“After two decades of regular and intimate contact with the ancient Buddhist manuscripts held at the British Library, Sam van
Schaik is uniquely positioned to write this book. The spell books
and amulets he presents are gateways into the Buddhism of daily
life. These are not the canonical works taught at monastic colleges and promoted around the world; they are the personally
held writings of individual Buddhists with particular concerns
living in specific locales. In van Schaik’s skilled hands, the category of Buddhist magic sheds light on a fundamental world of
Buddhist beliefs and practices that is rarely seen.”
—Jacob Dalton, author of
The Gathering of Intentions: A History of Tibetan Tantra
“An important book about what many don’t know about Buddhism in the Western world. Buddhist Magic affirms that there is
more to Buddhism than what meets the eye. With careful attention and confirming Buddhist scripture, van Schaik successfully
brings to light the shamanic realities of Buddhism that underline
every transmission. This book is a treasure.”
—Zenju Earthlyn Manuel, author of
The Way of Tenderness: Awakening through Race, Sexuality, and Gender
Buddhist
Magic
D I V I N AT I O N , H E A L I N G ,
AND ENCHANTMENT
THROUGH THE AGES
S A M VA N S C H A I K
Shambhala Publications, Inc.
4720 Walnut Street
Boulder, Colorado 80301
www.shambhala.com
© 2020 by Sam van Schaik
Cover art: Detail of Tamra phichai songkhram, British Library
Cover design: Kate White
Interior design: Claudine Mansour Design
All rights reserved. No part of this book may be reproduced
in any form or by any means, electronic or mechanical,
including photocopying, recording, or by any information storage
and retrieval system, without permission in writing from the publisher.
987654321
First Edition
Printed in the United States of America
oThis edition is printed on acid-free paper that meets
the American National Standards Institute Z39.48 Standard.
kThis book is printed on 30% postconsumer recycled paper.
For more information please visit www.shambhala.com.
Shambhala Publications is distributed worldwide by
Penguin Random House, Inc., and its subsidiaries.
Library of Congress Cataloging-in-Publication Data
Names: van Schaik, Sam, author.
Title: Buddhist magic: divination, healing, and enchantment
through the ages / Sam van Schaik.
Description: First edition. | Boulder, Colorado: Shambhala Publications, 2020. |
Includes bibliographical references and index.
Identifiers: LCCN 2019045976 | ISBN 9781611808254 (trade paperback)
Subjects: LCSH: Magic—Religious aspects—Buddhism.
Classification: LCC BQ4570.M3 V36 2020 | DDC 294.3/43—dc23
LC record available at https://lccn.loc.gov/2019045976
To Anne
CONTENTS
Preface ix
Introduction 1
Chapter 1. Magic across Cultures 17
Chapter 2. Magic, Medicine, and the Spread of Buddhism 43
Chapter 3. Sources of Magic in Buddhist Scripture 69
Chapter 4. Magic Users and Materia Magica 93
Chapter 5. A Tibetan Book of Spells 129
Afterword 167
Key Ter ms 171
Notes 175
Works Cited 198
Index 213
P R E FA C E
The first thing a student of magic learns
is that there are books about magic and books of magic.
And the second thing he learns is that a
perfectly respectable example of the former may be had for
two or three guineas at a good bookseller,
and that the value of the latter is beyond rubies.
—Jonathan Strange and Mr Norrell
Taking my cue from Suzanna Clarke’s novel about professional
magicians, I want to point out that this is a book about magic and
definitely not a book of magic. Or to be more precise, it’s a book
about a book of magic. That book is a Tibetan book of spells,
written over a thousand years ago. I first came across it while cataloguing the Tibetan tantric manuscripts with Jacob Dalton, in
a project that ran from 2003 to 2005, and owed much to Burkhard Quessel and Susan Whitfield at the British Library, as well
as Cathy Cantwell, Robert Mayer, and Ulrich Pagel. In 2009, I
wrote a couple of posts on my blog (https://earlytibet.com) exploring this manuscript, which started some interesting discussions
and helpful responses, in particular from Dan Martin and PéterDániel Szántó.
The next time I returned to this Tibetan book of spells was
under the aegis of a new research project, Beyond Boundaries:
Religion, Region, Language and the State, funded by the European Research Council in 2014. One of the themes of this project,
which I developed alongside Michael Willis at the British Museum
and Nathan Hill at the School of Oriental and African Studies,
was the way that Buddhist monks brought aspects of Indic culture
x
I
Preface
along the silk routes toward East Asia. In the course of the project
I began to see how the Tibetan book of spells could offer a way
in to an aspect of Buddhist history that has been neglected: the
role of magical practices in the spread of Buddhism beyond India.
Among the many researchers who contributed to the project, I
worked most closely with Gergely Hidas, whose writings—along
with many conversations at the Fitzrovia Tavern—about pragmatic rituals in Buddhism have deeply informed this book.
Another forum for discussing the rituals in the book of spells was
the series of workshops organized by Agnieszka Helman-Ważny at
the University of Hamburg on Bonpo manuscript cultures. I’d like
to thank her as well as Daniel Berounsky, Marc Desjardins, and
Bryan Cuevas for sharing their work on Tibetan magical literature. Jens Braarvig and Birgit Kellner also invited me to talk on
the books of spells at events in Oslo and Vienna that resulted in
lots of useful feedback. At the latter, Jonathan Samuels and Toni
Huber gave some very good advice on approaches to using the
word magic. Ulrike Roesler invited me to give a lecture at Oxford,
which provided an opportunity to think in more detail about the
interaction between Buddhist magic and Buddhist ethics.
During the course of writing this book, I worked at the British
Library alongside researchers on the Beyond Boundaries project,
Gethin Rees and Matt Kimberley, and had many productive discussions about Buddhist archaeology and medicine with them.
