Myths of the Dog-Man

  • 79 3,758 10
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Myths of the Dog-Man

MYTHS OF THE DOG-MAN Cynocephalic wearing striped trousers, Râby, Denmark. David Gordon White MYTHS OF THE DOG-MA

6,470 3,682 22MB

Pages 383 Page size 600 x 960 pts Year 2010

Report DMCA / Copyright

DOWNLOAD FILE

Recommend Papers

File loading please wait...
Citation preview

MYTHS OF THE

DOG-MAN

Cynocephalic wearing striped trousers, Râby, Denmark.

David Gordon White

MYTHS OF THE

DOG-MAN

T H E UNIVERSITY OF CHICAGO PRESS

Chicago and London

DAVID G O R D O N W H I T E

is a visiting professor in the Department of Religious Studies at the University of Virginia.

T H E UNIVERSITY OF CHICAGO PRESS, CHICAGO 6 0 6 3 7 T H E UNIVERSITY OF CHICAGO PRESS, LTD., LONDON

© 1991 by The University of Chicago All rights reserved. Published 1991 Printed in the United States of America 00 9 9 9 8 9 7 9 6 95 94 93 92 91

5432 1

LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA

White, David Gordon. Myths of the dog-man / David Gordon White, p. cm. Originally presented as the author's thesis (Ph. D.—University of Chicago, 1988) under tide: The other gives rise to self. Includes bibliographical references and index. ISBN 0-226-89508-4 (alk. paper). — ISBN 0-226-89509-2 (pbk. : alk. paper) l.Dogs—Mythology. 2. Dogs—Folklore. I. Title. BL443.D64W45 1991 291.2'12—dc20 90-43597

© T h e paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984.

CONTENTS

LIST OF ILLUSTRATIONS

ix FOREWORD

Wendy Doniger xi PREFACE

xiii 1 H E L L IS O T H E R PEOPLE

1 2 T H E CYNOCEPHALIC SAINT

22

3 T H E CYNOCEPHALIC H O R D E S

47

4 VlSvÂMITRA AND THE DOG-COOKERS

71

5 D O G - C O O K E R S AND O T H E R B O R D E R L I N E CASES IN ANCIENT AND MEDIEVAL INDIA

87

viii

CONTENTS

6 C E N T R A L A S I A : T H E VORTEX O F C Y N A N T H R O P Y

114

7 C H I N E S E D O G - M A N TRADITIONS: P'AN H U AND THE CH'ÜAN JUNG

140

8 BARBARIANS I N A N C I E N T C H I N A

161

9 F A C I N G U P TO O T H E R P E O P L E

180 NOTES

211 SELECTED BIBLIOGRAPHY

299 INDEX

317

ILLUSTRATIONS

Following page 86

1. Two Cynocephali from the tympanum of the Basilique de Notre Dame, Vezelay, France 2. St. Christopher "the Chananean," from a martyrology 3. Deceased person and his mummy protected by Anubis 4. Group of four dog-faced baboons worshipping the sun, from Egypt 5. Iskandar builds the Iron Rampart 6. inde et Indochine (a); LaTerre Neuve (b) 7. "Cy dist de Pisle de Seilan" 8. Cynocephalic wearing striped trousers, Râby, Denmark 9. "Le Cynocefale," from a French bestiary 10. Rama removing Râvana's spear from Lakşman 11. Bhairava riding his dog vehicle 12. Horned cynocephalic demon 13. Orientalism: Hindu myth of the Churning of the Ocean of Milk 14. Parisian clothing store mannequin

ix

FOREWORD

This is a remarkably wide-ranging book, so grand in scope that it is hard to realize that it is the author's first full-length publication. It is remarkable not only for its style, which is a sheer joy to read, full of word-play and irony and passion and weird anecdotes, but for the brilliance of its central thesis and the erudition with which that thesis is developed. It is also an important book, for it deals with a subject that is of central interest to anthropologists, historians, and historians of religion today—the subject of Otherness. An awful lot of frogwash has been written about Otherness and related subjects in India and elsewhere, but Myths of the Dog-Man approaches the subject with a fresh honesty that is much needed. With a serendipitous combination of instinct and wide knowledge, David White has fixed upon a brilliant point d'appui, dogs (polluting) and dog-men (liminal and monstrous). As he guides us through the permutations of this theme through the myths of several cultures, in texts, visual images, and social attitudes, it emerges as a deeply imbedded and powerful metaphor. In this book, David White has proved not merely the intellectual convergence of ideas, as all historians of religions attempt to do, but the geographic convergence of ideas. That is, he has not been content merely to demonstrate that three different cultures (European, Indian, and Chinese) share the same image of Otherness (a phenomenon that could be explained by some sort of Jungian, or even Eliadean, assumption of universality). He has gone on to demonstrate that these three cultures refract, in a negative inversion, the mythological self-image of one, single, actual culture that existed on the border of all three (a phenomenon that can be explained through the more accessible assumptions of history and anthropology). He argues for a central source for all three sets of myths, in Central Asia. A more cautious scholar would have been tempted to rest upon this geographically circumscribed achievement, but caution is not among David White's many virtues. He sallies forth again into the wider and more dangerous territory of universal human history, indeed universal human nature. The personal fascination, the love of the subject, that has made the book so alive throughout gives it, in the end, the power to reach down to a xi

xii

FOREWORD

deeper level of human concern, as he argues for an entirely new and politically enlightened reading of the problem of Otherness in the history of religions. Myths of the Dog-Man leaves us with a stunning and deeply moving vision of the pervasiveness, and destructiveness, of our own terror of the Other. Wendy Doniger Mircea Eliade Professor ofthe History of Religions

PREFACE

If, as Mircea Eliade wrote, the task of the historian of religions is to provide the modern West with alternative cultural or spiritual perspectives to what he called the "provincial" worldview of its own political ideologues and demagogues,1 then this book is an exercise in the history of religions method. If, as Jonathan Z. Smith has written, the historian of religion is to seek to make the strange seem familiar and the familiar seem strange "in order to enhance our perception of the familiar,"2 then this book is an exercise in the history of religions method. In terms of the first of Smith's criteria, the reader is well served here: this book will make the subject of monsters, specifically Dog-Men or DogHeaded Men, a familiar one. I will leave it to the reader to judge whether I have succeeded with this book in answering to the challenge of the latter of these criteria. While I confess that it was the wondrous strangeness of European, Indian, and Chinese mythic accounts of the monstrous races that first attracted me to this subject (and so I plead guilty here to latent orientalism), it is also the case that my principal motivation, throughout the writing of this book, has been my desire to render these fantastic hybrid creatures relevant, in some way, to modem western experience. I have written this book for as broad a readership as possible. In theoretical terms, this means that it is not so much for other scholars— classicists, medievalists, Indologists or Sinologists—as for all educated persons who consider the study of the history of ideas to be relevant to an understanding of their modem sociopolitical situation. So it is that I more often take issue with the language and built-in prejudices of modern political rhetoric than I do with the theories of a handful of historians specializing in the ancient and medieval world. In practical terms, this means that I have taken care in this book, in which a dozen or so foreign languages are employed, to define the foreign and technical terms I use wherever they first appear in the text and to limit their use as much as possible within the text itself. Discussions of a more technical nature as well as surveys of scholarly literature in the various fields I touch upon are found in the extensive endnotes to the text. In this way, the xiii

XIV

PREFACE

book may be read on two levels, the one in the text and the other in the notes. It is my hope that the book will satisfy the demands of both readability and scholarly rigor. It requires a certain amount of audacity for a historian of the ancient and medieval world and reader of myths to cast his interpretive hat into the ring of modern public discourse and address questions which, directed toward a nonspecialized readership, at times overstep the bounds of classical scholarship. It is therefore most appropriate at this juncture that I express my gratitude to my professors, my mentors who, apart from guiding me in my scholarly trajectory, have also served as role models, as examples of an unconventional and at times daring and controversial ideal of scholarship. I have already mentioned the first of these, Mircea Eliade, whose characterization of the history of religions—as a synthetic discipline that cuts across the divisions of a number of disciplines (philology, history, theology, anthropology, etc.) and cultural areas in order to generate an integral description of religion as a human phenomenon—remains a lively and provocative one. I also owe an extreme debt of gratitude, once again for the models and examples they have provided me through their own synthetic, innovative, and "popularizing" scholarly writings, as well as for the invigorating spirit of their scholarship, to my teachers Wendy Doniger, Charles Malamoud, and Madeleine Biardeau. A great number of people have aided me in the research and writing of this book. First among these is Rob Campany, who translated a number of passages for me from Chinese (a language I do not know) into English and who further helped me through his careful reading and corrections in matters of style of those portions of the book which deal with Chinese traditions. I also wish to thank, for their suggestions and support, Antonio Beltran-Hernandez, John Corrigan, Jean-Louis Durand, Jean-Charles Henry, Ravindra Khare, Lars Kjaerholm, Martine Pagnier, James Wilkerson, and, of course, my parents. I am lastly most grateful to the Woodrow Wilson National Fellowship Foundation, which awarded me a Charlotte Newcombe Foundation Dissertation Fellowship toward the writing of this book, in 1985-86. 1. Mircea Eliade, The Quest: History and Meaning in Religion (Chicago: University of Chicago Press, 1969), pp. 54-71, especially p. 69. 2. Jonathan Z. Smith, Imagining Religion: From Babylon to Jonestown (Chicago: University of Chicago Press, 1982), pp. xii-xiii.

O

N

E

HELL IS OTHER PEOPLE

This is a book about monsters, which Isidore of Seville identified as monstrations (monere) or warnings (monare) of divine will.1 The question arises: Of what were monsters monstrations, and whom were they intended to warn? The many different ways in which this question has been formulated and answered are the central concern of this study. As we follow the truly fantastic peregrinations of some four thousand years of monster traditions, we will also find ourselves obliged to grapple with the fundamental question of ideology. Here, we mean ideology not only in the sense of an explicit tool for propaganda or social manipulation,2 but also in the deeper, implicit sense of a structure or lens through which a human population experiences the world and gives expression to that experience.3 In all of the traditions we will address in these pages, monsters are ideologically (in both senses) construed as marginal groups that haunt the boundaries of human, civilized space. Their peripheral location does not imply, however, that they have been of marginal concern to humans living within such bounded spaces. The important, even central role played by these borderline creatures arises from the fact that their existence, real or imagined, has done nothing less than place in question the self-identity of humans. Where does the human begin and the monster leave off? Where does inside meet outside? Where do we become them? The perennial and ideal medium for discussion of these marginal yet intimately known beings has remained the genre of myth. Myth, demoted by the Presocratics in favor of testable, provable logos to the rank of rustic "bibble-babble," continues to suffer such a low estimation, in Western thought at least, down to the present day.4 Not only is the place of myth a marginal one in our own modem, scientific quest for truth, but the places that myth takes us, as its readers or hearers, are also located on the fringes of everyday experience. In the words of Northrop Frye, myth is "an example of thought working at the extremities of human possibility, a projection of 1

2

ONE

a vision of human fulfillment and of the obstacles that stand in the way to that fulfillment."5 Even when myths relate stories whose content is apparently superficial, however, they disclose an ongoing human quest for meaning and reveal a history of wresding with such essential human questions as self-identity, finitude, and death. This is the greatness, the power of myth: it opens us up to a heightened awareness of the paradox of our existential situation, of what it is to be human. 6 And because the human condition persists, so too do such questions, and so too do the myths that attempt to answer them. Retellings of myths, in myriad variant forms, are the rule rather than the exception. Central to our notion of who we are is the question of where we came from. This has given importance and prestige to a particular class of myths: the creation myths, which relate the beginnings of our world, ourselves, and our social and cultural institutions. The combative themes of such accounts is striking. Very often, the world is created and a sacred center founded after a cosmic struggle between opposing factions, be they gods, demigods, or heroes. The world is founded out of a dualistic situation that is resolved when one camp defeats the other. Victory in combat is often identified, in such accounts, with a victory over the dark forces of chaos by a cosmocrator. Such myths factor the two into one: after the defeat or annihilation of the enemy, the victors create one world. Our modern geopolitical situation will attest to the perenniality of this schema. The force of darkness, the adversary defeated in creation myths, is often portrayed as some form of chaos. But before there can be order, there has to be a space to be ordered. This space, this gap into which an ordered world can be slipped, is chaos; chaos is thus a necessary ground for creation.7 Once defeated by the forces of order, however, chaos and its minions are generally relegated to a marginal situation and are perceived as negative. No longer a primordial ground for being, chaos becomes the threatening antitype to an ordered cosmos, a sacred power that is "sacred in the wrong way."8 When we look more closely, however, we find that such ordering is little more than the fabrication of a cover story, a legal fiction used to describe the universe in ideal, yet wholly unrealizable terms. Here we return to the question of ideology. For persons living within a society, an ideology is as passively and unconsciously acquired as language; indeed, it is acquired along with language and forms the lens through which experience is refracted or filtered. The categories a given culture or people appropriates to

Hell Is Other People

3

express the content of its experience are always already there; moreover, their use is pretheoretical and subconscious.9 This perception of cosmos and chaos as polar opposites is the ideological ground for many of the phenomena we will be treating in this study. It will be useful here to compare Mircea Eliade's treatment of the ordering of space with Jonathan Z. Smith's: One of the outstanding characteristics of traditional societies is the opposition that they assume between their inhabited territory and the unknown and indeterminate space that surrounds it. The former is the world (more precisely, our world), the cosmos; everything outside it is no longer a cosmos but a sort of "other world," a foreign, chaotic space peopled by ghosts, demons, "foreigners" (who are assimilated to demons and souls of the dead) . . . on the one side there is a cosmos, on the other a chaos . . . "Our" enemies belong to the powers of chaos. Any destruction ofa city is equivalent to a retrogression to chaos. Any victory over the attackers reiterates the paradigmatic victory of the gods over the dragon (that is, over chaos).10 Chaos is never, in myths, finally overcome. It remains as a creative challenge, as a source of possibility and vitality over against, yet inextricably related to, order and the Sacred . . . [In whatever way] the cosmography is expressed . . . safety is not guaranteed. The walls of the "hamlet" are always vulnerable to attack, and man and the gods must ceaselessly labor to sustain, renew and strengthen the power, to keep the walls under repair.11 Eliade's is a statement regarding ideology, the self-validating nature of the sacred cosmos and the concomitant negativity of that which does not participate in its integrated hierarchy. Smith, on the contrary, is interested in that which resists integration, in those problematic areas of existence that elude the security umbrella of order that ideology offers.12 In this context, creation, rather than being a once-and-for-all proposition, becomes an ongoing struggle to maintain an established order against the entropy that is ever eroding its foundations. What are the implications of chaos, of this presence that is depicted as an absence? Even when chaos has become a mere antitype for self-sufficient order, a force threatening our paradigms and archetypes, it nevertheless remains the ground for the very creativity that a religious community normally associates with the sacred,

4

ONE

the "good" side of the cosmos-chaos polarity. In Smith's words, chaos is ultimately "a life-giving power" for he who would confront it and emerge victorious. Chaos, the undefinable in existence, is always already there in every experience of wholeness and cosmos. The contours, the margins of the world we feel we know to be our own can only stand out against a background of nonworld and nondefmition. Chaos precedes cosmos: therefore the world we create is created over and against a nonworld that ever threatens to swallow up our creation. One's cosmos, kingdom, or "enclave," the "world-for-man," must first be "founded" and then revitalized periodically to distinguish it from the undifferentiated chaos from which it was originally set apart. 13 A people may protect the vulnerable walls of its enclave from the untamed forces of the outside by the ordering acts of reiterating myths and performing rites—but it always the brooding presence of the undefined and unknown that incites it to this activity of world maintenance. Even before walls are built, there is an outside—otherwise there would never have been a perceived need for a wall. 1. SIMPLIFYING A COMPLEX SITUATION: MYTHS OF THE T H I R D WORLD

Ordering a chaos that won't go away involves, above all else, simplifying and placing under a single heading a great number of elements whose only similarity is that they are difficult to classify. A sterling example of this tendency, in our modern experience, is the category known as the Third World. In a world polarized into two camps vying for the legitimate universal rule that each claims for itself, the Third World is that group of nations whose only importance is as pawns in the power struggle between the two great powers. In an earlier time, these nations would have been depicted in other terms—as colonial sources of cheap labor and raw materials, markets for manufactured goods, or the "white man's burden." In a still earlier time, they were portrayed as races of subhuman creatures dwelling on the borders of the known world and the human imagination, potential objects for conquest or evangelization. As has befitted their place on the fringes of humanity, and their insignificant place in the great powers' plans for world domination, these ideological embodiments of "small change" (think of the Pygmies) have also been fated to dwell on the fringes of mythology and ideology. It is not until after the normalization of a people's own creation and foundation myths that that people seeks a mythological explanation for the existence of people wholly outside the primal dualistic situation. Third World my-

Hell Is Other People

5

thology, like its sociopolitical situation, is a tertiary, reductionist mythology, one that often serves to legitimate domination by the real powers in the world. Among the Greeks, it was only after the Olympian gods had defeated the Titans in heaven and after Greece had established its hegemony over its barbarian neighbors that attention turned to the bizarre Ethiopians who lived across the sea. The same was the case with ancient India's and China's "empirical others."14 In ancient times and modem times, Third World peoples have often served as curiosities, mere diversions for the "civilized world." Yet, even these peoples, these marginals, these monsters, have borne meaning for the self-proclaimed center with which they were contrasted, if only as negative exemplars of cultural or societal ideals. Accounts of them might even be said to constitute a sort of rearguard action on the part of a society's ideologues. For even as their importance is minimalized, the attention that is paid to them generally attempts to answer an extremely important, yet generally unvoiced question: If we are the people the gods and heroes brought into being when they won their cosmic battle and created the world, who are they? The mytho-logical answer to this question is transparent and shallow: they aren't really people; they used to belong to our world, but were cast out through disobedience; they are the bastard creations of antigods, etc. Such solutions rather beg the question, rendering it irrelevant, and so render the people themselves who constitute the "they" irrelevant. In such a context, the others existing outside of a centered sociopolitical system are generally condemned to a damned and meaningless existence in the chaotic space beyond the pale of civilization, a space haunted by exiled criminals, the insane, real and mythical beasts, and nearly all that has a negative valence in opposition to the positive values of "good society." Yet, as Mary Douglas and others have shown, it is only through a periodic eruption of energy from the outside that the inside of a system can remain vital. The ideological domestication of chaos may be cast in terms of how a society deals with dirt, with matter that is out of place. A self-centered ideology is most particularly a matter of shunting off fundamental and troubling ambiguities, of a studied compartmentalization of lesser details for the sake of establishing order, or a semblance of order. Here, the danger and power of ideological dirt lies in its potential for boundary transgression: uncategorized, it threatens to slip between the cracks to contaminate the heart of the system. Ordering is an endless task of differentiation and definition, through organizing activity.

6

ONE

In this task, exaggerating differences is an ideal way to create order and place ordered contours in high relief. The sort of wholeness that is the goal in this context can only be realized through an "obsessive"15 repetition of significant details, and in this respect, religious man's concern with maintaining a sacred cosmos most closely resembles the anal retentive's need for absolute cleanliness: "Holiness is exemplified by completeness. Holiness requires that individuals shall conform to the class to which they belong. And holiness requires that different classes of things shall not be confused."16 In this perspective, threatening chaos must be flushed out from within the ideal society as well. Antisocial elements, the scum of society, also need to be dealt with. The solution to these problems is physical expulsion, the voiding of the contrary member from the system by exile from the homeland. Exile has historically been the punishment for the terrible crimes of one member of society against the body politic. It is only in recent times, since the world has grown too small for exile to be practical, that penitentiaries and prisons have been set apart, within the national territory, in order to exile criminals from society. But it has not been such a very long time since Australia, Georgia, and Guiana, for example, were places of exile from British and French society. In earlier times, when exile (or incarceration) was less common than today, this punishment coincided with a highly developed symbolic system. When the pollution contracted by the commission of a violent or heinous act was inexpiable within the existing system, the only solution that remained was to cut off from the social order the offending member, both principle and bearer of dangerous contagion. The exiled criminal was called "sacred" (Latin sacer, Greek hieros) because the stain of his crime was such that execution within society might only spread the contagion. The criminal was thus expelled and led without arms or provisions to the frontier of the national territory. The task of killing him was left to foreigners or savage beasts.17 From here, it was but a small step to identify the outside and the foreigners who inhabited it with criminality, savagery, and impurity. This was quite explicit in the ancient Greek use of the pharmakos, the scapegoat, whose human representative, either a criminal or a slave, carried the dangerous contagion of a society's sins outside of the body politic and beyond the limits of its political borders. The same was even more explicit in medieval European ideologies regarding outlaws. Called werewolves, they were

Hell Is Other People

7

exiled to the regions where the wolf prowled and were open game to any who would encounter them. 18 Through a contagious play of associations, the outside, as a place of exile, comes to have a defiling effect on those who venture into it: living in a marginal space renders one marginal. Sophocles' Philoctetes is a fine example of the ambiguity of this relationship between inside and outside, between citizen and exile, between homeland and wilderness. In this drama, the punishment appears to occur before the crime—that is, an innocent Philoctetes is struck with an inhuman, polluting contagion that renders a continued "political" life an impossibility. Once exiled in the savage wasteland (agrios),19 he becomes as wild and inhuman as his environment, as his place or nonplace of exile. He has been condemned to hear no human voice, the desert land makes it impossible for him to cultivate any sacred grain, and he would be a victim of the ferocious beasts that he hunts were it not for his wonderful bow.20 Philoctetes' plight is that of the archetypal social pariah or barbarian whose wildness is as much a result of his exile as his exile is of his wildness, but whose presence at a distance is tolerable as soon as he proves useful to the center.21 In ancient and medieval India and China, as well as Europe, foreigners, generally termed barbarians, are in fact cast as the descendants of criminals exiled from society. These foreign populations are cut off from the benefits of the sacred center as the result of inexpiable crimes committed by their ancestors, against their people or gods, in a primordial time. As such, they deserve nothing from the centers which they or their forefathers have betrayed, and they live out their punishment in the nonplace of the wilderness. Any war against them is a therefore a just war, since they are the descendants of criminals, oudaws, and outcastes. In India, for example, outcastes are bom to their low station as the result of crimes committed either by an ancestral untouchable (usually a highcaste brahmin or king cursed to his fallen state by a brahmin), or by their parents or some forefather, through miscegenation, the dangerous mixing of "coded substances."22 However, once bom into the state of chronic, occupationally impure "commuter exile," these children of sin and crime are condemned to be further mired in dangerous contagion through the performance of their traditional duties, which consist of disposing of the dead and the excrements of the higher castes. In this case, a large segment of the population is voided from society but allowed to reenter the social sphere for the purpose of voiding that sphere of its own self-generated pollution.

8

ONE

For these people, these soulmates of Philoctetes, the teachings of the Buddha seem especially appropriate: existence is indeed a wound, and a festering one at that. 23 This can present problems in social or public settings where an ideology depicts society as a perfect system, a seamless whole. The Hindu solution has been to say at once that there are no blemishes on the perfect order; that the blemishes that exist on the perfect order are due to infractions of that order; that those who have stained society with their crimes are no longer a part of society; and that that part of society that voids the public sphere of its (nonexistent) detruitus does not really exist ("there is no fifth caste").24 This paradox, of an outside that is vital to the health of the inside even as it endangers the inside by its contagion, is part of the ambiguity of the spatial dimension of the wilderness, desert, forest, or ocean—so many Utopias (from the Greek ou topos, "nonplace")—for the traditions we are treating in this study. 2 . WILDERNESS AND UTOPIA

In many traditions, wilderness as a place of exile was identified with the desert. For the Jesus of the New Testament, the desert was a place of danger and of temptation, a place to face the enemy and engage in combat with him. In the time of the Apocalypse, the desert would become a place of refuge.25 And so it remained, a place of refuge but also one of testing, down to the fourth century, when saints like Anthony and Jerome ushered in a Christian tradition of hermetic (from the Latin eremus, "desert") monasticism. At the same time, the desert was a place of terrifying visions, of temptations, and of heroic combats with Satan.26 The religious hermit left civilization not just to find a place for quietistic contemplation, but ultimately to prepare for the final battle and for the world to come at the imminent end of time. As Christianity spread west, into the wilds of the European continent, the desert wilderness came to be replaced by the forest primeval, a forest so dense that light barely penetrated to its floor, a wilderness devoid of paths or guideposts. The forest wilderness held its own trials, not the least of which was the dread presence of predatory wolves. These, along with "savage" peasants, came to constitute the minions of the devil in descriptions of this new wilderness setting. 27 In this new context, the location of a sacred Utopia, the place in which one prepared for the end of time, became the edge of the forest. The word "forest" resonates with the Greek qgrios. It is etymologically related to the

Hell Is Other People

9

Latin word vastum, of which the principal significations are emptiness, aridity, and the absence of cultivation. Opposed to the world, to inhabited areas where human culture and society thrived, was the vast uninhabited emptiness of the forest. This polarity replaced the ancient urbs/rus (city/country) opposition in the European Middle Ages. In this context, savagery (Latin silvatica, from silva, forest) was not wholly inhuman, but was located at the absolute limit of human activity. Nevertheless, many crossed this fundamental boundary in the Middle Ages. Besides the monastic hermit, there were kings for whom the forest was a hunting preserve, errant forest-dwellers who eked out a foraging existence, social marginals, the criminal, and the insane. In courdy literature, the forest became a place of adventure, where heroes encountered wild men and savage beasts—and where the distinction between the two was quite blurred. The uninhabited forest, the medieval wilderness, was at once a place of exile, evangelistic mission, adventure, penance, and asylum; a place of terrible fascination to all who lived hemmed in by its dark presence.28 It was here, moreover, that most mythic accounts of monstrous persons or races were set: they had to be, since forays in the opposite direction would have constituted invasions of swarming hordes, a subject we will approach in due course. As in medieval Europe, the wild forest was the antitype of the inhabited world in epic and Puranic India. In ancient China, it was the untamed floodwaters that threatened ordered civilization; in the ancient Near East, as we have seen, the desert was identified with the wilderness. It is extremely important to note here that these spatial antitypes for culture, society, or civilization are—in spite of their ideological devaluation to nonlocations—natural givens that are always already there at the foundation of any human settlement, and are always part of the experience of those inhabiting the settlement once it has been founded. The desert blows in over the tilled fields or the thresholds of every sedentary community, covering it with its arid, nonarable sands: it must constandy be swept out, voided from inhabited places, lest it swallow them entirely. Likewise, the untamed floodwaters of the great Asian river systems brutally inundate and devastate all that human populations have cultivated and built during periods of respite between floods. The forest, when left untended, grows up over every field and habitation, completely engulfing them. One need only think of all the monuments of lost civilizations that have been uncovered from the midst of jungles and deserts or at the bottom of seas in the present century. The chaotic forces of the periphery precede, define, and constandy encroach upon the center's limits.

10

ONE

Medieval Europe was bounded, in fact, by two other sorts of yawning gulfs: the Indian Ocean to the east and the Atlantic Ocean to the west. More than physical or geographical boundaries, these oceans constituted the limits of the medieval imagination, symbolization, and ideology.29 It was on these blurred edges of the inhabited Christian world that waking and other sorts of reality met and penetrated one another. The Celtic saint Brendan sailed out to the islands of "Hell" and "Paradise" in the Irish Sea, where he met with monsters and wonders in his search for the ultimate. The mythic imagery of this and other Welsh and Irish journeys into the beyond gave rise, in the late twelfth century, to the idea of purgatory as a place. When, in the fifteenth century, these western and eastern oceans were braved and their limits charted, the crossing of these ocean frontiers did nothing less than end the Middle Ages.30 Much more powerful than the western ocean, however, was the Indian Ocean, in the words of Jacques LeGoff a "dream horizon" to the medieval mind. Not only was the Indian Ocean the sea across which were found the fabulous and horrible wonders of the East, but the ocean was itself an uncanny geographic entity. For, from the time of the ancients, and in spite of the works of the cooler heads of Ptolemy and Strabo, the Indian Ocean was considered a mare clausum, an enclosed body of water, an inland sea.31 The enclosed ocean was at once a without and a within, protean, möbius, a place of simultaneous claustrophobia and agorophobia. As a world turned in on itself, it was vertiginous in its internal immensity.32 The western shore of the Indian Ocean was at once a door that led to the outside and a door that led to another, a deeper, inside. And so it is with any marvel: the Latin mirabilia (marvels, wonders) tells us a great deal, for its root is the same as that of mirror—a marvel is a mirror to man's soul as well as to the corners of his imagination, a mirror that throws back a reverse or upside-down image of that which it reflects. The wonders of the East, on the far shore of the Indian Ocean, held all of these implications for the medieval mind. In this sense, our monsters and fantasies became internalized in the West centuries before Freud.33 In fact, to the east, the Indian Ocean was not the only limit, to the medieval mind, of waking reality. Another presence, this one political and religious, also closed off the world to the east. This was the Muslim world, which was by turns portrayed as the world of the demonic, of the army of the Antichrist against whom the Crusaders were sent to do battle; and a place of great wealth and refinement. In the first centuries of Islam, the Muslim world, with its capital at Baghdad, was portrayed as a perfect mir-

Hell Is Other People

11

ror image of Christendom in the West. Its sacraments were performed backwards, and its political and spiritual life was in every way upside down or reversed.34 Within these two horizons, then, the medieval West constructed its universe—its symbolization of space and of the ongoing struggle between cosmos and chaos, its ideology. 3. WALLS AND THRESHOLDS: A CANINE EXCURSUS

The parallel, in the world of animals, to the invading chaos of flooding rivers, advancing deserts, and creeping forests are creatures that swarm. This is a point that Mary Douglas makes at the end of her discussion of the dietary prohibitions of the Old Testament book of Leviticus.35 According to her, swarming creatures are the greatest of abominations, for the phenomenologically obvious reason that swarming is a particular characteristic of parasitic and carrion-feeding worms and insects. Swarming is thus associated with the contamination of death and putrefication, with the processes by which an integral being is reduced to nothingness, especially through the combined action of a great number of uncontrollable independent agents. The swarming creatures of the grave may well provoke a shudder of tremendumy perhaps even in the most positivist of analytic thinkers, because they represent the unbridled force of chaos that must inevitably swallow him up, in spite of his willing and acting. Our bodies are on loan from chaos; the integrity of our selves is but a fleeting moment between womb and tomb. The ethnic equivalent of swarming creatures in the ideologies of numerous traditions is the barbarian horde. Horde is a Mongol term, originally applied to a military encampment.36 In the West, however, horde quickly came to mean a swarming, uncontrollable, antinomian mass, generally of foreigners, of people who inhabit the upside-down worlds of the wilderness and lands far across the sea. The same peoples are called ganas, "hordes," in Hindu sources of the same period. In China, the barbarians are quite literally portrayed as a sea (or school) of fish37 constandy washing up against the borders of the middle kingdom. In this perspective, barbarians are not people who go on military campaigns and conquer foreign lands; rather, they overrun, wash over, flood into them. The language remains constant from one metaphorical system to another: chaos "bugs" cosmos and would undermine it as termites would a house, on the levels of body, enclave, kingdom, or universe. In this context, conquering the forces of chaos involves channeling its waters,38 sweeping out its drifting sands, cutting back its trees, and walling

12

ONE

out its savage minions. The nomadic, swarming races (ethne) share the lot of those members within society who turn away from the established order—the heretics (choosers of an ideology that is not the ideology), marginals, members of an underground, traitors, and madmen. In order to maintain its claims to exclusivity in the realm of ideology, the center must pursue and punish those outsiders who dwell within its borders. Until recently, death or exile were the classic punishments for crimes against society. In recent times, incarceration—walling outsiders inside a controlled environment outside of the social sphere—has become the norm. The ambiguous connection between walling in and fencing out has a mythohistorical precedent, one that we will examine in chapter 3 of this study. This was the wall built by Alexander the Great between two impassable mountain ranges that enclosed (within a mountain prison outside of the Hellenistic world) the impure races of central Asia. Here we move from the imprisonment of recalcitrant individuals within society to the effective imprisonment or forced exile of entire ethnic groups. An important parallel to this may be found in communities who perceive the reigning order as evil and who choose to exile themselves as a means to leading a meaningful existence. In the ideology of the center, such heretics are of the same ilk as exiled or incarcerated criminals or threatening foreign nations. 39 It is this sort of reduction that makes the forces of chaos, impurity, marginality, criminality, and foreignness all synonymous. Walling out becomes the common solution to all forms of swarming, both from within and from without. The permeable wall through which one can enter and exit the home is the door, which is marked by the threshold. In a temple, the prqfanum is the space before the sanctum itself; in a "profane" dwelling, this function is filled by the threshold and the porch. Now, it suffices to reflect for but a moment to recognize that the threshold of the house (domus) is universally associated with one particular creature, whose relationship with humans has always located it on the boundary between wildness and domesticity: we are of course referring to the dog, that creature with which mankind has enjoyed one of its most long-standing mutualist symbiotic relationships. This cohabitation with a great and changing variety of the canidae, dating from neolithic times, has no doubt played a significant role in the rise of homo sapiens to dominance over our planet, in the human transformation of environment into world. 40 We humans are in fact chauvinistic when we speak of the beginning of this relationship as the domestication of the dog, since in all probability it was the dog that originally took the initiative, insinuating itself into the

Hell Is Other People

13

human pursuit of game and accepting the scraps that were its due once the hunt was successfully concluded. The dog has remained with us ever since, exchanging its skills in hunting, herding, and protecting the home for food, shelter, and human companionship.41 We cannot overestimate the importance of this relationship to the "humanization" of the human species. Over the past ten to twelve thousand years, as we have completed our biological evolution through development of culture—an evolution parallel to that of the child in its formative years, when prenatal biological development is completed through acculturation 42 —we humans have grown up with dogs at our sides. It is by virtue of dogs' share in human evolution that our stories and histories, our activities and rituals, our dreams and waking interpretations of them, are so widely "peopled" by dogs, even today. Although "Fifi" and "Fido" are pale copies of their powerful ancestors Cerberus and Sarama, these somewhat inbred descendants are eminendy present to popular human culture. There is no creature who appears in human roles as often as the dog. We find dogs everywhere, from advertising and cartoons in which different breeds and varieties of dogs play the roles of different social classes and human personality types, to theological writings in which the relationship between man and God is compared to that between a dog and its human master.43 Indeed, according to a great number of myths, especially from central Asia, the first creatures that God placed on the earth were man and dog. 44 Nowhere is this primordial and essential relationship more transparent than in our use of language. One need only think of the scholar who, hounded by deadlines, works doggedly, poring over dog-eared manuscripts so as to avoid the cynosure of his peers (the sons of bitches). When the pressure is too great, he may end up in the doghouse, and only pull himself out with the hair of the dog that bit him. The English word dog is philologically unusual. Dog is an Anglo-Saxon term from a verbal form "to dog," which originally meant "to guard." It is the term hound that is the English form of an extremely widespread morpheme whose universality may be ascribed either to the domesticated dog's introduction into Eurasian cultures from a single central Asian source, or to the onomotopoetic sound of the dog's bark. 45 Whatever the case, the word for hound is etymologically quite constant not only among IndoEuropean languages, but also in the Semitic languages and throughout the languages of central Asia, from Hungary to Tartary, and perhaps even in Chinese.46

14

ONE

Parallel to the history of human culture, the canine species has evolved and constellated in its own ways, with varieties specially adapted to the hunt, the herd, and the home developing from a limited number of such common ancestors as the dingo, jackal, and wolf. 47 We cannot be certain which early "proto-species" of the canine came to be prized over others in the early days of this history of mutualism. Many scholars have identified the original territory of the "proto-dog" as central Asia, and some point to the Tibetan or Indian hound as the living ancestor of the race of Cams familiarise This common origin has further resonances in Indo-European mythology, where a pair of canine psychopomps, bearing similar names and functions, are a leitmotiv in a wide array of mythologies.49 Also universal, from Mexico to China, is the identification of the star Sirius with a celestial dog: this dog-star is responsible for the burning heat of the dog days of August throughout the northern hemisphere. This phenomenon also bears funerary associations, since in many traditions the heat of the dog days is identified with the opening of the gates of the infernal regions and the eruption of the dead into the world of the living. In a great number of cultures, then, the pastoral, cynegetic, and protective role of the dog is extended beyond the world of the living into the world of the dead. As such, psychopomps, guardians of the gates of hell, hellhounds, and the souls of the dead themselves are often depicted as canine. In fact, it is not so much that the dog's role extends beyond the world of the living into that of the dead, but rather that the dog's place lies between one world and another. Just as Sirius's location during the dog days is on the eastern horizon immediately before sunrise (whence the term "heliacal rising"), so the place of the dog in nearly all that it does in its relationship to man is liminal. The dog guards many celestial or temporal thresholds in Indo-European traditions, including dawn and dusk, the new and full moons, and the summer and winter solstices. This translates, in the night sky, into various canine constellations through which the sun passes on the ecliptic and, because the paths of the celestial vault—the Milky Way, the solar ecliptic, the moon's orbit, etc.—may be identified with the paths of the dead, celestial dogs are often stationed at dangerous heavenly crossroads and bridges. The dog either remains stationary, on the threshold between night and day or between indoors and outdoors; or it constitutes a moving periphery, enclosing the herd that it guards from savage predators (often its cousins, wolves) or human rusders, or providing a moving horizon between nature

Hell Is Other People

15

and culture as it pursues wild game, running ahead of its master, the hunter, who follows its bark.50 4 . WOLFMEN AND D O G - M E N : MONSTROUS HUMANS OR H O M I N I D MONSTERS?

Ultimately, the dog, with its ambiguous roles and cultural values, its constant presence in human experience coupled with its nearness to the feral world, is the alter ego of man himself, a reflection of both human culture and human savagery. Symbolically, the dog is the animal pivot of the human universe, lurking at the threshold between wildness and domestication and all of the valences that these two ideal poles of experience hold. There is much of man in his dogs, much of the dog in us, and behind this, much of the wolf in both the dog and man. And, there is some of the DogMan in god. The jackal is the canine scourge of the desert, whence the Egyptian Anubis, who in the Hellenistic age became domesticated into the cynocephalic Hermanubis or Mercury. The dread denizen of the forest primeval is the wolf, whose mythic pedigree goes just as far back as that of the Egyptian Anubis. From perhaps the second millennium B.C. onwards, the wolf was identified in Indo-European traditions with both outlaws and members of the warrior class.51 Elsewhere, many central Asian peoples, peoples with whom the Indo-Europeans were in early contact, trace their ethnic ancestry back to the union of a female human and a male wolf.52 These three themes no doubt interacted to produce the mythology of lycanthropy, or wolfmanism. While this tradition certainly influenced that of Dog-Men (cynanthropy) and of dog-headed men (cynocephaly), these mythologies drew, as we will show, on descriptions of such disparate creatures as the baboons of Somalia and the sea lions of Kamchatka. Thanks in great part to Hollywood, wolfmanism is more immediate to our popular mythology than cynanthropy or cynocephaly. Dog-Men, however, have historically enjoyed a much richer and more varied mythology than their lycanthropic cousins, no doubt as a result of the ambiguity of the dog itself as much as of cultural or historical trends. Just as the dog dwells on the boundary between domestication and savagery, so foreign or barbarian races have inhabited a space, in the human imagination, between the exotic and therefore fascinating, and the horrifying: animal freedom is as fascinating as animal savagery is terrifying, and the ethne of ancient and medieval mythology, geography, and historiography were as much animal races as they were human. 53

16

ONE

In the final analysis, "thinking with animals"54 leads us to a series of five categories that shade from wildness into humanity. These are (1) the wolf or wild dog, which is entirely savage; (2) the domesticated dog, which dwells on the threshold between the two; (3) the Wolfman, who oscillates between humanity and animal savagery—he is human inasmuch as he was bom into a given society, but animal in that he has willfully rejected social norms; (4) the Dog-Man, a hybrid creature who, while more human than the domesticated dog, is nonhuman in the sense that he belongs to an "other" or foreign race, yet human in his social behavior; and (5) civilized or political man, who is fully human in his physiology and sociability.55 A number of strategies for comparing thesefivecategories have persisted throughout history. The taxonomy of human and animal races has been an ongoing endeavor in India, China, and the West for nearly three millennia. Smith has shown that four sorts of comparison of the human races have been used in the West since antiquity. These are the ethnographic, the encyclopedic, the morphological, and the evolutionary, all of which are concerned with generating polarities like "we/they" or "man/not man." In every case, encountering the other gives rise to speculation about what is normal; the two extremes of such comparative pursuits are parallelomania and paradoxography.56 Paradoxography (the literature of paradoxes) belongs under the heading of encyclopedic comparison, and is the genre par excellence for descriptions of the cynocephalic monsters who will be the subject of the chapters that follow. Another term used for the study of freaks of nature is teratology, derived from teras, Greek for wonder, potent, sign, or monstrosity. Modern paradoxographical and teratological approaches can be found in the writings of Sir James Frazer, Ripley's Believe It or Not, and perhaps this study.57 Here, we will most closely follow what Smith calls the morphological model, with all the strengths and weaknesses that that model inherited from the Romantics.58 The French anthropologist Dan Sperber has also given some attention to the question of the normative and the aberrant in systems of classification. 59 Responding in particular to the writings of Mary Douglas, Sperber suggests that the ideal creatures of taxonomic classifications are precisely that: perfect animals that are not found in the real world. Here he introduces a brilliant metaphor for the various comparative and taxonomic approaches we have been discussing. The perfect exemplars of given animal races find their proper place, in the domesticated animal world, in zoos

Hell Is Other People

17

where they are displayed in a natural setting analogous to that over which they rule in the wild. At the circus, one finds perfect animals in a setting in which they are totally dominated by humans (the lion tamer in the center ring, for example); but one also finds, on the periphery of the big top, the menagerie, in which they are displayed in cages like framed museum pieces. The dolphin show is a case in which it is uncertain whether it is the tamer who is dominating the wild animal or vice versa. Finally, the traveling fair or carnival is the setting for freak shows, in which the monstrosity of anomolous creatures is the great attraction. (Although Sperber does not direcdy address the matter, it is likely that he would agree that the human freaks shown in circus sideshows play the same role as carnival freak shows). In the first and third cases, the animal occupies center stage and is exhibited as a perfect exemplar of its species or race. In the second (in the case of the menagerie) and fourth, the creature is out of its element, located on the fringes of a human setting and shorn of its dignity as an individual living creature. While the zoo and dolphin show exhibit animals in a manner that respects their integrity as creatures, the menagerie and especially the carnival freak show are venues for a strange sort of voyeurism.60 Such social forms of classification may be easily transferred to the ancient and medieval evaluations of the various human races. To the exemplary animals of the zoo we may compare the wild but noble Scythians who so fascinated yet troubled the Greeks. The nearly human, perhaps superhuman, dolphins reproduce the ancient and medieval view of the brahmins of India, who were considered by the West to be a race apart, and whose exemplary life placed them close to if not within the Earthly Paradise. The circus animals of the ancient and medieval world were the various barbarian races that the West sometimes had the good fortune to vanquish and dominate. And the monstrous and hybrid races are the freak show. In the annals of ancient and medieval paradoxography and teratology, no monstrous race was as perennially fascinating as the hybrid Dog-Men, and there is a remarkable similarity in accounts of these fantastic creatures throughout the mythologies of the world. What is striking in such myths is that they are nearly always reported by a second or third party, on the basis of second- or thirdhand information. These accounts, in fact, are often the earliest examples we have of ethnographies of foreign peoples, and thus are the start of a tradition to which we, as historians of culture or religion, or as members of societies interacting with other peoples, are the heirs. Also

18

ONE

noteworthy is the fact that the Dog-Men are the sole monstrous race that is zoocephalic: there are no cat- or lion- or horse-headed races in world mythology, although isolated creatures are so depicted at times. As we have mentioned, it is especially in central Asia that one finds ethnic peoples who trace their origins to the union of a female human and a male dog or wolf. For this reason, and because the great canine races also probably originated in this region, central Asia may be the primal vortex of the world's mythology of cynocephalic or cynanthropic races. In accounts of the canine ancestry of cynanthropic or cynocephalic races, one often finds the same cluster of major themes, which change very slighdy from one source to another: sometimes word for word correspondences betray a tradition of textual or oral transmission. What changes is the locus of these exotic creatures, and here too there is a similarity of differences. They are always located far away, beyond neighbors that are close enough to be known as enemies or allies, often over the last known mountain range or body of water. They always belong to another land, live under another sky, live according to other statutes (DogMen often cohabit and couple with gynecocratic Amazons), and speak (or bark) other tongues. 61 In the end, one arrives at a reciprocal situation. In the same way that the Europeans spoke of the riches and marvels of the East and the Chinese of the riches and marvels of the West, so each regional literature located its monsters outside, in the direction of another people, in a daisy chain of cynocephalic ethnography not unlike the intertwined beasts and monsters we find framing the pages of medieval manuscripts.62 Yet, even as they were exiled far from the self-proclaimed centers of civilization, monsters were very much present to the center as well. As stated at the beginning of this chapter, the Roman republic and empire used the term monstra to refer to all phenomena that were warnings (monitus) of the will of the gods. As such, monsters constituted a very important element of the Roman religion, which was obsessed with divining divine tempers through nature. The Roman monster, however, was not usually a member of a distant race or barbarian people. It was rather an omen or portent that manifested itself at the heart of human society, generally in the form of a monstrous or anomalous birth. 63 Such portentious appearances were terrifying to the ancienr Romans, and we find many examples of monstrous omens preceding devastating events in Latin literature, a tradition reflected in the opening act of Shakespeare's Julius Caesar.64 In Rome, the medium—the bearer of evil tidings or the vehicle of a por-

Hell Is Other People

19

was often confused with the message: messengers had short life expectancies in times of crisis, and infants born with the wrong number of fingers or graver deformities were subject to death. The Roman monster's fate was unhappy and short. In the fourth century of the common era, Rome became the seat of a Christian empire, and thus the pivot between a single almighty God and its imperial subjects. Under these changed conditions, in which the gods (and divine emperors) yielded to the one God, we should expect Roman attitudes towards monsters to change. This they did, and in a way that heralded great things for the monstrous races located on the edges of the ancient world—"saved" by Saint Augustine, the monstrous races eventually became Christendom's favorite foils in the lives of the missionary saints who pacified, converted, domesticated, and placed them in the service of the Catholic cause. For some twelve hundred years following Augustine's definitive statement on the place of the Cynocephali and other fantastic races described by pagan geographers and historians, a particular grouping of hybrid or fantastic creatures was a standard fixture in medieval and Renaissance encyclopedias. The heading "monster" nearly always included Cynocephali, Cyclops (also called Monoculi), Blemmyes (also called Acephali), Panotii (All-Ears), Sciapoda (Shadow-Feet), Antipodes, and Pygmies. A broad range of other hybrid or zoocephalic variations on the human form would vary from one source to another. Because this study is, in a very concrete sense, an encyclopedic compilation of all of such encyclopedic compilations to date, we will employ this broad application of the term monster in the same way our medieval Western forebears did. With the discoveries of the age of exploration, the West came to realize that the monstrous creatures at the limits of the known world were nearly as human as Europeans. So it is that the inhabitants of the Third World have come to be seen in the West more as foreign markets to be conquered, benighted heathens to be converted, sources of cheap labor, pawns in the greatest game, and—in cases of supreme hypocrisy—the "white man's burden." In the exclusive club that calls itself civilization, the Third World still eats in the kitchen and uses the back stairs.65 The Indians, in their depictions of peoples of problematic provenance, have eschewed the Sanskrit term for monster (adbhuta: literally, that which exceeds, ati-, being, bhü). Rather, such peoples are referred to either as demons or antigods (asuras, who are according to a folk etymology the opposites of the suras, the gods), or made out to be outcastes (antyajas, etc.), peoples at once inside and outside of human society. In general, the

tent

20

ONE

same terms are employed for both outcastes living on the periphery of the social system and for distant foreigners and foreign invaders or barbarians, a usage that betrays much of Indian category formation as regards its empirical others. These terms include babblers (Mlecchas),66 dog-cookers (Svapacas), and outsiders (Bâhyas). In India, a strong emphasis is placed on the impurity and defiling nature of such creatures, who are in fact not of the human race: this is the ideological ground for the class (varna) system. Ancient and medieval China, too, saw its distant marches and foreign seas populated by all manner of hybrid creatures. There doesn't seem to have been a Chinese term that direcdy translates as monster, perhaps because constant metamorphosis, rather than stasis, was the order of things. No creature was a monster in itself, since it existed somewhere along an ever-changing continuum between one sort of animal and another, or between humanity and animality.67 There was also a gradation in the categories imposed upon peoples living both within and without the imperial Chinese borders. Thus, there were outsiders who were considered barbarians of the inside (nei, whose borders actually touched those of the middle kingdom), and those who were called barbarians of the outside (wai). The barbarians of the outside shaded into animality, their names qualified by an "animal classifier" in their ideograms and their descriptions containing much of the zoomorphic. One particular and ancient Chinese term is most telling in this regard. This is kuei, whose earliest sense was foreign people, but which also came to mean a hybrid simian creature, demonic spirit of the dead, pupa, or, adjectivally, strange or bizarre.68 This brief overview gives a notion of the range of conceptual categories and symbolic modes that the monster incorporates in the three traditions we will discuss in the following chapters. From Iceland to Alaska, monstrous races are generally cast as oudandish, benighted, beasdike, damned, down, and dirty. Yet they are fascinating to the ideologues who depict them in these ways. It is useful to consider the constructive role that the monster plays in what Victor Turner has called "liminal situations." In his discussion of rites of passage among the Ndembu, Turner notes the tendency of that people to make certain natural and cultural features in their masks disproportionate: The outstandingly exaggerated feature is made into an object of reflection. Usually it is not a univocal symbol that is thus represented but a multivocal one, a semantic module with many components . . . If the exaggeration of single features is not irrational but thought-provoking, the same

Hell Is Other People

21

also may be said about the representation of monsters . . . Monsters are manufactured precisely to teach neophytes to distinguish clearly between the different factors of reality, as it is conceived in their culture . . . Monsters startle neophytes into thinking about objects, persons, relationships and features of their environment they have hitherto taken for granted. 69 Turner goes on to locate such uses of the monstrous in the broader perspective of the liminal in general. During rites of passage, the initiate is forced to reflect on a natural and cultural world whose givenness he had taken for granted. Such reflection renders the world at once more strange and more immediate. Thus the communication ofsacra that occurs in these initiations has three phases: the reduction of culture into recognized components; the recombination of the same into monstrous configurations; and finally their recombination in ways that make sense in the new state into which the neophyte will enter.70 We include these reflections on Ndembu religion not so much as a means of explaining the origin of hybrid monsters in world cultures through traditions of animal and monstrous totemism, as of showing how the symbol—when it is not reduced to the level of an object to be dominated— "gives rise to thought," just as the other gives rise to self.71 The balance of this study will be devoted to an examination of this phenomenon as it is exemplified in Western, Indian, central Asian, and Chinese Dog-Man traditions.

T

W

O

THE CYNOCEPHALIC SAINT

1. ABOMINABLE MEETS THE APOSTLES

In the chapter entitled "The Acts of Saints Andrew and Bartholomew among the Parthians," the Ethiopic Gadla Hawaryat (Contendings of the Aposdes) relates the following legend: Then did our Lord Jesus Christ appear unto (Andrew and Bartholomew) and say, ". . . Now depart into the desert, and I will be with you; and be not afraid, for I will send unto you a man whose face is like unto that of a dog, and whose appearance is exceedingly terrible, and ye shall take him with you into the city." And the Aposdes went forth into the desert, being exceedingly sorrowful because the men of the city had not believed; and they had only sat down for a litde space to rest themselves when they slumbered and fell asleep, and the Angel of God lifted up the Aposdes and brought them unto the City of Cannibals . . . There came from the City of Cannibals a certain man who was looking for a man to eat . . . And the Angel of God appeared and said unto him, "O thou whose face is like unto that of a dog, I say unto thee, Behold, thou shalt find two men sitting under a rock, and with them are their disciples; and when thou hast arrived at the place where they are let no evil thing befall them through thee (for they are the servants of God) . . . " And it came to pass that, when the man whose face was like unto that of a dog heard these things, he trembled exceedingly and he answered and said unto the angel, "Who art Thou? I know neither Thee nor thy God; but tell me who is the God concerning Whom thou speakest unto me?" And the angel answered and said unto him, "He Who hath created the heavens and the earth is God in very truth" . . . . . . Then the man with the face like unto that of a dog said 22

The Cynocephalic Saint 40

unto him, "I wish to see (some sign) so that I may believe in all His miraculous powers" . . . Then in that same hour fire came down from heaven and surrounded the man whose face was like unto that of a dog, and he was unable to withdraw himself therefrom, for he was standing in the midst of the fire . . . and he cried out with a loud voice saying, "O Thou God, Whom I know not, have compassion upon me, and save me from this tribulation, and I will believe in Thee." And the angel answered and said unto him, "If God saveth thee from the affliction of this fire, wilt thou follow the Aposdes into every place whithersoever they may go, and wilt thou hearken unto everything they shall command thee?" And the man with a face like unto that of a dog answered and said, "O my Lord, I am not like all the other men, and I have no knowledge of their speech . . . And if I be hungry, where shall I find men to eat? I should certainly then fall upon them and devour them . . . " And the angel said unto him, "God will give unto thee the nature of the children of men, and He will restrain in thee the nature of the beasts," and in that same hour the angel stretched out his hands and brought the man with a face like unto that of a dog from the fire and he made over him the sign of the cross and cried out unto him in the name of the Father, of the Son and of the Holy Ghost. Then straightway did the nature of the beast go forth out of him and he became as gende as a lamb . . . Then the man with a face like unto that of a dog rose up and went unto the place wherein the Aposdes were, and he was rejoicing and was glad because he had learned to know the right faith. Now his appearance was exceedingly terrible. He was four cubits in height and his face was like unto that of a great dog, and his eyes were like unto lamps of fire which burned brighdy, and his teeth were like unto the tusks of a wild boar, or the teeth of a lion, and the nails of his hands were like unto curved reaping hooks, and the nails of his toes were like unto the claws of a lion, and the hair had came down over his arms to look like the mane of a lion, and his whole appearance was awful and terrifying . . . And it came to pass that when the man with the face like unto that of a dog had come to where they had been, he found there (Andrew's) disciples who had become as it were

TWO

dead men through fear of him. Then he laid hold upon them with his hands, and said unto them, "Be not afraid, O my spiritual fathers," and thereupon God removed fear from their hearts . . . . . . Then (the Apostle) Andrew said unto him "May God bless thee, O my son! . . . But tell me, what is thy name." And the man with a face like unto that of a dog said unto him, "My name is Abominable [Hasum]." And Andrew said unto him, "Rightly thou speakest, for thy name is ever as thyself; but here there is a hidden mystery . . . for from this day onwards shall thy name be 'Christian.5" And on the third day they arrived at the city of Bartos and they sat down outside the city to rest themselves. Now Satan had gone before them unto the men of the city . . . And Andrew rose up and prayed, saying, "Let all the gates of the city be opened quickly." And as Andrew spoke these words the gates of the city fell down, and the Apostles and the man with a face like unto that of a dog entered in . . . Then the Governor commanded . . . (the townspeople) to bring hungry and savage beasts to attack them . . . and when he that had a face of a dog saw all that they were doing, he said unto Andrew, "O good servant of God, wilt thou command me to uncover my face (for he had covered it before entering)? And Andrew said unto him, "Whatsoever God commandeth thee, that do." Then he whose face was like unto that of a dog prayed saying, "I beseech thee, O my Lord Jesus Christ, Who didst take from me my vile nature . . . turn me back into my former nature . . . and strengthen Thou me with Thy power, so that they may know there is no other God beside Thee." And in that same hour his former nature returned unto him, and he became exceedingly wroth, and anger filled his heart, and he uncovered his face and looked at the people with great fury, and he leaped upon all the wild beasts that were among the multitudes of people who were gathered together, and he slew them forthwith, and tore out their bowels and devoured their flesh. And it came to pass that when the men of the city saw this act they feared exceedingly . . . and seven hundred men and three of the nobles of the city died . . . And God sent from heaven a great fire which surrounded the city, and not one of its people was able to flee from it. Then the people said, "We

The Cynocephalic Saint

25

believe and we know that there is no other God but your God, our Lord Jesus Christ, above the heavens and above the earth. And we entreat you to have compassion upon us, and to save us from this death and from the double affliction of the fire and of him whose face is like unto that of a dog" . • • And the Aposdes had compassion upon them . . . And the Aposdes drew nigh unto the man whose face was like unto that of a dog, and they laid their hands upon him, and said unto him, "In the name of our Lord Jesus Christ, let the nature of the wild beast remove itself from thee, and let the nature of the children of men return unto thee; what thou hast done is sufficient for thee, O my son, for behold, thou hast completed the purpose wherefore thou wast sent." And in the same hour the nature of the children of men returned unto him, and he became as gende as a lamb . . . And when the people and the governor saw this wonderful thing, they took olive branches in their hands and bowed before the Aposdes, and said unto them, "Have compassion upon us (and bless us) with your blessing and baptize us." And the Aposdes said unto them, "Preserve ye your souls in patience; behold, the grace of God hath descended upon you."1 This fourteenth-century Ethiopic text is a watershed of both Eastern and Western Christian legend which must ultimately be traced back to Gnostic and Manichaean sources originally written in Greek and Armenian. These first appear in Christian guise in fifth-century Nestorian legends of Bartholomew, which placed his acts and martyrdom in Lykaonia, in western Turkey, and in fourth-century Egyptian monastic legends of Andrew. The Councils of Ephesus (431) and Chalcedon (451) had exiled Nestorianism to Persia and alienated the Monophysite, or Coptic, church of Egypt from Christian orthodoxy. The Christianity of these apocryphal legends is therefore rather suspect from the outset.2 As we will show, the Nestorian church, centered in Persia and central Asia, was well situated to convey mythic traditions concerning cynocephalics, many of which originated in that region of the world. The apocryphal acts of Andrew and Bartholomew, of approximately fifth -century Nestorian origins, came to be transmitted—through Nestorian intermediaries and via the Greek language—into Syriac, Latin, Coptic, Armenian, Parthian, and Indian hagiographies.3 Drawing on addi-

26

TWO

tional Syriac and Coptic sources, the legend entered into the tradition of the Egyptian Jacobite church, later finding its way into the Arabic Synxavium (ca. 1 2 5 0 ) , from which it was translated into Ethiopic and included in the Gadla Hawaryat, approximately one century later.4 Over the same period, the apocryphal acts of Andrew and Bartholomew entered into Russian and Slavic hagiographies, and we even find echoes of them in medieval Islamic traditions.5 The origins of the above episodes are pre-Christian however, and may be traced back to cultic practices, myths, and legends originating in Egyptian, Persian, Near Eastern, Greek, Roman, Celtic, and Germanic traditions. In a broader context, this apostolic legend also draws on a number of IndoEuropean mythic motifs that recur from Iceland to India; central Asian traditions, from Siberia to the Crimea; and Pacific traditions, from Alaska to the Maldive Islands. This chapter is devoted, however, to European expressions of this truly cosmopolitan mythic motif; we shall therefore limit ourselves here, as much as possible, to a discussion of its Western and Mediterranean sources and lines of transmission. 2 . JACKAL, WOLF, AND D O G : WOLFMAN AND D O G - M A N IN ANCIENT EGYPT, EUROPE, AND ASIA

Before we begin this historical overview, however, it will be helpful to set out the salient features of this phenomenon as it has manifested itself in the West. We will show that it is possible to trace these central motifs back to the dawn of human culture, both in this world region and in general. Here, a glance at the ancient calendar would not be out of place. First, we should note that the death day of the St. Bartholomew of the Gadla Hawaryat, the first day of the Ethiopic holy year and of the month of Thot, falls at the end of the summer, on August 29. 6 In fact, nearly all of the saints of the Christian church whose hagiographies associate them with dogs or Cynocephali are celebrated between May and August, months identified with the setting and rising of the dog-star Sirius. The greatest concentration of these fall during the dog days, or canicular days, in the weeks following the heliacal rising of Sirius, which occurs on or around July 25. 7 This identification of Sirius and summer heat with things canine may be traced back, in Egypt, to Anubis, the jackal- or dog-headed guardian of the dead. Anubis's name is probably derived from ânup (desert jackal or dog), and it was under the name Anupu that he was revered at the cultic center of Cynopolis (modern Asyût, on the Nile) in the region of Cynopolitania: archaeological digs on this site have unearthed a great number of mummified jackals and dogs.8

The Cynocephalic Saint

27

Plutarch and other Hellenistic authors identify Anubis with the star Sirius and furthermore conflate him with the "canine" Greek god Hermes.9 The rising of Sirius was the pivotal moment in the Egyptian year, since the flooding of the Nile corresponded with the rising of Sirius following the sun's entrance into the zodiacal sign of Leo. Furthermore, the great Egyptian Sothic (the Egyptian name for Sirius) Year, of a duration of 1,461 solar years, was calculated according to Sirius's heliacal rising in the summer. 1 0

According to the Egyptian astrological system of decans, in which the daily march of the sun through the heavens was divided into thirty-six tendegree segments, the first decan, or the head of the decans, was the cynocephalic Hermanubis, a conflation of Hermes and Anubis.11 This identification of the threshold of both the year and the day with Cynocephali goes back to at least the time of Ramses II (about 1250 B.C.), upon whose obelisk at Luxor are figured four Cynocephali who greet the rising sun at the four doors of the celestial eastern horizon. These Cynocephali are in fact dog-faced baboons (Simia hamadryas\ as was the dog-faced Abominable of the Ethiopic Gadla Hawaryat.12 We will later discuss in greater detail the place of the dog-faced baboons and the importance of Egyptian astronomical motifs, since they brought together into a single symbol system temporal turning points, Cynocephali, dogs, death, and rebirth. A second mythic leitmotiv that contributed to western traditions of the Cynocephali is the lycanthropos: the wolfman or werewolf. This creature's origin apparendy derives from a social (or antisocial) phenomenon. From time immemorial martial brotherhoods existed among the (IndoEuropean) Greeks, Scythians, Persians, Dacians, Celts, and Germans in which initiates magically assumed lupine features. These brotherhoods, which were dreaded for their savage fury and their practice of sorcery, reappear, under different guises, throughout the entire span of Indo-European history and mythology, from the foundation myths of Rome and Persia, through the writings of Suetonius, Virgil, Geoffrey of Monmouth, and Snorri Sturlusen, down to Hollywood B-movies of the present century. Worthy of particular attention in this regard is the Iranian myth of the birth of Cyrus the Great, which Herodotus relates in his Historia. Here, the infant Cyrus is exposed on a mountainside by his parents at the order of his evil grandfather Astyages, the king of the Medians. He is found and raised by a shepherd and his wife. Cyrus's stepmother is in fact a dog: her name, Spakö, translated by Herodotus into the Greek as Kunö, means "dog." So it was that this hero and founder of a new dynasty was, according to tradi-

28

TWO

tion, suckled by a dog. This identification is furthered in another ancient source, the Cyropedia of Xenophon, which compares the youthful Cyrus to a young pedigreed hunting dog. 13 This is a mythic theme that bears dozens of parallels throughout central Asia, as well as in ancient Rome, where Romulus and Remus were suckled by a she-wolf.14 In Greece, we find references to lycanthropy in the myths of Lycaon and Dolon, in the toponym Mount Lycaon, and in the names of the gods Zeus Lykeios and Artemis Lykoton.15 In popular tradition, wolfmen gradually came to lose their fearsome aspects and to appear in a number of spring and summer carnivals as inoffensive or comic figures. Certain Christian saints as well came to be associated with wolves. These include St. Loup, St. Rocco, and, of course, St. Francis of Assisi, who tamed the ravening wolf of Gubbio. The wolfman's sinister side, however, has perennially outweighed any of his auspicious associations. In ancient Indo-European traditions, the lupine brotherhoods were generally classed together with thieves, whoremongers, wolves, demon-worshippers, and madmen.16 The medieval Church identified wolfmanism with satanism, and an eleventh-century law promulgated by Canute the Great employed the term verwulf for "outlaw." For such a person, punishment was banishment "beyond the limits of the places where men hunt wolves."17 The medieval career of the mythic "cousins" of these wolfmen, the Cynocephali, was quite a different one. First, a number of pre-Christian European peoples, from Scandinavia to the Balkans, knew of Dog-Men much closer to home: such is evidenced in the names of peoples (e.g. the Germanic "Hundigar") as well as individuals (e.g. the 12th-13th century Lombard king Cangrande).18 The cynocephalic races were much more frequently associated, however, with the Marvels of the East, and Western fascination with these faraway fairylands, places of "prelapsarian innocence,"19 insured the Cynocephali a quite happy place in the medieval European cosmos. The homeland of these Cynocephali was located on the edge of the world, in an India or Ethiopia that Western geographers considered to be a single land mass down to the fifteenth century.20 They shared this land, following the fourth-century B.C. writings of Ctesias the Cnidian,21 with a wide array of other hybrid races that included the Antipodes (Backwards-Feet), the Blemmyes (Acephalics, whose faces were set into their chests), Monoculi (Cyclops), Sciapoda (Shadow-Feet), Panotii (All-Ears), and Bragamanni (Brahmins). Although Ctesias believed he was reporting ethnological or geograph-

The Cynocephalic Saint

29

ical fact in his Indika, there can be no doubt that a number of his accounts were in fact inspired by Indian myths. Certain of these Indian myths shared a common heritage with Greek and other Indo-European monster traditions—whence perhaps their great attraction to the West. Whatever the case, the textual tradition inspired by Ctesias's account, together with the wealth of additional material reported by later travelers, came to take on a life of its own in the ancient and medieval Western world, and so it was, that for some fifteen hundred years following Ctesias, conquerers, missionaries, and adventurers looked for, and indeed said they found, Cynocephali in India.

It was not India's or Ethiopia's foreignness alone that made them prime candidates for the location of Dog-Men or canine hybrids. European dogs, or the choicest breeds at any rate, were imported from the east and south. The greyhound was Egypt's main contribution to the European breeds.22 But it was from the east that the greatest breeds, which most sparked the European imagination, came. These are first reported in a Persian context, when Herodotus relates that Xerxes invaded the Greeks with great numbers of Indian hounds in his army. He discusses the cultic importance of the dog for the Persians,23 who exposed their dead to dogs and to birds of prey. This last image undoubtedly evoked a canine breed already familiar to the West, a race which was certainly the least impressive of all such Eastern imports: the small, carrion-feeding, pariah dog, whose description, in the aftermath of battlefield scenes, is identical in Homeric and Indian sources.24 It was, however, the great Asian hounds that most impressed the West, especially the breeds known as Laconian and Molossian. The former, a light hunting variety, was probably represented in Homer by Odysseus's faithful hound Argos (Swiftfoot). The latter, more massive and used for hunting and herding, was likely descended from the shaggy Tibetan hunting dog. This ancestor of the modern bulldog lost most of its furry coat but none of its pugnacity once it adapted to the plains of Mesopotamia, where it was depicted in early bas reliefs.25 When the question was one of dogs, the Greeks identified the East with Persia or India, and this was the origin they assigned to the Molossian. Alexander's invasion of India (327-26 B.C.), while it was not the source of Greek oriental dog lore,26 certainly heightened the reputation of the Indian breeds. While a guest at the court of the Indian king of Saubhata (Punjab), Alexander was entertained with a staged combat between four dogs and a tiger. The tiger lost, but Alexander won, as a gift, 150 of these

30

TWO

tiger-hounds, which came to be known as the Söpeithes (Saubhata) breed. These dogs were said to have arisen from the cross-breeding of a female dog with a male tiger. Their pugnacity and disdain for wild animals, apart from the great cats, is described in Western sources from Ctesias to Marco Polo, as well as in native Indian traditions. 27 The ancient Greek uses of the dog were not unlike our modern uses: they were watchdogs,28 herding animals,29 beasts of war, status symbols,30 and, when not put to any use, general nuisances. But, whenever they came from India, dogs were always big: in fact, everything in India was big—it was the Texas of the ancient world. So Pliny the Elder, who devoted eleven chapters of his Natumlis Historia to praises of famous dogs, stated: "Animals grow the biggest in India. From India come dogs that are larger than all others."31 Immediately preceding this statement is another that is an apt encapsulation of classical attitudes on another, related, score: "India and Ethiopia are regions particularly abundant in miracles." And it is miracles that Pliny will go on to relate, for the next twelve paragraphs.32 Pliny was, in fact, the great pagan watershed for the mythology of the monstrous races: it was his uncritical collation of all that had preceded him that served theologians, encyclopedists, compilers of bestiaries, geographers, and cartographers throughout the Middle Ages. But it was St. Augustine who, in his De civitateDei, insured and even "sanctified" (16.8) the place of monsters, and of the Cynocephali in particular, in medieval Christianity. For Augustine took the Cynocephali and other monsters to be the inheritors of Cain's curse, and thereby the descendents of the disobedient offspring of the Biblical Noah (Ham and his sons) who were "spread over the earth" (Genesis 9.19). More importantly, he held that they were living proofs of the fact that the monstrous births that occurred in the Christian world were also part of God's great design.33 Thanks to Augustine, the Cynocephali were seen to be a part of the economy of salvation, albeit a fallen or exiled part; and so it was that they become widely allegorized and moralized as a quarrelsome, morally dumb, or even demonic race that was nevertheless redeemable.34 And indeed they were redeemed, from the fifth century onwards, by the apostolic heroes of a wide array of apocryphal traditions. Such was the case with "Christian" of the Gadla Hawaryat, whose story we reproduced at the beginning of this chapter. It was as Christianity's favorite fairyland monsters that the Cynocephali, at once within and without the borders of the universe, remained present to the medieval imagination (pi. 1).

