Table of Contents Title Page Copyright Page PREFACE: CHAPTER 1 - INTRODUCTION: WORKINGWITH AVAILABLE LIGHT CHAPTER 2 - TIME AND SPACE IN INDIA 50 Million to 50,000 BCE CHAPTER 3 - CIVILIZATION IN THE INDUS VALLEY 50,000 to 1500 BCE CHAPTER 4 - BETWEEN THE RUINSAND THE TEXT 2000 to 1500 BCE CHAPTER 5 - HUMANS, ANIMALS, AND GODS IN THE RIG VEDA 1500 to 1000 BCE CHAPTER 6 - SACRIFICE IN THE BRAHMANAS 800 to 500 BCE CHAPTER 7 - RENUNCIATION IN THE UPANISHADS 600 to 200 BCE CHAPTER 8 - THE THREE (OR IS IT FOUR?) AIMS OF LIFE IN THE HINDU IMAGINARY CHAPTER 9 - WOMEN AND OGRESSES IN THE RAMAYANA 400 BCE to 200 CE CHAPTER 10 - VIOLENCE IN THE MAHABHARATA 300 BCE to 300 CE CHAPTER 11 - DHARMA IN THE MAHABHARATA 300 BCE to 300 CE CHAPTER 12 - ESCAPE CLAUSES IN THE SHASTRAS 100 BCE to 400 CE CHAPTER 13 - BHAKTI IN SOUTH INDIA 100 BCE to 900 CE CHAPTER 14 - GODDESSES AND GODS IN THE EARLY PURANAS 300 to 600 CE CHAPTER 15 - SECTS AND SEX IN THE TANTRIC PURANAS AND THE TANTRAS 600 to 900 CE CHAPTER 16 - FUSION AND RIVALRY UNDER THE DELHI SULTANATE 650 to 1500 CE CHAPTER 17 - AVATAR AND ACCIDENTAL GRACE IN THE LATER PURANAS 800 to 1500 CE CHAPTER 18 - PHILOSOPHICAL FEUDS IN SOUTH INDIA AND KASHMIR 800 to 1300 CE CHAPTER 19 - DIALOGUE AND TOLERANCE UNDER THE MUGHALS 1500 to 1700 CE CHAPTER 20 - HINDUISM UNDER THE MUGHALS 1500 to 1700 CE CHAPTER 21 - CASTE, CLASS, AND CONVERSION UNDER THE BRITISH RAJ 1600 to 1900 CE CHAPTER 22 - SUTTEE AND REFORM IN THE TWILIGHT OF THE RAJ 1800 to 1947 CE CHAPTER 23 - HINDUS IN AMERICA 1900 - CHAPTER 24 - THE PAST IN THE PRESENT 1950 - CHAPTER 25 - INCONCLUSION, OR, THE ABUSE OF HISTORY Acknowledgements CHRONOLOGY GUIDE TO PRONUNCIATION AND SPELLINGOF WORDS IN SANSKRIT AND OTHERINDIAN LANGUAGES ABBREVIATIONS GLOSSARY OF TERMS IN INDIAN LANGUAGES AND NAMES OF KEY FIGURES NOTES BIBLIOGRAPHY: WORKS CITED AND CONSULTED PHOTO CREDITS INDEX ABOUT THE AUTHOR ALSO BY WENDY DONIGER Siva, the Erotic Ascetic The Origins of Evil in Hindu Mythology Dreams, Illusion, and Other Realities Splitting the Difference: Gender and Myth in Ancient Greece and India TRANSLATIONS: The Rig Veda The Laws of Manu and Kamasutra THE PENGUIN PRESS Published by the Penguin Group Penguin Group (USA) Inc., 375 Hudson Street, New York, New York 10014, U.S.A. • Penguin Group (Canada), 90 Eglinton Avenue East, Suite 700, Toronto, Ontario, Canada M4P 2Y3 (a division of Pearson Penguin Canada Inc.) • Penguin Books Ltd, 80 Strand, London WC2R 0RL, England • Penguin Ireland, 25 St. Stephen's Green, Dublin 2, Ireland (a division of Penguin Books Ltd) • Penguin Books Australia Ltd, 250 Camberwell Road, Camberwell, Victoria 3124, Australia (a division of Pearson Australia Group Pty Ltd) • Penguin Books India Pvt Ltd, 11 Community Centre, Panchsheel Park, New Delhi - 110 017, India • Penguin Group (NZ), 67 Apollo Drive, Rosedale, North Shore 0632, New Zealand (a division of Pearson New Zealand Ltd) • Penguin Books (South Africa) (Pty) Ltd, 24 Sturdee Avenue, Rosebank, Johannesburg 2196, South Africa Penguin Books Ltd, Registered Offices: 80 Strand, London WC2R 0RL, England First published in 2009 by The Penguin Press, a member of Penguin Group (USA) Inc. Copyright © Wendy Doniger, 2009 All rights reserved Acknowledgments for permission to reprint copyrighted works appear on page 754. Illustration credits appear on page 754. Library of Congress Cataloging-in-Publication Data Doniger, Wendy. The Hindus : an alternative history / Wendy Doniger. p. cm. Includes bibliographical references and index. eISBN : 978-1-101-02870-4 1. Hinduism--Social aspects--History. 2. Women in Hinduism--History. 3. Pariahs in Hinduism--History. 4. Hinduism--Relations. I. Title. BL1151.3.D66 2009 294.509--dc22 2008041030 Without limiting the rights under copyright reserved above, no part of this publication may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise), without the prior written permission of both the copyright owner and the above publisher of this book. The scanning, uploading, and distribution of this book via the Internet or via any other means without the permission of the publisher is illegal and punishable by law. Please purchase only authorized electronic editions and do not participate in or encourage electronic piracy of copyrightable materials. Your support of the author's rights is appreciated. KATHERINE ULRICH--student, friend, editor supreme-- and WILL DALRYMPLE--inspiration and comrade in the good fight INDIA'S MAJOR GEOGRAPHICAL FEATURES INDIA FROM 2500 BCE TO 600 CE INDIA FROM 600 CE TO 1600 CE INDIA FROM 1600 CE TO THE PRESENT PREFACE: THE MAN OR THE RABBIT IN THE MOON AN ALTERNATIVE HISTORY The image of the man in the moon who is also a rabbit in the moon, or the duck who is also a rabbit, will serve as a metaphor for the double visions of the Hindus that this book will strive to present. Since there are so many books about Hinduism, the author of yet another one has a duty to answer the potential reader's Passover question: Why shouldn't I pass over this book, or, Why is this book different from all other books? This book is not a brief survey (you noticed that already; I had intended it to be, but it got the bit between its teeth and ran away from me), nor, on the other hand, is it a reference book that covers all the facts and dates about Hinduism or a book about Hinduism as it is lived today. Several books of each of those sorts exist, some of them quite good, which you might read alongside this one.1 The Hindus: An Alternative History differs from those books in several ways. [TOP] The Mark on the Moon, [MIDDLE] Wittgenstein's Duck/ Rabbit, and [BOTTOM] The Rabbit in the Moon First, it highlights a narrative alternative to the one constituted by the most famous texts in Sanskrit (the literary language of ancient India) and represented in most surveys in English. It tells a story that incorporates the narratives of and about alternative people--people who, from the standpoint of most high-caste Hindu males, are alternative in the sense of otherness, people of other religions, or cultures, or castes, or species (animals), or gender (women). Part of my agenda in writing an alternative history is to show how much the groups that conventional wisdom says were oppressed and silenced and played no part in the development of the tradition--women, Pariahs (oppressed castes, sometimes called Untouchables)--did actually contribute to Hinduism. My hope is not to reverse or misrepresent the hierarchies, which remain stubbornly hierarchical, or to deny that Sanskrit texts were almost always subject to a final filter in the hands of the male Brahmins (the highest of the four social classes, the class from which priests were drawn) who usually composed and preserved them. But I hope to bring in more actors, and more stories, upon the stage, to show the presence of brilliant and creative thinkers entirely off the track beaten by Brahmin Sanskritists and of diverse voices that slipped through the filter, and, indeed, to show that the filter itself was quite diverse, for there were many different sorts of Brahmins; some whispered into the ears of kings, but others were dirt poor and begged for their food every day. Moreover, the privileged male who recorded the text always had access to oral texts as well as to the Sanskrit that was his professional language. Most people who knew Sanskrit must have been bilingual; the etymology of "Sanskrit" ("perfected, artificial") is based upon an implicit comparison with "Prakrit" ("primordial, natural"), the language actually spoken. This gives me a double agenda: first to point out the places where the Sanskrit sources themselves include vernacular, female, and lower-class voices and then to