Jump to ContentJump to Main Navigation
Tantric Traditions in Transmission and Translation$

David B. Gray and Ryan Richard Overbey

Print publication date: 2016

Print ISBN-13: 9780199763689

Published to Oxford Scholarship Online: April 2016

DOI: 10.1093/acprof:oso/9780199763689.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. Subscriber: null; date: 10 May 2021

Vicissitudes of Text and Rite in the Great Peahen Queen of Spells

Vicissitudes of Text and Rite in the Great Peahen Queen of Spells

(p.257) 6. Vicissitudes of Text and Rite in the Great Peahen Queen of Spells
Tantric Traditions in Transmission and Translation

Ryan Richard Overbey

Oxford University Press

Abstract and Keywords

The Great Peahen Queen of Spells (Mahāmāyūrīvidyārājñī) is a mercurial corpus of Buddhist narrative, incantation, and ritual practiced in South, Central, and East Asia from the early centuries CE to the present day. This chapter uses the six Chinese translations, dating from the fifth to the eighth centuries, to analyze the varieties of change in the grimoire. In the span of a few short centuries one finds a merging of recensions, a dramatic expansion in the pantheon, the incorporation of local Chinese spells, and increasingly intricate elaboration of ritual manuals. Close study of the many versions of the Great Peahen allows us to identify a gradual “esotericization” of its rituals, and to witness processes of textual development usually lost to history.

Keywords:   Mahāmāyūrīvidyārājñī, Great Peahen, pantheon, ritual, ritual manual, tantra


The Great Peahen Queen of Spells (Mahāmāyūrīvidyārājñī) is a dynamic corpus of scriptures, incantations, and rituals produced and practiced in South, Central, and East Asia from the early centuries CE to the present day.1,2 The text survives in many versions, including six Chinese translations, a Tibetan translation, in Sanskrit manuscripts, and as part of a later ritual collection called the Fivefold Protection (Pañcarakṣ‎ā).3 The earliest extant versions are Chinese translations dating to the late fourth century; the text coalesced into its final form by the early sixth century.

This large body of evidence serves as a resource not only for text-critical studies, but also for addressing questions in Buddhist intellectual and ritual history. By closely studying the transmission and transformations of the Great Peahen, one can learn more about the interaction of Buddhism with Chinese religions, about the rapid development and flexibility of Buddhist ritual practices, and about the appeal of Buddhist spells in much broader religious and cultural contexts.

On the Character and Classification of the Great Peahen

Characterizing a text like the Great Peahen is no easy task. Generally speaking, the versions of the Great Peahen contain a narrative frame and a series (p.258) of spells associated with gods, demons, and assorted numina. Many versions of the text also include ritual instructions as well as appended spells from other scriptures. The text has at least two early recensions, which later coalesced into a stable, expanded version (Sørensen 2006, 91–109). The text thus reads less like a standard Buddhist scripture and more like a grimoire. Sylvain Lévi quipped that the Great Peahen “ne doit son importance qu’à sa valeur magique; sa valeur littéraire est nulle” (1915, 25).4

Since the scripture is so difficult to characterize, I will begin with a translation of the simplest version, a short spell translated into Chinese in the late fourth century, entitled The Great Golden Peacock King Spell Scripture (T. 986):5

The Buddha told Ānanda, “Long ago, south of the king of the snowy mountains, there was a golden peacock king. The Buddha dwelled within him. Using this Great Peacock King Spell Scripture, he uttered it in the morning for self-defense, and the day would be peaceful. He uttered it at dusk for self-defense, and the night would be calm.” Then he uttered the spell, saying “Hu hu hu hu hu hu hu hu hu hu, le le le le dumba le le le vima le le huya huya vija vija thusa guru elā melā ili melā tili melā ili me tili ili tili mitte dumbe sudumbe tosu to golā velā capalā velā ca iṭṭ‎iri bhiṭṭ‎iri ṭ‎iri!6

“Ānanda, now I will utter for you the heart of the Queen of Spells of the Great Peacock King.” Then he uttered the spell, saying “ili kili . . . [&c.] 7 may the heavens quickly pour forth rain!

“Circumambulate the ritual area seven times and recite.” Then he uttered a spell, saying “nārayaṇ‎a pārayaṇ‎a . . . [&c.]

“Ānanda, this Great Peacock Queen of Spells was uttered according to Maitreya Bodhisattva.” Then he uttered a spell, saying “kili kili kili kili . . . [&c.]

“Ānanda, this Great Peacock Queen of Spells was uttered according to the god Brahmā, ruler of the Sahā World.” Then he uttered a spell, saying “hili hili hari . . . [&c.]

“Ānanda, this Great Peacock Queen of Spells was uttered according to Śākya, king of the gods.” Then he uttered a spell, saying “tala kṣ‎āntali . . . ili ha hi hu, may all calamity and evil submit! Seize their hands, feet, joints, and knuckles! May they never be able to move, even the Trayastriṃ‎śa gods! . . . [&c.]

“Ānanda, you should learn the names of the divine seers of yore.8 Now I will utter the names of those who have accomplished (p.259) the end of suffering, who have accomplished the observance of vows, who have accomplished the spell arts, who always practice ascetic practices, who dwell in the mountains and forests, who have potency and sovereignty, whose transformations are formidable, who have the five supernormal knowledges and the powers of accomplishing their wishes, who fly through the sky. They are called: Great Seer Aṣṭ‎amaka, Great Seer Vāmaka, Great Seer Vāmadeva . . . [Here follows a list of Greet Seers; the entire list contains 51 seers in total.]”

“Ānanda, these are the great seers of yore, who created the Four Vedas. Whatever they desire is entirely accomplished through the great power of their ascetic practices and glorious energy. May they also use this Great Peacock King Spell Scripture to protect so-and-so, and cause him to live for one hundred years, to get to see one hundred autumns!” Then he uttered a spell, saying “hari hari hari . . . [&c.]

“May the Four Kings of the Gods and the Great Demon Spirit Kings also use the Peacock Queen of Spells to protect so-and-so, and cause him to live for one hundred years, to get to see one hundred autumns!” Then he uttered a spell, saying “akaṭ‎e vikaṭ‎e . . . [&c.] Protect so-and-so, and cause him to live one hundred years, to get to see one hundred autumns!

“The Maṇ‎iratna Scripture and the *Mahāgandhabhikṣ‎u 9 rescue from sickness, suffering, and calamity. The Peacock King Spell eliminates enemies.”

Even with its many layers of later additions and emendations, the Great Peahen does not fundamentally deviate from this style of writing. It is a pragmatic text, a whirlwind tour of gods and demons and their associated spells and rituals. The text’s relative lack of narrative coherence, its jarring leaps from one spell to the next, make the process of expansion relatively easy. Adding another set of deities, spells, or ritual procedures would not radically change the nature of the scripture.

The Great Peahen has been called an “esoteric” or “prototantric” text.10 Two factors contribute to this classification. First, the text’s many versions contain numerous incantations and rituals, which remind scholars of the maṇḍ‎alas, mantras, and mudrās of later tantric scriptures.11 Second, scholars have shown that the Great Peahen was an important influence on other early Buddhist texts commonly regarded as prototantric.12

(p.260) In recent years some scholars have argued that words like “esoteric” or “prototantric” are simply not useful in describing Chinese Buddhist literature. The key argument for this position, made by Robert Sharf, is that the very concept of esoteric Buddhism is wholly the invention of late Japanese sectarians. Innovative ritual technologies like mantras and mandalas were coming to China from India, to be sure, but these new technologies did not occasion the formation of a new institution or Buddhist school called “esoteric Buddhism.” Discussions of esoteric Buddhism, for Sharf, smack of Japanese triumphalism, a teleological reading of history that regards the work and legacy of Kūkai as the apogee of authentic Buddhist practice, with the early history of Mahayana Buddhism serving as a mere prelude (2002, 263–278). Richard McBride, following Sharf, argued that the dhāraṇ‎īs and spells of East Asian Buddhism were quite normal, identified with no particular sectarian affiliation, and that we need not bother with categories like “prototantrism” when dealing with texts like the Great Peahen (McBride 2004, 2005).