The Cynocephalic Saint

31

3. APOSTOLIC MISSIONS TO ASIA: PARTHIANS, SCYTHIANS, AND O T H E R CANNIBALISTIC RACES

Having blocked out the three parameters of traditions of dog-men in Europe—their astronomical symbolism, their indigenous wolfman cousins, and their "Orientalism55—we now turn to the history of the cynocephalic phenomenon itself. It is our intention to write this history more or less in reverse and thereby, in a sense, to return to the "origins55 of the cynocephalic Christian hero who appears full-blown in the Gadla Hawaryat. By proceeding (or pre-ceding) in this manner, we will at once go back in time, and move from a complex sedimentation of symbols and socioreligious elements to simpler, less elaborate composites. We shall therefore first situate the Gadla Hawaryat acts of Bartholomew and Andrew in their Christian and European contexts, and then describe the general phenomenon of the Cynocephali in Western monster traditions. In so doing, we will also discuss the symbolism of the dog itself in Western pagan and Christian religion, culture, and society. This final analysis, of man's primordial expressions of his relationship with hi» first conversation partner from the animal kingdom, will permit us to intercept, as it were, certain very basic human categories in their embryonic formation. As we have stated, the cynocephalic Christian hero "Abominable-Christian55 has his origins in Eastern Gnostic, Coptic, and Nestorian traditions, later coming to be integrated into the hagiography of the medieval Church. In these legends, the cynocephalic hero Abominable constitutes an allegorical representation of the triumph of Christian ecumenicism over the benighted barbarism or paganism of foreign populations. Once converted to the faith, the monster 5s feral aspects are transmuted into human, Christian virtues—most particularly those of obedience and service. As is often the case with new converts, the domesticated cynocephalic becomes the most ardent of soldiers of the faith; and, in this role, even his savagery is sanctified. For, when the occasion arises to "evangelize55 barbarians or heretics through holy war, the cynocephalic military saint reassumes, as we have seen, his savage nature, so that he may better "baptize by the claw.55 But Abominable5s origins are anything but Christian, and this cynocephalic must in fact undergo many transformations before acceding into the privileged sphere of Christian hagiography. His legend intersects those of a great number of saints and aposdes whose apocryphal acts and martyrdoms recount, in often lurid tones, the early spread of Christianity. Included in this number are Bartholomew, Andrew, Matthias, Thomas, Mercurius, and

32

TWO

Christopher—with the last of these being the only truly cynocephalic Christian saint.35 But, as we have already intimated, many of the elements of these saints' legends are Christian adaptations of earlier, generally Egyptian, Persian, and central Asian, astronomical and mythic traditions, which were integrated into the traditions of a Hellenistic world in foment. Moving backwards along the chain of transmission from the GacUa Hawaryat, we find a version of the Coptic Acts of Bartholomew with a Latin codicil that both echoes the tide of the Gadla Hawaryat chapter and provides the necessary elements for bringing together the apocryphal hagiographies of several aposdes in their missions to various foreign peoples. The codicil reads: "These are the acts of Bartholomew who, upon leaving the land of Ichthyophages, went to Parthos with Andrew and Christianus, the cynocephalic man."36 A similar codicil is found much earlier, in the approximately fifth-century Syriac version of the acts of Matthew and Mar Andrew: it says that the aposdes converted the "City of Dogs, which is cIrqa," situated in or north of the Crimea.37 In the course of this and later chapters, we will show that central Asia, starting from the trans-Caucasus region between the Black and Caspian Seas—the center of Parthian (Parthos) strength in the Hellenistic age—was a major vortex for the West's monster lore. This same region was the heardand of the schismatic Nestorian church in Persia, which served as a changing-house for the transmission of central Asian mythology into Christian legend and hagiography. According to such early historians of the Church as Arnobius, Origen, Tertullian, and Eusebius, the missionary activities of the fledgling Church began very early in this region, in Scythia and Parthia. According to Eusebius, Thomas (who is most often identified with the mission to India) was sent to Parthia and Andrew to Scythia. In later traditions, it is Matthias who goes to the land of the Cannibals, where he is rescued by Andrew; or Andrew and Matthias who go together to the land of the Cannibals; or Andrew and Bartholomew who go to this land, which is called Parthia, but often placed in Africa.38 This hagiographic theme, of apostolic missions to a land of Man-Easters, is doubdess inspired by the classical writings of Herodotus and Aristode, who report that the Scythians were a cannibalistic race.39 This characterization certainly had more to do with Western lack of knowledge about these people than with central Asian tastes. Geographical knowledge of Scythia was equally vague. While the historical homeland of the Scythians was a specific region to the north of the Crimea throughout the Hellenistic

The Cynocephalic Saint

33

age, Scythia was a blanket term employed by geographers and historians to designate the great north that extended from the northern slopes of the Caucasus to the hyperborean edge of the world. 40 The Scythians were furthermore associated, at least from the time of Virgil, with the Amazon women, who were styled as the women warriors of Scythia.41 All of these elements, as we will show, came together in the later mythology of the Cynocephali. The Parthia whose name is attached to these acts of the aposdes is first and foremost the great empire that lay at Rome's eastern frontier for some four hundred years. The Parthians were, in fact, a central Asian people whose homeland, until the third century B.C., was restricted to an area around the southeastern shore of the Caspian Sea. In the following two centuries, they would expand outwards, serving as a bridge until the second century A.D. between a western Asia controlled by Rome, and a south and central Asia controlled by the Kushans and Sakas.42 While there is no doubt that Parthia was an important field for early Christian missionary activity, it is difficult to say, in the final analysis, who evangelized whom; for there is a distinct flavor of Zoroastrian (and, later, Manichaean and Gnostic) dualism to the apocryphal legends of the aposdes who passed through Parthia. In fact, the earliest Nestorian hagiographic sources on these apostolic missions were borrowed from the Gnostic periodoi and other hagiographies of the second to fourth centuries. Furthermore, the exiled Nestorian church, which went on to evangelize Persia after the fifth century, was declared heretical precisely because of its "gnostic" dualism on the subject of the nature of Christ. Every early hagiography of the acts of Andrew and Matthias—or of Andrew and Bartholomew—amongst the Man-Eaters locates them in the Black Sea region and involves crossing the Black Sea, from the area of Sinope on the south shore, north to the Crimea, which is generally identified as a "land of Cannibals" and thereby associated with the Scythians. The mention, in the Syriac hagiography of Matthias and Andrew, of a "City of Dogs" in 'Irqa, on the north shore of the Black Sea, is clearly the source of the Gadla Hawaryat account of the cynocephalic Abominable. While the name of the city changes, in later Greek hagiographies, to Myme, Myrmene, or Myrmecion, the Crimean venue nevertheless remains constant.43 Even though the Gadla Hawaryat relocates Parthos (or Bartos) to Africa, it is this same region, to the north of the Black Sea, that is being evoked.44 What changes, from the earlier Greek and Syriac to the later Coptic, Arabic, and Ethiopic versions of this legend, is the distance spanned in the

34

TWO

miraculous crossing made by one or more of the apostles. In the earlier versions, the crossing is of the Black Sea; later, the apostles are transported, in their sleep, from the trans-Caucasus to Ethiopia.45 This translation of the apostolic missions is made all the simpler by the fact that this region, of Colchis in the trans-Caucasus, was called Outer Ethiopia or the Second Ethiopia in this period.46 An alternative explanation for this geographical shift of the locus of the apocryphal acts of Bartholomew may be found in the biography of Nestorius. Nestorius, the former patriarch of Constantinople whose (Nestorian) creed was branded heretical at the Councils of Ephesus and Chalcedon, was himself exiled to "Oasis," in Egypt. In the Coptic, Arabic, and Ethiopic hagiographies of Bartholomew, this saint shuttles between the Black Sea region and a place called "Oasis," perhaps a mythic conflation of the biography of the patriarch Nestorius with the legend of the apostle Bartholomew.47 4 . CHRISTOPHER, THE CYNOCEPHALIC SAINT

Whatever the precise location of these apostles' acts may have been, there are certain recurring themes in their hagiographies which intersect to form a coherent cycle: the apostles' encounters with cannibalistic tribes (the Scythians or Parthians); the assistance they receive from a cynocephalic commended to them by God or Christ; and the Black Sea (or much later Ethiopian) regions that were the original field of their missionary activity. All of these motifs coalesce in the hagiographies of Bartholomew, Andrew, and Mercurius—and, to a lesser extent, Matthew, Matthias, and Thomas— around the acts of another figure. This is Saint Christopher—until recently the patron of motorcars and baby buggies48—whose life and acts, while wholly legendary, center on the Black Sea country of Bithynia and betray strong Gnostic influences. Christopher's Christian hagiography49 may be summarized as follows: He is a giant 50 belonging to a cynocephalic race, in the land of the Chananeans (the "Canines" of the New Testament [pi. 2] ), 51 who eat human flesh and whose only form of communication is barking (latrare).52 His original name, Reprebos (or Reprobus, Reprebus, or Deprebus), "the Condemned" corresponds to his nature: he is black, pagan, ferocious, and dreadful. One day, he asks God to show his power by giving him the gift of human speech: an angel sent in answer to his prayer strikes him and blows on his mouth, and he speaks Greek. Still a proud giant, he is cut down to size by the Christ child who commands that he carry him over a river.53 As he advances into the flood, the child becomes heavier and heavier: when

The Cynocephalic Saint

35

can go no further, the child reveals his true nature to him, and having learned his lesson, humbly devotes his life to Christian 4 service. He preaches the Gospel, but is especially an athleta dei fortissimo? who does battle against the armies of paganism. He is baptized by Saint Babylus at Antioch, when his appearance changes and his skin becomes white as milk.55 He is taken prisoner by the pagan Roman emperor Decius (A.D. 248251 ),56 and is martyred in Lycia, along with thousands of other Christians. It is fascinating to note that Christopher's and Bartholomew's martyrdoms are both connected with the "lupine" region of Lycia (called Cilicia in Bartholomew's martyrdom), in the southwestern region of modern Turkey. Bartholomew's tormenter was a king of Cilicia whose name, Astyages, is the same as that of Cyrus the Great's infanticidal grandfather, making these Christian martyrdoms possible retellings of a much earlier Indo-European mythology of lupine or canine ancestry.57 Christopher's martyrdom is celebrated in the Roman church on July 25, and in all of the Eastern churches on May 9. 58 It should be noted that Christopher's Greek hagiographies locate his martyrdom, at the order of Decius, not in Lycia, but rather in Marmaritarum, on the eastern border of Libya: a connection with the "Ethiopian" locus of Bartholomew's martyrdom, in Coptic traditions, is possible.59 Saint Christopher's cynocephaly is a constant theme in his Eastern iconography and hagiography, whereas he is only occasionally portrayed with a dog's head in Western traditions. His function and situation are nevertheless identical in both traditions: he is an apotropaic saint whom it suffices to gaze upon to be freed from the danger of male mort, death in a state of sin, whence his location, in gigantic representations, on bridges, church portals, and city gates.60 He is portrayed as a cynocephalic, in the iconography of the Eastern churches, down to the beginning of the nineteenth century;61 however, because the iconoclasts destroyed all early representations of the saint in the East, the earliest depictions of Christopher as a cynocephalic are to be found in Western manuscript illuminations.62 Christopher's origins are nevertheless most certainly Eastern: the earliest church dedicated to him was in Bithynia, in A.D. 452. 63 The way to Bithynia and Christendom was a long one, however, for Christopher, the only cynocephalic to ever have been granted the dignity of a name in the Western world; if we are to retrace his steps, we must go back to Persia and Egypt, by way of the Gnostics. The earliest source on the cynocephalic St. Christopher is a fragmentary Christopher

Christopher,

36

TWO

fifth to sixth century Coptic "Acts of Bartholomew" that has been reconstructed by the historian Konrad Zwierzina, on the basis of Greek sources. According to this Coptic text, which is clearly the source of the Gadla Hawaryat legend, a cynocephalic cannibal named Reprobus (or Adokimos), is sent by Jesus to Bartholomew, in the land of the Cannibals. He is converted by Bartholomew, who changes the cannibal's name to Christian; after defeating the Man-Eaters, he goes on with the aposdes to the land of the Parthians. In his conclusion Zwierzina states: "We may see here, in the 'pagan' name 'Reprobus' of this cynocephalic, that the 'Christian' of the original Coptic myth is none other than the St. Christopher of later hagiographies." This scholarly conclusion coincides with what was taken to be a commonplace in Christopher's eleventh-century Greek hagiography, and the Latin, Irish, and Provençale translations and adaptations thereof. These indicate that the saint originally belonged to a race of Cynocephali, and that his name, prior to his conversion by Babylus of Antioch, had been Reprobus. 64 Zwierzina concludes his study of these hagiographies by showing how they depicted a Gnostic martyrological motif of "indestructible life." Such "gnosticizing" themes were most often injected into Eastern Christianity through the intermediary of the Nestorian church, whose founder's Gnostic doctrine of the coexistence of radically divided divine and human persons and natures in Jesus Christ was declared heretical at the Councils of Ephesus and Chalcedon, in 431 and 451. The Nestorian church, however, thrived in a state of exile in Persia—in the same region that had been previously occupied by the Zoroastrian Parthians until the third century— down to the eleventh century. It is moreover through the lens of Nestorianism that the majority of myths of cynocephalic peoples come to be refracted throughout the Middle Ages.65 5 . T H E DARKER SIDE OF SIRIUS: DOG-STARS AND CANINE PSYCHOPOMPS IN THE ANCIENT AND HELLENISTIC WORLD

In the Hellenistic world, Gnosticism shared the stage with a broad array of syncretist cults, including Mithraism, apocalypticism, and the mystery religions. All of these were strongly influenced on the one hand by a Zoroastrian eschatology based on a dramatic struggle between powers of light and darkness, and on the other by Egyptian and Greek astrology. The forces of light and darkness, variously described as angels, archons, or spirits of light, were identified with or portrayed as controlling the stars and other heavenly bodies.

The Cynocephalic Saint

37

They were also depicted as theriomorphic. So it is that we find, in Origen's third-century apologetic work, Contra Celsus (6.30,33), a mention of the Gnostic Ophites' cynocephalic archon Erathoath, and of the Gnostic b e l i e f that "men . . . [after death] assumed the shapes of these theriomorphic spirits and were called lions, bulls, dragons, eagles, bears and dogs."66 The cynocephalic Hermanubis and Erathoath (Hermes-Thoth) are products of the Greco-Egyptian astrological tradition, and we may further glimpse, in the Coptic commemoration of Bartholomew's martyrdom on thefirstday of the month of Thot (August 29), an evocation of that animalheaded Egyptian deity whose symbolism was carried over into the Hellenistic world in the figure of Hermes Trismegistus.67 Second-century Alexandrian coins depict Hermes-Thoth together with cynocephalic apes and the caduceus,68 and another Ophite source, an "Abraxis" gemstone, depicts the cynocephalic Hermanubis holding a sceptre in each hand and standing between a half moon and a star; 69 on the other side is the archon Michael.70 These Hellenistic traditions were the sources of Christian zoomorphic depictions of the four evangelists as well: these first spread from Asia Minor and the Coptic Christians of Egypt into Sicily, Visigothic Spain, Merovingian southern France, and Celtic Ireland, with the cynocephalic Christopher hard on their tails.71 The Egyptian link to St. Christopher and cynocephaly in the Hellenistic age is the Coptic saint Mercurius, whose martyrdom falls on the same day (July 25) in the Coptic calendar as does Christopher's in the Latin church.72 Like Christopher, Mercurius suffers his martyrdom at the hands of the pagan emperor Decius. There exist, moreover, striking similarities between the acts of Mercurius related in the Ethiopic MashafaSenkesâr and those of Bartholomew and Andrew in other hagiographic sources. Mercurius's father is saved from two cannibalistic Cynocephali, who have already devoured his grandfather, when God changes their savage nature into ovine sweetness. These Cynocephali remain with him and, following Mercurius's birth, are converted to Christianity. They enter into the service of Mercurius's king, the ruler of the city of Asletes, and one of the two is martyred at the hands of an enemy king. Mercurius is made governor of Mardösaweyân after his father's death, and his loyal cynocephalic friend goes into battle on his side, at which times his original savage nature is restored to him so as to render him invincible.73 Mercurius's two cynocephalic retainers are sanctified with haloed heads in Coptic art. 74 This hagiography of St. Mercurius, or Abu Sefein, the great saint of the Coptic church, is unmistakably borrowed from the acts of Bartholomew

38

TWO

and Andrew.75 The most original and salient point of his legend is in fact the saint's name: Mercurius is of course the Latin Mercury, who is none other than the Greek Hermes, a god who, like Anubis and Christopher, was solicited by travelers for his protection. 76 The Egyptian psychopomp Anubis was a god closely identified with the dog-star Sirius, whose heliacal rising was of immense importance for the ancient world. The date of this astronomical event, the veritable pivot of the year, was fraught with both danger and promise: danger because it constituted an endpoint, the death of the year, a death that was made all the more immediate by the cruel midsummer heat of which it was believed to be the source. The canicular days were nothing less than an opening wide of the gates of hell—gates that were imagined as the maw of a dog 77 that belched forth the dead, given the form of dogs or guarded—or even herded—by dread watchdogs or cynocephalics. Indeed, the star was imagined to be a jet of flame belching forth out of the mouth of the heavenly dog, Canis major. 78 However, for the Egyptians, the rising of Sirius also heralded—with the rising of the Nile—a new agricultural cycle, a new year, new life, and even the resurrection or translation of the dead into another, happier existence: this was Sirius's promise. We will begin our telling of the dark side of this story in Egypt, where from perhaps the sixth millennium tomb entrances were guarded by black jackals.79 The archaic Book ofthe Dead (17) describes a dog-headed demon named Atum, who devoured the dead in the midst of a fiery sea. Also in Egypt, a dog- or jackal-headed divinity named Chenti Amentiu, "the first of those to the west," ruled over the subterranean necropolis of Abydos.80 In Homer, Sirius was the dog of Orion who, during the dog days "redoubled the fiery heat of the sun, bringing, in the afternoon, suffering to all living creatures."81 Elsewhere, several of the labors of Heracles may be also related to the heat of the dog days: these include the slaying of the Nemean lion (the son of Typhon, the principle of drought in Egypt; Leo is also the zodiacal constellation of the month of August); his harrowing of hell and theft of Cerberus; and his choking of Orthros, the herd dog of Geryon's red oxen. Cerberus and Orthros were brothers (along with Hydra and Chimera), the sons of Typhon and Echidna.82 In ancient Greece, the Heracleia festival fell on July 25; the same day also saw the celebration of the kunophontisy "the massacre of dogs," a sacrifice performed to appease the ancestors of Apollo's son, Linos, who was devoured by dogs. As we have seen, this is the day of St. Christopher in the Roman church calendar and that of Mercurius in the Coptic calendar.83'

The Cynocephalic Saint

39

Also on the same date, the Romans celebrated the Furinalia, with the sacrifice of russet (sea-) dogs to the goddess Furina: this goddess was none other than a Romanization of the Greek Furies (Erinyes) who, in the form of dread bitches, hounded the bodies and souls of criminals.84 In addition to the Erinyes, the Greek Ceres, as well as Hades and Hecate, the rulers of the underworld, were portrayed as dogs or cynocephalics. In a Euripidean tragedy, Queen Hecuba, having avenged the treacherous murder of her young son Polydoros by the Greek Polymestor, was cursed by the latter to end her life in the form of a dog with fiery eyes. When she died in this form, the promontory Cynos Sema, "Memorial of the Bitch," was named after her.85 In Rome the Lares, the ancestral divinities who protected the home and hearth, were conceived of as dogs or as men clad in dogs' skins; in the Middle Ages, these would come to be identified with flesh-eating cynocephalic "ghosts" (larv&\ the souls of the dead. The Roman goddess Genita Mara was offered dogs in sacrifice in order that none of the sacrificer's kin be seized by death. 86 The monstrous Scylla, "the Mangier" of the Odyssey whose name evokes the Greek term for puppy (skulax) and who howled like one (12.86), had dogs growing out of her extremities, and made dogfish her victims.87 In many Indo-European traditions, the dead are compared to a herd, shepherded or hounded by a divine or demonic dog, with a god of the dead as their herdsman. One such group was the red oxen of Geryon, who was killed, along with his red hound Orthros, by Heracles.88 So too, troops of lost souls were hounded by Hecate, in the form of a black dog; the same goddess also afflicted the living with insanity, epilepsy, disease, and despair.89 Hermes may have played the same role as psychopomp when, in the Boetian Homeric Hymn to Hermes (194-96), he stole in the night the fifty cattle of Apollo and took them to Mount Cyllene in Arcadia—leaving Apollo's four sheephounds behind!90 Indo-European dogs of the dead generally moved around in pairs, however: when they were not twin brothers, they could be represented as two dogs in one, as in the well-known case of Cerberus.91 Most often, one of these dogs was black and the other white, symbols of a path that led not only from a past life to a present death, but also from death to a future life. Among such canine twins may be counted the Indian Syâma and Sabala, the Germanic Gifr and Geri (or Frekr), the Armenian Spitak and Siaw, the two "devil dogs" of Celtic mythology, and an Iranian pair that is perhaps identified with Sraosa and Rasnu. The grey-red canine psychopomp of Welsh mythology may also be an instance of such pairing. The Indo-

40

TWO

European dead were exposed or buried with various sops to appease these dogs.92 The heliacal rising of Sirius in midsummer was not the sole dangerous pivot of the year for the ancient and medieval world. Another such moment, also fraught with canine and monstrous symbolism, was the winter solstice, which falls on December 21. In the holy year of the Roman church, this is the night of St. Thomas, yet another aposde whose legend intersects those of Andrew and Bartholomew. Thomas's earliest hagiography is the second-century Gnostic Periodoi Thöma, which depicts him as the aposde to the Parthians. All later traditions, likely of Nestorian origin, send him to India, where he is considered to be the founder of the Saint Thomas church.93 But there is another side to this Christian saint that, in popular traditions at least, transforms him into a multi-headed (three-, seven-, or nine-headed) monster who devours children on the longest and darkest night of the year. This monstrous aspect of the aposde may be traced, on the basis of artistic evidence, back to the third or fourth century; it has continued, down to the twentieth century, in Eastern European folk traditions. So it is that in Austria on St. Thomas's night people fashioned jack-o-lantern masks variously named Thomerl, Thomasschadel, Thomasnigl, Thomaszoll, Thomaskopf, or Zweibart (Double-Beard), composed of a monstrous central head around which are appended several smaller heads representing the children he devours. And, according to popular tradition, St. Thomas takes the form of a dog or cynocephalic on this night, when he is heard barking like a dog. 94 This tradition is, in fact, the conflation of at least three parallel motifs. St. Thomas represents Saturn, the monstrous figure of Time who devours his own children, and whose festival, the Saturnalia, was celebrated in Rome during the twelve nights of midwinter.95 He is also the god of the dead, depicted in Germanic traditions as the wild hunter Wode-Wodin who, accompanied by the souls of the dead in the form of dogs, roars through the dark forests during the twelve nights, the temporal microcosm of the twelve months of the year. Finally, several Indo-European parallels to this Germanic tradition will be discussed in chapter 5. The twelve nights act as a bridge between one year and another; in this context it should be recalled that the origins of the Christian St. Christopher—whose prime medieval locus was on bridges and doors— were in Bithynia, the southern shore of the Hellcspont, literally the bridge, between Europe and Asia. The name of the city of Cerberion in the same

The Cynocephalic Saint

41

region may evoke a similar motif of liminality.96 Christopher is also like Thomas in his association with lamps or lanterns. In his Western iconography, Christopher carries a lamp as he bears the Christ child across a great river. In his Ethiopic hagiography, his eyes are like great lamps; moreover, it is by unmasking his face and showing his terrible gaze that he routs the adversaries of the apostles Andrew and Bartholomew. For the French folklorist Claude Gaignebet, these are all images of the terrible burning gaze of Sirius in midsummer.97 Finally, it is a dog-headed Thomas who, in his Eastern European folk tradition, is invoked during the shaking of a tree. Children are believed to hide from the wild hunt of Thomas and his demonic minions in a tree; a great number of folktales from other regions of southeastern Europe describe how children flee from packs of cynocephalic demons by swinging across a forest from tree to tree.98 The first of these motifs, that of the child-eating Saturn, is also encountered in the case of St. Christopher who, in Hungary from at least the twelfth century onwards, was depicted as a child-devouring monster.99 Saturn himself is portrayed in medieval Europe as an evil child-eater. This role may be attributed to a confusion in the ancient world between Kronos, the Titan who castrated his father Ouranos and then swallowed his children, the Olympians, to insure that they would not usurp his power; and Chronos, the abstract divinization of time. Saturn, the Roman Kronos, inherits this confusion, and so becomes Father Time, the Grim Reaper who cuts down living creatures at the end of their allotted span of life. But Saturn is also identified with the planet Saturn, the dark, outermost planet whose revolutions, because they take the longest to complete, exert their (nefarious) effects over the longest durations. Dark Saturn, the childeater, is thus also the planetary divinity of melancholy throughout the medieval period and on into the Renaissance. Also due to his remoteness, Saturn is represented as the ruler of India and Egypt—precisely the loci of the apocryphal acts of Thomas and Christopher. In the interpretatio romana of Assyro-Babylonian traditions, it is Nergal, the ruler of hell, who is identified with Saturn. The Canaanite homologue of Nergal is Mot, the son of El, who rules hell and has a great appetite for children, who are sacrificed to him. Mot's deep throat is associated with his infernal realm; he is depicted as having "a serpent at his gate, and a dog at his scepter side."100 In Austrian popular tradition, the double-bearded St. Thomas—whose split beard, like that of Father Time, evokes both past and future—is closely associated with St. Nicholas. This saint, whose night is celebrated on either

42

TWO

December 5 or December 25, has for his animal emblem the horse, which, along with the dog and snake, was the Indo-European animal figure of the chthonic world of the dead. This saint is also depicted as a devourer of children, 101 a fact clearly lost on parents who bring their little ones to visit department-store Santa Clauses during the holiday season. St. Nicholas is, moreover, the Christian version of Wode-Wodin, the wild hunter of lost souls on the twelve nights in which the dead erupt from the underworld to mingle with the living. Numerous parallels to this motif may be found in other northern European traditions, down to the story of Little Red Riding Hood in the tales of Perrault. 102 Thomas's cynocephaly in this folk tradition, together with the motif of the trees in which his potential victims attempt to hide or escape from him, finds parallels in a great number of southeastern European medieval epic and modem folk traditions. These are a body of legends, concerning the Pesoglavci, "Dog-Heads," which are found across a wide swathe of eastern Europe, extending from the Balkans to the Baltic states. This is a living folk tradition, but its tales date from as far back as the fifteenth century—the period that the Ottoman Turks thrust themselves into eastern Europe in a time of concerted imperialism. In fact, a recurring theme throughout folktales concerning the Pesoglavci is the fact that they are dog-headed demon Turks, from the eastern edge of the world, who carry off, imprison, and eat Christian maidens and children. We will return to this identification of Cynocephali with Turks in the next chapter.103 6 . T H E BRIGHTER SIDE OF SIRIUS: T H E CYNOCEPHALIC APES OF SOMALIA

As dire as Sirius's symbolism was for the ancient Greeks, it was quite the opposite for ancient Egypt. The reason for this is not hard to find: the heliacal rising of the dog-star (Sothis) in midsummer coincided with the flooding of the Nile, Egypt's wellspring of life. 104 Therefore, Sothis-Sirius was identified with the cynocephalic psychopomp Anubis, as well as with Osiris, whose mythology identified him with the renewal of life and fertility.105 Just as the flooding of the Nile, announced by Sothis, brought about a renewal of life, so both of these deities led their followers to a new life beyond death. The Egyptian New Year began with the matinal appearance of Sothis, making it the pivot of the Egyptian year. More than this, it was the pivot of heaven and earth, and of time itself. The Sothic, or Great Year, of the ancient Egyptians, later termed the Perfect Year by Plato, was composed of

The Cynocephalic Saint

43

1461 (4 x 365.25) years and was named for the star that stood at its head, Sothis. As such, the Sothic Year was also called the "Year of the Dog." 106 And, just as the year could be magnified into the great Sothic Year, so too could it stand for the passage of a single day, as it did in the ancient Egyptian system of decans, the thirty-six ten-degree units into which the heavens were divided in charting the daily path of the sun. Here too, Sirius, deified as the dog-headed Anubis, played a leading role, as the first of the decans. In the later Hellenistic age, Anubis was fused with Hermes, the Greek god with whom he was identified, and the first of the decans was called Hermanubis. As the leader of the decans, this divinity heralded the dawning of a new day. His identification with Sirius is an obvious one: this brilliant star's position on or just below the horizon on the first morning of the new year placed it on the literal threshold of the sun's luminous entrance into the world. We also find other canine or cynocephalic divinities glorified as leading the dead or cultic initiates to a new life in the Hellenistic age. In the Mithraic cults that arose in the Mediterranean world in about the first century B.C., the tricephalic dog Cerberus played a pivotal role in the drama of initiates' rebirth into light. 107 The "Three-Horned" Hermes Trismegistus occupied a similar position in the Alexandrian mysteries. And as with the Gnostic Archons, the Decans—star spirits and rulers of the planets—were either theriomorphic or zoocephalic. As we have shown, the cynocephalic Hermanubis was a conflation of the dog-headed Greek psychopomp Hermes and the dog-star Sirius (pi. 3). 108 The greatest contribution to this Hellenistic composite, however, remained Egyptian. In the mysteries of Isis and Osiris, which remained in vogue down to the time of the Mithraic cults, Anubis was portrayed as one of the two sons of the great god, and bore the form of a jackal or dog. In the same context, we find at the cultic center Abydos gravestones inscribed: "the beauty of Upuat is the first that we see on leaving [i.e., upon resurrection]." Upuat (or Wepawat), too, is a wolf or jackal: his name means "WayOpener," since it is he who, in the Pyramid Texts (670), leads the dead over the Jackal Sea to a new life. 109 At Lycopolis, Anubis was also called the "Way-Opener" (Apherou, Oupherou): his appearance, as Sothis-Sirius, opened the way, at the beginning of the day and of the year, for Osiris.110 This brings us back, with a jolt, to the giant, cynocephalic, lamp-bearing St. Christopher with whom we opened this chapter. Christopher's name has been glossed, since the time of his earliest Western hagiographies, as the bearer of Christ (Greek Christo-phoros). There exists, however, an alternative reading of his name:

44

TWO

Christ-Apherou, "the way-opener of the Christ," a fusion of names and functions of the same order as Hermanubis!111 Christopher is not the sole medieval Christian saint to be closely associated with both the dog and the dog-star. Others include Dominic, James, Cucufat, and Guignefort. The first of these, the founder of che Dominican (Latin domini-canes) order, was bom, according to legend, after his mother had dreamt that she was carrying a little dog with a torch in its mouth in her womb. Claude Gaignebet, in his brilliant study of the esoteric in Rabelais, A plus hault sens, indicates that the torch in the dog's mouth was, for medieval commentators, a representation of the star Sirius in the mouth of the great dog, Canis major. Dominic's day is observed on August 4, 5 or 6. 112 The death day of two of the other three saints mentioned above falls on July 25, the day of St. Christopher as well as the day of the heliacal rising of Sirius. The first of these is James, the James whose pilgrimage route, to Compostello (Latin campus-stellae, "the field of stars"), was the most popular of the Middle Ages. Again according to Gaignebet, the church of St. James at Compostello, which is on the western shore of Galicia at a western extremity of the ancient world, is a Christianization of a pagan site. The "starry path" to Compostello was a trail toward the sunset, and thus toward the land of the dead. As such, it stood as the earthly homologue of the celestial Milky Way, whose gate—through which the westering sun passed— was guarded by the starry dogs, Canis major and Canis minor, and the dog stars Sirius and Procyon. On July 25, St. James of Compostello opened the gates of heaven to the souls of the dead. 113 The names of saints Cucufat and Guignefort share a common etymology. This is the low Latin root cania, "dog," which generates the prefixes Guigne-, Quinquen- and Cucu-. The connection between the two is made clear in the name Quinquenfat, which is a variant of Cucufat. Little is known regarding Cucufat, save that he was an African hermit who served as a guide to Christopher, and that his day falls, like Christopher's, on July 25. 114 Guignefort's legend is, on the contrary, highly developed and well-documented, with parallels from Wales to Thailand. In his thirteenthcentury Christian legend, documented by the inquisitor Etienne de Bourbon, Guignefort (Strong Dog) is none other than a greyhound, unjusdy martyred by his master after he has saved the life of his master's son from a serpent. St. Guignefort, although never canonized, has enjoyed a cult as a healer of childhood diseases in southeastern France (Dombes) and other parts of Europe down to the present century His motto is reminiscent of that of St. Christopher: "Great St. Guignefort, for life or for death." His day is generally celebrated on August 22, at the close of the dog days.115

The Cynocephalic Saint

45

Let us now return to Egyptian traditions. Anubis-Apherou, the wayopener for Osiris and the reinvigoration of life with the rising of the Nile, had his main cultic centers at Lycopolis, Cynopolis, and Dendera. This last site was, of course, the home of the famous Zodiac of Dendera, perhaps the greatest monumental depiction of the system of the decans ever constructed.116 Now, Anubis was originally a jackal who came to be transformed, no later than the eleventh century B.C., into a dog or a cynocephalic.117 We also know from the ancient authors that the dog— probably the greyhound, Egypt's contribution to the great canine races— was revered by the Egyptians. 118 The same ancient authors, however, also designate Ethiopia—just upriver from Egypt—as the land of the Cynocephali; but here they refer not to dog-headed men, but to a variety of baboon, the Simia hamadryas.119 The Egyptian tradition of Anubis, Osiris's way-opener, and the cynocephalic baboons, undoubtedly constitutes the earliest link in the entire evolution of this symbol system whose watershed, as we have indicated, was the Ethiopian GadlaHawaryat. In the Egyptian Book of the Dead and in inscriptions from the New Empire, there are fascinating references to these baboons, whose hieroglyph is the same as that used for the "Spirits of the East" (pi. 4). 120 Thus we read in an inscription from court 30 of Ramses Ill's Medinet Habu temple in Upper Egypt: It is the Cynocephali who herald the Sun when this great god is born in the hours of the Duat. They appear before him after he is born. They are in the Two Chapels of this god when he appears on the eastern horizon of the sky. And, in the "Book of the Day": The Spirits of the East are the four gods who worship the Sun. It is they who cause the Sun to rise when they open the latches to the four doors of the sky's eastern horizon, when light enters into the Two Chapels. They go out to meet the Sun when it rises, each day. Finally, on the tomb of Ramses VI, we find an image of the daily path of the sun which depicts the Cynocephali singing a hymn to the newly born god Re, with the inscription: These gods, Bentiu by name Have Punt for their city They are from the country of the "monkey-faces" near to the land called

46

TWO

Utenet and near to the Eastern Sea The eastern horizon is where they live. The hieroglyph for "monkey-faces" has for its exact equivalent the Greek term kunokephaloi (Cynocephali), employed by Herodotus to describe these baboons in his Historia (4.19). Punt is the coast of Somalia, of which the eastern tip is Cape Guardafiii;121 and it was on this site, at the edge of the world and the dawn of the world, that St. Christopher's spiritual ancestors, wholly unaware of the brilliant future that awaited them, instinctively greeted the new day with hoots and hops.

T

H

R

E

E

THE CYNOCEPHALIC HORDES

1. ANCIENT GEOGRAPHERS ON THE CYNOCEPHALIC RACES

In the entire history of the Western world, Reprobus-ChristianChristopher is the sole cynocephalic to ever have been dignified with a proper name. Yet, as the Gadla Hawaryat and other sources indicate, he was but one of an entire race of Dog-Men. One need only reflect, however, on the lot of monsters, enemies, outsiders, or others in our own contemporary context to realize that they nearly always constitute an anonymous horde. In the history of twentieth-century belligerence, for example, we have known, among others, the Krauts, the Japs, the Reds, and the Yellow Peril; closer to home, we find Spies, Wops, Niggers, and Kikes, to say nothing of the dread Wasp. Such caricaturizations, of racist or any other ilk, preclude all chances for the individual to emerge. A similar case may be found in the creations of the entertainment industry: the villain, the bad guys, invading aliens, the Blob, the Thing, and the Wolfman all remain nameless. Only occasionally is the other "redeemed" from his or her congenital curse of anonymity to be graced with a name (Dracula, King Kong, Godzilla), or at least an initial (E.T.)—which somehow renders him more endearing, because having a name is an important part of belonging to the human sphere. The nameless cynocephalic monsters of the ancient, Hellenistic, and medieval Western world lived on the same nebulous border between fascinans and tremendum as is inhabited today by the heavies or monsters of the Saturday matinee. But, in the former case, in which the two poles of interest may be qualified as "Asianism" and "Atticism,"1 the locus of the anonymous hordes was not outer space, but rather the far reaches of the nearest uncharted sea or wilderness—and all the Westerns were "Easterns." As we have seen, all of the saints and aposdes associated with the Cynocephali had their missions in Ethiopia or India, on the eastern or southern extremities of the earth. In truth, these two regions were the same for the ancient and medieval world. So it is that Homer can speak of both 47

48

THREE

African and Asiatic "Ethiopians" ("people with burnt faces"), while Herodotus, in the earliest surviving Greek reference to India, is able to make the following telling statement: "All Indians have a complexion closely resembling the Ethiopians. The seed they emit is not white, but black as their skin; the Ethiopians also emit a similar seed."2 With Herodotus, the die that fused India and Ethiopia into a single region was cast: the hom of Africa, the Arabian peninsula, and the Indian subcontinent were a single land mass, with the three regions named India Superior, Media, and Inferior, respectively. Slighdy later, Crates of Millos (approximately 150 B.C.) depicted the entire known world schematically, dividing it into four land masses separated by seas which extended along the equatorial zones and between the poles. In the northern hemisphere were the oikoumene and the perioikoi; to the south—"India"—were the antoikoi and antipodes.3 This ancient perspective would carry over into the medieval worldview. Isidore of Seville's Etymologiae, the definitive source of patristic geography, depicted the world as a disc surrounded by sea and divided into three unequal parts. At the center of this world lay Jerusalem, with Europe and Africa to the west and Asia to the east, and the Earthly Paradise at the eastern extremity of the earth, on the eastern edge of India.4 Through the intermediary of the medieval church, Herodotus's ancient perspective remained operative for some two thousand years. In the two centuries following Herodotus, however, new accounts reaching the West brought about a widening of geographical perspectives. These were the writings of Ctesias the Cnidian and Megasthenes, as well as annals of Alexander's conquest of India. While these new sources did not unstick the two regions, they made "India" the all-embracing term for the land of monsters and wonders, thus eclipsing the "Ethiopia" of earlier traditions. Herodotus's account of the Kynokephaloi may in fact have first been inspired by indigenous Greek traditions that depicted these creatures as bogey-men. We encounter such Cynocephali in Aeschylus's Prometheus Bound5 and Aristophanes's Equites.6 Hesiod mentions Hemikunes, perhaps in reference to the same creatures.7 Citations multiply after the time of Ctesias, with the cynocephalic bogey-man retaining his place in Greek and Slavic folklore down to the present century.8 Herodotus's identification of the Kynokephalos with the dog-faced baboon (Simia hamadryas) would, moreover, assure the place of Ethiopia in cynocephalic lore. His sparse references, however, pale before the fantastic and romanesque descriptions of Ctesias's Indika, which, recorded at the

The Cynocephalic Hordes

49

close of the fourth century B.C., truly launched the millenary career of the Indian Cynocephali, together with those of many other marvels of the East. The Indika is only known to us through the citations of later writers, of which the ninth-century Greek Patriarch Photius's epitome of the text, in his Bibliotheca, is the best-known and most complete rendering. It is an abridgement of Photius's summary of chapters 19-23 of the Indika that we present here:9 There is also a river that passes through India; it is not very long, but it is approximately two stadia wide. The name of this river is "Hyparchos" in Indian, and in Greek "bearer of all goods." 10 Thirty days each year, this river carries amber [in its current]; in effect, the Indians say that, in the mountains, there are trees that hang over its waters [for it flows down from the mountains] and there is a season in which tears fall from the trees . . . These tears fall into the river and harden. This tree is called in Indian "siptachöras," which, in Greek means sweet, pleasant. So it is that the Indians gather amber . . . It is said that there live in these mountains dog-headed men [anthröpous . . . kunos echontas kephalen]; they wear clothes made from animal skins, and speak no language but bark like dogs and recognize one another by these sounds. Their teeth are stronger than those of dogs, and their nails the same as theirs, but are longer and more curved. They inhabit the mountains down to the Indus; black is their skin and great their sense of justice, as with the other Indians with whom they have dealings; they understand the Indians' language but cannot speak with them; it is by means of cries and hand and finger signs that they make themselves understood, after the fashion of deaf mutes; the Indians call them Kalustrioi, which in Greek means Kynokephaloi.11 This people [ethnos] are numbered up to 120,000 . . . The Kynokephaloi living in the mountains do no work; they live by hunting, and when they have killed their game, they bake it in the sun; they also raise sheep, goats and asses. They drink the milk and sour milk of their sheep; they also eat the fruit of the "siptachöras" . . . they dry fruits and make fruit baskets, as do the Greeks with raisins. The Kynokephaloi make a raft on which they place a load of dried fruits, purple dye and amber worth one thousand talents; all of this is brought each year to the king of the Indians. They

50

THREE

also bring other products which they exchange with the Indians for bread, flour and cotton clothing; they also buy swords which they use to hunt wild beasts, as well as bows and javelins . . . they are invincible in battle because they live in the high mountains which are difficult of access. Every five years, the king makes them a present of 300,000 bows, 300,000 javelins, 120,000 light shields and 50,000 swords. These Kynokephaloi have no houses, but live in caves . . . The women bathe once a month, when they have their periods, and never at any other time. As for the men, they never bathe, but do wash their hands . . . They wear a light dress of trimmed hides, as thin as possible, men and women alike . . . They have no beds, but make pallets of leaves . . . All of these, both men and women, have a tail that hangs down from their hips like those of dogs, but longer and furrier.12 They couple with their women on all fours like dogs; to unite otherwise is a shameful thing for them. They are just [dikaioi], and the most long-lived [makrobiötatoi] of any human race, for they live for 170 years, and some of them for two hundred. These little black dog-faced montagnards, with their amber, dried fruits, and tiny javelins, their sun-cooked meats, leafy beds, and rustic dress, and especially their great sense of justice and remarkable longevity, were, so long as they remained on the eastern edge of the world (following Herodotus's location of India at that limit), Europe's original noble savages.13 One is tempted here to see the origin of the name of Diogenes of Sinope's fourth-century B.C. "philosophical sect," the Cynics (from kunikos, "little dog"): Diogenes and his followers professed a rigid ethics, even as they publicly behaved like dogs. More importandy, Ctesias's account of these and other wonders would have an immediate effect on the history of the ancient world, East and West: it would inspire a certain Alexander of Macedonia to begin a conquest that would bring him to the shores of the Indus in 3 2 6 B.C. 1 4

The accounts of Alexander's expedition and the travel books written by those who followed him would multiply and enrich the burgeoning Western mythologies of the East and, in a dialectical fashion, further eastward exploration and the consequent inventory of its wonders.15 Primary

The Cynocephalic Hordes

51

the early successors to Alexander was Megasthenes, who lived in the court of the Mauryan emperor Chandragupta near the beginning of the third century B.C. This diplomat, whose own writings are only known to us through citations in later sources, confused Indian mythology with ethnography, and so continued the tradition of the dog-headed race, changing their name from Kalustrioi to Kynokephaloi. Because Megasthenes was considered to be an eyewitness to these and other wonders, his account bore the stamp of authority—for the following eighteen hundred years! Among the pagan writers and works of the Hellenistic age that would further the taxonomy of the marvelous and monstrous were Aelianus (De natura animalium, 2d century A.D.), Eratosthenes (3d century B.C.), Strabo (Geographica, completed A.D. 19), Aulus Gellius (NoctesAtticae, 2d century A.D.), Pliny (Naturalis Historia, completed A.D. 77), Ptolemy (Geographica, 2d century A.D.), the anonymous Egyptian Physiologus (2d-5th century A.D.), Palladios of Helenopolis (Historia Lausaica, A.D. 420), Mardan of Heraclea (PeriplusMaris Erythreum, early 6th century A.D.) and the Liber Monstrorum (5th-6th century A.D.). 1 6 Of all the Hellenistic sources, the most prolific (and least discriminating), and the most influential for the Christian authors who would inherit these legends, was Pliny the Elder. Pliny was a Stoic, whose aim in writing his encyclopedic work was to describe nature's play in "creating singularities that are miracles to us."17 Pliny located his monsters in India and Ethiopia, which, like others before and after him, he considered to be the same region. In describing the Cynocephali, he faithfully follows Ctesias's account of that Indian race, but adds a reference—to their Ethiopian brothers—that conflates Ctesias with other ancient sources. These are the dogmilking cynocephalics, whom he calls the Cynamolgi. As we will show in the following chapter, Pliny may have been closer to the truth with the Indian Cynamolgi than he was with the Cynocephali. In the Indian epic and Puranic traditions, outcastes are commonly referred to by a term that may be construed as "Suckled by Dogs" (Svapâka).18 In addition to these cynocephalic Ethiopians, Pliny also refers to the king of Garamantes (modem Libya), in whose army the "dog soldiers" were literally dogs! 19 He also locates a race he calls Canarii, "Dog-People," in western Africa: here, he is drawing on a third-century B.C. reference to a People known as the Ptoemphani who, living in the Nile valley, purportedly had a dog for their king. We will discuss the Canary Islands (Islands of Dogs) of the western sea, off the coast of Africa, later in this chapter.20 among

52

THREE

2 . CYNOCEPHALI IN THE ALEXANDER ROMANCE: T H E FACELESS DOG-FACED H O R D E S

More than Ctesias and Pliny, the most influential medieval source of popular cynocephalic lore was the Alexander Romance. This legendary account of Alexander's adventures is attributed to Pseudo-Callisthenes, who is so named because the pre-fourth-century A.D. author of this work was believed to have been a nephew of Aristotle's named Callisthenes who had accompanied Alexander on his expedition.21 Alexander's encounters with many of the fantastic races, most of which appear in Book Three of the Pseudo-Callisthenes, are in fact late interpolations borrowed from Palladius of Helenopolis's fifth-century Lausaic Histories: Palladius's source was a second-century Cynic diatribe, which itself relied on Clitarchus, Megasthenes, and Onesicritos.22 This fabulous narrative, considered to be solid historiography throughout the Middle Ages, injects two important new elements into myths of the cynocephalic races: on the one hand, they are made out to be warlike, even cannibalistic giants; and on the other, it is here that they are first associated, albeit indirectly, with the Amazons, the women warriors of Scythia. In addition, an ancient commonplace—that the oikoumene, "the civilized world," was separated from the northern barbarians by a mountain range running from the Caucasus in the West and across all of Asia to an Eastern Sea, identified with the coast of India—becomes dramatized in the legend of Alexander's "walling out" of the barbarian races. In the course of the Middle Ages, these races would come to be identified with the Biblical armies of Gog and Magog. 23 The Alexander Romance, or the Pseudo-Callisthenes,24firstcomposed in Alexandria before the fourth century, was the blockbuster bestseller of the entire Middle Ages. It was quickly translated from the original Greek into Latin (4th century A.D.), Armenian (5th century), and Arabic (10th century); as well as into Pehlevi (6th century), Syriac (7th century), and Ethiopic (13th century). The most popular translation in the West was Leo of Naples's tenth-century Historia Alexandri magni, regis Macedonia, de prdiis, or the Historia deprdiis which, combined with an apocryphal Letter of Alexander to Aristotle (8th-9th century), was the source for a wide array of vernacular Alexander legends.25 In many ways, the Alexander Romance was the prime vehicle for the medieval lore of the monstrous races. As such, it also came to influence the legends of such Christian saints as Andrew and

The Cynocephalic Hordes

53

Bartholomew, soldiers o f Christ w h o traveled into distant lands to evangelize monstrous races.

With each new translation of the Alexander Romance and peripheral mythic cycles, with each new edition, novel material was added, ever enhancing the baroque content of the original. More than this, with every new invasion of some new barbarian people, from out of the wilds of eastem Europe or central Asia, the specific names of Alexander's monstrous enemies were revised to fit current events. In this way, the Cynocephali of the Alexander Romance would come to be serially associated with the Scythians, Parthians, Huns, Alans, Arabians, Turks, Mongols and a host of other real or imagined races.26 A prime example of such conflations may be found in the English cleric Thomas of Kent's thirteenth-century Roman de toute chevalerie. Drawing on the Pseudo-Callisthenes, Pliny's Naturalis Historia, and other sources, Thomas's tale seemingly combines every Eastern monster and wonder known to the medieval West. So we find, blended in with the acts of the "historical Alexander" in India, accounts of the long-lived Macrobiotic (chap. 84), "Candea where only women rule" (88), fish-eating Ichthyophagoi (89), "people who are half dogs and half men" (93), Sciapodes (90), "women who give birth at the age of five" (98), "Gog and Magog who eat people" (148), "people who eat moles and mice" (150), "Turks who eat people and dogs" (152), "the two queens of the Amazons" (159), and Alexander's battles against and imprisonment of Gog and Magog (163-64, 169-70). After Alexander enters "Ethiopia," the account discusses "people who made a dog their king" (180) and other Ethiopian monsters (181-88), as well as Alexander's contacts with Candace (205-6, 251-69), who is generally identified as queen of the Amazons. 27 In spite of these changing identifications, however, the Cynocephali remain a perennial fixture among the monstrous races, and are often designated in conjunction or juxtaposition with the Amazon women. Traditions are quite confused regarding the region in which Alexander encountered his Cynocephali. In certain versions of the Pseudo-Callisthenes, he meets them on his return journey from India; in the Historia deprdiis, they live on the shore of the easternmost river of India; in the Armenian version, they are somewhere "beyond the Taurus," on the "other side of the Median desert." Yet, as the Ethiopic version relates, Alexander leams of them on the edge of the world, between the "Foetid Sea" and the "Land of Darkness." Here, he asks the people that he meets if there are other races

54

THREE

living "over the edge." "Yes," he is told, "these are the nations living beyond. They are the Taftas, that is to say the Nagâshâwîyân, whose faces are like unto those of dogs . . . No man has ever been able to enter into their country, wherein there is nothing except lofty mountains. At the tops of these high mountains is Paradise, which is situated betwixt Heaven and Earth." 28 This last (and relatively late) source conflates a great number of mythic geographical motifs, including the eastern location of the Earthly Paradise, the eastern sea (located on the coast of India by Herodotus) out of which the sun rises, and Cynocephali living in the high peaks or on the extreme edge of the earth. We will return to this tangled web of themes in later chapters.29 In many of the extant versions of the Pseudo-Callisthenes, Alexander actually meets and fights a pitched battle with the cannibalistic dog-men, whom he routs: We came to a place where a delightful and abundant spring rose . . . Then there appeared to us, about nine or ten o'clock, a man as hairy as a goat. And once again, I was startled and disturbed to see such beasts. I thought of capturing the man, for he was ferociously and brazenly barking at us. And I ordered a woman to undress and go to him on the chance that he might be vanquished by lust. But he took the woman and went far away where, in fact, he ate her. And he roared and made strange noises with his thick tongue at all our men who had run forth to reach her and to set her free. And when his other comrades heard him, countless myriads of them attacked us from the brushes. There were 40,000 of us. So I ordered that the brushes be set afire; and when they saw the fire, they turned and fled. And we pursued them and tied up 400,000 of them, but they died since they refused to eat. And they did not have human reason, but, rather, barked wildly, like dogs. 30 In the same (B) versions of the Pseudo-Callisthenes (as well as in the Ethiopic version), Alexander is next said to have crossed over a land of darkness and come to a paradise on the eastern edge of the world. Sent back from there by "angels," he constructs, on his way out of the Taurus mountains, a great gate of iron between two lofty peaks to wall out or imprison the barbarian nations in the northern fastnesses. In the list of the twenty-two (or twelve, or twenty-four) races thus relegated beyond the northern limits we

The Cynocephalic Hordes

55

find mentioned "those cannibals who are known as the Kynokephaloi." Together with these are found the Amazons and other central Asian peoples.31 This image, of a hero plugging the sole gap between the civilized and savage races and worlds, was undoubtedly a very powerful one given the perenniality of this account across a wide array of legends and over several centuries. In addition to its many written editions, this episode of the Alexander Romance also inspired artistic creation. We find a stunning artistic rendition of the building of Alexander's Gate in a fourteenth-century Persian miniature entided Iskandar Builds the Iron Rampart. In this miniature, it is the break between civilization and savagery that is most evident. On the near side of the wall are men in brighdy colored clothes manning forges, machines, and other tools used in the wall's construction. On the far side of the wall are grey desert mountains with ragged underbrush in which crouching, hairy semi-human creatures may be made out, their bodies quite difficult to distinguish from their natural surroundings (pi. 5). 32 Where precisely Alexander's Gate was located was a perennial source of confusion for medieval writers. Because previous tradition located the wild and sometimes cannibalistic Scythians to the north of the Caucasus, and because the Caucasus and Hindu Kush were conflated into a single mountain range called Taurus, Alexander's Gate was located and relocated at (1) the Pass of Dariel, slighdy to the north of modem Teheran; (2) the Pass of Derbend, hard by the west coast of the Caspian Sea, in the Caucasus; (3) the Iron Gates in the Pamir system (between Bactriana and Samarkand); (4) northern Europe or the Urals; or (5) far eastern or northeastern Asia, in the region of Lake Baikal, to the north of Mongolia.33 It is in these medieval retellings of this episode in the Alexander Romance that we find the origin of a tradition of northern, Scythian, cannibalistic cynocephalics. This tradition would continue alongside and in confusion with those of the peaceable cynocephalic races of India and Ethiopia related in Ctesias and Pliny.34 Among the numerous early Christian sources that follow the various pagan traditions regarding the Cynocephali, it is the Revelations of a certain Armenian or Persian named Pseudo-Methodius (so named because this work is attributed to Methodius, bishop of Patara, who was martyred in the fourth century) that most closely parallel the Alexander legend. This author, writing in the midst of the mid-seventh-century Islamic Ummayad invasions of the near east and Persia, draws on a Syriac source, the Book of the Cave of Treasures, to construct an eschatological history which equates the barbarian nations

56

THREE

ultimately cast out by Alexander with the Muslims who would soon be cast out by one of his descendants, all in the name of the Christian God. In fact, Pseudo-Methodius's Persian and Syriac sources are the same as those used in the Pseudo-Callisthenes descriptions of the barbarian nations encountered by Alexander, and his account of Alexander's Gate may be the source of the story in later versions of the Greek Pseudo-Callisthenes, as well as of certain elements of the highly influential twelfth-century forgery, the Letter ofPresterJohn. So it is that we find the Cynocephali in the same incorrigible company here as in medieval editions of the Alexander Romance: among the names that recur in all translations of Pseudo-Methodius are the Garamantes, Amazons, and "the maneaters (anthröpophagfoi), who are called Kynokephaloi."35 The Christian Pseudo-Methodius, however, transforms this western Asian cycle of legends into a lesson in the economy of salvation. Like the many Christian theologians and historians who precede and follow him, Pseudo-Methodius places his account of the monstrous races in the second aetas of the world—the age following the Flood, in which the world was repopulated by the sons of Noah. And, integrating a number of earlier traditions into a single polemical work, he transforms Alexander the Great from a pagan empire-builder to an instrument of the Christian God. 36 What is original in Pseudo-Methodius, and what won him an enduring place of honor in Christian historiography, is the fact that he uses his narrative as a "proof" by which to repudiate the apparent victory of the "Turk" (i.e., the Muslim, the Hun, the Scythian) whose invasions had already been the terror of the West for centuries and who would continue to sweep over the bastions of Christendom for centuries to come. In so doing, he borrows from the Book of the Cave of Treasures a fourth son of Noah named Jonitus (or Yonton, Jonaton, or Maniton), whom the patriarch sends to the Land of Sunrise, far to the east. In his description of Jonitus's travels, Pseudo-Methodius closely associates his hero with both Nimrod, the builder of the Tower of Babel and founder of the giant race that ruled over Babylon, and Ham, the ruler of Pontipus, or Puntos. These toponyms evoke the "edge of the world" and the city of the cynocephalic baboons of ancient Egyptian and Hellenistic traditions. 37 Pseudo-Methodius's Revelations is greatly colored by Persian names and toponyms. And, once again, as in the hagiographic cycles of Christopher, Andrew, and Bartholomew, the Parthian element is very strong in the depiction of the (sometimes cannibalistic) Scythians, Huns, and Cynocephali. Pseudo-Methodius's argument, and the prime reason for the pe-

The Cynocephalic Hordes

57

rennial success of his work, is his prophecy that the armies of the Turk, rethis world "for a litde while" (Revelation 20.3) before the definitive defeat of Gog and Magog.38 Of course, by typologizing the peoples of central Asia in this way, Pseudo-Methodius also contributed to the sort of Western "citadel mentality" that would come to launch the Crusades and many other invasions against the standardized barbarian races of central Asia for centuries to come. More than this, his account underlies a number of twentieth-century folktales found in the Baltic and Balkan regions, folktales that, having passed through the lens of Russian encounters with invading Mongols, come to describe the invasion of eastern Europe by "Dog-Snouts" from the "eastern edge of the world."39 A twentieth-century example from Estonia illustrates this conflation of the medieval with the modern: 40 leased at last from behind Alexander's Gate, would only torment

The Dog-Snouts live at the edge of the world, where the earth comes to an end and heaven begins. They must stand guard at the edge of the world, so that no one may enter into heaven there . . . The Dog-Snouts at the edge of the world dwell behind a great mountain. This mountain forms the border between the land of the Dog-Snouts and that of men. It is necessary that a company of soldiers, indeed, according to general consensus, a company of Russian soldiers, stand watch at this mountain, lest the Dog-Snouts come into this country . . . Were the Dog-Snouts to succeed in coming over the mountain, they would tear every man in the world limb from limb. Their strength is so great that none can resist them . . . The Dog-Snouts are threatening the world of men with destruction . . . The end of the world will only take place when the Dog-Snouts succeed in coming over the mountain and breaking through into the rest of the world. One need not fear, however, so long as the Russian troops watch over the mountain and refuse to be trifled with. Victory is always with them . . . In some regions . . . it is held that each town in the land contributes proportionally to the maintenance of the mountain guard; this is because every town in all of society is anxious that the end of the world should never come. Nevertheless, there was once a time when the all of the DogSnouts did burst through, and might have destroyed the

58

THREE

world of men. They came to lay waste to our land, when fortunately a violent hailstorm commenced. The hail drove the Dog-Snouts back into their land, and now their mountain is guarded more strongly than ever. 3. AMAZONS AND CYNOCEPHALI AT THE WORLD'S CONFINES

From Virgil's day at least, the Amazons were identified with the woman warriors of "Scythia."41 As we shall see, these women came to be closely associated with the Cynocephali in later European, as well as Chinese and central Asian, traditions. In the Alexander legend itself, the association is one of juxtaposition: quite often, preceding or following his adventures with the monstrous races, Alexander comes to the land of the Amazons, ruled by their queen Kandake (Candace).42 In her exchange of letters with Alexander, Candace makes use of her brother, who serves as an emissary to Alexander on the behalf of the Amazon queen. This brother's name, Kandaules, is glossed in a Greek source as the Maeonian term for "dog-strangler," a likely reference to yet another case of a canine or lupine ethnos.43 Apart from his brief romance with queen Candace, Alexander has a number of other meetings with kingdoms or races of women, again associated with canine males, in the course of his peregrinations. In certain recensions of the Historia de prdiis, he and his army meet hairy, bearded women living in the forest on the near side of a river. Crossing the river, they encounter Cynocephali, whom they rout in battle.44 Later, in the thirteenth century, the Franciscan friar John of Piano Carpini gives an account of a pitched battle fought by the Tartars (whose own legendary accounts of Cinggis Qahan draw on the Alexander Romance) against a race of dogs who lived together with human women near the eastern ocean. These juxtapositions carry over into a modern Romanian folktale of two brothers named Cuckoo and Mugur. In describing Alexander's conquests of the northern lands of the Cynocephali, the tale says that it is the practice of male Cynocephali to raid southern lands, carrying off men whom they fatten and eat and women whom they marry to their king. Following his defeat of the Cynocephali, with the aid of Cuckoo and Mugur, Alexander comes upon a land of beautiful women, "where women take the place of men and men that of women." From here, Alexander travels onward to the gates of Paradise.45 To this point, the Cynocephali and Amazons whom Alexander met in close succession were generally located beyond the lofty Taurus range, hard by the eastern edge of the world, a land of the sunrise. For the medieval world, the world's western confines also harbored the same fantastic pair-

The Cynocephalic Hordes

59

ing of races. We mentioned above that Pliny located a race of Dog-Men, the Canarii, in western Africa. His location of these people in the west in fact goes back to another ancient tradition. The Canary ("Canine") Islands, off the west coast of Africa, were identified as the location of both the Garden of the Hesperides and a "Land of the Blessed Dead" from at least the first century B.C. down through the Middle Ages. In Greek traditions, the Hesperides were the daughters of Evening (hespera) or of the evening star (,hesperos), and it was into this red land of sunset and death that Heracles "descended" in order to win the apples of immortality. Like the red (dead) oxen of Geryon in Iberia, this garden was guarded by the great hound Orthros, the brother of Cerberus, whom Heracles had to defeat to reach his goal. Strabo locates the land of the dead in western Spain, and states that a temple of Baal-Satum, or Hercules-Melquart, is located there. Once again, as was the case with the Egyptian jackal-headed Chenti Amentiu, "the first of those to the west," the threshold of night and death is guarded by a dog.46 This motif is one which is even more widespread in Celtic tradition, for which the Canaries continue to double as the western "Isles of the Blest" in the legends of the sixth-century Saint Brendan. According to his Navigatio, which first appeared in Latin in the eleventh century, Brendan "dreamt" a blessed island in the western ocean, which he later visited. In his voyages, Brendan and his monks are met by a seemingly innocuous dog on an island that turns out to be inhabited by "little Ethiopian" demons; a meeting with dog- and pig-headed men is figured in a medieval German codex.47 The Canaries are called the "Isla Fortunate" in the eighth-century Beatus Map, and are identified with Brendan's islands on the thirteenth-century Hereford Map. 48 Before Brendan, the pagan Celts knew of a "Land of Sid," a paradise across the western sea, which was at once a "Land of Women" and a "Land of Apples" (Avalon). There also exist Irish legends of the invasion, from across the western sea, of a race of Dogheads (Coinceann):49 these are defeated by Cuchulainn, whose name means the "Dog of Culann" ("dog" is cu in Gaelic) in the same tradition. This invincible hero himself becomes the "dog" of the smith Culann (the "Dog of War") to atone for having killed Culann's hound, in self-defense, at age six. Later, Cuchulainn is rendered vulnerable and dies when he eats "his namesake's flesh": dog flesh. Ireland knew of other cynocephalic peoples, and had a number of legends concerning the cynocephalic St. Christopher.50 Elsewhere, as we have intimated and as we will discuss in detail in later chapters, the Cynocephali were located together with the Amazons in cen-

60

THREE

tral Asia. Perhaps the earliest Western source to bring these two races together in the "land of the Huns," to the north of the Caucasus and the Caspian Sea, is the sixth-century Syriac "Church History" of a certain bishop Zacharios ofMitylene, Pseudo-Zachariah the Rhaetor.51 This location, too, would become a commonplace through much of the Middle Ages, as would the northern shores of the Baltic. This conflation of Amazons and Cynocephali, together with a number of other terrible central Asian races, is most fully fleshed out in the eleventhcentury Gesta Hammaburgcnsis ecclesiae pontificum (4.19, 25) of Adam of Bremen:52 Along these shores of the Baltic53 it is said that there are Amazons . . . There are those who relate that they are made pregnant by merchants, or else by those that they keep as prisoners with them, or else by other monsters, which are not rare in this place . . . And, when they give birth, if the [offspring] is of the masculine gender, he becomes cynocephalic; if feminine, it is a woman of exceptional beauty . . . During their lives, these [Amazons] spurn the company of men who, if they approach them, are forcefully driven away. The Cynocephali are those who have their heads in their chests.54 In Russia, they are often held captive, and when they speak, their voices [are mixed with] barking [sounds] . . . Dogs defend their [the Amazons'] country. If it happens that it is necessary to do battle, they [put] lines of dogs in their ranks. In this country are found pale, vigorous and gigantic men that are known as the Husus; to conclude, there are those that are called Antropofagi [cannibals] who live on human flesh. Here are found many other monsters, who are spoken of by sailors as having been seen from close quarters, as incredible as this may sound to us. 55 4 . T H E CYNOCEPHALI IN MEDIEVAL GEOGRAPHY, CARTOGRAPHY, AND THEOLOGY

Following Adam of Bremen, the theme of male Cynocephali cohabiting with Amazon women would become a constant in the medieval geography and cartography of northern Europeans.56 Also in the late Middle Ages, the Baltic region would become yet another perennial locus of Cynocephali and Amazons, a state of affairs that would lead a sixteenth-century Swiss author to state that Crakow had replaced Africa as the prime venue of the

The Cynocephalic Hordes

61

monstrous races.57 Adam's amazing account, which brings together nearly all of the Western lore of the Cynocephali into a single confused mass, is innovative inasmuch as it suggests that the male offspring of the gynecocratic Amazons were the Cynocephali. This indeed becomes a leitmotiv in cynocephalic traditions not only in Europe, but also across the whole of Asia and as far east as Japan and the Eskimos of North America. While we will examine the burgeoning indigenous Asian traditions of this phenomenon in later chapters, here we will limit ourselves to Western accounts of it. We may begin by noting survivals of ancient indigenous European themes of cynocephaly, beyond those already mentioned in the context of Ireland and the Romanian tale of Cuckoo and Mugur, in contemporary northern and eastern European folklore. In northern Europe, we find Irish, Swedish, German, Russian, Lithuanian, and Latvian folktales in which a human princess unites with a dog, a union which at times produces a particular race or people. We also find accounts of the human race arising from the penis (or tail) of a dog who had stolen Adam's rib from the Garden of Eden!58 Paulus Deaconus, in his eighth-century Historia Lanpfobardorum (1.11), mentions fierce, blood-guzzling cynocephalic warriors in the army of the Lombards, a Germanic people then living on the east coast of the North Sea. The Lombards were ruled in the early fourteenth century by a king Cangrande, the "Great Dog"; indeed, the dog was an emblem of the Lombards.59 Cangrande was not the only king to bear a canine name in medieval Europe. Others who shared this distinction were Hundingus and Hundehoved, who are chronicled in the Scandinavian Eddas and writings of Saxo Grammaticus; and the "dog-headed" Kralj Pesoglav and Pes Marko, of Slavic tradition. Quite often, these kings were either rulers of peoples who identified themselves in some way with a canine totem (as did the Lombards), or who were leaders of elite troops who dressed and comported themselves in battle like dogs (Lombards) or wolves, with whom they identified themselves. The Icelandic Eddas name Hundingr as the king of Hundland, "Dog-land"; the pre-tenth-century Anglo-Saxon Widsith mentions the Hundingar as a dog-headed people; while the "werewolf" (ulfhednar) military brotherhoods of the Germanic tribes elsewhere fought alongside "half-dogs" (halfhundingas).60 With the exception of certaifc pre-fifth-century traditions relating to Pes Marko, nearly all southeast European legends on the subject of tynocephalics identify this dread, man-eating race as the Turks, who con-

62

THREE

quered the Balkans between the thirteenth and fifteenth centuries. It is nevertheless likely that the pesoglavci of Slavic tradition existed in popular myth and legend prior to the superimposition of the Turkish gloss. Bulgarian and Slovenian folktales relate the origins of the Turks from a union between a dog and a woman, and a nineteenth-century Danish source speaks of the Turks as being subdivided into dog- and pig-headed groups, of which the latter are man-eaters. Likewise, in a southern Slavic folksong, an old man laments the loss of his eight sons, his eight daughters-in-law, and his orphaned grandchildren, all of whom have been devoured by maneating dog-headed Turks. We will have much more to say on these associations in chapter 6. 61 Indeed, following Pseudo-Methodius, it is the Turkish trope that prevails in European accounts of Dog-Men. In spite of the many indigenous cynanthropic traditions we have passed in review, medieval geographers, historians, and theologians invariably located these monsters at the world's limits, even as they unconsciously projected their own indigenous traditions upon them. As such, medieval cartography was at once "historical" and typological. On the one hand, monsters were located at the ends of the earth, even beyond the lands of the "barbarians," because that is where ancient and Hellenistic writers had placed them. On the other, the logic of "zone maps" situated those races who were most extreme in their physiology or mode of life in regions of extreme climate, under the burning sun of "southern" Ethiopia or in the frozen wastes of the north. 62 Combining their own northern and southeastern traditions of cynocephaly with the southern and eastern traditions inherited from the ancient geographers, medieval cartographers placed the Cynocephali at three of the four corners of their own world maps. So it is that we find them pictured or listed in Ethiopia to the south, in India to the east, and Scythia to the north (pi. 6a). 63 Finally, at the close of the Middle Ages, new explorations would lead to the "discovery" of Cynocephali far to the west, in the New World (pi. 6b). 64 Following the lead of the ancient Egyptian and Greek sources we have already discussed at length, medieval cartographers located the Cynocephali in Ethiopia and India. Their localization of this race to the north, in Scythia, may be accounted for in three ways. First, many of Europe's own popular traditions of cynanthropy were northern. Second, such widely read works as the Alexander Romance and Revelations of Pseudo-Methodius placed the enclosed nations to the far north, beyond the Taurus range. Lastly, an ambiguous medieval toponym seemingly fused north and

The Cynocephalic Hordes

63

south: from the ninth century onwards, "Ethiopia" was used to designate the lands north of the Caucasus, between the Black and Caspian seas—the very location, in legend, of Alexander's Gate! 65 Among medieval maps that portrayed the Cynocephali on the southern edge of the world, in the land of Ham and to the left hand of Christ, are the tenth-century Cottoniana map, the Itinerarium Sigericus map, and the thirteenth-century London Psalter and Ebstorf maps. The Cynocephali are first situated in the extreme north in the seventh-century Cosmography of Aethicus Ister, who intimates that the cynocephalic St. Christopher came from such a northern race. They are located in the same region in the maps of Beatus (8th century) and the twelfth-century Heinrich of Mainz, as well as the thirteenth-century Hereford map and the works of the Muslim Qazwini.66 Yet, because Ctesias and the Alexander Romance generally located the monstrous races to the east, and because they were considered to be descended from Adam and Even exiled east of Eden, the Cynocephali and their monstrous brethren were sought after and generally found in India. Throughout the Middle Ages, historians would also locate Paradise in India, on the eastern edge of the world. And, following the twelfth-century anonymous Letter ofPrester John—which probably originated from a central Asian Nestorian Christian source—concerning the foundation of the (partially Christian) empire of the Kara Khitai, in central Asia, the kingdom of this apocryphal ruler of Paradise, too, would be situated in India. 67 However, as the world grew smaller, through exploration, evangelization, and commerce, the Cynocephali would come to be located in ever more obscure oriental settings. As we have seen, John of Piano Carpini (thirteenth century) finds them to the north of Tartary. At the end of the same century, Marco Polo 68 reports sightings of Cynocephali on the Andaman Islands of the Bay of Bengal (pi. 7). In the fourteenth century, Oderic of Pordenone finds them on the nearby Nicobars, Friar Jordanus in the ocean between Africa and India, and Ibn Battuta in Burma. At the end of the following century, the marvels of the East would be eclipsed by those of the New World: Columbus would find people he took to be cynanthropic "cannibals" (the origin of this term is the name these people gave to their island—Cariba, as in Caribbean—which Columbus confused with "cani-ba," the "canine" great Khan [confused with the Latin canis] of Cathay, China). On the map he drew of his travels, however, Cariba was located off the east coast of China, and thereby remained a far eastern island of the "Indies." Later, a great river to the south of the canine cannibals of

64

THREE

Cariba would be named Amazon for the tribal women living there. 69 Old myths die hard. As we have intimated, the major source on the monstrous races, for the Christian authors who followed him, was Pliny the Elder, in spite of the fact that his pagan predecessors and contemporaries (Eratosthenes, Strabo, and Ptolemy in particular) offered a more rational view of world geography than he did. 70 Christian encyclopedism on the monstrous races begins with the third-century Collectanea rerum memorabilium of Solinus and the fourth-century Commentarii of Servius, and continues into the late Middle Ages in the thirteenth-century Otialmperialia of Gervasius of Tilbury and Vincent of Beauvais's Speculum Historiale and Speculum Naturale; the fourteenth-century Gesta Romanorum; and the fifteenth-century Fulgentius Metaphoralis.71 In addition to these textual locations, the Cynocephali enjoyed great exposure, often together with other standardized monstrous races of the Middle Ages (the Cyclops, Blemmyes, Sciapoda, Panotii, etc.) in ecclesiastical art. Much more common than the massive images of the cynocephalic St. Christopher erected in this period are the humbler icons, stained-glass figures, frescoes, and bas reliefs of anonymous Cynocephali that ornamented medieval churches and cathedrals, from northern Denmark (pi. 8) to Constantinople, Russia, and Armenia.72 These sources are of secondary importance, however, in comparison with the two great Christian authors who truly set the medieval standard for the theology of the monstrous races, and who thus assured them a perennial place in medieval literature and art. These were the fifth-century St. Augustine and the seventh-century Isidore of Seville. In a general sense, the place of the monstrous races in medieval Christianity was one of mitigated exclusion. On the one hand, the Church constandy strove to censor irruptions of pagan mythology and beliefs within its fold, even though, as continues to be the case today, censorship only served to increase their notoriety. On the other hand, Christian authors and theologians, themselves prey to both thefascinans and the tremendum of the monstrous, could not help but theorize on their place in the universal economy of salvation. So it is that we read, in Augustine's meditations on the bewildering array of "human" races, that "those things which have no significance of their own are interwoven for the sake of things which are significant."73 Augustine's thinking was influenced both by classical philosophy and by Judeo-Christian theology. In the former, monsters—the satyrs, sileni, nymphs, and fauns—were neither particularly threatening nor important:

The Cynocephalic Hordes

65

they were mere curiosities. At the same time, they, along with those living outside the polis and the babbling barbarians living outside the Hellenic world, at several removes from the central One in the great chain of being, were not truly human. Because they lived under the wrong laws, or no laws, they were victims of the animal side of their own natures; but rather than being ethically immoral, they were viewed, like all creatures of the animal kingdom, as amoral. In the Judeo-Christian tradition, those who are outside the place and time of the economy of salvation are fallen and morally reprobate, since their situation is the result of either miscegenation or their refusal of the blessing offered to them by God. They are thus relegated to the wilderness, which is at once the cause and consequence of God's curse.74 Miscegenation begins with the union of the giant sons of God and the daughters of men (Genesis 6.1-4). The refusal of God's blessing begins in Eden, from which Adam and Eve are exiled, but especially with Cain who, because of his fratricide, is cast out by God and becomes physically changed into an animal.75 The postdiluvian Cain of the second aetas is Ham, the black son of Noah who disobediently looked upon his father's nakedness and later fathered the Ethiopian races. But it is Ham's son Nimrod, a "hunter against the Lord" in Augustine, whose outrage of the Tower of Babel caused the nations to be "scattered over the earth."76 In a Christian perspective in which genealogy was all-important, Nimrod becomes the ancestor of the monstrous races.77 So it is that Augustine, in the eighth chapter of the sixteenth book of his De civitateDei, a chapter entitled "Whether certain monstrous races of men were produced from the descendence of Adam of the sons of Noah," takes the Cynocephali as a case study in the economy of salvation. After a general description of this race, he states that man, the "rational, mortal animal," although he appears to be possessed of various unusual forms, colors, comportments. languages, essential characteristics, members, and natural qualities, nevertheless originated rous^x; illo unoprotoplasmo. He then opens heaven's gates to the monstrous races, saying that they were created by God who, in his wisdom, placed them in the world in order to reassure the Christian world that the monstrous births occurring within its own society were neither aberrations nor the work of a clumsy divine craftsman: Therefore, it should not seem absurd to us that, just as there are in each race certain men who. are monsters, there should also exist a race of monsters from among the entirety of the human race. And, to conclude with prudence and cir-

66

THREE

cumspection: either that which is written about these races is false; or they are descended from Adam, if they be men.78 For Augustine, then, the Cynocephali and their monstrous brethren indeed live within the City of God: signs of God's omnipotence for Christians in the present time, their humanity would be restored to them and their salvation insured with the final Resurrection.79 Isidore of Seville, in his monumental encyclopedia, the Etymolojjiae, discusses the interconnected matters of the genealogies of the races of man on the one hand, and the role of monsters as portenta on the other. Isidore makes the Indian races the descendants of Shem: they are the sons of Jectam, who was himself the son of Heber (the father of the Hebrews), the great-nephew of Shem. Jectam's son Sale is the father of the Bactrians, a people "exiled from the other Scythians." Chus, the son of Ham, is the father of the Ethiopians: his descendant Chanaam fathered the Afri, the Phoenices, and the ten races of the Chananeans. In later sources, this last designation is the name of the race from which the cynocephalic St. Christopher was originally descended.80 Isidore further groups the Amazons together with the Scythians, Parthians, and Bactrians.81 Here, too, a certain connection between Isidore's races and the legend of St. Christopher may be discerned. The Russian Basil's approximately A.D. 1000 Menologium, in which the apostle Andrew's cynocephalic assistant is apparendy transformed into St. Christopher, calls the land of the apostolic mission Baxthrios (or Bakhtischû or Boktjeschû), that is, Bactria (Bakhtrish, in Persian) and its capital of Balkh, which roughly corresponds to present-day northern Afghanistan.82 It is in the chapter "De homine et portends" (11.3) that Isidore presents his theology of the monstrous races, and so reveals a most fundamental side of the blend offascinans and tremendum that the monstrous held for medieval man. He begins by citing the Roman tradition, for which monstra, nearly identical with portenta, were limited to the phenomenon of monstrous births as warnings (monitus) or portendings (portendere) of divine ill will. Portenta, he adds, are not contrary to nature, but contrary to characteristic signs of nature, and so they may be seen as falling within the divine order.83 Following these definitions, Isidore goes on to offer a taxonomy of monsters which he bases on Aristotelian categories. In so doing, he "identifies and catalogues all the dangers that could threaten the human body. His classification represents the effort of Western man to reaffirm his own normality by confronting it, point for point, with the deformity of the

