.1. Jotei as Shaman-Queens ...... 266 6.3.2. Empresses and Amaterasu ...... 271 6.3.3. Living God(desses): Jotei and Images of Male Imperial Religiosity ...... 275 6.4. Nun-Empresses and World-Ruling Queens ...... 280 6.4.1. Buddhism, Gender, and the Image of the Ideal Monarch ...... 280 6.4.2. Confucianism and Hierarchy ...... 286 6.5. Empresses as Dowager-Regents, Empresses as Placeholder Monarchs ...... 290

CHAPTER 7: TOWARDS A NEW UNDERSTANDING OF WOMEN’S POWER IN ANCIENT JAPAN ...... 301 7.1. Conclusion of Findings ...... 301 7.1.1. Homogenisation and Diversity ...... 301 7.1.2. A Non-Linear Model of Patriarchy...... 303 7.2. Within the Wider Field ...... 304 7.3. Avenues of Further Study ...... 305

APPENDIX: A TIMELINE OF ANCIENT JAPAN ...... 308

BIBLIOGRAPHY ...... 318

CHAPTER 1: A RE-EXAMINATION OF THE “GOLDEN AGE” NARRATIVE

Within the field of Japanese women’s history, a particular narrative has formed throughout the late 19th and 20th centuries: one which depicts ancient Japan1, prior to the establishment of the ritsuryō2 bureaucracy in the seventh century CE, as a heyday of female power attained through women’s sacral roles and performances. Although more recent works within the field have complicated this image somewhat, its legacy on both Japanese women’s history, and adjacent fields such as Japanese religious history, remains considerable. This research will examine and re-evaluate this deteriorationist narrative of past glories replaced with patriarchal decline – henceforth dubbed the “golden age” narrative, to borrow a term from Hesiod3 – and its enduring threads within and

1 A timeline of ancient Japan, and the principle female figures from each era, is appended to the end of this thesis. “Ancient,” within this thesis, is used in broad alignment with the Japanese-language scholarship category of kodai (古代), and in particular the demarcation of that period from the third to 11th centuries C.E.

(Tsude 2011), with a particular focus on the early ancient period from Kofun to Nara. Brief forays into prehistoric (Jōmon and Yayoi) culture will also be made.

2 律令, “criminal, administrative, and civil codes”; referring to several Tang-inspired laws and codes beginning with the Taika Reforms of 645 andd continuing through the Asuka and Nara periods. “Ritsuryō” is also sometimes used in reference to the bureaucratic administration formed around these codes.

3 The ancient Greek poet Hesiod, in his Works and Days, presents a deteriorationist narrative of human history, beginning with an idyllic, utopian “golden age” and declining steadily until the warlike, impious contemporary

“age of iron” (Works and Days 11.109–201); the same concept was later echoed by Roman poet Ovid in the

Metamorphoses (Metamorphoses 1.89–150). The term has here been used to reflect an idealised image of past, lost glory. The term is one the author has applied themself; within scholarship, terms such boken (母権,

“matriarchy”) nyoji (女冶, “female governance”) or himehiko-sei (ヒメヒコ制, “the himehiko system”) are used to describe the notion of a primeval, female-centric era. These terms, their nuances, and their history in scholarship will be discussed in greater detail in Chapter 2.1 below.

8 without the sub-discipline of women’s history, through a close study of textual evidence concerning the relationship between women, religion, and power in ancient Japan.

It should be noted that “woman,” as a category, is a complex notion within scholarship. A full analysis of gender identity, gender performance, and the categorisation of onna (女) is beyond the scope of this thesis. For the most part, our identification of gender – who is considered a

“woman” for the purposes of this study – will be based on broad textual and linguistic markers such as gendered names (e.g. ones ending in -Hime) and explicit textual identification (e.g. someone being denoted as a “woman,” “daughter,” “sister”). It must be noted that, within ancient records, there are a number of figures whose gender is not clearly denoted in such a fashion; although these have often been normatively assumed male as default, we will in general attempt to refrain from such assumptions; instead, we will examine context clues and be guided by the analyses of other scholars. It is also necessary to note that “women” here does not refer to a single, cohesive category, united in oppression and having common goals and bonds, but rather to a broad, disparate social construct, one which both affected and was affected by other factors such as class, region, and the like4.

1.1. Rationale

The 1911 inaugural issue of the women’s literary journal Seitō5 opened with the following words by its founder, women’s activist Hiratsuka Raichō (平塚らいてう; 1886–1971):

In the beginning, woman was truly the sun; an authentic person. Now, woman is the moon.

Living in reliance on others, reflecting the light of others, she is the wan and sickly moon.6

4 Pandey 2004, 223–4.

5 青鞜; “Bluestocking”

Hiratsuka was not herself a historian, but the phrase spoke to a nascent trend within the emerging field of Japanese women’s history: the narrative of ancient Japan as a pre-patriarchal time of female power, a society mediated by women’s magico-religious performance until the importation of patriarchal values from the continent forced female authority into decline7. Ancient records – from

Yamato court chronologies to the travel records of Wei Chinese emissaries – offer up tantalising images of prominent religious women, such as the ancient princess-priestesses Yamato-Hime8 (倭比

売/倭姫) and Yamato-Totohi-Momoso-Hime9 (夜麻登登母母曽毘売/倭迹迹日百襲姫), the oracular Empress Jingū10 (神功), the cloistered “shaman-queen” Himiko (卑弥呼), even the splendorous image of the sun goddess Amaterasu (天照) herself in all her shining glory. The early

20th century further saw the translation into Japanese of the anthropological works of Lewis H.

Morgan (1818–1881) and Friedrich Engels (1820–1895), whose theories on human social development placed women at the centre of all prehistoric societies, with patriarchy as a later development11; theories which saw considerable influence in Japanese historical scholarship of the time, and would become something of a backdrop to the study of women in Japanese prehistory12.

6 “元始、女性は実に太陽であつた。真正の人であつた。今、女性は月である。他に依つて生き、他の

光によつて輝く、病人のやうな蒼白い顔の月である。” (“Genshi, josei wa makoto ni taiyō de atta. Shinsei no hito de atta. Ima, josei wa tsuki de aru. Hoka ni yotte iki, hoka no hikari ni yotte kagayaku, byōnin no yō ni aojiroi kao no tsuki de aru.”) (Hiratsuka 1911, 37).

7 Sakima 1926; Takamure 1966a, 1966b; Yanagita 1984.

8 Traditional birth and death dates unclear; appears as early as 5 BCE and as late as 121 CE.

9 Traditional birth and death dates likewise unclear; appears as early as 91 BCE and dies sometime after 88 BCE.

10 Regency traditionally dated 201–269 CE. As she is said to have been 100 years old at her death in 269, her traditional birth year must be around 169 CE.

11 Morgan 1871; Engels 1902.

12 Ijūin and Kawai 2017, 213.

The narrative of pre-ritsuryō ancient Japan as such a “golden age” of female power – in its scholarly branch, at least – was heavily developed by the works of historian Sakima Kōei (佐喜眞興

英; 1893–1925), folklorist Yanagita Kunio (柳田國男; 1975–1962), and especially women’s historian

Takamure Itsue (高群逸枝; 1894–1964). Emerging through their works was the story of prehistoric

Japan as a world ruled by women’s magic, where sacral power was believed to reside within the female, and the respect afforded to the ancient priestess was such that she was a leader of her community, her chiefdom, even her kingdom; sometimes sharing her throne with a male kinsman who handled the more mundane, secular tasks of rulership13. The narrative formed by these works went on to shape the discourse on gender in ancient Japan, and this last component – the theory of the ruling partnership of sacral female and political male, dubbed the “himehiko system” by

Takamure – has been particularly influential and enduring throughout the study of ancient Japan, seeing widespread reference and adoption even outside the specific subfield of women’s history14.

Subsequent waves of scholarship have, of course, adapted on, developed from, or even rejected aspects of this “golden age” narrative. Takamure’s theories on an initial matriarchy preceding himehiko, for example, have been heavily challenged15, and models of himehiko rulership itself have been debated and revised to include more nuance and complexity in the dual leaders’ respective roles16. Nevertheless, basic threads of the “golden age” narrative – the interpretation of empresses regnant and ancient priestesses as legacies of prehistoric female authority, or the existence of the “himehiko system” of dual rulership as a standard format of power – persist into

13 Sakima 1926; Takamure 1966a, 1966b; Yanagita 1984.

14 E.g. Sakima 1926, 51; Takamure 1966a, 69; 1966b, 96–117; Sekiguchi 1987, 16; Okano 1993, 28; Kuratsuka

1996, 28; Naoki 2008, 206.

15 E.g. Piggott 1994, 7; Tanaka 1996, 183, 228.

16 E.g. Sekiguchi 1987, 16–7; Yoshie 2007, 207–12; Maeda 2008, 17–9; Piggott 1997, 38–40.

11 modern scholarship, having hugely influenced the field of Japanese women’s history and making frequent appearance in other historical works touching on women and religion.

There remains, then, a deeper question to be asked: to what extent are the narrative construct of the “golden age” and its subsidiary threads helpful for us as scholars seeking to understand the lives and experiences of women, the systems of power and gender in ancient Japan?

The field of women’s history seeks to recover the “hidden truths” of ancient women, not as objectified background characters within a narrative, but as ancient people who lived long ago, reclaiming historical women’s humanity through reconstructing, as best we can, their lives and experiences17. Through the theories of historical hermeneutics, this can best be achieved through a constant back-and-forth dialogue between the evidence and our interpretations of it18. This work, therefore, seeks to return to the textual evidence as part of this “hermeneutical circle”19 (see

Ch.2.2.1) to develop on our understanding of it, examining and reconstructing the narrative lens through which we understand women’s experiences in ancient Japan.

1.2. Research Questions

This work is thus concerned with addressing the following questions:

1. Firstly, to what extent are this “golden age” narrative and its derivations useful lenses

for the historical study of female power in ancient Japan? How does this body of

scholarship fare in reclaiming the experiences and subjectivities of ancient women, and

where could it be improved in this regard?

17 Munslow 2000, 228–230; Schüssler Fiorenza 2013, 106–10; Faure 2003, 3.

18 Mazlish 1998, 91.

19 Mazlish 1998, 91.

2. Secondly – and similarly – to what extent is the “decline” or “silver age” portion of the

“golden age” narrative useful as a framework for understanding women’s experiences

and relationships to “power” – of all kinds – through the later ancient period? What

greater nuance could it encompass?

3. Thirdly, to what extent can later roles of female sacral or temporal power be considered

“transformations” of older positions of female authority? Are the high priestesses,

empresses, and the like of Asuka onwards best understood as a continuation of a

prehistoric tradition of female sacral rulership, or as roles developed within a particular

political and historical context, incorporating elements from various sources and societal

drives?

These assessments have been made through a critical reading of early textual sources, and the images of women, women’s power, and women’s religious performance contained therein, in comparison with their common framing in scholarship. In addition, the self-presentation of, reigns of, and circumstances surrounding the enthronement of empresses regnant within the records has been similarly studied. To examine the position of women within the family unit, research concerning census data – a considerable body of which survives from ancient Japan – has been consulted, tracking changes in marital customs, family structures, and the like.

1.3. Structure of Thesis

Our study of the “golden age” narrative and its legacies will be structured in roughly chronological order, beginning with a re-examination of the ancient period itself, where it will be seen that the “golden age” narrative, as it has appeared through various waves of women’s history, presents an overly-homogenous image of women’s position in society and religion which does not provide adequate space for the complexity and diversity of faith and gender across ancient Japan.

Next, it will trace the proposed “fall” into patriarchy, and the female figures – priestess and empress

– who have been traditionally interpreted as a legacy of female sacral rule. The changing of gender roles will be understood as a product of multiple factors, including a shifting political landscape and a developing social discourse both foreign and domestic. The role of priestesses, it shall be seen, is a mix of ancient customs and new developments caused by religious centralisation; empresses, meanwhile, are best understood as incoporations of many disparate political factors and influences rather than linear remnants of matriarchal power.

This thesis is comprised of seven chapters overall. This first chapter introduces the topic, giving a brief overview of the narrative of the “golden age” of female power in ancient Japan, how it has developed within scholarship, and the issues surrounding the adoption and continuation of this narrative and its subsidiary theories. The central research questions and structure of the thesis are here outlined, as will be preliminary notes on the sources used, translations, use of language and

Romanisation within this text.

The second chapter will outline the literature review, going into more detail about the history of the “golden age” narrative (and women’s history in Japan as a whole), its predecessor in

Morgan and Engel’s “prehistoric matriarchy” theories, and its modern developments and incarnations, including where it has and has not been challenged in modern scholarship, highlighting the gaps filled by this work. The theoretical underpinnings of this work will also be articulated, as will be the methodological processes involved in developing this research.

The third chapter is devoted primarily to addressing the first research question, examining the notion of pre- and early-Yamato20 Japan as a culture centring women and mediated by female

20 The official timeline of the Yamato Dynasty (大和王権, Yamato ōken) – the imperial line of Japan which survives to this day – places the beginnings of its dominion as far back as the seventh century BCE, but this dating is in considerable question; Chinese sources suggest the existence of a possibly-distinct “Yamatai”-

14 magico-religious power, by comparing the “golden age” narrative with textual sources on female sacral power and spirituality. Here, we will see that the neat dichotomy of empowerment versus subordination presented by the “golden age” narrative does not adequately encompass the complexities of gender relations in ancient Yamato, which appears as an androcentric society in which female spiritual power was attained through liminality and the notion of the woman as

“other,” yet still important to the functioning of society as part of a contrapuntal whole alongside male sacral roles and duties.

The fourth chapter continues on with the analysis of the “golden age” in the context of ancient Japan by examining images of female rulership: first, through the textual case studies of

“shaman-queen” Himiko and Empress Jingū; and second, through the broader theory of contrapuntal himehiko rule. Through this, we will see the complexities and tensions that exist within the examples of Himiko and Jingū that preclude their easy categorisation as signifiers of specifically feminine modes of rulership. Furthermore, we will see that overemphasis on the notion of the

“himehiko system” among many varied images of gender in rulership across time and region risks reifying normative assumptions concerning gender roles. It will also be noted that regional and cultural diversity – particularly as pertains to disparate groups such as the Hayato, or the various clans labelled tsuchigumo or kuzu – has been underinvestigated as a potential site of difference, perpetuating a homogenised image of the ancient archipelago.

The fifth chapter is dedicated to the second, and partially the third, research questions. Here the “decline” portion of the narrative is examined – what will here be termed the “silver age” of female power, following the earlier “golden”21 – with an overview on changing gender roles in both

headed confederacy in the third century CE (see Ch. 4.1), and Yoshimura Takehiko dates the current dynasty to the fourth or fifth century (Yoshimura 2010, 46).

21 The metaphor of the “silver age” derives also from Hesiod, representing something of a mid-point in his deteriorationist narrative of history in his deteriorationist narrative of human history: a marked decline from

15 secular and sacral spaces, under influences both foreign and domestic. The position of female spirituality and authority under the ritsuryō bureaucracy of the seventh century onward is examined through various feminine religious positions – miko both inside and outside the state cult, the Ise saigū, the Kamo saiin, and Buddhist nuns – in light of findings from Chapters 3 and 4, complicating the assertion that these roles stand as declined transformations of part female glories. From this, we will understand the ways in which the “golden age” narrative, even in its most recent incarnations and surviving “threads,” continues its program of homogenisation and struggles with a complex understanding of “power” and patriarchy, over-relying on access to specific avenues of authority as a broader marker of “power” and depicting patriarchy as a strictly linear decline.

The sixth chapter continues in a similar vein, this time challenging the notion that the empresses of the “Century of Empresses” (593–770) are likewise a remnant of “golden age” female power. The circumstances surrounding the enthronement of these women, the public presentation of their reigns, and their relationship to religion are all examined through textual evidence, showing that these women do not exist as pure declines from a previous image of feminine authority, but rather represent something more complex, a role created through various social factors and incorporating various elements and inspirations, each navigating their gender and deploying and maintaining their authority in unique and fascinating ways.

The seventh and final chapter summarises and concludes the findings of this work: that the narrative model of the “golden age” of female power and its derivations are ineffective means for

the preceding “golden age” but still superior to the contemporary “age of iron” (Works and Days 11.109–201).

Where the “golden age” narrative presents prehistoric Japan as idealised glory days of female power, the later ancient period – from around the fifth or sixth century CE to the twelfth – becomes the “silver age,” in which women’s power is lessened but their subordination not yet total. The term is one which the author has chosen to apply themself, and does not correspond to any particular phrase or terminology in existing scholarship, so much as a phase in the broader narrative.

16 understanding gender and religion in the past, oversimplifying and homogenising a complex and diverse issue. Instead, the need for a new, more nuanced narrative lens for understanding women’s power and religious performance in ancient Japan is articulated, as well as avenues through which it might be developed further, such as the historical study of regional female religious roles.

1.4. A Few Notes on Sources

The core of this study rests in the hermeneutical reading and interpretation of primary textual evidence, through processes outlined in Chapter 2.2. In particular, it will draw from the official, imperially-commissioned records of Japanese history that survive into the present: the

Kojiki22 (712 CE), the Nihon Shoki23 (720), the Shoku Nihongi24 (797) and, to a lesser extent, the Nihon

Kōki25 (840); the specific versions of these texts used are outlined in Chapter 1.5. This body of works

– the latter three comprising half of what is now known as the Rikkokushi26, or “Six National Histories”

– represents our best-preserved annals of ancient Japanese history and imperial court proceedings, and is thus indispensable for our study of monarchy and power, even though its own implicit biases and shortcomings must be taken into account (see Ch. 2.2). Several provincial gazetteers (風土記, fudoki) produced by court commission in 713 also survive in whole or in part to this day, and these too will be referenced where relevant. The Engishiki27, a 927 book of bureaucratic laws and regulations, including vital information on the management of the state cult, also provides us with

22 古事記; “Records of Ancient Matters”

23 日本補記; “Chronicles of Japan”

24 続日本紀; “Further Chronicles of Japan”

25 日本後期; “Later Chronicles of Japan”

26 六国史

27 延喜式; “Procedures of the Engi Era”

17 detailed accounts of systematised worship. These texts are the primary focus of reference for this thesis, but not the entirety of its sources; other ancient and mediaeval Japanese works are also drawn upon when relevant.

For Yayoi28 Japan (c. 300 BCE–300 CE), the existing Japanese histories are of doubtful veracity, being compiled long afterwards. For more reliable textual evidence of Japan’s pre-literate period, we must instead turn to the accounts of Chinese chroniclers. The earliest of these is the

“Treatise on the People of Wa”29 appearing in the Records of the Three Kingdoms30, compiled at the end of the third century CE, noted for the presence within the text of a female ruler, Queen Himiko.

Other early Chinese depictions of “Wa”31 (Japan) and Himiko appear in the Book of the Later Han32 (c. fifth century CE), the Book of Liang33 (635 CE), and the Book of Sui34 (636); these are all examined in their turn. Again, several other texts of Chinese origin have also been drawn from over the course of this research, and will be described in their place.

Although this research focuses on textual study, it also draws occasionally from the archaeological work of previous scholars in examining this ancient period. As histories of Japan are often heavily mythologised, physical findings will be important in helping to distinguish truth from legend. Unfortunately, the prevalence of wooden buildings and adoption of cremation result in a

28 1000 BCE–300 CE (in older categorisations, beginning at 300 BCE)

29 倭人傳; Wōrénzhuán

30 三國志; Sānguózhì

31 倭, Wa (Japanese) or Wō (Chinese); believed to be a pejorative Chinese exonym meaning “submissive people” or “short people” and later replaced by Japanese scholars with Wa (和, “peaceful”) which is still in use today.

32 後漢書; Hòu Hàn Shū

33 梁書; Liáng Shū

34 隋書; Suí Shū

18 dearth of archaeological remains for the later ancient period, limiting the usefulness of this method of “checking against.”

This work also makes use of secondary sources from the body of existing literature discussing ancient Japanese history. In particular focus will be the sub-fields of political history (with reference to the formation and centralisation of the early Japanese imperial state), women’s history

(with reference to the role of women in ancient Japanese society, and the impact their actions had on history), and religious history (with reference to the religious roles and beliefs of ancient Japan, and their impact on wider society), building upon the theories and studies of previous scholars.

Although primarily bound to the state of Japan (or, more specifically, “Japan” as defined by the ancient reach of Yamato authority, rather than its modern borders: broadly, the islands of Honshu,

Kyushu, and Shikoku, and excluding Hokkaido and the Ryukyu Islands) it will also incorporate elements of transnational history, examining the flow of both Buddhism and Confucian patriarchal ideals from China and Korea into Japan, and the impact of this importation on Japanese religious life and wider society.

Finally, it must be noted that this work was completed during the COVID-19 pandemic of

2020–1, much of which time was spent under strict lockdown laws in the Australian state of Victoria.

With libraries and institutions closed, and travel restricted, access to primary and secondary textual sources has been significantly limited during the final years of this research.

1.5. Notes on Language

One difficulty facing the Anglophone scholar of Japanese is the question of transliteration.

When Japanese terms and expressions are used, they will be rendered following the rules of

Modified Hepburn Romanisation, that being the current standard for Anglophonic writing. When

19 words in kanji are appended, modern forms of kanji will be used (e.g. 経 instead of 經), excepting in direct quotations, the handling of which is outlined below. Romanised transliterations will similarly use modern orthographies. Chinese names and terms shall be rendered in pinyin Romanisation, with tone markers omitted when the word is used within English text (as opposed to an appended footnote). Korean names and terms will be transcribed following the Revised Romanisation of

Korean system. Sanskrit names and terms will be transcribed according to the International Alphabet of Sanskrit Transliteration conventions.

Quotations from primary sources will be translated into English for the reader’s convenience, but appended with both original text and romaji (for Japanese) or pinyin (for Chinese) transliterations for the sake of transparency. All translations are the author’s own unless otherwise specified. Transliterations from the Kojiki, Nihon Shoki, Shoku Nihongi, Nihon Kōki, and Engishiki will be derived specifically from its rendering in Yoshikawa Kōbunkan’s Kokushi taikei series, albeit omitting annotations and reading guides. Kanji characters here so used will be the same as found in the published source, excepting those characters that do not as yet have Unicode assignations, in which case the author shall use best judgement in selecting the closest alternative. Characters noted in the source text as being assumed typographical errors will be omitted or altered according to customary interpretation. Romaji transliterations shall be based off the reading notes provided by the source text, but converting from ancient to modern orthography (e.g. 大 transcribed as ō instead of oho) for the sake of textual consistency. As reconstruction of ancient Chinese pronunciation is still very much an ongoing process, transliterations of Chinese will use modern pinyin readings for the sake of simplicity.

Japanese, Chinese, and Korean names will all be rendered surname-first (where applicable), excepting in the case of those modern scholars who typically publish surname-last. By convention of the era, the names of ancient Japanese figures will additionally take the possessive particle no between clan and personal name, e.g. Fujiwara no Kamatari (藤原鎌足), “Kamatari of the Fujiwara.”

Titles of office will be typically translated into their most widely used English approximation, e.g.

“Emperor/Empress” for “tennō”. One potential issue with this approach is the possibility of confusion between empresses regnant – who took the same tennō title as a male emperor – and empress-consorts, who took the entirely different title of kōgō; thus, it will be necessary to clearly differentiate between the two in context.

Court numerical ranks, or kurai (位), although customarily preceding an individual’s name in records, will be omitted except in direct quotation, as they are both cumbersome and also frequently subject to change both throughout an individual’s life and through the introduction of new ranking systems. Familial kabane (姓) titles, also ubiquitous in official records, were less malleable, but by scholarly convention will also be omitted except in quotation. The general honorific of -no-Mikoto (命 or 尊), broadly applied in earlier records, presents a challenge. Basil Hall

Chamberlain, in his translation of the Kojiki, renders it as “His/Her Augustness”35, but this quickly becomes cumbersome with repetition. For royalty, the text will instead use the English word describing their rank or status (e.g. “Emperor”/“Empress”/“Prince”/“Princess”) in order to delineate both reigning monarchs and relationships within the royal clan. For deities and for non-royal mortal elites, the honorific shall be omitted entirely for a naturalistic translation, its presence (or absence) to be noted only when relevant. The divine title of -no-Kami (神) will be likewise omitted, as it will suffice to make clear in the text that the figures in question are gods. Broadly speaking, the intention in translating terms will be to impart upon the reader their contextual meaning, rather than to cleave to specific word usage, unless that specific word usage holds particular significance to the argument (e.g. the labelling of Jingū as “empress consort” versus “empress regnant” discussed in Ch.

4.2.3).

35 Chamberlain 1882.

Many figures that will be discussed within this thesis are endowed with multiple personal names. An emperor, for example, may have one name at birth, another as they ascend their throne, and two more (one Japanese, and one Chinese-style) confirmed upon them posthumously. For monarchs, the Chinese-style posthumous name will be the standard point of reference, as the most common name by which these figures are historically known. For those empresses regnant with two regnal periods, and thus two imperial names, as an individual they shall be referred to as both at once: Empress Kōgyoku-Saimei (皇極/斉明) and Empress Kōken-Shōtoku (孝謙/称徳). For events particular to one regnal period, the name associated with that period shall be used. Kōken-Shōtoku will also be referred to as “Princess Abe” (阿倍) regarding events prior to her coronation, for consistency with quoted source texts. Other figures will be referred to by the name most commonly associated with them, e.g. “Crown Prince Shōtoku” (聖徳太子) instead of “Prince Umayado” (厩戸),

“Yamato-Takeru” (倭建/日本武) instead of “Prince Ousu” (小碓), and “Empress Kōmyō” (光明) instead of “Fujiwara no Asukabe-Hime” (藤原安宿媛). Monks and nuns will be referred to by monastic name. Japanese names will also be appended on first mention with their spelling in kanji; modernised, with the exception of a few figures for whom traditional characters are still customarily used. In ancient texts, kanji orthography was not yet standardised, and so an individual’s name may be written with multiple character combinations. Where multiple readings are given, the Kojiki spelling will precede the Nihon Shoki spelling, and fudoki variants will be entered last. Chinese names will be appended with traditional hanzi, and Korean names with hanja.

Some translators of these ancient texts have elected to translate names (of gods, people, places, etc.) into English in order to convey their meaning and narrative function36. Chamberlain’s

Kojiki takes this approach, although the resultant titles (such as “the Deity Master of the Great Land” for Ōkuni-Nushi-no-Kami (大国主神))37 are often cumbersome and not typically used in modern

36 Heldt 2014, xiii–xiv.

37 Chamberlain 1882; Heldt 2014, xiii–xiv.

22 scholarship. In Gustav Heldt’s 2014 Kojiki translation, he takes a more blended approach, using

Romanisations for the more recognisable figures and places and translations for the rest. However, as he points out, the etymologies of many of these names are the site of considerable scholarly debate and differing interpretation38. For clarity and consistency’s sake, name meanings will only be translated and discussed where they contain significance to the argument. Gods possessing multiple names will be referred to by their most commonly used one, with short-form versions favoured over long-form (e.g. “Ninigi” in preference of “Ame-Nigishi-Kuni-Nigishi-Amatsu-Hiko-Hiko-Ho-no-Ninigi”

(名天迩岐志国迩岐志天津日高日子番能迩々芸)) for ease of reading. Buddhist figures such as bodhisattvas will be referred to by their Japanese name, bracketed on first appearance in a given chapter with original Sanskrit name, e.g. “Kannon (観音; Avalokiteśvara)”. Emphasis on the Japanese name is done to highlight the fact that, through the transmission of these figures to Japan via China and Korea, they and their worship acquired specific characteristics not universal to Buddhist tradition; for example, the typically feminine depiction of Kannon in contrast to the typically masculine image of Avalokiteśvara in Indian statuary.

Over the last few decades, works by John Breen and Mark Teeuwen, Richard Bowring, and

Herman Ooms have called into question the use of the term “Shinto” to describe the religious practices of ancient Japan. In modern usage, “Shinto” calls to mind a single, organised faith, such as

Meiji39 State Shinto, or the modern Association of Shinto Shrines, whereas ancient Japanese religious traditions were highly varied and regionally divergent, especially prior to the founding of the

Jingikan40 in the eighth century41. Thus, although used frequently in older works of religious history42,

38 Heldt 2014, xiii–xiv.

39 The Meiji period extended from 1868 to 1912

40 神祇官, “Department of Divinity”; a department of the ritsuryō bureaucracy concerned with nation-wide oversight and administration of priests, rituals, and shrines.

41 Breen and Teeuwen 2000, 4; Bowring 2005, 3; Ooms 2009, 56–7, 166–7.

23 the term “Shinto” will be avoided for Japan’s ancient religious traditions, replaced by the appellation of “kami-worship,” which should itself be understood as an umbrella term representing an array of regionally, temporally, and socially disparate religious practices within the archipelago. The specific, centralised religious traditions promulgated by the Jingikan will be referred to as the “state cult” owing to its connection to, and use in support of, the Yamato imperial hegemony.

42 E.g. Anesaki 1930, Hori 1968.

CHAPTER 2: AN INTRODUCTION TO THE “GOLDEN AGE” NARRATIVE

We turn now to a more detailed examination of the “golden age” theory, its development, and its legacies within modern scholarship. Within this chapter, we will also examine our methodological means and outline the theoretical bases for this research.

Let us begin with a historiographical analysis of the academic debate on the position of women and female spiritual power in ancient Japan from the Meiji era – situated by Marnie S.

Anderson as the beginning of women’s history as we understand it in Japanese academia1 – to the modern day. This path leads us through various periods of Japanese women’s history: the first scholarly stirrings of the “golden age” narrative, casting ancient Japan as a world of gender equality and female magico-religious power2; the wartime and early post-war development and propagation of this narrative, and its subsidiary theories such as himehiko3; to the increasingly nuanced models of female authority presented by the modern wave from the 1980s onwards, developed in conversation between Western and Japanese scholars, which nevertheless retain aspects of “golden age” discourse such as himehiko theory4. It will also benefit us to visit the “universal prehistoric matriarchy” theories of anthropologists Morgan and Engels5 which, while speaking in generalities and not specific to the Japanese context, were highly influential in the development of this “golden age” narrative as pertained to Japanese women’s history6.

1 Anderson 2020, 63–4.

2 Tatsumi 1887; Sakima 1926; Anderson 2020, 64–5.

3 Takamure 1966a, 1966b; Yanagita 1984; Anderson 2020, 66–8.

4 Sekiguchi 1987; Piggott 1994, 1997; Kuratsuka 1996; Matsumura 1999; Kidder 2007; Yoshie 2007; Maeda

2008; Ambros 2015.

5 Morgan 1871; Engels 1902.

6 Anderson 2020, 67–8.

Through this thesis, we will examine this narrative and its developments against the context of the existing contemporary textual sources outlined in Chapter 1.4, as part of the dialogue between text and interpretation demanded by the “hermeneutical circle” of historical analysis7. This reading will further bring to light theoretical issues which complicate the narrative. For example, the concept of “ancient Japan” itself acts as once such conceptual issue, projecting a homogenised image of a multitude of ancient chiefdoms, conflating Yamato practice with the rest of the archipelago and vice versa.

How, then, shall we go about this undertaking? The answer: through a careful and close reading of primary textual evidence, taking into account each source’s own limitations, biases, and subjectivities, discerning the beliefs and norms of ancient society through the words that remain to us. Per the work of Friedrich Schleiermacher and Philip Gardner, this reading will be performed via a hermeneutic back-and-forth between text, context, and interpretation8; in particular, the texts will be read using Elizabeth Schüssler Fiorenza’s “hermeneutics of suspicion” and “hermeneutics of resistance”. These latter techniques were devised specifically to address the issue of studying historical women through androcentric source texts, drawing out information on ancient women’s lives and experiences9, a crucial element in our project of restoring and upholding ancient women’s humanity.

We must also keep in mind the fact that most, if not all, of the surviving texts (see Ch. 1.4) display a demonstrably androcentric worldview10. All with known authors or compilers were produced by men. The Kojiki, Nihon Shoki, and Shoku Nihongi overwhelmingly focus on the lives and

7 Mazlish 1998, 91.

8 Mazlish 1998, 91; Gardner 2011, 580.

9 Schüssler Fiorenza 2013, 106–10.

10 This appears through the heavy foregrounding of male figures (emperors, court officials, etc.) compared to female, and the frequent use of male-line ancestries and male relations as a means of defining women.

26 deeds of male actors, with the occasional empress or other royal woman as exceptions. Kinship is defined in largely patrilineal terms, particularly in later entries; women are frequently described in relation to men (Kojiki; Nihon Shoki), and myths call for wifely submission to husbands (Kojiki 1;

Nihon Shoki 1.4; a more detailed accounting of textual androcentrism and positioning of women will follow in Ch. 3). Androcentrism within modern societies also emerges as a concern for scholarship.

Although women’s history as a subfield in Japan has its roots within Meiji academia, these works were scattered and predominantly (until 1913, entirely) male-authored11; historians (women’s and otherwise) are not immune to projecting the patriarchal norms and assumptions of our own contemporary societies. Successive generations of women’s historians have engaged in critique of such normative assumptions within “mainstream” historical narratives12.

Androcentrism – on the part of both ancient sources and contemporary scholarship – can cause the relegation of ancient women to side characters in a historical narrative, not credited with the same internal agency or inner lives as male figures; even discussions of gender and patriarchy can risk reducing these women to passive victims, damsels in distress of the hegemony13, or else overwrite their subjectivities into one-dimensional role models, as Bernard Faure criticises in studies of the Chinese empress Wu Zetian14 (武則天). In order to engage with ancient women’s humanity, we must do our best to understand them not as characters within a story but as the fully realised, complex, actual human beings that they were. Thus, this analysis must take into account women as individuals, as holders of gendered positions, and as a broad social category: looking at once at both priestess (or empress, or nun) and person, avoiding the neat dichotomy of “empowered subject” versus “objectified victim”. It is also important – to avoid oversimplification, or projecting ideals of

11 Takamure 1966a, 5; Anderson 2020, 64.

12 E.g. Takamure 1966a, 6–7; Tsurumi 1982, 75; Yoshie 1993, 452, 462; 2013, 4; Ambros 2015, 14–6.

13 Faure 2003, 3; Meeks 2010, 12, 260; Ambros 2015, 4; Rothschild 2015, 75; DiLuzio 2016, 225.

14 Faure 2003, 189.

27 universal sisterhood onto the past – to be aware of the divisions that existed and still exist within womanhood, such as the intersection of class and gender15.

We must also engage with the complexities inherent within the concept of “power”: rather than a single, easily-quantified property, “power,” as we shall see, encompasses a wide variety of formats, relationships, and interactions within ancient Japanese society. This chapter will thus conclude by outlining various different formats of “power” that arise within our research, and examining the interactions – and tensions – that exist between them. First, though, we must begin by outlining one central concern of our research: what is the “golden age” narrative, and what enduring impact has it had on scholarship?

2.1. Japanese Women’s History and the “Golden Age” Narrative

2.1.1. The Beginnings of the “Golden Age” Narrative

In 1911, Japanese feminist Hiratsuka Raichō opened up the inaugural edition of women’s literary journal Seitō with an essay entitled Genshi, josei wa taiyō de atta16: “In the Beginning,

Woman was the Sun”. This impassioned work began:

In the beginning, woman was truly the sun; an authentic person. Now, woman is the moon.

Living in reliance on others, reflecting the light of others, she is the wan and sickly moon.17

Hiratsuka’s writing spoke to the lost glories of Japanese womanhood, and the hope of regaining spiritual liberation in the future. The likening of ancient womanhood to the sun brought obvious

15 Faure 2003, 8; Meeks 2010, 113; Ambros 2015, 2.

16 元始、女性は太陽であった

17 Hiratsuka 1911, 37. For a transliteration of the original Japanese text, see Chapter 1.1.

28 comparison to the lofty position of the sun goddess Amaterasu as a symbol of ancient female power and splendour18, although Hiratsuka herself later claimed the reference to be unintentional19.

Hiratsuka was a writer, not a historian, and yet the image she painted of primeval female liberation, the ancient period as a glorious golden age of female power before women suffered the thraldom of patriarchy, was an influential one, capturing the hearts of many Japanese feminists: a cadre that included prominent women’s historians such as Takamure, who would later go on to quote

Hiratsuka in her own argument for ancient matriarchy20. But the roots of this debate reach back even further, spreading far outside Japan itself, to the realm of Western anthropology in the late 19th century.

In 1871, American anthropologist Lewis H. Morgan published Systems of Consanguinity and

Affinity of the Human Family, which advanced his theory of universal prehistoric matrilineality for all human societies. Morgan argued that all peoples originally practiced matrilineal kinship, a system which centred women within the clan or gens. The further development of individual property through agriculture, Morgan claimed, lead to a push by the men of each society towards patrilineal inheritance and kinship, which centred men within the family and often resulted in the alienation of women from their natal families upon marriage. Patriarchy thus developed as a result of this disempowerment of women, as well as the strict seclusion, subordination, and relegation to a position of inferiority enforced upon them by men obsessed with guaranteeing the surety of their patriline21. Morgan further developed his theories in his 1877 work Ancient Society22, which was first

18 Takamure 1966b, 9–10.

19 Hiratsuka 2006, 160–1.

20 Takamure 1966b, 9–10.

21 Morgan 1871.

22 Morgan 1877.

29 translated into Japanese in 1924 by Takabatake Motoyuki (高畠素之) and Murao Shōichi (村尾昇

一)23.

In 1884, German sociologist Friedrich Engels built on Morgan’s theories in his similarly influential Origin of the Family, Private Property, and the State24. Engels, a communist and close contemporary of Karl Marx, focused heavily on economics and private property as the site of changing gender relations, similarly situating the switch from a universal prehistoric matrilineality to agricultural patrilineality as the source of patriarchy and female subordination25. Engels’ work was partially translated into Japanese by Sakai Toshihiko (堺利彦) in 1908 as Danjo kankei no shinka,

“The Progression of Gender Relations”26, the translated portions and title centring Engels’ work on shifting gender dynamics and familial/kinship structures, deemphasising his research on state- building.

Nowadays, Morgan and Engels’ theories no longer have the wide acceptance they once did27, and the considerable problems with using their work as a basis for matriarchal argument shall be examined in a later chapter (see Ch. 3.1.1). However, their materialistic view of history had a considerable influence upon the work of early 20th century Japanese historians28, and their arguments for universal prehistoric matriarchy came to serve as a foundation for the “golden age” narrative as pertained specifically to Japanese women29. Nor, too, is their work entirely defunct within the “golden age” narrative of ancient Japanese womanhood; as late as 2003, Morgan and

23 As Kodai shakai (古代社会, “Ancient Society”)

24 In the original German: Der Ursprung der Familie, des Privateigenthums und des Staats

25 Engels 1902.

26 男女関係の進化

27 Ueno 1987, 9.

28 Ijūin and Kawai 2017, 213; Anderson 2020, 67–8.

29 Anderson 2020, 67–8.

Engels’ theories on prehistoric matriarchy and the development of gender relations were applied to the Japanese context by Sekiguchi Hiroko (関口裕子)30.

In Japanese academia, “women’s history” as we know it is largely a product of the early 20th century. Before this, studies of specific female personages – the “exceptional women” of history – existed throughout the Edo period (1603–1868) as part of a broader trend of biographical histories

(列伝, retsuden), but women’s history as a discipline – studying not only the lives of these individual female figures, but also the broader position of women in ancient society – developed as an offshoot of cultural history, itself a new discipline emerging in the wake of the Meiji Restoration of 1868. Prior to this, the study of ancient Japanese womanhood was largely confined to those retsuden that happened to cover the lives of prominent female individuals: empresses, princesses, and the like – with little examination on the lives or social position of women as a whole31. Some such biographies were adapted into teaching materials, used to educate women on proper feminine moral conduct32 in the model of the Chinese Biographies of Exemplary Women33.

Some early studies of Japanese women’s history developed from this biographical trend, with Sudō Kyūma’s (須藤求馬; 1859–1917) three-volume Nihon joshi34, published in 1901, arranging a series of biographies across the eras – the first book focusing on the mythological/foundation era, the second on the age of the imperial court, and the third on the feudal period – appended with background information on contemporary systems and culture35. Like most of academia, however,

30 Sekiguchi 2003, 36–7; Sekiguchi 2003, 40.

31 Takamure 1966a, 4.

32 Takamure 1966a, 4.

33 列女傳, Liènǚ Zhuàn

34 日本女史, “Women’s History of Japan”

35 Takamure 1966a, 4.

31 women’s history would remain the province of male scholars until 191336, with the publishing of

Shimoda Utako’s (下田歌子; 1854–1936) Nihon no josei37. The development of the discipline in these early stages was was hampered by Meiji conservatism and rising nationalism, which brooked little challenge to the “official,” state-sponsored, androcentric narrative38.

In the years following the 1868 Meiji Restoration, furthermore, Japanese scholarship became increasingly concerned with fujin mondai (婦人問題); the “woman question” or “woman problem”: the issue of women’s status within Japanese society. Western nations used the oppression of Japanese women under patriarchy as a pretext to label Japan “uncivilised,” and thereby justify the unequal terms of treaties made. The issue of women’s rights thus became a battleground for Japanese intellectuals desiring parity with the West39.

One early work in this vein was Tatsumi Kojirō’s (辰巳小二郎; 1859–1929) 1887 Seiyō Nihon joken enkakushi40, which is notable for Tatsumi’s argument that, in the pre-Buddhist, pre-Confucian days of Japanese antiquity, women enjoyed equal status to men, and indeed, may have even been predominant. Citing evidence of more egalitarian terms in ancient marriages, as well as the appearance of powerful and respected female figures such as Jingū or Ame-no-Uzume (天宇受売/天

鈿女) in myth and history, Tatsumi argued that women had held considerable status in Japanese society before the influx of Buddhist and Confucian belief systems from the continent introduced patriarchal norms41.

36 Takamure 1966a, 3–4.

37 日本の女性, “Women of Japan”

38 Kidder 2007, 25.

39 Anderson 2020, 64–6.

40 西洋日本女権沿革史, “A History of the Development of Women’s Rights in Japan and the West”

41 Tatsumi 1887, 27–35.

Tatsumi’s work represents one of the earliest depictions of the “golden age” narrative of

Japanese womanhood in academia. The underlying impetus was the aforementioned “woman problem” that loomed over Meiji discourse on gender relations: by situating primeval Japan as an era of gender equality, Tatsumi thus positions women’s rights – and thus, “civilisation” in the

Western eye42 – as an intrinsic quality of native Japanese culture, present from its very beginnings; patriarchy is presented as the product of the continental (and thus, non-Japanese) values of

Buddhism and Confucianism; a foreign influence to blame43.

Building on these narratives of matrilineal prehistory and women’s pre-Confucian rights was historian Sakima Kōei, whose 1926 work Nyonin seiji kō44 further presented the image of ancient

Japan as a time of female splendour and authority; a “golden age” where women were not only respected, but elevated. Sakima argued for the presence, in ancient Japanese society, of not only familial matriarchy (母権, boken) and matrilineality (母系, bokei), but also female rulership (女冶, nyoji). This female rule, in Sakima’s thesis, was legitimated not by violence or strength of arms, as was often surmised of ancient (masculine) kingship, but rather through the magico-religious power45 attributed to women in days of old. Further, it often took the form of a “two-layered” monarchy, the theocratic priestess-queens assisted by a male king, lower in status46.

Sakima’s work thus introduced three key aspects to the “golden age” narrative of women’s power: firstly, the notion of women occupying a central, governing role in ancient Japanese society;

42 Indeed, Tatsumi compares the “equal” status of Izanagi and Izanami favourably with the Abrahamic creation story of Eve (woman) born from Adam’s (man’s) flesh (Tatsumi (1887), 28). As will be covered in Chapter 3.2.3, however, his egalitarian reading of the creator-god couple is not supported by the text.

43 Tatsumi 1887, 35.

44 女人政治考, “A Study of Women’s Governance”

45 霊的能, reiteki nōryoku, or 霊力, reiryoku

46 Sakima 1926.

33 secondly, that woman’s ancient authority and status was mediated through religious activity and the presumption of female spiritual predominance; and thirdly, that sacral women rulers were partnered with male co-monarchs who assisted them in mundane tasks of governance, the foundation to the theory of the “himehiko system” which will be introduced shortly.

Sakima drew heavily in his assertions from both Chinese records of the sacral queenship of

Himiko (see Ch. 4.1), as well as the traditional pairing of high priestess47 and king from the politics of the Ryukyu Kingdom (modern Okinawa Prefecture). The latter is presented within Sakima’s work as something of a crystallisation of ancient Japanese custom, a bridge between modern Japan and its lost prehistoric past48. The positioning of Ryukyuan priestesses as both model for and proof of an ancient Japanese “golden age” appears frequently throughout works of the early to mid- (and even, occasionally, late) 20th century49. The issues posed by this framing will be discussed in depth in

Chapter 2.2.4 below; for now, let us continue with the historiography of the “golden age” narrative.

2.1.2. The “Golden Age” Flourishing: Takamure and Yanagita

The emerging “golden age” narrative, begun in the works of Tatsumi, Sakima, et al, was further developed in the writings of Takamure Itsue: feminist, poet, and women’s historian; a contemporary of Hiratsuka, whose imagery of “woman as sun” is referenced within Takamure’s own studies on the “golden age” 50. In Takamure’s key works – the 1938 Bokeisei no kenkyū51, the 1953

47 In Japanese, the kikoe ōkimi (聞得大君); in Uchinaaguchi (Okinawan), the chifijin-ganashi-mē.

48 Sakima 1926.

49 Takamure 1966a, 69; 1966b, 112–5, 119; Okada 1970, 379–80; Ōi and Miyagi 1974, 17–9; Blacker 1982, 113–

4; Yanagita 1984, 33, 41, 309; Kuratsuka 1996, 17; Piggott 1997, 39.

50 Takamure 1966b, 9–10.

51 母系制の研究, “Studies in Matrilineality”

Shōseikon no kenkyū52, and the 1954 Josei no rekishi53 – ancient Japan is depicted first as a primeval matriarchy, mediated by theocratic women (akin to Sakima’s portrayal of magico-religious nyoji), and later shifting into the aforementioned system of “two-layered” monarchy. This latter partnership, based on the complementary unity of female (hime) priestess-queen and male (hiko) attendant-king, she dubbed the “himehiko system”54, building upon Sakima’s theories of “two- layered” monarchy to further detail the respective duties of hime and hiko and the change in their relationship over time, drawing example pairings from textual records55. The terminology of the

“himehiko system,” and its key assumption of male-female contrapuntal rulership as the standard format of power in ancient Japan, retains a significant presence within the fields of women’s and religious history56; research of later decades tends towards more complex and nuanced models of the hime and hiko, but still cleaving to the key assumptions of the concept57.

Ancient Japan, in Takamure’s work, manifests as a golden age of female power, a society in which women were respected, even celebrated; a far cry from the patriarchy and repression she fought in her own feminist activism. Patriarchy itself is, much as in Tatsumi’s work, positioned as the product of foreign ideologies such as Buddhism and Confucianism, saturating the Yamato court and spread through the latter’s centralisation of power. This, in Takamure’s narrative, leads to the subordination of the hime to her former hiko co-ruler, relegating her to the role of mere consort or

52 招婿婚の研究, “Studies in Uxorilocal Marriage”

53 女性の歴史, “Women’s History”

54 ヒメヒコ制, himehiko sei

55 Takamure 1966a, 1966b.

56 E.g. Tanaka 1996, 104; Matsumura 1999, 124; Maeda 2008, 3; Naoki 2008, 207–8.

57 E.g. Sekiguchi 1987; Piggott 1994, 1997; Kuratsuka 1996; Matsumura 1999; Kidder 2007; Yoshie 2007;

Maeda 2008; Ambros 2015.

35 sacral assistant (such as the saigū58 princess-priestess in remote Ise), before she disappears from the system of kingship entirely59. Takamure was also highly critical of androcentrism and projection of contemporary norms within the field of Japanese history, in particular the assumption of patriarchy as a universal, timeless phenomenon60. This narrative of pre-Confucian ancient Japan as a feminist idyll was undoubtedly linked to her own activism, refuting the idea of patriarchy as an intrinsic part of Japanese society (much like Tatsumi61) and thus serving as symbol of hope for contemporary feminists that the unnatural constraints of patriarchy could be removed and this state of female empowerment once again attained62 (much like Hiratsuka63), a nostalgic image of the distant past to provide inspiration for contemporary struggles64.

Unlike Engels and Morgan, however, who wrote of ancient matriarchy in universal terms,

Takamure positions this reverence for women as an almost uniquely Japanese phenomenon, contrasting reverence for the solar goddess Amaterasu with the patriarchy reflected in Christian,

Chinese, and Greek beliefs65 (see Ch. 3.1.2 for more on Amaterasu). In one sense Takamure’s work challenged the patriarchal nationalism of the pre-war Japanese state, presenting an alternative narrative to its relentless androcentrism. In another sense, however, through emphasising the nativity of female empowerment against the foreignness of patriarchy, through making this space of untarnished, authentic “Japaneseness” the site of glorious return, Takamure nevertheless replicated

58 斎宮,”Abstention Palace”, the Amaterasu-worshipping Sacral Princess of Ise; to be covered in more detail in

Chapter 4.3.

59 Takamure 1966a, 1966b.

60 Takamure 1966a, 1966b.

61 Tatsumi 1887, 27–35.

62 Anderson 2020, 66.

63 Hiratsuka 1911.

64 Tannock 1995, 454–5; Abrams 2005, 34; Pickering and Keightley 2006, 920–3.

65 Takamure 1966a, 1966b.

36 the nationalist dichotomy of the native-authentic-good against the foreign-artificial-bad; a nativist- feminism combining the fight against sexism with the nationalist bogeyman of the Other66. The

Japanese exceptionalism and nativist-feminism present in Takamure’s work – and in particular, her wartime support for the imperial regime – have left her with a complicated and controversial reputation within modern scholarship67.

The “golden age” narrative championed by Takamure, Sakima, and the like was not the only reading of Japanese womanhood in history. The 1946 work Nihon joseishi68, written by a contemporary of Takamure’s, Inoue Kiyoshi (井上清; 1913–2001), presented a more clear-cut narrative of gradual liberation over time, absent a “golden age”69. A similar image was conveyed in

Shiratori Kurakichi’s (白鳥庫吉; 1865–1942) 1935 examination of Chinese sources on Queen Himiko,

“Wa joō Himiko kō”70, which asserted the patriarchal nature of ancient Japan and Himiko’s status as an “exceptional woman” rather than a female leadership archetype71. This alternative narrative – of patriarchy as a constant force throughout Japanese history – was strenuously rejected by Takamure, who attributed it to a projection of historian’s own contemporary androcentric norms72.

66 Ryang 1998.

67 Ryang 1998; Anderson 2020, 66.

68 日本女性史, “Japanese Women’s History”

69 Anderson 2020, 67.

70 倭女王卑弥呼考, “A Study of Himiko, Queen of Wa.” Published in Tōa no hikari (東亜の光, “The Light of

East Asia).

71 Shiratori 1976.

72 Takamure 1966a, 6–7.

Another significant contributor to the development of the “golden age” narrative was renowned folklorist Yanagita Kunio, via his 1940 work Imo no chikara73. Here, Yanagita detailed the protective spiritual power accredited to women in ancient Japan and Ryukyu, from which, he argued, sprang the historic and folkloric pairing of brother-king and sister-priestess74. Unlike the female- centric social order put forth by Sakima and Takamure, Yanagita’s proposed image of ancient

Japanese rulership still held male kingship in pride of place, with priestesses as assistants and providers of magico-religious support rather than theocratic queens75. However, his writings on ancient female spiritual predominance and the ruling partnership of sacral female and secular male harmonised enough with Sakima’s nyoji and Takamure’s himehiko that they became incorporated into the broader discourse on the ancient power of women and priestesses, bolstering the narrative of the “golden age.”

The rise of second-wave feminism in the West and ūman ribu (“women’s lib”) in Japan saw a growing interest in women’s history as an academic discipline throughout the 70s and 80s76, bringing increasing engagement with Takamure’s work and the “golden age” narrative as a whole. The introduction to the 1974 Nihon joseishi77, edited by Ōi Minobu (大井ミノブ) and Miyagi Eishō (宮城

栄昌), directly referenced Takamure and replicated the “golden age” narrative, including her theories on the himehiko system and the identification of Ryukyuan practices with ancient Japan78. A number of religious histories were also produced during the post-war decades engaging with female

73 妹の力, “The Power of Women”; “imo” here is used in an archaic sense which can refer to wives as well as younger sisters

74 Yanagita 1984.

75 Yanagita 1984.

76 Anderson 2020, 70.

77 日本女性史, “Japanese Women’s History”

78 Ōi and Miyagi 1974.

38 religious roles and sacral power. Robert Ellwood and Okada Seishi (岡田精司) explored the role and history of the Ise saigū “princess-priestess”79, whereas Hori Ichirō (堀一郎) and Carmen Blacker examined female religious roles and shamanism in both ancient history and contemporary folk tradition80. The former works were entirely focused on central (royal) female religiosity, whereas the latter’s study of regional religious variation was limited to the modern context.

2.1.3. The “Golden Age” in Modern Scholarship

A new wave of the “golden age” narrative emerged in the late 80s, with the 1987 publication of a third, separate Nihon joseishi81, edited by Wakita Haruko (脇田晴子), Hayashi Reiko (林玲子), and Nagahara Kazuko (永原和子). The works contained therein presented revised images of this

“golden age” and the contrapuntal duties of hime and hiko, bringing increasing complexity to the concept. In “Himiko kara jotei e”82, Sekiguchi Hiroko refuted the previous division of hime and hiko into “sacral female and political male” (in the vein of Yanagita), while still supporting the broader theory of the “himehiko system” itself. Sekiguchi’s work also raised the issue of regional variation in women’s history, noting that the diffusion of patriarchal codes appeared slower in ancient Japan’s peripheries than in its centre83.

Yoshie Akiko’s (義江明子)1987 “Megami no botsuraku”84 similarly offered a more complex image of himehiko rulership and the “golden age” narrative, proposing a more fluid division of duties for hime and hiko, and positioning women’s religious involvement not as a complementary function

79 Ellwood 1967; Okada 1970; Ellwood 1972.

80 Hori 1968; Blacker 1982.

81 日本女性史, “Japanese Women’s History”

82 卑弥呼から女帝へ, “From Himiko to Empresses”

83 Sekiguchi 1987.

84 女神の没落, “The Fall of the Goddess”

39 with men’s mundane governance, but rather as a by-product of their increasing political disempowerment85. This argument was reiterated in her 2007 work Nihon kodai josei shiron86, along with a criticism of scholarship romanticising this political disenfranchisement under the misapplied notion of “balance”87.

Previously neglected in Anglophone academia, the study by Western scholars of Japanese women’s history emerged as a growing field throughout the 80s and 90s, allowing for cross-cultural conversations between Western and Japanese academics, and bringing the new Western turn towards “gender history” into Japanese academia88; one example being the 1994 collaborative work

Jendā no Nihonshi89, edited by Wakita and Susan B. Hanley. Joan R. Piggott’s contribution,

“‘Himehiko’ to ‘himemiko’: jendā hairākii to kokka keisei ni kansuru kōsatsu”90, as well as her 1997 work The Emergence of Japanese Kingship, further continued this modified “golden age” narrative, broadly accepting many of its key components – such as himehiko rulership, or the notion of ancient

Japan as an era of comparative female power and prominence – while introducing more nuance and complexity in its depictions of how female power manifested, or how contrapuntal rulership was structured91.

The incorporation of gender studies into examinations of women’s history has also provided new perspectives to the discourse on female divinities in myth. Tanaka Takako’s (田中貴子) 1996 Sei

85 Yoshie 1987.

86 日本古代女性史論, “A Historical Treatise on Women in Ancient Japan”

87 Yoshie 2007.

88 Anderson 2020, 71–3.

89 ジェンダーの日本史, “A Japanese History of Gender”

90 「ヒメヒコ」と「ヒメ王」ジェンダー・ハイラーキーと国家形成に関する考察, “‘Himehiko’ and

‘himemiko’: An Examination of Gender Hierarchy and National Formation”

91 Piggott 1994, 1997.

40 naru onna: saigū, megami, Chūjō-Hime92 (中将姫) and Matsumura Kazuo’s (松村一男) 1999 Megami no shinwagaku: shojo boshin no tanjō93 discuss the gendering of deities and what inferences about gender roles and attitudes can (and cannot) be inferred from goddess-myths94, their work standing as a challenge to Takamure’s assertion that the honoured position of the female deity Amaterasu reflects a similar reverence for mortal women95. Tanaka’s work also reclaims the saigū’s humanity through engaging not only with the historical status and function of the role, but also the inner lives and emotions of the women who occupied it96. A similar approach to reconstructing the subjectivities of historical women alongside the details of their post can be found in Bernard Faure’s

2003 The Power of Denial: Buddhism, Purity, and Gender, Lori R. Meeks’ 2010 Hokkeji and the

Reemergence of Female Monastic Orders in Premodern Japan, and Tokoro Kyōko’s (所京子) 2017

Saiō kenkyū no shiteki tenkai: Ise saikū to Kamo saiin no sekai97, devoting space to dealing with ancient women as actors within a system rather than as passive, voiceless victims of the hegemony98.

The last few decades have also seen a greater push towards interdisciplinary approaches engaging with intersectionality such as the relationship between gender and class, or gender and race99. Faure’s work also engages with the former point, reiterating the need for scholars to understand the ways in which class groups and family ties mould the power relations of patriarchy100,

92 聖なる女―斎宮・女神・中将姫, “Sacred Women: Saigū, Goddess, Chūjō-Hime”

93 女神の神話学―処女母神の誕生, “Mythography of Goddesses: The Birth of the Virgin Mother”

94 Tanaka 1996; Matsumura 1999.

95 Takamure 1996b, 7–10.

96 Tanaka 1996.

97 斎王研究の史的展開伊勢斎宮と賀茂斎院の世界, “Historical Development of Saiō Research: The World of the Ise Saigū and Kamo Saiin”; saikū is an alternate transliteration of 斎宮 (saigū) without rendaku

98 Faure 2003; Meeks 2010; Tokoro 2017.

99 Watanabe 1994, 75.

100 Faure 2003.

41 and Meeks’ with class as a factor in women’s mediaeval engagement with Buddhism101. The former’s work, however, has been criticised for a tendency to essentialise Buddhism into a monolithic force, and to characterise women as ‘for’ or ‘against’ patriarchal norms in a projection of modern expectations102. Yoshie’s 1993 article “Gender in Early Classical Japan: Marriage, Leadership, and

Political Status in Village and Palace,” published in Monumenta Nipponica, examined female authority and economic status at the rural village level, as well as at court103. Barbara R. Ambros’

2015 Women in Japanese Religions, a broad overview of women’s religious performance in Japan throughout history and against a changing backdrop of gender roles and customs, critiques the

“golden age” narrative for essentialising a diverse variety of ancient women’s experiences. Ambros further specifies class strata as a potential site of difference, and even brings up the disparate nature of ancient Japanese culture and rulership104. Rajyashree Pandey’s work not only engages with disparities in gendered experience, but also tackles the definition and understanding of gender as a social category within the ancient Japanese context105. Studies of folk religion, such as those of Hori and Blacker, also engage with regional diversity in religious tradition and female religious performance, although regional folk customs such as the itako of northern Honshu are primarily discussed in the modern, rather than historical, context106.

The majority of our surviving textual sources were centrally produced – in many cases, imperially commissioned – reflecting the norms and viewpoints of the Yamato (and later, Heian) capital and court. This overrepresentation of elite central voices can lead to their being taken as representative of ancient Japan in its entirety, homogenising the various polities, chiefdoms, and

101 Meeks 2010.

102 Pandey 2004, 240.

103 Yoshie 1993.

104 Ambros 2015, 2–3.

105 Pandey 2004; Pandey 2020.

106 Hori 1968; Blacker 1982.

42 even ethnic groups of the archipelago into a singular “Japanese culture”. Sekiguchi touches upon this issue by noting a discrepancy between central and regional women in attainment of official ranks and positions, with regional female elites attaining bureaucratic appointments well after the last such were given to their central counterparts107. Although this thesis will be focusing primarily on region as a site of difference, it is also important to note that social class and status – clan membership, family ties, and the like – also serve as a major point of variance in ancient women’s experiences. The viewpoints we see from Yamato are primarily those of the Yamato elite, and our understanding of the lives and attitudes of commoner women of Yamato is far hazier than that of the relatively prominent noblewomen. Our sources on regional women, too, tend to focus on local elites such as chieftain clans, and are not themselves reflective of the full range of womanhood in ancient Japan.

Many works of the last few decades do engage critically with the “golden age” narrative and its idealised image of ancient womanhood. Indeed, Haruko Okano expresses uncertainty on the existence of such an age108, and Barbara R. Ambros challenges the notion of a neat line from empowerment to oppression and back again entirely109. Nevertheless, the legacy of the “golden age” endures, surviving in threads running through the ever-shifting historical narrative. Discourse on the division of duties between hime and hiko, for example, still takes for granted the existence of the

“himehiko system” as a standard format of rulership in the first place (see Ch. 4.3). The goal of this research is to build on this critical re-examination of the “golden age” narrative through examining these enduring “threads” and concepts, with particular attention to regional disparity and diversity.

107 Sekiguchi 1987, 18–9.

108 Okano 1993, 27.

109 Ambros 2015, 2.

2.2. Methodology

Our exploration of the “golden age” narrative and its legacy-threads will see us making a close study of many primary historical texts in a process which Bruce Mazlish describes a dialogue between text and interpretation110: studying text in light of the “golden age” narrative, and “golden age” narrative in light of the texts. So what strategies for reading the text will serve us best in this endeavour? For this, we turn to the interpretive methodology of hermeneutics.

2.2.1. History and Hermeneutics

Hermeneutics, although first developed as a method of biblical exegesis, has also been adapted as an approach for the reading of history and historical texts. Friedrich Schleiermacher’s theories stress the existence of the “hermeneutical circle” of reading texts: the whole can only be understood in terms of its parts, and the parts can only be understood in terms of the whole, and thus the hermeneutical approach to reading a source work demands a constant back-and-forth between text and context111. Extrapolated to the reading of historical works, this requires us to be conscious not simply of the specific work or passage we seek to interpret, but also its various contexts: within (if applicable) a broader body of work; within the circumstances of its production and dissemination; within the point in culture and history that surrounded it at its inception. It is also crucial to engage with the varying interpretations of other readers, other scholars, which have accreted over the years, interpretations formed from perspectives and angles far different to our own. This allows us to refine and challenge our own interpretations, and broaden the limits of our own understanding112. As Philip Gardner puts it, the historian must approach a text as we would a

110 Mazlish 1998, 91.

111 Mazlish 1998, 91.

112 Mazlish 1998, 91; Bell 2011, 537.

44 person whom they are gradually coming to know, at once identifying the text as a constant entity while also coming to new understandings and new ways of looking at it113.

The vast majority of surviving texts that detail the early ancient period were produced by male authors and compilers, and reflect an androcentric worldview114, focusing heavily on the actions and deeds of ancient men. Women appear often as passing mentions; although prominent female personages exist, such as Empress Jingū or Yamato-Hime, these women are in the minority, and typically occupy elite positions, representing only a portion of the sum total of ancient Japanese womanhood. It can be difficult to reconstruct ancient women’s experiences and lives through texts which consistently overlook them. This research will therefore be adopting Elizabeth Schüssler

Fiorenza’s theories of feminist hermeneutics, devised specifically for the sub-discipline of women’s history. Schüssler Fiorenza breaks feminist hermeneutics into two categories: first, a “hermeneutics of suspicion” which analyses ideologies of gender within the source text, including underlying assumptions as well as explicitly-stated beliefs115; second, a “hermeneutics of resistance” which reads against the grain of these biases, redirecting the focus to women’s perspectives and actions – what is known, and what can be inferred – in defiance of a text’s own androcentrism116. Through these strategies, we may work to reconstruct the lives and experiences of ancient women in a far more comprehensive sense than the few glimpses of female elites afforded to us by a surface reading.

113 Gardner 2011, 580.

114 Shiratori 1976, 178.

115 Schüssler Fiorenza 2013, 106.

116 Schüssler Fiorenza 2013, 110.

2.2.2. Reading Texts in Context

The back-and-forth between text and context demanded by our hermeneutical reading requires that our primary sources be read with their proper context in mind: when were they produced; by whom; to what purpose117? One important context is authorial agenda: works such as the Kojiki and Nihon Shoki were commissioned by the imperial court, and so the images of ancient

Japan they present are affected by the narrative that the Yamato hegemony desired to propagate; one which affirmed Yamato dominion over the archipelago and centred Yamato tradition and belief.

Other issues of context also arise: it is crucial for us to consider, for example, the question of which voices are (and are not) reflected in a particular text, that we may avoid perpetuating the very homogenisation of ancient Japanese culture and women’s experiences that this research seeks to critique. Relatedly, we must also take into account the limitations imposed by the context of a work, such as the limits of the authors’ own knowledge or understanding of the subject. This consideration is particularly vital when reading a text detailing a culture foreign to the authors (such as the Records of the Three Kingdoms’ depiction of Yayoi Japan through the eyes of Chinese emissaries), or else a period of history ancient even at the time of the text’s writing (such as the earlier portions of the

Kojiki and Nihon Shoki, which describe events attributed to centuries before the works’ early eighth- century composition). It is important that we outline how we will deal with these issues of context.

First, let us examine the issue of historicity. The Kojiki and Nihon Shoki detail the history of ancient Japan from the creation of the cosmos, through the beginnings of the Yamato dynasty

(dated within the text to the seventh century BCE) up until the seventh century CE. The texts themselves, however, were only published in the early eighth century CE, casting doubt upon the reliability of their earlier historical records. This issue is further muddied by the improbably long

117 Mazlish 1998, 91.

46 reigns and life spans displayed by some of the earlier emperors, the elusive kesshi hachidai118, and discrepancies between the Kojiki and Nihon Shoki accounts and Chinese observations. In Japanese history, the reliability which can be afforded to these early records remains the site of considerable debate119, and so, treating with these texts requires careful deliberation.

Many historians have engaged with this dilemma of historicity by situating a specific imperial reign as a point of demarcation between (relatively) reliable later records and uncertain earlier accounts. Ellwood, for example, designates this point to be the reign of Emperor Kinmei120 (欽明), whereas Yoshimura Takehiko (吉村武彦) places it a little earlier with the fourth-century “Five Kings of Wa”121. But the debate is so interminable, its waters so muddied, that trying to narrow down a single emperor’s reign as the beginning of reliable records, trying to establish a clear point of demarcation where “myth” can be said to end and “history” to begin, may ultimately be a futile effort.

The earlier portions of the Kojiki and Nihon Shoki depict a mythologised history of Japan, a narrative based on the legends of creation and foundation circulated within and promulgated by the

Yamato elite, canonised as the official history of the Yamato dynasty and its kingdom. These myths played a social and political role in normalising and justifying existing customs, institutions, and power relationships122, such as the divine right of the emperors, or the traditional occupations of

118 欠史八代; the “Eight Generations Missing from History;” the eight emperors from Suizei to Kaika (reigns cumulatively dated from 581–98 BCE) who are now commonly believed to be later constructions (Yoshimura

2010, 35–8).

119 Yoshimura 2010, i.

120 R. 539–571 CE (Ellwood 1967, 40).

121 Five kings of Japan referenced in the Chinese Book of Song from 421 to 478 CE, generally identified as

Yamato-dynasty monarchs (Yoshimura 2010, 46).

122 Mazlish 1998, 115.

47 various clans. At the same time, however, myths also represent a means by which a society understands itself, its history, its values, and its place in the cosmos. To assume that the ancients viewed history in the same light of post-Enlightenment empiricism that we do, and deliberately obfuscated it with myth, is to perpetuate anachronisms and to fail at historical empathy123. Thus, our hermeneutical approach to the reading of myth will require a careful consideration of context, as well as attention to variation in different accounts of the same legend, in order to explicate the relationship between political utility and faith contained therein.

The recognition that myth presents a different way of understanding the past than the empirical “historicity” underpinning modern scholarship presents a further contextual nuance to incorporate into our reading. As mentioned above, early portions of the Kojiki and Nihon Shoki occupy a murky state of uncertainty that in many places is best understood as “myth-history”. In our hermeneutical reading of these accounts, we will draw from Mary Beard’s research124 on the myth- history of Ancient Rome: that our key concern should not be the attempt to extract a core of historical “fact” from a coating of myth; a process which, Beard argues, tends towards assumptions about what is “fact” or “fiction”. Rather, we should seek to understand the meaning conveyed by the myth; the perspectives, worldviews, and senses of identity that it reflects125. In the Japanese context, for example, Yoshimura’s analysis of the myth-history of Empress Jingū and her conquest of the Korean kingdoms of Baekje and Silla identifies the core “meaning” of this myth to be the assertion of Yamato’s rightful dominion in, and right to expect tribute from, the Korean Peninsula, rather than a literalist reading of a warrior-queen’s exploits126.

123 Mazlish 1998, 115; Naoki 2008, 208; 2009, 223.

124 Beard 2015.

125 Beard 2015, 71–3, 86.

126 Yoshimura 2010, 72.

The political statement embedded in Jingū’s tale reminds us of another important point: these “official histories” – the Kojiki, Nihon Shoki, et al. – cannot be considered impartial, unvarnished records even in their most chronologically reliable portions. They were, after all, commissioned by imperial order, compiled by central court noblemen, and they serve to transmit a narrative of history that it suited the Yamato court to propagate: one which, among other things, justified the paramount’s rule through the establishment of a continuous, unbroken line in receipt of divine mandate, endowed with droit divin127.Through these texts, the sphere of Yamato dominion is constructed as an imperial state, the epicentre of a world-spanning ritual and social order with the paramount-emperor at its heart128.

Yamato authority through Jinmu129 (神武) is cast as a direct follow-up to the primeval Age of

Gods130 (Kojiki 2.Jinmuki; Nihon Shoki 3). The conquests of the Yamato sovereigns are portrayed as swift, thorough, wide-reaching and, crucially, early; the realm is quickly unified in fealty, save for a few rebellious holdouts to be put down every other reign. This narrative served to perform what

Shmuel Noah Eisenstadt terms the “continual reconstruction of primordiality,” the process of solidifying a communal identity through affirming and reaffirming its antiquity, even as its boundaries change: a process which thereby homogenises all within131. Beard identifies a similar process as a common component of Roman myth-history, projecting later understandings of

“Roman culture” and “Roman identity” onto its distant past, reifying these concepts through the

127 Martin 1997, 14; Naumann 2000, 48; Yoshimura 2010, i; Ambros 2015, 22.

128 Duthie 2014. Duthie’s monograph contains a detailed explanation of the ways in which the Yamato court constructed their imperial self-image and projected power.

129 Legendary founding emperor of Japan; reign traditionally dated 660–585 BCE

130 神代, kamiyo or shindai; encompassing in the records the period from the beginnings of the cosmos to the birth of Emperor Jinmu and his conquests (Kojiki 1; Nihon Shoki 1–2).

131 Eisenstadt 2002, 37–62.

49 appearance of continuity132. Regional perspectives and diversities are thus subsumed into a central narrative which anachronistically projects the image of a singular, culturally and politically unified

“Japan” onto a far more fractionalised and disparate past133. When reading historical sources, we must be aware of the authorial perspectives and biases that filter through into the text, and consider how these inform our interpretation of the contents. To distinguish these, the hermeneutical approach recommends reading these sources with a critical eye, contrasting them against other sources (archaeological records, Chinese texts, etc.) where possible134: an especially important consideration when attempting to find the unvoiced perspectives of regional difference within a text.

Thus, this research also treats with Chinese sources depicting ancient Japan (see Ch. 1.4).

These works have the benefit of being older: the depiction of Himiko’s queendom in the Records of the Three Kingdoms was published scant decades after the events it portrays; whereas the Kojiki and

Nihon Shoki accounts were compiled centuries later. The Chinese writers’ position as external observers, less invested in the construction of a specific narrative for Japanese history, further confers an appearance of impartiality onto the text135. This does not mean, however, that a critical reading is unnecessary. For one thing, these works come from the perspective of outsiders, who did not necessarily have a full understanding of the society they described. Japan was framed in these records as a distant, barbaric backwater136, politically and culturally inferior, such that it is treated as an affront for the Japanese sovereign to dare place themself on an equal footing with the Sui emperor (Sui Shu 81, Woguozhuan). Clearly, the Chinese court records are hardly without their own

132 Beard 2015, 93–6.

133 Yoshimura 2010, 6.

134 Piggott 1997, 3.

135 Ambros 2015, 17.

136 “Wa” was classified as part of the “Eastern Barbarians” (東夷, Dōngyí), one of the “Four Barbarians” groupings used by the ancient Chinese to categorise non-Chinese peoples. Other “Eastern Barbarians” included the peoples of Ryukyu, Korea, and Manchuria.

50 biases and preconceptions137. Thus, the “hermeneutical circle” between text and context138 is just as crucial here as with Japanese sources.

Our hermeneutical reading must also take into account the issue of survivorship bias in regards to the available records. We as historians are dependent on the selection of texts that survive into the present day, which inform much of scholarship139. As a result, our perceptions and understandings of history will always be skewed by the limitations of what is available. What survives (and what does not) is not entirely haphazard, either; it is no coincidence that most of the best-preserved works from this period were records produced within, and often commissioned and/or propagated by, the central government, whereas regional works are patchier in their survival, making regional (and marginal) voices harder to reconstruct.

One must also consider the question of whose voices are absent in text altogether. The production of texts in itself was typically associated with the upper classes, and so even if every single ancient document was miraculously preserved to the present day, it would still represent only a handful of perspectives on history. This imbalance is felt particularly keenly in fields such as women’s history whose foci are typically on the margins, and thus, with the exception of elite women with positions at court, unlikely to have their voices transmitted to future generations140. It is this inequity of evidence that necessitates the use of Schüssler Fiorenza’s “hermeneutics of suspicion and resistance”141 (see Ch. 2.2.1) to “read between the lines” of what does survive and recover the hidden voices of women in history. Although we cannot analyse works that do not survive the ages, or were never written in the first place, we can still keep in mind, when engaging

137 Maeda 2008, 8, 25.

138 Mazlish 1998, 91.

139 Mazlish 1998, 125.

140 Munslow 2000, 228–230.

141 Schüssler Fiorenza 2013, 106–10.

51 with the texts that remain, that they represent only a portion of what there is to know, and not the sum total of history itself. By remembering the specific context of the texts we analyse, we can avoid the accidental homogenisation that can result from a limited selection of sources, and restore as many ancient women’s perspectives as we can to the historical narrative.

2.2.3. Reclaiming the Humanity of Historical Women

Such restoration is a key concern of the academic discipline of women’s history. With our surviving records mostly revolving around the lives and deeds of men, it is perhaps no surprise that the scholarly orthodox – aside from “golden age” narrativists such as Sakima or Tatsumi – has tended to follow suit. The result is a heavily androcentric historical narrative in scholarship, in which women appear mostly as background figures, passive. When women are present as influential figures, history-makers, they are often framed as “exceptional women” – the likes of Empresses

Jingū, Suiko142 (推古), Kōken-Shōtoku143 – going against the grain of gendered expectations and occupying the monarchic roles more typically associated with men. It is this aberration from the gendered norms that “earns” them their place in the historical narrative; the implication thus being that the lives of most women were too unimportant, too inconsequential to be worthy of study. In centring the actions and experiences of women as a whole, and their relationship to power and the state, this research challenges androcentric scholarship and thus situates itself within the field of women’s history144.