include, wherever possible, non-Sanskrit sources. The (Sanskrit) medium is not always the message;a it's not all about Brahmins, Sanskrit, the Gita . I will concentrate on those moments within the tradition that resist forces that would standardize or establish a canon, moments that forged bridges between factions, the times of the "mixing of classes" ( varna-samkara ) that the Brahmins always tried--inevitably in vain--to prevent. Second, in addition to focusing on a special group of actors, I have concentrated on a few important actions, several of which are also important to us today: nonviolence toward humans (particularly religious tolerance) and toward animals (particularly vegetarianism and objections to animal sacrifice) and the tensions between the householder life and renunciation, and between addiction and the control of sensuality. More specific images too (such as the transposition of heads onto bodies or the flooding of cities) thread their way through the entire historical fabric of the book. I have traced these themes through the chapters and across the centuries to provide some continuity in the midst of all the flux,2 even at the expense of what some might regard as more basic matters. Third, this book attempts to set the narrative of religion within the narrative of history, as a linga (an emblem of the god Shiva, often representing his erect phallus) is set in a yoni (the symbol of Shiva's consort, or the female sexual organ), or any statue of a Hindu god in its base or plinth ( pitha ). I have organized the topics historically in order to show not only how each idea is a reaction to ideas that came before (as any good old-fashioned philological approach would do) but also, wherever possible, how those ideas were inspired or configured by the events of the times, how Hinduism, always context sensitive,3 responds to what is happening, at roughly the same moment, not only on the political and economic scene but within Buddhism or Islam in India or among people from other cultures entering India. For Hinduism, positioning kings as gods and gods as kings, seldom drew a sharp line between secular and religious power. In recent years a number of historians of religions, particularly of South Asian religions, have contextualized particular moments in the religious history of the subcontinent.4 This book attempts to extend that particularizing project to the whole sweep of Indian history, from the beginning (and I do mean the beginning, c. 50,000,000 BCE) to the present. This allows us to see how certain ongoing ideas evolve, which is harder to do with a focus on a particular event or text at a particular moment. This will not serve as a conventional history (my training is as a philologist, not a historian) but as a book about the evolution of several important themes in the lives of Hindus caught up in the flow of historical change. It tells the story of the Hindus primarily through a string of narratives. The word for "history" in Sanskrit, itihasa , could be translated as "That's what happened," giving the impression of an only slightly more modest equivalent of von Ranke's phrase for positivist history: " Wie es [ eigentlich ] gewesen ist " ("The way it [really] happened"). But the iti in the word is most often used as the Sanskrit equivalent of "end quote," as in "Let's go [ iti ]," he said. Itihasa thus implies not so much what happened as what people said happened ("That's what he said happened")--narratives, inevitably subjective narratives. And so this is a history not of what the British used to call maps and chaps (geography and biography) but of the stories in hi-story. It's a kind of narrative quilt made of scraps of religion sewn in next to scraps of social history, a quilt like those storytelling cloths that Indian narrators use as mnemonic devices to help them and the audience keep track of the plot. The narrator assembles the story from the quilt pieces much as the French rag-and-bones man, the bricoleur , makes new objects out of the broken-off pieces of old objects (bricolage).5 Like any work of scholarship, this book rests on the shoulders of many pygmies as well as giants. I have kept most of the scholarly controversies out of the text, after laying out the rules of the game in these first two chapters of methodological introduction and in the pre-Vedic period (chapters 2 through 4), which might stand as paradigms for what might have been done with all the other chapters, as well as a few other places where the arguments were so loony that I could not resist the temptation to satirize them. Many a "fact" turns out, on closer inspection, to be an argument. There is another story to be told here: how we know what we know, what we used to believe, why we believe what we believe now, what scholars brought up certain questions or gave us the information we now have, what scholars now challenge that information, and what political factors influenced them. Those arguments tell a story that is interesting in itself but to which I merely allude from time to time. I also write in the shadow of a broad scholarship of theories about religion and history, and I will keep that too out of the text. I have tried to avoid setting my opinions against those with whom I disagree or using them as fall guys, beginning an argument by citing the imagined opponent. I have, rather, simply presented each subject in what I believe to be the best scholarly construction, in order to concentrate on the arguments about it within the Hindu texts themselves. Many crucial questions remain unanswered, and I hope that this book will inspire some readers to go back to the sources and decide for themselves whether or not they agree with me. The relevant materials can be found in the bibliography as well as in the notes for each chapter, which will also provide browsing material for those readers (I confess that I am one of them) who go straight to the back and look at the notes and bibliography first, reading the book like Hebrew, from right to left, to see where the author has been grazing, like dogs sniffing one another's backsides to see what they have eaten lately.b SANSKRITIZATION, DESHIFICATION, AND VERNACULARIZATION Sanskrit texts from the earliest period assimilated folk texts that were largely oral and composed in languages other than Sanskrit, vernacular languages. But even in the Vedic age, Sanskrit was not what has been called a kitchen language, c not the language in which you said, "Pass the butter."6 (Actually, Brahminsprobably did say, "Pass the butter," in Sanskrit when they put butter as an oblation into the fire in the course of the sacrifice, but those same Brahmins would have to have known how to say it in another language as well, in the kitchen.) At the very least, those male Sanskritists had to be bilingual in order to talk to their wives and servants and children.d It was through those interactions that oral traditions got their foot in the Sanskrit door. Henry Higgins, in George Bernard Shaw's Pygmalion , is said to be the author of Spoken Sanskrit , and many priests and scholars can speak Sanskrit, but no one ever spoke only pure Sanskrit. Sanskrit and oral traditions flow back and forth, producing a constant infusion of lower-class words and ideas into the Brahmin world, and vice versa. It must have been the case that the natural language, Prakrit, and the vernaculars came first, while Sanskrit, the refined, secondary revision, the artificial language, came later. But South Asianists often seem to assume that it is the other way around, that the dialects are "derived from Sanskrit," because Sanskrit won the race to the archives and was the first to be written down and preserved, and we