The Cynocephalic Hordes

67

imaginary races."84 Isidore's Cynocephali, "more like beasts than men in their constitution," are described immediately after the Giants and before the B l e m m y e s . 8 5

Such then, was the place of the faraway Cynocephali in medieval Christian theology: although they resembled beasts, they lived socially and were possessed of a rational soul, like men—and so their ultimate redemption was assured. It was therefore the task of the Church to evangelize the benighted Plinian races, and so it did, beginning with the apostolic missions to the Cynocephali we discussed in chapter 2. 86 5. MUSLIMS, PEASANTS, WILD M E N , AND BLACK PLAGUE: T H E CYNOCEPHALIC AS MEDIEVAL ANTITYPE

Not so happy as the Cynocephali were those others whose fate it was to dwell closer to home in the medieval sphere. This was particularly the case with the scourge of medieval Christendom, the Muslims. Beginning with Pseudo-Methodius, both religious and secular writers associated or identified the followers of the prophet Muhammad with the monstrous races. Throughout this protracted struggle (has it ever ended?), the black Saracen87 is often depicted as a minion of the devil, a child of fornication, a crude oudaw covered with hair and wielding a club. In medieval sources, Muhammad himself becomes a Christian heretic who is thrown on a dung heap to be devoured by dogs and pigs. The mid-thirteenth-century PseudoTurpin describes the Muslim Saracen warriors battling Charlemagne as "horned demonic Lares" (i.e., man-eating cynocephalics).88 In the medieval period, the society and "Church" of Islam was portrayed as the mirror image of Christendom and its institutions: a popular medieval work, the Covenant Vivien (w. 98-100) describes worship of the God Muhammad at a Pentecostal service that is in every respect backwards and upside-down. Yet, according to Christian tradition, it is precisely through the celebration of Pentecost that the Muslim infidel is redeemed. Through the Pentecost even black Nimrod, who fomented the confusion of tongues and the scattering of the races, is brought back into the fold. 89 It is in this context that we find in Byzantine and Armenian miniatures depicting the original Pentecost a black cynocephalic "Ishmaelite" occupying the place of the "nations" of the world. 90 Of course, the Muslims saw things differendy: the Persian Jenari, in his Pir asset Nama ("Book of Physiognomy"), portrays the Slavonians as cynocephalic descendents of Japhet, the son of Noah who, when his wife died, had his son suckled by a bitch. The boy transmitted his acquired traits,

68

THREE

and these were the cynocephalic Slavonians encountered by Alexander. The same Cynocephali, called the Burous, are located in the lands of the Prussians, to the northwest of Russia, by Ibn S'aid.91 Medieval Europe came to know an invasion of another sort, one which proved to be even more dreadful and fatal than the Muslim presence: the Black Plague or the Black Death. This epidemic, which decimated Western Europe from the fourteenth century onwards, lent itself naturally to the canine death imagery of the hellhound, blended together with that of the werewolf and cynanthropy. So in Messina, where the plague first appeared, "demons in the shape of dogs" moved among men. "A black dog with sword drawn in his paws appeared among [the people], gnashing his teeth and rushing upon them and breaking all of the silver vessels and lamps and candlesticks on the altar and casting them hither and thither."92 A final medieval European canine or cynocephalic trope was that applied to the peasant, the vilain antitype of the noble courtois of courdy literature; and even more to the Wild Man. This creature, a product of popular imagination and perennial indigenous and foreign monster traditions, was also no doubt inspired by very real hermits and marginals who lived outside the periphery of society. Whatever his origins, this creature is always portrayed, in both art and literature, in the company of dogs, whom he sometimes resembles in his physiognomy. This parallelism of the outcasts or dregs of society with monsters on the limits of the world is self-evident: the same spatial symbolism of center and periphery, of inside and outside, is operating here on a national or cultural scale, as a reflection of a parallel phenomenon that was effective on an international or universal scale. Just as Dog-Men live on the edge of the world, so Wild Men and their dogs live on the edge of the country, town, or manor. The theological and ideological explanations for these two sorts of social lepers were transparendy alike: social differences were the result of the degeneration or decadence of a race that was perfect at the time of its creation. In both cases, the punishment was exile, from good society or from the human sphere altogether. Yet, there was an important difference between these medieval representatives of "Asianism" and "Atticism." The Cynocephali, as we have seen, were promised a place in heaven after the resurrection; the vilain and Wild Man knew no such promise of redemption. 93 These living representatives of the "other within," while they constituted the same safety valve for fantasy and savagery in popular imagination as did the mirabilia of the monstrous races, were themselves

The Cynocephalic Hordes

69

caught in a no exit situation. So, too, was the Jewish infidel and, in a later time the black Pygmy: considered to be the highest of animals but nevertheless subhuman, Pygmies could and would be enslaved, as servants to man, without concern for the ethical implications of such an action.94 The peasant's lot was enslavement in everything but name. Furthermore, in the elite society of the noble clergy, which placed a high premium on leisure and the contemplative life, the peasant's manual labor was considered to reflect his irredeemable fallenness. In Salvianus of Marseille's list of the mortal sins, for example, the servi formed a group of sinners who were contrasted with the noble warrior or cleric. Suplicus Severus equates these rustici with the gentili, the pagani and the inculti.95 They are depicted in courdy literature as vicious sinners, drunkards, paupers, lepers, and heretics; as rude in manners, rough in speech, immoderate, inconsiderate, unaltruistic, unjust, stingy, ill-dressed, cowardly, gloomy, ugly, smelly, stupid, black, hideous, deformed, slothful, and criminal. This last trait earned the medieval peasant his designation of vilain: a derivate of the word "vile," it is the root of the term felon.96 Shut out from the noble's city walls or manor, the peasant's forest dwelling is the place of wild beasts, of monsters and animal immorality. In effect, this stock characterization of the peasant—identical to that of the faraway monsters—renders him an undifferentiated part of an anonymous horde, existing as a nameless antitype—in demeanor, behavior, vocation, ethics, and lifestyle—to courtly society. The peasant's "villainy" is the justification for his exclusion. This excursus has apparendy carried us far away from the subject of this chapter—a particular race of generally foreign monsters in medieval European traditions. But in fact, as we have shown, the peasants of medieval Christendom differed little, in theological and social theory, from its cynocephalic monsters. Furthermore, their description in medieval chansons degeste and courdy romances, the popular literature of the time, is almost identical to that of their Ethiopian or Indian homologues. So the vilain Dangiers of the Roman de la rose is "tall, black and brisding, with red eyes, a wrinkled nose, a hideous face and a humpback." In the same genre, the vilain is often portrayed as a giant and, in extreme cases, as an animalheaded monster.97 As was done with the Muslim "Saracens," these giant, black zoocephalic peasants are often made to hold a great club in one of t heir hands. At this point, peasants or vilains lack nothing to become completely identified with the late medieval Dog-Man. And indeed, a fourteenth-century French source calls the peasants living between Nantes

70

THREE

and Redon "Pengouets," or "Dog-Heads," because they have not been converted to Christianity.98 However, contrary to the general lot of the vilain, the Wild Man—the domesticated version of the monstrous race—is often sympathetically portrayed as the harmless object of curiosity that these monsters originally constituted for the ancient world and its Hellenistic and medieval heirs. As we draw near to the time of the Renaissance, the Wild Man becomes a cultural fashion, appearing in book illustrations and tapestries and as a popular carnival character. And, to close this circle that began with Abominable and the Kalustrioi, we find in a sixteenth-century French illustration (pi. 9) the figure of a finely attired "Cynocefale," staff in hand, standing in the midst of an idyllic country landscape, with a look of aristocracy and understanding etched into his features—a truly noble savage.99

F

O

U

R

VISVÂMITRA AND THE DOG-COOKERS

As we have noted in the previous chapters, a recurring theme in the Western mythology of the monstrous races was their "Indian" location. This tradition of the monsters and wonders of the East began with the Greek historians Ctesias and Megasthenes, who confused mythology with ethnology. Indeed, the epic Mahâbhârata is spangled with references to Cyclops, Sciapoda, and Panotii (2.28.44-50), as well as to Blemmyes, Cynocephali, and other zoocephalic hybrids (9.44.54-100). As such, it may well have been a, or the, source of Ctesias's and Megasthenes's confusion. In India itself, however, such fanciful creatures were never given the attention they came to receive in the West, nor did traditions concerning them take on a life of their own. India did, however, have categories of its own to which it was most attentive, categories according to which all of humanity, as well as a number of troubling hybrids, were classified. One social (or anti-social) group within (or outside of) the Indian classificatory system were the Svapacas or Svapâkas—the "Dog-Cookers," "DogMilkers," or "Dog-People." This group will be the focus of the following two chapters. In the "Südravarga" section of the sixth-century lexicon thcAmarakosa we find, in immediate succession, lists of synonyms for outcastes, hunters, and dogs.1 The first of these lists is entided "The Ten Names for Candâlas"—the Candâla bore the generic sense of outcaste in India from the time of the Vedas (12th-10th century B.C.). In this list we find the following names: Plavas (Floaters), Mâtarigas (Elephants or Matriarchates), Nişâdas, Svapacas (Dog-Cookers), Antevâsis (Dwellers of the Ends), Pulkasas, Bhedas (Sundered Ones), Kirâtas, Sabaras (Corpse-Men), Pulindas, and Mlecchas (Babblers). This list may be divided into three categories. The first, composed of those names for which no translation is given above—the Nişâdas, Pulkasas, Kirâtas, and Pulindas—are tribal or ethnic names that the Sanskrit tradition of the Amarakosa probably took piecemeal from the 71

72

FOUR

peoples' names for themselves. The second category, the Antevâsis and the Bhedas, are people defined by their exclusion, in this case, from the society of the four proper castes. Other such designations, not found in this list, include the Niravasita Südras (Expelled Südras or Unabidable Südras), Antyajas (Last Bom or Lowest Bom), Antyâvasf.yins (People of the Lowest Station or Final Remainder), Bâhyas (Outsiders) and Avarnajas (People Without Caste).2 The third category is that in which reference is made to what these people do: Plavas float, Mlecchas babble, Mâtarigas presumably deal with elephants, Savaras (or Sabaras) dispose of the bodies and personal effects of criminals or persons without family ties, and Svapacas cook dogs. It is this last member of the third category, especially inasmuch as it is juxtaposed to lists of hunters and dogs, that will be the focus of our attention throughout the next two chapters. In fact, the term Svapaca is—like Candâla and more than any other of the names given in the Amarakosa—a general term for all of the outcaste peoples of India. We may therefore deduce that the outcastes are in part defined by a particular relationship that they bear with the dog. There are very few references to dog-headed humans (svamukhas) in Indian traditions, and it is rather these outcastes, whose relationship to the dog is of another stripe, that will concern us in this chapter. There is, nonetheless, a connection between these Indian peoples and the "Indian" Cynocephali most graphically described in Ctesias and Pliny. Here we are referring to the race called Cynamolgi, or "Dog-Milkers," in the Western sources.3 In the Sanskrit of the Indian traditions, a variant rendering of the term Svapaca is Svapâka. While the two are considered to be synonymous compounds of the word dog (svan: sva- in compounds) with derivates of the root pac, "to cook," the second term may also be glossed as sva-pa-ka, from the root pa, "to nourish, suckle." In this reading of the term, the Svapâkas would thus be a race or people "Nourished by Dogs," "Suckled by Dogs," or even "Children of Dogs." This reading is supported by the fact that there are absolutely no mythic references to these people ever cooking dogs (sva-paca). Svapâka may thus be a synonym of Cynamolgi.4 Whatever the case, the outcastes of India were so closely identified with the dog that we find, in scattered sources, statements to the effect that among the animals, the dog is the Candâla. As outcastes of their respective races, the dog and the Svapaca were nearly interchangeable elements in the ideology and symbol system of post-Vedic India.5 Here too, "milking" is important. As we will show, the animal with which the priesdy brahmin

Visvâmitra and the Dog-Cookers

73

was most closely identified was the cow. As provider of the five purest substances known to the Hindus (the pancagavya, composed of milk, yogurt, clarified butter, dung, and urine), the cow epitomized brahmin purity. More than this, it was from the cow that the brahmin obtained the major part of his vegetarian diet: because milk and its by-products are "cooked" inside the cow, they are eminendy suited to ingestion by the pure brahmin. In this perspective, the two poles of Indian society, the wholly pure brahmins and the wholly impure Svapacas or Svapâkas, are contrasted in terms of their diet: brahmins live by the cooked milk of their pure cows, while outcastes live by thefleshor milk of their impure dogs. The danger of mixing the two is a leitmotiv in Hindu sources, in which a dog's or outcaste's contact with a sacrificial vessel (containing a milk product) is an abomination for which ritual expiations must be observed by the pure brahmin sacrificer.6 With few exceptions, India's Dog-Man tribes or races were located within geographical India, and were thus persons who lived within or interacted with Indian society even as they were ideologically excluded from or marginal to it. Their lot was, and remains, most similar to that of the European Wild Men and vilains. Like these, they constituted an other within, but were rarely qualified as actually being cynocephalic. As with the European traditions, India located its dog-headed men (Svamukhas or Sunomukhas) outside of its geographical borders. 1. INDIAN TROPICS OF DISCOURSE

Before we discuss particular accounts of India's monstrous others, both within and outside of its borders, we ought to look first at the place of myth in the Hindu tradition and the place of caste in the social ideology of India. A pervasive trope in Hindu mythology is the use of accounts of familial, social, or cosmic crises as means to teach or reiterate a basic worldview and ethos. Because the Indian tropics of discourse7 are so different from those we have reviewed in the West, a brief comparison of the two traditions is in order here. As fanciful as Christian hagiography might appear, it nevertheless maintains a certain concern for the historical element or perspective. This is to be expected in a religious tradition in which the universe rides upon an arrow of time whose flight is of finite duration. In the Western context, this flight, this divinely ordained linear history, ends with a division of the saved from the damned, for all eternity. But the strong determinist bent in the JudeoChristian history of salvation cannot compare with the post-Upanisadic

74

FOUR

Hindu worldview. In this tradition, all of history has already been played out from the start—and the start has itself been played an infinite number of times as well, since time is cyclic and circular. There are more important differences. Eternity, as opposed to time, only occurs at the end of time in the historical faiths; in India, time is encompassed—indeed engulfed and rendered inconsequential—by eternity. On the level of temporality, of existence, an individual's karma—the fruits of his or her acts as these are suffered or enjoyed in successive rebirths—is to a great extent determined by one's adherence, in past and present lives, to one's particular social and caste function (svadharma). This is of course Krşna's teaching to Arjuna in the Bhagfavad Gltâ: humans must grapple with their self-allotted karma in existence in order that God might play out his vast game plan (lilâ) on the eternal level. Rather than revolting, man is to offer his travail with devotion to God whose grace will free him. On a social level, a well-tempered balance of functions between the various classes, types, or colors (varnas) of society is necessary to the furthering of the divine plan of creation. An imbalance—provoked, for example, by ksatriya warriors who usurp the functions, practices and powers of brahmin priests—may bring on a corrective dissolution and the destruction of the human race. The great war of India's (and the world's) greatest epic, thcMahâbhârata, is one case of such a cosmic readjustment. The events that take place on earth are a reflection of and a catalyst to events that occur in heaven. But the divine sphere too is subject to the laws of karma. More than this, the gods are as helpless as are humans before the destiny (daiva, bhâgya) of the universe—which is to turn relendessly onward and to complete, like a water wheel, circle after circle of cosmic return.8 Therefore, in spite of various mystical or technical ways out of this impersonal and endless pulsing oscillation,9 in the end—and there is no end—the system is all that really is. It is in this vertiginously vast context that the Hindu myths and sociopolitical realities we will discuss in this and the next chapter are cast. With regard to the place of Dog-Men as the Indian other within, the myths are seemingly repetitive, a kaleidescope of variations. But in fact each restatement of the never-ending story adds or reshuffles elements of the whole, and in so doing offers a new perspective on a totality that resonates on so many diverse yet interconnected levels.10 A most comprehensive account of the lot of the Svapaca or Svapâka, the "native" Indian Dog-Cooker or Dog-Milker, is found in the post-secondcentury A.D. didactic portion of thcMahâbhârata. Here, Yudhişthira asks the slowly dying Bhlşma the way in which a king ought to rule "in times

Visvâmitra and the Dog-Cookers

75

when all living on earth has become dâsyu-fıed [slave-like]," and when "time has arrived at a low point."11 In answer to his question, Bhîşma relates to him the story of Visvâmitra in the village of the Dog-Cookers:12 Once, at the twilight of the Treta and Dvâpara ages, and according to the ordinances of daiva, there came to pass a terrible drought of twelve years' duration. It was the approach of the end of an age, and the thousand-eyed Indra sent no rain. Agriculture and herding had fallen into neglect, buying and selling had ceased, associations and assemblies were dissolved, and the great festivals had completely disappeared. The Earth was heaped up with bones and skeletons, with creatures and men in great confusion crying "Alas, alas!" The greatest cities were deserted, and towns and houses burnt down. Cows, sheep, and water buffalo fought one another for food. The twice-born were slain, and their princely protectors laid low. In that dangerous time in which dbarrna had wasted away, those who were dying, afflicted with hunger, ate one another. The seers, having forsaken their rules of conduct, left their divine fires unattended, and left their hermitages behind. Under these circumstances, the venerable seer Visvâmitra, homeless and hungry, wandered about. One day, he chanced upon a settlement of Svapacas, of those injurers and killers of living creatures, who were living in a certain forest. The setdement was scattered with broken pots, spread with heaps of dog hides, and piled with the bones and skulls of swine and asses. The clothing of the dead and garlands that had been stripped from them lay strewn about. Their hut temples were hung with iron bells and surrounded by packs of dogs. The great sage Visvâmitra, overcome by hunger, entered that place; but in his begging, he found nothing at all to eat, neither meat nor grain nor fruit nor any other food. And thinking "Oh, what misery has befallen me!" he fell to the ground. The muni saw in the house of a Candâla a broad piece of stringy meat from a dog that had been slain with a spear that very day. He then thought to himself, "There is truly no other available way to save my life at this point. In times of distress, theft from one's equal or inferior is permissible. Theft is not a sin when that which is taken belongs to outcastes: I shall therefore take that food." Having made up his mind, Visvâmitra fell asleep. Then,

FOUR

seeing that it was the dead of night, and that all was asleep in the Candâla dwelling, Visvâmitra slowly rose and went into the Candâla hut. There, an ugly, phlegmy-eyed Candâla, who had been lying as if he were asleep, said with a broken, rough and cracked voice, "Who is pawing at my stringy meat? While the Candâla settlement is asleep I am awake, and not sleeping. Away with you!" Such were his cruel words. The sage quickly answered, "I am Visvâmitra." The Candâla, hearing the words of that great seer whose soul had been purified, tripped out of his bed and drew himself to his feet. With tears welling up in his eyes and hands folded in great respect, he said to Visvâmitra, "O Brahmin! What is it that you wish to accomplish in the night?" Visvâmitra then said to the Mâtariga, in emollient tones, "I am starved, at my last gasp. I shall take that dog's hindquarters. My vital breaths are leaving me, and hunger is destroying my memory. Therefore, aware of my svadharma, I shall take that dog's hindquarters. The dharma of famine is one that corrupts. So I shall take that dog's hindquarters. Fire [Agni] is the mouth and priest of the gods, and therefore pure and clean. Just as he, that brahmin, is an eater of everything, in the same way will [eating this dog's hindquarters] be righteous in my case." Having heard him, the Candâla said, "O great sage, hear my words: The wise say that the dog is the vilest of all game. Of the body, the vilest portion is indeed the broad hindquarters. The act you have resolved upon, particularly the taking of that which is proper to Candâla-hood, of that which is not to be eaten, is a perversion. O sage, seek another means by which to stay alive. O great muni, let not the desire for meat ruin your austerities. You should know that this path is a forbidden one. The dharma of the intermingling of castes [varnasamkâra] ought not be practiced. Do not allow yourself, you who are the greatest knower of dharma, to be sundered from dharma." So addressed, Visvâmitra, afflicted with hunger, responded once again: "Life is better than death. One should follow one's dharma in life. I am alive and desirous of food: I shall indeed eat food." The Svapaca spoke: "I really do not dare give it to you. Nor can I allow my own food to be taken. We two would both be stained with our innate pollution for that, if I were the giver, and you, O twice-born, the receiver."

Visvâmitra and the Dog-Cookers

77

Visvâmitra spoke: "After today, when I have performed this sinful act, I shall lead an exceedingly pure life. Possessed of a cleansed soul, I shall hasten back to the righteous path. Speak truly, O guru! Is this a pure or a defiling act?" The Svapaca spoke: "One must be true to oneself in matters of this world. You know what stain lies in this. He who would consider dog meat to be proper food would, I think, find nothing repugnant in this world." Having spoken to Visvâmitra, the Mâtariga fell silent. Possessed of consummate understanding, Visvâmitra then took the dog's hindquarters. The great muni clung to the fivelimbed [dog], in order that he might survive. Together with his wife, the great muni ritually prepared it. Then he went into the forest. At that very moment, Indra sent rain. Revived, all creatures and plants came to life. And the great Visvâmitra, with all stain burned away by long austerities, attained the highest and most wondrous of powers. [Bhlsma then summarizes:] "Thus one who is expert and high-souled and a knower of solutions should, through every possible means, elevate his low-spirited self. Applying one's intellect, one should always strive for survival. Man thereby obtains a pure life and comes to realize laudable goals. Therefore, he who is without fault and in possession of his faculties ought to maintain a firm conviction regarding dharma and adbarma in this world."13 The myth of Visvâmitra in the Svapaca village is a self-consciously explicit account of a universal crisis that is both a reflection of and a catalyst for human disobedience vis-â-vis the cosmic order. It is also a myth whose conclusion throws its reader or hearer out of the temporal and into the eternal focus. Moreover, this is presented as a moral teaching, intended for the edification of kings, on the subject of âpad-dharma, moral order in times of calamity. Âpaddharma is the "dharma of thinking on one's feet," a moral order of expediency in calamitous times.14 The time that this myth depicts is one of both social anarchy and natural catastrophe, "when all living on earth has become dâsyu-fied." This is a twilight between cosmic ages or yugas (just as are the Râmâyana and Mahâbhârata wars), and the setting, by definition, for a corrective dissolution. This leads to a literal re-solution when Indra sends rain to the starving Visvâmitra and a burnt-out Earth. The casting of a Svapaca or Svapâka is perfecdy logical in this instance, since the entire universe has been reduced, as it were, to an outcaste exis-

78

FOUR

tence. Only in the upside-down world of these outsiders, where life is always hell, is there an atmosphere of business as usual. For the fall of manifest existence (samsam) has done nothing more than to put the entire world on a Svapaca "footing" (pada, as in â-pad-dharma). And in these extraordinary circumstances, it is perfecdy fitting that a truly ethical teaching should spring from the foul lips of such a fallen creature. So it is that a Svapaca can instruct Visvâmitra that he, a twice-born, should not eat dog, the vilest of all foods. But, of course, the Svapaca has it all wrong as usual because, living in a perpetual catastrophic state of âpaddharma, he is unable to comprehend a special law of relativity for those particular situations in which the general law must be superseded by extraordinary behavior. As for Visvâmitra, he is proven right even when he is wrong when, by ritually cooking (and eating, according to Manu Smrti 1 0 . 1 0 5 - 1 0 8 ) the dog's hindquarters, he causes Indra to bring the social and cosmic crisis to an end. 15 So it is that, having had recourse to a starding metaphor (if not an allegorical conceit), the orthodox ideology ultimately reasserts its definitive superiority and authority, by making the Svapaca wrong even when he is right and the rsi Visvâmitra right even when he is wrong. In this way, the social order of existence is upheld in a narrative whose expressed purpose is to allegorize the relativity of that order. 2 . CAN A K I N G BECOME A TECHNICIAN OF THE SACRED?

In this myth it is Visvâmitra, and no other rsi, and Indra, and no other god, who participate in this drama on the limits of reality and sanity.16 Visvâmitra himself is a quite unusual specimen: he might even be qualified as the stock "renegade ra" of the Hindu tradition. Of the seven primal Indian sages (the saptarsis) who, appearing as they do in the Rg Veda, are as old as India itself, it is Visvâmitra alone who originally usurps the status of rsi rather than being bom to it. This he generally does at the expense of Vasiştha, a highly orthodox rsi with whom he is constandy quarrelling. The earliest instance of Visvâmitra's and Vasiştha's rivalry is found in the R0 Veda,17 in which Visvâmitra, associated with an aboriginal people called the Kîtakas, finds his sacrifices for Sudâs ruined by Vasiştha's son Sakti. In the Taittiriya Samhitâ ( 3 . 1 . 7 . 3 ) , Visvâmitra and Jamadagni defeat Vasiştha's sacrifices with superior sacrifices; in the Jaiminiya Brabmana ( 2 . 3 9 2 ) andBrhaddevatâ ( 6 . 2 8 - 3 4 ) , their rivalry escalates into bloodshed, with Sudâs killing Vasiştha's one hundred and one sons.18 The rivalry between the two rsis is escalated to a cosmic level in the Devîbhögavata Purâna ( 6 . 1 2 . 3 2 - 3 4 ) , where it is said that the quarrel of the two, who have taken

Visvâmitra and the Dog-Cookers

79

the

forms of carrion-feeding birds, is the cause for the decay of the entire While this episodic theme is retold throughout the length and breadth of the Hindu literary tradition, the classic version of these two ros' rivalry is in thcMahâbhârata (1.165.1-44; 9.39.2,11-29).

c o s m o s !

Visvâmitra, the royal son of Gâdhi, becomes tired and thirsty while hunting in the waste lands [marudhanvasu]. He comes to the hermitage of the brahmin sage Vasiştha, where he sees Vasiştha's wish-fulfilling cow [kâmadhenu], Nandinî. Visvâmitra begs Vasiştha to give her to him, but Vasiştha refuses, because without a cow he, a brahmin, would be unable to perform the sacrifices proper to his station. When Visvâmitra attempts to wrest her from him by force, she resists. Nandinl's eyes become red with fury as she produces from her anus hordes of Pahlavas, from her excrements Sabaras and Sakas, from her foam, Yâvanas, and from her urine Pundras, Kirâtas, Dramadas, Sirhhalas, Barbaras, Daradas, and Mlecchas. These excrement-bom outcaste hordes rout the army of Visvâmitra. After his defeat, a chastened Visvâmitra decides that the greatest power in the world is not that of princely [ksatra] dominance, but the power of brahman proper to the priesdy [brâhmana] caste. He then performs austerities of such intensity that he accedes to brahmin-hood [brâhmanatvam avâpa]. This is the mythologization of what has become a post-Vedic commonplace: Visvâmitra is the ksatriya who usurps the status of rsi, a status proper only to the brahmin, as epitomized by Vasiştha. In spite of the fact that he was born a prince, and was thus entided by birth to acceed only to the status of royal sage (râjarsi), Visvâmitra becomes possessed of the powers of a divine or brahmin sage (devarsi or brahmarsi).19 While he wrests for himself the ascetic sage's power, he never gains the latter's sanctioned authority, which is a matter of brahmin birth. In a sense, this stru ggle between Visvâmitra and Vasiştha mirrors the social upheavals of the seventh to fifth centuries B.C., which occasioned the co-opting, by princes, of priesdy prerogatives in matters of sacred knowledge. We find ample evidence for this in the Upanişads.20 The mythology of the upstart Visvâmitra may be seen in this light as a nposte by the priesdy caste to a princely power play against its authority. T h i s changing relationship between the two elements of the Hindu power elite is also reflected in the mythology of the Vedic and Hindu gods. Here

80

FOUR

we see a gradual replacement of Varuna, the Vedic god who represents magical and contractual priesdy authority, by Indra, the royal god of military might: in later mythology, Visvâmitra becomes a double of the ksatriya Indra and Vasiştha a double of his father,21 the brahmin Varuna. And, as had been the case with Vedic Indra, so it comes to pass with Visvâmitra: after losing many battles against the brahmin orthodoxy, embodied by Varuna and Vasiştha respectively, in the end Visvâmitra and his iconoclastic ethic (of nivrtti over pravrtti, of renunciation over conventional Vedic sacrificial activity) win the war.22 The myths of Visvâmitra apparendy sanction the new order represented by the heroic ascetic, whose austerities endow him with the power necessary to assert his renunciant ideal over the old sacrificial ideology. Indeed, the old sacrificial order of brahmanic Hinduism came to be the casualty of two complementary trends: in tandem with the renunciant internalization of the sacrifice first broached in the Upanişads, there emerged the new Hindu devotionalism, the religion of bhakti. In this changed spiritual environment, which uncannily arose at the same time as did the Christian gospel of love in the West, devotion to a personal god (generally the "minor" Vedic deities Vişnu and Siva) was reciprocated by that god's love, protection, grace, and salvific intercession at death. The bhakti cults, which have characterized popular Hinduism down to the present day, thus generally bypassed the elitist mechanism of the sacrifice and democratized Hinduism, bringing participation in the "high religion" within reach of nearly all classes and castes of Indian society. This new egalitarian spirit is reflected in Krşna's teaching, in the Bhagavad Gitâ (5.18), that the "knower" should see the cow, brahmin, and Candâla as one and the same. But here we have leapt far ahead in the chronology of this mythologized power struggle between elites, for it is precisely the orthodoxy's ideological ammunition in response to such heterodox statements that is of primary concern to us. 3. O N THE O R I G I N S OF THE NONEXISTENT F I F T H CASTE

A variant mythological explanation for Visvâmitra's ambiguous (and therefore dangerous) status is indicative of the symbols that the orthodox ideology deployed in its portrayal of its nemesis, the nonruling, renunciant râjarsi. In the Mahâbhârata, Visvâmitra is said to have been bom from a ball of caru into which his brahmin "father" Rcîka had placed the energy of the supreme brahman, the essence of brahmin-hood. This ball of food was swallowed by a ksatriya princess named Satyavatl. Technically, this makes

Visvâmitra and the Dog-Cookers

81

the issue of a mixing of castes (varnasamkâra), by the logic of he would himself be an outcaste by birth. 23 If the textual tradition does not emphasize such a strong reading of this mV th, it nevertheless throws Visvâmitra together with "outcastes" whenever possible. In the classical origin of his rivalry with Vasiştha, related above, Visvâmitra is once again placed in contact with such peoples when they issue from Nandinî's excrements, even if this is the only case in which they are not on his side. In this myth, as in that of Visvâmitra in the Svapaca village, the ambiguous sage passes through a great wasteland or desert region stricken with drought. The wilderness is the place, or nonplace (ou topos), of the Svapaca: it is that undefined space which lies outside the limits of the royal capital, town, or priesdy hermitage, or between two such centers.24 In a sense, once one is in the wilderness, according to the ideological perspective of these traditions, there is no one to meet but outcastes. None but the pariah lives outside the society of either the royal town or the hermit's âsrama; therefore, none but the itinerant renouncer or the hunting or exiled king can ever come into contact with these nonpeople, in the literature. A primal myth that brings together the wilderness motif, the origins of the outcaste races, the râjarsi Visvâmitra, a renunciant Indra, and the symbolism of the dog, is the legend of Sunahsepa:25 Visvâmitra which

Hariscandra, the king of the Ikşvâkus, has 100 wives but no sons. On the advice of the sage Nârada [or of Vasiştha, in later variants], he pleads with Varuna, the god of Law and Order, to grant him a son. Varuna acquiesces, on the condition that the son born to him be immediately sacrificed to himself. A son is bom, to whom the name Rohita is given. Hariscandra withholds him from Varuna each time the god demands his due. This occurs five times over a period encompassing Rohita's youth. Before Varuna returns for a sixth time, Rohita runs away into the forest. Back in his capital, Hariscandra suffers Varuna's punishments. This god of the waters and of binding contracts distends the king's belly with dropsy. In the forest, Rohita is approached each year, over a period of five years, by Indra, who has put on the guise of a wandering brahmin ascetic.26 In this form, Indra counsels Rohita to live the same wandering life as he. In the sixth year of his forest exile, Rohita comes upon a brahmin named Ajîgarta who is starving in the wilderness with his wife and three sons,

82

FOUR

Sunapuccha, Sunahsepa, and Sunolârigüla [Dog-Tail, Dog-Penis, and Dog-Hindquarters]. Rohita offers the brahmin 100 cows for one of his sons, who would thus replace him in his father's contractual sacrifice to Varuna. Ajîgarta agrees, but refuses to give up his eldest; his wife refuses to give up her youngest. So, by process of elimination, the middle son, Sunahsepa, is chosen. Varuna accepts this surrogate victim, allowing that a brahmin is a worthy substitute for a ksatriya victim. Hariscandra decides to combine this sacrifice with his own anointment and royal consecration. In the sacrifice, the sacrificial priests are Visvâmitra, Jamadagni, Vasiştha, and Ayasya. None but Sunahsepa's father, Ajîgarta, will kill the boy; and as Ajîgarta advances on him with the sacrificial knife, Sunahsepa begins to sing prayers to the Vedic gods. 27 With his last hymn, to Uşas, the Dawn, his three sacrificial bonds fall from his body and Hariscandra is loosed from his affliction of dropsy Sunahsepa then himself completes Hariscandra's sacrifice with the "rapid-pressing" hymn which has just been revealed to him. The sacrifice completed, the sacrificial officiant Visvâmitra now offers to adopt Sunahsepa as the eldest of his 101 sons. Sunahsepa accepts, reviling his father as a südra [the lowest caste in India prior to the conclusion of this myth] when the latter tries to claim him back. Visvâmitra then announces the adoption of the boy to his sons, and gives Sunahsepa the new name of Devarâta [God-Given]. Of Visvâmitra's 101 sons, Madhucchanda, the middlemost, and his fifty younger brothers accept Devarâta as their elder, and are blessed by Visvâmitra. Madhucchanda's fifty elder brothers refuse, however, and are thus cursed by Visvâmitra with the words, "May your descendants obtain the ends [antân] as their lot." These brothers and their descendants became the Andhras, Pulindas, and Mütibas, who live at the northern limits [udantya].28 Visvâmitra is said to be the founder of most of the Dâsyu [non-Aryan, base, outcaste] races. While it is highly likely that the original Sunahsepa myth (alluded to in the Rgf Veda passages attributed to the rsi Sunahsepa) was in some sense an astronomical explanation for the fact that the three stars in the tail of Ursa Minor never fall below the line of the horizon,29 its reworkings in the Aitareya Brahmana and the Sâhkhâyana Srauta Sütra transform it into an

Visvâmitra and the Dog-Cookers

83

origin myth for India's outcastes or Dog-Cookers. The lot of Visvâmitra's elder sons closely resembles that of the biblical fallen races—humans who, although created perfect, were disobedient. By turning away from their creator and the norms of good society, they brought on their own punishment, which took the form of exile. In this Indian example, the same sort of insidious ideological agenda skews the story from beginning to end. It is the same ideology that informs the myth of Visvâmitra in the Svapaca village, but in that myth, the Dog-Cooker was depicted, for didactic purposes, as a noble savage much like the Abominable of Ethiopic Christian legend. 4 . CURSES AND COUNTERCURSES

There exists a series of uncanny and certainly calculated links between the myths of the Visvâmitra cycle; these generally involve "homeopathic curses"30 and their realization. We will begin our discussion with the curse that links the Sunahsepa myth with that of Visvâmitra in the Svapaca village. In thcMahâbhârata, it is said that after Visvâmitra had cursed his fifty eldest sons to Svapaca-hood for their refusal to accept Sunahsepa, he praised Indra: this god, pleased, released Visvâmitra from an earlier curse.31 This curse was one placed on Visvâmitra by the sons of Vasiştha (those killed by Sudâs, or by Sudâs's grandson Kalmâşapâda, with the blessing ofVisvâmitra: Mahâbhârata 1.165.1-1.167.5) that he would one day eat dog meat. This he in fact does—in a not uncommon epic tendency to present at least two sides to every story—in our core myth of Visvâmitra in the Svapaca village. But this is only a beginning. For, as we have seen in the earliest versions of the Sunahsepa legend, the fifty elder brothers of Madhucchanda are cursed, one after the other, to obtain the ends (antân) as their lot. Their refusal to accept Sunahsepa as their senior causes them to be placed upon or beyond the (northern) end (ud-antya) of the world.32 In the Râmâyana (1.61.16) version of the Sunahsepa myth, Visvâmitra's elder sons reject what they see as a usurpation of their primogeniture on the following terms: "We regard that [accepting Sunahsepa as their elder] to be like eating dog meat [svamahsam iva bhojana]." Visvâmitra curses them to suffer that very lot (to be dog-eaters, i.e., Svapacas) on earth for a thousand years.33 Vasiştha places a similar curse on a prince named Satyavrata, who, after having ravished a brahmin girl, has been exiled by his father to live in the forest like a Svapaca. There, during a drought, 34 this renegade ksatriya kills

84

FOUR

one of Vasiştha's cows to save Visvâmitra's wife and son Galava from starvation—after which he has the gall to ask Vasiştha to send him to heaven! Because of this heinous sin of bovicide, Satyavrata is cursed by Vasiştha to become a Dog-Cooker in essence, rather than merely suffering a DogCooker existence: he also changes Satyavrata's name to Trisariku (Triple Sting). This chain of curses is brought full circle when Trisariku, after having undergone severe austerities, enlists the help of Visvâmitra to elevate him to or beyond his former station. When Vasiştha's sons hear of this "conspiracy of renegade ksatriyas," they curse Visvâmitra that he will one day eat dog meat—which, as we have seen, he does. But Visvâmitra pronounces a countercurse upon Vasiştha's one hundred sons at this time: they are to be reborn for 700 lifetimes as the outcaste (Mrtapâ) race of the Muştikas. He then kills them with his irresistible tapas (yogic heat), and his curse is immediately realized.35 The Harivamsa version of this myth leads into a discussion of the Ikşvâku Satyavrata's descendants: his son is none other than the Hariscandra of the Sunahsepa legend,36 at the end of which this entire cycle of curses "began."37 It is, in fact, the very popular myth of Hariscandra, the Job of medieval India, which also closes, in a sense, the Visvâmitra cycle. Here, Hariscandra has angered Visvâmitra by accepting Sunahsepa as a surrogate victim for his own son Rohita. 38 He is then ensnared by the powerful sage into making him, Visvâmitra, the hotr (offertory priest) for his royal consecration: After Visvâmitra has performed Hariscandra's râjasüya, he demands as his sacrificial fee the entirety of Hariscandra's kingdom. This Hariscandra cedes to him, but Visvâmitra, still unappeased, asks for more. Hariscandra goes to Benares with his wife Saibyâ and his son Rohita, to seek a servile position as a means to finding the wherewithal to pay his debt to Visvâmitra. He first sells his wife and son into the service of a brahmin and gives the money to Visvâmitra; but this is still is not enough. He then looks to sell himself to some one, and makes his way to the terrible cremation grounds of the city. Thereupon, Dharma,39 having taken on Candâla disguise, comes for him. This Candâla is foulsmelling, ugly, rough, bearded, black-skinned, with a mass of fanglike teeth and a pendulous belly, bile-injected eyes, and vulgar in his speech. He carries a mass of birds, is adorned with garlands taken from corpses, with a skull and a thigh bone in his hands, horrible [bhairava]^ and surrounded by a

Visvâmitra and the Dog-Cookers

85

pack of dogs [svaganâbhivrttah].40 This Candâla is an executioner as well as scavenger of the shrouds or blankets of the dead. Hariscandra goes to work for this horrible version of Dharma in disguise, on the hellish cremation ground. There, while still alive, he enters into another birth, and suffers one hundred years in the space of one year. During this time, he falls into a dream, and sees himself reborn as a Pulkasa outcaste who falls into hell. There, after a day of tortures that is like one hundred years, he is reborn as a carrion- and vomiteating dog. After this, he takes rebirth in a series of creatures, until he is finally reborn, in his dream, as himself, Hariscandra. He then goes to heaven, but is dragged back down to hell by Yama's henchmen. There, Yama is told by Visvâmitra of Hariscandra's lot: Yama instructs Hariscandra to return to earth and finish out his twelve years of suffering, and has him hurled back to earth. At this point, Hariscandra, the king fallen to the station of a Pulkasa, reawakens from his horrible dream. Some time later, his wife Saibyâ brings the corpse of their son Rohita, who had been bitten by a snake, to the cremation ground. She does not recognize Hariscandra, who has become like a Svapaca. Hariscandra recognizes Rohita, however, and he and Saibyâ agree to throw themselves upon their son's pyre when it is ignited, to put an end to their sufferings. At this moment, the Candâla reveals himself to be Dharma, the deification of the moral order and, with the blessings of Visvâmitra, as well as of Indra and all the gods, he elevates Hariscandra, his wife, his restored son, and his entire city to the aerial star of Saubha.41 The "truth" of these romanesque allegories is nothing less than an elaboration of the ideology of caste exclusion. The Dog-Men they describe are cardboard characters extracted from an anonymous aggregate; as with the Cynocephali of European tradition, these Indian monsters are nearly always nameless. It is, moreover, likely that many of the names given to these outcaste or foreign groups are themselves pious inventions. Their importance, in any case, lies in the explanations offered for their existence: they are fallen kings and brahmins, cast out from noble society through the curses of all-powerful rsis. We conclude this chapter by illustrating the "mytho-logic" of this cause and effect scenario:

86

FOUR

Source

Mbh 1.165 AB 7 . 1 3 - 1 8

Kâm 1.61.16 D P 7.16-18

Hariv 9-10

Ram

1.58

Crime Visvâmitra tries to steal Vasiştha's cow NandinI Visvâmitra's elder sons refuse Sunahsepa as their elder brother Visvâmitra's elder sons say accepting Sunahsepa would be like eating dog meat Hariscandra approves of the sacrifice of Sunahsepa Satyavrata kills Vasiştha's cow to feed Visvâmitra's wife and son Visvâmitra helps the Svapaca Trisanku to gain an atmospheric station

Outcaste Group Produced all outcastes, produced from Nandinl's excrements elder sons become founders of udantya races, through Visvâmitra's curse elder sons become Svapacas through Visvâmitra's curse

Vasiştha's sons curse Visvâmitra to eat dog flesh

Visvâmitra effects a countercurse on Vasiştha's sons to become Dog-Cookers: Vasiştha's sons become Muştikas

Hariscandra is forced by Visvâmitra to become a Svapaca Satyavrata is condemned by Vasiştha to become the Svapaca Trisanku Vasiştha's sons curse Visvâmitra to eat dog flesh, which he nearly does in the Svapaca village: MBh 12.139

Ram

1.58

Plate 1. Two Cynocephali (detail) from the tympanum (12th century) of the Basilique de Notre Dame, Vezelay, France. The tympanum illustrates the missions of the apostles to distant peoples. Photo by Alex Cormanski.

Plate 2. St. Christopher "the Chananean" from a 12th-century martyrology. Hist. fol. 415, fol. 50r, Landsbibliothek Stuttgart, West Germany. Bildarchiv Foto Marburg.

Plate 3. Deceased person and his mummy protected by Anubis (detail), 3d century A.D. Musee du Louvre, Salle d'Art Copte, Paris. Photo by Antonio Beltran-Hernandez.

Plate 4. Group of four dog-faced baboons worshipping the sun, from the southern face of the obelisk of Ramses II, 19th dynasty, Luxor, Egypt, ca. 1250 B.C. Musee du Louvre, Salle Egyptienne (D.31), Paris. Photo by Antonio Beltran-Hernandez.

Plate 5. Iskandar builds the Iron Rampart, from a copy of the Shahnama of Abu'lQasim Firdawsi known as the Demotte Shahnama. Iran, ca. 1335-40. (no. S86.0104). Courtesy of the Arthur M. Sackler Gallery, Smithsonian Institution, Washington, D.C.

Plate 6a. inde et Indochine (detail), from Cosmojjonie Üniverselle (1555), of Jean le Testu, fol. 28. Cynocephali, walled in by mountains above the Himalayan range, are figured along with Blemmyes, Sciapoda, Pygmies, and "Cham, king of all Tartary." Ministere de la Defense—Service Historique de l'Armee de Terre, Paris.

Plate 6b. La Terre Neuve (detail), from Cosmogonie Üniverselle (1555), of Jean le Testu, fol. 5 7v. American Indians(?) are portrayed as dog-, lion-, and boar-headed monsters. Ministere de la Defense— Service Historique de l'Armee de Terre, Paris.

Plate 7. "Cy dist de Pisle de Seilan," illustrating the Cynocephali sighted by Marco Polo on the Andaman islands, Livre des Merveilles, 14th-15th century, MSS français 2810, fol. 76v. Cabinet des Manuscrits, Bibliotheque Nationale, Paris.

. Cynocephalic wearing striped trousers, Râby kirke, Râby, Denmark, 1510. Photo by David White.

pn m

A «Sh-fc.

Plate 9. "Le Cynocefale," from a 16th-century French bestiary originating from the Abbaye de Saint-Germain-des-Pres. Cabinet des Estampes (Jb22), Bibliotheque Nationale, Paris.

iCSV'sHJstls*

Plate 10. Râma removing Râvana's spear from Laksman (detail), from a late 15thcentury Moghul illuminated manuscript of thcRâmâyana (no. 07.271), fol. 265r. Courtesy of the Freer Gallery of Art, Smithsonian Institution, Washington, D.C.

Plate 12. Horned cynocephalic demon, 5th century. Musee Guimet (no. MG 17234), Paris. Photo by Antonio Beltran-Hernandez.

Plate 13. Orientalism: Hindu myth of the Churning of the Ocean of Milk portrayed as the worship of a Japanese god. The artist substitutes a "western" cynocephalic for a Hindu demon whose features he was unacquainted with, and the editor attributes a Japanese origin to the tableau. From H. Picart, Ceremonies et coutumes de tous lespeuples du monde (Amsterdam, 1723-37), vol. 6, no. 3.

Plate 14. Parisian clothing store mannequin, 1986. Photo by David White

F

I

V

E

DOG-COOKERS AND OTHER BORDERLINE CASES IN ANCIENT AND MEDIEVAL INDIA

1. MISCEGENATION AND EXILE

The myths of the Visvâmitra cycle constitute one explanation for the origins and existence of the fallen or exiled races of Dog-Cookers on the periphery of society and the borderline of human categories. There is, however, an alternative explanation, one that is found in both the didactic epic and the lawbooks of the beginning of the common era. This is the ideology of damnation through miscegenation, or more properly, through the mixing of castes (varnasamkâra\ the blurring of categories between castes in its most extreme form. But here, rather than a rsfs curse, it is the impersonal workings of cosmic order and fate that strike down recalcitrant humans. The crime is much graver than a mere flouting of occupational categories, as it involves the dangerous mixing of incompatible body fluids across caste lines. The miscegenator and his or her offspring dwell in a perpetual state of exile from the body politic; more than this, they are exiled from organized space and time as well. Their sexual crime, according to the orthodox ideology, condemns them to live in an undefined wilderness and under a perpetual yugic twilight.1 The genealogies of these fallen races is developed, with obsessive meticulousness, in the Mahâbhârata and the lawbooks.2 Thus we find: "A brahmin woman and a südra [give rise to] a Candâla, the fierce killer of criminals who lives on the outside [bâhyavâsina] . . . From the Candâlas spring a race called the Svapacas, who are guardians of the dead [tnrtapâs]."3 "From the südra spring, in the inverse order, three low-born sons: the Âyogava, the Kşattr, and the Candâla, the lowest of men . . . One born from a Kşattr man and an Ugra woman is called a Svapâka."4 The Mahâbhârata also offers various alternative cycles of rebirths into which high-caste persons may fall through their evil acts. If the individual falls to an animal level of existence, the dog is generally his last subhuman incarnation, or the Svapaca his first human incarnation. So, for example, a 87

88

FIVE

sinner is reborn first as a worm, then a hedgehog, mongoose, boar, deer, bird, Svapaca, südra, vaisya and finally as a ksatriya. The transmigratory fall is sometimes dramatized in the epic: a Candâla, whose "body is smeared with the dust raised by dogs and asses" had, in a past life, eaten food sprinkled with milk from the stolen cows of a brahmin. The dog that accompanies him had also been a human in his past life, and is presumably fallen for the same reason as his master. If one merely falls to a lower human level, rebirth is usually as some sort of outcaste: thus, a brahmin pupil who humiliates his teacher is reborn as a Svapâka or a Pulkasa.5 In the cycle of rebirths, the body into which a transmigrating soul is born in a given existence is one's due, on the basis of the sum total of one's past deeds {karma) with respect to dbarma. Svapaca wombs exist in a given number in order that those souls—even those of brahmins, as we have seen—who so merit it may suffer a base existence during which, if they follow their svadharma, they may so improve their karma as to gradually climb the ladder of rebirths up to the noble higher castes. The ideology of which these aphorisms, lists, and fables were expressions was perhaps not intended to be racist or jingoist: it was merely implicit to an elitist worldview. It was also hypocritical, for the so-called outcaste is as caught up in the social network of reciprocity and interdependence as is the brahmin. Even as they are excluded from society, the outcastes have their social functions to fulfill, without which the higher castes would not be able to maintain their ritual purity and distance from them. They have been, since time immemorial, the garbage men of India, carrying off all that is dead or left over. This includes the removal of refuse and excrements from streets and dwellings, the slaughter and disposal of the remains of dead animals, and the preparation of funeral pyres. Their situation and duties, dramatized in our core myth, are set out in legalist terms in the Manu Smrti (10.51-56): The dwellings of the Candâlas and Svapacas shall be outside the village: they shall be made to eat from separate or broken vessels [apapâtra] and their wealth shall consist of dogs and asses. The clothes of dead bodies shall be their dress; their ornaments shall be of iron; and they shall be constandy wandering. One who follows the law shall not seek intercourse with them . . . They shall not wander about in villages or cities during the night . . . They shall carry out corpses of peoples without relations . . . They shall always execute criminals . . . and they shall dispose of the clothes, beds and ornaments of those executed.

Dog-Cookers and Other Cases

89

Elsewhere, the Dog-Cooker and dog are grouped together as recipients of the Vaisvadeva sacrifice, in which the householder throws food out upon the ground twice daily.6 2. T H E ATMOSPHERE: A VERTICAL WILDERNESS

Dog-Cookers are relegated to the outer periphery of towns in the lawbooks, and to the unordered, Utopian wilderness in mythic sources. In the latter texts, myths of Visvâmitra end with that râjarsi elevating Mâtariga, Trisariku, and Hariscandra to an atmospheric—though not a celestial—plane. As in the myth of Sunahsepa, there is an astronomical motif underlying these myths, but this is overshadowed by the ideological, casteoriented purport of the Visvâmitra cycle. In these myths, the ambiguous Visvâmitra apparendy elevates Svapacas to a spatial locus "higher" than that of brahmins on earth. This elevation constitutes a kind of liberation from the enslaving network of servile duties to which these peoples are otherwise condemned. As a mythological change in status, it parallels to some extent that of the cynocephalic Abominable in medieval Christian traditions. While Visvâmitra is unable to raise these persons to heaven, or to moksa (liberation), he seems to improve their lot in a tangible way. From another perspective however, the "atmospheric rise" of the DogCookers is nothing but an insidious reassertion of the old order, a sort of mythological equivalent to public housing projects. The atmospheric level to which degraded Dog-Cookers accede is a mythological half-way house that, in certain respects, is nothing more than a locus that is as vertically excluded as had been their earthly locus on a horizontal plane. Just as the Dog-Cookers' forest haunts or peripheral location on the confines of towns were undefined zones located between the ordered mandala of one town or hermitage and another, so their new atmospheric haunt is the relatively undefined zone that falls, by default, between heaven and earth. The atmosphere, like the wild forest, is a place of transition and motion, and is the realm of the wind (deified as Vâyu and the Maruts) and of atmospheric or meteorological phenomena: these latter are deified in Indra, who, it will be recalled, instructed the wandering Rohita to "move" {cara) in the Sunahsepa myth. We have already seen that the undefined forest zone is the haunt of Svapacas and their ilk; it is also that of a dread class of wild animals, the svapadas, or "dog-footed" creatures.7 This juxtaposition of atmosphere, wild forest, and things canine goes further: Saramâ, the divine bitch and companion to Indra, is also associated with the wind. So too are the monkey-god Hanuman and the wild lupine Mahâbhârata hero Bhlma-Vrkodara, who is an incarnation of Vâyu. When Candâlas are first

90

FIVE

mentioned, in the pwmsamedha (human sacrifice) rite of the Tajur Veda, they are offered to Vayu.8 On the earthly plane, it is only the individual who ventures into the wild forest who is likely to happen upon Svapacas and svapadas. Such individuals may be divided into two major classes: the royal hunter, who pursues game together with his retinue and hunting dogs; and the wandering ascetic, who quite often has a dog for his sole companion. In the king's case, the wilderness is more than a hunting ground: each of his forays into it is a conquest of wild nature and an expansion of the mandala of his rule. In fact, mythic accounts of such expeditions were generally descriptions of a form of royal ordeal that symbolized, especially when it concluded with the establishment of a royal shrine, a cosmification of chaos.9 This is reflected, in a narrative replete with canine symbolism, in the Mahâvamsa account of the coming of Buddhism to Ceylon (Sri Lanka):10 Prince Vijaya lands on Lanka on the same day that the Buddha enters into parinirvana. Soon after his arrival, there appears, in the form of a bitch, ayakkhini who is an attendant of the great yakkhini Kuvannâ. One of Vijaya's men follows after her, thinking, "Only where there is a village are dogs to be found." He finds Kuvannâ sitting at the foot of a tree, spinning, as would a woman hermit. She captures the man and binds him with her magical powers; but because of the power of the magic thread he wears, she cannot devour him. She then throws the man and 700 of his soldiers into a chasm. Vijaya comes searching for his men, and immediately recognizes Kuvannâ to be a yakkhini. He rushes at her, and throws a noose over her neck crying, "Slave, give me back my men or I slay thee!" She frees his men, and Vijaya and Kuvannâ sup together. She then transforms herself into a beautiful sixteen year-old maiden and, making a bed at the foot of the tree, invites Vijaya to lie with her. And seeing this, the king's son [Vijaya], looking forward to the time to come, [takes] her to him as his spouse and [lies] with her blissfully upon that bed. Vijaya hears singing in the night, and is told by Kuvannâ that these are thcyakkhas who rule the island, and that their capital city of Sisiravatthu is close at hand. Vijaya slays all the yakkhas^ and himself puts on the garments of thcyakkha kings.11 The wilderness, as a locus of unordered movement (as opposed to the stasis of the royal capital or the brahmin's hermitage) and spatial nondefini-

Dog-Cookers and Other Cases

91

tion, is inhabited by such nonbeings as the Candâla or Svapaca peoples, or the Indian dryads or sileni, the yaksas, (oryakkhas in Pali). While the yaksas could be, and were, domesticated and civilized in myths like this one, the Svapacas were, conversely, tamed without ever being socialized on the mythological or ideological level. The Dog-Cooker does come to work within the social sphere, and even to live on its confines. As such, he is the archetypal migrant worker, serving the inhabitants of the town even as he is excluded from social intercourse with them. And, as we have seen, even when the antyâvasâyin is saved from his exclusion, it is only so that he may be shunted to another excluded level, this time the atmosphere. As for the wandering ascetic, his rapport with the wilderness-dwelling antyajas is generally more cordial than that of the king and his orthodox ideologues, and from time to time we even find the renouncer sharing prohibited food with his black Dog-Cooker friends while the sun sets on yet another cosmic age. What makes the forest the place of the renouncer, apart from the obvious fact that it is a good place in which to be antisocial? Symbolically, the forest, as a zone without boundaries or limitations, can in the renunciant ideology evoke liberation from the bonds of society and existence and thus from rebirth. Viewed as chaos or living hell by those engaged in the ordered existence of the center, the forest represents freedom for the renouncer. The wind of the intermediate regions is, in fact, of paramount importance to the ascetic: this we find as early as the Vedic hymn of the kesin, who flies with the winds (Rg Veda 1 0 . 1 3 6 ) . Mastery of the breaths is fundamental to yoga, the highest form of Indian asceticism. Here, breath control leads to perfect stasis, even to enstasis (samadhi\ which in turn reverses the entropic processes of aging and decay and affords great powers by which the yogin becomes master of the natural universe.12 This combination—of a transient lifestyle with a particular attention to the volatile element wind—earned for these antisocial iconoclasts the symbolic place that they came to inhabit in common with the nameless masses that society rejected. For, in the eyes of society, the yogic renouncer too was an "other within" whose perverse existence troubled the orthodox order of things: this is illustrated time and time again in the mythology of Visvâmitra. 3. ANTISOCIAL MOTION AND MOVEMENTS IN THE VEDIC PERIOD

The sixth-century Brhat Samhitâ of Varâhamihira, though it is mainly devoted to astrology, astronomy, and portents, is valuable in the implicit associations it makes between a certain class of humans and dogs. Thus we

92

FIVE

read, at the beginning of the section entided "The Circle of Dogs" (Svacakra): If a dog urinates on a man, horse, elephant, pot, saddle, milky tree, pile of bricks, parasol, pallet, seat, mortar, standard, cowrie, or a flowery or grassy site, and then goes in front of a traveler [yâtr], then his [the traveler's] undertaking will be successful. If it urinates on moist cow dung, sweet foods will come to him; if on dry cow dung, then dried grain, molasses, or sweets will be obtained. Again, if the dog urinates on or kicks a poisonous or thorny tree, firewood, a stone, a dried-up tree trunk, or the bones of a cremation ground, and goes ahead of a traveler [yayin], this portends undesirable events. Urinating on a bed of pottery vessels that are broken or intact means a blemish [dosa] will befall the daughters of [the traveler's] house; on vessels [as well as sandals] in use, the pollution of the woman of the house; by urinating on a cow, a dog portends miscegenation with an outcaste [avarnaja].13 This section continues with further omens, generally favorable, regarding the behavior of the dog vis-â-vis the traveler. Thereafter, however, follows a long series of portents related to dogs in houses and towns that are, with few exceptions, extremely inauspicious. A few points are worth immediate note in this text. First, the urinating dog is associated with a person who is defined by movement: this is theyâir oryayin (traveler). Second, there is a clear distinction made between omens issuing from the dog's passage through: (1) what is clearly an auspicious outside setting (milky tree), such as the tamed forest or a hermitage; (2) an inauspicious outside setting (poisonous tree), such as the wilderness or the world at the end of an age, and (3) the ordered inside space of the household. In the first outside case, all the omens are auspicious; in the third, in which the dog urinates on household items, and finally on a family cow, the results are extremely dire: the soilure of the women of the house, and ultimately miscegenation with an outcaste.14 The symbolism of a dangerous mixing of categories is transparent here: canine (as in Sva-paca) excrements mixed with human (as in brahmin) eating vessels, and with the source of pure brahmin food (the cow, provider of thepancagavya) is catastrophic. In the reigning ideology, dog and cow, outcaste and brahmin, excrement and food, are polar opposites that are to be stricdy segregated in every respect. Varâhamihira (54.84) also makes a tell-

Dog-Cookers and Other Cases

93

ing remark with regard to the cosmic place of foreigners and Dog-Cookers in the social microcosm: "The corners of towns, villages, and houses are attended with evils for those who abide there; but persons of the lowest tribes, the Svapacas and so on, will prosper in them." This passage echoes the Manu Smrtfs legalistic relegation of the outcaste to the periphery of towns. It also reproduces a certain Indian view of the cosmos as a circle whose circumference touches the sides of a square at the four cardinal points: the outside is constituted by the four corners outside the circle's circumference.15 Let us turn now to the traveler of these omens, as it is he who corresponds to the errant ascetic of the wilderness. Both of the terms employed for him in Varâhamihira's text are participial forms derived from yâ (to go). As such, both his names and canine associations evoke earlier and later avatars of the same phenomenon. Among the Vedic gods, it is Indra (together with Rudra, whom we will discuss shordy) who is most closely and frequendy associated with various groups of wandering ascetics: one of these is known as the yatis, from yâ, "to go," or from yat, "to strive." The yatis have a checkered career in brahmanic literature. They are referred to in the Atharva Veda (2.5.3) and the Âsvalâyana Srauta Sütra (6.3.1) without censure. In later brahmanic sources, however, it is related that Indra was once barred from a soma sacrifice for the threefold sin of having killed Vrtra, mistreated Visvarüpa, and thrown the Arurmqghas or yatis16 to the sâlavrkas (literally, "house-wolves").17 Commentaries on these sources identify the yatis with non-Aryan peoples, antibrahmanical sects, and evil renouncers. Yet, the Mahâbhârata states that the virtuous südra Vidura practiced yatidharma (errant asceticism) up to the end of his life.18 Another group whose name indicates a wandering existence are the yâtus, also derived from yâ, "to go." In the Vedas, yâtu and yâtudhâna are both generic terms for sorcerers, and both are associated with dogs. In the % Veda (7.104.20) and Atharva Veda (8.4.20, 22), we find references to Indra, who combats Svayâtus (Dog-Sorcerers or Sorcerers having the forms of Dogs). 19 Indra is, as we have seen, closely associated with dogs and Dog-Cookers in other sources. Apart from adopting a wandering ascetic's garb in the Sunahsepa legend, in which a boy with a canine name is to be sacrificed, it is Indra who receives, in the Rg Veda, a sacrifice that includes 100 black dogs. The same Indra is said to have offered a dog as sacrificial victim in a time of distress.20 Indra, the god of the intermediate atmosphere, in fact participates in nearly every brahmanic myth in which dogs come into play. He is the ally of Rjlfvan and Mâtzrisvan, and the rival of the

94

FIVE

Kâlakanja demons, of which two become dogs. 21 Most importandy Indra is, in the Vedic period, the master (when he is not her son) of the divine bitch Sarama,22 the mother of Yama's two hellhounds, the Sârameyau. Also in the Vedas, Indra slays the leader of the Yâtumatls (a generic term for "sorceress"), a group whose name Sayana glosses with the term Yâtudhâna.23 Indra and a witch named YâtudhânI are brought together, in the didactic portion of the epic (13.94.5-13.95.81), in another obscure myth set during a drought between two cosmic ages: A südra named Pasusakha [the Friend of Beasts] marries a girl named Gandâ [Dirty], and they work as servants for the Seven Rşis, who have nothing to eat in a time of famine. In a previous time, King Saibya Vrşâdarbhi24 had offered his son to the Rsis as a sacrificial present. Now the young prince has died, and the Rsis have cooked his body, with the intention of eating it. King Vrşâdarbhi appears, and offers them cows and wealth if they will eschew eating the flesh of his son. An argument ensues, which results in the Rsis leaving the boy's body uneaten. An angry Vrşâdarbhi then creates an ogress named Yâtudhânı, and instructs her to slay the Rsis. The Seven Rşis come upon a beggar named Sunahsakha [The Friend of Dogs] who, very fat and well-fed, is wandering about with a dog. Seeing him, Vasiştha says, "The sacred fire of this one is not like ours . . . it is for this reason that we see both him and his dog so well-developed." They come to a lake haunted by YâtudhânI, who will only permit them to harvest its lotus filaments for food after they have given her an etymological explanation for their names. This they all give, until it is Sunahsakha's turn. Saying that he is incapable of explaining his name, he strikes her with his "triple stick," reducing her to ashes. The Rsis then harvest the lotus filaments, but discover after a time that a portion of them is missing. They all pronounce a curse upon the thief, except Sunahsakha, who blesses him. Sunahsakha then reveals himself to be the thief; and reveals as well that he is Indra, who has come to earth to destroy YâtudhânI. After this, he causes the Seven Rşis to ascend to heaven, where they are visible as Ursa Major. The late and often corrupt SkandaPurâna (6.32.1-100) relates the same story, but in this version, Indra, the "foreigner" whose "sacred fire . . . is not like ours," is disguised as Sunomukha (Dog-Face). Among mythical

Dog-Cookers and Other Cases creatures,

95

it is most often the asuras who are portrayed in literary references or in art as being animal-faced, and certain of these are clearly dogheaded (pi. 10). Among Siva's gana hordes, which are also composed of various nocturnal or demonic creatures, there are some who are cynocephalic.25 We will discuss India's rare references to foreign cynocephalic humans in the following chapter. Myths concerning Indra's interactions with these various "goers" of the wilderness—especially that of Sunahsakha—appear obscure at first blush. A recent work by the German indologist Harry Faik, Bruderschaft und Würfelspiel, casts these, however, in a new interpretive light. Falk's work is primarily a discussion of the connection between ritual dice play in ancient India and a group known as the Vrâtyas. I have treated the first element of Falk's thesis in an earlier work, 26 and will therefore not touch upon it here. The Vrâtya element, however, merits extended discussion, inasmuch as it ties together many of the apparendy obscure themes we have been looking at in the preceding pages. More than this, it offers us a glimpse of what may well have been the earliest and most explicit Indian tradition of the DogMan.27 We have already noted a striking preponderance of canine imagery in the mythology of Indra: in addition to his role in the Vedic myths cited above, it is also Indra who, in Puranic myths, takes the form of Sunahsakha to slay the witch YâtudhânI, and who seemingly rewards Visvâmitra'for his sacrifice in the Dog-Cooker village. These latter two myths, like the early Rg Veda (4.18.13) reference, are set in a time of cosmic distress, of drought and famine. The Sunahsakha myth, moreover, involves the cooking of a human sacrifice, that of Vrşâdarbhi's son, one which is not unlike the near sacrifice of the "canine" Sunahşepa. It also portrays Indra Sunahsakha as extremely healthy and well fed, by virtue of an alternative form of errant asceticism that he is practicing. The canine elements in these myths of Indra, as well as that of an incongruous ascetic lifestyle, take on a new relief when approached in the light of the Vedic descriptions of the Vrâtyas, a group generally identified with Indra's (and Rudra's) troop of Maruts, who are themselves deifications of the wandering wind. Apart from being identified with the Maruts, the Vrâtyas are also called dogs28 in the Vedic sources—and it is here that we break through to the most archaic level of Indian Dog-Man traditions. Over the past century, a great deal of Western ink has been spilled, much of it in vain, on the subject of the Vrâtyas.29 Until the 1960s the general consensus had been that the Vrâtyas were a non-Indo-European, anti-

96

FIVE

brahmanic group of antinomian forest-dwelling ascetics, possibly the precursors of Saiva and Tantric sects. While this interpretation is not without foundation, it is in fact a gloss on an ex post facto brahmin cover story, which quite succeeded in obscuring the venerable origins of the Vrâtyas. In 1962, Jan Heesterman suggested in an important article30 that the Vrâtyas were in fact an extremely archaic component of Vedic sacrificial society whose role was gradually phased out with the rise of the brahmin varna as sacrificial specialists. These Vrâtyas were then degraded in the later literature and cast in an antinomian, anti-brahmanic mold, with their sattra rites surviving in Vedic initiation rites and in certain periods in the vrata, or vow of the brahmacârin, the Vedic student. 31 Falk gready expands on Heesterman's discussion in a number of ways, the salient points of which we sketch out briefly here. First, the Vrâtyas were priests who lived and sacrificed as a group or a troop—like the Maruts 32 themselves—taking the sacrificial present, daksina, for themselves, rather than offering it to a brahmin technician. One of their number, however, was primus inter pares and had the task of distributing the sacrifice among the group. Second, the Vrâtyas performed their rites in the forest—rather than in the civilized space of human settlements—and in secret. Their sacrifices were called sattras, "sittings," as opposed to brahmanic yajnas. These sacrifices involved bovine, or even human, victims, and were thereby considered violent. Last, and most important for the myths under study here, the Vrâtyas' sittings took place in the season of Sisira, the dead of winter—the darkest and leanest part of the year, a time when "domesticated animals become like wild animals in their skinniness and boniness."33 These sittings, of either twelve or sixtyone 34 days in duration, were styled as therapeutic sacrifices: through the offering of a victim, all of nature, including cattle, humans, and the sun itself, were reinvigorated and made healthy once more. Vrâtyas are referred to as dogs in a number of passages.35 The most striking of these is a passage in the Chândogya Upanisad, in which a certain Baka Dâlbhya chances upon a circle of hungry dogs. The passage is entided the "[Samavedic] Chant of the Dogs."36 Then Baka Dâlbhya . . . went out for personal recitation of the Veda. A white dog appeared to him. Other dogs gathered around it. They said, "May you, O Lord, sing to obtain food for us. Truly, we are hungry." He [the white dog] then said to them: "Tomorrow at

Dog-Cookers and Other Cases

97

dawn may you gather around me at this spot." So Baka Dâlbhya . . . kept watch.