Indeed, the relation of women to rulership is of particular interest to women’s history, as one oft-stated goal of the discipline is the critical re-examination of the existing scholarly narrative, with an eye to challenging – or, at least, reconsidering – previous normative assumptions of female

142 R. 592–628 CE

143 R. as Kōken 749–758 CE, again as Shōtoku 764–770 CE

144 Munslow 2000, 228–230.

52 political non-involvement145, as well as reclaiming histories and figures that have gone overlooked by previous scholarship146. Part of this process involves the reconceptualisation of women under patriarchy from passive subjects – the objectified victims of subjugation, “damsels in distress” of the hegemony – to actors within a system that disadvantaged them, but who held their own desires and aspirations and inner lives, and who could subvert, work within, toe the line, defy, or even propagate the patriarchal framework in which they existed, as fit their own beliefs and goals147. It is particularly important for Western scholars writing on Asian women’s history to be mindful of this, rejecting the

Orientalist stereotype of Asian women as passive, uncomplaining victims of patriarchal subordination, the adherence to which trope has been criticised by Meeks and Ambros148. As Pandey reminds us, however, it is also important to avoid reducing such a complex concept as “agency” to a simple matter of presence versus absence149. It is also important for us to be aware of how culture shapes perspective and outlook, and to keep in mind that “agency” as understood in ancient and mediaeval Japan differs from our own Western, liberal conception of it, including as actors such forces as gods and karmic destiny150.

Arguably, the restoration of women’s experiences and women’s subjectivities to the historical narrative constitutes a feminist act, and so it should come as little surprise that women’s history as a discipline has often found itself heavily connected with feminist movements. As we have seen, the beginnings of women’s history as a discipline in Japan intertwine with Meiji anxiety over contemporary women’s rights, the “woman problem”; the “golden age” narrative first appears as an

145 Takamure 1966a, 6–7; Yoshie 1993, 452; 2013, 4; Ambros 2015, 14–6.

146 Munslow 2000, 227.

147 Faure 2003, 3; Meeks 2010, 12, 260; Ambros 2015, 4; Rothschild 2015, 75; DiLuzio 2016, 225; Anderson

2020, 65.

148 Meeks 2010, 12; Ambros 2015, 4.

149 Pandey 2020.

150 Pandey 2020.

53 appeal to antiquity, an image of the past supporting the cause of women’s rights in the present151.

Takamure herself was active in the feminist movement well before she penned her first historical work, her own deployment of the “golden age” narrative serving as a symbol of hope for contemporary women’s movements by challenging the image of patriarchy as timeless and inherent to society152. Her feminist contemporary Hiratsuka drew from that same image of Japanese womanhood’s ancient glories as an inspiration to the modern movement; feminism-as- reclamation153. Nor is it any coincidence that the rise of women’s history as an academic discipline in the West – dedicated to the task of “restor[ing] women to history” – coincided with the height of second-wave feminism in the 1970s154.

There are, however, certain concerns that must be kept in mind in our reading of history through a feminist lens. First of all, we must take care not to project modern conceptions and ideologies anachronistically onto an ancient society155. For example, we must not assume that ancient women were united in some shared sense of sisterhood in response to male oppression.

Ancient Japan had its own share of complex intersections between identities, and two women of separate classes, for instance, may have seen more divisions between each other than commonalities: as Meeks puts it, noblewomen had a “cultural tendency to view themselves as aristocrats first and women second”156.

Another issue is that of the “role model.” We have seen in Hiratsuka and Takamure how images of past female strength and glory can inspire women in the present, be used as rallying cries

151 Tatsumi 1887; Anderson 2020, 63–4.

152 Takamure 1966a, 1966b; Anderson 2020, 66.

153 Hiratsuka 1911.

154 Anderson 2020, 70–1.

155 Faure 2003, 5.

156 Faure 2003, 8; Meeks 2010, 113.

54 and symbols of empowerment by feminist movements157. Indeed, research on the social function of nostalgia suggests the use of idealised images of the past as a constructive force in tackling present concerns and dissatisfactions158. At the same time, however, the project of “restoring women to history” ultimately demands that we take ancient women – their lives, and their experiences – as they were, rather than as we would like them to be159. For our goal, after all, is to restore and maintain the humanity of ancient women, understanding them as actual people endowed with the full range of human feeling and expression, rather than parade of silent bit-parts and victimised objects. To force these women into the narrative moulds of “role model” and “inspiration” is itself an act of objectification, a denial of full humanity that does a disservice to these ancient women. Our key concern, in the evaluation of the “golden age” narrative, is how well it reflects the experiences of historical women, for it is through understanding their circumstances and subjectivities that their humanity may be reclaimed.

2.2.4. Ryukyu as Model for Ancient Japan

In their attempts to understand the lives and experiences of ancient Japanese women, many early scholars turned to the religious and political position of women in mediaeval Ryukyu. The

Ryukyu Kingdom, active as a unified state from 1429 (with local fiefdoms stretching back well before this) until its annexation by Japan in 1879 (after which it became the modern Okinawa Prefecture), was both nearby to Japan and also noted for the prominent role taken by women in its religious life, with an emphasis on female practitioners that persists to this day160. Even prior to unification, the political organisation of Okinawa (or Uchinaa, in the Okinawan language of Uchinaaguchi) viewed feudal lords’ receipt of support from sister-priestesses as crucial to a lord’s maintenance of power,

157 Hiratsuka 1911; Takamure 1966a, 1966b; Munslow 2000, 227; Anderson 2020, 66.

158 Tannock 1995, 454–5; Abrams 2005, 34; Pickering and Keightley 2006, 920–3.

159 Faure 2003, 5, 189.

160 Sered 1999, 4.

55 and under the Ryukyu Kingdom, the centralised state hierarchy of priestesses was headed by the king’s own sister, the chifijin-ganashi-mē161 or kikoe-ōkimi162, who provided him with spiritual power and legitimacy, and who may have enjoyed some level of direct power of her own163. It is not hard to see where Ryukyu’s religious women may invite comparison with Japan’s own traditions, especially between the roles of kikoe-ōkimi and saigū as royal kin-priestesses.

The problem arises when these discussions on Japan and Ryukyu go beyond the realm of simple comparison. To theorise some relation between Ryukyuan and Japanese traditions is hardly, in itself, outlandish or objectionable: the two cultures are in close enough geographical proximity that some common roots to their beliefs would be entirely unsurprising. However, early scholarship in Japanese women’s history – including such prominent figures as Takamure and Yanagita – took this even further, presenting Uchinaa religion and the traditions of the Ryukyu Kingdom as some crystallised holdover of pre-patriarchal Japanese culture, from which insight into Japan’s own history could be gleaned164. Various contexts surround this presentation: the Meiji state’s annexation of

Ryukyu and Hokkaido into its constructed image of a modern, unified “Japan”; the early intelligentsia’s concern with articulating Japanese national identity following the upheavals of the

Restoration165; and the nationalistic discourse of the early 20th century, to which Takamure was a contributor166 . The depiction of Ryukyu culture as a preservation of ancient Japan thus arguably

161 Uchinaaguchi: 聞得大君加那志前. Lebra suggests that chifijin may be cognate with Japanese kifujin (貴婦

人, “noble lady”); ganashi and mē are very polite honorifics used for kami and high-ranking humans (Lebra

1966, 105).

162 Japanese: 聞得大君, the characters translating to something like “monarch/lord who can hear”.

163 Lebra 1966, 101–14.

164 Takamure 1966a, 69; 1966b, 112–5, 119; Yanagita 1984, 33, 41, 309.

165 Anderson 2020, 64–6.

166 Ryang 1998; Kidder 2007, 25; Anderson 2020, 66.

56 served as an act of “reconstructing primordiality”167, conflating Ryukyuan culture with ancient

“Japaneseness” in order to affirm its (forcible) inclusion in the Japanese Empire.

Nevertheless, the tendency to use Ryukyu as a model for prehistoric Japan would continue even into the post-war years168, the Ryukyuan kikoe-ōkimi still being portrayed within scholarship as a model or prototype for the Japanese saigū169. The issue here is that the kikoe-ōkimi is not the saigū, just as the Ryukyu Kingdom is not some perfectly preserved image of ancient Japan, forever ossified, but rather a distinct nation with its own unique history and traditions. Even if the two cultures do share common roots, they have both undergone change over time; simply in different ways, as Faure reminds us170. While it is certainly possible to draw parallels between the position of women in

Japanese and Ryukyuan religion, it is also important to take note of aspects in which the two differ171, and it is ultimately both reductive and condescending to Ryukyuan culture and religion to treat with it as nothing more than a fossilised version of ancient Japan. If the two belief systems do indeed share a common origin, then they are best understood as traditions that diverged and evolved differently, rather than one as simply a more archaic form of the other. Our redeveloped narrative of women’s power in ancient Japan must be done with respect to the indigenous traditions of Ryukyu, rather than positioning them as simply holdovers from an earlier period of Japan’s history.

2.2.5. Defining Power

Another important issue with which our redeveloped narrative of women’s power must grapple is the definition of “power” itself. We have seen that the “golden age” narrative situates

167 Eisenstadt 2002, 37–62.

168 Ōi and Miyagi 1974, 17–9; Blacker 1982, 113–4; Kuratsuka 1996, 17; Piggott 1997, 39.

169 Okada 1970, 379–80.

170 Faure 2003, 290.

171 Yoshie 1987, 35.

57 ancient Japan as an era of women’s “power,” but how, for the purposes of this analysis, should we read and evaluate this “power?”

The context of the “golden age” narrative – emerging amid discourse on contemporary women’s rights movements – contrasts this nebulous “power” with women’s subordination under patriarchal systems, equating “power” with women’s rights172 and freedom from the oppression of patriarchy173. At the same time, much of this “golden age” body of work grapples also with theories of female rulership (nyoji or himehiko) and female magico-religious power174, bringing together multiple interpretations and expressions of the broad concept of “women’s power” beneath a single aegis.

But such overarching use of the term “power” does not fully encompass the complexities of power relationships in Japan. “Power” cannot be defined in black-and-white terms, nor can it be so simply quantified; instead, it must be understood as a series of relations within a particular historical context175. In order to best understand ancient women’s experiences, we must engage with the complexities and intersections of these power relationships. To assist this, let us briefly outline several different concepts that arise within our examination of “women’s power.”

The first aspect is that of absence of patriarchy, a key concern for feminist historians and

Meiji scholars engaging with the “woman question.” In this sense, “women’s power” is understood as freedom from the demands and bonds of patriarchal subjugation. When we speak of “patriarchy,” we must understand that patriarchy itself encompasses a network of intersecting power relations: it does not require the subjugation of all women to all men, but rather specific women to specific men,

172 女権, joken

173 E.g. Tatsumi 1887; Takamure 1966a, 1966b.

174 Sakima 1926; Takamure 1966a, 1966b; Yanagita 1984; Kuratsuka 1996; Maeda 2008.

175 Dreyfus and Rabinow 1982, 184–5; Cousins and Hussain 1984, 225–8.

58 within parameters such as class and family176. Within the patriarchal societal context of mediaeval

Japan, for example, an aristocratic woman may still expect deference from a man of lower class status177, even as she was subordinated to a man of her own rank such as a husband or father. Thus, this “power-as-absence-of-patriarchy” is best evaluated through familial relationships, examining the position of women in relation to husbands and fathers through marriage customs, inheritance laws, and the like178.

The second aspect is that of temporal authority. Theories such as Sakima’s work on nyoji and Takamure’s himehiko focus on women as holders of hegemonic power, from small-scale community leadership to rulership of the country179. It is important, however, that we delineate female hegemonic rule from the “power-as-absence-of-patriarchy” concept outlined above.

Patriarchal systems are rarely so rigid that they cannot tolerate the occasional “exceptional woman” holding a position of influence typically not extended to her gender180; it would be misleading, for example, to take Queen Elizabeth I181 or Chinese empress regnant Wu Zetian182 as emblematic of wider female empowerment within their historical context. Our analysis of “women’s power” in the

“golden age” narrative must therefore take this nuance into account.

In Sakima’s theories of nyoji, ancient women accessed hegemonic power through the magico-religious power attributed to them183. Thus, sacral power – the mastery of an individual over

176 Faure 2003, 169–77.

177 Meeks 2010, 91–4.

178 Faure 2003, 6.

179 Sakima 1926; Takamure 1966a, 1966b; Fukutō 1987, 44; Sekiguchi 1987, 18–9.

180 Kidder 2007, 24.

181 R. 1558–1603 CE

182 R. 690–705 CE; also called Wu Zhao

183 Sakima 1926.

59 the spiritual world, credited efficacy of their rituals, and the like – emerges as another aspect of

“women’s power” contained within the “golden age” narrative. Although Sakima named sacral power as the source of hegemonic power, the two are by no means identical: Yoshie’s work even positions the two as almost oppositional in the Japanese context, an increased focus on women’s religious performance corresponding to their political disempowerment184. The discrepancy between sacral and hegemonic power is also seen in Yanagita’s work, in which the primary beneficiaries of women’s magico-religious power are their brother-kings whose hegemonic authority is legitimated through their sisters’ ritual performance185. Sacral power could even be a subversive force threatening hegemonic authority, as seen in ancient anxieties over curses and witchcraft (Shoku

Nihongi 7.Yōrō 1.4.Jinshin; Shoku Nihongi 10.Tenpyō 1.4.Kigai; Shoku Nihongi 18.Tenpyō-Shōhō

4.8.Kōin; Shoku Nihongi 19.Tenpyō-Shōhō 6.11.Kōshin; Shoku Nihongi 29.Jingo-Keiun 3.5.Heishin;

Shoku Nihongi 32.Hōki 4.10.Shinyū).

A distinction can further be drawn between the “active” sacral power generated by a ritualist’s talent and performance, and the “passive” sacral power, or “power of presence,” generated by the saigū high priestess of Ise. The former, being reliant on the ritualist’s own ability, is located within the individual; the latter, untied to any specific quality of the incumbent beyond their holding of the position, is best understood as a power attributed to the role and not to the individual holding it186. This discrepancy is emphasised in Tanaka’s work on the saigū and her counterpart at

Kamo, the saiin, which clearly distinguishes the saigū’s role as a sort of “living regalia” from “golden age” concepts of female spiritual power187.

184 Yoshie 1987, 2007.

185 Yanagita 1984.

186 Ellwood 1967, 43; Bock 1970a, 151.

187 Tanaka 1996, 103.

The strict cloistering of the saigū also brings up a final aspect of “women’s power”: the issue of personal autonomy. The tension between hegemonic or sacral power and individual autonomy is displayed in J. Edward Kidder’s study of Queen Himiko who, although queen of a confederation and endowed with considerable magico-religious charisma, appeared to live as a “virtual prisoner” ensconced within her palace188. Indeed, her loss of autonomy may have been seen as a necessary sacrifice for the generation of sacral power, as it was for the saigū189. Engaging with the aspect of autonomy is particularly important to the task of reconstructing the subjectivities of ancient women and their experiences, with works by Tanaka, Faure, and Tokoro studying the emotional responses of saigū and saiin to their lives of strict seclusion, humanising the women historically objectified as

“living regalia”190.

In this chapter, we have broadly traced the history and development of the “golden age” narrative, beginning with the prehistoric matriarchy theories of Morgan and Engels, to the Japan- specific nyoji and himehiko discourse of the early 20th century, to the more complex models of recent decades. Through this, we have also observed how, even in its more modern incarnations, the “golden age” narrative still participates in the homogenisation of a diverse array of relationships between gender, spirituality, and different kinds of power in ancient Japan. In order to uphold the humanity of historical women, we must address the ways in which previous scholarship has failed to fully encompass their experiences.

In analysing our texts, we will use the “hermeneutics of suspicion and resistance” (see Ch.

2.2.1 above), while taking note of the biases and limitations of our source texts, and the perspectives

188 Kidder 2007, 137.

189 Takamure 1966b, 78; Faure 2003, 298.

190 Tanaka 1996; Faure 2003; Tokoro 2017.

61 not represented within. This will help us achieve a more nuanced understanding of ancient women’s lives, avoiding homogenisation.

It is also critical that this thesis engage with historical women not as passive objects but as individual actors within a system, possessed of their own agency and their own inner lives, avoiding their dichotomous categorisation as solely “empowered subject” or “oppressed object,” which robs historical women living under patriarchy of their full agency and humanity. To challenge this, we must take into account the network of power relationships and expressions that may exist within a broader patriarchal system, and the ways in which women as actors may work within or defy societal strictures and expectations. This will further require a nuanced understanding of “power” as a complex and multifaceted concept, rather than a simple matter of “have” vs. “have not,” and the ways in which different modes of power may intersect, interact, and even countermand each other.

The following chapter may begin the analysis itself, using these theoretical and methodological bases in an examination of prehistoric Japanese religion and society, exposing the oversimplifications and homogenisations present in the “golden age” narrative’s accounting of this period.

CHAPTER 3: WOMEN IN THE “GOLDEN AGE”

In the earliest incarnations of the “golden age” narrative – the works of Tatsumi, Sakima,

Takamure (see Ch. 2.1.1) – ancient Japan is presented as a world of women’s power untouched by the subjugating effect of patriarchy1. This “power” appears in various interconnecting forms (see Ch.

2.2.5): as worldly authority, as spiritual acuity, as simple freedom from patriarchal oppression. More recent decades have brought forward more nuanced models of this “power” of women: Sekiguchi,

Yoshie, and Piggott, for example, have introduced new theories of gendered roles in rulership and ritual2, and Ambros has touched upon the issue of diversity in ancient women’s experiences based on points of difference such as class3.

It is this notion of difference and diversity which this research particularly seeks to explore in greater detail, re-examining the “golden age” narrative and its various legacies against a feminist- hermeneutical reading of the surviving textual evidence for the ancient period. By doing so, we may assess how well these narrative frameworks engage with these complexities and diversities of experience, and identify areas that the existing narrative frameworks have so far failed to encompass.

First, we will examine the image of prehistoric Japan as a female-centric, matrilineal society through observations of kinship and ancestry in the early texts. Here, we will also engage with the

“prehistoric matriarchy” theories of Morgan and Engels – a considerable influence upon early

“golden age” scholars4, and whose work has been used in defence of the narrative as late as 20035 –

1 Tatsumi 1887; Sakima 1926; Takamure 1966a, 1966b.

2 Sekiguchi 1987; Yoshie 1987, 2007; Piggott 1994, 1997.

3 Ambros 2015.

4 Ijūin and Kawai 2017, 213; Anderson 2020, 67–8.

63 examining the fundamental flaws within the application of this work to the Japanese context. We will further examine the scholarly use of goddess-figures as proof of a female-centric society, an argument that elides the complex relationship between deities and mortal gender.

From there, we will study the notion of “female spiritual superiority” within the ancient context, and its scholarly conflation with social centrality. We shall see that the narrative positioning of ancient women as having dominion over spiritual matters obfuscates a more complex system of gendered ritual roles with some indications of regional variance, and that women’s relationship to spiritual power reflects androcentric, rather than gynocentric, social norms. This androcentrism differs from later, more “conventional” images of patriarchy in its emphasis on the cooperative performance of central male and liminal female over wholesale male domination, a nuance not encompassed by the “golden age” narrative’s neat dichotomies of empowerment versus subjugation, egalitarianism versus patriarchy, and gynocentrism versus androcentrism.

3.1. Reading Prehistoric Society

3.1.1. Theories of Prehistoric Matrilineality

The “golden age” narrative of female power in Japan begins with an era of female predominance. Women – or so the story goes – were at the heart of Japanese religion and society, credited with magico-religious powers, and governed clans and communities with the authority of a priestess. These clans were matrilineal in structure, children sharing the kinship affiliations of their mother; an arrangement which placed women at the centre of the family unit, acting as heads of the clan groups around which society was structured6. Women, in this image of prehistory, were central

5 Sekiguchi 2003, 36–7, 40.

6 Takamure 1966a, 64.

64 to ancient society, respected and endowed with social status and spiritual acuity. This stands in stark contrast to the women of later eras, marginalised and subordinated under patriarchy and patrilineality, their religious activity reappropriated to be performed not for their own benefit but for the gain of their male kin7.

As we have seen (see Ch. 2.1), this image of ancient Japan as a matrilineal, female-centric society was particularly prominent in the earlier iterations of the “golden age” narrative, such as those of Sakima and Takamure. However, the notion is echoed in scholarship of the last few decades, a surviving thread of the “golden age” narrative: Piggott invokes the idea of matrilineal rulership preceding the Yamato dynasty8, whereas Sekiguchi asserts the matrilineal, female-centric quality of

Japanese family systems before patriarchal revisions9. Nor is this concept confined to the subdiscipline of women’s history: references to an ancient matrilineal past may also be found within the field of religious studies and religious history10.

One fierce proponent of Japan’s prehistoric matrilineality was Takamure, who argued for matrilineal kinship as the ancient societal default, the realisation of such only hindered by scholars’ reluctance to part with later androcentric norms11. Within her research, Takamure worked to trace hidden matrilines within the patrilineal records of the Kojiki and Nihon Shoki12, and posited an etymological connection between the words “ancestor” (親, oya) and “mother” (母, omo) to suggest

7 Sakima 1926; Takamure 1966a, 1966b.

8 Piggott 1994, 5–6.

9 Sekiguchi 2003, 32–7, 40.

10 E.g. Nelson 1996, 122; Naoki 2008, 206.

11 Takamure 1966a, 6–7.

12 Takamure 1966a, 7–8.

65 that traditional concepts of ancestry were based around the female line13. Male ancestor-figures were relegated only to smaller branch families within the larger, matrilineal clan14.

Similar arguments can be found within the works of other scholars within the “golden age” narrative. Sakima, for example, argued for the presence of matrilineal kinship (母系, bokei) and familial matriarchy (母権, boken) alongside female magico-religious governance (女冶, nyoji) in the society of prehistoric Japan15. Tatsumi cited ancient duolocal marriages16 as proof of a female-centric familial order17. A century on from Tatsumi, Naoki Kōjirō (直木孝次郎) speculates vestiges of prehistoric matriarchy retained in ancient Japan’s acceptance of female leaders18, and Sekiguchi cites the works of Morgan and Engels in her discussions of Japanese matrilineality, which she claims remained the dominant familial structure as late as the eighth century CE19.

Morgan and Engels (see Ch. 2.1) proposed matrilineal kinship as the universal initial state of human society20, with patrilineality – and the resulting marginalisation and subordination of women under patriarchy – as a later development21; theories that were extremely influential in Japanese historical scholarship and the formation of the early “golden age” narrative22. In more recent decades, however, while the occasional citation of Morgan and Engels’ work in favour of “golden age”

13 Takamure 1966a, 64.

14 Takamure 1966a, 8, 64.

15 Sakima 1926, 9, 11, 18, 24.

16 In which the wife remains with her natal family while the husband makes conjugal visits.

17 Tatsumi 1887, 27–8.

18 Naoki 2008, 206.

19 Sekiguchi 1987, 20; 2003, 32–7, 40.

20 Morgan 1877, 62, 66, 354; Engels 1902, 47.

21 Morgan 1877, 360, 365–6, 481–2; Engels 1902, 67–71.

22 Ijūin and Kawai 2017, 213; Anderson 2020, 67–8.

66 narrative still occurs23, these theories are largely considered outdated in current anthropological circles24.

Morgan and Engels’ arguments for ancient matrilineality are largely predicated on the antiquated theoretical conceit of a single path of development from “savagery” to “barbarism” to

“civilisation” along which all societies can be placed, based upon certain technological developments25. This fails, however, to take into account the different, branching ways in which societies grow and change in response to their environment. The racist and Eurocentric assumptions implied in the categorisations of “savagery” and “barbarism” are also present in full force, with

“savagery” represented by Indigenous Australians and “barbarism” by Native Americans such as the

Iroquois, whereas “civilisation” is predominantly equated with the Romans and Greeks26. Morgan’s argument of a universal matrilineal past was derived from his observations of matrilineality among peoples, such as the Gamilaraay of New South Wales, that he dubbed as “savage,” under the insulting assumption that they in some way represented a fossilisation of a stage which all societies passed through, and which they had yet to “grow out” of – in other words, his research was predicated on the belief that, through studying cultures labelled “primitive” by Europeans, he could gain knowledge of the ancient state of all humanity27.

Within the current academic world, however, many of Morgan’s theories on the linear progress of societal development are now understood as outmoded28, and it has been pointed out by later commentators that his studies on Indigenous Australian kinship systems – taken as

23 Sekiguchi 1987, 20; 2003, 36–7, 40.

24 Ueno 1987, 9; Gardner 2009, 286; Kelly and McConvall 2018, 27–8.

25 Morgan 1877, 12.

26 Morgan 1877, 48.

27 Morgan 1877, 48–9.

28 Fortes 1969, 219; Kelly and McConvall 2018, 27.

67 representative of “savagery”, the earliest state of human society – serve as poor representations of these cultures and kinship structures, the complexities of which preclude his categorisation of them as “savage” primordiality29. These works and their conclusions are thus fundamentally flawed, and cannot simply be transposed onto the Japanese context as an argument for an ancient matrilineal

“golden age.”

Having examined the notion of “universal prehistoric matrilineality,” let us return to the specific context of ancient Japan. In order to examine in greater depth this aspect of the “golden age” narrative, we will review the presentation of ancestry and lineage in the Kojiki and Nihon Shoki.

The genealogies within these records are predominantly patrilineal, with particular focus on the male-line descent of the imperial clan (Kojiki; Nihon Shoki). Although their divine ancestress – the female deity Amaterasu – at first appears as a striking exception to the male line, she is not, actually, a matrilineal ancestress: all descent from her, beginning with her son Masakatsu-Akatsu-

Kachi-Hayahi-Ame-no-Oshiho-Mimi (正勝吾勝々速日天之忍穂耳/正哉吾勝勝速日天忍穂耳), is reckoned from father to son (Kojiki 1; Nihon Shoki 1.6; Nihon Shoki 1.9.1); a patriline with a female figure at the apex, rather than an ancient line of maternal descent. There is also some uncertainty regarding the antiquity of Amaterasu’s role as main imperial ancestress, with Okada and Naoki holding the masculine deity Takami-Musubi (高御産巣日/高皇産霊) as an earlier co- or even sole focus of imperial ancestor-god devotion30. The suggestion of Amaterasu as a later addition to imperial worship further complicates her interpretation as a legacy of female-centric ancestral lineages.

Nor is Takami-Musubi without his own complications. While a masculine deity, he is not a strictly patrilineal ancestor, his relation to the imperial clan arising from his daughter’s marriage to

29 Gardner 2009, 285–6; Kelly and McConvall 2018, 27–8.

30 Okada 1970, 359, 399; Matsumura 1999, 114; Naoki 2009, 71–80; Emura 2012, 19.

Amaterasu’s son. Furthermore, whereas the Nihon Shoki presents varying accounts in which either

Takami-Musubi or Amaterasu is responsible for their descendants’ descent to earth, in the Kojiki accounting it is a joint command on both their parts (Kojiki 1; Nihon Shoki 2.9), the Divine Mandate of kingship originating from both maternal and paternal ancestors. While the imperial clan itself is primarily reckoned along a patriline, their relations to ancestral deities suggest a more bilateral approach to kinship with both maternal and paternal divine lineages accounted for; neither a simple patriline nor a straight matriline.

A bilateral system is in concordance with Yoshie’s studies on pre- and early ritsuryō elite genealogies, which suggest that, initially, both paternal and maternal ancestries were tracked, with personal clan affiliations often chosen based on political utility31. Yoshie in particular cites the case of

Mononobe no Yuge no Moriya (物部弓削守屋) who, while inheriting chieftaincy of the Mononobe clan from his father, was also closely affiliated with – and possibly chieftain of – the Yuge clan through his mother’s side32. Over the seventh to ninth centuries, paternal lineages became increasingly privileged over maternal, to the point where nobles petitioned the court to have their previous, “mistaken” female-line affiliations switched to the paternal side33. Examples of of maternal-side affiliation may be found in Prince Hikoimasu’s (日子坐/彦坐) consort Saho no

Ōkurami-Tome (沙本之大闇見戸売), daughter of Kasuga no Takekuni-Katsutome (春日建国勝戸売)

(Kojiki 2.Kaikaki)34; and Sujin’s35 (崇神) consort Tōtsu-Ayume-Makuwashi-Hime (遠津年魚目目微比

31 Yoshie 1993, 439.

32 Yoshie 1993, 439.

33 Yoshie 1993, 440.

34 “春日建國勝戶賣之女。名沙本之大闇見戶賣” (“Kasuga no Takekuni-Katsutome no musume. Na wa Saho no Ōkurami-Tome”).

35 Reign traditionally dated 97–30 BCE, first emperor following the “Eight Missing from History”

売/遠津年魚眼眼妙媛), who is listed as the daughter of the female chieftain Arakawa-Tobe (木国造

荒河刀弁/紀伊国荒河戸畔) (see Ch. 4.3.3) (Nihon Shoki 5.Sujin 1.2.Heiin)36.

Two more clans listed in the Kojiki and Nihon Shoki name female ancestral deities: the

Kagami-Tsukuri (鏡作) clan of mirror-makers, who hold Ishikori-Dome (伊斯許理度売/石凝姥) as their progenitor (Kojiki 1; Nihon Shoki 1.9.1), and the Sarume (猿女) clan of ritual dancers37, who trace a dual lineage from the union of the goddess Ame-no-Uzume and the god Saruta-Biko (猿田毘

古/猿田彦) (Kojiki 1; Nihon Shoki 1). While the further genealogies of these clans are not outlined beyond their divine ancestors, the Sarume clan’s dual reverence for Saruta-Biko and Ame-no-Uzume

– taking their name from both parties, and their traditional occupation from Ame-no-Uzume – also display a bilateral significance accorded to both male and female ancestor-figures, rather than a purely matrilineal or patrilineal conception of ancestry.

A careful examination of the early imperial records upholds this bilateral approarch to kinship. On the surface, imperial clan membership is depicted as patrilineal, all affiliates connected on the paternal line. At the same time, however, matrilineal affiliations clearly retain some significance of their own. Kawamata-Bime (河俣毘売/川派媛), daughter/ancestor (see below) of the

Shiki-no-Agatanushi (師木県主/磯城県主), gives birth to the next emperor Annei38 (安寧), whose personal name, Shikitsu-Hiko-Tamatemi (師木津日子玉手見/磯城津彦玉手看), in both records takes the same characters for “Shiki” as his mother’s family name39, the component “Shikitsu-Hiko” itself translating to “man of Shiki” (Kojiki 2.Suizeiki; Nihon Shoki 4.Suizei 2.1). Annei later marries his

36 “紀伊國荒河戶畔女遠津年魚眼眼妙媛” (“Kii-no-Kuni no Arakawa-Tobe ga musume Tōtsu-Ayume-

Makuwashi-Hime”).

37 Kuratsuka 1996, 14; Ambros 2015, 33–4.

38 Reign traditionally dated 549 – 511 BCE, second of the “Eight Missing from History”

39 In the Kojiki, 師木; in the Nihon Shoki, 磯城.

70 maternal cousin, another woman of Shiki-no-Agatanushi, and their third child is again named

Shikitsu-Hiko40 (師木津日子/磯城津彦) (Kojiki 2.Anneiki; Nihon Shoki 4.Annei 3.1.Jingo).

Another example of these maternal affiliations within the patrilineal clan structure can be found in Prince Sahohiko (沙本毘古/狭穂彦) and Princess Sahohime (沙本毘売/狭穂姫), who derive their names – and their connections to the Saho region, in the northern reaches of the

Yamato Basin – from their mother Saho-no-Ōkurami-Tome (Kojiki 2.Kaikaki). Further examples include Yasaka-Irihiko (八坂之入日子/八坂入彦), son of Yasaka-Furuama-Irobe (八坂振天某辺)

(Nihon Shoki 5.Sujin 1.2.Heiin), Mamuta-no-Ō-Iratsume (茨田大郎女/茨田大娘), daughter of

Mamuta no Sekihime (茨田連関比売/茨田連関媛) (Kojiki 3.Keitaiki; Nihon Shoki 17.Keitai 1.3.Kiyū),

Kasuga-no-Yamada-no-Iratsume (春日山田郎女), daughter of Kasuga no Nukako (春日之糠子)

(Kojiki 3.Kinmeiki; Nihon Shoki 19.Kinmei 2.3), and Prince Kasuga (春日), son of Kasuga no Ominago

(春日老女子) (Kojiki 3.Bidatsuki; Nihon Shoki 20.Bidatsu 4.1). These examples recall Yoshie’s case of the bilateral lineage of Mononobe no Yuge no Moriya41, with royal issue inheriting clan affiliation – and possible candidacy for the throne – from their fathers, but also maintaining strong connections with their maternal clans and territories, connections which were later exploited by the powerful clans of Soga and Fujiwara.

Let us return to the example of Kawamata-Bime, whose presence in the records is somewhat murky and ambiguous. In the Kojiki, she is portrayed as the principal consort to Emperor

Suizei42 (綏靖), and described as “ancestor (祖, oya) of the Shiki-no-Agatanushi” (Kojiki 2.Suizeiki)43.

40 The names of this prince’s two granddaughters, Hae-Irone and Hae-Irodo, may also derive from their maternal great-grandfather Shiki-no-Agatanushi Hae (Kojiki 2.Anneiki); however, the use of different characters for “Hae” makes this affiliation less clear.

41 Yoshie 1993, 439.

42 Reign traditionally dated 581–549 BCE; second recognised emperor and first of the “Eight Missing from

History”

In the Nihon Shoki, however, she is listed instead as a daughter (女, musume) of that same clan

(Nihon Shoki 4.Suizei 2.1)44, reconfiguring her from the central role of progenitor to the subsidiary role of descendant. The same pattern, with the same clan, is repeated two generations later for

Itoku’s45 (懿徳) possible empress, Futo-Mawaka-Hime/Iihi-Hime (賦登麻和訶比売/飯日比売/飯日

媛), who is alternately listed as ancestor of the Shiki-no-Agatanushi (Kojiki 2.Itokuki)46 or the daughter of Shiki-no-Agatanushi Futo-Mawaka-Hiko (磯城県主太真稚彦) (Nihon Shoki 4.Itoku

2.2.Kibō)47.

On the surface, these discrepancies appear as if they may be revisions, products of the gradual social devaluing of female-line ancestries which Yoshie describes48, female ancestors rewritten into mere affiliates. Another complicating factor arises, however: even in the Kojiki, where these women are marked “ancestors,” none of their own issue are designated as members of that clan, making their depiction as oya curious. Other genealogies show that Kawamata-Bime had at least one brother – Shiki-no-Agatanushi Hae (県主波延/磯城県主葉江) (Kojiki 2.Anneiki; Nihon

Shoki 3.1.Jingo) – endowed with the Shiki-no-Agatanushi name. This suggests that ideas of ancestry and genealogy were not entirely linear in ancient Japan, and women could be considered co- ancestors of their male relatives’ descendants, further muddying the assumption of lineages as strict, monogendered patrilines versus matrilines. Similar fraternal ancestry-affiliations can be seen with

43 “師木縣主之祖河俣毗賣” (“Shiki-no-Agatanushi no oya Kawamata-Bime.”)

44 “磯城縣主女川派媛” (“Shiki-no-Agatanushi no musume Kawamata-Hime.”)

45 Reign traditionally dated 510–476 BCE; third of the “Eight Missing from History”

46 “師木縣主之祖。賦登麻和訶比賣命。亦名飯日比賣命” (“Shiki-no-Agatanushi no oya. Futo-Mawaka-

Hime-no-Miko. Mata no mina wa Iihi-Hime-no-Mikoto.”)

47 “磯城縣主太眞稚彥女飯日媛” (“Shiki-no-Agatanushi Futo-Mawaka-Hiko ga musume Iihi-Hime.”)

48 Yoshie 1993, 440.

Sujin’s consort Ōama-Hime (意富阿麻比売/大海媛) (Kojiki 2.Sujinki)49, Yamato-Takeru’s50 consort

Miyazu-Hime (美夜受比売/宮簀媛) (Kojiki 2.Keikōki)51, and Keitai’s52 (継体) consort Wakahime (若

比売/稚子媛) (Kojiki 3.Keitaiki)53.

What we learn from these examples is twofold: firstly, ancient Japanese kinship presents as neither the singular patriline of later eras, nor as the singular matriline of the “golden age” narrative, but rather as a complex network of bilateral inheritances54. Within the Kojiki and Nihon Shoki, primary emphasis is placed on paternal affiliations, but maternal affiliations retain considerable significance. Images of “sister-ancestors” such as Kawamata-Bime further suggest that ancestry was not reckoned in terms of a direct line of exclusively-male or exclusively-female figures, as the

“patriline versus matriline” debate produced by the “golden age” narrative indicates, but rather encompassed a more complex network of affiliations and ancestor-figures within a paternal or maternal clan.

3.1.2. Goddesses and Women

Having examined patrilineality and matrilineality within the Japanese context, let us turn now to the figure of the imperial ancestress Amaterasu, and another “thread” of the “golden age” narrative: the image of ancient Japan as a female-centric, woman-revering society as signified by goddess-worship. In the cosmology of ancient Japan – or, more specifically, the imperially- sanctioned state religion thereof – the sun goddess Amaterasu appears as the centre of the

49 “尾張連之祖意富阿麻比賣” (“Owari-no-Muraji no oya Ōama-Hime”)

50 Conquering prince and folk hero of Yamato, son of Emperor Keikō (trad. r. 71–130 CE), c. 72–114 CE

51 “尾張國造之祖。美夜受比賣” (“Owari-no-Kuni-no-Miyatsuko no oya. Miyazu-Hime”)

52 Reign more reliably dated 507–531 CE

53 “三尾君䒭祖。名若比賣。” (“Mio-no-Kimi-ra no oya. Na wa Wakahime.”)

54 Yoshie 1993, 439–40; Matsumura 1999, 86; Ambros 2015, 19.

73 pantheon, supreme among the deities55. Early “golden age” scholars especially read this powerful and brilliant goddess as a symbol of the splendour of ancient Japanese womanhood, her supremacy among the gods and position in the important (and, among world cultures, typically masculine) role of solar deity proof of a matriarchy, of the respect and reverence accorded to women in ancient

Japan56. A related line of argument also emerged surrounding the feminine, often pregnant, dogū57 sculptures of the Jōmon period58, which one interpretation holds as representations of an

“Earthmother”-type fertility goddess; this hypothetical “mother goddess” has also been cast as intrinsically linked to a reverence for, and societal centring of, mortal womanhood59.

One core issue with this reading is its assumption that female deities are automatically and in all aspects representative of mortal womanhood, and thus goddess-reverence as a signifier of the respect afforded to mortal women within society60. Tanaka’s and Matsumura’s work on the gendering of goddesses has shown us that there is a far greater complexity to the relationship between deities and mortal beliefs about gender than a simple equation of goddess-worship to female-centrism encompasses61. Deific myths reflect and reinforce the underlying values and assumptions of their progenitor society, and a goddess-figure can just as easily represent an idealised image of womanhood through the male gaze, as she can an actual respect for women or a female-centric social organisation62. For a deeper understanding of this phenomenon, let us take a

55 Anzu 1986, 21–2; Emura 2012, 21.

56 Tatsumi 1887, 28; Sakima 1926, 68–9; Anesaki 1930, 27; Takamure 1966a, 8; 1966b, 9–10.

57 土偶, “earthen figure”

58 14,000 BCE–1000 BCE (in older categorisations, going as late as 300 BCE)

59 Sakima 1926, 72–3; Ōi and Miyagi 1974, 8; Ueno 1987, 6.

60 Tanaka 1996, 198, 227–8.

61 Tanaka 1996; Matsumura 1999.

62 Tanaka 1996, 198, 227–8; Matsumura 1999, 53–4.

74 closer look at the Greek goddess Athena, who Matsumura compares in depth with Amaterasu and the Virgin Mary63.

Athena is well-known as the patron-goddess of the Greek city-state of Athens, revered by the people as its protectress. She was a goddess of wisdom and craft, and like Amaterasu, also possessed a martial aspect, often painted and sculpted with arms and armour. A surface reading might therefore conclude that women occupied high status in ancient Athens (because, after all, a goddess was revered as their patron deity), that women were respected for their wisdom (because the god of wisdom is gendered female), and that women could be seen as having martial strength

(because that goddess was also a warrior).

Any study of Athenian society, however, will tell us that this is manifestly not the case. It is well-established by historical evidence that the mortal women of Athens were barred from positions of direct power and thoroughly subordinated to the men in their lives64, that they were often argued to be incapable of the kind of higher reasoning to which men had access, and that the only images of

“warrior-women” in Greek art and literature are the mythical Amazons, who appear as monstrous norm-inverting bogeywomen akin to Gorgons or harpies and similarly destined to be conquered by a bold masculine hero65. Athena’s warrior’s garb, conversely, serves to “masculinise” or “defeminise” her, distancing her from ideas of mortal womanhood and bestowing her with the sense of androgyny that permitted her to accumulate such respect66.

63 Matsumura 1999.

64 Beard 2017.

65 Beard 2017, 68–70.

66 Matsumura 1999, 98; Beard 2017, 68–70.

With that in mind, let us examine these passages from the Kojiki and Nihon Shoki, in which

Amaterasu, believing that her younger brother Susa-no-O67 (須佐之男/素戔鳴) means to usurp rule of the heavens from her, takes on a martial form to face him:

…she faced him in the stance of a great manly warrior, and demanded of him, “Why hast

thou come?”68 (Kojiki 1)

She shook the arm of her bow…and let out a great manly war cry.69 (Nihon Shoki 1.6)

So she took upon herself the appearance of a manly warrior…70 (Nihon Shoki 1.6.1)

The sun god said: “…Though I am a woman, how can I flee?” Thus she girded herself in the

manner of a warrior. 71 (Nihon Shoki 1.7.3)

Amaterasu’s own martial aspect is thus presented as an affectation of masculinity, a contrast to her own feminine gendering that androgynises her within this portion of the text. The strength and fearsome splendour she displays in facing her brother is not attributed to womanhood or femininity at all, but rather is a masculine quality which the goddess nevertheless has the ability to access.

Another aspect of Athena salient in her comparison with Amaterasu is that of her motherhood. Athena, while perpetually virginal, is also mother (sans intercourse) to a son,

Erechthonius. She fulfills the patriarchy’s contradictory demands of women by being a mother

67 God of storms; also Romanised as “Susanoo,” “Susanowo,” or “Susa-no-Wo”

68 “伊都…之男建…蹈建而。待問何故上來。” (“Itsu no otakebu fumi takebite. Nado nobori kimaseru to machi toitamaiki.”)

69 “振起弓彇。…奮稜威之雄誥。” (“Yuhazu wo furi tate…Itsu no otakebi wo furuwashi.”)

70 “乃設丈夫武備。” (“Sunawachi masurao no takeki sonae wo mōketamau.”)

71 “日神曰。…吾雖婦女何當避乎。乃躬裝武備…” (“Hi no kami notamawaku...Are taoyame nari to iu tomo nanzo saran ya to notamaite. Sunawachi mi ni takeki sonae wo yosou koto.”)

76 without the defilement of sex, at once a feminine ideal and defeminised, woman and not-woman, a reflection of Greek society’s own contradictory attitudes72. Although Amaterasu’s virginity is not so emphasised as Athena’s, she is never shown as having sex or bearing children, the circumstances of her own motherhood likewise utterly desexualised73. Indeed, Amaterasu’s androgyny is such that, although gendered female in the ancient period (“though I am a woman…”), in mediaeval commentary she was often interpreted as a genderless or even masculine deity, Amaterasu-

Ōmikami replaced by Tenshō-Daijin74.

The drawing of comparisons between Athena and Amaterasu is not to state that they or their cultural contexts are wholly identical; certainly, this is not the case. The example of Athena, a goddess from an ancient society whose attitudes towards women are clear and well-documented, is used here mainly to highlight that the presence of a powerful and revered goddess cannot be taken on its own as proof of a female-centric society, absence of patriarchal values, or indeed any particular regard for mortal women.

It is also important to remember that Amaterasu is far from the only female deity of ancient

Japan. The scholarly hyper-focus on Amaterasu, and the significance of her position among the gods, pushes out from the narrative the other goddess of the Kojiki and Nihon Shoki – such as Izanami (伊

邪那美/伊奘冊), Ame-no-Uzume, or Ōgetsu-Hime (大気都比売) – and what they may illustrate about gender beliefs in ancient Japan. Indeed, as Matsumura points out, the image of Izanami within

72 Matsumura 1999, 67, 98; Beard 2017, 68–70.

73 Matsumura 1999, 67, 98.

74 Both names use the characters 天照大神 (“Great Deity Amaterasu”); the former name using the Japanese- derived/kunyomi pronunciations and the latter the Chinese-derived/onyomi pronunciations (Tanaka 1996,

197; Bowring 2005, 45).

77 the official mythology stands in stark contrast to the glories of Amaterasu75; these myths, and their relation to pollution taboos, will be explored in greater detail below.

Furthermore, the conceit of the “official” mythology should also be remembered. The Kojiki and Nihon Shoki, as we have seen (see Ch. 2.2.2) are not unbiased representations of the full range of ancient Japanese beliefs and perspectives, but rather centrally-produced texts promulgating a specific worldview beneficial to the Yamato court, constructing Japan as an imperial state with the paramount-emperor at the centre of the world-realm representing ideal ritual order76, and this includes their mythic sections. It is no surprise that, as imperial ancestor-god, Amaterasu takes such pride of place within the pantheon in the official court corpus of myth, but we cannot simply assume that she was uniformly worshipped, let alone regarded with such reverence, across the entire archipelago. The “golden age” narrative’s engagement with Amaterasu as a goddess has thus typically been one of oversimplification of the role of gender in goddess-myths, and of homogenisation of diversity, through the conflation of Yamato royal beliefs with those of ancient

Japan in its entirety.

3.2. Women and the Spiritual World

So, then, what are we to make of the position of mortal women, if they are not exemplified by goddesses? Let us examine further the images of ancient womanhood within the “golden age” narrative. One key thread of the “golden age” narrative is the link between women’s temporal authority and sacral power: ancient Japan is configured as a society in which women, endowed with

75 Duthie 2014.

76 Matsumura 1999, 67, 98; Beard 2017, 68–70.

78 magico-religious abilities, occupy religious roles from which they derive considerable social status77.

This theory is based on two core suppositions: firstly, that women in ancient times were credited with greater spiritual acuity than men, and so were naturally preferred to mediate with the divine; and secondly, that ancient communities relied heavily upon the gods for guidance, and so those members responsible for ascertaining and transmitting divine will – the priestesses – were leaders of their communities. We shall focus now on how these assumptions are borne out in the ancient

Japanese context.

3.2.1. The Miko: Female Shamanism in Ancient Japan

The most prominent female role in the kami-worshipping traditions of Yamato is that of the miko78, also called the kannagi79. In English, “miko” is typically translated as “shrine maiden,” based on her role in modern Shinto: that of a virginal attendant whose primary duties are caretaking of the shrine and ritual dance, firmly subordinate to the male priests80. The modern miko is not a figure typically endowed with great spiritual power or importance, but this was not always the case. The ritual dances of the modern miko are derived from dances intended not only to entertain the gods, but to induce them to possess the miko, sending her into an ecstatic trance-state similar to that seen in other North and East Asian shamanic traditions81.

Although shamanism is now rarely seen in mainstream Shinto, its practice mostly limited to specific regional traditions, it is clear from archaeological evidence that the miko role has been an

77 Takamure 1966b, 119–20; Ōi and Miyagi 1974, 18; Yanagita 1984, 41; Kuratsuka 1996, 13; Matsumura 1999,

76.

78 巫女, “medium woman.” May also be spelled 神子, “divine child.”

79 巫, “pacifier of gods,” also rendered as kamunagi in the ancient orthography.

80 Blacker 1982, 30; Nelson 1996, 125; Ambros 2015, 62.

81 Blacker 1982, 27, 109; Kuratsuka 1996, 14–5.

79 important part of Japan’s religious traditions since prehistory82. Images identified as shamans have been found on ceramics dating back to the Jōmon and Yayoi periods. Bronze bells, also from Yayoi, depict shamans performing rites for hunting and agriculture, a spindle marking the shamans as female. Haniwa83 clay figures from the Kofun period84 have been unearthed of women, some in dancing poses, adorned with instruments of the miko’s trade: mirrors, headdresses or crowns of bells, and necklaces of comma-shaped magatama85 beads86. The miko gradually came to lose her shamanic role and mantic functions under the centralised, state-sponsored religious organisation of the Department of Divinity87 (see Ch. 5.2). Some regional seasonal festivals still have the miko act as a medium for a possessing god or spirit, but she is now largely a “passive” medium, her possession induced by a male yamabushi88 ascetic89.

Another female-gendered shamanic tradition that survives today, albeit in small numbers, is the role of the ichiko or itako, originating in north-eastern Honshu. Itako shamans are traditionally blind women90, initiated in cold-water purification rituals based off of Shugendō practice91, and credited with the ability to act as medium for gods and spirits of the dead92. There is a notable similarity between the name of the itako and other terms for shamanic roles found throughout

82 Hori 1968, 182, 191–9.

83 埴輪, “circle of clay”

84 300–538 CE

85 勾玉, “bent jewel/bead”

86 Blacker 1982, 104–5.; Ambros 2015, 13.

87 Hori 1968, 182; Blacker 1982, 30; Kuratsuka 1996, 13–5.

88 山伏; an ascetic practitioner of Shugendō mysticism, a mix of esoteric Buddhist, Taoist, and Shinto traditions.

89 Blacker 1982, 252–4.

90 Blacker 1982, 140–1.

91 Blacker 1982, 91.

92 Blacker 1982, 128, 151.

80 northern Asia, including the Ryukyuan yuta, northern Tungus idakon, the Kalmuck and Yakut udagan, the Mongol idugan, and the Buriat utygan93, although a clear line of descent is difficult to establish.

A regional tradition distant from ancient Yamato, from the homeland of the ancient Emishi94 people whose relation to modern Yamato Japanese is unclear, the itako does not appear have been involved with the state bureaucracy and its changes to the religious scene. Academic studies of itako traditions and social functions have typically focused on the contemporary context, with little investigation into the role’s presence in historical Tōhoku95. Two main commonalities unite the miko and itako: first, their role as shamanic mediums, and second, their gendering as female religious roles96. These qualities are also found in the Ryukyuan yuta, although she lacks the ritual involvement of the miko, that role instead fulfilled by the non-shamanic noro97. We will discuss the implications of shamanic gendering shortly (see Ch. 3.2.2); for now, let us examine how miko and shamanism are portrayed in the records.

The first act of shamanism is credited to the goddess Ame-no-Uzume, who in the Age of

Gods performs a dance to lure Amaterasu out from seclusion in the Heavenly Rock Grotto. She fashions a crown, sash, and staff out of vegetation, and bares her genitals in ecstatic dance, described in both the Kojiki and Nihon Shoki as “like she was possessed by a god”98 (Kojiki 1; Nihon

93 Hori 1968, 201; Blacker 1982, 27.

94 蝦夷; an ethnic group from north-eastern Honshu.

95 E.g. Hori 1968; Blacker 1982.

96 Hori 1968, 181; Blacker 1982, 22, 27; Kuratsuka 1996, 13. As will be seen below, although male shamanic practitioners existed, in Yamato culture the role was clearly associated with women.

97 Kuratsuka 1996, 113–4; Sered 1999, 5, 35, 195, 202–7. Sered does note the existence of male yuta, but suggests that they may be less respected as practitioners than their female counterparts.

98 Kojiki: “爲神懸而。” (“Kamugakari shite.”); Nihon Shoki: “顯神明之憑談。” (“Kamukakarisu.”). Both texts actually use the verb for “divine possession” but, given that Ame-no-Uzume is a deity herself, it is more likely that the texts refer here to the ecstatic state of the shaman’s dance rather than the literal act of possession.

Shoki 1.7). Ame-no-Uzume is heralded as the ancestress of the Sarume clan, known for its production of female ritual dancers99, her female descendants performing the gendered role first credited to their divine ancestress. When Amaterasu confronts her brother Susa-no-O in the Kojiki and Nihon Shoki, she brandishes a bow100 and adorns herself with magatama beads, shouting and stamping her feet (Kojiki 1; Nihon Shoki 1.6–7). Although contextually invoking the figure of a “manly warrior”, her use of beads may also be inspired by the garb and implements of a miko working herself into a trance state101, blending masculine and feminine imagery in androgynous performance.

Further images of shamanic women can be found after the Age of Gods, beginning with the reign of Sujin. The princesses Toyosuki-Irihime (豊鉏入日売/豐鍬入姫) and Nunaki-Irihime (沼名木

之入日売/渟名城入姫) are “entrusted” (託, tsuke) with the deities Amaterasu and Yamato-no-

Ōkunitama (日本大国魂), respectively, charged with worshipping these gods and moving their place of enshrinement away from the palace (Nihon Shoki 5.Sujin 6)102. In the reign of Sujin’s son Suinin103

(垂仁), Amaterasu is “entrusted” to Princess Yamato-Hime in much the same way, with the princess embarking on a pilgrimage through the provinces before enshrining the goddess in Ise on her instructions (Nihon Shoki 6.Suinin 25.3.Teigai)104.The verb 託, which can have the meaning of “to

99 Ellwood 1967, 39; Bock 1970a, 6; Kuratsuka 1996, 14; Yoshie 2007, 188.

100 Cherry birch-wood bows are common as ritual instruments in the dances of the modern miko (Hori 1968,

202; Blacker 1982, 106–7, 148).

101 Blacker 1982, 105.

102 “故以天照大神。託豐鍬入姬命。…亦以日本大國魂神。託渟名城入姬命令祭。” (“Kare Amaterasu-

Ōkami wo motte. Toyosuki-Irihime-no-Mikoto wo tsukematsurite…Mata Yamato-Ōkunitama-no-Kami wo motte. Nunaki-Irihime-no-Mikoto ni tsukematsurite matsurashimu.”)

103 Reign traditionally dated from 29 BCE–70 CE; much the same as his father in terms of debate on historicity.

104 “離天照大神於豐耜入姬命。託于倭姬命。” (“Amaterasu-Ōkami wo Toyosuki-Irihime-no-Mikoto ni hanachimatsurite. Yamato-Hime-no-Mikoto ni tsuketamau.”)

82 send via,” “to have someone deliver,” is used elsewhere within the Nihon Shoki to indicate divine possession, most notably for the oracle pronounced by Empress Jingū (Nihon Shoki 8.Chūai

8.9.Kibō)105; the kanji character forms the first half of the word “oracle” (託宣, takusen)106. Nunaki-

Irihime’s body, thin and losing hair, is too frail for the role of divine vessel (Nihon Shoki 5.Sujin 6)107.

Instead, the task of worship is given to the male priest Ichishi no Nagaochi (市磯長尾市), but without the language of “entrustment” (Nihon Shoki 5.Sujin 7.11.Kibō; Nihon Shoki 6.Suinin

25.3.Teigai)108, further suggesting that this shamanic “entrustment” was a gendered function.

Another passage within Sujin’s reign describes how he, troubled by the disasters in his realm, visits the plain of Kamu-Asajihara to consult the gods for advice. He brings his great-aunt Yamato-

Totohi-Momoso-Hime109, who is possessed by the god Ōmono-Nushi (大物主) and delivers the desired oracle (Nihon Shoki 5.Sujin 7.2.Shinbō)110. Yamato-Totohi-Momoso-Hime is also depicted as being in receipt of prophetic dreams (Nihon Shoki 5.Sujin 7.8.Kiyū), deciphering omens (Nihon Shoki

5.Sujin 10.9.Jinshi), and becoming the wife of Ōmono-Nushi (Nihon Shoki 5.Sujin 10.9.Jinshi). This last

105 “時有神託皇后而誨曰。” (“Toki ni kami mashite kisai-no-miya ni kakarite oshiematsurite notamawaku.”)

106 The second half, 宣, refers to a proclamation or edict.

107 “然渟名城入姬命髪落體瘦而不能祭。” (“Shikashi Nunaki-Irihime-no-Mikoto kamiochi yasekamite iwau koto yokarazu.”)

108 1: “又以長尾市。爲祭倭大國魂神之主。” (“Mata Nagaochi wo motte. Yamato-no-Ōkuni-Tama-no-Kami wo matsuri kamunushi to nasu.”); 2: “是以命大倭直祖長尾市宿禰。令祭矣。” (“Koko wo motte Yamato-no-

Atai no oya Nagaochi-no-Sukune ni mikotonori shite. Iwaimatsurashimu.”)

109 Traditional birth and death dates unclear; appears as early as 91 BCE and dies sometime after 88 BCE

110 “天皇乃幸于神淺茅原。而會八十萬神以卜問之。是時。神明憑倭迹迹日百襲姬命曰。” (“Sumera- mikoto imashi Kamu-Asajihara ni idemashite. Yasoyorozu no kamutachi wo tsudoete motte uratou. Kono toki ni.

Yamato-Totohi-Momoso-Hime-no-Mikoto ni kamukakari shite notamawaku.”)

83 point casts her as a divine bride, a mortal consort to the deity, another role later associated with the position of miko111.

Perhaps the most famous act of female shamanism, appearing in both the Kojiki and Shoki, is that of Empress Jingū’s ritual possession by several deities, delivering an oracle that leads to her conquest of Korea (Kojiki 2.Chūaiki; Nihon Shoki 8.Chūai 8.9.Kibō; Nihon Shoki 9.Sesshō Zenki Chūai

9.3.Jinshin)112. In the Kojiki telling of this story, Jingū’s trance states are not spontaneous; rather, she is specifically chosen for the role of medium, with her husband Chūai113 (仲哀) and the minister

Takeshi-Uchi-no-Sukune (建内宿祢/武内宿祢) assuming the ritual roles of male priests by playing the koto and interrogating the possessing deity. In the Nihon Shoki recounting, although her initial possession is possibly spontaneous in nature, the second is clearly intentional (Nihon Shoki 8.Chūai

8.9.Kibō; Nihon Shoki 9.Sesshō Zenki Chūai 9.3.Jinshin).

It should be noted that all of these stories come from the earliest periods of the records, and all of these figures are thus of decidedly uncertain historicity. But, at the very least, it shows a clear connection is drawn between women and the role of divine vessel in the minds of the early eight- century compilers. Although references to male priests also exist in the records, instances of male mediumship are comparatively rare. Men are occasionally depicted as receiving prophetic dreams, such as Takakuraji (高倉下) (Kojiki 2.Jinmuki; Nihon Shoki 3.Jinmu-Tennō Sokui Zenki

Bōgonen.6.Teishi) or Ōminakuchi-no-Sukune (大水口宿祢) (Nihon Shoki 5.Sujin 7.8.Kiyū), and in

111 Okada 1970, 380–1.

112 Kojiki: “其太后息長帶日賣命者。當時歸神。” (“Sono ōkisaki Okinaga-Tarashi-Hime-no-Mikoto wa. Sono kami kamigakari shitamaiki.”); Nihon Shoki 1: “時有託皇后而誨曰。” (“Toki ni kami mashite kisai-no-miya ni kakarite oshiematsurite notamawaku.”); Nihon Shoki 2: “皇后選吉日入齋宮。親爲神主。” (“Kisai-no-miya yokihi wo erabite iwai-no-miya ni iri. Mizukara kannushi to naritamau.”)

113 Reign traditionally dated 192–200 CE

84 three passages in the Nihon Shoki for the years 643 and 644, the term kannagi114 is used in a collective sense, written with characters implying the presence of both male and female shamans

(Nihon Shoki 24.Kōgyoku 2.2/Kōgyoku 3.6/Kōgyoku 3.7), but mediumship is typically attributed to female figures. Only two possible examples of male divine possession are recorded in the Nihon

Shoki, and in both cases the gender of the – apparently spontaneous – possessee is unspecified

(Nihon Shoki 15.Kenzō 3.2.Teishi/Kenzō 3.4.Kōshin)115.

A song recorded in the 12th-century collection Ryōjin Hishō116 illustrates how shamanism was, in the central perspective, gendered as female:

Are there no women

In the Eastern Provinces?

There are male shamans

And thus the gods

Possess the men.117 (Ryōjin Hishō 2.556)

In other words, although shamanism by men was not unknown, within central culture it was expected that a woman would hold the role118. A man performing the rituals of a miko was unusual enough to the central court that it provoked commentary, such as the cheeky question of whether there were simply no women available. Although male priests existed, with their own ritual duties,

114 In this instance, 巫覡, with 巫 representing a female and 覡 a male shaman or medium; the term is also rendered as kamunagi using the orthography of the period (Meeks 2011, 209).

115 1: “於是月神著人謂之曰。” (“Koko ni tsuki no kami hito ni kakarite iite notamawaku.”); 2: “日神著人。”

(“Hi no kami hito ni kakarite.”)

116 梁塵秘抄; “Songs to Make the Dust Dance on the Beams”

117 “東には女は無きか男巫さればや神の男には憑く” (“Azuma ni wa onna wa nakika otokomiko sarebaya kami no otoko ni wa tsuku”).

118 Hori 1968, 181–2; Yanagita 1984, 21; Nelson 1996, 123; Kimura 2017a, 4.

85 mediumship was typically a woman’s job. This is consistent with shamanic traditions seen elsewhere in northern and eastern Asia, where women are often thought to hold more sacral power, and thus be a more natural choice for the role of intermediary between humans and the divine119. An examination of this belief in the Japanese context will follow shortly.

This song, however, also raises another important point: that the way in which sacral power and sacral roles were gendered could vary regionally. Although the attitudes of the author treat the

“otoko-miko” (男巫, “man-miko”) of the Eastern Provinces120 as a bizarre inversion of the norm, the fact that there were male shamans in the east is itself a point of significance, suggesting that the gendering of sacral roles, such as mediumship, may have been more regionally variable than central records may lead us to believe. Studies in regional clan-god worship also suggest the presence of male shamans, with or without a female shamanic partner, despite the Yamato royalty’s evident preference for female kannagi only121. A further investigation of these “otoko-miko” would be useful in dispelling the Yamato homogenisation of scholarship and providing a clearer picture of the relationship between gender and sacral power throughout the whole of ancient Japan, and not just the centre; but such a task is beyond the scope of this research. For now, we must look at what we do know of the ancient miko and her role.

What function did the miko fulfil in society? In Yayoi art, they are shown performing rites for hunting and agriculture122. Fertility of hunt and harvest has been a primary concern for communities throughout human history, and it would be no surprise if the miko tradition began as a means of

119 Blacker 1982, 27; Kuratsuka 1996, 13.

120 The “Eastern Provinces” (東 azuma/東国 azuma-no-kuni, tōkoku) in ancient texts refers roughly to the modern Kantō region and parts of the Tōkai region, beginning with modern Shizuoka. The Tōhoku region of north-eastern Honshu was usually referred to separately as Michinoku or Mutsu.

121 Meeks 2011, 209.

122 Ambros 2015, 13.

86 securing divine blessings of abundance. Some interpretations also cast the early miko as the performer of rituals of healing and human fecundity123; also crucial spheres in an age before modern medicine or IVF. The medium or ritualist fulfilled a vital role in society: friendly gods and spirits could make a community prosperous through well-timed rain or lush growth, but, if angered, could also do untold harm through illness, natural disasters, and the like. The intercession of a shaman who could make contact with the supernatural was thus necessary to ensure protection and blessings for a community124. Folk rituals for miko at the village level show divinations primarily performed at the times of rice-planting and the harvest125, the key concerns for a small farming community, whereas court miko in the Engishiki were entrusted with protection of the realm, the palace, and the imperial person126. Although court and village miko served quite different purposes, this variation derives from the different priorities of their respective levels, each one fulfilling an important purpose within her context. The performance of the ancient miko was considered central to the prosperity of the community, potentially granting her a high status within it.

How does this status manifest? Within the traditional “golden age” narrative – such as the works of Sakima or Takamure – the miko is interpreted as a theocratic ruler, guiding communities via oracles, sometimes alongside a vice-ruler who handled the more mundane administrative duties127.

Himiko (see Ch. 4.1) may be treated as representative of such a tradition128. Maeda Haruto (前田晴

人) proposes a line of ancient priestess-queens stretching back into Japanese prehistory129, drawing from the “shaman-queen” images of early- and mid-20th century scholarship. Naoki instead

123 Ōi and Miyagi 1974, 16; Ambros 2015, 6.

124 Blacker 1982, 21.

125 Blacker 1982, 110.

126 Philippi 1990, 91.

127 Sakima 1926, 64; Takamure 1966b, 78; Hori 1968, 191; Ōi and Miyagi 1974, 17.

128 E.g. Ōi and Miyagi 1974, 17.

129 Maeda 2008, 3.

87 characterises the miko’s power as considerable but unofficial: the miko was rarely a formal leader, but nevertheless held a position of considerable influence through her pronouncements130; her authority existing outside “official” structures of rulership perhaps accounting for a lack of presence within central records, and creating a precariousness to her position explaining how it could be so easily lost in later eras.

The depiction of miko in the records is far closer to Naoki’s assessment than Sakima’s or

Takamure’s. In the Kojiki and Nihon Shoki, miko appear as oracles and diviners, sought by leaders desirous of divine advice. When Sujin brings Yamato-Totohi-Momoso-Hime to Kamu-Asajihara, he is seeking the reason and solution behind plague and social unrest causing upheaval in his realm

(Nihon Shoki 5.Sujin 7.2.Shinbō); when Chūai enlists the shamanic aid of Jingū, he wishes advice on a planned attack against the Kumaso131 (Kojiki 2.Chūaiki; Nihon Shoki 8.Chūai 8.9.Kibō). Although the stories themselves are of doubtful historicity, Naoki reminds us that they highlight the gravity with which oracles were treated in ancient Japan132. Divinations were almost certainly performed before major actions were taken, and served as the foundation of many political decisions133; Naoki compares the Japanese use of divination in politics to the political function of the Delphic Oracle of ancient Greece, who was famously consulted by kings for advice on governance134. Only after the formation of the Japanese ritsuryō bureaucracy in the mid-seventh century did the use of oracles in policy direction peter out135.

130 Naoki 2009, 219–20.

131 熊襲; a native people of southern Kyushu.

132 Naoki 2009, 223.

133 Kidder 2007, 127; Naoki 2009, 219.

134 Naoki 2009, 220.

135 Naoki 2009, 225.

Shamanic performance appears within the records as a means by which women may leverage status within their clan, acting as assistants and advisors to a male paramount. The influence which a miko might wield was presumably limited by her natal clan and region; the powerful shamanesses of central record are all of patrilineal royal descent, the roles they occupy not accessible to women of humbler birth. Furthermore, we must remember that, while individual women benefited from shamanic performance, not all women were miko. The “golden age” narrative is not incorrect, per se, in its assessment of a link between women’s magico-religious power and temporal authority, but there is a nuance here not encompassed in Sakima or

Takamure’s images of theocratic shaman-queens136. Rather, ancient Japan presents as a society in which, rather than being unambiguously revered and powerful as a result of magico-religious performance, women had the ability to access social capital and influence through religious activity, but the position was ephemeral and conditional, and the performance did not necessarily empower all women. This stands in contrast to the traditional line of “golden age” theories, in which women appear as universal beneficiaries of a cultural connection between the feminine and the spiritual137.

3.2.2. The Gendering of Sacral Power

Let us examine this “cultural connection” between women and the divine in more detail. It appears clear, at least, that the culture of the Yamato hegemony gendered mediumship as a female role, even if this was not the case throughout the entirety of ancient Japan138. Our sources, unfortunately, are rather elusive as to how this gendering was reckoned or interpreted. In the Ryōjin

Hishō song quoted above, the view of shamanism as a female occupation is cast as an entirely natural one, which the listener is expected to share, joining in with the singer in their wonder at these strange eastern “otoko-miko” (Ryōjin Hishō 2.556). The Engishiki is also quite clear in its

136 Sakima 1926, 64; Takamure 1966b, 78.

137 E.g. Takamure 1966b, 119–20; Ōi and Miyagi 1974, 18; Yanagita 1984, 41.

138 Meeks 2011, 209.

89 proscription that the court priestesses, or mikannagi, must be female, but again does not feel the need to explain this point (Engishiki 3.Jingi 3). Presumably, within Yamato culture (or, at least, the culture of the Yamato elite), the gendering of shamanism was simply taken for granted, the rhetoric behind it either seen as too obvious to bother writing down, or else lost throughout the ages as the weight of tradition and convention precluded the need for explanation.

The limitations of textual evidence on the matter of religious gendering make historical examination highly speculative, but one common explanation appears within the “golden age” narrative: that women, in ancient Japan, were considered to hold greater spiritual power than men, and thus naturally dominated the religious sphere; an argument which survives even into relatively modern scholarship139. We have already discussed in the previous section the complications introduced by this blanket statement regarding “ancient Japan,” in the face of evidence suggesting regional diversity of sacral gendering. But even within the specific context of Yamato, let us delve a little deeper into what this idea of female spiritual supremacy means, and how it comes about.

In Takamure’s vision of the “golden age” narrative, belief in the mystical powers of woman was once a worldwide phenomenon, surviving in the tales of witches, sorceresses, and other magical female figures found throughout a myriad of cultures140. Women, she claims, were once respected for these abilities, even revered as divine figures, but as patriarchy developed, what once was viewed as powerful divine possession was recast as female “hysteria”; the very mechanisms of social gender difference that had once afforded women respect were re-envisioned as negative qualities141.

In her argument, the respected sacral power from which women derived social status and centrality was transfigured into a subversive, destructive force representing a newfound sense of the female

139 E.g. Kuratsuka 1996, 13; Matsumura 1999, 76.

140 Takamure 1966b, 119.

141 Takamure 1966b, 118–9.

90 as the fearful, liminal “other”142, a position which recalls norm-inverting bogeywomen such as the gorgons or Amazons mentioned above143. Takamure draws a dichotomy between female sacral power – associated with a female-centric social order and a deep reverence for women – and subversive magical power or “witchcraft” – marked by a male-centric social order which disparages women as a fearful “other”.

This narrative of woman-reverence turned to woman-loathing contains a few issues, one of them being its conflation of “reverence” with social centrality. In Takamure’s iteration of the “golden age” narrative, this reverence for womanhood and female power exists within a matrilineal, matriarchal social structure with women at the centre144; a gynocentrism with which orthodox historians, in her view, struggled to come to grips145. Female spiritual power is cast as a natural result of women’s key position in society.

Let us examine, however, the argument for ancient female spiritual supremacy put forth by

Takamure’s contemporary, Yanagita, who states (rather condescendingly) that women were considered most suitable for the role of possessed shaman due to being more “emotional,” more likely to display “abnormal psychology,” and the hormonal changes caused by menstruation146.

Yanagita does not cast these as negative qualities, within this historical context; his narrative of ancient Japan holds that the spiritual acuity with which women were credited as a result of these traits as highly respected147, more akin to Takamure’s own images of female sacral power than the

142 Takamure 1966b, 118–9; Germanà 2010, 66.

143 Beard 2017, 68–70.

144 Takamure 1966b, 7–8.

145 Takamure (1966a), 6–7.

146 Yanagita 1984, 21.

147 Yanagita 1984.

“hysteria” and monstrous witchcraft she presented as an othering corruption of such148. However,

Yanagita’s work nevertheless rests upon what Tanaka critiques as a mysticised, othered image of femaleness that sacralises points of difference from the male body149 – menstruation, the childbearing capacity which he argues is linked to power in fertility rites150 – and in doing so, casts women as strange, mysterious divergences from a male “default”, their spiritual power a state of

“benign otherness,” an androcentric perspective quite at odds with “golden age” narrative’s notion of ancient Yamato as a female-centric society. As Faure points out, these issues of centrality, oppression, and temporal authority are never truly addressed within Yanagita’s work, brushed off in favour of a nostalgic image of a lost past151.

Scholarship of the last few decades has also raised the issue of male priests. Although men as shamans were clearly unusual within Yamato, this does not mean that ritual as a whole was purely the dominion of women; rather, as Yoshie’s research shows us, male priests often had their own duties acting in conjunction with miko152. Let us examine the Kojiki’s accounting of the divination of Empress Jingū:

The Emperor [Chūai] played upon a koto, with Takeshi-Uchi-no-Sukune-no-Ōmi acting as

interrogator, questioning the deity. Thus the Empress was possessed by a god…153 (Kojiki

2.Chūaiki)

148 Takamure 1966b, 118–9.

149 Tanaka 1996, 183.

150 Yanagita 1984, 22.

151 Faure 2003, 293.

152 Yoshie 1987, 35.

153 “天皇控御琴而。建內宿祢大臣居於沙。請神之命。於是太后歸神…” (“Sumera-mikoto mikoto hikitamaite. Takeshi-Uchi-no-Sukune-no-Ōmachigimi saniwa ni ite. Kami-no-mikoto wo koimatsuriki. Koko ni

ōkisaki kamigakari shitamaite…”)

While Jingū takes the role of medium and divine vessel, this is done in collaboration with two male figures, Chūai and Takeshi-Uchi-no-Sukune. Chūai’s role is to play the koto, assisting in the inducement of possession, while Takeshi-Uchi-no-Sukune acts as the saniwa154 – the “interrogator” who questions the deities or spirits speaking through the miko in her trance state – a role associated with male priests155. Yoshie’s studies on gender in ritual portray further gendered allotments of ritual performance with a basis in secular gendered division of labour, such as priestesses being charged with the preparation of rice offerings or the weaving of offered clothing156. Faure additionally casts a distinction between fertility symbolism of childbearing – which centres women – and fertility symbolism of sexual intercourse – which does not, arguing that it is the latter rather than the former which typifies agricultural rites157. Yoshie uses this contrapuntal religious performance to argue against “female spiritual power” as an ancient concept, rejecting the notion of mediumship as a role of greater spiritual acuity than a male priest’s duties, but rather that ritual efficacy was secured by a collaboration that mimicked secular labour divisions158.

Even among the Yamato imperial clan, such divisions are not strictly solid or universal.

Suinin’s daughter Ō-Nakatsu-Hime (大中姫) inherits mastery of the “sacred treasures”159 from her aging older brother Inishiki (五十瓊敷); the only issue in transferring this ritual role from man to woman is Ō-Nakatsu-Hime’s concern that her female body will be too weak to climb up to the

154 In the Kojiki, 沙 ; more typically, 審神者

155 Blacker 1982, 272; Yoshie 2007, 206.

156 Yoshie 2007, 160, 210–1.

157 Faure 2003, 293.

158 Yoshie 2007, 160, 212.

159 神宝, kantakara.

93 storehouse, a fear nullified by nothing more than a ladder (Nihon Shoki 6.Suinin 87.2.Shinbō)160. Here

Inishiki and Ō-Nakatsu-Hime, despite being of different genders, are capable of performing the same sacral role; some ritual duties appear as negotiable, rather than strictly gendered, and certainly not the sole domain of women.

The aspersions that modern work casts on the notion of sacral power as a specifically female endowment are persuasive, and yet there is one question still left open: why, within such division of ritual duties, was the specific position of shaman still gendered female in Yamato? In other words, why is it Jingū possessed by the deity, and not her husband? The miko was by nature a liminal figure, straddling the boundary between mortal and divine, acting as intermediary between gods and humans. In the ritualised exchange between gods and humanity, it is the masculine role of the saniwa who represents the latter, questioning the possessing deity or spirit on the mortal community’s behalf. The gendering of mediumship appears not as a question of “sacral power” as a feminine quality, but rather as a contrapuntal performance of a central/“default” male and a liminal/“other” female; an arrangement which implies an androcentric social order that nevertheless affords women a substantiative role, more nuanced than the “golden age” narrative conflict of early complete empowerment against later complete subordination would suggest.

160 “五十瓊敷命謂妹大中姬曰。我老也。不能掌神寶。自今以後。必汝主焉。大中姬命辭曰。吾手弱女

人也。何能登天神庫耶。五十瓊敷命曰。神庫雖高。我能爲神庫造梯。豈煩登庫乎。” (“Inishiki-no-

Mikoto iroto Ō-Nakatsu-Hime ni iite notamawaku. Ware oinu toshite. Kantakara wo tsukasadoru koto yokarazu. Ima yori nochi wa. Kanarazu imashi tsukasadore. Ō-Nakatsu-Hime inabite notamawaku. Are tayowame nari. Nanzo yoku ama-no-hokura ni noboran ya. Inishiki-no-Mikoto notamawaku. Hokura takaku to iedomo. Ware yoku hokura no tame ni hashi wo taten. A ni hokura ni noboru ni wazurai aran ya.”)