only encounter vernaculars much later. So we say that Sanskrit is older, and the vernaculars younger. But Sanskrit, the language of power, emerged in India from a minority, and at first its power came precisely from its nonintelligibility and unavailability, which made it the power of an elite group.7 Walt Kelly's Pogo used to use the word "Sam-skrimps" to describe highfalutin double-talk or manipulative twaddle. Many Euro-Americans mispronounce it "Sanscript," implying that it is a language without ( sans ) an (intelligible) script, or "Sand-script," with overtones of ruined cities in the desert or a lost language written in sand. The sociologist M. N. Srinivas, in 1952, coined the useful term "Sanskritization" to describe the way that Vedic social values, Vedic ritual forms, and Sanskrit learning seep into local popular traditions of ritual and ideology (in part through people who hope to be upwardly mobile, to rise by imitating the manners and habits, particularly food taboos, of Brahmins, and in particular avoiding violence to animals).8 Indian society, in this view, is a permanent floating game of snakes and ladders (or, perhaps, snakes and ropes, recalling that Vedantic philosophers mistake snakes for ropes and that you can climb up on ropes in the Indian rope trick), which you enter in a state of impurity, gradually advancing over the generations toward the goal of Brahminical purity, trying to avoid the many pitfalls along the way.9 Tribal groups (Bhils, Gonds, etc.) might undergo Sanskritization in order to claim to be a caste, and therefore, Hindu.10 But the opposite of Sanskritization, the process by which the Sanskritic tradition simultaneously absorbs and transforms those same popular traditions, is equally important, and that process might be called oralization, or popularization, or even, perhaps, Deshification (from the "local" or deshi traditions) or Laukification, from what Sanskrit calls laukika ("of the people" [loka] ). Let's settle on Deshification. The two processes of Sanskritization and Deshification beget each other. Similarly, through a kind of identificatio brahmanica ,11 local gods take on the names of gods in Sanskrit texts: Murukan becomes Skanda, a kind of Sanskritization, while at the same time there is an identificatio deshika , by which Sanskrit gods take on the characteristics of local gods, and to the people who worship Murukan, it is Murukan who is absorbing Skanda, not the reverse. "Cross-fertilization" might be a good, equalizing term for the combination of the two processes. "Written" does not necessarily mean "written in Sanskrit," nor are oral texts always in the vernacular (the Rig Veda, after all, was preserved orally in Sanskrit for many centuries before it was consigned to writing). We cannot equate vernacular with oral, for people both write and speak both Sanskrit and the vernacular languages of India, though Sanskrit is written more often than spoken. The distinction between Sanskrit and the vernacular literatures is basically geographical: Though there are regional Sanskrits, the vernaculars, unlike Sanskrit, are defined and named by their place of origin (Bangla from Bengal, Oriya from Orissa, and so forth), while the script in which Sanskrit is most often written allegedly has no particular earthly place of origin (it is called "the [script of the] city of the gods [ deva-nagari ]"). Once people departed from the royal road of Sanskrit literary texts, there were thousands of vernacular paths that they could take, often still keeping one foot on the high road of Sanskrit. The constant, gradual, unofficial mutual exchange between Sanskrit and the vernacular languages, the cross-fertilization, underwent a dramatic transformation toward the middle of the second millennium: Local languages were now promoted officially, politically, and artistically,12 replacing the previously fashionable cosmopolitan and translocal language, Sanskrit. Instead of nourishing and supplementing Sanskrit, the vernacular languages as literary languages began to compete with Sanskrit as the language of literary production. This process has been called, in imitation of Srinivas's "Sanskritization" (and in contrast with both Deshification and the more mutually nourishing, two-way process of cross-fertilization) vernacularization, "the historical process of choosing to create a written literature, along with its complement, a political discourse, in local languages according to models supplied by a superordinate, usually cosmopolitan, literary culture,"13 or "a process of change by which the universalistic orders, formations, and practices of the preceding millennium were supplemented and gradually replaced by localized forms."14 The great divide is between written and nonwritten, not between Sanskrit and the vernaculars, particularly as the Sanskrit corpus comes to be Deshified and the vernaculars eventually became Sanskritized themselves, imitating Sanskrit values and conventions, sharing many of the habits of the Sanskrit Brahmin imaginary, such as grammars and lexicons.15 The bad news is that some of the vernacular literatures are marred by the misogynist and class-bound mental habits of Brahmins, while the good news is that even some Sanskrit texts, and certainly many vernacular texts, often break out of those strictures and incorporate the more open-minded attitudes of the oral vernaculars. Many ideas, and in particular many narratives, seem to enter Sanskrit literature either from parts of the Sanskrit canon that have fallen away or from non-Sanskrit sources (two entirely nonfalsifiable speculations). It's an old joke among linguists that a language is a dialect with an army,e and this is sometimes used to explain the dominance of Sanskrit texts, since as usual, the victors wrote the history, and in ancient India, they usually wrote it in Sanskrit. (The earliest inscriptions were in Prakrit, not Sanskrit, but from about 150 CE, Sanskrit dominated this field too.) Sanskrit is perched on top of the vernacular literatures like a mahout on an elephant, like Krishna riding on the composite women that form the horse on the jacket of this book. SELECTIVITY AND SYNECDOCHE Such a luxurious jungle of cultural phenomena, truly an embarrassment of riches, necessitates a drastic selectivity. I have therefore provided not detailed histories of specific moments but one or two significant episodes to represent the broader historical periods in question.16 The result is not a seamless narrative that covers the waterfront but a pointillist collage, a kaleidoscope, made of small, often discontinuous fragments. Synecdoche--letting one or two moments in history and one or two narratives stand for many--allows us to see alternity in a grain of sand,17 taking a small piece of human history and using it to suggest the full range of enduring human concerns. These small fragments alternate with a few exemplary narratives quoted in considerable detail, where Hindus speak in their own words (in translationf). I have tried to balance my translations of the classic, much-translated texts with citations of more obscure, previously unnoticed texts, using as my framework the usual suspects that scholars have rounded up over and over, the basic curry and rice episodes of Hinduism, but moving away quickly, in each chapter, to a handful of lesser-known episodes, things usually left out of survey books on Hinduism. These are not the great imperial moments but episodes that give us an inkling of what religious life was like for some people, including ordinary people, in India long ago. I have also included a few episodes of interaction (both friendly and hostile) between Hindus and non-Hindus in India, such as