Then, just as [priests] here snake along hand in hand while chanting the Bahişpavamâna Stotra, even so did they snake along. Then they sat down together and began to pronounce the preliminary vocalizing [him]. They sang: "Om! Let us eat! Om! Let us drink! Om! May the gods Varuna, Prajâpati, and Savitr bring food here! O Lord of food, bring food here! Bring it! Om/" Baka Dâlbhya's name recurs in a number of other contexts that connect him with both Vrâtyas and dogs. These include the JaiminiyaBrâhmana,37 Châgaleya Upanisad,38 and an important myth in the Kâthaka Samhitâ.39 In these passages, Baka Dâlbhya is made out to be the sthâpati, the leader of a group of Vrâtyas, who are to be identified, according to Falk's thesis, with dogs. A locus for Baka Dâlbhya's Vrâtya rites is also given in a number of sources: this is the Naimişa, the "twinkling" forest. 40 The Naimişa forest is also the venue for a number of other obscure myths that bring Vrâtya dogs into play. One of these is the account of a sattra held by the Sunakas, the "Whelps," in the Naimişa forest, in the Châgaleya Upanisad;41 another is the sattra held by Janamejaya, in the presence of a certain sage named Sunaka, in the Naimişa forest. This myth serves as the outermost frame of the entire Mahâbhârata epic. Here, we are told that Janamejaya, the greatgrandson of the epic hero Arjuna, is holding a sattra in which the sacrificial victims are snakes. In the course of the sacrifice, he asks Sunaka to relate the deeds of his ancestors to him. At about the same moment, a Sârameya—that is, a son of the divine bitch Saramâ—approaches the sacrifice. The pup is beaten by three of Janamejaya's brothers. Saramâ, righdy pointing out that her son had not defiled the sacrifice since he had averted his gaze so as not to pollute the sacrificial butter, condemns Janamejaya for this gratuitous violence, saying that "fear would strike him" when he least expected it. The sacrifice is never completed.42 The sattras held by the Vrâtyas in the (Naimişa) forest in the dark dead of winter are portrayed as violent sacrifices in brahmanic mythology. Thus the Taittinya Samhitâ states: "the daksina of the sattra is 'itself,' atman . . . he who accepts [a daksina] at a sattra eats a corpse: a human corpse, [or] the corpse of a horse. Food is the cow."43 The Kâthaka Samhitâ refers to the same practices when it states that one should not sacrifice at a twelve-day

98

FIVE

sacrifice, since such is tantamount to eating human flesh.44 So it is that, in the sattra held in the Naimişa forest in Châgaleya Upanisad 2, the corpse of an officiant named Atreya is present. Elsewhere, the leader of the Vrâtyas in such a sacrifice is said to substitute the body of a cow for himself: thus he is "eaten" without dying himself.45 Due to a fundamental change in Indian society in the course of the Vedic age, as well as to a gradual movement away from the offering of blood sacrifices, these Vrâtyas and their secret forest sacrifices become degraded in later literature. This fall from grace is mythologized in a passage from the Pancavimsa Bmhmana, through a curse pronounced against them by Varuna:46 The god-worshipping Vrâtyas first held a Sattra with Budha as their as their sthapati. They consecrated the sacrifical ground without inviting King Varuna. Varuna cursed them: "I cut you off from a share in the sacrifice. You shall not know the way that leads to the gods."47 At that time there was neither juice in plants, butter in milk, nor fat in flesh, nor hair on skin, nor leaves on trees. But since the god-worshipping Vrâtyas performed this sixtyone-day sattra, creatures have been endowed with virility and have been full of radiance and vital fluids. This passage, besides explaining the fall of the Vrâtyas, also delineates the Vrâtyas5 ancient forte: this was to restore the natural world from a state of hunger and diminished strength to new health and vigor, through their twelve- or sixty-one- day sacrifices. According to Falk, the twelve-day sacrifices of the Vedic Vrâtyas were the ritual cognate of other Indo-European phenomena, including the Roman Lupercalia and the twelve nights of Christmas, in which the wild hunter Wode-Wodin roared through the forests of northern Europe. 48 The wild hunter of the Vedic forest was Rudra, the "howler" who, like Indra, was the lord of the Maruts, with whom the Vrâtyas identified themselves.49 We will return to Rudra-Siva's connections to these Dog-Men shordy. The sixty-one-day sattra, also performed by the Vrâtyas, and which took place over the entire period of the season of Sisira,50 is especially rich in "canine" allusions. First, there is the duration of the sacrifice itself: in nature, a dog gives birth after sixty-one days of gestation. This is also the duration of the season of Sisira: each of the six seasons of the Indian year are approximately sixty days in length. The temporal connection is especially important in the case of Sisira, however. This is because the name

Dog-Cookers and Other Cases 0f

99

the season itself is nearly identical with that of the male hound who is the divine bitch Saramâ's consort, whose name is Slsara.51 In this light, it is possible to generate a theory concerning the sacrificial behavior of the Vrâtya "dogs" in their sixty-one-day sacrifices in the season of Sisira. First of all, the sacrifice performed was one of a "human" cow,52 offered by "dogs" in a period of leanness and darkness, a sacrifice that served to revivify and restore nature and the lowering sun. The Vrâtyas' sixty-one-day sacrifice, by which they "won" the year at a time in which the sun was at its lowest point in its annual celestial movements, and in which human and natural life and vigor were at low ebb, was nothing less than the birthing of a new year with new life and a rising sun. And, because radical problems call for radical solutions, a secret bovine or human blood sacrifice, held in the wild forest by "dogs" over a "canine" time frame, bore a logical place in the Vrâtyas' archaic ritual. Ultimately, it was through a ritual canine gestation, on the part of a symbolic human dog (or group of dogs) nourished on the flesh of a cow, that the year and life came to be renewed in the winter darkness of the wild forest. 53 Here, the Vrâtyas' work was essential to the survival of nature and society: without their intervention, however terrible, all cows would die, the sun would not make its northward turn, and the universe would fall into chaos. However, with the emergence of a later socioreligious compact between brahmin sacrificial technicians and royal clients (yajamanas), the Vrâtyas became marginalized, together with their canine symbolism, to the profit of the brahmin and his sacred cow.54 It is in the light of the archaic sattras, held in the Naimişa forest by Vrâtya "dogs" who were at once human and canine, and whose victims are thus at once human, canine, and bovine, that many of the myths of the Visvâmitra cycle are to be understood. Visvâmitra's cooking of a dog in the Svapaca village, the near sacrifice of the canine Sunahsepa, Satyavrata's cooking of Vasiştha's cow, and perhaps the possessed king Kalmâşapâda's devouring of Vasiştha's sons, all hearken back to the practices of the Vedic Vrâtyas. Indeed, Visvâmitra's very nature, of being "betwixt and between" brahmin and ksatriya, also hearkens back to the ambiguous, and thereby dangerous, status of the Vrâtyas.55 However, the myth most open to elucidation through our discussion of the Vrâtyas is that of Indra-Sunahsakha. While this myth makes no explicit mention of the Naimişa forest, many of its themes evoke that place of Vrâtya ritual sacrifice. First of all, this is a period of great penury, a twelve year drought which may be identified with the twelve days of the shorter sattra. The rsis are starving; indeed, they are

100

FIVE

so hungry as to have been driven to cook the human body of prince Saibya Vrşâdarbhi.56 Later, when they meet Sunahsakha, "Friend of Dogs," and exclaim over his excellent health and his adiposity, each gives a reason for it, a reason that stresses his practice of non-Vedic rites: he does not offer the agnihotra offering at dawn and dusk, he is not perpetually burdened with their sacred precepts, and so on. 57 Sunahsakha, the "Friend of Dogs"—i.e. Indra, the leader of the Maruts, disguised as a non-brahmanic sacrificer living in the wild forest—is the sole well-fed creature in a time of general hunger. Moreover, when the rsis are tested by the witch YâtudhânI, whose own canine connections are numerous, it is Sunahsakha alone who is able to "answer" her riddle and thus not be struck down by her, a direct and highly meaningful reversal of the lot of the Vrâtyas in brahmanic mythology.58 At the end of this myth, after he has stolen the lotus stalks and parried the ms' curses, Indra raises them to heaven, and, presumably, good health—just as the Vrâtyas did for nature and the lowering sun. Indra Sunahsakha thereby stands as a Puranic reference to India's original Dog-Men, the Vrâtyas, whose socioreligious status was as lofty as that of the brahmins, once upon a time. 4 . ANTISOCIAL M O T I O N AND MOVEMENTS IN POST-VEDIC INDIA

As Indra's importance in the Hindu pantheon waned, his decline became translated into his degradation in epic and Puranic mythology. This is particularly the case with the epic emergence of the myth of Indra's brahminicide (brahmahatyâ). Here, because he has slain Vrtra, and in spite of the fact that doing so constituted a great service to the gods, he is punished for having murdered a brahmin. 59 It is to this feat that the Aitareya Brâhmana (7.28.1) refers, and the Mahâbhârata myth, in which Indra is pursued by an ogress named Brahmahatyâ, is an elaboration of this statement. Indra is not only degraded in later mythology: at times his place is co-opted, as if by a younger actor, by a less otiose god. This is the case with certain of the myths of Indra's wandering existence that we have reviewed, including that of his brahminicide. Taking over Indra's role on the heroic level is, as we have shown, the renegade sage Visvâmitra. On the divine level, however, his replacement is the enfant terrible (one of his names, Batuka Bhairava, literally means "terrible child") of Puranic mythology, Bhairava.60 Bhairava's mythology is not only that of Indra with a Saivite veneer, it is also the continuation of the original Vedic mythology of Rudra and a cast-

Dog-Cookers and Other Cases

101

ing into narrative of certain sectarian initiation rites. Moreover, it reflects a Vedic relationship between Indra and Rudra. First, the Maruts and the Rudras were nearly identical to one another, as denizations of such Indo-European confederations as the Vrâtyas. Second, Indra himself belonged to a general class of Vedic gods known as the Rudras. 61 T h e Vrâtyas themselves, as well as the later orthodox brah-

macârins who undertook a vrata, a vow, near the time of the winter solstice, clad themselves after the fashion of Rudra, wearing a black hide, turban, uncut hair, etc.62 When the Vrâtyas were cast as cow-slayers in their sattras, they were identified with Rudra Pasupati, who himself afflicted cows with disease and death—unless the Vrâtyas pacified him with their bovine victims.63 When the Vrâtyas slay a cow on Rudra's behalf, they are said to be his "dogs" or "wolves," and lupine and canine symbolism is nearly as abundant in the Vedic Rudra's case as it is in that of Indra. Thus the Sâhkhâyana Srauta Sütra (4.20.1) compares Bhava and Sarva, the sons of Mahâdeva (Rudra-Siva), to "snapping wolves who dash all around the wilderness." Dogs and leaders of dogs (svapatis) are extolled together in the Satarudrlya hymn.64 Parallels between the cults of Rudra, the wild hunter of the forest, and those of the Germanic Odin/Wodin, as well as the Iranian Aesma and a number of other Indo-European gods associated with the twelve nights of midwinter, are significant here.65 The mythology of Vedic Rudra bespeaks his fierce and terrible nature. It is said that he is first created from the fierce portions of the gods in order that he may punish the incestuous Prajâpati by shooting him with an arrow. In exchange for this feat, the gods make him Pasupati, the "Lord of Creatures." In later traditions, the myth takes the form of Siva's (or his surrogate Virabhadra's) destruction of his father-in-law Dakşa's sacrifice. In this myth, Siva's consort SatI, insulted because her husband has not been invited to participate in the rite, burns herself on the sacrificial pyre. After destroying the sacrifice and decapitating Dakşa, Siva wanders over the universe, carrying his beloved's corpse over his shoulder. Another variant on the Vedic original is the Lingodbhava (Manifestation of the Phallus) myth, in which we find Brahmâ and Vişnu arguing over who came first. Siva appears before them, stepping out of the endless linga of light that is his hierophany. He challenges the two to find either an upper or lower end to his linga. Brahmâ, whofliesupward as a swan, falsely claims he has found the top, but in the end, the two are obliged to recognize Siva's primacy and superiority.66

102

FIVE

The late Puranic myth of Bhairava's brahmahatyâ begins on the same note as the Lingodbhava: Brahmâ and Vişnu (who is in his Yajna form, as the boar who incarnates the sacrifice) are arguing over primacy. Siva appears in a ball of light. Brahmâ insults him with his fifth head, calling him Rudra (Crybaby). Siva then creates an enormous black man named Bhairava, whom he orders to cut off Brahma's offending fifth head. This Bhairava does (thus making five-headed Siva the god with the most heads), and, once Siva revives them, Brahmâ and Vişnu acknowledge his primacy. Bhairava, however, is left holding Brahmâ's head: in fact, it will not come loose from his hand. Siva then instructs him to undertake, for the edification of the universe, a twelve-year penance called the Kâpâlika-vrata (The Vow of the Skull Bearer). Bhairava, now accompanied by a dog, begins to wander the earth; because he has killed a brahmin, he is pursued by the ogress Brahminicide. Wearied by his penance, and terrified of the ogress, Bhairava eventually makes his way to Vişnuloka, where the god Vişnu tells him that he is himself Siva, and that he need only go to the tirtha of Kâshı (Benares) to stop playing his own game (lilâ). Bhairava arrives at the northern edge of Benares, and when he bathes in a tank there, Brahmâ's skull falls from his hand. He then establishes himself at the place where he was freed from the skull (Kapâlamocana) and becomes the "police chief" of the city.67 In Benares alone, Yama, the god of the dead, has no power over those who die. By Siva's grace, those who die in this holy city attain immediate liberation. It is Bhairava, however, who is the administrator of this unique system. As such, Bhairava is the lord of the Sârameyau, the two whelps of Saramâ and the dogs of death; moreover, his vehicle is the dog, and he is sometimes himself attributed canine characteristics. It is not only because he is Yama's replacement that Bhairava is the modern god most closely associated with the dog (pi. 11). This association goes back to the status of his devotees, as well as to the nature of the sect for which his twelve-year penance served as the mythological model. The two elements in fact go hand in hand: the sects that institutionalized Bhairava in Saivite Hinduism were generally anticlerical and anticaste, if not simply antinomian. They rejected caste distinctions, and some even prohibited brahmins from their fold. The other side of this egalitarian coin was their elevation of outcastes into authentic devotees of Bhairava: according to the late sâstra literature, Bhairava's are the only temples that may be founded by outcastes. His are also among the very few temples into which outcastes may enter, and the

Dog-Cookers and Other Cases

103

in which dogs are welcomed and even revered. It is likely that a aood number of the members of the Tantric sects were themselves outcastes bv birth, for this was the way they were depicted in the more orthodox puranic literature. In this perspective, the description of the Candâla for whom Hariscandra worked (even if he was Dharma in disguise) made him both a Savara (Corpse-Man) and a terrible (bhairava) skull-bearer (kapâladharin). Furthermore, the earliest literary references to Bhairava associate him with the Sabaras.68 The Sabaras are among the few indigenous peoples in India today for whom the dog plays an auspicious role. Pet dogs, whom they honor as brothers, are given burial rites by the Sabaras. The Ao Nâgas, another tribal group, also revere the dog, and are the sole people of the subcontinent to have a dog ancestry myth. One cannot discount the possibility that the orthodox association of these peoples with the dog had a certain ethnographic element.69 The sectarian origins of the cult of Bhairava are more direct and explicit than the ties it may have had with Svapacas or other indigenous peoples. The vow performed by Bhairava during his twelve years of wandering was called the Kâpâlika vrata. In fact, there was a terrible Saivite sect of the medieval period who called themselves the Kâpâlikas. Their supreme god was Bhairava, and their initiation followed that god's mythic model. The first reference to the Kâpâlika vrata actually predates both the formation of the Kâpâlika order and the Puranic Bhairava myths. This is the Tajnavalkya Smrti (3.243), a second- to fourth-century text in which the penances to be undergone by one who has killed a brahmin (brahmahatyâ) are legalistically described: he is to wander about for twelve years, staying at no place for more than five nights, carrying his victim's skull as his begging bowl and proclaiming his crime. A source contemporary with the Tajnavalkya Smrti is the Âpastamba Dharma Sütra, which states (1.28.21): "The slayer of a learned brahmin (or fetus), having put on the skin of a dog [svâjina] or the skin of the donkey with the fur facing outwards, carries the man's skull for drinking."70 What made the Kâpâlikas different from those who would have performed the penance described in the Tajnavalkya Smrti and elsewhere was that these iconoclasts purposely murdered a brahmin in order to have his head, perform their twelve-year initiation, and then continue to eat and drink from his skull. The notorious activities of these, the Hell's Angels of medieval India, are chronicled in a great number of sectarian and literary sole temples

104

FIVE

sources,71 including the seventh-century south Indian Mattavilâsa, in which a madman returns to a Kâpâlika his missing skull bowl, saying he had taken it from "a most respectable dog belonging to a Candâla.5'72 It is highly probable that the Kâpâlikas were responsible for the insertion of Bhairava's "passion" into the various Purânas in which it is presently found. They were, however, not alone in this work of sabotaging other sects' canonical texts for their own purposes. They were following the lead of at least one earlier Saivite sect, the Pâsupatas, followers of Rudra-Siva in his Pasupati (Lord of the Creatures) form. It is possible that this latter sect had historical ties with the Cynics (the Canines) of Diogenes Laertes: there are striking similarities in the practices of the two, and both were centered in port cities (Sinope on the Black Sea and Broach on the Arabian Sea) that had trade relations in the first centuries of the common era.73 For the past eight hundred years, there has existed a cult of Siva in Ujjain (an important stronghold of the Bhairava cult since at least the twelfth century). This cult, first attested to in the twelfth-century Sahkara Vijaya, is unusual inasmuch as devotion takes the form of cynanthropism.74 At the Mallâri temple there, devotees behave like dogs in every respect, barking and eating off the ground. The Mallâri cult remains strong in many parts of the Deccan, especially in Maharashtra, Karnataka, and Andhra Pradesh. In it, one may glimpse survivals both of the Pâsupata tradition of feigned madness and dissimulation,75 and, going even further back, of the Vrâtyas of Vedic tradition. Mallâri is himself closely identified with Mârtanda Bhairava, who is accompanied, in legend, by seven hundred dogs. These dogs are his devotees, who were first transformed by Mallâri/Mârtanda Bhairava from tigers to humans, and then instructed by him to behave like dogs. 76 On the ninth day of the Dasaharâ festival, the Vaggayyas, the "tiger dogs" of Mailar/Mallâri/Khandobâ/Mârtanda Bhairava, commemorate this mythic event, by which they helped Bhairava to defeat Malla, when they: 77 act and bark like dogs when pilgrims arrive . . . Formerly . . . they would run to their [begging] bowls, would howl and bark and quarrel amongst each other, and lie flat on the floor to eat like dogs . . . If food is offered into the bowls, they will fight like dogs trying to tear food from each other's mouth. According to Falk, the Mallâri's Dog-Men are to be placed in direct correspondence to the Vrâtyas of the Vedic period, with the ten days (dasa-

Dog-Cookers and Other Cases

105

the Dog-Men's festival corresponding to the ten days that in fact - o n s t i t u t e d the twelve nights of the Vrâtyas' dvadasaha ("twelve days"— actually ten nights of sacrifice, framed by two days) sattra. In this perspective, Bhairava-Mallâri's slaying of Malla on the tenth night of Dasaharâ s parallels the human sacrifice that ended the Vrâtyas' sattra7 So it is that the pedigree of tantric Bhairava's Dog-Men devotees is as venerable as that of the brahmin orthodoxy from which they generally distance themselves, extending back to Rudra's Vrâtyas, the Dog-Men of the Vedic age. In India, following the Pâsupatas and Kâpâlikas, the Nâth Siddhas, Dattâtreyins, and Aghoris—all medieval Saivite sects—spread the cult of Bhairava wherever they went. In this way, Bhairava's tantric cult was spread over the Deccan plateau, northern and western India, Bengal, and Nepal— in Nepal, Bhairava continues to be a highly important Buddhist divinity. The Nâths do not seem to emphasize Bhairava's canine connections, as none of the Himalayan Bhairava temples that they established show him with his canine vehicle. However, a Nepali tradition relates that Bhairava, having taken the from of a dog, barked "bu," by which Bungagaon was recognized to be the birthplace of Matsyendranâth, the great Nâth guru and patron deity (as a thinly veneered Avalokitesvara) of the Kathmandu Valley. Also in Nepal, the dog is one of the four creatures revered, with flower garlands, sweetmeats, and unguents, at the annual Tihâr (Diwall) festival. In certain north Indian folk rites of exorcism and the like, Bhairava, when called upon, appears in the form of a black dog. 79 The Vedic Indra and Yama and the Saivite Bhairava aside, Dattâtreya (a Siva-Vişnu hybrid when he is depicted as a god) is the only Hindu divinity closely associated with dogs. In his iconography, he is, in fact, surrounded by four dogs: these are styled as the four Vedas or the four human appetites, both of which the guru-god Dattâtreya mastered. Dattâtreya's cult is centered in Maharashtra, where later saints, such as Eknâth and Tukarâm, traditionally initiated by Dattâtreya, are also associated with dogs. 80 In a hagiographic source, the fourteenth- to fifteenth-century Sahkara Dig vijaya of Mâdhavâcârya, Siva is portrayed after the same fashion as Dattâtreya. Here, the orthodox scholar-saint meets, on a narrow path, a Svapaca accompanied by four dogs. He orders the outcaste to give way to him, at which point the Svapaca launches into a philosophical discourse, reminiscent of Krşna's Bhagavad Gttâ 5.18 teaching, on identity and difference! Sahkara acknowledges his error, whereupon the Svapaca reveals himself to be Siva, surrounded by the four Vedas. | ıar â) of

106

FIVE

Another account, also found in the SahkaraDigvijaya, turns this encounter into an armed sectarian confrontation. Here, a Kâpâlika named Ugrabhairava nearly wins the philosophical argument by cutting off Sarikara's head, but in the end he loses his own life at the claws of Sarikara's disciple Padmapada, in whom the Man-Lion Narasimha has taken a timely incarnation. A later Nâth version of the same legend, found in the seventeenth-century compilation known as the Gorafoasiddhântasamgraha, transforms the Kâpâlika Ugrabhairava into an incarnation of Bhairava himself. The dread god cuts off the heads of Sankara and four of his disciples, but later restores them, after which "true detachment arose."81 5 . PURITY AND POLARITY: RADICAL PROBLEMS AND SOLUTIONS

Leaving Saivite traditions of wandering ascetics behind, we will close this discussion of unordered or antinomian movement and space with the didactic myth of a brahmin who dreamed he was an outcaste and a king, found his dream to be confirmed by other peoples' testimony, and then learned from Vişnu that consensus does not verity make. This is the Yogavasistha myth of Gâdhi, which is evocative in at least two ways. First, we have already seen, in the Hariscandra myth, the theme of a person becoming a Dog-Cooker in a dream; and second, Gâdhi is one of the patronymics of Visvâmitra. In fact, one sees much of the renegade sage Visvâmitra in the Gâdhi of this quite orthodox Vaişnava text: By performing strict austerities, the brahmin Gâdhi obtains a boon from Vişnu. He asks that Vişnu explain mâya, illusion, to him. Some time later, while plunging himself into the water of a pond, he has a vision of his own death and cremation. He then sees himself reborn out of the stinking womb of a Svapaca woman. He then grows up as a Svapaca named Katanja, who lives in squalor and who hunts with dogs [sârameya]. He marries a SvapacI and has children by her. A terrible drought arises, during which his family dies. He begins to wander through other lands. There, in a country called Kira [Kashmir], he comes into a royal capital where the king has just died. Upon his arrival, an omen is fulfilled, by which he is named king of the country. He rules there under the name of Gavala for eight years, with no one knowing that he is a Svapaca, until a fellow outcaste from his old village chances into his capital and recognizes him, saying "Bho, Katanja!" When the word of Gavala's true identity spreads in Kira, it is, for the royal household and ministers,

Dog-Cookers and Other Cases

107

as if a corpse had been staying in their home. They run as if from a râksasa, to plunge themselves into a pool of fire. His kingdom now reduced to a state of "nontime at the end of a cosmic eon," Gavala too throws himself into the fire, and awakens from the fire into the water of the pond into which he had plunged as the brahmin Gâdhi. Some time later, Gâdhi receives a visitor who relates an account he had heard of an untouchable who had ruled for eight years in the northern regions before being unmasked . . . The perplexed Gâdhi begins a period of wandering to find out about "himself." He finds the village in which he had been Katanja, a truly ghoulish place in which the people smear the bones of the dead of whose bodies they dispose with bitch's blood and yellow coloring, wearing these as ornaments while they sing and dance on the cremation grounds. Leaving this land of ghoulish creatures [bhütamandala or "land of the Huns," Hüna-mandala82] behind, he comes to the paradisical Himalayas and the land of Kira, where he learns that "he" had ruled there as Gavala, who had committed suicide with most of his entourage twelve years previously. After some time, Vişnu comes to explain to him that his own and the Svapacas' and Kıra peoples' experience of him as Katanja and Gavala were all illusory.83 The doomsday imagery of the Bhütamandala, as well as of the Kıra holocaust, are reminiscent of the apocalyptic settings of the myths of Visvâmitra in the Svapaca village and Hariscandra on the Benares cremation grounds. The comparison of a Svapaca in the royal household to a corpse further evokes the death or doomsday imagery of the myths of Dog-Cookers. It also broaches the problem of implicit and explicit purity codes and violations vis-â-vis the outcaste. We have already seen concern for purity shown in the Dog-Cooker's admonition that Visvâmitra not eat dog meat, in Indra's horror at seeing the Dog-Cooker Trisariku elevated to heaven in his polluting body, and in the language of certain of the curses and countercurses of the Visvâmitra cycle. Kings, outcastes, dogs, untouchables, and the question of purity are brought together in a particularly surrealist account in Kalhana's twelfth-century Râjataramginî, the Chronicle of the Kings of Kashmir. In the fifth book of this, the only surviving "historical" chronicle of India, we find a romanesque account of the reign and fall of King Cakravarman, who ruled in A.D. 936-37. After winning his throne

108

FIVE

through a series of alliances and a decisive battle, Cakravarman becomes debauched by a pair of SvapacI or Dombl 84 dancing girls. They quickjv render the king their sexual plaything, while they and their family take o\ Cr the royal court. Svapaca ministers will only appoint to royal office those nobles who will eat the leavings (ucchistha) from the SvapacI courtesans' plates. These same ministers delight in wearing the Svapacls' mensesstained undergarments in court. Instead of noble Sanskrit, they speak their own base dialect; they enter and so defile temples reserved for the higher castes. Cakravarman commits a series of outrages of his own, in addition to shacking up with outcaste strumpets. He violates a fasting brahmin woman, builds a monastery for a tantric sect, and treacherously slays certain of his allies. His betrayed allies turn against him: one night, while Cakravarman is relieving himself in the Svapacl's toilet, assassins set upon him. They stab him to death in his lover's impure arms, and crush his knees with stones. The story ends thus: "In the night, that king who had been consumed by Svapâkas was killed by those murderers like a dog in [its own] shit."85 Cakravarman, who through his sexual relations with domineering outcaste women has himself become stained with their sin, is in the end reduced to the state of a dog. This is reminiscent, but on a much stronger register, of the crimes of Trisanku and Hariscandra, who are reduced to the station of Svapacas for their sins. There is, however, an emphasis in the Cakravarman account that, while present in other accounts of Svapacas, is not so strongly stressed in them: this is the important notion of purity and bodily contact. It should be noted that all of Cakravarman's and the Svapacls' abominations involve body fluids. In their union with the king, it is the mingling of sexual fluids that runs counter to nature; in the court, it is the saliva on the leavings of their food that violates the purity of the nobles forced to eat such horrors. The menstrual blood worn into court by the Svapaca ministers is so powerful and polluting as to stain the entire kingdom. It is the same power inherent to a Dombl's uterine blood that was so highly prized by the Tantrics, who were also recipients of Cakravarman's misguided largesse. We find similar language in the myth of Gâdhi-Katanja-Gavala: when it is discovered that an outcaste has been ruling the kingdom for eight years, it as if "a corpse had been staying in [the people's] homes." The only solution is dissolution: were they to go on living, they too would be outcastes, condemned to several base rebirths. Again, Hariscandra, by doing the work of an outcaste as he disposes of the dead in

Dog-Cookers and Other Cases

109

genares (where the present-day cremation monopoly is run by a Dom-Râj), himselfbecomes a Svapaca, as does Trisanku in the forest. In the Brhat Samhitffs treatment of omens, it is the contact of an outcaste dog's urine with the body of a brahmin cow that portends the most evil of fates. In the same vein, the post-Upanisadic cliche, "as disgusting as a sacrificial vessel that has been licked by a dog," reflects the danger of mixing polar opposites on the level of bodily excretions: here, the cow's milk or clarified butter would be brought into contact with canine saliva. In yogic traditions, the mechanism of skin contact as source of pollution is emphasized as well: the thousands of pores that cover the body must be purged regularly through yogic purification techniques.86 Ultimately, and on a more immediate level than the creation myths of the outcaste races we reviewed above, it is such notions and categories of purity that highlight and define the polarities of high-caste brahmins versus casteless Dog-Cookers in Hindu society.87 So it is that we find the legal texts spelling out the ideology of purity categories, as well as of the theoretical inviolability of the brahmin's body vis-â-vis other humans, in very clear terms. Thus we read in Manu (5.85): "When he has had contact with a Candâla, a menstruating woman, an outcaste, a woman in childbed, a corpse, or one who has touched a corpse, [a brahmin] becomes pure by bathing."88 Water, the universal solvent (except in extreme cases, where dissolution through fire proves necessary), cleanses away the stain inherent to the body, if not the ontological being, of the outcaste who has come into contact with the person of a twice-born. There are at least two features inherent to this concept of bodily purity and pollution. As far as the outcaste is concerned, polluted existence is a vicious circle indeed. Because of a dangerous and sinful mixing of sexual fluids and thus of social categories on the part of a parent or ancestor, he is condemned to birth in a body that is, a priori, polluted and polluting. As if this were not enough, his caste occupations condemn him to dwell in sin as well, working as he does killing creatures, flaying and dressing them out, or disposing of their mortal remains and waste products. But, of course, he is bound to engage in such occupations because he is the product of sin in the first place. His prohibition from social intercourse with the twice-born is a result of a violation of the same order by one of his forebears. One can understand the modern term "untouchable" in this light, even if its historical pedigree is a relatively short one. Yet, the manipulation of polluting substances is the outcaste's svadhanna, his place in the moral and social order from which he is de facto excluded.

110

FIVE

It is the brahmin who is most vulnerable to the Dog-Cooker's pollution in the social intercourse that he might have with the latter in the execution of their respective social duties. Manu (5.15-58) waffles on whether or not brahmins could eat meat, an indication that changes in this and other purity codes were in progress at the beginning of the common era. The root of this change lies mainly in the fact that even while they rejected the renunciant ideology as antisocial, the brahmins were in the process of co-opting many of its elements as means to maintaining a social and ethical distance from which to exert their authority.89 This was particularly the case with purity codes, and the concomitant result of the exclusion of polluting DogCookers from the lower end of the social pyramid was the self-exclusion of the pure, milk-drinking brahmin from the upper end. 90 This polarization is described in Manu (3.239-41) in highly legalistic and mechanistic terms: "A Candâla, a village pig, a cock, a dog, a menstruating woman, and a eunuch must not look at brahmins while they eat. Whatever any of these sees at a burnt oblation, at the giving of a gift, at a dinner given to brahmins, or at any rite in honor of the gods or ancestors, will not produce the intended result. A boar makes the rite useless by inhaling the smell of the offerings, a cock by the air of its wings, a dog by casting his eye on them, a low-caste man by touching them." This same polarity is cast in narrative form in the Puranic myth of an evil king named Vena (whose name is also that of an outcaste people), out of whose dead body the rsis churn out both the good and evil internal to him. From his left thigh arises a black Nişâda outcaste; from the right, the pure and virtuous King Prthu. 91 In one variant of this myth, a living Vena is made pure when the impure Nişâda is churned out of him. This cleansing away of the outcaste in him is his redemption. In the later literature of the Purânas in particular, we find that the spirit of the devotional religion of love (bhakti) offers easy purificatory solutions not only to evil kings, but also to the lowest of the low, the Dog-Cooker. Some texts go so far as to say that in the Kali Yuga, a Dog-Cooker who is devoted to a given god merits liberation more than does the brahmin nondevotee. It is in this context that we find parallels to the happy endings of the myths of Sunahsepa, Trisariku, and Hariscandra, endings rendered all the more baroque by the sectarian biases that pervade the Puranic accounts. So, in the Vâmana Samhitâ version of the Vena myth, that evil king comes to be reborn and, thirsting for liberation, lives in the precincts of the Saivite Sthânu tirtha, on the Sârasvatı River, whose waters are nevertheless prohibited to him. By chance, a dog who has swum

Dog-Cookers and Other Cases

111

there sprays Vena with the tvrthcCs waters when he shakes himself off. Through the agency of the dog and especially the purifying waters that dis-

solve all stain, Vena is cleansed. Siva then offers him a boon, and Vena requests that the dog be allowed to accede to Sivaloka. This request is granted, and Vena is himself later wholly purified when, reborn as Andhaka, he has his sin burned away by Siva, who burns him up. 92 In fact, all of the epic and Puranic myths involving the redemption, elevation, or salvation of a dog or Svapaca are set either at a tirtha or around a Unga. In some, there recurs the antibrahmanic or antinomian sentiment that, first evinced in the Visvâmitra cycle, takes its most extreme form in the Bhairava myths. So, in the Râmâyana, a dog pleads to Râma that he has been unjusdy struck on the head by a brahmin who had ordered him to yield to him on a street. When Râma argues that a brahmin is legally exempt from punishment, the dog replies that it had been his wish to pardon the brahmin and request a boon for him in any case. The boon, that the brahmin be made head of a certain Kâlânjana monastery, is granted, after which the dog cryptically predicts that the brahmin will, in the fullness of time, go on to hell by abusing his position. "The lustrous dog then returned to the place from whence it had come. Having been high-minded in previous births, it had fallen into a polluted state [apadusita] in that existence. It then undertook an expiation in Benares."93 Saivite myths of this sort, especially those found in the Skanda Purâna, generally have a greater air of serendipity to them: A certain king of the outcaste Kirâtas had been a dog in a previous birth. Fortuitously, on the night of Siva's greatest festival, the Mahâsivarâtrî, he had circumambulated a Siva temple while running from some pursuers. Praised be Siva, the dog was killed next to the temple's entrance. By virtue of this happy spatiotemporal coincidence, he is reborn into the Kirâta royal family. Because of his canine background, however, the Kirâta king is evil and uncontrolled in his passions, even in his present birth. He and his queen are instructed that, by worshipping Siva on Mahâsivarâtrî, they might both accede to the station of Siva after seven further rebirths. 94 In the sixth- to tenth-century Vdişrysv'ıtcBhâgavata Purâna, the message is no less radical than that of the Saivites, even if bathing at a tirtha occurs on a symbolic level: "By bathing, through the hearing of Hari's stainless eternal fame, the entire world, beginning with the Svapaca, is immediately purified." "Even a Svapaca becomes fit for a Soma festival" through devotion to god. 95 The crowning reversal of brahmanic purity codes is to be found in the

112

FIVE

five tantric antisacraments, the panca-makâray or five M's. Clearly a heterodox subversion of the panca-gavya, the five pure products of the brahmin's cow, th cpanca-makara consist of the use of flesh, fish, wine, parched grain, and sexual intercourse as means to the supernatural powers (siddhis) and bodily liberation (jivanmukti) so prized by the tantrics. In the practice of this final antisacrament, the outcaste Dombl or Svapaci is most prized as a sexual partner, in part due to the energy latent in her uterine blood. The reversal of pure and polluting—between what a brahmin drinks and what an "outcaste" woman secretes—could not be more explicit here. 96 The Buddhist tradition, contemporaneous with and generally contemptuous of that of Hindu bhakti, ridiculed Hindus as tirthikas—kneejerk pilgrims. It thus chose for its hyperbolic representation of the tirtha ideology the example of "a dog swimming in the Ganges [who nevertheless] is not considered pure." 97 The Buddhism of this period nevertheless did share the antiorthodox spirit of Hindu bhakti. Nowhere is this more apparent than in the eighth- to tenth-century Caryâpada songs: "Outside the city, O Dombl, is your cottage: go there touching only the brahmins and the shaven-headed."98 While a probable reference to the movement of vital energy in the yogin's body, the Dombl here is also an outcaste woman whose touch would be abhorrent to both brahmins and Buddhist monks. It is possible to see a parallel evolution of Buddhist and Hindu attitudes towards outcastes in references to both Candâlas and dogs. In reality, the Buddhist faith, which undoubtedly attracted outcastes into its fold with its original ethical rejection of the system of castes and other brahmanic abuses, came to share the Hindu ideology that excluded the outcaste from society. This was the case at least until the period that saw the emergence of bhakti in the Hindu tradition; then a revision of the old ideology followed that paralleled that of the Hindus. So we find, in some of the later Jâtakas (nos. 487 and 497), the promise of nirvana offered to Candâlas and Pukkasas. In another Jâtaka (no. 377), a Candâla who has become a bodhisattva teacher of the dhamma even kicks his brahmin pupils!99 An apparent parallel to the Hindu valorization of the ascetic life over worldly activity may be found in the Kukkura Jâtaka (The Dog Birth). Here, the Buddha takes birth as a cemetery dog in order to give ethical instruction to his disciple Ananda, who has himself taken birth as the king of Benares. The narrative ends with the king being shown by the dog-bodhisattva that his own royal hounds had perpetrated a crime for which the cemetery dogs had been blamed and put to death. The king becomes a benefactor of dogs. 100 This analogical tendency is carried to its limit in the person of Kukkurarâja, a

Dog-Cookers and Other Cases

113

of the tantric Buddhist tradition to which the Caryâpadas belonged. Kukkurarâja, whose name means "King of the Dogs," lived with a yogini who looked like a bitch. By day, he preached in the guise of a dog to one thousand vims anayojjints; by night, he would go to the cremation ground to observe the esoteric rites. 101 saint

S

I

X

CENTRAL ASIA: THE VORTEX OF CYNANTHROPY

1. DOG-FACED M E N , EPHTHALITE H U N S , AND KINGDOMS OF WOMEN IN INDIAN TRADITIONS

In India, we find a significant number of myths in which dogs stand as metonyms for their "outcaste" masters, who are themselves often called Dog-Cookers or Dog-Milkers. At this point, we ought to venture beyond ideology and typology and attempt to glimpse the ethnological or historiographical grounding of these uses. The Hindus and Buddhists could have chosen some other patendy revolting creature, such as the vulture, rat, or swine, to depict metaphorically their social pariahs. Yet, whenever outcastes are to be censured, reviled, or rejected, it is the canine trope that prevails. Could it be that, as with the Sabaras and Ao Nâgas, the DogCookers of these pious and didactic traditions really did maintain a relationship with their dogs that was something more than economical? When we review the many myths of Dog-Cookers and their dogs in Hindu mythology, it is striking to note how many locate these peoples to the north of India. We find this in the Sunahsepa myth, the earliest origin myth of the outcaste peoples, in which the disobedient sons of Visvâmitra are exiled to or beyond the "northern limit" (udantya). In the Yojjavasistha, Gâdhi rules as Gavala in the northern kingdom of Kira: when Gâdhi goes to learn "who he was," he goes to the "Land of the Huns" in the North. Cakravarman is a king of Kashmir who falls prey to the wiles of lascivious SvapacI or Dombl girls, also from the north. We also find in myths of Dog-Cookers a certain conflation of indigenous outcaste peoples together with foreign tribes or ethnic groups. To name but a few sources, we find such mixing of categories in the Sunahsepa myth, the Amarakosa lists, and the myth of Visvâmitra's defeat at the hands of the armies created from Nandinl's excrements. In a variant of the Vena myth, it is said that, in addition to causing people to marry outside of their castes, by which he gave rise to the outcaste races, Vena himself also fathers a son 114

The Vortex of Cynanthropy

115

named Barbara (foreign Barbarian). From this son are descended the foreign tribes.1 The question is, therefore: Which came first, the northern barbarian Dog-Cooker/Dog Milker, or the indigenous outcaste DogCooker? Do the terms Candâla, Svapaca, and Svapâka all refer to both barbarians and outcastes? Were they all perceived as one and the same people? What importance should we attach, beyond some ideological stamp, to these peoples' association, if not identification, with the dog? It should be recalled that, until the time of the epics and lawbooks, shordy before the beginning of the common era, there is very little evidence to indicate a devaluation of the dog in the Hindu ideology.2 It is precisely in this period that we first find reference to the peoples living north of India in the sacred texts. If a connection is to be drawn, then, between these two phenomena, we must look for it in the shadowy world of Indian historiography. Before we look at these Indian sources, a few remarks of a more general nature are in order. In Indian, as in European and Chinese, ideology the four corners of a square left uncovered by the circumference of an inscribed circle were the haunts of monstrous and barbarian races. The circle in the square was a conceptual model that all three employed. However, when we bring together the specific historiographic material of all three traditions, we find that: (1) the Europeans generally placed their Amazons, Cynocephali, and other monsters far to the east, in "India"; (2) the Indians placed their barbarians, "fallen ksatriyas," Dog-Cookers, and Kingdoms of Women to the north and west; and (3) the Chinese located many of their barbarian Dog-Men (and, as we shall see, Kingdoms of Women) to the south and west. By extrapolation, the point of intersection for all three of these traditions would have been central Asia, what is today Tibet, western China, Kazakhstan, Afghanistan, and Kashmir. Therefore, let us provisionally hypothesize an ethnographic or historical homeland for the Dog-Men of our three traditions in the general region of central Asia. Let us also note here that the Western, Indian, and Chinese traditions quite often agree with one another when they juxtapose their Dog-Men with homelands of the Amazons, Kingdoms of Women. We will show, in this and the following chapters, that this juxtaposition may have a central Asian ethnogenic myth or the practice of polyandry at its origin. We have already seen examples, in previous chapters, of what we called Asianism, the fear and dread, on the part of a Western "center," of that which is foreign. The case is no different in Asia: the wondrous peoples living beyond the northernmost or westernmost mountains provoke not only fascination but also terror. So it is that in India and China, as in the

116

SIX

West, there has been a general blurring of distinctions, not only between various barbarian invaders, but also between barbarian invaders and their more peaceable neighbors. Just as in Europe, where the tendency was to group Scythians, Huns, Parthians, and Muslims together under the heading of "Turks," "barbarians," or "incorrigible races," so too, the Indians often conflated the names and cultural uniqueness of their many northern and western barbarians, while the Chinese did the same for their barbarians of the north, west and south. Out of this confusion, there emerge two salient points about myths of the Dog-Man. The first of these is that a number of central Asian peoplesincluding the Ephthalite Huns and certain Turko-Mongol and Tibetan peoples—were polyandrous, with women playing a more prominent role in society than in the male-dominated traditions of the West, India, and China. The second involves the ethnogenic myths of many of these same peoples, who often identified themselves as the descendants of a primal union between a male dog or wolf and a female human. When these ethnological data were translated into the Western, Indian, or Chinese sources, a certain transformation occurred. On the one hand, the relative dominance of women within these central Asian societies gave rise to the perception, from afar, that they were gynecocracies. They were thus duly termed "Kingdoms of Women." On the other hand, these women's spouses, the men of these societies, were perceived both as cruel and lawless barbarians—because these peoples did in fact invade "civilization" at times—and as Dog-Men, even Cynocephali, by virtue of their indigenous dog ancestry myths. Central Asian peoples not possessed of such an ethnogenic myth may nevertheless have been thrown by their Indian, Chinese, and European interpreters together with their "semi-canine" neighbors. And, most importandy, their ethnogenic myths became skewed in the retelling, with the human woman and canine male of the original accounts being transformed, in the Western, Indian, and Chinese material, into Amazon women and Dog-Men. To be sure, we find ourselves incapable of judging with any degree of certainty the point at which ethnography ends and ideology and propaganda begin in these three great traditions' depictions of their barbarian DogMen/Amazon races. The reason for this is simple: with a few notable exceptions (the Mongols, Tocharians, and Uighurs), very few of the central Asian peoples who concern us here had a written language with which to document their own beliefs and practices. And even when they did have their own literary traditions, they were not heeded by the literati of the

The Vortex of Cynanthropy u

117

great traditions." For all intents and purposes, they themselves and the large have been held in thrall by Western, Indian, and Chinese interpreters of their culture and society, a state of affairs that is slowly being reversed in the present century through the efforts of scholars of central Asian studies. While we can be certain that these peoples did not locate themselves at the corners of their own maps of the universe,3 we remain in the dark concerning many of the details of their ancient worldviews. In the documents of the great traditions, they are ultimately dismissed as subhuman floods that briefly washed over the edifice of civilization before being definitely driven back by the armies of the same civilization. T h e y are a negativity, a blank space on the fringes of the conceptual map of each of these traditions' self-centered universes. Here, for the time period that concerns us (3d century B.C. to 6th century A.D.), India's great North passed through the hands of no less than five different peoples or empires. These were the Parthians (Sanskrit: Parâthas), Bactrian Greeks (Yâvanas), Indo-Scythians (Sakas), Tocharians (Kushanas, Tukharas), and Ephthalite Huns (Hünas).4 Of the five peoples in this series, all of whom conquered and ruled portions of northern and western India, the last two (the Kushanas5 and Hünas) may well have had indigenous myths of canine ancestry, myths they would have shared in common with a great number of non-Indo-European central Asian peoples.6 The Parâthas, Yâvanas, and Sakas were generally listed together with the Kushanas and Hünas and many other "fallen ksatriya" peoples of the Indian great North, with little discrimination made between their mores or practices. Generally, all were broadly identified with the outcaste or barbarian Mlecchas, Candâlas, or Svapacas, and with a group called the "five hordes," the panca-gana.7 Apart from the canine ancestry myth shared by many of these peoples, we also find in central Asia a living popular tradition of cynocephaly. First, there is in Tibet and the Chinese province of Tsinghai a rich indigenous tradition of dog-headed demons that may be magically created and manipulated to increase one's own wealth and to bring tribulations down upon one's enemies.8 This modern ethnology is complemented by archaeological evidence. In the 1920s, a French team of archaeologists unearthed, at Hadda, in eastern Afghanistan, a veritable treasure trove of sculpted figures. The site had been a Buddhist monastery located on the Silk Road in the first centuries of the common era, in the centuries prior to and contemporary with the Ephthalite Hun conquests of central Asia. The artifacts brought to light NV orld at

118

SIX

there attest to a truly cosmopolitan milieu: Greek-style Zeus figures jostle with Chinese Buddhas. But most amazing are a collection of demons' heads, in stucco, unearthed at Hadda and now on display at the Muscc Guimet in Paris. According to the French scholar Rene Grousset, these heads are stylized representations of various central Asian barbarians who are here, in a sculpted rendition of the Buddha's enlightenment, soldiers in the army of Mara. Some twenty heads represent demons of every stripe, with twisted physiognomies, heads with smaller heads in their mouths, and other deformities. Leading this ghoulish parade is a piece apdy identified as a "dog-headed demon." Perhaps here in Hadda is the archaeological missing link to central Asian traditions of cynocephaly and cynanthropy (pi. 12)» Tibet and Afghanistan are precisely the two regions in which medieval Indian sources located their Dog-Men and Kingdoms of Women, both of which they seem to identify, as we will show, with the Hünas, the Ephthalite Huns. These two "Indian" locations are, moreover, seconded by both Western and Chinese sources. Varâhamihira is the most systematic Indian source on this subject: in Brhat Samhitâ 14.21-27, he locates the Kingdom of Women (Stnrâjya) to the west, the Parâthas and Sakas to the northwest, and the Hünas and Svamukhas to the north of India. This last people, whose name means the "Dog-Faces" or "Dog-Heads," is one of the rare Indian references to a race of cynocephalic men. 10 Another instance of the same identification may be elicited from the Kubjikâmatatantra, which assigns Svamukhas to the Uttarâpatha, the "Northern Path."11 It is intriguing to find in the Indian material—as we have in the European context—a textual (if not geographical) juxtaposition of DogHeaded Men with both a Kingdom of Women (called the Amazons, in the West) and with the Parthians, Indo-Scythians, and Huns. In the West, this juxtaposition was reinforced by a tendency to conflate the latter three groups together with whoever was invading from the north or east at a given time. Furthermore, the five were all brought together in the lists of the inclusi that continued the traditions of Pseudo-Callisthenes and Pseudo-Methodius. In these sources, these peoples are walled outside of the oikoumene because of their cannibalism, abominable eating habits, and unspeakable impurity. This last element has a foreign ring to it in the Westem texts; yet, it fits very well with the Indian ideology of the outcaste or foreigner as impure pariah. With the Brhat Samhitâ, might we be in the presence of a parallel Indian source to this perennial motif of the Alexander legend? While the emphasis on impurity may very well be Indian, its "source" ought to be sought further to the north and east.

The Vortex of Cynanthropy

119

In the Indian context, we find more material concerning Kingdoms 0 f Women than we do explicit references to the Svamukhas, their DogHeaded neighbors. The latter may nonetheless be glimpsed, in the background, in Indian accounts of both northern and western Kingdoms of Women, as well as of such barbarian peoples as the Hünas. In his chronicle of the kings of Kashmir, the Rajataramgini, Kalhana makes a number of references to the women warriors of the northern kingdom of Strîrâjya. While he offers little more than a glimpse of their breasts, his emphasis on this element to the exclusion of all others, may also betray a familiarity with a more universal Amazon (One-Breasted) tradition. Kalhana also assigns a northern provenance to the two powerful women who caused the downfall of the Kashmiri king Cakravarman. These two girls are referred to as both Svapacls and Dombls, the name of another outcaste group. In fact, Doms (of which Dombl is the feminine form) entered into India from the north, during or shordy before the Gupta age. Elsewhere, Kalhana mentions the Strîrâjya in a list of foreign or barbarian peoples that more or less repeats earlier Mahâbhârata traditions. In one of its many barbarian lists, this epic source locates the following peoples outside (bahavah) of and to the west of India: the Yâvanas, Sakas, Harahünas ("Ruddy" or White Huns), Cinas (Chinese), Tukharas (Tocharians), Saindhavas (an Indus Valley kingdom), Jâgudas, Ramathas, Mundas, Strirâjyas (Kingdom of Women), Tariganas, and Simhalas (in this case, not the Singhalese). In Mahöbhâmta 7.95.38, the Kambojas (Gandharans), Sabaras (the outcaste people), Kirâtas, and Barbaras are added to this list. Some of these peoples—the Yâvanas, Kambojas (Gandharans), and Sakas—are among those referred to in the Harivamsa (10.21-45) as the "five hordes" who, along with the Haiheyas and Talajarighas, were the enemies of the Ikşvâku king Sagara. We cannot totally fault the Indian sources for the confusion in these lists: the names of the central Asian peoples represented in them were fluid, since theirs were constandy shifting confederations, alliances, and ethnic mixes, occurring in regions quite distant from the borders of India proper. So, for example, the Kushans were a federation of five tribes called the "Great Yüeh-chi." However, Chinese sources state that the Ephthalite Huns are descended from the Yüehchi, a group that also interacted or federated with other groups to form the Juan-juans (the Avars of European traditions) and T u chüeh (the Turks, called Turuşkas in Indian sources). Another account of a foreign people whose women dominated the social sphere is found in the Mahâbhârata description of the land of the Bâhlikas

120

SIX

(or Bâhikas).15 In this account, the Kaurava general Karna scorns the Bâhlikas by way of insulting Salya, a fellow officer and the king of the Madras (the Medians of Persia). Salya's sister is Mâdri, who is also called Bâhlikl. Bâhlika is Balkh, the capital of Bactria, or the name for the entire region (also called Gandhara), in modern Afghanistan. This account refers to the capital city of the Bâhlikas, however, as Sâkala (Mahâbhârata 8.30.14), the modern Sialkot, located on the Jhelum River in the Punjab region of modern day Pakistan.17 Two Bâhlika clan names are introduced in this passage: these are the Ârattas (8.30.36) and the Jartikas (8.30.14). This latter name refers to the Indian Jâts, a people said to be descended from the Hünas, the Ephthalite Huns in some sources—yet said to have defeated those same Hünas in another! 18 Both the Bâhlikas, as we shall see, and the Hünas bore a special relationship to their women. More than this, both are associated with the two cities of Balkh and Sâkala: these were Hüna capitals on the eve of the Hüna conquest of India in the late fifth century. Yet, the two are mentioned as separate peoples in the Brhat Samhitâ (5.80,10.7, 32.15).19 In his tirade against Salya, Karna spells out the abominations of his subjects: they are shunned (bahiskrta)20 by the Ganges, the five rivers, and the Himalayas; they eat beef and garlic and drink liquor; they are the product of the intermarriage of südra men with women of higher castes; and they eat from wooden and earthen vessels that have been licked by dogs. The greater part of Kama's scorn is reserved for the Bâhlika women, however. These, intoxicated and randy with drink, wearing neither clothing, garlands, nor cosmetics, dance wildly in the streets singing drunken ditties that resemble the braying of asses or camels: "Hâ hate hâ hate, slain my husband, slain my lord!"21 They are totally without restraint in sex and in every other pursuit. If the Mlecchas are the scum of mankind, then the Mauştikas are the scum of the Mlecchas, the Sandas the scum of the Mauştikas, and the Mâdrakas the scum of the Sandas: as for the Mâdraka women, they are the scum of womenkind. Then, to clinch his argument, Karna tells of a curse pronounced by a brahmin girl who had been violated by Bâhlika men. Because of this curse, Bâhlika women are unchaste, and it is their sisters' daughters and not their sons who are the heirs in Bâhlika families.22 There are two complementary points to be retained from this passage. First is the ad hominem attack made on Salya, which is in fact an attack ad mulieres, more on the subject of the women of this region than on Salya himself.23 Second is the unusual trait of matrilineal inheritance, coupled with the place names given in the narrative, which would indicate that the

The Vortex of Cynanthropy

121

Bâhlika or Mâdraka women were the (unchaste) spouses of the Ephthalite Huns.24 These are the dread Hünas of the Rajataramgini and other Indian sources, and the peoples whose lands were the Hüna-mandala of the Gâdhi mvth. That they were dreaded and despised enough to be made out as cuckolds in Indian traditions is a possibility. But that they were the unique matrilineal society of central Asia is supported by Chinese sources: "They [the Ephthalites] also have a custom by which elder and younger brother marry one wife. If one has no elder or younger brother, his wife wears a onehorned hat. If one has brothers, horns are added to the hat, according to their number."25 It is perhaps this social structure, combined with indigenous myths of dog ancestry as well as the behavior of what would appear to have been a very liberated group of women, that gave rise to the tradition of a Kingdom of Women in the same region as the lands of the barbarian Hünas and the Dog-Men. 26 Ultimately, the epic compilers may have split the Ephthalite Huns into male dogs and female Amazons, in accordance with their practices of polyandry and their possession of a dog ancestry myth, both of which would have been vaguely reported to the people of India. Such facts may also account for the theme of Scythians or Cynocephali living together with Amazons that we find in Western traditions, as well as a juxtaposition, in Chinese sources, of Su-p'i "Huns," a Kingdom of Women, and a Kingdom of Dogs. It is also possible that a tradition of independent but dissolute women living with terrible and despised men represents a fusion of two facets of India's reaction towards its "empirical others." On the one hand, the unknown is exotic and fascinating, but on the other it is fearsome and repugnant. These are the attitudes we called Asianism and Atticism in an earlier chapter: here they might be referred to as Hyperboreanism and Ephthalism. We have treated Ephthalism in the preceding pages; Hyperboreanism, which depicts the north as a blessed abode of redemption and pleasure, is a perspective also found in Indian sources, sources that are moreover supported by Chinese materials. This is the tradition of the Land of the Uttarakurus (Northern Kurus), the land lying to the north of the Himalayas and of Mount Meru, the central pillar of the universe. Indian evidence concerning a northern paradise of lithe and independent women—material that complements Kalhana's references to the northern Strîrâjya—is found at the end of the Mahâbhârata passage on the western Bahlika women. Following Kama's tirade, an unnamed warrior in Salya's army retorts by lamenting the fate that has carried him so far away from his beloved "jungles of Kuru"—identified, elsewhere in the epic, with the

122

SIX

Uttarakurus. There, he says, to the north of the six rivers, his beloved women await him. They are fair-skinned, tall and beautiful, with vulvas like great conch shells, dark flashing eyes ornamented with collyrium, women who wear only blankets and furs, and who play drums and sing in voices that, it must be allowed, resemble the braying of asses and camels. It is held, moreover, that the women of Uttarakuru are the sole remaining practitioners of the "ancient law" by which women, unbound by the constraints of marriage, take their pleasure with the men of their choice, as do "animal creatures . . . today."27 Yet another Indian portrayal of independent, nymphlike "girls from the north country" is found in the Buddhist Petavatthu. Here, female ghosts (petfs) live in isolation, luring humans to their distant lands by floating mangoes down their river. A prince who mounts the river to its headwaters and the petis' haunt falls in love with one of them, but finds that she is flayed alive every night—by a dog! 28 In the Kedâr Kalpa, an allegorical ascent of the Himalayas and of successive stages to yogic liberation, two heroic travelers pass through the mountain kingdom of the queen of Campika. Here they are offered, for their pleasure, 100,000 women and a 10,000-year life span. These they eschew, knowing that immortality awaits them on a higher plane.29 The Land of Uttarakuru is portrayed in thcRömöyana (4.43) as a land on the northern ocean that is so far to the north that the sun and moon shed no light on it. It is undoubtedly this land that inspired Ptolemy's Ottarakorra (which he located in China), Pliny's Attacori (whom he placed near the Tocharians), and Megasthenes's Hyperboreans. There, according to both Indian and Greek sources, people lived for a thousand years in a land strewn with gold and precious stones, a land where the women were particularly affectionate. In a Bengali recension of th cRâmâyana, however, the paradise of Uttarakuru is located "beyond the Sailoda River, in a place where ktchaka reeds grow." With this, we find what appears to be a direct correspondence between Indian and Chinese material on a northern Kingdom of Women. A number of Tang period (7th-9th century A.D.) sources from China describe the land called Su-pa-na (a transliteration of the Sanskrit Sauvarna, "Golden," or Suvarnagotra, "Golden Race")—located in the region of Khotan, to the north of Kashmir, the west of Tibet, and the south of a Kingdom of Dogs—which they identify as a Kingdom of Women (Nü kuo); they moreover conflate this Kingdom of Women with the land of the

The Vortex of Cynanthropy

123

Su-p'i "Huns." The Chinese sources further locate this Kingdom of Women on the Jo-shui or "Weak River," which scholars have identified with Ctesias's "Silas" and Megasthenes's "Side"—Greek transliterations of the 30 Sailoda of Indian traditions—and with the modern Khotan River. Thus, it is possible to extrapolate from the Indian and Chinese material a land to the north of India and the west ofTibet where Dog-Men, Amazons, and "Hun" barbarians lived side by side, when they were not simply identified with one another. This supplements Indian and Western traditions concerning a land to the northwest of India and far to the east of Europe, where a similar juxtaposition occurred. However, as we have said, the Great North of India and the Mysterious East of Europe often correspond to the Far West of China. We therefore turn to Chinese accounts of its western and northern barbarians. 2. CHINESE HISTORIOGRAPHY OF CENTRAL ASIAN PEOPLES

In China, we find ourselves in the presence of not one but two apparendy indigenous traditions of cynanthropy, traditions which the Chinese, moreover, conflate in their accounts of their barbarian neighbors. The peoples with whom these dog ancestry myths are identified vary widely in the Chinese sources, which, in addition to locating Kingdoms of Dogs, or of DogMen, beyond all four limits of its imperial lands, also locate Kingdoms of Women—sometimes juxtaposed with or in proximity to its Kingdoms of Dogs—in no less than three and as many as seven different liminal locations. Also unique in the Chinese case is the fact that there are indigenous accounts that 'Verify" the Chinese material, in which the ethnic peoples themselves claim descent from the union of a human and a dog or wolf. At first blush, then, the Chinese material seems to have some basis in solid ethnography; however, the possibility of a Chinese "pizza effect" must not be wholly discounted.31 We will return to the ethnographic features of Chinese accounts of cynanthropic barbarians shordy, but before doing so, we must first oudine, in a preliminary fashion, what happens when one tradition interprets another tradition that has no historic or literary voice of its own. Of the two DogMan races of Chinese tradition, one was never able to tell its own story, and the second has only recendy done so through the writings of modern ethnographers. The first of these traditions is that of the Hsiung-nu,32 China's northern and western barbarians; the second is that of the Yao, Liao, or Man, China's southern barbarians.

124

SIX

In both cases, at least down to the past century, nearly all that was known of these peoples was refracted through the writings of the Chinese annalists or historians. As we will see, these Chinese commentators, in their writings, had certain "barbarian genres" to respect, and their portrayals of these peoples were thus necessarily skewed. On the one hand, these barbarians were considered to be ethnic or racial inferiors by the Chinese, living in a perpetual state of exile; on the other, they were incorporated—even if their inclusion amounted to ideological exclusion—into Chinese cosmology, history, and cosmography. The result of this double deformation of these peoples' identity was the mythologization of their history, society, and culture. In this, the Chinese treatment of their others was similar to the two other great traditions we have already reviewed. In all three cases, a number of people, conflated into a single, generally undifferentiated group, were depicted as semihuman monsters with subhuman behavior and social practices. These foreign or barbarian others were generally defined by their exclusion from or their opposition to the categories of the civilized sphere of the great traditions. At worst, they were the bestial enemies or prey of the traditions in question; at best they were their slaves, servants, vassals, or social inferiors. The imperial attitude towards barbarians is made most clear in the opening pages of the Man shu, an A.D. 864 source on the Man barbarians of Yunnan and the oldest Chinese geographical record. While its author, Fan Ch'o, presents a quite detailed account of the tribes and subgroups of the barbarians of the province he administers, the general note of disdain is most apparent: 33 If, then, we do not invade and attack them [the Southern Man] from all four sides, they are violent and bad persons, and difficult to reform. That is why I record their cities and garrison-towns, their river-valleys and plains—mere dust defiling the audience-screen of the Imperial apartments. Perhaps one might hope to wipe out their host of ant-swarms, and purge forever [these] . . . barbarian rebels. Parallels between the three traditions are particularly tantalizing with regard to accounts of cynanthropy. As we have intimated, there is a certain intersection in these three traditions of the geographic localizations of the Dog-Man races: the vortex of all three would be the vast region of central Asia or Turkestan.34 This region has historically been the home of the central Asian nomads, for the most part the Altaic-speaking Turko-Mongols. While the area has been inhabited by dozens of distinct peoples, these have

The Vortex of Cynanthropy

125

enerally been grouped under various all-purpose names that have changed over the ages; Scythian, Parthian, Hun, Turk, and Mongol are the terms most commonly employed by all three great traditions. We may infer, then, that it has been the central Asian nomads who have been the prime candidates, among the Europeans, Indians, and Chinese, for identification with their Dog-Men. We have also noted that in all three interpretive traditions, these "cynanthropized" central Asian peoples have quite often been juxtaposed with an Amazonian race of women. We may suggest another parallel between these three traditions, this time a philological one; it may be that the terms for "dog" or "hound" in the three languages are linguistic cognates. This is certainly the case with European and Indian terms, as both share a common Indo-European vocabulary, leaving no question that the Sanskrit svan is direcdy related to the Greek kuön, the Latin canis, and so on. But what of the Chinese term, ch'üan? The word is structurally very similar to the Indo-European terms. Cultural diffusion on a linguistic level, either between Chinese and the Indo-European languages or from a central Asian source into both Chinese and Indo-European vocabularies (it should be recalled that the dog originated from central Asia) ought not to be rejected out of hand. 35 This philological parallelism is supported by similarities in the cultural uses made of dogs among both the archaic Chinese and their barbarians. In both cases, the dog bore strong associations with death. The Shang emperors may have been buried together with their watchdogs; dog figurines were buried with the Han emperors, probably to protect them, as watchdogs, after death. The absence of such figurines between Shang and Han times may be attributed to an interim use of straw dogs buried in place of the animals themselves; straw dogs are still used among China's northern ethnic populations today. In both the ancient Chinese and the more recent barbarian cases, the dog serves as psychopomp through the underworld. 36 3. TOWARDS A DERIVATIVE HISTORY OF THE AHISTORICAL PEOPLES OF CENTRAL A S LA

The time has come to trace, as well as possible, the history of the central Asian peoples whose erstwhile fate it was to become the Dog-Men of the three traditions we have been examining. In addition to the Chinese historical annals, which are quite detailed and objective,37 we can also draw upon Chinese epigraphy, archaeology, and philology as well as on Western and Indian historiography. In the earliest Chinese references to the barbarians they would later call

126

SIX

the Hsiung-nu, this people is called Hsien-yun or Hun-yu and made one of three subgroups of a northern barbarian people known as the Jung. The Jung were apparendy a perennial threat to the Chinese from the very beginnings of their imperial history. As we will detail in the next chapter, the earliest (mythic) conflict between the Jung and the Chinese, recorded in a fourth-century A.D. source, is set in the time of the Hsia emperor Kao-hsin (traditionally dated to the 24th century B.C.). It is doubtful that either Kao-hsin or the war that he fought in this account were anything but legendary. However, it is noteworthy that Kao-hsin's adversaries in this myth are designated as the Ch'üan Jung, or the "Dog Jung." Following the Hsia, the Shang (whose traditional dates are 1783-1122 B.C.), too, are said to have been harassed by the Jung: while references to the Jung of this period do not explicidy qualify them as dogs, there is much philological evidence for such an implicit assumption on the part of the Chinese. All Chinese barbarians had animal radicals of one kind or another in their ideograms, since the barbarians were, in the eyes of the empire, subhuman.38 Following the Shang era, the Dog Jung were vassals to the Chou dynasty (1122-255 B.C.). That they were not always loyal tributaries to their imperial lords is made clear in the Shih chi, the "Historical Memoirs" of Ssu-ma Ch'ien. According to this source, the Dog Jung, on several occasions, alone or in alliance with pretenders to the imperial throne, attacked Chou. After one such attack, in which the Dog Jungs had been enlisted in the cause of a certain Duke of Shen, Chou, after reasserting its power, moved its capital east, from Hao to Loyang. This shift, in 770 B.C., marks the division of this dynasty into its western and eastern periods: the Chinese capital would remain in Loyang down to the early fourth century A.D., when northern barbarians would once more force the empire into retreat. Shih Huang-ti, the founder of the short-lived but powerful Ch'in dynasty (225-206 B.C.), was a native of the feudal state of Ch'i, a state with a high proportion of Dog Jung in its population.39 Following Ch'in, the Dog Jung seem to disappear from the imperial frontiers; however, their disappearance more or less coincided with the rise of the Hsiung-nu (Ferocious Slaves).40 The Great Wall of China was erected, in the third century B.C., with the expressed purpose of walling the Hsiung-nu out and stopping their invasions from the north. In fact, the Chinese themselves had at least one "Pseudo-Methodian" tradition that prophesied a reversal of Hsiung-nu encroachments on the Middle Kingdom: so, in the Hou Han shu (119), we read, "The oracle used to say that in

The Vortex of Cynanthropy

127

the ninth generation of the Hsiung-nu the northern barbarians would be driven back over an extent of 1,000 li. Surely this must be it?"41 The four centuries that followed saw the rise and fall of great empires on both sides of the Wall. To the two Han dynasties of China (206 B.C.—A.D. 220), there corresponded the two empires of the Hsiung-nu (209 B.c.A.D. 246). Throughout this period and in spite of their perennial rivalry, these two great eastern powers came to respect one another, to influence each other culturally and politically, and to cooperate with one another to the point of exchanging their princesses and intermarrying their princely families. Towards the beginning of the common era, the Hsiung-nu split into northern and southern kingdoms, after which time the Han-Hsiung-nu stalemate was broken in favor of Han. A last Hsiung-nu hurrah came in the early fourth century A.D., when a short-lived southern Hsiung-nu dynasty—which called itself the northern Han dynasty—occupied Loyang (to which the Chinese had been pushed by the Jung in Chou times) and pushed the Chinese further to the south. It was at this time that the imperial capital was shifted to present day Nanking, on the southern shore of the Yangtze River, in A.D. 318. This southward dislocation of the Chinese also pushed the Man or Yao peoples south and east from their original homelands in Hunan and Chekiang into Fukien, Annam, and Yunnan.42 Throughout the many centuries of their imperial rivalry, the strategic movements of Han and Hsiung-nu set up repercussions that rippled to the west and south from northern Turkestan (Kashgaria, Tibet, the Tarim Basin) into the expatriate Greek kingdoms of Bactria, Sogdia, and Farghana; the Indian regions of Gandhara, Kashmir, the Punjab, and Sind; and, eventually, the eastern and western Roman Empire, Gaul, Spain, and North Africa. We will limit ourselves here to describing the movements of peoples, provoked by Sino-Hsiung-nu interactions, into India and Europe. The earliest of these was the northern Hsiung-nu attack, in about 165 B.C., on the Yüeh-chi, a people living to the west of them. This pushed the Yüeh-chi south and west; when they crossed over the Oxus River into Bactria, they conquered the Tocharians living there. This in turn dislodged a people called the Sacae (the Sakas), who pushed southward into the Sind and on into central India. In the meantime, the Yüeh-chi split into five principalities (jabgus). One of these, thcjabgu of Kuei-shang (so-called because he was from Gandhara, modern Kandahar) conquered his four f e l l o w ^ -

128

SIX

gus and founded the dynasty that would be known in northern India as the Kushan (Kuei-shang) dynasty. The changing central Asian scene translated, in India, into successive waves of invasion. First the Sakas rushed in from the northwest, to be followed by the Bactrian Greeks (Yâvanas), and then the Kushans: these three foreign groups (the peoples called Mlecchas in the epics and Purânas) ruled parts of India from about 200 B.C. to A.D. 220. Even after these empires fell to the Guptas, some of them maintained small kingdoms in northern and central India for many centuries.43 In the fifth century, the Kushans, driven out of northern India, reconquered the five Yüeh-chi tribes north of Gandhara. During this time, a Yüeh-chi people known as the Hua rose to prominence in Tocharistan, establishing their capital at Badakhshan (near present-day Faizabad). These came to be known throughout the Eastern and Western world as the (H)ephthalites, or White Huns. Their push into India, and later into western Asia, was apparendy precipitated by pressure exerted from the north by the Avars, a people to whom they had been vassals of sorts. Apart from their Ephthalite vassals, the Avars were also lords of the T u chüeh, the Turks, who would overthrow them in S52.44 The Ephthalites themselves were either a branch of the Yüeh-chi, or a people originating from Tu'fan (Tibet), the same region the Dog Jung were said to inhabit from the Han period onwards. They first moved through Kashgaria and Sogdia, and then occupied Bactria, establishing their capital near modern Herat, in western Afghanistan. In this way, the Ephthalites became neighbors of the Sassanid Empire (which had brought down the Parthian empire in the early third century) of Persia, with whom they were constandy at odds from about A.D. 430 to 510. Towards the end of this period however, Ephthalite attention was turned more and more towards India, to the south. Their arrival on the Indian scene, in about A.D. 470, sounded the end of the classic age of the Guptas. Ruling from their capital in Sakala (Sialkot), they tyrannized the subcontinent for some fifty years, until their empire was overthrown by a popular revolt led by Yasodharman of Mâlava. They continued, however, to rule over scattered kingdoms in India well into the eleventh century.45 As we discussed earlier in this chapter, the Ephthalite Huns were unique, among central Asian peoples known to the Indians, in their practice of polyandry, under which a single wife married a group of brothers. The Ephthalites were also unique among "Huns" in that they were likely of Indo-European, rather than Turko-Mongolian stock. They apparendy

The Vortex of Cynanthropy

129

spoke a variant of Tocharian, an Indo-European language, and were called White" Huns for their skin color, according to Western sources.46 The same sorts of pressures as experienced in south Asia, originating with movements of the Hsiung-nu in the east, came to change the face of the Western world in about the same period. It was the ensuing westward migrations that fueled the Western traditions of "northern" barbarian and monstrous races discussed in chapters 2 and 3. In fact, the names of most of these nomadic people who seemingly caromed across Asia like so many bumper cars would come to be enshrined among the inclusi of the Alexander legend, not only in Pseudo-Methodius, but also in many later Western sources.47 In A.D. 89-91, a series of defeats suffered by the northern Hsiung-nu at the hands of their southern brothers and the Chinese forced them into central Asia, where they disappeared. It has long been thought that it was the same northern Hsiung-nu who reappeared as the Huns of European tradition, who conquered their western neighbors, the Alani, in the middle of the fourth century. We will return to this question in due course. The Huns themselves first appeared on the shores of the Don, on the threshold of Europe, in 374. In the hundred years that followed, the pressure of the Huns on the eastern border of the Roman Empire forced both Germanic tribes and other central Asian peoples to move into Roman territories. So it was that, in 410, the Visigoth king Alaric sacked Rome and effectively brought an end to Roman imperial pretensions. At the same time, the displaced Suevi, Vandals, and Alani pushed into Gaul and Spain, and eventually North Africa. For a very brief time, the Huns themselves controlled a vast territory stretching from the Caspian nearly to the North Sea. However, following Attila's untimely death in 453, they were driven back eastward into southern Russia48 and the north shore of the Black Sea, where they were perhaps transformed, in ecclesiastic traditions, into the Man-Eaters encountered by such aposdes as Andrew and Bartholomew.49 It is certain that the Huns (Hunni) were already well on their way, in the fifth century, to being conflated with a host of other barbarians from the e ast, including Pliny's Thuni, Ptolemy's Chunoi, and a number of other peoples previously grouped under the general heading of "Scythians." If nothing else, this brief sketch of the role played by the central Asian Peoples in world history—that is, in the history of the three great traditions of China, India, and Europe—is intended to show just how varied those people were whom our historical sources generally group under one heading. Another important implication is the constant "presence at a dis-

130

SIX

tance" in the historiography, and perhaps the presence tout court, of these barbarian peoples to the conceptual continuity of the great traditions. I n fact, when we examine the chronologies of Europe, India, and China, we find sizeable chunks of history in which barbarians were the outright lords of "civilization." However, the invasion and occupation of any of these civilizations by the Huns per se—the Huns in Europe, the Hünas in India, and the Hsiung-nu in China—were of extremely limited duration, especially in comparison to the polemic subsequendy raised against them. The historical breakdowns of Hun occupation are as follows. Northern China fell under the domination of the southern Hsiung-nu or some Altaicspeaking people for about three hundred years, from the late third century to A.D. 581. Between the fall of Pang in 908 and the formation of the People's Republic in 1949, large parts of China were also ruled by nonChinese for over four hundred years. The Hsiung-nu, however, remained outside of China, in Mongolia, during their period of relative strength, between 202 B.C. and A.D. 216, and only occupied northern China proper between A.D. 3 1 1 and 3 4 9 . 5 0 Much of northwest India was ruled by central Asians for the four centuries straddling the dawn of the common era, and by foreign barbarians of one sort or another from the tenth to the twentieth centuries, but here too, Hüna domination per se was limited to a period of thirty years in the first half of the sixth century.51 The Roman Empire fell to barbarian onslaughts in the fifth century, with the Huns ruling parts of Europe, mainly under Attila, for twenty-eight years, from A.D. 441 to 469. Soon after this time in the West, most of Europe's barbarians gained a new common identity, an identity rooted in their conversion to Christianity and in the Christianized reinterpretation of the Roman Empire they had brought down. No sooner had this conversion occurred, however, than did Europe, or Christendom, find itself besieged by a new sort of infidel, the Muslim, with whom its conflict has never truly ceased. Two questions must be posed, questions we will address at length in the final portion of this book: First, why were the "Huns" singled out as the bogey-men of these traditions, in spite of the extremely brief periods of their occupation of the three great civilizations? Second, how much were the self-identities of these great traditions, for all their ideological bluster and self-assuredness, shaped by the presence of "others" (often wrongly identified with the "Huns") on their borders—on the perceived limits of their "selves"? 4 . T H E D O G JUNG AND CENTRAL ASIAN D O G ANCESTRY TRADITIONS

The earliest case of a Chinese identification of a barbarian group with the canine species is that of the Dog Jung. This barbaric people (Jung means

The Vortex of Cynanthropy

131

"wild, warlike, barbaric") are located, in the "Geography of the Tribute of yü " to the northwest of Shang China. Later, in the approximately fourthcentury B.C. Shan-hai ching, they are situated in the "great northern wilderness" (corresponding to the modern provinces of Shansi and Shaanxi). These Dog Jung are said, in the same source, to be descended from a pair of white dogs, or from a white dog with two heads. The Dog Jung ancestry myth is our earliest Chinese source for northern and western barbarian traditions of dog ancestry. Here we are referring, in general terms, to central Asian peoples in contrast to China's southern or eastern barbarians, some of whom had similar ethnogenic myths. The northern and western traditions thus comprise the ancestry myths of such diverse yet ethnically or culturally related peoples as the Kirghiz, Kazakhs, Mongols, Tungus, Uighurs, Ch'i-tans, Turks, Tibetans, Koreans, Ainus, and Eskimos—to which we might add the Hungarians, Finns, and Estonians, as well as the ancient Scythians reported by Herodotus, of different ethnic backgrounds. In a general sense, these are Altaic peoples, descended from the proto-Turks, proto-Mongols, and proto-Tungus of central and northern Asia.53 These myths are most similar to—probably because they were the source of—the European myths of Amazon women and their cynocephalic men. 54 An Altaic myth belonging to this northern complex probably lies at the source of a report brought back to the West by the Franciscan missionary John of Piano Carpini in 1247, following a stay among the Mongols. Piano Carpini's account mentions "Tartars": these are the Mongols who presumably told him the story.55 Here, their rivals, the Dog-Soldiers of Nochoy Kadzar (nochoyghajar means "Dog Country" in the Mongol language), are peoples living to the north of them: 56 The Tartars, however . . . proceeded to the southeast, and marching for more than a month through the desert reached the Land of Dogs, which in Tartar is called Nochoy Kadzar; for nochoy means dog in Tartar, and kadzar means land. They found only women there without men, and taking two ofthese prisoner they waited by theriverwhichflows through the middle of the country. They asked the women where and of what kind the men were, and they replied that they were dogs by nature, and had crossed over theriveron hearing of the enemy's approach. On the third day all the dogs in the country were seen to be gathering; and when the Tartars made mock ofthem, they crossed theriverand rolled themselves in the sand, which owing to the coldness of the weather then froze. For a second and third time they did the same, and as the dogs were shaggy

