Founding Territorial Cults in Early Japan

Front matter

Founding Territorial Cults in Early Japan

Traces of a Forgotten Ritual in Ancient Myths and Legends

Gaudenz Domenig

Contents

Preface 1x ListofFiguresand Tables xi

Introduction 1 The Problem of the Pre-Shinto Cults 1 Territorial Cults 3 The Focus on Early Japan 7 Japan’s Protohistory 8 Innovations Introduced by the Taika Reform 10 Different Versions of the Same Story in Nihon Shoki The God Age Mythology 13 The Fudoki Mythology 14 The Method of Interpretation 16 The Theoretical Model 18 The Structure of the Book 20 Romanisation 20 Quoting from Ancient Texts 21

1 Divination, the Crucial Rite 23 Divining with Things Thrown and Falling Down 26 Divining the Place for Founding a Shrine 29 Absurd Uses of the Falling Motif 31 Realistic Methods Exaggerated 32 Land Divination Typically Performed in Front 33 Divining with Things Cast Overboard 33 Floating a Wisteria Twig to Find the Right Place 34 Letting a Cooking Set Float to Enemy Land 36 Susanoo and the Floating Chopsticks 38 Kisakahime and the Lost Bow and Arrow 4 Articles to Play on the Sea 43 Floats Used for Divining 43 Divining in Boats 45 The Religious Use of Wood Drifted Ashore 47 Conclusion 48

Preface

This book carries the readers into an early time when it was still possible to claim and appropriate new land in the Japanese archipelago. When taking possession of ownerless land, early settlers had to deal with ‘nature’, and I argue that they did it by making a deal with the nature spirits which they assumed were inhabiting and owning the land they intended to use. It was this deal that allowed the settlers to expel the spirits from their land, but the same deal also required them to then worship these spirits in an area beyond their land and to keep this other area sacred as long as they held the land they had opened up.

The theoretical model of this ritual has originally imposed itself in a study about the spatial organisation of settlements in early Japan that was carried out with a general interest in the history of human space-ordering. The model has already been tested in a few articles, but in the present book, it is for the first time used as an analytical tool in exploring the territorial aspects of the cults to which many of the relevant sources refer. The title of this book highlights the foundation ritual because it is the key to an adequate understanding of the territorial cults, but the book also includes chapters that deal with various other characteristics of territorial cults and with their correspondences in the traditional 'grove beliefs' (mori no shinko) and in the worship of estate deities (yashikigami).

The main conclusion is that there was a time when religion was still a rather earthly affair and the dwelling area of gods was not yet imagined to be far away in the sky vertically above the earth. The ideal type of a settlement had its own local ‘heaven’, as it were, in the form of a sacred area that was often situated in an adjacent mountain side and was reserved as dwelling space of the local guardian deities. Trying to make this evident by dealing with the territorial aspect of religion in early Japan, this book aims to contribute to a better understanding of territorial cults and the basic spatial organisation of settlements founded in the way described.

Numerous scholars and friends have inspired me on the way and I express my sincere gratitude to all of them, but having written the book in retirement age and with almost no exchange with other researchers, I will only mention here the few who were most important for me during my early years in Japan. Professor Masuda Tomoya (1914-1981) of Kyoto University introduced me into various fields of Japan's traditional culture. His Saturday lectures for foreign architects were legendary, and for me, it was particularly helpful that he once also lectured on Shinto, partly basing himself on an interesting work by Tsukushi Nobuzane. Ten years later I first met Professor Harada Toshiaki

X PREFACE

(1893-1983), the noted historian and ethnographer of Japanese religions, and had the privilege to spend an unforgettable day with him visiting numerous shrines and temples in the Nara area. On this occasion and on a number of later meetings in Tokyo and at his home, he taught me his views of the territorial nature of early Japanese village cults. I admired Professor Harada as a great scholar and for his persistence in emphasising the territorial character of the early Japanese village cults against the then more popular ancestor worship theory. To know him personally has encouraged and confirmed me in my own studies. Herbert Plutschow (1939-2010), professor of Japanese Studies at UCLA, I first met in 1970 and we later often met in Japan, in Zurich and in Los Angeles, when I was lecturing there in 2001. Herbert was my best informed and most honestly interested friend when it came to speaking about ancient Japanese texts and about Japanese religions. Our discussions often lasted for hours and hours.

