From Deer Bones to Turtle Shells: The State Ritualization of Pyro-Plastromancy during the Nara-Heian Transition

Abstract

This article reviews received and recovered evidence of divination with bone and fire in early Japan to identify and investigate a shift from deer scapulae to turtle shells that took place during the Nara-Heian transition, particularly within the state cult. It questions why this shift occurred and analyzes a detailed explanation of it found in a purportedly early Heian treatise on the divinatory cracking of turtle plastrons known as the Shinsen kisoki (Newly compiled record of turtle omens). The Shinsen kisoki claims to have been authored by a group of men descended from a common genealogical line of ancestral kami associated with divination. It not only reveals much about why members of a handful of related clans would have promoted a change from scapulimancy to plastromancy at this point in history, but also much about how the state ritualization of the latter affected, and was affected by, other changes in state and local religion and politics during the late Nara and early Heian periods.

Headnote

This article reviews received and recovered evidence of divination with bone and fire in early Japan to identify and investigate a shift from deer scapulae to turtle shells that took place during the Nara-Heian transition, particularly within the state cult. It questions why this shift occurred and analyzes a detailed explanation of it found in a purportedly early Heian treatise on the divinatory cracking of turtle plastrons known as the Shinsen kisoki (Newly compiled record of turtle omens). The Shinsen kisoki claims to have been authored by a group of men descended from a common genealogical line of ancestral kami associated with divination. It not only reveals much about why members of a handful of related clans would have promoted a change from scapulimancy to plastromancy at this point in history, but also much about how the state ritualization of the latter affected, and was affected by, other changes in state and local religion and politics during the late Nara and early Heian periods.

keywords: deer-bone pyro-scapulimancy-pyro-plastromancy-Shinsen kisoki- Urabe and Nakatomi clans-state and local kami cults-Tsushima and Iki

(ProQuest: ... denotes non-US-ASCII text omitted.)

The earliest known description of divination performed on the Japanese archipelago comes from the Waren zhuan ??? (Traditions of the people of Wa) of the late third century ce Sanguo zhi (Chronicles of the three kingdoms). This relatively short compilation of traditions or transmissions describes the lands and customs of a foreign people named Wa thought to inhabit parts of the southern Japanese archipelago during the third century ce. It states that the Wa customarily performed divination with fire and bones:

...

They scorch bones for cracks, using them to prognosticate auspiciousness and inauspiciousness. The initial declaration of what is to be cracked resembles words used in "charging the turtle" methods, [methods] which involve the careful observation of fire-produced fissures to prognosticate [the auspiciousness and inauspiciousness of ] crack omens.1 (Sanguo zhi 3: 30, 856)

Because the divinatory practices of the people of Wa are said to involve "bones" and resemble Chinese "charging the turtle" methods, there is good reason to argue that the passage is referring to something other than divination with turtle shells. Received and recovered evidence can be used to defend and expand on this point.

Archaeologist Kanzawa Yuichi's (1987, 6) oft-cited work on early Japanese bone divination lists 158 divined bones and shells from thirty-four different sites. The list includes seventy-five deer bones, twenty-one wild boar bones, forty-six dolphin bones, and four loggerhead turtle (akawamigame ??????) plastrons ranging in date from roughly 100 bce to 800 ce. Ninety-eight examples are dated to the Yayoi (ca. 300 bce-250 ce) period, seventeen to the Kofun (ca. 250-538), and forty-seven to the Nara (710-784) period. Only about half of the total are deer scapulae, but they are found at a majority of the listed sites (26/34), indicating widespread use of deer bones for divination before the Nara period. Kanzawa's list includes only four examples of turtle shells: three late Kofun plastrons from the Maguchi Cavern ??? ruins on the Miura Peninsula in Kanagawa, and one late Kofun to early Nara plastron from the Shitaru ??? ruins in northwest Tsushima.

A more recent list of archaeologically recovered divined turtle shells compiled by Saso (2006, 105) adds twenty-seven late Kofun to early Nara examples from six different sites. A vast majority of these shells postdate 600 ce and come from the present-day prefectures of Kanagawa, Chiba, and Nagasaki (including Tsushima and Iki). According to the temporal distribution of the shells in Saso's survey, pyro-plastromancy came to be practiced just as frequently as pyroscapulimancy by the early eighth century of the Common Era.

The archaeological record, as it stands today, indicates that the most common pyromantic media in Japan were deer bones until, or shortly before, the Nara period, when turtle shells start to appear just as frequently. Received Japanese court-sponsored texts, however, suggest a late Nara or early Heian shift from bones to shells. Deer-bone pyro-scapulimancy (rokuboku ??) is portrayed as the standard technique for royal and divine divinatory consultations in the early Nara Kojiki (Record of ancient matters) and Nihon shoki (Chronicles of Japan), and in other early mytho-histories like the Kujiki ??? (Chronicle of ancient things). Pyro-plastromancy (kiboku ??; turtle-shell cracking [with fire]) is never mentioned in these works, but it is clearly the dominant form of divinatory cracking in early Heian court-sponsored historical and bureaucratic texts such as:

1. Shoku Nihongi (Chronicles of Japan continued) [ca. 797 ce]

2. Ryo no gige (Explanations of the codes and ceremonies) [ca. 833]

3. Nihon koki (Later chronicle of Japan) [ca. 840]

4. Shoku Nihon koki (Later chronicle of Japan continued) [ca. 869]

5. Nihon Montoku tenno jitsuroku (The veritable records of Montoku of Japan) [ca. 879]

6. Nihon sandai jitsuroku (The veritable records of the three reigns in Japan) [ca. 901]

7. Engi shiki (Procedures of the Engi era [901-923]) [ca. 927]

All of these late eighth to early tenth century works portray the divinatory cracking of turtle plastrons as a state rite managed by certain clans in the state department or bureau of religious affairs, the Jingikan ??? (Office of the spirits of heaven and earth). Clan associations are seldom stated, but all seven texts contain references to pyro-plastromancy,2 and the divinatory cracking of deer bones never seems to be mentioned in any of them.

The latest Nara text in the list above, the Nihon shoki, mentions deer-bone pyro-scapulimancy in the first two fascicles of the text. The former (1, 30) cites a long passage from an anonymous source introducing the undesirable offspring produced by Izanagi ??? and Izanami ??? after they had created much of the Japanese archipelago. This kami couple ascends to Heaven to inquire about their defective progeny. The heavenly spirits deliberate before divining the matter with "the grand [pyro-scapulimantic] prognostication" (futomani ??). They determine that the problem was caused by the couple's improprieties, so they urge the couple to right the wrong by performing proper marriage rites on earth. They then divine an auspicious date-presumably using "the grand prognostication" once again-for Izanagi and Izanami to return to the earthly realm. In the second fascicle (2, 136), heavenly kami Takamimusubi ???? dispatches fellow spirits to help pacify the earthly spirits of the Central Land of Reed Plains ????. Some of these heavenly spirits are ordered to help protect his great-grandson, who is to rule. Others, like Futodama ?? and Ame no koyane ???, are asked to conduct divinatory and sacrificial rites on behalf of the new ruler's descendants. Ame no koyane is put in charge of "the customary sources for spiritual affairs" (shingi no sogenja ??????), including "divination procedures pertaining to the grand prognostication" (futomani no bokuji/shinji ?????). Pyro-scapulimancy is the go-to means of divination for the heavenly spirits in the Nihon shoki, just as it is in the earlier Kojiki and, depending on how it is dated, the Kujiki.3 Pyro-plastromancy is never mentioned in these mytho-histories, but according to Heian court-sponsored documents, it was the exclusive method of divinatory cracking in the Jingikan.

The divinatory cracking of turtle plastrons is mentioned a number of times in the first Heian text listed above, the Shoku Nihongi. The earliest and latest dated entries containing descriptions of the technique appear in the compound "turtle and yarrow" (kizei), a synecdoche for pyro-plastromancy and achilleomancy.4 The earliest entry recording this compound is dated 708 ce. It summarizes part of an imperial decree declaring that the time to relocate to a new capital in Heijo-kyo (Nara) had arrived. Among the many auspicious signs introduced in the decree, both "turtle and yarrow" are said to have approved the move, which was carried out in 710.5 The latest entry recording the compound claims that a memorial was submitted to the throne in late 782 requesting that "turtle and yarrow" be consulted to inquire about a rash of calamities suffered by the nation (Shoku Nihongi 5: 37, 244). While "turtle and yarrow" may have been consulted in 710 and 782, the rare and situational nature of these consultations prevents us from claiming that either technique had become a court convention.

The ca. 833 ce Ryo no gige is one of the earliest works to overtly present plastromancy as a state standard. The divinatory cracking of turtle plastrons is depicted as a duty of the urabe?? (divination guild or department) in the Jingikan, and appears to have completely eclipsed the art of scapulimancy. A comment on the term bokucho?? (crack omen) in the Ryo no gige reads:

...

The character refers to the scorching of a turtle shell. The character ? refers to horizontal and vertical patterns caused by the scorching of a turtle shell. Generally, the scorching of turtle shells for the divinatory prognostication of auspiciousness and inauspiciousness is an enterprise controlled by the urabe.

( Ryo no gige 1: 29)6

The need for additional explanation in both cases implies that plastromancy was not yet a fully recognized court convention, yet it seems to have already replaced scapulimancy as the preferred method of state divinatory cracking.

The Ryo no gige clearly states that pyro-plastromancy was administered by the urabe. The term urabe already appears in the early eighth century Ritsuryo (regulatory and administrative codes) as a reference to state diviners, but their exact responsibilities are never described. It is only with early ninth century texts like the Ryo no gige-and the perhaps earlier Shinsen kisoki-that the pyroplastromantic duties of Jingikan urabe are made explicit.7 Although received courtsponsored texts point to a mid-to-late eighth century shift from deer bones to turtle shells, neither archive explains why this occurred. We could simply write off the matter as one of many aspects of Tang (618-907) Chinese culture adopted by the early Heian court, but in doing so, we would be missing a good opportunity to clarify something about how specific appropriations were carried out, legitimized, and maintained. The Shinsen kisoki is particularly helpful in this regard, as it offers a clear description-and defense-of the Nara-Heian incorporation of plastromancy into the religious branch of the theocracy. The text reveals much about why a certain group of men acted to privilege and elevate one form of divination over another, and much about how historical circumstances and cultural strategies were used to achieve the state incorporation and ritualization of pyro-plastromancy.8

The History and Content of the Shinsen kisoki

In the late summer of 830 ce, a team of writers and editors headed by Urabe Totsugu... (fl. 830 ce) purportedly presented a treatise on pyro-plastromancy titled Shinsen kisoki to the throne.9 The earliest extant edition is a 1620 ce manuscript copied by Bonshun ?? (1553-1632),10 which was subsequently transmitted within a larger collection known as the Kiboku sho ??? (Notes on turtle divination).11 Bonshun's copy records only part of the first fascicle of what is said to have originally been a four-fascicle text, and is presently held in the archives of the University of Tokyo's Department of Religious Studies. All published editions of the Shinsen kisoki are based on this text.12

The Shinsen kisoki, as mythologist and historian Tsubaki Minoru (1957, 41) contends, is an extremely valuable text for three major reasons: it is the earliest known treatise on pyro-plastromancy ever produced in Japan, it is one of the earliest texts recording fragments of Japanese norito ?? (words for reciting spells; incantations),13 and it is one of the earliest texts to draw from multiple sections (the preface, beginning, middle, and end) of the Kojiki.14 I would add that the Shinsen kisoki also provides a valuable record of the mytho-histories and corresponding genealogies recorded-and at least partially constructed-by certain members of the Nakatomi ??, Urabe??, and Atai ? clans to help legitimize distinct but related ancestral lines extending back to kami named in the early eighth century state mytho-histories. These kami are associated with incantation, divination, and ritual purification; skills that defined the sacerdotal duties of members of the Chinese-inspired Jingikan during both the Nara and Heian periods.15 What remains of the Shinsen kisoki distinctly shows that members of the Nakatomi and Urabe clans relied on kami myths and Chinese precedents to redefine their duties and roles in the early Heian theocracy.

But is the Shinsen kisoki an early Heian text? Most twentieth century Japanese scholarship on the work argues or assumes it is. Although text-critical research on the dating of the text is rare, some of this work is introduced in the following discussion. A few preliminary and general defenses of the early-Heian dating of the text include the fact that it claims to be an early ninth century work. All Chinese and Japanese works cited in the text pre-date 830 CE, and the strongest motives for its composition are bound to early Heian geopolitical history, particularly to the intense competition for political and religious power that took place between powerful clans during the Nara-Heian transition.

Tsubaki (1957), Akimoto (1978, 111), and Kudo (2005, 6-7) all argue that, as a collection of statements regarding the divine descent and sacerdotal duties of the Urabe and Nakamoti clans, the Shinsen kisoki closely resembles Inbe Hironari's ???? (ca. 803-807) early ninth century Kogoshui (Collected traces of ancient tales).16 A collection of myths and legends set before the reign of the legendary Emperor Jinmu ???? (great-grandson of Amaterasu), the Kogoshui was reportedly offered to the throne in 807 ce. It borrows from the Kojiki and Nihon shoki, but rearranges and supplements accounts in those works with materials passed down hereditarily (and orally) through the Inbe ?? clan. As a record of clan achievements that selectively draws from and attempts to fill in blanks in the early Nara myths of state to establish divine descent and mastery of theocratic rites, the Kogoshui seems to have provided a clear and nearly contemporary precedent for the Shinsen kisoki. It might have even inspired the Shinsen kisoki as a response to its list of complaints about the Inbe's loss of traditional ritual responsibilities to other clans.17 Both texts extend and rewrite orthodox myths of state to establish links between state kami and the ancestral kami of the authors of each text. Both also represent responses to perceived threats against the sacerdotal authority of the kinship groups responsible for their compositions.

Could the Shinsen kisoki have been composed as late as the early seventeenth century? It is possible, but it would have been a brilliant fabrication. Not only would the author/s have refrained from completing the text, but they would have also created a work that perfectly fit into a much earlier historical context. Finally, we would expect a combination of mythical persistence and technical innovation in an early seventeenth century copy of an early ninth century text on divination, and that is exactly what we find in Bonshun's copy of the Shinsen kisoki. These points are further defended below.

The received edition of the Shinsen kisoki was transmitted within a larger collection of works known as the Kiboku sho (Notes on turtle divination; also the title of a separate work in the collection). The Kiboku sho dates no later than the early seventeenth entury. Bonshun's copy includes the following six titled works, three of which were likely initially composed before the end of the tenth century ce (#3, #4, and #6).