3.2.3. The Female Body as Source of Pollution

The liminal, othered status of women asserts itself in another aspect of Yamato tradition: pollution taboos surrounding menstruation and childbirth. Menstrual blood and parturition blood, later referred to euphemistically as “red uncleanness”161, were regarded as sources of pollution, offensive to the gods162 and antithetical to sacral power163. It was not simply a matter of blood taboos, either; blood from any source was polluting, but blood that was shed through injury resulted in a much shorter period of ritual contamination than blood that was shed through menstruation or childbirth164.

This notion of menstrual pollution was also seen within Japanese Buddhism, syncretically adopted from local spiritual beliefs and later harmonised with continental discourses on the “Blood

Bowl Hell”165. In Japanese Buddhism, menstrual taboos became an argument for women’s inherent moral inferiority and were used to justify the complete exclusion of women from ritual spaces166.

Blacker argues that this sense of moral judgement is absent from the understanding of taboo in kami-worship, pointing out that many other sources of ritual impurity – such as illness, injury, or proximity to death – were unavoidable “facts of life” and were not intended as negative reflections upon those subject to them167. It is true that, from what we can see in kami-worship, the exclusion

161 赤不浄, aka fujō

162 Faure 2003, 68.

163 Blacker 1982, 42.

164 Blacker 1982, 42.

165 血の池地獄, chi no ike jigoku. As described in the Chinese Blood Bowl Sutra (血盆教, Ketsubon- kyō/Xuèpénjīng) that came to Japan in the mediaeval period, a special posthumous punishment for women for the pollution caused by menstruation (Bodiford 1993, 206; Meeks 2010, 18; Ambros 2015, 85).

166 Okano 1993, 28–9; Faure 2003, 223–4.

167 Blacker 1982, 42.

95 was only temporary – as shown by the extensive involvement of women in ritual – and any accidental “contamination” could be resolved fairly easily through a purification ceremony168.

Discourses also appeared at various points exempting certain women from the taboo entirely: the saigū, or royal priestess of Ise, for example, was sometimes said to have a special white menses, free from pollution, for as long as the gods accepted her; her child-assistants, the kora, traditionally ended their term of service with their menarche, but it was held that, through force of will, they could delay this until the age of 20 or even 30169. These dialogues provided a means by which certain female practitioners might sidestep the usual considerations of menstrual impurity, and cast these women as special beings “above” the ordinary pollution of menstruation.

Even so, these menstrual and parturition blood taboos reflect, as Faure puts it, the “fear of female otherness”170: an androcentric social order in which the male body is seen as the “default,” and the points where the female body diverges from this norm – menstruation, pregnancy, childbirth – become mysterious, alien, even grotesque171. The positioning of menstruation and childbirth as sources of ritual pollution, anathema to divine power, suggests that the spiritual traditions of Yamato were themselves shaped within an androcentric framework. The sacral world appears not as a space of female dominance (per Takamure, Yanagita, and other early-type iterations of the “golden age” narrative172) nor as a space of contrapuntal egalitarianism (per

Yoshie173), but rather as a cooperation of central male and liminal female ritualists. The liminality of the miko is itself a complex thing, at once the source of her spiritual power – and the social clout that could be obtained thereby – but also linked to pollution, subversion of the spiritual. While

168 Faure 2003, 71.

169 Tanaka 1996, 117–8; Faure 2003, 300.

170 Faure 2003, 68.

171 Faure 2003, 293.

172 Takamure 1966b, 119–20; Ōi and Miyagi 1974, 18; Yanagita 1984, 41; Kuratsuka 1996, 13.

173 Yoshie 2007, 160, 212.

Takamure draws a distinction between the image of the “respected priestess” and the “reviled witch” as representing opposing social positions of womanhood174, are the two truly so contradictory? The sacral and the subversive both appear as aspects of the “power-of-otherness” attributed to this liminality.

In the Yamato mythic canon, these tensions of pollution and otherness are embodied not only in the sexless motherhood of Amaterasu, but also in the sexuality of Izanami. Let us recall the virgin motherhood of Amaterasu and Athena, and how their mythic ability to be mothers without sex or childbirth exemplifies the contradictory ideals of a male-dominated society that at once valorises motherhood as a female role, while viewing the necessary preliminaries of menses, sex, and/or childbirth as polluting or defiling175. Whereas Amaterasu in the royal canon is presented as an idealised “virgin mother,” an androgyne born (in some accounts) not of traditional labour but of a male god’s ablutions176, a royal ancestress yet unpolluted with the realities of sex and childbirth177, the creator-goddess Izanami is at once mother of the entire archipelago and many of the gods, and at the same time bearing the brunt of androcentric revulsion towards childbirth and female sexuality178.

In the official Yamato corpus of myth, the creation of the earth, the islands of Japan, and many of the key gods is attributed to a pair of deities: the male god Izanagi (伊邪那岐/伊奘諾) and

174 Takamure 1966b, 118–9.

175 Matsumura 1999, 67, 98.

176 In the Kojiki, and a variant account of the Nihon Shoki, Amaterasu is born when Izanagi cleanses himself following a visit to the underworld, although motherhood may in some sense still be attributed to Izanami (as is the case with her younger brother Susa-no-O) (Kojiki 1; Nihon Shoki 1.5.6).

177 Matsumura 1999, 97–8, 102.

178 Matsumura 1999, 102.

97 the female god Izanami. When they – the first gods to possess sexed bodies179, as may be surmised from their initial wonderment over their genitalia, or the Nihon Shoki’s note that this marked the first conjugal union (Kojiki 1, Nihon Shoki 1.4) – are first attempting to conceive, it is Izanami who takes the sexual initiative; in the Kojiki, the sickly children that result are explicitly blamed on the impropriety of this role-reversal, and only when she allows the man to take the sexual lead are they able to conceive fruitfully:

Thus the two deities then said to one another: “Only poor children are being born to us. We

ought to seek an explanation from the Heavenly Deities.” So the pair headed up to call upon

the Heavenly Deities and ask them why this was the case. The Heavenly Deities divined the

answer by burning a stag’s scapula and observing the cracks, and informed the couple: “The

children are poor because it is the woman who has spoken first.”180 (Kojiki 1).

179 Copeland 2018, 14.

180 “於是二柱神議云。今吾所生之子不良。猶冝白天神之御所。卽共參上請天神之命。尓天神之命以。

布斗麻迩尓…卜相而詔之。因女先言而不良。” (“Koko ni futahashira no kami hakarite notamawaku. Ima aga umeru miko yokarazu. Nao amatsu-kami no mimoto ni mōsubeshi. Sunawachi tomo ni mainoborite amatsu-kami-no-mikoto wo koitamaiki. Koko ni amatsu-kami-no-mikoto mochite. Futomani ni…urabete noritamawaku. Omina no koto sakidachishi ni yorite yokarazu.”).

In the Nihon Shoki, this connection of female initiative with fertility troubles is not made, but

Izanami is still chided by her husband when she makes the first advance:

The male god was most displeased, saying: “I am a man. It stands to reason that I ought to

speak first. How can it be that the wife speaks first instead? This is a most inauspicious

thing.”181 (Nihon Shoki 1.4).

These passages enshrine as the natural order a specific attitude towards gender roles held and promoted by this Yamato mythic corpus: initiating sex is the man’s proper role, and passively responding to a man’s advances is the woman’s. The man leads, the woman follows; the man acts, the woman responds. As in religious ritual, success is achieved through man and woman cooperatively playing their respective roles, and yet the nature of these roles centres the male.

The two deities go on to birth the islands of Japan and a multitude of gods, but – as seen in the Kojiki and a variant text of the Nihon Shoki – ultimately the process of childbirth ends poorly for

Izanami, with the birth of the fire god Hi-no-Kagutsuchi (火之迦具土/軻遇突智):

Next was born Hi-no-Yagi-Hayao-no-Kami…also called Hi-no-Kagutsuchi-no-Kami. In giving

birth to this child, Izanami-no-Mikoto’s vagina was burned, and she became greatly sickened

and lay down.182 (Kojiki 1).

Next was born the god of fire, Kagutsuchi. At this time, Izanami-no-Mikoto was burned by

him, and passed away.183 (Nihon Shoki 1.5.2).

181 “陽神不悦曰。吾是男子。理當先唱。如何婦人反先言乎。事旣不祥。” (“Ogami yorokobazushite notamawaku. Are wa kore otoko nari. Kotowari masa ni mazu tonau beshi. Ikanzo taoyame no kaeshite koto saidatsu ya. Koto sude ni saganashi.”)

182 “次生火之夜藝速男神。…亦名謂火之迦具土神。…因生此子。美蕃登…見炙而病臥在。” (“Tsugi ni Hi- no-Yagi-Hayao-no-Kami wo umimashiki…Mata no mina wa Hi-no-Kagutsuchi-no-Kami to mōsu…Kono miko wo umimashishi ni yorite. Mihoto…yakaete yami koyaseri.”).

As she lies dying, she excretes the unclean substances of urine, faeces, and vomit, which themselves generate deities (Kojiki 1; Nihon Shoki 1.5.2). Death in (or resulting from) childbirth was unfortunately commonplace in the ancient world, and this myth reflects such fears; it also, however, links Izanami giving birth to her illness and death, and the pollution thereof.

Izanagi descends to the underworld, Yomi, to rescue his wife, but looking upon her (against her wishes) he finds her body further touched by pollution, “decaying and swarming with maggots”184 (Kojiki 1; Nihon Shoki 1.5.6). Here Izanami transitions from the role of life-giving creator goddess to chthonic death-goddess: shamed at being seen in her rotted state, Izanami furiously pursues her former husband, vowing to slay a thousand people in his realm each day185, taking the title Yomotsu-Ōkami (黄泉大神), the “Great Goddess of the Underworld” (Kojiki 1; Nihon Shoki

1.5.6). This realm is associated with death, and thus with uncleanness and pollution, all things despised by the gods; a “vile, hateful land of filth”186 from whose taint Izanagi must purify himself after visiting (Kojiki 1; Nihon Shoki 1.5.6).

183 “次生火神軻遇突智。時伊弉冊尊爲軻遇突智所焦而終矣。” (“Tsugi ni ho no kami Kagutsuchi wo umu.

Toki ni Izanami-no-Mikoto Kagutsuchi ga tame ni kayarete kamisarimashinu.”)

184 Kojiki: “宇士多加礼斗呂々岐弖。” (“Uji takare tororogite.”) ; Nihon Shoki: “則膿沸虫流。” (“Sunawachi unawake uji takare.”)

185 Kojiki: “伊耶奈美命言。愛我那勢命。爲如此者。汝國之人草。一日絞殺千頭。” (“Izanami-no-Mikoto mōshitamawaku. Utsukushiki aganase-no-mikoto. Kaku shitamawaba. Imashi no kuni no hitokusa. Hitohi ni chikashira kubirikoroshi ni korosan.”); Nihon Shoki: “時伊弉冊尊曰。愛也吾夫君言如此者。吾當縊殺汝所

治國民日將千頭。” (“Toki ni Izanami-no-Mikoto notamawaku. Uruwashiki aganase-no-mikoto kaku shinotamawaba. Are masa ni imashi ga shirasu kuni no hitokusa hitoi ni chikōbe kubirikorosan.”)

186 Kojiki: “伊那志許米志許米岐…穢國” (“Ina shikome shikomeki…kitanaki kuni”); Nihon Shoki: “不須也凶目

汚穢之處” (“Ina ya shikomeki kitanaki tokoro”).

100

Izanami is shamed for her sexual forthrightness, and villainised in her anger at Izanagi’s peeking187. Her death in childbirth links her pregnancy and labour to her corruption into a monstrous death-bringer, presiding over all things polluted. Her legend thus underscores both notions of female initiative (especially sexual initiative) as against the natural order and in need of containment, and the connection between the female-sexed body, sex, childbirth, and ritual defilement188. Yet at the same time, she is also mother-goddess, creator-goddess, bringer of life as well as death; even her unclean excretions birth deities. The legend of Izanami as it appears in the Yamato mythic corpus, promulgates androcentric narratives of parturition pollution, but also underscores the tensions and contradictions inherent in the “othering of femaleness,” Izanami’s capacity for childbirth both wondrous and profane.

Another Yamato legend of the polluted female body concerns the agricultural goddess

Ōgetsu-Hime or Ukemochi (保食), who is capable of producing a variety of abundant foodstuffs from her orifices. This power of generation elicits the disgust of a male deity who slays her in his revulsion189:

Haya-Susa-no-O-no-Mikoto stood and wondered at this act, thinking it most filthy in nature.

Thus he slew Ōgetsu-Hime-no-Kami.190 (Kojiki 1).

187 Copeland 2018, 14; compare this also to the tale of Yamato-Totohi-Momoso-Hime’s marriage to Ōmono-

Nushi, in which the wife’s violation of her god-husband’s command not to look at him results in divorce and ultimately, her death (Nihon Shoki 5.Sujin 10.9.Jinshi).

188 Copeland 2018, 19–22.

189 Matsumura 1999, 105.

190 “速湏佐之男命立伺其態。爲穢汙而奉進。乃殺其大冝津比賣神。” (“Haya-Susa-no-O-no-Mikoto sono shiwaza wo tachiukagaite. Kitanakushite tatematsuru to omōshi. Sunawachi sono Ōgetsu-Hime-no-Kami wo koroshitamaiki.”).

101

Tsukiyomi-no-Mikoto was enraged by this, and said, flushing with anger: “How filthy, how

base! How darest thou offer up to me that which thy mouth has vomited forth?” Thereupon

he drew his sword and cut her down.191 (Nihon Shoki 1.5.11).

Unlike Izanami, however, Ōgetsu-Hime/Ukemochi retains her connection to abundance even in death, grain sprouting from her corpse. In the Nihon Shoki variant text; Tsukiyomi is censured by

Amaterasu for this act of violence; in the Kojiki, it is committed by the known troublemaker Susa-no-

O (Kojiki 1; Nihon Shoki 1.5.11). Neither her fertility nor her perceived pollution are connected explicitly to sex and childbirth, as Izanami’s are; her vagina is, perhaps surprisingly, not one of her cornucopian orifices. In Izanami’s tale, however, the female body is more explicity linked with pollution and uncleanness, death and disease192; female sexual submission is depicted as the natural order of things, required to keep cosmic balance193.

Even with the multiple versions of tales found across the Kojiki and Nihon Shoki, of course, these stories reflect only the specific perspective of the Yamato elite. We do not know if there exist earlier, or regionally disparate, variants of these stories which portray a very different sense of gender relations. Although we have evidence of regional variance in the gendering of shamanism, most of what we do have is filtered through a Yamato perspective, and thus our understanding of this regional diversity is fragmentary and limited. But the textual and ritual evidence which does survive today displays an androcentric social and religious order in the polity of Yamato. This androcentrism manifests in ways quite different from the later patriarchal structures of the ritsuryō

191 “是時月夜見尊忿然作色曰。穢哉。鄙矣。寧可以口吐之物敢養我乎。廼拔劍擊殺。” (“Kono toki ni

Tsukiyomi-no-Mikoto ikari omōterishite notamawaku. Kegarawashiki kana. Iyashiki kana. Mushiro kuchi yori tagureru mono wo motte aete are ni kau beken ya tonotamaite. Sunawachi tsurugi wo nuite uchikoroshitsu.”).

192 Ambros 2015, 29.

193 Ambros 2015, 30.

102 state, such as a stronger emphasis on the cooperation of the contrapuntal gender roles even if those roles are not equal in centrality. Rather than the linear transition from female-centric or wholly egalitarian social structure to oppressive patriarchy described by the “golden age” narrative, the story of womanhood in ancient Yamato alone is a far more complex construction than can be encompassed by the neat dichotomy of “empowerment” versus “subjugation.”

103

CHAPTER 4: FEMALE RULERSHIP IN THE “GOLDEN AGE”

But what of the women at the apex of the ancient Japanese social hierarchy: Himiko, Jingū, and other female chieftains and paramounts? Within the “golden age” narrative, these figures appear as representative of a broader tradition of female sacral rulership1: temporal authority obtained via magico-religious performance2. Yet our previous discussion (see Ch. 3.2) of the status of the miko suggests a more complex relationship between female spiritualism and access to temporal authority`: a more conditional, ephemeral relationship than the glorious priestess-queens of the

“golden age” narrative. Where does that leave Himiko, Jingū, or the enduring image of the hime co- chieftain? In the previous chapter, we looked at images of female shamanism and sacral performance; now, let us examine ancient female rulership, and how its appearance in the records compares to that within the “golden age” narrative and its subsidiary threads.

First, we turn to the elusive figure of Himiko, the apparent shaman-queen of third-century

Japan who appears within Chinese sources. Absent from Japan’s own timeline of rulership, Himiko has provoked considerable speculation among historians, particularly within the sub-discipline of women’s history. Variously interpreted as an erased ancestress of the Yamato regime, a Kumaso chieftain unrelated to Yamato, a mere sacral figurehead, a co-ruler alongside her brother, or a misinterpreted reference to some royal woman already extant within the Rikkokushi – Jingū,

Yamato-Hime, Yamato-Totohi-Momoso-Hime – little concrete is actually known about the “Queen of

Wa.” Building upon the works of Kidder and Yoshie, we will examine the intricacies of her rule and its broader socio-cultural context that have largely gone unexamined in her “golden age” narrative positioning as an uncomplicated emblem of female power. These include the complex interweaving

1 Takamure 1966b, 114; Hori 1968, 191; Ōi and Miyagi 1974, 17; Blacker 1982, 27; Maeda 2008, 83.

2 Sakima 1926; Takamure 1966b; Yanagita 1984; Sekiguchi 1987; Kuratsuka 1996; Naoki 2008.

104 of liminality and centrality; of temporal authority, sacral power, and personal autonomy; and the relation of her rule to the broader position of women in third-century Japan.

Next, we will make a similar study of the figure of Empress Jingū, appearing within the Kojiki and Nihon Shoki, whose regency and oracular performance has also led to her use as proof-positive of female sacral rulership.But as we shall see, she too is not an uncomplicated example of female power and splendour, her position within Yamato myth-history that of facilitator of the patriline and establisher of divine right over Korea; the latter role both presented as unusual for a woman and credit for it at least partially bestowed upon her unborn son. The mythic image of Jingū is one of an

“exceptional woman;” the antiquity of her legend in considerable doubt, raising uncertainty on its ability to reflect gendered rulership of ancient Yamato in the first place.

Finally, we will build upon our study of regional diversity in ancient Japanese customs and gender roles. In Chapter 3.2.1 above, we noted regional disparity in sacral traditions and the gendering thereof, clashing with the “golden age” narrative-derived interpretation of female magico-religious performance. Here, we will take a closer look at the possibility of regional diversity in gendered rulership traditions, against the backdrop of one of the most enduring threads of the

“golden age” narrative: the concept of the “himehiko system,” and its positioning of contrapuntal male-female rule as the standardised format of rule in pre-Yamato dominion Japan. Through this, we shall see how the “golden age” narrative’s presentation of gendered power in “ancient Japan” homogenises a variety of regionally disparate rulership customs and elite male/female pairs into a single universal structure, the “himehiko system.”

Through examining the complexities not encompassed by the “golden age” narrative – the tensions of power buffed out into examples of female glory, the regional diversity conflated into a singular image of “Japanese culture” – we may work towards building a narrative framework that allows space for these issues to be examined in all their depth.

105

4.1. Himiko, Queen of Wa

4.1.1. An Introduction to Himiko

For our first study, we turn to the enigmatic figure of Himiko, the third-century “Queen of

Wa.” Within the “golden age” narrative, this apparent cloistered priestess-queen, elided from the annals of Japanese history by Yamato dynasty records, she appears as a key figure, the characteristics of her rule inspiring Sakima’s theories of nyoji3. But just who was Himiko, really?

Queen Himiko is first recorded in the third-century Chinese text, the Records of Three

Kingdoms, in its “Treatise on the People of Wa”. This section describes Chinese diplomats’ impressions of the land of “Wa” (Japan), and its relations with the Chinese state of Wei. Wa is presented as a collection of polities (chiefdoms), most gathered under the authority of “the country of Yamatai, where the Queen dwells in the capital”4, but some independent. The “Treatise” is the earliest known written depiction of Japan – predating the Kojiki by nearly five centuries – and is thus of huge importance to the study of early Japan. Of particular interest in the context of the “golden age” narrative is the figure of this Queen, a female ruler largely absent from the official Yamato records, and with apparent magico-religious qualities to boot:

The land once had a man as king, who abided for seventy or eighty years. Then Wa was

thrown into disorder, warring for years amongst themselves. Eventually, they came together

to install a woman, named Himiko, on the throne. She knew the way of spirits, and could

enchant the people. Even as she grew older, she never married. She had a younger brother

3 Sakima 1926, 3.

4 “邪馬壹國、女王之所都”. (“Xiémăyīguó, nǚwáng zhī suŏ dū.”) Rendered as 邪馬台國 (Xiémătáiguó) in the

Book of the Later Han and 邪靡堆 (Xiémĭduī) in the Book of Sui.

106

who assisted her in ruling the country. The Queen herself was rarely seen. She had a

thousand maids attending her, and only one man to bring her food and drink and convey her

words. The palace in which she lived was surrounded by strong walls and watchtowers, and

was kept guarded by many soldiers.5 (Sanguozhi, Wei Shu 30, Worenzhuan).

Himiko is shown as living in strict ritual seclusion, a virgin whose only male contact was her younger brother, who is typically identified as the “one man” who brings her food and transmits her proclamations. Her cloistering is reminiscent of the later saigū priestesses of Ise, whose own strictures will be discussed in greater detail in Chapter 5.3. Like the saigū, her seclusion was presumably for the sake of maintaining ritual purity, avoiding pollutants to ensure the efficacy of her spiritual power.

The terminology used for Himiko’s magico-religious practices is guidao or guishendao, the

“way of spirits/ghosts”6. Similar references to “ghost” worship appear in the Records of Three

Kingdoms7, the Book of the Later Han8, the Book of Sui9, and the Book of Liang10 in reference to the

5 “其國本亦以男子爲王、住七八十年、倭國亂、相攻伐歷年、乃共立一女子爲王、名曰卑彌呼、事鬼道、

能惑衆、年已長大、無夫婿、有男弟、佐治國、自爲王以來、少有見者、以婢千人自侍、唯有男子一人、

給飲食。傳辭出入居處。宮室・樓觀・城柵嚴設、常有人持兵守衛。” (“Qí guó bĕn yì yĭ nánzĭ wéi wáng, zhù qī bāshí nián, Wōguó luàn, xiāng gōngfá lìnián, năi gòng lì yī nǚzĭ wéi wáng, míng yuē Bēimíhū, shì guĭdào, néng huò zhòng, nián yĭ zhăngdà, wú fūxù, yŏu nándì, zuŏ zhìguó, zì wéi wáng yĭlái, shăoyŏu jiàn zhĕ, yĭ bì qiānrén zì shì, wéiyŏu nánzĭ yīrén, jĭ yĭnshí, chuán cí chūrù jūchù. Gōngshì, lóuguàn, chéngzhà yán shè, cháng yŏu rén chí bīng shŏuwèi.”)

6 鬼道, guĭdào or 鬼神道, guĭshéndào.

7 “祭鬼神，又祀靈星、社稷。…常以五月下種訖，祭鬼神，群聚歌舞，飲酒晝夜無休。…立大木，縣鈴

鼓，事鬼神。” (“Jì guǐshén, yòu sì língxīng, shèjì…Chángyǐ wǔyuè xià zhòng qì, jì guǐshén, qúnjù gēwǔ, yǐnjiǔ zhòuyè wúxiū…Lǐ dà mù, xiàn línggǔ, shì guǐshén.” “[The people of Goguryeo] worship spirits, and also make offerings to sacred stars and the gods of the earth…Every fifth month, when the seeds are sown, [the people of

107 religious practices of Korean Peninsula nations, in particular Goguryeo. These included harvest festivals, feasting, dancing, bells, and drums (Sanguozhi, Wei Shu 30, Gaogoulizhuan; Sanguozhi, Wei

Shu 30, Hanzhuan; Hou Han Shu 85, Gaogoulizhuan; Hou Han Shu 85, Sanhanzhuan; Sui Shu 81,

Gaolizhuan; Liang Shu 54, Gaogoulizhuan). The use of this term may indicate that the spiritual practice of Himiko was deemed similar in character by the Chinese observers, and was perhaps even related11. The limitations of these records must also be considered, however. The Chinese observers were not only outsiders to these cultures, as Maeda and Yoshie point out, but also approaching them from a position of assumed cultural supremacy12; furthermore, given Himiko’s reclusive nature, it is unlikely that the ambassadors would have had firsthand experience of her ritual performance. It is thus uncertain how much significance can be apportioned to the use of the term guidao to describe Himiko’s magico-religious power13.

Himiko’s realm was apparently formed from a confederation of smaller states, which maintained some level of independent identity even as they came under the authority of the

Korea] worship the spirits, and gather together for singing and dancing, and eat and drink all day and night without end…They hang belled drums from a tall tree to worship the spirits.”)

8 “好祠鬼神、社稷、零星…常以五月田竟祭鬼神.” (“Hào cí guǐshén, shèjì, língxīng…Chángyǐ wǔyuè tián jìng jì guǐshén.” “[The people of Goguryeo] worship well the spirits, the gods of earth, and the sacred stars…Every fifth month, when the rice is planted, [the people of Korea] worship the spirits.”)

9 “敬鬼神，多淫祠。” (“Jìng guǐshén, duō yín cí.” “[The people of Goryeo] revere the spirits and are most excessive in their worship.”)

10 “祭鬼神，又祠零星、社稷。” (“Jì guǐshén, yòu cí língxīng, shèjì.” “[The people of Goguryeo] worship the spirits, and also make offerings to the sacred stars and the gods of the earth.”)

11 Piggott 1997, 26; Kidder 2007, 132; Ambros 2015, 13.

12 Maeda 2008, 8, 25; Yoshie 2013, 8.

13 Sakima 1926, 65; Maeda 2008, 25; Naoki 2008, 77; Yoshimura 2010, 25; Yoshie 2013, 8.

108 monarch in “Yamatai”14. Judging by later regional power structures, these constituent “states” would have been territories – hierarchies of hamlets – held by powerful extended clans and administered by a chieftain15. It is likely that Himiko was elected from among these chieftains16, possibly as a compromise candidate after a long period of political unrest17. It is unclear whether Himiko was also chieftain of “Yamatai,” or whether the two roles were discrete18. The unrest both before and following Himiko’s reign do not suggest a particularly solidified political system, implying that the confederation was still in a nascent stage during this time19.

Although the text does not give the exact dates or length of Himiko’s reign, she is known to be in power as early as 238 CE – when she is dubbed a “Friend to Wei”20 – and as late as 247 CE, when she is said to be at war with a neighbouring state. After Himiko’s death, a male king is said to be enthroned, but is not accepted by the populace, resulting in further conflict. This is ultimately resolved by the installation of Iyo (壱与), a 13-year-old kinswoman of Himiko, as the next Queen of

Wa (Sanguozhi, Wei Shu 30, Worenzhuan)21. The relationship of the intended male successor to

Himiko and Iyo is not explicated, nor the reason for his inability to hold power.

14 Sekiguchi 1987, 15; Yoshimura 2010, 6.

15 Martin 1997, 25; Piggott 1997, 17.

16 Naoki 2008, 203.

17 Kidder 2007, 131.

18 Yoshimura 2010, 23–4.

19 Yoshimura 2010, 6, 18.

20 “親魏倭王卑彌呼”. (“Qīn Wèi Wō Wáng Bēimíhū.”)

21 “更立男王、國中不服、更相誅殺、當時殺千餘人。復立卑彌呼宗女壹與年十三爲王、國中遂定。”

(“Gēng lì nán wáng, guózhōng bùfú, gèng xiāng zhūshā, dāngshí shā qiānyú rén. Fù lì Bēimíhū zōngnǚ Yīyŭ nián shísān wéi wáng, guózhōng suì dìng.”)

109

The Book of the Later Han, compiled in the fifth century, contains a very similar account of the ancient queen. The details and phrasing of this passage closely resemble the extract from the

Records of the Three Kingdoms – presumably drawing upon it as a base – but also including new details which suggest an additional source. These include the dating of Wa’s pre-Himiko chaos to the

“reigns of Huan and Ling”22, and a further note about the strictness of Himiko’s laws. The Book of the

Later Han passage more closely links Himiko’s spiritual acuity to her queenship, presenting her accession as a result of her bewitchment of the populace. No mentioned is made within this account of Himiko’s successor (Hou Han Shu 85, Wozhuan)23. The Book of Sui, completed in 636, contains an almost identical description to the Book of the Later Han, with the key difference that Himiko is said to be attended by two men, rather than the “one man” seen in every other text. The inconsistency between this point and every other account suggests that it may be the result of erroneous copying rather than additional information (Sui Shu 81, Woguozhuan) 24.

22 Emperor Huan of Han (漢桓帝) (r. 146–168 CE) and Emperor Ling of Han (漢靈帝) (r. 168–189 CE)

23 “桓・靈間、倭國大亂、更相攻伐、歷年無主。有一女子、名曰卑彌呼、年長不嫁、事鬼神道、能以

妖惑衆、於是共立為王。侍婢千人、少有見者、唯有男子一人、給飲食、傳辭語、居處・宫室・樓觀・

城栅、皆持兵守衛、法俗嚴峻。” (“Huán, Líng jiān, Wōguó dà luàn, gèng xiāng gōngfá, lìnián wú zhŭ. Yŏu yī nǚzĭ, míng yuē Bēimíhū, niánzhăng bù jià, shì guĭshéndào, néng yĭ yāo huò zhòng, yúshì gòng lì wéi wáng. Shì bì qiānrén, shăoyŏu jiàn zhĕ, wéiyŏu nánzĭ yīrén, jĭ yĭnshí, chuán cíyŭ, jūchù, gōngshì, lóuguàn, chéngzhà, jiē chí bīng shŏuwèi, făsú yánjùn.”)

24 “桓・靈之間、其國大亂、遞相攻伐、歷年無主。有女子、名卑彌呼、能以鬼道惑衆、於是國人共立

爲王。有男弟、佐卑彌理國、其王有侍婢千人、罕有見其面者、唯有男子二人、給王飲食、通傳言語。

其王有宮室・樓觀・城柵、皆持兵守衛、爲法甚嚴。” (“Huán, Líng zhījiān, qí guó dà luàn, dì xiāng gōngfá, lìnián wú zhŭ. Yŏu nǚzĭ, míng Bēimíhū, néng yĭ guĭdào huò zhòng, yúshì guórén gòng lì wéi wáng. Yŏu nándì, zuŏ Bēimí lĭ guó, qí wáng yŏu shì bì qiānrén, hănyŏu jiàn qí miàn zhĕ, wéiyŏu nánzĭ èrrén, jĭ wáng yĭnshí, tōng chuán yányŭ. Qí wáng yŏu gōngshì, lóuguàn, chéngzhà, jiē chí bīng shŏuwèi, wéi fă shèn yán.”)

110

The Book of Liang, however, completed in 635 CE, contains an account with more detailed dating and information about Himiko’s succession. The chaotic, leaderless time before Himiko’s accession is attributed to the Guanghe Period (178–184 CE) of Emperor Ling of Han’s reign. Himiko’s death is dated to some point within the Zhengshi Period (240–249 CE), during the reign of Emperor

Cao Fang of Wei (曹芳); given that the Records of the Three Kingdoms attests events of her life as late as 247 CE, this narrows down her date of death to within a roughly two-year period.

The events following Himiko’s death – the failed accession of a male ruler, rejected by the populace, followed by the successful enthronement of Himiko’s kinswoman – proceed in the Book of

Liang much the same as in the Records of the Three Kingdoms. In the Book of Liang, however, this kinswoman’s name is given as “Toyo” (臺與) rather than the “Iyo” (壹與) seen in the Records of the

Three Kingdoms. This discrepancy presumably derives from a recording error concerning the two very similar characters 壹 (i) and 臺 (to); it is not clear which phoneme is the “correct” one. The final addition contained in the Book of Liang account adds that, following Toyo’s reign, “a man was again made king.” (Liang Shu 54, Wozhuan)25. Following from this are the “Five Kings of Wa,” beginning from the reign of Emperor An of Jin (晉安帝) (r. 382–419 CE): “San” (賛), “Mi” (弥), “Sai” (済), “Kō”

25 “漢靈帝光和中、倭國亂、相攻伐歷年、乃共立一女子卑彌呼爲王。彌呼無夫婿、挾鬼道、能惑衆、

故國人立之。有男弟佐治國。自爲王、少有見者、以婢千人自侍、唯使一男子出入傳教令。所處宮室、

常有兵守衛。…正始中、卑彌呼死、更立男王、國中不服、更相誅殺、復立卑彌呼宗女臺與爲王。其後

復立男王”. (“Hàn Líng Dì Guānghé zhōng, Wōguó luàn, xiāng gōngfá lìnián, năi gòng lì yī nǚzĭ Bēimíhū wéi wáng. Míhū wú bù jià, xié guĭdào, néng huò zhòng, gù guórén lì zhī. Yŏu nándì zuŏ zhìguó. Zì wéi wáng, shăoyŏu jiàn zhĕ, yĭ bì qiānrén zì shì, wéi shĭ yī nánzĭ chūrù chuán jiàolìng. Suŏchù gōng shì, cháng yŏu bīng shŏuwèi…Zhèngshĭ zhōng, Bēimíhū sĭ, gēng lì nán wáng, guózhōng bùfú, gèng xiāng zhūshā, fù lì Bēimíhū zōngnǚ Táiyŭ wéi wáng. Qíhòu fù lì nán wáng.”)

111

(興), and “Bu” (武), generally considered to refer to Yamato dynasty monarchs26 (Liang Shu 54,

Wozhuan)27.

Himiko’s significance to the “golden age” narrative of female power derives from two key points: first, her status as a female ruler over a portion of ancient Japan, and secondly, her practice of “the way of spirits”, the magical means by which her accession was leveraged. Given the gendering of sacral roles in ancient Yamato – as we explored last chapter – this “way of spirits” is commonly interpreted as shamanism, with Himiko as a talented miko invested with temporal authority, ruling through oracular pronouncements28. Her younger brother, in this model, takes the male priestly role of saniwa29 (see Ch. 3.2.3); her “thousand maids” including sacral assistants akin the kora of the saigū or the junior priestesses of the Ryukyuan kikoe ōkimi30. The name “Himiko” itself is sometimes interpreted as a title labelling the ancient queen as a miko31. In this sense, Himiko appears as something of an apex of the mythic “priestess-advisors” (such as Yamato-Totohi-

Momoso-Hime) of the Kojiki and Nihon Shoki; she is not the sacral advisor to a male monarch, but rather is endowed with monarchic authority in her own right.

26 Yoshimura 2010, 46.

27 “晉安帝時，有倭王贊。贊死，立弟彌；彌死，立子濟；濟死，立子興；興死，立弟武。”. (“Jìn Ān Dì shí, yǒu Wō Wáng Zàn. Zàn sǐ, lì dì Mí; Mí sǐ, lì zĭ Jì; Jì sǐ, lì zĭ Xìng; Xìng sǐ, lì dì Wǔ.”)

28 Bock 1970a, 5; Piggott 1997, 25; Kidder 2007, xii; Maeda 2008, 3; Ambros 2015, 13.

29 Maeda 2008, 20, 28; 2009, 219.

30 Sakima 1926, 66; Naitō 1976, 204.

31 Shiratori 1976, 174; Piggott 1997, 39; Maeda 2008, 3; Ambros 2015, 12–3. This competes with another argument that “Himiko” is a corrupted transliteration of himemiko (皇女/姫御子, roughly “princess”)

(Takamure 1966b, 74; Satō 2006, 141; Ambros 2015, 12),

112

The characteristics of Himiko’s reign were a key basis for Sakima’s theories of primeval nyoji, ancient female rulership mediated by magico-religious ability32. Indeed, Sakima presents Himiko as representative of a broader system of female sacral rulership, equated with the kikoe ōkimi high priestesses of Ryukyu33; her brother is read as a secondary ruler, assisting with mundane governance34. Within Takamure’s oeuvre, Himiko’s reign is similarly read as the typical format of ancient chieftaincy, the two-layered himehiko35 rule of sacral female and assistant male36.

This motif of Himiko as representative of systematic male-female paired rule is echoed within modern scholarship: Matsumura interprets her rule as the standard, and her male predecessor as an aberrant warlord37; Maeda similarly situates her within a line of sacral queens later overwritten by the Yamato patriline38. Sekiguchi also takes Himiko as proof of ancient queenship, although points out that she represented Wa in political matters, her rule sacerdotal in character but not confined to the religious sphere39. Yoshie alone takes a dissenting view, characterising her sacral performance as the standard expected for any ruler, male or female; her spiritual power not the source of her authority, but rather the product of it, unbound from her gender40. Himiko’s presence as a sacral queen, not only equal to but apparently higher in status than her brother, has cast her as a key figure within the “golden age” narrative, in both earlier and more

32 Sakima 1926, 3.

33 Sakima 1926, 65–7.

34 Sakima 1926, 65.

35 The notion of himehiko will be explored in greater detail in Section 3 below.

36 Takamure 1966a, 69; 1966b, 77, 83.

37 Matsumura 1999, 121–6.

38 Maeda 2008), 2–4, 16.

39 Sekiguchi 1987, 16.

40 Yoshie 2013, 8–11.

113 recent incarnations; the scant details about her reign perhaps only fuelling her mystique as a symbol of the lost glories of womanhood in ancient Japan.

4.1.2. Situating Himiko in History

Furthering Himiko’s mystique is her uncertain position within the history of the Yamato state, for despite her reoccurring presence within Chinese records, she is absent from the Kojiki and Nihon

Shoki, save for portions of the Chinese records cited in the latter. The years of her reign are, in the

Nihon Shoki, attributed to Jingū’s regency (Nihon Shoki 9), but the identification of Himiko with the empress-regent is unsatisfying, the two appearing as very different figures apart from their gender and magico-religious performance (see Ch. 4.2.2). This discrepancy complicates matters for us as scholars, for situating Himiko within her proper historical context is crucial for understanding how her presence interacts with and reflects upon broader Japanese history. Assessing her historical veracity, for example, will suggest to us whether we ought to read her as an extant example of female rulership – with the implications that might have for ancient gender roles as a whole – or as a myth-historic construct, existing to fulfil a particular role in a narrative. Determining her location will also be important as we grapple with the concept of regional diversity in gender roles and rulership traditions: which part of Japan did she hail from, and how does her example reflect on, harmonise with, or conflict with other images of rulership from that region?

First, let us address the question of historicity. Within the “golden age” narrative, Himiko is typically read as an extant, if somewhat mysterious, historical personage. Nor is the assumption of

Himiko’s historicity unique to the discipline of women’s history: Yoshimura, for example, in his study of the early kingship of Yamato, similarly engages with Himiko and Iyo/Toyo as historical monarchs, treating the Records of the Three Kingdoms as reliable documentary evidence41. Although Naoki occasionally calls into question the reliability of the text based on the limited perspectives of the

41 Yoshimura 2010, 4–32.

114

Chinese observers, Himiko’s queenship itself is not challenged42. As we shall see in the following section, the same does not hold true for Jingū’s reign, the veracity of which has come under considerable scholarly doubt.

Himiko’s veracity as a historical figure is supported by the surrounding context of the textual evidence: the Records of the Three Kingdoms, the first work citing her, was compiled during the third century, mere decades after Himiko’s death. The presence of dated entries regarding Himiko’s envoys to Wei suggests that the compiler, Chen Shou (陳壽), was working from contemporary court documents which, as the Chinese court had long been highly literate and bureaucratic, would doubtless have existed then even if they no longer survive today. The chronological discrepancy between the contents of a record and its production date are a key point in assessing a text’s reliability (see Ch. 2.2.2). Conversely, the oldest surviving source for the competing Yamato narrative is the Kojiki, produced nearly 500 years later; as Yoshimura reminds us, the timeline of literacy in

Japan precludes any internal written records for some 200 years after Himiko’s death43. In this respect, the dating and descriptions of the Chinese sources appear considerably more reliable.

Who, then, was Himiko to the emperors? If “Yamatai” was, indeed, Yamato, then Himiko could potentially be an imperial ancestress – and yet, her existence would also completely contradict the official chronology of the Yamato line44. The location of Himiko’s “Yamatai,” and whether it was truly identical with the imperial province of Yamato, has thus been a driving question for scholars of this uncertain period of ancient Japanese history.

For the purposes of this research, however, the significance of locating Himiko is somewhat different: our key concern is not the specifics of her relationship to the Yamato monarchy, but rather,

42 Naoki 2008.

43 Yoshimura 2010, 104.

44 Naoki 2008, 2–3; Ambros 2015, 12.

115 the aspect of regionality. Sakima, in asserting Himiko’s typicality as an ancient magico-religious female ruler, argued that her actual location did not matter, only the fact that she was “Japanese”45.

However, this assumes a homogenous practice of gendered leadership and spirituality across the archipelago; as we have seen in the previous chapter, regional variation is very much evident in gendered spiritual roles, and below (see Ch. 4.3) we will observe similar variance in formats of rulership. To engage with Himiko in her regional context – to examine her against other images of rulership and female sacral performance within her region – opens up new avenues of exploration into regional trends and diversity in ancient Japanese gender roles and rulership systems.

Furthermore, it helps us understand her, and the traditions of rulership surrounding her to which she may tie in or subvert.

Himiko’s location has, within Japanese academia, been a source of some controversy. The directions given in the Records of Three Kingdoms are unhelpful: if followed on a modern map, they appear to send the reader straight into the ocean. A slight adjustment – the ancients were, after all, not immune from copy errors – could indeed lead the reader to Yamato Province, in the Kinai46.

Alternatively, a different slight adjustment could guide the reader to Kyushu where, further complicating matters, there is in the modern day another location called “Yamato”47. Although early historians initially equated Himiko with Jingū, as the Kojiki and Nihon Shoki compilers had done, from the Edo Period (1603–1868) on new theories emerged – spearheaded by such prominent scholars as

Kojiki-den (“Commentaries on the Kojiki”) author Motoori Norinaga (本居宣長; 1730–1801), and associated with the nationalistic Kokugaku school – relocating “Yamatai” to Kyushu and recasting

45 Sakima 1926, 65.

46 畿内, the Capital Region; comprised of Yamato (modern Nara Prefecture) and its surrounding provinces:

Yamashiro (southern Kyoto Prefecture), Settsu (south-eastern Hyōgo and northern Osaka Prefectures),

Kawachi (eastern Osaka Prefecture), and Izumi (south-western Osaka Prefecture).

47 Kidder 2007, xi; Ambros 2015, 12.

116

Himiko as a local chieftain, possibly of the Kumaso, safely away from the official imperial patriline48.

In 1910, Shiratori’s arguments to that same effect vied with Naitō Torajirō’s49 (内藤虎次郎; 1866–

1934) defence of Himiko in Honshu50.

Discussion of Himiko was stymied, however, by contemporary politics. In the late 19th and early 20th centuries, historians walked a fine line when they treated with the official narrative of the imperial line. Straying too far from the approved history could be regarded as a challenge to the emperor’s divine mandate, and scholars in such a position faced loss of tenure or even incarceration51. Himiko was not only a challenge to the official imperial narrative, but also a tributary to Wei, both her and Wa cast in the Records of the Three Kingdoms as subservient to China52. It is not hard to see how, amid the atmosphere of Edo and Meiji nationalism, such a history would prove quite an unpopular concept. Chinese sources discussing Himiko were banned from school classrooms prior to 194553. The early “golden age” work of Takamure, with its insistence on Kinai

Theory, came as a challenge to the scholarly orthodox. Takamure viewed the refusal of scholars to accept Himiko within the Kinai as the work of academic misogyny: Himiko was rejected because male scholarship found the female-dominated origins of Japan unpalatable54. For Takamure, situating Himiko in the Kinai constituted an act of resistance against scholarly androcentrism.

48 Shiratori 1976, 166; Kidder 2007, 22–4; Naoki 2008, 1.

49 Commonly known as Naitō Konan (内藤湖南)

50 Shiratori 1976, 171; Naitō 1976, 208; Naoki 2008, 1.

51 Kidder 2007, 25–7.

52 Kidder 2007, 22; Ambros 2015, 12.

53 Miller 2018, 69.

54 Takamure 1966b, 74.

117

In the current day, Kinai Theory appears dominant, favoured by Kidder, Naoki, Maeda, and

Yoshimura, to name a few55. Naoki notes that Kyushu Theory was prevalent in the 90s56, and appears to be favoured by Sekiguchi (writing in 1987) and Piggott (1997), albeit with a nod to the ongoing uncertainty of the matter57.

Nevertheless, several strong arguments in favour of Kinai Theory have emerged within the last few decades. One is a matter of linguistics: although grouped together in the modern language, several phonemes were divided into two different pronunciations – “A-type” and “B-type” – in ancient Japanese, represented phonetically by different kanji58. Furthermore, although ancient

Chinese pronunciation is difficult to reconstruct, there is evidence that it in many ways was quite different than modern Mandarin. The character used for the “tai” of “Yamatai” in the records would not have been pronounced as such contemporarily, but rather was phonetically associated with the

“B-type” pronunciation of the phoneme to. The Kyushu “Yamato” – Yamato District in Chikugo

Province (modern Fukuoka Prefecture), and Yamato Village in Kikuchi District, Higo Province

(modern Kumamoto Prefecture) – were both spelled with the “A-Type” to; the to in Yamato Province, however, was “B-Type”59.

The other argument is cartographical: the Records’ directions send the reader straight into the sea when traced on a modern map, made with the benefit of satellite imaging, but the ancients did not have the technology for such an objective geographical sense. This discrepancy can be clearly seen in the Honil Gangni Yeokdae Gukdo Ji Do60, a Korean atlas produced in 1402 based on Chinese

55 Kidder 2007, xii; Maeda 2008, 46; Naoki 2008, 6; Yoshimura 2010, 10–7.

56 Naoki 2008, 6.

57 Sekiguchi 1987, 15–17; Piggott 1997, 15, 20–1.

58 Yoshimura 2010, 10–2.

59 Yoshimura 2010, 10–2.

60 混一疆理歷代國都之圖; “Map of Integrated Lands and Regions of Historical Countries and Capitals”

118 maps from the Yuan Dynasty (1271–1368). Here, the orientation of Japan relative to China appears completely different, with Kyushu far more northerly and Honshu tilting to the south. According to

Yoshimura, tracing the Records directions onto this map would place Yamatai quite confidently in the Kinai61.

The direct relationship between Himiko and the Yamato dynasty still remains unclear. The

Book of Liang is the only text to even mention the succession after Iyo/Toyo, and it remains extremely vague about the man who apparently followed her, or his connection to the “Five Kings of

Wa” (Liang Shu 54, Wozhuan). It is possible, given the instability that followed her death, that she belonged to a preceding dynasty, erased from the official Yamato histories in order to support their own claims to primeval divine mandate62. For the sake of this research, however, the key point is that the current evidence supports a tentative reading of Himiko as a monarch of the Yamato region, linking her to central traditions and practices. The gendering of her reign may be read in light of other images of Yamato rulership in order to examine her “typicality”; furthermore, if her sacral seclusion was truly as presented by the Chinese observers, it would tie in with other, later, Yamato religious practices (see Ch. 5.3).

Finally, let us examine the images of rulership surrounding Himiko to situate her typicality within third-century Yamato. Was Himiko – as the “golden age” narrative presents her – representative of a broader ancient tradition of female rulership63? Were she and Iyo/Toyo

“exceptional women,” afforded through their ritual performance special access to positions of authority more typically bestowed upon men64? Or, as Yoshie argues, was their gender unrelated to

61 Yoshimura 2010, 16–7.

62 Yoshimura 2010, 40, 52–3.

63 Sakima 1926, 65; Takamure 1966b, 77, 83; Hori 1968, 31, 191; Ōi and Miyagi 1974, 17; Matsumura 1999,

121–2; Maeda 2008, 2–4, 16.

64 Shiratori 1976, 178.

119 their monarchic status and performance, the role of sovereign attainable by men and women alike, the “special” characteristics of Himiko’s rule not a feminine model of sacral rulership but rather a common, gender-neutral practice read out of context through a foreign gaze65?

Himiko’s rule was preceded by a man; a man was also intended as her successor, though his failure to successfully wield this authority necessitated the enthronement of the young Iyo/Toyo

(Sanguozhi, Wei Shu 30, Worenzhuan). This suggests a preference for male rulership among the

Yamatai elite, although one which was not necessarily shared by the general populace66. The enthronement of a man after Iyo/Toyo appears to have been more successful, with no reports of further disturbance and an eventual transition to the Yamato dynasty’s “Five Kings of Wa” (Liang Shu

54, Wozhuan).

Although little record on Wa leadership survives outside of Himiko, the Book of the Later

Han does refer to a “King Suishō of Wa” (倭国王帥升) ruling the region in 107 CE67 (Hou Hanshu 85,

Wozhuan). Although this monarch’s gender is not specifically mentioned, the interest drawn by

Himiko as a female ruler suggests that Suishō was (or was assumed to be) male. Himikoko (卑弥弓

呼), a contemporary of Himiko’s ruling over the neighbouring and hostile kingdom of Kunu68, is explicitly referred to as a “male king” (男王, nán wáng)69 (Sanguozhi, Book of Wei 30, Worenzhuan).

Little is known of Suishō, and Himikoko ruled a separate region from Yamatai, with potentially different rulership traditions. But these examples do not suggest a specific preference for a feminine

65 Yoshie 2013.

66 Takamure 1966b, 77, 83; Piggott 1994, 7.

67 “安帝永初元年，倭國王帥升等獻生口百六十人，願請見。” (“Ān Dì Yŏngchū yuánnián, Wōguó wáng

Shuàishēng dĕng xiàn shēngkŏu băi liùshí rén, yuàn qĭng jiàn.”)

68 狗奴國, Kunukoku/Kunu-no-Kuni (Japanese) or Gŏunúguó (Chinese). The precise location of this kingdom is unknown.

69 “狗奴國男王卑彌弓呼” (“Gŏunúguó nán wáng Bēimígōnghū”).

120 model of rulership, as the “golden age” interpretation of Himiko would indicate; rather, they display a possible preference for a masculine sovereign, but not so strict a preference that the role was inaccessible to a woman. This is in keeping with our findings on sacral gendering (see Ch. 3.2.2), which suggest a broadly androcentric society within which women may still attain considerable status.

4.1.3. Himiko and Tensions of Power

The uncertain significance of gender of Himiko’s rule is not the only complication in her relationship to power. Although Himiko is presented within the “golden age” narrative as a figure of

“female power,” the broad label of “power” holds within it a great variety of aspects and power relationships. Indeed, the characteristics of Himiko’s rule serve as a good illustration of the tensions that can exist within the notion of “power.”

Returning to the types of power outlined previously (see Ch. 2.2.5) – temporal authority, sacral power, personal autonomy, and a nebulous “absence of patriarchy” – Himiko, as Queen of Wa, is clearly invested with temporal authority. Although her use of that authority was mediated through her younger brother – leading to speculation that, in point of fact, the authority of rulership was actually shared between them70 – it is ultimately Himiko, not her younger brother, who represents Wa on the international stage71, receiving the title “Friend of Wei” (Sanguozhi, Wei Shu

30, Worenzhuan); from the perspective of the Chinese observers, at least, it was in Himiko’s hands, not her unnamed brother’s, that monarchic authority was concentrated72. Aside from her gender and her sacral performance, within the Chinese sources she appears endowed with the full suite of

70 Takamure 1966b, 78; Matsumura 1999, 123; Kidder 2007, 133; Naoki 2008, 207; Yoshimura 2010, 26;

Ambros 2015, 14.

71 Maeda 2008, 20.

72 Yoshimura 2010, 26; Yoshie 2013, 11.

121 monarchic authority73. Indeed, Yoshie argues that Himiko’s image in scholarship has been plagued by normative assumptions of female political uninvolvement, pigeonholing her into a purely sacral function and elevating an otherwise unremarkable male assistant to the position of “true” monarchic authority74.

Himiko’s sacral performance further endowed her with sacral power, possibly honed by her life of strict ritual seclusion. This sacral power is not wholly separate from her temporal authority; in the Chinese sources, Himiko’s magic is presented as the basis of her reign. The explanation to which the Chinese onlookers came – that Himiko used her abilities to ensorcell the populace into enthroning her (Sanguozhi, Wei Shu 30, Worenzhuan; Hou Han Shu 85, Wozhuan; Sui Shu 81,

Woguozhuan; Liang Shu 54, Wozhuan) – is unlikely, but may spring from the core understanding that her powers were foundational to her rulership, the potential benefit of a spiritually powerful oracle- queen qualifying her for the throne, her sacral power leveraged to provide herself with temporal authority. In Yoshie’s interpretation, the relationship is reversed, her sacral performance being typical of a paramount and unrelated to her gender75. In either case, the two axes of “power” are seen to intersect without being wholly identical.

Yoshie’s interpretation, furthermore, rejects the notion of Himiko’s sacral performance as a special female quality in the first place, attributing this reading instead to normative assumptions of irrational female rule against rational male rule, and therefore the affective nature of shamanism as a female quality76. Other instances in the records do suggest shamanic mediumship as a feminine role within Yamato tradition (see Ch. 3.2), but Yoshie’s point remains salient: the Chinese sources reveal very little about the nature of Himiko’s sacral performance or the gendering thereof, and so

73 Maeda 2008, 20.

74 Yoshie 2013, 10–11, 16.

75 Yoshie 2013, 8–11.

76 Yoshie 2013, 8.

122 we cannot conclusively say that her position of sacerdotal rulership was part of a specifically female tradition.

Within the Chinese sources, Himiko appears as a reclusive figure, her temporal authority having to be exerted through use of her brother as an intermediary, the same spiritual power which allowed her access to queenly authority at the same time complicating her free use of it. This life of seclusion also heavily restricted Himiko’s personal autonomy. In the Records of the Three Kingdoms, she lives cloistered within her heavily-guarded palace, rarely (if ever) leaving its confines, her human contact mostly limited to her brother and a small army of maids (Sanguozhi, Wei Shu 30,

Worenzhuan). Although the queen of a confederacy, she lives, as Kidder puts it, as a “virtual prisoner”77. Although this ensconcement at first might appear a contradiction to her considerable sacral power and monarchic authority, they may, in fact, be intimately connected, the ritual seclusion serving to enhance the very spiritual abilities which serve as a foundation for Himiko’s queenship. In other words, Himiko’s sacral power – and, thereby, her temporal authority – may have been leveraged through the loss of her personal autonomy, her power as a shaman-queen bought at the price of her individual freedom of movement and action. Himiko’s reign presents a complex interrelationship of various forms and axes of “power,” which “golden age” narrative’s neat dichotomy of female empowerment versus subjugation is insufficient to encompass.

Yoshie argues differently: Himiko’s “seclusion,” she suggests, was not a true confinement, but rather a standard practice of Yamato paramounts concealing themselves from envoys, misinterpreted by the Chinese observers, as seen with Suiko78 (see Ch. 6.2.2). However, other passages from the Nihon Shoki suggest a more direct meeting between Empress Jingū, the future

Emperor Ōjin79 (応神), and envoys from Silla80 (Nihon Shoki 9.Jingū Kōgō Sesshō 47.4), and later

77 Kidder 2007, 137.

78 Yoshie 2013, 11.

79 Reign traditionally dated 270 – 310 CE

123 between Emperor Kinmei and envoys from Baekje81 (Nihon Shoki 19.Kinmei 13.10), complicating this assertion.

Further complicating our understanding of Himiko as an emblem of female power is that she, ultimately, is but a single queen of third-century Yamato, not a microcosm of all ancient Japanese women. She exists within her own specific context, as a woman of the Yamato elite raised to queenship following a period of domestic turmoil. Her experiences and relationships to power are not inherently representative of all women across all classes and regions of the Wa confederation.

The Records of the Three Kingdoms provides some further information on the women and marital customs of Wa, although it is not made clear from which region(s) this information is derived:

There is no difference in seating for meetings, regardless of age or gender…By custom, the

great men of the nation each have four or five wives, and even the lower households have

two or three. Their women are not lewd or jealous…If one breaks the law, if the charge is

light he will lose his wives and children, and if it is serious, the rest of the household as well82.

(Sanguozhi, Book of Wei 30, Worenzhuan).

Here Wa is described as a polygynous society, whereas women’s sexual monogamy is implied through the statement that they are not “lewd”83. As Sakima reminds us, polygyny is not an intrinsically patriarchal social structure: wives cohabiting with and being dependent upon a single husband might place them in a disadvantageous position, but wives remaining in their natal families,

80 新羅, Shiragi to the Japanese; a kingdom covering the southern and central parts of the Korean Peninsula. In both Kojiki and Nihon Shoki accounts, it is the primary target of Jingū’s conquests.

81 百濟, Kudara to the Japanese; a kingdom covering the south-western Korean Peninsula.

82 “其會同坐起，父子男女無別…其俗，國大人皆四五婦，下戶或二三婦。婦人不淫，不妒忌。…其犯法，

輕者沒其妻子，重者滅其門戶。” (“Qí huìtóng zuò qĭ, fùzĭ nánnǚ wú bié…Qí sú, guó dàrén jiē sì wŭ fù, xiàhu huò èr sān fù. Fùrén bù yín, bù dùjì…Qí fànfă, qīngzhĕ mò qí qīzĭ, zhòngzhĕ miè qí ménhù.”)

83 Sakima 1926, 66.

124 with husbands as conjugal visitors, may yet retain a high position within the family84. The Book of the

Later Han’s note that mothers, fathers, and siblings reside separately would seem to indicate the latter marital arrangement85 (Hou Hanshu 85, Wozhuan). The Book of the Later Han simply says that

“all other men” – and not households – may have two or three wives86 (Hou Hanshu 85, Wozhuan), suggesting that the Records of the Three Kingdoms’ grouping of wives into a “household” is more representative of the virilocal norms of the Chinese observers than actual Wa marital practices. Such arrangements align with Yoshie’s assertion that marital unions of this time were loose and fluid, easily dissolvable and usually not involving cohabitation, the consolidated, patriarchal family unit having not yet emerged87. The Book of Sui adds that wives do not take new names upon marriage and implies that matches form based on mutual interest rather than political arrangement88 (Sui Shu

81, Woguozhuan), further suggesting that women held a degree of agency within their marriages.

The vague reference to a man “losing” wives and children as penalty for a crime does, however, hint at limits to women’s agency. Ambros interprets this “loss” to mean the enslavement or even execution of wives and children as a penalty for the crimes of the father, an arrangement which, counter to Yoshie, speaks to a thoroughly patriarchal family structure, with wives and children treated as extensions of a central male pater familias89. This structure is, however, hard to reconcile with the reference to separate habitation within the Book of the Later Han. With little textual evidence beyond these hints from outsiders’ perspectives, it is difficult to recreate with

84 Sakima 1926, 108.

85 “父母兄弟異處，唯會同男女無别。” (“Fùmŭ xiōngdì yì chŭ, wéi huìtóng nánnǚ wú bié.”)

86 “其餘或两或三。” (“Qíyú huò liǎng huò sān.”)

87 Yoshie 1993, 442–5.

88 “女多男少，婚嫁不取同姓，男女相悅者卽爲婚。” (“Nǚ duō nán shăo, hūnjià bù qŭ tóng xìng, nánnǚ xiāng yuè zhĕ jí wéi hūn.”)

89 Ambros 2015, 19–20.

125 certainty the exact structures of ancient Wa’s dynamics of family, gender, and power. But what these contrasts and complications may suggest is a space and time that shows marital and family structures quite different from the clear, virilocal patriarchy of the mediaeval era, a Wa which affords women greater agency than later period but at the same time does not neatly fit into the

“golden age” narrative image of uncomplicated pre-patriarchal empowerment.

4.2. Jingū, Empress-Regent

So far, we have examined the tensions of power and gender present in the reign of Himiko,

But Himiko and her kinswoman are not the only images of female rulership appearing in Yayoi Japan.

Our examination of tensions of power within early examples of female rule takes us now from

Chinese records to the Yamato official narrative, with the figure of Empress Jingū.

4.2.1. An Introduction to Jingū

Empress Jingū, also known as Okinaga-Tarashi-Hime (息長帯比売/気長足姫), appears within the Kojiki and Nihon Shoki as the wife of Emperor Chūai and mother of Emperor Ōjin. Her tenure – officially, a regency following Chūai’s early death – is traditionally (if improbably) dated from 201 to 269 CE, covering the years attributed by Chinese sources to Himiko’s reign. Like Himiko,

Jingū appears as a shamanic queen, assisting governance with her oracular performances. However, her presentation in the records is quite different: she is also a war leader, a conquering heroine spearheading an invasion of the Korean Peninsula.

As a dynamic, powerful female ruler, Jingū stands out in the early Kojiki and Nihon Shoki chapters, which otherwise portray the Yamato dynasty as relentlessly male until the accession of

Suiko in 592. It is perhaps unsurprising that she should draw such attention within the “golden age” narrative. In one of the earlier “golden age” works, Tatsumi presents Jingū alongside the goddess

126

Ame-no-Uzume as an emblem of the power and respect afforded to women in ancient Japan90. To

Takamure, and later, E. Patricia Tsurumi, Jingū represents the former grandeur of the imperial consort role91; to Hori, Ōi, and Miyagi, she is but one of many theocratic shaman-queens of ancient

Japan, a holdout who survived in legend92, a theory echoed later by Maeda93. Although Naoki reads

Jingū herself as a mythic construct, he argues that she was grounded in existing custom, her format of power – sacral co-rulership with her husband, and later, the minister Takeshi-Uchi-no-Sukune – typical of the time94.

Unlike the vague allusions to Himiko’s magico-religious performance contained in the

Chinese records, the Kojiki and Nihon Shoki are far more explicit and detailed concerning Jingū’s own shamanic performance, with her assuming the role of oracle on at least two key occasions. The Kojiki account of the first oracle proceeds as thus:

The Empress Okinaga-Tarashi-Hime-no-Mikoto was possessed by a god at that time. When

the Emperor was dwelling in Kashii Palace in Tsukushi95, and was preparing to attack the

land of the Kumaso, he played upon a koto, with Takeshi-Uchi-no-Sukune-no-Ōmi acting as

interrogator, questioning the deity. Thus the Empress was possessed by a god, and

instructed them: “There is a land to the west, containing many treasures dazzling to the eye;

90 Tatsumi 1887, 32.

91 Takamure 1966b, 114; Tsurumi 1982, 72.

92 Hori 1968, 195; Ōi and Miyagi 1974, 17.

93 Maeda 2008, 4.

94 Naoki 2008, 208.

95 Modern Fukuoka Prefecture.

127

treasures of all sorts, including gold and silver. I bestow that country upon thee.”96 (Kojiki

2.Chūaiki)

During the ritual, Chūai accuses the god of lying about this mysterious, foreign land, and is struck dead for his temerity (Kojiki 2.Chūaiki; Nihon Shoki 8.Chūai 8.9.Kibō). Jingū assumes the mantle of leadership, acting as regent for her unborn child, and quickly holds a follow-up divinatory session to learn more about this oracle (Kojiki 2.Chūaiki; Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai

9.3.Jinshin).

Jingū in the Kojiki takes on the role of a miko, providing guidance to the Emperor through divine possession. The minister Takeshi-Uchi-no-Sukune takes on the male priestly role of saniwa interrogator, while Chūai himself plays music to assist Jingū in inducing a trance state. Jingū’s turn as medium is in the Kojiki presented as intentional and expected, almost an ordinary occurrence, her oracular abilities integrated into the mechanisms of government. Unlike Himiko, however, she is not treated as the monarch herself, but rather as his helpmate, only stepping into the shoes of kingship when her husband’s untimely death demands it.

96 “其太后息長帶日賣命者。當時歸神。故天皇㘴筑紫之訶志比宮。將擊熊曽國之時。天皇控御琴而。

建內宿祢大臣居於沙。請神之命。於是太后歸神言教覺詔者。西方有國。金銀爲本。目之炎耀。種々

珎寶。多在其國。吾今歸賜其國。” (“Sono ōkisaki Okinaga-Tarashi-Hime-no-Mikoto wa. Sono kami kamigakari shitamaiki. Kare sumera-mikoto Tsukushi no Kashii-no-Miya ni mashimashite. Kumaso-no-Kuni wo utan to shitamaishi toki ni. Sumera-mikoto mikoto hikitamaite. Takeshi-Uchi-no-Sukune-no-Ōmachigimi saniwa ni ite. Kami-no-mikoto wo koimatsuriki. Koko ni ōkisaki kamigakari shitamaite koto oshiesatoshi noritamaeru wa. Nishi no kata ni kuni ari. Kogane shirogane wo hajime toshi. Me no kagayaku. Kusagusa no uzu no takara.

Sawa ni sono kuni ni ari. Ware ima sono kuni wo yosetamawan.”)

128

In the Nihon Shoki, her first oracle begins somewhat differently, with none of these ritual preparations:

The Emperor ordered that the retainers were to convene to suppress the Kumaso. At this

time the Empress was possessed by a god and instructed them thus.97 (Nihon Shoki 8.Chūai

8.9.Kibō)

Jingū’s possession appears within the Nihon Shoki as a spontaneous, unexpected occurrence; it is only her second divination that is ritualised, with her undergoing appropriate abstentions and with male saniwa and koto-player appointed (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai 9.3.Jinshin).

The casting of her first divination as the result of spontaneous possession, rather than standard ritual performance, may serve to emphasise Jingū as an extraordinary figure, divinely selected for the role.

Chūai having angered the gods with his disbelief, the gift of conquest of the fabled land of

Korea is granted instead to another:

“If thou, oh king, will not believe in what I say, then thou shalt never gain hold of this land.

Only the child newly formed in the Empress’s belly shall obtain it.”98 (Nihon Shoki 8.Chūai

8.9.Kibō)

97 “詔群臣以議討熊襲。時有神託皇后而誨曰。” (“Maetsu-no-kimitachi ni mikotonori shite Kumaso utsu koto wo hakarashime tamau. Toki ni kami mashite kisai-no-miya ni kakarite oshiematsurite notamawaku.”)

98 “其汝王之。如此言而遂不信者。汝不得其國。唯今皇后始之有胎。其子有獲焉。” (“Sono imashi mikoto. Kaku notamaite tsui ni uketamawazuba. Imashi sono kuni wo ezu. Tada ima ōkisaki no hajimete harami maseri. Sono miko etamau koto ari.”)

129

A similar proclamation is made within the Kojiki during the second divinatory session:

After this Takeshi-Uchi-no-Sukune once more stood as interrogator and questioned the deity.

The instruction proceeded much like the previous day: “Altogether this land is a land to be

ruled by the child in thy womb.” Then Takeshi-Uchi-no-Sukune said: “I am frightened, oh

great deity. What kind of child is it that resideth within the divine womb?” The deity

answered: “It is a boy-child.”99 (Kojiki 2.Chūaiki)

In both texts, the intended beneficiary of the conquest of Korea – dominion bestowed by several gods, including the imperial ancestress Amaterasu herself (Kojiki 2.Chūaiki; Nihon Shoki 9.Jingū Kōgō

Sesshō Zenki Chūai 9.3.Jinshin) – was Chūai but, brought low by his hubris, it is now passed on through the patriline to the unborn Ōjin. Jingū herself appears as the intermediary, rather than the recipient, of this divine blessing.

Jingū’s role as intermediary of the conquest of Silla is not limited to her oracular performance, either. With the force of divine will at her back, the empress-regent crosses the sea and leads her armies to the Korean Peninsula, conquering the lands there to serve as tributary states.

The gods assist her at every turn, driving her ships forward and protecting her person (Kojiki

2.Chūaiki; Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai 9.9.Kibō; Nihon Shoki 9.Jingū Kōgō Sesshō

Zenki Chūai 9.10.Shinchū); in the Nihon Shoki account, she also dispatches soldiers to achieve the victory her husband sought over the Kumaso, further enhancing the Yamato dynasty’s territorial holdings (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai 9.3.Jinshin). After this, the heavily-pregnant

99 “亦建內宿祢居於沙。請神之命。於是教覺之狀。具如先日。凢此國者。㘴汝命御腹之御子所知國

者也。尓建內宿祢白恐我大神。㘴其神腹之御子何子歟。答詔男子也。” (“Mata Takeshi-Uchi-no-Sukune saniwa ni ite. Kami-no-mikoto wo koimatsuriki. Koko ni oshiesatoshitamau sama. Tsubusa ni saki no hi no gotoku nite. Ōkata kono kuni wa. Na-ga-mikoto no mihara ni masu miko no shirasan kuni nari. Koko ni Takeshi-

Uchi-no-Sukune kashikoshi a ga ōkami. Sono kami no mihara ni masu miko wa nani miko zo to mōseba. Hiko miko zo to kotaetamaiki.”)

130

Jingū returns to Japan and gives birth, ensuring that the Emperor is not born on foreign soil; through clever strategy, she further quells an attempted usurpation by Ōjin’s older half-brothers, defending his imperial birthright from the children of a lower-ranking consort (Kojiki 2.Chūaiki; Nihon Shoki 9).

She continues to act as Ōjin’s regent until her death, ostensibly in 269 CE.

4.2.2. Situating Jingū in History

As we have seen, of course, situating this 69-year regency – its length already improbable – within a chronological timeline is rendered problematic by the overlap of her regency in the Nihon

Shoki100 and the dates attributed to Himiko’s queenship in the Records of the Three Kingdoms. The compilers of the Nihon Shoki were clearly aware of the Records of the Three Kingdoms’ account, incorporating all dated entries into Jingū’s reign as annotations (Nihon Shoki 9.Jingū Kōgō Sesshō

39/40/43), effectively equating the two queens.

There is, however, nothing else within Jingū’s chapter to link her with Himiko. The three annotations appear in years otherwise completely blank of recorded events. The quoted portions refer to emissaries sent to Wei by the “Queen of Wa,” but there is no mention of them in the rest of the text, nor any other mention of Jingū treating with China (Nihon Shoki 9). Nor are the other dated events in the Chinese sources reflected in Nihon Shoki: in the Book of the Later Han, “King Suishō” sends tribute to China in 107 CE, a year completely devoid of entry in the Nihon Shoki (Nihon Shoki

7); the unrest preceding Himiko’s reign is similarly absent, the years attributed to it in the Book of

Liang appearing in the Nihon Shoki as largely empty and uneventful (Nihon Shoki 7). The events following Himiko’s death are also not incorporated into the Nihon Shoki, which has Jingū still alive, well, and politically active for well past the Book of Liang’s dating of Himiko’s passing (Nihon Shoki 9).

100 The Kojiki typically only gives dates for imperial deaths, and does not reference the Records of the Three

Kingdoms at all (Kojiki 2.Chūaiki)

131

It is possible that the compilers of the Nihon Shoki did not have access to the other Chinese sources, but the problems in equating Himiko with Jingū do not stop at dating discrepancies.

The characterisations of the two queens, within the records, show very little in common between them besides their sacral performance. A considerable list of discrepancies have been noted between the two women in 21st century scholarship: Kidder and Ambros contrast Himiko’s status as reclusive hermit-queen, cloistered in her palace and barely seen by her own people, with

Jingū’s role as bold general, personally leading an invasion force across the sea to Korea, moving openly throughout her realm and hosting banquets101. Yoshimura and Ambros point also to Himiko’s perpetual maidenhood: childless and explicitly unmarried, her contact with men appears restricted to her kin. Jingū, however, is both wife and mother, pregnant during her conquests and divinations, directly interacting with male soldiers and retainers102. Kidder and Yoshimura observe also a difference in international connections: Himiko pays tribute to Wei China, with no known link to

Korea; Jingū secures tribute from the Korean Peninsula, with no link to China outside of extracts from the Records of the Three Kingdoms103. Finally, Kidder also points to the matter of Iyo/Toyo: whereas Himiko’s (accepted) successor is female, Jingū’s successor is her son Ōjin, from whom exclusive male succession continues until Empress Suiko in 592104. As shaman-queens go, the two could hardly be more different.

These discrepancies suggest that the Records of the Three Kingdoms’ account of Himiko and the Nihon Shoki’s account of Jingū are not the same story of a sacral queen, told from different perspectives; the variation is simply too great. Rather, it appears that the compilers of the Nihon

Shoki, having access to the Records of the Three Kingdoms, selectively incorporated passages from

101 Kidder 2007, 21; Ambros 2015, 16.

102 Yoshimura 2010, 53; Ambros 2015, 16.

103 Kidder 2007, 22; Yoshimura 2010, 53.

104 Kidder 2007, 22–3.

132

Himiko’s portion of the Records into Jingū’s chapter of their own work, but only insofar as they could be harmonised with the pre-existing story of Jingū’s reign. Elements that conflicted with this narrative – such as Himiko’s seclusion, the chaos preceding her enthronement, or her female successor Iyo/Toyo – were apparently disregarded entirely. As Yoshimura points out, the Nihon

Shoki’s equation of Himiko with Jingū appears as a tenuous assumption based on little more than their gender and time period105. Thus, despite their identification by the compilers, we engage with

Himiko and Jingū as separate figures with separate textual sources, rather than different accounts of the same ancient queenship.

If Jingū was not Himiko, then where does she fit into history? Takamure, in the early “golden age” narrative, interprets her as an extant historical figure, albeit one who has undergone considerable mythologisation over the centuries106; this notion is further echoed by Tsurumi107. Hori, in his work on Japanese religion, also interprets her as a historical figure, presenting her conquests as having a concrete impact on the development of ancient Japanese society108. Kidder uses Jingū’s conquests to propose a martial aspect to ancient miko109, and Maeda situates her within a line of shaman-queens, although as a different person to the Himiko of the Chinese sources110. Through reading Jingū as a historical figure, her role as a medium-queen is presented alongside Himiko’s as representatives of an ancient, half-lost tradition of female shamanic rule.

As we have seen, however, it is not so simple to situate Jingū within the already-murky history of the Yamato dynasty. Even her personal name, Okinaga-Tarashi-Hime, is in question,

105 Kidder 2007, 21–3; Yoshimura 2010, 53; Ambros 2015, 16.

106 Takamure 1966b, 114.

107 Tsurumi 1982, 72.

108 Hori 1968, 195.

109 Kidder 2007, 131.

110 Maeda 2008, 4.

133

“Tarashi” being identified by Yoshimura as an anachronistic seventh-century flourish111. The notion of Jingū being a purely mythic construct has gained increasing favour within scholarship of the last few decades, theorised by Matsumura as early as 1999112. Faure refers to Jingū’s story in the Kojiki and Nihon Shoki as “obviously the product of a much later period” 113; similarly, Yoshimura identifies her given name as one of several anachronistic elements within the account114. Naoki’s religious histories similarly treat with Jingū as a mythic figure, although argue that her depiction was nevertheless grounded in pre-existing traditions of rulership115.

Matsumura puts forth the argument that Jingū in the records is a mythologisation based on the appearance of Himiko in Chinese texts. The compilers of the Kojiki and Shoki could hardly ignore the presence of Himiko in Chinese histories, but at the same time she represented a clear contradiction with a pre-existing court chronology, which had backdated the Yamato imperial clan’s patrilineal reign into the Age of the Gods116. The figure of Jingū, Matsumura suggests, was meant to harmonise Himiko with the Yamato dynasty, her role as regent explaining her access to monarchic authority in a way that still maintained the official patriline117. This interpretation, however, does not account for the considerable discrepancies between the two shaman-queens as outlined above. If

Jingū’s figure was based directly on Himiko, a much greater degree of overlap between them would be expected. The incidental and ill-fitting identification within the Nihon Shoki instead suggests

Jingū’s prior existence within Yamato myth-history.

111 Yoshimura 2010, 71.

112 Matsumura 1999, 126.

113 Faure 2003, 293–4.

114 Yoshimura 2010, 71.

115 Naoki 2008, 208; 2009, 223.

116 Matsumura 1999, 126.

117 Matsumura 1999, 126.

134

So then, what does it mean for Jingū to be a figure of myth-history? How does this impact our reading of her tale? Let us recall again Beard’s theories on the reading of myth-history (see Ch.