Curatorial colleagues at the library were generous with their
knowledge of relevant literature, especially Mélodie Doumy, who
pointed out interesting sources and issues in the Chinese Dunhuang manuscripts, and Eyob Derillo, who shared his work on
magic in Ethiopian manuscripts
As with everything I have written about Buddhism, I owe a
debt of gratitude to Lama Jampa Thaye for sharing his profound
understanding of the tradition. In this case, I also must thank
him specifically for our discussions about the use of and attitudes
Preface
I
xi
toward magical rituals in contemporary Tibetan Buddhism. Finally, I want to thank Shambhala Publications, and Nikko Odiseos
in particular, for accepting this slightly unusual work for publication, and Matt Zepelin for his spot-on editorial comments and reflections on the place of magic in contemporary Buddhism, which
have truly improved the book.
As a final note, I would like to emphasize that the spells in this
book are offered for information only. There is no longer a living lineage of transmission and instruction for these particular
spells, which in the Buddhist context means that they are no longer “alive” as a practice. And in the absence of such a lineage of
instruction, my own translation will always be provisional. It is
perfectly adequate to help us understand the tradition of Buddhist magic, but entirely unsuited to instruction in the practice of
magic. So I would not recommend that anybody actually try these
spells, and if they did, I would not expect them to work as they
were originally intended. This is, after all, a book about magic.
Buddhist Magic
INTRODUCTION
In the hills of northeastern Tibet, a Buddhist monk stands in the
midst of a hailstorm, swinging a sling above his head, and looses
its missile into the clouds while reciting a fierce mantra. The pellets he fires from the sling are made from the soil of a ruined castle
wall, the tooth of a mad dog, and the ash of a burned musk deer.
Far away in a Theravada monastery near Bangkok, a monk tattoos a symetrical design with spiraling letters flanked by dragons
onto the arm of a soldier. This sacred Buddhist symbol will confer
power and protection upon its owner. And in a Buddhist temple in
Kyoto, row upon row of small pouches called omamori are offered
for sale. A woman picks up one made specifically to ensure a successful childbirth, which she will wear throughout her pregnancy.
These practices, and many others like them, are seen throughout the Buddhist world today. They are nothing new: the practices
and sacred texts on which they are based go back centuries into
the Buddhist past. I have been aware of this side of Buddhism
from my own experiences living in Asia, but it was only when I
started working with the ancient Buddhist manuscripts from the
Silk Road site of Dunhuang that I began to realize that such practices had been part of Buddhism for a long, long time. This book is
an exploration of this lesser-known side of our Buddhist heritage,
an exploration of how Buddhists have used what I will be calling
“magic” to address the everyday needs of monks, nuns, and laypeople. While these practices are found across and beyond Asia,
they address the same concerns: protecting travelers, pregnant
women, children, and the sick; knowing when to begin a journey
or start a business venture; bringing on the rains or sending away
2
I
Introduction
hailstorms; making somebody fall in love or breaking two lovers
apart; silencing a critic or even killing a threatening enemy; and
acquiring the tempting powers of magical persuasion, invisibility,
and flight.
There are books of magic in most Buddhist cultures, including
Theravadin ones, but they have rarely been studied. If we are to
understand the way Buddhism has worked in the past, the way it
still works now in many societies, and the way it can work in the
future, we need to look at these aspects of Buddhist practice that
scholarship has tended to overlook. For centuries, Buddhist monks
and nuns have offered services including healing, divination, rainmaking, love magic, and more to local clients. This is how Buddhist monastics developed and maintained relationships with their
local communities, and this is why magic and healing have played
a key role in Buddhism’s flourishing.
THE TROUBLE WITH MAGIC
Using this word magic allows us to focus on a particular kind of
activity that went on, and still goes on, in Buddhist societies.
These practices are well known to Buddhist monks and laypeople
in Asian countries and are described in detail in manuscripts, yet
they are far less apparent in the scholastic and doctrinal texts that
have been the focus of much Buddhist scholarship. This holds true
even in the tantric literature, which, despite its sometimes transgressive imagery, remains focused on the ultimate Buddhist state
of enlightenment. Thinking in terms of magic allows us to bring
into view a group of practices that have little to say about Buddhist
concepts of virtue, merit, and enlightenment, practices that are
entirely intended for the everyday concerns of the present life.
Despite the importance of magical practices in Buddhism, they
are still one of the least studied aspects of the religion. I suspect
that one of the main reasons for this is the idealized image of
Introduction
I
3
Buddhism as a rational religion, essentially free from superstition
and ritual. Though this ideal is belied by any experience of the
practices of modern Buddhist communities in Asia, such practices are often just taken as evidence that Buddhism degenerated
from its early ideals, the pure Buddhism taught by Gotama and
practiced by his disciples. Yet as early as the evidence can take us
back, spells and other magical practices are there. Figures of ogres
adorn early Buddhist monasteries, and the Pali canon includes accounts of Buddhist monks dealing directly with such beings. The
canonical texts include spells for protection from nonhuman beings—and sometimes also for summoning them. Though the Pali
canon is from relatively late sources, the earliest manuscripts containing Buddhist texts, found in Afghanistan, also contain magical
practices.1
Magical powers also feature heavily in early Buddhist literature.
As Bryan Cuevas points out, the achievement of magical powers
was not only accepted but was a crucial characteristic of the Buddha and other advanced practitioners:
In some of the earliest Buddhist sūtras from the Pāli
canon, we find standardized lists of the various magical
powers possessed by the Buddha and certain other practitioners advanced in meditation. . . . The first category, the
wonder-working powers, encompasses the widest array
of paranormal abilities, including the powers of physical
transformation and multiplication of the body, as well as
the ability to appear and disappear at will, to pass unhindered through walls, mountains, and other solid objects
and surfaces, to walk on water, to fly cross-legged through
the air, to manipulate the elements (earth, water, fire and
air), to touch the sun and moon, and to travel to the heavenly realms.2
4
I
Introduction
The idea that the spiritual path included cultivating powers
like these was also accepted by Mahayana Buddhist monks and
the practitioners of tantric Buddhism, where the powers gained
through the practice of magical scriptures can be divided into
three types. The highest powers are the attainment of the status
of a magic user (vidyādhara), supernormal cognitive powers, and
perfect knowledge of treatises. The middle powers are invisibility, youthfulness, and speed-walking. The lower powers are controlling, killing, and exorcising. As these last three powers in the
list show, discussion of tantric texts is distinguished by an open
acknowledgment of the practice of aggressive magic.3 These magical powers were not the ultimate aim of the Buddhist path, but
they were part of it, whether they arose as an effect of meditation
or were cultivated in their own right.