Scholars should be grateful that these recent debates are shedding new light on the ubiquity and tremendous influence of spells and liturgical manuals for early Chinese Buddhism. However, much of the debate about what constitutes “esoteric” or “prototantric” Buddhism rests on semantic quibbles.13 In this chapter, I will call the Great Peahen “prototantric” because (1) its ritual manuals show a clear trend toward increasing complexity, with many procedures and instructions reminiscent of later, recognizably tantric Buddhist ritual manuals; and (2) the preface to Amoghavajra’s translation of the text explicitly calls the Great Peahen an “esoteric teaching” (mìjiào 密教‎; T. 982, 19:415a16). This means, at the very minimum, that the earlier versions of the Great Peahen furnished raw material from which Amoghavajra could assemble what he viewed as an esoteric scripture and ritual manual, complete with a maṇḍ‎ala, mantras, and mudrās.14 The labels “prototantric” or “tantric” need not refer to unique institutional identities or monastic lineages—we may instead use them to describe a set of historically and textually related ritual techniques.

Much research on the Great Peahen has already been conducted. Since the late version of the text contains long lists of yakṣ‎as and the lands they rule, Sylvain Lévi studied the text as a resource for South Asian geography (1915). Lambert Schmithausen studied the spell in the context of Buddhist encounters with dangerous natural phenomena like cobras and famine, and delved into its many precursors, including the Āṭ‎ānāṭ‎iyasutta (p.261) of the Dīghanikāya, the Cullavagga of the Vinayapiṭ‎aka, the Morajātaka, the Upasenasūtra in the Saṃ‎yuktāgama, and the Bhaiṣ‎ajyavastu of the Mūlasarvāstivādavinaya (1997, 11–23, 53–57). J. F. Marc DesJardins explored the text’s pantheon, its connection to South and East Asian spells more broadly, and its later developments in China and Japan (2002). Most recently, Henrik Sørensen studied the differences between the Chinese versions in detail, situating the text within the broader history of esoteric Buddhism in East Asia (2006).

While scholars have explored the text’s South Asian antecedents and its development in East Asia, relatively little has been written on the possible significance of the Great Peahen’s vicissitudes. In this chapter I will analyze the ways in which the Great Peahen grew and changed, and how these different kinds of change might be usefully interpreted by historians of religion.

Types of Change in the Great Peahen

The wealth of surviving versions of the Great Peahen offers scholars rare insight into the complexity of Buddhist textual and ritual change. The growth of the Great Peahen was not a simple or linear process. Thinking about textual development as a conjunction of overlapping processes, one can isolate at least four discrete types of textual change: (1) the merging of recensions, (2) a sudden expansion of the pantheon, (3) the addition of spells and passages from Chinese Buddhist texts, and (4) continuous changes in ritual procedures.

Merging Recensions

The extant versions of the Great Peahen may be divided into four main recensions, which have been explored in some detail by Henrik Sørensen (2006, 91–109).

  1. 1. T. 986. Dà jīnsè kǒngquè wáng zhòu jīng 大金色孔雀王呪經‎ [Great Golden Peacock King Spell Scripture]. Attributed to Śrīmitra. Probably late fourth century.

  2. 2. T. 987. Fóshuō dà jīnsè kǒngquè wáng zhòu jīng 佛說大金色孔雀王呪經‎ [Scripture Spoken by the Buddha on the Great Golden Peahen Queen Spell]. Attributed to Śrīmitra. Early fifth century.

  3. 3. T. 988. Kǒngquè wáng zhòu jīng 孔雀王呪經‎. [Peahen Queen Spell Scripture]. Attributed to Kumārajīva. Early fifth century.

  4. (p.262) 4. T. 984. Kǒngquè wáng zhòu jīng 孔雀王呪經‎. [Peahen Queen Spell Scripture]. Translated by Sēngqiépóluó 僧伽婆羅‎ [*Saṅ‎ghabhara]. Early sixth century.

The text of T. 984 is basically identical to the two later Chinese translations (T. 985, translated in 705 by Yìjìng 義淨‎, and T. 982, translated in the eighth century by Amoghavajra), to the surviving Sanskrit manuscripts, and to the Tibetan translation (Tōhoku 559, translated ninth–tenth century by Śākyaprabha and Ye shes sde). The two earliest recensions feature the same spells, but have entirely different narrative frames. T. 986, translated earlier in this chapter, is centered on the story of the peacock king named Suvarṇ‎āvabhāsa, who recites a spell to keep him from danger. T. 987 narrates the story of a monk named Svāti who, when bitten by a venomous snake, is healed through the power of the vidyā. Both recensions contain a core set of spells: the hu hu hu spell, the heart spell, and the spells to Maitreya, Brahmā, Indra, the Great Ṛṣ‎is, and the Devarājas. T. 988, on the other hand, shares the narrative frame of T. 987 but almost none of the spells found in the earliest two recensions. It shares with its predecessors only the spell of the Devarājas.

It is in T. 984 and its later manifestations that one can detect the merging of the earliest two recensions. Not only are the spells of both texts found in T. 984, but both narrative frames of the peacock king and the monk Svāti are found in the text. The unique elements of T. 988 are entirely missing from T. 984 and all later texts. Figure 6.1. shows what a (p.263) provisional stemma based on the surviving versions of the Great Peahen would look like. Table 6.1. collates shared material across the earliest versions.

Vicissitudes of Text and Rite in the Great Peahen Queen of Spells

Figure 6.1. Relationship of Extant Versions of the Great Peahen

The most striking fact to keep in mind here is the remarkable consistency of what I label the “core spells” across versions, with the exception of the anomalous T. 988. The main recensions of the Great Peahen vary mainly in their narrative frames, and by the early sixth century translation these recensions coalesced in the version of *Saṅ‎ghabhara, from which all later translations would not substantially diverge.

This recensional history can tell us some important facts. First, even though the text changes a great deal, the spells of the Great Peahen exhibit remarkable stability given the linguistic and geographic distances that must have been involved in the text’s transmission from South to Central and East Asia. One may readily consult the Sanskrit or Tibetan versions for guidance in reconstructing the transcribed spells of T. 986 or T. 987. The remarkable accuracy with which the vidyā was transmitted through the centuries testifies to the importance given by this tradition to the proper utterance of spells. Second, the Great Peahen gives us at least one way of thinking about what how prototantric literature works: the narrative frames for these texts are secondary to the spells and rituals they transmit. The pragmatic and modular nature of Buddhist prototantrism is thus reflected in the Great Peahen’s textual history.15

Expanding Pantheon

The core of the Great Peahen features spells connected with Maitreya, Brahmā, Indra, the Great Seers, and the Devarājas. Over time, new recensions of the Great Peahen expanded the range of featured deities, demons, and numina. The earliest signs of this expansion are found in T. 987, which contains a list of fourteen demonesses (luóchànǚ 羅剎女‎, rākṣ‎asī) who are propitiated for protection.16 In T. 984, the pantheon radically expanded, swelling to include long lists of the past Buddhas, yakṣ‎as, demonesses, dragon kings, river queens, mountain queens, constellations, and even poisons. Each section of this expanded version lists the names (and sometimes the locations) of these beings and gives a protective spell.