2.2.2): our examination of myth-history should not focus on the literalist task of extracting particles of historical fact from the mire of legend, but rather on understanding the meaning her story held to those who transmitted it118. Jingū within the chronicles is a fascinating figure, not as a crystallised image of primeval sacral queenship, but rather as a figure conveying the beliefs, norms, and self- identity of the Yamato court. The question we seek to answer now is not “what about Jingū’s story is authentic?” but rather “what can Jingū’s story tell us about gender, power, and spiritual performance in the eyes of the Yamato compilers?”

4.2.3. Power and Gender in the Jingū Myth

Two key episodes from Jingū’s myth-history are of particular interest here: firstly, her oracle and subsequent invasion of Silla, which serves to establish the Yamato dynasty’s dominion over

Korea119; and secondly, her defeat of the rebellious princes Oshikuma (忍熊) and Kagosaka (香坂/麛

坂), protecting her son’s patrimony over a lesser consort’s line (Kojiki 2.Chūaiki; Nihon Shoki 8–9).

As for the first point, there is little in Korean history to prove the presence of Jingū herself.

However, evidence does exist of Japanese assaults on the peninsula conducted at a later date. The controversial Gwanggaeto Stele, erected in 414 CE in Gungnae-seong120, the then-capital of the

Korean kingdom of Goguryeo121, appears to list a Japanese invasion of the southern Korean kingdoms of Baekje and Silla in 391 CE, although the fragmentary nature of the surviving text makes

118 Beard 2015, 71–3, 86.

119 Yoshimura 2010, 71–2; Yoshie 2013, 17; Kimura 2017b, 2–3.

120 Modern Ji’an City in Jilin Province, China

121 高句麗, Kōkuri to the Japanese; also called Goryeo (高麗/Kōrai); a large kingdom covering the northern and central Korean Peninsula and southern to central Manchuria.

135 conclusive interpretation difficult122. The 12th century Korean chronicle Samguk-Sagi123 says that in

397 CE, King Asin (阿莘王) of Baekje was friendly with Japan, and places Japanese invasions of Silla in 400, 401, and 405 CE124. Not only are these invasions unrecorded in the Nihon Shoki, but no mention of Silla whatsoever is made for the years between 365 and 414 CE (Nihon Shoki 11–13).

Jingū’s conquest appears to represent a mythologised conglomeration of several different invasions and tributary negotiations throughout early Yamato dynasty history.

Through this mythical conquest, pre-existing tributary relationships are explained and cast as divine mandate, and a series of invasions are rewritten into a single, glorious charge resulting in swift and easy victory for the Japanese forces, allowing for the self-presentation of the later Yamato kings as a line of heroic, divinely-sanctioned conquerors. Furthermore, the story asserts the Yamato monarchs’ right to military control over the kingdoms of Baekje and Silla, granted through the conquests of their ancestors and the King of Silla’s vow. The oracular scenes further present this dominion as a matter of divine right Korea and its treasures are bestowed upon Ōjin, through Jingū, by the very gods themselves125 (Kojiki 2.Chūaiki; Nihon Shoki 8.Chūai 8.9.Kibō). The invasion force led by Jingū also receives very palpable divine support, impressing upon the King of Silla the divine might behind the Japanese army and leading to a quick victory (Kojiki 2.Chūaiki; Nihon Shoki 9.Jingū Kōgō

Sesshō Zenki Chūai 9.10.Shinchū). The legend can thus be called nationalistic in character, asserting the divine mandate of the Yamato kings over the kingdoms of Korea, with the shaman-empress Jingū as the means for transmitting and enacting the gods’ will.

122 Yoshimura 2010, 68–9.

123 三國史記, “History of the Three Kingdoms”. This text was compiled from earlier records which are no longer extant.

124 Yoshimura 2010, 70.

125 Yoshie 2013, 17; Kimura 2017b, 2–3.

136

The attribution of these conquests to a female leader is, indeed, striking against the otherwise thoroughly patrilineal myth-history of Yamato. But let us recall the passages quoted earlier: “Altogether this land is a land to be ruled by the child in thy womb.” (Kojiki 2.Chūaiki)126. In other words, dominion over Korea is attributed not to Jingū herself, but rather to the unborn

Emperor Ōjin in her womb, his mother Jingū facilitating this dominion as an intermediary. A later passage in the Nihon Shoki referencing Jingū’s conquests reiterates this, stating that while Jingū

“designated the first imperial territories”127 in Korea, the gods “first granted the foreign Lands of

Gold and Silver…unto Emperor Homuda [Ōjin] in the womb.”128 (Nihon Shoki 17.Keitai 6.12). Ōjin was born into the world ruling not only Japan, but also counting a large swathe of the Korean

Peninsula as his vassal-states. In this, he fits the image that the later Yamato monarchs desired for themselves and their ancestors: that of the tenka no ō129, the “king of the realm/world,” drawing from the Chinese concept of the emperor ruling the tianxia130.

Jingū’s role within the story is one of facilitation of the patriline and the increase and solidification of its holdings. The second key episode of Jingū’s myth – in which she defeats through cunning strategy the attempted uprising of princes Oshikuma and Kagosaka, thereby guaranteeing the patrimony of her son, the rightful heir – only reinforces this (Kojiki 2.Chūaiki; Nihon Shoki 9).

Although Jingū wields both temporal authority (as regent) and considerable sacral power (as skilled

126 “凢此國者。㘴汝命御腹之御子所知國者也。” (“Ōkata kono kuni wa. Na-ga-mikoto no mihara ni masu miko no shirasan kuni nari.”)

127 “故大后氣長足姬尊。…每國初置官家。” (“Yue ni ōkisaki Okinaga-Tarashi-Hime-no-Mikoto. Kuni goto ni hajime ni miyake wo okite.”)

128 “夫住吉大神。初以海表金銀之國。…授記胎中譽田天皇。” (“Sore Suminoe-no-Ōkami. Hajime ni wata no hoka no kingin no kuni…Hara no uchi ni mashimasu Homuda-no-Sumera-Mikoto ni sazuke matsureri.”)

129 天下の王

130 天下, tiānxià; “under heaven”, a term which could alternatively refer to the realm or the world entire, particularly in reference to the sphere of imperial dominion.

137 medium), both are leveraged to the benefit of male Yamato kings: first Chūai, then Ōjin. This complicates her portrayal within the “golden age” as a representation of the power attributed to ancient Japanese women: not only is her ability to typify “ancient women” in question through her uncertain historicity, but her power exists within, and for the sake of, a broader patrilineal hegemony131.

Naoki contends that, even if Jingū herself is a mythical figure, she must have a basis in pre- existing images of ancient female sacral rulership; in particular, Naoki cites the role of the hime in the himehiko system (see Ch. 4.3). In other words, the appearance within legend of a shamanic female ruler reflects actual traditions of female rulership in Yamato’s distant past132. However, this does not satisfactorily explain Jingū’s uniqueness within the records: if she is a mythic legacy of an earlier system of female rulership, then why does the rest of the Yamato monarchy – both before and after her regency – appear as so relentlessly patrilineal? Why does only this image of female kingship survive?

Within the records themselves, Jingū-as-regent is instead presented as something of an exceptional woman, her military prowess directly contrasted in the Nihon Shoki with her gender:

“Although I am a woman, and moreover young, I must now for a time assume the form of a

man, and unleash strong and heroic strategy.”133 (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki

Chūai 9.4.Kōshin)

She garbed herself in the dress of a man and made war on Silla.134 (Nihon Shoki 9.Jingū Kōgō

Sesshō Zenki Chūai 9.12.Shingai)

131 Faure 2003, 293–4.

132 Naoki 2008, 208.

133 “吾婦女之加以不肖。然蹔假男貌。强起雄略。” (“Are taoyame ni shite mata onashi. Shikashi shibaraku masurao no sugata wo karite. Anagate ni ōshiki hakarigoto wo tatsu.”)

138

This passage recalls – and possibly derives from135 – the figure of Amaterasu in battle-garb (see Ch.

3.1.2): assumption of a martial form presented as a masculinising process, standing in contrast to her feminine gender (Kojiki 1; Nihon Shoki 1.6; Nihon Shoki 1.7.3). Kimura Saeko (木村朗子) interprets this cross-dressing as a ritualised performance undertaken by women136. This is reminiscent of Yoshie’s assessment of gendered labour in ritual, which suggests a certain amount of fluidity or vaguery instead of clear and defined borders about who may perform what function137. In other words, Kimura’s interpretation implies a society in which martial activity is a “male” task which is nevertheless accessible to a woman so long as she briefly assumes the mantle of masculinity; gender itself rendered fluid alongside gender roles. In this case, Jingū’s positioning as representing a specifically feminine mode of rulership does not encompass this nuance of gender fluidity; rather, such a custom would suggest that the gender of the sovereign was negotiable.

At the same time, within the Yamato myth-historic corpus, Amaterasu and Jingū appear as the only female figures to explicitly perform this action, making Jingū in the chronicles appear as something of an “exceptional woman,” who alone is provided access to this fluidity of gender roles.

Male Yamato sovereigns also appear to eschew ritual cross-dressing on their part138, with the closest example in the records being Prince Yamato-Takeru’s use of cross-dressing in an infiltration plan

134 “則皇后爲男束裝。征新羅。” (“Sunawachi ōkisaki o yosoi shite. Shiragi wo uchitamau.”)

135 Kimura 2017b, 3.

136 Kimura 2017b, 3.

137 Yoshie 1993, 459.

138 The Tsūshō (通証) commentary on the Nihon Shoki, produced in 1762 by Tanikawa Kotosuga (谷川士清;

1709–1776), does interpret Jinmu’s retainer Michi-no-Omi (道臣) as a man who situationally assumes femininity – including the feminine name or title Itsuhime (厳媛) – in order to perform a typically female sacral role (Aston 1896, 137). Within the text, however, Michi-no-Omi is never explicitly gendered outside of this title, and so this presumption of masculinity is not conclusive (Nihon Shoki 3).

139

(Kojiki 2.Keikōki; Nihon Shoki 7.Keikō 27.12). These suggest that the Yamato imperal clan configured certain monarchic duties (such as warfare) as specifically masculine, but gendered roles could be ritually negotiated, ultimately providing little practical obstacle for a female sovereign. Rather than exclusively masculine or feminine models of kingship, this implies that the role of Yamato sovereign was normatively a man but nevertheless fulfillable by a woman, the boundaries of gender roles less defined than in later eras.

The Kojiki and Nihon Shoki are insistent upon the terminology of “regency” (摂政, sesshō) for

Jingū’s reign. In the text, she is referred to either with her marital title of “Empress-Consort” (皇后, kōgō) – even into widowhood – or is referred to as the “Empress Dowager” (太后, taikō/皇太后, kōtaigō), as appropriate for the mother of a current monarch; she never takes the tennō139 title as later Empresses Regnant did (Kojiki 2.Chūaiki; Nihon Shoki 9). In the Hitachi Fudoki, the regional chronicle of Hitachi Province (modern Ibaraki Prefecture) produced shortly after the Kojiki, she is indeed named as tennō (Hitachi Fudoki, Ibaraki-gun 1), as well as in the 12th century Fusō Ryakuki140, a chronology of emperors produced by an Enryakuji monk (Fusō Ryakuki 1), but never in central court records. The Nihon Shoki gives her a section of her own, like an emperor, but is careful to mark it as a regency and never address her as tennō; the Kojiki does not even do this, merely appending her actions onto Chūai’s paragraph. In both central texts, she appears clearly delineated from the official imperial line; instead, her role as facilitator of the patriline, preserving the throne for the unborn/infant Ōjin, is emphasised, her titles reflecting her relationships to Chūai and Ōjin rather than her own seat of power. The Kojiki appending her actions to her husband’s reign only further communicates her position as intermediary of patrilineal authority.

139 天皇, “Heavenly Sovereign”; the title of a reigning emperor regardless of gender.

140 扶桑略記; “Abridged Chronicles of Fusang [Japan]”

140

The language – or, indeed, practice – of regency is not used elsewhere within the ancient portion of the Nihon Shoki, further calling the historicity of Jingū’s records into question. Her regency lasted until her death: an improbable 69 years, according to the official chronology, and yet no contradiction is noted in a “regent” holding the throne until the actual monarch is nearing 70. Nor is

Ōjin officially a “monarch”: despite the Nihon Shoki’s insistence on avoiding “official” imperial title for Jingū, Ōjin himself is explicitly granted the rank of “Crown Prince” (Nihon Shoki 9.Jingū Kōgō

Sesshō 3.1.Boshi)141, positioning him as Jingū’s heir rather than a boy-king for whom she is merely a temporary guardian. In this sense, she as a figure resembles the later Empress Suiko, an interim empress enthroned to solidify an otherwise precarious male successor’s claim – Ōjin being little more than a foetus at his father’s death – but existing before any custom of abdication was established allowing her to cede the throne to him, hence the need for her to reign until death142.

The portrayal of Jingū’s rulership could also have been influenced by the Han Chinese practice of dowager regencies, in which empress dowagers acted as regents until their sons were old enough to assume the responsibilities of emperor for themselves, a process which has also been argued as an inspiration behind Suiko’s enthronement143.

Furthermore, although mythically attributed to the third century, the actual age of the Jingū legend itself is uncertain; it cannot simply be assumed as an accurate reflection on third-century

Japanese society and gender norms144. Recalling the myth-history reading methods of Beard, absent any clear historical proof to the historicity of Jingū, her story is instead best understood in terms of what it meant to the eighth-century Yamato courtiers telling the tale145. To them, she appears as a

141 “立譽田別皇子。爲皇太子。” (“Homuda-Wake-no-Miko wo tachite. Hitsugi-no-miko to nasu.”)

142 Yoshikawa 2011, 33.

143 Piggott 1994, 19.

144 Faure 2003, 293–4.

145 Beard 2015, 71–3, 86.

141 facilitator of the patriline, a loyal wife and mother whose considerable talents – in mediumship and rulership, war and strategy – are deployed wholly in service to the patriline of the Yamato dynasty.

Rather than a vestige of an older, broader tradition of female power and rulership, she appears as an exceptional woman, fair and virtuous in war and victory (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki

Chūai 9.9.Kibō; Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai 9.10.Shinchū), her role as a powerful female figure and mother of emperors bestowing dominion upon her son’s line reminiscent of the role of Amaterasu as legitimating presence in Yamato myth.

At the same time, Jingū’s ability to negotiate through gendered occupations suggests a certain fluidity to the boundaries of gender in ancient Japan; the debate on her titulary possibly representing a tension among the eighth-century central court concerning how to interpret these elements in the face of changing court culture and models of kingship. The “golden age” narrative’s reading of her simply in terms of her femininity (sans androgyny) and sacral and temporal power

(sans service to the patriline) fails to engage with the complexities introduced to both these attributes by her context within the myth-historical canon of the Yamato dynasty: the negotiation of this gender within a patrilineal hegemony, and the deployment of this power for the benefit of male rulership.

142

4.3. The Himehiko System

Our third case study of female rulership refers not to a single example, but rather to a collection of examples, whose common feature is (apparent or assumed) male-female co-rulership, and which are interpreted as part of a broader practice of contrapuntal rule: the “himehiko system”146.

4.3.1. An Introduction to the “Himehiko System” within the “Golden Age” Narrative

The “himehiko system” first appeared within the “golden age” narrative to indicate a society built on the contrapuntal performance of male (hiko) and female (hime) rulers maintaining separate, gendered spheres of control. The term “himehiko system” was first coined by Takamure, although similar rulership structures are also depicted within the works of Sakima and Yanagita147. Previously we have discussed how, within the early “golden age” narrative, religious performance was seen as a female-dominated sphere (see Ch. 3.2); accordingly, the role of the hime was read as one of sacral rulership, performing oracles and leveraging magico-religious power; the figures of Himiko and Jingū portrayed as prominent examples of the hime-type. The hiko, conversely, was assumed to handle the mundane governance of the clan, territory, or kingdom, translating the divine pronouncements of the hime into political action; a role accordingly attributed to Himiko’s brother, or to Chūai before his untimely death. Himehiko is particularly associated with brother-sister partnerships, the co-rulers bound by natal kinship rather than marital ties148, presumably due to the more fluid nature of

146 Sakima 1926, 51; Takamure 1966a, 69; 1966b, 96–117; Sekiguchi 1987, 16; Okano 1993, 28; Kuratsuka 1996,

28; Matsumura 1999, 128; Naoki 2008, 206.

147 Sakima 1926, 51; Takamure 1966a, 69; 1966b, 96–117; Yanagita 1984, 308.

148 Sakima 1926, 51; Takamure 1966a, 69; 1966b, 96–117; Yanagita 1984, 308; Okano 1993, 28; Kuratsuka 1996,

28; Matsumura 1999, 128.

143 marriages in pre-ritsuryō Japan, where conjugal unions were often temporary arrangements dissolved with relative ease149.

The notion of the “himehiko system” represents one of the more enduring threads of the

“golden age” narrative, appearing even within scholarship which challenges the broader conceit of the “golden age” as a whole150. The forms in which himehiko appears in modern scholarship are more variable, however, than their earlier incarnations, broadly divisible into two major camps. The first of these camps largely cleaves to the “sacral female/political male” division of gendered roles described by Takamure et al. Okano reads the hime’s temporal authority as heavily grounded in her sacral performance, such that increasing secularisation of the central government was a significant factor in her decline151. Kuratsuka Akiko (倉塚曄子) asserts a similar division of duties, but with the hime’s fall attributed not to secularism but to male usurpation of the religious sphere, her traditional dominion152. Using the rule of Himiko and her brother as a model Matsumura presents a himehiko system based on contrasting but complementary roles: the male ruler not only handles mundane affairs, but is associated with the “outer” world, a la Himiko’s brother transmitting her will and acting as her “public front”; the female ruler is not only sacrally-based but also associated with the

“inner” world, the domestic sphere, indicating some form of cloistering as typical of the hime153.

Naoki diverges from the trend somewhat by arguing that, in ancient Japan, religion and politics were equated rather than contrasted. Nevertheless, his image of the division of duties between hime and hiko still separates the two between the sacral and the mundane: the hime, with the special

149 Yoshie 1993, 442–4.

150 Eg. Okano 1993, 27.

151 Okano 1993, 278.

152 Kuratsuka 1996, 17, 28.

153 Matsumura 1999, 122–8.

144 connection to spirituality afforded by her gender, pronounces oracles; the hiko enacts mundane governance on the basis of these oracles154.

Competing with this traditional depiction of sacral female with political male, another camp has emerged, preserving the conceptual core of himehiko – the systematised contrapuntal rule of a male-female pair as a widespread practice in ancient Japan – but with more nuanced images of hime and hiko’s respective roles. Yoshie’s research on the gendering of spirituality suggests that hime and hiko initially performed religious rites in concert, their roles reflecting gendered divisions of labour in the mundane world; increased emphasis on the hime’s religious function was a product of her later political disenfranchisement, Yoshie argues, and should not be romanticised as an image of balanced authority155. Maeda presents hime and hiko as sharing political and religious authority in a two- layered rulership system: the hiko, typically, the primary ruler, with the hime as his secondary chieftain; Himiko’s rulership representing a reversal of the trend156. Sekiguchi concurs, pointing to

Himiko’s clear political role, and grave goods suggesting industrial and military functions for female chieftains rather than clear dichotomy between the sacral and the political157. Ambros further argues for joint sacral performance, adding that the notion of religion and politics as clearly distinct

“spheres” is an anachronistic concept, not reflective of ancient perspectives on governance158.

However, such debate on the format of himehiko, of the precise division of duties between hime and hiko, still cleaves to the basic notion of himehiko as the standardised system of rulership in ancient Japan, complicating certain aspects while still rooted deep within the conceptual outlook of the early “golden age” narrative. Within this section, we will take a critical look through the records

154 Naoki 2008, 207.

155 Yoshie 1987, 35; 2007, 160, 205, 212; 2013, 16.

156 Maeda 2008, 17–9.

157 Sekiguchi 1987, 16–7.

158 Ambros 2015, 21.

145 at the concept of a “himehiko system” as a whole, and observe the textual pairs presented as examples in order to assess how well it encompasses rulership dynamics across the ancient archipelago.

We have already explored the interpretation of Himiko and her brother as such a male- female pair (see Ch. 4.1.1), and how it has been challenged by modern scholarship as an over- estimation of her brother-assistant’s authority (see Ch. 4.1.3). We have also looked at Empress Jingū, who has been variously read as written into a himehiko relationship with either her luckless husband

Chūai, and/or Takeshi-Uchi-no-Sukune159, a minister who provided her with considerable support during her regency and even shares the credit of the conquest of Korea in a later passage of the

Nihon Shoki (Nihon Shoki 17.Keitai 6.12); this reading will be discussed in greater detail below (see

Ch. 4.3.4). But these only represent two among many potential examples of paired rulership in ancient Japan, and both connected specifically to the Yamato paramountcy. To examine the notion of the “himehiko system” as a standard cultural format of rule across the archipelago, we must look at a broader selection of male-female pairs in the records.

In keeping with our theme of examining regional diversity in gender and authority, these have been divided into three broad geographic categories to compare and contrast with each other: pairs attributed to Kyushu, pairs attributed to the Eastern Provinces, and pairs attributed to Yamato and other provinces of central Honshu. We will also take a look at other leaders and formats of power attributed to these areas, in order to gain a broader understanding of gendered powership structures by region, and how “systematic” the scholarly conceit of the “himehiko system” truly appears.

159 Hori 1968, 191; Matsumura 1999, 125; Kidder 2007, 133; Naoki 2008, 208.

146

4.3.2. Himehiko and Formats of Power in Kyushu

First, we look to the island of Kyushu, whence appears one of the earliest textual examples of paired rulership past the Age of the Gods: Usatsu-Hiko (宇沙都比古/菟狭津彦) and Usatsu-Hime

(宇沙都比売/菟狭津媛), who have been named as an example of himehiko rulership by Sekiguchi and Maeda160. In both Kojiki and Nihon Shoki, these two figures appear as joint chieftains of Usa161, greeting the conquering Emperor Jinmu162 on his arrival in their territory and constructing for him a palace (Kojiki 2.Jinmuki; Nihon Shoki 3.Jinmu Tennō Sokui Zenki Kōin.10.Shinyū)163. The Nihon Shoki account gives this passage the questionable dating of 667 BCE, and further notes the pair as ancestors of the Usa-no-Kuni-no-Miyatsuko (菟狭国造) clan, the hereditary chieftains of Usa

District; Usatsu-Hime is married off to Jinmu’s retainer Ame-no-Taneko (天種子), solidifying the nascent hegemony’s ties to the region.

The pair’s presence within the annals of Emperor Jinmu poses something of an issue, however, due to the legendary founder’s own dubious historicity. Yoshimura, for example, reads

Jinmu as a mythical construct, the tales of his conquests by extension forming a myth-historic foundation story164. Within this context, Usatsu-Hiko and Usatsu-Hime – their names simply meaning

160 Sekiguchi 1987, 16; Maeda 2008, 17.

161 In modern Ōita Prefecture

162 Legendary founding emperor of Japan; reign traditionally dated 660–585 BCE

163 Kojiki: “故到豊國宇沙之時。其土人名宇沙都比古。宇沙都比賣。…二人。作足一騰宮而獻大御饗。”

(“Kare Toyokuni no Usa ni itarimaseru toki ni. Hito na wa Usatsu-Hiko. Usatsu-Hime…Futari. Ashi-Hitotsu-Agari- no-Miya wo tsukurite ōmiae tatematsuriki.”); Nihon Shoki: “行至筑紫國菟狹。…時有菟狹國造祖。號曰菟狹

津彥。菟狹津媛。乃於菟狹川上。造一柱騰宮。而奉饗焉。” (“Yukite Tsukushi-no-Kuni no Usa ni itarimasu…Toki ni Usa-no-Kuni-no-Miyatsuko no oya na wo Usatsu-Hiko. Usatsu-Hime to ii haberi. Sunawachi

Usa no kawakami ni. Ashi-Hitotsu-Agari-no-Miya wo tsukurite. Miae tatematsuru.”)

164 Yoshimura 2010, 36–9.

147

“man of Usa” and “woman of Usa” respectively – could be read as allegorical figures representing

Usa as a whole, symbolically enacting the capitulation and fealty of the Usa elite towards the Yamato dynasty.

A similar couple, identified by Maeda and Naoki as himehiko rulers165, appears later in the

Nihon Shoki, during the further conquests of Emperor Keikō166 (景行). Entering the land of Aso167 and wondering at its emptiness, the emperor is then confronted by two figures: Asotsu-Hiko (阿蘇都彦) and Asotsu-Hime168 (阿蘇都媛) (Nihon Shoki 7.Keikō 18.6.Heishi)169. They are identified in the text as both “gods” (神) and “people” (人); blurring of the lines between the two is not uncommon in the earlier portions of the imperial records, where figures such as Emperor Jinmu straddle the boundary between mortal and divine. The two, however, are not depicted as chieftains, per se, residing within/embodying a land otherwise described as empty and unpopulated. Their narrative role is again one of regional fealty; although asserting their presence before Keikō, the two do not object to the inclusion of Aso into the Yamato imperial holdings.

Both pairs above – Usatsu-Hiko/Usatsu-Hime and Asotsu-Hiko/Asotsu-Hime – appear as representations of a particular region and its fealty to the paramount within Yamato dynastic myth- history. This apparent derivation from Yamato mythic tradition – rather than that of northern

165 Maeda 2008, 17; Naoki 2008, 204–6.

166 Reign traditionally dated 71–130 CE

167 In modern Kumamoto Prefecture

168 Again, “Man of Aso” and “Woman of Aso”

169 “到阿蘇國也。其國郊原曠遠。不見人居。天皇曰。是國有人乎。時有二神。曰阿蘇都彥。阿蘇都媛。

忽化人以遊詣之曰。吾二人在。何無人耶。” (“Aso-no-Kuni ni mashimasu. Sono kuni nohara no hiroku tōkite. Hito no ie mirezu. Sumera-mikoto notamawaku. Kore kuni ni hito aran ya. Toki ni futahashira no kami ari. Asotsu-Hiko. Asotsu-Hime to iu. Tachimachi ni hito ni nari itarite notamawaku. Are futari ari. Nazo hito nakaran ya.”)

148

Kyushu, to where they are attributed – complicates their ability to truly serve as exemplars of regional rulership tradition and practice. Their appearance as male/female pairs may potentially reflect an understanding of such as a standard form of regional control. However, given in particular the Aso pair’s identification as “gods” (Nihon Shoki 7.Keikō 18.6.Heishi), it may also be linked to the occurrence within Yamato creation myth of male/female paired deities170, a phenomenon which is in the Nihon Shoki explicitly linked to (imported) Daoist notions of yin/yang dualism (Kojiki 1; Nihon

Shoki 1)171. Further, although interpreted as brother/sister partnerships172, in both pairs the actual relationship between hime and hiko is left largely undefined.

Jingū’s chapter of the Nihon Shoki attributes another male/female pair to northern Kyushu, in another narrative of imperial dominion: the “rebel” (tsuchigumo173) leader Taburatsu-Hime (田油

津媛) and older brother Natsuha (夏羽), situated in Yamato Territory174. Their rebellion is quashed when Empress Jingū moves on Kyushu and executes Taburatsu-Hime, further solidifying the imperial hold over the region; Natsuha, despite having raised an army, flees upon hearing of his sister’s execution, his fate after this left unknown (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai Tennō

170 E.g. the creator gods Izanagi the estuary gods Haya-Akitsu-Hiko and Haya-Akitsu-Hime, and the hearth gods

Okitsu-Hiko and Okitsu-Hime. Other unpaired gods also occur, however.

171 The Nihon Shoki explains that the first gods were exclusively masculine, for heaven/yang formed first; later, when earth/yin was formed, gods began to appear in male/female pairs (Nihon Shoki 1).

172 Maeda 2008, 17.

173 土蜘蛛, “earth-spider”; a term used for both a mythical spider-monster as well as rebels and local clans hostile to the Yamato regime. This pejorative exonym is associated with the ideology of Yamato imperialism, portraying conquered opponents as “barbaric aliens” (Yoshie 1993, 463); they are sometimes depicted as having tails, further portraying them as animalistic (Kojiki 2.Jinmuki).

174 In modern Miyama City, Fukuoka Prefecture; not to be confused with Yamato Province

149

9.3.Heishin)175. Although Natsuha acts as a military leader, Taburatsu-Hime appears as the dominant presence within the fraternal relationship, positioned as the centre of defiance of Yamato authority, singled out for execution. Her death effectively dissolves the rebellion, the fleeing Natsuha apparently not accorded enough importance for Yamato forces to pursue, suggesting that the primary leadership role may have fallen to her.

Despite appearing within the historically murky context of the Jingū Regency, Taburatsu-

Hime and Natsuha have a clearly defined relationship – older brother/younger sister – and distinct personal names. Although appearing in a narrative of imperial dominion, the two do not symbolically enact regional fealty, but rather appear as chieftains of a “renegade” clan sprouting up in a region already loosely under Yamato dominion. These details lend Taburatsu-Hime and Natsuha a stronger grounding in historicity than the Usa and Aso pairs, more likely to reflect power dynamics within the elite clans of Kyushu. In this case, the two appear not so much as a balanced co-rulership role, but rather as a primary female chieftain with a secondary male ruler. Taburatsu-Hime displays no particular sacral function; her military involvement is unclear. Information on the pair is too sparse to do more than guess as to whether their partnership was based around notions of complementary male and female rulership roles.

The notion of contrapuntal performance as the preferred format of rulership is further complicated by other examples of chieftaincy in Kyushu. The Nihon Shoki contains these three male/female pairs, but also attributes two sole female chieftains to the Kyushu region: Kamu-

Natsuso-Hime (神夏磯媛), appearing in an unspecified region of north-eastern Kyushu (Nihon Shoki

175 “轉至山門縣。則誅土蜘蛛田油津媛。時田油津媛之兄夏羽。興軍而迎來。然聞其妹被誅而逃之。”

(“Yamato-no-Agata ni utsurimashite. Sunawachi tsuchigumo Taburatsu-Hime wo tsuminau. Toki ni Taburatsu-

Hime ga irose Natsuha. Ikusa wo okoshite mōku. Shikashi sono iroto no korosaruru koto wo kikoete nigen.”)

150

7.Keikō 12.9.Boshin)176, and Hayatsu-Hime (速津媛) of Hayami Village, Okita (modern Ōita

Prefecture), who is also mentioned in the Bungo Fudoki (Nihon Shoki 7.Keikō 12.10; Bungo

Fudoki.Hayami District 1)177. The early seventh-century regional gazetteers, or fudoki, of Kyushu add several more female chieftains to this list, each attributed to “rebel”/tsuchigumo clans. In modern

Saga Prefecture, the Hizen Fudoki lists “Yasome” (八十女)178 of Mt. Otome (Hizen Fudoki.Kishima

District 1)179, Mirukashi-Hime (海松橿媛) (Hizen Fudoki.Matsura District 3)180, and the double-female chieftaincy of Ōyamadame (大山田女), and Sayamadame (狭山田女) (Hizen Fudoki.Saga District

1)181. In modern Nagasaki Prefecture, the same text names Hayakitsu-Hime182 (速来津姫) and

176 “爰有女人。曰神夏磯媛。其徒衆甚多。一國之魁帥也。” (“Koko ni onna ari. Kamu-Natsuso-Hime to iu.

Sono yakara nie nari. Hitokuni no hitokonokami nari.”)

177 Nihon Shoki: “到碩田國…到速見邑。有女人。曰速津媛。爲一處之長。” (“Ōkita-no-Kuni ni itaru…Hayami-no-Mura ni itaru. Onna-no-hito ari. Hayatsu-Hime to iu. Hitotokoro no osa to nasu. ”); Bungo

Fudoki: “時於此村有女人。名曰速津媛。為其處之長。” (“Toki ni kono mura onna-no-hito ari. Nazukete

Hayatsu-Hime to iu. Sono tokoro no osa to nasu.”)

178 It is unclear from the text whether 八十女 refers to an individual named Yasome or yaso no omina, “eighty women,” suggesting the women of the clan as a whole. Yoshie takes the latter interpretation (Yoshie 1993,

464).

179 “同天皇行幸之時。土蜘蛛八十女又有此山頂。常捍皇命。不肯降服。” (“Onaji sumera-mikoto yuku toki ni. Tsuchigumo no Yasome [yaso no omina; see note above] mata kono yama no itadaki ni ari. Tsune ni

ōmikoto wo sakaete. Matsuroi aezu.”)

180 “北昔者此里有土蜘蛛。名曰海松橿媛。” (“Mukashi wa kono sato ni tsuchigumo ari. Nazukete

Mirukashi-Hime to iu.”)

181 “于時有土蜘蛛大山田女。狹山田女。” (“Toki ni tsuchigumo Ōyamadame. Sayamadame ari.”)

182 “遣此郡速來村。捕土蜘蛛。於茲有人。名曰速來津姬。” (“Kono kōri Hayaki-no-Mura ni tsukawashite.

Tsuchigumo wo torashimu. Koko ni hito ari. Nazukete Hayakitsu-Hime to iu.”)

151

Ukianawa-Hime183 (浮穴沫媛) (Hizen Fudoki.Sonogi District 1). The Bungo Fudoki adds another female tsuchigumo chieftain, Itsuma-Hime (五馬媛) of modern Ōita Prefecture (Bungo Fudoki.Hita

District 1)184. Sole female rule is also observed among the Hayato185, a cultural group of Kyushu distinct from the Yamato Japanese. As late as 700, the Shoku Nihongi lists female chieftains Satsuma-

Hime (薩末比売), Kume (久売), and Hazu (波豆) as rebellious ringleaders186 among the Hayato of

Satsuma, now part of modern Kagoshima Prefecture (Shoku Nihongi 1.Monmu 4.6.Kōshin)187.

Examples of sole or pair male rule may also be found within Kyushu. The aforementioned

Hayatsu-Hime names five tsuchigumo chieftains among her enemies, including one with the clearly masculine name of Kunimaro (国摩侶) (Nihon Shoki 7.Keikō 12.10; Bungo Fudoki.Hayami District 1);

Kamu-Natsuso-Hime is similarly opposed by four local “bandit” (賊, ata) chieftains (Nihon Shoki

7.Keikō 12.9.Boshin). In Chūai’s day, male chieftains Wani (熊鰐) and Itote (五十跡手) appear as powerful figures in northern Kyushu (Nihon Shoki 8.Chūai 8.1.Jingo; Fudoki Itsubun.Chikuzen

Fudoki.Kōrui.Ito District); in Keitai’s day, the Kyushu elite Tsukushi no Iwai (筑紫石井/筑紫磐井) conflicts with imperial forces (Kojiki 3.Keitaiki; Nihon Shoki 17,Keitai 21.9.Kōgo; Fudoki

183 “有土蜘蛛。名曰浮穴沫媛。捍皇命甚無禮。即誅之。” (“Tsuchigumo ari. Nazukete Ukianawa-Hime to iu. Ōmikoto wo sakaete hanada iyanashi. Sunawachi koroshiki”)

184 “昔者此山有土蜘蛛。名曰五馬媛。” (“Mukashi wa kono yama ni tsuchigumo ari. Nazukete Itsuma-Hime to iu.”)

185 隼人; a native people of southern Kyushu.

186 Yoshie 1993, 465.

187 “薩末比賣。久賣。波豆。衣評督衣君縣。助督衣君弖自美。又肝衝難波。從肥人䒭持兵剽劫覓國使

刑部眞木䒭。於是勑竺志惣領。准犯决罸。” (“Satsuma-no-Hime. Kume. Hazu. E-no-kōri no kami E-no-Kimi

Agata. Kōri-no-suke E-no-Kimi Tejimi. Mata Kimotsuki no Naniwa. Kurahito-ra wo shitagae hyō wo chi shite kunimagi no tsukai Osakabe no Maki-ra wo obiyakasu. Koko ni oite Tsukushi no sōryō ni choku shite. Han ni jun shite ketsubatsu seshimu.”)

152

Itsubun.Tsukushi Fudoki.Otsurui.Iwai-no-Haka). Other sole male tsuchigumo chieftains are mentioned in the regional fudoki, including Utsuhio-Maro (欝比表麻呂) (Hizen Fudoki.Sonogi District

1), Ōmi (大身) (Hizen Fudoki.Matsura District 3), and Shinokaoki (小竹鹿奧) (Bungo Fudoki.Ōno

District 1).

Paired male rulership is also observed, even among regionally representative pairs, whose hiko/hime dichotomy is replaced by e (兄, older brother) and oto (弟, younger brother): Hinamori

(modern Kobayashi City, Miyazaki Prefecture) is represented by the brothers E-Hinamori (兄夷守) and Oto-Hinamori (弟夷守), who act as scouts for the emperor during his stay (Nihon Shoki 7.Keikō

18.3)188. Kuma (modern Kuma District, Kumamoto Prefecture) is represented by two brothers, both called Kumatsu-Hiko (熊津彦); whereas the elder, also dubbed Ekuma (兄熊), dutifully responds to the emperor’s summons, the younger, Otokuma (弟熊), refuses and is executed (Nihon Shoki 7.Keikō

18.4.Kasshi)189. The fudoki further mentions two pairs dubbed tsuchigumo – Ōmimi (大耳) and

Tarimimi (垂耳) (Hizen Fudoki.Matsura District 3) and Ōhashi (大鉗) and Ohashi (小鉗) (Fudoki

Itsubun.Himuka Fudoki.Kōrui.Chiho Village) – as well as a tsuchigumo triad, the three brothers

Ōshiro (大白), Nakashiro (中白), and Oshiro (少白) (Hizen Fudoki.Fujitsu District 1). Paired and sole male rulers are also seen among the Kumaso people of southern Kyushu, who during Keikō’s reign are led by the male pair of Atsukaya (厚鹿文) and Sakaya (迮鹿文) (Nihon Shoki 7.Keikō 12.12.Teiyū), and later by the “Brave of Kawakami,” Toroshikaya (取石鹿文) (Nihon Shoki 7.Keikō 27.12). A

188 “乃遣兄夷守。弟夷守二人令覩。” (“Sunawachi E-Hinamori. Oto-Hinamori futari wo tsukawashite.”)

189 “到熊縣。其處有熊津彥者兄弟二人。天皇先使徵兄熊。則從使詣之。因徵弟熊。而不來。故遣兵誅

之。” (“Kuma-no-Agata ni itari. Sono tokoro ni Kumatsu-Hiko to iu ani oto futari ari. Sumera-mikoto mazu

Ekuma wo mesu. Tsukai no shitagaite mairitari. Yorite Otokuma wo mesu. Shikashite mōkozu. Kare ni ikusa wo tsukawashite tsuminau.”)

153 brother-pair of Kumaso “braves” (建, takeru) also confront Yamato-Takeru in the Kojiki (Kojiki

2.Keikōki)190.

Among the chieftains of Kyushu, we do see at least one example of primary female/secondary male rule, and potential examples of male/female paired rule on equal footing.

However, we also see examples of sole male, sole female, paired male, paired female, and in one case, triple male chieftaincies. The enduring notion of the “himehiko system” posits a prefence for paired male/female rulership, based on notions of gender complementarity: a “masculine” and

“feminine” mode of chieftaincy working together in harmony. However, our examination of the records shows a considerable variety of rulership formats; the same is reflected in excavations of chieftain burial mounds191. Rather than having specifically “masculine” and “feminine” modes of leadership, expected to act as a set, the role of rulership appears gender-variable. This is consistent with archaeological findings described by Yoshie and Seike Akira (清家章), grave goods of which suggest no significant discrepancy in the roles of male and female chieftains, noting tools and weapons alongside ritual objects in female burials192.

Although the notion of the “himehiko system” is not entirely inaccurate, in that male/female paired rulership did exist within ancient Japan, it overemphasises only one among the many formats of power observed outside the Yamato imperial clan. Indeed, the overemphasis on this particular pairing, and the notion of contrapuntal gendered performance built around this, projects later normative assumptions of gender roles and gendered spheres onto ancient society, and does not give these ancient female chieftains full credit for the power and authority they – often solely – wielded, and their involvement in military and economic matters.

190 “尓熊曽建兄弟二人。” (“Koko ni Kumaso-Takeru ani oto futari.”)

191 Sekiguchi 1987, 17–8; Yoshie 1993 462–5; Seike 1998.

192 Yoshie 1993 462–5; Seike 1998.

154

Cultural diversity also presents itself as an underappreciated point: we see several names of female chieftains among the Hayato, a distinct cultural group from the ancient Yamato; many more examples of female rule are attributed to “tsuchigumo” clans of Kyushu. Although an ambiguous, loaded designation, the labelling of a clan as tsuchigumo served to delineate them not only from the

Yamato political hegemony, but also from Yamato culture and customs, as a barbarous “Other”. This angle of cultural and regional variation in women’s histories and experiences is one that has unfortunately gone largely overlooked, the “golden age” narrative and its legacies instead projecting a homogenous image of the archipelago with Yamato culture as the universal standard.

4.3.3. Himehiko and Formats of Power in the Eastern Provinces

Having looked at the chieftaincies of Kyushu, we now turn to what information survives concerning the Eastern Provinces, the land of the otoko-miko. What can we learn about formats of rulership there, and how do they correspond to those of Kyushu?

Piggott presents another male/female regional pair as an example of the “himehiko system”:

Kitsu-Hiko (寸津毘古) and Kitsu-Hime193 (寸津毘売) from Kitsu Village194 in the Eastern Province of

Hitachi. As with the Usa and Aso pairs, the two exist within a narrative of imperial dominion: in a tale recounted in the Hitachi Fudoki, the hero-prince Yamato-Takeru – Keikō’s son, and great conqueror for the Yamato regime –encounters the pair during his famous eastward expansion, where they are described as rebellious local chieftains, opposed to the Yamato dynasty. Yamato-Takeru slays Kitsu-

Hiko in battle, but to Kitsu-Hime he grants mercy following her surrender and acceptance of

193 “Man of Kitsu” and “Woman of Kitsu”

194 In modern Namegata City, Ibaraki Prefecture

155 subordination, she and her sisters pledging their service and hospitality to the prince (Hitachi

Fudoki.Namegata District 4)195.

Piggott further reads Kitsu-Hime’s submissive “hospitality” as sexual in nature, reflecting a process of Yamato kings and princes marrying regional elite women to solidify their territorial hold, breaking up pre-existing himehiko relationships196 (see Ch. 5.1.1). It is also reminiscent, however, of similar tales within the Yamato canon concerning older brother-younger brother pairs: the stories of

Ekuma/Otokuma (see Ch. 4.3.2), E-Ukashi (兄宇迦斯/兄猾)/Oto-Ukashi (弟宇迦斯/弟猾), and Eshiki

(兄師木/兄磯城)/Otoshiki (弟師木/弟磯城) (see 4.3.4) each depict a sibling pair, one of whom pledges allegiance to the Yamato paramount while the other, rebellious, is slain. Oto-Ukashi also displays hospitality for the victorious soldiers, throwing them a banquet (Kojiki 2.Jinmuki; Nihon

Shoki 3.Jinmu Tennō Sokui Zenki Bogo.8.Itsubi/11.Kishiki; Nihon Shoki 7.Keikō 18.4.Kasshi). Kitsu-

Hime’s narrative role may be akin to that of Oto-Ukashi; her gender does not inherently imply a sexual component to her hospitality.

Kitsu-Hiko and Kitsu-Hime are also described as kuzu (國栖), a term which is in the Hitachi

Fudoki both defined as “native inhabitants of the land” and equated with the rebellious

195 “藝都里。古有國栖名曰寸津毗古寸津毗賣二人。其寸津毗古。當天皇之幸。違命背化。甚無肅敬。

爰抽御劍。登時斬滅。於是。寸津毗賣。懼悚心愁。表舉白幡。迎道奉拜。天皇。矜降恩旨。放免其房。

更迴乘輿。幸小拔野之頓宮。寸津毗賣。引率姊妹。信竭心力。不避風雨。朝夕供奉。天皇。其慇懃惠

慈。” (“Kitsu-no-Sato. Inishie ni kuzu ari nazukete Kitsu-Hiko Kitsu-Hime futari to iu. Sono Kitsu-Hiko. Sumera- mikoto idemashi ni atari. Mikoto ni tagai omomuke ni somuku. Ito iya naki. Koko ni mitsurugi wo nukite.

Sunawashi kirikoroshi. Koko ni oite. Kitsu-Hime. Osori ureete. Shiroki hata kakagete. Michi wo mukaete orogami tatematsuri. Sumera-mikoto. Awaremi mimegumi kudarite. Sono ie hanachi yurushi. Sara ni mikoshi megurite. Onukino no karimiya ni idemashi. Kitsu-Hime. Irone iroto wo hikiite. Makoto ni kokoro wo tsukushite.

Fūu sakezu. Asayū tsukaematsuriki. Sumera-mikoto. Sono nemokoro uruwashimi wo mede.”)

196 Piggott 1994, 8; 1997, 38.

156 tsuchigumo197 (Hitachi Fudoki.Ibaraki District 1). Although the kuzu of Hitachi are distant from the tsuchigumo of Kyushu, this identification delineates both groups culturally from the people of

Yamato, a nuance which is elided within the “golden age” narrative in favour of a singular, culturally homogenous “ancient Japan”.

Within the rest of the Hitachi Fudoki, other examples of female chieftains (paired or sole) are few, with most prominent kuzu appearing as male (Hitachi Fudoki); paired male chieftaincies are also seen among the kuzu, with Yasakashi (夜尺斯) and Yatsukushi (夜筑斯) in Namegata (Hitachi

Fudoki.Namegata District 3). One exception is Abura-Okime (油置売) of Mt. Ashio in modern Ibaraki

Prefecture, whose identification as a “bandit” (山賊, yama-no-nishimono) may associate her with

“renegade” native groups such as the tsuchigumo, hostile to Yamato authority (Hitachi

Fudoki.Niibari District 1)198. Archaeological evidence suggests female chieftains existing in many parts of the archipelago199, but few references to them survive in textual form, particularly in contrast to the many female chieftains of Kyushu. Our knowledge of the Eastern Provinces is particularly focused Hitachi, due to the fragmentary survival of regional fudoki; in the sixth century, the Nihon Shoki describes two male leaders fighting over the chieftaincy of Musashi (modern Tokyo,

Saitama, and Kanagawa), with assistance from a male elite of Kamitsukeno (Gunma) (Nihon Shoki

18.Ankan 1.Urū 12), but information in this chronicle is also relatively sparse compared to Kyushu.

However, the angle of potential cultural distinction exists here as well, with the kuzu clans and with the Emishi people of Tōhoku who remain as a distinct ethnic group well into the Shoku Nihongi.

197 Kuzu is another pejorative exonym by the Yamato regime intended to imply “barbaric aliens” (Yoshie 1993,

463); as with the tsuchigumo, kuzu are occasionally depicted with tails (Nihon Shoki 3.Jinmu Tennō Sokui Zenki

Bogo.8.Itsubi).

198 “古老曰。古有山賊。名稱油置賣命。” (“Furu okina iwaku. Inishie ni yama-no-nishimono ari. Nazukete

Abura-Okime-no-Mikoto to shōsu.”)

199 Yoshie 1993 462–5; Seike 1998.

157

4.3.4. Himehiko and Formats of Power in the Kinai

Now let us turn to the central provinces, the heartland of Yamato control. Piggott further raises the example of Tomibiko (登美毘古)/Tomi-no-Nagasune-Biko200 (登美能那賀須泥毘古/長髓

彦) and Tomiya-Bime (登美夜毘売)/Nagasune-Hime (長髓媛) as a himehiko partnership, likening them to Kitsu-Hiko and Kitsu-Hime as a regionally-bound pair divided by conquest and marriage201.

As the dynastic ancestor Jinmu makes his eastward push into what would later become the imperial heartland of Yamato, Tomibiko appears as an adversary to his forces. The deity Nigi-Hayahi (迩芸速

日/饒速日), descended from the heavens to assist Jinmu, slays Tomibiko in combat, and takes to wife his younger sister Tomiya-Bime (Kojiki 2.Jinmuki; Nihon Shoki 3.Jinmu Tennō Sokui Zenki Bogo

4.Kōshin/12.Heishin)202. Tomiya-Bime’s position, however, is ambiguous: she is not listed along her brother, nor is any co-rulership function attributed to her, only appearing as a bride for Nigi-Hayahi, conjugal spoils of war. The assumption that, as Tomibiko’s sister, she acted as his contrapuntal

200 Tomi is located in moern Ikoma City, Nara Prefecture. According to the Nihon Shoki, “Nagasune” was the former name of Tomi (Nihon Shoki 3.Jinmu Tennō Sokui Zenki Bogo.12.Heishin)

201 Piggott 1997, 38–9.

202 Kojiki: “經浪速之渡而。泊青雲之白肩津。此時登美能那賀須泥毗古…興軍。待向以戰尓。…故迩藝速

日命娶登美毗古之妹。登美夜毗賣” (“Namihaya no watari wo hete. Aokumo no Shirakata-no-Tsu ni hatetamaiki. Kono toki Tomi-no-Nagasune-Biko…Ikusa wo okoshi. Machi mukaete tatakai shikaba…Kare Nigi-

Hayahi-no-Mikoto Tomibiko no imo. Tomiya-Bime ni miaite”); Nihon Shoki 2: “時長髓彥乃遣行人言於天皇曰。

…號曰櫛玉饒速日命。…是娶吾妹三炊屋媛。〈亦名長髓媛。亦名鳥見屋媛。〉” (“Toki Nagasune-Hiko no tsukai wo matate sumera-mikoto ni mōshite mōsaku…Nazukete Kushitama-Nigi-Hayahi-no-Mikoto to mōsu…Kore a ga iroto Mikashikiya-Hime wo yometote. (Mata no na wa Nagasune-Hime. Mata no na wa

Torimiya-Hime.)”)

158 partner is premature given the appearance in text and archaeology of multiple formats of rule, practised alongside each other203.

Later, another pair emerges in Yamato Province: Sahohime and her elder brother Sahohiko, linked to the Saho region in the northern Yamato Basin204 through their mother, Saho-no-Ōkurami-

Tome. Unlike the previous examples, the two are not explicitly regional chieftains, instead being of imperial clan descent. However, their relationship has nevertheless been read by Matsumura and

Maeda as a reflection of regional himehiko power structures disrupted by an interloping Yamato paramount205: in this case, Emperor Suinin, who takes Sahohime as his principal consort. The plotting

Sahohiko attempts to convince his sister to assassinate her husband, promising her that the two of them will rule together once he has taken the throne206. Sahohime, caught between her familial and

203 Sekiguchi 1987, 17–8; Yoshie 1993 462–5; Seike 1998.

204 Associated with the Saho River of northern Nara

205 Matsumura 1999, 123; Maeda 2008, 17.

206 Kojiki: “沙本毗賣命之兄沙本毗古王問其伊呂妹曰。孰愛夫与兄歟。答曰愛兄。尓沙本毗古王謀。曰

汝思愛我者。將吾与汝治天下而。” (“Sahobime-no-Mikoto no irose Sahobiko-no-Miko sono iromo ni toite iikeraku. Oto to irose to izureka hashiki zo. Irose zo hashiki to kotaetamaiki. Koko ni Sahobiko-no-Miko hakarigochite. Imashi makoto ni a wo hashiku omōsaba. Masa ni are to imashi to ame-no-shita shirasan to iite.”); Nihon Shoki: “皇后母兄狹穗彥王謀反欲危社稷。因伺皇后之鷰居而語之曰。汝孰愛兄與夫焉。於

是。皇后不知所問之意趣。輙對曰。愛兄也。則誂皇后曰。…是以冀吾登鴻祚。必與汝照臨天下。則高

枕而永終百年。” (“Kisaki-no-miya no omo no harakara Sahohiko-no-Ō mikado katamuken to hakarite kuni wo ayabumen to omou. Yorite kisaki-no-miya utakushi ni mashimasu wo ukagaite katarite iwaku. Imashi irose to oto to izureka utsukushiki. Koko ni. Kisaki-no-miya tou kokoro wo shiroshimesazu shite. Sunawachi kotaete iwaku. Konokami utsukushi. Sunawachi kisaki-no-miya wo atoraete iwaku…Kore wo mote koinegawaku wa are amatsu-hitsugi shiraseba. Masa ni imashi to ame-no-shita ni teri nozomite. Sunawachi makura wo takamarite hitafuru ni momotose wo oen koto.”)

159 marital loyalties, can neither bring herself to murder her husband nor to merely watch her brother’s execution, dying alongside Sahohiko at the last (Kojiki 2.Suininki; Nihon Shoki 6.Suinin 4.9.Bōshin).

The co-rulership of brother and sister proposed by Sahohiko suggests an acceptance of kin- based pair rulership as a potential format of power by the ancient Yamato, though contrasting with the imperial clan’s emphasis on marital bonds. At the same time, however, Sahohiko and Sahohime are themselves imperial kinsfolk, cousins to Suinin, with Sahohiko taking the title of a lesser prince

(Kojiki 2.Kaikaki); the region they represent is one already under the sphere of Yamato control, with no hint of a ruling relationship between them prior to Sahohime’s marriage. Yoshimura instead interprets the pair as an allegory for an uprising of Saho, with Sahohiko and Sahohime personifying the region and its divided loyalties207. Their proposed co-rulership, in this case, could reflect a sense of conflict between “old” regional rulership traditions and “new” imperial models, without tying the

“old” style of kin-based partnership to Saho specifically.

Piggott proposes Prince Take-Haniyasu-Hiko (建波迩夜須毘古/武埴安彦) and his wife

Atahime (吾田媛) as another example of himehiko found amongst the imperial clan rebels208. Take-

Haniyasu-Hiko, an uncle of Emperor Sujin, plots rebellion against his nephew. Atahime first performs a ritual for their success, taking soil from Mount Kago in Yamato and binding it in her neckerchief with the incantation, “This signifies the spirit of Yamato.”209 Husband and wife then split their forces, each leading a contingent of soldiers to make a two-pronged assault on the Yamato capital. Atahime

207 Yoshimura 2010, 61.

208 Piggott 1994, 8; Piggott 1997, 38.

209 “武埴安彥之妻吾田媛。密來之取倭香山土。裹領巾頭。而祈曰。是倭國之物實。” (“Take-Haniyasu-

Hiko no tsuma Atahime. Hisoka ni kitarite Yamato no Kagoyama no tsuchi wo torite. Hire no hashi ni tsutsumite.

Inorite iwaku. Kore Yamato-no-Kuni no monoshiro.”)

160 and her soldiers are intercepted and slain by imperial forces at Ōsaka210, while her husband suffers a similar fate at Kusuba211 (Nihon Shoki 5.Sujin 10.9.Jinshi).

While Atahime does perform ritual activity, both she and her husband are active in leading soldiers, her role military as well as ritual. Although the two form a working partnership, the significance of gender here is unclear, their cooperative military leadership challenging the notion of separate but complementary roles implied in the contrapuntal “himehiko system”. As with Sahohiko and Sahohime, husband and wife emerge from within the imperial clan, existing within the bounds of Yamato imperial hegemony rather than being clearly tied to pre-existing regional systems of control. Despite their “rebel” status, neither pair is named tsuchigumo; they are not culturally delineated from the imperial line.

As with Chūai and Jingū, the pairing of Take-Haniyasu-Hiko and Atahime depicts an imperial male figure receiving considerable support from a wife-assistant, one who partakes in both magico- religious activity and warfare. The “traditional” model of himehiko, however, is based around the partnership of natal kin – brother and sister212 – rather than the less stable marital relationship of husband and wife. Takamure posits that brother/sister pairings were the earlier version of himehiko, with husband/wife pairings as a later replacement, a change which privileged husbands through their natal links to the clan and acted as an intermediary stage between women’s empowerment through himehiko and subjugation under patriarchy213. However, the timelines suggested by the records complicate this: Take-Haniyasu-Hiko and Atahime are attributed to an earlier period than

210 Not the modern metropolis of Ōsaka, but rather the Ōsaka region of modern Kashiba City, Nara Prefecture

211 Modern Kuzuha District, Hirakata City, Ōsaka Prefecture

212 Sakima 1926, 51; Takamure 1966a, 69; 1966b, 96–117; Yanagita 1984, 308; Okano 1993, 28; Kuratsuka 1996,

28; Matsumura 1999, 128.

213 Takamure 1966b, 112.

161 the brother/sister partnership of Sahohiko and Sahohime, or Taburatsu-Hime and Natsuha; Jingū and Chūai are presented as contemporaneous with the latter.

To add more complication into the mix, Jingū’s other proposed himehiko partner, Takeshi-

Uchi-no-Sukune, was neither kin nor husband; rather, he was a storied (and heavily mythologised) minister who also appears prominently in the service of male emperors Keikō, Chūai, Ōjin, and

Nintoku214 (仁徳). Matsumura theorises that his close association with Jingū may have been influenced by the relationship of Himiko and her brother in the Records of the Three Kingdoms215, reflecting a specific previous example of paired rule rather than a broader convention. We may also recall, however, Yoshie’s notes on the scholarly treatment of Himiko’s younger brother: that, while male sovereigns are often in receipt of assistance from male allies and retainers, it is only in the case of a female sovereign that such aides are elevated into the role of assumed co-ruler216.

As in Kyushu, a variety of rulership formats are attributed to Yamato, including sole male, sole female, paired male, and paired male-female. As with Hitachi, the sole chieftains tend to skew male (Kojiki; Nihon Shoki), although a lack of surviving fudoki for the Kinai hinders our gathering of data. A sole female chieftain, Niiki-Tobe (新城戸畔), is credited as ruler of Soekami in Yamato; she too is designated as tsuchigumo for her opposition to Jinmu (Nihon Shoki 3.Jinmu Tennō Sokui Zenki

Kibi.2.Shingai)217. Her name appears linked to a cluster of female chieftains in Kii Province (modern

Wakayama): Nagusa-Tobe218 (名草戸畔), Nishiki-Tobe219 (丹敷戸畔), and Arakawa-Tobe220 (Nihon

214 Reign traditionally dated 313–399 CE

215 Matsumura 1999, 126.

216 Yoshie 2013, 10–11, 16.

217 “是時層富縣波哆丘岬有新城戶畔者。” (“Kono toki ni Sō-no-Kōri Hata-no-Okazaki ni Niiki-Tobe to iu mono ari.”)

218 “軍至名草邑。則誅名草戶畔者。” (“Miikusa Nagusa-no-Mura ni itarite. Sunawachi Nasuga-Tobe wo tsumisu.”)

162

Shoki 3.Jinmu Tennō Sokui Zenki Bogo.6.Teishi/Nihon Shoki 5.Sujin 1.2.Shingai); possibly indicating a tradition of female chieftaincy originating in this region and spreading eastward. Himiko herself is tentatively placed in this region, and dual male chieftaincies – E-Ukashi and Oto-Ukashi221, and Eshiki and Otoshiki (Kojiki 2.Jinmuki; Nihon Shoki 3.Jinmu Tennō Sokui Zenki Bogo.8.Itsubi/11.Kishi/Jinmu

2.2.Itsushi) – are also observed.

A description of “braves” in Uda depicts a female army being raised against Jinmu’s forces, quartered separately from the male army, although the leadership of this cohort is uncertain (Nihon

Shoki 3.Jinmu Tennō Sokui Zenki Bogo.9.Boshin)222; this “women’s army” contrasts with Jingū’s affectation of masculinity before warfare (see Ch. 4.2.3), suggesting some level of cultural variation in the gendering of such tasks.

Within the imperial clan, a trend does arise of early Yamato paramounts assisted by their priestess kin. In the Nihon Shoki, Atahime and Take-Haniyasu-Hiko’s plot is revealed through the omen interpretations of Sujin’s great-aunt Yamato-Totohi-Momoso-Hime, a wise priestess who is also shown to perform divinations and enter a divine marriage with the god Ōmono-Nushi (Nihon

Shoki 5.Sujin 7.2.Shinbō/7.8.Kiyū/10.9.Jinshi). Sujin also receives the sacral assistance of his daughters Toyosuki-Irihime and Nunaki-Irihime, who are entrusted with the worship of Amaterasu and Yamato-no-Ōkunitama, respectively (Nihon Shoki 5.Sujin 6); Suinin’s daughter Yamato-Hime,

219 “帥軍而進至熊野荒坂津。〈亦名丹敷浦。〉因誅丹敷戶畔者。” (“Miikusa shite susumite Kumano no

Arasaka-no-Tsu ni itarimasu. (Mata no na wa Nishiki-no-Ura.) Yorite Nishiki-Tobe wo tsuminau.”)

220 “又妃紀伊國荒河戶畔女遠津年魚眼眼妙媛。” (“Mata no kisaki wa Kii-no-Kuni no Arakawa-Tobe ga musume Tōtsu-Ayume-Makuwashi-Hime.”)

221 “天皇使徵兄猾及弟猾者。…是兩人菟田縣之魁帥者也。” (“Sumera-mikoto E-Ukashi oyobi Oto-Ukashi wo mesashimu…Kono futari wa Uta-no-Kōri no hitokonokami nari.”)

222 “時國見丘上則有八十梟帥。…又於女坂置女軍。男坂置男軍。” (“Toki ni Kunimi-no-Okano no ue ni sunawachi yaso takeru ari…Mata Mesaka ni me no ikusa wo oki. Osaka ni o no ikusa wo oki.”)

163 mythical founder of Ise Shrine and priestess of Amaterasu, likewise both assists her father and provides sacral aid and support to the conquests of her nephew Yamato-Takeru (Kojiki 2.Keikōki;

Nihon Shoki 6.Suinin 25.3.Teigai; Nihon Shoki 7.Keikō 40.10.Bogo).

This pattern of emperors (or conquering hero-princes) receiving assistance from priestess- kinswomen has been interpreted as a manifestation of himehiko dynamics by various scholars across the years, including Takamure, Kidder, and Emura Hiroyuki223 (榎村寛之). In these relationships, however, the female figure (or figures, in the case of Sujin’s daughters) is configured within the text as an adjunct to a male ruler, providing him with sacral assistance, not as a partner in rulership; a clear hierarchy divides the two. Recall the dynamics of Himiko’s reign, with the female as the clear centre of monarchic power, yet the male assistant frequently assumed as co-ruler224; the interpretation of priestess-assistants as hime co-rulers arguably represents an inversion of this trend, in both cases the presumption being that any assistant of a different gender to the monarch must in fact be a hidden partner.

Furthermore, the priestesses’ duties appear restricted to their sacral function; Yamato-Hime provides Yamato-Takeru with advice, but does not co-command his armies. This “sacral female, political male” division is, while a traditional foundation of the “himehiko system” theory, a point of divergence from overt textual and archaeological appearances of paired rulership, which do not evince a strong distinction in the spheres governed by male versus female rulers. The “shaman- queen” Himiko represents her kingdom in international diplomacy, Jingū is a war leader as well as a shamaness, and while Atahime’s ritual performance at Mount Kago may indicate a magico-religious function, both husband and wife act jointly as military commanders, leading their own contingents of troops225. Other female co-chieftains – such as Kitsu-Hime or Taburatsu-Hime – show no particular

223 Takamure 1966b, 112; Naitō 1976, 204; Yanagita 1984, 308; Kidder 2007, 133; Emura 2009, 10.

224 Yoshie 2013, 10–11, 16.

225 Piggott 1994, 8; 1997, 38.

164 sacral role at all; burial mounds show a range of grave goods for female chieftains, with no particular pigeonholing226.

The kin relationships between emperor and assisting priestess are also highly variable, lacking the standardisation attributed to himehiko: father and daughter (Suinin/Yamato-Hime), father and two daughters (Sujin/Toyosuki-Irihime/Nunaki-Irihime), nephew and aunt (Yamato-

Takeru/Yamato-Hime), grandnephew and great-aunt (Sujin/Yamato-Totohi-Momoso-Hime). This trend of king/priestess partnerships is certainly worthy of discussion for what it can show us about gender, religion, and power in the ancient Yamato dynasty. However, it is hasty to assume that these relationships formed an overarching, systematic pattern with and functioned identically to all other examples of paired chieftaincy throughout the archipelago.

Indeed, it seems overhasty to declare that paired male-female chieftaincy alone held a special systematic status in the traditions of ancient Japan. Certainly, we have seen examples of such, but we have also seen a great diversity in formats of power, across both text and archaeology227: sole male, sole female, paired male, paired female, paired male-female, primary female and secondary male, primary male and secondary female, even triple-male rule. The contrapuntal performance of male and female does not appear as a site of particular importance in these narratives of rulership; indeed, male-female ruling pairs often receive similar narrative treatment to dual-male pairs. “Himehiko” appears as only one among a great variety of rulership systems appearing throughout the archipelago, and does not seem to receive particular preference. Rather than a single, systematised format of power based on notions of gender complementarity and

226 Yoshie 1993 462–5; Seike 1998.

227 Sekiguchi 1987, 17–8; Yoshie 1993 462–5; Seike 1998.

165

“masculine” versus “feminine” modes of rulership, chieftaincy appears as diverse and highly variable, with gender as a negotiable (if skewing masculine) factor.

Some elements of regional diversity do appear worthy of further study: accounts of ancient

Kyushu, for example, depict a relatively sizeable amount of female chieftains, particularly among the

Hayato and those clans labelled as “tsuchigumo.” These cultural distinctions have been elided throughout the “golden age” narrative in favour of a singular, homogenous image of pre-dynastic

Japanese culture and customs. The enduring narrative thread of the “himehiko system,” a product of the “golden age” narrative, represents a homogenisation of what appears as a diverse and fascinating array of rulership formats and relationships to gender appearing throughout the ancient archipelago. In doing so, it projects a specific image of contrapuntal gender roles – based on imperial clan custom and/or historians’ own normative assumptions – as the norm across the entirety of ancient Japanese society.

The Yamato imperial clan, conversely, appears to demarcate gender roles differently, or at least more clearly, than other clans, with warfare as a masculine pursuit and shamanic/oracular assistance as feminine, although these roles are yet negotiable. Through the images of Himiko and in particular Jingū we see not a specific “feminine” model of kingship but an androcentric, male- preferred notion of the sovereign, nevertheless accessible to women; a complex relationship between gender and authority not easily described with neat categories of “patriarchy” versus

“empowerment”.

166

CHAPTER 5: WOMEN IN THE “SILVER AGE”

So far, we have looked mainly at images of women, female spiritual performance, and female rulership in the pre-ritsuryō period, the alleged “golden age”. But the “golden age” narrative is not only a story of an idyllic, nostalgic, empowered past, but also the story of its loss: how the inexorable encroachment of patriarchy stripped away the power and prestige once afforded to womankind. In this chapter, we turn from “golden age” to “silver” (see Ch. 1.3), to the “deterioration” phase of the deteriorationist narrative, examining the impact of Yamato centralisation and ritsuryō codes on women’s social position. Through this, we shall see if the later ancient period truly evinces the decline in female power posited by the “golden age” narrative, or whether this, too, is a more complex and diverse matter than a linear deteriorationist narrative gives credit.

First, we will examine the impact of Yamato political centralisation on regional elite women: the Yamato monarchy’s use of chieftain-class women as sites of political control through “stranger king” marriage and uneme requisitioning, and the centralisation of regional governance around

Yamato political norms. We will also study the Yamato elite’s adoption of patriarchal Chinese thought, such as Confucianism and Tang legal codes, and its promulgation through central legislature.

Through this we will see that, while some reduction can be observed in regional female status as a result of central legislation and ideology, the issue is muddied by a disparity between official positions and procedures (de jure) and actual practice (de facto), with women continuing to maintain social standing and autonomy in village hierarchies and marital structures for some time in an

“unofficial” capacity. We will also see that androcentric norms are also present in the culture of the

Yamato imperial clan from early records, and that placing sole blame upon Confucianism for patriarchal spread promulgates a nationalist paradigm of the “foreign” as a corrupting influence.

167

Our examinations of the Yamato programme of religious centralisation, and its impact on female religious practitioners, will show a similar discrepancy. Although typically cast as a narrative of pure decline – the state-sponsored jinja-miko lacking bureaucratic rank and reduced to the bare bones of her ritual function, the unofficial miko controlled and persecuted through state suppression

– the experiences of miko under the ritsuryō government appear as multifaceted and complex, rather than total marginalisation. The miko remained an important figure within folk religious practice, and had access to social capital outside of the bureaucratic hierarchy; a point elided by the

“golden age” narrative’s overemphasis on bureaucratic legislation and central elite perspectives.

From this we will take a close look at the royal sacral roles of Ise saigū and Kamo saiin, princess-priestesses who are, within the “golden age” narrative, frequently positioned as depowered hime, former partners of the emperor stripped of all rulership function and now acting as mere spiritual support. Here, we will see that the relationship of saigū and saiin to “power” is again a complex one, encompassing many different understandings of the term in conflicting ways. The roles of saigū and saiin were loci of considerable sacral and political power that was not extended to them on an individual level, their ritual efficacy gained through a loss of personal autonomy which ties them closer to the apparent qualities of Himiko than a broad, ill-defined notion of a sacrally- powerful co-ruling “hime”. The mythic progenitors of the saigū are not symbols of an earlier, empowered era, but later constructs serving a particular role within a narrative of Yamato dominion.

Rather than a narrative of direct decay from ancient glories, detached of surrounding context, the saigū and saiin must be understood within the specific political and religious circumstances that formed them.

Finally, we will discuss the early status of Buddhist nuns in Japan, whose relative prominence and frequent ritual partnership with monks have also invited himehiko comparisons. Through these examples, we shall observe the complex factors of political and religious context that went into the

168 development of these roles, and the tensions held within the notion of “power” that are not encompassed within the “golden age” narrative.

5.1. Centralisation and Gender: The Impact of Yamato Control

5.1.1. The Process of Yamato Centralisation

The story of how one chieftain clan in Yamato rose to a position of supremacy over most of the archipelago is still a murky one. The imperial clan’s own founding legends attribute much of it to the conquests of such powerful figures as Jinmu or Yamato-Takeru, but historians today still struggle to sort the historic from the mythic in these tales. We shall see, however, that this process of centralisation appears to have had considerable impact upon pre-existing power structures.

Previously (see Ch. 4.3.3), we discussed Piggott’s interpretation of the tale of Kitsu-Hiko and

Kitsu-Hime: that Kitsu-Hime’s surrender to Yamato-Takeru included her becoming his consort1, indicated by the description of she and her sisters serving him “from morning to night” (Hitachi

Fudoki.Namegata District 4)2. Piggott situates this within a broader tradition of Yamato princes and paramounts becoming centralised “stranger-kings” tied to conquered regions through union with local elite women. The women, in turn, embody their clan’s submission through the bearing of the

“stranger-king’s” children, while at the same time solidifying the clan’s position by forging maternal- line relations with their imperial offspring3. A system of uneme4, in which regional elite women were

1 Piggott 1994, 8; 1997, 38.

2 “寸津毗賣。引率姊妹。信竭心力。不避風雨。朝夕供奉。” (“Kitsu-Hime. Irone iroto wo hikiite. Makoto ni kokoro wo tsukushite. Fūu sakezu. Asayū tsukaematsuriki.”)

3 Piggott 1994, 8; 1997, 38.

4 采女

169 requistioned by the Yamato court to serve as ladies-in-waiting, is also attested in ancient records5.

These uneme may, as Kuratsuka argues, have had a secondary function as political hostages, handed over to the central court as a gesture of subservience6. Similar hostage relationships are certainly evident in the Yamato court’s tributary relationships with the kingdoms of Korea, with Korean princes and noblemen as the political bargaining chips (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai

Tennō 9.10.Shinchū; Nihon Shoki 9.Jingū Kōgō Sesshō 5.3.Kiyū; Nihon Shoki 23.Jomei 3.3.Kōshin;

Nihon Shoki 24.Kōgyoku 1.8.Heishin; Nihon Shoki 24.Kōgyoku 2.7.Shingai; Nihon Shoki 26.Saimei 1).

In both “stranger-king” and uneme models, regional elite women are made into battlegrounds of regional control, means by which the central Yamato hegemony may leverage power.

Under the traditionalist (“sacral female/political male”) interpretation of the “himehiko” system, a further dimension is read into the Yamato use of women in their centralisation process:

Takamure, Kuratsuka, and Matsumura have argued that stranger-king” marriages and uneme recquisitions served as a means of undermining male chieftains by subordinating their hime – and the religious legitimisation the hime provided – to the centre7. As we observed in the last chapter, however, this “sacral female/political male” division is not in particular evidence outside the specific norms of the Yamato imperial clan.

Despite the ambiguity in Kitsu-Hime’s case – a marriage to Yamato-Takeru not actively stated within the text (Hitachi Fudoki.Namegata District 4) – other examples do appear within

Yamato conquests narratives of princes and retainers marrying local women. Yamato-Takeru himself is shown entering into other, more peacefully arranged, marriages with local elites, such as Miyazu-

Hime from the chieftain family of Owari8 (Kojiki 2.Keikōki; Nihon Shoki 7.Keikō 40), or the two

5 Takamure 1966b, 113; Sekiguchi 1987, 19; Piggott 1994, 14; Kuratsuka 1996, 9–10; Matsumura 1999, 126–7.

6 Kuratsuka 1996, 10.

7 Takamure 1966b, 111; Kuratsuka 1996, 9–10; Matsumura 1999, 124.

8 Modern western Aichi Prefecture

170 daughters of the regional chieftain of Mino9 (Kojiki 2.Keikōki; Nihon Shoki 7.Keikō 4.2). Jinmu’s right- hand deity Nigi-Hayahi is married to Tomiya-Bime, younger sister of the regional elite Tomibiko, slain by Nigi-Hayahi himself as part of Jinmu’s campaign of conquest (Kojiki 2.Jinmuki)10; Usatsu-Hime is similarly married off to Jinmu’s retainer Ame-no-Taneko (Nihon Shoki 3.Jinmu Tennō Sokui Zenki

Kōin.10.Shinyū)11.

These “stranger-king” marriages were not a universal policy regarding female elites, however: rebellious female chieftains, such as Nagusa-Tobe, Nishiki-Tobe, Niiki-Tobe, or Taburatsu-

Hime, were executed in much the same manner as a male adversary (Nihon Shoki 3.Jinmu Tennō

Soku Zenki Bogo.8.Itsubi/Kibi.2.Shingai; Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai Tennō

9.3.Heishin). Conversely, sole female chieftains on good terms with the Yamato monarchs, such as

Kamu-Natsuso-Hime and Hayatsu-Hime, appear to retain their position without “stranger-king” marriage (Nihon Shoki 7.Keikō 12.9.Boshin/10); the marital union may have been reserved for the female relatives of a male (sole or dual) ruler.

Matsumura reads the story of Sahohiko and Sahohime (see Ch. 4.3.2) as a reflection of the tensions caused by the “stranger-king” model: an elite male feeling his own power base threatened by the paramount’s requisition of his kinswoman, and an elite female torn between old familial and new marital loyalties12 (Kojiki 2.Suininki; Nihon Shoki 6.Suinin 4.9.Bōshin–Suinin 5.10.Kibō).

Transformed through marriage into an intermediary between her natal clan and the imperial throne,

9 Modern Gifu Prefecture

10 “故迩藝速日命娶登美毗古之妹。登美夜毗賣” (“Kare Nigi-Hayahi-no-Mikoto Tomibiko no imo. Tomiya-

Bime ni miyaite.”)

11 “是時。勑以菟狹津媛。賜妻之於侍臣天種子命。” (“Kono toki ni. Mikotonori wo mote Usatsu-Hime wo.

Ōmachikimi Ama-no-Taneko-no-Mikoto ni awasetamau.”)

12 Matsumura 1999, 123.

171

Sahohime herself – representing the women of elite clans – is made into the unwilling site of disputes between regional elites and the Yamato dynasty.