Buddhists, Jainas, Sikhs, and Muslims, though without paying direct attention to those other religions in their own right. Beginning with a minimal backbone or infrastructure of basic historical events and concepts that many people would agree upon (data never value free but still valuable), we can then move from this point outward to other points, and from social history to literary texts, to search for narratives of and about alternative people. That selectivity makes this book alternative in another sense, in that it leaves wide open a great deal of space for others to select from in writing their histories, alternative to mine. Someone else would make different choices and write a very different book. This is a history, not the history, of the Hindus. THEMES AND VARIATIONS The central actors and their actions are threads around which the great narratives of Hinduism coalesce like crystals in a supersaturated solution. The actors and actions connect in various ways: Sanskrit texts usually regard women and hunted animals as primary objects of addiction, and the senses that cause addiction are likened to horses; animals often represent both women and the lower classes; the tension between sexuality and renunciation results in an ambivalence toward women as mothers and seductresses; and violence is first addressed largely in the form of violence against animals. Violence and tolerance also interact in attitudes not only to other religions but between the upper and lower castes, between men and women, and between humans and animals. I will highlight in each period those moments when intrareligious (including intercaste) or interreligious interactions took place, marked by either tolerance or violence, the deciding factor between the two options often being historical circumstances. Each chapter deals with several themes, but not every chapter has instances of every theme or treats the same theme in the same way (chapter 12 for instance, is about women more than about goddesses, while chapter 14 is about goddesses more than about women), and indeed I have often noted the activities of women in other contexts, without explicitly highlighting their gender. But wherever the evidence allows, I will organize each chapter around these central themes. (NON)VIOLENCE In the Introduction (chapter 1), I spell out the assumptions behind my attention to history and to the particular actors in this story (women, lower classes and castes, and animals). Here let me just say a few words about the central action: (non)violence. The term "nonviolence" ( ahimsa ) originally applied not to the relationship between humans but to the relationship between humans and animals. Ahimsa means "the absence of the desire to injure or kill," a disinclination to do harm, rather than an active desire to be gentle; it is a double negative, perhaps best translated by the negative "nonviolence," which suggests both mental and physical concern for others. The roots of ahimsa may lie in Vedic ritual, in animal sacrifice, in the argument that the priest does not actually injure the animal but merely "pacifies him"; the primary meaning of ahimsa is thus to do injury without doing injury, a casuist argument from its very inception. In the Rig Veda ' (the earliest Sanskrit text, from c. 1200 BCE), the word ahimsa refers primarily to the prevention of injury or violence to the sacrificer and his offspring, as well as his cattle (10.22.13).18 The problem is exacerbated by the fact that the verb on which ahimsa is based, han , is ambiguous, meaning both "to strike or beat" and "to kill." Ahimsa , therefore, when applied to cows, to take a case at random, might mean refraining either from beating them or killing them--quite a difference. In any case, ahimsa represents not a political doctrine or even a social theory, but the emotion of the horror of killing (or hurting) a living creature, an emotion that we will see attested from the earliest texts.g Arguments about whether or not to kill, sacrifice, and/or eat animals were often at the heart of interreligious violence, sometimes the grounds on which human beings attacked other human beings (usually with words, though occasionally with blows).h Arjuna, the heroic warrior of the Mahabharata , the great ancient Sanskrit poem about a tragic war, excuses the violence of war by saying, "Creatures live on creatures, the stronger on the weaker. The mongoose eats mice, just as the cat eats the mongoose; the dog devours the cat, your majesty, and wild beasts eat the dog. Even ascetics [ tapasas ] cannot stay alive without killing" [12.15.16-24]. The text here justifies human violence by the violence that is rampant in the animal world. Yet the most common sense of ahimsa refers to humans' decision to rise above animal violence. Vegetarianism, both as an ideal and as a social fact in India, challenges Arjuna's belief that animals must inevitably feed on one another and attempts to break the chain of alimentary violence simply by affirming that it is not, in fact, necessary to kill in order to eat. Nonviolence became a cultural ideal for Hindus precisely because it holds out the last hope of a cure, all the more desirable since unattainable, for a civilization that has, like most, always suffered from chronic and terminal violence. Non-violence is an ideal propped up against the cultural reality of violence. Classical Hindu India was violent in ways both shared with all cultures and unique to its particular time and place, in its politics (war being the raison d'être of every king); in its religious practices (animal sacrifice, ascetic self-torture, fire walking, swinging from hooks in the flesh of the back, and so forth); in its criminal law (impaling on stakes and the amputation of limbs being prescribed punishments for relatively minor offenses); in its hells (cunningly and sadistically contrived to make the punishment fit the crime); and, perhaps at the very heart of it all, in its climate, with its unendurable heat and unpredictable monsoons. Hindu sages dreamed of nonviolence as people who live all their lives in the desert dream of oases. It is against this background that we must view the doctrine of nonviolence. The history of Hinduism, as we shall see, abounds both in periods of creative assimilation and interaction and in outbursts of violent intolerance. Sometimes it is possible to see how historical circumstances have tipped the scales in one direction or the other. Sometimes it is not. In their ambivalent attitude to violence, the Hindus are no different from the rest of us, but they are perhaps unique in the intensity of their ongoing debate about it. THE MAN/RABBIT IN THE MOON I have organized several of these tensions into dualities, for dualism is an (if not the) Indian way of thinking, as the folklorist A. K. Ramanujan pointed out, speaking of his father: "I (and my generation) was [sic] troubled by his holding together in one brain both astronomy and astrology. I looked for consistency in him, a consistency he didn't seem to care about, or even to think about. . . . 