1. Kiboku shidai (Rules for turtle divination) <no line numbers>

2. Kiboku sho <lines 1-361>

3. Shinsen kisoki <lines 362-605> [ca. 830 ce]

4. Kiso bubun (Section on turtle omens) <lines 606-648> [ca. 973 ce]18

5. Mondo bubun (Question and answer section) <lines 649-819>

6. Saigi kankei kiji (Records related to sacrifices and ceremonies) <lines 820- 871> [ca. 806 ce]

The Shinsen kisoki records thirty section titles for fascicle 1 and titles for the now-lost second through fourth fascicles of the work. Fascicles 2-4 are titled:

1. <untitled>

2. "On the Identification and Interpretation of Earth [Graphs]" (Setsu chi no shoko ...)

3. "On the Identification and Interpretation of Heaven [Graphs]" (Setsu ten no shoko ...)

4. "On the Interpretation and Identification of the Three Line Forms: Spirit, Human, and Omen [Graphs]" (Setsu shijincho sanka shoko ????????)

Based on these titles, we can assume that fascicles 2-4 featured omens and oracles based on five basic types of pyro-plastromantic cracks or line-graphs.19 The content of fascicle 1, however, differs, as it provides a much more general introduction to the text as a whole.

Each of the following numbered entries translates, transcribes, and records one of thirty section titles from the Shinsen kisoki table of contents for fascicle 1. Corresponding content in the body of the Shinsen kisoki-or in other works found in the Kiboku sho collection-is cited in brackets at the end of each entry.

1. "The Two Kami Izanagi and Izanami Give Birth to Onogoro Island" (Izanagi Izanami ryojin sho Onogoro ...) [Shinsen kisoki lines 408-416 (hereafter cited as Ssksk 408-416]

2. "The Two Kami Give Birth to the Nation and Establish Marriage Ceremonies and Fire Pacification Rites" (Ryojin sho kokudo cho fugi kachin sai ???????????) [Ssksk 416-429]

3. "The Fiery Well and the Metamorphosis of the Trinity of Kami" (Kasei sanjin shokei ??????) [Ssksk 429-444]

4. "Izanagi Orders Each of the Trinity of Kami to Rule Sun, Moon, and State, but Sentences Susanoo to be Exiled, and More" (Izanagi no mikoto sanjin haicho hitsugi kokushu shika harai Susa no mikoto to ?????????? ????????) [Ssksk 444-485]

5. "The Eight Million Kami Fine Susanoo a Thousand 'Tables' [of Reparations], Exile Him, and More" (Yaoyorozu no kami Susa no mikoto chikura no okido harae to ?????????????) [Ssksk 475-485]

6. "The Heavenly Kami Send Down and Install a Ruler of the Nation" (Tenjin kokyu kokushu ??????) [Ssksk 485-498]

7. "The Nakatomi and Inbe Clans Take Control of Crack Omens and the Offering of Silk, and More" (Nakatomi Inbe ryoshi sho bokucho hanpei to ????????????) [Ssksk 501-503]

8. "The Heavenly Descendant Descends to Rule the Chiho Peaks in Himuka" (Tenson kozei Himuka Chiho shin ???????[?]?) [Ssksk 503-505]

9. "Emperor Izahowake sumera and His Illustrious Younger Brother Mizuhawake no mikoto Kill Sobakari After the Divine Deed Is First Carried Out During His Reign" (Izahowake sumera mikoto Miyo kotei sai Sobakari wa kamigoto saki kaijo ??????? ??????????????? ????) [Ssksk 506-518]

10. "Reigning Sovereign Emperor Onagatani Prohibits the Placement of 'Fish Logs' on Human Abodes and Standardizes Gift Offerings and Restitution Paid with Rolls of Silk" (Onagatani tenno Miyo kin seido jinkyo okujyo katsuogi horai kaenusa ??????????????? ??????) [Ssksk 518-520]

11. "The Invasion of Silla Ordered by Emperor Tarashina hiko's Great Empress Okinaga tarashi" (Tarashina hiko tenno no okisaki Okinaga tarashi hime no mikoto shu Shinra ??????????????????) [Ssksk 520- 527]

12. "Short Explanations from the Classic of Turtles and the Major Purports of Pyro-plastromancy" (Ryakujutsu Kikyo oji kame taii ????????) [Ssksk 527-533]

13. "Tabooed Days According to the Branches of the Four Seasons and FiveColored Turtles from the Classic of Turtles" (Dokyo shiji shiyo goshiki kame kijitsu ???????????) [Ssksk 534-536]

14. "An Explanation of the Pledge of the Turtle" (Jutsu Kisei ???) [Ssksk 536-567]

15. "The Principal Shrine of Turtles, the Hahaka Wood Shrine, and the Use of Water in [Divinatory] Scorching and Cracking" (Kame honsha Hahakaki no jinja shakuki yosui ????????????) [Ssksk 567-572]

16. "The Supreme Ancestors of the Urabe of the Four Provinces Oversee and Formally Offer Up Pyro-plastromantic Omens" (Shikoku Urabe kamiso shiho bokucho ??????????) [Ssksk 573-578]

17. "The Urabe Clan of the Four Provinces" (Yokuni Urabe uji ?????) [Ssksk 581-584]

18. "Tsushima Island Comes to Be Comprised of Two Provinces" (Tsushima shima sho ryokoku ...) [Ssksk 584-587]

19. "Cases of Oral Transmission from the Record of Ancient Matters" (An Kojiki yo kuden ???????) [Ssksk 587-598]

20. "The Original Recognition of Superior Supervisors of Divinatory Cracking" (Shinin boku no chojo ?????) [Ssksk 598-602]

21. "Purification Ordinances for Divinatory Cracking" (I boku seikai ????) [Kiboku sho line 606]

22. "Prayerful Words for the Cracking of Scapulae" (I bokukata kitsushi ??? ??) [Kiboku sho lines 606-608]

23. "Specifications for the Division and Use of Turtle Shells" (Bunyo kiko josu ??????) [Kiboku sho lines 608-613]

24. "The Presentation of Offerings to the Two Planks of Shell Used in Royal Divinatory Cracking" (Gubu miura yoko niban ????????) [Kiboku sho line 613]

25. "Methods for the Raising of Fire and the Use of Water in the Scorching and Cracking" (Shakuki juka yosui kata ???????) [Kiboku sho lines 614-617]

26. "Chief Principles of the Five Branches of Cracks: Earth, Heaven, Spirit, Human, and Omen" (Jiten shinjin cho goshi juji ?????????) [Kiboku sho lines 618-628]

27. "Explanations of the Twenty-Nine Earth and Heaven Line-Graphs, the Thirty-Eight Spirit and Human Line-Graphs, and the Three Omen LineGraphs" (Setsu jiten kaku nijukyu ka shinjin kaku sanju hachi ka, cho sanka kotai ???????, ??????, ?????) [Kiboku sho lines 628-648]

28. "Incantatory Methods for Beseeching Cracks to Divine Miscellaneous Matters" (Boku zatsuji kitsu bokushi kata???????) [Mondo bubun lines 649-819]

29. "Increases and Decreases in the Number of Fires in Royal Pyroplastromantic Consultations for Sacrificial Rites" (Gubu omima no ura kasu zogen ?????????) [Saigi kankei kiji lines 820-854]

30. "The Recognition and Observation of Auspiciousness and Inauspiciousness in Royal Divinatory Consultations for Sacrificial Rites" (Gubu omima no ura kikkyo shoko ?????????) [Saigi kankei kiji lines 854-868]

The first twenty titles in the contents perfectly correspond to what remains of the Shinsen kisoki; titles 21-30 do not. These last ten section titles match up remarkably well with other works compiled in the Kiboku sho collection. Section titles 21-27 correspond (in perfect order) to the contents of the Kiso bubun.20 Section title 28 mirrors the content of the Mondo bubun,21 and section titles 29-30 more or less transcribe the 806 ce or earlier Saigi kankei kiji.22 According to the table of contents, the first fascicle of the Shinsen kisoki is comprised of three approximately equal parts devoted to the retelling of Kojiki myths,23 the historicization of pyro-plastromancy in Japan,24 and descriptions of pyroplastromantic rites and techniques.

Kudo (2005, 174) argues that although the Shinsen kisoki gradually collapsed onto a faithfully transmitted base (fascicle 1, sections 1-20) after the mid-ninth century, it also generated separate transmissions that expanded, but never completely shed their original contents. I agree with Kudo's assessment, but I see little evidence to support his conclusion that content from fascicles 2-4 was incorporated into the body of fascicle 1. Instead, I follow others who argue that fascicles 2-4 are now lost, but that technical portions of fascicle 1 (sections 21-30) were separately transmitted and periodically reappended to a diminished version of the Shinsen kisoki over time.25 A very similar hypothesis has been proposed and well defended by Kinoshita (2000, 145-47). The present study adopts Kinoshita's view that the mythological and historical sections of fascicle 1 were more or less faithfully transmitted, while most of the technical material in the final third of fascicle 1 was compromised and corrupted-yet never completely lost-between the early ninth and seventeenth centuries.

What remains of Shinsen kisoki fascicle 1 provides a bit of mytho-historical background, introduces Chinese precedents for the divinatory cracking of turtle plastrons, and stakes claims about the related ancestral kami and sacred skills of the Nakatomi and Urabe clans. It suggests that influential members of the Fujiwara ?? and Nakatomi clans played significant roles in the state transition from deer-bone pyro-scapulimancy to pyro-plastromancy, yet it clearly presents the Urabe as the "frontmen" in this transition.26 The Urabe are portrayed as descendants of the Nakatomi in the Shinsen kisoki, but many early texts and modern secondary studies depict them as the forefathers of both the Nakatomi and Fujiwara clans. Although Urabe Hiramaro????? (807-881) is traditionally remembered as the first person to have been officially rewarded the surname Urabe by the throne,27 quite a few Japanese studies convincingly show that local occupational and kinship groups adopted the surname long before the ninth century (for example, Inoue 1980; Yokota 1982; Nagadome 1984). Many of these works also argue that the surnames Nakatomi and Atai should be traced back to early Urabe kinship groups in Tsushima and Iki.28

The Shinsen kisoki's detailed descriptions of the Urabe clan's early history and achievements are often used to defend the aforementioned claims. For example, the preface to the Shinsen kisoki (lines 362-369) begins by informing us that "turtle techniques" (kijutsu ??) originated in the high heavens and have been esteemed through the ages.29 It argues that, just like a number other mantic arts, an "original canon" (hongyo ??) of "plastromantic omens" (kiso ??; literally, "turtle figures") still exists in lineage transmissions. The best transmissions, claims the preface, were received by the Urabe clan. Urabe lineages and their connections to the divinatory cracking of turtle plastrons, as we will see below, are both traced back to mythical times in the Shinsen kisoki.

The term urabe?? was used in different ways during the late eighth and early ninth centuries. As an official title, it referred to divination and ritual specialists employed under the Nakatomi supervisors of the Jingikan. As a surname, it referred to kinship groups associated with specific locales, kami, and crafts. Like the Nakatomi, the Urabe were known for their expertise in incantation, but unlike their Nakatomi descendants (or ancestors), their namesake or eponym was directly tied to divination (ura?). The Shinsen kisoki establishes and promotes genealogical ties between the Urabe and prominent kami in the early Nara state histories by weaving what appears to have been a local Urabe cult complex devoted to pyro-plastromancy into the state cult. One of the major goals of the present study is to determine the effects that this local cult-real or imagined-might have had on the late Nara and early Heian theocratic cult of jingi. A brief review the history of the Urabe clan, however, is needed to set up our analysis.

The Urabe Before, In, and After the Shinsen kisoki

Modern Japanese scholarship on the origins and early achievements of the Urabe is extensive (for example, see Hirano 1966; Inoue 1980, 126-87; Yokota 1982, 221-302; Nagadome 1984; and Kudo 1994.). This work is important because relations between the Urabe, the Nakatomi, and the Fujiwara clans had a great impact on the geopolitical and religious history of Japan. Little of this work, however, has made its way into modern Western scholarship. An article by Allan Grapard (1992b) on the Shinto of Yoshida Kanetomo ???? (1435- 1511) is one of the few exceptions. Grapard introduces some of the most substantial modern Japanese research on the early history of the Urabe, and is one of the only detailed English-language discussions on the subject (outside of encyclopedia entries) published to date.

Grapard's brief history of the "Urabe diviners" (1992b, 27-33) opens with Gari Ledyard's notion of a maritime state or power called the "Thalassocracy of Wa" (1975, 230-32). This thalassocracy, or loose confederation of thalassocratic chiefdoms, is supposed to have covered the southern portions of the Korean Peninsula and the southwestern coastal areas of Japan during the first half of the first millennium ce. Ledyard argues that this "empire at sea" was upset by an influx of continental culture during the centuries following the fall of the Eastern Han dynasty (25-220), and Grapard dates the birth of the Urabe to this era's largescale eastward migrations. Grapard extends this thesis to contend that Urabe diviners settled in the coastal areas of Wa, where they "customarily accompanied travelers on both land and sea, ... were held responsible for their 'projections' on direction, weather, time, and the outcome of the travel," and were "closely associated with the rulers, for they [also] made predictions and offered projections regarding battles, auspicious days and sites, and crop conditions" (1992b, 29). Plausible as they may be, none of these claims are defended, and the others that are rely on a combination of conclusions drawn from relatively late texts like the Engi shiki and the late fourteenth century Sonpi bunmyaku (Genealogies of the venerable and base).

The Urabe, according to Grapard (1992b, 30), continued to follow the expansion of Wa over organized and well-travelled sea routes to the eastern reaches of present-day Honshu. By the end of the fifth century, some had settled in places like the eastern province of Izu,30 where their expertise in divinatory practices- particularly the art of deer-bone pyro-scapulimancy-was recognized and awarded by the nascent Yamato court. Most of these points are borrowed from Inoue (1980) and other contemporary Japanese scholars whose studies on the early development of the Urabe are more fully introduced below.

Grapard (1992b, 33) goes on to contend that the rising status of the Urabe during the Nara and early Heian periods should be attributed to the support they received from the ruling political clan at that time, the Fujiwara. The Fujiwara shared the same ancestral kami as the Nakatomi and Urabe,31 and by the mid-ninth century, all three clans were prominent players in state religion and politics. The Urabe further distinguished themselves as authorities on the Kojiki and the Nihon shoki during the early Heian, and their control of Yoshida Shrine ???? in Heian-kyo-from its founding in the mid- to late ninth century- eventually led to the establishment of Yoshida Shinto (Grapard 1992b, 33).