I also express my honest thanks to Martin Repp, professor of religious studies. He took an active interest in my first representative article on the subject of this book (1988) and, as editor of the journal Japanese Religions, later made it possible that an English translation could appear in that journal (2008). His lasting interest was one of the reasons that have prompted me to revisit the topic and deal with it in the wider context of the present book.

I am also grateful to the Swiss National Science Foundation for having funded the open access publication of this book, after having already supported a considerable part of my Japanese studies before the mid-1980s. The editors of Brill's Japanese Studies Library have contributed valuable comments and taken care of transforming the manuscript into a book. A special thank I express to Patricia Radder who patiently guided me through the editing process, making me feel good although transdisciplinary studies of this kind naturally involve additional problems.

My wife Mioko has experienced the development of the thoughts presented in this book from the very beginning. She has accompanied me on almost all field trips and often discovered interesting publications related to my research interests. She also contributed by writing several related articles in Japanese, helped with the bibliography, and read the text when it was finished. I can think of no better way to thank her than to dedicate this book to her.

Figures and Tables

15 16

17 18

The reconstructed Yayoi period settlement of Yoshinogari in northern

Kyushu 9

Theoretical model of the two-phase ritual performed when taking possession of land in ownerless regions 18

Semantic relations between Chinese characters and the Old Japanese concepts of claiming and divining 24

The area along the Akashi River where the territory mentioned in Sumiyoshi taisha jindaiki was established between the coast and two ‘god mountains’ 36 Geographical relation of Izumo to Korea and Yamato 40

Shimane Peninsula with the estuary of the ancient Hinokawa and the caves where the Great God of Sada was said to have been born 42

Map showing the valley where the story of Matachi and Yato no kami was probably meant to beset 53

Two different interpretations of the story of Yato no kami 57

A horned snake represented on a wall in a Late Kofun period burial cave 61 Straw snake at the entrance to the Tsuna no goo Shrine 61

A dragon (tatsu) made of straw hanging in the tree branches behind a

small shrine 62

A new snake rope above many old ones on the foot of a tree in the precinct of the Omoto Shrine of Kita Hiroshima-chō 62

A sacred tree with a new snake rope on top of countless old ones that are no longer distinguishable 63

A new straw snake on top of many old ones heaped up around the hidden stump of a tree whose upper part was no longer extant 64

View from the south towards the dam of the present pond seen in figure 16 67 The location of the pond Ubagaike (blue) and of three places mentioned in the story of Yatonokami 68

Map showing the location of the ancient provinces of Harima and Tajima — 73 The main places along the Ibo River where the story of Ame no Hiboko and Ashihara no Shikoo is set 74

The hill traditionally identified with the Iibo Hill mentioned in Harima fudoki seen from the south 78

The forest of the Iwa Jinja in the upper course of the Ibo River seen from the ancient Iwa Nakayama kofun 87

Situation of the Iwa Shrine in the midst of a plain surrounded by

mountains 87

XII

23 24

29 30 31 32

36 37

FIGURES AND TABLES

Map showing the geographic locations of Cape Miho, Miwa, and the beach of Inasa 93

Mt. Miwa at the eastern side of the Nara basin 99

Innovations introduced during the reigns of emperors Sujin and Suinin, according to Nihon shoki uo

The identification of the mythic Cape Kasasa with Cape Noma at the south-western end of Kyushu. Inset: Cape Noma and the valley of

Oura-chó 124

Mt. Bukkyo, the ancient kannabiyama of Izumo District, seen from the north-west 141

Map showing the ancient Mt. Kannabi (right) in relation to the Izumo Grand Shrine at Kizuki (left) 144

The Shimane Peninsula and its four parts mentioned in the land-pulling story 148

The four kannabi of four districts mentioned in [zumo fudoki 151

The inner precinct of the Izumo Taisha at Kizuki 153

Mythically important places along the Hinokawa in relation to the Shimane Peninsula and Ou in eastern Izumo 155

The open-air sanctuary in the precinct of the Matsushita Shrine discussed in chapterg 159

The Takihara no miya, a high-ranking shrine (betsugü) of the shrine complex of the Inner Shrine at Ise 159