The exclusion of magical practices and powers from most discussions of Buddhism in the modern era can be seen as part of the
appropriation of Buddhism by Europeans and Americans, and
also as a result of modernization movements in Asia and within
Asian Buddhism.4 We see this same pattern of idealizing a religious tradition, focusing on the rational to the exclusion of ritual,
in other traditions as well. For example, the study of Jewish magic
has only recently developed after more than a century of scholarship had largely ignored its existence. The author of Early Jewish
Magic, Gideon Bohak, has discussed how a cache of ancient Jewish manuscripts found at the the Cairo Genizah—the storeroom
of the Ben Ezra synagogue in Cairo—was studied in terms of its
literary and documentary value, while the thousands of magical
texts in the collection were treated as if they did not exist. Musing
on this, Bohak writes:
In the study of Judaism, the assumption that magic has
nothing rational about it had one obvious implication—
magic is an intrinsically un-Jewish activity. If some Jews
Introduction
I
5
dabbled in it, they must have been of the lower classes, the
uneducated masses, those Jews whose unseemly practices
the enlightened religious establishment was grudgingly
forced to endure.5
There has been a similar dynamic in the study of Buddhism.
Scholars have had their own prejudices and preferences for the rational philosophical text and the documentary source. Many Buddhists too have been keen to present their own religion to modern
outsiders as rational and free from superstition. Where magical
practices cannot be avoided, they have often been dismissed as
corruptions of the original, pure Buddhism or as accommodations
to the superstitious needs of the uneducated Buddhist masses.
Even those who have observed magical practices performed by
Buddhist monks at firsthand have often assumed that these were
local traditions that had been absorbed by Buddhism, rather than
an integral part of Buddhist practice.
Another barrier to understanding how fundamental magic is
to mainstream Buddhist practice is academic specialization. Buddhist studies, relying heavily on transmitted texts, usually from the
canon, have not engaged deeply with manuscripts except where
these help with understanding the transmission of canonical literature. The observation of the actual practices of Buddhists “on the
ground” has been left largely to anthropologists, who rarely engage
deeply with the history and texts of Buddhism.6 So there has been
very little study of Buddhist magic users or their books; however,
this is changing. In a recent book on Buddhist magical practice in
Thailand, Justin McDaniel makes the case for incorporating magical practices embedded in vernacular Thai Buddhism into the
mainstream study of the religion, observing that “in the twentieth
century, with the rise of distinct academic fields such as religious
studies and Buddhist studies, scholars in these fields have almost
universally ignored the study of vernacular Thai literature.”7
6
I
Introduction
McDaniel also points out that the rough-and-ready manuscripts
that often contain magical texts have been neglected in favor of
more expensive, elite manuscripts, though both were produced
and used by Buddhist monks.8 In this book I have taken a Buddhist manuscript, one of the earliest surviving Buddhist books of
spells, as my key reference point for discussing Buddhist magic. I
am not trying to use this shift in focus to ritual and magic to minimize Buddhism’s rational, philosophical, and ethical nature. It is
a rational, philosophical, and ethical religion—and much more
besides. In traditional Buddhist cultures, our current categorical
distinction between reason and superstition did not exist. Buddhists could and did practice philosophy and meditation at the
same time as using magical talismans.
For readers who already have some familiarity with the types
of practices examined in this book, I want to point out that “Buddhist magic” as I use it here is not meant to replace more specific
categories of practice such as divination, healing, or dealing with
the world of spirits and demons. It denotes all of these, gathered
together as a package that can be physically present in a book of
spells or in the person of the magic user who provides any of these
services for those who need them. I’m suggesting that thinking
in terms of “magic” instead of these more specialized topics is
actually truer to the world of Buddhist practice. While scholarly
articles usually focus on a genre like divination, in practice divination is just one part of the wide range of expertise offered by those
who deal in worldly rituals, whether they are Buddhist monks or
lay specialists.
W H AT I S M A G I C A N Y W AY ?
The word magic, if we are aware of what we are doing with it, can
help us to focus on a specific group of Buddhist texts, artifacts, and
practices. But it is important to proceed with caution when we use
Introduction
I
7
this word, because it comes loaded with a history of connotations
that may distort the very thing we are trying to understand. In
the modern study of magic as an academic subject, one of the
most influential works is The Golden Bough, published in 1890 by the
Scottish anthropologist and folklorist James G. Frazer.9
In his attempt to describe all mythological and religious systems,
Frazer identified magic as the first primitive stage in mankind’s attempt to understand and control the world. According to Frazer,
magic evolved into religion, a more sophisticated system that relied on supernatural beings. Religion in turn was superseded by
science. Frazer proposed that magic is based on the mistaken belief “that things act on each other at a distance through a secret
sympathy,” so that power can be extended over great distances by
use of a figurine representing a person or through the magician
possessing one of that person’s belongings.10 We can apply this
principle of sympathy to some Buddhist rituals, but certainly not
to all of them, so I would not want to use it as the definition of
Buddhist magic.
Frazer’s approach to magic was that of a nineteenth-century
Englishman: puzzled, patronizing, and trying to fit everything into
his own grand theory. The philosopher Ludwig Wittgenstein criticized him for exactly that:
Frazer’s account of the magical and religious views of
mankind is unsatisfactory: it makes these views look like
errors.
Was Augustine in error then, when he called upon God
on every page of the Confessions?