132

SIX

the mixture of ice and sand froze a hand's-breadth thick. This done, they charged upon the Tartars, who laughed and began to shoot them with their arrows, but succeeded in killing very few, as it was impossible to wound them except in the mouth or eyes. But the dogs ran swifdy up, throwing a horse to theground with one bite and throttling it with the next. The Tartars, seeing that neither arrow nor sword could hurt the dogs, took to flight; and the dogs pursued them for three days, killed very many, dismissed them from their country, and so had peace from them ever after. One ofthe Tartars even told Friar Benedict that hisfather was killed by the dogs at that time; and Friar Benedict believes beyond doubt that he saw one of the dog's women with the Tartars, and says she had even borne male children from them, but the boys were monsters. The aforesaid dogs are exceptionally shaggy and understand every word the women say, while the women understand the dogs' sign language. If a woman bears a female child, it has a human form like the mother, while if the child is a male it takes the shape of a dog like the father. Piano Carpini (5.30-31) also mentions a Mongol account of a people living to the north of the Samoyeds. This too is recorded, with variations, in the "Tartar Relation": The people are called the Ucorcolon, that is Ox-feet, because ucor is Tartar for ox and colon for foot, or otherwise Nochoyterim, that is Dog-heads, nochoy being Tartar for dog and terim for head. They have feet like oxen from the ankles down, and a human head from the back of the head to the ears, but with a face in every respect like a dog's; andfor that reason they take their namefrom the part ofthem which is monstrous in form. They speak two words and bark the third, and so can be called dogs for this reason also.57 A Western account, of the same century and no doubt the same origin as Piano Carpini's, is that attributed to a certain Armenian King Het'un I, who travelled to the Far East in 1255 and 1266: There exists beyond the Khatai a land where the women have a human face and are possessed of human reason, and where the men have the form of dogs, and are great and hairy and without reason. These dogs allow no one to enter into their territory, and nourish themselves and their women with the game they catch. The males born from the com-

The Vortex of Cynanthropy

133

merce of these dogs with their women resemble dogs and the females (human) women. 58 Earlier than either of these Western sources is an account by a Chinese named Hu Chiao, who was retained among the Turko-Mongol Ch'i-tan (the Khatai of the preceding account, who ruled over an area corresponding to modern Manchuria) in the middle of the tenth century, and gave the following account of a Kingdom of Dogs to the north of the Ch'itan:59 Further to the north is the Kingdom of Dogs [Kou km], where the inhabitants have the bodies of men and the heads of dogs. They have long hair, they have no clothes, they overcome wild beasts with their bare hands, their language is the barking of dogs. Their women have a human form and can speak Chinese; when they give birth to males, they have the form of dogs; when they give birth to females, these have the form of humans. They intermarry, live in caves and eat their food raw; their women and girls are man-eaters. It is said that a man from the Middle Kingdom once came into this country; the woman he met there had pity on him and helped him to escape. She gave him a dozen chopsticks and advised him to drop one every ten li; the male dogs pursued him, but, seeing their household utensils, they stopped to pick them up with their mouths. In this way, they were unable to pursue him. Such is the story that is told. 60 traveler

These three accounts—of the Italian Piano Carpini, the Armenian Het'un, and the Chinese Hu Chiao—all concern the same peoples (the Ch'i-tan and their northern neighbors) and the same region (Manchuria and points north). 61 In his account, Piano Carpini describes the DogSoldiers of Nochoy Kadzar, and explains that nochoy ghajar means "Dog Country" in the Mongol language; there was, in fact, a Mongol people who called themselves nokai [nochoy] ku'on, "Sons of Dogs." This name is moreover identical with that of a tenth-century Ch'i-tan ritual that celebrated, on the eighth day of the eighth month (August, the dog days of northern China as well), the Dog's Head Festival (nai-ho in the Ch'i-tan language; tan-ho nai in Chinese), a commemoration of the "canine" ancestor of the race, whose name was Nai-ho ([dog-] skull). On this day, the niler of the country killed a white dog seven feet from his sleeping tent. He buried the dog so that only its muzzle was left uncovered. After seven days, he transplanted his tent over the head of the buried dog. This canine found-

134

SIX

er of the race, named Nai-ho (related, in the Mongol languages, to the term nokai!nochoy), was worshipped as a skull on this day. The name of the Ch'i. tan imperial family was in fact No-hai I-liu-tsü: their name meant the "I-lju Dogs.5562 A source from the Sung era (A.D. 960-1280) further identifies the I-Hu. tsü, the name of the Ch'i-tan royal family, with a northern people called the It Barak. It Barak (literally "Shaggy Dog55), or Barak, is also the name of a fantastic dog "born from the last two eggs laid by a goshawk ÇAsturpalumbarius)." According to the Muslim author Rashid-ed-din, It Barak was the adversary of Kara Barak (Black Dog) and gave his name to a kingdom. This author describes the men of It Barak as "dark brown and of mongrel visage" and the women as "cadike in shape and doglike in manner.55 The same author gives an account of Oghuz Qahan, a legendary ancestor of the Turks and Mongols, who defeats a group called the Kil Barak, in spite of a strategem that was nearly identical to that of the dogs whom the Tartars faced in Piano Carpini5s and de Bridia5s accounts: the Kil Barak natives prepare their heroes by dipping them in glue and sand to make them invulnerable to arrows.63 It is through a strange twist of fate that the lot of the Ch5i-tan remains intertwined with that of Dog-Men in later traditions as well. After the fall of their Manchurian empire early in the twelfth century, the Ch5i-tan were pushed far to the south and west into the region between Lake Baikal and the Jaxartes River, where they converted to become Nestorian Christians and came to be known as the Kara Khitai. It was this people and their ruler, Wang Khan, who were identified, in rumors spread westward by the Nestorians around A.D. 1145, with the Kingdom of Prester John, an earthly paradise in which, it will be recalled, the peaceable Dog-Men were said to live!64 The Tibetans reproduce an account of northern provenance which bears a certain resemblance to those we have so far reviewed: this is a report given by five Uighurs of their travels through central Asia. This eighth-century account combines a myth of canine ancestry with other themes we have already observed, including polyandry:65 The first dogs were descended from heaven, a red dog and a black dog. These had a she-wolf for their wife, but were unable to have children by her. They then ravished a Drugu [Turk] girl. From their union with this girl, the male children were dogs and the females were of the human race and were true girls. The race born from the red dog was called Ge-zir

The Vortex of Cynanthropy

135

gu-su. The race born from the black dog was called Ga-ra gusu. 66 The girl was mistress of all of their wealth, their herds and their food. These dogs and their women speak in the Drugu language and through the use of signs. The same Tibetan account concerning these Uighurs relates how two from the army of Zama khagan became lost i n a desert of dunes. They wandered into the Land of Dogs, where a young woman (the wife of the dogs from the above account) and her dog husbands came to their rescue and helped them to return to Drugu country.67 There are several accounts from later, mainly Chinese, traditions that parallel those we have reviewed so far, inasmuch as they juxtapose or identify lands of Amazon women with those of Dog-Men. They differ, however, on one important point: nearly all of these later accounts of Dog-Men and Amazon women locate them far to the east, either on the far northeastern coast of "Tartary" or across the Eastern Sea. This shift may be ascribed to a poverty of Chinese geographical knowledge, parallel to that of Western traditions of the Alexander legend, which also tended to identify eastern Siberia with the Northern or Eastern Sea. Alongside this later northeastern maritime location, the Chinese continued to maintain traditions of Kingdoms of Dogs (often connected in some way to Kingdoms of Women) to the west, in Tibet and central Asia, and to the south, in southeastern China and Indochina. Before discussing these Chinese variants, however, we would do well to first examine the "indigenous" central Asian origins of the themes we have seen thus far. In the Uighur travel account cited above, we are told that the women who live with dogs are found in the land of the Turks (Drugu) or to the north of this land. In Hu Chiao's account, the Land of Dogs is also to the north, beyond the Ch'i-tan. Finally, Piano Carpini's and Het'un's narratives place the Dog-Men to the north of the land of the Tartars (Mongols). In fact, all of these peoples—the Turks (T\ı chüeh), Tibetans, Ch'i-tan, and Mongols—as well as many of their neighbors, have their own dog or wolf ancestry traditions. The earliest of these, transmitted by the Chinese shortly before the beginning of the common era, is that of the Wu-huan, who were at that time living north of the Jaxartes River, to the northwest of Tibet. One very early version of the legend is worth reproducing in full, because it bears a tantalizing resemblance to Herodotus's account of the birth of Cyrus, and thus perhaps betrays a connection with Indo-European tradition.68 According to this account, the Wu-huan were forced, shordy soldiers

136

SIX

after the birth of a young prince, to seek refuge in the kingdom of the Shan-yu (Hsiung-nu). The child's preceptor fled into the desert, taking the infant with him. One day, when he returned from begging for food, he found a she-wolf suckling the child and a crow with a piece of meat in its beak flying over him. He then felt the child to be holy, ard resolved to take him to the Hsiung-nu. The Hsiung-nu loved him and raised him. When he had grown up, they gave him the people that had belonged to his father and placed him at the head of an army. In the centuries that follow, Chinese sources cite other wolf ancestry myths, some involving the union of human women with male wolves, with regard to a number of central Asian peoples. A very precious source for such traditions is the Bugut stele, which, dating from A.D. 581, is the oldest archaeological artifact of the T'u chüeh (Turks). This stele is crowned by an image of a she-wolf suckling an infant. Upon it is inscribed the ancestry myth of the P u chüeh, according to which the last survivor of the ancestors of the T'u chüeh, themselves a branch of the Hsiung-nu, was a ten year-old boy. He was left to die by enemy soldiers, who cut off his feet. A she-wolf found him and fed him with meat. He grew up and united with her. She went to a cave in a mountain to the north of Tu'fan (Tibet). There she gave birth to ten boys, who took wives from the outside. From their union, the race was founded. 69 The Hsiung-nu may themselves have been descended from a dog or wolf, if we are to trust accounts given by the Chinese. Indeed, the Peishih (History of the North) and the Wei shu (History of the Wei Dynasty) offer the following account of the origins of the Hsiung-nu: 70 There is a folktale to the effect that a Hsiung-nu named Tan Yü [Father] begat two daughters who were so beautiful that people took them for goddesses. Tan Yü said, "With these two daughters, how can I find husbands good enough for them? I'll present them to heaven." So he built a tall tower in the north of the country where no one lived, and put his two girls on top of it, saying, "I invite heaven to receive them" . . . After a year, an old wolf [lang] came and prowled about the base of the tower in the deep of the night and howled . . . The younger daughter said, "Our father put us here in order to present us to Heaven. Now that this wolf has come, perhaps it is a divine being, sent by Heaven." And with this she started to climb down. Her elder sister cried, "That's a wild animal, not fit to become a mother-in-law with!" But the younger girl disregarded her words, and de-

The Vortex of Cynanthropy

137

to become the wolf's wife. She gave birth to a son. In later times this progeny multiplied into a whole country. That is why these people like to sing songs which sound like a wolf's howl. scended

The lupine myth of the Turks must still have been current in the time of c i n g g i s Qahan, for we find it again, somewhat altered, in the fourteenthcentury Tüan ch'ao pi shih (Secret History of the Mongols), which opens with these lines: "The origin of Cinggis Qahan: There was a bluish wolf [Börte Chino] which was born having his destiny from Heaven above. His spouse was a wild she-dog [Ko'ai Maral]." And some verses later, it is said that Cinggis Qahan himself was descended from the union of Alan K'oa, a human woman, with "a shining yellow man" who, when he moved, resembled a yellow dog. The yellow man/dog was considered to be an incarnation of Heaven.71 Another fourteenth-century Mongol source, the Altan Tobciy states that Alan K'oa gave birth to three sons, whose conception she described in the following terms: "In the dark night, a shining young man entered, his brilliance penetrating into my tent. When he rubbed my belly, he became a black [> kaljin] dog, and departed, licking his chops."72 Perhaps by virtue of a Mongol influence, the lupine ancestry myth is one that is widespread among the peoples of central Asia; but so too is the canine ancestry myth. Whether the latter is a variant of the former or an independent tradition that became conflated with the former, as in the Mongol tradition, is an insoluble problem.73 Let us mention, in this context, a Bulgarian account of the Turks, whom they say sprang from the union between a princess and a dog: now, according to a folk song, the cynocephalic Turks are eating Bulgar children.74 A number of northern Siberian peoples have canine ancestry myths: they include the reindeer Tungus of Manchuria (who claim descent from a bitch and a man who descended from Heaven), the Koryaks, and the C h u k c h e e . 7 5 The Ainu of Sakhalin and Hokkaido also have a dog ancestry myth: their name, moreover, like that of the North American Eskimos, the Inuit, may mean "Dogs" (oinu, inu). The earliest extant literary fragment from Japanese tradition, the eighth-century Kojiki, relates a dog legend concerning the fifth-century emperor Juliaku.76 We have already noted the many Chinese, Indian, and Western references to Dog-Men and Kingdoms of Women in central Asia and Tibet, far to the west of China. Once again, these refer to various peoples of Turko-

138

SIX

Mongol background, and of the Altaic language group, whose homelands shifted—across the great north and far west of China—according to the movements of their neighbors.77 An example we have already noted is that of the Ch'i-tan of Manchuria, who in the twelfth century were forced thousands of miles to the south and west, into Transoxiania. Another people living to the west of the Middle Kingdom are the Kirghiz, whose indigenous dog ancestry myth also serves as a folk etymology for their name (kirk £fiz means "the forty girls"). Here, the forty companions of a khaness return from an excursion to find their camp devastated. The only living creature they find is a red dog. A year later, and presumably through the efforts of the dog, the little colony has doubled in size.78 The Kazakhs and Tanguts of the same region also possess a dog ancestry myth. Still further to the west, the Persians and Armenians have similar traditions.79 In this context, let us also recall the dog-headed demons of Tibet and the pre-fifth-century stucco head of such a demon unearthed at Hadda, discussed at the beginning of this chapter.80 Lasdy, let us recall the Chinese and Indian traditions of a Strîrâjya or Kingdom of Women, located above Kashmir, to the west of Tibet, and to the south of a Kingdom of Dogs (in T'ang Chinese sources), women whose men were identified with the Su-p'i "Huns." These western Chinese traditions, of a Kingdom of Women and a Kingdom of Dogs in the region of Tibet, complement those related to the northern Ch'i-tan in which Dog-Men lived together with Amazon women, in northeastern Tartary or across the Northern Sea. Earlier than nearly all of these traditions are Chinese accounts of eastern seas, or more properly speaking, southeastern seas. This third center for Chinese Dog-Man/Amazon traditions points to China's southern barbarians, as well as to related peoples from Indochina. The earliest mention of eastern Amazons is found in the second-century B.C. Huai-nan tzu.sl The first full-blown narrative, concerning eastern Amazons and Dog-Men, is an account, attributed to Hui-shen, of an encounter that purportedly occurred in A.D. 507: 82 More than 1000 li east of Fu-sang (itself said to be more than 20,000 li to the east of the northeasternmost country known to the Chinese), there is the Kingdom of Women [Nü kuo]. These women are beautifully shaped, and of a pure white color. Their body is hairy, and the hair of their heads is so long that it trails on the ground. In the second or third moon, they eagerly enter into the water, and become pregnant; and give birth to children in the sixth or seventh moon. These women have no breasts on the chest, but, at the nape

The Vortex of Cynanthropy

139

of the neck, hair grows on white stems; in these hairs is a juice with which they suckle their children . . . In the year 507, a man of Chin-an [Fukien province], while crossing the sea, was blown adrift by the wind to an island, and went ashore. People lived there. The women were like those of China, although their language could not be understood. The men had human bodies, but dogs' heads, and their voice was a sort of barking . . . They built up walls of earth of a round shape, the door of which was like a dog entrance [koutou]. Hui-shen's Kingdom of Dogs that is also a Kingdom of Women (Nü kuo) has been identified by certain scholars with the Ainu of northeastern peninsular and insular Asia. A most imaginative interpretation belongs to Gustave Schlegel, for whom this account refers to the matriarchal Kurile fishing communities of Kamchatka, Korea, or Hokkaido, with the DogMen husbands of these women being none other than sea lions!83 This maritime venue of a Kingdom of Women is in fact a perennial one in traditions from all over the world, an illustrious Western example being Adam of Bremen's description of the land of the Amazons in his Gesta Hammaburgensis. In Adam's account, it will be recalled, the Terra feminarum was located on an eastern shore of the Baltic. Here, Amazon women, living together with cynocephalic men, conceived either by mere exposure to blowing wind, or by drinking water. There are, in fact, a great number of Asian sources that provide the same description, of "Amazon" women living together with Dog-Men, who often conceived by bathing in a well or river, or by exposing themselves to the wind. 84 If these similarities are indices of a cultural diffusion, one wonders by what route and through which traditions the account traveled. As we shall see in the next chapter, the location of this land off a southeastern coasdine corresponds to that of a certain Dog Fief Country to which P'an Hu, the canine ancestor of the southern Man barbarians, was sent together with his wife. One of the earliest sources on P'an Hu is the fifth -century Hou Han shu, a text that bears certain parallels with many other sources we have reviewed, inasmuch as it juxtaposes its description of the Dog-Men descended from P'an Hu with that of a Kingdom of Women. So it is that the 115th chapter of the Hou Han shu is consecrated to the Kingdom of Women (which it locates on an eastern island), and the 116th to the myth of P'an Hu and the canine ancestry of the southern Man barbarians.85 We now turn to this latter tradition.

S

E

V

E

N

CHINESE DOG-MAN TRADITIONS: P'AN HU AND THE CH'ÜAN JUNG

1. THE HouHanshuMYTH

OF P'AN HU

No doubt due to their exposure to both the Altaic and the Tibeto-Burmese language families, but also to their longstanding tradition of imperial "scribalism,"1 the Chinese offer by far the world's richest sampling of cynanthropic mythology. More than this, Chinese accounts are often accompanied by annotated bibliographies of sorts. What we find in the Chinese material are at least two levels of "indigenous barbarian" traditions refracted through the lens of this tradition's particular interpretive apparatus. In this way, central Asian and southern Chinese ancestry myths arc often transformed into cosmogonic myths or mythic representations of various economic relationships between vassal and empire. We will begin our approach to these many strata of meaning with the Hou Han shu (History of the Latter Han) myth of P'an Hu. This account is the foundation myth for a tributary relationship that the Middle Kingdom maintained with the Man, a group of southern barbarians, from perhaps the beginning of the common era. This myth, if it is of Man origins, is one which borrows heavily from Chinese "barbarian genres" in the Chinese (re-)telling of it. As we will show, however, there is much of the barbarian in China's selfidentity, in spite of the voluminous evidence that Chinese tradition musters to the contrary. Much of this and the following chapter will be devoted to showing how the Man, in borrowing a mythological pedigree from the Chinese, were ultimately "borrowing from themselves" at several removes from a primordial situation in which "Chinese" and "barbarian" had not yet been distinguished from one another. The P'an Hu myth is found at the beginning of chapter 116 of the Hou Han shu, where it introduces an appendix to the main body of the work. This appendix, several chapters in length, is entided "Nan man chuan" (Biography of the Southern Barbarians). While Fan Yeh, the author of the Hou Han shu, wrote in the fifth century A.D., it is certain that the mythic 140

Chinese Dog-Man Traditions :rial of this

141

account was already known to the Chinese by the third cenA.D., if not before.2 Formerly, in the time of Kao-hsin, that emperor was troubled over the banditry and depradations of the Dog Jung, but his attacks on them were unsuccessful. A chieftain of this tribe named General Wu was especially formidable. So volunteers were solicited from all over the empire: whoever could obtain the head of a certain General Wu, a leader of the Dog Jung, would be given a thousand jy* of gold, a portion of land large enough to support ten thousand families, and the emperor's youngest daughter. Now at that time, the emperor had a domesticated dog whose fur was of five colors, called P'an Hu. After this proclamation had been given, P'an Hu, holding a human head in his mouth, came up to the imperial palace. The courtiers all wondered at it, and when they examined it, they found it to be General Wu's head. The emperor was gready pleased. His intention was to reward the dog, but he did not know how to do so appropriately, since P'an Hu could not very well be given a girl as wife or granted rank and emolument. But when his daughter heard of all this, she felt that the emperor could not go back on what he had proclaimed, so she asked permission to put the command into effect. The emperor then had no choice but to give his daughter to P'an Hu. P'an Hu, having obtained a wife, bore her on his back to Nanshan ["Southern Mountain," in the Lu-chih district of Hunan province], stopping at a stone chamber. That place was extremely treacherous: no human tracks went up to it. There the girl took off her clothes, made a pu-chien (dogcoiffure) hair arrangement, and donned a tu- li (dog-tailed) dress. The emperor was sorrowful and missed her, and sent envoys to search for her, but these met with winds, rains, earthquakes, and darkness, and could advance no further. 3 Three years after this marriage, the girl gave birth to twelve children, six of each sex. Then P'an Hu died. These children then took each other as husbands and wives. They wove tree bark and dyed it with vegetable juices; they liked five-color clothing, and the fashion and cut of their clothes always left a tail at the end. Their mother later returned to court, in order to present an appeal to Kao-hsin, the White Emperor, whereupon he summoned his grandchildren to his court.

142

SEVEN

Their clothing was of variegated color, and their speech sounded like zhuli.4 They preferred living in mountain and ravine areas and did not care for flat, open country. The emperor acceded to their wishes, bestowing upon them spacious mountains and wide marshes. Their descendants multiplied, and have since been known as Man-i, or barbarians. They are cunning but appear silly; they are conservative regarding the old customs and love to stay at home. Because their ancestral father [P'an Hu] achieved merit and their ancestral mother was an emperor's daughter, they are exempt from taxation even though they make commercial use of their fields. They have villages and lords and chieftains; all of the latter are given [or: bestow to their own ministers] seals of office. Their caps are made of otter skin. Their great leaders are called cbing-fu (spirited fellow); they call each other hang-tu.s The Man people of present day Wu-ling in Ch'ang-sha [Hunan] are a branch of these. 2 . P'AN H U AND THE DOG JUNG, M A N , AND YAO

The P'an Hu myth places us in the time of the legendary twenty-fourthcentury B.C. Hsia emperor Kao-hsin. His rivals are none other than the Dog Jung, China's—if not the world's—most ancient cynanthropic race. Therefore their canine pedigree goes back much further than that of P'an hu's Man, since their origin myth first appears in the early mythological compendium called the Shan-hai ching (Records of Mountains and Seas), in the period of the Warring States (403-221 B.C.). According to this source, the Dog Jung were descended from a pair of white dogs. Their lineage ultimately goes back to the Yellow Emperor (Huang-ti), the civilizing hero of Chinese tradition. Huang-ti was the father of Miao-lung, grandfather of Yung-yu, and great-grandfather of Bian Ming, who was the progenitor of the two white dogs who founded the Jung race.6 There is an alternate Dog Jung ancestry myth, however: the earliest references to P'an Hu, the third-century A.D. Wei Lüeh and the fourth-century Sou-shen chi (In Search of the Supernatural), state that the Jung are the descendants of the same P'an Hu who decapitated the Dog Jung general in the Hou Han shu\ A portion of this chapter will be devoted to untangling this confused situation.7 Certain other features connect the Jung and the Man, as well as the Hsiung-nu, with things canine in the Chinese context. The Chinese

Chinese Dog-Man Traditions

143

deograms for Jung and Hsiung-nu are written with "dog" radicals, a usage addition to the Jung and their Hsiung-nu cousins, the southern Yao, Man, and Miao peoples—whose origin myth of P'an Hu we have just presented—bear simile associations in Chinese orthography.8 This is not all. In the Chinese sources, barbarians in general, and P'an Hu in one of his myths in particular, are closely associated with the mythic theme of chaos. This interconnectedness of themes and peoples is quite explicit in Chinese terminology. The terms for Hsiung-nu or Hun (hun-i), the dog ancestor P'an Hu (Platter Gourd) and hun-tun ("Chaos Gourd," the original unformed mass from which the cosmos was shaped), and "halfbreed" (huri) are etymologically connected to one another. This cluster is further related to the orthography and symbolism of K'unlun, the western mountain (in modern Tsinghai or Kokonor province, on the northern border of Tibet) that is schematically represented, in later periods, as an inverted gourd placed mouth to mouth over an upright gourd. The K'un-lun mountain range is the locus of a flood (chaos) myth of P'an Hu; it also comes to be identified as the homeland of the Dog Jung. It would appear that the Dog Jung did in fact move to the west in the Han (206 B.C.—A.D. 220) and Chin dynastic periods, and came to intermingle with the T'u-fan (Tibetan) and Ch'iang peoples of that region. Yet, according to the twelfth-century B.C. Yü pen chi, (Geography of the Tribute of Yu,) K'un-lun was located, at the close of the Shang dynastic age, in precisely the region occupied by the Dog Jung at that time. Their historic migration may therefore have occasioned a mythic transfer of the nexus of chaos itself, since the K'un-lun range has been situated, since at least the second century B.C., far to the west of where it was in the late Shang era, in Tsinghai province. We will return to these important themes in due course.9 The Shan-hai ching links K'un-lun to another set of canine toponyms: "To the north of K'un-lun mountain there is a man called Earl Ta-hsing. To his east lies Dog Fief Country (Ch'üan Feng kuo). Dog Fief Country is also called Dog Jung Country (Ch'üan Jung kuo)."10 Herein hangs a tale. Kuo P'u (A.D. 276-324), an important early commentator on the Shan-hai thing, ascribes the identification of Dog Fief Country and Dog Jung Country to the phonetic similarity of the Chinese termsfeng andjung. In fact, the center of Dog Fief Country, K'uai-chi (on the east coast of Chekiang province), is far to the south and east of Dog Jung Country.11 We may thereby surmise that the Shan-hai ching was already conflating two distinct that betrays much of Chinese ideology regarding its barbarians. In

144

SEVEN

dog ancestry traditions, one of northwestern (Jung) and one of southeastern provenance. Kuo P'u is aware of this fact; however, in attempting to square the Shanhai ching data with a reference from the Hou Han shu, he places Ch'üan Feng kuo, "Dog Fief Country," in a region across the Eastern Sea from the Chinese mainland. This he does in his commentary on this Shan-hai ching (12. la-12.2a) passage, in which he relates a sequel to the P'an Hu myth related above. Here he says that "because Kao-hsin could not tame P'an Hu, he made him swim out into the Eastern Sea oft'the coast of K'uai-chi. There P'an Hu and his wife were granted a fief [feng] of300 li in area. When his wife and her descendants gave birth to males, they were dogs; when to females, they were women. This is Dog Fief Country."12 ThcHsüan chung chih and Kuangpo wu chih (Enlarged Miscellaneous Notes) reproduce Kuo P'u's material, but introduce slight changes: the land to which they are sent is called Kou-min kuo (Dog People Country) rather than Ch'üan Feng kuo, is 300 square li in size, and is located 21,000 li off the coast of K'uai-chi.13 In defense of Kuo P'u, we should mention that he wrote his commentary to the Shan-hai ching on the basis of his familiarity with the Sou-shen chi, attributed to Kan Pao of the Chin dynasty (A.D. 265-419). He was, however, unacquainted with the Hou Han shu version of the P'an Hu myth. The Sou-shen chi version, recorded about a century before that of the Hou Han shu (but in fact later than the oral tradition of the myth recorded in the Hou Han shu), has P'an Hu enfeoffed as the Marquis of Kuei-lin (Kwangsi)— and not of K'uai-chi (Chekiang)—and makes him the ancestor of the Dog Jung. 14 Kuo P'u evinces further confusion when he situates the center of the P'an Hu tradition not in Kuei-lin or on the Wu-shan mountain (in Kwangsi and Hunan, in west central China, and the locus of the Man cult of P'an Hu), but rather in K'uai-chi (in Chekiang, on the southeast coast of China). He ultimately transfers to this latter toponym all of the dog ancestry traditions he knows—those of the Dog Jung as well as of the Man. There is a concrete reason, however, for Kuo P'u's choice of K'uai-chi, on the southeast coast of China, over the Man regions of Kuei-lin and the Jung regions, "near K'un-lun mountain." This is the existence of another southern people called the Yao, who, as we will show in detail, had a P'an Hu cult of their own. Their cultic center was, moreover, located at the very toponym indicated by Kuo P'u: the Yao cult of P'an Hu was based at K'uaichi, in Chekiang, on the southeast coast of China. So, in fact, Kuo P'u fuses three traditions (Yao, Man, and Jung) into one. 15 As we will see, the Chinese remained confused on this subject, but we

Chinese Dog-Man Traditions

145

indicate that their confusion may have been occasioned in part by a series of migrations effected by the Man and Yao peoples from Shaanxi, Hunan, or Anhwei provinces south into Yunnan, and southeast into Chekiang, as well as into Fukien and other coastal areas.16 It is also possible that the Yao and other southeastern peoples may have "inherited55 their ancestry myth from the Chinese (who would have learned it from some northern Altaic or Turko-Mongol people) and made it their own, in a case of the "pizza effect.5517 In their later retellings of these "barbarian55 ancestry myths, the Chinese employ the terms Yao or Man more or less interchangeably when they describe those southern Chinese peoples who claim the dog P5an Hu for their ancestor. Man is the term the Hou Han shu employs at the end of its account of these peoples; Yao is another blanket term the Chinese have used, since perhaps the eleventh century, for their southern barbarians: in Chinese, Yao means "(children of the) dog.55 It is noteworthy, however, that in the Yao language, the term for human being is man, the same as the name of the other "barbarian55 people (the Man) claiming descent from P5an Hu. 18 Regardless of the origins and tangled lines of tranmission of the dog ancestry myths, we can be quite certain that the Man or Yao peoples, even though they considered themselves to be descended from a dog, nevertheless did not call themselves dogs. Only the Chinese would have done so. Once again, as was the case with the Hsiung-nu and so many other peoples, the foreigner is rendered anonymous, an indistinguishable splash in a protean sea of outsiders, by being given the same (animal) signifier as every other foreigner. In fact, there continue to exist in modern day China a great number of peoples or tribes who, in spite of their broad classifications as "Man,55 "Yao,55 "Miao,55 or "Liao,55 vary significandy from one another in their interpretation of their own traditions, as well as in their social structure. Among these may be counted the Hsia-min, the Hsia-bo, the Lolo, the Li, the Shan, the Shaka, and the Lung-chia.19 Edouard Mestre is very explicit in delineating the historical distinctions that have developed between different Yao peoples, as well as the differences between the Yao and the Man: 20 should

I f . . . we bear in mind the fact that certain of the Yao divide themselves into tribes and others into clans, of which certain are sedentary while others practice slash and burn cultivation (with these latter being the descendants of P5an Hu's eldest son), that certain speak the so-called Tao language (in which the term for "human being55 is man, mien, or

146

SEVEN

mun), while others speak a southern Chinese dialect, and still others a T'ai dialect, there can be no doubt that we are in the presence of a mixture of peoples. . . . It is their concern to maintain their independence that most clearly distinguishes [the Yao] from the more ancient Man. These latter were organized into chiefdoms whose tided individuals received their investiture from the Chinese government. They regularly offered tribute [to the Chinese] and, on such occasions, they frequendy requested that they be assimilated into the Chinese people proper, a request that was gradually granted to them . . . It is following this fusion of the majority of their population into the Chinese [system] that we see them disappearing, over time, with their place being taken, in the Sung period, by the Yao who themselves remained "apart," as they say in their "charters." Nevertheless, the present-day core of the Yao population is composed of elements of the Man who preferred to fall back and associate themselves with other ethnic groups as a means to preserving their old ways. Under such conditions, we should not be surprised to find the cult of P'an Hu appearing under various forms, in accordance with changing places and times. Within mainland China, both Yao and surviving Man populations are presently clustered along the southeast coast and China's southern border with Vietnam, Laos, Thailand, and Burma. While toponyms related to P'an Hu are found in some of these regions, the greatest concentrations of these are found in the more northern and interior province of Hunan. These, moreover, testify, in spite of the confusion introduced by Kuo P'u, to the northern locations of the Man at the time of imperial China's first contacts with them. For the period chronicled in thc Hou Han shu, Hunan, then called Ching province, was very much a southern region, as were all territories south of the Yangtze River. While Han did in fact manage to penetrate south of the Yangtze as early as 127 B.C., it never had significant contacts with its southern barbarians. It was only after the fall of the Latter Han in A.D. 220 that a truly "Chinese" kingdom was established in southern China. This was the short-lived kingdom of Han Shu (221-263), which fought over the remains of the old Han empire with the northern Wei (221-280) and the southeastern Wu (221-280), during the period of the

Chinese Dog-Man Traditions

147

Three Kingdoms, the beginning of the Chinese "middle ages." It is, in fact, in this period that we first hear of the Man being mentioned together with p'an Hu, and it is precisely in Hunan that both are located. So it is that we find in a Chinese source, the Ch'ien-pao chi, an account of the Man barbarians: "In Wu-ling [in the Ch'ang-sha and Lu-chiang Comnianderies of Hunan] are the Yi [I], P'an Hu's descendants. They live here and there in the Five Streams area. P'an Hu was able to clamber up onto dangerous mountain fastnesses, where others were often injured. They [the I] mix up fish and meat in a trough and utter cries, and thus they make sacrifice to P'an Hu. They are said by [Chinese] commoners to have red buttocks and cross-belted skirts. These are precisely P'an Hu's descendants." This account is nearly identical to a slighdy earlier source, the Chinchi (Annals of the Chin Dynasty: A.D. 265-419). This text, reconstructed from later citations, agrees with the ChHen-pao chi on Man customs, and adds that they "always cause a great deal of trouble." The same customs are associated, in a twelfth-century source, with the Yao of the same region.21 The mountain area to which P'an Hu takes his new wife is called Nanshan (Southern Mountain) in the Hou Han shu\ it is identified with Wushan mountain in southwestern Hunan province, a mountain described by Huang Min, in his sixth-century Wu-ling chi (Record of Wu-ling), as tens of thousands of feet tall. In its midst is "P'an Hu's stone chamber," which can accommodate several tens of thousands of people. Inside it there is a "stone sheep" and other stone animal shapes; and ancient tracks and the marvellous and strange are particularly abundant. Looking at the caves from some distance, each appears about as large as three sitting rooms; and from afar, one of the stones resembles a dog's shape: the people say it is a "Man," and that it is P'an Hu's image.22 A similar cave is described—this time at a place called Ch'ien-chung, in Wu-ch'i, slighdy to the northwest of Wu-ling and Ch'ang-sha—in two T'ang period documents. One of these, the Man shu, states that the skin and bones of P'an Hu are visible at this site. The sixth-century Shui-ching chu (The Annotated "Classic of Waters"), states that P'an Hu's center is found at the source of the Wu River. This region itself is known as Wu-ch'i, and its Man inhabitants are called the Wu-ch'i Man. 23 A great number of other names and toponyms containing the phoneme " Wu" are found across much of southern China. Many are also found in the

148

SEVEN

P'an Hu traditions, where they evoke General Wu, whose head P'an Hu bit off and presented to Kao-hsin. Indeed, the complex relationship between P'an Hu and Wu in geography and myth will be our key to interpreting the history of the Man and Yao peoples in their relations with one another, as well as with the Chinese, in this period of early contact If we are to understand the history of contacts between these peoples on the one hand, and the P'an Hu tradition that has been left to us on the other, we ought to first review three variations on Fan Yeh's fifth-century Hou Han shu myth of P'an Hu, the myth we related at the beginning of this chapter. These are the approximately fourth-century Sou-shen chi of Kan Pao, and the quite recent "P'an Hu charters" of the Hsia-bo of Fukien province, and the Hsia-min of Chekiang. We begin with the Sou-shen chi version: Once upon a time the emperor Kao-hsin was perturbed by the rebellion of King Fang. In order to preserve the integrity of his empire, he proclaimed that whoever could bring him the head of King Fang would be awarded a thousand ounces of gold and beautiful girls. The emperor had a favorite dog named P'an Hu, who disappeared on the day of Kao-hsin's proclamation. It went to King Fang, who was very much pleased and held a great feast in honor of the dog. On that night he was very drunk and slept soundly. The dog bit off his head and went with it back to his master. Then the emperor saw King Fang's head, he was overjoyed with its meritorious deed, and ordered that it be fed with abundant meat and rice, but the dog did not eat any. A day passed without the dog even answering to His Majesty's call. When the Emperor said to him, "Why don't you eat? I suppose you hate me for not having rewarded your merit. Now I shall reward you as promised. Would that make you happy?" As soon as P'an Hu heard the Emperor's words, he jumped up with joy. Accordingly, the Emperor established him as the Marquis of Kuei-lin and gave him five beautiful girls and the tribute of one thousand households of Kuei-lin Commandery. With these beautiftd girls, three sons and six daughters were born. Although they looked like human beings, they still retained a dog's tail. As time went on, his descendants increased in great numbers, known as the Ch'üan Jung or "Dog Tribes." Emperor Yü of the Chou dynasty was killed by these Ch'üan Jung. Now the aboriginals of the said commandery were offspring of P'an Hu. 24

Chinese Dog-Man Traditions

149

Both the Hsia-min and the Hsia-bo (Yao peoples of southeast China 25 ) have their own P'an Hu "charters," records which, although they are indisputably later than that of the Hou Han shu, are nonetheless highly useful since they synthesize many disparate traditions into meaningful wholes. In the Hsia-bo26 variant, P'an Hu brings the head of General Wu to the king (named Pu-k'u rather than Kao-hsin). At this point, the king ordered at once that the dog should be rewarded, given a high rank and better food and accommodation, for it was a meritorious dog. But the dog was not happy at all. The dog then ran into the princess's room with the bloody head and would not come out. Remembering his promise, Pu-k'u then told the dog that it was not fitting that he should marry the princess. "For," said the king, "you are a dog." As soon as the dog heard this, he put his head down on the floor and said to the king, "Put me under a bell for forty-nine days and do not allow anyone to peep in during that time." This was done, but on the forty-eighth day, they thought the dog must be dead of hunger and took away the bell.27 Then they saw a man where the dog had been, but he had not yet had quite enough time to change himself entirely and still had the head of a dog. Pu-k'u then felt obliged to keep his word and married the princess to this dog-headed husband. People called her the dog-headed princess. The princess was ashamed of her dog-headed husband, so they went far away and lived in the forest. The princess was afraid to look at her husband's face, so she dressed her hair up to the top of her head and tied a piece of red cloth to it. When her husband came near, she pulled the red cloth down over her face so as not to see him. The princess had three sons, to whom Pu-k'u gave the names Lui, Pan, and Lan respectively.28 The Hsia-min account, drawn up in 1803, constitutes this people's genealogical charter, by virtue of which they became recognized as Chinese citizens by the central government: 29 The army of Wu was harassed by them. They invaded the country, there was war, and the people suffered death and destruction. None of the vassals could help. (There was) received a command from the hall of heaven that Lu Chinkou (Lu, the Golden Dog) should descend to earth and tame the moon-heart-tiger. Now the Golden Dog descended into the palace of the emperor Kao-hsin. And the doctors

150

SEVEN

brought forth a golden insect. All the ministers of the court covered P'an Hu with lotus leaves and nourished him for several days.30 Soon he was transformed into a dragon-dog. He had 120 spots on his body. The emperor grew afraid and said, "We must make a proclamation: if there exists a brave warrior who can restore peace to the country, then I will give him to the third princess and appoint him as imperial son-in-law." Only the dragon-dog answered the proclamation, saying he could catch the enemies. He jumped over the ocean and turned into a golden dragon. After seven days and seven nights, he came to the hall of the King of Yen. When the king was lying completely drunk on the bed, the golden dog came in and bit off his head. He was able to change himself into a human being, except as yet for the head. The princess wanted to become his wife. The empire having been made secure, all the countries brought tribute and came year after year to be received in audience. Our [the Hsia-min's] ancestor, however, was granted the tide of a son-in-law as an imperial favor, and received his seal as King P'an Hu. He had three sons and a daughter. The first son was granted the surname of P'an, the second Lan, and the third Lei. They lived on Mount K'uai-chi and were free of all taxes.31 3. M A N AND YAO IN THE PERIOD OF THE THREE KINGDOMS

There are a great number of contradictory elements in the versions of the P'an Hu myth we have reviewed here. Principal among these is the name of the dog P'an Hu's emperor, as well as that of the enemy general, and the locus of P'an Hu's cult and descendants. We present these variations along with certain supplementary material culled from other P'an Hu charter myths. Source

Emperor's Name

General's Name

Hou Han shu

Kao-hsin (Ti K'u)

Wu

Man shu32

Kao-hsin

Wu

Sou-shen chi

Kao-hsin

King Fang

P'an Hu Cult Locus Wu-ling and Ch'ang-sha (Hunan) Ch'ien-chung (Hunan/ Kiangsi) Kuei-lin (Kwangsi)

Chinese Dog-Man Traditions Hsia-bo Hsia-min Yao 34

Pu-k'u Kao-hsin protector of Wu P'ing

King of Yen33 Kao

151

K'uai-chi (Chekiang) K'uai-chi (Chekiang)

Let us consider these three elements one by one. First, the emperor is Kao-hsin in both the Hou Han shu and Sou-shen chi (this latter is the source followed by Kuo P'u in his conflations of numerous traditions, related above). In the Hsia-min version, the emperor's name is Kao-hsin, but he is now the lord, rather than the enemy, of General Wu—with a certain King of Yen taking the place of Wu as Kao-hsin's adversary. Even more bizarre, in certain Yao charters, the enemy general himself is named Kao, while the name of the P'an Hu's emperor is P'ing. Furthermore, it will also be recalled that P'an Hu is himself said to be the ancestor of the Dog Jung (whose General Wu P'an Hu beheads in the Hou Han shu version of the myth) in the third-century Wei Lüeh and Sou-shen chi, as well as the ninthcentury Man shu.35 Finally, there is disagreement between the various documents over the locus of the P'an Hu cult: it is either located in the west (southern Hunan shares a common border with northern Kwangsi; Ch'ien Chung is located to the northwest of Wu-ling, in Hunan) or the east (K'uai-chi is in Chekiang). One key to this puzzle is the name "General Wu." Edouard Mestre, who devoted years of research to this subject, presents the material as follows. "General Wu" was a general, named P'an Hsün, from the kingdom of Wu (A.D. 221-280). Not far to the south of P'an Hu's Nan-shan mountain and Kuei-lin, this "general" received sacrifices similar to those received by P'an Hu on his birthday, as related in the Ch'ien-pao chi. In the case of General Wu, or P'an Hsün, this figure was offered sacrifices annually, in appreciation for his defeat of the Man of the Five Streams region, many of whom he decapitated. In the case of P'an Hu, he received, in addition to the usual meat, fish, rice, and wine, an offering called "the head of General Wu": the practice of offering a real human head was oudawed in the third century, but a replacement made of sago flour, pork, and mutton was used at least until the seventeenth century. In the period of the Six Dynasties (A.D. 220-589), the Liao, neighbors of the Man in this region, offered the heads of dogs to their god, who was represented as a human head. This reversal parallels, in a sense, those of the mythic variants we have so far discussed.36 This is not all. Apart from an apparent opposition between Man and Wu named

152

SEVEN

peoples across an east-west axis approximately equivalent to the modern border of Hunan and Kwangsi provinces, there was another "sensitive" border to which medieval sources refer. This was the east-west border between the present day provinces of Hunan and Kweichow. This was a border shared or disputed between the Man of Hunan and various Yao peoples of Kweichow. So it is that we find P'an Hu called, in one variant of his myth, "Count Border-Fixer."37 So far so good. But what are we to make of a reference to the Miao, a Yao people living in Kweichow, who sacrificed, in their new year celebrations in the tenth month, to P'an Hu, "the exterminator of the Man of the FiveStreams regionn> On the basis of this source, we would have to conclude that P'an Hu was the name of a Yao ancestral hero who defeated the Man, just as had the General of Wu, P'an Hsün, in other accounts. This is supported in several other sources. So, for example, the Yao maintain that P'an Hu snuffed out the kingdom of the Man-i when he carried off the head of Kao. In the same vein, the Miao of Kweichow call themselves the descendants of P'an Hu, who exterminated the Man of Wu-ch'i. Also, a fifteenth-century Ming source, the Ta Ming yi fung chihy states that the eastern region of Kweichow was an extension of Wu-ch'i, in western Hunan: as such, it "holds P'an Hu [i.e. the Man of Ch'ang-sha, in the Five Streams region] by the throat." Finally, we find a Yao version of the P'an Hu myth, originating from Kwangsi and Kwangtung to the south, that has the dog behead the general of the enemy, here referred to as the Ken Yung ("Mountain Bears," a transformation of Dog Jung), in the service of Kao-hsin. Once again, we find a systematic reversal of the values assigned to various peoples and characters in our sources.38 The history of the period of the Three Kingdoms fortunately untangles many of these problems. For some sixty years following A.D. 221, what had previously been Han China was divided into three separate kingdoms which vied with one another for control over the whole of the imperial territories. The strongest of these was Wei, which ruled China north of the Yangtze River. South of the river was (Han) Shu, which ruled the southwest from Szechwan, and Wu, which ruled the southeast from Nanking. These two latter kingdoms were the first historical instances of "Chinese" kingdoms being ruled from southern capitals. Both were conquered and ruled by generals from the north and occupied by their armies. Wei, to the north, was probably the most "legitimate" successor to the Latter Han, but it was Shu that claimed direct descent from the Han lineage as justification for its imperial ambitions: the Han ancestor it claimed

Chinese Dog-Man Traditions

153

vvas in fact very distantly related to its princes, and the connection was tenuous. Shu was the weakest of the three kingdoms, and fell in A.D. 263. The kingdom of Wu consisted mainly of marshy plains and narrow valleys. When the rulers of Shu and Wu moved into the south, they found themselves obliged to give up northern wheat for southern rice cultivation, which their conquered T'ai subjects were already practicing; their Yao subjects were hill-dwelling hunters and primitive farmers.39 Wu mounted perhaps two military operations during its sixty-year reign. The first was a successful conquest of certain eastern Shu lands. This was precisely the campaign of the "mythic" General Wu, or P'an Hsün, who, in the course of pushing back the Shu borders, decimated the Man populations of the Five Streams area. This theme—whether it is of historical or mythic origins—dates from the third-century sources of Fan Yeh's Hou Han shu. The second Wu attempt at greatness did not meet with the same success as its first campaign. A new power in southern Manchuria, the kingdom of Yen, enticed Wu into an alliance as a strategy for attacking Wei from north and south simultaneously. The gambit failed, and in 237, Wei wiped out Yen: Wu went the same way in 280. 40 We can now generate a schema of two diametrically opposed traditions that tell two sides of the same story. On the one hand, there is the tradition of the Hou Han shu, which concludes with the statement, "The Man people of present day Wu-ling in Ch'ang-sha are a branch of these [descendants of P'an Hu]." This we will call the "Man" version. On the other hand, we have a group of traditions that make the Man the victims of Wu or of the Yao: this we will call the "Yao" version. Man P'an Hu is the ancestor of the Man P'an Hu is the pet and defender of Emperor Kao-hsin Kao-hsin's enemy is General Wu of the Dog Jung P'an Hu is established as "Count Border-Fixer" of the Five Streams area P'an Hu's victory over General Wu is celebrated with an annual sacrifice of the "head of General Wu"

Tao P'an Hu is the exterminator of the Man P'an Hu is the defender of emperor P'ing P'ing's enemy is Kao General P'an Hsün of Wu decimates the Man of the Five Streams area and pushes back the Shu border. General P'an Hsün's victory over the Man is celebrated with sacrifices to him; sacrifices of dogs are offered to a head (of General Wu?)

154

SEVEN

The obvious conclusion, and here we follow Mestre once again, is that there were two different "P'an Hu peoples," one Yao and one Man, located on opposite sides of at least one border, if not of two. Furthermore, there would have been some collusion, in this early period of Chinese contacts with southern barbarian peoples, between the Man and Shu on the one hand, and the Yao and Wu on the other. 41 The sole "truth" of the entire group of narratives is that the people who were "exterminated" in the Yao accounts were the original Man of the Five Streams region. "Ghettoized" is probably a better characterization, inasmuch as Fan Yeh's Hou Han shu version concludes with the Man followers of P'an Hu living high in the mountain fastnesses of Hunan, with ties of fealty to the Chinese. The Man of Hunan had, in fact, been living on the fringe of Han China since the time of the Han emperor Wu (157-87 B.C.), a sovereign not to be confused with either the small kingdom or the general of the same name. The Man had, since that time, been generally favorable to sinicization, whence their alliance, in the period of the Three Kingdoms, with Shu, the sole kingdom that claimed hereditary descent from Han. This is the source of the identification of P'an Hu's master with the primordial emperor Kao-hsin: in fact, it is to the ruler of the kingdom of Shu that the myth refers. Opposed to the Man, if only because they were opposed to sinicization, were the Yao, another southern people whose homelands, further from the borders of the Han empire, had remained quite insulated from Chinese influence down to this period.42 The Yao became tributaries of Wu, whose general P'an Hsün overran the border-dwelling Man to conquer the eastern part of the Shu kingdom. These Yao peoples had their own cult of a dog ancestor, whom they too called P'an Hu. Their cult, however, was centered in K'uai-chi, near to the Wu capital of Nanking in northern Chekiang. This was, of course, the same region as the Dog Fief Country (Ch'üan Feng km) that Kuo P'u conflated with Dog Jung Country (Ch'üan Jung km) in his fourthcentury commentary on the Shan-hai ching. It should also be noted that the first Chinese mention of P'an Hu is found in the third-century A.D. Wu-yin li-nien chih (Chronology of the Five Periods), which was written by a subject of the kingdom of Wu, Hsü Chen. In the context of the ongoing conflicts of the period of the Three Kingdoms, conflicts that had repercussions among the various southern peoples of China, certain other of the mythic elements of the P'an Hu traditions become meaningful, if not clear. This is particularly the case with the Yao variants that make Kao(-hsin) the enemy, rather than the protagonist, in

Chinese Dog-Man Traditions

155

the myth. The name of the emperor who replaces Kao-hsin, P'ing, may be explained by an alliance that was struck between Wu and the Manchurian kingdom of Yen: P'ing is the name of a state located on the northern boundary of China during this period.43 More to the point in this regard is the very late Hsia-min variant that, besides placing Kao-hsin and an army 0 f Wu on the same side, makes the enemy the king of Yen, the very kingdom with which the kingdom of Wu sided in A.D. 232! The next question that must be answered is who or what P'an Hu represented for two rival "P'an Hu peoples," the Man and the Yao. Was P'an Hu the name of the ancestral dog of both rival peoples? This was certainly the case as far as the Chinese were concerned. In fact, the Yao are not even mentioned by name until the seventh-century Sui shu (History of the Sui Dynasty), where they are called the Mo-yao. They are not explicidy distinguished from the Man until an eleventh-century source, which speaks of the Man-yao, who are to be identified with the Yao-jen, but who are different from the "Cheng-Man," the "true Man."44 This would explain the third-century references to a cult of P'an Hu in K'uai-chi, the conflation of Dog Jung Country with Dog Fief Country, and the mistaken identification of P'an Hu as the ancestor of the Dog Jung (i.e., the Yao peoples of Dog Fief Country, or northern Chekiang). In the meantime, the Man had their own cult of P'an Hu, centered in the Ch'ang-sha region of Hunan, as related in the Hou Han shu version of their ancestry myth or charter. In a slighdy later period, from perhaps the time of ninth-century Man shu onwards, the Chinese began to differentiate between the Yao and the Man, gradually reversing earlier policy and shifting from calling both "Man" to calling both "Yao." With the gradual eclipse of the Man, whose absorption into both Yao and Chinese populations had been ongoing for centuries, the variants of the myth come to bear less and less of a connection with historical peoples, with the result that the Hsia-min, a Yao people, come to ally Kaohsin with Wu against Yen in their "charter." Elsewhere, the Man shu is able to say, in a single passage, that P'an Hu killed General Wu of the Dog Jung, and that the Dog Jung are P'an Hu's descendants.45 This long historical interpretation of the P'an Hu myths, if it has reconstituted a proper chronology of their development and spread, has advanced but litde our understanding of its meaning—the meaning it has borne for the peoples for whom it has remained a genealogical charter, down to the present century. Here, we are speşüdng of the difference between langue and parole, context and text, system and event. What did the

156

SEVEN

event of the P'an Hu myth, of the promulgation of a genealogical charter, mean for the southern barbarians who gradually developed ties with the Chinese empire? In the Hsia-min and other contexts, the P'an Hu ancestry myth becomes that non-Chinese people's "passport" into imperial Chinese politics, society, and culture. In this way, an original and indigenous myth is ultimately placed in the service of a great tradition's imperialistic ideology: P'an Hu is "sinicized" (and "cyno-cized") into a subhuman functionary of the emperor of heaven. He performs a military execution for his emperor, for which he is enfeoffed and given an imperial princess for his wife. His descendants become the "indigenous" Man or Yao tribes, who sometimes enjoy a preferential tax status. In this euhemerized context, neither P'an Hu nor the Man or Yao even existed, in the Chinese perspective, before the former was brought down into the world to perform a service for Chinese heaven. The same sort of sinicizing tendency is already visible in the "original" Hou Han shu version of the P'an Hu myth: the founder of the Man race is the Hsia emperor's obedient dog. Thus these peoples' "borrowings" of their own ancestry myth—from a Chinese literary tradition of southern barbarian genealogical charters—have at once constituted cases of the "pizza effect" and the continuation of indigenous traditions. In the course of some 1,500 years, the two have become intertwined so as to be quite indistinguishable from one another. 4 . RAW AND COOKED BARBARIANS, BARBARIAN AND IMPERIAL RICE

Regardless of the positive or negative valences of the various characters in the P'an Hu traditions, there is one factor that remains constant. This is the rewarding of P'an Hu by his emperor for the service he has rendered to the Empire. In every case, the hero P'an Hu is enfeoffed and wedded to an imperial princess, upon whom he sires sons and daughters. These he (or his wife) brings before an imperial audience, and the emperor gives them lands and tides—and thus their very existence in the Chinese perspective—in recognition of P'an Hu's deeds. These multiple versions of the P'an Hu legend betray a state of affairs that existed, in the Chinese imperial ideology if nowhere else, for nearly two millennia. The practice that lay behind this ideology was the policy of dividing the Middle Kingdom's barbarians into two major classes, the "inner" (net) and "outer" (wai) barbarians.46 Often this distinction was geographical: those peoples living on or near the borders of the empire were generally more sinicized, and were thus "inner" in both location and

Chinese Dog-Man Traditions

157

culture vis-â-vis those who were further removed from the center. It was also a reflection of a classification system applied to China itself: under the Former Han, the central provinces of the Middle Kingdom were the "inner" provinces, while those sharing frontiers with barbarian peoples were called "outer" provinces. This system is also explicit in early Chinese cosmologies, a point to which we will return. Before we discuss how these people were different from one another, we will first review what they shared in common according to the Chinese ideology. In general, all barbarians were subhuman creatures, living outside of the bounds of humanity, which corresponded to the frontiers of the empire. As we have seen, the names of some of these peoples corresponded to terms for "half breed" or "mixed race." Furthermore, according to the earliest mythological sources, all were descendants of one or another disobedient ancestor, exiled for his crimes. We find a strong resemblance to Indian ideologies regarding its Svapâkas on both of these points. 47 All barbarians were represented in Chinese characters with an animal (quadruped or insect) "classifier" or radical of one sort or another. The dog radical ch'üan precedes the names of such southern peoples as the Liao, Khih, Ling, Miao, Nao, Yao, Lolo, Li, and Chung. Another canine term, ti9 was used in designating China's northern barbarians from the eighteenth to at least the first century B.C. Because theirs was a realm of chaos and nondefinition, animals, demons, and barbarians were indistinguishable. The Chinese term kuei is most telling in this regard: it means "strange anthropoid simian creature," "people or race of alien origin," "fear, strangeness, large size, cunning," "pupa," and "ghost of the dead." So it was that when princes or generals set out to conquer barbarian lands, their campaigns were considered more as hunting expeditions than as military operations.48 Yet, a distinction was made from a very early time between those barbarians who had submitted to the Chinese and those who refused. One set of terms for the former and latter was "cooked" and "raw." It will be recalled that in certain Chinese traditions of barbarian cynanthropy (that of Hu Chiao, for example), it is the males who are canine and who eat their meat raw, whereas their women are human and cook their food. The men are thus "raw" wildness and the women "cooked" nature. Yet, in certain versions of the P'an Hu myth, it is his human wife who bars his final transformation into a human being when she lifts his bell one day too early.49 The Man were precisely the sort of hybrid that Chinese ideology characterized as "inner" or "cooked." They were people the emperor could rely

158

SEVEN

upon to wage war against his enemies—as part of the i-ifa-i (using barbarians to attack barbarians) policy—in exchange for nominal Chinese honors and tides and a preferred tax or tribute status. A prime example of an outer barbarian people is the Hsiung-nu, against whom the Great Wall was constructed in the third century B.C. In fact, the two are perhaps invoked together in a 177 B.C. report that the Hsiung-nu had invaded (modern) Shaanxi and plundered the Pao-sai Man-i, or "Frontier-guarding barbarians" there. Besides the Man, the Ch'iang—those peoples with whom the Dog Jung came to be identified after their dispersal by the Chou—were also called "frontier-guarding" in Han documents.50 Whatever the case, there were two general trends in relations between the Chinese government and their barbarians throughout history. First, exemption from taxes could quickly be transformed into a levying of taxes, according to the whims of the imperial court. Second, "inner" status often effected a change in the lifestyle of the barbarian peoples: many nomadic herders, indeed even the Hsiung-nu, became agriculturalists. This brings us to the matter of rice. Until the Chinese first moved south of the Yangtze River, in the period of the Three Kingdoms, they knew nothing of rice. The northern staple was wheat, and if the Chinese entry into the south was a political or military conquest, the victory was won at the expense of their agricultural and eating habits. The Chinese learned rice cultivation from the barbarians they encountered in their newly won principalities. We might surmise that the Man were among these peoples, since the early Sou-shen chi version of the P'an Hu myth has the emperor feeding his devoted dog rice when he returns with the head of King Fang.51 It might also be assumed, in the ceremonial exchanges that took place periodically between the Chinese court and its tax-exempt vassals, that rice would have been a product offered to the king in exchange for manufactured goods and finery, rather than the opposite. In this light, let us examine an interpretation offered by Edouard Mestre of the "original" Hou Han shu version of the P'an Hu legend: The animation of agrarian rites went hand in hand not with the development of agriculture, but with the generation of the exchange of precious objects. In ancient times, when these exchanges were the occasion for what were sometimes risky visits, if not expeditions, the barbarians, notably the Man, were truly an integral part of the Chinese world. Their ancestor P'an Hu the dog, a dancing dog ac-

Chinese Dog-Man Traditions

159

cording to one version, is above all else the personification of a tribute offered to the sovereign and redistributed by him in the form of rewards, honorary distinctions and emblems.52 The P'an Hu myth of the southern barbarians, in this perspective, is nothing other than a dramatization of exchanges of gifts made between the imperial capital and a vassal people. The Man side of this transaction—that is, what the Man would have given in return for the rewards, honors, and emblems coming to them from the Chinese—would have been agricultural products, with rice figuring prominendy.53 More than this, the Yao of extreme southern China—Tonkin, Kwangsi, and Kwangtung—refer to their ancestor not as P'an Hu, but as P'an Wang (King P'an). This same tide, "King P'an," is applied to three other mythic figures. One of these is P'an Ku, the cosmic man of Taoist tradition, whose mythology we will discuss shordy. Another is "the dragon of the mountain," of whom P'an Hu would have been the dog—a theme we glimpse in the Hsia-min charter. Finally, this tide is also employed to designate thegod who broughtriceto mankind.54 This identification between dog and rice god is quite explicit in other southern Chinese and southeast Asian traditions. The Lolo of Yunnan (the same region as that chronicled in the Man shu\ for example, in their new year rites commemorate the dog who saved them from starvation by bringing them seed grains down from heaven: "It is the dog who brought us the things that we eat; it is he that we ought to feed first."55 Also, in Hunan, it is to the dog that "first fruit" sacrifices are offered, in the sixth month. 56 There exists, in fact, throughout southern China, a myth of a dog who is one of the domesticated pets of the god of heaven. After a great flood, humanity is extremely poor, and the god of heaven wishes to improve humanity's lot by sending his animals down to the earth with rice. Of all the creatures sent down with the grain, it is only the god's dog who manages to swim over the floodwaters, carrying a grain of rice on its tail so that human culture may start anew.57 This reproduces a "native Chinese" tradition, dating from the Shang dynasty, that connects a dog with the ancient rice god Shang-ti, and a Ch'in and Han period sacrifice called the lei (a term for which the Chinese characters are "dog," "rice," and "head") that involved the offering of dog's flesh and rice, by which a dismembered Shangti was ritually reintegrated and resurrected. Under the Chou, Shang-ti becomes conflated with Tien (Heaven), and is accompanied, in his apotheosis, by his dog, who now becomes the Hound of Heaven, a star in Draco. The Shih chi represents Shang-ti as "Ti-Dog," the ancestor of the

160

SEVEN

Hou Chih and Pai peoples. According to the Chou-li, the two subministers of Justice who raised and chose the sacrificial dogs, and who officiated at Chou-period dog sacrifices, were called the "Dog Officials" (ch'üan-jen We have already alluded to the meanings, on a cosmic level, of the names of emperor P'ing (flat earth-maker) and his enemy Kao (heaven or sun) in Yao versions of the P'an Hu myth. In this final cluster of symbols, we also find themes of a great flood (and thus the waters of chaos), the dog as an envoy to earth from the god of heaven (the "hound of heaven"), and the dog as the founder of riziculture (and of southern, i.e., "barbarian," agriculture), as well as features vaguely reminiscent of Turko-Mongol traditions of divine ^marriage.59 With these themes, we break through to a most ancient layer of canine symbolism, one that is at once Chinese and "barbarian"—if such a division is possible.

E

I

G

H

T

barbarians in a n c i e n t china

1. T H E BARBARIAN FLOODWATERS OF CHAOS

Among the southern peoples of China, there are a good number who continue to commemorate their canine ancestor or culture hero by celebrating their new year as P'an Hu's birthday, by wearing the "canine" puchien hair arrangement or the "dog-tailed" tu-li tunic, and by reenacting P'an Hu's founding adventures in drama and performing dances they associate with him. In fact, such dances go back to the Chin dynasty (A.D. 265419). The dynastic dance of Chin was called the p'an dance: it will be recalled that P'an Hu is referred to as a "dancing dog" in certain sources. This dance included a gesture that evoked turning a platter, surmounted by a bowl, upside-down. A dance performed by the Liao called the Hun-tun dance is also related to this.1 P'an Hu's name means "platter gourd," and from this it is possible to draw several important connections between P'an Hu; Hun-tun, the "chaos egg" or "chaos vessel" (or those lumpy little dumplings in Chinese soup); P'an Ku, the cosmic man and demiurge of Taoism; the flood motif; chaos; and the southern barbarians.2 First, Hun-tun too is a dancer, one, moreover, who resembles a bear-like, long-haired dog: the source that describes Hun-tun as such says that he lives on Mount K'un-lun. The first use of the term K'un-lun was to designate the Jung peoples.3 This is not all. Hun-tun is said to have a son; in most traditions, this son's name is P'an Ku, which is tantalizingly similar to P'an Hu. Moreover, there are several parallels between the P'an Ku, P'an Hu, and Hun-tun themes, all of which originated in south China. Now just how does P'an Hu connect with Huntun? Hun-tun is basically an amorphous and passive lump. In certain Taoist Writings, most notably a passage from Chuang Tzu, there is a myth of Huntun in which two Confucians try to poke him to life. What they in fact do is "bore him to death": 4 161

162

EIGHT

The Emperor of the South was called Shu. The Emperor of the North was called Hu. And the Emperor of the Center was called Hun-tun. Shu and Hu at times came together and met in Hun-tun's territory. Hun-tun treated them very generously. Shu and Hu, then, discussed how they could reciprocate Hun-tun's virtue, saying: "Men all have seven openings in order to see, hear, eat and breathe. He alone doesn't have any. Let us try boring him some." Each day they bored one hole, and on the seventh day, Hun-tun died. For Chuang Tzu, this is certainly a parable on the subject of fundamental differences between Taoists, for whom chaos was an optimum potential or force and a desideratum (the unhewn stone, the uncut cloth, and the Tao itself all precede regimenting creation), and the Confucians, who were meddling busybodies. We will return to this sectarian gloss shordy; here, however, we must discuss the rituals that lay behind this myth. In very early Chinese mythology, we find accounts of King Wu-i of the Shang dynasty who played chess with a doll that represented the god of heaven. When the doll lost the game, the king filled a leather sack with blood, hung it up, and shot it. This was apparendy a ritual practice enacted in ancient China during pivotal times in the year and in connection with lightning. A seventh-century source relates a Miao or Man myth of P'an Hu, from southern Hunan and Hupei, that draws similar themes together: "P'an Hu went hunting and met a 'stone ram' who gored him and threw him into the branches of a tree. His children followed the cawing of crows to find where his corpse lay." This same source notes a funerary rite practiced in this region called "piercing the Great Bear" in which the people reenact the movements that would have been necessary to lower P'an Hu down from the tree in which his body had been caught. Finally, a similar practice is reported among the Wu-ning Man peoples of southern Szechwan in the Tu-yang tsa-tsu, a source from the Tang period. These people, who "pierce the sky with a needle" at funerals, say they laid P'an Hu in a tree when he died and pricked him with a needle.5 Besides his depiction as a chaos vessel, a sack of blood, or a mass of flesh, Hun-tun is also portrayed as an egg, an empty grotto-cave, a split- or double-tree, the land of the dead, a meditation chamber, a vase, and a cocoon. All of these forms are interrelated: we will concentrate on the last, the cocoon, in its relationship to the P'an Hu mythology. In a passage from the third-century Wei Lüeh, cited in the commentary to the Hou Han shu P'an Hu myth, we find the earliest account of P'an Hu's birth:

Barbarians in Ancient China

163

Kao-hsin had an old woman who lived in his imperial palace. She contracted a sickness of the ear, and by piercing into it something was taken out that was as large as a cocoon. The lady in waiting placed it in a gourd [hu] and covered it with a platter [p'an]. After a while it was transformed into a dog with five-color hairs. And so it was named P'an Hu, "Platter Gourd."6 In this short narration, we encounter not only P'an Hu's first association with Kao-hsin and with one of the symbols for the piercing of Hun-tun, but we are also offered an etymology for his name and a prefiguration of the bell under which the dog P'an Hu is placed in order that he become a human, or, as it turns out, a cynocephalic. Most striking is the term used for cocoon in this account: while the common Chinese term is chien, here the cocoon inside the old woman's head is called hun-tun7 P'an Hu is thus the "son" of Hun-tun. Not only is he born out of chaos, but in this case, the location of chaos may also be construed as a chaos: in Taoist thought, the body is conceived as a double-chambered system that at once corresponds to a two-headed distilling drum used in alchemy and the "double" K'unlun mountain that has the form of two gourds (hun-tun) whose mouths are joined together. The inside of the old woman's head would, in this context, be identified with the inside of a chaos gourd. This theme is further elaborated in a variant account of the birth of P'an Hu's children by Kao-hsin's daughter. In the Kuang i chu, it is said that after P'an Hu had won her, "the princess gave birth to seven lumps of flesh. On cutting them open, there were seven males."8 This same source describes the birth of P'an Hu as follows:9 In the time of Kao-hsin, in a man's family a dog was born. Its master was shocked at it and abandoned it in the ditch by the road. After seven days, it was not dead. The animals suckled it. Its form grew bigger and bigger day by day. Its master took it back. At the time when he first abandoned it below the roadside, he had contained it in a plate and covered it with leaves. So regarding it as auspicious, he then offered it to the Emperor; and it was called P'an Hu [PlatterGourd]. So it would appear that the earliest origin myths of P'an Hu would have him primordially made of the same stuff as Huh-tun, the original chaos vessel. Another important chaos motif in this cycle is that of a great flood out of which a mass of solid ground—sometimes in the form of an egg—

164

EIGHT

eventually appears. We have already encountered one example of this theme in the context of the dog who brings rice to mankind from heaven, swimming over the waters of the flood with a grain of rice on the tip of his tail. A variant of this myth, one which brings P'an Hu into play, is also one of a great number of southern myths in which a brother and sister float, often in a gourd, over the waters of a great flood to become the ancestors of the human race. Here, P'an Hu is saved from the chaotic deluge by mounting a platter supported by a huge gourd. After this he becomes the ancestor of newly created mankind.10 It should also be recalled that in at least one version of the P'an Hu myth, the dog ancestor floats his princess out across the sea to an island where she bears and raises the children of their union. In this and most of his myths, P'an Hu's children intermarry to give rise to a particular people, or to repopulate the entire earth. We ought not to discount possible ties between this theme and the accounts of Amazons and their cynocephalic or canine mates who lived on the shore of or across the sea.11 Floodwaters are also associated with barbarian peoples who wash up against, and sometimes over, the bastions of the civilized world. We have already seen this in Indian and European depictions of barbarians as swarming hordes. This is portrayed quite literally in several Chinese sources. While it may be argued that these are Confucian glosses on earlier myths, it is still a recurrent theme from the earliest mythology, and was therefore not invented out of whole cloth by later commentators. In the words of Norman Girardot, "the so-called 'southern' cultures . . . are crucial in the development of Chinese civilization. Furthermore, such 'foreign' cultural elements may have also played a role in the evolution of the earliest Chinese neolithic and Shang traditions. It is sufficient to emphasize that if the hun-tun theme possesses 'le caractere meridionale,' this does not mean that it is foreign or peripheral to the mainstream of Chinese tradition."12 In a general sense, then, floodwaters are barbarians and civil engineers are Confucian rulers in the mythology. The latter motif is particularly well represented by the figure of Yü the Great, whom we will discuss shordy. As for the former motif, it is built into Chinese cosmology: surrounding the earth, the square top of a truncated pyramid, are the "four seas" (ssu-haî), which are equated with the barbarians of the four directions.13 The Chinese knew of traditions of southern cultural heroes, and they presented certain problems: if these individuals ordered the chaotic barbarians but were themselves barbarian, were they part of the flood or were they dam-

Barbarians in Ancient China

165

builders? Hun-tun and several of his mythological cognates for chaos are portrayed as rebel barbarians or monsters themselves. P'an Hu plays the role of founder of a vassal barbarian people: he floats over the floodwaters. Another such southern barbarian hero was the founder of the kingdom of pu-nan in Indochina. He is named Hun-hui (Troubled Waters-Overflowed River) or Hun rien (Troubled Waters-Kingdom).14 We have already discussed the cluster of Chinese terms that derive from the root hun: troubled waters, chaos vessel, two-tiered mountain, barbarian people, half-breed, and P'an Hu's last name. We also know that p'an, the first name of the dog ancestor of the Man and Yao, means "platter." He shares this name with another chaos creature, and another son of Hun-tun: P'an Ku. P'an Ku's last name is a mystery. It has been suggested that it is a deformation of the name P'an Hu, a deformation that, according to Mestre, was an intentional attempt by the Yao (or the Liao) to distance themselves from a non-human ancestor. However, according to David Yu, the Yao and Miao in fact call P'an Ku "P'an Hu." 15 It is generally agreed that P'an Ku's origins, in textual traditions at any rate, are slighdy more recent than those of P'an Hu. Moreover, the earliest P'an Ku cultic centers were located in Kwangsi province and on the east coast of China, the same regions as those inhabited by the Man and Yao in the time of the Three Kingdoms. It has also been argued that the origins of the P'an Ku myth are to be sought to the west of the southern Man or Yao peoples, in the region of the Pa culture. The Pa, in addition to having ties with the Yao, were also connected to the Jung—that other ancient non-Chinese people said to be descended from a dog, one of whose names was related to a very early use of the term K'un-lun. 16 P'an Ku's birth from Hun-tun is described in much the same way as was that of P'an Hu: 1 7 Heaven and Earth were in the chaos condition [hun-tun] like a chicken's egg, within which was born P'an Ku. After 18,000 years, when Heaven and Earth were separated, the pure yany formed the Heaven and the murky yin formed the Earth. P'an Ku stood between them. His body transformed nine times daily while his head supported the Heaven and his feet stabilized the Earth. Each day Heaven increased ten feet in height and Earth daily increased ten feet in thickness. P'an Ku who was between them daily increased ten feet in size. After another 18,000 years this is how Heaven and Earth came to be separated by their present distance of90,000 li.