Sylvain Lévi already showed how useful this was as a resource for early medieval Indian geography. He investigated the long list of yakṣ‎as and their domains in detail, and pointed out the relative importance of Pāṭ‎aliputra and the high frequency of locations in the northwest of India (p.264) (1915, 116–117). Lévi dated the list to the first few centuries of the Common Era, in the time between the Kuṣ‎āṇ‎a and the Gupta empires.17

While Lévi’s contribution was immensely valuable, he did not write about the religious and cultural implications of the Great Peahen’s long lists of deities, demons, and numina. This necessary work was begun by DesJardins, whose dissertation devoted an entire chapter to the pantheon of the Great Peahen, tracing the Indian antecedents of the various deities (2002, 70–157). For DesJardins, the lists allow one to read the Great Peahen as a comprehensive inventory or census of the Buddhist spiritual world.18

But how might one read the broader meaning of such an inventory? One could see the list of yakṣ‎as as the product of Buddhist expansion through the subcontinent and into the northwest, but the exact nature of the relationship between the Buddhist community and the yakṣ‎as for this text is ambiguous. Not only do the deities chant the spells of the Great Peahen, but the text in turn implores the deities to recite their spells on behalf of the reader, and wishes long life for the deities. This is no simple appropriation or conquest of local cults.19 The Great Peahen reflects a complex interplay of Buddhist accommodation to or incorporation of local cults, as well as an attitude of superiority and conquest over these mundane spirits.20

Much as Lévi found the text most useful for thinking about Indian geography, I find the Great Peahen useful for thinking about Buddhist cosmology as a practical religious concern. When I teach undergraduates about Buddhist cosmology in the classroom, it is all too easy to think of and teach this system as a simple topographic “map,” a spatial model that helps orient the bewildered students as they try to make their way through the confusing details in the nidāna of a Buddhist text. But Buddhist cosmology is of course much more than a spatial model—it is a political map as well.

As the Great Peahen’s root text develops, we see increasing density and detail in its mapping. The earliest text has a broadly vertical arrangement; Brahmā, Maitreya, Indra, and the Seers all live at different altitudes. The later versions of the Great Peahen expand our view horizontally within Jambudvīpa proper, mapping out communities of rākṣ‎asas and yakṣ‎as, mountains and rivers, poisons and stars, all with great density and detail. This cosmology is less about the broad outlines of spatial organization, and much more about the sociopolitical world of the Buddhist practitioner. The practitioner’s immediate concerns are simple: Which agents are responsible for my sickness? Which agents (p.265) can I petition or coerce for redress? The developments of the Great Peahen allow us to trace the growth and increasing density of a Buddhist practical cosmology.

As the text moved from South to East Asia, the reception of these lists must have drastically changed. Chinese monks were surely less familiar with the intricate details of the local numina of the Indian subcontinent. Rather than reminding the listener of the deities and places of one’s homeland, the Great Peahen in Chinese would have likely communicated a vast and largely incomprehensible list of powerful alien beings. But even if the names and locations of the Indian demons were exotic and unfamiliar, the Chinese audience would have seen some resonance with their own practices. The notion of addressing demons by name to cure illness has been a constant feature of Chinese lived religion since at least the Warring States.21 Any Daoist in the Celestial Master tradition during the fourth or fifth centuries would surely have understood the gist of the Great Peahen, as its long lists of demons closely resembled early demonographic literature such as the Demon Statutes of the Lady Azure (Nǚqīng guǐlǜ 女青鬼律‎).22 The similarity of approach would have made it quite easy for Chinese practitioners to adopt the Great Peahen and augment it with their own materials.

Incorporation of Chinese Material

While the South Asian material may have been opaque at best to the East Asian readers of the Great Peahen, some of the Chinese versions contain lists of deities and excerpts from texts that would have been much more familiar. These Chinese additions reveal the fluidity and flexibility of the early medieval grimoire.

The mention of the Maṇ‎iratna Scripture and the *Mahāgandhabhikṣ‎u at the end of T. 986 is the earliest example of a Chinese intervention into the Great Peahen. The *Mahāgandhabhikṣ‎u is now lost. The Maṇ‎iratna Scripture, which survives only in a late fourth-century Chinese translation, is a short catalog of the names of demons (guǐ 鬼‎).23 T. 986 makes only the briefest reference to these scriptures, noting their ability to “rescue from sickness, suffering, and calamity.” The Great Peahen, on the other hand, is given an offensive mission: to “eliminate enemies” (T. 986, 19:478c27–28). This is powerful evidence that by the fifth century at the latest, the Great Peahen existed within a rich network of Chinese Buddhist ritual texts, and that the circulating copies of these texts contained instructions about their own use.

(p.266) T. 987 was even more explicit in forging connections between the Great Peahen and the growing Chinese corpus of Buddhist spells. The end of the scripture contains verbatim excerpts of Chinese translations of the dhāraṇ‎īs of Buddhist texts like the Bandit Scripture, the Lotus Scripture, and the Nirvāṇ‎a Scripture.24 This appendix of spells shows the reader how dhāraṇ‎īs could be detached from their literary context and circulated through Buddhist communities. Spells from the Lotus Scripture were separable lemmata from which the redactors of the Great Peahen could draw to create their new grimoire. The fact that these spells were taken from Chinese translations shows us that Chinese practitioners were actively engaging with Buddhist texts to build their own scriptures and traditions. T. 987 should be considered not just a distinct “version” or “recension” of the Great Peahen; it should be considered a new work—a Chinese liturgy—with the Great Peahen as its primary ingredient.25

T. 988 takes this trend to its extreme, stripping away many of the Great Peahen’s “core spells” and prepending the text with invocations to the Dragon Generals of the five directions, and uttering the “Mahāprajñāpāramitā Spell” (T. 988, 19:481c25–482b19). This version closes with the inclusion of a completely different scripture, the Āṭ‎avaka Spell Scripture.26 The radical differences between T. 988 and the other versions of the Great Peahen only reinforce the conclusion that the notions of “transmission” and “translation” in prototantric Buddhism are dangerously imprecise. Even while one might identify shared elements in most texts over time, these elements could be tossed out in favor of new texts and new spells. Besides speaking to the flexibility and fluidity of the Buddhist tradition, the Great Peahen’s inclusion of explicitly Chinese material shows how much freedom Chinese translators and editors could exert when fashioning their texts. This creativity and continuous innovation was reflected not only in the base text, but in the continuously changing ritual manuals as well.

Changes in Ritual Procedures

Beginning with T. 988, all versions of the Great Peahen append ritual manuals. Even as the base text more or less solidified and remained static after the translation of *Saṅ‎ghabhara, the ritual manuals continued to expand and transform, often quite radically.

The manual of T. 988 does not outline any specific ritual procedures. Instead, it reads like a shopping list: (p.267)

The Peahen Queen Spell arena: use cow dung to smear the ground. Use scattered seven-colored flowers, forty-nine pennants, four blades, four mirrors, one hundred arrows, one bow, seven vessels filled with broth, sixty lengths of black wool rope, twenty-five small cakes. Burn seven oil lamps. One dish of cream, one dish of grits, one dish of rice, one dish of small cakes. Set down one dish of honey and one dish of flowers. (T. 988, 19:484c4–8)

Though there is not much to go on here, the ritual undoubtedly would have involved propitiation with food and flowers, as well as decoration of the ritual space with pennants, mirrors, weapons, and lamps.

A ritual along similar lines was more thoroughly elaborated at the end of *Saṅ‎ghabhara’s translation. T. 984 lists similar ingredients, but also gives explicit instructions for the ritual procedure. The ritual involves the creation of a circular arena, invocation of the guardians of the directions for protection, and the ritual burning and beating of demons. The text even closes with a helpful diagram showing the arrangement of the lamps, swords, mirrors, pennants, and censer (T. 984, 19:458c13ff.) This ritual is strikingly simple. It involves no Buddha images, not even a bodhisattva or other central deity. Instead, the practitioner engages in a direct, violent battle with demonic forces within a delimited, ritually protected space.

The next translation by Yìjìng in the early eighth century features a much more complex ritual repertoire. The end of the text features no fewer than six distinct rituals: (1) a mandala rite involving the propitiation of painted images; (2) a mid-month rite featuring ingested metal beads and a fire ritual; (3) an abbreviated protective ritual for use at crossroads or in cemeteries; (4) emendations to the main ritual in order to care for a sick male child; (5) a heart spell; and (6) instructions on setting up a Great Peahen image in one’s own household (T. 985, 19:476a10ff.) The mandala rite is the most involved. In the center of Yìjìng’s mandala are a Buddha statue and a clay statue, painting, or iconic representation of the Peahen Queen. These figures are flanked by a kneeling Ānanda and an image of Vajrapāṇ‎i. Flowers are scattered; the Buddha and the Peahen Queen are offered food, drink, and sweets; incense is burned; and the guardian deities of the four directions are invoked for protection and are bribed with food and drink. The guardian deities stand on the four continents of the Sumeru world-system: Videha, Jambudvīpa, Godānīya, and Uttarakuru. Finally, the mandala rite prescribes a procedure for protecting the body in (p.268) which a young virgin girl weaves a rope, enchants it with the spell, knots it 108 times, and ties the spellcaster to herself with the rope.