'Don't you know, [he said,] the brain has two lobes?"19 But some of the most interesting developments take place in the combinations of the two cultural lobes, whether we define them as Brahmin and non-Brahmin, written and oral, or male and female. One medieval Hindu philosophical text defined a great teacher as someone with the ability to grasp both sides of an argument.20 It is, I think, no accident that India is the land that developed the technique of interweaving two colors of silk threads so that the fabric is what they call peacock's neck, blue if you hold it one way, green another (or sometimes pink or yellow or purple), and, if you hold it right, both at once. Another metaphor for this sort of double vision is the dark shape visible on the moon: many Americans and Europeans (for convenience, let us call them Euro-Americans) see the face of a man in the moon (whom some Jewish traditions identify as Cain, cursed to wander), and other cultures see a woman, a moose, a buffalo, a frog, and so forth. But most Hindus (as well as Chinese, Japanese, and Aztecs) see a hare.i (I am calling it a rabbit to avoid the unfortunateEnglish homonym "hare/hair," another bit of double vision, though calling it a rabbit lands me in the middle of a rock group called the Rabbit in the Moon). The man's right eye can be read as the rabbit's ears, his left eye the rabbit's chest, and his mouth the rabbit's tail. (There was a time, in the 1930s, when some people in India saw the image of Gandhi in the moon.21) The Buddhists tell how the moon came to have the mark of a rabbit: THE RABBIT IN THE MOON The future Buddha was once born as a rabbit, who vowed that he would give his own flesh to any beggar who came to him, in order to protect the beggar from having to break the moral law by taking animal life. To test him, Indra, the Hindu king of the gods, took the form of a Brahmin and came to him; the rabbit offered to throw himself into a fire and roast himself so that the Brahmin could eat him. Indra conjured up a magical fire; when the rabbit--who first shook himself three times so that any insects that might be on his body would escape death--threw himself into the fire, it turned icy cold. Indra then revealed his identity as Indra, and so that everyone would know of the rabbit's virtue, he painted the sign of a rabbit on the orb of the moon.22 The convoluted logic of the rabbit's act of self-violence, in his determination to protect anyone else from committing an act of violence against any other animal, is a theme that we will often encounter. The rabbit in the moon is one of so many ideas that Hinduism and Buddhism share. As an approach to the history of Hinduism, seeing both their rabbit and our man in the moon means maintaining an awareness both of what the tradition says (the insider's view) and of what a very different viewpoint helps us to see (the outsider's view). Hindus may approach their scriptures as a part of their piety or as scholars who study Hinduism as they would study any other human phenomenon, or both simultaneously.j There are certainly things that only a Hindu can know about Hinduism, both factual details of local and private practices and texts and the experiential quality of these and other, better-known religious phenomena. This is what inspires interreligious dialogue, an often interesting and productive conversation between individuals who belong to different religions.k But there are also advantages in a more academic approach, such as a religious studies approach, to which the religion of the scholar in question is irrelevant. I would not go so far as some who would insist that a Hindu is not the person to ask about Hinduism, as Harvard professor Roman Jakobson notoriously objected to Nabokov's bid for chairmanship of the Russian literature department: "I do respect very much the elephant, but would you give him the chair of zoology?" Nor would I go to the other extreme, to insist that a Hindu is the only person to ask about Hinduism. For no single Hindu or, for that matter, non-Hindu can know all of the Hinduisms, let alone represent them. So too there are many different ways of being an academic: Some are careful with their research, others sloppy; some make broad generalizations, while others concentrate on small details. Nowadays most non-Hindu scholars of Hinduism strike the familiar religious studies yoga posture of leaning over backward, in their attempt to avoid offense to the people they write about. But any academic approach to Hinduism, viewing the subject through the eyes of writers from Marx and Freud to Foucault and Edward Said, provides a kind of telescope, the viewfinder of context, to supplement the microscope of the insider's view, which cannot supply the same sort of context.23 Always there is bias, and the hope is that the biases of Hindus and non-Hindus will cancel one another out in a well-designed academic study of any aspect of Hinduism. The ancient Persians (according to the Greek historian Herodotus, c. 430 BCE) would debate every important question first drunk, then (on the next day) sober or, as the case may be, first sober, then drunk (1.133). So too, in our scholarly approach, we need to consider the history of Hinduism first from a Hindu viewpoint, then from an academic one. Different sorts of valuable insights may come to individuals both inside and outside the tradition and need not threaten one another. To return to those elephants, you don't have to be an elephant to study zoology, but zoologists do not injure elephants by writing about them. To change the metaphor and apply it more specifically to Hindu texts, a story is a flame that burns no less brightly if strangers light their candles from it. To return to my central metaphor, once you've seen the rabbit (or hare) in the moon, it's hard to see the man anymore, but the double vision is what we should strive for. This means that when we consider, for instance, the burning of living women on the pyres of their dead husbands (which we call suttee, to distinguish it from the woman who commits the act, a woman whom the Hindus call a sati), we must try to see their rabbit, to see the reasons why some Hindus thought (and some continue to think) that it is a good idea for some women to burn themselves to death on their husbands' funeral pyres, while other Hindus strongly disagree. On the other hand, we cannot, and need not, stop seeing our American man (or, perhaps, woman) in the moon: the reasons why many Americans think that suttee is not a good idea at all. The philosopher Ludwig Wittgenstein pointed out that the image of a duck-rabbit (also, actually, a duck-hare) was either a rather smug rabbit or a rather droopy duck24 but could not be both at once.l But this is precisely the goal that a non-Hindu should have in studying Hinduism: to see in the moon both our man and their rabbit. YOU CAN'T MAKE AN OMELET . . . Hindus nowadays are diverse in their attitude to their own diversity, which inspires pride in some, anxiety in others. In particular, it provokes anxiety in those Hindus who are sometimes called Hindu nationalists, or the Hindu right, or right-wing Hindus, or the Hindutva ("Hinduness") faction, or, more approximately, Hindu fundamentalists; they are against Muslims, Christians, and the Wrong Sort of Hindus. Their most powerful political organ is the BJP (Bharatiya Janata Party), with its militant branch, the RSS (Rashtriya Swayamsevak Sangh), but they are also involved in groups such as Hindu Human Rights, Vishwa Hindu Parishad, and the ABVP (Akhil Bharatiya Vidyarthi Parishad). I will generally refer to them as the Hindutva faction or the Hindu right. This book is also alternative to the narrative of Hindu history that they tell. There's a personal story that I should tell about my relationship with this group of Hindus here at the start, in the interest of full disclosure. In the middle of a lecture that I gave in London on November 12, 2003, chaired by William Dalrymple, a man threw an egg at me.25 (He missed his aim, in every way.) A message that a member of the two-hundred-strong audience posted the next day on a mailing list Web site referred to a passage I had cited from Valmiki's Ramayana in which Sita, the wife of Rama, accuses her brother-in-law, Lakshmana, of wanting her for himself. The Web message stated: I was