But—one might say—if he was not in error, surely the
Buddhist holy man was—or anyone else—whose religion
gives expression to completely different views. But none of
them was in error, except when he set forth a theory.11
8
I
Introduction
What this means, and I agree, is that magical practices are not
based on any explicit theory, and the need to develop a theory to
explain them is a peculiarly modern Western impulse. There are
philosophical theories in Buddhism, but Buddhist magical practices do not invoke or rely on them, and since Buddhist magic
is not based on theories (such as Frazer’s theory of sympathy), it
can’t be some mistaken version of religion or science. It is worth
remembering as well that Buddhist philosophy has its own movements against theory, and Wittgenstein’s comments are reminiscent of Nāgārjuna’s classic statement, “If I had a theory, then that
error would apply to me; but as I have no theory, I am free from
error.”12
The distinction between magic and religion was also adopted
by the French sociologist Émile Durkheim, who saw religion as a
shared set of beliefs held by a social group, while magicians were
lone agents, whose important relationships were with their clients.
Thus he famously concluded, “There is no church of magic.”13
Again, this very general statement does not really apply to Buddhism, where we find practices such as weather control, amulet-making, and divination practiced by monks and nuns as well
as laypeople. So, rather than reviving the idea that magic stands in
opposition to religion, we can see it as having a specific role in the
wider context of Buddhist practice.
In general, I think it’s best to avoid grand theories of magic,
whether they are attempting to describe the psychology of magical acts, as Frazer does, or their social setting, as Durkheim does.
These theories have a tendency to conflate or homogenize what
are actually diverse and distinct practices. We would do better to
look at what makes magic a useful concept in the context of Buddhist societies. We also need to make sure we are not simply imposing an entirely foreign concept upon the material we are studying;
our understanding of what we want to call “magic” should come
from the sources we are using—in this case, manuscripts.
Introduction
I
9
A BUDDHIST BOOK OF SPELLS
The key manuscript for this book is a Tibetan Buddhist book of
spells from the Silk Road site of Dunhuang. The book was written
and used in the tenth century, before being consigned to a cave
containing thousands of other manuscripts and paintings. This
cave, originally built as a shrine to an eminent Buddhist monk, was
almost completely full by the time it was walled up at the beginning
of the eleventh century. It remained hidden until its rediscovery
at the beginning of the twentieth century, after which its contents were dispersed to museums and libraries around the world.
This manuscript from more than a thousand years ago is the
inspiration for this book, and in the fifth chapter I have provided a
complete translation of it along with my commentary.
The contents of the Tibetan book of spells provide an insight
into the scope and history of Buddhist magic:
1. Spells to cure psychotic episodes and seizures; ailments
of the urinary tract and womb; and diseases caused by
dragons, such as arthritis, tumors, and leprosy
2. Various spells as a practice of the deity Bhṛkutī, including curing illnesses caused by demons, controlling the
flow of springs and rivers, finding treasure, and manipulating the emotional reactions of others (love and hate
magic)
3. Spells as a practice of the deity Garuda, the most important being the mirror divination ritual called prasena
4. Spells as a practice of the deity Avalokiteśvara, which
are mainly medical practices and prescriptions from
the Dharani of the Blue-Necked One, including several for
10
I
Introduction
ailments of the eye, and miscellaneous spells for summoning spirits, escaping from bonds, and reconciling a
husband and wife
5. A single ritual for making rain based on the Great Cloud
Sutra, with several further steps to be taken if the ritual
is unsuccessful
6. Spells for clairvoyance, fortune-telling, and invisibility
7. Spells for curing disease in animals, loosely based on the
Dharani of Parṇaśabarī
8. Spells for conceiving a child and protecting a pregnancy,
based on the Sutra of the Great Destroyer of the Universe
9. Closing spells and prayers—including spells for the ritual specialist to aid in study and travel and protection
from infection—and a final prayer to Mañjuśrī
The various spells found under these headings are usually brief,
with clear instructions. For example, this is an invisibility spell
from the sixth section of the manuscript:
Defeather the head of a crow and fill it with seeds, then
grow them in dark soil. Then standing in front of it, pour
in the milk of a dun cow and rainwater. Once the fruits
have ripened, cut the flowers and fruits, and tie them carefully. Mash them with the milk of a dun cow and anoint
your eyes. You will become invisible.
Strange as spells like this may seem, they have their roots in
Buddhist literature found across Asia. Some of these are the
Introduction
I
11
tantras, but the sutras and other nontantric scriptures provide rich
sources of spells as well. Mantras, mandalas, and ritual objects like
protective threads were all part of Buddhist ritual practice before
the tantras became popular. For example, a spell to help a woman
to conceive a child involves setting up a mandala and leading her
to the center of it:
Clean her body and dress her in new clothes, ornament
her with jewelery, and take her into the center of the mandala. Place mustard seeds on top of her head, and stay
until midnight reciting aspirational prayers and confession. Recite the mantra of the Great Destroyer of the Universe,
holding a five-colored thread and tying a knot for each
recitation of the mantra, then untying it again. Once you
have knotted and untied the thread 108 times, write the
mantra and the name of the mudra of Kinnara, great
captain of the ogres, on paper and attach it to her.
The basic elements of Buddhist spells such as this one may be
familiar to many from Vajrayana, or tantric Buddhism. This is
not a coincidence; the literature of spells predated the Vajrayana,
and the creators of Vajrayana rituals took aspects of these magical
practices—the mantras, mandalas, and ritual implements—and
integrated them into a system of salvation. In the tantras, the deity, their mandala, the recitation of their mantra, and the use of
hand gestures and special implements are all part of a path toward
awakening. It is the power in these methods, and the recognition
of the practitioner’s identity with the deity through visualization
and recitation, that makes the Vajrayana the swift path to enlightenment.
Yet the Buddhist tantras also contain rituals for protection,
wealth, and love, not to mention destruction. This mixed nature
of the tantras, encompassing the most worldly goals as well as the
12
I
Introduction
highest aspirations, is partly rhetorical, bringing enlightenment
down to earth. But it is also the result of how the tantras developed historically, drawing on the powerful magical practices of
earlier Buddhist sutras and dharanis. Mantras, for example, began
as spells taught by the Buddha for protection from snakebite and
other threats to monks and nuns. Later, mantras became linked
to specific deities, representing the power of each deity in sound.