One striking feature of this rite is the provision of substitute flowers and fruits: if no white arka flowers may be found, one may substitute them with lotus flowers. If no oleander flowers are available, one may use white apricot flowers instead (T. 985, 19:476a21–24).

The arena itself takes on a broader cosmological significance when compared with the simple circle in T. 984. While both rituals invoke the directional guardians, Yìjìng places the practitioner explicitly into the center of the Sumeru world system. The presence of the Buddha and Ānanda in the mandala mimics the narrative frame of the scripture itself, as does the inclusion of the Great Peahen and the great yakṣ‎a Vajrapāṇ‎i. If the root text of the Great Peahen reflects an increasingly dense practical cosmology, we can see in this ritual manual a concrete enactment of that cosmology.

As complex as this mandala might seem, it would become even more intricate in the ritual manual of Amoghavajra. This mandala features the Great Peahen in the center on an eight-petalled lotus blossom, riding a golden peacock. The eight petals of the lotus contain the seven Buddhas of the past and Maitreya, Buddha of the future. Outside the lotus petal are four pratyekabuddhas in the four cardinal directions and the four great śrāvakas (Ānanda, Rāhula, Śāriputra, and Maudgalyāyana) in the intermediate directions. Outside the inner sanctum are the “Celestial Kings of the Eight Directions” (Indra, Agni, Yama, Rākṣ‎asa King, the Water God, Vāyu, Vaiśravaṇ‎a, and Śiva). The third level contains the twenty-eight yakṣ‎a generals, the constellations, planets, and the twelve spirits of the zodiac. All groups of deities in the three levels of the mandala are propitiated with incense, food, and drink, and the practitioner is enjoined to set down a lectern with the scroll of the Great Peahen. Making offerings of flowers to the scripture, he and his assistants should read it aloud continuously for days on end. The practitioner smears his hands with incense and forms various mudrās while uttering short mantras like “oṃ‎ samaya sattvam!” (T. 983a).

Here we see the Great Peahen integrated into a recognizably esoteric ritual context. The practitioner is encouraged to supplicate the thirty-seven deities of the Diamond Realm mandala, thus linking the Great Peahen to one of the fundamental texts of East Asian esoteric Buddhism.27 The deities in this mandala stand in clear hierarchical relation to one another. As in the previously discussed rituals, they are given food and drink as a way of establishing and sealing off the ritual boundary. However, instead of violently beating the demons or simply reciting the spell, the practitioner is ordered to read the scripture itself with a team of assistants. While the formation (p.269) of the Great Peahen’s text may have involved the remixing and fluid circulation of various Indian and Chinese spells, by Amoghavajra’s time we can see the emergence of the Great Peahen as a discrete text, a ritual object with its own particular efficacy. No longer are the practitioners simply repeating oral formulæ; they are reading from a ritually consecrated scroll.

The rapid changes in ritual manuals in every version of the Great Peahen is striking, especially given the stability of the base text from the early sixth century forward. Once the text coalesced into its final form, it became a carrier of innovative ritual procedures. It is in these ritual procedures, the detailed instructions for describing the world of gods and demons, propitiating and commanding them, that we can see how rapidly the traditions of prototantric Buddhism were developing in the fifth through the eighth centuries. The ritual manuals show increasing systematization and detail, an expansion of their underlying cosmology, and ultimately an assimilation of the Great Peahen into more influential esoteric ritual systems.


That year [746 CE] the entire summer was exceedingly sunny. The Emperor asked the Great Master to enter the palace and pray for rain. His edict read, “The duration should not be long, and the rain should not be heavy.” The Great Master formally requested the mandalic rite of the Great Peahen Spell Queen Scripture. Before three days had passed, they were completely drenched with precipitation. The Emperor greatly rejoiced. Personally taking up his treasure chest, he bestowed upon the Great Master a purple kaṣ‎āya-robe. The Emperor opened it up for him to put it on. He also bestowed upon him two hundred bolts of silk.28

This passage, written by Zhàoqiān 趙遷‎, a disciple of Amoghavajra, seems at first glance a hagiographic cliché. But we know that Amoghavajra translated the Great Peahen. We know that the text instructs the reader to recite the spell in case of famine. We know that some of the Indian spells in the text contain exhortations to the gods to let rain pour forth. We also know that Amoghavajra wrote two appendices to this text, one a list of the spells to be recited in Siddham characters (T. 983b), the other a set of instructions for the images and mandala of the Peahen Queen rite (T. 983a). Given Amoghavajra’s interest in this spell, his prominent position in the Táng court, and the spell’s clear association with the production of rain, it seems wise not to view the account of Zhàoqiān with excessive suspicion.

(p.270) The scenario described by Zhàoqiān is even more striking when we turn our gaze southwest, to the great city of Kanyakubja. Here reigned Harṣ‎a, the Indian king who in 641 opened diplomatic ties with the Táng court after his celebrated meeting with the Buddhist pilgrim Xuánzàng 玄奘‎. In the Harṣ‎acarita, an account of the king’s deeds by the great poet Bāṇ‎a, the reader finds an amusing tale of Harṣ‎a’s father Prabhākaravardhana overcome by a grave illness, and the frantic efforts of his subjects to find a remedy:

Slowly, the guardians of the gates bowing to him, he entered the palace of the king, whose grievous condition was being discussed by the lords of men who had entered the courtyard. Their minds were aggrieved because they had not seen their master. They learned news from the members of his close retinue who emerged from the palace interior. Bathing, eating, and sleeping had become a mere fiction to them. Their clothes were filthy from rejecting personal hygiene. There were motionless, like drawings in a picture. They were with their outside retinue, who had formed a circle on the terraces, talking day and night of sorrow and misery in whispered discussions. Someone was pointing out the faults of the doctors. Someone was reciting passages concerning the characteristics of incurable diseases. Someone was talking about his bad dreams. Someone was telling a story about goblins. Someone was elucidating the teachings of astrologers. Someone was singing portents. Another was meditating on impermanence, condemning saṃ‎sāra, lambasting the decadence of the Kali age, and censuring fate. Another was angry with the dharma, and scolding the deities of the royal family. Another was complaining about the fortune of the afflicted gentleman. In the palace all possessions were being offered. Devotional rites for the family deities were being performed. The cooking of nectar and barley had begun. The fire-sacrifice of six oblations was being performed. Tremulous shoots of dūrvā grass smeared with bits of ghee and curd were being offered. The Great Peahen was being recited, the house pacification rites were being performed, and the performance of the protective offerings to the bhūtas was completed. The utterance of the Saṃ‎hitā propounded by the sages had been offered. The temple of Śiva was resounding with the Eleven Verses to Rudra being recited. The bathing of Virūpākṣ‎a with thousands of pitchers of milk was being carried out by the very pure devotees of Śiva.29

(p.271) There are clear differences between these two accounts. While Amoghavajra performs a mandala ritual (tánfǎ 壇法‎), the palace officiants in Kanyakubja were reciting (paṭ‎hyamāna) the Great Peahen. There can also be no doubt that the literary record of these two moments was shaped by the limitations of the hagiographical genre and by the agenda of the authors. Zhàoqiān’s account is unambiguously heroic, while Bāṇ‎a’s tale is much more ambivalent. In the Harṣ‎acarita, the Great Peahen is but one of a panoply of rites performed across the spectrum of Indian religious traditions. The Great Peahen—along with all the other grain offerings and statue bathings and fire sacrifice and verses to Rudra—is so lamely ineffective that the servants outside the palace resort to darker thoughts, reflecting on the nature of incurable diseases or the inevitable corruption of our wicked Kali age. Despite these differences, there remains a remarkable continuity: in both seventh-century India and eighth-century China, the Great Peahen Queen of Spells was an eminently practical text, used at the highest levels in imperial courts to combat dire calamities.