Grapard relates much more about the early history of the "Urabe diviners" to set up the central topic of his article. His conclusion offers a brief synopsis:

[I]t seems that in very early times the Urabe and Nakatomi sacerdotal lineages existed in the western part of the isles, and that they became closely associated during the sixth or seventh century when the Yamato court evolved in central Japan. The Urabe were also present in the Izu Islands, although they do not appear in records of that area before the year 746.32 And, finally, the Urabe and Nakatomi sacerdotal lineages also settled in Kashima and Katori, in the province of Hitachi. The Urabe diviners played an important role during the Heian period; according to Engi shiki, a member of the Urabe lineage was to accompany all embassies bound for China. (Grapard 1992b, 33)

Grapard's history of the early development of the Urabe is eloquent and succinct, but his supporting evidence is rarely cited. The Kojiki, the Nihon shoki, the Engi shiki, the Kujiki, and the Sonpi bunmyaku, along with secondary studies by Hirano (1966) and Inoue (1980), are referenced to support some of the details in his summary of Urabe history. While I find the grand narrative that Grapard proffers to be both viable and compelling, I am also left with many questions. Where did all of this information about the Urabe come from? When, why, and how did Urabe plastromancy come to be favored over scapulimancy in Japan, and where can one find evidence to help defend answers to these questions? The Shinsen kisoki-a text never mentioned by Grapard-provides at least one set of contingent answers. However, before investigating exactly what it relates about the Urabe, we survey a few of the earliest occurrences of the term in Japanese texts.

Some claim that the earliest recorded mention of the term urabe?? appears in the 718 Yoro ritsuryo ???? (Regulatory and administrative codes of the Yoro reign period [717-724]).33 This, however, is difficult to substantiate because the Yoro Codes were lost early on, and the Ryo no gige, which provides commentary on a version of them, was not completed before 833. Others argue that the earliest dateable mention of the term urabe should be attributed to one of a number of eighth century occurrences of terms like bokusha??, Ura no uji??, and urabe ??, which are all commonly read as variants of??.34 The first definite occurrence of the term in received Japanese texts does not seem to surface until the mid-eighth-century Man'yoshu (Collection of a myriad leaves).35 The poem in which the term appears almost certainly predates this period, but it does not provide any evidence that would allow a more exact dating. The clear allusion to the divinatory cracking of turtle shells it contains, however, points to a clear connection between Urabe diviners and the art of pyro-plastromancy.

At least one other text predating the 830 Shinsen kisoki mentions the term urabe??. The Shoku Nihongi records the term once (3: 18, 104), in an entry dated to the fourth lunar month of 750. It records part of an edict describing rites for the purification and renewal of the nation and a great imperial amnesty (taisha ??) meant to initiate the process. The end of the entry adds that, "midlevel crimes committed by Nakatomi-Urabe-Ki no Okiko no maru were cleared" (? ? ? ? ? ? ? ? ? , ? ? ? ? ). We are never told what these crimes were,36 but the entry is informative in other ways. It lends credence to at least two important claims made in the first Shoku Nihongi entry: that the term urabe?? was used in a surname during the mid-eighth century, and that connections between the Urabe and Nakatomi were close.

The Shinsen kisoki first mentions the term urabe?? in line 366 of the preface, where "mid-level functionaries in the guild of official diviners" (urabe shokujo ????) are said to have composed the text to illustrate their authority in the art of pyro-plastromancy. Ten of a total of twelve occurrences of the term in the body of the text appear in the following passage (roughly corresponding to titled entries 15-17 in the table of contents). This portion of the Shinsen kisoki introduces specific sites, kami cults, and clans associated with biannual "pyro-plastromantic consultations for the royal body" (omina no ura ????). It reads:

...

The base shrines of the Futonorito Shrine complex are in the three provinces.37 Today, [royal] sacrificial rites are performed at the base shrines of the Urabe-managed Kushimachi Shrine complex in the two provinces.38 Sacrificial rites [for the royal body] are also performed at the Urabe-housed Koba Shrine complex.39 Most of the time, water is used in scorching and cracking. The apical ancestor of the Urabe of Iki Island is Amahitotsuwashira no mikoto. The apical ancestor of the Atai on Tsushima Island is Chinsen no mikoto.40 Both assisted Ame no koyane in pyro-plastromantic consultations. [What is determined about] the good or bad fortune of the royal body [in these consultations] is valid for a three-year span. Today it is the Urabe who report to Ame no koyane and handle memorials submitted to the [spirit of] the First Sovereign Kamuyamato Iwarehiko (Emperor Jinmu) at the Ooyama totoyo akitsushima no miya (a shrine devoted to the island of Honshu). Sovereign Ikumeiri no Isachi (Emperor Suinin [trad. r. 29-70]) decreed awards for border provinces that established [branch] shrines for the spirits of heaven and earth. When the male attendants of the Futonorito Shrines [in Tsushima and Iki] were first informed of these decrees, they established another [Futonorito Shrine] in the province of Yamato. During the reign of the Sovereign Tarashinakatsu hiko (Emperor Chuai [trad. r. 192-200]), the Great Lord of Thunder (Ikatsu omi), twelfth-generation descendant of Koyane, was placed in charge of the administration of divine affairs. He declared that everyone with the surname Urabe in the eastern provinces were his descendants. He had the Urabe of Izu administer the handling of divinatory matters. Today, [these responsibilities] have come to be associated with the Urabe of the four provinces, including five members of the Urabe of Izu, five members from one Urabe clan of Iki Island, and ten members from two Urabe clans of Tsushima Island;41 altogether twenty people from three clans. (Shinsen kisoki, lines 567-84)

The network of Urabe-managed shrines described at the beginning of the passage introduces established cults in Tsushima and Iki devoted to the spirits of turtles and hahaka wood ???.42 The underlying message of the whole passage is clear: the Urabe of the four provinces are bound together and bound to the state cult because of their traditional ties to the divinatory cracking of turtle plastrons.

The centripetal force of the increasingly centralized Yamato chiefdom appears to have drawn pyro-plastromancy into the Kinai Plain, where it contended with deer-bone pyro-scapulimancy as the preferred method of state divination. The Shinsen kisoki passage translated above explains that the Urabe of the eastern provinces were descendants of Ikatsu omi ??? and shared lineage ties to Ame no koyane ???/???? (apical ancestor of the Nakatomi; distant ancestor of Ikatsu omi and Futonorito ???).43 The eastern Urabe of Izu may have been the first group to win imperial recognition in Japan, and they may have also been the first to crack turtles in the Jingikan, but they were not the first people to use the surname Urabe, and were certainly not the first to crack turtle plastrons in Japan. The Shinsen kisoki constructs a genealogy meant to elevate the standing of the eastern Urabe at court.44 The value of these claims depends on our ability to discern the intentions behind their expression. This task is continued below, but knowing that there is a substantial body of Japanese scholarship devoted to unraveling the history of the Urabe, I will continue with a general overview of it. Much of this research focuses on the Urabe of Tsushima and Iki. Without the space to introduce more than a few insights from the most seminal and oft-cited studies, I will be brief.

In perhaps the most detailed modern study on the Urabe to date, historian Inoue Tatsuo (1980) emphasizes that Urabe lineages in different locales must be examined separately to understand the early history of the clan. Inoue explains that members of Urabe lineages in different parts of Japan were purposefully brought together in the Jingikan sometime during the late seventh and early eighth centuries, but local Urabe lineages and kinship groups extend much further back in time and have their own histories. Inoue attempts to reconstruct these histories with hundreds of references to works dating to a wide variety of different time periods. He concludes that the term urabe?? was already a surname used by different local clan-based guilds in fifth century Tsushima and Iki, and that certain members of these groups moved to the eastern reaches of the empire where they were awarded the name Nakatomi in the sixth century. The Nakatomi maintained close relations with the eastern Urabe, emerged as leaders of the Jingikan during the seventh century and, with political support from their Fujiwara descendants, propagated shared genealogies and mytho-histories to bind all three clans together during the eighth century. Inoue concludes that by the mid-ninth century, the Urabe had been recognized as a divine clan, plastromancy had displaced scapulimancy as the orthodox method of state divination, and the Nakatomi had become the sole supervisors of state sacerdotal matters in the Jingikan (Inoue 1980, 183-87).

Historian and mythologist Yokota Ken'ichi (1982) reexamines the origins of the Urabe through the lens of myth. In one sequence of comparisons, he draws from the Shinsen kisoki, the late fourteenth century Sonpi bunmyaku, and the late eighteenth century Matsunoo shake keizu (Diagramed genealogy of the shrine lineages of Matsunoo) to trace the mythical origins of the Urabe back to Ikatsu omi and the reign of Emperor Chuai.45 Yokota admits this evidence is relatively late, but his comparative exercise is meant to identify traditions that can help explain fragmented records from earlier times. Like Inoue, he argues that the Nakatomi branched off from an Urabe family before the sixth century CE and employed members from eastern Urabe families in and around the capital during the sixth century as official diviners. The Nakatomi were written into the early Nara mytho-histories as descendants of the already recognized state kami Ame no koyane, and they turned to the Urabe of the four provinces during the eighth century to promote pyro-plastromancy. Yokota also contends that scapulimancy and plastromancy were both performed from the mid-fourth to the mid-sixth century, and the gradual transition from deer to turtles in state divinatory consultations is ultimately attributable to the branching out of the Japanese aristocracy that occurred during the eighth and ninth centuries (Yokota 1982, 231-35).46 In short, little in the mythical accounts mentioned above contradicts previous research on the history of the Urabe. Yokota concludes that the Urabe originated sometime before the sixth century, were associated with divination and incantation from the start, and gained prestige during the late Nara and early Heian periods as their Nakatomi patrons outmaneuvered competing clans (like the Inbe) for dominance in the Jingikan.

Archaeologist and historian Nagadome Hisae (1984, 90-110) argues that pyro-plastromancy entered Tsushima long before the earliest dates of transmission to Japan proposed by scholars like Inoue.47 According to Nagadome, the technique was transmitted directly from China, and its earliest appearance in Japan coincides with the emergence of the Urabe of Tsushima and Iki sometime during the fourth or fifth centuries ce. Three insights are offered to defend these views. First, the long history of plastromancy in China and the transmission of continental cosmological schemes through places like Tsushima and Iki before the middle of the first millennium ce both suggest that the technique would have likely entered those areas much earlier.48 Second, the divined shell recovered from the Shitaru ruins in northwest Tsushima should be dated much earlier than previously claimed (to the third or fourth rather than the sixth century).49 And third, there are no archaeologically recovered turtle shells from the Korean peninsula.50 Nagadome claims that pyro-plastromancy likely arrived in Tsushima and Iki during the third century, giving rise to specialists by the fifth century. He further speculates that by the sixth century, Urabe lineages in the east were performing pyro-scapulimantic consultations with deer bones, but this practice was gradually displaced by the divinatory cracking of turtle plastrons during the Nara period,51 and a new Urabe genealogy was constructed to serve Nakatomi interests sometime during the late Nara and early Heian periods (Nagadome 1984, 98).52 The Nakatomi relied on the services of the urabe because of genealogical and professional relations, and because diviners-particularly pyro-plastromancers-played a significant role in the continental Chinese theocratic models promoted by both the Nakatomi and their Fujiwara relatives. While Nagadome pushes the history of the Urabe much further back in time than Inoue or Yokota, there is very little received or recovered evidence to support this theory.

Finally, an article by Kudo (1994) introduces an interesting hypothesis on the origin of Kamezu hime no mikoto ?????, kami of turtles and emanation of the divine apical ancestor of the Urabe clan in the Shinsen kisoki. Kudo begins with the following entry from the Nihon shoki, dated to the seventh lunar month of 477 ce:

...

Mizunoe no Urashimako of Tsutsugawa in the Yosa District of Tanba Province (north of present-day Kyoto, near the coast) boarded a boat and went fishing. He ended up catching a large turtle that transformed into a woman. Thereupon, Urashimako was moved and made her his wife. They entered the sea together, went to Mount Horai (Ch. Penglai), and mingled with the transcendents there. This account is recorded in other texts. (Nihon shoki, 2: 14, 206)

Urabe Kanekata's???? (fl. 1274-1301) late thirteenth century Shaku Nihongi (Commentary on the Nihon shoki) cites a fragment from the long-lost mideighth-century Tango no kuni fudoki (Regional gazetteer of Tango Province) recording the name of Urashimako's ??? wife as Kamezu hime ???? (5, 112). Though it is impossible to say for certain whether the Shinsen kisoki appropriated from or contributed to this legend (Kanekata could very well have added the name of Urashimako's wife to the Tango no kuni fudoki account based on a likely familiarity with the Urabe traditions described in the Shinsen kisoki), Kudo cites the account as evidence that Kamezu was neither created by the authors of the Shinsen kisoki nor exclusively bound to the Urabe. He argues that Futonorito was traditionally a major object of the Urabe cult, while Kamezu and Ikatsu omi were appropriated and elevated in status by the authors of the Shinsen kisoki. Of the three, Futonorito is most directly associated with incantation-the one skill that the Urabe and Nakatomi seem to have consistently held in common. Other associations (Futonorito as Ikatsu omi, as Atomimi, or as a primary form of Kamezu) would have been easier to justify and sustain given this long-standing skill-based commonality.53 In short, Urabe ties to Futonorito and spells appear to predate more consciously constructed links to spirits explicitly associated with plastromancy.

The Nakatomi gained official clan status by the mid-seventh century and are regularly depicted as sacerdotal supervisors of the Jingikan in eighth and ninth century texts.54 The Urabe, on the other hand, are not formally recognized as a state clan until the mid-ninth century-many decades after the earliest descriptions of their official pyro-plastromantic duties in the Jingikan. The earliest extant record of the pyro-plastromantic duties of the urabe in this office likely appears in the Shinsen kisoki, which claims that the title "superior supervisor of divinatory cracking" (boku no chojo???) was instituted by the throne in 774 (line 598). Moreover, the character ura/boku? refers to pyro-plastromancy throughout the Ryo no gige (for example, 1, 29; 2, 81; and 4, 155), where all occurrences of the term urabe refer to diviners in the Jingikan.