Later still, the use of mantras was taught as a method for enlightenment itself, rather than just for magical effect.14 Therefore one
of the benefits of understanding Buddhist magic is a better understanding of the nature of the Buddhist tantras.
MANUSCRIPTS, PEOPLE,
AND PRACTICES
I have come to see the Tibetan book of spells from Dunhuang as
a way to understand the history of Buddhism from the bottom
up. Rather than trying to understand Buddhist texts and doctrines
through the scriptures and a few key events (usually the actions of
rulers and their armies), manuscripts like this book of spells offer
insights into the everyday, local events of Buddhist monks and laypeople. The elite level of society surely had an effect on these dayto-day activities, but it did not determine or radically change the
concerns of pregnant women, farmers, and traveling merchants.
This is succinctly pointed out in George R. R. Martin’s A Game of
Thrones, when a would-be queen tells her knight that the people
are praying for her return to the throne:
“The common people pray for rain, healthy children,
and a summer that never ends,” Ser Jorah told her. “It
is no matter to them if the high lords play their game of
thrones, so long as they are left in peace.” He gave a shrug.
“They never are.”15
Introduction
I
13
The Buddhist rituals I am calling “magic” concern childbirth,
crop yields, illness, and drought, and they are aimed directly at the
concerns of ordinary people. Often administered by monks and
nuns, they also provide a key part of the interface between monastics and the laity. Since these relationships are not apparent in
most Buddhist texts, in order to perceive them, we need to use all
the information we can get our hands on—not just from the study
of individual manuscripts, but also from the evidence of archaeological sites and anthropological observations, and the close study
of individual manuscripts.
But isn’t this very specific? What can observations about specific manuscripts, ancient sites, or Buddhist communities say
about a concept as broad as Buddhist magic? This question, this
tension between specific and general, local and translocal has been
discussed in the past as the tension between the micro and macro
levels of analysis. Historians have usually considered it insufficient
to focus on documents concerned with a single person or event, an
approach sometimes called microhistory. For many historians, this
looks too much like antiquarianism, a dubious fascination with the
past for its own sake.16 But the value of looking closely at the local
has been well expressed by one of the key figures in the movement,
Carlo Ginzburg, who wrote that “any social structure is the result
of interaction and of numerous individual strategies, a fabric that
can only be reconstituted from close observation.” The difficulty
arises, he admits, when the microhistorian is challenged to apply
these local insights to the broader, macrohistorical picture:
The results obtained in a microscopic sphere cannot be automatically transferred to a macroscopic sphere (and vice
versa). This heterogeneity, the implications of which we are
just beginning to perceive, constitutes both the greatest difficulty and the greatest potential benefit of microhistory.17
14
I
Introduction
At the time Ginzburg and his colleagues were writing, solutions
to the apparent divide between the micro and the macro were
already being discussed in the field of sociology. Randall Collins
proposed a solution in his work on “interaction ritual chains,” that
is, local rituals that have a chaining effect through being reproduced across social networks. He argued that the macro level is
just the view from a distance of multiple micro-level events that
are repeated and therefore provide consistency over space and
time:
It is at the micro level that the dynamics of any theory
must be located. The structures never do anything; it is
only persons in real situations who act. It is on the micro
level that we must show energizing processes, both those
that cause structural change and those that are responsible
for maintaining and reproducing the structures from one
occasion to another (that is to say, the “glue” that holds
structures together).18
Collins’s point is that the concepts of social structure are based
on, and nothing without, local and individual people and actions.19
This brings us back to the question of how the study of individual manuscripts can reveal something about wider historical
questions. Collins’s interaction ritual chains can help us to conceptualize how Buddhist manuscripts were involved in the dissemination and alteration of the religion as it moved across Asia. If
we can start to follow the interactions between people and their
things across space and time, we start to see how manuscripts and
other artifacts can help us to move beyond generalizations about
Buddhist history and practice.20
In this book, when I move from specific manuscripts to the
wider world of canonical texts, or between rituals written down
a thousand years ago and those practiced only recently, it is in an
Introduction
I
15
attempt to create connections between the general and the specific,
to contribute to an interdependence between micro and macro.
The ritual practices of Buddhist monks at Gandhara, Gilgit, and
Dunhuang were all local and all different, yet they established patterns of practice and became part of wider Buddhist networks. If
it is not yet possible to fully explain these connections, I would argue that it is only because we still cannot see much of the network.
By pointing out correspondences across space and time, we can at
least start to imagine what that network might look like.
The kinds of Buddhist practice found in the Tibetan book of
spells and other Buddhist compendiums are an example of how
the local, micro level of interaction has created and maintained
networks that may not be immediately evident in the official scriptural collections, biographies, and histories. Buddhist monks, nuns,
and other ritual specialists used worldly pragmatic rituals to offer
services, including medicine, divination, rainmaking, love magic,
and more to local clients, in this way embedding themselves in
local economies.
These practices are still flourishing across the Buddhist world,
in Theravada Buddhist societies as much as in Mahayana and
Vajrayana ones. If we are to understand the way Buddhism has
worked in the past, the way it still works now in many places, and
the way it can work in the future, we need to pay attention to all
our sources and be willing to see the wider picture, the aspects of
Buddhist practice that most scholarship has overlooked or actively
avoided. Magic has played a key role in Buddhism’s flourishing;
understanding this, we may be able to get a better idea about its
role in Buddhism’s future.
CHAPTER 1
Magic across Cultures
Using the word magic in a Buddhist context is controversial. The
word is evocative but troublesome—most of us have an idea of
what it means, but it is notoriously difficult to define. There is no
word that exactly corresponds to magic in Asian Buddhist cultures,
so why talk about “Buddhist magic” at all?1 For a start, magic is
a lively and productive concept in the study of rituals found in
various cultures around the world. We see this in the way the word
magic is used in studies of manuscripts from all kinds of traditions,
which are often noncanonical sources that were not necessarily
meant to be preserved, and usually survive only by luck. Based on
these manuscripts, the study of Mesopotamian magic, Hellenistic
magic, Jewish magic, and the European magical practices of the
Middle Ages are very much alive and well. So why not Buddhist
magic too?