166

EIGHT

The sixth century Shu-i chi in turn describes how the world in its diverse parts arose from P'an Ku's body when it disintegrated and was dispersed His eyes became the sun and moon, his four limbs the four quarters, his teeth and bones the metals and stones, his marrow gold and precious stones, and the parasites on his body, impregnated by the wind, became human beings. In a Pang period source, the Kuang Hung-ming-chi, the cosmic giant is no longer P'an Ku, but Lao Tzu, the founder of Taoism: in this source, however, humans are said to arise from his genitals.18 This identification with Lao Tzu insured the demiurge P'an Ku a perennial place in Chinese mythology. His myth is, in fact, highly Taoist in its theology. It is the chaos that precedes the moment of creation, the unformed mass, that has the greatest creative potential. Differentiated creation arises when the body of this chaos creature disintegrates and atrophies—a very different sort of creation than that of the Confucians. In this perspective, too, the floating of P'an Hu over the ocean in a special boat recalls a deluge theme of returning to chaos, but this chaos is nothing other than the primordial wholeness of the Tao, the "something that was chaotic yet complete which existed before the creation of heaven and earth" (Tao te ching 2 5 . 1 ) . Huntun, P'an Hu, and P'an Ku are in fact stages of beginning and return in a Taoist cosmology that is characterized by an alternation between differentiated creation and undifferentiated chaos. Yet, while the mythical links for the many motifs we have discussed in this section were probably first forged in an explicit way by the Taoists, they nevertheless interlock with a great number of similar "Confucian" themes.19 In order to discuss these common points between Taoist and Confucian elements in, or glosses of, cosmogonic myth, we must investigate the mythic and mythological origins of Hun-tun, the father of P'an Hu the dog and P'an Ku the giant. 2 . CONFUCIAN VERSIONS OF CHAOS

We have already indicated, on several occasions, that Chinese ideology threw its barbarians together with the elements of floods, chaos, and monsters, all of which it viewed with a jaundiced eye. This ideology, which governed institutional attitudes towards the "outer limits," might be characterized as Confucian, while speculation on the structure of chaos was a properly Taoist concern. Thus, when the Confucians discussed creation and the ordering of the universe, they were talking about the origins and institutional structure of their China, a structure of "radiating polity" (to borrow Stanley Tambiah's term) centered on the imperial court. The Tao-

Barbarians in Ancient China

167

ts on the other hand, antiestablishment to the core, dwelt on the rimordial Tao that could not be named as a means to recovering, on a microcosmic level and without the interfering apparatus of society and state, a primal and salvific unity. In the preceding section of this chapter, we alternated in our discussion between Taoist and Confucian interpretations of early creation myths. In the last chapter, we were mainly concerned with mvthologized relations between barbarian populations and a Chinese state that was almost by definition Confucian. In this section, we will discuss the earliest Chinese mythological sources along with their primary (Confucian) euhemeristic or historiographic interpretations, and in so doing glimpse the origins of the "barbarian ideology" even as it grew out of a preConfucian mythic nexus. Norman Girardot has identified six strata of Chinese cosmogonic myth. These are: (1) the precosmic condition of chaos (hun-tun), most frequendy symbolized by the image of the cosmic egg or gourd; (2) the splitting of the cosmic egg and the birth of the cosmic giant (P'an Ku), who gives form to the prehuman cosmos; (3) cataclysmic disasters following the creation of man, including a great flood; (4) a reestablishment of order, with human society in a precivilized barbarian state; and (5) a first and (6) a second period of civilizing heroes.20 The Taoists, as we have seen, were most interested in what happened before the institutionalization of eternity into time and imperial history. The Confucians, on the other hand,

while intensively searching the past for models, never went beyond the bounds of historicity which coincides with the rise of (civilized) human society. The time of the sage kings constitutes an absolute limit for them. It was they who had taught men to distinguish themselves from "four-legged creatures and birds" . . . The idea of turning one's back on this world, either temporally by an attempt to return to the period predating the inception of civilization, or spatially by an escape into the wilderness, not only seemed barbaric to them but was simply unimaginable.21 In a word, the Confucians, with their rites and etiquette, their mandarins and sage kings, were fuddy-duddies. But there was method to their fuddyduddiness, and this was the euhemeristic method. By making historical figures, generally sage kings, out of the improbable monsters and other creatures who populated the ancient mythology, the Confucian scholars

168

EIGHT

could teach, through their annals, models of government. (Of course, the same ideology made a reverse procedure de rigueur for the Chinese authors when they wrote about their barbarians: because these were "animals,55 the human founders of their cultures and societies had to be mythologized into subhuman creatures like P'an Hu). The most illustrative case in point here is the use made of the Chou period (approximately fourth century B.C.) mythology of the Shu Ching, as well as of the slighdy later Shan-hai ching, by the father of Chinese historiography, Ssu-ma Ch5ien (who died in approximately 80 B.C.). A few words about the man and his times, as expressed in his writing, are in order here. Ssu-ma Ch5ien wrote his Shih chi at a time when the Hsiung-nu were at the foot of the Great Wall that had been erected to shut them out of China. He himself lived in the Lungmen area (modern Shansi province), quite close to the Wall. His preoccupation with these dangerous barbarians is perhaps echoed in his "scientific55 discussion of them—one of Ssu-ma Ch5ien5s many "firsts55 in Chinese literature—in chapter 110 of his work. This was in fact one of six special monographs that he included in his Memoirs.22 The Lungmen area was also the locus of a great number of ancient mythic traditions of sage kings and emperors who founded Chinese civilization. Ssu-ma Ch5ien handled these with perspicacity, taking care to avoid what he considered unworthy of historiography. So he says in his work, "as for the marvels found in the Yü pen chi and the Shan-hai ching, I dare not recount them at all.5523 The other great extant source of ancient mythology, the Shu Ching, providentially survived the book burnings of the Ch5in, with its surviving portions being reassembled in about the time of Ssu-ma Ch'ien. The hermeneutical approach employed by Ssu-ma Ch5ien and others in compiling a "critical edition55 of this incomplete work is very revealing of the reigning world view.24 China and the universe were a single harmonious order that, in the writings of the early historiographers and their followers, would extend even to imperial succession. Theirs was thus a symbolic chronology that was above all else attentive to the "quint-essential55 nature of the universe, in which everything was reducible to parallel series of five. Corresponding to thefiveelements were, for example, the four gates of the imperial palace with the palace at the center, thefivedirections,fivecolors, five musical tones, five seasons, five temperaments, five animals, etc.25 Therefore, the annalists did not so much deny their mythic heritage as fit it to their pre-existing and highly elaborate world system. In this way they

Barbarians in Ancient China

169

v^ere quite like the pre-Socratics: they could no longer conscience the bizarre gods that had been bequeathed to them, but were moved to see higher principles in them. However, whereas Plato continued to employ mythological genres to discuss these principles, the Chinese transformed them into emperors. When these principles are translated into spatial and vectoral constructs, w e have the ordering of the Chinese cosmos. In its most fundamental form, the universe has the form of an imperial chariot or dwelling, with a square floor (earth) and a round domed roof (heaven) overhanging the base and supported by pillars at its four corners.26 Heaven is composed of nine vertical levels. In the beginning, and according to the harmonious movement of the primordial fivefold elements, this universe was a well-ordered dance, choreographed by enlightened sage kings and their hierarchized functionaries. But an act of disobedience threw the elements out of their proper order, and the entire system was rocked by a catastrophe. Not only did this catastrophe change the order of the universe, but it also gave rise to China's barbarians and other dread beings too numerous to mention. This event fits into the Taoist model of a periodic return to a chaos state. It is the third of Girardot's six strata, the myth of the flood provoked by the wrathful Kungkung.27 In the beginning, Kung-kung is a close relative of the five original sage kings who founded and began to order the earth. Kung-kung is, however, not wise, and this is his and the earth's fall. Although fire logically succeeds wood in the order of thefiveelements, Kung-kung, whose element is water (his tide is "Superintendant of Public Works"), forces his way into power upon the demise of Nü-kua, a goddess identified with wood. Kung-kung's illegitimate rule results in an excess of the watery element. The earth is completely flooded, and there is danger that the floodwaters of earth will rise up to the upper waters of heaven, thus confounding all and throwing the universe into a state of total chaos. Of the four mountain pillars that support heaven, only the northwestern peak, Pu-chou, is singled out in the Shu Ching for particular attention. Kung-kung places himself atop this peak to do battle with the hero and w ind god Chuan-hsü, who has been sent to attack him. Chuan-hsü throws Kung-kung down from the peak, and when the monster falls, one of his horns smashes the base of the mountain, which collapses. With the shatterln g of this pillar, the planes of heaven and earth, previously parallel, become skewed; now that the Pole Star has slipped from the center of the sky, heav-

170

EIGHT

en tilts down towards the northwest and earth towards the southeast, and the heavenly bodies flow from east to west while the rivers of the earth flow from west to east. Strangely, it is Kung-kung's own son who, in certain traditions, repairs the damage the recalcitrant monster has caused. Kou-long reorders the earth and reinstitutes agriculture: he is rewarded by being named Hou-t'u, Sovereign Earth, and Ta-shi, Great Lord of the Empire. Kung-kung had also been an ancient lord of the earth and soil: his origins may in fact be found in the runoff basins of houses in ancient China, which were situated in the northwest quarter. These drained to the subterranean waters that were associated both with the waters of the four barbarian oceans and the land of the dead. 28 In other variants of this myth, the name of the "civil engineer" who rights Kung-kung's wrongs changes. A classic variant is Yü the Great, who, like Kung-kung has the tide of "Superintendant of Public Works" and who, like Kung-kung's son, is also the son of a monster: Yü's father is H'uan-t'ou or Kun, the monster of the southern direction. Through the acts of this cultural hero, the floodwaters of chaos are drained and dammed, a feat that is equated with the exiling of illegitimate monsters beyond the bounds of the Middle Kingdom's nine provinces (which correspond to the nine levels of heaven).29 Kung-kung's exile is only one of a group of four: to the four directions there correspond four disobedient monsters whose exiles constituted the origins of the barbarians of each direction. According to the Shih chi, Kung-kung is sent away to the north, the land of the northern Ti barbarians. The other incorrigible monsters are Huan t'ou, whose southern exile corresponds to the southern Man barbarians; San-miao, whose western exile corresponds to the western Jung barbarians; and Kun, whose eastern exile corresponds to the eastern I barbarians.30 Of course, there are variations of this ordering. In the earlier Shu Ching, the barbarians of the four directions are the Yu to the east, the Nan-chiao to the south, the Liu-ku to the west, and the Yu-tu to the north. San-miao, the monster of the west in the Shih chi, becomes the name of a particularly difficult barbarian people: the San-miao san-yao are "winged men" whose sinister behavior makes it necessary to break the channel between earth and heaven. They are exiled to the west, to a place near the K'un-lun mountains, according to the Shu Ching; in the Shih chiy it is said that they are m o v e d from the south to the west for repeated insubordination. Hun-tun, the chaos vessel, is said to be one of the San-miao; but elsewhere he is the son

Barbarians in Ancient China

171

0 fTi-hung

or of Huang-ti, the "Yellow Emperor" who founded human civilization. Finally, he is identified with Huan-t'ou, the monster exiled to the south. In this last case, he is exiled to Pa (Szechwan) "as a protection against evil spirits." The Jung of the west are identified with the Hsiung-nu in the time of the Chou dynasty: in the Shang period, their Mongolian homelands are called Kuei-feng, "the region of demons."31 In these variations, one sees a series of attempts to square a fourfold ideological schema with a changing political map of China and its neighbors: the movements of the San-miao are particularly telling. In spite of changes in content or location, structure remains constant throughout. So, too, does ideology, in this case that regarding exiled criminal monsters, as is readily apparent in a Shih chi account of Yü the Great's many acts of ordering the human universe. Yü is urged by the emperor Shun to punish "the Man and I who trouble our fair land—[who are] thieves and assassins, enemies and rebels . . . [with] the five punishments." These five are various forms of capital punishment; they may, however, be commuted to exile from the capital, with three concentric zones of exile delineated according to the seriousness of the crime. Later commentators on this passage saw in it the implication that barbarians, apart from persons who infringed the civil laws, were also considered to be, a priori, criminal enemies of the state. 32 In this way, draining swamps and exiling criminal monsters and human criminal barbarians become the same act. The walls are made strong and the center is held. 3. THE DOG IN ANCIENT CHINA

Nevertheless, Chinese treatments of outsiders often echo domestic policies as well as Chinese religious and cultural themes, and in this way make them meaningful to the Chinese themselves. For example, we saw how in ancient and medieval Indo-European traditions, the werewolf was a criminal whose punishment was exile, and we saw how this individual exile could be universalized for an entire disobedient people, as in the Old Testament traditions of Cain's offspring. A parallel in China may be found in a Li chi (Book of Rites, 100 B.C. to A.D. 200) injunction concerning a royal counselor who is unable to share the unanimous opinions of his colleagues. The only solution in this case is self-exile from the kingdom. He breaks all connections with his country and his ancestors:33 Once he has crossed over the border (of his country), he smoothes out a piece of ground and piles up a mound there. He turns his face towards his country and cries out his lam-

172

EIGHT

entations . . . He wears . . . (mourning clothes) . . . slippers of untreated leather, the axle of his wagon is covered with the hide of a white dog . . . He ceases to cut his nails, beard or hair. When he takes his meals, he abstains from offering any libation (he has broken his contacts with the gods) . . . After three months, he may put on his ordinary clothing again. The dog was, for the Chinese as well as for some of their northern and western barbarian neighbors, an animal that bore strong associations with death. The Shang emperors may have been buried together with their watchdogs; dog figurines were buried with Han emperors, probably to protect them, as watchdogs, after death. Dogs were also buried under palace foundations, in Shang era building sacrifices.34 The absence of such figurines between Shang and Han times may be attributed to an interim use of straw dogs that would have been buried in the place of the animals themselves.35 This use of straw dogs, which remains current among certain of China's northern peoples, has an ancient pedigree. The classic text concerning this practice is of course the Tao tc ching, in which the wise man is advised to regard human society as so many straw dogs. 36 The use of strawfigurinesis extremely ancient in China; the use of straw dogs, is described explicidy in the second-century B.C. Huai-nan-tzu (Work of Prince Huai-nan): "The straw dogs and the clay dragons are painted blue and yellow, wrapped in various silks, and tied with red string . . . straw dogs [used] against epidemics, during their time, are kings.55 The first-century A.D. Lun-heng says straw dogs were used at funerals, a practice that remains current among the Solon of Manchuria. In these contexts, the dog is a guide to the dead in their peregrinations to and through the underworld. This belief was also current among the Ch5i-tan, Tungus, and later Wu-huan, for whom the dog led the dead to the Red Mountain. A parallel practice, among the Jurchen, was the impaling of a white dog upon a long pole: the dog served as an intermediary between heaven and earth, as a sacrifice to Heaven, and as a means to averting epidemics. This last element brings us back to straw dogs, which were also employed in a rite described in the fourteenthcentury A.D. Yüan-shih: during the second decad of the first moon, human and canine figurines are made from straw. Multicolored threads represent their intestines. Arrows are shot until the dog is totally destroyed. An offering of wine is made. The participants take off their clothes and have them propitiated by a shaman to drive away all unhappiness.37 Shooting at straw dogs also evokes the shooting of sacks of blood, and

Barbarians in Ancient China 0f

173

p'an Hu placed in a tree and pierced with needles after his death, as well as theriteknown as "piercing the Great Bear." These are further related to a Chinese myth that has a "southern" hero/god named "I" shoot nine disob e d i e n t suns from the sky with arrows. This I becomes, in more recent times, conflated with a god named Erh-lang. Erh-lang, a god originating from Szechwan, is accompanied by a dog in his fights against demons. However, his dog is in fact the same creature as the Hound of Heaven çpien kou\ a hound that is the target of arrows shot into the air at the birth of children in the Amdo border region of northeastern Tibet: this custom, w h i c h originated in China, was observed as a means of countering the nefarious effects of this dog on newborn and young infants.38 The hound of heaven may have its origins in the Shang rice god Shangti, who, when he came to be identified with Heaven (Tien) by the Chou, was transformed into two distinct beings, a celestial emperor and his canine companion.39 This hound of heaven, depicted as dwelling beneath the (raised) door to the celestial palace of Heaven, has been identified with a star in the tail of the constellation of the Dragon. 40 The star Sirius, the dog-star of Western tradition, is called the Celestial Wolf (Tien-lang) in China. A common Chinese theme describes three stars in Canis Major and one star from Argo as an arrow pointed at the Celestial Wolf. There is no direct evidence to link this case of shooting the wolf of heaven with the shooting of various mythic or sacrificial dogs. The hound and wolf of heaven are brought together, however, in their function: both guard the celestial palace of Shang-ti, the constellation of Ursa Major, itself a land of the heroic dead in ancient Chinese thought. The guardian is called the hound of heaven in the Shih chi; it is called the wolf of heaven in the Chao hun. In this latter source, "a wolf with piercing eyes comes and goes and slowly throws men into the air and plays ball with them. He tosses them into a deep abyss, in obedience to his Lord, and then he may go to sleep."41 The hound of heaven was more often a shooting star or comet, a harbinger of pestilence and war and a bringer of suffering to the people. It was also identified with cruel representatives of governmental authority sent to prey on the common people. Government agents are reported in the Vang shu as having dressed in dog skins on two occasions (in the seventh and eighth centuries A.D.) to tear the hearts and livers out of people as offerings to the dog of heaven. These minions of cruel officials were termed ck'engcfreng, who are said, in an eighth-century source, to be "Hairy Demons" or "Hairy Men," originating from Hunan. 42 Such human sacrifices to a dog were counterbalanced by dog sacrifices

174

EIGHT

for the benefit of humans. One such sacrifice was the ju ritual, instituted in 676 B.C. to drive away, through a canine intermediary, the evil effects of the dog days on humans. In this sacrifice, dogs were torn apart at the gates of the city, thus serving as surrogates for humans who would otherwise be struck down by the ills attendant to the heat of midsummer.43 The fu ceremony was perhaps an elaboration of one of the four lei sacrifices performed in Chou, if not Shang times. All of these sacrifices involved the dismemberment of dogs as offerings to the rice god Shang-ti. The four winds are invoked in certain of these sacrifices, one of which entails driving a carriage over a dog prior to setting out on a journey. Both of these are indications that the dog in China, as in India, was associated with movement and the atmospheric region. In all of these cases, these were apotropaic sacrifices, serving to appease gods and ward off demons, disease, evil, and pestilence.44 4 . HOUSE, TOMB, AND UNIVERSE IN ANCIENT CHINA

There is a connection between dogs sacrificed at the gates of the city and the dog of heaven who, located beneath the palace of heaven, like a watchdog beneath the door of a house, was at times the object of human sacrifices.45 This connection may be found in the ideology underlying ancient Chinese architecture—exemplified in the construction of houses, tombs, and ceremonial edifices—as such ideology informed the shape of the universe. As we have shown, the ancient Chinese generally ordered the inhabited earth into concentric zones. Surrounded by a square earth beneath a round heaven, this inhabited space was organized into a series of Chinese boxes, at the center of which stood a gnomon, a central tree, around which the capital was constructed, with gates at each of the four cardinal points. We have already seen that the place of the barbarian peoples was at the periphery of this system. As such, they were located in time and space, but their time was the primordial time that preceded the civilizing heroes, and their space lay beyond the limits of the world regulated by Heaven (outside of the THen-hsia). At the center of this universe was the palace of the king; the mythic parallel of this palace was a cosmic tree that the ancient rulers could use to mount to heaven and descend to earth. From the beginning of the Han period onwards, the symbolism of the king as the axis of the world was made explicit through the ritual construction of the ming-fang, an edifice that symbolically reproduced the universe at the center of the king's palace.46

Barbarians in Ancient China

175

The ming-fang and the ordered universe may have originally been modeled after archaic Chinese dwellings. There were three models for the archaic dwelling, first mentioned in the eighth-century B.C. Shih-ching (Book of Songs). These were (1) a sort of pumpkin or gourdlike dwelling that drew on the same cosmological model as the hun-tun "chaos gourd"; (2) raised tumuli or houses on pilings or stilts called "nests"; and (3) subterranean abodes, perhaps in the form of potters' ovens, called "pits" or "holes." Such dwellings had passed out of use by the time of the Han. The classic Han house had a central courtyard surrounded by four buildings opening onto it, with but one door opening to the exterior. The ensemble was therefore a system of nine (3 x 3) "squares," a system that mirrored the nine original provinces of China, the arrangement of Shang tombs, the Han ming-fang, and of the Confucian "well-field" (chHng-tien) system.47 While these dwellings retained elements of the ancient symbolism, the conscious prototypes the Chinese used for the raised nests and subterranean holes were the dwellings of their barbarian neighbors. The former were the dwellings of the T'ai and Tibetan peoples, whose high towers are referred to in Chinese sources from the seventh century onwards; the latter were first drawn from the dwellings of the Tungus and Manchurians, whose multistoried cave or hollowed-out hillside abodes are mentioned in the Hou Han shu.48 Whether the barbarians' constructions constitute anachronistic survivals of an ancient Chinese model or are the products of their own genius is a moot question. Nonetheless, it is clear that the Chinese were using a barbarian model to construct—to interpret—a modality of their own origins. In fact, the Shih-ching passage to which we have referred locates these constructions in proto-Chou times (1796-1327 B.C.), in an area corresponding to modern Shaanxi province. The close neighbors of these proto-Chou would have been the barbarian Ch'iang, from whom the rulers of Chou would have primordially descended. Near the Ch'iang of this time were another barbarian people who would eventually be assimilated into the Ch'iang after the rise of Han: the Dog Jung. 49 The Chinese also knew of Man nests and holes, which are referred to at the end of the Man shu version of the myth of P'an Hu. After relating how P5an Hu carried his new wife into the high mountains and fathered six sons and six daughters by her, it goes on to say that the twelve children paired off with each other and mated . . . When the Emperor bestowed on them the Southern mountains, they

176

EIGHT

proceeded to raise their dwellings on high "lofts" [i.e., proppiled houses] and stop there. Their descendants multiplied and spread, so that they formed a kingdom of their own. 50 The raised and subterranean ancient dwellings were ideally multi-level affairs. If the house or tumulus was raised on stilts, it ideally had nine stories; if it was hollowed out of the earth or into a hillside, it ideally had nine subterranean levels. In both cases, we find the number nine, the number of the levels of heaven in ancient Chinese cosmology. Combined, these two sorts of dwellings symbolized the mythic hourglass-shaped double mountain, Mount K'un-lun, and here once again we see architecture informing cosmology. In the Shib-cbing, the term k'un-lun is employed for a threestoried subterranean or hillside house. It will be recalled here that K'un-lun is a double mountain, in the form of two gourds whose "mouths" are joined at the level of the earth: here, we see that Mount K'un-lun was modeled after a structure that combined the raised tumulus with the subterranean hole. And, of course, this double imagery finds its microcosmic parallel in the human head and torso, the seats of the two souls in Chinese physiology.51 It is the vertical axis of these constructions that most concerns us here. Generally, access to the levels of these dwellings was through a central hole threaded by a tree trunk, in the fashion of North American firehouses of a bygone day. In the case of the subterranean dwellings, the roof hole was the door of the house; for the raised structures, the roof hole opened onto the sky In both cases, however, the pivot of the structure was the central pillar. This was identified with the cosmic tree by which shamans and ancient rulers were able to ascend to Heaven or descend to the underworld. At the top of this world tree were the principal heavenly bodies—the sun or the Pole Star. At the base was the central drain, the hog wallow, the latrine, and the door to the underworld. Important to this schema is the archer I who, it will be recalled, shot nine disobedient suns from the sky: he is also credited with having killed a man-eating monster named Ya-yu. In the religious organization of the house, Ya-yu becomes the guardian of the household sewer at the base of the central pillar, while I becomes the guardian of the center of the house itself, and is associated with the divine Director of Destiny, Ssu-ming.52 As we have shown, the dog had associations with both cosmic antipodes—the Pole Star and the land of the dead—in Chinese mythology. On the one hand, the Hound of Heaven sat beneath the door to the palace

Barbarians in Ancient China

177

of heaven as its watchdog; on the other, dogs guided the dead from this world to the world of the dead. Both of these elements come together, in a somewhat confused way, in the organization of the house. The connection is curiously made via the northwest quarter. This was the direction of the cosmic pillar of Pu-chou that was smashed by Kung(oing when that disobedient monster was thrown down from above. It will be recalled that Kung-kung's damage—of nearly flooding the earth with the waters of chaos—was undone by his son, Kou-long, who became the god of the soil. We should also bear in mind the fact that Kung-kung and other disobedient monsters were exiled for their crimes, and that barbarians were perceived as living in perpetual exile. The northwestern corner of the Chinese house was called the chong-liu, and served as a gutter by which water ran off the house roof. 53 As such, its identification with the smashed northwestern mountain, towards which the heavens tilted, is apparent. This corner was moreover associated with the runoff basin. The basin was thus the "place" of Kung-kung (or one of his minions) in the household cosmos; it was moreover associated with the underground land of the dead. To further complicate this picture, the northwest was also the direction of Mount K'un-lun, itself closely associated with the primordial Dog Jung. 54 It would appear that the northwestern chong-liu later came to be identified with the center of the house: this simplified schema is already in place in the Li chi. Moreover, the guardian of the chong-liu, of the center of the house, was also identified with the god of the soil, the center of the social life of the village and country. This is not all. At the base of this microcosmic center was the central drain that was at once the hog wallow and household latrine, around which the people lived in ancient times. The central sewer, at the base of the central pillar, then, was the door to chaos and the land of the dead, as well as a hun-tun. This in turn "lined up" with the Pole Star (called Wen-ch'ang from the Han period onwards), the roof hole, the hearth, the Director of Destiny, and the god of the soil.55 We may now return to the place of the dog in all of this. The ming-fang of the Han period schematically reproduced the construction of the great Shang royal tombs, which were themselves constructed as dwellings for the dead. These tombs, of which one type was constructed in the shape of a cross, consisted of a central pit with four or eight holes located at the cardinal points. These holes may be compared to the pillars of the Chinese house, with the central hole corresponding to the central pillar in its multileveled symbolism. In the central hole of one such cross-shaped Shang tomb was found a dog that had been sacrificed and buried there. This corre-

178

EIGHT

sponds to the classical Chinese practice of using the dog as the sacrificiaj animal of the principal door or gate (men) of the house or city. Yet, in fact the watchdog is kept either beneath the door of a raised house or attached to the central pillar of subterrean dwellings. These Chinese elements are reproduced among peoples of Manchuria, Siberia, Tibet, and the southern "barbarians" from at least the Han period down to the present. In this perspective, then, the image of P'an Hu suspended in a tree, and the wolf or dog of heaven keeping watch below the entrance to the palace of Heaven, the dogs sacrificed at the gates of the city, the Ch'i-tan burial of a dog's skull beneath a tent, and so many of the other canine motifs we have reviewed all take their origins in the architecture of the non- or proto-Chinese nest or hole house which itself was the model for much of both cosmos and chaos in the Chinese universe.56 With this, we have exhausted the Chinese mythology of the dog and its Dog-Men. Yet it is in this tradition, more than in those of Europe or India, that we find ourselves truly at sea with regard to the separation of civilization and barbarian peoples. If we began this section of our study with the impression that all barbarians were one in the ancient and medieval world, (that is, that it was the same central Asian peoples who were accountable for all the monster mythology of the civilized world), we leave it without a clear notion of where the barbarians leave off and civilization begins. In the end, the scrupulous and scribalist Chinese seem to betray their own chaotic and barbarian origins. This interpenetration occurs on a great many levels and throughout Chinese historiography and mythology. There is as much of chaos as there is of cosmos to China itself, and many of its cosmogonic motifs may have been borrowed from its southern barbarians. The Shang tombs are not unlike the barbarian or proto-Chinese raised or subterranean houses. The ancestor of the Chou dynasty was a Ch'iang, a cousin of the Jung peoples; and the founder of the Ch'in dynasty which succeeded Chou was a native of a Jung area. The Chinese give their southern and western barbarians the demeaning name of Dog-Men; yet Shang-ti, the ancient Shang rice god, later elevated to become the god Heaven (Tien) by the Chou, may have originally been a Tai dog. Finally, even in the case of the barbarians, there is little certainty about who was who and who received which traditions from whom. Our "historical" description of the interpenetrating Man and Yao traditions, reached through induction, has certain problems; and the relations between these "southern" peoples and the "western" Jung is a possibility that we have hardly broached.

Barbarians in Ancient China

179

In the words of Wolfram Eberhard, whose approach we have followed here, there was a period in which no Chinese culture existed, in the sense of a culture belonging to a specific group of people different from those living around them and identified by cultural traits peculiar only to them. Originally there were in the geographical area of China a number of Local Cultures . . . In the course of time, parts of early Local Cultures were fused to what then became the high-Chinese culture. Other parts, further away from the points of fusion and friction, remained Local Cultures . . . The process of fusion, which probably began before 2000 B.C., led after several centuries to the emergence of the high-Chinese culture, which has continued down to the present day.57 Perhaps what the Chinese material teaches us then, above all else, is the interconnectedness of all that its elite tradition has so carefully attempted to classify, separate, and hierarchize. Not only are China's barbarians constandy interacting with one another, but China is also constandy interacting with them. Yet, for all this, China's barbarian chaos is clearly constituted by a great number of distinct peoples that cannot be conveniendy decanted into a single chaotic canine group, be they called Hsiungnu, Tartar, Turk, Man, or anything else. Which, in a way, brings us full circle in our considerations, if circles are what we have been going around in.

N

I

N

E

f a c i n g up t o o t h e r p e o p l e

we could be invaded at any moment by a twofold flood of Barbarians, those from without and those from within

Marcel Proust, A U Ombre desjeunesfillesen fleurs 1. REDUCTIONISM AND INTERNAL CONTRADICTIONS

In the Chinese material discussed in the last chapter, we found traces of a primordial unity of inside and outside, of human and barbarian, upon which were superimposed multiple layers of "legal fictions" that transformed unity into polarity. The original sage kings and their clan lineages were a single, harmonious social unit until a willfiil act of disobedience shattered that order. For the perpetrator of the crime—in this case Kungkung, who attempted to impose the element water in a cycle that called for fire—the punishment was exile to the confines of the world. His mythological exile constituted the origin of an equally "criminal" barbarian race.1 The same theme, of exile as punishment for disobedience, recurs throughout all the mythologies we have reviewed. In India, the fifty disobedient sons of Visvâmitra are sent to the northern ends of the earth for having refused to accept the adopted Sunahsepa as their senior.2 The ksatriya races who have turned away from the righteous path of Hinduism are soundly defeated by Sagara and reduced, like Visvâmitra's elder sons, to the status of outcaste Dog-Cookers.3 In the Christian West, the primal theme of Adam's disobedience, fall from grace, and exile from Paradise resonates throughout myth and literature in the genealogies of the biblical (and pagan) races. The barbarian, heretical, or monstrous races are the descendants of the fratricide Cain, or of the post-diluvian Ham, who looked upon his father's nakedness.4 More to the point, the incorrigible, vile, and abominable descendants of Nimrod (himself the grandson of Ham), are walled in—by Alexander, a hero whose conquest of the East quickly takes on apocalyptic overtones in Western historiography—beyond the limits of civilization.5 180

Facing up to Other People

181

However, these accounts are all, as we have said, legal fictions, so many creations of a simple duality, "us versus them," from a much more complex and interconnected reality.6 The internal contradictions of such simplifications are apparent even on a mythological level. So, for example, the West, which branded the Cynocephali a monstrous race and lycanthropy an antisocial crime, had its own dog- and wolf-men in the Roman Romulus and Remus, the Achaemenian Cyrus the Great, and the biblical Japhet, all of whom were suckled, like the ancestors of so many central Asian nomadic races, by dogs or wolves.7 Pliny the Elder, the great if often credulous encyclopedist of the ancient world, was a Stoic who sought to find, through the fabulousfloraand fauna of the planet, reflections of the Logos by which the cosmos was ordered. The Neoplatonists too, used foreign myths to explain their own tradition. Medieval Europe borrowed its lost pagan gods back from the East, under "Oriental disguise," in much the same way as it did Aristode, through renewed contacts with the Muslims in Spain.8 And when it served its purposes, Christianity could brook a cynocephalic saint of its own, in St. Christopher. Similarly, in India, where Dog-Cooker outcastes were generally the products of miscegenation, such great sages as Vasiştha and Parâsara were themselves of mixed caste; the pedigree of the Vrâtya "Dog-Men" may have been as venerable as that of the Brahmin; and even the god Indra was (according to Yâska) a son of the bitch Saramâ.9 In China, Hun-tun, the chaos dumpling, was a San-Miao barbarian, and the father of the barbarian Man/Yao dog ancestor P'an Hu as well as of the cosmic giant P'an Ku; but he was also the son of Huang-ti, one of the sage kings who founded the Chinese world order. 10 These mythological contradictions are but a pale image of the glaring socio-political conundrums they exemplified. The West traded, had political relations with, and was at times governed by representatives of its barbarian or monstrous races. The dread Huns held Europe in their grasp for a time in the sixth century, traders plied the Indian Ocean in quest of the wealth and wonders of the East, and the Muslim infidel came to be respected for his culture and refinement. The Holy Roman Empire, medieval bastion of Western Christendom, was founded and ruled by people of nearly the same northern European barbarian background as those who had called themselves "Dog-Men" (Hundingar) a few centuries before. In India, a legalist ideology of exclusion was but a thin veneer over a de facto system that embraced the Dog-Cooker outcaste into its fold as the necessary means of voiding good society of its inevitable self-generated pollutions. And, if miscegenation between castes was so loudly and persistendy

182

NINE

condemned in the legal texts, this was no doubt due in part to the fact that it remained a common practice.11 China, which began building the Great Wall in the third century B.C. to contain the barbarian Hsiung-nu, found itself allied, trading, and exchanging princes and princesses with them in the centuries that followed. The Chou and Ch'in dynasties (the latter being the source of the name "China") were founded and ruled by clans originating from districts possessed of a high proportion of Dog-Jung inhabitants.12 In every case, we find that the destiny of the great tradition or civilization13 remained inextricably entwined with that of its barbarians. In a general sense, the first phase of the legal fiction by which these civilizations asserted their own independent existence, as well as their primacy over and against all else, was a process of fission: out of a primordial unity, ambiguous in its internal complexity, a simplified duality was posited. The second phase of this process, also a crude simplification, may be characterized as fusion. By this we mean casting into a single undifferentiated mass or horde a great number of distinct and diverse peoples.14 This tendency is most apparent in China, the Middle Kingdom. First, we encounter the term kuei, which at once signifies anything strange or uncanny, a pupa, a foreigner, or a class of demons.15 China was also unique among the three great traditions we have reviewed in its development of an explicit classificatory system for its barbarians, assigning as it did a quadruped, insect, or reptile qualifier to the Chinese ideogram for the peoples of each direction. While the "dog classifier" was theoretically reserved for the southern and western barbarians—since this was the direction from which China encountered its first ethnic group with a dog ancestry myth— it gradually came to be applied to barbarians from every point of the compass.16 This global identification of such diverse peoples, while it allowed the general classificatory system to stand as a whole, gave rise to a great number of internal contradictions. Many of these turned around the character of P'an Hu: the same Dog-Man was at once the ancestor of the Man, the Yao who "exterminated" them, and the Dog Jung who, although they had long since disappeared as a race, were inserted as the imperial (and P'an Hu's) adversary in the Man "charter myth."17 The same confusion, arising from an inability or refusal on the part of the Chinese to distinguish between their others, situated the P'an Hu myth and cults to the east (K'uaichi, in Chekiang) and south (Kuei-lin, in Hunan and Kwangsi) simultaneously. This was further conflated with the homeland of the Dog Jung,

Facing up to Other People

183

Ch'üan Jung kuo, far to the west. 18 Rather than respecting the distinctions among the three peoples—as they would, for example, that between the provinces of Shansi and Shaanxi within the Middle Kingdom—the Chinese fused these into a single alien entity. Such broad identifications were not limited to the ancient Chinese sphere. In India, the terms Svapaca and Candâla were universally applied not only to the servile underclass, but also to tribals within India's own cultural and political sphere, as well as to such varied foreign peoples as the Kushans, Sakas, and Hünas, who ruled much of north India for several centuries. In the case of foreign peoples, it is quite impossible, when we follow the classical sources, to make head or tail of which nations lived where and who were their neighbors, allies, or enemies. All were indifferendy placed outside (bakya) of the social sphere, even if most became "cannon fodder" in the great epic wars. So too, in Europe, the vilain, the peasant, found himself walled out of his lord's estate, the Church, and society in much the same way as were Europe's monstrous races, especially Alexander's inclusi. The peasant's vileness extended to his misshapen or monstrous physiognomy, and his segregation barred him even from future redemption, a fate even the monstrous were spared once Augustine's speculations became normative. We will return to this theme in due course. As far as the monsters of European traditions were concerned, it is telling that the Germanic term for monster, Unjjeheuer; had for its original sense "one who does not belong to the tribe."19 And, as was the case in China, Europe located its monsters and Dog-Men in every direction, depending on which way the winds of invasion and conquest were blowing. This is most evident in the conflation of Ethiopia and its peoples with Africa and India, and even with the great north of Russia or Scythia.20 Here, we would do well to review another such identification, one we introduced provisionally in chapter 6, that central Asia was the point of intersection of Western, Indian, and Chinese accounts of Dog-Men, Amazon women, and other barbarian races. We intimated a connection between the works of Pseudo-Methodius and other Western sources, which posited a far eastern location of their Amazons, Cynocephali and other inclusi; of Indian accounts of northern Dog-Faces, the Kingdom of Women, and the polyandrous Ephthalite Huns; and of Chinese reports of Amazons and Dog-Men beyond its western borders.21 A closer look at these accounts, however, reveals that many of these correspondences are the fruit of ere-

184

NINE

ative geography on the part of our ancient and medieval sources, of attempts to force a variety of geographical locations into a single central Asian mold. Of all the traditions of Dog-Men and Amazon women we have reviewed, the best documented sources are those on the Man of south China, and the proto-Turkic Ch'i-tan and their neighbors, of northern Manchuria. The first of these traditions is corroborated by Ibn Battuta, as well as by medieval Bengali sources which speak of a Kingdom ofWomen in Assam. When we look more closely at the second of these two traditions however, we find a number of odd twists on the locations of these peoples. Here, our earliest sources are Chinese, dating from the early sixth century, which locate these people on a distant shore of northeast Asia. This is corroborated in a number of later Chinese as well as Uighur and Tibetan sources. It is in this region that the Land of Dogs and their human women ought to have been situated in Piano Carpini's retelling of a Tartar account, but in a twist that smacks of a "pizza effect" emanating from Mongol exposure to the Alexander legend (in which Cinggis Qahan is made out to be the "Mongol Alexander"), Cinggis Qahan and his army find this land to the southeast, rather than the northeast, of their own Mongolian homelands.22 This would correspond to a location somewhere in the region of the Hindu Kush, the central Asian region where Alexander's historians, Arrian and Quintius Curtius Rufiis, tell us that Alexander turned back in his conquest of the world. It was here that, in the high peaks on the eastern edge of the world, his men refused to advance any further. This would correspond, approximately, to Indian locations of the Svamukhas, the Kingdom of Women, and the Hünas. Furthermore, it is in this broad region of "southern Turkestan" that we find the greatest number of Turko-Mongol peoples with a dog ancestry myth. The Gates of Iron, located in this region in a strategic defile between the valley of the Jaxartes and Samarkand, would correspond well to the ancient and medieval ethnography and mythology of Alexander's Gate. However, for the medieval West, with its canons of patristic geography to respect, the Hindu Kush was still much too far to the east to correspond to its traditional location(s) of Alexander's Gate. So it was that this barrier, by which the legendary hero walled out the cannibalistic Cynocephali, Amazons, and some twenty other barbarian peoples, was generally identified with either the Pass of Dariel or the Derbend Pass, in the Caucasus, to the south and west of the Caspian Sea. In the end, the Alexander legend makes the northern (or eastern) ocean a

Facing up to Other People

185

mere twelve day march from the Caucasus, along the eastern shore of the Caspian which, it was assumed, was a bay or arm of the same eastern sea. pseudo-Methodius's tampering with myth to fit geographical prejudices is the most flagrant. This author has Alexander find the unclean nations in the east. He then transfers them to the (west and then?) north before enclosing them: in other words, before walling them in, he brings them back from the Hindu Kush region to the trans-Caucasus.23 It is only by accepting such geographical tampering—which, when the Caspian is made to be a great arm of the northern sea, would reduce the Asian land mass by half—that it is possible to maintain our contention that central Asia was the vortex of all myths of cynocephaly. Here, we are perhaps better served by the term "Turkestan," as employed by McGovern, for the world region in which Dog-Men, Amazons, and their barbarian, cannibalistic brethren would have been located. In so doing, however, we give up any claims we may have made to pinpointing the original locus of traditions of cynocephaly: Turkestan, the lands of the nomadic Turko-Mongols, covers most of Asia north of 30° latitude and east of 40° longitude, from Turkey to Siberia, and from the Baltic Sea to western China! Of the various Turko-Mongol peoples who had a dog ancestry myth, those for whom we have the best documentation are the Ch'i-tan, who ruled northern China from their Manchurian homeland until the early twelfth century. It is an irony of history that these same Ch'i-tan—who had, in addition to their canine totem and ancestry myth, certain Amazon motifs in their mythology—came to be identified, after their conversion to Nestorian Christianity and throughout the twelfth century, with the kingdom of Prester John. As such, they came to be viewed as lost children of the Earthly Paradise, constituting an eastern outpost of Christendom in a barbarian world. We should also note the role of the Nestorians in this context. As the sole medieval Christian presence in the Orient, the Nestorian church was an important changing house for religious, cultural, and mythological traditions throughout the medieval period. Under these conditions, the Nestorians played an intermediary (if not an original) role in the propagation of legends ofthe acts of aposdes Thomas, Bartholomew, and Andrew, as well of as Prester John, Alexander, and others. 24 2 . D O G - M E N AND AMAZONS ON THE LIMITS OF HUMAN IMAGINATION

Regardless of their "true" geographical locations, the classic location of monsters in Western, Indian, and Chinese sources remained the no man's land of the trans-Caucasus or trans-Taurus, the Great North ofTurkestan or

186

NINE

central Asia. As we have demonstrated, it was there that dozens of different peoples, from Siberia to the Caucasus and from the Baltic to Burma, were nearly fused, ideologically, into a single horde. These were the inclusi of European tradition, the Indian Mlecchas or bâbyas, and the Chinese Jung, Man, and Hsiung-nu. In different ages, the names would change, from Scythian to Parthian to Hun to Turk in the West and from Saka to Parâtha to Hüna to Turuşka in India—but the homogeneity of these foreigners, in the great traditions' ideologies, would remain a constant. The reasons for these conflations are quite clear. First, while the three civilizations were aware of and recognized each others' power and greatness from at least the second century B.C. onwards, the rest of the world (with the exception of Persia) was denied such a pedigree. In the absence of a "civilized" tradition of interaction with them, these "uncivilized" others were a cipher on the European, Indian, or Chinese maps of the world. An apt modern comparison would be the nineteenth-century maps of Africa, in which uncharted lands appeared as empty spaces, yawning gaps in a geographical and conceptual continuity. The three civilizations did, of course, have contacts with their central Asian others. The Silk Road, the main East-West trade route for a millenium, spanned much of Turkestan; and central Asians at times (but not as often as they were accused) invaded, occupied, pillaged, conquered, and even ruled parts of Europe, India, and China. Such contacts naturally colored the great traditions' accounts of them. It should be noted in this regard that while none of these three civilizations ever conquered each other before the Western conquests of India and China in the nineteenth century, all three were subject at different times to the military onslaughts of various central Asian peoples. This image of the latter as savage and warlike nomads was aggravated by the fact that few of these peoples had a written or literary tradition of their own with which to defend themselves before their literary accusers.25 It is nonetheless certain, however, that at least one ethnographic detail concerning the central Asian peoples was grasped and retained, albeit in skewed way, by Europe, India, and China. This was the dog- or wolfancestry myth that was common to so many of them. Most fascinating are the different transformations this myth underwent in its appropriation by the three great traditions. China, which bore the closest and longest-standing contacts with central Asia, reported such an ancestry myth quite direcdy, on the subject of the Dog-Jung, in the early Shan-hai ching.26

Facing up to Other People

187

However, with time, the account came to take on more baroque accretions as it was applied to a broader and broader aggregate of "barbarians." India's recognition of its central Asian neighbors' dog ancestry is late, fragmentary, and, as with all Indian geography and historiography, quite muddled. More to the point are its wholly mythologized accounts of the exiled ancestors of the Svapacas. The standard gloss of this term is "DogCooker" (fromsvan + pac). Yet, nowhere in Indian literature does one find an account of Svapacas actually cooking dogs. One may nevertheless glimpse in the alternative form Svapâka a name hearkening to the central Asian dog ancestry myth. The root pâ means "to suckle, nourish, drink," whence the possible meaning "Suckled by a Dog." The same approach may also be applied to the Plinian race known as the Cynamolgi, "Dog Milkers," located (no doubt as a result of the geographical confusion which reigned in the ancient world) in both India and Ethiopia. 27 In every case, a dog ancestry myth—of the order of the legends of Romulus and Remus, Cyrus, and others—may indeed lie at the origins of these traditions. Much later, from some time before the ninth century onwards, the Chinese Man or Yao peoples, who claimed descent from P'an Hu, moved into Yunnan province and so came into contact with Indian culture on the borders of modern Assam and Burma.28 Such contacts would have fueled a continuing Indian Svapaca tradition and served as a source for the fourteenth-century Arab traveler Ibn Battuta's account of Cynocephali and Amazons in the region of Burma.29 Perhaps most striking of all is the juxtaposition, in Europe, India, and China, of races of Dog-Men with Amazons, or with a Kingdom of Women. Central Asian dog ancestry myths would be operative here as well, since they generally involve a human woman who couples with a canine or lupine male to found a given race. This leitmotiv is found in European accounts of eastern, western, and northern monstrous races, as well as in a great number of Chinese descriptions of peoples from every point of the compass. The Indian material merely intimates a connection, locating as it does the Kingdom of Women (Strîrâjya) in the same general direction as its Dog-Faces (Svamukhas), and its Women of Kuru together with Dog-Cookers.30 However, with time, the central Asian locus of the Amazons (and their Dog-Men) would ripple outward to touch such far-flung regions as the Baltic and the Caribbean Seas and the Indian Ocean. In fact, when we look closely at many of the world's traditions of the ancestral union of a male dog with a human woman, we find certain recurrent

188

NINE

motifs that point more to mythic themes than they do to hard ethnography. Most tantalizing is the constant location of these women on an island or across a river, which insulates them from the masculine world of their (often cynanthropic) neighbors. This is a theme we find in the Alexander legend, the "Tartar Relation," the Ceylonese Mahâvamsa, the Indian Petavatthu, Adam of Bremen, and the Chinese sources on Nü kuo, P'an Hu's wife, and elsewhere.31 The juxtaposition of races, genders, and functions, of cynocephalic males and Amazon females is, moreover, a fascinating example of the systematic distortion of ethnographic data. The central Asian dog ancestry myths simply say that a human woman coupled with a male dog or wolf to found a given race, or that a male child was suckled by a dog or wolf bitch, with whom he later united with the same results. The European and Chinese sources do not say this at all. Rather, they describe an ongoing practice, one continuing down to their present time, of miscegenation between two heterogeneous species that remain distinct yet capable of fruitful sexual union with one another. All males are born cynocephalic and all women human. Two separate races, if not two separate species. It is rarely said whether such miscegenation is realized through the cunning of nature or contra naturani, a most convenient oversight in light of the uses made of such accounts.32 This ambiguity made it possible to portray the union of two different races or species as monstrous, savage, or inhuman. In Europe 33 and India, miscegenation is the cause of a fall from grace, which results in exile. In India, the Svapaca outcaste is the exiled servant of the four castes because of an act of miscegenation on the part of ancestral partners in crime. The Svapaca is, moreover, subhuman, belonging neither to his parents' dissimilar castes nor to the human race. For, as Manu's commentator Kullüka Bhâtta writes, just as a mule is neither a horse nor an ass, so a Candâla is of neither of his parents' castes: there is no fifth varna34 In China as well, where the women, but not the men, of "inner barbarian" peoples may intermarry with the Chinese, the cohabitation of human women with male dogs reduces their offspring to the level of animals.35 This first facet of this misrepresentation of the dog ancestry myth, then, was its paradigmatic use as a case of nonrespect of caste, class, or race boundaries. Mixing with the wrong people results in punishment and exile, in dehumanization. This was at least the form that elite propaganda took, even if the power of the elites' position made them immune to such sanctions when they themselves broke the rules.

Facing up to Other People

189

The Chinese case leads us to the other side of the Dog-Man/Amazon phenomenon. Perhaps before all else, and in spite of moralistic trumpetings to the contrary, it was a source of fantasy for males of the great traditions. The lure of "native girls" is not a new one, and reducing the males of a barbarian population to the level of animals would allow one to have it both ways—as indeed the Chinese, upper-class Hindus, and European nobles have done; and as ethnographers (like Bruno Malinowski in the Trobriand Islands) and sailors and soldiers—down to the U.S. Army (and armies of tourists) in Thailand—continue to do down to the present. That is, they have no compunctions about appropriating for themselves the potential wives of "animal" husbands. And, in the absence ofthe real thing—an Amazon for one's very own—myths on their subject might have been a form of ancient and medieval soft: pornography One need only think of the conchshaped vulvae of the women of Kuru, or the intimations of the lover's tryst Alexander had with Queen Candace, to grasp the point we are making here.36 The same holds for the Amazons' account of their mating habits in the Armenian Pseudo-CaUisthenes: We live on the Amazon River, but on this side, on an island. It takes a year's journey to travel around it. And around it there is a river which has no beginnings; and there is no entrance. We 200,000 warrior virgins who have no experience of men live here, and there is no male amongst us. Our men live on the other side of theriver.The state of the land is such that we live in happiness. And from year to year we have a ceremonial assembly . . . and we cross to the other side of the river to the men and they mate with us in marriage, for thirty days. And those of us who wish to be violated cross over and live among the men. 37 In Adam of Bremen's adaptation of this passage, as well as in the Romanian folktale of Cuckoo and Mugur, the Amazons' "men" are the Cynocephali. While it was only in medieval sources that Amazons and their cynocephalic men were used to portray "exotic" sex and violence, we would nevertheless argue that the coupling of male dogs with human women continues to constitute an archetypal fantasy, a sort ofmale projection of sexual domination onto an image of bestiality. This perennial fantasy is one that recurs in present day pornography, as well as in gender art forms—one recent avatar is a certain "Kill Tete de Chien" who appears in a French comic strip series both as a sexual athlete in great demand among aristocratic

190

NINE

women, and as a well-meaning, if impetuous and gullible, warrior.38 In the latter r e s p e c t , he most resembles the cynocephalic St. Christopher. 3 . MONSTERS AS VORTICES OF FASCINATION AND H O R R O R

As it should by now be clear, the monster genre, bereft of Amazon titillations, constitutes a perpetual source of horrified fascination for humans. This is evident in the transparent borrowings between one tradition and another of myths and accounts of monstrous races. The most striking case of such borrowing may be seen in the similarities between the accounts offered by Adam of Bremen and by Hu Chiao of races of cynocephalic men living together with human women. Even the fleeting glimpse the Râjataramgini offers of the exposed breasts of the northern women warriors may be a reference to the Greek and European Amazon ("Missing One Breast") tradition, as may also the Chinese account of the breasdess women who suckle their young from white hair roots at the nape of their-necks. But more than such point for point similarities, the three traditions bear witness to a constant thirst for monster lore that seemed to be best sated with tales transported over great distances by travelers from afar. This is the foundation of Western Orientalism, but also of Chinese Occidentalism and Indian Hyperboreanism (pi. 13).39 Myths found in Indian epics influenced more than a thousand years of European monster lore through the intermediary of Megasthenes; these may in turn have contributed to the monster traditions of the Shan-hai ching. In the medieval West, Pliny's accounts of monstrosities of nature and humanity, although proscribed by the Church, were dutifully commentated by a long series of geographers and theologians well into the Age of Exploration. Back in Turkestan, the Alexander legend undoubtedly contributed to the formation of the Mongol "Romance of Cinggis Qahan."40 How each tradition approached and treated its respective monsters, whatever their provenance, betrays its own preoccupations, category formations, and anxieties regarding its own place—that is, the place of "civilized man"—in the universe. In their meditations on the "limit situation" of the monstrous, the souls of the three great traditions were laid bare in a way not encountered elsewhere. Whatever the case, the "self" that spoke of its monstrous "others" was a troubled one, much less sure of itself than the more common fare of self-assertive ideology would lead one to believe. We willfirstdiscuss the ways in which foreign monster mythologies were translated into the three great traditions. In every case, foreign monsters

Facing up to Other People

191

vvere interpreted according to indigenous categories.41 This is especially apparent in the Western myths of the cynocephalic races. In an ancient Greece that saw nonpolitical man as a mere curiosity, and not even truly human, Ctesias's diminutive Kalustrioi and Kynokephaloi are adorable fantasy creatures living on the edge of the world. When they reappear in the various traditions of the Alexander legend, they have been transformed into gigantic, cannibalistic, and bellicose creatures. This transformation occurred as a function of changes in both political attitudes and ancient ethnology. The democratic ideal of the Greek and Roman republics had given way to imperial ideologies, and the Alexandrian conquests were the new political paradigm.42 Neutral populations could no longer exist in the context of the imperial worldview, and those who would not willingly surrender had to be militarily defeated. So the Cynocephali of the Alexander legend are made out to be hostile and warlike vis-â-vis the civilizing hero of the new order. Ancient ethnography and geography had greatly advanced, albeit in a Frazerian way, between the time of Ctesias and the early versions of the Alexander legend. Megasthenes, Strabo, and especially Pliny each compiled encyclopedic compendia of the wide varieties of natural and cultural phenomena to be found in the known world. Pliny, the most credulous of the ancient geographers, was also the most prolific and most widely read. Although he was not the only one of his profession to do so, Pliny was the most widely cited by his successors when they followed him in fusing together the African and Asian continents into a single land mass known as either Ethiopia or India. So it was possible for Pliny to equate the Cynocephali of India (for whom Megasthenes, following Ctesias, was the source43) with those of Africa, who had previously been mentioned by Herodotus. The latter are without question the cynocephalic apes (Simia hamadryas) of East Africa. As tall as a man but of a much more massive build and a canine physiognomy, these could indeed have been construed as great, hairy, dog-headed men. Even greater than these apes themselves were the bas relief images of them, sculpted on the Upper Nile in the thirteenth century B.C. These giant carved Cynocephali originally stood on the eastern edge of the Egyptians' world, where they greeted the dawn from the Somalian coast (pi. 4). 44 So it was that Alexander would come to meet his giant Cynocephali on another eastern edge of the world, in India. These dog-faced ape men were further conflated, when such Egyptian themes were appropriated by European traditions, with another sort of hybrid. This was Anubis, the jackal-headed (and later dog-headed) Egyp-

192

NINE

tian god of the dead, whose role dovetailed with that of the psychopomp Hermes/Mercury. So it was that the Mediterranean world sometimes depicted Hermes as a cynocephalic and associated St. Mercurius with the same race.45 At the same time, the association of "India" with both the Land of Sunrise and the land of Dog-Men, from the time of Herodotus (Historia 3.98) and Ctesias, also led geographers and historians to seek and find them there, whether the east coast of "India" was in fact Somalia in east Africa, Kamchatka in eastern Siberia, or the island of Cariba that Columbus discovered in the "Indies," far to the east of Cathay. Piano Carpini's retelling of the Tartar campaigns to the Land of Dogs and the Land of Darkness—which is also the place of sunrise—smack of Alexander legend traditions of the walling in of the inclusi, as well as of Alexander's expedition to the edge of the world. 46 Chinese accounts of Dog-Men living on the coast of the eastern sea portray them as eating their food raw, while their human women eat cooked food. 47 This brings us to one of the horrible facets of the Cynocephali's monstrosity: their feral eating habits, if not their cannibalism (this term originated with Columbus's encounter with the man-eaters of Cariba, whom he confused with the "canine" followers of the great Khan). This we find in the Pseudo-Callisthenes, the Revelations of Pseudo-Methodius, the Gadla Hawaryat, Adam of Bremen, and other sources. The source of this characteristic trait may be sought, in the European case, much closer to home than central Asia. Here, the ancient martial brotherhoods of "wolfmen," for whom initiation consisted of dressing in wolf's hides and tearing a chosen victim apart with one's bare "claws," would have served as an obvious model and source of inspiration.48 In this way, the terrible character of the "European" Cynocephali may be seen to have been the product of indigenous traditions, or of uncritical encyclopedic compilations, that fused Egypt, India, and Europe, as well as jackal, ape, dog, and wolf. To these may be added the sea lion, if we are to trust Schlegel's interpretation of the Dog-Men who lived with the Chinese Amazons on the Eastern Sea, in the account of Hui-shen on Nü Kuo, the Kingdom of Women.49 4 . THREE WESTERN THEOLOGICAL APPROACHES TO MONSTERS

In the seventh-century writings of Pseudo-Methodius, we encounter quite a different wrinkle on the cynocephalic and monster traditions of the West. On the one hand, the races or nations of Alexander's inclusi are said to have been shut in because of their bellicosity, incorrigibility, terrible impurity, and their bizarre eating habits (fetophagy, cynophagy, etc.).50 More impor-

Facing up to Other People

193

tant than purity categories for the perennial popularity of PseudoMethodius throughout the Middle Ages was his insertion of the inclusi tradition into the framework of the Christian genre of salvation history. The acts of the pagan king Alexander—who was nevertheless made out to be the precursor of Christ triumphant (and of a certain Jonaton, the fourth son of Noah)—were now inserted into a great cosmic drama that brought all the world's nations and historical ages into play, as divinely ordained preparations for the final battle between Good and Evil, between God and the Adversary. This dualistic eschatological vision was a logical consequence of the biblical creation story in which God ordered chaos—Tehom, the Deep, the equivalent of the Leviathan, or of the Babylonian Tiamat, all sea monsters—and in which the talking serpent of the Garden was transformed into a seven-headed dragon in the last days. It was also the logical consequence of a Zoroastrian gnosticism upon which Pseudo-Methodius, in his Nestorian context in upper Mesopotamia, would have been weaned. What provoked Pseudo-Methodius to adapt the Alexander legend to a Christian context was undoubtedly the rise and imperial conquests of a new world power, the heretical Muslim "Turk." Pseudo-Methodius's was a twopronged rejoinder to the rise of Islam. First, it gave hope to a beleaguered Christian citadel: the release of the Alexandrian inclusi had been preordained by God since the beginning of time. 51 Their eventual defeat by Jonaton would herald the final victory of Christ the King. Second, the inclusi were not merely incorrigible, warlike, and impure; they were also heretics, willfully disobeying God's commandments and the redemption offered them through Christ. Such a categorization would, of course, have been impossible before the emergence of Heilsgeschichte as a historiographical genre, and the appearance upon the stage of history of the dread infidel Turk. With Pseudo-Methodius as well, a new page was turned in the generic nomenclature of the monstrous and enclosed races: no longer Scythians, Parthians, or Huns, they were now gathered together indifferendy under the all-purpose category of (Muslim) Turks, the sons of Umea (the Ummayads)—a category that has, for all intents and purposes, remained with us down to the present in the Christian West. Pseudo-Methodius's was not, however, the final word on the Cynocephali and other monstrous races, nor was his citadel mentality the only Western reaction to the invasions from the East. As we discussed in chapter 3, the classic theology of the monstrous races is that of St. Augustine. Writing some two hundred years before Pseudo-Methodius, this theologian, too, was forced to come to terms with a besieging of Western

194

NINE

civilization by Eastern nomads, and his De civitateDei was a direct response to the agonizing question of how it was possible for God to have allowed His chosen empire and Holy See to be sacked by the barbarian Goths. Yet how different from Pseudo-Methodius's his conclusions were! Rather than dividing the world into two incommensurable camps, and condemning the one as that of the devil (how familiar this sounds to American ears) and elevating the other as that of the Lord, Augustine looked at both Christian and barbarian, human and monster, as citizens of a greater whole, the City of God. In his ecumenical perspective, Rome, the City of Man, was incidental to a much greater plan, whose grandeur and subdetv mere human creatures were incapable of fathoming. Augustine did not presume to know God's will, and it was his humility in this regard that allowed him to think more clearly than his contemporaries as well as most of his successors, and not distort his view of the history of salvation by imposing his own immediate desires and interests upon his theological project, his interpretation of God's will as disclosed through history. Rather than indulging himself and playing to the gallery of popular opinion and the political rhetoric of the day, rather than promulgating Pseudo-Methodian pat answers to the troubled times in which he lived, Augustine insisted on asking difficult, self-searching questions. If God was indeed abandoning the seat of Christendom, was this not perhaps because Rome was in need of cleaning its own house? Augustine sought to redefine the "self" of Christendom in the face of a brutal eruption of a barbarian "other." This he did, instead of glibly weaving a protest of self-justification in the name of a God made to conform to his limited human perspective at the expense of the intrinsic value and dignity of an other who was also a creature of the same God. Augustine's question was, "If God is good and omnipotent, who are we in terms of these others created by Him in His infinite goodness and wisdom?" In the context of the monstrous races, his conclusions were nothing short of revolutionary: God must have created races of monsters and placed them on the extremities of the earth in order that the Christian community, located at the earth's center, might learn from the existence of such creatures that the occasional monstrous births occurring at the heart of their own city were also a part of His great and unfathomable plan. The difference, then, between monsters and humans—if monsters were, as the biblical tradition seemed to indicate, descended from the "one protoplasm" of Adam's stock—was not one of kind, but of degree. Some of us are like them just as all of them are like some of us. Augustine's astonishing ecumenicity in this

Facing up to Other People

195

regard not only redeemed the monstrous race by allowing them a potential niche in the City of God, but it also gave dignity and a right to life to social

pariahs and medical anomalies living within the Christian sphere. Some centuries after Augustine lived and wrote, there remain a great many self-proclaimed leaders who, like the propagandist PseudoMethodius, fall far short of Augustine's example of humanity, foresight, sixteen

and compassion.

In terms of the scope of this study, the immediate effect of Augustine's meditations was the redemption of the monstrous races and a promise that they too would find a place in heaven after the final trump. Augustine's conclusions, moreover, opened the way for further theological speculation on their fate, thus institutionalizing such reflection as a Christian theological genre. While many of his successors would infantilize the Augustinian example for their own sensationalist or propagandists ends, there were others who conscientiously continued Augustine's ecumenical project. That is, they were courageous and humble enough to venture outside the limits of their own institutional order to ask challenging existential questions about their place in the world, in the face of the rest of the world. Isidore of Seville, a Spanish theologian more or less contemporary with Pseudo-Methodius, was one of these. Isidore, like Augustine, had to grapple with a situation of social and spiritual anomie. Of aristocratic Latin stock, Isidore was troubled by a continuing barbarian presence in the heart of his, the last bastion of the Roman empire. In his Etymologiae on the races of men, Isidore preoccupied himself with the conundrum of the basic humanity of man in the face of the prodigious and troublesome "swarming" of a great variety of human races and ethnic groups. Were Christian and barbarian, human and monster, all men? In his treatment of the question, Isidore began by applying the ancient (from the time of the Babylonian EnumaElish at least) interpretive device of genealogical reasoning, of deducing an order from a family tree of the world's populations. He thus began his chapter "On the Names of the Races" with a genealogy of the foreign and monstrous races, all of which he traced back to venerable biblical ancestors. If these descendants of Adam and Noah were all monsters, then were all the descendants of Adam—that is, all humans—not monsters in some way? Here, Isidore turned to that hermeneutical technique for which his great encyclopedia was named: etymology. "Monsters" were, from the time of pagan Roman religion, a "showing forth" of the divine will, or a divine warning to humanity. Their existence was therefore not contra naturam, but a part of nature, even of

196

NINE

human nature, for which man had to account in his reckoning of himself if he were to generate a coherent anthropology.53 In Isidore's synthesis we see a self-conscious yet self-effacing attempt to derive a deductive definition of humanity. It is doubtful that Isidore, any more than Augustine before him, would have been moved to pose the difficult, self-searching questions that he did had there not been an oppressive barbarian presence at the very heart of his universe. The physical presence of social, cultural, and political others provoked in him what we would today call an existential crisis. Rather than falling apart or striking back in counter-aggression, Isidore allowed those others to inform his world, alter his life, and place his own beliefs in question in order to define his truth more completely. In his writings are laid bare not only the precariousness of the medieval image of man—vis-â-vis other men as well as the other world of divine and demonic powers—but also the problematic position of man in a hostile and threatening world. 54 5 . ENLIGHTENED MISSIONARIES TO BENIGHTED HEATHENS

The responses of Isidore and Augustine were exceptions to a general rule in interpersonal and international relations, as well as in many of the methods employed by the social and positive sciences. In most instances, one's own truth—be it in the form of individual self-assertion, political propaganda, a normative philosophical concept, or a scientific proof—is asserted hypocritically, in order to allow for a hypercritical reduction of the other to a negativity, a counterinstance, foil, opposition, or antitype. Augustine and Isidore, although exceptional, are indicative of what really happens in an interpersonal encounter: the "self" is a reaction to the question which the other poses by its very presence. More than this, the self is posited, in every case, as a reaction to existence itself. Without an outside world—a world in which death dances with both the meek and the mighty—to stimulate, frustrate, and shock, there could be no self-consciousness. We cannot see ourselves save in the mirror of the world, in spite of the cold, quantitative calculations of the hard and political sciences, which are vain denials of the primacy of the other to any self's willing and doing. The troublesome presence of the unexpected, the imperfecdy known, is the motor to all one does, even when one says "I think, therefore I am."55 In the course of time, with the reinforcement of the medieval Church, albeit under protectors who would have been branded barbarians a few centuries before,56 a new era of European expansionism began. Now, under the banner of Christendom, European imperial pretensions and

Facing up to Other People

197

interests took the form of a mission to redeem the world. Now, than barbarian adversaries or inhuman monsters (Augustine's writings had seen to this) to be humbled, these became benighted heathens to whom a truly "catholic" Church could reach out to bring into its fold. As servants to their enlightened masters, of course. The pious fiction of the simple heathen glimpsing, through an act of divine grace, the superiority of the Christian path is one that has served missionary colonialism down to the present day. Only the names of the heathens (the Third World), adversaries ("religious extremists," or the "evil empire" of the communist bloc) and saviors ("democracy," "free trade") have changed. The medieval propagandists had much of the raw material of their gospel ready to hand in the accounts of the monstrous races bequeathed to them by the ancients. These were the peoples, in the medieval myths, who were first missionized by the aposdes (Bartholomew, Andrew, Thomas, Matthew, etc.), following the Pauline evangelization of the Mediterranean world. The legend of the cynocephalic Abominable is a prime instance of this new medieval genre. 57 The monster remains the same—only the hand holding the mighty sword has changed: instead of Alexander and his army, it is God himself who, through the works of his military saints, causes the wild savage to surrender himself. Abominable, who would wolf down the aposdes Andrew and Bartholomew and their disciples in a trice, is here first ordered by a disembodied voice instead to find and protect them. When the cynocephalic man falters, pleading that his wild nature would only drive him to eat them the first time hunger came upon him, God tames the beast by casting a wall of fire around him. Thus terrorized, Abominable becomes as gende as a lamb when God sends an angel to make the sign of the cross over him, and he is "glad because he [has] learned to know the right faith." Andrew then bestows upon him the name of "Christian"—but to what end? In order that he might revert to his savage, cannibalistic nature at a difficult moment, so that his cannibal cousins of Bartos might "know that there is no other God beside Thee." Christian then slays, disembowels, butchers, and devours seven hundred men and three nobles, to the glory of God; in the end the survivors, too, see the light. Christian now returns to his sweet, converted nature, which moves the cannibals of Bartos to demand baptism for themselves forthwith. 58 It is surprising that the Japanese have not yet made a Godzilla-style movie, or the Americans a melodramatic mini-series, from this account. All economic rather

198

NINE

the necessary elements—the good-hearted monster who only becomes monstrous when it is a question of killing other monsters, together with a happy ending, neady tied together—are explicidy present. Behind the bathos of Bartos, however, is a piece ofecclesiastical propaganda that depicts not only the triumph of Christianity over a backwater of the universe, but also the neat use of one monstrous or barbarian convert to destroy other barbarians.59 All that remains is for the aposdes to baptize the survivors. This sort of tactic has remained common down to the present da\; through recent empires whose military forces (with a complement of "native" soldiers) worked in tandem with missionaries to bring the heathen to submission, and through current tactics of supplying both sides of regional conflicts with weapons to destroy one another. It is also a policy that was employed by the Chinese Han dynasty: i-ifa-i was the pitting of one barbarian people against another to the benefit of the Middle Kingdom. 60 Such willing submission to a higher order on the part of a lesser creature is carried to an even further extreme in the Indian and Chinese core myths of chapters 4 and 7. The Chinese case, while less hypocritical and sanctimonious than that of Abominable-Christian, is more cynical. Here, an "inner barbarian" people, the Man, in the form of the emperor's pet dog P'an Hu, kills the general of the "outer barbarian" Jung (or Fen) and defeats his minions. In exchange for their fealty—that of a loyal puppy—the Man are granted a certain status vis-â-vis the imperial court. The Indian myth of Visvâmitra in the Dog-Cooker village is, more like the Ethiopic Abominable myth, a study of jingoist, self-serving propaganda. It is only, of course, because this is a parable on royal realpolitik in times of social and natural upheaval that an outcaste is given any voice whatsoever in this authoritative source. And, in this topsy-turvy context, the nameless Candâla speaks the "truth"—that is, the orthodox teaching on licit and illicit foods. In so doing, he justifies his own group's baseness. The punch line comes when his particular situation proves to be the sole exception to the rule he has been parroting. Visvâmitra cooks dog meat and is blessed, while the Svapaca, who is not so much as thanked for his solicitude, is out a meal. In a sense, only an outcaste could be cast here since, in the role he plays, a heinous sinner proves him wrong even when he is right. This would have been an outrage to a member of any other caste. Visvâmitra's Svapaca plays the same role as the happy go-lucky "pore dumb nigger" or "injun" in early and mid-twentieth-century American cinema. Outcastes continue to provide comic relief in modern Hindi cinema. No doubt the conversation

Facing up to Other People

199

between Visvâmitra and the Svapaca would be turned into high camp by modern Bombay film moguls. 6 . T H E SELF-SURRENDER OF THE NOBLE SAVAGE

In all three of these core myths, apart from their overriding hypocrisy and cynicism and the leitmotiv of the domination of an inferior people by a great tradition, we can also clearly see the seeds of myths of the noble savage. In India, apart from the Svapaca met by Visvâmitra in the midst of a cosmic crisis, the monkey servant-god Hanumân, whose many heroic exploits further the cause of Râma and Sltâ, may be the mythic representative, along with his monkey army, of an aboriginal Indian people who chose the right cause.61 The same may be said of the protagonist of another myth, the yoksa Harikesin, who turns against his fellowyaksas to become Dandapâni, a servant of the god Siva. Dandapâni's submission entails his dedication to Siva of a linga, a phallic icon, named Dandapânesvara. This too was a commonplace—that an Indian "folk god" surrender to the god of the great tradition by consigning his own shrine to that god. In the case of Siva, the god's name is retained in that of the linga dedicated by him, with the Saivite suffix -isvara appended to it. 62 Parallel to Harikesin's example is that of the Sri Lankan yaksini Kuvannâ, who gives herself up to the civilizing prince Vijaya.63 In fact, yaksas, yaksinis, and other sorts of aboriginal demons are constandy surrendering to gods in Hindu and Buddhist accounts. On the one hand, this is a continuation of a Vedic theme which identified the indigenous enemies of the invading Indo-Aryan nomads with demonic asuras; on the other, such myths document the "Sanskritization" of non-Hindu divinities—in this case of the archaic tree-spirits, the Indian complements to the aquatic nagas—and their shrines, and the entrance of these divinities into the great tradition as servants of the Hindu gods. In the realm of the plastic arts, the yaksas" submission translates, in early Buddhist architecture, into their transformation into column supports. These big-bellied and toothy dwarves, compacted as it were beneath the tons of stone they uphold, suffered a fate not shared by their female counterparts. Yaksinis remain lithe, sinuous, voluptuous Amazons, ornamenting the vertical supports of entrance gates—at the Buddhist stupa of Sanchi for example—in alluring poses.64 Their submission to civilization is sexual, while that of their men is servile. Of course, we also encounter the converse in Hindu mythology. An outcaste or tribal may prove to be a god

200

NINE

in disguise, as is the case with Siva, who tests Arjuna's mettle by appearing to him, in the form of an aboriginal Kirâta, accompanied by a dog. 65 Such myths are good inducements to attitudes of hospitality and equanimity any stranger one meets may be a god in disguise. This theme, which is anything but self-centered (although it may have been abused by begging "holy men" at times), testifies to the largesse of Hindu society towards strangers on a personal level, a quality which never fails to impress visitors to India. Apart from the odd cynocephalic, Europe's original and perennial noble savages were members of that monstrous race known as the Bragamanni. These were the holy men of India whose comportment and life philosophy so impressed Alexander's chroniclers and many who followed them. Although "Bragamanni," the Latinization of the Sanskrit term brahmana (brahmin), was employed as a blanket term in the West, it was not the members of the highest caste or the Hindu priesthood (the group that produced the sort of generally propagandistic mythology we have been discussing) that were intended. Rather, the European traditions were referring to India's renunciant ascetics, those very heroes whose equanimity, selfeffacement, and acts of self-mortification threatened brahmin authority in post-Vedic India. 66 Such cases—of one orthodoxy's bane being the paragon of virtue in the eyes of a brother orthodoxy—can be seen today when one group's "terrorists" are another's "freedom fighters." Whatever the case, the Bragamanni, because of their saintliness, were often located by Western geographers in a paradise situated at the eastern extremity of the world. 67 Living a simple life in the bosom of nature, and possessed of extraordinary powers, the Bragamanni straddled the line between the noble savage and the wonders of the East. Stranger than paradise itself was one of the uses the West made of them. As the perennial rivalry between Christendom and the Islamic world continued without end, one of the many literary forgeries—forgeries which seemed to be a medieval speciality—brought this Eastern paradise to the fore. This was the socalled Letter ofPresterJohn, written in the twelfth century to the Byzantine Emperor Manual. According to this letter, somewhere on the far reaches of the world—in India, central Asia, or Ethiopia—Prester John ruled over a prelapsarian paradise that the West had forgotten. Now Prester John and his kingdom of saints (among whom the Bragamanni and the Amazons were counted) was under siege by the armies of the infidel. This association is all the more remarkable when we note that the historical people most closely identified with the kingdom of Prester John throughout the twelfth