Neither the early Nara ritsuryo nor the early state mytho-histories ever mentions the divinatory cracking of turtle shells. This appears to have left the door open for different clans to stake claims to the technique during the mid- to late eighth century, and attempt to monopolize it as their own during the early Heian period.55 The Nakatomi must have seen an opportunity to increase their power and legitimacy in court by replacing the older, imperially-sanctioned deerbone scapulimancy with a "new" brand of pyro-osteomancy revered in China and familiar to the urabe in their employ. The urabe, as a professional group of mantic specialists subordinate to-but working closely in tandem with-an imperially-recognized uji ? or clan, would have been at least partially bound to the clan kami (ujigami ??) of the Nakatomi, but they also had their own kami cults. It was not their place to demand an officially recognized lineage connection to the royal clan, but this does not mean that they did not relish an opportunity to construct a mythical foundation for it with Nakatomi support. The Shinsen kisoki promotes Nakatomi aims, but it veils them in the more transparent and immediate intentions of certain members of Urabe kinship groups. Both groups benefitted from these claims, and the Nakatomi and Urabe clans continued to play important roles in state religion and politics for the rest of the Heian period and beyond.56

Pyro-plastromancy in the Shinsen kisoki

What remains of the first fascicle of Bonshun's Shinsen kisoki provides a number of interesting hints about how pyro-plastromantic consultations might have been conducted in early Heian Japan, and how this technique's displacement of deer-bone scapulimancy was justified. For example, lines 536-566 draw from a lost work titled Kisei, or The Pledge of the Turtle, to record an account of a pledge or promise made by a kami of turtles to a retinue of heavenly spirits.57 The covenant expressed through this pledge is said to have been responsible for the replacement of the heavenly art of scapulimancy with a mantic technique better suited for both heaven and earth.

The account of the "pledge of the turtle" begins with a meeting called by Kami rogi ??? (that is, Amaterasu) and Kami romi ??? (that is, Takamimusuhi no kami ??????). Deliberations end in a decision to send down Mima no mikoto ??? (great imperial grandson of Amaterasu) to rule over the land. As Mima is leaving, he asks the kami above who they will dispatch to assist him in political administration and religious rites. A white and truly famed deer (shiro manaka ????) of Heavenly Mount Kagu ??? steps forward and declares: "I will assist and offer myself up. Reach in, pull on, and extract my scapulae. Inquire about matters with a fire-induced crack omen" (lines 541-542).58 Afterwards, Futonorito steps forward to say:

...

The white and true deer can discern matters above the nation, but how can it discern matters below the ground? I am able to discern above the nation and below the ground, the matters of the heavenly deities and the earthly spirits.

( Shinsen kisoki, lines 543-545)

Mima no mikoto then descends from the heavens for a tour of the underwater realm of Futonorito. Futonorito, who we learn later is in the manifested form or substitute body of a turtle spirit (Kamezu hime), proclaims the following processes to which she is to be subjected:

...

Dry my "eight-cross bone" (that is, shell) by exposing it to the sun. Use an axe to chop the "thousand-like dividers" (that is, the bridges).59 "Thousand" refers to the [two] dividers separating the upper shell from the lower shell. Take the "perfectly clear mirror" (that is, the lower shell or plastron) and prepare it. With a heavenly blade, hollow out machi on, slice,60 and sweep it.61

( Shinsen kisoki, lines 552-554)

Kamezu hime sacrifices her turtle body for pyro-plastromantic consultations expected to be conducted by kami diviners in the heavenly retinue of Mima no mikoto.

The rest of this section of the Shinsen kisoki (lines 554-566) provides brief directions for how to start a fire and how to select hahaka wood to scorch a shell. In both cases, materials are to be procured from Heavenly Mount Kagu and are associated with named kami connected to an organized cult managed by the Urabe. Two of the most significant kami named in this part of the Shinsen kisoki are Hikiribi ?? /???, a kami of fire, and Hahakaki no kami ????, kami of hahaka wood pyromantic pokers (and a manifestation of Kushimachi ???). Details on the technical processes involved in pyro-plastromantic cracking begin in lines 527-536 of the Shinsen kisoki. This section of the text provides a brief synopsis of technical information purportedly drawn from a no longer fully extant medieval Chinese work entitled Guijing ?? (Classic of turtles), including comments on nine basic types of turtles and five colors of turtles; taboo days for pyro-plastromantic consultations; and associations between seasons, quarters of turtle shells, and the proper placement of hollows for cracks. Much of this information appears to have been directly appropriated from the early to mid-eighth-century Tang liudian (Six institutes of the Tang) (14, 412). Some of these concepts might have been adopted in practice, but a majority of the specifics do not match processes detailed in the remainder of the Shinsen kisoki. For example, unlike the Guijing (a text traditionally attributed to Liu Shilong ??? [442-491]),62 the Shinsen kisoki consistently affords a separate place to the center of the shell. The plastron was not simply quartered (as it was in Han and medieval China), but was marked with a five-line symbol known as a machi ? or machigata ?? ( machi form). The machi symbolically divided the surface of the medium into five separate parts corresponding to each of the five phases: 1. top vertical fire; 2. left horizontal wood; 3. middle vertical soil; 4. right horizontal metal; and 5. bottom vertical water. The machi symbol is not found on the Maguchi Cavern or the Shitaru shells, and is not attested outside of Japan. This convention almost certainly originated in Japan, probably with the Urabe.

The use of the machi is not carefully explained in what remains of the Shinsen kisoki, but the table of contents provides just enough information to conclude that the five omen and oracle lists in fascicles 2-4 were likely organized according to the five lines comprising this symbol. Each of the five machi lines is named, correlated with one of the five phases, and associated with a particular class of crack omen in lines 623-628 of the Kiso bubun. The table above records these names, correlations, and associations.

Details about the fabrication and interpretation of pyro-plastromantic crack omens must have been much better documented in the earliest editions of the Shinsen kisoki. Although the table of contents details certain aspects of these processes, little of this information appears in what remains of the text. For example, section titles 26 and 27 in the table of contents advertise the production and reading of crack omens, but the body of the Shinsen kisoki says very little about them. They are, however, mentioned in the Kiso bubun,63 where a total of 137 crack forms are introduced.64 It is impossible to determine the extent to which these lists might have matched the content of fascicles 2-4 of the Shinsen kisoki, but a connection is hard to deny.

The Shinsen kisoki, the Kiso bubun, and the Saigi kankei kiji collectively describe pyro-plastromancy as an interpretive art based on cosmological schemes adopted from medieval Chinese texts. They also depict it as a mechanical technique based on an innovation called a machigata, and as an Urabe-managed ritual and cultic practice devoted to the kami of turtles, hahaka wood, fire, and spells. None of these texts link specific divinatory topics to descriptions of crack forms or prognoses, but, as mentioned above, the Kiso bubun includes detailed comments on the production and reading of cracks. It is possible that the Kiso bubun dates later than the Shinsen kisoki, but the latter must have contained omens and oracles sharing much more in common with what was eventually transmitted as the Kiso bubun than with any version of a Chinese Guijing. Names of the five machi lines fill the Kiso bubun lists, while terms like head (shou ?/tou ?), body (shen ?), and foot (zu ?)-common in Han and medieval Chinese descriptions of pyro-plastromancy-never appear.65 The Shinsen kisoki plainly reveals that the Urabe drew from mid-Tang Chinese sources to construct a revised brand of pyro-plastromancy with its own cult, consecution, and semiotic system.

I submit the following outline of what a pyro-plastromantic consultation would have entailed according to information drawn from the Shinsen kisoki, the Kiso bubun, and the Saigi kankei kiji.

1. Preparation of the shell:

Dry a turtle shell in the sun. Cut the bridges connecting the plastron to the carapace with a small ax, dividing the shell in two. Cut each half of the plastron into six pentagonal plates,66 and carve or chisel a machigata on the surface of all twelve of them.67

2. Preparation of the fire and poker:

Gather firewood, ignite fires with this wood, and light the tips of hahaka wood pokers with these fires.68

3. Preparation of the diviner:

Observe ritual purifications for a number of days and bathe.69 Two diviners take their proper places at dawn.70 They begin by invoking Amatsu norito no Futonorito no mikoto ???????? (spirit of turtles).71 One diviner holds the two sheets of the shell and recites pledges to them. Each diviner uses a separate fire and poker to scorch machi on their respective sheets.

4. Fabrication of the crack:

First scorch from earth below to heaven above, then douse the plastron with water. Next, scorch from east to west, then once again douse it with water.72 Repeat this process until a crack forms.

5. Interpretation of the crack:

Match crack omens to five separate lists based on the machi line from which each crack emanates. Interpret the oracles in these lists to determine auspiciousness and inauspiciousness.73

Pyro-plastromantic consultations performed in the Heian Jingikan likely involved the scorching of strips of plastrons to manufacture cracks that were interpreted as divine figures, images, or omens. The transmission of the technique from China, to the islands of Tsushima and Iki, and into central Honshu transformed the practice in significant ways. The plastron came to be divided into placard-like strips for divinatory consultations, and the round, oval, square, and rectangular hollows typically found on Chinese shells were replaced by the machigata. We have already touched on a number of other changes, and while much more could be said about the functions, the ritual uses, and the technical processes involved in ninth century and later Japanese practice,74 detailed explanations would take us much further from our expressed focus on the transition from deer bones to turtle shells in late Nara and early Heian Japan.

Conclusion

Japan is the only place in the world where the art of pyro-plastromancy is still widely practiced. Transmitted in Shinto lineages throughout the second millennium ce, this variety of divination continues to be performed under imperial auspices and at local shrines throughout Japan.75 The present study traces the early rise and development of divination with fire and plastron in Japan. It details the local and state clans, spirits, and cults involved in the technique's transmission and growth, and reveals much about the Nara and Heian state and local religious traditions that would eventually contribute to the advent and growth of Shinto.

Scholars of Japanese religion continue to question Shinto's view of itself as the religion of the Japanese people since time immemorial. Some regard the Kojiki as the beginning of Shinto, and some argue that the term should be reserved for post-Meiji Reformation (1868) state religion. Mark Teeuwen, a well-published advocate of the view that Shinto was born when people began to talk about it as a distinct tradition, dates its emergence to sometime between the thirteenth and fifteenth centuries.76 Much of his work on the advent of "Shinto" is devoted to the historicization of earlier traditions recognized as building blocks for this native Japanese "Way." Teeuwen frames Shinto as a medieval creation that drew much from earlier traditions like local kami cults, the state cult of jingi ?? (spirits of heaven and spirits of earth), and the Way of the Buddha. While Buddhism was already well established and quite pervasive by the end of the Nara period, the present study has focused much more intently on the first two traditions.

The Shinsen kisoki provides evidence that local kami cults in places like Tsushima and Iki continued to redefine and shape the state cult of jingi long after the Taika Reforms, the inception of the ritsuryo system, and the writing of the Kojiki and the Nihon shoki. The adoption of Chinese imperial models by the Nara and Heian courts was an ongoing and selective process motivated by the contingent needs and aims of particular groups of people with the power, capital, and predilection to make specific innovations work.77 The Kojiki and Nihon shoki appear to institutionalize the divinatory cracking of deer scapulae in the early Nara imperial court by writing the technique into state myth. Both of these early eighth century mytho-histories name divine progenitors of the art, and both present these progenitors as apical ancestors of specific clans. The Shinsen kisoki submits a rewrite. It substitutes pyro-plastromancy for pyro-scapulimancy and proposes a new, named, and streamlined set of divine progenitors and associated clans. Its compilers tell us that pyro-plastromancy was adopted into the late Nara imperial ritual system on a very limited basis, only to be elevated and more definitively institutionalized-along with its associated deities and clans- by the early Heian court. The Shinsen kisoki justifies the incorporation of a clanbased cult devoted to kami of turtles, hahaka wood, and fire into the Japanese theocracy. It weaves this allegedly local, Urabe-managed, shrine-based, pyroplastromantic kami cult complex into the then-contemporary mythology of the Japanese imperium. Although much purposeful reimaging must have occurred in the composition of the text, there is good reason to believe that plastromantic cults and kami existed long before the early Heian period.78

The Japanese theocracy experienced a number of major changes during the late eighth and early ninth centuries. In 784 ce, the capital and court were moved to Nagaoka-kyo, and in 794, they were moved to the nearby Heian-kyo. Nevertheless, at the start of the Heian period, members of the Fujiwara clan continued to hold many of the most important political posts in the Office of the Imperial Bureaucracy (Daijokan ???), members of the Nakatomi clan continued to supervise the Office of the Spirits of Heaven and Earth, and members of the Urabe of the four provinces continued to crack turtle plastrons as urabe officials under the direct supervision of the Nakatomi. The official shift from deer-bone pyro-scapulimancy to pyro-plastromancy benefitted and empowered certain clans (at the expense of others) within the imperial hierarchy.

Members of Urabe clans may have led the turn from deer scapulae to turtle plastrons, but the Nakatomi seem to have been just as invested in the change. As supervisors of the Jingikan, and as relatives of the Fujiwara, the Nakatomi were in an excellent position to reestablish their authority at the beginning of the Heian period. A change in the conventional method of divining state affairs would have provided them with a way to redefine and establish their own sacerdotal duties, distinct from those of competing clans. It would also have allowed them to identify more closely with the Tang Chinese imperial models then in vogue at court, and to elevate the status of relatives in their employ who had strong geopolitical ties to both eastern and western Japan, and could be fairly easily woven into state myths. The Nakatomi used Fujiwara political power and Chinese-inspired Urabe pyroplastromancy to help gain control of the Jingikan, and the Nakatomi and Urabe clans rose to new heights during the Heian period as kinship groups with common professions and related ancestral kami. The ritualization and official adoption of pyro-plastromancy for state divinatory consultations empowered all three of these clans.

The Shinsen kisoki is the earliest known treatise on pyro-plastromancy ever produced in Japan. It describes and defends the state's use of turtle plastrons rather than deer scapula, includes some of the earliest recorded examples of norito, and is the earliest text recording passages from different sections of the Kojiki. Look a bit more closely at both its content and the context in which it claims to have been produced, and the text takes on additional significance as an early history of the Urabe, the Nakatomi, and the Jingikan; as a rare record of eighth century local cults; and as a window into the development of late Nara and early Heian state religion and politics. It also plainly promotes many of the lineages and traditions that would, centuries later, coalesce into a Japanese "Way" known as Shinto.

Footnote

* This article is dedicated to the memory of my remarkably fearless and generous teacher and ally, John R. McRae. I hope that it brings to light something you saw when you suggested looking further into the question of deer bones and turtle shells not too long ago.

1. Chinese "charging the turtle" methods involved the "charging" (ling ?) or "mandating" (ming ?) of a spirit or spirits before scorching a point on bottom half of a turtle shell (that is, the plastron) with the glowing tip of a wood poker. The resultant crack was then interpreted to determine whether the spirit/s deemed the "charge" to be auspicious or inauspicious. Known today as pyro-plastromancy, the technique essentially involves the localized scorching of the surface of a turtle plastron and the interpretation of the resultant crack as a divine image, figure, or omen. Classical Chinese terms for the technique include guibu ?? (turtle [shell] cracking), bu? ([turtle shell] cracking), and bujia?? (the cracking of [turtle] shells). There is a massive corpus of scholarship on pyro-plastromancy in ancient China. For work on the history and development of the technique in medieval China (coinciding with the time period under investigation in the present study), see Liu (1992, 403-409); Bai (1998); and Kory (2012).