The vicissitudes of the Great Peahen provide a rich set of data for thinking about the development of Buddhist tantra. The preservation of so many recensions in the Chinese canon is not simply a testament to the popularity and enduring value of the Great Peahen. It may also reveal the kinds of textual and ritual developments generally applicable in Buddhist history—developments usually obscured by a lack of data. Texts were not only circulating in easily remixable fragments, but even relatively stable texts could become a site of startlingly rapid ritual innovation. Instead of thinking of the Great Peahen as a strange aberration, an embarrassingly inconsistent or unphilosophical work, it is far more useful to think of it as a uniquely detailed example of the construction and evolution of a Buddhist scripture. We see the movement from a hodgepodge grimoire, a series of loosely joined lemmata, toward a more stable macroform whose singular identity is the focus of ritual veneration. Would this process be so different in the case of the Lotus, the Flower Garland, or the Nirvāṇ‎a Scripture?

One might imagine a critical difference between the Great Peahen and a more “literary” or “philosophical” Mahāyāna scripture: perhaps it is the pragmatic nature of the Great Peahen that leads to such easy alteration, adaptation, and expansion. I am tempted to read the explosion of the Great Peahen’s pantheon and the mercurial shifts in its ritual manuals as a manifestation of the same kind of urge to comprehensiveness at work in Prabhākaravardhana’s court. The denizens of Kanyakubja wielded every possible rite and every possible devotional practice in the service of (p.272) their ailing king. An apotropaic text like the Great Peahen would be subject to the same universalizing impulses. The scripture’s hodgepodge nature and its lack of literary coherence allowed it to rapidly spread throughout South and East Asia with an ever-expanding pantheon and ritual repertoire. It would later be used as a rain-summoning ritual even further East, in Heian and Kamakura Japan (Trenson 2003; de Visser 1919–1920, 385–387). The Great Peahen would also find its way into the Pañcarakṣ‎ā, a twelfth-century anthology of five protective spells popular in Tibet and Nepal.30 Vajrācāryas to this day recite and transmit the Great Peahen as part of this ritual corpus, which is valued for its comprehensive powers of protection (Lewis 2000, 154ff.).

The Great Peahen did not confine itself to specifically Buddhist contexts. The scripture and many of its deities were absorbed into Daoist texts and rituals in the Míng 明‎ dynasty, appearing in the Scripture Uttered by the Celestial Venerable Tàishàng Yuánshǐ on the Holy Mother Celestial Venerable Empress of Precious Moonlight and the Peahen Queen of Spells (Tàishàng Yuánshǐ tiānzūn shuō Bǎoyuèguāng huánghòu shèngmǔ tiānzūn Kǒngquè míngwáng jīng 太上元始天尊說寶月光皇后聖母天尊孔雀明王經‎), the Invocation Rite for the Venerable Scripture of the Holy Mother and the Peahen Queen of Spells (Shèngmǔ Kǒngquè míngwáng zūnjīng qǐbó yí 聖母孔雀明王尊經啓白儀‎), and the Invocation Text Uttered by Celestial Venerable Tàishàng Yuánshǐ on the Peahen Queen Scripture (Tàishàng Yuánshǐ tiānzūn shuō Kǒngquèjīng bówén 太上元始天尊說孔雀經白文‎). Just as in the Buddhist Great Peahen, the texts feature an expansive pantheon of demons, dragons, and deities, but the specific names have changed, and the Buddhist elements have been fused into a Daoist context.31

The Great Peahen’s popularity extends even beyond the religious sphere. The text and its ritual power served as inspiration for a Japanese manga serialized from 1985–1989 called Peacock King (Kujaku Ō, 孔雀王‎), which later was adapted into an animated series, several video games, and even a live action kung-fu film.32 In this new manifestation the Buddhist monk Kujaku, a reincarnation of Mahāmāyūrī, battles all manner of evil spirits using his esoteric mantras and mudrās. While we may be tempted to dismiss these popular incarnations of the Great Peahen as trivialities, the long history of this spell’s transmission and its rapid changes should caution us from clinging too tightly to any monolithic vision of the tradition. Whether in art, literature, or liturgy, the only consistently unchanging feature of the Great Peahen has been its startling capacity for ready transmission and rapid transformation.

Table 6.1. Detailed Outline of the Earliest Recensions of the Great Peahen

T. 986

T. 987

T. 988

T. 984

Skt. [Takubo]

Directional guardians


Asking bodhisattvas for help


Kings of various divinities


Intention to utter the spell


Homage to spell, realms, and kings


Homage to the Buddha, etc.





Exhortation to demons to listen










Story of Svāti




Buddha recommends MMVR to Ānanda





Protective metrical verse: rātrau svasti divā svasti…





Spell iḍ‎i viḍ‎i kiḍ‎i…





Alliance magic (maitrī) verses for the Nāga Kings



Story of the Golden Peacock King




Spell “hu hu hu…” #1





Spell “hu hu hu…” #2



Spell siddhe susiddhe



Spell huci guci muci svāhā



T. 986

T. 987

T. 988

T. 984

Skt. [Takubo]

Heart spell ili mitti tili mitti…





Spell ilia mile kili…



Spell Citra mule Citra Citra male…



Buddhas and their trees



Great Yakṣ‎as



Kings of the Directions



Great Yakṣ‎a Generals



28 Yakṣ‎a Generals



Vaiśravaṇ‎a’s brothers



Great Piśācīs



Great Rākṣ‎asīs



Nāga Kings



Spells of the Buddhas



Maitreya spell





Brahmā spell





Indra spell





Four Deva Kings



River Queens



Mountain Kings






Great Ṛṣ‎is





Great Poisons



Spell of Devarājas and demons: “akaṭ‎e vikaṭ‎e…”






Plug for the Maṇ‎iratna and *Mahāgandhabhikṣ‎u scriptures


Spell of Deva Kings


Āṭ‎āvaka Spell Scripture


Lineage and benefits of spell



Ānanda cures Svāti



Ritual manuals




Spell “akaru arule…”


Bandit Scripture


Lotus Scripture spells


Nirvāṇ‎a Scripture spell


(p.273) (p.274) (p.275)





Bibliography references:

Chou Yi-liang [Zhōu Yīliáng 周一良‎]. 1945. “Tantrism in China.” Harvard Journal of Asiatic Studies 8 (3–4): 241–332.

Davidson, Ronald. 2002. Indian Esoteric Buddhism: A Social History of the Tantric Movement. New York: Columbia University Press.

Decaroli, Robert. 2004. Haunting the Buddha: Indian Popular Religions and the Formation of Buddhism. New York: Oxford University Press.

DesJardins, J. F. Marc. 2002. “Mahāmāyūrī: explorations sur la création dune scripture prototantrique.” PhD dissertation, McGill University.

Dutt, Nalinaksha, ed. 1984. Gilgit Manuscripts. 2nd edition. 4 vols. Bibliotheca Indo-Buddhica 14–19, 22–24. Delhi: Sri Satguru Publications.

Giebel, Rolf W. 2001. Two Esoteric Sūtras. BDK English Tripiṭ‎aka, vol. 29-II, 30-II. Berkeley, CA: Numata Center for Buddhist Translation and Research.

Harper, Donald. 1985. “A Chinese Demonography of the Third Century B.C.” Harvard Journal of Asiatic Studies 45 (2): 459–498.

Harper, Donald. 2004. “Contracts with the Spirit World in Han Common Religion: The Xuning Prayer and Sacrifice Documents of A.D. 79.” Cahiers d’extrême-asie 14: 227–267.

Hoernle, August Friedrich Rudolf, ed. 1893–1912. The Bower Manuscript: Facsimile Leaves, Nagari Transcript, Romanised Transliteration and English Translation with Notes. Archæological Survey of India New Imperial Series 22. Calcutta: Superintendent of Government Printing.

Kāṇe, Pāṇḍuraṅga Vāmana, ed. 1918. Harshacharita of Bāṇ‎abhaṭṭ‎a (Uchchvāsas I–VIII). Bombay.