The Nihon Shoki version of this tale further suggests a sense of instability felt also by the elite women subject to such arrangements. When Sahohiko tries to tempt Sahohime into helping him usurp power, he implies that her position as consort is tenuous, reliant on the vagaries of the paramount’s sexual favour, and that ruling alongside her kin will grant her a far more solid base of power:

“Thou servest thy husband with thy beauty, but should thy beauty fade, so too will his

affection. There are many beauties in the realm, who each seek his affections for their own;

how wilt thou rely on thy looks forever? But if I may ascend the Heavenly Sun Lineage, then

thou and I may rule the realm together, and thy pillow shall be raised high for a hundred

years hence; does that not please thee?”13 (Nihon Shoki 6.Suinin 4.9.Bōshin)

Read through the lens of conflict between regional and central elites, this passage suggests a certain precariousness to the position of “stranger-king’s” wife, recognised even by the Yamato compilers.

The later custom of “double royalty,” in which emperors’ nieces and half-sisters were favoured as principal consorts over women from other clans, further eroded the position of the regional consort,

13 “夫以色事人。色衰寵緩。今天下多佳人。各遞進求寵。豈永得恃色乎。是以冀吾登鴻祚。必與汝照

臨天下。則高枕而永終百年。亦不快乎。” (“Sore kao wo mote hito ni tsukawaba. Kao otoroete megumi yamu. Ima ame-no-shita ni kao-yoki onna sawanari. Onoono takai ni susumite megumaren koto wo motomu. A ni hitafuru ni kao wo tanomu koto eru ya. Kore wo mote koinegawaku wa are amatsu-hitsugi shiraseba. Masa ni imashi to ame-no-shita ni teri nozomite. Sunawachi makura wo takamarite hitafuru ni momotose wo oen koto. Mata kokoro yokazaran ya.”)

172 stripping her of the opportunity to forge maternal ties between her clan and future emperors and concentrating authority firmly within the imperial family14.

The weakening of regional female authority is further attested through archaeological evidence, which shows the presence of female chieftain burials in the early Kofun period15, but decreasing as the period went on16. Outside of uneme and “stranger-king” marriages, Seike raises another possible factor in the decline of female chieftaincies: the militarisation of the chieftain’s role as a result of the centralisation process, resulting in an increasing number of sole-male rulership structures17. Although the previous two chapters have shown some evidence of female military involvement – such as Atahime – the textual positioning of Jingū’s and Amaterasu’s own martial turns do suggest a broad gendering of warfare as a masculine occupation, at least within the eyes of the Yamato administration (Kojiki 1; Nihon Shoki 1.6–7; Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai

9.4.Kōshin). As late as 746, a woman of the former Owari chieftain clan, Owari no Ogura (尾張小倉), occupied the role of kuni-no-miyatsuko, a former chieftaincy position now relegated to provincial ritual coordinator (Shoku Nihongi 17.Tenpyō 19.3.Boin)18, but by then she was a rare exception in a post both male-dominated and devoid of previous rulership qualities.

This decline of women’s access to authority, however, was not universal across all levels of society. Yoshie’s research on female economic management at the village level shows a great deal of status and authority accorded to the sato toji (里刀自), or female village leader. Unearthed communiqués show that, while legislature attributed village headman status only to the male sato

14 Piggott 1994, 14–5.

15 300–538 CE

16 Seike 1998, 43–5.

17 Seike 1998, 43–5.

18 “命婦從五位下尾張宿祢小倉授從四位下。爲尾張國々造。” (“Myōbu Jūgoi-no-ge Owari-no-Sukune

Ogura Jūshii-no-ge wo sazukete. Owari-no-Kuni no kuni-no-miyatsuko to nasu.”)

173 osa (里長), in practice regional bureaucrats also coordinated with sato toji as local elites and independent economic managers19. Although the “silver age” did see a decline in women’s authority in the upper echelons of regional society, at the village level women continued to hold considerable de facto status and autonomy, a gulf emerging between official bureaucratic legislature and actual practice. The disparity between elite and common custom, between code and practice, emerge as underappreciated points within the “golden age” framing; a linear deteriorationist narrative focused primarily on the doings of the central bureaucracy insufficient to cover the diversity of women’s experiences in ancient Japan.

5.1.2. Confucianism, Patriarchy, and the End of the “Golden Age”

Another such gulf between legislation and practice appears when we consider the

Confucianism-inspired marital laws of the ritsuryō20 state. Formed during the Asuka period21, the ritsuryō bureaucracy was a further centralisation effort by the Yamato monarchy, creating a new system of bureaucratic postings (including centrally-appointed gubernatorial bureaus, further supplanting regional power structures) and standardised laws and procedures along Yamato-drawn lines. These codes were based off Tang Chinese models, themselves heavily influenced by Confucian morality, which proved an issue: Chinese legal codes were based around a rigidly patriarchal family mode, and Confucianism actively promulgated patriarchal values. Confucian texts – which were

19 Yoshie 1993, 452.

20 律令, “criminal, administrative, and civil codes”; referring to several Tang-inspired laws and codes beginning with the Taika Reforms of 645 and continuing through the Asuka and Nara periods. “Ritsuryō” is also sometimes used in reference to the bureaucratic administration formed around these codes.

21 538–710 C.E

174 taught to bureaucratic officials as mandated by the ritsuryō codes22 – argued for the subordination of women to men as the natural order of society.

The concept of the “Three Obediences”23 of women is seen in its oldest form in Bu Shang’s commentary on the Book of Etiquette and Ceremonial24, dating to the fifth century BCE:

For women, there is the principle of the Three Obediences, rather than one specific path. As

a maiden, a woman obeys her father; as a wife, her husband; as a widow, her son. The

father is as heaven to the child, and the husband is as heaven to the wife.25 (Yili 11)

A similar passage is contained in Dai the Greater’s Book of Rites26, an edited copy of the Confucian classic Book of Rites attributed to around the first century BCE:

A woman is called fùrén27, because she is a person (rén28) who submits (fú29). A woman thus

has no autonomy, but rather follows the path of the Three Obediences: while at home, she

22 Sekiguchi 2003, 29.

23 三從, sanjū/sāncóng

24 儀禮, Yílǐ

25 “婦人有三從之義，無專用之道，故未嫁從父，旣嫁從夫，夫死從子。故父者子之天也，夫者妻之天

也。” (“Fùrén yǒu sāncóng zhī yì, wú zhuānyòng zhī dào, gù wèi jià cóng fù, jì jià cóng fū, fū sǐ cóng zǐ. Gù fù zhě zǐ zhī tiān yě, fū zhě qī zhī tiān yě.”)

26 大戴禮記, Dà Dài Lǐjì

27 婦人

28 人

29 伏

175

obeys her father; as a married woman, she obeys her husband; as a widow, she obeys her

son; at no point is she self-contained.30 (Da Dai Liji, Benming)

The Lessons for Women31, written in the first century CE by the female intellectual Ban Zhao (班昭) for her daughters, contains similar admonitions for women to be submissive to men:

Of old, three days following the birth of a girl, she would be placed below the bed, and given

a potsherd to play with, and an offering would be made to the ancestors. She was placed

below the bed so that she may be aware of her own lowliness and weakness, and know to

humble herself before others. She was given a potsherd to play with so that she will know to

make a habit of toil, and view industry as her primary duty…These three customs cover the

proper behaviour of a woman and the teachings of correct etiquette. A woman should be

modestly deferential and respectful, and should put others before herself. She should never

announce her good deeds, nor should she deny her flaws. She should turn the other cheek,

and always behave as fearful and meek; this is what it means to humble herself before

others.32 (Nü Jie 1)

30 “故謂之婦人。婦人，伏於人也。是故無專制之義，有三從之道。在家從父，適人從夫，夫死從子，

無所敢自遂也。” (“Gù wèi zhī fùrén. Fùrén, fú yú rén yě. Shìgù wú zhuānzhì zhī yì, yǒu sāncóng zhī dào. Zài jiā cóng fù, shìrén cóng fū, fū sǐ cóng zǐ, wú suǒ gǎn zìsuì yě.”)

31 女誡, Nǚ Jiè

32 “古者生女三日，臥之牀下，弄之瓦塼，而齋告焉。臥之牀下，明其卑弱，主下人也。弄之瓦塼，明

其習勞，主執勤也。…三者蓋女人之常道，禮法之典教矣。謙讓恭敬，先人後己，有善莫名，有惡莫辭，

忍辱含垢，常若畏懼，是謂卑弱下人也。” (“Gǔ zhě shēng nǚ sān rì, wò zhī chuáng xià, nòng zhī wǎ zhuān,

ér zhāi gào yān. Wò zhī chuáng xià, míng qí bēiruò, zhǔ xià rén yě. Nòng zhī wǎ zhuān, míng qí xí láo, zhǔ zhíqín yě…Sān zhě gài nǚrén zhī chángdào, lǐfǎ zhī diǎnjiào yǐ. Qiānràng gōngjìng, xiān rén hòu jǐ, yǒu shàn mò míng, yǒu è mò cí, rěn rǔ hán gòu, cháng ruò wèijù, shì wèi bēiruò xià rén yě.”)

176

Texts such as these propagated an ideology of female subordination and inferiority referred to as danson johi (男尊女卑): “respect the male, revile the female”33.

The influence of patriarchal family ideals may be seen in ritsuryō legal codes treating with marriage and inheritance laws. Early texts and censuses show a variety of marital structures, including virilocal marriages (in which the wife came to live with the husband), uxorilocal marriages

(in which the husband came to live with the wife), duolocal/tsumadoi34marriages (in which husband and wife resided separately, with the husband making conjugal visits), and even neolocal marriages

(in which husband and wife established their own household separate from either natal family)35.

The Taihō and Yōrō Codes of the eighth century codified virilocal, yometori36 marriages as the standard37, an arrangement which centred the husband within the family unit and placed the wife at a disadvantage through separation from her natal kin. This was appended with a corpus of marital laws privileging husbands and placing wives in a subordinate position: adultery was punishable for wives but not husbands, divorce was easily attainable for husbands but not wives, a husband was permitted the use of (non-lethal) violence against his wife, and offences by a wife against her in-laws

33 This phrase (in Chinese, nánzūn nǚbēi) dates back to at least the fourth century CE, appearing within the

Daoist text Liezi (Liezi, Tianrui).

34 妻問い, “wife-visiting” (McCullough 1967, 105; Wakita 1984, 83).

35 Faure 2003, 171.

36 嫁取り, “bride-taking”; in which the wife becomes a member of her husband’s family, distinct from “virilocal” with its emphasis on kinship affiliation over locality. The reverse of this is mukotori 婿取り, “bridegroom- taking,” typically associated with uxorilocal marriage. The terms shōseikon (招婿婚, “groom-inviting marriage”) and boshokon (母処婚, “matrilocal marriage”) are also used within Japanese-language scholarship for marital location and kinship arrangements centring the wife’s household (McCullough 1967, 105; Wakita

1984, 83).

37 Faure 2003, 174; Ambros 2015, 24.

177 were held more severe than offences by a husband against his38. Inheritance was codified as strictly patrilineal39, undermining women’s ability to own independent property, making women more dependent on men for financial security, and rendering widows and divorcées financially vulnerable.

There was, however, a gulf in the adoption of these laws. Existing customs preferred a much looser system of marriages, favouring duolocal arrangements, allowing fluid transitions between marital types and easy dissolution of bonds, and independent female inheritance and ownership of wealth and property40. In the eighth century, a husband and wife living in the same village were considered divorced after a scant three months with no contact41; legal distinctions were not made between married and unmarried women in Japan, as they were in China42; and women of means managed property independently regardless of marital status43. Although census records were based around the patriarchal, virilocal household, or ko (戸), it appears that these were mostly something of a legal fiction, created to harmonise the taxation system of the ritsuryō code – built around the assumption of a Tang-style patriarchal family – with the fluidity of actual marital custom44. For some time a considerable discrepancy existed between de jure dictates of ritsuryō marital law, and the de

38 Ambros 2015, 24.

39 Umemura 1987, 26; Faure 2003, 174; Ambros 2015, 24.

40 Wakita 1984, 81; Umemura 1987, 25–6; Yoshie 1993, 441–6; Faure 2003, 172; Sekiguchi 2003, 36–9; Ambros

2015, 56.

41 Yoshie 1993, 442–3.

42 Yoshie 1993, 443–4; Yoshie, Ijuin, and Piggott 2013, 362.

43 Yoshie 1993, 455.

44 Yoshie 1993, 445.

178 facto practice of actual marriages, observable even amongst the elite45; the position of women not so far fallen as a study of legal codes alone would suggest.

The mid-ninth century, however, did see an increasing social emphasis on the stable conjugal unit, the relationship between husband and wife beginning to take priority over that of wife and natal clan46. Records across the Heian period47 display a gradual transition from the “old” marital and inheritance customs to the “new,” with virilocal marriages becoming increasingly common and ultimately the standard, uxorilocal marriages apparently fading out of practice entirely, and a significant decrease in female property ownership. The gulf that, in the early ritsuryō state, existed between legislation and practice began to narrow, culminating in a thoroughly and unambiguously patriarchal model of the Japanese family emerging at the dawn of the Kamakura48 period49.

In Tatsumi’s early “golden age” work, this coming of the patriarchal family system is attributed to the Yamato adoption of continental patriarchal belief which, diffused across the kingdom through legislation, overturned a previous culture of female importance and empowerment50. This narrative is continued by Takamure, who contrasts the patriarchal family systems of ancient China with the image of female empowerment in her reckoning of the “golden age”, and asserts the role of Buddhism and Confucianism in introducing Chinese patriarchy to the archipelago51, a view supported later by Ōi and Miyagi52. Piggott cites a similar process as

45 Umemura 1987, 26; Yoshimura 1993, 441–6; Faure 2003, 170–4; Sekiguchi 2003, 33, 40; Ambros 2015, 24–5,

58–9.

46 Yoshie 1993, 442–3.

47 794–1195 CE

48 1185–1333 CE

49 Faure 2003, 172–4; Sekiguchi 2003, 40; Ambros 2015, 58–9, 77–8.

50 Tatsumi 1887, 35.

51 Takamure 1966b, 219–20.

179 responsible for the delegitimisation of female rule: saturation of Chinese texts and legal codes – with implicit patriarchal ideology – into the Yamato elite, followed by their propagation on the national stage through a program of centralisation53.

Where the early “gulf” is reckoned with, it is often framed in terms of “native,” empowered traditions resisting the encroachment of “imported” patriarchy. Sekiguchi positions this as a two- way relationship, the discrepancy between ritsuryō law and actual practice undermining court fervour for Chinese ideology54. Maeda also positions patriarchal belief as a Chinese import, but situates its adoption much earlier on the timeline – around Himiko’s reign – but held at bay by the respect still afforded to female ritualists, and the importance afforded to their labour by the agricultural society of Yayoi and Kofun55.

In other narratives, the power of female ritualists is shown as undermined by continental belief structures: Hori, in his study of shamanism in Japan, cites a change from ancient matrilineality to Chinese-influenced patrilineality as leading to a significant reduction in the social role and status of the miko56. Tanaka similarly holds the centralised Yamato state, and its foundation in legal codes imported from China, responsible for stripping the hime of her power and central role57; calling to mind a similar process in the later Ryukyu Kingdom, where the influence of Confucian politicians in the 17th century contributed to reforms undermining the status of the national hierarchy of

52 Ōi and Miyagi 1974, 30–2.

53 Piggott 2003, 50.

54 Sekiguchi 2003, 40.

55 Maeda 2008, 19–20.

56 Hori 1968, 187.

57 Tanaka 1996, 73.

180 priestesses58. Thus, the “golden age” is ended by the Yamato adoption of patriarchal beliefs from the continent.

Certainly, the early Yamato elite display an adoption of Chinese thought – in particular,

Buddhism and Confucianism – by the dawn of the seventh century. The “Cap and Rank System” 59 created in 603 CE, the first in a series of measures aimed at establishing a clear hierarchy within the nobility, derived the names of its ranks from six of the traditional Confucian virtues60, firmly grounding it in Confucian ideology (Nihon Shoki 22.Suiko 11.12.Jinshin). Scarcely four months later, in 604, Crown Prince Shōtoku introduced the Seventeen-Article Constitution, a document which explicitly promoted Buddhism (Article 2) but was also heavily saturated in Confucian morality, promoting precepts such as obedience to fathers and authority figures (Articles 1 and 3), propriety

(Article 4), sincerity (Article 9), and meritocracy (Article 11) (Nihon Shoki 22.Suiko 12.4.Boshin). The

Taika Reforms of the mid-seventh century saw the formation of a bureaucratic administration based on both Tang Chinese statecraft and Confucian principles of hierarchy and meritorious conduct61; later reforms, such as the Taihō and Yōrō Codes of the early eighth century, were similarly based on

Chinese principles62. Old ways were denounced as wicked by a reform process that saw itself as a

“civilising” force63.

58 Lebra 1966, 103, 119; Sered 1999, 91.

59 冠位, kan’i.

60 Namely, virtue (德; toku/dé), benevolence (仁; jin/rén) propriety (禮; rei/lĭ), sincerity (信; shin/xìn), justice

(義; gi/yì), and knowledge (智; chi/zhì); all in “greater” and “lesser” form.

61 Sekiguchi 2003, 28–9; Yoshikawa 2011, 78–9.

62 Sekiguchi 2003, 30; Ambros 2015, 23.

63 Yoshikawa 2011, 80.

181

There is some argument that the more explicitly sexist teachings of Confucianism, such as the Three Obediences, remained largely unknown in Japan until much later64. Indeed, another “gulf” appears in this regard; they did not truly flourish in the popular discourse until the mediaeval period, where they became heavily entangled with Buddhist teachings65. Nevertheless, the writings of

Yamato elites show an awareness of these concepts much earlier. A commentary on the Queen

Śrīmālā Sutra ostensibly made by Crown Prince Shōtoku himself mentions the Three Obediences, although Ambros notes that this “commentary” may have been simply a copy of a Chinese original66.

The concepts do, however, appear in writing of the Nara period67: a poem by Yamanoue no Okura

(山上憶良) in the eighth century Man’yōshū68 compilation directly references not only the Three

Obediences, but also the “Four Virtues” of womanhood69 espoused by Confucian texts such as the

Rites of Zhou70 and the Lessons for Women (Man’yōshū 5.794). While not yet taken up by the broader populace, there was a clear awareness of patriarchal ideology among the Yamato court intelligentsia.

But was this Confucian belief truly the wellspring of all Japanese patriarchy, as the “golden age” narrative would imply? How well does this narrative correspond to our own investigations, over the last two chapters, of gender in the pre-ritsuryō age? We have seen that the image of female empowerment presented by the “golden age” narrative is an oversimplification of a complex and

64 Meeks 2010, 4–5; Ambros 2015, 25, 46–7, 53.

65 Faure 2003, 63, Meeks 2010, 4–5; Ambros 2015, 25, 53.

66 Ambros 2015, 46–7.

67 710–784 CE

68 万葉集; “Collection of Ten Thousand Leaves”

69 四德, shitoku/side; those being Feminine Ethics (婦德, fùdé), Feminine Speech (婦言, fùyán), Feminine

Countenance (婦容, fùróng) and Feminine Works (婦功, fùgōng).

70 周禮, Zhōu Lǐ

182 variable network of relationships between women and power across the ancient archipelago. We have noted in the previous chapter that – the remarkable figures of Himiko and Iyo/Toyo aside – the elite of “Yamatai” appeared to show some level of preference for male rulership, the two shaman- queens bookended by men and with a failed enthronement of a male high king occurring between them71. Wa contact with China, at this time, appears sporadic, confined to a few scattered records of diplomatic missions; could Confucian thought really have so rapidly insinuated itself amongst the elite? Or is it more likely that this patriarchal strand, this preference for male rulership, emerged from within Yamato culture?

A similar preference for male rulership is evidenced throughout the Yamato dynastic line, whose own histories also show an insistence on a strict male line following Amaterasu, and whose centralisation processes (see Ch. 5.1.1) connected Yamato dynastic expansion to the spread of sole- male rulership structures. The mythology of the Yamato imperial clan also promotes wifely submission to husbands in the name of gender balance and complementarity, through the marriage of Izanagi and Izanami (Kojiki 1; Nihon Shoki 1.4); although the marital customs of Yamatai are less clear, they were recorded as polygynous (potentially centring men within the marital unit) and dubbed virtuous by the standards of Chinese observers (Sanguozhi, Wei Shu 30, Worenzhuan). The apportion of blame for sexist attitudes in ancient Japan purely to Chinese influence disregards these factors – the culpability of native, pre-Confucian belief – slipping instead into a nationalistic dichotomy of “foreign/bad/oppressive” against “native Japanese/good/empowering”, a nativist- feminist fantasy of return to tradition72.

Let us recall the diversity of gender relations and gendered power structures native to the archipelago observed in the previous two chapters. We see images of female rulership mixed in with scripts of male dominance, regionally variable roles of gender in ritual performance, complex

71 Piggott 1994, 7; Kidder 2007, 132; Ambros 2015, 19–20.

72 Ryang 1998, 19, 23.

183 interweavings of authority, autonomy, and liminality. This variety and nuance challenges the “golden age” narrative framing of a singular “ancient Japan,” a singular relationship between gender, authority, and spirituality in the ancient archipelago. To cast a neat division between an earlier state of native empowerment and a later, Chinese-influenced patriarchy is to present a homogenised image of gender in a singular pre-Confucian “ancient Japan,” an image which fails to account for the tensions of power and variety of custom apparent in the ancient archipelago.

This is not to say that continental belief did not have an impact on ideologies of gender. We have seen how the influence of Confucianism and Tang legal codes on the Yamato ruling classes led to the creation and dissemination of legislature – the ritsuryō codes – which codified and normalised patriarchal social structures73. At the same time, however, the preference for male leaders evinced in early Yamato, as well as the androcentric social order contained within the liminality of the miko

(as evinced in Ch. 3.2), suggests that at least a kernel of these patriarchal values were already present within the Yamato elite. Later works, such as by Ambros, have sought to complicate this narrative of “decline,” acknowledging and tracking the changes in ancient gender roles and social position of women, while eschewing the nativist fantasy of a purely-empowered “golden age” lost to the ravages of foreign belief systems74.

Patriarchy is, as we have discussed, not a matter of total empowerment versus total subjugation, but a complex network of power relations expressed in a variety of ways75; quantifying patriarchy thus becomes more than a simple question of “presence” versus “absence”. Within the

“golden age” narrative, arguments such as Takamure’s76 or Maeda’s77 position the influence of

73 Umemura 1987, 25–6; Faure 2003, 172; Sekiguchi 2003, 36–9; Ambros 2015, 56.

74 Ambros 2015, 2.

75 Faure 2003, 6, 170.

76 Takamure 1966b, 219–20.

77 Maeda 2008, 19–20.

184 continental ideology as the imposition (“presence”) of patriarchy onto a state of previous “absence,” but this casts patriarchy as an absolute, a concrete property either fully there or fully not. What we see, however, when we delve into records and practices of Yamato before the ritsuryō state is something more complicated; social orders which may perhaps be dubbed quasi-patriarchal. The troubled accession of Himiko and the gender negotiations within Jingū’s tale suggests a preference for male rule among the Yamato elite as far back as the early Kofun period, but with female rule nevertheless permitted. Reports from Chinese emissaries and the male-line genealogical traditions that informed the Kojiki and Nihon Shoki centre men within marital and family units, but eight- century de facto practices show that women nevertheless maintained a certain degree of independence, even autonomy, in their marriages. The religious tradition of the Kofun miko and saniwa present a model of gender complementarity informed by androcentric ideals. In one sense, a shift may be tracked, Tang legal codes and Chinese patriarchal norms influencing Yamato court culture towards a more rigidly hierarchical gender ideology emphasising the male-dominated family unit. But this is a more complex process than the simple transition from “empowerment” to

“patriarchy” tracked by the “golden age” narrative.

5.2. Priestesses in the “Silver Age”

5.2.1. Ritsuryō and Religious Centralisation

The religious sphere was not spared the impact of Yamato centralisation. The same ritsuryō codes that mandated patriarchal family systems and female disenfranchisement also promoted the establishment of a centralised, court-controlled religious bureaucracy: the Jingikan, or Department of Divinity (神祇官). Established under the Yōrō Code of 718, the Jingikan, partly inspired by the

185

Tang Chinese Ministry of Rites (禮部, Lǐbù)78, formed a centralised administrative hierarchy overseeing shrines, priests, miko, diviners, and other ritualists79. Under Jingikan auspices, religious performance was corralled into the service of the central hegemony, a state cult with the Yamato emperor as its chief ritual coordinator and his ancestor-gods (and the divine mandate they bestowed him) at the head of the pantheon80. Even provincial shrines fell under Jingikan dominion, central priests sometimes sent to staff them to further reinforce central government control over the religious life of the periphery81.

In Jingikan hierarchy and procedure, official status and ritual importance was concentrated on the central (male) role of liturgist, whereas liminal (female) shamans and mediums often found themselves lacking in rank compared to their male counterparts, their ritual roles diminished, or else risking legal suppression operating outside the bounds of the state cult82. Within the “golden age” narrative, the lofty position of the priestess is emblematic of the empowerment of ancient women; in some tellings, the sacral is designated as her domain. The marginalisation of priestesses under

Jingikan control – the apparent deterioration of the miko from powerful shamaness to shrine attendant – thus becomes symbolic of the loss of the “golden age,” the Jingikan priestess a vestige of women’s ancient sacral power, in whose remaining functions a glimpse of the ancient shamaness’s glories may be read. But how well does this narrative serve to represent the lived experiences of these ancient women? Let us examine some female sacral roles – both within and without the state cult – and their relationship to sacral and temporal power.

78 Naumann 2000, 47.

79 Martin 1997, 39–40.

80 Anzu 1986, 16; Piggott 1997, 144–5, 208–10; Yoshie 2007, 172.

81 Ooms 2009, 114.

82 Kuratsuka 1996, 17; Meeks 2011, 213.

186

5.2.2. Priestesses and Power within Hegemonic Systems

The most basic female role within the Jingikan was that of jinja-miko83, or “shrine priestess,” attached to a specific shrine under the auspices of the central religious bureaucracy. Although a derivation of the miko tradition, within the context of the state cult, the role of the jinja-miko appears a far cry from the power and mystique of the ancient shamaness. Her ritual dances – originally an inducement to divine possession, or onset of ecstatic trance-state – became increasingly ceremonial, losing their mantic function and transforming into mere entertainment for court and god84. The decline in the state miko’s shamanic nature led to a significant reduction in her ritual importance, her role often relegated to little more than a shrine attendant, housewife of the gods85. Furthermore, she lacked the official bureaucratic ranking of her male priestly counterparts86.

In this way, the average jinja-miko was divested not only of (official) temporal authority, but also of sacral power.

Within the “golden age” narrative these two forms of power are frequently homogenised into a single, amorphous concept of “women’s power,” manifesting in the ancient glory days but stripped away by the ravages of patriarchy and the increasing sidelining of female religious practitioners; a clear, linear decline, a straight downwards fall from power to patriarchy. Earlier (see

Ch. 3.2.1), we complicated the notion of “women’s power” as pertaining to the ancient miko, examining their role in light of various types of power: sacral power, temporal authority, personal autonomy, and a nebulous “freedom from patriarchy”. Let us recall them again as we make a study

83 神社巫女 (Hori 1968, 182; Kuratsuka 1996, 14); also called kanmiko (官巫女, “bureau priestess”) (Yoshie

2007, 181)

84 Blacker 1982, 30; Kuratsuka 1996, 13–5; Meeks 2011, 220.

85 Faure 2003, 302; Meeks 2011, 214; Ambros 2015, 93–4.

86 Kuratsuka 1996, 17; Meeks 2011, 213.

187 of the palace kannagi – the most central of the Jingikan priestesses – and the complications of their own relationship to power.

Six palace kannagi were employed by the Department of the Divinity in protection of the imperial family: four attending to rituals of the imperial palace, and one each for the palaces of the empress and crown prince (Engishiki 3.Jingi 3). Of the four imperial palace kannagi, the mikannagi or

ōmikannagi87 was entrusted with the worship of a suite of deities charged with protecting the imperial person and the emperor’s daily life, including his speech, his meals, and his priestesses. The mikado-no-kannagi88 made worship to the gate-gods of the imperial palace, and the ikasuri-no- kannagi89 to the gods of the land on which it sat, both their efforts ensuring its safety and stability.

Finally, the Ikushima-no-kannagi 90 worshipped the protective gods of the realm as a whole91

(Engishiki 9.Jingi 9.Norito.Toshigoi-no-Matsuri).

These kannagi could be no mere shrine attendants; their rituals, and the divine protection they secured, were too crucial to the interests of the imperial hegemony. Their spiritual power must be considerable, and indeed, this manifests as one of their selection criteria: the mikannagi, mikado- no-kannagi, and Ikushima-no-kannagi, along with the kannagi for the empress’s and crown prince’s palaces, were all to be selected from the ranks of common women displaying particular spiritual acuity (Engishiki 3.Jingi 3)92. The requirement of sacral power suggests the palace kannagi as an

87 御巫/大御巫; “honoured medium/great medium.”

88 御門巫; “medium of the [imperial] gates.”

89 座摩巫; “medium of residence.”

90 生嶋巫; “medium of [the gods of] Ikushima.”

91 Philippi 1990, 91; Naumann 2000, 57; Yoshie 2007, 159–60, 162–3.

92 “凡御巫。御門巫。生嶋巫各一人。〈其中宮。東宮唯有御巫各一人。〉取庶女堪事充之。但考選准

散事宮人。” (“Subete no mikannagi. Mikado-no-kannagi. Ikushima-no-kannagi onoono hitori. (Sono chūgū.

188 active participant in ritual; no passive sacerdotal placeholder, she. The ikasuri-no-kannagi had slightly different selection criteria, focusing more on her kinship ties than her spiritual power: she must be a young, unmarried daughter of the noble Tsuge-no-Kuni-no-Miyatsuko93 (都下国造) clan, no fewer than seven years of age94. Although young, the minimum age requirement for this kannagi suggests an active ritual role; the priestess had duties requiring a certain level of physical capability.

The more passive noble-priestess position of the saigū had no such minimum95, as we shall see below.

So we may see that the palace kannagi were invested with sacral power. In our analysis of power forms of the ancient miko (see Ch. 3.2.1), we observed that sacral power could be conditionally leveraged to access temporal authority. For the palace kannagi, however, that access does not manifest within the bureaucracy. Whereas male Jingikan priests were accorded bureaucratic rankings, giving them official government position, the same was not true of the palace kannagi96 who, even as the most central miko within the Jingikan hierarchy, received only the minor, non-administrative rank of unassigned palace staff (Engishiki 3.Jingi 3). Within the context of the state cult and bureaucratic hierarchy, the kannagi was marginalised outside of her ritual context, her sacral power subordinated to the needs of the imperial state and its male-dominated institutions97.

Tōgū tada mikannagi onoono hitori ari.) Shojo koto sugurete torite mitasu. Tadashi kōsen sanji kyūjin toshite junsu.”)

93 Headquartered in Tsuge, in the southeast of modern Nara City.

94 “凡座摩巫。取都下國造氏童女七歲已上者充之。若及嫁時。申辨官充替。” (“Subete no Ikasuri-no- mikannagi. Tsuge-no-Kuni-no-Miyatsuko-no-uji no dōjo shichisai ijō ni torite mitasu. Moshi totsugu toki ni oyobite. Bekan mitashikaeri wo mōsu.”)

95 Ellwood 1967, 43; Bock 1970a, 151.

96 Yoshie 1987 35–6.

97 Yoshie 2007, 175.

189

What we may note, however, is that this appropriation of female sacral activity for the benefit of masculine imperial power is not an entirely new phenomenon. Recall the Yamato myth- history of Empress Jingū (see Ch. 4.2): it was noted that, despite Jingū’s considerable spiritual power and military skill, all these talents are ultimately deployed, within the narrative, for the benefit of the imperial patriline, the dominion of her husband and son. Recall also the partnership of king and princess-priestess (see Ch. 4.3.4): the assistance provided to Sujin by Yamato-Totohi-Momoso-Hime, in the form of oracles and omen interpretations; to Sujin again by Toyosuki-Irihime and Nunaki-

Irihime in their worship of the royal gods, who are brought out from the palace by the girls so that

Sujin need not be crowded by their power (Nihon Shoki 5.Sujin 6); to the conquering prince Yamato-

Takeru by his priestess-aunt Yamato-Hime, delivering unto him the tools which she has predicted will lead him to victory (Kojiki 2.Keikōki; Nihon Shoki 7.Keikō 40.10.Bogo).

In the “golden age” narrative, these women appear as active, powerful figures, standing in stark contrast to the marginalisation of the jinja-miko. Yet their sacral performance is still ultimately deployed for the benefit of the male Yamato paramount, even if they appear to leverage a greater degree of official authority from the transaction98. The relationship between throne and kannagi appears not as a complete turnaround from a previous age of sacral glory, but rather as an amplification of a pre-existing Yamato programme of appropriating female sacral power for the benefit of male kingship, codified by a male-centric bureaucratic ranking system which afforded little status to the ephemeral influence of the priestess-advisor. The mostly-common backgrounds of the kannagi only made their subordination easier, fitting in with a pre-existing hierarchical bloodline

98 Additionally, each of these prominent shamanesses is of royal blood, holding ranks of princess or empress even outside the context of sacral performance, their access to temporal authority at least partially enabled by birthright. This begs the question: how much power could a common woman leverage through sacral performance, even in the ancient period? The role of class in sacral women’s access to power may also prove fruitful ground for study.

190 structure. Although an increasing culture of female subordination and the homogenisation of regional religious practices through the state cult contributed to this environment, it was not the total inversion of previous Yamato norms presented by the “golden age” narrative.

5.2.3. Priestesses and Power: Beyond the Bureaucracy

But the state cult of the Jingikan was not the entirety of ancient Japanese religiosity, nor the bureaucracy the entirety of society. Did the miko’s lack of access to bureaucratic rank, under ritsuryō, mean her total disempowerment? And what, for that matter, of the miko performing outside the auspices of the Jingikan: the sato-miko and the aruki-miko? To answer these questions, let us look into the miko’s performance and power outside of the state cult.

Not all miko fell under the purview of the Jingikan. The sato-miko, or “village priestess”99, was attached to a village shrine unaffiliated with the state cult; the aruki-miko, or “walking priestess”100 was not attached to a particular shrine at all, an itinerant priestess who made her living from private, freelance ritual services101. Unlike the increasingly domesticated role of the Jingikan miko, these “unofficial” priestesses retained more of the miko’s shamanic and mantic functions102; indeed, mediumship and divination served as the primary livelihood for aruki-miko103. The narrative of the “golden age” depicts a clean, linear decline of the miko from powerful shamaness to passive state priestess or criminalised itinerant104. The sato-miko and aruki-miko enter into focus mainly in the context of state suppression: official measures aimed at curbing religious performance outside

99 里巫女 (Hori 1968, 182; Yoshie 2007, 181)

100 歩き巫女 (Hori 1968, 202; Meeks 2011, 212)

101 Meeks 2011, 223.

102 Hori 1968, 182–3; Kuratsuka 1996, 14; Faure 2003, 302; Meeks 2011, 213.

103 Meeks 2011, 224.

104 Meeks 2011, 215.

191 the bounds of Jingikan authority, rendering miko both inside and outside the state cult marginal figures stripped of their power, a “power” which is equated with bureaucratic rank.

The state did make several attempts to curb the activities of miko and other shamans outside of the state cult, concerned about their potential to influence the populace. The state cult existed to centralise religion in service to the needs of court and emperor; spiritual practitioners not under its control constituted a threat, a potential source of opposition105. In 752, 17 shamans from the capital were rounded up and banished to the distant outer provinces (Shoku Nihongi 18.Tenpyō-

Shōhō 4.8.Kōin)106; a severe punishment ranking just below execution in the ritsuryō penal code. In

795, Kamitsukeno no Ekunime (上毛野兄国女), who called herself “Shoten” (諸天), was herself exiled from the capital to far Tosa Province (modern Kōchi Prefecture) for “beguiling the populace with her prophecies” (Nihon Kōki 3.Enryaku 14.5.Kishi)107. The following year, Ikue no Iemichime (生

江家道女), the “Lay Devotee of Koshi,” was forcibly returned to her home in Echizen Province

(modern northern Fukui Prefecture) for “talking irresponsibly about sin and blessing and confusing the populace”; in other words, unsanctioned preaching (Nihon Kōki 5.Enryaku 15.7.Shingai)108.

Two points of interest arise among these examples. The first is that the 752 banishment targeted not only female miko but also male shamans, as indicated through the use of the character

105 Kuratsuka 1996, 14; Faure 2003, 302–3; Yoshie 2007, 171, 181; Naoki 2009, 232; Ambros 2015, 62–3, 93–4.

106 “捉京師巫覡十七人。配于伊豆。隱伎。土左䒭遠國。” (“Keishi no kannagi tō-amari-nanatari wo toraete. Izu. Oki. Tosa-ra no tōtsu kuni ni haisu.”)

107 “右京人上毛野兄國女流土佐國以自稱諸天妖言惑衆也。” (“Ukyō no hito Kamitsukeno no Ekunime mizukara Shoten to shōshite yōgen ni shū wo madou koto wo mochite Tosa-no-Kuni ni rusu.”)

108 “生江臣家道女遞送於本國。家道女。越前國足羽郡人。常於市。妄説罪福。眩惑百姓。世號曰越

優婆夷。” (“Ikue-no-Omi Iemichime wo moto no kuni ni teisō su. Iemichime wa. Echizen-no-Kuni Asuwa-no-

Kōri no hito. Tsune ni ichitana ni. Midari ni zaifuku wo toki. Hyakusei wo genwaku su. Yo wa gō shite Koshi-no-

Ubai to iu.”)

192

覡109 (see Ch. 3.2.1). In other words, it was not only the gendered role of miko that was targeted by state suppression, but the unsanctioned practice of shamanism as a whole. The second point is

Iemichime’s title of “Laywoman”: ubai (優婆夷), a Japanese transliteration of the Sanskrit upāsikā, a female lay devotee of Buddhism, suggesting that her unlicensed religious activity was primarily grounded in Buddhism. Similar state sanctions were placed on Buddhist activity and practitioners: monks and nuns without official certification, transmitting unsanctioned teachings, and/or selling spells and divinations as a side business (Shoku Nihongi 7.Yōrō 1.4.Jinshin; Shoku Nihongi 10.Tenpyō

1.4.Kigai). The state desire for religious monopoly was by no means a specific patriarchal marginalisation of the ancient female power of the miko, but rather a sweeping suppression of all prominent religious figures and beliefs outside of state control in the pursuit of religious centralisation, with the unlicensed miko as just one of its casualties.

At the same time, however, the miko was not a purely marginal figure, even outside the bounds of the Jingikan. Although state suppression occurred, it was not constant and overarching; the state’s position on unsanctioned shamans was flexible110, as suited its needs, and official sources made occasional distinctions between “true” and “false” folk mediums, only the latter recommended for suppression111. Not all elite opinions towards miko were negative, either; some aristocrats clearly evinced belief in the powers of their oracles, and held benign attitudes towards the priestesses; elite condemnation of miko as “deceivers” was often connected to the specific motive of state religious monopoly112. A gulf also appears, in aristocratic writings on the miko, between elite and folk customs and religious practices, many traditional rituals viewed by educated,

109 Meeks 2011, 209.

110 Kuratsuka 1996, 14; Yoshie 2007, 171.

111 Yoshie 2007, 171.

112 Meeks 2011, 215–7, 229.

193

“civilised” elites as unsophisticated holdovers of an unrefined past, an attitude that perhaps influenced the jinja-miko’s increasing focus on more respected arts such as kagura dance113.

Indeed, even miko outside the auspices of the Jingikan received occasional patronage from the elite114. In 805, for example, a shamaness from the the old capital of Heijō (Nara), credited with delivering an oracle, was invited to the palace to perform a Chinkonsai ceremony for the illness of

Emperor Kanmu115 (桓武) (Nihon Kōki 12.Enryaku 24.2.Kōjutsu)116. Pilgrimages were particularly common occasions of elite miko patronage117: Emperor Toba118 (鳥羽) was said to have consulted a medium for oracular advice on his pilgrimage to the Kumano shrine complex (Hōgen Monogatari 1).

Hyperfocus on the miko’s exclusion from official bureaucratic ranking makes the error of conflating one specific relationship to temporal authority – position in regards to the state administrative hierarchy – with the broader concept of “power” as a whole. Furthermore, discourse on the negative attitudes held by elite writers towards miko risks conflating elite opinion with broader social attitudes and realities119. We must remember the diversity present within ancient

Japan, including diversity of custom based on class and region. Research by Meeks has shown that,

113 Meeks 2011, 246.

114 Meeks 2011, 241.

115 R. 781–806 CE

116 “聞平城松井坊有新神託女巫。…召彼女巫。令鎭御魂。女巫通宵忿怒。託語如前。” (“Heijō no

Matsui-bō ni atarashiki kami arite onna-miko wo kakotsukeru koto wo kikite…Kano onna-miko wo meshite.

Mitama wo shizumerashimu. Onna-miko yoi tōshite fundo su. Takugo saki no gotoku.”)

117 Meeks 2011, 241. Also see Ruch (2002) for a similar analysis of the mendicant “Kumano bikuni” as representing a range of female religious practitioners with disparate activities and class backgrounds, conflated into a singular tradition in the scholarly imagination.

118 R. 1107–1123 CE

119 Meeks 2011, 216.

194 despite the poor opinion of miko expressed by certain elites and their lack of official bureaucratic ranking, miko on both state-sponsored and unofficial levels could hold considerable status at the community level120. Meeks uncovers a diversity of experiences within the category of “miko,” encompassing a multifaceted range of social positions from wandering, marginalised aruki-miko beggars to well-connected respectable, independently wealthy miko with substantial marriage prospects, calling into question the validity of “miko” as a single category of social status and function121.

Three key points arise from this examination of the miko’s status. The first is that the “silver age” miko did not function as a singular social category, but rather comprised an array of female practitioners with diverse experiences, social standings, relationships to state authority, and religious functions122. The second is that these qualities were shifting, the role of the miko a “fluid site” constantly renegotiated and redefined123, both by the miko herself and the underlying social tensions – between court and folk culture, state cult and folk religion – on whose faultlines she lay.

The third is that “authority,” in the ritsuryō state, was not a monolithic concept, represented entirely by bureaucratic ranking, but rather manifested in a variety of relationships; thus, a miko could lack official rank but still hold considerable social capital at the community level. “Power” – even in the specific form of temporal authority – is a heterogenous concept, and the marginalisation of miko by the state cult far from monolithic.

120 Meeks 2011; 216–31.

121 Meeks 2011, 224–5, 231. Meeks’ work recalls a similar conclusion by Ruch in her studies of the Kumano bikuni of the Edo period, in which she concludes that, rather than a single itinerant position, the term “Kumano bikuni” encompassed a wide range of female practitioners of varying statuses and religious activities (Ruch

2002).

122 Meeks 2011, 224–5, 260.

123 Meeks 2011, 217.

195

5.3. Women, Religion, and Imperial Dominion: Tensions of Power within the Ise Saigū and Kamo Saiin

We have thus observed how the miko under the ritsuryō system existed not in a state of utter decline and complete marginalisation, as the “golden age” narrative framing presents it, but rather a complex web of tensions between “official” status and “unofficial” social capital, between elite ideology and folk custom. But what of the most prominent priestesses of the ritsuryō, those who held core roles within the state cult and the Yamato process of centralisation yet existed outside of the Jingikan structure entirely: the royal Ise saigū and Kamo saiin?

The saigū and, to a lesser extent, the saiin, occupy key positions within the “golden age” narrative, where they appear as decayed remnants of the ancient power of women: former hime stripped of their co-rulership function and carted off to distant Ise, ur-examples of the miko’s decline, their existence a tantalising hint to the glorious past of female spirituality while at the same time exemplifying its degradation under patriarchy124. How, then, does this narrative fare in light of a careful examination of their role, and the tensions of power within? We have seen that there was far more to the ritsuryō miko than the narrative of pure decline held space for; does the same hold true for the Sacral Princesses of the Yamato throne?

5.3.1. The Ise Saigū and Kamo Saiin: Historical Background

To understand the Ise saigū and Kamo saiin, let us first situate them within their historical context. The Yamato imperial clan took as its titulary deity the sun goddess Amaterasu, their brilliant, powerful ancestress who endowed her descendants with the divine mandate of kingship. But the traditional cultic centre of Amaterasu’s worship was (and still is) not headquartered in Yamato at all,

124 Takamure 1966b, 96, 219; Yoshie 1987, 34; Tanaka 1996, 73; Matsumura 1999, 128–31.

196 but rather in the easterly Ise Province (modern Mie Prefecture), at the Grand Shrine of Ise

(specifically, at the Inner Shrine). Hence, the establishment of the Ise saigū or itsuki-no-miya125, also called the Sacral Princess126, the most prominent role of female religiosity within the imperial family during the ancient period: a virgin daughter of the Yamato emperor, sent to distant Ise to worship the clan’s divine ancestress and secure her blessings upon her descendants, living a life of strict seclusion to maintain her ritual purity.

The antiquity of the saigū role is rather uncertain, tied into the myth-historic canon of the eighth-century Yamato court. The mythic prototype and legendary founder of the saigū role is

Yamato-Hime127, daughter of Emperor Suinin, who is also credited with the founding of the Inner

Shrine of Ise in a chronicle officially (and unreliably) dated to 5 BCE (Nihon Shoki 6.Suinin 25.3.Teigai).

Toyosuki-Irihime, daughter of the previous emperor Sujin, also appears in myth as a sort of proto- saigū, a princess charged with the worship of Amaterasu while the goddess was still in Yamato

(Nihon Shoki 5.Sujin 6). The specifics of these tales – and their role within the Yamato dynasty’s campaign of religious centralisation – will be discussed in more detail later, as we examine the relationship between the saigū and the Yamato program of political and religious centralisation.

Following these two progenitor-figures, evidence for early saigū is little and sporadic128.

Princess Iono (五百野), daughter of Emperor Keikō, is “sent to worship” Amaterasu in an article attributed to 90 CE (Nihon Shoki 7.Keikō 20.2.Kōshin)129, but even outside the unreliable datings, this

125 斎宮, “sacral/abstention palace”; this term can refer to either the priestess herself or her residence at Ise.

126 斎王, saiō or 斎内親王, itsuki-naishinnō.

127 Traditional birth and death dates unclear; appears as early as 5 BCE and as late as 121 CE

128 Kuratsuka 1996, 55–6; Emura 2009, 5.

129 “遣五百野皇女令祭天照大神。” (“Iono-no-Himemiko wo tsukawashite Amaterasu-Ōkami wo matsurashimu.”)

197 is generally read as a later addition, meant to strengthen the antiquity of the saigū role130. Another princess, Iwashima (伊和志真), is attested to the role of saigū during Chūai’s reign in the Nisho

Daijingū Reibun131, an internal record of Ise Shrine from 1304, but she is absent from earlier imperial records, leaving her own historicity in doubt (Nisho Daijingū Reibun 15). Next is Takuhata-Hime (栲

幡娘姫), the ill-fated daughter of Emperor Yūryaku132 (雄略) (Nihon Shoki 14.Yūryaku 1.3)133, who is said to commit suicide in 459 after becoming the victim of cruel slander on the subject of her sacral chastity (Nihon Shoki 14.Yūryaku 3.4). Takuhata-Hime, whose genealogy (but not service) is further attested in the Kojiki (Kojiki 3.Yūryakuki), is a murky figure, her historicity supported by prominent saigū scholars Okada and Naoki, but disbelieved by others such as Emura134.

The first saigū attested as such in both Kojiki and Nihon Shoki is Princess Sasage (佐佐宜/荳

角), daughter of Emperor Keitai (Kojiki 3.Keitaiki; Nihon Shoki 17.Keitai 1.3.Kiyū)135, situating the saigū role with the early sixth century, with more regular appointments to follow136: Emperor

130 Kuratsuka 1996, 56; Emura 2009, 12.

131 二所大神宮例文; “An Illustration of the Two Great Shrines”

132 Reign traditionally dated 456–479 CE

133 “與稚足姬皇女。〈更名栲幡娘姬皇女。〉是皇女侍伊勢大神祠。” (“To Waka-Tarashi-Hime-no-

Himemiko to. (Mata no mina wa Takuhata-Iratsu-Hime-no-Himemiko.) Kono himemiko Ise no Ōnkami no iwai ni haberi.”)

134 Okada 1970, 345; Emura 2009, 13–5; Naoki 2009, 11, 76.

135 Kojiki: “次佐佐宜王者。拜伊㔟神宮也。” (“Tsugi ni Sasage-no-Miko wa. Ise no Kami-no-Miya wo itsuki matsuritamaiki.”); Nihon Shoki: “生荳角皇女。…是侍伊勢大神祠。” (“Sasage-no-Himemiko wo umu…Kore

Ise-no-Ōkami no iwai ni haberi.”)

136 Emura 2009, 19.

198

Kinmei’s daughter Princess Iwakuma137 (磐隈) and Emperor Bidatsu’s138 (敏達) daughter Princess

Uji139 (菟道), both dismissed from the post due to loss of virginity (Nihon Shoki 19.Kinmei 2.1; Nihon

Shoki 20.Bidatsu 7.3.Jinshin). Between them, the role was apparently fulfilled by a woman named

“Princess Miyako” (宮子), a daughter of the high priest presumably granted later honourary title, described in Kamakura records as a “staff substitute for the princess” (內親王御杖代, naishinnō no mitsue-shiro) (Nisho Daijingū Reibun 15; Toyouke Daijingū Negi Honin Shidai). The last of the early saigū, Yōmei’s140 (用明) daughter Princess Sukate-Hime (酢香手姫), is credited with serving 37 years before retiring, through the reigns of her father, Emperor Sushun141 (崇峻), and Empress Suiko

(Nihon Shoki 21.Yōmei Tennō Sokui Zenki Bidatsu 14.9.Jinshin)142; the later custom of saiō dismissal on parental death evidently not yet in practice143. She is mentioned in passing in the Kojiki under the

137 “其二曰磐隈皇女。〈更名夢皇女。〉初侍祀於伊勢大神。後坐姧皇子茨城解。” (“Sono futari wo

Iwakuma-no-Himemiko to iu. (Mata no mina wa Ime-no-Himemiko.) Hajimete Ise-no-Ōnkami no matsuri ni haberi. Ato wa miko Muharaki ni okasaretaru ni yori tokenu.”)

138 R. 572–585 CE

139 “以菟道皇女侍伊勢祠。即姧池邊皇子。事顯而解。” (“Uji-no-Himemiko wo motte Ise no iwai ni haberi.

Sunawachi Ikenobe-no-Miko ni okasarenu. Koto arawarete tokenu.”)

140 R. 585–587 CE

141 R. 587–592 CE

142 “以酢香手姬皇女拝伊勢神宮奉日神祀。〈是皇女自此天皇時逮于炊屋姬天皇之世。奉日神神祀。自

退葛城而薨。見炊屋姬天皇紀。或本云。三十七年間奉日神祀。自退而薨。〉” (“Sukate-Hime-no-

Himemiko wo motte Ise-no-Ōnkami-no-Miya ni meshite hi-no-kami no matsuri ni tsukamatsurashimu. (Kono himemiko kono sumera-mikoto no mitoki yori Kashikiya-Hime-no-Sumera-Mikoto no miyo ni oyobu made. Hi- no-kami no matsuri ni tsukamatsuru. Onozukara Kazuraki shirizokite miusemashinu. Kashikiya-Hime-no-

Sumera-Mikoto no mimaki ni mieru. Aru moto ni iwaku. Miso tose amari nana tose no aida ni hi-no-kami no matsuri ni tsukamatsurite. Onozukara shirizokite miusemashinu.)”)

143 Emura 2009, 17–8.

199 name Suga-Shiroko (須賀志呂古) (Kojiki 3.Yōmeiki), and in the Jōgū Shōtoku Hō’ō Teisetsu144 as

Sukateko (須加氐古) (Jōgū Shōtoku Hō’ō Teisetsu), several sources attesting to her existence and occupation of her position145. A hiatus appears to have ensued following Sukate-Hime’s retirement in 622, with no further references to saigū appointments.

In 673, upon his accession to the throne following the Jinshin War, Emperor Tenmu146 (天武) designated his daughter Ōku (大来) as his saigū, ending a 50-year vacancy in the position. Ōku spent a year and a half purifying herself in the Hatsuse Abstention Palace before she embarked upon her journey to Ise Shrine (Nihon Shoki 29.Tenmu 2.4.Kishi 3.10.Itsuyū)147. She held this position until 686, when the forced suicide of her brother Ōtsu (大津) on charges of treason necessitated her departure from the post (Nihon Shoki 30.Jitō Tennō Sokui Zenki Shuchō 1.10.Kōgo/11.Jinshi), the ritual pollution caused by the death of a close family member rendering her unsuitable for sacral duty. Tenmu’s reinstatement of the saigū system is often read as a gesture of gratitude towards Ise Shrine and

Amaterasu, who he credited for his success against his nephew in the Jinshin War; in more practical terms, it was also a method of maintaining control over the region, as we shall examine shortly148.

The oldest sections of wall excavated from the ruins of the old saigū residential complex (now in the town of Meiwa, Mie Prefecture) also date back to this time149. Ōku herself is well-attested,

144 上宮聖徳法王帝説; “Imperial Chronicles of Dharma King Jōgū Shōtoku”

145 Emura 2009, 17.

146 R. 673–686 CE

147 1: “欲遣侍大來皇女于天照大神宮。而令居泊瀨齋宮。是先潔身。稍近神之所也。” (“Ōku-no-

Himemiko wo Amaterasu-Ōmikami no miya ni haberashimemu to hosu. Shikashite Hatsuse no iwai no miya ni mashimu. Kore wa mazu mi wo sayamete. Yaya kami ni chikaki tokoro nari.”); 2: “大來皇女自泊瀨齋宮向伊勢

神宮。” (“Ōku-no-Himemiko Hatsuse no iwai no miya yori Ise-no-Kami-no-Miya ni mōtsu.”)

148 Emura 2009, 18–9; 2012, 74; Naoki 2009, 80; Ooms 2009, 58.

149 Emura 2009, 18–9.

200 appearing in the Man’yōshū poetry collection and several mokkan150. From Ōku onwards, the saigū gradually became a regular, systematised position, whose office and place within the state cult was clearly codified by the 10th century (Engishiki 5.Jingi 5).

Then what of the Kamo saiin? In 794, Emperor Kanmu, seeking to establish a base of power away from the influential Buddhist institutions of Nara151, moved the capital to Heian-kyō (modern

Kyoto) in Yamashiro Province, where it would stay for over a thousand years. This shift – which began the Heian period of Japanese history – altered the relationship between the imperial dynasty and the major shrines of Yamashiro. The Kamo Shrine complex – whose large festival crowds were previously viewed with suspicion by the Yamato authorities (Shoku Nihongi 1.Monmu 2.3.Shinshi;

Shoku Nihongi 2.Taihō 2.4.Kōshi; Shoku Nihongi 5.Wadō 4.4.Itsubi) – was reenvisioned as a vital site for securing the divine protection of the court152, transformed from a regional shrine with few

Yamato connections to a major cultic centre in close proximity to the palace. In Heian period records, it comes to sit firmly alongside Ise as a major shrine of royal patronage (Nihon Kōki 24.Kōnin

6.8.Shinchū)153.

During the reign of Emperor Saga154 (嵯峨), the post of Kamo saiin155 was created: a counterpart to the Ise saigū in the new heartland, with similar ritual duties and strictures. The first saiin was Saga’s daughter, Princess Uchiko (有智子). Uchiko’s appointment as the first saiin does not

150 Emura 2009, 18–9.

151 Martin 1997, 50.

152 Yoshie 1987, 34; Tanaka 1996, 67; Emura 2009 4, 111–2.

153 “遣使奉幣於伊勢大神宮并賀茂大神。以霖雨不晴也。” (“Tsukai wo tsukawashite mitegura wo Ise-no-

Ōkami-no-Miya narabi ni Kamo-no-Ōmiya ni tatematsurashimu. Rin’u harezaru koto wo motte nari.”)

154 R. 809–823 CE

155 斎院, “Sacral Institution”; like the saigū, this could refer to both the priestess and her complex, with the term saiō also used for the former.

201 appear in surviving or reconstructed passages from the Nihon Kōki, alkthough the Kamo Chūshin

Zakki156 dates this to 810 (Kamo Chūshin Zakki 4). The first mention of the role of “Sacral Princess of

Kamo” comes upon her retirement in 831, when she is replaced by Princess Tokiko (時子) due to her advancing age (Nihon Kōki 39.Tenchō 8.12.Jinshin)157. Like the revived saigū, there was an ostensible justification of imperial gratitude towards the shrine explaining the position of princess-priestess.

For the Kamo saiin, it was supposedly Emperor Saga’s way of thanking the gods protecting the palace for his victory against his older brother Heizei158 (平城) in the Kusuko Incident, by offering up his daughter to their worship159. In actuality, much like Tenmu’s saigū, this probably functioned as little more than a rationale for the institution of a new mechanism of religious centralisation and undermining of the sacral influence of regional elites, as we shall see shortly. First, however, let us examine how the figures of the saigū and saiin have been viewed through the lens of the “golden age” narrative.

5.3.2. Hime in Decline: Saigū as Sacral Partner

Within the “golden age” narrative and its surviving threads, the relationship between saigū and emperor is often read in terms of himehiko dynamics. In Takamure’s work, the saigū appears as a sort of decayed hime, fully subordinated to her hiko, stripped of her co-rulership function and shunted off to be cloistered in distant Ise, lest she wield any power or influence that might interfere with the wants of the patriarchal ritsuryō state. Her spiritual acuity empowers her no longer, her

156 賀茂注進雑記; “Miscellaneous Reports on Kamo”

157 “替賀茂斎內親王。…皇大神〈乃〉阿禮乎止賣〈爾〉進〈禮留〉內親王。齢〈毛〉老。身〈乃〉安

〈美毛〉有〈爾〉依〈弖。〉令退出〈留〉代〈爾。〉時子女王〈乎〉卜食定〈弖〉”. (“Kamo-no-Itsuki- no-Himemiko wo kawaru…Sume-Ōkami no are-otome ni susumereru himemiko. Toshi mo oi. Mi no yasumi mo ari ni yorite. Shirizoki izuru kawari ni. Tokiko-no-Joō wo urabete tabite sadamete”.)

158 R. 806–809 CE

159 Tokoro 2017, 42.

202 rituals instead performed purely for the benefit of her hiko’s authority; she goes from co-ruler to objectified asset160. A similar theme is presented in Yoshie’s work, the saigū the product of a reversal in relationship between hime and hiko: the hime, who once spoke with the voice of a god, reconfigured into a priestess worshipping the divine lineage of the hiko, who was now equated with his ancestor-deity161. Piggott similarly interprets the saigū as an “artefact of earlier contrapuntal structures”162, and Matsumura as a declined form of the hime163. The saigū is thus placed within the broader narrative of the depowering of sacral women across the later ancient period, with the priestess serving as a barometre of women’s social status as a whole.

The prototypical saigū, Yamato-Hime, acts as a median, in this narrative, between glory and decline: a progenitor of the cloistered princess-priestess, her deeds performed for the sacral benefit of her male kin, but also an active figure in her own right, advisor to princes and mistress of the Ise

Shrine, still retaining traces of her hime function164. Her mastery over the Grand Shrine is displayed in the Nihon Shoki, when she expels a group of “ill-mannered” Emishi from the shrine precinct165

(Nihon Shoki 7.Keikō 51.8.Jinshi); standing in contrast to the codified saigū of Tenmu’s day, who was

160 Takamure 1966b, 96. Takamure specifically describes a process by which the hime was alienated from her hiko, installed in “remote” Ise, and Yamato was configured as a hiko-centric chieftaincy with a “legal fiction of a hime” (擬制姫) in the hiko’s principal spouse. In this way, the saigū is depicted as a remnant of hime traditions, stripped of her co-rulership function and partnership with the hiko, whereas the hiko’s consort is made into a sort of figurehead-hime in the dependent role of wife.

161 Yoshie 1987, 34.

162 Piggott 1997, 212.

163 Matsumura 1999, 128–31.

164 Takamure 1966b, 112; Naitō 1976, 204; Yanagita 1984, 308; Sekiguchi 1987, 18; Kidder 2007, 133; Emura

2009, 10; Ambros 2015, 17.

165 Emura 2009, 9–11.

203 neither high priestess nor administrator of Ise but rather distanced from the shrine’s hierarchy entirely166.

In both Kojiki and Nihon Shoki, Yamato-Hime provides aid and advice to her nephew, the conquering hero-prince Yamato-Takeru, who fulfils the role of her hiko alongside her father Suinin167, the “Princess of Yamato” and the “Brave of Yamato”. In the Kojiki, she grants him first her garments, to aid him with his cunning cross-dressing plot (Kojiki 2.Keikōki)168; later, in both texts, she presents him with the holy Grass-Cutting Sword – Kusanagi-no-Tsurugi (草薙剣), one of the Imperial Regalia – as well as a fire-striker, both of which later save his life in an emergency. In the Kojiki text, the young hero confesses his fears to his aunt, the priestess serving as his confidant as well as source of advice and prophetic support (Kojiki 2.Keikōki; Nihon Shoki 7.Keikō 40.10.Bogo)169. Within the text, Yamato-

166 Emura 2012, 65–7.

167 Naitō 1976, 204; Kuratsuka 1996, 56; Emura 2009, 9–10.

168 “尓小碓命給其姨倭比賣命之御衣御裳。以釼納于御懷而幸行。”; “Koko ni Ousu-no-Mikoto sono oba

Yamato-Hime-no-Mikoto no mikeshi mimo wo tamawarite. Tsurugi mifutokoro ni irete idemashiki.”

169 Kojiki: “故受命罷行之時。參入伊㔟大御神宮。拜神朝。卽白其姨倭比賣命者。天皇旣所以思吾死

乎。何擊遣西方之人䒭而。返參上來之間。未經幾時。不賜軍衆。今更平遣東方十二道之人䒭。因

此思惟。猶所思看吾旣死焉。患泣罷時。倭比賣命賜草那藝釼。…亦賜御嚢而詔。若有急事。解茲嚢

口。” (“Kare mikoto wo makaridemasu toki ni. Ise no Ōmikami-no-Miya ni mairite. Kami no Mikado wo orogamitamaiki. Sunawachi sono mioba Yamato-Hime-no-Mikoto ni mōshitamaeraku wa. Sumera-mikoto wa hayaku are shine ya to omōsuran. Ikanareka nishi no kata no ashiki hitodomo wo uchi ni tsukawashite. Kaeri mainobori koshi hodo. Imada ikuda mo hezaru ni. Ikusabitodomo wo mo tamawazute. Ima sara ni himugashi no kata tōmarifuta michi no ashiki hitodomo wo kotomuke ni tsukawasuran. Kore ni yorite omou ni. Nao are hayaku shine to obōshimesu narikeri to. Urei nakite makaritamau toki ni. Yamato-Hime-no-Mikoto Kusanagi- no-Tsurugi wo tamai…Mata mifukuro wo tamaite noritamawaku. Moshi tomi no koto araba. Kono fukuro no kuchi wo tokitamae.”); Nihon Shoki: “抂道拝伊勢神宮。仍辭于倭姬命曰。今被天皇之命。而東征將誅諸

叛者。故辭之。於是倭姬命取草薙釼。授日本武尊曰。慎之莫怠也。” (“Ise-no-Ōmiya wo yogirite ogamu.

204

Hime appears as a sacral assistant to her kinsman – whose name is geographically linked with hers – providing him with divine blessing and legitimation. It is perhaps unsurprising for this story to be read in terms of himehiko dynamics. Furthermore, Yamato-Hime is the prototype for her office, the saigū that later saigū seek to emulate170, retracing the steps of her journey to Ise on their own initiatory pilgrimage171; does this indicate that, as the “golden age” narrative asserts, the passive, cloistered saigū has its roots in the powerful, respected hime?

The myth-historic status of Yamato-Hime and her predecessors complicates their use within the “golden age” narrative as symbols of pre-ritsuryō female sacral splendour. As we discussed with the image of Empress Jingū in the previous chapter, these figures exist as constructs within the

Yamato canon, serving the interests of the patriline172. As Jingū’s story serves to assert Yamato tributary dominion over Korea, so too does Yamato-Hime’s story serve to assert Yamato cultic dominion over Ise Shrine; a facet which we shall explore in greater detail shortly.

As for Yamato-Hime’s support of Yamato-Takeru, her intervention for his benefit provides his conquests with divine sanction; furthermore, its ties Ise Shrine to a program of eastward Yamato expansion173. The connection between Yamato-Hime and Yamato-Takeru is particularly emphasised in the Kojiki, where she assists him on two occasions – both western and eastern conquests – and the “Yamato” portion of their names is written with the same character, 倭. In the Nihon Shoki,

Sunawachi Yamato-Hime-no-Mikoto ni makari mōshitamaite iwaku. Ima sumera-mikoto no omikoto wo uketamawarite. Higashi ni yukite masa ni moro no somuku mono wo tsuminaen to su. Kare itomamōshisu.

Koko ni Yamato-Hime-no-Mikoto Kusanagi-no-Tsurugi wo torite. Yamato-Takeru-no-Mikoto ni sazukete iwaku.

Tsutsushimite naokotari zo.”)

170 Ellwood 1967, 37; Kuratsuka 1996, 55.

171 Tanaka 1996, 72.

172 Maeda 2008, 7.

173 Naoki 2009, 9.

205 however, Yamato-Hime assists only with the eastern conquests, and her “Yamato” (倭) is distinct from Yamato-Takeru’s more modern “Yamato” (日本)174, creating a gap between them. Instead,

Yamato-Hime’s connection to Ise is prioritised, through the story of her pilgrimage and her expulsion of the Emishi, both absent from the Kojiki.

Emura’s work on the founding mythology of the saigū has presented us with means of interpreting this discrepancy: the story of Yamato-Hime aiding Yamato-Takeru subtly serves to link

Emperor Tenmu – who sought aide from Ise Shrine to win the Jinshin War, who re-instituted the position of saigū with his own daughter Princess Ōku upon his accession, and who initially ordered the compilation of the Kojiki – with the conquering hero, symbol of Yamato strength175. The Nihon

Shoki’s emphasis instead on Yamato-Hime’s Ise connections may instead mean to connect her to

Tenmu’s widow and successor, Empress Jitō176 (持統), whose own travels to Ise were highly unusual for a reigning monarch177. With both the myth-historic and political contexts behind these passages in mind, the extent to which they can be literally read as accurate reflections of early Yamato priestess power dynamics is cast into considerable doubt.

As for the later saigū, their interpretation as female sacral counterparts to a male monarch, legacies of an ancient “himehiko” tradition, is similarly problematic. The saigū itself is, certainly, a gendered role, intended specifically for a priestess. We have observed (see Ch. 3.2.1) how the language used for the appointments of Toyosuki-Irihime and Yamato-Hime indicated a shamanic and/or divinatory quality to their role through the use of the character 託 (tsuke, “to entrust”, “to

174 倭, the character used for “Wa” in the Records of the Three Kingdoms, can be used to mean the entirety of

Yamato dominion as well as the province of Yamato specifically. 日本, here also read “Yamato,” is a later name for the dominion, surviving to this day as Nihon/Nippon – “Japan”.

175 Emura 2009, 12, 21.

176 R. 686–697 CE

177 Emura 2009, 12, 20–1.

206 send via”); we also observed how that role was particularly gendered female in Yamato culture.

Although the later saigū did not perform the traditional activites of the miko role – neither delivering oracles nor dancing kagura178 – the shamanic language used for her mythic progenitors suggests the awareness or expectation that the post was grounded in the gendered religious traditions of Yamato, conceived of as a miko even if that function was later lost.

But the concept of the hime – and the “golden age” narrative framing of the saigū as the remnant of such – invokes more than just the presence of gendered religious positions. The

“himehiko system” specifically centres on the idea of contrapuntal co-rulership between male and female figures179; the saigū decayed from a co-ruler to a mere sacral assistant. We have discussed in the previous chapter the issues present within the “golden age” framing of himehiko, and in particular the inclusion of the assistant-priestesses of early Yamato canon within the notion. The first concern is that of homogenisation: as we have seen, the concept of himehiko casts a wide net, encompassing a disparate variety of rulership formats and gendered relationships to authority.

The spiritual aspect is particularly tenuous: the saigū-as-hime holds a purely sacral role, compared to the mundane politics of her hiko; other women presented as hime displayed sacral functions alongside posts of temporal rulership; yet others evinced no apparent religious function whatsoever. It is hard to connect a direct line of derivation from, for example, regionally disparate female chieftains – such as Taburatsu-Hime of the tsuchigumo – who display no particular sacral function, to clan priestesses such as the saigū who display little rulership activity. The two roles have little in common, aside from the broad categorisation as female figures working in concert with a kinsman.

178 Ellwood 1967, 38; Tanaka 1996, 72; Ooms 2009, 192; Emura 2012, 66.

179 Yoshie 1987, 35; Yoshie 2007, 160, 212; Maeda 2008, 17–9.

207

Let us examine the saigū’s relationship with the paramount in greater detail: the theorised himehiko model posits a male and female figure, of roughly equal standing, working in counterpoint, dividing their duties based on societal gender norms. Can traces of this complementarity be found between saigū and emperor?

The first complication to the “saigū-as-hime” narrative is the mythic position of Nunaki-

Irihime, the failed counterpart to the proto-saigū Toyosuki-Irihime. Sujin’s initial plan was to have two of his daughters each take on worship of one of the royal gods. If the Yamato court of the late seventh and early eighth centuries conceived of the saigū as a contrapuntal partner to the emperor, representing the duality of male and female roles, then where does Nunaki-Irihime fit into the equation? Is Emperor Sujin meant to be understood as a hiko with two hime; three, if one counts his sacral assistance elsewhere from Yamato-Totohi-Momoso-Hime? Rather than a specific male-female pair embodying contrapuntal duality, Sujin receives sacral assistance from multiple female relatives.

Instead of himehiko, this suggests a broader tradition of clan-priestesses supporting the rule of the clan head, without breaking off into single male-single female partnerships.

What, then, if the clan head is a woman? If the saigū was conceived of as the sacral-female counterpart to the masculine hiko-sovereign, then would not such a partner be unnecessary when the sovereign was female? At first glance, this might seem to be the case: despite succeeding Tenmu,

Empress Jitō lacked a saiō of her own; Empresses Genmei180 (元明) and Kōken-Shōtoku also lack records of saigū appointments within the Shoku Nihongi. As Tokoro and Emura note, this apparent absence has oft been read as a signifier that a saigū, as a hime-derivative, was considered unnecessary for a female monarch, who did not require a female intermediary to obtain sacral power181.

180 R. 707–715 CE

181 Emura 2009, 17; Tokoro 2017, 29.

208

A look at the records, however, challenges this interpretation. Sukate-Hime, one of the earliest historically-reliable saigū, had a term of service that ended well into the reign of the first official Empress Regnant, Suiko (Nihon Shoki 21.Yōmei Tennō Sokui Zenki Bidatsu 14.9.Jinshin); within the Shoku Nihongi, Princesses Kuse182 (久勢) and Inoue183 (井上) hold the post during the reign of Empress Genshō184 (元正) (Shoku Nihongi 7.Yōrō 1.4.Itsugai; Shoku Nihongi 8.Yōrō

5.9.Itsubō), the latter continuing her duties into the reign of her father Shōmu185 (聖武).

Furthermore, although absent from the Shoku Nihongi, the Kamakura chronology Ichidai Yōki186 lists two saigū – Princesses Chinu (智努) and Madokata (円方) – appointed during Genmei’s reign187, and another – Princess Oyake (小宅) – for Empress Kōken-Shōtoku’s first reign (as Kōken)188 (Ichidai Yōki

1.44/46). We may thus see that a female saigū was neither redundant nor contradictory to a female tennō, undermining the notion of the saigū role being understood as a gender-complementary partner of the emperor.

182 “遣久勢女王侍于伊勢太神宮。” (“Kuse-no-Joō wo tsukawashite Ise-no-Ōkami-no-Miya ni haberashimu.”)

183 “以皇太子女井上王爲齋內親王。” (“Hitsugi-no-miko no musume Inoue-no-Ō wo motte Itsuki-no-

Himemiko to nasu.”)

184 R. 715–724 CE

185 R. 724–749 CE

186 一代要記; “Essential Chronicles by Reign”

187 “神祇記云是時齋王不定信田方内親王多貴内親王各一度参入次智努女王次圓方女王各一度参入云云”

(“Jingiki iwaku kono toki saiō wo sadamazaru makoto ni Takata-Naishinnō Taki-Naishinnō onoono ichido sannyū tsugi ni Chinu-Joō tsugi ni Madokata-Joō onoono ichiso sannyū unun.”)

188 “小宅女王〈天平勝寶元年九月六日以從三位三原王女小宅女王〉” (“Oyake-Joō (Tenpyō-Shōhō gannen nagatsuki muika wo motte jūsanmi Mihara-Ō no musume Oyake-Joō).”)

209

As for the remaining empresses, Kōgyoku-Saimei189 ruled during the saigū hiatus between

Suiko and Tenmu, making her lack of saiō an unremarkable fact with no particular gendered connotations. Kōken-Shōtoku’s second reign (as Shōtoku) was heavily saturated with Buddhism: having taken monastic vows between reigns, she referred to herself as a nun upon her coronation and appointed monks as her advisors (Shoku Nihongi 26.Tenpyō-Jingo 1.Urū 10.Kōin/11.Kōshin).

Shōtoku’s reign prioritised Buddhism over kami-worship, and her approach to traditional imperial ritual was one of syncretisation, having monks and nuns at her Daijōsai190 – an unprecendented occurrence191 – and building a Buddhist shrine-temple (神宮寺, jingūji) at Ise (Shoku Nihongi

26.Tenpyō-Jingo 1.11.Kōshin; Shoku Nihongi 27.Tenpyō-Jingo 2.7.Heishi). Shōtoku’s lack of saigū appears as less of a reflection of the position being redundant for a female tennō, and more of a characteristic of Shōtoku’s distinctive Buddhist emphasis, establishing control over Ise via construction of Buddhist institutions rather than Amaterasu-worship192. Bowring also suggests Ise’s role in the legitimation of monarchy may have been reduced at this point in time, citing the selection of Usa Hachimangū instead as the potential kingmaker for the Dōkyō Incident as a possible indicator193. Only Jitō’s lack of a saiō appears as unusual within context, and even this may be explainable by both the nascent state of the saigū revival, as well as Jitō’s own unusual trips to Ise

(Nihon Shoki 30.Jitō 6.2.Shinbi), which served to directly impress the might of the royal court upon the peripheries of the empire, rather than through the intermediary of a saigū194. The gendering of

189 R. as Kōgyoku 642–645 CE, again as Saimei 655–661 CE

190 The Niinamesai (新嘗祭) or Festival of First Fruits was a harvest festival, which doubled as a rite of protection of the imperial lifespan. The first Niinamesai of an emperor’s reign was called the Daijōsai (大嘗祭), and also served as something of a sacral coronation rite.

191 Ooms 2009, 195.

192 Emura 2009, 43–4; 2012, 83.

193 Bowring 2005, 96.

194 Bowring 2005, 37.

210 the saigū is not, within the records, presented as a deliberate counterpoint to a male tennō; rather, it reflects a broader gendering of priestess and paramount, without the two specifically configured as a contrapuntal pair.

Indeed, within her ritual context, the saigū is positioned as a stand-in for the emperor, rather than a partner working in complimentary harmony; in literature, she is referred to as his “staff substitute”195, representing emperor and court at Ise, before the ancestor-goddess196. More specifically, she and her attendants represented the court as it wanted to be seen by Amaterasu: as a crystallised haven of sacral purity and kami-worship, free from the pollution of the mundane world or the influence of foreign religions197; reproducing at Ise the Yamato court’s own idealised image of its pre-Buddhist past, its own narrative of a “golden age”. The identification of the saigū as the emperor’s substitute suffused the daily life and construction of her compound. The ritual calendar maintained by the saigū complex was based on that of the Yamato court, rather than that of the Ise

Shrine198; the construction of the saigū’s residence itself was based on the royal palace199, received the same blessing rites200, and was even referred to as a rikyū (離宮), a term normally reserved for royal villas201.