2. Pyro-plastromancy is mentioned at least fifty times in the five post-Nihon shoki state histories constituting the Rikkokushi (Six state histories). It is mentioned in tandem with yarrowstalk divination (a mantic technique introduced in note 4) in a majority of these instances. For example, Shoku Nihongi entries for the years 708 and 782 ce record "turtle and yarrow" (kizei ??; that is, pyro-plastromantic and achilleomantic) consultations. Both of these accounts are briefly described below. Nihon koki entries for the years 796, 802, 805, 806, 823, 831, and 832, and Shoku Nihon koki entries for 842 (2 occurrences) and 844 (3) document instances of "crack and yarrow" (bokuzei??; another synecdoche for pyro-plastromancy and achilleomancy); and a Nihon Montoku tenno jitsuroku entry for 858 records a "turtle and stalk" (kizaku ??) consultation. The Nihon sandai jitsuroku mentions "crack and yarrow" in entries for 863, 865, 866, 870, 873, 874, 875, 876, 878 (2). The same text mentions "yarrow and turtle" (shiki ??) in entries for 864, 866 (2), 870, 871, 874, 878 (2), and 881. Two of the examples listed here are particularly significant for the present study. The 806 Nihon koki entry (23rd day of the 3rd lunar month [????]) explicitly connects the term boku? (crack; divination) with turtles, and the 858 entry (10th day of the 4th lunar month [????]) from the Nihon Montoku tenno jitsuroku depicts members of a particular Urabe?? clan lineage as masters of the arts of pyroplastromancy and achilleomancy.

3. For work on the dating and authenticity of the Kujiki, see Bentley (2006), who argues that the text must have been compiled before the mid-eighth century. The Kujiki mentions the divinatory cracking of deer scapulae in a number of different places. See Sendai kuji hongi (1.23, 2.97, 7.387); Bentley (2006, 125, 151-53, 269). This trinity of examples contains: 1. an account of Izanagi and Izanami turning to the heavenly kami for divinatory consultations; 2. an account of Amaterasu charging Ame no koyane with divination and exorcism; and 3. comments on the supervision of royal rites shared by the Nakatomi and Inbe ?? during the early Nara period.

4. Achilleomancy involves the casting and counting of achillea (yarrow) stalks to determine numbers, which are calculated or interpreted as omens. The technique represents a form of cleromancy or sortilege. It is commonly used in tandem with the oracular system canonized in the Yijing ?? (Classic of changes), and it is frequently mentioned alongside pyro-plastromancy in early China (for example, in the compounds "crack and yarrow" [bushi??] and "turtle and yarrow").

5. See Shoku Nihongi (1: 4, 130). For more on the long-standing Chinese tradition of consulting turtle plastrons to test a date or site selected for the relocation of an imperial capital, see Kory (2012, 251-61).

6. See Ryo no shuge (13: 2, 31-33); and Ritsuryo (2, 157). A similar clarification appears later in the Ryo no shuge: "The scorching of a turtle [shell] is called boku/ura" (?????) (4, 155).

7. This is also stated in Nakano (1975, 178); Shigematsu (1986, 325-39); Grapard (2003, 84).

8. My use of the term "ritualization" is adopted from the work of Catherine Bell, who generally describes it as a process of "differentiation and privileging" (1992, 204); more specifically as "a matter of various culturally specific strategies for setting some activities off from others, for creating and privileging a qualitative distinction between the 'sacred' and the 'profane,' and for ascribing such distinctions to realities thought to transcend the powers of human actors" (1992, 74).

9. According to line 605 of the Shinsen kisoki, the completed text was offered to Emperor Junna ?? (r. 823-833) on the eleventh day of the eighth lunar month (5 September) of 830 ce by Urabe Totsugu, an acting junior eighth-rank lower-grade superior supervisor of divinatory cracking (boku no chojo???). Lines 602-605 of the Shinsen kisoki list the following co-compilers and commentators: 1. Urabe Katsumuro?????, a great scribe (daishi ??) and senior sixthrank upper-grade official; 2. Urabe Shimatsugu????, a junior sixth-rank upper-grade official; 3. Masunaga ?? of the Izu Island Atai [clan] ????, a senior sixth-rank upper-grade official; and 4. two members of the Iki Atai clan ???? (the director of the Seisai Shrine ???, and a junior seventh-rank lower-grade official of Kamizukasa Shrine ??? named Hirokichi ??). None of these individuals appear elsewhere. For more on the system of ranks used here (first rank highest), see Hall (1966, 71-72).

10. Bonshun, also known as Ryugen ?? or Jinryu ?? (after the Jinryu Shrine ??? in Kyoto where he once studied), served as head abbot of Toyokuni Shrine ???? for a number of years, indicating membership in both the Yoshida clan ??? (a branch of the royal clan) and the priesthood of Yoshida Shinto ????. More on connections between the Urabe and Yoshida clans is discussed below, particularly connections detailed in Grapard (1992b). Remembered as a prolific writer, Bonshun's most famous work is the lengthy Shun kyuki ??? (Shun's records of former times).

11. Kudo (2003, 144) traces the transmission of the Shinsen kisoki from its presentation to the throne in 830 ce, to a 973 copy by Urabe Masanobu???? (fl. 973 ce), to a mid-sixteenthcentury edition attributed to Funabashi Nobukata ???? (1475-1550) and his son Funabashi Narikata ???? (1499-1566), and finally to Bonshun's 1620 handwritten copy. For more on the transmission of the text between 830 and 1620, see Kudo (2005, 176-78).

12. I primarily rely on Kudo's (2005, 22-62, lines 362-605) typeset, annotated, and collated edition of the Shinsen kisoki throughout the present study. Kudo (2005, 10-14) critically compares Bonshun's copy of the Shinsen kisoki with fourteen later editions. Eight are dated (1691, 1713, and six Meiji); six are not. Oft-cited modern editions of Bonshun's Shinsen kisoki include a photographic facsimile prepared by Tsubaki (1957, 111-64), a hand-copied and annotated edition by Akimoto (1978, no page), and a typeset and annotated edition by Akimoto (1992, 13: 177-237). At least five editions of the Shinsen kisoki have been published since 1913, and its full title appears in at least thirty academic Japanese articles published since 1938.

13. For more on norito, see Philippi (1959); and Bock (1970-1972, 2: 57-105).

14. Connections between the Kojiki and Shinsen kisoki are detailed in Kurano (1959); Kudo (2005, 213-43 and 318-19).

15. Discussion on the early history of the Jingikan can be found in Bock (1970-1972, 1: 17-24).

16. For more on the aims, historical value, dating, and transmission of the text, see Kato and Hoshino (1972, 1-12).

17. A significant portion of the Kogoshui is devoted to eleven complaints about the Nakatomi. For a translation of these complaints, see Kato and Hoshino (1972, 63-66). For a clear and concise description of them, see Grapard (2002, 224-25). The official responsibilities of the Nakatomi and Inbe clans are clearly stated in lines 501-503 of the Shinsen kisoki, where Ame no koyane and Futodama are both summoned to serve the early divine rulers of the nation. The heavenly kami invest Ame no koyane-apical ancestor of the Nakatomi clan-with control and management of divination and divine invocations, while Futodama-apical ancestor of the Inbe clan-is afforded control and management of matters involving sacrificial offerings and the handling of the royal regalia. The Kogoshui, however, states that biannual royal divinatory consultations (miura ??) were instituted and controlled by the Inbe clan in the mid-seventh century; see Kogoshui (1: 63); translated in Kato and Hoshino (1972, 43).

18. Akimoto (1992) presents the Kiso bubun (lines 606-648) as part of the Shinsen kisoki, and this text accompanies all fifteen editions of the Shinsen kisoki compared in Kudo (2005, 10-14 and 174). Because of this, I maintain-following Akimoto and Kudo-that the Kiso bubun was bound to the Shinsen kisoki long before Bonshun's copy, likely in or before Urabe Masanobu's 973 edition. Considered along with the court-commissioned Saigi kankei kiji (a record of the "royal cracking for the royal sovereign" [omima no miura ????] allegedly instituted by Emperor Heizei ?? [r. 806-809] in 806 ce), these three works seem to provide a fairly detailed and consistent view of pyro-plastromantic practice in the early- to mid-Heian court. A description of the technique based on these works is presented below. For more on royal plastromantic rites during the Heian period, see Engi shiki (18: 1, 23-24); Bock (1970-1972, 1: 79); and Inoue (2008).

19. This claim is based on the likelihood that the titles of fascicles 2-4 collectively mention all five parts of a machi ? or a machigata ?? (machi form). The machi was carved or chiseled on shells to direct scorching and to help control the direction of cracking. The five categories featured in the titles for fascicles 2-4 refer to the five lines that constitute a machi. This strongly suggests that all three fascicles were devoted to pyro-plastromantic omens and oracles associated with machi lines. All five lines, along with the image of a machi, are introduced in table 1.

20. Nearly a third of the section titles for fascicle 1 correspond to what is now recorded in the Kiso bubun, and all extant editions of the Shinsen kisoki are directly followed by the Kiso bubun. A closer bond exists between these works than any others in the Kiboku sho collection.

21. The Mondo bubun (lines 649-819) is a compilation of questions and answers on the fabrication and interpretation of pyro-plastromantic cracks to divine a wide variety of different topics.

22. The Saigi kankei kiji or Records Related to Sacrifices and Ceremonies outlines the history of, procedures for, and participants involved in early ninth century "royal divinatory [consultations]" (omima no ura ???) to determine auspicious dates for the Divine Offering of Foods (Jinkonjiki/Kamuimake ???). The Jinkonjiki was a sacrificial rite in which specially prepared rice was offered to the reigning thearch who was to receive it in the company of Amaterasu. For more on the rite's early history, see Kudo (2005, 180).

23. Three of the most significant myths, for our purposes, are Kojiki (1, 34 and 50-66; and 2, 242-44). The first includes a description of pyro-scapulimantic consultations (futomani ??? ?) performed by the heavenly kami on behalf of Izanagi and Izanami (a Nihon shoki version of this account is introduced above). The second includes a description of pyro-scapulimantic consultations performed by the kami of heaven to help lure Amaterasu ?? out of the heavenly stone chamber (ame no iwaya ???). The third myth describes the Great Purification Sacrifice (Oharae sai ???) organized by Empress Jingu ?? (trad. r. 209-269) after the death of Emperor Chuai ?? (trad. r. 192-200). The Shinsen kisoki faithfully retells these myths, but it adds information on the urabe-managed Fire Pacification Sacrifice (kachin sai ???) to the second mythical account (lines 428-429), and adds information on the Great Purification Sacrifice and related sacrificial and divinatory rites to the third (lines 477-485 and 501-502). The Pacification of Fire Sacrifice (chinka sai ???), according to commentary in the mid-ninth-century Ryo no gige (2, 77-78), was held in all four corners of the palace precincts in the sixth and twelfth lunar months, was managed by "the urabe and others" (???), and involved the burial of old fires and the lighting of new ones for purification and protection. Later descriptions of special Urabe techniques for gathering kindling, lighting and extinguishing fire, and invoking spirits to help control fire can be found in Tsugita (1927, 334). The Shinsen kisoki introduces the Great Purification Sacrifice along with part of a norito used in the rite. It also comments on related rites like the offering of silk to Ise shrines and the divinatory determination of abstinent imperial princesses confined to Ise. The Inbe clan is said to have been in charge of the former; the Nakatomi the latter. For more on the history of the Great Purification Sacrifice, see Kudo (2005, 276-91); Bock (1970-1972, 2: 84-87); and Ooms (2009, 181-82).

24. This historicization is examined in more detail below, but note that lines 587-598 of the Shinsen kisoki draw from the preface of the Kojiki for a concise history of the transmission of texts in ancient Japan (including texts on plastromancy). The Shinsen kisoki comments on early attempts to chronicle the history of the royal clan, a dependency on oral transmissions in these efforts, and the eventual compilation of a written record entitled Kojiki in 711 ce (712 according to the Kojiki). It explains that dates begin to be noted in the Kojiki during the reign of Emperor Kinmei ?? (r. 539-571), and that "the Buddha's dharma" (buppo ??) was first presented to the throne from Paekche ?? during Jinmei's reign. It also states that a monk from Paekche named Kanroku ?? (fl. 602-614) presented calendrical, astrological (specifically astro-calendrical tonko ?? [Ch. dunjia] divination-table methods), medical, and other technical texts to the throne during the reign of Emperor Suiko ?? (r. 593-628). Finally, the Shinsen kisoki argues that Japan began to produce its own texts in earnest from the mid-seventh century, most of them based on oral transmissions. "Pyro-plastromancy"-according to a concluding statement-"was also recovered in this manner" (??????).

25. The practical nature of technical portions of the Shinsen kisoki (final third of fascicle 1 and fascicles 2-4) would have made these materials especially attractive, but especially susceptible to loss. Historian Liu Lexian (2002, 13-14) argues that both innovation and continuity, both "love of the new and satiety with the old" (xixin yanjiu ????), are essential for the long-term perpetuation of mantic techniques. Mythical foundations for mantic arts tend to remain relatively resilient through time, while technical aspects of practice demand constant reformulation to keep up with the times. The textual history of the Shinsen kisoki well reflects this combination of mythical continuity and technical change.

26. For an article linking the Fujiwara, Nakatomi, and Urabe clans to Tsushima and to one another, see Umehara (1990).

27. For early accounts of the accomplishments of Hiramaro, see Nihon sandai jitsuroku (9: 39, 507); and Dai Nihon shi (8: 226, 161). Both works describe Hiramaro as a native of Izu skilled in pyro-plastromancy who visited Tang China in 838 before an illustrious career in court. They also claim that he received the surname Urabe from the throne in the mid-ninth century. Hiramaro is traditionally remembered as the patron of the Urabe clan and as an exemplar in the history of Yoshida Shinto.

28. Iki Island is just a few kilometers off the northwest coast of Kyushu. Tsushima Island is situated almost a hundred kilometers north-northwest of Iki, in the middle of the Korea Strait.

29. Kudo (2003, 144) points out a number of similarities between the preface and body of the text to stipulate that the former was either part of the original Shinsen kisoki or was added to it sometime before or in a 973 ce edition attributed to Urabe Masanobu. For similar views, see Tsubaki (1957, 86) and Kudo (2005, 177-78).