Landforms along the Isuzu River where the Isuzu no Miya, the predecessor of the Inner Shrine (Naikü), was founded at the confluent of the two upper courses 167

View of the Isuzu when looking upstream towards the southern mountains where the Inner Shrine is situated 168

The southern mountains of the Isuzu Valley seen from north-east 171

The Inner Shrine (Naikü) within the wide catchment area that belongs to the shrine as ‘shrine precinct forest’ (kyuikirin) 175

View of the fenced inner precinct of the Inner Shrine (Naiku), showing the roof of the main hall and close behind it that of a treasure house 176

The southern part of the ancient Ise Province. The districts Iino,

Take and Watarai formed the legendary shinkoku or ‘sacred territory’ oftheIseShrine 180

Sarutahiko leading a procession at the summer festival of Kosuge 181 Tomb of the deity worshipped at the Mononobe Jinja of Oda-shi, Shimane Prefecture 196

An ancient tomb in the shrine forest of the Izumo Daijingü of

Kameoka-shi 197

FIGURES AND TABLES XIII

43 The upper part of a wooden torana used in a festival at Sanchi (India) 205

44 A torii at the coast of Naoshima in the Seto Inland Sea. Divining is done by throwing a stone and trying to get it to rest on top of the torii. 206

45 The entrance to a sacred grove dedicated to a frog deity 207

46 The Yamato basin (Nara basin) and its four main entrances mentioned in the two chronicles. (A) Sumisaka (east); (B) Osaka (west); (C) Nara (north); (D) Ki or Kii (south) 2u

47 Iwarebiko’s detour in the story of his conquest of Yamato 212

48 The chigi cross on the roof of a shrine in Izumo Taisha style 214

49 Amarushime during construction 215

50 Sacred grove of a yashikigami in Kagoshima Prefecture 224

51 The straw shrine of the deity that was worshipped in this grove 225

52 Field sketch of the plan of this grove 225

53 Entrance to the sacred grove of Hoki shown in figure 54 228

54 The sacred grove of Hoki seen from the east 229

55 The gard yama of Hirayama Mukai 233

56 Inside view of the sacred grove shown in figure 55 233

57 The sacred grove called o-ta no mori belonging to the sacred rice fields of the Homan Jinja in Tanegashima 234

58 Sketch of Miwayama seen from the west, indicating the forbidden zone behind the straight fence and the torii of the Omiwa Jinja 236

59 The Matsushita Shrine in its protected shrine forest 240

60 Plan of the Matsushita Jinja and plan of its public part (right) 241

6ı The main hall of the Matsushita Shrine in its wooden fence 242

62 The cult place of Somin Shorai in November1973 242

63 The cult place of Somin Shoraiin March 1994 243

64 Changing positions of the sakakimaki dedicated to Somin Shorai between 1994 and 2015 243

65 A granary from the protohistoric Kofun period reconstructed in the Miyoshi Fudoki-no-oka Park 244

66 Illustration from Orikuchi 1915-1916 (adapted, with words written in romanletters) 247

67 A structure called hashiramatsu (‘pillar-pine’), two samples of which were used in a divining ritual 247

68 Two marushime set up in the back of the Ukehi Shrine on the third day of the New Year. Compare figure 49, which shows another type of marushime. 248

69 A type of ohake (here called hakke) in the form of a conical hut (left) with a higher central bamboo pole that carried a white gohei at its top (right) 249

70 Temporary altar on the dam of a small irrigation pond — 250

XIV FIGURES AND TABLES

71 Decorative torches (kazari taimatsu) used at the fire festival of the Shinoda Shrine 251

72 A shimenawa with paper shide and tassels hung around the foot of the cult tree in a sacred grove that was dedicated to a snake deity 252

73 Offering-sticks (gohei) set up in front of a shimenawa forbidding access to a straw snake hanging in the lowest branches of a tree 252

74 Atobiyashiro in the precinct of the Chinjü Jinja of Funagi, together with two stones marked as sacred and two stone shrines (hokora) 255

75 Tobiyashiro from five different places of Hirado Island, Kyushu 255

76 AShinto priest worshipping in front of a modern himorogi at a jichinsai celebrated on a building site 260

77 The same himorogi, consisting of a sakaki tree tied to a wooden post and left standing on the site until construction time 260