Facing up to Other People

201

-entury were t h e same Ch'i-tan ( n o w called t h e Kara Khitai) w h o h a d been portrayed as a race o f Dog-Men in Manchuria two centuries before. Having s * n c c ' o s t t ' l c ^ r e m P i r e been pushed into central Asia and converted to Nestorian Christianity, they are n o w made o u t to be t h e children o f Paradise.68

An important result of the Letter ofPrester John was the continuation of the Crusades into the thirteenth century. While the Muslim powers to the east were quite unaffected by such exercises in ecclesiastical intemperance, the fabled lands of the noble savages were very much affected. If the Crusaders failed in wresting the Holy Land or any other lands back from the Seljuq Turks or other Islamic empires in this period, their contact with the Oriental world brought a sharp rise in the West's demand for such luxury items as spices and silk. Within a few centuries, the Mediterranean became the center of great trading empires. Under this new commercial impetus—one that was nonetheless coupled with a continuing missionary effort by the Church—the fabled lands of the East were explored throughout the Middle Ages and the Age of Discovery. During this period, the frontiers beyond which the monstrous races dwelt were constandy rolled back. As India became quite well known to the West, the monstrous races came to be located further and further afield, in Tartary, Burma, and the islands of the East and West Indies. 69 Such exploration brought about the discovery and conquest by the West of peoples unheard-of in the writings of the ancients, even if the ancient names and categories continued to dominate medieval, Renaissance, and later thought. These human races became subject, like the Cynocephali and cannibals of medieval hagiography, to evangelization, colonization, and economic exploitation by the West in the centuries that followed their discovery.70 The uplift of the benighted heathen, and the task of turning him from his oudandish beliefs to the straight and narrow path, became a prime apologetic for the imperial expansion of the Europeans, one which came to engulf much of India and China. It was the Indians and natives of the Americas and Africa, however, and not the subjects of these doddering and decadent empires, who most excited the European imagination. In the wave of anti-institutional thinking (to which this study is indirecdy heir), it was these "nature folk" who were identified, because they had never been tainted by internal or external political domination, as the last remnants of a perfect Edenic humanity, an ideal toward which a neurotic modern world might aspire to return. Enlightenment thinking on the noble savage, starting with Rousseau

202

NINE

and others of his stamp, carried its own ideological baggage, and thus complemented the rise of a positivist scientific method which placed the researcher, newborn as it were to each of his experiments, at the center of the natural world which he in turn bent to his will. This idealization and "domestication55 of the noble savage has undergone a great number of reincarnations and reinterpretations down to the present day—"flower power55 and the return to nature of ecological movements are contemporary avatars of the phenomenon. The noble savage is only one post-medieval hypostasis of the monstrous races in Western traditions. Along with him must be counted the Wild Man, that denizen of the European forests who, half-peasant and halfmonster, has been domesticated over the past five hundred years into a benign carnival character and a romantic illustration motif. The domesticated Wild Man, Europe's indigenous noble savage, became further conflated with certain monstrous races who had also received their "titreş de noblesse.55 Among these were the Cynocephali, admirably depicted in a sixteenth-century French illustrated bestiary (pi. 9). 71 The disappearance of the Wild Man from the land left a void in the European mind, however, for it deprived the imagination of an "empirical other5572 upon which to externalize, in a concrete way, European fears and anxieties. It is not surprising that, with the "disenchantment55 of Western man, there arose, in the late nineteenth century, the enlightenment science of psychology (with its therapeutic applications in psychoanalysis and psychiatry), which selfconsciously internalized the history of world mythology into a system of complexes, cathexes, and archetypes. The Frazerian march of progress, from magical to religious to scientific thinking, rendered monsters and Wild Men members, albeit wayward ones, of a self-sufficient psychological whole. The other was made to be a backwater of self 7 3 7. H U N S , REDS, AND THE M A D DOG OF THE MIDDLE EAST

Another hypostasis of the medieval other was and remains foreign peoples. One of the terms employed in English legal usage for a "non-naturalized" citizen clarifies this point: the term alien is derived from the Latin alienus, "that which belongs to another.5574 If we have harped at great length upon the ancient and medieval uses of political, ethnic, or racial others, it now behooves us to do the same with more modern examples. It will be recalled that the cynocephalic races, once they burst into the European world picture, came to be located at every point of the compass, often in accordance with the provenance of Europe's invaders or the course of European inva-

Facing up to Other People

203

sions or evangelization. So they were located in Ethiopia because this was Herodotus said they were, and in India because this is where Ctesias and Alexander found them. Later, with the influx of Germanic and central Asian peoples into the heart of European "civilization," the great North, the ancient homeland of the Scythian, became the direction of choice for such races. Finally, for the Irish, on the Western reaches of Europe, the Cynocephali invaded from the West, from across the sea. Now, if we move from the dread monstrous races to the nomadic peoples who perhaps inspired accounts of them, we run up—as did the Europeans, Indians, and Chinese—against the Huns. Indeed, Huns, cannibalistic Cynocephali, and Amazons are often identified or juxtaposed in Western traditions. Further to the east, the nemesis of the Chinese were a northern Altaic people whom they called the Hsiung-nu (Ferocious Slaves). It was in order to control their rampages that the building of the Great Wall was undertaken in the third century B.C. It was not, however, until early in the fourth century A.D., when the practice of pitting barbarians against one another and the process of Sinicization once again prevailed, that the Chinese at last freed themselves of them. At this point, the southern Hsiung-nu had been neutralized, even sinicized. Their northern cousins were driven into retreat—into Chinese central Asia and out of Chinese historiography—late in the first century A.D. 7 5 NOW, it happens that in the waning years of the fourth century there appeared, on the eastern marches of Europe, a people who came to be known as the Hunni; and that Indian sources report, starting in the late fifth century, the existence of northern invaders known as the Hünas (whom they sometimes qualify as Sveta, "white"). Western sources as well refer to the White (Ephthalite) Huns, who are light-skinned and have "Persian" features, in contrast to the Huns tout court.76 For the past two hundred years—since the time in which many of the disparate sources on the Western "Huns," Indian "Hünas," and Chinese "Hsiung-nu," as well as on the White or Ephthalite Huns have been brought together by scholars—it has been assumed that these were all one (or at most two) people. Yet, when one examines the grounds for such an identification—at least between the Huns who were the scourge of late antiquity in the West, and the Hsiung-nu who troubled China for so many centuries—it becomes evident that solid evidence is lacking. First of all, there is the troublesome question of what became of these people during the century in which nothing is heard of either the Hsiung-nu or the Hunni (this problem is compounded by Chinese accounts of their having %vhere

204

NINE

exterminated the Hsiung-nu, to a man).77 Can it be possible that this most outgoing of nomadic groups simply withdrew to a retreat removed from all external contacts—in a region crossed by the Silk Road, a region that stood as a perennial changing-house for dozens of different cultures? If thls were the case, what would have suddenly moved them, after a century to strike out in a western direction and conquer everything in their path, all the way to the North Sea? Among the scholars who have worked to debunk the "pan-Hun" theory. Otto Maenchen-Helfen has perhaps been the most ruthless deconstructionist. 78 On no less than four grounds—linguistic, historical, archaeological, and ethnographic—this scholar has shown that Hsiung-nu and Hunni do not a Hun make. For example, the Hunni were beardless while the Hsiung-nu had beards; on the scant linguistic evidence at our disposal, there is no resemblance between the languages of the Huns and Hsiungnu; and so on. In the final analysis, it becomes difficult to maintain that either "Hunni" or "Hsiung-nu" were a single racial, linguistic, cultural, or political group. Both were composed of a great mixture of peoples, one which changed over history. Maenchen-Helfen's contention is that modern historians can fall into the same reductionist trap that medieval theologians and geographers did, of giving all barbarians the same face and attaching to this undifferentiated, swarming mass the name of whatever group is presendy invading "civilization" (although in the modern case, it is not a present invader, but the fusion of two nomadic peoples of the past that is being effected). The Hsiung-nu were the most incorrigible of Han China's neighbors and the Hunni the bane of fifth and sixth century Europe. Therefore all are reduced to "Huns." "Hun" is a term with modern pejorative uses, too. This was the case with the application of the term to the German people—a most ironic appellation considering that their own ideologues, during the Nazi era, portrayed the Germans as the purest scions of the "Aryan" race. What is most meaningful about this designation is, once again, the blanket use of a particular name—Hun, barbarian, horde, even "Middle Ages"—to cover a wide variety of disparate peoples, movements, or phenomena. We see this today most glaringly in the use, by conservative ideologues around the world, of the terms "communist" or "Marxist-Leninist conspiracy" for any popular movements that run counter to their ideal of "freedom"—generally the notion of "free" expansion of trade.79 Under this monolithic heading, p e a s a n t revolts against feudal landowners and Stalinist military bureaucrats are all thrown, indifferendy, into a single threatening mass, against which the cit-

Facing up to Other People adel

205

of freedom must defend itself—through systematic disinformation, ^ s t a b i l i z a t i o n , or decimation. Following the same logic, the threat, first perceived as an external one, shows itself to be internal as well. So it was that the American senator Joseph McCarthy was able, in the 1 9 5 0 s , to brand all persons whom the extreme right feared—intellectuals, artists, hom o s e x u a l s , labor leaders, foreigners, Jews—as "communists," and include their names on imaginary lists of subversives. Such demagogery generally t e n d e d to numb the public to the fact that the political and military bureaucracies of both the Eastern communist bloc and Western democracy bore m ore similarities to one another than differences: both sought, above all else and in spite of their ideological trumpetings to the contrary, to further and increase their own powers and privileges. Apart from the Red Peril, there is another perceived menace to world democracy and freedom which is most germane to this study. This is militant, politicized Islamic fundamentalism. In the past decade, Western Christendom—as well as oil and military interests—has faced in the new avatar of its perennial adversary, the Saracen infidel, a mirror image, as it were, of its own ideological intolerance. One of the flashpoints of this new avatar of an ancient conflict has been the North African state of Libya, a region that was conquered by the barbarian Vandals, Alemanni, and others shordy after the sack of Rome. Libya's ruler, Moammar Qaddafi, has made no secret of his hostility towards the "superpowers," an attitude that has won him charges, from the usual sources, of being a communist sympathizer, international terrorist, psychotic, and so forth. The former President of the United States, Ronald Reagan, shordy before he launched a modern-day crusade against the Muslim infidel—an air strike that destroyed petroleum-related and civilian targets—referred to Libya's leader as the "mad dog of the Middle East."80 Let us take this statement as a contemporary example of ancient and medieval reductionism, as a naive appropriation of a perennial image, used to propagandistic purposes. As we have seen in the historical and ethnographic portions of this study, "dog" as a term of abuse has a long pedigree in Europe, India, and China.81 To call a human being a dog is to deny his humanity in a particular way. The dog is "man's best friend," and the creature with which humans have had the longest standing symbiotic relationship. However, it may be said that familiarity breeds contempt between species as well as between individuals. More than this, the domesticated dog always bears some vestige of its wild ancestors—the wolf, jackal, or dingo—and it is precisely this feral aspect that surfaces

206

NINE

when dogs become rabid or distempered. The "mad dog" bites its master (like a savage beast), and in so doing is especially dangerous to humans since it attacks from the very heart of the domesticated world—the family hearth, the main street of our town—from within the social sphere rather than from the wilderness without. This tension is inherent in the medieval treatment of oudaws as "wolfmen."82 The mad dog is therefore a traitor to the trust that humans have placed in the canine species for some ten thousand years. Its crime of assault is com» pounded by that of treason. The penalty is death: mad dogs must be destroyed. President Reagan referred to Colonel Qaddafi as a mad dog so that he might be put to death with impunity. Unfortunately but not uncharacteristically, the American air force's "surgical strike" missed its target, and instead snuffed out the life of Qaddafi's fifteen-month-old adopted daughter.83 Somehow, we are told, this murder, committed in the interest of freedom, was not an act of terrorism. So much for a phenomenology of this particular animal term of abuse. When we move from this domain to that of history, the modern and ancient parallels are quite astonishing. Ethiopia was, of course, the land of the cynocephalic baboon, whom many ancient geographers conflated with the Indian race of Cynocephali. Africa was also the land of the Ptoemphani, who placed a dog on their throne during transitional periods between human rulers. Most to the point, however, is the kingdom of Garamantes, whose ancient borders correspond to those of the modern state of Libya: according to Pliny, the King of Garamantes literally had two hundred "dogs of war" in his army! To be sure, neither Reagan nor his advisors consulted Pliny before their bungled attempt to assassinate the present day ruler of Garamantes. The point here is that, contrary to developments in many other fields of human endeavor, political rhetoric and its implementation have not, in many cases, emerged from the Middle Ages—at the expense of all too many babies. 8 . T H E FACE OF THINGS AND THE FACE OF THE OTHER

Forces of chaos, minions of the devil, monsters, barbarians, lower evolutes of the species, the "others" against whom the apologists of the great traditions define themselves—all seem to have no ontological value whatsoever, save as antitypes to the categories of their civilized superiors. Or so it would appear on the face of things. The question that must be posed, once again, is whether the face at issue is not a two-dimensional legal fiction concocted

Facing up to Other People

20 7

mask some deeper, more multidimensional and interconnected human realityThis has been, of course, the thesis of this study. Ours has been an exercise in reading between or behind the lines—that is, an application of a hermeneutics of suspicion to ecclesiastical and political systematic distortions on the subject of other people. Through this interpretive approach, we feci we have come to a certain understanding of the existential and psychological grounds for more than two millennia of sound and fury. DogMen have been portrayed in many of the traditions under study here as "speak[ing] two words and bark[ing] the third," 84 this failure of words thus betraying their inhumanity. Perhaps, in the end, our commentators and ideologues have been barking up the wrong tree. As we demonstrated in the first chapter, the polarized categories used in political and religious propaganda are ideal types, not unlike the perfecdy controlled environment of the circus, that bear little resemblance to the world as we normally experience it. 85 We further pointed out the need for regular injections of the chaotic into the neady ordered cosmic categories of a given tradition to spare it from total stagnation and inertia. Without a tension from without and an osmotic exchange across boundary lines, "the center cannot hold."86 Its structures and institutions become ossified and incapable of withstanding inevitable change. One of the ways in which we have allowed our world to be charged with new energy and life has been through our fascination with the monstrous and horrible, a fascination that today translates into the great success of war, horror, and science-fiction films. In our timorous quest for meaning, we are moved to push to its very limits our experience of human reality, in spite of all the logical, metaphysical, institutional, social, and cultural defense mechanisms that shelter us from such questions. This sporadic existential brinksmanship implies having the courage to face the limits of our own humanity—the courage to gaze into the troubled nightside of existence, into the world of dreams and nightmares, of our own dark passions and impulses, and to reach a satisfactory compromise by which we may carry on with our lives. Ultimately we cannot understand what it means to be human without having had some experience of the inhuman. Yet, as we have shown throughout this study, this authentic human response to the nothingness that "lies coiled at the heart of being, like a worm"—a response that has taken the form of continuing fascination, if not an obsession, with the ant0

208

NINE

cient and medieval traditions of monstrous races—has been appropriated and used to propagandistic ends by various ruling institutions and their ideologues. There has nearly always been some authenticity to popular fears of foreign or monstrous races, even when such fears have been exacerbated by propaganda. It has only been in the past century, however, that all the blank spaces or black holes in our maps of the world have been filled in with both physical and demographic data. In the pre-twentieth century context, the question of whether the "missing link" continued to live somewhere 011 the globe was still a viable one. So too, for that matter, were speculations regarding monstrous, even cynocephalic races. It was still theoretically possible that, beyond the last known mountain range or inland sea, there still existed some unknown land whose people defied all existing taxonomic ordering. In this sense, the fear of monstrous races was a valid one until quite recendy. And indeed, in those parts of the world where maps and photographs and teaching have not sufficiendy penetrated to lend a cosmopolitan view of the planet, such fears are still understandable. It is likely that most of the planet continues to live in a conceptual world in which many such blank spaces remain. Until a certain cultural literacy is attained, in whatever society one inhabits, the world remains a place of disconnected "bytes" in which ignorance, in the form of disinterest, gossip, racism, bigotry, etc., fills the gaps. Such popular ignorance has been the greatest ally of the ideologues of superpowers and empire builders throughout history. These ideologues' propaganda has consisted of nothing less than the systematic distortion—and the reduction, into "bibblebabble"—of other peoples' myths; that is, of the very foundations of other peoples' self-understandings, of their geography, history, and worth. The old empires are dead; sadly, imperialist thinking is not. And here we come to the reason for which we feel it necessary to apply the lessons of the monstrous races to our contemporary global context. We find much of the Pseudo-Methodian—the citadel mentality and jingoist rhetoric—among the political ideologues of our own time and place. Conspicuously absent are the deeper, self-questioning attitudes of Augustine and Isidore. This modern approach is most paradoxical in light of the fact that economic, transportation, and communications networks have transformed the planet into a global village. There no longer is an outside world or chaos lying beyond the last known body of water or shadowy mountain to harbor real or imagined monsters. If monsters exist, they must be sought on Mars, or on some Star Wars planet in an unpronounceable galaxy—or in the

Facing up to Other People

209

human subconscious. In this last case, it is the psychoanalyst who has become the modern monster-hunter and exorciser. Either our world, the entire world, is a human world or, to paraphrase St. Augustine, "barbarians 'r' us" (pi. 14). 87 Why then do so-called world leaders find themselves obliged to rally us against such bogey-men, even to root them out from our very midst? Are they all psychotics, to a man? Of course not. They are, like Proust's arrogant, chowder-headed diplomat, de Norpois, cited in the epigraph at the beginning of this chapter, merely possessed of attitudes that are willfully provincial,88 as opposed to being open, ecumenical, or cosmopolitan. Such persons deliberately reduce the complexities of the interconnectedness of existence to a limited self-referential scale. Not only does this create artificial polarities of national versus foreign interests, but it also creates a perceived need for walls to defend against the latter. With time, these walls have grown higher and higher and more and more costly even as they have become less credible. At the heart of this perverse state of affairs lies the ancient problem of "us, not them." They are the mad dogs, the great satans, the empires of evil, while we are the self-proclaimed divinely ordained dog-catchers or saviors of a threatened world. It is such logic which has permitted such human monsters as Hider and McCarthy to present themselves as defenders of the general welfare against generally non-existent "lists of names" and anonymous aggregates. How are we to respond to such violence to others, to ourselves? In the end, there can be no clear divisions between "us" and "them," between humans and monsters, or civilization and barbary. Only through an openness to meaningful encounter, to dialogue, and to interaction can we hope to find a path to authentic self-understanding, and hope for the continued existence of our fragile blue planet. So it is that we must let ourselves "see God in the face of the other" and allow the other to figure in the reckonings we make of ourselves—just as we would the God who created us. 89

notes

CHAPTER 1

1. Isidore of Seville, Etymologiae 11.3.1-3. This is discussed in Claude Lecouteux, Les monstres dans la litterature allemande du MoyenAge: Contribution a I'etude du merveilleux medieval, 3 vols., Göppinger Arbeiten zur Germanistik, no. 330 (Göppingen: Kümmerle Verlag, 1982), vol. 1, Etude, pp. 2 - 3 . Lecouteux's is perhaps the most encyclopedic study of European monster traditions written to date. See also vol. 2 (Dictionnaire), pp. 2 1 - 2 7 . 2. A contemporary reading of this sort of ideology may be elicited from Jürgen Habermas's Theory and Practice (tr. John Viertel [Boston: Beacon Press, 1973], pp. 272-73), in which he gives a critique of positivist critiques of ideology: "The [third level of rationalization] extends to strategic situations, in which rational conduct in the face of an opponent who also acts rationally is to be calculated. Those acting do not only wish to gain control technically over a specific field of events . . . but also to gain the same control over situations of rational indeterminacy; they cannot inform themselves about the conduct of the opponent empirically in the same manner as about processes of nature, by means of lawlike hypotheses; their information remains incomplete . . . What interests us . . . [is] the particular technical compulsion which . . . such strategic situations exercise upon value systems. A basic value enters into the technical task itself, namely, successful self-assertion against an opponent, the securing of survival." 3. Louis Dumont, Homo hierarchicus: le systeme des castes et ses implications (Paris: Gallimard, 1966), p. 15, n. l a : "The word 'ideology' commonly designates a more or less social group of ideas and values . . . It is obvious that there exists a fundamental ideology, a kind of mother-ideology, that is linked to a common language or to society as a whole." On the disclosure of ideology through myth, Georges Dumezil writes that "the function o f . . . myths is to express dramatically the ideology under which a society lives; not only to hold out to its conscience the values it recognizes and the ideals it pursues from generation to generation, but above all to express its very being and structure, the elements, the connections, the balances, the tensions that constitute it; to justify the rules and traditional practices without which everything within a society would disintegrate" (The Destiny ofthe Warrior; tr. AlfHiltebeitel [Chicago: University of Chicago Press, 1970], p. 3). 4. Jaan Puhvel, Comparative Mythology (Baltimore: Johns Hopkins University Press, 1986), pp. 1 - 2 ; Michel Meslin, "Breves considerations sur l'histoire de la recherche mythologique," Cahiers intemationaux de symbolisme 36 (1978): 194-95. 211

212

NOTES TO PAGES 2 - 7

5. Northrop Frye, cited without reference in Hayden White, "The Forms of Wildness: Archaeology of an Idea," in The Wild Man Within, ed. Edward Dudley and Maximillian E. Novak (Pittsburgh: University of Pittsburgh Press, 1972) p. 30. 6. Jonathan Z. Smith, Map Is Not Territory: Studies in the History of Religions (Leiden: Brill, 1978), pp. 134-36. 7. Jean-Pierre Vernant, Mythe et pensee chez les Grecs (Paris: Maspero, 1965) 1:191. 8. Smith, Map, p. 97. 9. An excellent elaboration of this notion is Peter Berger, The Sacred Canopy: Elements of a Sociological Theory of Religion (Garden City, N.Y.: Anchor, 1969), p. 4 on the concurrent socializing processes of externalization, objectivation, and internalization. See also Jonathan Z. Smith, Imagining Religion: From Babylon to Jonestown (Chicago: University of Chicago Press, 1982), p. 50; and Mary Douglas, Purity and Danger: An Analysis ofthe Concepts of Pollution and Taboo (London: Routledge & Kegan Paul, 1966; reprint 1980), pp. 5, 36-37. 10. Mircea Eliade, The Sacred and the Profane: The Nature of Religions, tr. Williard R. Trask (New York: Harcourt, Brace & Jovanovich, 1959), pp. 29 and 48: emphases are Eliade's. Cf. idem, Images etsymboles: essais sur le symbolisme magico-religieux (Paris: Gallimard, 1952,1984), pp. 48-49. 11. Smith, Map, p. 97; 136. 12. Ibid., p. 109. Cf. Edmund Leach and Alan Aycock, Structuralist Interpretations of Biblical Myths, 2ded. (Cambridge: Cambridge University Press, 1983), pp. 5, 16, 37,41, on the biblical wilderness; and Wendy Doniger O'Flaherty, Dreams, Illusion, and Other Realities (Chicago: University of Chicago Press, 1984), p. 302, on myths that point to a "more relevant reality." 13. Smith, Map, pp. 98, 109. 14. Charles H. Long, "Primitive/Civilized: The Locus of a Problem," History of Religions 20, nos. 1 - 2 (August/November 1980): 45: "I use the phrase 'empirical others' to define a cultural phenomenon in which the extraordinariness and uniqueness of a person or culture is first recognized as a negativity." 15. Smith, Imagining Religion, pp. 38-39; and Douglas, Purity, pp. 4, 161. 16. Douglas, Purity, pp. 53,160-65. 17. See Roger Caillois, Vhomme etlesacre (Paris: Gallimard, 1950), pp. 56-57, on the contagion of both the criminal and the executioner in traditional societies. 18. Mircea Eliade, Zalmoxis, the Vanishing God, tr. Williard Trask (Chicago: University of Chicago Press, 1972, 1986), p. 4: In medieval Europe, the term "wolf" was applied to: (1) adolescents who, during their initiatory probation, had to hide farfromvillages and live by rapine; (2) immigrants seeking a new territory in which to settle; (3) outlaws or fugitives seeking a place of refuge. Under the laws of Edward the Confessor (ca. A.D. 1000), the condemned criminal had to wear a wolfheaded mask (wulfeshened), and the gibbet was called the "wolf's-head tree" in Anglo-Saxon. 19. This term for desert wasteland, the antitype of the polis, is echoed in the name of a people Strabo describes who live at the extreme southern end of the earth.

Notes to Pages 13-15

213

These people, called the Agrioi, are also called Cynamolgi (Dog-Milkers, from the GreekKunamolgoi): Strabo Geographica 16.4.10. 20. Sophocles Philoctetes 691-717. 21. In Aristotle, no one outside the periphery of the polis could ever fully realize his humanity, since the conditions of life unregulated by law—the law of the inhabitant of the polis, the zöon politikon—precluded such: Aristotle Politica 1252a. 1 - 2 3 . 22. McKim Marriott, "Hindu Transactions: Diversity Without Dualism," in Transaction and Meaning: Directions in the Anthropology of Exchange and Symbolic Behavior, ed. Bruce Kapferer (Philadelphia: Institute for the Study of Human Issues, 1976), pp. 109-14. 23. Collection of the Middle-Length Sayings (Majjhima-Nikaya), 3 vols., tr. I. B. Homer (London: Luzac & Co., 1959), vol. 1, disc 2. Cf. Thomas Mann, The Magic Mountain, tr. H. T. Lowe-Porter (New Yqrk: Vintage, 1969), pp. 274-76. 24. Manu Smrti 10.4, in TheManusmrti with the ( Commentary of Kullüka Bhâtta, ed. Pandit Gopala Sastri Nene, Kashi Sanskrit Series, no. 114 (Varanasi: Chowkhamba Sanskrit Series Office, 1970), p. 535. 25. Revelation 12:6-14. 26. Jacques LeGoff, Uimaginaire medieval (Paris: Gallimard, 1985), pp. 6 1 - 6 4 . 27. Ibid. pp. 6 4 - 6 5 . 28. Ibid., pp. 6 6 - 7 4 . On the wild forest/ordered city polarity in India, see Charles Malamoud, "Village et foret dans l'ideologie de l'Inde brahmanique," in Cuire le monde: rite et pensee dans l'Inde ancien (Paris: Editions de la Decouverte, 1989), pp. 9 5 - 9 6 . 29. Long, "Primitive," pp. 4 6 - 4 7 , in his discussion of the overlap between the late medieval wild man and ideas of insanity, cites Michel Foucault (Madness and Civilization, tr. Richard Howard [New York: Random House, Vintage Books, 1973], p. 11) on "ships of fools": "But water adds to this dark mass its own values; it carries off, but it does more, it purifies. Navigation delivers now to the uncertainty of fate; on water, each man's voyage is at once a rigorous division and an absolute Passage. In one sense, it simply develops, across a half-real, half imaginary geography, a madman's liminal position on the horizon of medieval concern." 30. LeGoff, Uimaginaire, p. 64; cf. "Aspects savants et populaires des voyages dans I'au-dela au Moyen Age," pp. 103-19. On the development of the idea of Purgatory, see pp. 1 2 8 - 3 2 and idem, Pour un autre Moyen Age: Temps, travail et culture en Occident (Paris: Gallimard, 1977), "L'occident medievale et l'ocean Indien: un horizon onirique," p. 298. On the effects of crossing beyond the ocean horizon of the Atlantic and Indian Oceans, see Marie-Dominique Chenu, Nature, Man and Society in the Twelfth Century, tr. and ed. Jerome Taylor and Lester K. Litde (Chicago: University of Chicago Press, 1968, Chicago: Phoenix Books, 1979), p. 181. 31. LeGoff, Uimaginaire, "Le merveilleux dans l'Occident medieval," pp. 1 7 - 3 9 and idem, Pour un autre, pp. 282,292. The Indian Ocean is depicted as a mare clausum in the maps of Ptolemy, for example. 32. Gaston Bachelard,LapoetiquedeI'espace (Paris: Presses Universitaires, 1974), pp. 198-99. A similar metaphor for an "internal immensity" is the Taoist "chaos gourd": see below, chapter 8, part 1.

214

n o t e s t o pages 1 0 - 1 3

33. LeGoff, Uimaginaire, pp. 17-39, especially pp. 17-18, 24, 28-29; and White, "Forms of Wildness," pp. 35-36. 34. Philippe Senac, Ulmage de IAutre: Histoire de VOccident medievale face a lyls. lam (Paris: Flammarion, 1983), pp. 55-57, 66, 71-72; LeGoff, Pour un autre, p. 24. The Muslims depicted the West in a similar way, also believing the Indian Ocean to be an enclosed sea: LeGoff, Pour un autre, p. 284, n. 16; p. 289.

35. Douglas, Purity, p. 56. 36. The Concise Oxford Dictionary of the English Language, 6th ed. (1976), s.v. "horde" and "Urdu." Cf. R. A. Skelton, Thomas A. Marston, and George D. Painter, The Vinland Map and the Tartar Relation (New Haven: Yale University Press, 1965), p. 87, n. 1 to par. 38. 37. Remi Mathieu, Etude sur la mythologie et Vethnologie de la Chine ancienne: Traduction annotee du Shanhaijing, 2 vols., Memoires de PInstitut des Hautes Etudes, vol. 22 (Paris: College de France, Institut des Hautes Etudes Chinoises, 1983), l:xliv. 38. Smith, Map, pp. 134-35. 39. And, as Michel Foucault has so convincingly shown, the mentally ill too are condemned to the inside-outside of the asylum. The earliest mental institutions, established at the end of the Middle Ages, were often converted leprosaria; before this, the insane were left to wander outside the walls of the city, or placed at the mercies of the elements on "ships of fools": Foucault, Madness and Civilization, cited in Long, "Primitive," pp. 46-47. 40. Mircea Eliade (.History of Religious Ideas, 3 vols., tr. by Williard R. Trask (Chicago: University of Chicago Press, 1978), vol. 1, p. 33 places the origin of this commensalist relationship at 7500 B.C.; Edward Tryjarski ("The Dog in the Turkic Area: An Ethnolinguistic Study," Central Asian Journal 23:3-4 [1979]: 298, n. 4), proposes the period between 8000 and 6000 B.C. 41. Saara Lilja, Dogs in Ancient Greek Poetry, Commentationes Humanarum Litteratum, no. 56 (Helsinki: Societas Scientiarum Fennica, 1976), p. 12; Wilhelm Köppers, "Der Hund in der mythologie der zirkumpazifıschen Volker," Wiener Beitrage zur Kulturgeschichte und Linguistik 1 (1930): 381. 42. Clifford Geertz, The Interpretation of Cultures (New York: Basic Books, 1973), "The Impact of the Concept of Culture on the Concept of Man," pp. 33-54 and "The Growth of Culture and the Evolution of the Mind," pp. 55-84; Berger, Sacred, p. 4: this is not a simple case of ontogeny recapitulating phylogeny. 43. Spuds MacKenzie, the bull terrier "party dog" who sells a popular brand of American beer is but one avatar of a perennial type. On theological uses of the dog, see Friedrich von Hügel, Essays and Addresses on the Philosophy of Religion, First Series (New York: E. P. Dutton & Co., 1923), pp. 102-3, cited by John H. Hick (Philosophy of Religion, 2d ed. [Englewood Cliffs: Prentice-Hall, 1973], p. 71). 44. In certain of these myths, man is originally the fur covered animal and the dog naked, with the hide changing owners at a crucial moment, often through the trickery of the devil or some other adversary: Manabu Waida, "Central Asian Mythology of the Origin of Death: A Comparative Analysis of its Structure and History," Anthropos 77:5-6 (1982): 670 and passim; Siegbert Hummel, "Der

Notes to Pages121-23

26 7

Hund in der religiösen Vorstellungswelt des Tibeters," Paideuma 10, no. 8 (November 1958): 499-500. 45. Bruce Lincoln, "The Hellhound," Journal of Indo-European Studies 7 : 3 - 4 (Fall/Winter 1979): 273-85. 46. Ibid., especially p. 274, n. 3, on Mayrhofer's acceptance of Kuiper's theory that the Sanskrit svan is derived from an Austroasiatic source. See also Tryjarski, "Dog" pp. 304-10; Oswald Szemerenyi, "Iranica," Zeitschrift fur Deutsche Morgenldndichen Gesellschaft 101 (1951): 2 1 0 - 1 1 ; Wolfram Eberhard, The Local Cultures ofSouthern and Eastern China (Leiden: Brill, 1968), p. 464 and China und seine westlichen Nachbarn (Darmstadt: Wissenschafdiche Buchgesellschaft, 1978), p. 149; Eduard Erkes, "Der Hund im alten China," ToungPao 38 (1944): 199, nn. 2 and 3; and Fritz Flor, "Haustier und Hirtenkulturen: Die Hundesucht," Wiener Beitrage zur Kulturgeschichte und Linguistik 1 (1930): 49. 47. Erkes, "Hund," pp. 187-98; Lilja,Dö^ p. 11. 49. Bernfried Schlerath, "Der Hund bei den Indogermanen," Paideuma 6, no. 1 (November 1954): 36; Lincoln, "Hellhound," pp. 274-85. The canine psychopomp is also common to central Asian mythology: Eberhard, Local, pp. 4 6 2 - 6 5 ; and Hummel, "Hund," pp. 500-501. The excellent work of Mahasti Ziai Afshar, The Immortal Hound: The Genesis and Transformation of a Symbol in Indo-Iranian Traditions (Ph.D. diss., Harvard University, 1988), presents a witty and illuminating discussion of the symbolism of the dog in Vedic and Avestan, as well as in a wide array of other ancient and medieval Indo-European traditions. 50. As Bruce Lincoln has so apdy pointed out, dogs represent the threshold between speech and silence: their growls and barks are perhaps the onomotopoetic origin of certain Indo-European terms for dog: "Hellhound," pp. 184-85. 51. Hittite law says of the proscribed man that he "has become a wolf": Eliade, Zalmoxis, "The Dacians and the Wolves," p. 4. Ulfhedhnar, "wolf-skin men," was the term used for members of Germanic military fraternities; these are called "twopawed wolves" in Iranian traditions. According to Wikander, the Achaemenidian ancestral family was called saka haumavarka, "those who change themselves into wolves in the ecstasy brought on by haoma," Stig Wikander, Der Arische Mannerbund (Lund: H. Ohlsson, 1938), p. 64, cited in Eliade, Zalmoxis, p. 9. On werewolves in Herodotus, Petronius, and other ancient sources, as referring to annual ceremonies, initiations, etc., ibid., p. 17. Herodotus (Historia 4.107) discusses the Neuri, a nation of werewolves in Asia. Later, the Huns are identified with the Neuri, by the fifth-century Christian historian Philostorgius: Otto MaenchenHelfen, The World of the Huns, ed. Max Knight (Berkeley: University of California Press, 1973), p. 8. On anthropophagic lycanthropy in India, Iran, Greece, North America, Africa, etc., see Jan Przyluski, "Les confreries des loups-garous dans les societes indo-europeennes," Revue de lyHistoire des Religions 121 (1940): 128-45; cf. Mary R. Gerstein, "Germanic Warg: The Oudaw as Werwolf," in Myth in IndoEuropean Antiquity, ed. Gerald James Larson (Berkeley: University of California Press, 1974), especially pp. 143-56; Hans-Peter Hasenfratz, "Der indogermanische 'Mannerbund,'" Zeitschriftfur Religions- und Geistesgeschichte 34 (1982):

216

n o t e s t o p a g e s 98-100

148-63; and Louis Gernet, Anthropologic de la Grece ancienne (Paris: Maspero 1968; reprint Flammarion, 1982), "Dolon le loup," pp. 201-23, on lycanthropy i n the Iliad. 52. Eliade, Zalmoxis, p. 1: According to Strabo (Geographica 7.3.12), the name of the Dacians is derived from daoi, the Phrygian word for wolf a term related to the verb "to strangle." In the Alexander Romance, the name of Lydian Kandaules (kunagches in Greek), the brother of Queen Candace (Kandake), literally means "dog-strangler," a term which has a Sanskrit cognate (Manfred Mayrhofer, Kurzgefasstes etymologisches Worterbuch des Altindischen (A concise etymological Sanskrit dictionary), 4 vols. [Heidelberg: Carl Winter, 1956-80], 3:403). Related to the term daoi was that employed for the Scythians dwelling to the east of the Caspian, who were called Dahae and Daoi by Latin and Greek geographers, both cognates of the Iranian (Saka) term for wolf, dahae. The Greek Hyrcania (the region near the Caspian Sea from which came the prized Hyrcanian hound of antiquity) is a Hellenization of the Iranian Vehrkana or Varkana, "country of wolves," whose peoples were called Hyrkanoi, "wolves," by Greco-Roman authors. The same pattern is apparent in the names of the Lycaones of Arcadia, Lycaonia or Lucaonian in Asia Minor, and Apollo Lykagenes ("he born of the she-wolf" i.e., Leto in the shape of a wolf). Remus and Romulus are sons of the wolf god Mars who are suckled by the she-wolf of the Capitol. Another parallel to this theme is found in the myth of the birth of Cyrus, who was nursed, according to Herodotus, by a woman named Spakö ("She-Dog"): Eliade, Zalmoxis, pp. 1 - 2 0 . See below, chapter 2, note 14. 53. So for the Greeks, there was a split vision of foreigners which vacillated between a fascination for "Scythian nomadism" and a dread fear of "barbarianism" (Smith, Map, p. 131), or between what John Block Friedman calls Asianism and Atticism (The Monstrous Races in Medieval Art and Thought [Cambridge: Harvard University Press, 1981], p. 35). 54. On this theme, see Stanley J. Tambiah, "Animals are Good to Think about and Good to Prohibit," Ethnology 8 (1969): 4 2 3 - 5 9 ; and Dan Sperber, "Pourquoi les animaux parfaits, les hybrides et les monstres sont-ils bons â penser symboliquementrUHomme 15, no. 2 (April-June 1975): 5 - 3 4 . 55. This shading of categories, from the animal to the human, is especially transparent in the Hindu ranking of castes. Members of the lowest of the four conventional castes, the südras, "are to hold themselves at a distance from the higher castes, in a space that is difficult to qualify as either human or subhuman, at a level in the chain of being that undoubtedly locates them below that of domestic animals. Thus, one consequence of the ritual origin of the hierarchy of beings is that there is no clear break between humanity and animality": Madeleine Biardeau, UHindouisme: Anthopologie d'une civilisation (Paris: Flammarion, 1981), p. 44. 56. Smith, Map, pp. 2 4 1 - 6 4 . 57. Ibid., pp. 2 5 0 - 5 2 . 58. Ibid., pp. 2 5 3 - 5 9 : this approach has, according to Smith, held the field in the "science" of comparative religions since the nineteenth century, with the ahistorical, morphological emphasis drawing direcdy on the science of comparative anatomy. Goethe's use of such models, transparent in his writings on the Urpflanze,

Notes to Pages121-23

26 7

has moreover exerted much influence on the spirit of Eliade's systematic approach "patterns in comparative religion." 59. Sperber, "Pourquoi," pp. 5 - 3 4 . 60. Ibid., p. 27: "One too many paws means one hundred more visitors; one too many heads and the sky's the limit. The half-animal half-human monsters, the mermaid, ape-man, etc. bring in the highest receipts." 1. A perennial description of Dog-Men has them barking rather than speaking, or barking after a few words of human speech: this echoes the place of the hellhound in Indo-European mythology, whose growl places it on the threshold between silence and speech. In Sophocles' play, the noble Philoctetes craves human speech as much as he does anything else in his desire to free himself from savage nature. On the "high languages" of the ancient and medieval worlds (Greek, Latin, Sanskrit, Chinese), and writing as elite institutions, see Carlo Ginzburg, The Cheese and the Warms: The Cosmos of a Sixteenth Century Miller; tr. John and Anne Tedeschi (New York: Penguin, 1984), pp. xxi-xxiv, 59. 62. However, a tenth-century geography composed in the world region that was the major vector for such mythology—i.e., Afghanistan in central Asia—finds no monsters whatsoever within or outside o f its borders: Hudüd al-Alam, aThe Regions ofthe World": A Persian Geography; 372 A.H.—982 A.D., tr. and explained by V. Minorsky (Karachi: Indus Publications, 1980; reprint of 1937 translation). 63. Isidore of Seville, Etymologiae 11.3.1-3. The German term for monster, Ungeheuer; suggests the sorts of layered meanings we have found in the other terms we have examined. Literally meaning "not belonging to the tribe," Ungeheuer means, by extension, anything eerie or strange, a monster: Michel Meslin, gen. ed. Le merveilleux: LHmaginaire et les croyances en Occident (Paris: Bordas, 1984), p. 87. 64. William Shakespeare, Julius Caesar; act 1, sc. 3, lines 1-36. 65. Long, "Primitive," pp. 4 8 - 4 9 and passim. 66. Properly speaking, the Mleccha barbarians are the Greco-Bactrian (Tavana), Indo-Scythian (Saka), and Yue-chi (Kushan) conquerers of India of the 2d century B.C. to the 2d century A.D.: Romila Thapar, History of India (New York: Penguin, 1968), vol. 1, pp. 97,184. 67. Mathieu, Shanhai jing l:xl-xlv. Cf. p. xliii: "II n'est en effet pas rare qu'un animal soit percu tantot comme un oiseau, tantot comme un quadruped ou qu'un animal aquatique soit classe parmi les quadrupeds ou les poissons selon ses pattes ou son cri. Cependant, la contusion des genres sur laquelle nous nous attarderons le plus est celle qui se presente â l'interieur du regne animal entre l'homme et les betes . . . II est clair que les anciens Chinois percoivent de l'homme chez Panimal et de la bete chez Phomme, ils n'ont done pas de raison de les distinguer ou de les opposer . . . De meme, au niveau social, les barbares ont-ils des 'sentiments d'oiseau et de betes sauvages' et le peuple est-il semblable â un banc de poissons." 68. Arthur P. Wolf, "Gods, Ghosts and Ancestors," in Religion and Ritual in Chinese Society\ ed. Arthur P. Wolf (Stanford: Stanford University Press, 1974), p. 174; and Mathieu, Shanhai jing, 1:1. 69. Victor Turner, The Forest of Symbols: Aspects ofNdembu Ritual (Ithaca: Cornell University Press, 1977), "Betwixt and Between: The Liminal Period in Rites de Passage," pp. 103-5. to

218

n o t e s t o p a g e s 98-100

70. Ibid., pp. 105-6. 71. The phrase, "the symbol gives rise to thought," belongs to Paul Ricoeur (The Symbolism of Evil [Boston: Beacon Press, 1967], pp. 347-57). Smith (Map, pp. 294 299) has offered a variation on Ricoeur that is germane to our concerns: "incongruity gives rise to thought." CHAPTER 2

1.E. A. T. Wallis Budge, The Contending of the Apostles (London: Oxford University Press, 1935), pp. 171-77. 2. On Andrew, see Zofia Ameisenowa, "Animal-headed Gods, Evangelists, Saints and Righteous Men," Journal of the Warburg and Courtauld Institutes 12 (1949): 42; and Francis Dvomik, The Idea ofApostolicity in Byzantium and the Legend ofthe Apostle Andrew (Cambridge: Harvard University Press, 1958), p. 203. On the Nestorian origins of the Passio Bartholomaei, see Dictionnaire d'Histoire et de Geographic Ecclesiastiques (Paris: Letouzey, 1932), s.v. "Barthelemy, Apötre." For a very detailed discussion of the hagiographies of Bartholomew and the other saints of these traditions, see R. A. Lipsius, Die Apokryphen Apostelgeschichten und Apostellegenden, 3 vols. (Braunschweig: C. A. Schwetschke und Sohn, 1883-90), 2, pt. 2, pp. 54-106. The post-conciliar "diaspora" of the Coptic and Nestorian churches is mapped out in The Times Concise Atlas ofWorld History (rev. ed., gen. ed. Geoffrey Barraclough [London: Times Books Ltd., 1984; Maplewood, N.J.: Hammond, 1984]), pp. 38-39. 3. Lipsius, Apokryphen Apostelgeschichten, 2, pt. 2:54-71. 4. The later transmission of this cycle of legends is oudined in the finest of recent works on the monster traditions. This is Friedman, Monstrous Races, p. 228, n. 33. See also E. A. T. Wallis Budge, The Book ofthe Saints ofthe Ethiopian Church, 2 vols. (Hildesheim and New York: Georg Olms, 1976), l:xi-xix, 2 - 3 . 5. Basil II, Emperor of the East (976-1025) is the author of the Menologium through which these apocryphal legends were introduced into Russia. It was Basil's "forced conversion" of Prince Vladimir of Kiev that made Russia a Christian kingdom. See Henri Gaidoz, "Saint Christophe â la tete de chien en Irlande et en Russie," Memoires de la Societe Nationale des Antiquaires 76 (1924): 2 0 3 - 4 ; and Konrad Richter, Der deutsche S. Christoph: Eine historisch-kritische Untersuchung (Berlin: Mayer und Müller, 1896), p. 25. For Muslim traditions, see Rudolf Wittkower, "Marvels of the East: A Study in the History of Monsters," Journal of the Warburg and Courtauld Institutes 5 (1942): 175. 6. The Ethiopic Gadla Hawaryat was innovative in the sense that its martyrologies were compiled in order that they be read on the death-days of the saints. Consonant with this usage, there is an increased emphasis on the miraculous in the saints' acts: A. F. J. Klijn, The Acts of Thomas (Leiden: Brill, 1962), p. 18. On the dates for Bartholomew's martyrdom, see also Lipsius, Apokryphen Apostelgeschichten, 1:222. In Budge, Saints ofthe Ethiopian Church, 1:2, the date given is the first day of Maskaram, or September 7. In Rome, Bartholomew's day was celebrated on August 24 or 25: Dictionnaire d'Histoire et de Geographic Ecclesiastiques, s.v. "Barthelemy, Apötre." 7. July 25, the date of the heliacal rising of Sirius and the beginning of the can-

Notes to Pages121-23

26 7

icular, or dog days, was one of the two pivots of the ancient and medieval year (the other pivot being February 3). A masterful study of this phenomenon, viewed through the lens of the works of Rabelais, is Claude Gaignebet, A plus hault sens: I'esoterisme spirituel et charnel de Rabelais, 2 vols. (Paris: Maisonneuve, 1986), especially 1:253-336; 4 5 8 - 6 9 . On the same subject, see Giorgio Diez de Santillana and Hertha von Dechend, Hamlefs Mill: An Essay on Myth and the Frame of Time (Boston: Gambit, 1967). See also Jean-Claude Schmitt, Le saint levrier: Guignefort guerisseur d'enfants depuis leXIII'siecle (Paris: Flammarion, 1979), pp. 206-10. For the particular cases of saints Christopher and Mercurius, see Pierre Saintyves, Saint Christophe, successeur d'Anubis, dyHermes etd'Heracles (Paris: Emile Nourry, 1936), pp. 4 1 - 4 5 . 8. These toponyms are Ptolemy's interpretation of Egyptian terms: Claudii Ptolemaei Geographica, ed. Carl Müller (Paris: Firmin Didot, 1883), p. 717 and notes, according to which the Egyptian sacred toponym is Pi-anup (or Ha-ânup), "city of Anubis the jackal." The passage is in Geographica 4.5.29; there follows, in 4.5.31, a reference to Lycopolites, located nearby. On the same identification, see Plutarch De Iside et Osiride 380b; Strabo Geographica 17.1.19, 22,40; Pliny Naturalis Historia 5.49. On the mummified jackals and dogs found in Cynopolites, see Freda Kretschmar, Hundestammvater undKerberos, 2 vols. (Stuttgart: Strecker und Schroder, 1938), 2:218. See also Pauly-Wissowa, vol. 2 3 [ = 12.1] (1924), s.v. "Kynopolis." A second town of the same name, also dedicated to Anubis, was locted in the Nile delta: William Smith, ed., Dictionary ofGreek and Roman Geography (London: n.p., 1856), s.v. "Cynopolis." 9. Plutarch, De Iside et Osiride 14. See also Schlerath, "Hund," p. 29; Gaidoz, "Saint Christophe," p. 210, and Ameisenowa, "Animal-headed Gods," p. 45. 10. Aelianus De natura animalium 10.45 and Pliny NaturalisHistoria 5.57 and 18.167. Egyptian traditions dating from the third millennium identify the heliacal rising of Sothis (Sirius) with the flooding of the Nile: Danielle Bonneau, La crue du Nil, divinite egyptienne (332 av.-641 ap.J.-C.) (Paris: Klincksieck, 1964), pp. 3 2 33. On the Sothic year and Sirius, see also Giorgio Diez de Santillana and Hertha von Dechend, "Syrius as a Permanent Center in the Archaic Universe," in Eternitd e storia, ed. Enrico Castelli (Florence: Vallecchi, 1970), pp. 153-55. On the identification of Sothis with Isis, in the context of the Osiris myth cycle, see Edmund Leach, "Why did Moses Have a Sister?" in Edmund Leach and D. Alan Aycock, Structuralist Interpretations, p. 48. 11. Ameisenowa, "Animal-headed Gods," pp. 2 6 - 2 9 , 4 4 - 4 5 . 12. Besides New Empire epigraphic descriptions, such cynocephalic apes are also described in chapters 15 and 16 of the Book ofthe Dead. Plate 4 of E. Naville and Sir Peter LePage Renouf, eds., The Book ofthe Dead (Paris, Leipzig: n.p., 1907), pp. 3 5 - 3 6 reproduces an illustration, from a papyrus (Leiden Museum, no. 2), of six cynocephalic apes flanking the sun (god) Osiris who rises between his sisters Isis and Nephthys. An image of such a Simia hamadryas was worshipped in Egyptian Hermopolis: Strabo Geographica 17.1.40. See also A. Piankoff, "Saint Mercure, Abou Sefein et les Cynocephales," Bulletin de la Societe dArchaeologie Copte 8 (1942): 2 2 - 2 3 . See plate 4. 13. Herodotus Historia 1.110-22; Xenophon Cyropedia 1.4.15, cited in Carla

220

n o t e s t op a g e S98-100

Mainoldi, Ulmage du loup et du chien dans la Grece ancien d'Homere d Platon (Paris: Ophrys, 1984), pp. 150-51. 14. In Nestorian tradition, Saint Bartholomew is martyred at the order of a certain Astyages, in Cilicia (Turkey), a toponym also possessed of a "lupine" etymology: Dictionnaire d'Histoire et de Geographie Ecclesiastiques, s.v. "Barthelemv, Apötre." See also Lipsius, Apokryphen Apostelgeschichten, vol. 2, pt. 2, pp. 67-68, 71-72. In the light of the Cyrus myth and other sources, Bartholomew's martyrology may be a transposition of ancient Persian mythology. On the perenniality of this theme in Indo-European traditions in general, see the excellent work of Mahasti Ziai Afshar {Immortal Hound, pp. 117-63). See above, chapter 1, nn. 5 1 - 5 2 ; below, notes 37,46, and 57 to this chapter; and chapter 6, n. 6. An Indian instance of lycanthropy may be glimpsed in the monstrous epic hero Bhlma's epithet of Vrkodara (Wolf-belly). The Mahâbhârata (8.27-30) also associates Cyrus's Medians (if they are to be identified with the Madras) with dogs and women in an unusual way: see below, chapter 6. For later Hellenistic and medieval references to this phenomenon, see Meslin, Le merveilleux, pp. 32 and 136-37; and Claude-Catherine and Gilles Ragache, Les loups en France: Legendes et realite (Paris: Aubier, 1981), p. 8. 15. Ragache, Les loups, p. 6. The account of the "wolfman" Dolon's night raid is found in the Iliad 10.208-488. On this, see Gemet, Anthropologic, pp. 201-23. A very detailed treatment of both cynanthropy and lycanthropy in ancient Greece is W. H. Rorscher, "Das von der 'Kynanthropie' handelnde Fragment des Marcellus von Side," in Abhandlungen der Philologisch-historische Klasse der Königlich Sachsischen Gesellschaft der Wissenschaften (Leipzig) 17 (1897): 1 - 9 2 . 16. They were called "two-legged wolves" (vahrka-bizangra) in the Avesta (Tasna 9.18, 21): see Hasenfratz, "Der indogermanische 'Mânnerbund,'" p. 151. 17. Meslin, Le merveilleux, pp. 137-38. 18. Leopold Kretzenbacher, Kynokephale Dâmonen Südosteuropâischer Volksdichtung: Vergleichen Studien zu Mythen, Sagen, Maskenbrâuchen und Kynokephaloi, Werwolf und siidslawische Pesoglavci (Munich: Rudolf Trofenik, 1968), pp. 85, 100. Kretzenbacher's excellent monograph is the most complete discussion of indigenous European cynocephalic traditions written to date. 19. Friedman, Monstrous Races, p. 165. 20. On Ethiopia and India as Lesser and Greater India, respectively, in the context of patristic geography, and on the division of India into "minor," "media," and "superior" divisions, see Wittkower, "Marvels of the East," p. 161, n. 4, and Skelton, VinlandMap, pp. 68,133. 21. Ctesias wrote his Indika (summarized by Photius in Rene Henry, Ctesias, le Perse, ITnde: Les Sommaires de Photius [Brussels: Lebegue, 1947]) during an eightyear period of service as physician to the court of the Persian Achaemenidian Artaxerxes Mnemon II, in 3 9 8 - 9 7 B.C.: Fernando Tola and Carmen Dragonetti, "India and Greece Before Alexander," Annals of the Bhandarkar Oriental Research Institute 67 (1985): 172. Ctesias's account is not based on personal experience, but rather on tales told by merchants and other visitors to the court. In many cases, Ctesias or his informants accepted myth as ethnology. See below, chapter 3, part 1.

Notes to Pages121-23267 22. Lilja, Dogs, p. 11. 23. Herodotus Historia 1.140. 24. Such dogs are mentioned thirty times in the Iliad (Lilja, Dogs, p. 17) and in Mahâbhârata 5.139.51,6.95.50 and elsewhere. 25. Lilja, Dogs, p. 11. In Tibet, it may have been these same dogs who were identified, by Europeans following Megasthenes, with the griffins who guarded the "gold-digging ants" (Tibetan miners!): J. W. McCri ndle, A ncient India as Described m Classical Literature (Westminister: A. Constable, 1901; reprint New Delhi: Oriental Reprints, 1979), p. 138, citing Aelianus,IV natura animalium 4.19, 27; 8.1; 10.45. 26. Indian or Persian dogs are already mentioned in Herodotus Historia 7.187; Aristotle Historia Animalium 606b.29; Xenophon De venatione 9.1, 10.1; and Ctesias Indika 5. 27. Megasthenes, who lived in India shordy after the Alexandrian conquests, refers to these dogs asHartspai, a probable Hellenization (via the Persian medium) of the Sanskrit *hari-sva: Joseph Marquart,Die Benin-Sammlung des Reichsmuseumfur Volkerkunde in Leiden, Veröffendichungen des Reichsmuseums fur Volkerkunde in Leiden, ser. 2, no. 7 (Leiden: Brill, 1913), p. 207. Among other ancient authors, Aelianus De natura animalium 4.19, 8.1, Diodorus of Sicily Bibliotheca Historica 17.92, and Strabo Geographica 15.1.31 describe these tiger hounds, who later come to be associated with the Cynamolgi (Dog-Milkers). Indian accounts of such dogs are found in Jaimintya Brâhmana 2.442 and Râmâyana 2.70.21. 28. Hesiod Opera et Dies 6 0 4 - 5 . Cerberos began as a simple watchdog of the house of Hades: Lilja, Dogs, p. 39. 29. Hesiod Homeric Hymn to Hermes 193-95. Geryon, the herdsman of the red oxen (identified with the dead) in Iberia, had a watchdog named Orthros: Hesiod Theogonia 309. 30. Lilja, Dogs, pp. 36,43. 31. Pliny NaturalisHistoria 7.1.21; 8.142-53. 32. Ibid., 7.1.21-32; these chapters on "monsters" precede the proper subject of this book, the races of (normal) human beings. 33. Wittkower, "Marvels of the East," pp. 167-68. 34. Judith Kidd, "A Medieval Legacy of Aesthetic Conflict," Theology (G.B.) 85 (1982): 23; and Wittkower, "Marvels of the East," p. 177. 35. On the many traditions of the legends of Bartholomew and Alexander, see Lipsius, ApokryphenApostelgeschichten, vol. 2, pt. 2, pp. 54,65, 7 6 - 8 6 , 9 0 - 9 1 ; and J. Flamion, Les Acts Apocryphes de lApötre Andre, Receuil des Travaux, no. 33 (Louvain: Bureau de Recueil, 1911), p. 313. The Arabic Synxarium version of the legend, nearly identical to the Ethiopic but a century earlier, is translated in Agnes Smith Lewis, Mythological Acts of the Apostles (Cambridge: Cambridge University Press, 1904), pp. xix-xxv, 1 , 1 1 - 2 4 . The many pre-Arabic apocryphal acts of those aposdes—Andrew, Bartholomew, Matthias, Matthew, Peter, and Thomas—whose missions brought them into contact with Cynocephali or canine minions of the devil are found in Wilhelm Schneemelcher, ed., New Testament Apocrypha, 2 vols., tr. Edgar Hennecke (Philadelphia: Westminster, 1965), 2:178-79, 291-95, 3 9 0 400, 576-77; Montague Rhodes James, The Apocryphal New Testament: Being the

222

n o t e s t o p a g e s 98-100

Apocryphal Gospels, Acts, Epistles andApocalypses with Other Narratives and Fragments (Oxford: Clarendon Press, 1960), pp. 3 3 7 - 3 9 , 4 5 3 - 5 8 , 4 7 1 - 7 2 ; and K l i j n ^ o y ^ Thomas, 13-29. See also Adolf Jacoby, "Der hundsköpfıge Dâmon der Unterwelf' (Archiv für Religionswissenschaft 21 [1922], pp. 219-25). See plate 3. 36. Piankoff, "Saint Mercure," p. 18. 37. William Wright, Apocryphal Acts ofthe Apostles, 2 vols. (London: Williams & Northgate, 1871; reprint Amsterdam: Philo Press, 1968), 2:115. This Syriac version's mention of a "City of Dogs" is clearly the earliest traceable source of the later accounts of the cynocephalic Abominable. This version is dated to the fourth to sixth centuries (1 :xiv), the same period in which other Syriac sources begin to mention Cynocephali living to the north of the Caucasus: see, for example, the sixthcentury "Church History" of Pseudo-Zachariah the Rhaetor, bishop of Mitylene (Die sogenannte Kirchengeschichte des Zacharios Rhaetor, ed. and tr. Karl Ahrens and G. Kriiger, Teubneriana, vol. 241, pt. 3 [Leipzig: Teubner, 1899], p. 253). 'Irqa's location would have corresponded to the northern frontier of the kingdom of Armenia. "Indigenous" Armenian traditions of dog ancestry or divine cynocephaly are identified by Chungshee Hsiun Liu, "On the Dog Ancestor Myth in Asia," Studia Serica 1 (1941): 283 [hereafter referred to as Liu (1941)]; and Kretzenbacher, KynokephaleDamonen, p. 91, n. 27 and p. 126. fIrqa, the "City of Dogs," becomes, in Ethiopic hagiographies of Andrew, Axis, Archaias, or Achaia: Lipsius, Apokryphen Apostelgeschichten, 1:621. 38. Lipsius, Apokryphen Apostelgeschichten, 1:14-15; and Dvornik, Idea of Apostolicity, p. 199. The earliest sources on Andrew and Bartholomew place their martydoms in Acradis (Kurdistan, south of Armenia) and Lykaonia (west-central Turkey) respectively (Lipsius, Apokryphen Apostelgeschichten, 1:28,603-4,621; and 2, pt. 2:54-57, 71-72). According to Eusebius, Andrew evangelized Scythia (1:14), while Thomas's mission was to Parthia, Matthew's to Ethiopia, and Bartholomew's to "Citerior India" (Historia Ecclesiastica 3.1.1 and 5.10.3). Other traditions send Andrew and Bartholomew to Parthia and Ethiopia, Thomas to India, and Matthew or Matthias to Parthia and Scythia: Schneemelcher, New Testament Apocrypha, pp. 179, 379, 576; James, Apocryphal New Testament, pp. 3 3 7 - 3 8 , 4 5 3 , 4 7 1 ; K\i]n, Acts of Thomas, p. 25. 39. Herodotus Historia 4.18 and 4.106; and Aristode Politico, 1338b.20-22. Soon after the Alexandrian conquests, Megasthenes gives an early account of this same region, alleging that the men and women of the Caucasus have sexual intercourse in the open and eat the bodies of their kinsmen: Strabo Geographica 15.1.55-56. 40. Pliny Naturalis Historia 2.135. 41. VıvgûAenead 1.490 and 11.659. Herodotus Historia 4.18 mentions the Scythae together with Alazones. 42. Parthia and Rome shared a common border following the Roman annexation of Armenia and Assyria to their already existing eastern provinces around the Black Sea, in A.D. 114-17. The western portion of the Parthian empire disappeared in A.D. 224, with its territories shrinking back to the original southern Caspian homeland. During their heyday, the Parthians also expanded eastwards to Taxasila (ca. A.D. 18) and had Kushan India on their eastern doorstep from about

Notes to Pages a .D.

121

-23

26

7

60 to 224. For the cartography of the Hellenistic age, see Nicholas G. L. Hammond, Atlas of the Greek and Roman World in Antiquity (Park Ridge, N.J.: Noyes, 1981) and Times Concise Atlas of World History. 43. Wright, Apocryphal Acts of the Apostles, 2:115; Flamion, Act es Apocryphes de l'Apötre Andre, p. 313; and Salomon Reinach, "Les apotres chez les Anthropophages," Revue d'Histoire et de Litterature Religieuses 9:4 (1904): 316. The Greek "Acts of Andrew and Matthias" are found in Maximillian Bonnet and R. A. Lipsius, A eta Apostolorum apocrypha, 2 vols. (Leipzig: 1891-93; reprint Hildesheim: Georg Olms, 1972), vol. 2, pt. 1:65-116. 44. For Ptolemy's precise locations of the toponym Bartos (Barthos, Parthos), on the Armenian shores of the Black Sea, see Müller, Claudii Ptolomaei Geographica, pp. 798, 874 and 950 (notes to Strabo Geographica 5.1; 5.6 and 5.12). 45. Dvornik, Idea of Apostolicity, pp. 199, 203; Lipsius, Apokryphen Apostelgeschtchten, 2, pt. 2:72-86, 132; and Reinach, "Les Apotres chez les Anthropophages," pp. 307-16. In the Greek and Syriac traditions, it is Andrew who rescues Matthias from the man-eaters; in the Coptic, Arabic, and Ethiopic sources, Andrew and Bartholomew, together with Abominable, confront the man-eaters. 46. On the development of the Bartholomew legend, the connection between an Ethiopian Barthos with the Parthians of the Black Sea region, Colchis as Outer Ethiopia, and Bartholomew's martyrdom by Astyages (who may be identified with the Assyro-Babylonian death god Nergal), see Lipsius, Apokryphen Apostelgeschichten, 1:603-4, 621, and 2, pt. 2:57-71, 76-85, 9 0 - 9 2 , 134-36, 177. See above, n. 14, and below n. 100 to this chapter, and chapter 3, n. 34. On the Andrew legend, see Dvornik, Idea of Apostolicity> pp. 2 0 1 - 7 ; and James, Apocryphal New Testament, pp. 4 5 3 - 5 8 . H. Hartmann (review of Die Ptoembari und Ptoemphani des Plinius by Paul Buchere, in Zeitschrift Jur Ethnologie 2 [1870]: 138) refers to a native Ethiopian myth of dog ancestry, in which human women unite with dogs to produce Dog-Men. 47. Maenchen-Helfen, World ofthe Huns, pp. 94,121,480. See also Lipsius, Apokryphen Apostelgeschichten, 2, pt. 2:76-84; Lewis, Mythological Acts of the Apostles, pp. xx, 11. According to Lipsius (2:2, p. 84) and Lewis (xx), Oasis is to be identified with the region of Asyût, the ancient Cynopolitania (see above, n. 8 to this chapter), in the general region of the later kingdom of Alwâh. On the Nestorian origins of these and other hagiographic traditions, see above, nn. 2 and 14 to this chapter. Coins struck in Asyût, dating from the late Roman empire, depict a cynocephalic ape: B. V. Head, HistoriaNumorum (Oxford: Clarendon Press, 1887), p. 724. 48. Vies des Saints et des Bienheureux, 12 vols., by the Benedictine Fathers of Paris (Paris: Letouzey, 1949), 7:613. 49. The earliest Christian hagiographic sources on Christopher are a fifthcentury Syriac martyrology (Richter, Der deutsche S. Christoph, p. 18), and an Eastern Passio and a Spanish Mozarabic missal and breviary, both dating from the sixth century (Ameisenowa, "Animal-headed Gods," p. 42; and Walter Loeschcke, "Darstellung des Kynokephalen hi. Christophorus," Forschungen zur Osteuropaischen Geschichte 5 [ 1957]: 39). Among his Eastern hagiographies are that of Basil II (his ca. A.D. 1000Menologium [3.89], reproduced in Gaidoz, "Saint Christophe," pp. 2 0 3 - 4 ) ; and an eleventh-century Greek life of St. Christopher reproduced in

224

n o t e s t op a g e S98-100

Analecta Bollandıana 1 (1882): 121-48. This last work, Christopher's "Easterrf hagiography, was later translated into Latin, Irish, and Provençale; J. Schwartz, propos de l'iconographie orientale de S. Christophe," Le Museon: Revue d'Etudes Orientates 67 [ 1954]: 93. His earliest Western hagiography is to be found in an 8th-century manuscript preserved in the Wiirzburg University library (Mp. Th. F28, cited in Loeschcke "Darstellung," p. 39). A scholastic treatment of Christopher's cynocephaly Js Ratramnus of Corbie's ninth-century Epistola de Cynocephalis (J. P. Migne, Patrologia cursus completus . . . Series latina [hereafter referred to as Migne, PL], vol. 121, cols. 1153-56). The greatest (and most convoluted) of Christopher's Western hagiographies is Walter of Speyer's tenth-century Vita et Passio Sancti Christopher Martyris, in monumenta Germaniae Historica (inde ab anno Christi 500 usque ad annum 1500) Poetae Latini Medii Aevi, ed. Karl Strecker and Gabriel Silagi, vol. 5 (Leipzig: Monumenta Germaniae Historica, 1937-39; reprint Munich: Monumenta Germaniae Historica, 1978), 5:66-78. It is also found in Jacobus de Voragine's thirteenth-century Legenda A urea, no. 95 (tr. by Jean-Baptiste M. Rose as La leğende doree, 2 vols. [Paris: Flammarion, 1967], 2:7-11) and in the Bollandists' Acta Sanctorum for July 25 (Antwerp: n.p., 1729; reprint Brussels, n.p., 1965-70), 32:125-49. An eighth-century Irish hagiography is found in Oengus the Culde's Felire (cited in Gaidoz, "Saint Christophe," p. 195); another Irish source is the Passio Sancti Christophori, reproduced in M. J. Fraser, "The Passion of Saint Christopher,"Revue Celtique 34 (1913): 307-25. An Old English Passion of St. Christopher is the first portion of the so-called eleventh-century Beowulf Manuscript: Kenneth Sisam, Studies in the History of Old English Literature (Oxford: Clarendon Press, 1953), pp. 6 3 - 7 2 . His Coptic hagiography is found in the Synxarium for April 7 (Budge, Saints of the Ethiopian Church, 2:776-77). For the wide geographical spread of the medieval cult of St. Christopher in Western Europe, see the maps at the end of HansFriedrich Rosenfeld, Der HI. Christophorus: Seine Verehrung und seine Legend, Acta Academiae Aboensis, Humaniora 10, no. 3 (Âbo: Âbo Akademi, 1937), following p. 552. 50. In Walter of Speyer's hagiography (Vita et Passio, p. 76), he is twelve cubits (nearly twenty feet) tall. In the Beowulf Manuscript, he is "twelve fathoms high": Sisam, Old English Literature, p. 65. Christopher's gigantism becomes a commonplace in later hagiography. 51. Matthew 15.21-29. In this biblical passage, Jesus makes a pun in Latin, to which the Canaanite woman makes a timely response. Apart from its biblical origins, the land of the Chananeans (terra Chananea: Walter of Speyer, Vita et Passio, p. 67) may be a translation of Pliny's Cynopolitania, on the Nile in Egypt (this is the thesis of Jacobus de Voragine: Schmitt, Le saint levrier, pp. 204-5). An alternative interpretation may be derived from Isidore of Seville's seventh-century genealogy of the races (Etymologiae 9.2.11-12), which names the African Chananae as descendants of the biblical [C]Ham. As Kretzenbacher (Kynokephale Damonen, pp. 6 8 69) points out, a Latin identification of chananea with canis, "dog" and caninea, "dog-men," could not have been made prior to the medieval vulgarization of the Latin language found in the eighth-century passio upon which Walter of Speyer

Notes to Pages121-23

26 7

drew for his depiction of Christopher as a cynocephalic. Cf. Gaignebet,^4 plus hault 1:312. See plate 2. 52. Walter of Speyer, Vita et Passio, p. 67. Barking cynocephalics are located in Ethiopia in Philostratus of Lemnos's third-century Vita Apollonii 6.1; in India by numerous authors, including Pliny (NaturalisHistoria 7.23); and in central Asia by the author of the Tartar Relation (Skelton, VinlandMap, p. 87). 53. This theme only occurs in Chrisopher's Western hagiographies, which sometimes emphasize his gigantism to the omission of his cynocephaly: Vies des Saints et des Bienheureux, 7:613. Rosenfeld (Der HI. Christophorus, p. 370) suggests a Buddhist origin for this facet of the Christopher legend, in the Mâbâsüttasoma Jâtaka (no. 537) account of King Brahmadatta who is carried over a river by an animalfaced, man-eating demon named Kalmâşapâda. See below, chapter 6, n. 22. 54. Walter of Speyer, Vita et Passio, p. 75. 55. St. Babylus appears in Basil IPs Menologium (Gaidoz, "Saint Christophe," p. 204) and an Irish martyrology (Fraser, "Passion," p. 313). Christopher's change of color is described by Walter of Speyer (Vita et Passio, p. 68): "vultus . . . per sacri chrismatis inunctionem candidior lacte resplenduit." 56. Decius is called Dagnus in certain texts, the latter name being in fact connected to a "lost" Greek legend that was conflated with that of Christopher's martyrdom by the hand of Decius, in later sources. See Konrad Zwierzina, "Die Leğenden der Martyr von unzerstörbarem Leben," in Innsbrucker Festgruss von der philosophischen Fakultât (Innsbruck: n.p., 1909), pp. 140-44. 57. Herodotus Historia 1.108 and Dictionnaire d'Histoire et de Geographic Ecclesiastiques, s.v. "Barthelemy, Apotre." 58. Exceptions are the Egyptian Synxarium (Budge, Saints of the Ethiopian Church, 2:776), which celebrates Christopher's martyrdom on April 7, and certain Western sources which give the date April 28 (Richter, Der deutsche S. Christoph, p. 19). The dates May 9 and July 25 correspond to the setting and rising of Sirius, whence Pierre Saintyves's ("L'origine de la tete de chien de Saint Christophe," Revue Anthropologique [1924]: 383) identification of Christopher with the Canicular days. 59. Schwartz, "Iconographie orientale," p. 95. It should be recalled that Pliny the Elder locates the kingdom of Garamantes in this same region: the king of Garamantes had several hundred "Dog Soldiers" in his army. 60. Histoire des saints et de la saintete chretienne, 11 vols. (Paris: Hachette, 198688), s.v. "Christophe," 2:104-7; Ameisenowa, "Animal-headed Gods," p. 42. Christopher's motto is "Look at St. Christopher, and go safely." 61. Loeschcke reproduces fourteen examples of the cynocephalic St. Christopher from Eastern icons and wall paintings: "Darstellung," plates 1 - 5 . Kretzenbacher (Kynokephale Dâmonen, pp. 5 8 - 6 7 ) and Schwartz ("Iconographie orientale," p. 94, n. 5) give comprehensive accounts of the Eastern iconography of the cynocephalic St. Christopher between the fifteenth and the nineteenth centuries. 62. This is the twelfth-century manuscript illustration, portraying a giant cynocephalic St. Christopher standing at a high city gate, from the Swabian Codex of the Zweifalten monastery (Stuttgart Landsbibliothek, Hist. fol. 415, fol. 50r). scnS}