30. Izu refers to the Izu Peninsula and nearby islands.

31. The apical ancestral kami of the Nakatomi clan is Ame no koyane ????. The Fujiwara clan is traditionally traced back Nakatomi no Kamatari ???? (614-669), who was officially awarded the surname Fujiwara by Emperor Tenji ?? (r. 661-671). See Kojiki (1, 116); Nihon shoki (1: 1, 82 and 1, 128); and Ryo no gige (2, 78, and 79). The Fujiwara dominated late first and early second-millennium ce Japanese politics by intermarrying with the royal clan and by serving as regents to the throne. For more on the early history of the clan, see Kitagawa (1987, 98-116); Nomura (1995); and Takashima (1999).

32. Here, Grapard equates the term Urabe?? with Urabe ??. He is referring to a Heizeikyu mokkan ????? (Heijo Palace wooden tablet) dated 746 that mentions two different branches of the Urabe ?? of Izu. This tablet is introduced and quoted in Inoue (1980, 163).

33. See Ryo no gige (1, 29-30), that simply states that twenty urabe served in the Jingikan. The same number of urabe is listed in lines 581-584 of the Shinsen kisoki, which state that five members from each of the four Urabe provinces served the throne in the capital. Comments on the Yoro ritsuryo are recorded in both the court-sponsored Ryo no gige of 833 and the privately compiled Ryo no shuge (Collected expositions on the civil codes) of the mid-to-late ninth century.

34. For example, the term bokusha?? (diviner) occurs in the 720 Nihon shoki (2: 20, 490), the 721 Hitachi no kuni fudoki (Regional gazetteer of Hitachi Province) (364), and the early ninth century Nihon ryoiki (Record of numinous anomalies in Japan) (2: 5, 231).

Another term commonly read as Urabe is Ura no uji?? (Ura clan). It occurs twice in the following Hitachi no kuni fudoki (392) passage translated by Grapard: "On the tenth day of the fourth moon of each year a ritual feast is held and rice-wine is served. Members of the Urabe sacerdotal lineage [??] assemble men and women, and day after day, night after night, people deport themselves in drinks, songs, and dances.... The Urabe [??] dwell in the immediate surroundings of the shrine" (Grapard 1992a, 38-39). The term?? is not??, but there are reasons to associate the former with the latter. An earlier Hitachi no kuni fudoki entry, for example, records a tale about Ikatsu omi ???, apical ancestor of the Urabe clan. This account portrays Ikatsu omi as the son of the Great Kami of Kashima ????. Also known as Atomimi no mikoto ??? or Oshimi no sugune no mikoto ?????/????, Ikatsu omi is said-in certain traditions-to have received the surname Urabe because of his or her expertise in plastromancy. The proximity of these accounts strongly suggests a connection between?? and??, yet they remain distinct terms.

Finally, the term urabe ?? appears on the 746 Heijo Palace wooden tablet introduced in note 32. It is also frequently read as a variant of??, but both terms surface in close proximity in later state histories like the Shoku Nihongi, the Nihon Montoku tenno jitsuroku, and the Nihon sandai jitsuroku.

35. For the Japanese, see Man'yoshu (4: 16, 28-30 [#3811]). For English translations, see Nihon Gakujutsu Shinkokai (1940, 110-11); Bates (2005, 37-38).

36. However, knowing that the amnesty mentioned here was declared in preparation for the unveiling of the Great Buddha ?? of Todai-ji ??? in 751, these mid-level crimes could very well have involved criticisms of the project, or of Emperor Shomu's ?? (r. 724-749) lavish support of Buddhism during this era.

37. According to the in-text commentary, this refers to a set of shrines located 1. in the Iki district ??? of Iki island; 2. in the Soekami district ??? (in Nara) of Yamato province; and 3. in the Kamiagata ?? and Shimoagata ?? districts of Tsushima Island. The early tenth century Engi shiki (18: 9, 181 and 10, 320) only records the Soekami and Shimoagata shrines in its lists of state shrines.

38. Commentary explains that this is where Kushimachi ???, manifest as Hahakaki no kami ???? (kami of the wood used for pyromantic pokers), was worshipped. It also claims that Kushimachi shrines were located on Iki island, and that Kushimachi was worshipped at Ame no Kaguyama shrine ???? in Yamato province's Toochi district ??? (in Nara). The Ame no Kaguyama shrine is listed as a state shrine in the Engi shiki (18: 9, 181).

39. In particular, at the central shrine in Yamato province's Heguri district ??? (Nara) where, as the commentary explains, hikiribi no kami ??? (kami of kindling) or hikirigi no kami ??? ? (kami of firewood) was worshipped. This particular Koba Shrine must refer to Ikomaniimasu Ikomatsuhiko Shrine ??????????? (better known as Ikoma Shrine ????, Nara), where Ikomatsu hiko no kami ??????? and Ikomatsu himeko no kami ???????, both known as kami of fire and kindling, are still worshipped. Exactly how they relate to Kagutsuchi ???/???? is too complex a question to answer here, but the Kojiki (1, 40) refers to Kagutsuchi as "Hinokaga biko no kami" ??????, a designation that shares much in common with the name Ikomatsu hiko no kami. Finally, although we are told that the central Koba shrine was located in Yamato's Heguri district, it likely originated elsewhere. Given the origins of the other shrines introduced here, and what we have already identified as traditional associations between fire and the Urabe (see note 23), the most probable points of origin are Tsushima, Iki, or Izu.

40. Lines 573-578 of the Shinsen kisoki introduce the divine ancestors of Iki and Tsushima. Both are said to have assisted Ame no koyane in ancient pyro-plastromantic consultations. The supreme ancestor of the Urabe clan of Iki is identified as Amahitotsu hashira no mikoto ??????, and the supreme ancestor of Tsushima's Atai clan is identified as Chinsen no mikoto ???. The Kojiki account (1, 36) of the birth or creation of major islands in the Japanese archipelago (by Izanami and Izanagi) lists the same Amahitotsu hashira no mikoto ????? as the kami of Iki Island, but it names Ame no sadeyori hime ??????? rather than Chinsen no mikoto as the kami of the island of Tsushima. Neither of the latter two kami is, as far as I can tell, connected to divination, Ame no koyane, the Urabe, or the Atai in the Kojiki, and neither is mentioned elsewhere in the Shinsen kisoki. In the Shinsen kisoki, the term "four provinces" (yokuni ??) refers to Upper and Lower Tsushima, Iki, and Izu (versus the more commonly recognized set of Tsushima, Iki, Izu, and the capital). For more on the history of the four Urabe provinces, see Kawabata (1988).

41. The Shinsen kisoki (lines 585-587) explains that Tsushima is traditionally divided into two provinces: a northern or "upper" province named Kamiagata, and a southern or "lower" one named Shimoagata.

42. Hahaka ... is an ancient name for the Japanese bird cherry (uwamizu zakura ???) tree. The wood of the tree is hard, and is often identified as "metal-hard cherry" (kongo zakura ???). The wood was commonly used for tool handles and for pyro-osteomantic pokers used to scorch and crack bones.

43. The earlier Nihon shoki (1: 1, 50) alleges that Ikazuchi ? was born from a piece of the fire kami Kagutsuchi ..., who Izanagi cut up with his sword after its birth caused Izanami's death. Ikazuchi no mikoto Shrine ???? is listed in the Engi shiki (18: 10, 320) as one of thirteen major state shrines in Lower Tsushima.

44. These lineage claims are, as John McRae has brilliantly pointed out with regard to traditional Chan Buddhist lineages, likely "just as wrong as they are strong" (McRae 2003, xix).

45. The Shinsen kisoki reports that it was Emperor Chuai who first put Ikatsu omi in charge of the administration of divine affairs. Ikatsu omi subsequently declared that everyone with the surname Urabe in the eastern provinces were his descendants and put the Urabe of Izu in charge of divination. The Sonpi bunmyaku (58: 1, 23) and Matsunoo shake keizu (181, 238a) both explain that Ikatsu omi no mikoto ???? mastered the art of plastromancy and was granted the surname Urabe by the court of Emperor Chuai. The Matsunoo shake keizu not only introduces Ikatsu omi as a descendent of the same Ame no koyane worshipped by the Nakatomi, Oonakatomi ???, Urabe, Yiki ??, and Fujiwara clans, but it also clearly states that the Urabe lineage originated when his original surname, Nakatomi, was changed to Urabe. For additional comments on this passage, see Yokota (1982, 237-38). For more on the Sonpi bunmyaku passage, see Inoue (1980, 137); Yokota (1982, 237); and Grapard (1992b, 31). All three texts posit associations between Emperor Chuai, Ikatsu omi, Ame no koyane, the Nakatomi, and the Urabe. Connections between Ikatsu omi and the Nakatomi clan were established in or before the 815 Shinsen shojiroku ????? (Newly compiled register of surnames and clans); see Saeki (1962, 2, 212; 231; 259; and 334).

46. Yokota argues that the official duties of prominent sacerdotal clans in the Jingikan become far more specific in the transition from the Kojiki to the Nihon shoki. The Kojiki depicts Ame no koyane and Futodama jointly engaged in pyro-scapulimantic consultations and incantations used to help lure Amaterasu out of the cave, but the closest corresponding account in the Nihon shoki (1: 1, 82) records that Ame no koyane oversaw divine incantations, while Futodama-ancestor of the Inbe-controlled the presentation of offerings. By the end of the eighth century, the Nakatomi had begun to claim both incantation and divination as their own exclusive duties. This becomes clear in early ninth century texts like the Kogoshui, the Shinsen kisoki, and the Engi shiki, all of which are referenced in Yokota's lengthy discussion on how the Inbe lost-to the Nakatomi-the authority to recite state liturgy (that is, perform incantations) during the late Nara and early Heian periods (Yokota 1982, 260).

The Kojiki claims that both Ame no koyane (ancestor of the Nakatomi no Muraji ???) and Futodama ??? (ancestor of the Inbe no Obito ???) were charged with extracting the shoulder blade of a "true male deer" (mao no shika ???) from Heavenly Mount Kagu. They were also charged with selecting hahaka wood from the same mountain for scapulimantic consultations on how to lure her out. Lines 462-463 of the Shinsen kisoki draw from the Kojiki to state that, "Ame no koyane reached in, pulled on, and extracted the scapulae of a true male deer of Heavenly Mount Kagu and selected bark from the hahaka [trees] of Heavenly Mount Kagu. Ame no koyane then divined a fire-produced [scapulimantic] crack omen." Note that Futodama is absent from this retelling of the myth. Shinsen kisoki, lines 463-475, continue with an account of enticing Amaterasu out of the cave. The last few lines tell us that when she exited, "The multitude of deities were joyful, and having witnessed the emergence of divinatory cracking (scapulimancy), it was repeatedly done in this manner."

47. At least two other studies on pyro-plastromancy in Tsushima explore similar hypotheses. See Iwasa and Harada (1992, 182-201); Fujii (2008).

48. The connection that Nagadome draws between pyro-plastromancy and the transmission of Chinese cosmological and correlative systems like yin-yang and the five phases is not explained in detail, but the technique was certainly connected to these systems in China by the early medieval period. For more on these connections, see Kory (2012, 148-68).

49. Nagadome contends that the Shitaru plastron from Tsushima predates the three Maguchi Cavern shells, which are usually recognized as the earliest known examples of divined shells in Japan. The Shitaru shell, claims Nagadome, must predate the Kofun period because it was discovered next to ceramic objects dating to the mid-Yayoi. While the Shitaru shell might be slightly older than the Maguchi Cavern shells, I remain unconvinced of Nagadome's radical re-dating of the former, though it does better reflect the main direction of transmission in early Japan (sw to ne).

50. Divined turtle shells have not yet been discovered on the Korean peninsula, but there is a substantial archive of mid-first millennium ce textual evidence claiming that practice was performed there. For example, see Lee (1981, 67-70). For more on the early transmission of plastromancy from China to Korea to Japan, see Mishina (1970, 114, note 63); Obayashi (1977, 54); and Nitta (1977). Nitta combines textual and archaeological evidence with folklore and ethnographic studies to conclude that pyro-scapulimancy was a Tungustic practice transmitted to Japan during the late Yayoi or early Kofun period (route of transmission unknown), while pyro-plastromancy was a Chinese practice transmitted via the Korean peninsula during the late Kofun. Similar conclusions are defended in Kanaseki (1982, 69-93, especially 84-87); Kanzawa (1976, especially 24; and 1987).

51. Nagadome (1984, 92) asserts that Tsushima and Iki remained important during the Heian, and that pyro-plastromantic consultations are still being conducted there. To support the former claim, Nagadome points out that almost half (53/107) of the imperial shrines listed under "Western maritime [regions]" ?? in the Engi shiki (18: 10, 313-320) are located in Tsushima and Iki. He also reminds us that Ban Nobutomo's ??? (1773-1846) mid-nineteenth-century Seiboku ko (On correct divination) opens with a section on "correct" or orthodox forms of pyroplastromantic consultations attributed to the Urabe of Tsushima. See Seiboku ko (1, 447a-448a). Perhaps the most radical claim in Nagadome's article is that the royal Yamato Sun line and its imperial cult are at least partially based on the early solar and lunar cults of Tsushima and Iki. Nagadome (1984, 98-99 and 102) theorizes that prominent deities of sun and moon, and a prevalence of spirit mediums devoted to spells and dependent on these natural and divine phenomena in Tsushima and Iki provided a model for the creation and legitimization of the imperial sun line of Amaterasu. Much of the evidence used to support this hypothesis is drawn from relatively late texts like the Sonpi bunmyaku and Matsunoo shake keizu. For a study that focuses on earlier works to help untangle the processes leading to the creation of the imperial sun line of Amaterasu, see Kirkland (1997). Connections to Tsushima are briefly mentioned on page 149, note 70.

52. Like Yokota, Nagadome identifies Ikatsu omi as the apical ancestor of the Urabe, and largely relies on post-Heian texts to do so. Unlike Yokota, he emphasizes post-Heian textual traditions depicting Ikatsu omi as a prominent invocator (saniwa ???) and female spiritmedium (miko ??) once in the service of Empress Jingu.

53. See Kudo (2005, 295-98). The late thirteenth century Shaku Nihongi also identifies Futonorito as the daughter of Ame no anmochi and manifest in turtle form as Kamezu hime. For more on all of these kami and their associations with the Urabe, see Nakayama (1930, 143-52). The whole Kisei synopsis recorded in the Shinsen kisoki perfectly matches a long fragment from a Kicho den ??? (Traditions of pyro-plastromantic omens) cited in the Shaku Nihongi (5, 112- 113). Immediately preceding this fragment is a brief passage asking "What is ura ??" After telling us to equate it with ura?, we are told that it referred to the cracking of deer scapulae in ancient times, but later came to more commonly refer to the cracking of turtle plastrons. The origins of this innovation, says the passage, remain unclear, but the Kicho den fragment (perhaps originally drawn from the Kisei) is offered as a relatively early and illustrative record of this change.