78 Boundary sign made with high-growing kaya grass, prohibiting entry into a forest where mushrooms were growing 265

79 Boundary sign made of the leafy tops of two young trees joined together, bent down and tied with straw 265

80 Sign made of brushwood and two downwards-bent tree branches tied with straw to the stem ofa tree 266

81 Prohibition sign made of a bunch of grass that was folded and tied to a stick with the top leaves hanging down 266

82 Map of Thorolf’s territory at Thorsnes, according to Eyrbyggia Saga 274

83 Schematic view of an older and a newer concept of 'heaven' in the myth of the river Sarasvati in Vedictexts 284

Tables

Table1 Two versions of the land-ceding myth compared with the story of Yato nokami 107

Table 2 Descent myths of Korea, based on text in Waida1973 n5

Table 3 New land names and titles of their governors or chiefs (left column) whose answers to Yamatohime's question “what is the name of your land?” reveals the old name of the land (right column) 162

Introduction

The remarkable progress of Shinto studies over the last decades has led to a situation where many scholars no longer believe that it makes sense to speak of Shinto when meaning the religious beliefs and practices of people living on the Japanese archipelago before the seventh or sixth century. The more this becomes clear, the more we may feel that the traditional understanding of earlier religious cults in the Japanese archipelago might be inadequate so long as we only orient ourselves on the archaeological record and on what appears to have survived of such cults in combination with parts of later Shinto.

The present book argues that territorial cults must have been common in pre-Taika times and tries to understand them on the basis of a theoretical model of the ritual procedure applied in taking possession of land in ownerless regions. The model is a result of comparative studies carried out earlier, but itis used here for the first time as an analytical tool in a study of numerous ancient myths and legends of Japan. In chapter 10, it is also applied to a few relevant sources from other cultures.

The Problem of the Pre-Shinto Cults

Dealing with the pre-Shinto cults of early Japan is usually thought to be almost impossible because of the lack of relevant written sources (Grapard 2002: 228-29; Breen and Teeuwen 2010: 27). Mori Mizue, for example, thinks that the late date of the written sources "renders it very difficult to reconstruct concrete instances of kami cults before the formation of the state cult" As she points out, only the archaeological record is reliable as a source of information on early cults. She refers to archaeological sites where "fetishes carved from wood, precious objects such as jewels, and containers for food" have been found in combination with large rocks, and she notes that such sites “are generally understood to have been places of kami worship". However, the source value of material objects for questions of religion being limited, Mori cannot help but interpret the archaeological record in terms of ideas based on the textual sources and religious traditions of later times. Thus, she states, for example, that the kami were imagined as invisible spirits and had no permanent dwelling places - ideas that cannot be deduced from material objects such as those known from archaeological excavations (Mori 2003: 15).

This is an open access chapter distributed under the terms of the CC BY-NC-ND 4.0 license.

Similar ideas were already held by Oba Iwao (1970), the pioneer of Shinto archaeology, and by other scholars who assumed that Shinto somehow developed from animistic roots. Matsumae Takeshi (1993: 325), for example, suggested that in the prehistoric Jomon and Yayoi periods "the roots of Shinto developed from animistic forms of nature worship" whereas in the next period (4th and 5th centuries) “animistic spirits were personalized, honoured at special places of worship, and frequently adopted as tutelary deities of local clans".

It is of course correct to stress the lack of written sources from the protohistoric period and the failure of the archaeological record to inform us well about beliefs, but this does not force us to sketch early beliefs by following more or less common patterns of speculation about the nature of early religious practices and experiences. In addition to focusing on historical sources and archaeological evidence, we can also study the rich material in the form of myths and legends that were written down in Japanese texts of the early eighth century, notably in Kojiki and Nihon shoki and in the various fudoki reports from the provinces. Rather than attempting to extract historical truths from such sources or fill in the gaps with tentative sketches of early cults and beliefs, we can focus research on matters that can be learned from these texts.