struck by the sexual thrust of her paper on one of our most sacred epics. Who lusted/laid whom, it was not only Ravan who desired Sita but her brother-in-law Lakshman also. Then many other pairings, some I had never heard of, from our other shastras were thrown in to weave a titillating sexual tapestry. What would these clever, "learned" western people be doing for a living if they did not have our shastras and traditions to nitpick and distort?26 After a bit more of this, the writerm added: Her friends and admirers certainly made their applause heard, Muslims among them. In the foyer before the lecture I shook hands and asked a Muslim if he had attended the other lectures in the series and if he was ready for conversion. He said that someone (did he name Vivekananda in the hubbub?) at a similar sort of function had taken off his clothes and asked the audience if they could tell if he was a Hindu or a Muslim. The deeper political agenda of the author of the posting was betrayed by that second set of remarks, particularly by the gratuitous reference to Muslim conversion, and I am grateful to the unnamed Muslim in this vignette for so aptly invoking the wise words of Vivekananda (or, as the case may more probably be, Kabir). My defense now, for this book, remains what it was in the news coverage then, about the lecture (and the egg): The Sanskrit texts [cited in my lecture] were written at a time of glorious sexual openness and insight, and I have often focused on precisely those parts of the texts. . . .The irony is that I have praised these texts and translated them in such a way that many people outside the Hindu tradition--people who would otherwise go on thinking that Hinduism is nothing but a caste system that mistreats Untouchables--have come to learn about it and to admire the beauty, complexity and wisdom of the Hindu texts.27 And, I should have added, the diversity of the Hindu texts. To the accusation that I cited a part of the Hindu textual tradition that one Hindu "had never heard of," my reply is: Yes!, and it's my intention to go on doing just that. The parts of his own tradition that he objected to are embraced by many other Hindus and are, in any case, historically part of the record. One reason why this book is so long is that I wanted to show how very much there is of all that the egg faction would deny. And so I intend to go on celebrating the diversity and pluralism, not to mention the worldly wisdom and sensuality, of the Hindus that I have loved for about fifty years now and still counting. CHAPTER 1 INTRODUCTION: WORKING WITH AVAILABLE LIGHT SEARCHING FOR THE KEY Someone saw Nasrudin searching for something on the ground. "What have you lost, Mulla?" he asked. "My key," said the Mulla. So they both went down on their knees and looked for it. After a time the other man asked: "Where exactly did you drop it?" "In my own house." "Then why are you looking here?" "There is more light here than inside my own house." Idries Shah (1924-96), citing Mulla Nasrudin (thirteenth century CE)1 This Sufi parable could stand as a cautionary tale for anyone searching for the keys (let alone the one key) to the history of the Hindus. It suggests that we may look for our own keys, our own understandings, outside our own houses, our own cultures, beyond the light of the familiar sources. There's a shortage of what photographers call available light to help us find what we are looking for, but in recent years historians have produced studies that provide good translations and intelligent interpretations of texts in Sanskrit and other Indian languages and pointers to both texts and material evidence that others had not noticed before. I have therefore concentrated on those moments that have been illuminated by the many good scholars whose thick descriptions form an archipelago of stepping-stones on which a historian can hope to cross the centuries. This book tells the story of Hinduism chronologically and historically and emphasizes the history of marginalized rather than mainstream Hindus. My aims have been to demonstrate: (1) that Hindus throughout their long history have been enriched by the contributions of women, the lower castes, and other religions; (2) that although there are a number of things that have been characteristic of many Hindus over the ages (the worship of several gods, reincarnation, karma), none has been true of all Hindus, and the shared factors are overwhelmingly outnumbered by the things that are unique to one group or another; (3) that the greatness of Hinduism--its vitality, its earthiness, its vividness--lies precisely in many of those idiosyncratic qualities that some Hindus today are ashamed of and would deny; and (4) that the history of tensions between the various Hinduisms, and between the different sorts of Hindus, undergirds the violence of the contemporary Indian political and religious scene. History and diversity--let me lay them out one by one. HISTORY: AVAILABLE LIGHT The first European scholars of India believed that Hindus believed that everything was timeless, eternal, and unchanging ("There always was a Veda"), and so they didn't generally value or even notice the ways in which Hindus did in fact recognize change. We now call their attitude Orientalism (a term coined by Edward Said in 1978, in a book by that name), which we may define for the moment--we will return to it when we get to the British Raj--as the love-hate relationship that Europeans had with the Orient for both the right and the wrong reasons--it's exotic, it's erotic, it's spiritual, and it never changes.n Like many of the Indian branch of Orientalists, Europeans picked up this assumption of timeless, unified Hinduism from some Hindus and then reinforced it in other Hindus,2 many of whom today regard Hinduism as timeless, though they differ on the actual dating of this timelessness, which (like Hindu scholars of earlier centuries) they tend to put at 10,000 BCE or earlier, while the British generally used to put it much later. The "eternal and unchanging" approach inspired Orientalist philologists to track back to their earliest lair some concepts that do in fact endure for millennia, but without taking into account the important ways in which those concepts changed or the many other aspects of Hinduism that bear little relationship to them. The so-called central ideas of Hinduism--such as karma, dharma, samsara--arise at particular moments in Indian history, for particular reasons, and then continue to be alive, which is to say, to change. They remain central, but what precisely they are and, more important, what the people who believe in them are supposed to do about them differ in each era and, within each era, from gender to gender, caste to caste. And many new ideas arise either to replace or, more often, in Hinduism, to supplement or qualify earlier ideas. Some Hindus always knew this very well. Many Hindu records speak of things that happened suddenly, without precedent ( a-purva , "never before"), right here, right now; they are aware of the existence of local dynasties, of regional gods, of political arrivistes. The Hindu sense of time is intense; the importance of time as an agency of change, the sense that things that happen in the past come to fruition at a particular moment-- now --pervades the great history ( itihasa ) called the Mahabharata. That sense of history is different from ours, as different as Buddhist enlightenment is from the European Enlightenment (what a difference a capital E makes). But in India, as in Europe, human beings compose texts at some moment in history, which we strive with varying degrees of success to discover, and those texts continue to develop and to