54. The Fujiwara clan, for example, was awarded uji status in the middle of the seventh century as a branch of the Nakatomi; see note 31. The Taiho ritsuryo ???? (Regulatory and administrative codes of the Taiho reign period [701-704]) of 701 is purported to have established the Nakatomi as supervisors of the Jingikan, and the Yoro Codes is said to have listed positions for bokucho?? (diviners of crack omens) and "twenty urabe" serving under the Nakatomi as subordinate specialists; see note 33. The Taiho Codes can be partially reconstructed from a commentary entitled Koki ?? (Ancient records), which is quoted throughout the Ryo no shuge's commentary on the Yoro Codes, but neither set of codes is fully extant.

55. Nagadome (1984, 113) submits a much earlier date for the start of pyro-plastromantic dominance at court. He clearly illustrates a nearly half-century gap in Nihon shoki passages referring to divinatory prognostication (bokusen??) between the reigns of Ingyo ?? (r. ca. 412- ca. 453) and Keitai ?? (r. ca. 507- ca. 531), and argues that a radical change in technique must have taken place during this time. However, neither the absence of terms like? or ?, nor minor changes in divinatory topics over the course of this period convincingly demonstrates that a major transition from eastern scapulimantic prognostication to western plastromantic divination transpired between the mid-fifth and early sixth centuries. The present study uses a combination of received and recovered evidence to date this shift between the mid-Nara and early Heian periods.

56. Later texts, like 927 ce Engi shiki, continue to afford significant roles to the Nakatomi and the Urabe in the Jingikan. Both clans are also given prominent places in developing branchbureaus outside of the department but still within the chimeric imperial cult complex. For example, see Bock (1970-1972, 1: 20, 116 note 359, and 151-185; 2: 9 note 15, 74 note 341, 86 note 385, and 89 note 410).

57. As a whole, this section of the Shinsen kisoki promotes a transition from bones as a medium for the spirits of heaven, to shells as a medium for the spirits of both heaven and earth. The term sei/ukehi ? refers to "a pledge" or "vow" (often glossed as ukehi/inori ?). For more on the term in a religious context, see Nakayama (1930, 139-40). Kudo (2005, 300-301) points out that the first few lines (536-539) of the Kisei closely match a number of "Nakatomi no yogoto" ???? (Words for the longevity of the Nakatomi clan; for example, norito for the Great Purification Sacrifice) recorded in book eight of the Engi shiki. Later pyro-plastromantic treatises, like the Kiboku shidai, use this whole "explanation" as "sacrificial liturgy" (saimon ??).

58. The hide of a "truly famed deer" or "true stag" is mentioned in one of the Nihon shoki accounts of luring Amaterasu out of the cave (1: 1, 80). The translation in Aston (1956, 47), reads, "[Ishi kori dome] stripped off in one piece the hide of a true stag, and made of it Heavenly bellows" ([???]? ?? ?? ? ?,??? ??).

59. The two walls of bone flanking the sides of a turtle shell, between the fore and hind limbs, which connect the lower shell or plastron to the upper shell or carapace, are known as bridges or dividers.

60. Section titles 23 and 24 for Shinsen kisoki fascicle 1 specify that shells were divided into at least two plates (han/ita ?) for pyro-plastromantic consultations. The Kiso bubun (lines 608- 611) provides a more detailed description of this process. Plastrons were to be divided into two sheets or plates (mai ?), which were each sliced into six strips (suji ?). These strips are said to resemble "tallies" or "insignia" (kei/tama ?), suggesting that they shared the same "?"-shape commonly used for Heian court tallies. The Engi shiki (18: 1, 24) records that four knives and chisels were needed for pyro-plastromantic consultations, further supporting the view that shells were cut and sliced into sheets and strips carved or chiseled with machi during the early to midninth century. While later texts, like Ban Nobutomo's Seiboku ko, stipulate that plastrons were to be sawn into pentagonal planks, this practice is never brought up in the Shinsen kisoki. Frequent mention of the slicing of shells and use of the machi in the Kiso bubun, however, strongly imply that tally-like strips of plastron were used by the mid-Heian period.

61. The phrase kore o harau ?? (sweep it) can be understood in a number of different ways. It might refer to the cleaning or ritual purification of the shell or a site for pyro-plastromantic consultations, it might describe the act of running a poker along carved machi lines, or it might allude to the act of writing with a brush.

62. For more on Liu Shilong, see Kory (2012, 149, note 83). For more on extant fragments attributed to Liu Shilong's Guijing, see Kory (2012, 148-53 and 454-60).

63. For instance, the Kiso bubun states that, "there are twenty-nine line graphs for Earth, the same for Heaven" (?????????) (line 629). All twenty-nine are then named. Line 639 states that "there are thirty-eight line graphs for Spirit and the same for Human" (?????? ???). The names of all seventy-six of these omens follow. Finally, line 647 claims that, "there are three line graphs for Omen" (????). They are: 1. subordinate crack (jiboku ??) ... Spirit above, Human below (?? ??); 2. even crack (soboku ??) ... Earth to Heaven, and Spirit to Human, like the written character "?" (????????); and 3. superior crack (shoboku ??) ... Spirit below, Human above (??? ?).

64. For explicit mention of this total, see Kiso bubun line 629 and Kiboku sho line 12. The total of 137 includes 29 Earth, 29 Heaven, 38 Divine, 38 Human, and 3 Omen [line-graphs]. For additional comments on these lists, see Kudo (2005, 179-80). Modern reconstructions of traditional Japanese plastromantic interpretive methods tend to rely on Ban Nobutomo's Seibokuko. See, for example, Blacker (1981, 68-69); and Nakajima (2002-2006).

65. The "head-body-foot" division of pyro-plastromantic cracks is employed in texts like the early to mid-first-century bce Shiji (Records of the grand scribe-astrologer), "Guice liezhuan" ???? (Arrayed traditions of turtles and stalks), the late fifth century Guijing, and the mideighth-century Taibai yinjing ???? (Secret classic of Grand White [that is, Venus]) or Shenji zhidi Taibai yinjing ???????? (Secret classic of Grand White, divine trigger and controller of enemies); see Kory (2012, 121-91). Terms in the Kiso bubun (lines 630-647) commonly used to describe major crack lines include tada ? (straight), kai ? (curling back), shin ? (rising up), kei ? (extending out), tan ? (broken), setsu ? (cut off ), and ki ? (ascending).

66. See Kiso bubun (lines 608-610), and note 60. The backside (that is, flesh side) or base (moto ?) of each strip is for scorching; the green topside (sue ?) or outer surface is for reading. Lines 611-617 record names for each of the six strips produced from each half of the plastron.

67. For a mid-to-late-ninth-century description of how the machi was to be applied to the shell, see Jogan gishiki (3, 218).

68. Lines 820-825 of the Saigi kankei kiji explain that a total of 160 fires (80 for auspicious pyro-plastromantic inquiries; 80 for inauspicious inquires) were used before the end of the Enryaku ?? era (782-806). However, in the first year of the Daido ?? era (806-810), the total number of fires was decreased to 48 (24/24). The next ten lines provide details concerning the diviners (two urabe officials from each of the four Urabe provinces), liturgists (Nakatomi officials), and palace representatives expected to participate in pyro-plastromantic consultations concerning dates for the Divine Offering of Foods Rite. Lines 834-854 focus on the roles of Nakatomi representatives in the consultations, along with ritual preparations. They also record the names of the spirits to be summoned, the invocations used to summon them, and some of the other persons and objects used in the rite. Lines 855-868 of the Saigi kankei kiji describe how a total of 48 fires were used to produce 48 cracks by four urabe pyro-plastromancers. Four plates (mai ?) of plastron were to be selected by the superior supervisors of cracking and the "Invocatory Words for the Divinatory Cracking of Scapulae" (bokukata kitsushi????) was to be recited. After four different diviners cracked each plate (or plank), the process was repeated twelve times by each diviner until all 48 decisions could be tabulated to determine a date (the number 48 suggests that this particular description dates to the Daido era [806-810] when the same number of fires are said to have been used). The final decision was to be offered to the throne conjointly by the Urabe diviners and the Nakatomi liturgists. The former was responsible for recording which of the five basic types of line-graphs or crack omens (Earth, Heaven, Spirit, Human, Omen) had been obtained in each consultation. The last three lines of the Saigi kankei kiji (lines 869-871) record the number of turtle shells presented to the throne from different provinces for the Divine Offering of Foods Rite: 17 from Kii no kuni ??? (present-day Wakayama and Mie prefectures), 13 from Awa ?? (present-day Tokushima prefecture), and 10 from Tosa ?? (present-day Kochi prefecture). The total of 40 plastrons (each divided into twelve strips) would have been more than sufficient to conduct 48 consultations.

69. Line 606 of the Kiso bubun explains that one must fast, bathe, be clear and focused, and be without deviant thoughts to conduct an efficacious and accurate pyro-plastromantic consultation.

70. Lines 606-608 of the Kiso bubun state that two diviners were to be seated for royal offerings at dawn. Each was to take hold of the two halves of the plastron and recite a pledge to them.

71. Commentary in line 836 of the Saigi kankei kiji claims that Futonorito (shell) and Kushimachi (hahaka-wood poker) must be invoked for pyro-plastromantic consultations. The same statement is found in Engi shiki (18: 1, 23). Lines 567-584 of the Shinsen kisoki (translated above) associate both Futonorito and Kushimachi with a shrine-based Urabe-managed pyro-plastromantic cult centered in Tsushima and Iki, but it is important to point out that Futonorito appears to have been the primary kami summoned in the rite. Kiso bubun lines 606-608, for instance, explain that while Amatsu norito no Futonorito had to be invoked, the summoning of Kushimachi was optional.

72. Lines 615-617 of the Kiso bubun record detailed directions for how to scorch a machi and when to douse water on the shell. First, the poker is to be rubbed along the vertical axis of the machi (bottom to top). Water is then doused on the plastron before the poker is rubbed along each of the horizontal lines of the machi (first east then west). Finally, water is once again used to douse the shell. This process is to be repeated until a crack forms. Lines 619-620 of the Kiso bubun claim that in ancient times, major crack lines were divided into two basic types (up/ down ??), but later, use of the machigata gave rise to five different categories of line graphs or crack omens.

73. A basic list of directions, seasons, colors, and a number of other associations for cracks produced from each of the five machi lines is recorded in lines 618-623 of the Kiso bubun. Lines 623-628 describe "the principles of the five branches of crack omens."

74. Some of the functions listed in the Seibokuko, for example, include the selection of places, dates, offerings, and agents to be used in rites; the identification of curses; and the interpretation of natural anomalies. For general comments on premodern Japanese pyro-plastromantic topics and functions, see Blacker (1981, 66-71).

75. Pyro-plastromancy is still practiced in Japan under imperial auspices and in local shrines (particularly Yoshida Shinto shrines). Consultations were reportedly carried out for Emperor Hirohito's ?? (r. 1926-1989) formal enthronement ceremony in February of 1928, and most recently for Emperor Akihito's ?? (r. 1989-) enthronement in November of 1990. For informed descriptions of these ceremonies see, respectively, Holtom (1972, 75-78); and Demura (1991, 75). For more on the divinatory cracking of turtle shells in contemporary Japanese shrines, see Fujino (1974) and Nitta (1977).

76. See Teeuwen (2002, 257; 2007, 386-900. Both articles provide excellent synopses and comparisons of contemporary views on the history of the development of Shinto. In the former, Teeuwen argues that, "Together, kami worship, the jingi system, and Shinto form a triangle consisting of three closely related but at the same time very different sides, each with its own historical development. If we fail to distinguish between them, we lose sight of the dynamics that formed all three of them" (2002, 259). In the latter, he contends that, "the jingi cult offered a concrete model for the institutionalization of an autonomous Shinto" (2007, 399).

77. The Urabe managers of the Futonorito cult outlined in the Shinsen kisoki and Ryo no gige may have been cracking turtle shells and offering cult to kami associated with the technique before the official enactment of the ritsuryo system, before the official arrival of Buddhism, and even before the consolidation of the Yamato empire. Textual support for these claims, however, is either lacking or late. According to the early tenth century Engi shiki (18: 9, 181 and 187), a shine-based cult in Tsushima devoted to Futonorito was replicated in Yamato Province during the Asuka, in Nara-kyo during the Nara, and in Heian-kyo during the Heian period. Both Inoue (1980, 129 and 196) and Grapard (1992b, 32) cite this information as evidence of the appropriation of peripheral kami cults into the early imperial Japanese theocracy.

78. Like the Shinsen kisoki, the Ryo no gige clearly identifies pyro-plastromancy as the preferred form of state divination, and it is hard to believe that the technique would not have been established for some time before these works were presented to the throne. Moreover, the Shinsen kisoki contains the earliest evidence of the use of the machi-form, and it is doubtful that a campaign to fabricate a local tradition from scratch-for incorporation into the theocratic cult-would have used anything but Chinese versions of the technique.

References

references

abbreviations

KT Kokushi taikei ????. 60 vols. Tokyo: Yoshikawa Kobunkan, 1962-1967.

SNKBT Shin Nihon koten bungaku taikei ..... 100 vols. Tokyo: Iwanami Shoten, 1989-2005.

SSKSK Shinsen kisoki no kisoteki kenkyu: Kojiki ni ikyoshita saiko no kibokusho ... ed. Kudo Hiroshi ???. Tokyo: Nihon Edita Sukuru Shuppanbu, 2005.

primary sources

Dai Nihon shi ????. Comp. Tokugawa Mitsukuni ???? (1628-1700). 17 vols. Ibaragi-ken: Yoshitsune Seitan 300-nen Kinenkai, 1928.

Engi shiki ???. KT, vol. 26.

Hitachi no kuni fudoki ??????. In Fudoki ???, ed. Uegaki Setsuya ?? ??, 353-421. Tokyo: Shogakkan, 1997.

Jogan gishiki ????. Ed. Kada Arimaro ???? (1706-1751). Tokyo: Gendai Shichosha, 1980.

Kiboku shidai ????. SSKSK, 134-169.

Kiboku sho ???. SSKSK, 96-126 <lines 1-361>.

Kiso bubun ????. SSKSK, 80-82 <lines 606-648>.

Kogoshui ????. Comp. Inbe Hironari ???? (fl. 808), ed. Kato Genchi ?? ??. Tokyo: Iwanami Shoten, 1929.