Kieschnick, John. 1997. The Eminent Monk: Buddhist Ideals in Medieval Chinese Hagiography. Kuroda Studies in East Asian Buddhism 10. Honolulu: University of Hawai’i Press.

Kieschnick, John. 2003. The Impact of Buddhism on Chinese Material Culture. Princeton, NJ: Princeton University Press.

(p.281) Lai Chi-tim 黎志添‎. 2002. “The Demon Statutes of Nüqing and the Problem of the Bureaucratization of the Netherworld in Early Heavenly Master Daoism.” T’oung-Pao 88 (4): 251–281.

Lévi, Sylvain. 1915. “La catalogue géographique des Yakṣ‎a dans la Mahā-māyūrī.” Journal Asiatique 5 (1): 19–138.

Lewis, Todd T. 2000. Popular Buddhist Texts from Nepal: Narratives and Rituals of Newar Buddhism. Albany: State University of New York Press.

Lopez, Donald S. 1996. Elaborations on Emptiness: Uses of the Heart Sutra. Princeton, NJ: Princeton University Press.

McBride, Richard Dewayne II. 2004. “Is There Really ‘Esoteric’ Buddhism?” Journal of the International Association of Buddhist Studies 27 (2): 329–356.

McBride, Richard Dewayne II. 2005. “Dhāraṇ‎ī and Spells in Medieval Sinitic Buddhism.” Journal of the International Association of Buddhist Studies 28 (1): 85–114.

Ol’denburg, Sergej Fedorovič, ed. 1899. “Otryvki kašgarskixʺ i sansritskixʺ rukopisej izʺ sobranija N. F. Petrovskago.” Zapiski Vostočnago Otdělenija Imperatorskago Russkago Arxeologičeskago Obščestva 11: 207–264.

Needham, Rodney. 1975. “Polythetic Classification.” Man n.s. 10 (3): 349–369.

Orlando, Raffaello. 1981. “A Study of Chinese Documents Concerning the Life of the Tantric Buddhist Patriarch Amoghavajra (A.D. 705–774).” PhD dissertation, Princeton University.

Orzech, Charles D. 1996. “Maṇḍ‎alas on the Move: Reflections from Chinese Esoteric Buddhism circa 800 C.E.” Journal of the International Association of Buddhist Studies 19 (2): 209–244.

Orzech, Charles D. 1998. Politics and Transcendent Wisdom: The Scripture for Humane Kings in the Creation of Chinese Buddhism. University Park: Pennsylvania State University Press.

Orzech, Charles D. 2002. “Metaphor, Translation, and the Construction of Kingship in The Scripture for Humane Kings and the Mahāmāyūrī Vidyārājñī Sūtra.” Cahiers d’Extrême-Asie 13: 55–83.

Overbey, Ryan Richard. 2012. “On the Appearance of Siddhis in Chinese Buddhist Texts.” In Yoga Powers: Extraordinary Capacities Attained Through Meditation and Concentration, edited by Knut Axel Jacobsen, 127–144. Leiden: Brill.

Ruegg, David Seyfort. 2008. The Symbiosis of Buddhism with Brahmanism/Hinduism in South Asia and of Buddhism with ‘Local Cults’ in Tibet and the Himalayan Region. Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Sitzungsberichte, 774 Band. Beiträge zur Kultur- und Geistesgeschichte Asiens 58. Wien: Verlag der Österreichischen Akademie der Wissenschaften.

Schipper, Kristofer. 1994. “Purity and Strangers: Shifting Boundaries in Medieval Taoism.” T’oung-Pao 80 (1–3): 61–81.

Schipper, Kristofer, and Franciscus Verellen, eds. 2005. The Taoist Canon. Chicago: University of Chicago Press.

(p.282) Schmithausen, Lambert. 1997. Maitrī and Magic: Aspects of the Buddhist Attitude Toward the Dangerous in Nature. Österreichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Sitzungsberichte, 652 Band. Veröffentlichungen zu den Sprachen und Kulturen Südasiens 30. Wien: Verlag der Österreichischen Akademie der Wissenschaften.

Seidel, Anna. 1987. “Traces of Han Religion in Funeral Texts Found in Tombs.” In Dōkyō to shūkyō bunka道教と宗教文化‎, edited by Akizuki Kan’ei秋月觀暎‎, 21–57. Tōkyō 東京‎: Hirakawa Shuppansha平河出版社‎.

Sharf, Robert. 2002. Coming to Terms with Chinese Buddhism: A Reading of the Treasure Store Treatise. Kuroda Institute Studies in East Asian Buddhism 14. Honolulu: University of Hawai’i Press.

Skilling, Peter. 1992. “The Rakṣ‎ā Literature of the Śrāvakayāna.” Journal of the Pali Text Society 16: 109–182.

Sørensen, Henrik H. 2006. “The Spell of the Great, Golden Peacock Queen: The Origin, Practices, and Lore of an Early Esoteric Buddhist Tradition in China.” Pacific World third series 8: 89–123.

Strickmann, Michel. 1996. Mantras et mandarins: le bouddhisme tantrique en Chine. Paris: Éditions Gallimard.

Strickmann, Michel. 2002. Chinese Magical Medicine. Ed. Bernard Faure. Stanford, CA: Stanford University Press.

Takubo Shūyo 田久保周誉‎, ed. 1972. Ārya-Mahā-Māyūrī Vidyā-Rājñī. Tōkyō 東京‎: Sankibō Busshorin 山喜房佛書林‎.

Takakusu Junjirō 高楠順‎ and Watanabe Kaikyoku 渡邊海旭‎, eds. 1924–1932. Taishō shinshū daizōkyō 大正新修大藏經‎. 85 vols. Tōkyō 東京‎: Taishō issaikyō kankōkai 大正一切經刊行會‎.

Trenson, Steven. 2001–2003. “Une analyse critique de l’histoire de Shōugyōhō et du Kujakukyōhō: rites ésotériques de la pluie dans le Japon de l’époque de Heian.” Cahiers d’Extrême-Asie 13: 455–495.

de Visser, Marinus Willem. 1919–1920. “Die Pfauenkönigin (K’ung-tsioh ming-wang, Kujaku myō-ō) in China und Japan.” Ostasiatische Zeitschrift 8: 370–387.

Waddell, Laurence Austine. 1912. The “Dhāraṇ‎ī” Cult in Buddhism: Its Origin, Deified Literature and Images. Berlin: Österheld.

Watanabe Kaikyoku 渡邊海旭‎. 1907. “A Chinese text corresponding to part of the Bower Manuscript.” Journal of the Royal Asiatic Society of Great Britain and Ireland 1907: 261–266.

Texts from the Taishō shinshū daizōkyō 大正新修大蔵経‎

T. 982. Fómǔ dàkǒngquè míngwáng jīng 佛母大孔雀明王經‎ [Scripture on the Goddess Great Peahen Queen of Spells]. Translated 8th c. by Amoghavajra.

T. 983a. Fóshuō dà kǒngquè míngwáng huàxiàng táncháng yíguǐ 佛說大孔雀明王畫像檀場議軌‎ [Ritual Manual Spoken by the Buddha on the Images and Maṇḍ‎ala for the Great Peahen Queen of Spells]. Compiled 8th c. by Amoghavajra.

(p.283) T. 983b. Kǒngquè jīng zhēnyánděng fànběn 孔雀經真言等梵本‎ [Sanskrit Volume of Mantras &c. for the Peahen Scripture]. Compiled 8th c. by Amoghavajra.

T. 984. Kǒngquè wáng zhòu jīng 孔雀王呪經‎. [Peahen Queen Spell Scripture]. Translated early 6th c. by Sēngqiépóluó 僧伽婆羅‎ [*Saṅ‎ghabhara].

T. 985. Fóshuō dàkǒngquè zhòuwáng jīng 佛說大孔雀呪王經‎. [Scripture Spoken by the Buddha on the Great Peahen Queen of Spells]. Translated 705 by Yìjìng 義淨‎.

T. 986. Dà jīnsè kǒngquè wáng zhòu jīng 大金色孔雀王呪經‎ [Great Golden Peacock King Spell Scripture]. Probably late 4th c. Attributed to Śrīmitra.