be transformed through commentary, interpretation, and translation. Hinduism does not lend itself as easily to a strictly chronological account as do some other religions (particularly the so-called Abrahamic religions or religions of the Book, or monotheisms--Judaism, Christianity, Islam), which refer more often to specific historical events. Many central texts of Hinduism cannot be reliably dated even within a century. Since early Buddhism and Hinduism grew up side by side in the same neighborhood, so to speak, historians of Hinduism have often ridden piggyback on historians of Buddhism, a religion that has for the most part kept more precise chronological records; the historians of Buddhism figure out when everything happened, and the historians of Hinduism say, "Our stuff must have happened around then too." Historians of early India have also depended on the kindness of strangers, of foreign visitors to India who left reliably dated (but not always accurately observed) records of their visits. The chronological framework is largely imperialistic--dates of inscriptions, battles, the endowment of great religious institutions--because those are the things that the people who had the clout to keep records thought was most important. And though we no longer think that kings are all that matter in history (siding more with D. D. Kosambi, who urged historians to ask not who was king but who among the people had a plow), kings (more precisely, rajas) do also still matter. They are, however, no longer all that we would like to know about. The crucial moments for cultural history are not necessarily the great imperial moments, as historians used to think they were, the moments when Alexander dipped his toe into India or the Guptas built their empire. For some of the richest and most original cultural developments take place when there isn't an empire, in the cracks between the great dynastic periods. And although the historical records of inscriptions and coins tell us more about kings (the winners) than about the people (the losers), there are other texts that pay attention to the rest of the populace. When we cannot date events precisely, we can often at least arrange things in a rough but ready chronological order, though this leads to a house of cards effect when we are forced to reconsider the date of any text in the series. The periodizations, moreover, may give an often false suggestion of causation.o We cannot assume, as philologists have often done, that the texts line up like elephants, each holding on to the tail of the elephant in front, that everything in the Upanishads was derived from the Brahmanas just because some Upanishads cite some Brahmanas. We must also ask how the new text was at least in part inspired by the circumstances of its own time. Why did the Upanishads develop out of the Brahmanas then ? What about the stuff that isn't in the Brahmanas? "Well (the speculation used to go), maybe they got it from the Greeks; it reminds me of Plato. Or perhaps the Axial Age, sixth century BCE and all that? Or how about this? How about the Indus Valley civilizations? Lots of new ideas must have come from there." Since there is no conclusive evidence for, or against, any of these influences, before we look to Greece we must look to India in the time of the Upanishads to find other sorts of factors that might also have influenced their development--new forms of political organization, taxation, changes in the conditions of everyday life. Even an imported idea takes root only if it also responds to something already present in the importing culture;3 even if the idea of reincarnation did come from Greece to India, or from Mesopotamia to both Greece and India (hypotheses that are unlikely but not impossible), we would have to explain why the Indians took up that idea when they did not take up, for instance, Greek ideas about love between men, and then we must note how different the Upanishadsare from Plato even in their discussion of ideas that they share, such as reincarnation. Moreover, to the mix of philology and history we must add another factor, individuality. The question of originality is always a puzzle, in part because we can never account for individual genius; of course ideas don't arise in a vacuum, nor are they nothing but the sum total of ideas that came before them. Individuals have ideas, and those ideas are often quite different from the ideas of other people living at the same time and place. This is particularly important to keep in mind when we search for the voices of marginalized people, who often achieve as individuals what they cannot achieve as a group. People are not merely the product of a zeitgeist; Shakespeare is not just an Elizabethan writer. In Indian history, individuals have turned the tide of tolerance or violence even against the current of the zeitgeist. The emperors Ashoka and Akbar, for example, initiated highly original programs of religious tolerance, going in the teeth of the practices of their times. Someone with a peculiar, original, individual bent of mind wrote the "There Was Not" ( nasadiya ) hymn of the Rig Veda , and the story of Long-Tongue the Bitch in the Jaiminiya Brahmana , and the story of Raikva under the cart (one of the earliest homeless people noted in world literature) in the Upanishads. And these individual innovators in the ancient period did not merely compose in Sanskrit. They also lurked in the neglected byways of oral traditions, sometimes in the discourses of women and people of the lower classes, as well as in the broader-based Sanskrit traditions that those local traditions feed. For original ideas are rare both among people who have writing and among those that do not. Public individuals too--such as Ashoka, Akbar, and Gandhi, on the one hand, Aurangzeb, Brigadier General Reginald Dyer, and M. S. Golwalkar, on the other, to take just a few at random--brought about profound transformations in Hinduism. The question of flourishing is less puzzling than the question of innovation, and we can often ask how a particular king (or political movement, or climactic change) helped the horse sacrifice (or the worship of a goddess or anything else) to survive and thrive. Often history can explain why some ideas take hold and spread while others do not; ideas take root only when they become important to people at a particular time, when they hitch on to something that those people care about. An understanding of the social context of the Upanishads, reintroducing the world into the text, may go a long way to explain not who first thought of the story of Raikva but why the Brahmins were willing to include his story in their texts despite the ways in which it challenged their social order. MYTH, HISTORY, AND SYMBOLISM In addition to understanding the history of the texts, we need to understand the relationship between records of historical events and the construction of imaginary worlds as well as the symbolism that often joins them. To begin with the symbolism of physical objects, sometimes a linga is just a linga--or, more often, both a linga and a cigar. Numerous Sanskrit texts and ancient sculptures (such as the Gudimallam linga from the third century BCE) define this image unequivocally as an iconic representation of the male sexual organ in erection, in particular as the erect phallus of the god Shiva. So too a verse from the "Garland of Games" of Kshemendra, a Brahmin who lived in Kashmir in the eleventh century, refers to the human counterpart of the Shiva linga: "Having locked up the house on the pretext of venerating the linga, Randy scratches her itch with a linga of skin."4 The first linga in this verse is certainly Shiva's, and there