As a kinswoman of the emperor, the saigū performed the same role within the same rituals along the same calendar – and may have even worn the same clothing while doing so202 – from

195 御杖代, mitsue-shiro.

196 Emura 2010, 159; Tokoro 2017, 2.

197 Ellwood 1967, 42–4, 60; 1972, 50; Ooms 2009, 192–3; Tokoro 2017, 3.

198 Ellwood 1967, 42, 54; 1972, 50; Ooms 2009, 192; Emura 2012, 93; Tokoro 2017, 14.

199 Emura 2009, 27.

200 Emura 2010, 157–9.

201 Emura 2012, 16.

202 Emura 2010, 149–51.

211 within an enclave of absolute purity at the cultic centre of the imperial ancestor-goddess. This synchronous performance of ritual bolstered its efficacy, securing divine blessings and protection of the imperial reign203. Contrast this with the image of Jingū’s divinations (see Ch. 4.2.1): the feminine role of medium clearly delineated from the masculine role of saniwa interrogator, each position a necessary half of a ritual whole, a performance based around the notion of gender complementarity.

The saigū, however, acts not as the contrapuntal partner to the emperor, but rather his ritual duplication. Although the gendering of the saigū appears to have its roots in the Yamato miko tradition, her function evinces no notions of gender complementarity or contrapuntal performance, absent the outdated assumption (see Ch. 4.3.1) of “sacral female/political male” pairings as the standardised form of rulership in ancient Japan.

5.3.3. Saigū and Saiin as Mechanisms of Regional Control

Another issue with the presentation of saigū-as-hime is the deteriorationist narrative that underlies it: the saigū, in this understanding, can only exist as a remnant of something greater, a decayed form of a previous image of female power. Yet Piggott notes a problem with this framing: the saigū’s presence in Ise was heavily intertwined with a Yamato program of regional control204. For the saigū’s role as representative of court and emperor was not restricted only to the ritual context.

Her posting, as a royal princess and imperial clan-priestess in the major cultic centre of Ise, also served as a sacral reinforcement of imperium: impressing upon the provinces the emperor’s authority, and the divine mandate behind it205. She represented the court at Ise not just before the gods, but also before the regional elites, reminding those distant from the Yamato court of its power

203 Ooms 2009, 192–3.

204 Piggott 1994, 15.

205 Piggott 1997, 212.

212 and reach, a “personified symbol” of the empire206; a “living regalia”, almost, to borrow a term from

Tanaka, a symbol and proof of imperial power in (objectified) human form207.

The exact age of the Inner Shrine of Ise – the royal nexus, housing Amaterasu – is uncertain, although Okada proposes a dating of 477 CE, during the reign of Yūryaku208. The Yamato monarchy of this time was heavily concerned with control and administration of the provinces, especially in the wake of new systems of social organisation introduced by the central government209. Ise was an important hub for management of the Eastern Provinces210; the story of Yamato-Takeru and Yamato-

Hime, in which Yamato-Takeru visits his aunt in Ise before going to conquer the east, indicates that a link may have been made between the shrine and the Yamato dynasty’s eastward expansion211. Ise

Shrine had a pre-existing worship of a local solar deity whose cult could be easily syncretised with

Amaterasu’s212; furthermore, its location east of Yamato Province may have led to it having pre- existing solar associations in Yamato religious tradition213.

This latter aspect of religious syncretisation also speaks to the position of the saigū as a mechanism of the state cult. Yamato political centralisation was contiguous with religious centralisation, placing Amaterasu – royal ancestress and symbol of their divine mandate – at the

206 Ellwood 1967, 43.

207 Tanaka 1996, 103. While Tanaka specifically discusses the notion of “living regalia” as a product of mediaeval discourse on the saigū, arguably her use as a symbol of imperium casts her in this position even in the ancient period.

208 Okada 1970, 346; Piggott 1997, 61; Naoki 2009, 12.

209 Okada 1970, 342, 347–8.

210 Naoki 2009, 3.

211 Naoki 2009, 9.

212 Okada 1970, 335, 350, 352; Kuratsuka 1996, 53; Naoki 2009, 17; Emura 2012, 18–9.

213 Okada 1970, 337.

213 head of a hierarchy of clan-cults, with local religious beliefs subordinated beneath214. Within the

Yamato myth of the Kojiki and Nihon Shoki, Amaterasu is positioned as supreme among the divinities, rightful ruler of the heavenly gods (Kojiki 1; Nihon Shoki 1.5.6)215, provided of the mandate of rulership to her descendants (Kojiki 1; Nihon Shoki 2.9.1)216, and greater in stature than the ancestor- gods of other clans217. The state cult ingrained contemporary clan hierarchies and duties into its mythic canon, with Amaterasu and her progeny, representatives of the imperial clan, at the apex.

The dedication of Inner Shrine and saigū to the worship of Amaterasu – within a provincial cultic centre previously dedicated to the worship of local elites – constituted a Yamato assertion of political dominance with religion as its medium. The worship of the saigū was not simply an internal

214 Anesaki 1930, 29.

215 Kojiki: “賜天照大御神而。詔之汝命者所知高天原矣。事依而賜也。” (“Amaterasu-Ōmikami ni tamaite.

Na-ga-mikoto wa Takama-no-Hara wo shiroshimese to noritamaite. Koto yosashi tamaikinari.”); Shoki: “已而

伊奘諾尊勑任三子曰。天照大神者可以治高天原也。” (“Sude ni shite Izanagi-no-Mikoto mihashira no miko ni koto yosashite noritamawaku. Amaterasu-Ōnkami motte Takama-no-Hara wo shirasu beshi nari.”)

216 Kojiki: “尓天照大御神。高木神之命以。詔太子正勝吾勝々速日天忍穗耳命。今平訖葦原中國之白。

故隨言依賜。降㘴而知看。…是以隨白之科詔日子番能迩々藝命。此豊葦原水穗國者。汝將知國。言依

賜。故隨命以可天降。” (“Koko ni Amaterasu-Ōmikami. Takagi-no-Kami no mikoto motte. Hitsugi-no-miko

Masakatsu-Akatsu-Kachi-Hayahi-Ame-no-Oshiho-Mimi-no-Mikoto ni noritamawaku. Ima Ashihara-no-Nakatsu-

Kuni wo kotomuke wo hen to mōsu. Kare koto yosashi tamaishi manimani. Kudarimashite shiroshimese…Koko wo mote mōshitamau manimani Hikoho-no-Ninigi-no-Mikoto ni. Kono Toyo-Ashihara-no-Mizuho-no-Kuni wa.

Imashi shirasan kuni nari to. Koto yosashi tamaeri. Kare mikoto no manimani amorimasubeshi to mikoto

ōsetamaiki.”); Nihon Shoki: “因敕皇孫曰。葦原千五百秋之瑞穗國。是吾子孫可王之地也。宜爾皇孫就而

治焉。行矣。寶祚之隆當與天壤無窮者矣。” (“Yorite sumemima ni mikotonori shite notamawaku.

Ashihara-no-Chiioaki-no-Mizuho-no-Kuni wa. Kore a ga umi no ko no kimitarubeki kuni nari. Yoroshiku imashi sumemima yuite shirase. Sakiku. Amatsu-hitsugi no sakaemasan koto sani to ametsuchi to kiwamari nakarubeshi.”)

217 Okada 1970, 349–50; Wakita 2003, 13; Naoki 2009, 3.

214 act of clan-god worship, performed by a priestess on behalf of a male paramount; it was also a public proclamation, within the provinces, of the supremacy of the Yamato kings, and the divine right to rule with which they were bestowed.

In other words, while the deteriorationist narrative favoured by Takamure presents the saigū role as a simple matter of a former hime, distanced from the centre of power by her installation in far Ise218, her placement there was by no means an afterthought. Rather, it was part of a new, active strategy of power consolidation by the Yamato court, the specific context of which is interwoven with the saigū’s role, activities, and founding mythology; points which are missed through the framing of the saigū post purely in terms of deterioration from the hime.

The saigū’s role as a bastion of imperial power in the provinces is baked into her founding mythology. Let us return again to the beginnings of the saigū in the Yamato myth-historic canon: in the Nihon Shoki accounting, Amaterasu was once enshrined not at Ise, but within the royal palace itself, alongside the god Yamato-no-Ōkuni-Tama who embodied the land of Yamato. Sujin, fearful of their power, was uneasy residing alongside these two deities, and so entrusted them to two of his daughters: Amaterasu to Toyosuki-Irihime, and Yamato-no-Ōkuni-Tama to Nunaki-Irihime.

Amaterasu was safely enshrined in a village elsewhere in Yamato Province (Nihon Shoki 5.Sujin 6)219.

Toyosuki-Irihime continues her service in this role well into into the reign of Sujin’s successor, Suinin.

At this point, the entrustment of Amaterasu is passed on to Suinin’s daughter Yamato-Hime, mythic

218 Takamure 1966b, 96.

219 “先是。天照大神。倭大國魂二神。並祭於天皇大殿之內。然畏其神勢共住不安。故以天照大神。託

豐鍬入姬命。祭於倭笠縫邑。仍立磯堅城神籬。” (“Kore yori saki. Amaterasu-Ōkami. Yamato-no-Ōkuni-

Tama futahashira no kami wo. Sumera-mikoto no miaraka no uchi ni narabi ni iwaimatsuru. Shikashi sono kami no miikioi wo osorete tomo ni sumitamau ni yasumazu. Kare Amaterasu-Ōkami wo motte. Toyosuki-Irihime-no-

Mikoto wo tsukematsurite. Yamato no Kasanui-no-Sato ni iwaimatsuru. Sunawachi Shikataki Himorogi wo tatete.”)

215 progenitrix of the saigū and founder of the Ise Shrine complex. Here, Yamato-Hime embarks upon a journey across the provinces searching for a suitable location to enshrine the goddess. Arriving in Ise,

Amaterasu personally instructs her to build a shrine on the banks of the Isuzu River: the Inner Shrine of Ise (Nihon Shoki 6.Suinin 25.3.Teigai)220.

The purpose of these tales, within the myth-historic canon of Yamato, is threefold: firstly, they explain the question of how the imperial clan-god’s221 cultic centre should be located so far from their seat of authority in Yamato. Secondly, they serve to establish a primeval connection between Ise Shrine and the Yamato monarchy, crediting a Yamato princess (indeed, one named after

Yamato) with the very founding of the shrine complex. Finally, they act as a foundation myth for the institution of saigū as well as the shrine, inextricably linking the two together: the Inner Shrine of Ise is presented by the Nihon Shoki as having always been a centre for royal worship of Amaterasu, and always been presided over by a saiō, since the shrine’s very foundation.

The coalescent core of Yamato-Hime’s myth-historic role, as the Yamato chroniclers understood it, was not as an image of feminine sacral power, or even a particular attempt to

220 “離天照大神於豐耜入姬命。託于倭姬命。爰倭姬命求鎮坐大神之處。…東迴美濃到伊勢國。時天照

大神誨倭姬命曰。是神風伊勢國。則常世之浪重浪歸國也。傍國可怜國也。欲居是國。故隨大神教。其

祠立於伊勢國。因興齋宮于五十鈴川上。是謂磯宮。則天照大神始自天降之處也。” (“Amaterasu-Ōkami wo Toyosuki-Irihime-no-Mikoto ni hanachimatsurite. Yamato-Hime-no-Mikoto ni tsuketamau. Koko ni Yamato-

Hime-no-Mikoto ōkami wo shizume masasen tokoro wo motomete…Higashi-no-kata ni Mino wo megurite Ise- no-Kuni ni itaru. Toki ni Amaterasu-Ōkami Yamato-Hime-no-Mikoto wo oshiete notamawaku. Kono kamikaze no Ise-no-Kuni. Sunawachi tokoyo no nami no shikinami yosuru kuni nari. Katakuni no umashi kuni nari. Kono kuni ni on to omou. Kare ōkami no oshietamau manimani. Sono hokora wo Ise-no-Kuni ni tatetamau. Yorite iwai-no-miya wo Isuzu-no-Kawakami ni tatsu. Kore wo Iso-no-Miya to iu. Sunawachi Amaterasu-Ōkami no hajimete ama yori kudarimasu tokoro nari.”)

221 氏神, ujikami.

216 faithfully recreate gendered religious and power dynamics in the Yamato dynastic past. Rather, her function is to assert the imperial clan’s “ownership” of Ise Shrine, amid a program of religious centralisation contemporaneous with the chronicles’ creation, of which Ise was a prominent battleground. The story serves as an alibi for the royal appropriation of Ise Shrine as an act of regional control, a revised history which presents the Yamato dynasty as the shrine’s founders by divine fiat. The Outer Shrine – still held by the regional Watarai clan (度会) – is presented as a later addition to the complex, lesser in significance, backdating Yamato control over the site and undermining regional elites’ connection to their own ancestral seat of worship222. The relationship between the later saigū and the earlier, more active and powerful, image of Yamato-Hime in Yamato myth-history, is not a neat one-way narrative of decline; rather, it is at the very least a two-way relationship, the role of the later saigū informing and shaping the myth-historic depiction of Yamato-

Hime. To interpret the later saigū only in terms of “deterioration” from the image of their mythic progenitrix is to misinterpret the relationship between the two, in an overly-literal reading of

Yamato myth-history.

The “golden age” narrative’s treatment of the saigū is rooted in an essentialist, overly linear reading of patriarchy, which assumes that a female religious role could not organically emerge out of the Yamato court of this period, that it could only exist during this time as a deterioration of something earlier and greater. Piggott notes the importance of the saigū as a mechanism of regional control, although characterises it as a transformation of a presumed hime antecedent223. But let us take this a step further: the role of the saigū and the later saiin are intrinsically intertwined with contemporary court politics and strategies of regional control. To divorce them from this context, to define the role of the saigū purely in terms of encroaching patriarchy and “decline” from a “golden age” ideal, is insufficient to understand the history and purpose of the role.

222 Okada 1970, 346.

223 Piggott 1994, 15.

217

As a political mechanism, the importance of the saigū’s role was variable based on the changing needs of the state. Naoki suggests that the saigū hiatus between Sukate-Hime and Ōku corresponds to a period in which relations between Ise and the centre were sufficiently harmonised, the Yamato monarchs’ hold on the region sufficiently secure, that the use of the saigū as a means of regional control became (temporarily) obsolete224. The saigū revival occurred during Tenmu’s reign, following his victory in the Jinshin War succession conflict of 672, in which the successful occupation of Ise Shrine had proved a major coup for Tenmu’s forces’ morale, taken as a sign of Amaterasu’s favour225; shrine authorities ostensibly favoured him over his rival Prince Ōtomo (大友)/Emperor

Kōbun226 (弘文). Given the turbulence of his rise to the throne, it is unsurprising that his rule should come with a renewed focus on Ise as both symbol of his divine mandate and base of regional control.

Ise was highly regarded as a sacred site by many eastern elites, and thus maintaining a good relationship with the shrine was also advantageous to Tenmu’s political relationships227; later,

Emperor Kanmu would similarly use an emphasis on the saigū to stabilise relations with Ise in the wake of the changing capital228.

When the Yamato royalty moved their capital to Heian, they continued this program of state-governed religious centralisation with the establishment of the Kamo saiin. She too existed as a mechanism of regional control: this time not of the periphery, but rather of the new imperial capital, where Kamo Shrine was commandeered much in the same manner as Ise. Prior to this, the shrine was tended to by the chief clan of the region, the Kamo-no-Agatanushi, for whom the Kamo

224 Naoki 2009, 77.

225 Piggott 1994, 26; Bowring 2005, 45; Emura 2009, 18; Naoki 2009, 3, 79.

226 Piggott 1997, 128. “Emperor Kōbun” (r. 671–672) was only given this name and title after the Meiji

Restoration; contemporary chronicles such as the Nihon Shoki do not recognise him as a legitimate monarch, instead attributing this period to the beginning of Tenmu’s reign.

227 Piggott 1997, 144–6.

228 Emura 2012, 85.

218

Shrine was a centre of ancestor-god worship229. Central to the shrine calendar was the hierogamic ritual of Miare230, in which a male priest (are-otoko) was possessed by the Kamo god and enacted a rite of divine marriage with a priestess (are-otome), from which union the Kamo god was symbolically born231. The celebrations surrounding this rite – the Kamo Festival – was a major regional social event, attracting a large crowd. The Yamato court’s attempt to regulate these gatherings displays the shrine’s importance to the centre as a lynchpin of regional control (Shoku

Nihongi 1.Monmu 2.3.Shinshi; Shoku Nihongi 2.Taihō 2.4.Kōshi; Shoku Nihongi 5.Wadō 4.4.Itsubi)232.

During Kanmu’s reign, as the Yamato court began preparations to relocate to Yamashiro, the

Kamo Shrine took on increased importance as an impending central religious institution. Hereditary kabane titles were granted to local elites (including a senior priest of the shrine) (Shoku Nihongi

36.Hōki 11.4.Kōshin)233, offerings were made to inform the local deities of the capital’s move (Shoku

Nihongi 38.Enryaku 3.6.Jinshi)234, and both Upper and Lower Kamo Shrines were promoted to Junior

229 Yoshie 1987, 33; Ōwa 2009, 127.

230 御阿礼/御生, “Birth/Appearance.”

231 Faure 2004, 295; Yoshie 2007, 207–8; Meeks 2011, 209; Ambros 2015, 63.

232 1: “禁山背國賀茂祭日會衆騎射。” (“Yamashiro-no-Kuni Kamo no Matsuri no hi ōki atsumete kisha suru koto wo kinzu.”); 2: “禁祭賀茂神日。徒衆會集執仗騎射。唯當國之人不在禁限。” (“Kamo-no-Kami wo matsuru hi. Tada shūkai atsume tsuwamono wo torite kisha suru koto wo kinzu. Tada tōkoku no hito kin no kagiri ni arazu.”); 3: “詔。賀茂神祭日。自今以後。國司每年親臨檢察焉。” (“Mikotonori suraku. Kamo-no-

Kami matsuri no hi. Ima yori ikō. Kuni-no-tsukasa maitoshi mizukara nozomi kensatsu seyo.”)

233 “山背國愛宕郡人正六位上鴨祢宜眞髪部津守等一十人賜姓賀茂縣主。” (“Yamashiro-no-Kuni Otagi-no-

Kōri no hito shō-rokui-no-jō Kamo no negi Makabe no Tsumori-ra totari ni kabane wo Kamo-no-Agatanushi to tamau.”)

234 “遣參議近衛中將正四位上紀朝臣船守於賀茂大神社。奉幣。以告遷都之由焉。” (“Sangi Konoe-no-

Chūjō shō-shii-no-jō Ki-no-Ason Funamori wo Kamo-no-Ōkami-no-Yashiro ni tsukawashite. Mitegura wo tatematsurashimu. Sento no yoshi wo tsugeru wo motte nari.”)

219

Second Rank235 (Shoku Nihongi 36.Enryaku 3.11.Teishi)236. Eventually, the shrine itself was appropriated by the imperial clan through the installation of the saiin. The saiin was ostensibly a royal version of the are-otome, an imperial princess worshipping the gods of Kamo237. In practice, however, she was a replication of the saigū on the new stage of the capital, a political playing piece, rather than a continuation of local religious practices. The old ritual of the are-otome was appropriated by the ruling class, local worship once more absorbed into the imperial cult; the are- otoko declined in importance, his role effectively removed through the replacement of his former hierogamic partner with a cloistered vestal of the imperial cult. Through the saiin, the imperial clan laid claim to Kamo, as they had done with Ise via the saigū, subsuming the local religious institutions and forging their own connections to the presiding deities238. A reconstituted Kamo Festival239, sans

Miare rite, remained the biggest social event in the Heian calendar for commoners and courtiers alike, the key focal point the saiin’s procession with an elaborate parade, gilded carriage, and silk- clad attendants (Makura no Sōshi 208.1/4): a very public spectacle of splendour and grandeur, and at the centre was the saiin, symbol of imperial religious dominion240, a reminder to all and sundry of the emperor’s political and religious hold over the region.

The late ninth and tenth centuries saw a lessening of central concern with the regional government. The saiin, located in the capital, thus came to be prioritised over the more peripheral

235 Within the ritsuryō bureaucracy, shrines were accorded rankings along a similar system to the kan’i of aristocrats.

236 “敘賀茂上下二社從二位。” (“Kamo Jōge no futayashiro wo jū-nii ni josu.”)

237 Yanagita 1984, 31.

238 Emura 2012, 136; Ambros 2015, 63–4.

239 This festival survives in some form today, as the modern Aoi Matsuri (葵祭, “Hollyhock Festival”) held each

May in Kyoto; the role of the saiin is played by a saiō-dai (斎王代), or “saiō substitute”.

240 Bock 1970b, 6; Ambros 2015, 63–4.

220 saigū, with the strongest princess-candidates going to Kamo over Ise241; the relationship between political circumstance and attention given to these positions displays the implicit role of saigū and saiin as mechanisms of imperial control. In mid- to late-Heian, imperial authority weakened under a succession of boy-kings and regency governments; the saiō, so inextricably linked to imperial power, experienced a similar loss of significance242, although both saiin and saigū endured until the

Kamakura period.

The role and character of the saigū was thus heavily based in the clan politics and eastward expansion of the early Yamato Dynasty; in particular, the shift towards a centralised state cult undertaken by Tenmu. The role of the saiin was similarly interwoven with the regional power dynamics involved in the move to Heian-kyō, and the imperial court’s solidification of their hold on their new base of power. This context is interwoven with the saigū and saiin’s duties and purpose; the portrayal of the saiō as “just” a depowered hime, shunted off to far Ise to remove her from the mechanisms of power243 or holed up in Kamo as an ornamental shell of an ancient shaman-queen, is a reductive one that fails to take into account the specific qualities and significance of her role. It is crucial to understand the saiō within her historical context, and not as a sort of generic successor of a broadly-defined hime role.

5.3.4. Tensions of Power within the Saiō

We see therefore that the saigū was not a once-respected hime relegated to Ise to sever her from the mechanisms of power; rather, as Piggott notes, she was placed in Ise to become a nexus of imperial power in the provinces, a constant reminder of the imperial clan’s divine lineage244. Possibly,

241 Emura 2009, 112, 115.

242 Tanaka 1996, 96.

243 Takamure 1966b, 96; Yoshie 1987, 34; Tanaka 1996, 73; Matsumura 1999, 128–31.

244 Piggott 1994, 15.

221 the image of the saigū as a depowered hime owes at least in part to the tensions of power within the saiō role itself; tensions that we shall here seek to examine in the context of the saigū’s dynastic function.

In one sense, the saiō was accredited with considerable sacro-political importance, particularly during periods of increased court concern with regional control. Ostensibly, she functioned as a clan priestess of the Yamato paramount, occupying a special position outside the hierarchy of the state cult. But the significance of the position does not appear to have translated into any tangible form of power held by the saiō as an individual. In contrast to the Nihon Shoki’s depiction of Yamato-Hime as mistress of Ise Shrine (Nihon Shoki 7.Keikō 51.8.Jinshi), the saigū had little formal or concrete authority in the governance of the complex, let alone the wider sphere of the kingdom. Her autonomy was heavily curtailed by the strict demands of ritual purity imposed upon her. Even the sacral acuity attributed to an individual saiō was limited, despite the apparent sacral importance of the role.

The saiō was a priestess, but her activities were very different from those of a court kannagi or village miko: the traditional dances and offerings were performed not by the saiō herself but by a bevy of assistants245; oracular pronouncements were few and extraordinary246, not part of the saiō’s typical repertoire247. The saiō had a key role in rituals, but not as a ritualist; she did not so much perform or wield spiritual power as she did passively generate it through her presence. Whereas court kannagi (see Ch. 5.2.2) were typically required to be skilled, active, potent ritualists (Engishiki

3.Jingi 3), the saiō was often appointed as a very small child248; all that was required from her was to

245 Ellwood 1967, 38; Tanaka 1996, 72.

246 Tanaka 1996, 94–5; Ooms 2009, 192.

247 Tanaka 1996, 72; Ooms 2009, 192; Emura 2012, 66.

248 Ellwood 1967, 43; Bock 1970a, 151; 1970b, 9.

222 simply occupy her position while retaining her ritual purity249. The saiō was in one sense a site of considerable sacral power, a particular ritual efficacy derived from her life in an enclave of complete purity (for the saigū, at Amaterasu’s own cultic seat); a power of “being,” of presence in time and space, rather than one of specific self-assertion250. In another sense, however, this power was contained within the role, not the individual holding it251; it was neither her particular talent nor something she could actively wield.

We have discussed (see Ch. 4.3.1) how Himiko’s power as queen and priestess may have been bought at the cost of her personal autonomy, a life of strict seclusion necessary to provide her with spiritual acuity. A similar transaction lies in the construction of the saiō role: the saigū left her compound only thrice yearly, to attend major rites at the main Ise Shrine complex; she would not leave Ise until dismissed from her post. A two-year period of ritual seclusion and frequent purifications was required before the saigū could even begin her role (Engishiki 5.Jingi 5)252; for the

249 Although the ikasuri-no-kannagi was chosen more on basis of clan affiliation, the existence of a minimum age requirement for her duties suggests an active ritual role with at least some necessary level of physicial capability (Engishiki 3.Jingi 3).

250 Sered 1999, 161; Bellamy 2008, 42.

251 Ellwood 1967, 36; Tanaka 1996, 96.

252 “凡齋內親王定畢。卽卜宮城內便所。爲初齋院。祓禊而入。至于明年七月。齋於此院。更卜城外淨

野。造野宮畢。八月上旬。卜定吉日。臨河祓禊。卽入野宮。自遷入日。至于明年八月。齋於此宮。九

月上旬。卜定吉日。臨河祓禊。參入於伊勢齋宮。” (“Subete no Itsuki-no-Naishinnō sadamite owaru.

Sunawachi kyūjō no uchi no binjo wo urabite. Shosaiin to nasu. Misogi shite hairu. Myōnen no fuzuki ni itaru.

Kono miya ni itsuku. Sara ni jōgai no kiyono wo urabite. Nomiya tsukurite owaru. Hazuki no jōjun. Yokihi wo urabite sadamu. Kawa nozomite misogi su. Sunawachi nomiya wo hairu. Utsuri hairu hi yori. Myōnen no hazuki ni itaru. Kono miya ni itsuku. Nagatsuki no jōjun. Yokihi wo urabite sadamu. Kawa nozomite misogi su. Ise

Saigū wo mairi hairu.”)

223 saiin, this lasted three years (Engishiki 6.Jingi 6)253. Even the speech of the princess was constrained in this quest for purity, with the saiō observing several linguistic taboos throughout her term of service. Words relating to ritual contaminants were forbidden, as were references to Buddhism, with euphemistic replacements for each: “sweat” for blood, “dyed paper” for Buddhist sutras, and so on

(Engishiki 5.Jingi 5)254.

The saigū lived isolated from her family in distant Ise, often separated from them at a very young age, and many died in office before ever returning home. The Kamo saiin suffered many of the same disadvantages as her saigū sister, although she, at least, was not so isolated from her family or the court social scene. Some saiin – such as the famous “Great Saiin” (大斎院), Princess

Senshi (選子), an acclaimed poetess who served the impressive term of 975 to 1031 – made noblewomen’s literary salons of their compound, an option permitted by their proximity to the court and unavailable to the distant saigū255. During her term as saiin, Uchiko was visited by her father at

253 “凡定齋王畢。卽卜宮城內便所。爲初齋院。卽先臨川頭。祓潔乃入。…凡齋王於初齋院三年齋。畢

其年四月始將參神社。先擇吉日。臨流祓禊。” (“Subete no Saiō sadamite owaru. Sunawachi kyūjō no uchi no binjo wo urabite. Shosaiin to nasu. Sunawachi mazu kawa no hotori ni nozomite. Misogi shite hairu…Subete no Saiō Shosaiin ni mitose itsuku. Owarite sono toshi no uzuki hajime ni masa ni jinja ni mairan to su. Mazu yokihi wo erabite. Nagare ni nozomite misogi su.”)

254 “凡忌詞。內七言。佛稱中子。經稱染紙。塔稱阿良良岐。寺稱瓦葺。僧稱髮長。尼稱女髮長。齋稱

片膳。外七言。死稱奈保留。病稱夜須美。哭稱鹽垂。血稱阿世。打稱撫。宍稱菌。墓稱壤。又別忌詞。

堂稱香燃。優婆塞稱角筈。” (“Subete no imikotoba. Uchitsu nanatsu koto. Hotoke wo nakago to shōsu. Kyō wo somegami to shōsu. Tō wo araragi to shōsu. Tera wo kawarafuki to shōsu. Sō wo kaminaga to shōsu. Ama wo me-kaminaga to shōsu. Imoi wo katajiki to shōsu. Totsu nanatsu koto. Shi wo naoru to shōsu. Byō wo yasumi to shōsu. Naku wo shiotare to shōsu. Chi wo ase to shōsu. Utsu wo natsu to shōsu. Shishi wo kusahira to shōsu. Haka wo tsuchikure to shōsu. Mata betsu no imikotoba. Dō wo koritaki to shōsu. Ubasoku wo tsunohazu to shōsu.”)

255 Tanaka 1996, 68; Emura 2009, 115; Tokoro 2017, 43.

224 least once, although the fact that he wrote a poem expressing his joy over the occasion suggests that it was not a common occurrence (Nihon Kōki 30.Kōnin 14.2.Kichū)256.

Recent decades have seen an increasing body of work reclaiming the emotional subjectivity of the saiō257: her feelings about her post, her joys and sorrows, her experiences and lived history,

“reading against” the objectified passivity with which she has been treated both outside of the

“golden age” narrative and within. In the remaining poetry of the saiō, we may catch a glimpse of their thoughts and feelings of individuals. Through the Chinese-style poetry of Uchiko, we may glimpse her loneliness and the sense of isolation produced by her ritual seclusion258 (Kamo Chūshin

Zakki 4.Kasuga sansō shi)259; through the works of the “Great Saiin” Senshi, we may see her sorrow at the inability to express her deep Buddhist faith, an alienation enforced by linguistic taboos260:

Though I think about it, it is taboo, a thing not to be said,

and so all that I can do is turn in that direction and weep.261 (Kamo Chūshin Zakki 4.Shikashū

12; trans. E. Kamens.)262

256 “幸无品有智子內親王山庄。上欣然賦詩。群臣獻詩者衆。” (“Muhin Uchiko-no-Himemiko no sanshō wo yuki su. Kinzen shite fushi wo ageru. Kunshin shi woba moromoro kōzu.”)

257 Kamens 1990, 9, 15; Tanaka 1996, 68–71; Faure 2003, 295–9; Tokoro 2017, 42–3, 131–2.

258 Kamens 1990, 15–6.

259 “寂々幽莊迷樹裏、/仙輿一降一池塘/棲林孤鳥識春澤、/隱澗寒花見日光、/泉聲近報新雷響/山色高

晴陰雨行/從此更知恩顧涯、/生涯何以答穹蒼。” (“Sekiseki taru yūsō ju no uchi ni mayō,/Senyo hitotabi kudaru ichi chitō/Hayashi ni tsumu kochō shuntaku wo ori,/Tani ni kakururu wo kanka nikkō wo miru,/Sensei chikaku hōjite shinrai hibiki/Sanshoku takaku harete in’u tsuranaru/Kore yori sara ni shiru onko no atsui wo,/Shōgai nani wo motte ka kyūsō ni kotaen.”).

260 Kamens 1990, 17; Ambros 2015, 64; Tokoro 2017, 126–30.

261 Kamens 1990, 16.

225

Such works depict the ways in which saiō themselves felt keenly the constraints of their position.

The temptation, therefore, may be to see the saiō as a sort of fallen Himiko, stripped of the authority and personal charisma that went alongside her seclusion, and moved from the palace to the ancestor-shrine at Ise, subordinating her to the paramount. The eminence of the saiō, in this interpretation, may be read as a legacy of the role’s hime-progenitrix, a cultural significance remaining in the paramountcy’s kin-priestess even as her prior rulership and ritual functions are gutted.

But this assumes that the gulf between the saiō’s symbolic importance and her own lack of authority is “unnatural,” in a sense, existing as a conflict between earlier female empowerment and later patriarchal disenfranchisement. Let us consider, instead, that the role of saiō was, from the outset, that of “living regalia”263, impressing imperial might upon the provinces through her presence, generating power for the monarch’s sake. The saiō holds significance as a symbol of divine monarchy, of dynastic might; as an individual, she is objectified, her autonomy and personal desires sacrificed to furnish her with the public image of inviolable ritual purity. She is furnished with little to no temporal authority, for her role is inherently passive; her individual performance is irrelevant for this is a role to be occupied, more than fulfilled.

The passivity of the saiō, as with other key aspects of her role – her location in Ise and Kamo, her ritual calendar, her associated myth-history – must be examined within the context of the program of Yamato political and religious centralisation that created her. To remove her from this political background, with the intent of creating a detached, linear narrative of “sacral female

262 “おもへどもいむとていはぬことなれば/そなたにむきて音をのみぞなく” (“Omoedomo imu to te iwanu koto nareba/Sonata ni mukite ne wo nomi zo naku.)”)

263 Ellwood 1967, 43; Tanaka 1996, 103.

226 decline” is to stymie full exploration of the role, its history, and its relationship to “power” in various forms.

5.4. The Buddhist Nun as Hime

So far, we have examined the positioning of various priestess roles within a framework of himehiko dynamics. But miko, kannagi, saiō, and other kami-worshipping priestesses were not the full extent of women’s religious roles in ancient Japan. To close out our analysis of the religious women of the ritsuryō era and their relation to a supposed “golden age” of female power, let us briefly explore the place of female ritualists within the newly-arrived Buddhist faith. Through doing so, we will observe that these female ritualists, and their relationships to both institutional and unofficial power, are more complex than can be encompassed within a simple narrative of

(himehiko-derived) glory-to-decline.

A quick note must be made regarding terminology in this section. The term “nun” (尼, ama or ni) was used very broadly to refer to a range of women ascetics and world-renouncers (出家者, shukkesha) across time and place, both within and outside of “official” systems of ordination264. As with the miko (see Ch. 5.2.3 above), the “nun” was not so much a singular female religious position as it was a broad range of female Buddhist roles and practitioners, including state-sponsored, officially ordained conventuals (官尼, kanni); privately or unofficially-ordained, home-dwelling

“widow-nuns” (後家尼, goke-ama); and a variety of mendicants, itinerant fund-raisers, and beggar- nuns265.

264 Groner 2002, 69; Katsuura 2002, 119.

265 Groner 2002, 67; Katsuura 2002, 109.

227

Within the “official” conventual system, a number of vinaya-derived categories existed: bikuni (比丘尼, from Sanskrit bhikṣuṇī), referring to a fully-ordained nun; shikishamana (式叉摩那, from śikṣamāṇā), referring to a postulate/probationary nun; shamini (沙弥尼, from śrāmaṇerī), referring to a novice nun; and ubai (優婆夷, from upāsikā), referring to a lay devotee266. Even these categories, however, could be variable depending on context: privately-ordained nuns might be referred to as shamini regardless of actual precepts taken, for example, and these categories were rarely delineated outside the official monastic system, with phrases such as “nun” or “world- renouncing” (出家, shukke) not specifying precepts taken or actual religious practice or lifestyle267.

When we discuss Buddhist nuns, we must keep in mind that we are referring to a large and disparate group of religious women, whose practices, lifestyles, and social standings could be highly variable across time and class, and who existed both within and without formal temple hierarchies268.

5.4.1. Nuns and Monks as Replication of Hime and Hiko

In 552, King Seongmyeong (聖明王) of the Korean kingdom of Baekje made a gift for the

Yamato court of several sutras and a statue of Shaka (釈迦; Śākyamuni, Siddhārtha Gautama,

Gautama Buddha). After heated debate amongst the retainer clans concerning the adoption of this foreign faith, the ministerial Soga clan – headed by Soga no Iname (蘇我稲目) – were charged with trialing its worship (Nihon Shoki 19.Kinmei 13.10). In 584, Iname’s son Umako (蘇我馬子), having succeeded his father as head of the clan and its religious practices, arranged for the ordination of

Japan’s first native monastics: the nuns Zenshin-ni (善信尼; lay name Kuratsukuri no Shima 鞍部嶋),

Zenzō-ni (禅蔵尼; Ayahito no Toyome 漢人豊女), and Ezen-ni (恵善尼; Nishikori no Ishime 錦織石

266 Meeks 2010, 6.

267 Groner 2002, 67–9.

268 Groner 2002, 85; Katsuura 2002, 110.

228

女) (Nihon Shoki 20.Bidatsu 13)269. Zenshin-ni was 11 at the time, a daughter of Soga retainer and

Chinese immigrant Shiba Datto (司馬達等); the other two nuns’ ages are not specified, but they are described as Zenshin-ni’s “disciples” (弟子, deshi), suggesting that they were of similar age, if not younger. The three novitiates were instructed by Hyepyeon (恵便; Eben), a retired immigrant monk from the Korean kingdom of Goguryeo (Nihon Shoki 20.Bidatsu 13).

The introduction of Buddhism remained controversial among the retainer clans, and the first nuns fell into the firing line of clan conflicts, suffering defrocking, imprisonment, and flogging at their tender age (Nihon Shoki 20.Bidatsu 14.3.Heijutsu). Eventually, however, the Soga won out, and the young nuns – having taken formal precepts in Baekje – were installed in Sakurai Temple (Nihon Shoki

21.Sushun Tennō Sokui Zenki Yōmei 2.6.Kasshi/Sushun 3.3). That same year, at least 12 more women became nuns, many themselves of Chinese or Korean descent (Nihon Shoki 21.Sushun 3)270. By the year 624, an official census recorded that Japan boasted 569 nuns among its clergy, alongside 816

269 “蘇我馬子宿禰請其佛像二軀。乃遣鞍部村主司馬達等。池邊直氷田。使於四方訪覔修行者。於是唯

於播磨國得僧還俗者。名高麗惠便。大臣乃以爲師。令度司馬達等女嶋。曰善信尼。〈年十一歲。〉又

度善信尼弟子二人。其一漢人夜菩之女豐女。名曰禪藏尼。其二錦織壺之女石女。名曰惠善尼。”

(“Soga no Umako-no-Sukune sono hotoke no mikata futahashira wo masete. Sunawachi Kuratsukuri-no-Suguri

Shiba Tatsuto to. Ikenobe-no-Atai Hita wo tsukawashite. Yomo ni shite okonai hito wo toimotomeshimu. Koko ni tada Harima-no-Kuni ni kaeri-no-mono wo eru. Na wa Koma no Eben. Ōmi sunawachi motte norinoshi to shite. Shiba Tatsuto ga musume Shima wo ieseshimu. Zenshin-no-Ama to iu. (Toshi tōmari-hitotose.) Mata

Zenshin-no-Ama no deshi futari wo iete. Sono hitori wa Ayahito no Yabo ga musume Toyome. Na wo Zenzō-no-

Ama to iu. Sono futari Nishikori no Tsufu ga musume Ishime. Na wa Ezen-no-Ama to iu.”)

270 “度尼大伴狹手彥連女善德。大伴狛夫人。新羅媛善妙。百濟媛妙光。又漢人善聰。善通。妙德。法

定。照善。智聰。善智惠。善光等。” (“Iedeseru ama Ōtomo no Sadehiko-no-Muraji ga musume Zentoku.

Ōtomo no Koma-no-Iroehito. Shiragi-hime Zenmyō. Kudara-hime Myōkō. Mata Ayahito Zensō. Zentsū. Myōtoku.

Hōjō. Shōzen. Chisō. Zenchie. Zenkō-ra.”)

229

Buddhist monks (Nihon Shoki 22.Suiko 32.9.Heishi)271. Although male monastics quickly came to outnumber female, the gender discrepancy seen within early Japanese Buddhism is minor in comparison to that of later eras, or of other Asian countries, where the role of the nun was much reduced272. Within Takamure’s “golden age” narrative, the selection of women for the country’s first monastics has been framed as a reflection of the power once held by spiritual women, and the relative prominence of the early Buddhist nun in Japan as a replication of himehiko dynamics, transposed onto the new male-female pair of monk and nun273.

Monks and nuns have, in early Japanese records, been configured as a paired set. In 741,

Emperor Shōmu unveiled a new undertaking: the installation of a court-sponsored monastery and nunnery in each province. These provincial monasterys (kokubunji274) and nunneries (kokubunniji275) existed to hold ceremonies praying for the protection of the state, the province, and the emperor, and, much like the Jingikan, existed as part of a national, centralised religious regime under court oversight (Shoku Nihongi 14.Tenpyō 13.3.Itsushi). The nunnery and the monastery are treated as a pair, state Buddhism explicitly requiring and sponsoring both male and female monastics

271 “校寺及僧尼。具錄其寺所造之緣。亦僧尼入道之緣。及度之年月日也。當是時。有寺四十六所。僧

八百十六人。尼五百六十九人。幷一千三百八十五人。” (“Tera oyobi hōshi ama wo shirabete. Tsubusa ni sono tera tsukureru koto no moto. Mata hōshi ama no okonau yoshi. Oyobi iedeseru toshi tsukihi wo shirusu.

Kono toki ni atarite. Tera yoso-amari-mutokoro. Hōshi yao-amari-tō-amari-mutari. Ama itsuo-amari-muso- amari-kokonotari. Narabite chi-amari-mio-amari-yaso-amari-itsutari ari.”)

272 Takamure 1966b, 219; Faure 2003, 29.

273 Takamure 1966b, 219.

274 国分寺; formally called the Temple for the National Protection of the Four Heavenly Kings of the Sublime

Golden Light (金光明四天王護國之寺, Konkōmyō-Shitennō-Gokoku no tera).

275 国分尼寺; formally called the Temple of Atonement of the Lotus of the Dharma (法華滅罪之寺, Hokke-

Metsuzai no tera).

230

(collectively known as 官僧尼 kansōni)276 working towards the divine protection of the realm.

Similarly, monks and nuns were invited to the palace together for sutra readings and meditation retreats, carrying out similar tasks277; alms were granted to both in apparently equal measure (Nihon

Shoki 25.Hakuchi 2.12.Tsugomori; Nihon Shoki 29.Tenmu 9.10.Itsushi/12.Summer/14.4.Kōshi; Shoku

Nihongi 1.Monmu 4.10.Jinshi). The term “monks and nuns” (僧尼, sōni) appears frequently within the texts, grouping the two together as a set.

However, this pairing is not necessarily a reflection of himehiko dynamics, as the “golden age” narrative understands them. We have seen that recent scholarship favours the idea of contrapuntal male-female performance as a common ritual format in kami-worship, such as the Miare rite of are- otoko and are-otome. Japanese Buddhism was, in its early stages, still heavily influenced by the lens of native religious traditions278. The early nuns may have in their duties more resembled native priestesses worshipping this new deity than continental nuns279, a syncretic amalgamation of local and foreign tradition incorporating aspects of native ritual performance into the imported faith. In other words, the configuration of monks and nuns as a paired set may reflect specific ritual traditions of kami-worship, which have in the “golden age” narrative been brought under the aegis of “himehiko”; but this is not the same thing as reflecting himehiko as a whole.

Furthermore, while kokubunji monks and kokubunniji nuns were portrayed as a set collectively, they were not paired on an individual level. As we saw in the 624 census results above, monks were consistently and considerably more numerous than nuns; nuns were relatively prominent, in comparison with mediaeval Japan and with other countries, but they were by no means equal to monks in number or stature. Although kokubunji were paired as institutions with

276 Groner 2002, 67.

277 Bowring 2005, 80.

278 Ōsumi 2002, xxxi; Bowring 2005, 22.

279 Faure 2003, 29; Ambros 2015, 44.

231 kokubunniji, the two were not in balance, 20 monks to each kokubunji compared with only 10 nuns per kokubunniji. Monasteries also received the lion’s share of financial support, although presumably the personnel differences account for at least some of this discrepancy (Shoku Nihongi

14.Tenpyō 13.3.Itsushi; Shoku Nihongi 17.Tenpyō 19.11.Kibō). Monastic codes subordinated nuns to monks280, and positions of power in the Office of Monastic Affairs (僧綱, Sōgō) were typically held by men. Although conceptually the two came as a set, the monks were more emphasised, more numerous, and higher in the temple hierarchy than their female counterparts. This disparity is similar to that noted between Jingikan priests and jinja-miko (see Ch. 5.2.1).

It is hardly outlandish to suppose that the relationship between early nuns and monks reflected broader sacral gender dynamics of ancient Japan. However, the notion that this relationship was a specific replication or transformation of the himehiko system – and thus, serves as a form of proof as to the latter’s existence – is complicated by two factors. The first issue is the discrepancy in numbers: theories of himehiko have always focused on the partnership of individual men – priests, kings – with individual women – shamans, queens. Although monks and nuns, in the plural sense, seem to have been envisioned as a set, a himehiko-style partnership on the individual level is highly unlikely given the discrepancy of numbers; the kokubunji monks, for example, outnumbered the kokubunniji nuns two-to-one by design. Had monk and nun been seen as equal complements, one would expect attempts to maintain a more even balance. The second is the broader issues of homogenisation and conflation present within the “golden age” conceptualisation of himehiko as a whole (see Ch. 4.3). The configuration of nuns and monks as a pair in the plural sense may relate to specific ritual traditions of contrapuntrality without necessarily connecting with ancient formats of rulership or gendered social roles; “himehiko” only in the loosest sense.

280 Nagata 2002, 284; Faure 2003, 23–4; Meeks 2010, 21–2.

232

5.4.2. Nuns and Power: A Brief Note

The position of nuns within Japanese Buddhism is a complex subject, with considerable fluctuations over time, between different schools, and on the boundaries of “official” versus

“unofficial” ordinations. A complete study of the nuances of nuns’ relationships to power is well beyond the scope of this research. Let us briefly address, however, some of the tensions of power occurring within the early nun as they relate to other priestessly positions; through this, we may further explore the insufficiencies of the “decline” narrative of the female ritualist in accounting for these complexities.

The question of individual agency has been raised in the case of the first Soga nuns. Japan’s first monastics may have been female but, as the Nihon Shoki points out, Zenshin-ni was only 11 at the time of her ordination; the other two possibly even younger (Nihon Shoki 20.Bidatsu 13). Given their youth, their agency in their adoption of a monastic lifestyle, or within the clan conflicts that followed, is in considerable question. Faure and Ambros argue that, like the child-saiō, they were placed into service by their families for the sake of clan politics; another sacral sacrifice, who suffered flogging and imprisonment as part of the interclan conflict over the new religion281 (Nihon

Shoki 20.Bidatsu 14.3.Heijutsu). A later text, the Gangōji Garan Engi282, holds that the girls, already students of an elderly immigrant nun, requested ordination of their own accord; Zenshin-ni is here

17 rather than 11, an age with somewhat more possibility of autonomy. This version also depicts solidarity between religious women with Suiko (then Bidatsu’s Empress Consort) acting as their sponsor283 (Gangōji Garan Engi 1)284; as a later work, however, its reliability as a source is in some

281 Faure 2003, 29; Ambros 2015, 42.

282 元興寺伽藍縁起, “Origins of the Gangōji Monastery.”

283 Ambros 2015, 44–5.

284 “但是時針間國。有脱衣高麗老比丘名惠便與老比丘尼名法明。時按師首達等女斯末賣年十七在。阿

野師保斯女等已賣。錦師都瓶善女伊志賣合三女等。就法明受學佛法在。倶白。我等爲出家。欲受學佛

233 question. In the earlier accounts from the Nihon Shoki, though, the young nuns are portrayed as child-priestesses sequestered for the sake of clan politics, whose activities served the interests of the male heads of family rather than empowering the girls themselves285; kokubunniji nuns were older, but, like the Jingikan priestesses or the saiō, their rites similarly revolved around service to the state and to the patriline of emperors286.

The question of individual agency is also debated as regards later nuns. Meeks argues that scholarship has tended far too much towards portraying nuns as passive figures in their own ordinations, stripping them of agency; instead, sheand Faure contend, the role of nun could afford a woman a certain degree of autonomy, as patriarchal family structures took fuller hold, freeing her from the demands of wifedom and motherhood287. Katsuura notes that the diverse nature of ancient nuns, and the variable reasons they had for taking tonsure, bely any simple categorisation of the nun and her agency. For example, she argues, voluntary renunciation could indeed allow a woman to release herself from the expectations of female sexuality, but forcible tonsuring was also used as a way to punish sexually wayward women, stripping them of their sexuality and sexual rights and

法白。大臣即喜令出家。〈嶋賣法名善信。等己賣法名禪藏。伊志賣法名惠善。〉爾時大臣大ゝ王池邊

皇子二柱歡喜請櫻井道揚令住。” (“Tada kono toki ni Harima-no-Kuni ni. Datsue no Kōrai no oi biku na wa

Eben to oi bikuni na wa Hōmyō ari. Toki ni Kuratsukuri-no-Obito Tatsuto ga musume Shimame toshi tōmari- nana ari. Aya no Hoshi ga musume Toshime. Nishikori no Tsufura ga musume Ishime awasete mitari no musume-tachi. Hōmyō ni tsukite butsuhō wo uke manabite ari. Tomo ni mōsaku. Warera shukke shite. Butsuhō wo uke manaban to omō to mōshiki. Ōomi sunawachi yorokobite shukke seshimu. (Shimame wa hōmyō

Zenshin. Toshime wa hōmyō Zenzō. Ishime wa hōmyō Ezen.) Shikari shite toki ni ōomi ōōkimi Ikenobe-no-Miko no futahashira yorokobi koite Sakurai no dōyō ni sumawashimeki.”)

285 Ambros 2015, 42–3.

286 Piggott 1997, 217.

287 Faure 2003, 45; Meeks 2010, 12.

234 forcing them out of secular life and relationships288. Furthermore, class played a role: a woman who was supported by the economic stability of a home renunciation or a private chapel would have a very different experience of nunhood than those women to whom these avenues were not open, and who might therefore find themselves in the precarious position of itinerant, or else reliant on the male monastic community for support289. Nuns are thus perhaps best understood as representing a range of experiences and relationships to agency and autonomy. Pandey’s critique of

“agency” as a concept within Anglophone women’s history is also salient here: the reduction of taking tonsure into an act of empowerment versus one of subjugation (as these concepts are understood in modern feminist thought) is a reductive one which risks projecting modern Western ideologies of female liberation onto the actions of ancient women, rather than understanding those women’s own inner lives, perspectives, and thoughts on their own experiences290.

Another complication to the linear narrative is presented by elite female lay patrons: rarely categorised alongside nuns in discourse on female religious positions, and yet a great increase in female patronage is contemporaneous with a lack of monastic support for “official” nun ordinations, this transformation of female involvement in Buddhism often overlooked in favour of simple tale of decline291. Katsuura Noriko (勝浦令子) similarly notes a tendency to focus analyses of the nun and her social standing predominantly on the state-sponsored, “officially”-ordained kanni and their decline, to the exclusion of the “unofficial”, extra-institutional nuns who maintained presence throughout the Heian period292, and Barbara Ruch details a conflation of various lineages of

288 Katsuura 2002, 125–6.

289 Katsuura 2002, 125–6.

290 Pandey 2020.

291 Faure 2003, 25; Meeks 2010, 99.

292 Katsuura 2002, 110.

235 mendicant-nun into a singular model of “Kumano bikuni” (熊野比丘尼)293. This emphasis on “official” status, and difficulty grappling with the religious experiences of women outside of hegemonic structures other than in terms of subordination, echoes the neglect and assumed marginalisation of the sato-miko and aruki-miko within the “golden age” narrative.

Finally, the Buddhist nun, like the saiō, has also been subject to objectification within scholarship, relegated to the passive role of damsel in distress, denied full expression of her subjectivity. Faure points out the complexity of Buddhist discourses on and attitudes towards women and gender: Buddhist doctrine, he notes, has certainly tended towards misogyny, but at the same time allows for considerable flexibility and even contradiction; to cast it as a purely oppressing force to female devotees fails to engage with this multiplicity294. Meeks has similarly pointed out the ways in which women – both nuns and elite female laity – engaged very differently with gender than monks and laymen, evincing their own understanding of their gender within Buddhism rather than passively accepting every sexist doctrine295. This multivocality is also invoked by Karen Derris, in a critique of scholarship of Buddhism presenting Buddhism as a monolithic institution and failing to account for the ways in which Buddhist women engage with their own faith296.

As Meeks, Derris, and Ambros have pointed out, the failure of many discussions of Buddhism and gender to engage with Buddhist women’s own opinions, readings of doctrine, and understanding of their place in the faith, erases women’s voices from the historical narrative and falls into sexist and Orientalist assumptions of Asian women as passive subjects of oppression

293 Ruch 2002.

294 Faure 2003, 9.

295 Meeks 2010, 7, 255.

296 Derris 2014, 61–2.

236 without opinion or agency of their own297. Nuns were women with their own agency, actors within a patriarchal system, and – like the saiō – their historical humanity is deserving of reclamation.

The deteriorationist “silver age” phase of the “golden age” narrative tells the story of decline: the fall of Japanese womanhood from the respected, spiritually and socially powerful hime of the “golden age” to the subordinated victims of ritsuryō patriarchy. The female religious roles of the later ancient period – the priestesses of the Jingikan, the itinerant miko, the princessly roles of saigū and saiin, the emerging orders of Buddhist nun – are placed along a linear chart of deterioration, all deriving from a single sacral female role, differentiated by time and the presence or absence of a broadly-defined notion of “power”.

But this narrative does not serve to encompass the full breadth of female experiences and relationships to spiritual and temporal power across the archipelago. We have seen how the “golden age” narrative’s coalescence of all female political and spiritual power into a single systematised figure – the “hime” – homogenises a diverse array of examples of female roles across various regions and points in time, and the tensions of power within. Furthermore, the construction of ancient Japan as a primeval space of “native” empowerment, untouched by the corrupting influence of foreign patriarchy, falls too easily into a nationalistic paradigm.

The presentation of all female ritualists under ritsuryō as decayed remnants of the hime continues this scholarly conflation. The specific historical/political/religious contexts shaping the roles and experiences of religious women are elided, their function within the “golden age” narrative reduced to an unfavourable comparison against an idealised construct of a female-centric past. This is evident in the “golden age” narrative treatment of the Ise saigū, her supposed descent from a primeval “hime” born out of the assumption that roles of female religious prominence could only

297 Meeks 2010, 12; Derris 2014, 66; Ambros 2015, 2.

237 exist during this time as a deteriorated form of a previous, greater role of female power, along a strict narrative of decline. This interpretation both fails to encompass the historical factors other than gender which go into the formation and shaping of these roles – such as, in the saigū’s case, her role within a broader Yamato program of regional control – and further does not account for the way in which the image of “ancient Japan” in the records was shaped to suit the needs of the later court.

Simple comparisons of “power” are further rendered inadequate by the breadth and complexity contained within the concept. Rather than an easily quantifiable trait, a clear notion of presence or absence, we have identified several types and expressions of power within our study of religious women (see Ch. 2.2.5). Within the “golden age” narrative, notions such as temporal authority, spiritual acuity, social standing, and individual autonomy are conflated into a single, amorphous notion of “power”; a framing which fails to encompass the tensions of “power”, the multifaceted and even conflicting ways in which various kinds of power can be held and expressed by women operating within various systems and social frameworks. Nor are these tensions and conflicts new aspects of womanhood under ritsuryō – the product of imported patriarchy, in a quasi- nationalist paradigm of the foreign-as-corrupting – but rather they appear in some form even in the earliest images of female religiosity and power that survive: Himiko, Jingū, Yamato-Hime, the doyennes of the “golden age.”

The image of the later miko – both the jinja-miko of the state cult and the unofficial sato- miko and aruki-miko – within the “golden age” narrative also presents this oversimplified image of

“power,” fixating on official bureaucratic status, legitimation in the eyes of court and aristocracy, as the single marker of a ritualist’s “power,” while failing to take into account the ways in which priestesses could hold status at a smaller, community level. Derisory attitudes of (some) aristocrats towards miko are enshrined as the opinion of not only the entire aristocracy, but Japanese society as

238 a whole; ancient Japan is taken as a homogenous society with court culture as its barometre, and bureaucratic status the only form of “power” there is.

Issues are also seen in the presentation of Buddhist nuns as passive figures, “damsels in distress” of patriarchal structures, without regard to how ancient women themselves interpreted the place of their gender within their Buddhist faith, or the ways in which nunhood could serve their goals. This oversimplification of the lives and statuses of female ritualists of the ritsuryō age, this failure to engage with the complexity of their relationships to “power,” overwrites their lived experiences as historical women with a simplified narrative of decay. Within the “golden age” narrative, they are rendered into passive figures, objectified symbols of female subordination. To truly engage with these women’s experiences, to reclaim their historical humanity, we must work towards a narrative framework which understands them within the historical and religious context that shapes them, which allows space for the complexities and tensions of various kinds of “power” associated with their position, and which also regards, on an individual basis, their inner subjectivities and emotional lives, engaging with them as actors within systems.

But priestesses (and nuns) only form one half of the story of decline, one branch of descendancy from the hime-construct posited within the “golden age” narrative. From here, we turn to the image of the jotei, or empress regnant, configured within “golden age” discourse as another transformation of the hime, a position deriving from and enabled by an earlier tradition of shamanic queenship in a distant, female-centric past. We shall see the tensions of power that exist also within the figure of the jotei, how various empresses interacted with, worked within, or even defied the systems of power surrounding them, and the further simplifications and objectification of women under patriarchy contained within their “golden age” narrative framing.

239

CHAPTER 6: EMPRESSES IN THE “GOLDEN AGE” NARRATIVE

The period of Japanese history from 593 to 770 is sometimes referred to as the “Century of

Empresses,” for it saw six women ruling as tennō for a total of eight reigns, a frequency of female monarchs not seen before or since. Suiko, ruling 593 to 628, stands as the first officially recognised empress regnant, followed by the twofold reign of Kōgyoku-Saimei (r. 642–645, 655–661), Jitō (r.

687–697), Genmei (r. 707–715), Genshō (r. 715–724), and finally Kōken-Shōtoku (r. 749 – 758, 764–

770). After Kōken-Shōtoku’s death, Empresses Regnant became vanishingly rare: Meishō (明正), the seventh, would not take the throne until the 17th century; Go-Sakuramachi (後桜町), the eighth and final, in the mid-18th.

The “golden age” narrative framing of these jotei 1in many ways resembles that of the saigū, or the Buddhist nun: as a transformation of the hime, a remnant of once-widespread female rulership, the last gasp of an ancient “power of women” in the face of the burgeoning patriarchy2.

Contained therein is the assumption that positions of female sacral power could only exist as decayed legacies of an earlier, greater role.

The notion of jotei as a transformation of an ancient, specifically female rulership position is, however, complicated not only by the uncertain historicity of the hime (see Ch. 4.3) but also by the gendering of the tennō, and the rhetoric surrounding it. For the most part, the jotei of the Century of

1 Japan’s empresses regnant held the same monarchic title – tennō/sumera-mikoto (天皇) – as their male counterparts; within Japanese-language scholarship, the terms jotei (女帝) or josei-tennō (女性天皇) are used to delineate them. In Anglophone scholarship the term “female emperors” is sometimes used in place of

“empresses regnant” to reflect this titulary (Tsurumi 1982, 71).

2 E.g. Takamure 1966b, 112–7; Ōi and Miyagi 1974, 17; Tsurumi 1982, 72; Sekiguchi 1987, 19–20; Piggott 1994,

15–6; Matsumura 1999, 125–7; Naoki 2008, 205–8.

240

Empresses adopt the same self-presentation as their masculine counterparts, the script of the tennō remaining largely identical. Rather than drawing on some ancient tradition of hime, some special notion of female rulership distinct from the male, the gender of the jotei appears as something requiring careful negotiation, even justification. When a jotei’s gender is leveraged in support of her rule, this is done not through appeal to past images of female power, but through tropes of feminine virtue which display her devotion to the men of the imperial clan, and thus, her utility to the patriline.

The sacral performance of the female tennō further shows little connection to notions of the hime’s ritual role, or indeed, even to images of a female contrapuntal ritual role. “Golden age” scholarship on the sacral performance of jotei can fall into the trap of interpreting an empress’s actions through the preconceived lens of her as a remnant of an ancient “shaman queen,” reading particulars of her reign (such as Jitō’s trips to Yoshino) as traces of shamanic practice which may then be used as proof of shamanic origin. The jotei also evince no special connection to or identification with her other alleged hime-descendant, the Ise saigū: while a female tennō could use the image of Amaterasu as a symbol of power by subtly likening herself to the goddess, her worship of the ancestral deity was otherwise largely similar to that of a male emperor, her gender providing no special function. Indeed, jotei and saigū could comfortably co-exist as roles. We shall see instead that the jotei’s religious performance, and use of religion as a source of legitimation, largely follows that of the male tennō, holding the same ritual role, and promulgating similar discourses of divine will, omens of divine approbation, and imperial deification. In the sacral realm as well, jotei legitimate themselves through replicating the (masculine) image of kingship of the tennō, rather than drawing on female-gendered traditions of sacral queenship.

Jotei similarly tended to deploy the newer faiths – Buddhism and Confucianism – as sources of legitimation in much the same way as male tennō, regardless of any doctrinal issues their gender might be expected to cause. Jotei presented themselves in the image of the Buddhist saviour-king, or

241 cakravartin, much as male emperors did, despite androcentric doctrines which blocked women from that position. Similarly, female tennō promoted Confucian notions of filial piety and obedience to hierarchy, as male tennō did, to support their rule, while simply not engaging with concepts such as the Three Obediences which preached female subordination. Not only did jotei follow a “male” script of rulership identical to their masculine counterparts (rather than the uniquely female script implied by the “empress-as-hime” notion), but in some sense their position as tennō discursively overrode their femininity, delivering them a sort of monarchic androgyny. Where the jotei did acknowledge their gender, and use religious tradition to justify it, the images they selected were not ancient shaman-queens of native tradition, but Buddhist texts portraying wise queenship, or discussing androgyny and the transient nature of gender. Legitimation, for jotei, came not so much from an ancient, empowered “golden age” of female rulership, but rather from a careful negotiation of their gender in relation to their position.

Ultimately, the image of the jotei in the records does not suggest a holdover of an ancient custom of female rulership, so much as a distinct role arising within the patriarchal framework, and requiring considerable negotiation of gender; akin to, and possibly incorporating elements of, the

Chinese dowager regencies. However, this does not mean that jotei were passive figureheads; rather, they were agents within a system, and even when intended as placeholders were capable of leveraging their monarchic powers to considerable effect. Rather than being a transformed remnant of the hime, a holdover from a past “golden age” in an era of patriarchal decline, the jotei of the

“Century of Empresses” present as a distinct structure of power for (a limited subset of) women, navigating their way through the complexities of the patriarchal social order in which they lived.

242

6.1. Empresses as Hime: Traditions of Female Rule

Japan’s “Century of Empresses” stands out amidst the kingdoms of ancient East Asia.

Although several Chinese dynasties boasted a tradition of “dowager regencies” – in which a widowed empress consort would serve as regent for a child-emperor – in two millennia of empire,

China saw only one official empress regnant: Wu Zetian3, whose reign took place nearly a full century after Suiko. The Korean kingdom of Silla preceded this slightly with Queen Seondeok4 (善德

女王) and Queen Jindeok5 (眞德女王), and later Queen Jinseong6 (眞聖女王), but through the

“Century of Empresses” alone ancient Japan displays both the greatest number and the earliest examples of female monarchs in the region7, to say nothing of the figures of Himiko and Iyo/Toyo.

The six empresses from Suiko to Kōken-Shōtoku have thus drawn considerable interest from scholars, who seek to understand what this period implies about the relationship between women and monarchic power in ancient Japan.

In the “golden age” narrative, the traditional framing of this period is as a transformed manifestation of himehiko rulership. The empress regnant becomes a latter-day hime, her accession enabled by ancient traditions of female leadership back in the “golden age” persisting against the inexorable encroachment of patriarchy, a last holdout of women’s ancient authority8. The jotei-as- hime-remnant is differentiated by the saigū-as-hime-remnant by her retention of the hime’s proposed co-rulership function, compared to the sacral function of the saigū; this split of duties

3 R. 690–705 CE; also called Wu Zhao (武曌)

4 R. 632–647 CE

5 R. 647–654 CE

6 R. 887–897.

7 Yoshie 2007, 209.

8 Takamure 1966b, 112–7; Ōi and Miyagi 1974, 17; Tsurumi 1982, 72; Sekiguchi 1987, 19–20; Piggott 1994, 15–

6; Matsumura 1999, 125–7; Naoki 2008, 205–8.

243 presumably enabling within the narrative their contemporaneous existence. The jotei is also taken as representative of a later manifestation of himehiko dynamics, in which hime and hiko are configured as husband and wife rather than natal kin; the “Century of Empresses” a remnant of himehiko co- rulership which generally placed the consort-hime below the emperor-hiko in status but allowed her to assume his duties should he predecease her9.

Takamure interprets the imperial consort, or kisaki (后/妃), in this vein of latter-day hime, representing a turning point in the himehiko system. Consorts, she explains, replaced kinswomen as the male ruler’s contrapuntal partner, which shifted the power dynamics of the pair in the male’s favour. Takamure further argues that consorts would have had a sacral role, positing an etymology of kisaki as deriving from hisaki (日前), or “before the sun,” casting the consort-hime as a solar priestess much as Takamure claims for the kinswoman-hime10. The kisaki, she explains, still maintained some power – as shown in her ability to assume the imperial seat of a deceased hiko11 – even though it was more in the shadows than her that of her predecessors, and, like Takamure’s model of the hime of old, it was largely focused in religious performance and ritual; the end of the

Century of Empresses marked the end of monarchic himehiko in the face of new imperial patriarchy12. Sekiguchi, too, views the Century of Empresses as a period of flux, where consorts who typically served as hime-like “secondary monarchs” could still occasionally take the throne, should political circumstances make a male candidate untenable13. Piggott, too, names the consort as a kind of hime, subordinate but nevertheless holding religious authority and capable of continuing her rule

9 Takamure 1966b, 112–6; Sekiguchi 1987, 19–20; Piggott 1994, 16.

10 Takamure 1966b, 112.

11 Takamure 1996b, 114–6.

12 Takamure 1966b, 117.

13 Sekiguchi 1987, 19–20.

244 past the death of her mate14. The notion of the Century of Empresses as a transformation of himehiko is further echoed by Matsumura15 and Naoki16, the latter of whom even posits the acceptance of female rulers during this time as possible remnant of a primeval matriarchy.

This theory of empress as consort-hime brings to mind the legendary reign of Empress Jingū

(see Ch. 4.2): a helpmeet-empress, of imperial ancestry herself, stepping into the primary monarchic role upon her paramount-husband’s untimely death. As we have seen, some records clearly titled her as an empress regnant (Hitachi Fudoki, Ibaraki-gun 1; Fusō Ryakuki 1) despite the Kojiki and

Nihon Shoki’s insistence that her term of power was a regency, not a true reign. Jingū has been posited as a clear example of the consort-hime, a precedent to the “Century of Empresses” written out of the official imperial line17. However, this reading rests on the assumption of Jingū as a factual historical figure, possessed of a “true nature” outside the Yamato narrative; an assumption that, as discussed in Chapter 4, is very much in question. Furthermore, the tennō/sumera-mikoto title with which she is endowed in the Hitachi Fudoki and Fusō Ryakuki is itself an anachronism, not in use for emperors until the seventh century and applied to earlier monarchs retroactively18. Its appearance in records shows that the authors conceived of Jingū as an empress regnant, but this may simply reflect a different translation/modernisation of the generic respectful suffix of -no-Mikoto, as is used for

Jingū in the Kojiki.

14 Piggott 1994, 16.

15 Matsumura 1999, 127.

16 Naoki 2008, 205–8.

17 Takamure 1966b, 114–5, Matsumura 1999, 125–6; Naoki 2008, 208.

18 Yoshikawa 2011, 30.

245

One more potential precedent of female rulership within the Yamato dynasty19 may be found in the enigmatic figure of Princess Iitoyo (飯豊), also called Aomi (青海) or Oshinumi (忍海).

Iitoyo is somewhat a reverse of Jingū’s situation, hailing from a more reliable period of the chronicles but with very little written about her. In the Kojiki, she is named as the daughter of

Emperor Richū20 (履中) and Kurohime (黒比売) (Kojiki 3.Richūki); the Nihon Shoki repeats this, but elsewhere claims her to be Richū’s granddaughter, sister to Emperors Kenzō21 (顕宗) and Ninken22

(仁賢) (Nihon Shoki 12.Richū 1.7.Jinshi; Nihon Shoki 15.Kenzō Tennō Sokui Zenki). In both the Kojiki and Nihon Shoki accountings, the line of succession is in confusion following the death of Emperor

Seinei23 (清寧), whereupon Iitoyo appears to take the throne for a short time. In the Nihon Shoki accounting, it is more clear that she assumes imperial authority, “holding court” (臨朝秉政, mikado no matsurigoto shitamau) and calling herself by the regnal name of Oshinumi-no-Iitoyo-no-Ao (忍海

飯豊青); however, her term only lasts for half a year before she passes away (Kojiki 3.Seineiki; Nihon

Shoki 15.Kenzō Tennō Sokui Zenki)24.

19 Excluding, of course, the examples of Himiko and Iyo/Toyo, whose relationship to the Yamato dynasty and its practices remains unclear.

20 Reign traditionally dated 400–405 CE

21 Reign traditionally dated 485–487 CE

22 Reign traditionally dated 488–498 CE

23 Reign traditionally dated 480–484 CE

24 Kojiki: “故天皇崩後。無可治天下之王也。於是問日継所知之王。市邊忍齒別王之妹。忍海郎女。亦名

飯豊王。㘴葛城忍海之高木角刺宮也。” (“Kare sumera-mikoto kamusarimashite nochi. Ame-no-shita shirasubeki miko mashimasazariki nari. Koko ni hitsugi shiroshimesan miko wo tō ni. Ichinobe-no-Oshiha-Wake- no-Miko no imo. Oshinumi-no-Iratsume. Mata no mina wa Iitoyo-no-Miko. Kazuraki no Oshinumi no Takaki no

Tsunosashi-no-Miya ni mashimashiki nari.”); Nihon Shoki: “白髪天皇崩。是月。皇太子億計王與天皇讓位。

久而不處。由是天皇姊飯豐青皇女於忍海角刺宮臨朝秉政。自稱忍海飯豐青尊。…冬十一月。飯豐青尊

246

Iitoyo has been variously interpreted as an empress regnant or a short-term regent. In the

Kojiki, she is depicted as a regent, appointed only the miko (王) title of a lesser prince or princess. In the Nihon Shoki, she is never referred to as tennō/sumera-mikoto; instead, she appears as a self- styled, “unofficial” empress, holding court and taking the mantle of leadership upon herself, her regnal name described as “self-appointed” (自称). In neither text is she given a section of her own.

Only in the Fusō Ryakuki is she listed as an “official” empress, a paragraph dedicated to “Itoyo-Tennō”

(Fusō Ryakuki 2). Within the “golden age” narrative, Iitoyo has been read as another “hidden empress,” a female ruler from the ancient tradition of such that allowed for the accession of Suiko, written out of the official Yamato patriline25.

Iitoyo, however, does not fit within a “consort-hime” interpretation. A brief passage in the

Nihon Shoki indicates that she may have been married, but states that this is unclear (Nihon Shoki

15.Seinei 3.7)26. Her predecessor, Seinei, was not her husband, nor was he closely linked to her in any way; even their shared familial relations were relatively distant, being either second cousins or first cousins once removed. Iitoyo’s role is not portrayed as systematic, but instead as an unusual access to regnal authority provided by the circumstance of a disrupted line of succession. Nor does she evince any ties to an idea of “himehiko” rulership, displaying only weak familial connections to the preceding monarch. While her role as an interim monarch legitimated by circumstance may

崩。” (“Shiraka-no-Sumera-Mikoto kamusaritamau. Kono tsuki. Hitsugi-no-miko Oke-no-Miko to sumera- mikoto to kurai wo yuzuri.Hisashikute etamawazu. Koko ni yorite sumera-mikoto ga irone Iitoyo-no-Ao-no-

Himemiko Oshinumi no Tsunosashi-no-Miya ni oite mikado no matsurigoto shitamau. Mizukara Oshinumi-no-

Iitoyo-no-Ao-no-Mikoto to nanoritamau…Fuyu no shimotsuki. Iitoyo-no-Ao-no-Mikoto kamusaritamau.”)

25 Ōi and Miyagi (1974), 17; Piggott (1994), 15–6; Naoki (2008), 205–6.

26 “飯豐皇女於角刺宮與夫初交。謂人曰。一知女道。又安可異。終不願交於男。〈此曰有夫未詳

也。〉 ” (“Iitoyo-no-Himemiko Tsunosashi-no-Miya ni oite maguwai shitamau. Hito ni katarite notamawaku.

Hitohashi omina no michi wo shiranu. Mata izukunizo kenaru bekemu. Oi ni otoko wo awan koto wo horiseji.

(Kore ni oto ari to ieru koto mada tsubahiraka narazu.)”)

247 certainly reflect on the later “Century of Empresses” – as we will explore later – her appearance within the records does not suggest a systematised tradition of ancient female rulership within the

Yamato dynasty.

The application of this “widow-hime” theory onto the “Century of Empresses” is also problematic for several reasons. The first is the results of our exploration of the base concept of himehiko (see Ch. 4.3): that the notion of a singular “himehiko system” conflates and homogenises a diverse array of male-female pairings within the records, and the systematised tradition of paired rule presented within the “golden age” narrative is not reflected in textual evidence. The second is that the framing of jotei as widow-hime succeeding their late hiko is also an oversimplification, the circumstances surrounding their accessions more varied and complex. Suiko was the widow of a previous emperor, Bidatsu, but not his direct successor, with two imperial reigns separating them.

Kōgyoku-Saimei’s first reign fits the proposed pattern, but her second accession complicates the issue: her male successor Kōtoku27 (孝徳), whose death prompted her return to the throne, left behind a widow of his own – Kōgyoku-Saimei’s own daughter, Hashihito (間人) – and yet it was

Saimei who succeeded him, engineered by her son. Genmei was the mother, not the wife, of her male predecessor; her late husband Kusakabe (草壁), although nominated as Crown Prince, died before becoming emperor. Genshō and Kōken-Shōtoku never married at all, the former granted the throne by her abdicating mother Genmei, and the latter named as heir by her father, becoming

Japan’s first Crown Princess. Only Jitō’s accession neatly fits the notion of jotei as widow-hime stepping into the shoes of a late hiko; otherwise, there is little evidence for the hypothesised common role as co-ruler prior to first accession.

The example of Iitoyo shows that there was, within the ancient period, some potential provision for female rule depending on circumstance. However, this is not the same thing as an

27 R. 645–654 CE

248 ancient tradition of female rule which legitimates later empresses. The theories of connection between hime and jotei proposed within the “golden age” narrative are both reliant upon the questionable existence of a “himehiko system,” and also fail to cover the differing positions of the jotei prior to accession, which themselves appear circumstantially variable rather than systematised.

Furthermore, the notion of the empress regnant as a pre-existing tradition of female power is contradicted by the unique and complex ways in which a female tennō’s gender was negotiated, as we shall examine in detail now.

6.2. Gendering the Tennō

6.2.1. Empresses and Femininity

The theory of empress-as-hime positions the empress-regnant as a continuation or transformation of a native tradition of female leadership. The role of tennō, however, was typically occupied by a man. In order to best understand how the female tennō of the Century of Empresses factor in to the broader status of female leadership in ancient Japan – as representatives of an ongoing tradition, or as exceptional figures permitted by circumstance to occupy a role usually barred to them – we must examine the public images of these sovereigns carefully cultivated within court records, and how gender factors into these presentations: in recorded edicts, actions, and rhetoric surrounding their accessions. Did they link themselves to ancient traditions and images of female rule? Did they follow the same script of rulership as a male emperor, or construct something distinct to themselves? How was the gender of a female tennō reckoned in contemporary discourse?