T. 987. Fóshuō dà jīnsè kǒngquè wáng zhòu jīng 佛說大金色孔雀王呪經‎ [Scripture Spoken by the Buddha on the Great Golden Peahen Queen Spell]. Probably early 5th c. Attributed to Śrīmitra.

T. 988. Kǒngquè wáng zhòu jīng 孔雀王呪經‎. [Peahen Queen Spell Scripture]. 5th c. Attributed to Kumārajīva.

T. 1034. Zhòu wǔshǒu jīng 咒五首經‎ [Scripture of Five Spells]. Translated 664 by Xuánzàng 玄奘‎.

T. 2056. Dà Táng gù dàdé zèng sīkōng dàbiàn zhèng guǎngzhì Bùkōng sānzàng xíngzhuàng 大唐故大德贈司空大辨正廣智不空三藏行狀‎ [Account of the Deeds of the Late Great Bhadanta Amoghavajra Trepiṭ‎aka of the Great Táng, to Whom Was (Posthumously) Bestowed the Titles Sīkōng, He of Great Discrimination, He of Correct Vast Wisdom]. Compiled 8th c. by Zhàoqiān 趙遷‎.

T. 2061. Sòng gāosēng zhuàn 宋高僧傳‎ [Sòng Biographies of Eminent Monks]. Compiled late 10th c. by Zànníng 贊寧‎.

T. 2145. Chū sānzàng jìjí 出三藏記集‎ [Collected notes on the translation of the Tripiṭ‎aka]. Compiled circa 515 by Sēngyòu僧祐‎.


(1.) My thanks to Janet Gyatso for comments on the first draft of this article, originally begun as a term paper for a graduate course in Buddhist Studies at Harvard University. I would also like to thank the graduate students at the University of California, Berkeley, who took my seminar on early Buddhist ritual literature in the autumn of 2014. Their critical and enthusiastic engagement with the Great Peahen corpus was inspiring. Special thanks to James Marks, whose gimlet eye spotted some potentially embarrassing errors in my comparative table. All that is good in this chapter comes from my teachers and students; all infelicities are mine alone.

(2.) The translation Great Peahen Queen of Spells for Mahāmāyūrīvidyārājñī is tentative. Lambert Schmithausen, noting that the text draws from an old tale about a peacock king named Suvarṇ‎āvabhāsa, precisely translates Mahāmāyūrīvidyārājñī as the “Queen of Spells [called] the Great One of the Peacock” (Schmithausen 1997, 53). However, most iconography of Mahāmāyūrī in South and East Asia depict the figure as a female bodhisattvā, sometimes riding on a peacock. So should the reader take Mahāmāyūrī adjectivally (feminine in agreement with the noun vidyārājñī) or nominally (referring to the female embodiment of the spell itself)? Charles Orzech noted that the first explicit stance on this question was taken by Amoghavajra, whose eighth-century Chinese translations refers to Mahāmāyūrī as a Buddhist goddess (fómǔ 佛母‎; Orzech 2002, 82–83). As we shall see, not all versions of the text draw from the story of Suvarṇ‎āvabhāsa. I use Schmithausen’s interpretation when the context and grammar of the Chinese sources are clear, but when there is no clear answer I will, in deference to the iconography, refer to the spell as the Great Peahen.

(3.) The Chinese texts may be found in T. 982–988. Fragments of the Sanskrit text were found in the Bower Manuscript, on which see Hoernle (1893, 234ff). These fragments were first connected with the Chinese texts in Watanabe (1907). The earliest full Sanskrit edition may be found in Ol’denburg (1899). The most thorough Sanskrit edition to date is Takubo (1972). For detailed overviews of the many versions of the Great Peahen, see Waddell (1912); de Visser (1919–1920); Schmithausen (1997); DesJardins (2002, 20–28); and Sørensen (2006). DesJardins’s dissertation remains the most comprehensive source on the antecedents and later reception of the Great Peahen.

(4.) This was put more charitably by J. F. Marc DesJardins in his dissertation on the Great Peahen: “Le MMVR est un ouvrage difficile à aborder à cause de son hermétisme, de son aspect de grimoire médiéval, et de son manque total de philosophie. Difficile à interpréter, il ne semble y avoir rien qui puisse valider idéologiquement sa popularité chez les prêtres de l’ésotérisme. En cela, le language du MMVR en est un de rituel, de magie, et de pouvoir. Le rituel fait le MMVR” (2002, 6).

(5.) Dà jīnsè kǒngquèwáng zhòu jīng 大金色孔雀王呪經‎. The Chinese title’s use of jīnsè 金色‎ “Golden” probably translates Suvarṇ‎āvabhāsa (“Golden Light”), the name of the Peacock King also found in the Mūlasarvāstivādavinaya (Dutt 1984, 3.1: 285–288). The text of T 986 varies in its titles considerably. The heart spell is entitled the “Heart of the Queen of Spells of the Great Peacock King” (Dà kǒngquèwáng zhòuwáng xīn 大孔雀王呪王心‎; T. 986, 19:477c16). Later the spell is called “Queen of Spells of the Great Peacock” (Dà kǒngquè zhòuwáng 大孔雀呪王‎; T. 986, 19:478a5, a12, a18, &c.), and later still the scripture itself is referred to as the “Scripture of the Spell of the Great Golden Peacock King” (Dà kǒngquèwáng zhòu jīng 大金色孔雀王呪經‎; T. 986, 19:478b19).

(6.) This spell (vidyā) contains mostly incomprehensible syllables transcribed into Chinese. When possible, I refer to the later Sanskrit version to interpret the Chinese transcriptions—often they are remarkably close. All the translations of the Great Peahen contain Chinese transcriptions of these Indian magical utterances, although occasionally the spells contain comprehensible injunctions that are translated into Chinese. For example, in the Sanskrit version the spell beginning with akaṭ‎e vikaṭ‎e . . . (Takubo 1972, 23–24) ends with the injunction priyaṃ‎kare rakṣ‎a māṃ‎ sarvasatvāṃ‎ś ca jīvatu varṣ‎aśataṃ‎ paśyatu śaradāśataṃ‎! In T. 986 this portion of the spell is translated, not transcribed: yōnghù mǒujiǎ, lìng déshòu bǎisuì, déjiàn bǎiqiū! 擁護某甲‎, 令得壽百歲,得見百秋!‎ (T. 986, 19:478c26).

(7.) In this translation I use . . . [&c.] to indicate a series of largely nonsensical Indic sounds phonetically transcribed into Chinese. I omit most of the translation for the sake of brevity.

(8.) The “Great Seers” of this spell are found in the later Sanskrit edition as maharṣ‎i, the celebrated hearers of the Vedas. Sørensen claimed that this passage was one of the earliest references to siddhis in an East Asian text (2006, 95), but this claim rests on a dubious translation of the passage. For a detailed analysis of this passage see Overbey (2012, 132–135).

(9.) The characters Móhēqiántuó bǐqiū 摩訶乾陀比丘‎ probably refer to the scripture in one scroll, now lost, whose title in the early sixth-century Chū sānzàng jìjí 出三藏記集‎is recorded as Móhēqiántuó wéiwèiluó jìnxìn bǐqiū děngdù jīng 摩訶揵陀惟衛羅盡信比丘等度經‎. See T. 2145, 55:19c4–6. Sørensen (2006, 96) erroneously translates the characters 摩訶乾陀比丘‎ in the vocative as “Mahāmaudgalyāyana Bhikṣ‎u,” but nowhere in the canon is 摩訶乾陀‎ attested as a translation of Mahāmaudgalyāyana.

(10.) For example, Michel Strickmann called the Great Peahen “l’un des textes fondamentaux du prototantrisme” (1996, 296).

(11.) See Chou (1945, 241–245) for the argument that the various Buddhist spells and rituals translated in China during the first centuries CE constituted a kind of incipient tantrism.

(12.) See Strickmann (1996, 117–118); Sørensen (2006, 109–110).