Buddhist magical practices do not fit neatly into existing theories of magic. As we saw in the previous chapter, whenever we try
to apply a grand definition such as James Frazer’s theory of sympathetic magic or Émile Durkheim’s distinction between magic
and organized religion, the Buddhist practices we are looking at
slip away. Yet there is definitely a subset of Buddhist practice that
fits very well with what people mean when they talk about magic.
So I suggest we take a “family resemblence” approach to understanding the word magic: rather than looking for a single unifying
18
I
Buddhist Magic
feature of all the practices that have been called magical, we can
look for a set of overlapping features that they share.
The idea of family resemblence as a method of forming loose
definitions through overlapping features comes from Ludwig Wittgenstein’s Philosophical Investigations. Wittgenstein’s argument is
that any definition of a word that attempts to identify an essential
meaning shared by all its uses inevitably fails. The need to identify
an essential meaning in any concept goes back to Plato and has
been a potent force in the history of Western philosophy, and its
influence is felt in nineteenth-century theories of magic.2
Arguing against all essential meanings, Wittgenstein suggested
a different approach, taking games as his example:
Consider the proceedings that we call “games.” I mean
board-games, card-games, ball-games, Olympic Games,
and so on. What is common to them all?—Don’t say:
“There must be something in common, or they would not
be called ‘games’”—but look and see whether there is anything common to all. For if you look at them you will not
see something that is common to all, but similarities, relationships, and a whole series of them at that. To repeat:
don’t think, but look!3
We can apply the same principle to the word magic, looking at
some of the more famous sources from other cultures and times
that have been studied under the general heading of magic to see
whether there is anything they share. This will help us to understand how the concept of magic has been and is still applied in
different contexts and to get a clearer view of what we mean when
we talk about magic in the Buddhist context.4
So this chapter is a tour through some of the traditions of the
world that have been studied under the name of magic. We start
with the Atharvaveda, one of the main sources for magical and
Magic across Cultures
I
19
medical practice in the Indic world, then move on to Chinese
magical rituals found in ancient wooden documents. From there
we proceed to Mesopotamian practices recorded on clay tablets
found in the old city of Nineveh, Greco-Egyptian spells from
papyri found in Alexandria, Jewish magic in manuscripts found
in a medieval repository, and some of the best-studied European
grimoires, or books of spells. This is not a complete review of
magic across the globe; the point of looking at these traditions
is that they have been and still are classified as magic. Does this
mean there is a family resemblence between them? Let’s see.
I N D I C M A G I C : T H E AT H A R V A V E D A
The earliest collection of magical spells in Indic literature is the
Atharvaveda, the fourth and last of the Vedas. The Atharvaveda
comprises a large collection of texts that have been added to over
many centuries. The core verses (saṃhita) are a series of brief rituals for various purposes that influenced many later Indic traditions, including Buddhism. These core verses are thought to date
back to the second millennium b.c.e. and to have been collected
together by the end of the second or beginning of the first millennium b.c.e. However, all the extant manuscript copies of the
Atharvaveda are much later than this, most having been written in
the eighteenth and nineteenth centuries.5
Though the Atharvaveda does contain hymns to Indo-Aryan
deities that are similar to those in the Rig Veda, most of the core
verses are brief chants and mantras aimed at specific worldly ends.
Maurice Bloomfield, who translated these parts of the text in the
late nineteenth century, organized them according to traditional
Indic categories in this way:6
• Charms to cure diseases and possession by demons of disease (bhaiṣajya). This category contains the largest number
of spells, including charms against fever, headache and
20
I
Buddhist Magic
coughing, diarrhea and constipation, hereditary diseases,
leprosy, and skin diseases. This category also contains
spells against worms, snakebite, and poison; spells to promote hair growth and virility in general; spells to drive
away ghouls; and several panaceas.
• Prayers for long life and health (āyuṣya). There are fewer
spells in this category. Most are general long-life chants,
though there is also a recipe for a salve and descriptions of
both a pearl in its shell as an amulet and an amulet made
of gold.
• Imprecations against demons, sorcerers, and enemies
(abhikārika and kṛtyāpratiharaṇa). This category includes several different spells generally aimed at repelling sorcerors
and averting their spells, as well as instructions on making amulets against sorcery. A few spells target ghouls, but
most seem to be concerned with human sorcerors. There
are also spells to counter more mundane enemies and
their designs.
• Charms pertaining to women (strīkarma). Here there are
many spells concerned with courtship, marriage, and
childbearing. There are spells to obtain a husband or wife,
spells to arouse passion in a man or woman or to secure
their lasting love, and spells against rivals in love. There
are spells to be recited by a bride and groom, spells to ensure the conception of a child and to ensure that the child
is male, and spells to prevent miscarriage.
• Charms pertaining to royalty (rājakarma). These spells
address the desires of rulers. As well as chants for consecration, there are spells for a king’s success, superiority,
Magic across Cultures
I
21
and/or restoration to power, and a variety of spells to ensure success in battle. Though the subject of these spells
might seem much more exalted than the ordinary folk addressed by many of the other spells, it is worth remembering that a rāja (“king”) in ancient India would often be a
relatively minor local ruler.
• Charms to secure harmony, influence in the assembly, and
the like (sāmmanasya and others). Here are several general
spells for securing harmony, allaying discord, and appeasing anger. There is a spell against an opponent in debate,
another to ensure success in an assembly, and yet another
to cause a person to submit to one’s will.
• Charms to secure prosperity in house, field, cattle, business, gambling, and kindred matters. In this category,
Bloomfield gathered spells for a variety of ends, including
for the successful construction of a house; the sowing of
seed; protection of a crop from bad weather; the harvest;
the protection and ensurance of the prosperity of cattle;
success in trade and gambling; protection against robbers,
snakes, and other dangers; and the finding of lost property.