Myths and legends usually lack historicity, but by comparing ancient Japanese stories we can observe in them recurring themes, motifs and story patterns that open up perspectives on ritual behaviours that are likely to have been characteristic of numerous local cults in early Japan. Mark Teeuwen is probably right when he notes that during the Kofun period (c.250-700 CE) there was so much foreign influence and regional diversity that it is unthinkable that people in the archipelago “could have shared a single religion at any time during this period”. If he adds that “any attempt to pin down ‘what the ancient Japanese believed' is doomed to fail" (2012: 68), he may think of complex belief systems; but we can expect that territorial cults were widely distributed although other beliefs and ritual practices may have been specific and restricted to particular groups of people. It is precisely the fact that the protohistoric centuries of Japan witnessed numerous immigrations and movements of people within the archipelago that calls for studies trying to understand the way in which the local cults of pre-Taika times used to be founded.

Traditionally, Shinto was and is often still thought to have deep roots in Japanese protohistory and even prehistory, but Kan'ichi Asakawa (1873-1948) wrote more than a hundred years ago in his Early Institutional Life of Japan that "Shinto is, even in its earliest records, found to be a composite body of beliefs which in general fall in two classes, namely, the lower religious ideas common with other primitive races of like stage of culture and the higher ones peculiar to the Japanese people" (Asakawa 1903: 31). According to Asakawa, the higher religious ideas peculiar to the Japanese are those which centre on Amaterasu as the ancestral deity of the imperial family, on the myth of the conquest of Japan by the heavenly ancestors and on the emperor as the high priest of the nation whose political sovereignty rests on religious sanction (1903: 32-33). Asakawa believed that these specifically Japanese aspects of Shinto predate the advent of Buddhism in the middle of the sixth century. This can no longer be assumed today, but whereas the composite body of Shinto beliefs also includes Buddhist, Taoist and Confucian ideas and characterises Shinto through recorded history, some of the not specifically Japanese beliefs and practices of Shinto can indeed be assumed to have deeper roots in the proto- and prehistory of Japan.

By no longer classifying the early, prehistoric and protohistoric cults of Japan as forms of Shinto, the problem of understanding these cults presents itself in a new light. It now seems more likely that these earlier cults had characteristics that make them comparable to early cults and indigenous religions in other parts of Asia and the world.

Ito Mikiharu (1998), for example, has proposed that we focus on certain aspects of indigenous religions in Southeast Asia, using this perspective as an analytical framework for dealing with the concept of kami in ancient Japan. More recently, Mark Teeuwen (2007b) looked at a problem of Japanese religious history in a comparative perspective on Southeast Asia. Interested in identifying cases in other Asian cultures where interactions of certain earlier features of religion with Buddhism led to the emergence of something comparable to Shinto, he notes that certain features of Shinto — “worship of deities associated with water, trees and mountains in sacred groves, worship of snakes as messengers or manifestations of deities, myths about the descent of heavenly deities to earth, and so forth” — are also found in other religions of Asia and the world and “that it makes little sense to label their occurrence in Japan as ‘Shinto’, as though they share a common origin in a specific Japanese tradition" (2007b: 375).

Territorial Cults

In light of such new developments, this book looks at Japanese myths and legends related to the founding of territories in pre-Shinto times and seeks to understand how settlers ritually proceeded when settling down in a new place and what characterised the primary spatial organisation of their territories as well as the corresponding ‘territorial cults’.

The study of territorial cults has a long history, but the use of the English term as an analytical concept seems to have come up only in the 1970s in works about African religions.! Japanese scholars rather used compounds like chien saishi ("land-related rituals"), chükiteki saishi (‘regional rituals’) or chüki shükyo (‘region cults’). The importance of territorial cults in early Japan has been recognised by Tsuda Sokichi (1964), Aruga Kizaemon (1954) and others, but it was Harada Toshiaki (1893-1983) who most consistently emphasised this point.? As early as 1941 he wrote in an article that appeared in English translation:

The fact that the Japanese had to depend from ancient times on husbandry as the means of their livelihood has given a specific character to the Japanese community since old. It may, therefore, be safely said that farming went a long way, even in the case of clans and families, in the formation of territorial units of people. Accordingly, the Ujigami or village-tutelary-deities, worshipped by such units of people, came to acquire and manifest a territorial character, regardless of whether or not the said groups of people were united by ties of blood. In other words, it is the locality which characterizes the Ujigami, and this fact marks it off from other deities of this country. (Harada 1941: 2)