Furthermore, as the hime model of governance often credits female authority to religious performance, we must look at the ways in which religion was deployed by female emperors, and how it differed, or did not differ, from the strategies of their male counterparts.

249

The Soga ministerial clan, lead by Soga no Umako, enjoyed considerable status during the reign of their maternal relative, Empress Suiko, who was Umako’s niece. But in 624, when Umako petitioned the crown for rulership of Kazuraki District, the Empress rebuffed him thusly:

“But if in our reign, we were to suddenly lose this territory, later rulers would say: ‘A foolish,

stupid woman ruled the realm, and suddenly threw away that territory!’…It would disgrace

us in the eyes of later generations.”28 (Nihon Shoki 22.Suiko 32.10.Kibō)

The attribution of the phrase “a foolish, stupid woman” (愚痴婦人, orokanaru me-no-ko) to Suiko here is interesting, in the emphasis it places upon her gender and the ways in which it may define her rule to future generations. For the most part, the edicts attributed to Suiko display no gendered difference from those of a male tennō: she uses the standard imperial first-person pronoun of 朕29 rather than a feminine pronoun such as 妾30, and is referred to as ōkimi (大王), “great king”; for the most part, she uses the same terms of reference as a male tennō, her monarchy in a sense overriding her femininity. The juxtaposition of this elusive reference to herself as a woman31 with

28 “然今當朕之世。頓失是縣。後君曰。愚痴婦人臨天下。以頓亡其縣。…是後葉之惡名。” (“Shikamo ima are no yo ni atarite. Hitaburu ni sono agata wo ushinaite wa. Nochi no kimi no notamawaku. Orokanaru me-no-ko ame-no-shita ni nozomite. Mochite hitaburu ni sono agata wo horobu…Kore nochi no yo no ashiki na naran to iu.”)

29 Chin or ware; a pronoun exclusive to the tennō, often translated into English as the “royal we”.

30 Warawa; an archaic feminine first person pronoun.

31 The introduction to her chapter does make explicit mention of her gender – as “daughter,” “younger sister,”

“empress-consort” – but in the context of her relationships to past emperors, as is standard for the opening of every emperor’s reign within the text. Although it is noted that she is “fair of form and face” (姿色端麗), similar phrases are used to describe the male emperors Nintoku (貌容美麗) and Hanzei (容姿美麗), casting doubt as to whether the descriptor was gendered (Nihon Shoki 11.Nitoku Tennō Sokui Zenki; Nihon Shoki

12.Hanzei Tennō Sokui Zenki; Nihon Shoki 22.Suiko Tennō Sokui Zenki).

250 the broader context of anticipating future criticism and belittlement suggests a certain level of derision was expected, that Suiko’s womanhood may have formed part of the criticism against her, be presented as something rendering her unfit for the throne. Amidst the backdrop of a largely gender-neutral monarchic self-presentation, this singular self-reference to womanhood in the context of insult may indicate contemporary resistance to, or uncertainty concerning, Japan’s first official accession of an empress regnant; a resistance which would indicate the jotei as something negotiated and unconventional, rather than the remnants of a former norm.

The notion of the jotei as unconventional is further reinforced by the unusual accession of

Empress Kōken-Shōtoku. Determined to have a child of himself and the principal consort Empress

Kōmyō take the throne, Emperor Shōmu bucked with convention and declared their daughter,

Princess Abe, Crown Prince(ss)32 in 728; the first such in Japanese history33 (Shoku Nihongi 13.Tenpyō

10.1.Jingo)34. By this stage, of course, Japan had already had at least five women ruling as tennō, but none had been designated as the imperial heir beforehand, their accessions situational rather than preordained. Abe’s – later known as Kōken-Shōtoku’s – appointment as hitsugi-no-miko was particularly surprising, as Shōmu at the time had a son – Prince Asaka (安積) – by a lesser consort,

Agata-no-Inukai no Hirotoji (県犬養広刀自). A woman was thus not only the intended, non- placeholder heir, but was prioritised over a male candidate.

In 743, a banquet was held before all the lords in which Abe herself danced the gosechimai, a traditional dance attributed to Tenmu. This event was very much a performance intended to legitimise her station as Crown Princess before the assembled retainers, accompanied by speeches

32 Abe was granted the title of kōtaishi/hitsugi-no-miko (皇太子); as with tennō, the same title as a male counterpart.

33 Tokoro 2017, 19.

34 “立阿倍內親王爲皇太子。” (“Abe-Naishinnō wo tachite hitsugi-no-miko to nasu.”)

251 from Shōmu and Retired Empress Genshō (Shoku Nihongi 15.Tenpyō 15.5.Kibō). Shōmu’s in particular directly links Abe with Tenmu’s legacy of skilled governance:

“The sacred Emperor who ruled the Great Eight-Island Land from Kiyomihara Palace in Asuka,

who we tremble even to mention, thought to rule and pacify the realm. To put all levels of

society in harmonious order, and keep the realm quiet and unmoving, was what he thought

best for a long and peaceful reign that combined propriety and joy. Thus he created this

dance, to be passed down in the heavens and on earth without end. The Crown Princess has

reached the pinnacle of her studies of this monarch, and now makes her offering before

us.”35 (Shoku Nihongi 15.Tenpyō 15.5.Kibō)

In Genshō’s speech, she agrees with Shōmu and asserts the importance of the dance in teaching principles of hierarchical deference, to which Shōmu – this exchange obviously having been scripted in advance – heartily concurs36. The speeches further link Abe’s study and performance of this dance with Emperor Tenmu’s legacy as a skilled monarch, implying that the Crown Princess, having

35 “掛〈母〉畏〈岐〉飛鳥淨見御原宮〈尓〉大八洲所知〈志〉聖〈乃〉天皇命天下〈乎〉治賜〈比〉

平賜〈比弖〉所思坐〈久〉。上下〈乎〉齊〈倍〉和〈氣弖〉无動〈久〉靜〈加尓〉令有〈尓八〉礼

〈䒭〉樂〈䒭〉二〈都〉並〈弖志〉平〈久〉長〈久〉可有〈䒭〉隨神〈母〉所思坐〈弖〉此〈乃〉舞

〈乎〉始賜〈比〉造賜〈比伎䒭〉聞食〈弖〉与天地共〈尓〉絕事無〈久〉弥繼〈尓〉受賜〈波利〉行

〈牟〉物〈䒭之弖〉皇太子斯王〈尓〉學〈志〉頂令荷〈弖〉我皇天皇大前〈尓〉貢事〈乎〉奏。”

(“Kakemakumo kashikoki Asuka no Kiyomihara-no-Miya ni Ōyashimakuni shiroshimeshishi hijiri no sumera- mikoto ame-no-shita wo osametamai tairagetamaite omōshimasaku. Kamishimo wo totonoe yawaragete ugoki naku shizuka ni arashimuru ni wa rai to gaku to futatsu narabeteshi tairakeku nagaku aru beshi to kamunagara mo omōshimashite kono mai wo hajimetamai tsukuritamaiki to kikoshimeshite ametsuchi to tomo ni tayuru koto naku iyatsugi ni uketamawari yukan mono toshite hitsugi-no-miko ono miko ni narawashi itadaki motashimete wa ga ōkimi sumera-mikoto no ōmae ni tatematsuru koto wo mōsu.”)

36 Ooms 2009, 121.

252 dutifully studied his governance, possessed the potential for equally able leadership (Shoku Nihongi

15.Tenpyō 15.5.Kibō). Asaka passed away in 744, further cementing Abe’s position as successor

(Shoku Nihongi 15.Tenpyō 16.Urū 1.Teichū).

Nevertheless, more public justification was forthcoming in the form of Shōmu’s 749 edict that daughters ought to be able to inherit their fathers’ roles alongside sons:

“Further, for the children of ministers who have served the court as chancellors, although

the sons have been able to follow into service and govern various things, the daughters have

not been permitted to govern. Thus it seems as if only sons have truly been able to bear

their fathers’ names, whereas for daughters, it becomes as a nonexistent thing, yet it seems

that there is good reason for the two to serve together… Thus, the heart of the ancestor

should exist also in the child. If this heart is not lost, then son and daughter should govern

together in service with a pure heart.”37 (Shoku Nihongi 17.Tenpyō-Shōhō 1.4.Kōgo)

Three months later, Shōmu abdicated in Abe’s favour; the first abdication by a male emperor, and probably performed to ease the unconventional transition. That same day, Abe took the throne as

Empress Kōken (Shoku Nihongi 17.Tenpyō-Shōhō 1.7.Kōgo).

37 “又爲大臣〈弖〉仕奉〈部留〉臣〈多知乃〉子䒭男〈波〉隨仕奉狀〈弖〉種種治賜〈比ツ礼䒭母〉

女不冶賜。是以所念〈波〉男〈能未〉父名負〈弖〉女〈波〉伊婆〈礼奴〉物〈尓〉阿礼〈夜〉。立双

仕奉〈自〉理在〈止奈母〉念〈須〉。…故是以子〈波〉祖〈乃〉心成〈伊自〉子〈尓波〉可在。此心

不失〈自弖〉明淨心以〈弖〉仕奉〈止自弖奈母〉男女幷〈弖〉一二治賜〈夫〉。” (“Mata ōomi toshite tsukaematsuraeru omi-tachi no kodomo onoko wa tsukaematsuru sama ni shitagaite kusagusa osametamaitsuredomo menoko wa osametamawazu. Koko wo mote omōseba onoko no mo chichi no na oite menoko wa iwarenu mono ni are ya. Tachi narabi tsukaematsurushi kotowari nari tonamo omōsu…Kare koko wo mote ko wa oya no kokoro nasu ishi ko ni wa arubeshi. Kono kokoro ushinawazu shite akaki kiyoki kokoro wo mote tsukaematsure toshite namo onoko menoko wasete hitori futari osametamau.”)

253

These occurrences were highly visible performances designed to legitimate to the Yamato court the selection of a princess as heir to the throne. Behind them, therefore, is the implication that such justifications were needed; that a female tennō required some form of negotiation to be accepted among the central nobility. Quite apart from the notion of empresses regnant being accepted as a transformation of earlier female rule, the circumstances of Kōken-Shōtoku’s accession depict the concept as something quite radical in the court’s eye, even despite the five previous, and fairly recent, empresses regnant.

Piggott views the justifications preceding Kōken’s rule as drawing from an ancient history of male-female paired rule – the notion of daughters inheriting positions and serving alongside sons; the gosechimai dance performed before both a male current emperor and a female retired one38 – arguing that the early Nara period was “an era where notions of gender complementarity and male- scripted rulership were both salient”39. Shōmu’s second edict, however, shows no reference to or awareness of ancient traditions of contrapuntal rule, or particular notions of complementarity; rather, it presents the notion of female rulership as a novel proposal: “the daughters have not been permitted to govern”. Instead, the edict makes hagiographical reference to the diligent court service of Abe’s maternal grandparents, Agata-no-Inukai no Michiyo (県犬養三千代) and Fujiwara no Fuhito

(藤原不比等) (Shoku Nihongi 17.Tenpyō-Shōhō 1.4.Kōgo), indirectly asserting Abe’s qualifications through her lineage. No connection is made to past images of female rule, even previous jotei; only a broad notion of talent as a hereditary quality.

The unusual nature of her reign is further highlighted by a speech attributed to Kōken herself, as her relationship with her initial successor Junnin40 (淳仁) soured, in which she declares that her initial accession was mandated to continue the line of Prince Kusakabe “even should it

38 Piggott 2003, 54.

39 Piggott 2003, 56.

40 R. 758–764 CE

254 mean a girl-child taking the throne”41 (Shoku Nihongi 24.Tenpyō-Hōji 6.6.Kōjutsu); the use of “even”

(are tomo) here is telling. The dialogue surrounding Kōken-Shōtoku’s enthronement specifically presents her gender as an obstacle to her accession, a controversy to be ameliorated through careful public performance; performance with never draws upon images of some ancient hime, or other ancient tradition of female rule, but rather likens Kōken to the masculine figure of Tenmu and to her ministerial grandfather.

It is surprising, perhaps, that this controversy took place in Kōken’s day, given the five jotei of the Century of Empresses preceding her. Perhaps what is different, in her case, is the title of

Crown Princess, and the intention behind it. As we shall see, there is a close link between previous jotei and young male intended successors; if the accession of previous jotei was enabled through their presentation as intermediaries, guides and facilitators for the next male tennō, then Shōmu’s choice of Kōken as his Crown Princess ahead of a viable male candidate represented something different, something more challenging to tradition. The controversy behind Kōken’s accession suggests a court climate in which female tennō could be accepted as befit the broader needs of the patriline, but the specific selection of a woman as heir remained tenuous. Female rule is accepted not so much as a tradition but a utility.

41 “朕御祖大皇后〈乃〉御命以〈弖〉朕〈尓〉告〈之久〉岡宮御宇天皇〈乃〉日繼〈波〉加久〈弖〉

絕〈奈牟止〉爲。女子〈能〉繼〈尓波〉在〈止母〉欲令嗣〈止〉宣〈弖〉此政行給〈岐〉。” (“A ga mioya ōkisaki-no-mikoto mochite are ni tsugetamaishiku Oka-no-Miya ni ame-no-shita shiroshimeshishi sumera-mikoto no hitsugi wa kakute taenan to su. Ominako no tsugi ni wa are tomo tsugashimemu to noritamaite kono matsurigoto okonaitamaiki.”)

255

This is not to say that the previous jotei did not put forward their own justifications for female rulership. In 712, Empress Genmei declared an edict commending two widows of court ministers:

The Empress ordered: “When their husbands were around, Iehara no Onna, wife of Minister

of the Left Senior Second Rank Tajihi-no-Mahito Shima, and Ki-no-Ason Onna, wife of

Posthumous Minister of the Right Junior Second Rank Ōtomo-no-Sukune Miyuki, consulted

with them on the best path for the nation, and after being widowed, they both firmly

protect the wish to be buried in the same grave. We feel a deep admiration for their fidelity.

They are each to be granted property of fifty households.” Iehara no Onna was also granted

the rank of muraji.42 (Shoku Nihongi 5.Wadō 5.9.Kishi)

42 “詔曰。故左大臣正二位多治比眞人嶋之妻家原音那。贈右大臣從二位大伴宿祢御行之妻紀朝臣音那。

並以夫存之日。相勸爲國之道。夫亡之後。固守同墳之意。朕思彼貞節。感歎之深。宜此二人各賜邑五

十戶。其家原音那加賜連姓。” (“Mikotonori shite notamawaku. Kare sadaijin shō-nii Tajihi-no-Mahito Shima ga tsuma Iehara no Onna. Sō-udaijin jū-nii Ōtomo-no-Sukune Miyuki ga tsuma Ki-no-Ason Onna. Narabi ni otto zonseru hi wo mote. Kuni wo osamuru michi wo ai susume. Otto bō suru no ato wa. Kataku dōfun no kokorozashi wo mamoru. Are a ga teisetsu wo omoite. Kantan suru koto shinshi. Yoroshiku kono futari ni onoono mura gojūko wo tamau. Sono Iehara no Onna ni wa muraji no kabane wo kuwae tamau.”)

256

On the surface, this reward was centred around these women’s fidelity to their late husbands, their virtuous fulfilment of the roles of wife and widow. What is interesting, however, is the acclamation that, while their high-ranking husbands were alive, they “consulted with them on the best path for the nation”. Within the framework of “virtuous wife,” this casts them as able helpmeets, but is also reminiscent of the Nihon Shoki’s description of Empress Jitō, who was said to be a capable co-ruler and helpmeet, following Tenmu into hiding him and aiding him with his military endeavours in the

Jinshin War (Nihon Shoki 30.Jitō Tennō Sokui Zenki Tenmu 1.6). During her stint as Empress Consort, the Nihon Shoki notes:

From the beginning until now, she had assisted the Emperor in ruling the realm. During the

period she was serving, she often assisted in advice and government.43 (Nihon Shoki 30.Jitō

Tennō Sokui Zenki Tenmu 2)

Upon Tenmu’s death in 686, she continued to handle matters of court (Nihon Shoki 30.Jitō Tennō

Sokui Zenki Shuchō 1.9.Heigo); acting as effective Empress Regnant before her formal accession in

690 (Nihon Shoki 30.Jitō 4.1.Boin). Genmei had herself been married to Prince Kusakabe, Jitō’s intended successor.

Jitō’s public persona was that of dutiful wife, described in the Nihon Shoki as “virtuous as a mother”44 (Nihon Shoki 30.Jitō Tennō Sokui Zenki). During Tenmu’s reign, she was depicted as an able helpmeet assisting her husband in governance (Nihon Shoki 30.Jitō Tennō Sokui Zenki Tenmu 2), and thus perhaps the closest empress of the Century to the model of hime. Her mourning for the late Tenmu is publicised through a “sorrowful and painful” edict upon making an almsgiving of

43 “皇后從始迄今。佐天皇定天下。每於侍執之際。輙言及政事。多所毘補。” (“Ōkisaki hajime yori ima ni made tamau ni. Sumera-mikoto wo sasōte ame-no-shita wo sadamaru. Tsukanmatsuritamau goto no aida ni.

Sunawachi koto matsurigoto ni oyobite. Tasuke oki tokoro ōki.”)

44 “有母儀德。” (“Omotaru ikioi arisu.”)

257 monastic robes from the fabric of the former emperor’s clothing (Nihon Shoki 30.Jitō 1.8.Kibi).

Furthering her public image as a dutiful widow, Jitō was buried alongside Tenmu instead of in her own mausoleum (Shoku Nihongi 3.Taihō 3.12.Kiyū); being buried alongside a late husband was, in

Genmei’s edict, portrayed as a sign of wifely fidelity and a source of praise (Shoku Nihongi 5.Wadō

5.9.Kishi).

Both Genmei’s edict and Jitō’s portrayal in the Nihon Shoki valorise the notion of a wise wife who consults with her husband on policy, explicitly legitimising Jitō’s succession (and implicitly

Genmei’s) through their status as widows of political men. There are shades of the model of empress-as-hime in this, featuring the idea of empresses as substitute rulers for a deceased male partner, although the context suggests this to be exceptional, rather than standard. One can imagine, however, that the significance of a female tennō awarding an official commendation to other women for their assistive role in government would not have gone unnoticed: through this proclamation, Genmei positions female political involvement as not only acceptable but praiseworthy – although whether the intent was to uplift (noble)women as a class or focused on validating her own rule cannot be known. Notably, both Shōmu’s edict and Genmei’s cast female political acumen in terms of women’s relationships to men: in Shōmu’s female aptitude is passed down by a father, in Genmei’s it is displayed in marriage to a husband. In both edicts, it is presented in terms palatable to an androcentric, patrilineal court society.

The motif of co-burial is also interesting, as it appears more commonly with female emperors than male. Ankan45 (安閑) is written as having been buried together with his Empress

Kasuga-no-Yamada (Nihon Shoki 18.Ankan 2.12), but it was she who was buried with him, outliving him by at least four years (Nihon Shoki 19.Senka 4.12.Kōshin). Senka46 (宣化), too, was buried with his Empress Tachibana (橘) and their young child, although the order of their deaths is not clear

45 R. 531–535 CE

46 R. 535–539 CE

258

(Nihon Shoki 18.Senka 4.11.Heiin). In Ankan’s case, the tomb is primarily constructed for his sake, and it is his wife who is later interred alongside him; it is probable that Senka’s burial followed a similar pattern. Bidatsu, who was buried alongside his late mother Ishihime (石姫) (Nihon Shoki

21.Sushun 4.4.Kasshi), is perhaps the only male monarch of this period to be interred in a late relative’s tomb rather than having his own.

In her posthumous edict, Suiko declared her wish to be buried not in her own mausoleum, but alongside her eldest son, Prince Takeda (武田) (Nihon Shoki 22.Suiko 36.4.Boshi). Her stated reason was mercy – not troubling a famine-stricken populace with construction works – but this act also served to publically memorialise her as a loving, grief-stricken mother. Kōgyoku-Saimei wept openly for her deceased grandson, Prince Takeru (健), and composed six poems of mourning with instructions that they be recorded for the ages; she too expressed a wish to be buried in the same tomb (Nihon Shoki 26.Saimei 4.5/4.10.Kasshi). It is unclear if this latter wish was carried out, although she is known to have been buried together with her daughter Hashihito (Nihon Shoki

27.Tenji 6.2.Bogo). Genmei’s accession speech makes mention both of her status as mother of the previous emperor, Monmu47 (文武), as well as her Confucian duty to “rule the realm with benevolence as a parent raises and governs their child” (Shoku Nihongi 4.Keiun 4.7.Jinshi)48, indirectly casting her maternal experience as a qualification for virtuous rule. Kōken’s extensive program of mourning prayers commissioned for her late father were explicitly depicted as expressions of filial piety (Shoku Nihongi 19.Tenpyō-Shōhō 8.5.Teichū/6.Jinshin/6.Kōin/11.Teibō), constructing her image as a dutiful daughter.

In these ways, female tennō used tropes of feminine virtue to construct a positive public image. At the same time, however, this virtue was derived from their relationships to men – fathers,

47 R. 697–707 CE

48 “人祖〈乃〉意能賀弱兒〈乎〉養治事〈乃〉如〈久〉治賜〈比〉慈賜來業” (“Hito no oya no ono ga wakugo wo yashinai hitasu koto no gotoku osametamai megumitamai kuru waza”)

259 husbands, sons, grandsons – in the form of fidelity, devotion, filial piety. Their qualifications as women rulers are equated with the support they provide to the men of the imperial clan, their devotion to their male kin. The virtue of a female tennō served as apologia for her reign and her fidelity to the patriline; a model of female authority negotiated within a patriarchal system rather than resisting it as a holdout of some glorious matrilineal past or ancient image of queenly power.

6.2.2. Empresses and Masculinity

According to the Chinese Book of Sui, the “King of Wa” (俀王, tuǐwáng) sent envoys to the

Chinese court in 600 CE, the 20th year of Kaihuang, and again in 607, the third year of Daye; the following year, China sent Pei Shiqing (裴世清) to Wa in return (Sui Shu 81.Woguozhuan). These diplomatic missions are corroborated in the Nihon Shoki, although the Japanese narrative differs considerably in many aspects: in the Sui account, Emperor Yang (隋煬帝) is most offended by the effrontery of the “barbarians” (蠻夷, mányí) in approaching them as equals in 207, and Pei’s visit in

208 finds the Japanese monarch humble and self-effacing, defusing the situation (Sui Shu

81.Woguozhuan); in the Japanese account, the two countries communicate warmly and graciously, with no hint of offence (Nihon Shoki 22.Suiko 16.4/6.Heishin/8.Jinshi/9.Shinshi). Nevertheless, both accounts agree that Pei Shiqing arrived in Japan as an envoy in 208, leaving the broader historicity of this occurrence in little doubt; the envoys named by Pei, “Ahodai” (阿輩臺) and “Katabi” (哥多毗)

(Sui Shu 81.Woguozhuan), are probably corruptions of the names and titles of two actual envoys dispatched to meet the visitors, Ōshikōchi-no-Atai Nukade (大河内直糠手) and Nukatabe-no-Muraji

Hirabu (額田部連比羅夫) (Nihon Shoki 22.Suiko 16.6.Heishin/8.Kibō).

Another striking difference does present itself between these accounts, however. All three of these missions – 600, 607, and 608 – took place during the reign of Empress Suiko, the first female sovereign of the Yamato dynasty; and yet, the Book of Sui account appears to be under the continuing impression that Japan at this time was ruled by a man. It is not simply a matter of gender-

260 neutral language – Himiko, who is discussed immediately preceding the 600 mission, is explicitly referred to in terms of her gender – but they give a specific name for this man, in what is clearly a phonetic rendering of a Japanese name or title: Ame Tarishi-Hiko Ohokimi49 (阿每多利思北孤阿輩

雞彌) (Sui Shu 81.Woguozhuan).

“Ohokimi” is clearly a transliteration of ōkimi, or “great king” (大王), but the rest is unusual.

“Ame-Tarashi-Hiko” would not sound out of place as a Japanese-style (和風, wafū) monarchic name, but it does not at all resemble Suiko’s wafū name of Toyomike-Kashikiya-Hime (豊御食炊屋比売/豊

御食炊屋姫) (Nihon Shoki 22), and furthermore the inclusion of “-Hiko” firmly and specifically genders the name as masculine, opposed to the feminine “-Hime”. Nor does any preceding emperor fit the moniker; Sushun was Hatsusebe (長谷部/泊瀬部), Yōmei Tachibana-no-Toyohi (橘豊日)

(Nihon Shoki 21), Bidatsu Nunakura-no-Futo-Tamashiki (沼名倉太珠敷/渟中倉太珠敷) (Nihon Shoki

20), and Kinmei Amekuni-Oshiharaki-Hironiwa (天国押波流岐広庭/天国排開広庭) (Nihon Shoki 19).

The closest to that name found in the records is Prince Ame-Tarashi-Hikokuni-Oshihito (天足彦国押

人), a non-reigning son of the (presumed mythical) Emperor Kōshō50 (孝昭) whose lifetime is attributed to the mid-fifth century BCE (Nihon Shoki 4.Kōshō 29.1.Heigo). Mentions are also made of the title of kimi (雞彌, jīmí) for the king’s wife, a harem of 600–700 women (presumably including maids and not simply consorts), and a Crown Prince named Rikamitafuri (利歌彌多弗利,

Lìgēmíduōfúlì) (Sui Shu 81.Woguozhuan).

It is possible that the name was, in fact, a general poetic title for an emperor, which could be applied to Suiko despite its gendering, although the Chinese clearly interpreted it as a name; the other details may also have been interpreted as general information rather than pertaining to a specific ruler, although the meaning of “Rikamitafuri” is unknown, bearing little apparent relation

49 “俀王姓阿每，字多利思北孤，號阿輩雞彌” (“Tuǐwáng xìng Āměi, zì Duōlìsīběigū, hào Ābèijīmí.”)

50 Reign traditionally dated 475–393 BCE

261 either to the typical Japanese term for a royal heir – hitsugi-no-miko – or any known name or title of

Crown Prince Shōtoku. Complicating the issue further is that the Book of Sui notes of the 608 mission that “the king met with Qing”51; the use of the word xiāngjiàn (相見, lit. “see each other”) suggesting an in-person meeting (Sui Shu 81.Woguozhuan). Nevertheless, no mention is made of this “king’s” gender, in notable contrast to the interest that Queen Himiko drew as an oddity to the patriarchal

Chinese. In the Nihon Shoki version, however, although Pei visits court, he does not appear to have ever directly met Suiko, his message being conveyed via a letter passed on from Pei to Abe no Tori

(阿倍鳥), from Tori to Ōtomo no Kui (大伴齧), and from Kui to the empress’s quarters (Nihon Shoki

22.Suiko 16.8.Jinshi). It is possible, therefore, that Pei mistook a senior official for an emperor, or that the entire meeting with a contrite Japanese monarch was a fabrication intended to display the might of China through the “barbarians’” subservience, but either way, it appears that Pei returned to his home country still under the apprehension that Japan was ruled by a man.

The question, then, is how this misunderstanding came about. Was it simply a normative assumption from the Chinese side that, unless clear statements were made to the contrary, a monarch would of course be a man, or was there some deliberate obfuscation of Suiko’s gender carried out on the part of the Japanese? It is quite possible that this was simply a misunderstanding on the part of the Chinese: the titles of tennō and ōkimi are not inherently gendered, even if the role they represented was a typically masculine one, and the language employed by female tennō in the

Nihon Shoki and Shoku Nihongi is identical to that of a male ruler. It may well be that the Chinese envoys, having received no overt sign of Suiko’s femininity, merely assumed that of course it would be a man in charge. Yoshikawa Shinji (吉川真司), however, has also speculated that Japan may have intentionally sought to hide Suiko’s true gender, presenting instead the facsimile of a man on the

51 “其王與清相見” (“Qí wáng yǔ Qīng xiāngjiàn.”)

262 throne, motivated perhaps by insecurity fuelled by the understanding of a female monarch as something difficult for their neighbouring kingdoms to accept52.

Earlier (see Ch. 3.1.2), we discussed the gendering of Amaterasu, how her martial aspect facing Susa-no-O was depicted in masculine terms – sometimes directly contrasted with her feminine identification – creating a sort of androgyny that could exempt her from the negative associations of womanhood and permit her to occupy the high status which she did53. It appears that this androgyny could be leveraged by mortal queens as well. According to the Nihon Shoki, before leading her troops for her invasion of Korea, Jingū is said to have declared:

“Although I am a woman, and moreover young, I must now for a time assume the form of a

man, and unleash strong and heroic strategy.”54 (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki

Chūai 9.4.Kōshin)

Compare this with the numerous instances in the Kojiki and Nihon Shoki of Amaterasu’s warrior aspect being referred to in explicitly masculine terms: as “the stance of a manly warrior”55 (Kojiki 1),

“acting as a manly warrior”56 (Nihon Shoki 1.6), taking on “the appearance of a manly warrior”57

(Nihon Shoki 1.6.1). Compare it also with the famous, if probably equally mythical, quote attributed to Queen Elizabeth I as she mustered troops at Tilbury to repel the Spanish Armada:

52 Yoshikawa 2011, 20.

53 Matsumura 1999, 98, 102, 105; Beard 2017, 68–70.

54 “吾婦女之加以不肖。然蹔假男貌。强起雄略。” (“Are taoyame ni shite mata onashi. Shikashi shibaraku masurao no sugata wo karite. Anagate ni ōshiki hakarigoto wo tatsu.”)

55 “伊都…之男建…蹈建而。” (“Itsu…no otakebi…fumi takebite.”)

56 “奮稜威之雄誥。” (“Itsu no otakebi wo furuwashi.”)

57 “乃設丈夫武備。” (“Sunawachi masurao no takeki sonae wo mōketamau.”)

263

“I know I have the body of a weak, feeble woman; but I have the heart and stomach of a king,

and of a king of England too.”58

Jingū’s declaration, much like the depictions of Amaterasu, associates warfare with masculinity; although female, they must take on the attributes of a man to act as a warrior or take on board the martial duties of a wartime monarch. This speaks to a broader understanding of kingship as a gendered quality; Jingū’s gender, like Elizabeth’s, must be negotiated for her to be seen as capable of carrying out the full suite of kingly duties, a process apparently unnecessary for a male monarch. The women of the Century of Empresses were less quick to actively declare themselves men/masculine, but they nevertheless occupied the same ritual positions as their male counterparts59, with little deference to their femininity. Did they, too, seek to develop a similar state of queenly androgyny?

The notion of gendered labour was echoed even by Empress Genshō, who in a 715 edict repeated the Confucian Chinese idiom of “men plough and women weave” (男耕女織, nángēng nǚzhi)60, asserting the gendered division of labour as natural and correct: “For the men toil in ploughing the fields, and the women work at their spinning and weaving, and their households have food and clothing in abundance.” (Shoku Nihongi 7.Reiki 1.10.Itsubō)61. The specific gender division implied by this phrase was probably not part of ancient Japanese society, borrowed as it was from another culture62, but the rhetoric itself links correct performance of gendered labour to a nation’s prosperity and, only slightly further on, moral character; a surprising proclamation from a woman

58 Beard 2017, 22.

59 Emura 2010, 150.

60 Ko 2012, 170.

61 “故男勤耕耘。女脩絍織。家有衣食之饒。” (“Yue ni otoko wa kōun wo tsutome. Omina wa jinshoku wo osamete. Ieie ishoku no yutaka arite.”)

62 Yoshie 1993, 458–9.

264 holding a typically male office, but perhaps the lack of distinction from what a male tennō might say was the point.

The role of an empress-regnant was not, in nature, contrapuntal; although, as we have seen, a female tennō’s rule was clearly impacted by her gender, her powers, ritual duties, and court roles were officially identical to that of a male tennō. She was not envisaged as the occupant of a separate, complementary role taking on additional duties as required by circumstance; her involvement with the divine was, as we shall see, largely the same as that of her male counterparts – with a few noteworthy exceptions – a far cry from the notion of the jotei as a transformation of some ancient shaman-queen performing a specifically feminine ritual role. Rather than drawing from a historical model of female leadership, empresses-regnant largely drew from the self-presentation and rulership traditions of their male peers. Where their gender does appear, it is more often as something in need of justification, an obstacle to accession rather than a facilitator. It is hard to validate the theory of the female tennō as derived from a female-specific rulership position based on ideas of gender complementarity, when a jotei’s gender was so often downplayed, her performance of her role so often identical to a male ruler’s. If the male tennō was hiko, then the female tennō was not hime, but rather a woman, for a time, stepping into the hiko’s shoes. This is particularly apparent if we look at female emperor’s performance in that field which is often claimed to be the bread and butter of the hime, the focus of her role: the religious sphere.

265

6.3. Empresses, Gender, and Kami-Worship

6.3.1. Jotei as Shaman-Queens

Traditional gendered divisions of labour did play a role in religious performance, informing the development of contrapuntal male and female ritual duties63 (see Ch. 3.2.2). An empress- regnant, however, performed the same court rituals in the same fashion as a male emperor64; in other words, she took the “masculine” role, rather than a “feminine” counterpoint. The one exception to this is Jingū’s performance of the typically female oracular role (Kojiki 2.Chūaiki; Nihon

Shoki 8.Chūai 8.9.Kibō; Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai 9.3.Jinshin), but this is fraught for several reasons. For one, Jingū’s own historicity is doubtful, weakening her use as a model of actual ancient practice. For another, within the official history, Jingū is strictly delineated as a regent- kōgō rather than a female tennō, discursively denied full access to the role of reigning monarch despite effectively acting as such. Thirdly, of her two divinations, one is performed while her imperial husband yet lives, and another following his death but preceding her official “accession,” within a time still credited to Chūai’s reign (Nihon Shoki 8.Chūai 8.9.Kibō; Nihon Shoki 9.Jingū Kōgō

Sesshō Zenki Chūai 9.3.Jinshin). Jingū’s assumption of the “female” shamanic role in divinatory ritual is, strictly speaking, a precursor to her rule rather than a quality of it.

In 607, Empress Suiko pronounced an edict placing herself in charge of kami-worship, effectively declaring herself the highest ritual coordinator (Nihon Shoki 22.Suiko 15.2.Boshi). Piggott reads this pronouncement as a gendered act deriving from an ancient connection between women and sacral power65, but this connection has been heavily contested within modern scholarship (see

63 Yoshie 2007, 160, 209–11; Maeda 2008, 17–9.

64 Emura 2010, 150.

65 Piggott 1994, 21.

266

Ch. 3.2.2). Furthermore, as Piggott notes elsewhere, the construction of the tennō as chief ritual coordinator was inherent to the process of religious centralisation occurring during the Asuka period and its concepts of sacral kingship66. In this respect, Suiko’s edict seems less of a “hime-esque” assumption of a feminine sacral role, and more of an adherence to an existing relationship between the (masculine) role of emperor and ritual coordination. The establishment of the Jingikan, begun by the male monarch Tenmu and carried on by his female successor Jitō in largely identical ways67, can arguably be viewed as a continuation of this process of centring emperor and court in worship.

Empress Kōgyoku is also credited with a religious role, derived from a rainmaking prayer episode in the Nihon Shoki. During the first year of her reign, in response to a baleful drought that both kami-worship and Buddhist sutra readings had failed to end, Kōgyoku’s own prayers apparently met with greater success:

The Empress visited Kawakami in Minabuchi, and kneeled in each of the four directions, and

looked up to the sky and prayed. Suddenly, there was thunder, and heavy rain fell. It rained

for five days, and all the realm was watered. (One record has it that it rained continuously

for five days, and the nine grains all ripened.) The common people of the realm all exclaimed

in joy: “This is an Empress of superb virtue!”68 (Nihon Shoki 24.Kōgokyu 1.8.Kōshin)

Prayers and offerings for reign in response to dry spells were commonly recorded throughout the

Nihon Shoki and Shoku Nihongi, but unlike Kōgyoku’s ceremony, these were not normally performed

66 Piggott 1997, 210.

67 Ooms 2009, 113.

68 “天皇幸南淵河上。跪拝四方。仰天而祈。卽雷大雨。遂雨五日。溥潤天下。〈或本云。五日連雨。

九榖登熟。〉於是。天下百姓倶稱萬歲曰至德天皇。” (“Sumera-mikoto Minabuchi no Kawakami ni yukite.

Hizamazukite yomo wo ogamu. Ame ni aogite koitamau. Sunawachi ikazuchi nari hisame furu. Tsui ni ame furu koto itsuka. Ame-no-shita ni uruoshitsu. (Arui fumi iwaku. Itsuka hisamete. Kokonotsu no tanatsu mono nari akaran.) Koko ni. Ame-no-shita no ōntakara tomo ni yorokobite ikioi mashimasu sumera-mikoto to iu.”)

267 by the tennō in person; instead, prayers were typically commissioned from ritualists, and offerings dispatched through messengers. Kōgyoku’s prayer was dubbed by Hori as “dimly shamanic”69, interpreted as a surviving trace of an ancient, more overt model of shamanic queenship epitomised by Himiko and later Jingū70. To Kidder, these “shamanic talents” are what enabled her second reign as Saimei71, although Naka-no-Ōe’s (中大兄) political manoeuvrings seem a more supported explanation. Yoshie, conversely, links Kōgyoku’s performance with Himiko, but also with male monarchs such as Yūryaku and Tenmu, arguing that ritual activity was part of the standard suite of monarchic duties and not linked with any particular gender72. Nevertheless, Kōgyoku’s direct performance of religious ritual is unusual in the monarchic context, and worthy of further investigation into the role of gender in rainmaking ceremonies.

Hori similarly connects Jitō’s frequent visits to Yoshino – the region of the sacred Mount

Yoshino – with sacral performance73, and notes an overall trend of the Asuka-Nara empresses

“frequently [visiting] the sacred mountains, sacred waters, and sacred hot springs”, a tradition he speculates may be linked to Himiko’s model of shamanic queenship74. Textual descriptions of Himiko

(see Chapter 4.1), however, paint her as a recluse, not a traveller to sacred sites of her realm; the particulars of her ritual and performance beyond this are largely unclear. Furthermore, hot springs in ancient Japan were sometimes credited with healing properties (Nihon Shoki 26.Saimei 3.9), and were patronised also by male tennō on occasion, with Jomei75 in particular being a frequent visitor

(Nihon Shoki 23.Jomei 3.9.Itsugai/10.10/11.12.Jingo; Shoku Nihongi 2.Taihō 1.10.Teibi). There is little

69 Hori 1968, 197–8.

70 Hori 1968, 191, 194.

71 Kidder 2007, 133.

72 Yoshie 1993, 8–10.

73 Hori 1968, 163–4.

74 Hori 1968, 166.

75 R. 629–641 CE

268 proof of a deeper symbolic meaning behind the visits, or any ties between Himiko and “sacred waters”; indeed, the latter is most often theorised to have practised solar worship rather than ritual centred on mountains or rivers76. Kidder interprets Jitō’s travels to Yoshino as a ritual communion with the spirit of Tenmu77, whereas Wada Atsumu (和田萃) and Michael Como connect them with worship of the rainmaking goddess Nifu78 (丹生). Como has also speculated a relationship between

Yoshino, legends of women consuming immortality herbs, and Daoist medicinal practices79.

The issue with all of these interpretations of Jitō’s travels is the fact that her visits are never, within the Nihon Shoki itself, contextualised as any sort of religious performance. Their frequency is unusual compared to other monarchs, perhaps, but there is nothing overtly cultic noted within the text. Indeed, historians not writing on religious performance tend to interpret the visits as political in nature, with Yoshino presented in the Nihon Shoki as the starting point of Tenmu’s successful campaign in the Jinshin War. Jitō’s visits are thus viewed as a commemoration of Tenmu’s victories, within Yoshino’s significance deriving from contemporary imperial politics, rather than ancient shamanic practices80.

Yoshino, Torquil Duthie claims, was constructed in the Tenmu-Jitō courts as a symbolic location of imperial conquest and dominion, as well as the legitimacy of the Tenmu line, appearing

(inspired by Tenmu’s Jinshin campaign) in the Kojiki and Nihon Shoki accounts of Jinmu’s eastward conquests, the submission of the “wild” kuzu to Ōjin, and other narratives of dominion in Yūryaku’s chapters81. Duthie further argues that Yoshino was the location of something akin to a “Tenmu cult”,

76 Kidder 2007, 133.

77 Kidder 2007, 133.

78 Wada 1995, 5–17, 129–252; Como 2009, 63.

79 Como 2009, 75, 82; Ooms 2009, 148.

80 Duthie 2015, 193.

81 Duthie 2015, 198–201.

269 celebrating Tenmu as the divine founder of a new regime and political order, to which Jitō was inheritor82. Its association with Sinic immortal legends – already extant in Japan, but not as yet regionally connected to Yoshino – was a later construct apparently spearheaded by Fujiwara no

Fuhito, with the establishment of a link between legitimacy and Yoshino unrelated to Tenmu advantageous to Fuhito’s own political goals83.

Jitō’s visits to Yoshino were thus rituals of legitimation and dominion, but derived from a contemporary symbology of imperial power, rather than hearkening to an ancient tradition of

“shaman-queens”. Rather than an act of rulership through “sacral power,” as the discourse of empress-as-hime implies, the visitations served to bolster Jitō’s temporal authority, legitimating it through the victories of her husband-predecessor and his imperial line. Even the possibility of sacral activity in the form of a “Tenmu cult” retains these qualities: it would both be a recently-constructed form of monarchic religiosity (as opposed to a legacy of a sacral-female “golden age”), and also act primarily as an enhancement of Jitō’s temporal authority rather than placing her in a feminine- specific role of “sacral power.”

Indeed, the assumption that empress’s activities must link to some ancient female-specific religious tradition speaks to a kind of circular logic based on the preconceived notion of female tennō as a descendant of the shamanic hime: if we start from the belief that the role of the empress has shamanic roots, then it naturally follows that unusual particulars of her reign will be interpreted in a shamanic light; these events may then be presented as further proof of the jotei’s shamanic origins. Other cultic activies of the jotei are read as hearkening to a specifically feminine sacral role, even when they strongly resemble the sacral involvement of male tennō, such as Suiko and Tenmu’s role as chief ritual coordinators. The model of jotei-as-hime projects the assumption of a specificially feminine model of sacral queenship upon the jotei’s actions, but her actual performance in the

82 Duthie 2015, 193, 215.

83 Duthie 2015, 221–4.

270 records cleaves more to the male tennō’s monarchic ritual involvement than any contrapuntal hime- image.

6.3.2. Empresses and Amaterasu

In the Kojiki account of Jingū’s act of shamanic communion, Amaterasu is named as the principal god speaking through the empress (Kojiki 2.Chūaiki). In the Nihon Shoki, this deity’s name is given as Tsuki-Sakaki-Itsu-no-Mitama-Amazakaru-Mukatsu-Hime (撞賢木厳之御魂天疎向津媛), although due to the inclusion of the feminine “-Hime” and the goddess’s stated enshrinement in

“Isuzu Shrine” (the Inner Shrine of Ise, the cultic centre of Amaterasu by the time of the Nihon

Shoki’s compilation) (Nihon Shoki 9.Jingū Kōgō Sesshō Zenki Chūai 9.3.Jinshin), it is reasonable to conclude that the two are one and the same. Through the medium of a woman, Jingū, the royal ancestress delivers the treasures of Korea unto her patrilineal heir; an echo, perhaps, of her own role of granting the divine mandate to the male line of kings.

It is perhaps no surprise, then, that some pre-war scholars identified the powerful, brilliant

Amaterasu with the warrior-empress Jingū84; indeed, earlier in this very chapter (see Ch 6.2.2.), we compared their deployment of masculinising language. Ōwa Iwao (大和岩雄) argues that, as mother of the deified Ōjin, Jingū is herself equated to Amaterasu, mother of the imperial line85. However, both Jingū and Amaterasu’s appearances in the Kojiki and Nihon Shoki should not be simply accepted as unbiased, unvarnished retellings of myth-historical tradition, but must be considered – as with

Jitō’s visits to Yoshino – within the specific historical and political context of the works in which they appear.

84 Maeda 2008, 7.

85 Ōwa 2009, 287.

271

Okada views the prominent position of Amaterasu as linked to Suiko’s reign, a female deity elevated in status under a female monarch legitimising female rule86. Amaterasu’s importance to the throne was solidified by the Tenmu lineage, the goddess credited with his Jinshin War victory87.

Other scholars have cited the influence of Jitō’s reign as a major factor in the presentation of stories featuring female rulers, the roles of figures such as Amaterasu and Jingū perhaps emphasised by the compilers in a bid to gain favour with the female tennō88. The backdated image of an always-royal

Ise Shrine appearing in the Yamato-Hime founding myth may have drawn from Jitō’s journey to Ise in its depiction of the princess’s pilgrimage89; Ooms goes further, suggesting that Amaterasu herself was intended to evoke the image of Jitō, the image of a powerful female deity deployed in a similar way to that which Okada envisioned for Suiko: as a symbol justifying female rule90. As N. Harry

Rothschild explored in the religious self-presentation of the Chinese empress Wu Zetian, “affiliation with ancient gods…played a vital role in establishing political legitimation” through the narrative of a present ruler returning an ancient golden age91. For the jotei, the feminine figure of Amaterasu could provide special legitimation, a way in which her mode of rule was distinguished from that of the male tennō, who might emphasise his connection to or favour with the sun goddess but did not liken himself directly to her.

One angle of the jotei-as-hime notion equates the jotei to the saigū, another proposed transformation of the hime. The suggestion is that jotei had no need to send a saigū to Ise: a male tennō or so the theory goes, required a kinswoman to serve the ancestral deity in his stead, replicating the religious function attributed to the ancient hime, whereas a female tennō, herself a

86 Okada 1970, 394.

87 Piggott 1997, 145–6.

88 Piggott 1994, 27; Kidder 2007, 3.

89 Emura 2009, 12, 20–1.

90 Okada 1970, 394; Ooms 2009, 32.

91 Rothschild 2015, 9–10.

272 hime, could forge this spiritual connection directly92. We have seen (see Ch. 5.3.2), however, that this was not typically the case: despite their dual interpretation as transformations of the hime, the jotei and the saigū could and did exist side-by-side. The jotei were also, by and large, very different in nature to the saigū. The saigū was invariably virginal and spent her entire period of service in a secluded state meant to preserve a strict regime of ritual purity. Four of the six jotei, however, were already widows and mothers upon taking the throne, and many suffered the loss of a parent or close relative during their term in office, whereas the pollution of mourning demanded a saigū’s retirement from service. The tennō further existed in the central, mundane world of court, involved in its daily affairs, a far cry from the numinous, cloistered space inhabited by the saigū at Ise. Indeed, few jotei even seem to visited Ise; apart from Jitō’s famous travels (Nihon Shoki 30.Jitō

6.2.Teibi/3.Shinbi), Genmei served once as a “princess-messenger” under Tenmu’s reign (Nihon

Shoki 29.Tenmu 4.2.Teigai), but both women’s visits were brief, and Genmei’s well preceding her actual reign.

Jitō’s visit to Ise may have had a spiritual component of Amaterasu-worship93, but it is also important to consider the political context of her visits. Above (see Ch. 6.3.1), we discussed the symbolic importance of Yoshino, a frequent destination of hers, through its associations with the victorious Jinshin War campaign. Ise Shrine was also a significant location in that conflict, and thus

Jitō’s travel there may have had similar connotations of reinforcing the dominion of the Tenmu-line regime, directly impressing the might of the centre upon the peripheral province without the intermediary of the saigū94. It also served as a key ritual location in Tenmu’s new state cult, formed as a process of religious centralisation which Jitō was active in continuing95.

92 Emura 2009, 17; Tokoro 2017, 29.

93 Emura 2009, 20.

94 Bowring 2005, 37.

95 Piggott 1997, 144–6; Ooms 2009, 113.

273

Indeed, other than her direct journey there, Jitō mostly deployed similar strategies in her relationship with Ise to her male predecessor, her femininity not seeming to affect her involvement with the sacral sphere. Jitō continued Tenmu’s process of weakening local elites: as Tenmu divided the shrine into “Inner” and “Outer”, Jitō appointed the Arakida (荒木田) clan to hold priesthood over the Inner Shrine – the sanctum of Amaterasu – relegating the local Watarai to the less centrally relevant Outer Shrine96. Jitō’s interactions with Ise and the state cult, on the whole, appear less as the assumption of a contrapuntal female sacral role to Tenmu’s male, and more as the continuation of the program he started, much as would be expected of a chosen successor.

The essential component of the saigū – her continued occupation of a particular space in both geographical and ritual terms, an enclave of sacral purity on the goddess’s doorstop – was absent in the jotei; who performed court rituals identically to her male counterparts, whether invested with a saigū or not97. There is little evidence for female tennō fulfilling the role of the saigū in addition to their imperial duties; indeed, this theory seems to rest upon the assumption that, if the saigū and jotei are both remnants of the sacral hime, then the roles must be understood as simultaneously occupiable despite their differing characteristics. Not only does this assume the presence of a common lineage for the two roles, but also takes for granted that the Asuka and Nara courts would recognise this commonality and identify the two roles with each other, despite no such identification being made in the text. Amaterasu could provide a legitimating symbol for a female tennō, but jotei evince no special function in worship of the ancestral goddess, distinct from the ritual of a male tennō; her position and duties are clearly distinct from her other alleged hime- counterpart, the saigū.

96 Okada 1970, 335.

97 Emura 2010, 150.

274

6.3.3. Living God(desses): Jotei and Images of Male Imperial Religiosity

Thus we come to the salient point: much has been made of the ways in which empresses- regnant did, or even simply could’ve, differed from their male counterparts in their deployment of religion. But how are we to read the ways in which they didn’t, when the female tennō followed the same script as the male? For example, when Genmei took the throne in 707, her accession speech is recorded as beginning thus:

“Oh imperial princes, lesser princes, ministers, and officials, citizens of the realm, hark to the

edicts of Empress Yamato-Neko, the Living God who rules over the Great Eight-Island

Land.”98 (Shoku Nihongi 4.Keiun 4.7.Jinshi)

Compare this with the accession speech of her son, Emperor Monmu, 10 years earlier:

“Oh attendant princes, lesser princes, and bureaucrats, citizens of the realm, hark to the

edicts of your Emperor, the Living God who rules over the Great Eight-Island Land.”99 (Shoku

Nihongi 1.Monmu 1.8.Kōshin)

These opening statements are near-identical, and both deploy the official vocabulary of the emperor as a “living god” (現つ神; akitsu-kami, “shining god” in Piggott’s translation100), an appellation used

98 “現神八洲御宇倭根子天皇詔旨勑命。親王諸王諸臣百官人䒭天下公民衆聞宣。” (“Akitsu-mikami to

Ōyashima-Kuni shiroshimesu Yamato-Neko-Sumera-ga-Ōmikoto rama to noritamau ōmikoto wo. Miko-tachi

ōkimi-tachi omi-tachi momo-no-tsukasa no hito-tachi ame-no-shita no ōmitakara moromoro kikoshimesae to noru.”)

99 “現御神〈止〉大八嶋國所知天皇大命〈良麻止〉詔大命〈乎。〉集侍皇子䒭王䒭百官人䒭。天下公

民諸聞食〈止〉詔。” (“Akitsu-mikami to Ōyashima-Guni shiroshimesu sumera-ga-ōmikoto rama to noritamau ōmikoto wo. Ugonawareru miko-tachi ōkimi-tachi momo-no-tsukasa no hito-tachi. Ame-no-shita no

ōmitakara moromoro kikoshimesae to noru.”)

100 Piggott 1997, 160.

275 with some frequency in official proclamations as recorded in the Shoku Nihongi onwards, and deployed as well by Shōmu (Shoku Nihongi 9.Jinki 1.2.Kōgo), Kōken (Shoku Nihongi 21.Tenpyō-Hōji

2.8.Kōshi), Junnin (Shoku Nihongi 21.Tenpyō-Hōji 2.8.Kōshi), Kōnin101 (光仁) (Shoku Nihongi 31.Hōki

1.11.Kasshi), Heizei (Nihon Kōki 17.Daidō 4.4.Heishi), Saga (Nihon Kōki 20.Kōnin 1.9.Kōjutsu),

Junna102 (淳和) (Nihon Kōki), and many others henceforth. This concept was first established under

Tenmu’s reign103, and the similar descriptor of the monarch as kamunagara (神ながら, “as a god”) appears in reference to Jitō, in a poem by Kakinomoto no Hitomaro (柿本人麻呂) in the

Man’yōshū104 (Man’yōshū 1.38).

It is unclear whether such terminology implies a literal worshipful understanding of the tennō as a god-king, a la the Egyptian pharaoh, or was primarily a flattering title emphasising the divine origins of the imperial line105, but the important point for our purposes is that this concept, devised under the auspices of a male emperor, was employed in equal measure by female tennō as well. Rather than borrowing from some hime-like tradition of female-specific sacral rule, the jotei occupied the same ritual space and largely used identical tactics to connect themselves to the divine as their male counterparts, leaning into the same framework of the tennō as an embodied deity.

Gender here is neither negotiated nor distinguished; a jotei’s femaleness is no obstacle to her intrinsic divinity (kami are, after all, by no means limited to male), but nor does it lend her the special sacral status, or the tradition of uniquely female means of handling the religious sphere implied in the empress-as-hime theories of Takamure, Hori, or Kidder.

101 R. 770–781 CE

102 R. 823–833 CE

103 Ooms 2009, 58.

104 “安見知之吾大王神長柄神佐備世須登” (“Yasumishishi/wa ga ōkimi/kamunagara/kamusabi sesu to”)

105 Piggott 1994, 26; 1997, 160, 211; Bowring 2005, 45–6.

276

The same pattern is replicated in other recorded edicts engaging with the divine, with male and female tennō alike borrowing from a similar repertoire of phrase and rhetoric. As well as denoting the tennō as a living god themselves, edicts also depicted their rule as a matter of divine will. Let us compare once again the accession speeches of Monmu:

“Hear the Emperor who says that the realm and tributary lands shall be granted order and

peace, and the citizens of the realm blessed and cared for, and all be done according to the

will of the gods.”106 (Shoku Nihongi 1.Monmu 1.8.Kōshin)

And Genmei:

“Hear, oh assembly, that we shall honour and treasure the will of heaven and earth as we

accede this solemn post in awe… Hear our decree that we shall permit no bend or

movement of those unchanging laws of our land that stand long and distant as the heavens

and the earth…It is our duty to rule the realm with benevolence as a parent raises and

governs their child, in accordance with the will of the gods.”107 (Shoku Nihongi 4.Keiun

4.7.Jinshi)

106 “此〈乃〉食國天下〈乎〉調賜〈比〉平賜〈比。〉天下〈乃〉公民〈乎〉惠賜〈比〉撫賜〈牟止奈

母〉隨神所思行〈佐久止〉詔天皇大命〈乎〉諸聞食〈止〉詔。” (“Kono osukuni ame-no-shita wo todotoetamai tairagetamai. Ame-no-shita no ōmitakara wo megumitamai nadetamawan to namo kamunagara omōshimesaku to noritamau sumera-ga-ōmikoto wo moromoro kikoshimesae to noru.”)

107 “此重位〈尓〉繼坐事〈乎奈母〉天地心〈乎〉勞〈美〉重〈美〉畏坐〈左久止〉詔命衆聞宣。…又

天地之共長遠不改常典〈止〉立賜〈覇留〉食國法〈母。〉傾事無〈久〉動事无〈久〉渡將去〈止奈母〉

所念行〈左久止〉詔命衆聞宣。…人祖〈乃〉意能賀弱兒〈乎〉養治事〈乃〉如〈久〉治賜〈比〉慈賜

來業〈止奈母〉隨神所念行〈須〉。” (“Kono ikashi kurai ni tsugimasu koto wo namo ametsuchi no kokoro wo itōshimi ikashimi kashikomimasaku to noritamau ōmikoto wo moromoro kikoshimesae to noru…Mata ametsuchi no muta nagaku tōku kawarumajiki tsune no nori to tatetamawaru osukuni no nori mo. Katabuku

277

Genmei’s phrasing emphasises the image of the tennō as parent of the realm, perhaps invoking her status as the late emperor’s mother; the core conceit, however, is the same. The tennō has the duty of carrying out the will of the gods in their reign, but conversely, the language of emperors – “the

Heavenly Sun Lineage”, the “Living God”, the frequent invocations of imperial ancestors – reminds the subjects of their divine mandate: it is the gods’ will that they take the throne in order to carry out the gods’ will, as made particularly clear in Shōmu’s abdication edict, which stresses that the tennō’s rule is according to the gods’ command (Shoku Nihongi 17.Tenpyō-Shōhō 1.7.Kōgo)108. The language of the divine mandate is the same regardless of the gender of the tennō; the jotei do not leverage any special feminine connection to the divine, or to spiritual power – as implied by the model of jotei-as-hime – but rather use identical oratory tactics to their male counterparts.

We see this pattern throughout the edicts of the Asuka and Nara emperors. Male and female tennō alike publicised auspicious omens as signs of divine favour: auspicious clouds109, mysterious turtles110, and strange appearances of kanji characters proclaiming peace in the realm were all dutifully noted in court records, often directly attributed to divine approbation of the

koto naku ugoku koto naku watari yukan to namo omōshimesaku to noritamau ōmikoto moromoro kikoshimesae to noru…Hito no oya no ono ga wakugo wo yashinai hitasu koto no gotoku osametamai megumitamai kuru waza to namo kamunagara omōshimesu.”)

108 “高天原神積坐皇親神魯棄神魯美命以。吾孫〈乃〉命〈乃〉將知食國天下〈止〉言依奉〈乃〉隨遠

皇祖御世始而天皇御世御世聞看來食國天〈ツ〉日嗣高御座〈乃〉業〈止奈母〉隨神所念行〈佐久止〉

勑天皇〈我〉御命〈乎〉衆聞食勑。” (“Takama-no-Hara ni kamuzumarimasu sumera ga mutsu kamurogi kamuromi-no-mikoto mote a ga mima-no-mikoto no shirasamu osukuni ame-no-shita to kotoyosashi matsuri no manimani tō sumerogi no miyo wo hajimete sumera ga miyo miyo kikoshimeshi kuru osukuni amatsu- hitsugi no taka-mikura no waza to namo kamunagara omōshimesaku to noritamau sumera-ga-ōmikoto wo moromoro kikoshimesae to noru.”)

109 慶雲, keiun; clouds of multiple colours

110 霊亀, reiki; turtles of unusual size, colour, and/or markings

278 sovereign111 (Nihon Shoki 24.Kōgyoku 2.1.Jingo; Nihon Shoki 27.Tenji 9.6; Nihon Shoki 29.Tenmu

10.9.Shinchū; Shoku Nihongi 3.Keiun 1.5.Kōgo; Shoku Nihongi 6.Wadō 6.12.Itsushi/Reiki

1.1.Kōshin/8.Teichū; Shoku Nihongi 9.Yōrō 7.10.Kibō; Shoku Nihongi 10.Tenpyō 1.6.Kibō; Shoku

Nihongi 20.Tenpyō-Hōji 1.3.Boshin; Shoku Nihongi 28.Jingo-Keiun 1.8.Kishi/9.Boshin; Shoku Nihongi

31.Hōki 1.10.Kichū; Nihon Kōki 9.Enryaku 20.6.Kishi; Nihon Kōki 10.Enryaku 20.7.Heijutsu; Nihon Kōki

19.Daidō 5.8.Boin; Nihon Kōki 34.Tenchō 3.7.Shinshi). Kōken-Shōtoku in particular incorporated the symbology of divine favour into her public image, delivering an edict upon one such omen:

“We too think it a major and mystical omen, and have often heard that such is a

manifestation of heaven and earth’s approval of the virtuous reign of a sacred monarch. We

do not dare think that we could have so moved heaven and earth by our virtue.”112 (Shoku

Nihongi 28.Jingo-Keiun 1.8.Kishi)

Despite her apparent modest denial of the title of “sacred monarch” 聖皇, seiō/hijiri-no-sumera) – maintaining a public performance of humility and thereby, virtue – the edict served to impress that very idea upon the court. The term “sacred monarch” first appeared within the records in reference to Emperor Suinin (Nihon Shoki 6.Suinin 2/3.3), and later to Crown Prince Shōtoku (Nihon Shoki

23.Suiko 36.9), both male figures; it was linked to the normatively-masculine role of the sumera (皇).

Again, we see the jotei cleaving to largely the same traditions and symbologies as a male sovereign, rather than supporting or legitimising their rule through connection to any ancient images of female

111 Piggott 2003, 57.

112 “然朕念行〈久〉。如是〈久〉大〈仁〉貴〈久〉奇異〈尓〉在大瑞〈波〉聖皇之御世〈尓〉至德

〈尓〉感〈天〉天地〈乃〉示現〈之〉賜物〈止奈毛〉常〈毛〉聞行〈須〉。是豈敢朕德〈伊〉天地

〈乃〉御心〈乎〉令感動〈末都流倍岐〉事〈波〉無〈止奈毛〉念行〈須〉。” (“Shikaru ni a ga omōshimesaku. Kaku ōki ni tōtoku kusushiku koto ni aru ōki shirushi wa hijiri-no-sumera ga miyo ni itaru miutsukushimi ni kamakete ametsuchi no arawashi tamau mono to namo tsune mo kikoshimesu. Kore a ni aete a ga utsukushimi i ametsuchi no mikokoro wo ugokashimatsuru beki koto wa nashi to namo omōishimesu.”)

279 authority; a trend which continues even in the sacral sphere, the supposed domain of the ancient hime.

Rather than a distinctly female script of rulership, derived from some ancient tradition of female authority via contrapuntal rule, empresses regnant generally drew upon the same tropes as a male tennō. That, in itself, acted as a form of legitimisation, smoothing away their differences from a male monarch with the performance of a familiar model of kingship. While the “golden age” narrative holds that jotei’s reigns were enabled by previous traditions of female rulership – often held to be sacrally-based, existing in counterpoint to a male sovereign – the self-presentation of jotei shows them not turning to ancient, feminine traditions of (sacral and temporal) power for their legitimacy, but rather performing kingship in the same way as the masculine tennō. The jotei is not conceived of as the contrapuntal partner to a hiko-king, a transformation of the hime role; rather, she occupies the same position as a male monarch in both temporal authority and ritual status.

6.4. Nun-Empresses and World-Ruling Queens

6.4.1. Buddhism, Gender, and the Image of the Ideal Monarch

The jotei leveraged religious belief and performance in ways largely identical to those of male tennō, and this replication was not limited to the sphere of kami-worship. As the emperors of

Asuka and Nara also faced a change in the traditional script of kingship, with the influx of continental faith – Buddhism, Confucianism, Daoism – into the archipelago, jotei made use of the new beliefs mostly in the same way as their male counterparts.

Although the Yamato monarchs were at first hesitant to adopt the Buddhist faith into court practice (Nihon Shoki 19.Kinmei 13.10), through the influence of the Soga clan, Buddhism soon came to join kami-worship as part of the emperor’s public devotions (Nihon Shoki 21.Yōmei 2.4.Heigo).

280

Beyond the realm of personal piety, Buddhism offered several benefits to the tennō. Buddhist texts presented their own image of an ideal monarch: the cakravartin (in Japanese, tenrin-jōō (転輪聖王),

“sacred wheel-turning king”), the world-ruling Buddhist king who would convert the universe into a peaceful, unified Buddhist state113. This presented a new image of divinely-ordained monarchy for the tennō to construct themselves as: as the native tradition of the Heavenly Sun Lineage cast kingship as a divine mandate granted by ancestry, so to did the Buddhist cakravartin stand as a cosmically legitimated, enlightened monarch, proven through pious deeds, and the monarchs of

Asuka and Nara – male and female alike – deployed both concepts with gusto.

There is some question as to which emperors were deliberately invoking the mantle of cakravartin, and which were merely setting themselves up at the head the national Buddhist cult –

Piggott, for example, places Tenmu’s devotions in the former category (compared to Suiko’s reign, which she places in the latter)114, whereas Ooms situates him firmly in the second category115 – but arguably the distinction between those two concepts is not so clear-cut. The cakravartin was, after all, envisaged as the creator and coordinator of a universal Buddhist state, and likewise the concept of the cakravartin legitimised the monarch’s position at the centre of the Buddhist hierarchy. The cakravartin provided the pinnacle for a broader Buddhist ideal of pious, virtuous kingship, into which the tennō of Asuka and Nara – male and female alike – played to varying degrees.

Within East Asian Buddhist, the Humane King Sutra116 in particular was connected to the notion of Buddhist kingship and its associated ideals, equating the bodhisattva’s path to

113 Piggott 1997, 96.

114 Piggott 1997, 215.

115 Ooms 2009, 167.

116 In full, the Humane King Perfection of Wisdom Sutra or Humane King State-Protection Perfection of Wisdom

Sutra; in Japanese Ninnō-Hannya-Haramitsu-Kyō 仁王般若波羅蜜経 or Ninnō-Gokoku-Hannya-Haramitsu-Kyō

仁王護国般若波羅蜜経 or Ninnōgyō 仁王経 for short

281 enlightenment with the king’s path to rulership117. It was, accordingly, frequently deployed by monarchs wishing to present themselves as enlightened Buddhist rulers akin to the cakravartin.

Imperial promotions of the sutra – such as ceremonies, lectures, and commissioned readings – appear throughout late Asuka to early Heian, under a slew of tennō, including Saimei, Jitō, and

Kōken-Shōtoku: half of the “Century of Empresses”’ cohort of jotei (Nihon Shoki 26.Saimei 6.5; Nihon

Shoki 29.Tenmu 5.11.Kōshin; Nihon Shoki 30.Jitō 7.10.Kibō; Shoku Nihongi 10.Tenpyō 1.6.Kōshin;

Shoku Nihongi 16.Tenpyō 18.3.Teibō; Shoku Nihongi 17.Tenpyō 19.5.Kōin; Shoku Nihongi 18.Tenpyō-

Shōhō 2.5.Itsubi; Shoku Nihongi 19.Tenpyō-Shōhō 5.3.Kōgo/8.12.Kōin; Shoku Nihongi 20.Tenpyō-Hōji

1.7.Kōgo; Shoku Nihongi 22.Tenpyō-Hōji 4.2.Kōshin; Shoku Nihongi 30.Hōki 1.1.Boin; Shoku Nihongi

32.Hōki 3.6.Kasshi; Nihon Kōki 3.Enryaku 13.9.Kiyū/Kigai; Nihon Kōki 16.Daidō 3.3.Kibi/Kōjutsu;

Nihon Kōki 21.Kōnin 2.10.Shinshi; Nihon Kōki 26.Kōnin 9.4.Kōshin; Nihon Kōki 33.Tenchō 2.Urū

7.Kōin). Both Kōken-Shōtoku and her father Shōmu were particular patrons of this sutra and the notion of cakravartin behind it, the former frequently invoking the language of salvation and world- rulership, constructing herself as something akin to a bodhisattva-queen118 (Shoku Nihongi

18.Tenpyō-Shōhō 3.10.Jinshin; Shoku Nihongi 20.Tenpyō-Hōji 1.7.Bogo/11.Jin’in /12.Shingai; Shoku

Nihongi 28.Jingo-Keiun 1.1.Kibi; Shoku Nihongi 29.Jingo-Keiun 3.5.Heishin). Piggott argues that Suiko also made use of the cakravartin image in self-representation119.

The cakravartin, however, was not a gender-neutral being in Buddhist teachings. Within

Japanese Buddhism was found the doctrine of the Five Obstacles (五障, goshō), which listed five types of rebirth unattainable for a woman: as the god Bonten’ō (梵天王; Brahmā), as the god

Taishakuten (帝釈天; Śakra), as the demon Mara (魔羅, also called Maō 魔王; Māra), as a Buddha,

117 Bowring 2005, 67.

118 Bowring 2005, 97.

119 Piggott 1997, 96.

282 and as a cakravartin120. Naturally, the understanding that a woman could not even be directly reincarnated as a cakravartin, let alone be one herself, could be expected to obstruct a jotei’s self- presentation as one. In practice, however, this does not appear to have been the case. Although the gender of the tennō in general sometimes required negotiation – as seen with Kōken-Shōtoku’s accession – her adherence to the same acts of religious performance as a male ruler show little more accounting for gender in Buddhism than they do in native ritual.

Meeks and Ambros argue that the concept of the Five Obstacles had not yet taken hold in

Japanese Buddhism of this time, appearing only in later discourse121; the cakravartin was not yet conceived of as more strictly gendered than the broader concept of kingship, i.e. masculine, but not male-exclusive. Faure instead asserts that the doctrine was present in early Japanese Buddhism, but was only invoked in the sense of a strict exclusion from around the ninth century onwards122. This is consistent with the mediaeval image of the emperor, which more explicitly equated the tennō with the cakravartin123, and which was also markedly bereft of female rule124. Writings from pious women themselves – including the famous Kamo saiin Senshi – suggest some female belief in the Five

Obstacles, but also narratives about how these obstacles may be overcome125. The sole Chinese empress-regnant, Wu Zetian, also depicted herself as cakravartin in spite of doctrine126; other elite women of Nara, such as Kōken’s mother Empress Kōmyō, expressed interest in feminine/female depictions of bodhisattvas, talking past any notion of obstruction127. Thus, even if the concept of the

120 Ōsumi 2002, xxviii; Nagata 2002, 280; Faure 2003, 62; Ambros 2015, 53.

121 Meeks 2010, 4–5; Ambros 2015, 53, 84.

122 Faure 2003, 63.

123 Ooms 2009, 130.

124 Faure 2003, 192.

125 Faure 2003, 64–6.

126 Faure 2003, 189; Bowring 2005, 81.

127 Lowe 2011, 21.

283

Five Obstacles was broadly recognised during the Century of Empresses, it does not automatically follow that a female tennō would have considered herself subject to it. Although the strategy of self- presentation as cakravartin was at least theoretically masculine, if not male-only, female tennō deployed it in a manner largely identical to their male counterparts. Here, also, the jotei work along the same script of kingship as male monarchs, rather than using some ancient example of hime or a

“golden age” to bolster their rule.

The points where jotei did diverge from their male counterparts in use of religion are also illuminating: in building a legitimating image of female rulership, jotei appear to have turned more to Buddhist examples than to images from Japan’s own past. The following two passages occur during the reign of Empress Suiko:

T he Empress requested that the Crown Prince lecture on the Lion’s Roar of Queen Śrīmālā

Sutra. The explanation was finished after three days.128 (Nihon Shoki 22.Suiko 14.7)

The Crown Prince also lectured on the Lotus Sutra at Okamoto Palace. The Empress was

overjoyed. 129 (Nihon Shoki 22.Suiko 14)

Twice in short succession, Crown Prince Shōtoku lectured on sutras on Suiko’s behalf; he is also known to have written or sponsored a commentary on the Vimalakīrti Sūtra, probably also for

Suiko’s sake130. The choice of sutras is not coincidental, either. The Lion’s Roar of Queen Śrīmālā

128 “天皇請皇太子令講勝鬘經。三日。說竟之。” (“Sumera-mikoto hitsugi-no-miko masete Shōmangyō wo tokaseshimu tamau. Mitsuka ni. Toki oenu.”)

129 “皇太子亦講法華經於岡本宮。天皇大喜之。” (“Hitsugi-no-mikoto mata Hokkekyō wo Okamoto-no-Miya ni toku. Sumera-mikoto ōyorokobi tamaite.”)

130 Ambros 2015, 46.

284

Sutra131 depicts the titular Queen Śrīmālā as a wise saviour-queen whose intelligence earns even the

Buddha’s praise132. The Lotus Sutra133, one of the seminal texts of Mahayana Buddhism, speaks of the universality of Buddha nature, and contains in the “Devadatta” chapter an account of the spontaneous transformation and enlightenment of a young nāga princess, an incident which has been interpreted as showing the potential for enlightenment in a female body134. Indeed, the poetry of Senshi draws upon this very chapter as a source of hope for her own enlightenment135; it was similarly popular as a subject for copying by the court women of Heian136. The Vimalakīrti Sūtra137 contains a passage in which a goddess upstages the Buddha’s disciple Śāriputra, teaching him of the illusory nature of gender and sex distinctions by transforming herself into a male and him into a female138. Suiko having Shōtoku lecture or commentate on these texts both formed a part of her self-presentation as a Buddhist saviour-queen akin to Śrīmālā, and justified her accession as a female monarch in Buddhist terms139.

Kōken’s accession, in contrast, made use of Buddhist texts – mobilised by her mother Kōmyō

– which stressed androgyny, her “male attributes” in the (normally masculine) role of Crown

Prince(ss) constructed and publicised to legitimate her accession ahead of a male competitor, Prince

131 Śrīmālādevī-siṃhanāda Sūtra; in Japanese Shōman-Shishikui-Ichijō-Daihōben-Hōkōkyō 勝鬘師子吼一乗大

方便方広経 or Shōmangyō 勝鬘経 for short

132 Piggott 1994, 22; 1997, 95; Bowring 2005, 20; Ambros 2015, 46–7.

133 Saddharma Puṇḍarīka Sūtra, “Sutra on the Lotus of the Sublime Dharma”; in Japanese Myōhō-Rengekyō 妙

法蓮華経 or Hokkekyō 法華経 for short

134 Piggott 1994, 22; 1997, 96; Ambros 2015, 47.

135 Faure 2003, 65.

136 Bowring 2005, 128.

137 Vimalakīrti Nirdeśa; in Japanese Yuimakyō 維摩経

138 Faure 2003, 120; Ambros 2015, 47.

139 Piggott 1997, 95–6.

285

Asaka. In particular, the Golden Light Sutra140, which spelled out a theory of women transforming into male bodhisattvas to transmit Buddhist teachings, was used to provide religious contextualisation for this deliberate deployment of androgyny; a method also invoked by Wu

Zetian141. This provided another model of female rule made acceptable through and within

Buddhism, slightly different in character to Suiko’s but using the same broad strategy: emphasis on

Buddhist texts that presented a more favourable image of female rule, or womanhood in general, and constructing a public image in light of them, while avoiding or simply ignoring more misogynistic readings. Arguably, Buddhist texts and discourse were more used in (specific) support of female rule than the native religious tradition of kami-worship; a distinction that challenges the “golden age” notion of jotei as a holdout of female-empowering native tradition in the face of foreign-derived patriarchy.

6.4.2. Confucianism and Hierarchy

The seventh century saw an increasing adoption by the elite of and reorganising of court around Confucian beliefs, with a particular emphasis on notions of hierarchy142. Imperial promotion of Confucian beliefs in the virtue of filial piety and obedience to authority appears as early as the

Seventeen-Article Constitution of Crown Prince Shōtoku (Nihon Shoki 22.Suiko 12.4.Boshin). It is not hard to see why the latter quality, in particular, was one which a monarch might seek to promote.

Gender was ingrained as another site of hierarchy within Confucian texts: man/woman – in the form of father/daughter, husband/wife, son/widowed mother; the “Three Obediences” (Yili 11; Da Dai Liji,

Benming) – existed as a hierarchical dichotomy alongside father/son and ruler/subject. Concepts which encouraged the subordinate position of women, such as those “Three Obediences” in addition to the “Four Virtues” of womanhood were, if not widespread amongst the populace, at least known

140 Suvarṇaprabhāsa Sūtra; in Japanese Konkōmyō-kyō 金光明経

141 Ooms 2009, 213.

142 Sekiguchi 2003, 28–9; Yoshikawa 2011, 78–9.

286 to the court intelligentsia of Asuka and Nara, as seen by their appearance in poetry (Man’yōshū

5.794) and commentaries on sutras143.

One may expect, therefore, that Confucianism would prove a nuisance to an empress- regnant, a challenge to her authority to be talked around, subverted, or ignored entirely. The dowager-regents of China had enough of a complicated position within Confucianism – a widowed mother was an ambiguous figure, her expected subordination to her son as per the Three

Obediences warring with the respect afforded to her as a parent144 – and they still ruled “as a woman”145, officially the stewards of their sons’ thrones, never (but for the exceptional figure of Wu

Zetian) crossing over into the masculine territory of the crown itself. The jotei of Japan did not even necessarily have the authority of the mother to fall back upon, their relationships with their intended successors not limited to the maternal or grandmaternal.