• Charms in expiation of sin and defilement. Here there
are a few chants for the expiation of sins, which might
seem more an aspect of general religious ritual than the
specific worldly aims of magical spells. However, this category is equally composed of more specific spells against
evil omens in the form of birds, including those that come
in the form of pigeons or owls.
• Prayers and imprecations in the interest of Brahmins.
This category comprises spells against those who oppress
22
I
Buddhist Magic
Brahmins; a prayer to be chanted by a Brahmin upon receiving a gift; and various chants relating to the importance
of giving shelter, cows, and food to Brahmins. Though
many of the spells in the Atharvaveda may not originally
have been the province of the Brahmins, the spells in this
category suggest that the master of the Atharvan spells
must also be a Brahmin.
To provide a feeling for these spells, here is one intended to remove the anger that someone else has toward oneself:
As the bowstring from the bow, thus do I take off thy anger
from thy heart, so that, having become of the same mind,
we shall associate like friends. Like friends we shall associate—I take off thy anger. Under a stone that is heavy do
we cast thy anger. I step upon thy anger with my heel and
my fore-foot, so that, bereft of will, thou shalt not speak,
shalt come up to my wish.7
Chants like this one in the Atharvaveda often suggest ritual
actions without clearly stating what is to be done. The specific
instructions may have been handed down orally, and some are
found in the Vedic commentaries, the Brahmanas. These rituals
may not have originally been specific to the Brahmins, and there
are clues in the texts themselves. A significant number of spells in
the Atharvaveda concern the traditional needs of women in love,
sex, pregnancy, birth, and child care. While in some of them the
woman is addressed in the third person, in many her voice is the
voice of the spell caster. For example, in a spell for the bride to
chant to the bridegroom, she says, “I envelope thee in my garment that was produced by Manu (the first man), that thou shalt
be mine alone, shalt not even discourse of other women!”8
On the other hand, there is no doubt that the Atharvaveda is
Magic across Cultures
I
23
the result of worldly rituals being brought under the sway of the
male priestly caste of Brahmins. As Bloomfield argued:
The Atharvan hymns as well as the Gṛhya-rites present
themselves in a form thoroughly Rishified and Brahmanized; even the mantras and rites of the most primitive
ethnological flavor have been caught in the drag-net of
the priestly class and made part of the universal Vedic
religion.9
The Gṛhya rites mentioned here are household rituals, which
were considered part of the domestic, female sphere, before they
were brought into the orthodoxy of Brahmin ritual.10 By the first
millennium b.c.e., there were experts in the Atharvaveda, known
as Atharvavedins, providing a vast range of ritual practices across
Indian society, from rulers to villagers.11 With the rising influence
of the Brahmins at the beginning of the first millennium c.e.,
the Atharvaveda became the province of the Brahmins, so that
Atharvavedins were usually Brahmins as well.12 Nevertheless, the
Atharvaveda was also studied in Buddhist monasteries and seems
to have been used as a sourcebook for medical practice by Buddhist monks in India.13
CHINESE MAGIC:
THE SHUIHUDI BAMBOO SLIPS
Among the most important sources for understanding early Chinese beliefs and practices are the bamboo slips excavated from
an ancient tomb in Shuihudi, in central China, dating back to
217 b.c.e. These slim strips of bamboo were tied together with
thread and rolled up, a form of the book that later developed into
the scroll after the invention of papermaking. The rolls include
many legal documents and books of law, as well as almanacs concerned with other matters of everyday life, including war, business,
24
I
Buddhist Magic
agriculture, marriage, childbirth, sickness, and death. While the
legal documents have received a great deal of scholarly interest,
the almanacs have often been passed over as examples of “primitive religion.”14
Much of the content of the almanacs concerns calculations
based on astrology, numerology, and other methods to identify
good and bad days to perform particular activities. Sometimes
instructions are given on averting problems with simple ritual procedures. One section, called jie and translated as “Spellbinding”
by Donald Harper, gives instructions for dealing with problems
caused by spirits and demons. The phrase “spirits and demons”
(guishen) covers a wide range of beings, often deceased people who
have not truly passed away. Sometimes the terms are explained
as spirits being the incorporeal presence of the deceased, while
demons are bodies of the deceased that have not been properly
put to rest.15
The problems addressed in the Spellbinding bamboo slips vary
from unwelcome visitations by these beings to everyday problems
that may or may not be caused by them. For example, sometimes
the demons are very insistent:
When a demon continually drums on a person’s door at
night, singing or wailing to be admitted by the person—
this is the Malevolent Demon. Shoot it with straw arrows.
Then it will not come.16
At other times, the influence of spirits or demons is only implicit:
When a person is anxious without cause. Make a peachwood figurine and rub it. On a gui (the tenth Celestial
Stem) day at sunset, throw it into the road and quickly say,
“So-and-so will avoid anxiety.”17
Magic across Cultures
I
25
Serious problems, such as the death of an infant, are also addressed; in the following case, the unquiet spirit of a drowned
child has to be dealt with:
When a person continually loses a newborn infant—this is
the Child Who Perished in Water who has taken it. Make
an ash house to imprison it. Hang a scrub-brush outside.
Then it will be captured. Slash it with the scrub-brush.