Kojiki ???. Ed. Yamaguchi Yoshinori ... and Konoshi Takamitsu .... Tokyo: Shogakkan, 1997.

Man'yoshu ???. Ed. Satake Akihiro ????. 5 vols. Tokyo: Iwanami Shoten, 1999-2004.

Matsunoo shake keizu ??????. In Zoku gunsho ruiju ..., ed. Hanawa Hokinoichi ???- (1746-1821), vol. 7, 236-259. Tokyo: Keizai Zasshisha, 1911.

Mondo bubun ????. SSKSK, 86-95 <lines 649-819>.

Nihon koki ????. KT, vol. 3.

Nihon Montoku tenno jitsuroku ... Comp. Fujiwara Mototsune ???? (836-891). KT, vol. 3.

Nihon ryoiki ?????. By Keikai ?? (ca. 787- ca. 822). SNKBT, vol. 30.

Nihon sandai jutsuroku .... Comp. Fujiwara Tokihira ???? (871- 909). KT, vol. 4.

Nihon shoki ????. Ed. Kojima Noriyuki ???? et al., 3 vols. Tokyo: Shogakkan, 1994-1998.

Rikkokushi ???. Ed. Saeki Ariyoshi ????. 12 vols. Tokyo: Asahi Shinbunsha, 1940-1941.

Ritsuryo ??. Comp. Fujiwara no Fuhito ... (659-720) et al. In Nihon shiso taikei ??????, ed. Inoue Mitsusada ???? (1917-1983) et al., vol. 3. Tokyo: Iwanami Shoten, 1976.

Ryo no gige ???. Comp. Kiyohara Natsuno ???? (782-837) et al. KT, vol. 22.

Ryo no shuge ???. Comp. Koremune no Naomoto ???? (fl. 859-930). KT, vols. 23-24.

Saigi kankei kiji ??????. SSKSK, 83-85 <lines 820-871>.

Sanguo zhi ???. Comp. Chen Shou ?? (233-297), comment. Pei Songzhi ?? ? (372-451). 5 vols. Beijing: Zhonghua Shuju, 1959.

Seibokuko ???. Comp. Ban Nobutomo ??? (1773-1846). In Ban Nobutomo zenshu 5, vol. 2. Tokyo: Kokusho Kankokai, 1907-1909.

Sendai kuji hongi ??????. Comp. Urabe Kanenaga???? (1467-1536), ed. Yokota Ken'ichi ???-. 1521; Tenri: Tenri Daigaku Shuppanbu, 1978.

Shaku Nihongi ????. Comp. Urabe, Kanekata ???? (ca. 1274-1301). KT, vol. 8.

Shenji zhidi Taibai yinjing ... By Li Quan ?? (fl. mid-eighth century CE), ed. Qian Xizuo ??? (ca. 1801-1844). In Baibu congshu jicheng ??????, ed. Yan Yiping ?-? et al. Taibei: Yiwen yinshu guan, 1965- 1968.

Shiji ??. Comp. Sima Tan ??? (ca. 180-110 BCE) and Sima Qian ??? (145- 85 BCE), suppl. Chu Shaosun ??? (ca. 104- ca. 30 BCE). Beijing: Zhonghua shuju, 1959.

Shinsen kisoki .... Comp. Urabe Totsugu???? (fl. 830 CE) et al. SSKSK, 22-62 <lines 362-605>.

Shoku Nihongi ????. Ed. Kazuo Aoki ???? et al. 6 vols. Tokyo: Iwanami Shoten, 1989-2000.

Shoku Nihon koki ?????. KT, vol. 3.

Sonpi bunmyaku ????. KT, vols. 58-60.

Tang liudian ???. Comp. and comm., Li Linfu ??? (d. 752), ed. Chen Zhongfu ???. Beijing: Zhonghua shuju, 1992.

Tango no kuni fudoki ??????. In Fudoki ??? , ed. Uegaki Setsuya, 479- 483. Tokyo: Shogakkan, 1997.

secondary sources

Akimoto Yoshinori ????

1978 Shinsen kisoki no kenkyu: Honkoku no bu .... Seisen Joshi Daigaku kiyo ... 26: 111-64.

1992 Shinsen kisoki ?????. In Shinto taikei koten hen ???????, Murata Masashi ????, Akimoto Yoshinori, and Makabe Toshinobu ????, eds., vol. 13. Tokyo: Shinto Taikei Hensankai.

Aston, W. G., tr.

1956 Nihongi: Chronicles of Japan from the Earliest Times to A.D. 697. 2 vols. London: Allen & Unwin. (Originally published 1886)

Bai Bin ??

1998 Chongqing Yunyang Mingyueba yizhi chutu Tangdai bujia de chubu yanjiu .... Sichuan daxue xuebao ?????? (Zhexue shehui kexue ban ???????) 4: 106-10.

Bates, Evan, tr.

2005 Japanese Love Poems: Selections from the Man'yoshu. Mineola, NY: Dover Publications.

Bell, Catherine M.

1992 Ritual Theory, Ritual Practice. New York: Oxford University Press.

Bentley, John R.

2006 The Authenticity of Sendai Kuji Hongi: A New Examination of Texts, with a Translation and Commentary. Leiden: Brill.

Blacker, Carmen

1981 Japan. In Divination and Oracles, Michael Loewe and Carmen Blacker, eds., 63-86. Boston: George Allen & Unwin.

Bock, Felicia Gressitt, tr.

1970-1972 Engi-Shiki: Procedures of the Engi Era. 2 vols. Tokyo: Sophia University.

Demura Katsuaki ????

1991 Yoshida Shinto ni okeru kiboku kenkyu ni tsuite: Nobunata, Kanemigi fude "Kiboku den," Kanemigi "Kiboku shidai hei matsuri bun" o chushin toshite ????????????????-???????????, ??? ? ?? ? ? ? ? ? ?? ? ? ??? . Shinto shi kenkyu 39: 74-126.

Fujii Hiroaki ????

2008 Tsushima, Iki ni okeru umigame no minzoku: Kiboku no sato to umigame .... Minzoku bunka 20: 181-240.

Fujino Iwatomo ????

1974 Senboku ni kansuru ni-san no mondai: Waga kuni o chushin to shite .... In Uno Tetsuto sensei hakuju shukuga kinen Toyogaku ronso ?????????????????, ed. Uno Testuto ????, 1007-21. Tokyo: Uno Tetsuto Sensei Hakuju Shukuga Kinenkai.

Grapard, Allan G.

1992a The Protocol of the Gods: A Study of the Kasuga Cult in Japanese History. Berkeley, CA: University of California Press.

1992b The Shinto of Yoshida Kanetomo. Monumenta Nipponica 47: 27-58.

2002 Shrines registered in ancient Japanese law: Shinto or not? Japanese Journal of Religious Studies 29: 209-32.

2003 The source of oracular speech: Absence? Presence? or Plain treachery? The Case of Hachiman Usa-gu gotakusenshu. In Buddhas and Kami in Japan: Honji Suijaku as a Combinatory Paradigm, Mark Teeuwen and Fabio Rambelli, eds., 77-94. London: RoutledgeCurzon.

Hall, John

1966 Government and Local Power in Japan: 500 to 1700. Princeton, NJ: Princeton University Press.

Hirano Hiroyuki ????

1966 Tsushima, Iki Urabe ni tsuite: Sono seiritsu ki ni tsuite no ni, san no mondai .... Kodai bunka 17: 65-73.

Holtom, Daniel Clarence

1972 The Japanese Enthronement Ceremonies: With an Account of the Imperial Regalia. Tokyo: Sophia University.

Inoue Tatsuo ????

1980 Kodai oken to shukyoteki bumin .... Tokyo: Kashiwa Shobo.

Inoue Wataru ?? ?

2008 "Omima no miura" to Shinsen kisoki ... . Higashi ajia bunkakanryu ???????? 1: 40-64.

Iwasa Kyochi ???? and Harada Minoru ?? ?

1992 Tsushima "kiboku shinji" .... In Yamataikoku tettei ronso: Yamataikoku mondai o kiten to shite kodaishi toron shinpojiumu ..., ed. Furuta Takehiko ????, 182-220. Tokyo: Shinsensha.

Kanaseki Takeo ????

1982 Bokkotsu dangi????. In Koko to kodai-Hakkutsu kara suirisuru ..., ed. Kanaseki Takeo, 64-89. Tokyo: Hosei Daigaku Shuppankyoku (originally published in 1975).

Kanzawa Yuichi ???-

1976 Yayoi jidai Kofun jidai oyobi Nara jidai no bokkotsu bokko ni tsuite .... Sundai shigaku ???? 27: 1-25.

1987 Nihon no bokkotsu ?????. Kokogaku janaru 281: 4-9.

Kato, Genchi, and Hikoshiro Hoshino, trans.

1972 Kogoshui: Gleanings from Ancient Stories. London: Curzon Press.

Kawabata Moriyuki ????

1988 "Shinsen kisoki" to yokuni no Urabe... Shinto shi kenkyu ????? 36: 112-26.

1990 Fotonorito no mikoto ko: Ama no koyane no mikoto doshin setsu o megutte .... Shinto shi kenkyu 38: 21-36.

Kinoshita Fumitada ????

2000 "Shinsen kisoki" no kosei ni tsuite ... Kojiki nenpo 42: 144-57.

Kirkland, Russell

1997 The sun and the throne: The origins of the royal descent myth in ancient Japan. Numen 44: 109-52.

Kitagawa, Joseph M.

1987 On Understanding Japanese Religion. Princeton, NJ: Princeton University Press.

Kory, Stephan N.

2012 Cracking to Divine: Pyro-plastromancy as an Archetypal and Common Mantic and Religious Practice in Han and Medieval China. PhD dissertation, Indiana University.

Kudo Hiroshi ?? ?

1994 Shinsen kisoki to Urabe shi no densho keisei ... Jodai bungaku ???? 73: 67-77.

2005 Shinsen kisoki no kisoteki kenkyu: Kojiki ni ikyoshita saiko no kibokusho ... Tokyo: Nihon Edita Sukuru Shuppanbu.

Kudo Takashi ?? ?

2003 Shinsen kisoki ga kataru kodai ... Rekishi dokuhon ???? 48: 144-47.

Kurano Kenji ????

1959 Shinsen kisoki to Kojiki .... Nihon jokoshi kenkyu ??? ???? 27: 41-45.

Ledyard, Gari

1975 Galloping along with the horseriders: Looking for the founders of Japan. Journal of Japanese Studies 1: 217-54.

Lee Hyeong Koo ???

1981 Bohai yan'an zaoqi wuzi bugu zhi yanjiu (xia): Jianlun gudai Dongbeiya zhu minzu zhi bugu wenhua ... Gugong jikan ... 16: 55-81.

Liu Lexian ???

2002 Jianbo shushu wenxian tanlun ... Wuhan: Hubei Jiaoyu Chubanshe.

Liu Yujian ???

1992 Zhongguo gudai guibu wenhua .... Guangxi: Guangxi Shifan Daxue Chubanshe.

McRae, John R.

2003 Seeing Through Zen: Encounter, Transformation, and Genealogy in Chinese Chan Buddhism. Berkeley, CA: University of California Press.

Mishina Shoei ????

1970 Yamataikoku kenkyu soran ... Osaka: Sogensha.

Nagadome Hisae ????

1984 Urabe no seiritsu ni tsuite: Tsushima kara no shiten.... In Shinto shi ronso: Takigawa Masajiro Sensei beiju kinen ronbun shu ... ed. Takigawa Masajiro Sensei Beiju Kinen Ronbunshu Kankokai, 87-115. Tokyo: Kokusho Kankokai.

Nakajima, Hiroko

2002-2006 Kiboku. In Encyclopedia of Shinto <http://eos.kokugakuin.ac.jp/ modules/xwords/entry.php? entryID=1311> (accessed 1 December 2012).

Nakano Hatayoshi...

1975 Hachiman shinkoshi no kenkyu ... Tokyo: Yoshikawa Kobunkan.

Nakayama Taro ????

1930 Nihon miko shi ?????. Tokyo: Ookayama Shoten. (Reprinted by Kokusho Kankokai, 2012)

Nihon Gakujutsu Shinkokai, tr.

1940 The Manyoshu: One Thousand Poems Selected and Translated from the Japanese. Tokyo: Published for the Nippon Gakujutsu Shinkokai by Iwanami Shoten.

Nitta Eiji...

1977 Nihon shutsudo bokkotsu no shikaku ... Kodai bunka 29: 27-42.

Nomura Tadao...

1995 Naracho no seiji to Fujiwara shi ... Tokyo: Yoshikawa Kobunkan.

Obayashi Taryo ...

1977 Yamataikoku ... Tokyo: Chuo Koronsha.

Ooms, Herman

2009 Imperial Politics and Symbolics in Ancient Japan: The Tenmu Dynasty, 650- 800. Honolulu: University of Hawai'i Press.

Philippi, Donald L.

1959 Norito: A New Translation of the Ancient Japanese Ritual Prayers. Tokyo: Institute for Japanese Culture and Classics, Kokugakuin University.

Saeki Arikiyo ..., ed.

1962 Shinsen shojiroku no kenkyu (Honbun hen) ... Tokyo: Yoshikawa Kobunkan.

Saso Mamoru...

2006 Koko shiryo kara mita kodai no kiboku: Bokko to Urabe ... In Kiboku: Rekishi no chiso ni himerareta uranai no waza o horiokosu ..., ed. Higashi Ajia Kaii Gakkai ..., 103-33. Kyoto: Rinsen Shoten.

Shigematsu Akihisa...

1986 Hachiman Usagu gotakusenshu ... Tokyo: Gendai Shichosha.

Takashima Masato...

1999 Nara jidai no Fujiwara shi to chosei ... Tokyo: Yoshikawa Kobunkan.

Teeuwen, Mark

2002 From Jindo to Shinto: A concept takes shape. Japanese Journal of Religious Studies 29: 233-63.

2007 Comparing perspectives on the emergence of Jindo and Shinto. Bulletin of SOAS 70: 373-402.

Tsubaki Minoru...

1957 Shinsen kisoki (Todai hon) kaidai ... Manuscript.

Tsugita Uruu ...

1927 Norito shinko ... Tokyo: Meiji Shoin.

Umehara Takashi ...

1990 Nihon to wa nani ka (#95): Tenno, Kamatari no shutsuji: Tsushima yori kiboku o motte miyako e nobotta ?????? (95) ... Asahi Journal 32.46 (Nov. 16): 74-77.

Yokota Ken'ichi ...

1982 Nihon kodai shinwa to shizoku densho .... Tokyo: Hanawa Shobo.