Nevertheless, jotei engaged with Confucianism not as an obstacle but just as their male counterparts did: as a source of legitimacy, with little negotiation or justification of gender; as if the holding of the position of tennō in some sense superceded their womanhood. The Three Obediences and Four Virtues appears wholly absent from the image of Confucianism promoted by the Yamato regime, which instead focused heavily on the relationship between ruler and subject, and ideals of fealty and obedience to one’s lord.

The Seventeen-Article Constitution, which is credited to Crown Prince Shōtoku during

Empress Suiko’s reign, devotes Articles One and Three to mandating obedience to lords and fathers, and stressing the proper performance of hierarchy as necessary to both natural and social order

(Nihon Shoki 22.Suiko 12.4.Boshin). That the highest-ranking lord of this time was a woman does not

143 Ambros 2015, 46–7.

144 Yang 1961, 47; Chaffee 2001, 22.

145 Psarras 1993, 29.

287 appear as a matter of concern. Although Crown Prince Shōtoku is the most associated with this promulgation of Confucian hierarchy, Suiko herself appears also as an active participant in this process, revising the rites of court to include kowtowing upon entry through the palace gates, signifying subordination to the throne (Nihon Shoki 22.Suiko 12.9). Similar etiquette revisions emphasising hierarchical relationships within the court were made by Jitō (Nihon Shoki 30.Jitō

4.7.Kōshin/Kichū), during whose reign the Confucian scholar Ue no Kudara (上百済) was rewarded for his studies by the court (Nihon Shoki 30.Jitō 7.3.Kōgo). This strict observance of hierarchy was also seen during the regency of Dowager Empress Liu (劉) of Song China146.

Empresses Genmei and Genshō appear as particular patrons of Confucian morality.

Genmei’s accession speech invokes the image of rulership as a duty, carried out benevolently for the benefit of the subjects, a picture which harks to the ideal Confucian relationship of lord and vassal as well as the native divine mandate of the Heavenly Sun Lineage147 (Shoku Nihongi 4.Keiun 4.7.Jinshi).

One of Genmei’s earliest proclamations was a declaration of the absolute importance of “propriety”

(禮; rei/lĭ) to proper governance – “propriety” being one of the cardinal Confucian virtues – in an edict saturated with Confucian language and ideology, and invoking the submissive gesture of kowtowing as an example of lost propriety in need of restoration (Shoku Nihongi 4.Keiun

4.12.Shinbō). Both Genmei and Genshō also promoted Confucian family values via the granting of official rewards to “filial children” (孝子; kōshi), “obedient grandchildren” (順孫; junson), “righteous husbands” (義夫; gifu), and “faithful wives” (節婦; seppu) (Shoku Nihongi 6.Wadō 7.6.Kibi; Shoku

Nihongi 7.Reiki 1.9.Kōshin/Yōrō 1.11.Kichū). These categories all revolve around Confucian family hierarchies: obedience to elders and husbands, and the duty of care of the husband/father towards his faithful dependants.

146 Chaffee 2001, 16.

147 Piggott 1997, 228.

288

Empress Shōtoku, too, made similar rewards (Shoku Nihongi 28.Jingo-Keiun 1.8.Kishi); before, as Empress Kōken, she called for all the subjects to be “filial” in the mourning of her father, the Retired Emperor Shōmu (Shoku Nihongi 19.Tenpyō-Shōhō 8.6.Kōin), and in her repudiation of

Kiyomaro (和気清麻呂) at Dōkyō’s (道鏡) request, she emphasised the duty of the retainer to follow his ruler (Shoku Nihongi 30.Jingo-Keiun 3.9.Kichū). Her gosechi dance as Crown Princess may also have been a display of filial piety through the principles of hierarchy underpinning the dance, constructing her own filial virtue as a sign of her leadership credentials148. Lack of filial piety was similarly used as a pretext to depose her intended successor149.

The relationship between ruler and subject was often likened in Confucianism to the relationship between parent and child: as a child had a duty of obedience to their parent, so too did a subject have a duty of obedience to their ruler; likewise, as a parent had a duty of care and nurture towards their children, so too did a ruler have a duty of care and nurture towards their subjects. The state was in this way a macrocosm of the family, and the family a microcosm of the state. The virtue of filial piety (孝; kō/xiào) was thus indirectly but closely linked with obedience to state authority150.

It is no surprise, therefore, that it was an ideal the emperors and empresses of Nara were keen to promote151. Indeed, imperial edicts often continued the trend of equating the two relationships, comparing their obligations to their subjects with a parent’s love for a child (Shoku Nihongi 4.Wadō

1.2.Boin). In one edict, Genshō quotes directly from the Classic of Filial Piety (孝經, Xiàojīng), a seminal text concerning the Confucian notion of filial piety and its equation with obedience to a ruler

148 Piggott 2003, 54, 57; Sekiguchi 2003, 35; Ooms 2009, 121.

149 Piggott 2003, 56, 59.

150 Piggott 2003, 47; Sekiguchi 2003, 27–9; Ambros 2005, 24.

151 Sekiguchi 2003, 35.

289

(Xiaojing). That the texts themselves were written around male rulers and male parents152 prompts no apparent change in their usage by female tennō.

Again, we see that the language of monarchy and the imperial deployment of religion are by and large the same for jotei as for their male counterparts, even when the doctrine behind that religion would seem to contradict their rule. The jotei is rarely delineated from the male tennō by her gender; and when that delineation does occur, when women’s rule must be justified, it is not on ancient icons of feminine authority, images of some contrapuntal hime-role, that the jotei draws.

Instead, the jotei’s legitimation incorporates elements of Buddhist scripture, notions of heredity and affiliation with a male ancestor, some abstract state of monarchic androgyny. Even when Genmei uses the image of wife-as-helpmeet in support of her reign (see Ch. 6.2.1), it is through the

Confucian framework of the “virtuous wife” rather than any notion of hime and hiko partnership.

6.5. Empresses as Dowager-Regents, Empresses as Placeholder Monarchs

So if empresses-regnant were not a legacy of the ancient hime, then what were they? What about the political and social context of Asuka and Nara allowed for this “Century of Empresses” if not an ancient tradition of female rulership? Let us take another look at the presentation of jotei in the records.

First we shall start with the uncertain figures of Jingū and Iitoyo. We have written before about how, outside from certain fudoki, Jingū’s status as empress-regent rather than regnant was emphasised, but, given that her “regency” was said to last until her death, not only lasting until her son’s adulthood but well into his own old age (traditional dating claims that Ōjin did not take the

152 Piggott 2003, 47.

290 throne until age 70), despite the insistence on titular differentiation it is hard to see the qualitative distinction between her reign and that of a pre-abdication jotei such as Suiko153.

Nevertheless one fact is always made clear: Jingū takes the throne for Ōjin’s sake, acting as a preserver of the patriline. Her most famous deeds as regent are also presented as undertaken on behalf of her son: the conquest of Korea is explicitly said to be a gift delivered by the gods to the unborn emperor; his mother is the facilitator of this divine fiat, not its beneficiary (Kojiki 2.Chūaiki;

Nihon Shoki 8.Chūai 8.9.Kibō). Jingū’s defeat of the rebel princes Oshikuma and Kagosaka is similarly written as her safeguarding not her own position of power, but her son’s birthright (Kojiki 2.Chūaiki;

Nihon Shoki 9.Jingū Kōgō Sesshō 1.2). The image of Iitoyo in the chronicles, too, is similarly positioned as a temporary “compromise candidate” holding court due to uncertainties in the patriline, rather than a legitimate reign; in the Nihon Shoki, especially, her status as a “self-styled” monarch is emphasised (Kojiki 3.Seineiki; Nihon Shoki 15.Seinei 5.1).

For Suiko, the first jotei to receive official recognition as such, the complications were slightly different. Unlike Jingū, Suiko was understood as a reigning monarch, not a regent; and yet, she had a “regent” of her own, in the form of her designated successor Crown Prince Shōtoku:

Prince Umayado-no-Toyoto-Mimi was made Crown Prince. He acted as regent, and was

entrusted with the multitudes of government.154 (Nihon Shoki 22.Suiko 1.4.Kibō)

What it meant for him to be “regent”, and exactly how the power of governance was divided between him and Suiko, is unclear155. It has sometimes been interpreted to mean that Crown Prince

153 Takamure 1966b, 115.

154 “立廄戶豐聰耳皇子爲皇太子。仍錄揶政。以萬機悉委焉。” (“Umayado-no-Toyoto-Mimi-no-Mikoto wo tachite hitsugi-no-miko to nasu. Yorite matsurigoto torifusane kawarashimu. Yorozu matsurigoto wo mote kotogoto ni yudanenu.”)

155 Bowring 2005, 20.

291

Shōtoku was the true governing power of this court, and Suiko a mere figurehead monarch existing to legitimise his position, bereft of real authority of her own; Daniel Clarence Holtom, for example, credits only Shōtoku and not Suiko for the court adoption of Buddhism at this time156, despite passages from the Nihon Shoki which present Suiko actively promoting the new faith (Nihon Shoki

22.Suiko 2.2.Heiin). This is not, however, the impression given within the chronicles themselves, which depicts both empress and crown prince as governing entities, often working in tandem (Nihon

Shoki 22), a format of power through which other scholars such as Okada, Piggott, and Yoshikawa have understood their rule157.

Tsurumi has argued that the assumption that the presence of a male power figure, such as

Shōtoku, overrode the authority of a female monarch, relegating her to the mere status of figurehead, is based in sexist norms which are quick to assume female subordination or delegitimise the position of power held by jotei158. Certainly, textual records depict Suiko as active in statesmanship, far from a mere sacerdotal figurehead159. At the same time, however, it is important to recognise the unique position held by Shōtoku as crown prince, assuming a level of involvement in governance far above what was typical of a hitsugi-no-miko, even one who, like the young

Empress Kōken, was being actively promoted as a successor. This has invoked comparison with the theory of co-rule of the hiko and hime160, but it is not replicated by later female tennō with their intended successors, casting doubt on the idea of this as a pre-existing systematised format of rule.

Instead, their rule is perhaps best understood in its historical context, as the deliberate placement of two Soga-connected monarchs – one current and one (intended) future – on the

156 Holtom 1938, 33.

157 Okada 1970, 394; Piggott 1997, 80–1; Yoshikawa 2011, 17.

158 Tsurumi 1981, 43, 49; Tsurumi 1982, 72–5.

159 Piggott 1997, 80–1.

160 Piggott 1994, 19; 1997, 81.

292 throne by this powerful clan, in the precarious political state following the assassination of the previous incumbent. Both Suiko and Shōtoku had solid connections to previous emperors – Suiko was Kinmei’s daughter and Bidatsu’s widow, and Shōtoku was Yōmei’s son – and each helped to legitimise the other’s claim161. Suiko’s unusual accession as a female monarch was bolstered by the supporting presence of a male co-ruler, who also served to assuage any fears about uncertain succession. Shōtoku’s future accession was smoothed through the current position as Crown Prince, his assumption of the title of “regent” probably intended to further ease his later transition to power, making him something of a “vice-monarch” now so that he would naturally inherit the full monarchy later. This tactic resembles that used by the Roman emperors who, in the absence of an official patrilineal succession, granted their intended heirs the role of co- or junior rulers to legitimise their later assumption of full imperial authority162.

Suiko’s gender was in fact beneficial to Shōtoku’s future accession; being a woman, she could not establish her own patriline to challenge his position. That he predeceased her was an unintended development, and perhaps a contributing factor to the later custom of jotei abdicating in favour of their intended male successor, abdication being not yet customary during Suiko’s time163.

In some sense, Suiko does act as a “placeholder” for Shōtoku’s intended accession, but she was by no means a passive one; she is still apparently endowed with full imperial authority and the will to use it. This distinction is key to understanding the position of the jotei: the systematic intention of their enthronement does not necessarily govern their actual agency on the throne.

The next jotei, Kōgyoku, did indeed abdicate in favour of a male successor164, setting the precedent; Yoshikawa characterises this in circumstance as more a dethronement than a willing

161 Piggott 1994, 18; 1997, 79–80.

162 Corbett 1974, 91.

163 Yoshikawa 2011, 33.

164 Martin 1997, 38–40.

293 abdication165, but agrees that Kōgyoku’s enthronement was intended as an interim appointment, holding the position until her son Naka-no-Ōe – who faced strong competition – came of age166. In her second rule as Saimei, Naka-no-Ōe – who had chosen to remain éminence grise for the time being – appears as much of the true power with his mother as his figurehead, a stark contrast to the cooperation of Suiko and Shōtoku167. Kōgyoku-Saimei was still, in many regards, an active monarch and not purely a passive placeholder, but unlike Suiko her ability to exercise the authority of her position was constrained by surrounding court politics.

Jitō too was an abdicant, an interim monarch holding the throne until her infant grandson

Monmu was old enough to assume the mantle of kingship – although, as Piggott reminds us, this by no means automatically negates her own agency or possession of imperial authority168. Indeed, she continued to hold unofficial authority as a guiding co-ruler even after her abdication169, negotiating continued influence against the demands of the patriline that intended her authority be temporary and ceded to a male heir as soon as possible.

165 Yoshikawa 2011, 57. The selection of Kōtoku as Kōgyoku’s initial successor is an odd one; as Kōgyoku’s younger brother, his claim to the throne was otherwise fairly weak. In the Nihon Shoki, he is presented as the selection of Naka-no-Ōe and Nakatomi (later Fujiwara) no Kamatari (中臣鎌足) based on his seniority as the former’s uncle (Nihon Shoki 25.Kōtoku Tennō Sokui Zenki Kōgyoku Tennō 4.6.Kōjutsu). Yoshikawa argues that

Kōtoku had gained considerable court influence during his sister’s reign, and that this, combined with Naka-no-

Ōe’s relative youth, permitted his accession (57–8).

166 Yoshikawa 2011, 50.

167 Tsurumi 1981, 48; Tsurumi 1982, 73; Yoshikawa 2011, 84.

168 Martin 1997, 43; Piggott 1997, 230–1; 2003, 52–3; Yoshikawa 2011, 163.

169 Yoshikawa 2011, 166.

294

Genmei and Genshō were also placeholder monarchs, as made clear by Genmei’s abdication edict:

“We thus wish to cede these sacred objects to the Crown Prince, but he is still young in years,

not yet able to leave the inner palace and attend to the manifold daily duties of state.”170

(Shoku Nihongi 6.Reiki 1.9.Kōshin)

Genshō was thus enthroned in his stead, for the explicit purpose of holding the throne for the young

Shōmu’s sake. Nevertheless, Genmei and Genshō were also active monarchs in their own right, making such momentous decisions as the founding of Nara (Shoku Nihongi 4.Wadō 1.2.Boin). Their status as placeholders was in no way an obstacle to their ability to exercise the full authority of their role; it by no means automatically relegated them to the role of passive figure.

Although Kōken was not initially intended as a placeholder, her gender nevertheless posed problems, as she could not herself produce a patrilineal heir. Perhaps this is why, in his posthumous edict, her father the Retired Emperor Shōmu designated Prince Funado (道祖) as her successor

(Shoku Nihongi 19.Tenpyō-Shōhō 8.5.Itsubō). Funado, however, never succeeded Kōken, who had him removed from the position on charges of improper conduct (Shoku Nihongi 20.Tenpyō-Hōji

1.3.Teiyū); this could be read as an exercise of her own agency against the constraints of the role of placeholder171. Under pressure from Fujiwara no Nakamaro (藤原仲麻呂) and her mother, Kōken did eventually abdicate in favour of Funado’s replacement, Junnin, but continued to hold authority as a co-ruler in defiance of her relegation to the role of placeholder172.

170 “因以此神器。欲讓皇太子。而年齒幼稚。未離深宮。庶務多端。一日萬機。” (“Yorite kono jinki wo mote kōtaishi ni yuzuran to suru tomo. Shikashite nenrei yōchi ni shite. Mada shingū wo hanarezu. Shomu tatan ni shite. Ichinichi banki ari.”)

171 Piggott 2003, 56.

172 Piggott 2003, 57; Bowring 2005, 95.

295

After her mother’s death, Kōken began to actively retake much of the power she had ceded, ousting Junnin and Nakamaro and re-enthroning herself (Shoku Nihongi 24.Tenpyō-Hōji 6.6.Kōjutsu;

Shoku Nihongi 25.Tenpyō-Hōji 8.10.Jinshin). It is perhaps no surprise that, ruling again as Shōtoku, she designated no heir; although some have interpreted her advisor-monk Dōkyō as the true power behind the throne, as with Crown Prince Shōtoku, he appears more of a vice-monarch than anything else. While Dōkyō evidently held considerable power, was greatly trusted by the empress, and was extremely influential in court, there is no evidence that Shōtoku was at all a passive figurehead or ceded her authority wholly to him, other than the assumption that the real authority must be solely held by a man173. Although many of the jotei negotiated the rule around the role of placeholder- monarch, Kōken-Shōtoku alone seems to have actively rebelled against it.

But how are we to understand the jotei in the broader political context of Asuka and Nara

Japan, if not a legacy of hime? Let us turn briefly to the example of imperial China. Although China had only one empress regnant throughout its history, within the imperial system – particularly during the Han, Wei, Yuan, and Qing dynasties – there was a strong tradition of “dowager regencies”, in which a widowed empress, rather than a male minister, would act as regent for a boy-king174, much like Jingū’s mythic role. Indeed, unlike the Fujiwara regents of Heian, male regents were rare in Chinese imperial history175; perhaps it was feared that they might, in the manner of Richard III of

England, prove unwilling to relinquish authority once given to them. Women, being unable to establish a competing patriline, were less of a danger to the dynasty and, if the mother of the boy- king, had a vested interest in protecting his position. Such regencies were even systematised during the Later Han, with specific procedures in place for the institution of a dowager-regency176.

173 Tsurumi 1982, 75; Piggott 2003, 62.

174 Yang 1961, 48–60.

175 Yang 1961, 54.

176 Yang 1961, 55.

296

What has this to do with Japan? The Yamato monarchy, during the Asuka and Nara eras, was in a period of flux. The old model of high kingship was being replaced by the Sinified image of the emperor177. The “Century of Empresses” occurred during the awkward middle stages of this transition, when new preferences in succession brushed up against a native idea of kingship which would not yet permit a child to take the throne. Thus, the interim enthronement of a woman to hold the position while the male heir matured178; a translation of the Sinic dowager-regency onto the

Japanese monarchic system179, with the key difference being that the jotei, unlike the regents, had full official access to monarchic authority in their own right.

The Century of Empresses stemmed from much the same impulses and intended purposes as Chinese dowager-regencies – the idea of a woman as a “safe” interim candidate while an intended male successor matured – but superimposed upon a different monarchic system, which apportioned a greater amount of authority to the female incumbent through its insistence that she hold the throne in the meantime. In this way, a broader patriarchal social order – one which mandated patrilineal inheritance of the throne – gave rise to a specific position of female authority; the jotei was not so much mediated out of ancient acceptance of female power, as she was created by the demands of the patriarchal court. The changing court landscape of this time can also be seen in the practice of abdication to pave the way for a successor: unheard-of in the beginning of the

Century of Empresses, becoming normalised among the jotei beginning with Kōgyoku, and later appearing in the first male example with Shōmu.

Indeed, Kōken’s unusual accession is something of the exception that proves the rule: the intense program of justifications and legitimations embarked upon by the royal family prior to her accession, up to and including Shōmu’s abdication in her favour, is not seen with previous empresses,

177 Sekiguchi 1987, 19–20.

178 Piggott 1997, 230–1, 239; 2003, 53–4, 65; Yoshikawa 2011, 166; Ambros 2015, 32.

179 Piggott 1994, 19; 1997, 81.

297 suggesting a greater anxiety over her position than with previous jotei. That Kōken alone of the

Century of Empresses was (initially) not enthroned as an interim candidate at all but as an heir in her own right, even in preference to an acceptable – if lower in maternal rank – male successor, would explain why her accession was that much more precarious than the five female tennō preceding her180. A woman as a stopgap monarch was acceptable, a woman as a designated heir still required justification; the acceptance of female tennō was conditional. It would also explain why their examples do not appear to have been used as justifications for her own rule: if they were understood as placeholders and she was not, then their script would not be salient for her.

Conversely, that so many of the jotei abdicated in favour of their successor, whereas Shōmu’s doing so as a male emperor was a first not only for this period but in Japanese history, only highlights the separate status that the jotei – for whom abdication was the norm – held from the male emperors, whose tenures were at this time expected to end only with their deaths181.

It is important, however, to remember that an interim monarch is not the same thing as a figurehead. Although the reigns of the jotei may have often been intended as temporary, there is no evidence that during their tenure they were invested with anything less than full regal authority – and most were apparently quite possessed of the will to use it. To recognise that these women were not drawing upon some ancient tradition of female empowerment in their reigns should not mean denying them of their agency within an admittedly male-centric system, or erasing their very real and significant actions within the scope of Japanese history182. We must still consider the jotei not as passive playing-pieces in patriarchal politics, but as actors within – and sometimes in opposition to – a broader socio-political system: this understanding being the core of women’s history183.

180 Piggott 2003, 65.

181 Yoshikawa 2011, 166.

182 Piggott 1997, 231; Kidder 2007. 137.

183 Faure 2003, 5; Alameen 2013, 14–5; Rothschild 2015, 15, 75; DiLuzio 2016, 239.

298

Within the linear path of decline charted by the “golden age” narrative, female relationships to power – sacral or temporal – from the Asuka period onwards are always configured in terms of transformation, of descent; women’s access to power only coming through remnants of earlier, greater roles of female authority. The underlying assumption is that positions of female power could not naturally emerge out of a society which was in other aspects becoming more patriarchal in character; they could only exist, in this context, as vestiges of a pre-patriarchal era, surviving against the tide. In this model, patriarchy appears as a monolithic institution, subordinating all women equally in all respects, in a direct line of descent from empowerment to subjugation.

But as we have seen, the jotei of the “Century of Empresses” did not align themselves with ancient traditions of hime-queens, or female sacral rulership; rather, they alternated delicately between two basic tactics, the first being replicating the ritual and self-presentation of male tennō, thereby cloaking themselves in a sort of monarchic androgyny. This process shows itself heavily in the sacral sphere, where jotei assumed the same sacerdotal role as their male counterparts (rather than a contrapuntal female position) and used the same rhetoric of cakravartin and Confucian virtue despite its apparent exclusion of women. The second tactic was the introduction of discourses legitimating female rulership, such as Suiko’s promotion of sutras with positive images of women and queenship, or the edicts surrounding Kōken’s accession arguing (directly and indirectly) for her suitability as a successor to Shōmu. Where jotei cleaved to tradition, they primarily stuck to the masculine traditions of the Yamato paramount, rather than any feminine hime position; where they did present arguments for female rule, it was mainly through “newer” concepts such as Buddhist sutras rather than an appeal to the traditions of a distant past.

The assumption within the “golden age” narrative-derived view of the jotei-as-hime is that the jotei represented a unified, systematic category, enthroned with the same goals and parameters, exercising the same level of agency. But what we see from an examination of these women and their

299 reigns is variety. We see a range of different accession circumstances and relationships to previous monarchs: Suiko’s dual candidacy with Shōtoku, Kōgyoku-Saimei’s accessions engineered amongst coups, Jitō’s carrying on the work of her late husband, Genmei and Genshō’s baton-passing stewardship of the throne, and Kōken-Shōtoku’s appointment to Crown Princess and active dethronement of her successor. We see different strategies of legitimating rule, from Suiko’s queenly sutras to Jitō’s “virtuous wife” partnership to Kōken’s concerted pre-accession campaign.

We even see different behaviours, ranging from the relatively passive figure of Kōgyoku-Saimei to the bold, norm-defying reigns of Kōken-Shōtoku. Rather than a clean, cohesive category, the jotei appear as a highly variable construction, the product of a transition period between old and new images of rulership.

Rather than a remnant of an ancient hime, the jotei of the “Century of Empresses” appears as a situationally-formed gestalt; a role of female authority born out of a patriarchal social order at a turning point between two systems of monarchy. There is no contradiction in this, in circumstance requiring a new role of empress regnant within a broader patriarchal framework, for patriarchy, as we have seen, is not a single social force but a complex network of relationships; societal change, likewise, is not always strictly linear, and a spike may exist in the graph, a position of female power unbound to a previous image. The jotei, while women, were also royalty, at the apex of the class system, their social status allowing them considerable authority even within a patriarchal society; furthermore, their accessions often served the goals of powerful clans such as the Soga or the

Fujiwara. Despite the court intrigues surrounding their enthronements, however, and despite the patriarchal society in which they lived, it is clear that the jotei nevertheless appear as active, intriguing women, navigating their own way through monarchic authority and displaying their own unique personalities and strategies: “actors within a system”.

300

CHAPTER 7: TOWARDS A NEW UNDERSTANDING OF WOMEN’S POWER IN ANCIENT JAPAN

Over the course of this research, we have examined the “golden age” narrative and its various surviving threads against textual evidence. Through this process, we have identified many areas where greater nuance can be encompassed in our studies of ancient Japanese women, and modern avenues of scholarship which complicate previous understandings and narrative threads.

7.1. Conclusion

7.1.1. Homogenisation and Diversity

One issue we encountered frequently with the “golden age” narrative and its framing is that of homogenisation. The “golden age” narrative and its subsidiary theories – such as the “himehiko system,” the notion of female spiritual predominance, or the interpretation of the saigū as the declined remnant of a once-greater role – present a single model of female “power” across the archipelago, a single relationship between women and spirituality, women and authority, for all of

“ancient Japan”.

One complication which presents itself is the issue of regional diversity. Ancient Japan was not always a single, unified state, after all, and included many cultural groups and practices distinct from Yamato. We see, for example, in the Eastern Provinces a tradition of male shamanism which runs counter to Yamato’s own feminine gendering of the role. We see among those clans of Kyushu designated tsuchigumo – delineated from Yamato authority – a comparatively large number of female chieftains appearing in the records, clashing with the Yamato imperial clan’s own patriline.

We see women appointed to provincial bureaucratic posts well after central women lost access to

301 similar positions of status. Such examples remind us that ancient Japan – particularly before the

Yamato program of centralisation – was not a monolith, but rather was culturally and societally variable. With this in mind, we see that a single model, a single narrative of women’s status and relationship to temporal and spiritual power, struggles to fully encompass the full breadth of power and gender structures across the ancient archipelago.

The disparity between provincial and central custom also highlights another point of diversity in ancient Japan, existing even under Yamato centralisation: the division between centre and periphery, elite and common, “official” and “unofficial.” We mentioned above a provincial disparity in women’s access to bureaucratic status; we have seen also that female village chiefs continued to hold authority and manage economic projects despite lacking the “official” bureaucratic status of the village head. Within the “golden age” narrative, the attribution of temporal authority purely to positions recognised by the official bureaucracy has stymied full exploration of the complexities of women’s status under the ritsuryō state. We have seen that patriarchal marital laws often did not correspond to actual practice; we have also seen that many miko continued to enjoy considerable social standing and belief in their powers despite condemnation from (a subset of) the Yamato elite. The “golden age” narrative and its derivatives’ tendency to hyperfocus on “power” at the central elite level obscures the diversity of women’s experiences across social strata, and homogenises power and social status into a singular, centrally- granted quality.

Such homogenisation of power into a singular quality also results in the conflation of various

“types” of power, such as temporal authority and spiritual acuity, as seen in discourse related to the

“himehiko system”: the conflation of, for example, a lauded priestess with a female chieftain who otherwise evinces no sacral function. The notion of the “himehiko system,” an enduring thread of the “golden age” narrative, elides the nuances of various female figures’ power relationships – to sacral power, to temporal authority, and between the two – in favour of a singular, monolithic image

302 of prominent women in Japanese history. It is important that we encompass the diversity of these women’s lives, paths to prominence, and formats and structures of “power” (in whatever sense applicable); a multiplicity that the singular model of the hime in her singular “himehiko system” is insufficient to contain.

7.1.2. A Non-Linear Model of Patriarchy

We also see considerable complexity in the matter of gender in ancient Japan, a complexity which defies a neat dichotomy between earlier complete empowerment versus later complete patriarchy. Across the archipelago, we see variability in gender norms. Within Yamato culture, the liminality of women is both the source of sacral power for the miko, and the source of defilement for menstrual blood pollution. Marriages appear to require female (but not male) monogamy, but their fluidity, preference for duolocal arrangements, and prioritisation of familial over marital bonds all ameliorate their capacity for patriarchal oppression. Inheritances are bilateral, with an apparent preference for the paternal side but considerable significance nevertheless accorded to maternal kinship. The paramountcy, normatively a masculine occupation, can nevertheless be held by a woman, with varying amounts of negotiation necessary. The path of female empowerment to patriarchal decline charted by the “golden age” narrative does not serve to encompass these nuances of gender in pre- and early-ritsuryō Japan, which appears as neither matriarchal nor purely gender equal, whilst still remaining clearly distinct from the later model of the mediaeval patriarchal household.

Finally, we note that the “golden age” narrative-derived discourse on the saigū, saiin, and jotei positions them as transformations of a singular hime-archetype, albeit lessened in scope and/or authority due to the degradation of patriarchy. This, too, is an insufficient framework for understanding these roles: firstly, because of the already-explored issues with the idea of a singular hime-figure; and secondly, because this linear narrative of patriarchal decline – in which all female roles are equally and strictly declined over time, and no forms of female power may exist within a

303 patriarchal system except as remnants of something earlier and greater – does not satisfactorily explain the particulars of these roles. Rather, the saigū and jotei are perhaps best understood as incorporating a variety of disparate elements and traditions. For the saigū, these include earlier

Yamato models of female shamanism and imperial clan-priestesses, Yamato programs of territorial expansion, power consolidation, and religious centralisation, and stringent concerns over ritual purity possibly deriving from a very ancient practice. For the jotei, these include Yamato attitudes to ancient female chieftaincies and paramountcies, a period of flux amidst a changing dynastic system, inspiration from Chinese dowager regencies, and political manoeuvring by both imperial and ministerial clans.

7.2. Within the Wider Field

Increasingly, modern scholarship is shifting to encompass the complexities of women’s history in ancient Japan. Okano, Tanaka, and Ambros have questioned the broader existence of a

“golden age” of female power1. Yoshie’s work on women village chiefs and Meeks’ research on miko under the ritsuryō system, for example, have raised the importance of distinction between official bureaucratic position and social status as a whole2. Sekiguchi, Faure, Meeks, and Ambros have articulated the need for more varied study on women’s lives and experiences in ancient Japan, noting class and region as possible sites of difference3.

This research, then, continues in that vein, calling for greater encompassing of the nuance and variety appearing within the lives and records of ancient Japanese women. Through this thesis, we have identified several areas where the complexities and points of difference in this history have

1 Okano 1993; Tanaka 1996; Ambros 2015.

2 Yoshie 1993; Meeks 2011.

3 Sekiguchi 1987; Faure 2003; Meeks 2011; Ambros 2015.

304 perhaps gone unexplored. In doing so, we build upon the work of previous scholars in critiquing the

“golden age” narrative and its depictions of ancient womanhood, incorporating their conclusions and developments into our work. In particular, we have developed on Yoshie and Meeks’ work on the complexities within the notion of “power” and the danger of conflating central elite perspectives with ancient Japan entire, applying these concepts across the broader study of ancient women’s lives and positions. We have also expanded upon Sekiguchi, Meeks, and Ambros’s points on the multiplicities of women’s experiences across space and class, identifying several elements of regional and cultural difference worthy of further study, and taken up Faure’s theoretical work on the nuances within patriarchy as a lens through which to study the ritsuryō miko and jotei.

This research also highlights the issues contained within enduring “threads” of the “golden age” narrative such as the notion of the contrapuntal “himehiko system” as the singular widespread format of authority in ancient Japan, or the idea of saigū and jotei as pure transformations of some ancient feminine role of authority distinct from the masculine. By challenging these “threads,” we hope to open up space for greater nuance in our understanding of women’s lives and histories in the ancient period.

7.3. Avenues of Further Study

Over the course of this research, we have identified several new avenues of study to further develop this more nuanced understanding of female power and status. For example, we have identified regional disparity as a complication oft-overlooked by the traditional “golden age” narrative, with Kyushu and the Eastern Provinces standing out in particular. Many of the records’ apparent male-female ruling pairs, used as examples of himehiko by the traditional narrative, were attributed within the text to Kyushu, and in particular those clans labelled tsuchigumo: the pairings of Usatsu-Hiko/Usatsu-Hime, Asotsu-Hiko/Asotsu-Hime, and Natsuha/Taburatsume-Hime, as well as

305 the sole female chieftain Hayatsu-Hime, for example (see Ch. 4.3.4). This discrepancy is significant enough to merit further investigation, combining textual evidence with archaeological studies of regional chieftain burials to investigate the role of gender in rulership customs for specific regions, rather than a monolithic focus on “ancient Japan” as a whole.

The Eastern Provinces (modern Kantō) are also of interest due to an elusive reference in the

Ryōjin Hishō to male “otoko-miko,” performing a ritual role that was in central culture gendered female: that of divine possession (Ryōjin Hishō 2.556). Although the astonishment shown by the central commentator to the notion has been used as proof of the female gendering of shamanic performance in ancient “Japanese” culture4, the other suggestion contained within the passage – that this gendering could be regionally variable – also presents a ripe field for further exploration.

Developing a more thorough and nuanced understanding of the relationship between gender and sacral power in ancient Japan requires that these disparities and diversities, and what more can be gleaned from them about regional traditions, be properly explored and encompassed.

A fairly extensive body of scholarship does exist on many regional religious roles, such as the itako, the ogami-san, or the gomiso of Tōhoku, and their ritual performance and social function in contemporary Japan5. Given the importance, to our restructured narrative, of appreciating regional diversity in female religious performance and social status, the challenge therefore is to extend this work to the historical context, tracing the origins, ritual roles, and social significance of such positions throughout history. Tōhoku is of particular interest here, as a site very much on the frontiers of the ancient court’s sphere of control, as well as the home of a distinct ethnic group – the

Emishi – whose own traditions and cultural legacy have gone overlooked in the study of Japanese religious history. Investigations into the cultures and religious practices of other lost or assimilated ethnic groups of ancient Japan – such as the Kumaso and Hayato of southern Kyushu, or the

4 Hori 1968, 181–2.

5 Hori 1968, 201–14; Blacker 1982, 128, 140–177, 238–40, 252–71.

306

Mishihase of Sado Island – could also add nuance to the narrative and provide further context for a study of regional diversity.

In broad summation, it can be seen that the “golden age” narrative of female power in ancient Japan, and the narrative “threads” such as himehiko that it has introduced to the scholarly discourse on gender and power, are heavily flawed, oversimplifying complex issues of power relations and social dynamics and homogenising a diverse array of examples and ancient women’s experiences. Instead, our research calls for a critical re-examination of the tropes of the “golden age” narrative still present in contemporary scholarship, and the development of a new narrative lens that holds space for these complexities and multiplicities. Through doing so, we may attain a deeper, more nuanced understanding of the historical realities of women in ancient Japan, reclaiming (rather than overwriting) the “hidden voices” of history.

307

APPENDIX: A TIMELINE OF ANCIENT JAPAN

JAPANESE PALEOLITHIC (旧石器時代): 40,000–14,000 BCE

Human habitation in Japan has been conclusively proven since at least 35,000 BCE, although surviving evidence other than stone tools is scarce. Prior to the discovery of the first Paleolithic site in 1946, it had long been assumed that human habitation of Japan began with Jōmon, and thus this period is not represented in older historical works; even in modern scholarship, the paucity of information presents a challenge, and renders this era largely outside the scope of this research.

JŌMON PERIOD (縄文時代): 14,000–1000 BCE

Japan’s Mesolithic/Neolithic period, marked by small hunter-gatherer communities and known primarily through archaeological findings. In earlier reckonings, this period extends as late as 300

BCE. The Jōmon period is largely absent from the official imperial timeline of the records, which situates everything prior to 667 BCE within the numinous “Age of Gods” (神代), the divine age of creation and establishment, in which first appear divinities such as Izanami (伊邪那美/伊奘冊),

Amaterasu (天照), and Ame-no-Uzume (天宇受売/天鈿女).

YAYOI PERIOD (弥生時代): 1000 BCE–300 CE

Marked by a new wave of migration from mainland Asia and the emergence of wet rice agriculture; in earlier reckonings, this period begins as late as 300 BCE. Power structures based on small networks of hamlets. Japanese records for this period are in considerable doubt, and dates especially spurious.

308

Women of Yayoi

667 BCE: Usatsu-Hiko and Usatsu-Hime (宇沙都比売/菟狭津媛) receive future Emperor Jinmu in

Kyushu

663 BCE: female chieftains Nagusa-Tobe (名草戸畔) and Nishiki-Tobe (丹敷戸畔) are killed by

Jinmu’s forces in Kii (modern Wakayama); Tomiya-Bime (登美夜毘売)/Nagasune-Hime (長髓媛) is married off following the defeat of her brother in Yamato

662 BCE: female chieftain Niiki-Tobe (新城戸畔) is mentioned in Yamato

97 BCE: possible female chieftain Arakawa-Tobe (荒河刀弁/荒河戸畔) of Kii Province is mentioned

92 BCE: Toyosuki-Irihime (豊鉏入日売/豐鍬入姫) is charged by her father Emperor Sujin with the worship of Amaterasu, becoming the proto-saigū; Nunaki-Irihime (沼名木之入日売/渟名城入姫) is similarly charged with worship of Yamato-no-Ōkuni-Tama but is deemed unfit

91 BCE: Yamato-Totohi-Momoso-Hime (夜麻登登母母曽毘売/倭迹迹日百襲姫), daughter of

Emperor Kōrei and great-aunt of reigning emperor Sujin, performs a divination on Kamu-Asaji Plain

88 BCE: Yamato-Totohi-Momoso-Hime interprets an oracle for Suinin, concerning the rebellion of

Take-Haniyasu-Hiko and his wife Atahime (吾田媛); after this, she is married to the god Ōmono-

Nushi, but dies after offending the deity

28 BCE: Sahohime (沙本毘売/狭穂姫) becomes Suinin’s empress

26 BCE: Sahohime’s brother Sahohiko attempts to cajole his sister to assassinate her husband with promises of co-rulership

25 BCE: Sahohime cannot go through with the plot, but chooses to perish alongside her brother

309

5 BCE: Yamato-Hime (倭比売/倭姫), daughter of Suinin, takes over worship of Amaterasu from

Toyosuki-Irihime, travelling to Ise and founding the Inner Shrine

58 CE: Ō-Nakatsu-Hime (大中姫) takes over stewardship of the “sacred treasures” from her brother

Inishiki

82–90 CE: the westward travels of Emperor Keikō around Kyushu, during which time many female tsuchigumo chieftains are placed in the fudoki, including Yasome (八十女), Mirukashi-Hime (海松橿

媛), Hayakitsu-Hime (速来津姫), and Ukianawa-Hime (浮穴沫媛)

82 CE: female chieftains Kamu-Natsuso-Hime (神夏磯媛) and Hayatsu-Hime (速津媛) attested in

Kyushu

88 CE: Emperor Keikō encounters Asotsu-Hiko and Asotsu-Hime (阿蘇都媛) in Aso Province, Kyushu

90 CE: Keikō’s daughter Iono (五百野)/Kusu-Hime (久須姫) possibly sent to saigū, but absent from

Kojiki and conflicting with Yamato-Hime’s stated tenure

97 CE: Yamato-Takeru is sent to fight the Kumaso; in the Kojiki accounting, first conferring with

Yamato-Hime

110 CE: Yamato-Takeru confers again with Yamato-Hime before heading to the Eastern Provinces, during which campaign he kills Kitsu-Hiko and achieves the surrender of Kitsu-Hime (寸津毘売)

192–200 CE: a saigū for Chūai, Iwashima (伊和志真), is attested in the Nisho Daijingū Reibun but absent in central records

193 CE: Jingū (神功)/Okinaga-Tarashi-Hime (息長帯比売/気長足姫) becomes Chūai’s empress

200 CE: Empress Jingū delivers two oracles, executes the female tsuchigumo leader Taburatsu-Hime

(田油津媛), and leads the conquest of Korea

310

201–269 CE: regency of Empress Jingū

238–247 CE: dates attested by Chinese sources to the reign of Queen Himiko (卑弥呼), competing with Jingū’s place in the Yamato timeline (Records of the Three Kingdoms)

238 CE: Himiko is named “Friend to Wei”

247 CE: Himiko is at war with neighbouring country of Kunu

247–249 CE: rough time range for Himiko’s death, upon which she is succeeded first by an

unsuccessful male candidate, and then by Iyo (壱与)/Toyo (台与) (Book of Liang)

Other tsuchigumo and kuzu women appear in fudoki records without specific time periods attributed to them, but often linked to the broader program of Yamato expansion. These include Ōyamadame

(大山田女), Sayamadame (狭山田女), and Itsuma-Hime (五馬媛) of Kyushu, and Abura-Okime (油

置売) of the Eastern Provinces.

KOFUN PERIOD (古墳時代): 300–538 CE

This period is named for large keyhole-shaped burial mounds, which show the presence of powerful regional chieftains, among which the Yamato paramountcy reigned supreme. This period also marks the arrival of written language in Japan, leading to increasing reliability of records and dates.

Women of Kofun

457: Takuhata-Hime (栲幡娘姫)/Waka-Tarashi-Hime (若帯比売/稚足姫), daughter of Yūryaku, becomes saigū

459: falsely accused of loss of virginity, Takuhata-Hime commits suicide and is proven innocent

484: possible (if short-lived) reign of Iitoyo (飯豊), between Emperors Seinei and Kenzō

311

507: Sasage (佐佐宜/荳角), daughter of Keitai, becomes saigū

ASUKA PERIOD (飛鳥時代): 538–710 CE

Power of the Yamato Dynasty already evident by this time, with power increasingly centralised.

Emergence of ritsuryō bureaucracy, a reordering of government and legislature based on Tang

Chinese models. Considerable increase in surviving textual evidence; records and dates for this period considered historically reliable. This period also saw the rise and fall of the powerful Soga

Clan, the adoption and propagation of Buddhism by the Yamato court, and the first half of the

“Century of Empresses”.

Women of Asuka

541: Iwakuma (磐隈), daughter of Kinmei, is appointed saigū but dismissed for loss of virginity

541–579: tenure of “stand-in” saigū Miyako (宮子), daughter of High Priest Kogoto, according to

Kamakura records (Nisho Daijingū Reibun; Toyouke Daijingū Negi Honin Shidai)

576: Suiko (推古)/Nukatabe (額田部) becomes Bidatsu’s empress

578: Uji (菟道), daughter of Bidatsu, is appointed saigū but dismissed for loss of virginity

584: Japan’s first Buddhist nuns – Zenshin-ni (善信尼)/Kuratsukuri no Shima (鞍部嶋), Zenzō-ni (禅

蔵尼)/Ayahito no Toyome (漢人豊女), and Ezen-ni (恵善尼)/Nishikori no Ishime (錦織石女) – are established by the Soga clan

585: Sukate-Hime (酢香手姫), daughter of Yōmei, is appointed saigū

312

592–628: reign of Empress Suiko/Toyo-Mike-Kashikiya-Hime (豊御気炊屋比売/豊御食炊屋姫), first officially-recognised Empress Regnant of Japan

592: Suiko takes the throne following the assassination of Sushun, beginning the “Century of

Empresses”

593: Suiko’s nephew Crown Prince Shōtoku is appointed regent

622: Sukate-Hime retires her post; Shōtoku passes away

628: Suiko passes away, succeeded by Jomei

630: Kōgyoku (皇極)/Takara (宝) becomes Jomei’s empress

642–645: reign of Empress Kōgyoku/Ame-Toyo-Takara-Ikashi-Hitarashi-Hime (天豊財重日足姫)

642: Kōgyoku ascends the throne following her husband Jomei’s death, and performs a

successful rainmaking ceremony

645: Kōgyoku abdicates in favour of her brother Kōtoku, who swiftly institutes the Taika

Reforms, marking the beginning of the ritsuryō state

655–661: Kōgyoku reinstalled on the throne as Empress Saimei (斉明) following Kōtoku’s death

657: Jitō (持統)/Uno-no-Sarara (鸕野讃良) becomes the principal consort of the future

Emperor Tenmu

661: Saimei passes away, succeeded by her son Tenji

673: Ōku (大来), daughter of Tenmu, installed as the first of a revived saigū program

313

686–697: reign of Empress Jitō/Takama-no-Hara-Hirono-Hime (高天原広野姫)

686: Jitō ascends the throne following her husband Tenmu’s death

697: Jitō abdicates in favour of her grandson Monmu; continues to hold significant political

role

700: Satsuma-Hime (薩末比売), Kume (久売), and Hazu (波豆) named among ringleaders of a

Hayato rebellion

703: Jitō passes away

NARA PERIOD (奈良時代): 710–794 CE

This period is marked by the foundation of Japan’s first permanent capital, Heijō-kyō (modern Nara), in 710. Abundance of literary evidence increases. This period represents the height of court Buddhist fervour and patronage, with powerful temples flourishing in the new capital. The Fujiwara Clan becomes increasingly prominent. The second half of the “Century of Empresses” covers the bulk of this period.

Women of Nara

705–715: reign of Empress Genmei (元明)/Ahe (阿閇)/Yamato-Neko-Amatsu-Miyo-Toyokuni-

Narihime (日本根子天津御代豊国成姫); tenure of saigū Chinu (智努) and Madokata (円方) according to the Ichidai Yōki

705: Genmei ascends the throne following her son Monmu’s death

712: Genmei rewards the virtuous widows Iehara no Onna (家原音那) and Ki no Onna (紀

音那)

314

715: Genmei abdicates in favour of her daughter Genshō (元正)/Hidaka (氷高)

715–724: reign of Empress Genshō/Yamato-Neko-Takamizu-Kiyo-Tarashi-Hime (日本根子高瑞浄

足姫)

715–721: Kuse’s (久勢) tenure as saigū (parentage unknown)

721: Genmei passes away

721–744: Shōmu’s daughter Inoue (井上) made saigū, continuing into her father’s reign

724: Genshō abdicates in favour of her nephew Shōmu

738: Princess Abe (阿倍) (future Empress Kōken-Shōtoku (孝謙称徳)) is named Shōmu’s Crown

Princess, the first such in Japanese history

741: Shōmu orders the construction of provincial Buddhist temples and monasteries, the latter perhaps spurred on by his wife Empress Kōmyō (光明)/Fujiwara no Asukabe-Hime (藤原安宿媛)

743: Abe dances the gosechimai before Shōmu and Genshō

746: Court lady Owari no Ogura (尾張小倉) appointed kuni-no-miyatsuko

748: Genshō passes away

749: Shōmu makes a pronouncement declaring the right of daughters to inherit position from fathers

749–758: reign of Empress Kōken (孝謙)/Takano-Hime (高野姫)

749: Shōmu abdicates the throne to Kōken, smoothing the way for her accession

749–752: tenure of saigū Oyake (小宅), according to Ichidai Yōki

315

757: Kōken rejects successor chosen by Shōmu, appoints third cousin Junnin instead

758: Kōken abdicates in favour of Junnin

762: Kōken takes monastic precepts at Hokkeji as a rift widens between her and Junnin

764–770: Kōken seizes back the throne, reigning again as Shōtoku (称徳)

764: Kōken-Shōtoku is victorious against Fujiwara no Nakamaro, retaking her throne

768: oracular crisis centring around Shōtoku’s favourite, the monk Dōkyō

770: Shōtoku passes away, ending the “Century of Empresses”

HEIAN PERIOD (平安時代): 794–1185 CE

Marked by the foundation of the new imperial capital of Heian-kyō (modern Kyoto) in 794, this period is famous for its splendorous and intellectual but insular court culture, producing many renowned works of literature and poetry but increasingly detached from provincial matters, with focus heavily on the centre. A gradual but noticeable decrease of women’s status in marriage, inheritance, etc. may be tracked as popular custom slowly falls in line with pre-existing patriarchal legal codes. The role of the “official” Buddhist nun is in decline with the rise of the Tendai sect and its practice of nyonin kekkai. As a transition point between ancient and mediaeval Japan, and well past the “golden age” of Jōmon, Yayoi, and Kofun, this period features only sporadically in this thesis.

316

Women of Heian

795: Kamitsukeno no Ekunime (上毛野兄国女)/Shoten (諸天) exiled to Tosa for unlawful prophecy

796: Ikue no Iemichime (生江家道女), the “Lay Devotee of Koshi,” sent back to Echizen for unlawful preaching

810: Uchiko (有智子), daughter of Saga, becomes the first appointee to the new role of Kamo saiin

823: Saga makes a rare visit to his daughter, the first reference to her in the Nihon Kōki

831: Uchiko retires from the post of saiin, replaced by Ninmyō’s daughter Tokiko (時子)

975–1031: the lengthy tenure of Senshi (選子), the “Great Saiin”; an acclaimed poetess, Senshi runs a literary salon out of her compound and is contemporary with female writers Sei Shōnagon (清少納

言; The Pillow Book) and Murasaki Shikibu (紫式部; The Tale of Genji)

KAMAKURA PERIOD (鎌倉時代): 1185–1333 CE

Marked by the establishment of the military government of the Kamakura Shogunate and increasing irrelevance of the imperial court, this period is frequently defined as the beginning of Japan’s mediaeval era (中世), ending the ancient period (古代). Hierarchic patriarchal values have by now firmly and clearly set in. This period is outside the scope of this thesis.

317

BIBLIOGRAPHY

PRIMARY SOURCES

Anonymous, Bungo-no-Kuni Fudoki (Gazetteer of Bungo Province), Shinpen Nihon koten bungaku

zenshū Vol. 5, Shōgakukan

Anonymous, Fudoki Itsubun (Extracts of Lost Fudoki)

Anonymous, Gangōji Garan Engi narabi ni Ruki Shizaichō (Origins of the Gangōji Monastery and its

Assets), accessed at http://www004.upp.so-net.ne.jp/dassai1/gangoji/gangoji_frame.htm on

08/01/2018

Anonymous, Hitachi-no-Kuni Fudoki (Gazetteer of Hitachi Province), Shinpen Nihon koten bungaku

zenshū Vol. 5, Shōgakukan

Anonymous, Hizen-no-Kuni Fudoki (Gazetteer of Hizen Province)

Anonymous, Hōgen Monogatari (The Tale of Hōgen), Iwanami Bunko

Anonymous, Ichidai Yōki (Essential Chronicles by Reign), Hiramatsu Bunko

Anonymous, Jōgū Shōtoku Hō’ō Teisetsu (Imperial Chronicles of Dharma King Jōgū Shōtoku),

accessed at http://www004.upp.so-net.ne.jp/dassai1/taisetsu/frame/taisetsu_frame.htm on

08/01/2018

Anonymous, Kamo Chūshin Zakki (Miscellaneous Reports on Kamo), Kamo Wake-Ikazuchi Jinja

Anonymous, Nisho Daijingū Reibun (An Illustration of the Two Great Shrines), Rekisai Shoko

Anonymous, Toyouke Daijingū Negi Honin Shidai (Order of Priestly Appointments at the Grand Shrine

of Toyouke), Iwase Bunko

318

Anonymous, Yili (Book of Etiquette and Ceremonial), accessed at https://zh.wikisource.org/zh-

hant/%E5%84%80%E7%A6%AE on 08/02/2019

Ban Zhao, Nü Jie (Lessons for Women), accessed at https://zh.wikisource.org/zh-

hant/%E5%A5%B3%E8%AA%A1 on 08/02/2019

Chen Shou, Sanguozhi (Records of the Three Kingdoms), Wei Shu, Zhonghua Book Company

Dai De, Da Dai Liji (Dai the Greater’s Book of Rites), accessed at

https://zh.wikisource.org/wiki/%E5%A4%A7%E6%88%B4%E7%A6%AE%E8%A8%98 on

08/02/2019

Fan Ye, Hou Han Shu (Book of the Later Han), Heibonsha

Fujiwara no Otsugu, Minamoto no Tokiwa, Fujiwara no Yoshino, and Fujiwara no Yoshifusa, Nihon

Kōki (Later Chronicles of Japan), Kokushi taikei Vol. 3, Yoshikawa Kōbunkan

Fujiwara no Tokihira and Fujiwara no Tadahira, Engishiki (Procedures of the Engi Era), Kokushi taikei

Vol. 26, Yoshikawa Kōbunkan

Fujiwara no Tsugutada and Sugano no Mamichi, Shoku Nihongi (Further Chronicles of Japan), Kokushi

taikei Vol. 2, Yoshikawa Kōbunkan

Cloistered Emperor Go-Shirakawa, Ryōjin Hishō (Songs to Make the Dust Dance on the Beams),

Kōdansha

Hesiod, Erga kai Hēmerai (Works and Days)

Kōen, Fusō Ryakuki (Abbreviated Chronicles of Japan), Kokushi taikei Vol. 6, Keizai Zasshisha

Lie Yukou, Liezi (The Writings of Master Lie), accessed at

https://zh.wikisource.org/wiki/%E5%88%97%E5%AD%90 on 01/09/2020

319

Ō no Yasumaro, Kojiki (Records of Ancient Matters), Kokushi taikei Vol. 7, Yoshikawa Kōbunkan

Ōtomo no Yakamochi, Man’yōshū (Collection of Ten Thousand Leaves), Nishi Honganji-bon

Ovid, Metamorphoseon libri (Book of Metamorphoses)

Sei Shōnagon, Makura no Sōshi (The Pillow Book), Yūhōdō Shoten

Prince Toneri and Ō no Yasumaro, Nihon Shoki (Chronicles of Japan), Kokushi taikei Vol. 1, Yoshikawa

Kōbunkan

Wei Zheng, Sui Shu (Book of Sui), Iwanami Shoten

Yao Silian, Liang Shu (Book of Liang), Bashu Shushe

SECONDARY SOURCES

Abrams, L. 2005. Myth and Materiality in a Woman’s World. Manchester: Manchester University

Press.

Ambros, B. R. 2015. Women in Japanese Religions. New York: New York University Press.

Anderson, M. S. 2020. “The Forgotten History of Japanese Women’s History and the Rise of Women

and Gender History in the Academy.” Journal of Women’s History 32.1:62–84.

Anesaki, M. 1930. History of Japanese Religion: With Special Reference to the Social and Moral Life of

the Nation. London: Kegan Paul International.

Anzu, M. 1986. Shintō to Nihonjin. Jinja shinpō bukkusu 1. Tokyo: Jinja Shinpōsha.

Aston, W. G., trans. 1896. Nihongi, Chronicles of Japan from the Earliest Times to A.D. 697. London:

Kegan Paul, Trench, Trübner & Co.

320

Beard, M. 2015. SPQR: A History of Ancient Rome. London: Profile Books.

Beard, M. 2017. Women & Power: A Manifesto. London: Profile Books.

Bell, A. 2011. “Re-constructing Babel: Discourse Analysis, Hermeneutics and the Interpretive Arc.”

Discourse Studies 13.5:519–68.

Bellamy, C. 2008. “Person in Place: Possession and Power at an Indian Islamic Saint Shrine.” Journal

of Feminist Studies in Religion 24.1:31–44.

Blacker, C. 1982. The Catalpa Bow: A Study of Shamanistic Practices in Japan. 2nd ed. London: Allen &

Unwin.

Bock, F. G., trans. 1970. Engi-shiki: Procedures of the Engi Era. Vol. 1, Books I–V. Tokyo: Sophia

University.

Bock, F. G., trans. 1970. Engi-shiki: Procedures of the Engi Era. Vol. 2, Books VI–X. Tokyo: Sophia

University.

Bodiford, W. M. 1993. Sōtō Zen in Medieval Japan. Honolulu: University of Hawaii Press.

Bowring, R. 2005. The Religious Traditions of Japan, 500–1600. Cambridge: Cambridge University

Press.

Breen, J. and Teeuwen, M. 2000. “Shinto Past and Present.” In Shinto in History: Ways of the Kami,

eds. J. Breen and M. Teeuwen, 1–12. Honolulu: University of Hawai’i Press.

Chaffee, J. 2001. “The Rise and Regency of Empress Liu (969–1033).” Journal of Song-Yuan Studies

31:1–25.

Chamberlain, B. H., trans. 1882. A Translation of the “Ko-ji-ki” or Records of Ancient Matters.

Yokohama: R. Meiklejohn and Co.

321

Como, M. 2009. Weaving and Binding: Immigrant Gods and Female Immortals in Ancient Japan.

Honolulu: University of Hawai’i Press.

Copeland, R. 2018. “Kirino Natsuo Meets Izanami: Angry Divas Talking Back.” In Diva Nation: Female

Icons from Japanese Cultural History, eds. L. Miller and R. Copeland, 13–33. Oakland: University of

California Press.

Corbett, J. H. 1974. “The Succession Policy of Augustus.” Latomus 33.1:87–97.

Cousins, M. and Hussain, A. 1984. Michel Foucault. Hampshire: MacMillan Education.

Derris, K. 2014. “Interpreting Buddhist Representations of Motherhood and Mothering.” Journal of

Feminist Studies in Religion 30.2:61–79.

DiLuzio, M. J. 2016. “The Vestal Virgins in Roman Politics.” In A Place at the Altar: Priestesses in

Republican Rome, 223–39. Princeton: Princeton University Press.

Dreyfus, H. L. and Rabinow, P. 1982. Michel Foucault: Beyond Structuralism and Hermeneutics.

Sussex: The Harvester Press.

Duthie, T. 2014. Man’yōshū and the Imperial Imagination in Early Japan. Leiden: Brill.

Duthie, T. 2015. “Yoshino and the Politics of Cultural Topography in Early Japan.” Monumenta

Nipponica 70.2:189–235.

Eisenstadt, S. N. 2002. “The Construction of Collective Identities and the Continual Reconstruction of

Primordiality.” In Making Sense of Collectivity: Ethnicity, Nationalism and Globalisation, eds. S.

Malešević and M. Haugaard, 33–87.

Ellwood, R. S. 1967. “The Saigū: Princess and Priestess.” History of Religions 7.1:35–60.

Ellwood, R. S. 1972. “Shinto and the Social Order.” Journal of Church and State 14.1:43–58.

322

Emura, H. 2009. Ise saigū no rekishi to bunka. Tokyo: Hanawa Shobō.

Emura, H. 2010. Ise saigū no saishi to seido. Tokyo: Hanawa Shobō.

Emura, H. 2012. Ise Jingū to kodai ōken: Jingū, saikū, tennō ga orinashita roppyakunen. Tokyo:

Chikuma Shobō.

Engels, F. 1902. The Origin of the Family, Private Property and the State. Translated by E. Untermann.

Chicago: Charles H. Kerr & Company.

Faure, B. 2003. The Power of Denial: Buddhism, Purity, and Gender. Princeton: Princeton University

Press.

Fortes, M. 1969. Kinship and the Social Order: The Legacy of Lewis Henry Morgan. Chicago: Aldine

Publishing Company.

Fukutō, S. 1987. “Josei no chii to sōzokusei.” In Nihon joseishi, eds. H. Wakita, R. Hayashi, and K.

Nagahara, 44–8. Tokyo: Yoshikawa Kōbunkan.

Gardner, H. 2009. “‘By the Facts We Add to Our Store’: Lorimer Frison, Lewis Henry Morgan and the

Spread of Kinship Studies in Australia.” Oceania 79.3:280–92.

Gardner, P. 2011. “Hermeneutics and History.” Discourse Studies 13.5:575–81.

Germanà, M. 2010. “Witches, Demon Lovers and Female Monsters.” In Scottish Women’s Gothic and

Fantastic Writing: Fiction since 1978, 60–97. Edinburgh: Edinburgh University Press.

Groner, P. 1994. “Nara jidai makki oyobi Heian jidai shoki no nisō jukai no henyō.” In Jendā no

Nihonshi. Vol. 1, Shūkyō to minzoku, karada to seiai, eds. H. Wakita and S. B. Hanley, 3–26. Tokyo:

Tokyo Daigaku Shuppankai.

323

Groner, P. 2002. “Vicissitudes in the Ordination of Japanese “Nuns” during the Eight through the

Tenth Centuries.” In Engendering Faith: Women and Buddhism in Premodern Japan, ed. B. Ruch,

65–108. Ann Arbor, MI: Centre for Japanese Studies, the University of Michigan.

Hane, M. 1988. Reflections on the Way to the Gallows: Rebel Women in Prewar Japan. Berkeley:

University of California Press.

Heldt, G., trans. 2014. The Kojiki: An Account of Ancient Matters. New York: Columbia University

Press.

Hiratsuka, R. 1911. “Genshi, josei wa taiyō de atta.” Seitō 1:37–52.

Hiratsuka, R. 2006. In the Beginning, Woman Was the Sun: The Autobiography of a Japanese Feminist.

Translated by T. Craig. New York: Columbia University Press.

Holtom, D.C. 1938. The National Faith of Japan: A Study in Modern Shintō. Tokyo: Kenkyusha.

Hori, I, 1968. Folk Religion in Japan: Continuity and Change. Chicago: University of Chicago Press.

Ijūin, Y. and Kawai, S. 2017. “Gender and the family in the archaic and classical ages.” In Routledge

Handbook of Premodern Japanese History, ed. K. F. Friday, 202–15. Abingdon and New York:

Routledge.

Kamens, E. 1990. The Buddhist Poetry of the Great Kamo Priestess: Daisaiin Senshi and Hosshin

Wakashū. Ann Arbor, MI: Centre for Japanese Studies, the University of Michigan.

Katsuura, N. 2002. “Tonsure Forms for Nuns: Classification of Nuns according to Hairstyle.”

Translated by V. Skord Waters. In Engendering Faith: Women and Buddhism in Premodern Japan,

ed. B. Ruch, 109–29. Ann Arbor, MI: Centre for Japanese Studies, the University of Michigan.

Kelly, P. and McConvell, P. 2018. “Evolving Perspectives on Aboriginal Social Organisation: From

Mutual Misrecognition to the Kinship Renaissance.” In Skin, Kin and Clan: The Dynamics of Social

324

Categories in Indigenous Australia, eds. P. McConvell, P. Kelly, and S. Lacrampe, 21–40. Canberra:

ANU Press.

Kidder, J. E. Jr. 2007. Himiko and Japan’s Elusive Chiefdom of Yamatai. Honolulu: University of

Hawai’i Press.

Kimura, S. 2016, 25 November. “Dainikai: Sono ero wa dare no mono ka.” Mōsō Koten Kyōshitsu.

http://www.webchikuma.jp/articles/-/393.

Kimura, S. 2017, 23 June. “Daikyūkai: Onna no suberu reiteki sekai.” Mōsō Koten Kyōshitsu.

http://www.webchikuma.jp/articles/-/670.

Kimura, S. 2017, 28 July. “Daijūkai: Ushinawareta monogatari wo motomete.” Mōsō Koten Kyōshitsu.

http://www.webchikuma.jp/articles/-/721.

Ko, D. 2012. “Epilogue: Textiles, Technology and Gender in China.” East Asian Science, Technology,

and Medicine 36:167–176.

Kuratsuka, A. 1996. Kodai no onna: shinwa to kenryoku no fuchi kara. Tokyo: Heibonsha.

Kyōraku, M. 1994. “Heian jidai no josei to shukke sugata: toshi to josei.” In Jendā no Nihonshi. Vol. 1,

Shūkyō to minzoku, karada to seiai, eds. H. Wakita and S. B. Hanley, 27–58. Tokyo: Tokyo Daigaku

Shuppankai.

Lebra, W. P. 1966. Okinawan Religion: Belief, Ritual, and Social Structure. Honolulu: University of

Hawaii Press.

Lowe, B. 2011. “Texts and Textures of Early Japanese Buddhism: Female Patrons, Lay Scribes, and

Buddhist Scripture in Eighth-Century Japan.” Princeton University Library Chronicle 73.1:9–36.

Maeda, H. 2008. Kodai joōsei to tennō no kigen. Osaka: Seibundō Shuppan.

325

Martin, P. 1997. The Chrysanthemum Throne: A History of the Emperors of Japan. Honolulu:

University of Hawai’i Press.

Matsumura, K. 1999. Megami no shinwagaku: shojo boshin no tanjō. Tokyo: Heibonsha.

Mazlish, B. 1998. The Uncertain Sciences. New Haven: Yale University Press.

McCullough, W. H. 1967. Harvard Journal of Asiatic Studies 27:103–67.

Meeks, L. R. 2010. Hokkeji and the Reemergence of Female Monastic Orders in Premodern Japan.

Honolulu: University of Hawai’i Press.

Meeks, L. R. 2011. “The Disappearing Medium: Reassessing the Place of the Miko in the Religious

Landscape of Premodern Japan.” History of Religions 50.3:208–260.

Miller, L. 2018. “Searching for Charisma Queen Himiko.” In Diva Nation: Female Icons from Japanese

Cultural History, eds. L. Miller and R. Copeland, 51–76. Oakland: University of California Press.

Morgan, L. H. 1871. Systems of Consanguinity and Affinity of the Human Family. Washington:

Smithsonian Institution.

Morgan, L. H. 1877. Ancient Society. Chicago: Charles H. Kerr and Company.

Munslow, A. 2000. The Routledge Companion to Historical Studies. London: Routledge.

Nagata, M. 2002. “Transitions in Attitudes toward Women in the Buddhist Canon: The Three

Obligations, the Five Obstructions, and the Eight Rules of Reverence.” Translated by P. B. Watt. In

Engendering Faith: Women and Buddhism in Premodern Japan, ed. B. Ruch, 279–96. Ann Arbor,

MI: Centre for Japanese Studies, the University of Michigan.

Naitō, T. 1976. “Himiko kō.” In Meiji shiron shū. Vol. 2. Meiji Bungaku Zenshū No. 78, ed. E.

Matsushima, 192–209. Tokyo: Chikuma Shobō. Originally published in Geibun 1.2–4 (1910).

326

Naoki, K. 2008. Yamatai-koku to Himiko: Naoki Kōjirō kodai wo kataru 2. Tokyo: Yoshikawa Kōbunkan.

Naoki, K. 2009. Ise Jingū to kodai no kamigami: Naoki Kōjirō kodai wo kataru 4. Tokyo: Yoshikawa

Kōbunkan.

Naumann, N. 2000. “The State Cult of the Nara and Early Heian Periods.” In Shinto in History: Ways

of the Kami, eds. J. Breen and M. Teeuwen, 47–67. Honolulu: University of Hawai’i Press.

Nelson, J. K. 1996. A Year in the Life of a Shinto Shrine. Seattle: University of Washington Press.

Nishino, Y. 1987. “Kaikyū no nai shakai to yakuwari buntan.” In Nihon joseishi, eds. H. Wakita, R.

Hayashi, and K. Nagahara, 1–5. Tokyo: Yoshikawa Kōbunkan.

Ōi, M. and Miyagi, E., eds. 1974. Nihon joseishi. Tokyo: Yoshikawa Kōbunkan.

Okada, S. 1970. Kodai ōken no saishi to shinwa. Tokyo: Hanawa Shobō.

Okano, H. 1993. “Women and Sexism in Shinto.” Japan Christian Review 59:27–31.

Ooms, H. 2009. Imperial Politics and Symbolics in Ancient Japan: The Tenmu Dynasty, 650–800.

Honolulu: University of Hawai’i Press.

Ōsumi, K. 2002. “Historical Notes on Women and the Japanization of Buddhism.” Translated by C.

Segawa Seigle and B. Ruch. In Engendering Faith: Women and Buddhism in Premodern Japan, ed.

B. Ruch, xxvii–xlii. Ann Arbor, MI: Centre for Japanese Studies, the University of Michigan.

Ōwa, I. 2009. Jinja to kodai minkan saishi. Tokyo: Hakusuisha.

Pandey, R. 2004. “Medieval Experience, Modern Visions: Women in Buddhism.” Review of

Engendering Faith: Women and Buddhism in Premodern Japan, ed. B. Ruch, and The Power of

Denial: Buddhism, Purity, and Gender, by B. Faure. Monumenta Nipponica 59.2:223–44.

327

Pandey, R. 2020. “Rethinking the politics of gender and agency: an encounter with the ‘otherness’ of

medieval Japan.”Japan Forum 32.4:461–83.

Philippi, D. L. 1990. Reprint. Norito. Princeton: Princeton University Press. Original edition, Tokyo:

Institute for Japanese Culture and Classics, University Kokugakuin, 1959.

Pickering, M. and Keightley, E. 2006 “The Modalities of Nostalgia.” Current Sociology 54.6: 919–41.

Piggott, J. R. 1994. “‘Himehiko’ to ‘himemiko’: jendā hairākii to kokka keisei ni kansuru kōsatsu.” In

Jendā no Nihonshi. Vol. 2, Shutai to hyōgen, shigoto to seikatsu, eds. H. Wakita and S. B. Hanley,

3–35. Tokyo: Tokyo Daigaku Shuppankai.

Piggott, J. R. 1997. The Emergence of Japanese Kingship. Stanford: Stanford University Press.

Piggott, J. R. 2003. “The Last Classical Female Sovereign: Kōken-Shōtoku Tennō.” In Women and

Confucian Cultures in Premodern China, Korea, and Japan, eds. D. Ko, J. H. Kim Haboush, and J. R.

Piggott, 47–74. Berkeley: University of California Press.

Psarras, S. K. 1993. “The Political Climate of the Later Han.” Journal of Asian History 27.1, 16–29.

Rothschild, N. H. 2015. Emperor Wu Zhao and Her Panthon of Devis, Divinities, and Dynastic Mothers.

New York: Columbia University Press.

Ruch, B. 2002. “Woman to Woman: Kumano bikuni Proselytizers in Medieval and Early Modern

Japan.” In Engendering Faith: Women and Buddhism in Premodern Japan, ed. B. Ruch, 537–80.

Ann Arbor, MI: Centre for Japanese Studies, the University of Michigan.

Ryang, S. 1998. “Love and Colonialism in Takamure Itsue’s Feminism: A Postcolonial Critique.”

Feminist Review 60:1–32.

Sakima, K. 1926. Nyonin seiji kō: jinrui genshi kihan no kenkyū. Tokyo: Oka Shoin.

328

Satō, Y. 2006. Nihon kodaishi nyūmon: Nihonjin to Nihonkoku no aidentiti wo motomete. Tokyo:

Bungeisha.

Schüssler Fiorenza, E. 2013. “Feminist Hermeneutics.” In Changing Horizons: Explorations in Feminist

Interpretation, 101–18. Minneapolis: Augsburg Fortress.

Seike, A. 1998. “Josei shuchō to gunjiken.” Machikaneyama ronsō 32:25–47.

Sekiguchi, H. 1987. “Himiko kara jotei e.” In Nihon joseishi, eds. H. Wakita, R. Hayashi, and K.

Nagahara, 15–20. Tokyo: Yoshikawa Kōbunkan.

Sekiguchi, H. 2003. “The Patriarchal Family Paradigm in Eighth-Century Japan.” In Women and

Confucian Cultures in Premodern China, Korea, and Japan, eds. D. Ko, J. H. Kim Haboush, and J. R.

Piggott, 27–46. Berkeley: University of California Press.

Sered, S. 1999. Women of the Sacred Groves: Divine Priestesses of Okinawa. New York: Oxford

University Press.

Shiratori, K. 1976. “Wa joō Himiko kō.” In Meiji shiron shū. Vol. 2. Meiji Bungaku Zenshū No. 78, ed. E.

Matsushima, 164–80. Tokyo: Chikuma Shobō. Originally published in Tōa no hikari 5.6–7 (1935).

Takamure, I. 1966. Reprint. Bokeisei no kenkyū. Takamure Itsue zenshū 1. Tokyo: Rinronsha. Original

edition, Tokyo: Kōseisha Kōseikaku, 1938.

Takamure, I. 1966. Reprint. Josei no rekishi ichi. Takamure Itsue zenshū 4. Tokyo: Rinronsha. Original

edition, Tokyo: Kōdansha, 1954.

Tanaka, T. 1996. Sei naru onna: saigū, megami, Chūjō-Hime. Kyoto: Jinbun Shoin.

Tannock, S. 1995. “Nostalgia Critique.” Cultural Studies 9.3:453–64.

Tatsumi, K. 1887. Seiyō Nihon joken enkakushi. Tokyo: Tetsugaku Shoin.

329

Tokoro, K. 2017. Saiō kenkyū no shiteki tenkai: Ise saikū to Kamo saiin no sekai. Tokyo: Bensei

Shuppan.

Tsude, H. 2011. Kodai kokka wa itsu seiritsu shita ka. Tokyo: Iwanami Shinsho.

Tsurumi, E. P. 1981. “Japan’s Early Female Emperors.” Historical Reflections 8.1:41–9.

Tsurumi, E. P. 1982. “The Male Present Versus the Female Past: Historians and Japan’s Ancient

Female Emperors.” Bulletin of Concerned Asian Scholars, 14.4:71–8.

Ueno, H. 1987. “Dogū, sekibō kara miko he.” In Nihon joseishi, eds. H. Wakita, R. Hayashi, and K.

Nagahara, 6–13. Tokyo: Yoshikawa Kōbunkan.

Umemura, K. 1987. “Ritsuryōhō to josei no chii.” In Nihon joseishi, eds. H. Wakita, R. Hayashi, and K.

Nagahara, 25–9. Tokyo: Yoshikawa Kōbunkan.

Wacker, M. 2003. “Onarigami: Holy Women in the Twentieth Century.” Japanese Journal of Religious

Studies 30.3/4:339–359.

Wada, A. 1995. Nihon kodai no girei to saishi, shinkō. Vol. 3 Tokyo: Hanawa Shobō.

Wakita, H. 1984. “Marriage and Property in Premodern Japan from the Perspective of Women’s

History.” Translated by S. Gay. The Journal of Japanese Studies 10.1:73–99.

Wakita, H. 2003. Tennō to chūsei bunka. Tokyo: Yoshikawa Kōbunkan.

Watanabe, K. 1994. “Japanese Women’s Studies.” Women’s Studies Quarterly 22.3:73–88.

Yanagita, K. 1984. Reprint. Imo no chikara. Tokyo: Kadokawa Bunko. Original edition, Tokyo:

Sōgensha, 1942.

Yang, L. 1961. “Female rulers in ancient China.” Harvard Journal of Asiatic Studies 23:47–61.

330

Yoshie, A. 1987. “Megami no botsuraku.” In Nihon joseishi, eds. H. Wakita, R. Hayashi, and K.

Nagahara, 33–6. Tokyo: Yoshikawa Kōbunkan.

Yoshie, A. 1993. “Gender in Early Classical Japan: Marriage, Leadership, and Political Status in Village

and Palace.” Translated by J. R. Goodwin. Monumenta Nipponica 60.4:437–79.

Yoshie, A. 2007. Nihon kodai josei shiron. Tokyo: Yoshikawa Kōbunkan.

Yoshie, A. 2013. “Gendered Interpretations of Female Rule: The Case of Himiko, Ruler of Yamatai.”

Adapted by H. Tonomura. Translated by A. A. Takata. U.S.-Japan Women’s Journal 44:3–23.

Yoshie, A., Ijuin, Y., and Piggott, J. R. “Gender in the Japanese Administrative Code, Part 1: Laws on

Residence Units.” Teikyō shigaku 28:317–418.

Yoshikawa, S. 2011. Asuka no miyako. Tokyo: Iwanami Shoten.

Yoshimura, T. 2010. Yamato ōken. Tokyo: Iwanami Shoten.

331

Minerva Access is the Institutional Repository of The University of Melbourne

Author/s: McKay, Natalie Louise

Title: Shaman-Queens and Sacral Princesses: A Re-Examination of the “Golden Age” Narrative of Female Sacral Power in Ancient Japan

Date: 2021

Persistent Link: http://hdl.handle.net/11343/279355

File Description: Final thesis file

Terms and Conditions: Terms and Conditions: Copyright in works deposited in Minerva Access is retained by the copyright owner. The work may not be altered without permission from the copyright owner. Readers may only download, print and save electronic copies of whole works for their own personal non-commercial use. Any use that exceeds these limits requires permission from the copyright owner. Attribution is essential when quoting or paraphrasing from these works.