Then it will die. If boiled and eaten, it will not be harmful.18
These simple methods of exorcism addressing domestic problems are written in such a way that they can be practiced by ordinary householders rather than ritual specialists. As Harper writes:
The exorcistic methods described in Spellbinding were intended to be employed by anyone. . . . Before the discovery
of Spellbinding we did not know that this kind of magic
was part of common practice, and that it was not limited
to the shamanic specialists or the officiants of statesponsored rituals or cults. The very nature of Spellbinding
as a text that teaches people to identify demonic phenomena and deal with them expeditiously broadens our perspective on magico-religious traditions before Buddhism
and religious Daoism.19
The preceding examples show the main characteristics of
these exorcistic spells: the use of arrows to repel demons, spirit
traps to confine and kill them (thrown into the road after use),
and some brief verbal recitations. Another important strand of
early Chinese magical practice was the use of writing, with characters—and more elaborate designs based on characters—being
inscribed on wooden stakes as talismans. Instructions for making
26
I
Buddhist Magic
talismans feature in the bamboo slips from Shuihudi and other
early sources.20
Such practices predate the introduction of Buddhism to China
and the formalization of Daoism as a religious system, which happened in part as a reaction to the success of Buddhism. Both Buddhism and Daoism incorporated these early magical practices into
their texts, with the use of character-based talismans in particular
becoming a feature of Chinese Buddhism. Many books of divination, talismans, and spells for dealing with the world of spirits were
found in the library cave at Dunhuang, which was mainly formed
from the collections of local Buddhist monks and monasteries.21
M E S O P O TA M I A N M A G I C :
T H E L I B R A R Y O F A S H B U R N I PA L
Mesopotamia, situated in the Tigris-Euphrates river basin, is famous for being the place of origin of many aspects of human culture and technology. So it is not surprising that some of the earliest
magical texts are from this region. Many individual magical rites
survive in the Akkadian language, impressed on clay tablets in cuneiform script. There are also catalogs of magical texts that show
how these rituals were grouped together in a corpus.
The clay tablets containing these Akkadian magical texts are
mostly from an ancient collection, the library of Ashburnipal, who
was king of Assyria in the seventh century b.c.e. The site, the ancient city of Nineveh, near the modern city of Mosul in Iraq, was
excavated in the 1850s, and more than thirty thousand tablets and
fragments were brought from there to the British Museum. This
and other smaller collections of cuneiform texts have been the key
resource for the reconstruction of ancient Mesopotamian religion,
culture, science, and technology. As a result, we know more about
Mesopotamian society than any other ancient civilization.22
A significant proportion of the tablets contain texts categorized
as divinatory, magical, and medical. The earlier Akkadian texts
Magic across Cultures
I
27
of this kind, dating from the second millennium b.c.e., include
spells to be chanted against disease; spells to overpower and exorcise demons; spells to remove a mote from the eyes; spells against
gas, constipation, and joint pain; spells against the bites of dogs,
snakes, and scorpions; and spells against diseases in animals.
There are spells for pregnancy, birth, and newborn infants, as well
as numerous love charms to be recited by both men and women.
To give one example of the way these incantations are worded,
the following spell to overpower a demon is from a tablet found in
the royal palace at Mari in Syria:
Let me speak, may my speech be stronger than your
speech.
As wild beasts are stronger than cattle,
So may my command be stronger than your command.
As heaven is stronger than earth,
So may my command be stronger than your command.
You have tied your nose to your anus.
So there! Have I not slapped you in the face?23
This kind of ritual practice came to be the special duty of experts called āšipu, usually translated as “exorcists.” A list of rituals
with which they were meant to be familiar was widely copied and
shows how these magical texts were learned as a corpus. The list
begins with rituals for the foundation of temples and statues, and
rites to protect the king. It then moves on to spells against demons,
witchcraft, and curses, before proceeding on to more general
magico-medical spells:
The list continues with measures against evil-portending
dreams, rituals against impotency, and texts pertaining to
pregnancy, childbirth, and infants (ll. 14–15). The next
section (ll. 16–18) lists rituals and incantations against
28
I
Buddhist Magic
diseases affecting specific body parts, against nose-bleeds,
vomiting, and diarrhea, followed by incantations against
snake bite, scorpion sting, and the sāmānu (“redness”) disease (l. 19). Measures for the protection of a man’s house,
especially against epidemics, are followed by rituals to ensure the acceptance of offerings (l. 20). After ceremonies
pertaining to settlements, houses, fields, gardens, and canals come rituals against storm damage and field pests (ll.
21–22).24
This list is important because when we compare magical texts
across cultures, the way the rituals are grouped together is as significant as the content of the rituals themselves. These Mesopotamian rituals, recorded in the seventh century b.c.e. but going back
several centuries earlier in some cases, involve recitations invoking
deities or addressing demons, as well as describing ritual actions
and substances. Clearly, most of these spells are protective in nature, but aggressive magic was also practiced in Mesopotamia to
gain power over others in love, law, and trade.
As well as these recitations in Akkadian, foreign languages were
used for spell incantations, sometimes garbled so as to obscure
their meaning.25 Along with the power of these exotic incantations,
other actions accompanied the spell: (1) offerings for specific deities
arranged on and around a portable altar, a censer, and a libation
vessel; (2) purification rites such as washing and donning a clean
garment; (3) rites that symbolized the transfer of impurity and evil
onto materials that could be eliminated or substitutes that could be
permanently removed; and (4) preventive measures, such as amulets, apotropaic figurines, fumigations, and medicine pouches.26
It is impossible to clearly separate the categories of magic
and medicine in these ancient texts. The line between the two is
blurred, especially when illnesses are personified as supernatural
beings. Rituals against animal bites and stings often demonize
Magic across Cultures
I
29
those animals, and the distinction between them and demons,
which also live in the wilderness, is not clear-cut.27 As Markham
Geller has said, this is another case where our modern categories
are misleading when applied to the ancient world:
Our use of the terms “magic” and “medicine” is somewhat misleading, since ancient Babylonian scholars used
no such terminology. Healing therapy consisted of a combination of therapeutic recipes and incantations, since
recipes were often accompanied by incantations.28
As we will see in the next chapter, these words describe the case
in Buddhism equally well.
The official practitioners of magic and medicine in Mesopotamia were drawn from the priesthoods of various temples. They
held a relatively high position in society, acting at times as advisers to kings. Yet they would also have attended to people at
lower social levels, and there were probably many unofficial practitioners of magic of whom no record remains. The practice of
divination was carried out by another kind of expert not drawn
from the priesthood; most of those known from the cuneiform
tablets were working directly for the court, the local government,
or the army.29
Though these posts were generally held by men, several spells–
especially those concerning sex, pregnancy, and childbirth—appear to be written for women to recite. Many concern childbirth,
including the following lines to be recited by the midwife to the
baby:
Run hither to me like a gazelle,
Slip out to me like a little snake!
I, Asalluhi, am the midwife,
I will receive you!30