226

n o t e s t o p a g e s 98-100

Ameisenowa ("Animal-headed Gods," p. 42) mentions a fifth-century representation of a cynocephalic St. Christopher at the Mount Sinai monastery. However, he also states (p. 43) that the earliest Eastern representations of St. Christopher as a cynocephalic date from the fifteenth century, due to the ravages of the iconoclasts See plate 2. 63. Vies des Saints et des Bienheureux, 7:613; Kretzenbacher, Kynokephale Ddmonen, p. 62; and Ameisenowa, "Animal-headed Gods," p. 42. 64. Zwierzina's is the most detailed and convincing article on the subject: "Die Leğenden," 130-40, especially pp. 134, 138-39. See also Dvornik, Idea ofApostolicity, p. 203; Lipsius, Apokryphen Apostelgeschichten, 2, pt. 2:54, 76-78, 85; Schwartz, "Iconographie orientale," p. 93; and Ameisenowa, "Animal-headed Gods," pp. 4 2 - 4 3 . A number of hymns found in the Spanish Mozarabic liturgy, from the sixth century, intimate Christopher's cynocephaly: Kretzenbacher, Kynokephale Ddmonen, p. 67. For the Irish hagiography in which this identification is made, see Fraser, "Passion," pp. 309, 315; and Kretzenbacher, Kynokephale Ddmonen, p. 62, who suggests Ireland as a bridge between Egyptian Coptic and Western European traditions. This identification is taken for granted by Louis Reau: Iconographie de lArt Chretien, 3 vols. (Paris: n.p., 1955-58), vol. 3, pt. 1:304. A fifteenth-century handbook of iconography, authored by Dionysios of Mount Athos, describes Christopher as "Reprobos, one of the Kynokephaloi": Kretzenbacher, Kynokephale Ddmonen, p. 67. 65. Times Concise Atlas of World History, pp. 3 8 - 3 9 ; Henry Bettenson, cd., Documents ofthe Christian Church, 2d ed. (London: Oxford University Press, 1963), pp. 4 6 - 4 8 . On the possible Asian origins of the Gnostics' cynocephalic imagery, see Kretzenbacher, Kynokephale Ddmonen, p. 64. 66. Ameisenowa, "Animal-headed Gods," pp. 2 3 - 2 7 . See also Jacoby, "Der hundsköpfıge Dâmon de Unterwelt," p. 223. 67. Jean Seznec, The Survival of the Pagan Gods: The Mythological Tradition and Its Place in Renaissance Humanism and Art, tr. Barbara Sessions (Princeton: Princeton University Press, 1972), pp. 162,166. 68. Head, Historia Numorum, pp. 722-24. A similar representation is a bas relief, from the important Greco-Parthian town of Hatra in Mesopotamia, of Nergal. Here, this anthropomorphic divinity is identified with the Greek Cerberus (Cerberus is a two- or three-headed dog in Greek mythology: here, the anthropomorphic Nergal-Cerberus is accompanied by a three-headed dog). Nergal is depicted with "the axe of Hadad," "the eagle of Zeus," and a sword and caduceus topped by a trident. His body is ornamented and surrounded by snakes and scorpions: ReneDussaud, "Melquart," Revue de VHistoire des Religions 151 (1957): 14. See above, note 47, on similar numismatic evidence from Cynopolitania (Asyût) on the Upper Nile. 69. Ameisenowa, "Animal-headed Gods," pp. 23-24. The same author reproduces Jewish apocalyptic sources from the same period, which depict certain dwellers in heaven as cynocephalic (pp. 26-27). 70. In later Eastern Christianity, Michael the Archangel becomes the "torchbearer" who guides the souls of the dead to the judgment seat of God and defends them from the devil's onslaughts: he is their advocate in the other world

Notes to Pages121-23

26 7

(Ameisenowa, "Animal-headed Gods," pp. 25, 44-45). In their Coptic hagiographies, Andrew and Bartholomew are accompanied by Michael on their mission to save Matthew from the cannibals (Lipsius, Apokryphen Apostelgeschichten, 1:54653; 2, Pr- 2:76-86, especially p. 80). He is depicted together with the cynocephalic 5t. Christopher on Eastern icons: Loeschcke, "Darstellung," pp. 41, 56-57. 71. Ameisenowa, "Animal-headed Gods," pp. 39-42. 72. Saintyves, Saint Christophe, p. 44. However, Budge (Saints of the Ethiopian Church, p. 278) and Piankoff ("Saint Mercure," p. 17), place his day on November 22 and 21, respectively, in Ethiopian and Egyptian traditions. August 24, the day of Bartholomew in the Roman church, was that of a St. Mercurius of Apulia: Hippolyte Delahaye, Les legendes grecques des saints militaires (Paris: Picard, 1909), p. 92. 73. Piankoff, "Saint Mercure," p. 18. 74. Schwartz, "Iconographie orientale," p. 98. 75. The earliest source of the Mercurius legend is a local tradition from Cappodocia, not far from Bartholomew's Lycia. The earliest known artistic representation of Mercurius and the cynocephalics is a fresco from the St. Anthony monastery in Egypt, which dates from 1232-33 (Loeschcke, "Darstellung," p. 43). A similar example, in Cappodocia, is St. Andrew of the Cynocephali at Kokarkilise: Petro B. T. Bilaniuk, "The Holy Spirit in Eastern Christian Iconography," Patristic and Byzantine Review 1 (1982): 102. A Cappodocian image also depicts a halfcanine, half-human figure bearing the Christ. This monster is not identified with Christopher; rather it is called, in Romance and Italian traditions, Pulicane, and is born of the union of a dog and a Christian woman from Cappodocia: Kretzenbacher, Kynokephale Ddmonen, pp. 22, 34. The eleventh-century tympanum of the Vezelay church, located at the crossroads of the Rome and Compostello pilgrimage routes in France, also shows a scene from the apostolic missions to the Cynocephali: Francois Vogade, Vezelay (Vezelay: n.p., 1974), plates 8 , 9 , 1 9 ; and Friedman, Monstrous Races, pp. 78, 230, and n. 52. See plate 1. 76. On the shift from the jackal-headed Anubis of ancient Egypt to the dogheaded Hermes or Mercury of Hellenistic and medieval times, see Seznec, Survival of the Pagan Gods, p. 162; Vogade, Vezelay, notes to plate 19; Gaidoz, "Saint Christophe," pp. 2 0 9 - 1 0 ; and Saintyves, Saint Christophe, pp. 38, 4 1 - 4 4 . In The Golden Ass, Lucius Apuleius describes a priest of Isis who plays the role of Anubis. He has a dog's head and carries a scepter and a palm branch. A priest representing Mercury figures in the same context: Apuleius, The Golden Ass, tr. William Adlington and ed. Harry C. Schnur (New York: Collier, 1962), p. 268. See plate 3. 77. Saintyves, Saint Christophe, p. 45; and Manfred Lurker, "Hund und Wolf in ihrer BeziehungzumTo&e?Antaios 10 (1969): 200. 78. It was especially the unburied dead whose souls were considered to be cynomorphic in the ancient world: Lilja, Dogs, p. 35. In Roman religion, the Lares—ancestral divinities—are also protectors of streets and crossroads, vectors of danger in numerous traditions since they were conceived as paths or doors opening onto the infernal regions, and were thus haunted by wandering spirits: Lurker, "Hund und Wolf," p. 203; and Seznec, Survival of the Pagan Gods, p. 294. In medieval Europe, dogs and cynocephalics are associated or identified with the Devil, who is the lord of hell: Jacoby, "Der hundsköpfıge Dâmon der Unterwelt," pp. 2 1 9 -

228

n o t e s t o p a g e s 98-100

25; and Martha Paul, Wolf, Fuchs und Hund beı der Germanen, Wiener Arbeiten zur Germanischen Altertumskunde und Philologie, 13 (Vienna: Karl M. Halosar, 1981), p. 252. On the imagery of Sirius in the mouth of the heavenly Great Dog, see Gaignebet,^4 plus hault sens, 1:259, citing Manilius, Astronomica 1:646. 79. Saintyves, Saint Christophe, pp. 30-31. 80. Lurker, "Hund und Wolf," pp. 208, 211, and idem "Der Hund als Symboltier fiir den Ubergang vom Diesseits in das Jenseits" Zeitschriftjur Religions- und Geistesgeschichte 35 (1980): 138. See also Jacoby, "Der hundsköpfıge Dâmon de Unterwelt," pp. 2 1 9 - 2 5 . 81. See Homer Iliad 22.29; Hesiod Opera etDies 583-95, 609; and Aeschylus Agamemnon 967. In a reference to the heat of midsummer, Virgil calls Orion "the cruel hunter" (Aenead 7.719). Sirius is also identified with the bitch Maira, a dog of Icarus, as well as one of the hounds of Actaeon: Gaignebet,^ plus hault sens, 1:263. Cf. Pliny Naturalis Historia 2.40, 18.68. The word seirios is derived from the verb seiriaö (blaze up, burn, from the heat of the sun): Anton Scherer, Gestirnnamen bei den Indogermanischen Volkern (Heidelberg: Carl Winter, 1953), p. 112. 82. Hesiod Theogonia 2 8 9 - 9 4 , 306-12; Pindar Isthmian Odes 1.13; Saintyves, Saint Christophe, p. 43; and Mainoldi, Ulmage du loup, pp. 4 2 - 4 3 . 83. According to Pausanius (Descriptio Graeciae 1.43.7-8), the Kunophontis were observed to calm violent storms. Midsummer is of course the period in which rabies epidemics are most widespread, whence perhaps the term dog days and the qualification of Sirius as the dog-star. On Sirius and rabies, see Homer Iliad 8.299 and Pliny Naturalis Historia 2.107,123 and 8.153. Pliny (29.32) also offers a remedy for rabies that is nothing other than the calcinated "hair of the dog"! The Greek term for rabies, lussetera, is divinized as Lyssa, the goddess of frenzy whose name is etmologically related to lukos (wolf). Lyssa is identified with the bitch Maira in Euripides Heracles 1024: elsewhere in Greek mythology, Maira helps Erigone to discover her father Icarios's corpse (Lilja, Dogs, pp. 21,64; and Dictionnaire des mythologies s.v. "Dionysos," by Marcel Detienne). According to Scherer {Gestirnnamen, pp. 114-15), Maira is the Greek personification of the height of the dog days: he also notes that the planet Mercury is often associated with Sirius in Indo-European traditions. For his general discussion of Sirius as the dog-star of nearly every Indo-European tradition, see pp. 109-16. Walter Burkert (HomoNecans: The Anthropology of Ancient Greek Sacrificial Ritual and Myth, tr. by Peter Bing [Berkeley: University of California Press, 1983], pp. 108-9) demonstrates a symbolic connection between the dog days, the myths of Apollo's son Linos and of Actaeon, the myths of the birth of Cyrus and of Romulus and Remus, and Apollo Lykeios. See above, chapter 1, n. 52. 84. Lilja, Dogs, pp. 57, 65; Schlerath, "Hund," p. 25; and Schmitt, Le saint levrier, p. 206. May 9, the day of Christopher's commemoration in the Greek church, corresponds to the Roman festival of the Christian martyrs. May was the Greco-Roman month of the dead (Majores): Ovid Fasti 5.427; and Saintyves, Saint Christophe, p. 41. 85. Lilja, Dogs, pp. 23, 52, 65; Kretschmar, Hundestammvater, 2:214; and W. Klinger, "Hundsköpfıge Gestalten in der antiken und neuzeidichen Uberlieferung," Bulletin International de lAcademie Polonaise des Sciences et des Lettres

Notes to Pages121-23

26 7

(Classe d'Histoire et de Philologie) (Cracovie, July-December 1936): 122. Hades' cynocephaly is discussed in Mainoldi, LTmage du loup, p. 43. On Euripides' Hecuba, see Martha C. Nussbaum, The Fragility ofGoodness: Luck and Ethics in Greek Tragedy and Philosophy (Cambridge: Cambridge University Press, 1986), pp. 398,414,416. Gaignebet identifies Hecuba transformed into the bitch with Maira, a mythologization of Sirius: A plus hault sens, 1:266. The vengeful regicide Clyternnestra is called "dog-faced" (kunöpis), also a reference to the dog days: 1:335. 86. Schlerath, "Hund," p. 27; and Lurker, "Hund und Wolf," pp. 203,211,215. For medieval parallels, see Friedman, Monstrous Races, p. 125; and Kretzenbacher, Kynokephale Ddmonen, p. 48. 87. Lilja, Dogs, p. 40. Scylla, as well as Cerberus and Orthros and other Greek death monsters, have snakes, another great chthonic creature, coiling out of their necks. This is equally the case with Nergal of Hatra and the Indian Siva, whose canine connections are also numerous. 88. For a discussion of Lithuanian and other Indo-European evocations of the land of the dead as a pasture, see Vlatcheslav Ivanov and Vladimir Toporov, "Le mythe Indo-Europeen de I'orage poursuivant le serpent: reconstruction du schema," p. 1191, and A. T. Greimas, "La quete de la peur," p. 1219, in Echanges et Communications: melanges ojfertes a Claude Levi-Strauss, ed. Jean Pouillon and Pierre Maranda (Paris: Mouton, 1970). When the hellhound is not a herd dog, he is often portrayed as a watchdog on the threshhold of two worlds. The best-known example of this is Cerberus; this is also the case of Odin's dog Garmr, who guards the entrance to Niflheim in Germanic mythology: Lincoln, "Hellhound," p. 276. 89. Schlerath, "Hund," p. 25; and Lilja, Dogs, p. 80. Frau Holle (Mrs. Hell), the Germanic Hecate, barks like a dog: Frank Jenkins, "The Role of the Dog in Romano-Gaulish Religion," Latomus 16:1 (January-March 1957): 62. 90. Apollo himself is "canine" when he is Telmissos, Krimisos, or Alkestes (Kretschmar, Hundestammvater, 1.166) and "lupine" when he afflicts humans with disease, in the form of Lykeios (Lurker, "Hund und Wolf," p. 214). On Hermes as psychopomp in later traditions, see Seznec, Survival of the Pagan Gods, pp. 162, 166,180-81,199-200. 91. Cerberus is generally figured as bicephalic on Attic pottery and tricephalic in Greek tragedy and in Roman sources; he is fifty-headed in Hesiod Theogonia 3 1 1 12: Seznec, Survival ofthe Pagan Gods, p. 207. He and his brother constitute a pair, but Orthros himself is cited in a dual declension in Pindar (Isthmian Odes 1.13, cited in Lilja, Dogs, p. 51). On the conflation of Cerberus with the Babylonian Hades, named Nergal—and by extension with many other gods of death and the underworld—see Dussaud, "Melquart," p. 14 and Cristiano Grottanelli, "Cosmogonia e sacrificio II," Studi Storico Religiosi 5:2 (1981): 173-96. Cf. Lipsius, Apokryphen Apostelgeschichten, 2, pt. 2:66-67, 71-72. See above, n. 46; and below, n. 100 to this chapter. 92. The best sources on the Indo-European dog's role as intermediary between the worlds of the dead and the living are Schlerath, "Hund," pp. 2 7 - 3 9 ; Lurker, "Hund als Symboltier," pp. 132-44; and Lincoln, "Hellhound," pp. 274-85. On the use of a sop to appease the hellhound, see J. Bernolles, "A la poursuite du chien de la mort d'Asie steppique en Occident et en Afrique Noire," Revue de I'Histoire des

230

n o t e s t o p a g e s 98-100

Religions 173 (1968): 3 4 - 8 4 , and other references listed in Afshar, Immortal Hound, p. 30. 93. Thomas's hagiography is found in Andre-Jean Festugiere, Les Actes Apocryphes de Jean et de Thomas (Geneva: Patrick Cramer, 1983) and Klijn, Acts of Thomas, especially pp. 13-29. A discussion is George Huxley, "Geography in the Acts of Thomas," Greek, Roman and Byzantine Studies 24:1 (Spring 1983): 71-81. On Thomas's Gnostic hagiography, see Lipsius,Apokryphen Apostelgeschichten 1:28— 29, 2 2 5 - 2 9 , 2 7 8 - 8 0 . The original source of Thomas's Indian hagiography mav have been a Buddhist legend: Rose, La leğende doree, 1:57-64. 94. Ernst Burgstaller, "Uber einige Gestalten des Thomasbrauchtums m Überösterreich," Mitteilungen der Anthropologischen Geselbchaft in Wien 95 (1965) 308. 95. Ibid., pp. 308-21. 96. The Gallo-Roman Tarascon is also depicted as a child-eater in an image preserved in the Musee Calvet, in Avignon. In astrology, Saturn rules the sign of Capricorn, in which the winter solstice falls: ibid., pp. 320-21. See also p. 317 and Lurker, "Hund als Symboltier," pp. 132-33; cf. Schlerath, "Hund," p. 26, who associates this motif with lycanthropy. A Celtic parallel, that of Gwynn am Nudd, is cited in Lurker, "Hund und Wolf," p. 203. 97. See above, p. 23 of this chapter, for Christopher's Ethiopic iconography. Gaignebet's discussion is A plus hault sens, 1:310-14. See also Kretzenbacher, Kynokephale Damonen, p. 30; Histoire des saints et de la saintete chretienne, 11 vols., sous la direction d'Andre Mandouze (Paris: Hachette, 1986-88) s.v. "Christophe," by Jean Delumeau, vol. 2, p. 105. A "Lamp-bearing Cynocephalic" is placed, by Greek astronomers—perhaps drawing on a ninth-century source from Khorasan, in eastern Iran—in the "Sphaera Barbarica": Scherer, Gestirnnamen, p. 210. 98. Burgstaller, "Thomasbrauchtums," pp. 316-17; Kretzenbacher, Kynokephale Damonen, pp. 16-18, 2 4 , 4 4 - 4 5 , 1 2 1 - 2 2 . 99. Richter, Der deutsche S. Christoph, p. 220. 100. On Saturn and India, see Friedman, Monstrous Races, p. 51; and Seznec, Survival ofthe Pagan Gods, p. 162, citing the fourteenth-century Muslim cosmography of Qazwini. On the ancient and medieval shift from Kronos and Chronos to Saturn, see Seznec, pp. 61, 159, 162, 176 and Raymond Klibansky, Erwin Panofsky, and Fritz Saxl, Saturn and Melancholy (London: Thomas Nelson & Sons, 1964; reprint Nedeln, Lichtenstein: Kraus-Thomson, 1979), pp. 131-33, 139, 185 and 208. On Saturn and Nergal, see pp. 136, 159. On Canaanite parallels, see Grottanelli, "Cosmogonia e sacrificio," pp. 1 8 3 - 8 7 , 1 9 1 - 9 2 . Compare Mot's symbolism with that of Nergal: Samuel Noah Kramer, Mythologies of the Ancient World (Garden City, N.Y.: Double day Anchor, 1961), p. 202; and Dussaud, "Melquart," p. 14. On Nergal and Astyages, the "king" who exposes Cyrus and who martyrs Bartholomew, see above, part 1 and nn. 13 and 46 to this chapter. Strabo (Geographica 3.1.4) locates a temple to Baal-Saturn, or HerculesMelquart, at the western edge of the ancient world, on the west coast of Spanish Galicia, the later site of St. James of Compostello: Gaignebet, A plus hault sens, 2:429. 101. Lutz Röhrich, "Hund, Pferd und Schlange als symbolische Leitgestalten in

Notes to Pages121-23

26 7

Volksglauben und Saga," Zeitschriftfar Religions- und Geistesgeschichte 3 (1951): 71, ^ d Burgstaller, "Thomasbrauchtums," pp. 317, 320. 102. In French and Welsh Celtic traditions, the god Cernunnos and the goddesses Nehalennia and Rhiannon, all psychopomps, are associated with dogs or horses. When the Celts were christianized, these and other divinities with canine or equine attributes came to be associated with St. Nicholas. On this subject, see Jenkins, "Role of the Dog," pp. 6 0 - 7 4 ; Kretzenbacher, Kynokephale Ddmonen, pp. 87-88; and Elyane Gorsira-Ronnet, "La Fete de St. Nicolas au Pays-Bas,"Melanges offerts a Andre Varignac (Paris, E.P.H.E., 1971), pp. 341-53. See also Paul, Wolf Fuchs und Hund, pp. 240-41. Little Red Riding Hood is the sun devoured by the night of midwinter (the wolf): Lurker, "Hund und Wolf," p. 212. 103. For a discussion of the term pesoglavci and its alternate forms in various Slavic languages, see Kretzenbacher, Kynokephale Ddmonen, pp. 5 - 6 . On the principal regions of southeastern Europe in which pesoglavci traditions are found, and their connections with the "Turks," see pp. 8 - 2 6 , 115-34. Kretzenbacher's principal aim in his excellent work is to establish a European foundation for these southeast European folk traditions. 104. G2i\%pebtx.,A plus hault sens, 1:439,461-64. 105. Ibid., 1:310 and 2:426, citing Diodorus of Sicily Bibliotheca Historica 1.11. 106. Ibid., 1:458-59,468, citing Proclus, Porphyry, and Chalcidius. 107. Ameisenowa, "Animal-headed Gods," p. 22; Seznec, Survival of the Pagan Gods, pp. 120,178; and Apuleius, The Golden Ass, p. 268. 108. Ameisenowa, "Animal-headed Gods," pp. 26, 44; and Lurker, "Hund als Symboltier," p. 142. On the relationship between the planet Mercury (Hermes, Mercurius) and the star Sirius, see Scherer, Gestirnnamen, pp. 109-16. For the identification of star groups with cynocephalics in other astronomical systems, see pp. 171, 200-201 and 2 1 0 - 1 1 . See also Jacoby, "Der hundsköpfıge Damon," p. 223. 109. On Anubis's kinship with Osiris, and the place of the dog in the cults of Isis and Osiris, see Plutarch, De Iside et Osiride 356F; and Diodorus of Sicily Bibliotheca Historica 1.18, 87. See Kretschmar, Hundestammvater, 1:183-86, on catabasis and resurrection as a European folk motif. On Upuat, see Lurker, "Hund als Symboltier," pp. 142-43. 110. Gaignebet, A plus hault sens, 1:314. 111. Ibid., 1:315 and 2:433, citing Grayet, Itineraire de la Haute-Egypte, pp. 108-9. 112. Willehand Eckert, "Der Hund mit der Fackel und andere Attribute des hlg. Dominikus," Symbolon 5 (1966): 31-40; and Gaignebet, A plus hault sens, 1:266. A fourteenth-century Florentine fresco, attributed to Andrea de Bonaiuto, depicts the Dominican brothers as a group of black-spotted white dogs chasing away heretics in the form of wolves: 1:320. 113. Gaignebet, A plus hault sens, 1 : 7 4 , 2 9 7 - 3 0 4 , 4 6 3 - 6 4 ; 2:429. 114. Ibid., 1:316-17, 327. 115. Ibid., 1:316,325-34, discusses Guignefort's legend at great length, tying it in to astronomical phenomena. He takes issue on a number of matters with Schmitt {Le saint levrier, pp. 6 1 - 1 7 1 , 199-213), whose recent work addresses the legend

232

n o t e s t o p a g e s 98-100

and cult of Guignefort directly. See also Pierre Saintyves, En marge lie la Leğende doree: songes, miracles et survivances; essai sur la formation de quelques themes hagiographiques (Paris: Emile Nourry, 1931), pp. 4 1 2 - 2 8 . 116. Rosenfeld, Der HI. Christophorus, p. 354. The Zodiac of Dendera is preserved in the Egyptian Collection, Musee du Louvre, Paris. 117. He is called latrator in Virgil Aenead 8.698. On his transformation in the Hellenistic age see Gaidoz, "Saint Christophe," p. 210 and Schlerath, "Hund," p. 29. 118. Herodotus Historia 2.65-67; Strabo Geographica 17.1.40; and AelianusD? natura animalium 10.45. In north African Islam, the greyhound is the only dog who is possessed of baraka; in Dante and in Dürer, it is the messenger, or precursor, of the Second Coming: Dictionnaire des symboles: mythes, reves, coûtumes,gestes,formes,figures, couleurs, nombres, gen. ed. Jean Chevalier (Paris: Robert Laffont, 1969), p. 198. 119. Herodotus, who identifies the Kynokephaloi as dog-faced baboons, locates them in Libya (Historia 4.191); Aelianus places them in both Ethiopia and India (De natura animalium 4.46, 10.25); Aristotle (Historia Animalium 502a.l9) and Strabo (Geographica 17.1.40) say they are sacred animals in Egypt. See also Lilja, Dogs, p. 40, and Liddell and Scott, A Greek-English Lexicon, s.v. "kunokephalos." See plate 4. 120. Piankoff, "Saint Mercure," p. 23. 121. These inscriptions are recorded and translated ibid., pp. 2 2 - 2 3 , and in Andre Piankoff and Etienne Drioton, Le livre duJouretdelaNuit (Cairo: Imprimerie de l'Institut Français d'Archeologie Orientale, 1942), pp. 4, 8 4 - 8 5 , 88-89. Artemidoros (cited in Strabo Geographica 16.4.14-16) mentions Nötou Ceras (Cape Guardafui), the southern and eastern edge of Africa, as a watering place for cynocephalic baboons: "[after] the Watering Place of the Cynocephali . . . the last promontory on this coast [is] Nötou-Ceras [Horn of the South]. After rounding this promontory, we no longer . . . have any record of harbors and places, because the promontory is not known from here on." CHAPTER 3

1. Friedman, Monstrous Races, p. 35. 2. Homer Odyssey 1.23-24, 7.70; Herodotus Historia 3.101. It is an interesting coincidence that Herodotus's source on India was a sixth-century B.C. Greek traveler named Skylax of Karyanda (Wittkower, "Marvels of the East," p. 159, n. 2): in Greek, Skylax's name means "puppy." 3. Skelton, Vinland Map, pp. 153-54. All of these terms, excepting antipodes ("backwards feet") are based on the Greek term oikos, "house, home." Oikoumene may thus be roughly translated as "home"; periokoi, "(people) around the home"; and antoikoi, "(people) opposite the home." 4. Wittkower, "Marvels of the East," p. 176 and Skelton, Vinland Map, p. 146. For the ancient and medieval world, each of the four extremities of the earth constituted a "paradise" of sorts: Gernet, Anthropologic, pp. 185-88, 193, 200. Ethiopia to the south was Homer's paradise; Herodotus (Historia 1.163) is highly impressed by the longevity of the inhabitants of Tartessos to the west (Spain) while

Notes to Pages121-23

26 7

the northern land of the Hyperboreans was a kind of paradise for Megasthenes and later writers (see below, chapter 6, part 1). India was the eastern edge of the world: "the Indians live farthest to the east and the sunrise, of all the inhabitants of Asia" (Historia 3.98). India was identified with Eden, paradise, and the Kingdom of Prester John well into the Middle Ages: Skelton, VinlandMap, pp. 48, 69, 119, 131, 245; John Kirdand Wright, Geographical Lore ofthe Time of the Crusades, American Geographical Society: Research Series, no. 15 (New York: American Geographical Society, 1925), pp. 72, 261-65; and Wittkower, "Marvels of the East," p. 181. 5. There are no extant manuscripts of this play. This is a fragment reported in Strabo Geographica 7.3.6. 6. Equites 415, cited in Klinger, "Hundsköpfıge Gestalten," p. 119. 7. Reported in Strabo Geographica 7.3.6. 8. Kretschmar, Hundestammvater 1:136,2:169. In the latter traditions, these are the pesoglavci, of which the first member of the compound may be a metasthesis of the Indo-Iranian svan- or span (dog): Kretzenbacher, Kynokephale Ddmonen, pp. 5, 22, 26. A Russian term, kobeljs ("manlike dog") may be a cognate of the Sanskrit sabala, one of the two Indian hellhounds (Max Vassmer, Russisches etymologisches Worterbuch, 3 vols. [Heidelberg: Carl Winter, 1950-58], 1:582). 9. The portion of Ctesias's Indika in which the Kynokephaloi are discussed is recorded in Photius's Bibliotheca 72.19-23 [=47a.l9-48b.4]. This is translated in Henry, Ctesias, pp. 73-79. Another edition of Ctesias, a collection of fragmentary citations by writers other than Photius, is Carl Müller, ed., Herodoti Historiarum LibriLK (et) Ctesiae Cnidii (Paris: Firmin Didot, 1857), pp. 99-100. An English translation is found in J. W. McCrindle,Ancient India as D escribed by Ktesias the Knidian (London: Trubner, 1882). 10. Hyparchos (or Hyperbaros) may be derived from a Persian transliteration (*wispabara) of an unknown Sanskrit toponym: Marquart, Die Benin-Sammlung, p. 204. 11. Ctesias's Kalustrioi, which Photius, following Megasthenes, translates as Kynokephaloi, may be a Greek transliteration of the Persian sa-dauzstr (ibid., p. 209) and the Sanskrit sva-duhitr, which signify "Dog-Milker." Megasthenes' identification of Kalustrioi with Kynokephaloi—"Kalountai de hupo tön Indön Kalustrioi hoper estin Hellenisti Kunokephaloi"—is cited in Solinus, Collectanea rerum memorabilium 52.27, and reproduced in Photius's Bibliotheca 47b.30-31: see Henry, Ctesias, p. 75. 12. Ouran . . . hoianper kuön (a tail like a dog). The Greeks knew of a Mycenean goddess (Dictionnaire des mythologies, s.v. "Crete et Mycenes," by Paul Faure), a Spartan tribe, and a race of dogs named Kynosoura, Kynosouros, and Kynosourides (Dog-Tails), respectively. See Pauly-Wissowa, vol. 23 [ = 12.1], s.v. "Kynosura." The Sanskrit cognate of Kynosoura is Sunahsepa (Dog Tail or DogPenis), the name of a hero of a very popular myth which probably arose as an explanation for the astronomical motions of Ursa Minor, the "bear with a dog's tail": cf. Scherer, Gestirnnamen, pp. 176-77; and Lilja, Dogs, p. 80. The myth of Sunahsepa is discussed in chapter 4. 13. After Ctesias, Aelianus (De natura animalium 4.46; 10.25) is the sole author to define the Cynocephali as animals rather than men: this because they do not

234

n o t e s t o p a g e s 98-100

speak, but bark. Alongside the Cynocephali, the other "noble race" of the East were the Gymnosophists (the naked philosophers) or Bragamanni, the brahmin ascetics of India immortalized in Arrian's account of Alexander's Indian campaigns (Anabasis 7.1). 14. Strabo Geographica 15.1.5, 15.2.5. 15. Wittkower, "Marvels of the East," pp. 162,180-81. 16. See Pliny Naturalis Historia 7.1.23; Aulus Gellius Noctes Atticae 9.4.9; Strabo Geographica 1.2.35, 7.3.6; Aelianus D* natura animalium 4.41,46, 5.3; and Liber monstrorum de diversisgeneribus 16 (reproduced in Moriz von Haupt, Mauricu Hauptii Opuscula, 3 vols. [Leipzig: Hirzel, 1876], 2:228). Tzetzes, author of the Chiliades, is a much later (12th-century) source, but much of the material he cites is archaic: in Chiliades 7.629,713-768, he mentions the Kynokephaloi together with the Hesiod's Hemikunes, and cites the third-century B.C. Simmias Eliagicus, Apollodorus (frag. 159-60), and Ctesias. Tzetzes locates the Kynokephaloi in India, Egypt, and Ethiopia. According to Marquart (Die Benin-Sammlung, p. 257), the author of the Peripulus Maris Erythreum (65) and Pliny are describing the mountain-dwelling Kirâtas of India when they speak of Cynocephali. On the Physiologus, see Meslin, Le merveilleuXy p. 30. On Palladius, see J. Duncan M. Derrett, "The History of Talladius on the Races of India and the Brahmans,'" Classica et Mediaevalia: Revue Danoise de Philologie et d'Histoire 21 (1960): 6 4 - 9 9 . 17. Pliny Naturalis Historia 7.32. See also Friedman, Monstrous Races, p. 8. 18. Pliny Naturalis Historia 6.195. Indian Dog-Milkers are also mentioned in Aelianus (De natura animalium 16.31), who is probably following Ctesias. On the theory that Ctesias's term Kalustrioi is itself derived from a Persian or Sanskrit term for Dog-Milker, sa-dauzstr or sva-duhitr, see Marquart, Die Benin-Sammlung, pp. 2 0 7 - 9 . Ethiopian Dog-Milkers are first reported, in the second-century A.D., by Agatharcides of Cnidos: Friedman, Monstrous Races, p. 27 and Lecouteux, Les monstres dans la litterature allemandey 2:22. Strabo (Geographica 16.4.10) locates them in Ethiopia. See below, p. 72. 19. Pliny Naturalis Historia 8.142. Two hundred of these dogs came to Garamantes' rescue, freeing him from slavery in the hands of his enemies: T. H. White, ed., The Book ofBeasts: Being a Translation from a Latin Bestiary (New York: G. P. Putnam's Sons, 1954), p. 62. 20. Pliny Naturalis Historia 6.192: Pliny is drawing on a tradition that may be traced back to Aristocreon (3rd century B.C.) through Callimachus, Hermippes, and Aelianus (De natura animalium 7.40). This tradition, also reported in Ptolemy and Solinus, may have its basis in ethnographic fact: Ptolemy Geographica 4.7.10 and notes in Müller, Claudii Ptolemaei Geographica, p. 784. See also H. Hartmann, review of Die Ptoembari, pp. 136-38. On reports, going back to the first century B.C. Juba II, of the (Ba)-canari as the name of a Mauritanian people, see Robert Schilling, Pline VAncien, HistoireNaturelle, Livre 5, pp. 139-40 (n. 2 to Pliny, Naturalis Historia 5.16); and Hans Biedermann, "Die 'Hunde-Inseln' im Westmeer," Almogaren 3 (1972): 102-3. 21. On the propagation of the Pseudo-Callisthenes tradition, see Andrew Runni

Notes to Page 52

235

Anderson, Alexander's Gate, Gog and Magog, and the Inclosed Nations (Cambridge, Mass.: Medieval Academy of America, 1932), p. 30; and Paul Meyer, Alexandre le Grand dans la litterature francaise du moyen age, 2 vols. (Paris: Vieweg, 1886), 2:1-7. 22. Derrett, "History of'Palladius,'" pp. 6 4 - 9 9 . Palladios 2 . 3 - 5 7 is the source of Pseudo-Callisthenes (3.6-16) and Julius Valerius (3.17) on the monstrous races of India. 23. On the history of the various manuscript traditions of the Alexander Romance, Anderson, Alexander's Gate, pp. 15-43; Reinhold Merkelbach, DieQuellen des greichischen Alexanderromans, Zetemata Monographaien zur klassischen Altertumswissenschaft, no. 9 (Munich: Beck, 1977), p. 142; E. A. T. Wallis Budge, The Alexander Book in Ethiopia (London: Oxford, 1933), pp. xv-xxv; and Albert Mugrdich Wolohojian, The Romance ofAlexander the Great by Pseudo-Callisthenes (New York: Columbia University Press, 1969), pp. 1 - 7 . On the origins of the Cynocephalic-Scythian-Amazon identification, see Anderson, Alexander's Gate, p. 15. On the ancient geography of the Taurus range, which, it was held, separated north from south across Eurasia, see Strabo Geographica 15.1.1; Skelton, VinlandMap, pp. 6 3 - 6 4 ; and Anderson, Alexander's Gate, pp. 1 5 57, passim. 24. It is generally accepted that there exist four groups of texts that stemmed from the Greek original. These are: (A), a Greek manuscript (no. \7\\,fondsgrec of the Bibliotheque Nationale [Paris]), together with the fourth-century Latin rendition of Julius Valerius, the fifth-century Armenian translation, and an eleventhcentury Arabic translation; (B), no. 1685, fondsgrec of the BN and the majority of extant Greek manuscripts, developed from reworkings of the (A); the (C) text (no. 113, suppl.grec) group, designated by Müller as C 1 , which is an expansion of (B), with interpolations; and the (D) group, including a Syriac version, an Ethiopic version (see below, note 28), and the Greek source used by Leo of Naples for his Historia deprdiis. The development of the legend in the East owes much to a lost Pehlevi source, upon which the seventh-century Syriac version was based: Wolohojian, Romance of Alexander, pp. 2,5; Meyer, Alexandre le Grand, 2:4; and Anderson, Alexander's Gate, pp. 32,92. 25. J. W. McCrindle, The Invasion of India by Alexander the Great (Westminster: A. Constable, 1893), p. xvii. On the Historia deprdiis and Letter ofAlexander to Aristotle, see Wittkower, "Marvels of the East," pp. 179-80. The editions which we have used for these sources are Vita AlexandriRegisMacedonum, ed. J. Trumpf, Bibliotheca Scriptorum Graecorum et Romanorum Teubneriana (Stuttgart: Teubner, 1974), pp. 103, 125-26, 146; Arriani Anabasis et Indica . . . Pseudo-Callisthenes Historiam Fabulosum, ed. Carl Müller (Paris: Firman Didot, 1866), pp. 87, 89,216 (the last page cited contains the EpistolaAlexandriMacedonis adAristotelem 40); Anderson, Alexander's Gate, pp. 3 8 - 4 1 ; andDerAltfranzösische ProsaAlexanderroman nach der Berliner Bilderhandscrift nebst dem lateinischen original der Historia de Proliis (Rezension J2), ed. Alfons Hilka (Halle: Niemeyer, 1920), p. 234 (this is chapter 119 of the Old French source). A valuable general source on the history of the Alexander Romance from its Greek sources through its Latin and vernacular French

236

n o t e s t o p a g e s 98-100

recensions is Meyer, Alexander le Grand, especially volume 2, which discusses the history of the legend. 26. Anderson, Alexander's Gate, pp. 1 2 - 1 4 and passim. Among these, the Huns are described, in Jerome's late fourth-century Hebraicae quaestiones in librogeneseos as the offspring of fallen angels and human sorceresses, or of daemonia immunda: Maenchen-Helfen, World ofthe Huns, p. 5. 27. Meyer, Alexandre le Grand, 1:177-93; 2:273-99, especially p. 283. 28. Helmut van Thiel, ed., Leben und Taten Alexanders von Makedonien: der GreischischeAlexanderroman nach derHandschriftL (Darmstadt: 1974), p. 192; Historia Alexandri Magni (Historia de preliis): Rezension J1, ed. Alfons Hilka and Karl Steffens (Meisenheim am Glan: Anton Hain, 1979), pp. 170,174; Wolohojian, ilomance of Alexander, p. 112 (chapter 209 of the Armenian version, which corresponds to 2 . 2 3 - 4 3 in manuscript BC of Müller's edition); and Budge, Alexander Book in Ethiopia, p. 140. The Ethiopic version, drawing as it does on a seventhcentury Syriac source, draws this episode, and much of the latter part of its account, from an approximately sixth-century Pehlevi version, of Persian origin: Budge, Alexander Book in Ethiopia, p. xxvii. According to Piankoff ("Saint Mercure," p. 19) this location of the Cynocephali beyond the edge of the world, in mountains "where the sun appears ten times larger than in Greece," is inspired by Ctesias (Indika 5, in Henry, Ctesias, p. 63), who was himself inspired by classical reports of Ethiopia that ultimately went back to the Egyptian dog-faced baboons of the New Empire inscriptions! See above, chapter 2, part 6; and below, no. 37 to this chap. 29. Skelton, VinlandMap, pp. 6 3 - 6 4 , 1 3 6 , 1 5 4 . In an earlier version of the Alexander Romance, the Syrian Christian Legend Concerning Alexander (ca. A.D. 514), Alexander is told that the nations living beyond the mountains are the Huns, who are related to the sons of the Biblical Japhet. In subsequent traditions, it is the armies of Gog and Magog who are said to dwell beyond the mountains at the "place where the sun riseth" (Koran 18): lists of the nations included in this army are given, but vary from source to source. In Jerome's Hebraicae quaestiones in libro geneseos, the Huns are identified with the armies of Gog and Magog. This development is detailed in Anderson, Alexander's Gate, pp. 2 0 - 3 4 ; and Maenchen-Helfen, World of the Huns, p. 5. 30. Wolohojian, Romance ofAlexander, pp. 1 1 3 - 1 4 (chapter 209 of the Armenian Pseudo-Callisthenes). Elsewhere, these are described as having great black bodies, heads like horses (!), enormous teeth, clawlike hands, and fire shooting out of their mouths: Hilka, Prosa-Alexanderroman, p. 234. "Horse-headed" Cynocephali are probably the fruit of a medieval iconographic tradition which was quite incapable of producing a realistic representation of a dog. Many such cynocephalics have heads of indeterminate species affiliation, varying between those of dogs, horses, asses, or apes. On the dog-horse hybrid, see Kretzenbacher, Kynokephale Damonen, pp. 6, 22. Manuscript traditions of the Historia de praeliis relate the same battle without mentioning horses' heads or spouting flames: Hilka and Steffens, Historia Alexandri Magni, p. 174. 31. Anderson, Alexander's Gate, pp. 3 5 - 5 7 ; andTrumpf, Vita Alexandri, p. 146. Anderson presents parallel texts of early Greek manuscript accounts, the Greek Pseudo-Callisthenes 3.29 in manuscript BC of Müller's edition, and the Byzantine

Notes to Pages121-23267 Alexandrou, w. 5712-5799. The origin of the list of the races is the fourthcentury Syrian work of Ephram Syrus, Sermo de Fine Extreme: Anderson, Alexander's Gate, pp. 16, 38-42. The name anouphagoi, "sometimes referred to as dogmen," alternates with kynokephaloi in several versions: p. 42, n. 1. The Armenian Pseudo-Callisthenes mentions the gates, in passing, at the end of chapter 209: "And I put the gates together and carefully sealed up the place": Wolohojian, Romance of Alexander, pp. 155-56. 32. This is an illustration to an edition of the fourteenth-century Shahnama collected by Demotte, now on display in the Sackler Museum of the Smithsonian Institution, Washington D.C. (no. S. 86.0104). See plate 5. 33. The classical source on the history of the shifting locations of Alexander's Gate (as well as of the Caucasus range and the Caspian Sea) is Anderson, Alexander's Gate, pp. 1 - 1 0 4 , passim. Cf. Skelton, Vinland Map, pp. 30, 6 3 - 6 4 , 118, 134; Minorsky, Hudüd al-Alam, p. 112; and Rene Grousset, Sur les traces du Bouddha (Paris: Plon, 1957; reprint Librarie Academique Perrin, 1977), pp. 8 3 - 8 5 , 2 3 6 . Flavius Josephus is the first author to speak of Alexander's Gate, in his firstcentury A.D. Bellum Judaicum 7.7A. In A.D. 399, Jerome (Epistulae 77.8) identifies the peoples held behind the Caucasus by Alexander's Gate as the Huns: MaenchenHelfen, World of the Huns, p. 4. It is the Syrian Christian Legend Concerning Alexander which, drawing on a lost Persian account, locates the gates in extreme northeast Asia (Anderson,Alexander's Gate, pp. 28,91-95), which is the source for most, if not all of the Arabic as well as Tartar traditions of Alexander's (and later Cinggis Qahan's) conquest of Far East Asia. 34. The Cynocephali first appear on the northern edges of world maps in the Cosmography of the seventh-century Aethicus Ister: Anderson, Alexander's Gate, pp. 51-53. An indication of the confusion that reigned over their location is the use of the term "Ethiopia" for the trans-Caucasian hinterlands north of Colchis from the ninth century onwards: an eleventh-century source names the "Scythian" Iberians, Soussians, Phoustes, and Alani as inhabiting this northern "Ethiopia": Dvornik, Idea of Apostolicity, p. 225. "Indian kings who hunt with women" and men who inhabit the Caucasus, who have intercourse with their women in the open and eat the bodies of their kinsmen are already mentioned in Megasthenes, cited in Strabo Geographica 15.1.55-56. 35. "Anthröpophagoi, hoi legomenoi Kunokephaloi." Identical descriptions are found in Greek and Latin translations of Pseudo-Methodius (Anderson, Alexander's Gate, p. 48), as well as in the Greek text (39.7) of Vita Alexandri (Trumpf, p. 146). Garamantes (who also have important canine connections in Pliny), Amazons, and Cynocephali are numbered 16,17, and 18 in parallel lists generated from various translations of chapter 8 of Pseudo-Methodius by Ernst Sackur, Sibyllinische Texte und Forschungen (Halle: Miermeyer, 1898), p. 37. In chapter 7 of the Latin version of Pseudo-Methodius, these evil races are said to eat, among other things, raw dog meat: this corresponds to the Indian traditions regarding the Svapacas, who are also located, in early sources, "beyond the northern pale." See below, chapter 4, part 3. Translations and discussions of the pertinent passages of Pseudo-Methodius are Bios

238

n o t e s t o p a g e s 98-100

in Anderson, Alexander's Gate, pp. 2 6 - 3 8 , 4 4 - 5 1 ; Sackur, Sibyllinische Texte, pp 6 0 - 9 6 . On another (6th-century) Syriac source that reproduces the same information, see Ahrens and Krüger, Sogenannte Kirchengeschichte, p. 253. Another sixthcentury Syrian source, the Christian Legend ofAlexander, gready influenced the Ethiopian History of Alexander: Anderson, Alexander's Gate, p. 32. A translation and history of the Syriac Book of the Cave of Treasures, dated between the fourth and sixth centuries A.D., is E. A. T. Wallis Budge, The Book ofthe Cave ofTreasures (London: n.p., 1928). See also Sackur, Sybillinische Texte, pp. 167 and 183. A very early reference to the Huns is Ammianus who, in his late fourth-century Res Gestae, names the Huns together with the Anthropophagi and Amazones: MaenchenHelfen, World ofthe Huns, p. 18. 36. Anderson,Alexander's Gate, pp. 2 4 - 2 5 , 4 4 - 5 1 ; Budge, Cave ofTreasures, pp. 114-35; and Wittkower, "Marvels of the East," pp. 167,183. 37. Jonaton (Yonton) is found in Budge, Cave ofTreasures, pp. 142-44. Compare Flavius Josephus (Antiquitates Judaicae 1.38), who, in his discussion of the division of world between Noah's offspring, states that the easternmost portion of the earth fell to the thirteen sons of Yoqtan, himself the great great grandson of Shem: Francis Schmidt, "Entre Juifs et Grecs: Le modele indien," in L'Inde et I'Imaginaire, ed. Catherine Weinberger-Thomas, Purusartha 11 (Paris: Editions de l'E.H.E.S.S., 1988), p. 33. See also Sackur, Sibyllinische Texte, pp. 5 - 7 , 15-20. Puntos, Ham's capital, evokes the Punt of the dog-faced baboons of Egyptian New Empire inscriptions: see above, chapter 2, part 6. The Cynocephali Alexander encounters in the Historia de prdiis are said to live across the last river of India ("ad ultimas silvas Indie . . . iuxta fluvium"), which thus would have constituted the eastern edge of the world: Hilka and Steffens, Historia Alexandri Magni, pp. 170, 174. Nimrod himself is identified with the constellation Orion, whose associations with the dog days and the dog-star Sirius were oudined in chapter 2. Gaignebet, A plus hault sens, 1:297. 38. Sackur, Sibyllinische Texte, pp. 19-22. 39. Kretzenbacher, Kynokephale Ddmonen, pp. 115-24. 40. Ibid., pp. 119-20. 41. Virgi\Aenead 1.490,11.659. In Aeschylus's Eumenides (685-88), the Amazons are women warriors who besieged Athens in a former time. The Scythians are mentioned a few verses later (703). 42. Kandake, "The Candace," was the tide of the queen of Ethiopia: Thiel, Leben und Taten Alexanders, p. 191; and Merkelbach,Quellen, pp. 1 4 6 , 1 6 3 - 6 4 , 2 1 5 . See also Hilka and Steffens, Historia Alexandri Magni, pp. 148-52. Lipsius (Apokryphen Apostelgeschichten 1:621) and Flamion (Actes Apocryphes de I'Apotre Andre, p. 314) locate Candace in Lydia, in western Anatolia. In the Greek (B) version of the Pseudo-Callisthenes, Alexander's dealings with Candace and the Amazons are in 3.18-24; he encounters the Cynocephali in 3.28.2: Thiel, Leben und Taten Alexanders, pp. 154-55. Other such juxtapositions, in Greek and Latin sources, are discussed in Anderson, Alexander's Gate, pp. 34-41. In the Armenian Pseudo-Callisthenes, he meets the former in chapters 225-54. A cynocephalic race other than the Dog-Men he battled in chapter 209 (see above, n. 30 to this chapter), is men-

Notes to Page 58

239

tioned in chapter 259, in a place "where neither sky nor earth could be seen": Wolohojian, Romance of Alexander, p. 146. 43. Kandaules was the name of a historical ruler of the dyansty of Sardis, in western Anatolia, adjacent to Lydia, through which Alexander passed on his outward journey to the East. More important is a fragment from the Greek poet Hipponax: "O Hermes, dog-strangler [kunagches], in the Maeonian language Kandaules, fight with me!" The import of this passage is twofold. First, an equivalence is made between the Greek god Hermes, the "dog-strangler," and a Maeonian god named Kandaules. More than this, we are in the presence of an etymological homologization: in Maeonian, kan- was the term for "dog," and daula- meant "strangler." The Maeonian language belonged to the Indo-European family, in which the root for "strangle" is *dhau. It is this same root that generates the ethnic names Daoi (the Dacians, whose homeland corresponds to present-day Romania), who appear to be related to the Dahae (who lived on the eastern shore of the Caspian, the later homeland of the Parthians): the Daoi and Dahae were thus "Stranglers." As Mircea Eliade has demonstrated, the "stranglers" with which these peoples identified themselves were wolves (dhaunos, in Illyrian). A number of other peoples with "wolfish" names (the Lycians, for example) are found in the same Balkan and Anatolian regions; the Sardis of the historical Kandaules is one of these. While the relationship between Kandaules, the brother of the queen of the Amazons, the dogstrangling Hermes of the Maeonians, and the wolfish stranglers of Dacia is unclear, it is the canine trope that unites all of these elements. See T. Burrow, "A Note on the Indo-Iranian Root Kan- 'Small' and the Etymology of Latin Cants 'Dog,'" Transactions ofthe Philological Society (1983): 161; Olivier Masson, Lesfragments dupoete Hipponax (Paris: 1962), pp. 104-5 and n. 6; Mainoldi, Ulmage du loup, p. 76; Eliade, "The Dacians and the Wolves," in Zalmoxis, pp. 1 - 2 . The term kunagches is also employed, in Greek, to describe "l'angoisse qui serre la gorge des pendus": Gaignebet, A plus hault sens, 2:327 and Mainoldi, Ulmage du loup, pp. 78-79. Gaignebet further relates this term to the name Guignefort, the greyhound who was "sainted" in southeastern France before the fourteenth century. See also Mayrhofer, Concise Etymological Dictionary, s.v. svaghnin (3:403), for the Sanskrit cognate of kunagches. See above, chapter 1, note 52. 44. In the Historia Prdiis for example, Alexander comes into contact with Candace and the Amazons in 3.25-26 and the bearded women and Cynocephali in 3.27: Hilka and Steffens, Historia Alexandri Magni, pp. 148-52, 172-74. 45. On Piano Carpini's account, see Skelton, VinlandMap, pp. 44, 49, 70-72. According to this source, Cinggis Qahan, the emperor of the Tartars, was the Mongol Alexander, and many of his wondrous feats are clear borrowings from the Alexander Romance: see below, chapters 6 and 9. On Cuckoo and Mugur, see Moses Gaster, Rumanian Bird and Beast Stories, Publications of the Folk-Lore Society, no. 75 (London: Sidgwick & Jackson, 1915), pp. 284-87. A Magyar variant names the brothers Hunor and Magor. Here, the two have no connection with Alexander or the Amazons. Rather, they are hunters who follow a doe who leads them eastward from their homelands and into the fertile plain ofMaeotis, where they settle. Later, they carry off the wives and daughters of the region and unite with them to found

240

n o t e s t o p a g e s 98-100

the Hun and Magyar races. This may have originally been a Hun myth: Eliade, Zalmoxis, pp. 142-44. 46. On the Greek Island of the Hesperides, see Gemet, Anthropologic, pp. 194^ 95, 200. On the Canary Islands, see Skelton, VinlandMap, pp. 122, 137,154-59; Lurker, "Hund als Symboltier," p. 136; Biedermann, "Die 'Hund-Inseln,'" pp. 9 9 - 1 0 8 ; and William H. Babcock, Legendary Islands ofthe Atlantic, American Geographic Society: Research Series, no. 8 (New York: American Geographic Society, 1922), pp. 34-43. On hespera and hesperos see the Iliad 10.317-18; and Scherer, Gestirnnamen, p. 83. Isidore of Seville divides Ethiopia between the Hesperii to the West, the Garamantes in the center, and the Indii to the East: Etymologiae 9.2.12728. Much later, the first-century B.C. Diodorus of Sicily, the most imaginative and least reliable of all the ancient historians, located the land of the "Western Amazons" on the island of Hesperia: Diodorus of Sicily Bibliotheca Historica 1.3.4. See Babcock, Legendary Islands, p. 16. Strabo's reference to Spanish Galicia as the land of the dead is in Geographica 3.1.4, cited in Gaignebet, A plus hault sens 1:297-98 and 2:429, who draws the connection between the ancient gods and the Christian St. James of Compostello. 47. The Voyage ofSaint Brendan: Journey to the Promised Land, ed. and tr. John J. O'Meara (Dublin: Dolmen, 1978), pp. 11-14; and Kretzenbacher, Kynokephale Ddmonen, pp. 4 5 - 4 6 and plate 7, p. 84, which reproduces Navigatio Sancti Brandani, Cod. Pal. Germ. 60, p. 177. Plutarch is perhaps referring to this tradition when he states in Peri ton Prosöpon that according to barbarian mythology, Zeus imprisoned Kronos-Satum on an island far to the west of Britain: Gaignebet, A plus hault sens, 1:463,2:459. 48. Skelton, VinlandMap, pp. 122-23; 137-38; 154; 159; 232-34. 49. Yann Brekilien, La mythologie celtique (Paris: Marabout, 1981) pp. 13, 22, 1 2 9 , 2 2 5 - 2 7 , 2 9 2 , 3 7 4 ; and Duanaire Finn: The Book ofthe Lays ofFionn, 2 vols., ed. and tr. Gerard Murphy (London: Irish Texts Society, 1933), 2:23-27. In all of these "indigenous" European traditions, the Cynocephali are giants. The Irish tradition may have been inspired by the Alexander legend: Friedman, Monstrous Races, p. 228, n. 40. See also Raymond-Riec Jestin, "Le poeme d'En-me-er-kar/'i^iw de PHistoire des Religions 151 (April-June 1957): 214. 50. The Cuchullin Saga in Irish Literature, ed. Eleanor Hull (London: Nutt, 1898), pp. 139-42; 2 5 4 - 5 5 . See also Fraser, "Passion," pp. 307-25, and Gaidoz, "St. Christophe," p. 197. 51. Ahrens and Krüger, Sogennante Kirchengeschichte, p. 253. Paul Pelliot ("Femeles [Island of Women]," in Notes on Marco Polo, 2 vols. [Paris: Imprimerie Nationale, 1963], 2:676) suggests that this source is the earliest textual connection between Eastern and Western Amazon-Cynocephali traditions. 52. Adam of Bremen, Gesta Hammaburgensis ecclesiae pontificorum, in Scriptores rerumgermanicarum in usum scholarum exMonumentis Germaniae Historica, 3d ed., ed. Bernhard Schmeidler (Hannover: Hahnsche Buchhandlung, 1917), 2:24248, 256. 53. Adam of Bremen's source was the Danish king Svein Estriddson, whose kingdom was located on the western end of this sea; Kretschmar, Hundestammvater

Notes to Pages121-23

26 7

1:126; and Skelton, VinlandMap, p. 172. A scholiast specifies that the Cynocephali and Amazons lived on the islands of Oeland, Gothland, and Oesel. 54. Here, Adam conflates the Blemmyes with the Cynocephali, perhaps due to a misreading of a source citing Herodotus, who mentions the two races (he calls the headless Blemmyes akephaloi) in the same passage (Historia 4.191) or of PseudoCallisthenes (3.28) or Isidore of Seville (Etymologiae 11.3.15-17), who do the same. According to Maenchen-Helfen (World of the Huns, p. 143), the Blemmyes were a historic people living in the Sudan, one of whose priests dedicated an inscription to Isis and Osiris in A.D. 452. 55. A scholiast to this text relates an expedition, made by King Emond of Sweden into the lands of the Scythians, that was annihilated by Amazon women, who poisoned their water. Adam also suggests that the virginal Amazons conceive by drinking water. This is a widespread mythic theme, found across central and East Asia. See below, chapter 6, part 4. 56. The late tenth-century Persian geography, the Hudüd al-'Âlam 4.17 (Minorsky, pp. 5 8 - 5 9 ) speaks of an Isle of Women in the Baltic region. According to Minorsky's notes (p. 191), this appellation arose from Germanic confusion over the name of a Finnish tribe called the Kwen-en and the word kwen, which means "woman" in Scandanavian languages (in the modem Danish kvinde, for example). 57. Friedman, Monstrous Races, p. 232, n. 72. 58. Kretschmar, Hundestammvater, 1:127, 131, 137, 145, 155-65; 2:183-84. 59. Kretzenbacher, Kynokephale Damonen, pp. 8 3 , 9 9 - 1 0 1 . 60. Ibid., pp. 15, 2 4 - 2 5 , 34, 8 5 - 8 9 , 101-2. See also Lecouteux, Les monstres dans la litterature allemande, 2:23, and Paul, Wolf Fuchs und Hund, pp. 9 6 - 1 0 1 . 61. The southern Slavic folksong is reproduced and translated into German by Kretzenbacher in Kynokephale Damonen (p. 21), the authoritative work on southeastern European traditions of cynocephaly. See especially pp. 5 - 2 6 , 99-134. See also Kretschmar, Hundestammvater, 1:134; and Bengt Holbek and lorn Pio, Fabeldyr og sagnfolk (Copenhagen: Politikkens Forlag, 1979), p. 258. Accounts of Mongol women living with cynocephalic men reached Europe through the A.D. 1245 travel account of the Franciscan friar John of Piano Carpini, the Historia Mongalorum [sic] quos nos Tartaros appellamus (Description of the Mongols whom we Call Tartars): Skelton, VinlandMap, pp. 2 2 - 2 3 . See below, chapter 6, part 4, and notes 55 and 56. 62. Friedman, Monstrous Races, pp. 3 7 - 4 5 ; Anderson, Alexander's Gate, pp. 9 6 100. 63. Paulus Deaconus's De Gesta Longobardiensis (1.11), a text which draws on a seventh-century source, places the Cynocephali in Western Europe itself, in the service of the king of the Lombards. The same source mentions the Amazons: Konrad Miller, Mappae Mundi: die altesten Weltkarten, 6 vols. (Stuttgart: Roth, 1895-96), 4:18; and Maenchen-Helfen, World of the Huns, p. 127. Benedict, the Polish friar who accompanied Piano Carpini on his travels to the court of Cinggis Qahan, locates the Cynocephali in Russia: Wittkower, "Marvels of the East," p. 196. 64. Jacqueline Bolens-Duvernay, ("Les Geants Patagons ou l'espace retrouve: Les debuts de la cartographie 2xnc\2x'\czri\stz? VHomme 28: 2 - 3 [April-September

242

n o t e s t o p a g e s 98-100

1988], pp. 162-63) reproduces sixteenth- and seventeenth-century English accounts of cynocephalic giants sighted in Patagonia and "between the Equator and the Tropic of Capricorn." 65. Dvornik, Idea of Apostolicity\ p. 225. The Syriac version of the PseudoCallisthenes, especially the later books in which the accounts of Dog-Men and Amazons are found, drew on a sixth-century Pehlevi source: Budge, Alexander Book m Ethiopia, p. xix, and Wolohojian, Romance of Alexander, pp. 2,5. See above, notes 34 and 51 to this chapter. 66. The classic source on ancient maps is the six-volume work of Miller (Mappae Mundi, 1:48, 3:33,4:18, 5:49, and plates, on the Cynocephali). Another excellent source is Skelton, VinlandMap, especially pp. 109-239. Ithacus Ister, the first cartographer to locate the Cynocephali to the north, was influenced by PseudoMethodius: Anderson, Alexander's Gate, pp. 51-53. See also Anderson, pp. 9 1 103; Lecouteux, Les monstres dans la litterature allemande, 2:26; Friedman, Monstrous Races, pp. 46, 84,231, and nn. 6 8 - 7 0 ; Wittkower, "Marvels of the East," pp. 173-175; and Kretschmar, Hundestammvater, 1:125. Cynocephali are placed immediately to the right of the magnificent Christ of the twelfth-century Vezelay tympanum: here, they are represented as the first of the monstrous races, situated at the limits of the earth, to whom the apostles were sent in their missions. See plate 1. 67. Prester John was variously identified with Cinggis Qahan himself and with the Nestorian Wang Khan of the Keraits. The transfer of his kingdom from central Asia to India occurred as early as 1177. From here, it would be relocated in Ethiopia in Africa, starting with Jordanus in 1321. It is this Ethiopian avatar of Prester John whom Jamaican Rastafarians see in the deceased emperor Haile Salassi. See Skelton, VinlandMap, pp. 48, 69, 131, and Wittkower, "Marvels of the East," pp. 181, 197. The Karakhitai had previously ruled northern China from Manchuria. Here the imperial family saw itself, in the tenth century, as being descended from a canine ancestor. We will return to this theme in chapter 6, part 4. 68. The same Marco Polo claims to have met a certain George, a Nestorian Christian who was sixth in the line of descent from Prester John. He also says that Prester John's kingdom was located in Manchuria (the homeland of the Ch'i-tan), that his Chinese name was Ung Khan (Wang Khan), that he had been defeated and killed by Cinggis Qahan, and that Ung and Mungol (Prester John's and the Mongol's kingdoms) were the same as Gog and Magog: Marco Polo, Travels (New York: Grosset and Dunlap, n.d.), pp. 7 1 - 7 5 , 87-88. 69. Wittkower, "Marvels of the East," p. 196. Ibn Battuta's account (Voyages, tr. C. Defremery and B. R. Sanguinetti, 3 vols. [Paris: Maspero, 1982], 3:291) may have a Chinese origin. The ninth-century Man shu, an account of the Man "barbarians" living on the Chinese side of the Burmese border, knows of Burma and Assam, and of the Indian Ocean. The reverse is also the case: Assam and Burma were aware of the southern Chinese aboriginals living to the north of them: Grousset, Sur les traces du Bouddha p. 188. These included the Man of Yunnan, who had an indigenous dog ancestry myth. See below, chapter 7. On the origins of the term cannibal, see Holbek and Pİ0, Fabeldyr, p. 258. On the Portuguese's naming of the Amazon River in Brazil, see Pelliot, "Femeles," p. 674. 70. Wittkower, "Marvels of the East," pp. 166-67.

Notes to Pages121-23267 71. On these sources, see Bruno Roy, "En marge du monde connu: les races des nionstres," in Guy Allard, Aspects de la marginalite au Moyen-Age (Montreal: Aurore, 1975), pp. 7 1 - 8 0 ; Seznec, Survival of the Pagan Gods, pp. 89, 94, 168, 171; and Meslin, Le merveilleux, pp. 30, 87,96. 72. Many of these are inventoried in Kretzenbacher, Kynokephale Ddmonen, pp. 46-61. The most renowned of these is the cynocephalic pair figured to the left of Christ on the twelfth-century tympanum at Vezelay, in Burgundy, France. An especially attractive example of such a medieval work is a fresco, dating from the first decade of the sixteenth century, painted in vivid colors on the vault of the Râby church in eastern Judand, Denmark. See plate 8. 73. Augustine D e civ itate Dei 16.2. 74. The agrios, "wilderness," was, in the Greek tragedies, the place of exile, of death to society, of the unburied dead, and of such carrion-feeding animals as the dog: outside of both the polis and Attica, this was the ou topos of the apolitical, the asocial, and thus the inhuman. On Greek attitudes towards monsters, barbarians, and non-citizens, see Friedman, Monstrous Races, pp. 2 6 - 3 5 and White, "Forms of Wildness," pp. 1 1 , 1 8 , 2 0 , 2 4 . 75. Augustine De civitateDei 15.20. As they were exiled east of Eden, it follows that the Orient should become the region in which their exiled offspring are found. Among the Plinian races, however, the Gymnosophists/Bragamanni were considered to be located in Eden itself (Freidman, Monstrous Races, p. 167). In rabbinical literature, it is said that a time of general degeneracy will be characterized by people who "will have the face of a dog": Sotah 9.15, cited in Afshar, Immortal Hound, p. 45. See also Friedman, Monstrous Races, p. 93. In the Zohar; Cain, born from the union of the Devil with Eve, gives rise to the cannibalistic races: pp. 9 6 - 9 9 . 76. Friedman, Monstrous Races, pp. 9 9 - 1 0 1 ; and Augustine De civitate Dei 16.4-5. 77. Friedman, Monstrous Races, p. 99, citing a Middle Irish treatise on the Six Ages. 78. Augustine, De civitate Dei 16.8. 79. Friedman, Monstrous Races, pp. 119, 121. A ninth-century theologian, Ratramnus of Corbie, addressed a letter (Epistola de Cynocephalis) to Rimbert of Hamburg in which he raised the same points regarding the humanity and thus the soteriology of the Cynocephali, and brought the Alexander Romance and the cynocephalic St. Christopher into his discussion. 80. Isidore of Seville Etymologiae 9.2.5, 10-11, within Isidore's general discussion of distant barbarian peoples (9.2.1-135). Isidore's general discussion of monsters and/as omens and portents is 11.3.1-3, with the special cases of the Cynocephali, Cyclops, and Giants discussed 11.3.12-15. Another medieval author who mentions dog-headed men in his discussion ofportenta is Hrabanus Maurus (784-856) [De Universo 7.7: Migne PL 111, col. 198], cited in Kretzenbacher, Kynokephale Ddmonen, pp. 34-35. On Christopher and the Chananeans, see chapter 2, part 4. See plate 2. 81. Isidore of Seville Etymologiae 9.2.62-66. 82. The original source was Syriac: Richter, Der deutsche S. Christoph, p. 37. Bactria and Balkh were the center of the Ephthalite Huns' power for some two hun-

244

n o t e s t o p a g e s 98-100

dred years following the fifth century. The (Ephthalite) Huns and their role in the mythology of Dog-Men are discussed in chapters 6 and 9. 83. Isidore of Seville Etymologiae 11.3.1-3. Isidore's etymology of monstra is derived from the verb monere "to warn": other authors derive it from monstrare, "to show," but also "to warn." 84. Roy, "En marge," pp. 75-76. 85. Isidore of Seville Etymologiae 9.3.12-15. In so doing, Isidore follows an ordering that goes back to Herodotus and the Alexander legend, and which may have contributed to the tradition of giant Cynocephali. See Klinger, "Hundsköpfıge Gestalten," pp. 121-22; and Wittkower, "Marvels of the East," p. 175. 86. Friedman, Monstrous Races, pp. 60-61. 87. The Saracen is so called because he is descended from Abraham's concubine Sarah. He is also called the Ishmaelite, because he is descended from the hairy wild man Ishmael (Isidore of Seville Etymologiae 9.2.6). His monstrous aspect is a conflation of the "Turk" with the "Scythian," an identification made classic by PseudoMethodius. 88. Senac, Ulmage de I'Autre, pp. 5 5 - 5 7 , 6 6 , 7 1 - 7 2 ; Friedman, Monstrous Races, p. 67; and Kretzenbacher, Kynokephale Damonen, pp. 90-91, citing Pseudo-Turpin 22. Kretzenbacher (p. 90) also mentions a ninth-century Carolingian source (the Gesta Caroli of Notker Balbulus) that depicts the Norman enemies of the Frankish Charlemagne as Cynocephali. Lecouteux (Les monstres dans la litterature allemande, 2:27) lists other medieval sources which describe battalions of Cynocephali in enemy armies. See above, chapter 2, part 5, for the medieval identification of Lares with man-eating cynocephalic spirits. 89. Senac, Ulmage de I'Autre, pp. 97-99. On Nimrod, see Friedman, Monstrous Races, p. 64. Baptism also "whitened" the black Saracen: pp. 65, 164,172,175. In an interesting ideological reversal of roles, Wycliffe, in attacking the Church in the fourteenth century, said, "We are the Muhammeds of the West": Senac, Ulmage de VAutre, p. 141. 90. Friedman, Monstrous Races, pp. 6 3 , 6 7 , 2 2 5 - 2 7 , nn. 6 - 9 , 2 4 - 2 5 ; and Hugo Buchthal, "A Miniature of the Pentecost from a Lectionary in Syriac " Journal ofthe Royal Asiatic Society ofGreat Britain and Ireland (1939): 613-15 and plates xiv-xvi. Buchthal describes three miniatures executed in the Byzantine style in which, in the place of the "nations" in the foreground, are depicted (a) a king and an animalheaded figure, (b) two figures, of which one is bicephalic and the other zoocephalic, and (c) a three-headed figure, of which one head is meant to represent the Ishmaelites. The tricephalic figure may be a Christian figuring of Cerberos or Hermanubis. A ninth-century icon depicts two Cynocephali flanking the Christ: these represent the Jews, who are to be redeemed at the second coming of Christ (Kretzenbacher, Kynokephale Damonen, p. 56). 91. Henri Cordier, Les Monstres dans la leğende et dans la nature: Nouvelles etudes sur les traditions teratologiques, vol. 1, Les cynocephales (Paris: Lafolye, 1890), pp. 1 5 17. The name "Burous" indicates a probable Buriat origin: the Buriats were a central Asian people whose name, Buri, means "wolf" in Turkish (Skelton, Vinland Map, p. 72). Japhet is held to be the father of the enclosed nations, and of the armies of Gog and Magog, in certain sources: Anderson, Alexander's Gate, p. 22.

Notes to Pages121-23267 92. G. G. Coulton, The Black Death (New York: J. Cape and H. Smith, 1930), pp. 2 2 - 2 7 , cited in Barbara Tuchman, A Distant Mirror: The Calamitous Fourteenth Century (Harmondsworth: Penguin, 1980), p. 104. This image evokes lycanthropy (wolfmanism), which continued to constitute a real threat to the medieval mind and which came to know a renaissance in the sixteenth century (Ragache, Les loups, pp. 104-11, and Meslin, Le merveilleux, pp. 136-38), and cynanthropy, associated from ancient times with disease and melancholia (Rorscher, "'Kynanthropie,'" pp. 1-50, passim). In another context, the inauspicious tarot card "The Fool" is portrayed as a strangely dressed man (perhaps one of Diogenes' Cynics) accompanied by a dog. Diogenes is himself called a "heavenly dog" in an elegy by Chersite of Megalopolis: Gaignebet, A plus hault sens, 2:438. For medieval attitudes towards the dog in general, see Louise Gnâdiger, Huidan und Petitcreiu: Geştalt undFigur des Hundes in der mittelalterlichen Tristandichtung (Zurich: Adantis, 1971), pp. 9 16.

93. The vilain is barred from both heaven and hell: in the latter case, this is due to his stink (Tuchman, Distant Mirror, p. 175), or because he mistakes the place of exit of the soul from the body—i.e., his soul comes out through his lower end (Stanley Leman Galpin, Cortoisand Vilain [New Haven: Ryder's, 1905], p. 84)! A Francoprovençal passio, on the other hand, offers redemption for both indigenous European and foreign Cynocephali (Friedman, Monstrous Races, p. 229, n. 42). The indigenous peasant or Wild Man would not receive his "titres de noblesse" until the writings of Rousseau and Marx. 94. On the pygmies, see Friedman, Monstrous Races, pp. 194-96, and Pietro Janni, Etnografta e mito: Lastoria deiPigmei (Rome: Ateneo & Bizzarri, 1978). On the Jews, see Friedman, Monstrous Races, pp. 61, 69, and Kretschmar, Hundestammvater, 2:183-84. 95. Jacques LeGoff, "Les paysans et le monde rural dans la litterature du Haut Moyen Age (V C -VI C siecles)," in Agricoltura e mondo mrale in occidente neWaltoMedioevo, Settimane di Studio del Centro Italiano di Studi sull'alto Medioevo, no. 13 (Spoleto: 1966), pp. 731, 734-38. 96. Galpin, Cortois and Vilain, passim, especially p. 87. 97. Ibid., p. 77 and Micheline de Combarieu, "Image et representation du vilain dans les chansons de geste," in Exclus et systemes dyexclusion dans la litterature et la civilisation medievales, Colloque, Aix-en-Provence, 4 - 6 mars 1977 (Aix-enProvence: n.p., 1977), pp. 12-14. Wild vilains figure in the medieval Tristan et Tseult, in Aucassin et Nicolette (24.11.12-24), and in Chretien de Troye's Yvain (293-324). 98. Vita sancti Geosnovici, cited in Lecouteux, Les monstres dans la litterature allemande, 2:24. 99. Meslin, Le merveilleux, p. 86. CHAPTER 4

1. In the same section (Amarakosa 2.10.44-46) are found lists of gambling terms. For a detailed study of the symbolic relationship between dogs and dicing in India, see my article "Dogs Die," History ofReligions 23, no. 4 (May 1989): 2 8 3 303. Another term for outcaste, antyavâsâyin, is technically applied to the offspring

246

NOTES TO PAGES 98-100

of a Candâla man by a woman of either the Kşattr, Süta, Âyogava, Vaidehika, or Mâgadhay^s, or subcastes. The first three of thcscjâtis play important roles in the sacrifice of the four-eyed or "four-dice" dog in the context of the asvamedha rite. 2. Pânini (2.4.10) uses the term niravasita südra, whom Patanjali identifies, in his Mahâbhâsya (1.475), as the Pulkasas, Pulindas, Candâlas, and Mrtapâs, of which all, with the exception of Candâlas, are tribal or ethnic names. Antyavâsâyin is a term employed in Manu Smrti [hereafter referred to as Manu] 10.39; and the Mârkandeya Purâna [hereafter referred to as MP] 25.34-36. Manu also uses babya and bâhyavâsina as general terms. Avarnaja and antarala are used in the same sense in the Brhat Samhitâ 89.1 .kha and Yao pi shih, 137 Yüeh-chi. See Kushans Yunnan province, 124, 127, 145, 159, 187,242,278-80,282 Tüpen chi, 131,143,158,168 Yü the Great, 164,170-71,279, 287; son of a monster, 170

index

Zacharios of Mitylene, 60,294 Zodiac. See Astronomy and astrology, Greco-Egyptian Zoocephaly, 18, 19, 43, 59, 62, 71, 244, 274. See also Jackals and jackal-headed gods Zoroastrianism. See Iranian religion and mythology Zwierzina, Konrad, 36

j'

j»ij

J