(13.) For terminological reflections on the applicability of words like “tantra” to particular phenomena, see Lopez (1996, 83–104); Davidson (2002, 118ff). Both scholars rely on “polythetic classification” to define the word. This is a fancy way of saying that, given two different sets of defining characteristics for tantra A and B, one may choose to call x tantric if x ∈ (AB), rather than requiring x ∈ (AB). The term “polythetic classification” was introduced to anthropologists in the journal Man by Rodney Needham, who had noticed the similarity between Wittgenstein’s family resemblances and the polythetic taxonomies of contemporary biologists (Needham 1975). Needham’s article was written about one century after George Cantor invented naïve set theory.

(14.) McBride goes quite far in arguing against applying labels like “esoteric” or “tantric” to rites and scriptures labeled mìjiào 密教‎, since the term mìjiào often was used as a polemical device to announce the superiority of one’s own tradition or lineage (McBride 2004, 354–356). On this reading, the scholar should always view any deployment of the term mìjiào with suspicion.

(15.) For reflections on the modular structure of esoteric Buddhism, see Orzech (1996).

(16.) Sørensen (2006, 99) discusses and translates the names of these fourteen demonesses. Unfortunately, his translations from the Chinese did not benefit from a comparison with the Sanskrit edition. To give one example, Sørensen’s translation of “Naked Body” for Báijù 白具‎ is surely mistaken, as the name of the demoness in Sanskrit versions reads Śaṃ‎khinī. In English the Sanskrit would be translated as “She with the Conch.” This should, in turn, lead us to emend the Chinese to Báibèi白貝‎, “White Shell.” For the relevant Sanskrit passage, see Takubo (1972, 2).

(17.) See Lévi (1915, 118). Lévi also ventures a much bolder hypothesis in his article. By dating the list in the Great Peahen, and pointing out similarly named locales in the Mahābhārata, Lévi guessed that the Mahābhārata itself was finally redacted around the same time as the Great Peahen (122).

(18.) “Il est un document officiel, un registre d’état civil du monde spirituel bouddhique” (DesJardins 2002, 155).

(19.) A repeating formula throughout the text reads: “May they [the numina] protect the mendicant Svāti, myself, and all beings with this Great Peahen Queen of Spells! May they preserve, protect, surround, guard, pacify, bless, rescue from punishment, rescue from weapons, counteract poison, destroy poison, mark off the ritual arena, and mark off the earth! May they live a hundred years! May they see a hundred autumns!” See for example, Takubo (1972, 7, 10, 13, 15, etc.).

(20.) For archæological evidence on the complex relationship between Buddhist monasteries and yakṣ‎a cults, see Decaroli (2004). For methodological reflections on the relationship between Buddhism and South Asian deities generally, see Ruegg (2008, especially pp. 19–29).

(21.) See Harper (1985, 2004). For further reflections on the demonology and spell literatures of Buddhism and Daoism in early medieval China, see Strickmann (2002, 58–122).

(22.) On the Nǚqīng guǐlǜ, see Seidel (1987), Schipper (1994), and Lai (2002).

(23.) See T. 1393. Sørensen (2006, 95–97) argues that the Maṇ‎iratna Scripture served as a liturgical complement to the Great Peahen, the former providing a list of demons and the latter describing the necessary rituals. Sørensen views T. 986 not as a full-fledged scripture, but rather as a short ritual guide. For an overview of the Maṇ‎iratna Scripture, see Strickmann (2002, 109–113).

(24.) T. 987, 19:481a25–c17. See Sørensen (2006, 100–101) for more details on these excerpted spells. Sørensen argues that this sort of textual “mining” is evidence that “the formation of esoteric Buddhism as a distinct—although not separate—tradition in Chinese Buddhism was gradually taking shape” by the time of the text’s redaction.

(25.) Many such grimoires are found throughout the Chinese Buddhist canon. One notable example, attributed to Xuánzàng 玄奘‎, is the Scripture of Five Spells (Zhòu wǔshǒu jīng 呪五首經‎; T. 1034), which contains five mantras completely stripped away from their original literary contexts. I am speaking here only of the materials that made it into the Chinese Buddhist canonical catalogues; the Dūnhuáng archives are full of non-canonical, practical ritual manuals. See Dalton’s contribution to this volume (Chapter 4) for examples in the Tibetan materials from Dūnhuáng.

(26.) T. 988, 19:484a13ff. Sørensen (2006, 104) correctly notes that this version of the Āṭ‎akava Spell is most useful for assigning a date of the early fifth century to the text. See Strickmann (2002, 143–151) for a description of an early scripture of magical seals devoted to Āṭ‎avaka.

(27.) For a convenient diagram and list of these deities, see Giebel (2001, 12–14).

(28.) T. 2056, 50:293a21–24. This is from the biography of Amoghavajra written by his lay disciple Zhàoqiān 趙遷‎. Zhàoqiān’s description of this incident was copied nearly verbatim in the hagiographical collection edited by Zànníng 贊寧‎ (919–1001), the Sòng Biographies of Eminent Monks (T. 2061, 50:712c14–17). For further details on the life and work of Amoghavajra, see Chou (1945, 284–307); Orlando (1981); and Orzech (1988, especially 135–167). For more details on the use of the purple robe in China, see Kieschnick (2003, 100–103).

(29.) See Kāṇ‎e (1918, ucchvāsa 5, 21.25–22.7). This long passage translates a single sentence from Bāṇ‎a’s verbose Sanskrit prose. The sentence reads: mandaṃ‎ mandaṃ‎ dvārapālaiḥ‎ praṇ‎amyamānaś ca dīyamānasarvasvaṃ‎ pūjyamānakuladevataṃ‎ prārabdhâmṛ‎tacarupacanakriyaṃ‎ kriyamāṇ‎aṣ‎aḍ‎āhutihomaṃ‎ hūyamānapṛṣ‎adājyalavaliptapracaladurvāpallavaṃ‎ paṭ‎hyamānamahāmāyūrīpravartyamānagṛ‎haśāntinirvartyamānabhūtarakṣ‎ābalividhānaṃ‎ prayatavipraprastutasaṃ‎hitājapaṃ‎ japyamānarudrâikadaśīśabdāyamānaśivagṛ‎ham atiśuciśaivasamāpadyamānavirūpākṣ‎akṣ‎īrakalaśasahasrasnapanam ajirôpaviṣṭ‎aiś cânāsāditasvāmidarśanadūyamānamānasair abhyantaraniṣ‎patitanikaṭ‎avartiparijananivedyamānavārtair vārtībhūtasnānabhojanaśayanair ujjhitâtmasaṃ‎skāramalinaveśair likhitair iva niścalair narapatibhir nīyamānanaktandivaṃ‎duḥ‎khadīnavadanena ca praghaneṣ‎u baddhamaṇḍ‎alenopāṃ‎śuvyāhṛ‎taiḥ‎ kenacic cikitsakadoṣ‎ān udbhāvayatā kenacid (p.280) asādhyavyādhilakṣ‎aṇ‎apadāni paṭ‎hatā kenacid duḥ‎khasvapnān āvedayatā kenacit piśācavārtāṃ‎ vivṛṇ‎vatā kenacit kārtāntikādeśān prakāśayatā kenacid upaliṅ‎gāni gāyatânyenânityatāṃ‎ bhāvayatā saṃ‎sāraṃ‎ cāpavadatā kalikālavilasitāni ca nindatā daivaṃ‎ côpālabhamānenâpareṇ‎a dharmāya kupyatā rājakuladevatāś câdhikṣ‎ipatâpareṇ‎a kliṣṭ‎akulaputrakabhāgyāni garhayatā bāhyaparijanena kathyamānakaṣṭ‎apārthivâvasthaṃ‎ rājakulaṃ‎ viveśa.

(30.) For more on rakṣ‎ā literature generally, see Skilling (1992).

(31.) For summaries of these works, see Schipper and Verellen (2005, 1233–1234).

(32.) The manga, written by Makoto Ogino 荻野真‎, was serialized in Shūkan yangu jampu 週刊ヤングジャンプ‎ [Weekly Young Jump] magazine between 1985 and 1989.