Harada's insistence that the village deity (ujigami) was originally a deity guarding the village as a territorial unit was partly provoked by the contrary theory of Yanagita Kunio, who, based on ethnographic research, argued that the village deity was originally a deified ancestor and the rituals were therefore blood-related (ketsuenteki saishi). This controversy regarded only the ‘original’ beliefs and rituals in proto and prehistoric times. Like many Japanese scholars today, Harada thought that the original concept of kami (spirit, deity) referred to no more than a mysterious and vague presence (Harada 1970: 100) and that the worship of human persons came up only after the Taika Reform

1 Schoffeleer 1972, cited in Van Binsbergen 1979: 47. See also Ranger 1973. French scholars used to deal with such cults by referring to a famous study by Edouard Chavannes (1910), which deals with the 'god of the soil' in ancient China. Recently, however, French researchers dealing with populations living on the margins of state power in parts of Asia also prefer the term territoire (Schlemmer 2012).

2 Harada is known for his 'thesis of the essentially territorial nature of religious ceremonies in Japan’ (Nihon shükyo no chiensaishi honshitsuron). For a discussion of his work, see for instance Yoshitani 1986.

(645 CE). He argued that ujigami was originally the name of the clan deity as a territorial village guardian deity that “was worshipped there alone" and “did not have much individuality to be distinguished from others by means of a proper name”. Whereas “most of the present shrines [...] are dedicated to deities referred to in the myths" most of the shrines registered in Engishiki (ninth century) “are named after places and a few after natural phenomena, and only a very small number of them located in districts near the capital, Izumo and a few other provinces, have names" (Harada 1959: 216).?

Harada dealt with the territorial nature of the village deity, but to my knowledge he did not also focus research on the question of how territories were founded. The territorial founding rites have not yet become a subject of intensive research: not generally and not in Japan. Although countless Japanese scholars have commented on the key story discussed in chapter 2 and on other stories discussed in this book, I do not know of a comprehensive study of the available sources that would propose a theoretical model of the typical procedure followed in taking possession of land in early Japan. In an article dealing with senyu girei (occupation rites) based on fudoki stories, Komatsu Kazuhiko (1977) comes to the conclusion that in the myths, the claiming of land was done by gods who placed their staff at the place where a sanctuary was to be founded, in which the claiming deity was then worshipped (1977: 125). Komatsu mentions several earlier scholars who have dealt with the subject in the context of general works on Japanese mythology, ancient politics or culture (1977: 138m), but he does not mention an established theory regarding this subject. The same can still be said of a recent article by Sakae Wataru (2013), which deals with land-claiming myths from Harima fudoki. Considering how little the subject has been studied in other parts of the world, this is not surprising, although the ancient texts of Japan are particularly rich in stories about the making of territories.

We only see what we already know and understand: Goethe's phrase written in a letter to a friend two hundred years ago is a truth that is sometimes also responsible for the neglect of a research topic. Knowing about the function of oracles and other forms of divining in the founding of ancient Greek colonies, for example, it has been easy to see that the usual interpretations of the ancient Japanese sources about land-claiming might miss something of importance as long as they ignore the role which divination played in the making of territories. Here lies the heuristic value of cross-cultural comparative research; for sometimes we indeed see something only if we already know

3 Referring to early village cults, Harada would sometimes even speak of a sort of local ‘monotheism' (personal communication).

it from somewhere else. Divination has become a major subject in this book, because knowing what it implied in the case of opening up land for agriculture and settlement means to understand that it was a necessary rite that was intended to stipulate a contract with the local spirit-owners of the land. This contract was a typical feature of early territorial cults, whereas the later Shinto cults are usually not said to be based on a contract with the deity worshipped. Divination is therefore dealt with at length in chapter 1, but the subject pops up again and again in other chapters as well.

The cross-cultural comparative aspect is not central in this book, but it is important enough that chapter 10 is devoted to it. Although the ancient texts of Japan first inspired me to study rites performed for taking possession of land, I would not have spent so many years on them if my interest had not also been a more general one. As a researcher in the anthropology of space with interests in protohistory, prehistory and human evolution, I thought from the beginning that different cultures may have used the same basic procedure for taking possession of land, although they enacted it in many different, culturally specific ways.