is an implied parallelism, if not identity, between it and the second one, which could be either a leather dildo or its human prototype, attached to a man. And many Hindus have, like Freud, seen lingas in every naturally occurring elongated object, the so-called self-created ( svayambhu ) lingas, including objets trouvés such as stalagmites. The linga in this physical sense is well known throughout India, a signifier that is understood across barriers of caste and language, a linga franca , if you will. The Gudimallam Linga. But other texts treat the linga as an aniconic pillar of light or an abstract symbol of god (the word means simply a "sign," as smoke is the sign of fire), with no sexual reference. To some, the stone lingas "convey an ascetic purity despite their obvious sexual symbolism."5 There is nothing surprising about this range; some Christians see in the cross a vivid reminder of the agony on Calvary, while others see it as a symbol of their God in the abstract or of Christianity as a religion. But some Hindus who see the linga as an abstract symbol therefore object to the interpretations of those who view it anthropomorphically; their Christian counterparts would be people who refuse to acknowledge that the cross ever referred to the passion of Christ. Visitors to the Gudimallam linga in the early twenty-first century noted that while the large linga as a whole remains entirely naked, with all its anatomical detail, the small image of a naked man on the front of the linga was covered with a chaste cloth, wrapped around the whole linga as a kind of total loincloth (or fig leaf) simultaneously covering up the middle of the man in the middle of the linga and the middle of the linga itself. Here is a fine example of a tradition driving with one foot on the brake and the other on the accelerator. We need to be aware of both the literal and symbolic levels simultaneously, as we see both the rabbit and the man in the moon. Similarly, we have to be careful how we use history and myth to understand one another. In this context I would define a myth as a story that a group of people believe for a long time despite massive evidence that it is not actually true; the spirit of myth is the spirit of Oz: Pay no attention to the man behind the curtain. When we read a text that says that a Hindu king impaled eight thousand Jainas, we need to use history to understand myth--that is, we need to understand why such a text was composed and retold many times; that means knowing the reasons for the tensions between Hindus and Jainas at that time (such as the competition for royal patronage). But we cannot use the myth to reconstruct the actual history behind the text; we cannot say that the text is evidence that a Hindu king actually did impale Jainas. To take another example, when the Ramayana speaks of ogres (Rakshasas), it may be simultaneously constructing an imaginary world in which evil forces take forms that can destroy us and using ogres as a metaphor for particular types of human beings. But it does not record an actual event, a moment when people from Ayodhya overcame real people in India (tribals, or Dravidians, or anyone else), nor does the story of the building of the causeway to Lanka mean that Rama and a bunch of monkeys actually built a causeway to (Sri) Lanka. Such myths reveal to us the history of sentiments rather than events, motivations rather than movements. The history of ideas, even if not a source of "hard" history, is still a very precious thing to have. For stories, and the ideas in stories, do influence history in the other direction, into the future. People who heard or read that story about the impaled Jainas may well have acted differently toward Jainas and/or Hindus (better or worse) as a result. More often than not, we do not know precisely what happened in history, but we often know the stories that people tell about it. As a character in a Garrison Keillor novel remarks, "There are no answers, only stories."6 In some ways, the stories are not only all that we have access to but all that people at the time, and later, had access to and hence all that drove the events that followed. Real events and sentiments produce symbols, symbols produce real events and sentiments, and real and symbolic levels may be simultaneously present in a single text. Myth has been called "the smoke of history,"7 and my intention is to balance the smoke of myth with the fire of historical events, as well as to demonstrate how myths too become fires when they do not merely respond to historical events (as smoke arises from fire) but drive them (as fire gives rise to smoke). Ideas are facts too; the belief, whether true or false, that the British were greasing cartridges with animal fat started a revolution in India. For we are what we imagine, as much as what we do. DIVERSITY Is there a unique and distinct phenomenon worth naming that covers the religion(s) of the people from the Veda (c. 1200 BCE) to the Hare Krishnas in American airports and that tells us where Hinduism ends and Buddhism begins? It is useful to distinguish the objection that there is no such thing as Hinduism in the sense of a single unified religion, from the objection that the people we call Hindus lack a category, or word, for Hinduism and identify themselves not as Hindus but as Indians or as Bengali Vaishnavas (worshipers of Vishnu, living in Bengal). That is, we may ask: (1) Is there such a thing as Hinduism?; (2) is that the best thing to call it?; and (3) can we do so even if Hindus didn't/don't? These are related but separate questions. Let's consider the phenomenon and the name one by one. ARE THERE SUCH THINGS AS HINDUS AND HINDUISM? There are several objections to the use of any single term to denote what, for the sake of argument, we will call Hindus and Hinduism.p Hindus did not develop a strong sense of themselves as members of a distinct religion until there were other religions against which they needed to define themselves, like the invisible man in the Hollywood film who could be seen only when he was wearing clothing that was not a part of him. Until as late as the seventeenth century, many Indian rulers used titles that identified them with a divinity or with their preeminence over other rulers or with their personal qualities or with all their subjects, but not merely with the Hindus. Cultures, traditions, and beliefs cut across religious communities in India, and few people defined themselves exclusively through their religious beliefs or practices; their identities were segmented on the basis of locality, language, caste, occupation, and sect.8 Only after the British began to define communities by their religion, and foreigners in India tended to put people of different religions into different ideological boxes,9 did many Indians follow suit, ignoring the diversity of their own thoughts and asking themselves which of the boxes they belonged in.10 Only after the seventeenth century did a ruler use the title Lord of the Hindus (Hindupati).11 Indeed most people in India would still define themselves by allegiances other than their religion.12 The Hindus have not usually viewed themselves as a group, for they are truly a rainbow people, with different colors ( varna s in Sanskrit, the word that also designates "class"), drawing upon not only a wide range of texts, from the many unwritten traditions and vernacular religions of unknown origins to Sanskrit texts that begin well before 1000 BCE and are still being composed, but, more important, upon the many ways in which a single text has been read over the centuries, by people of different castes, genders, and individual needs and desires. And this intertextuality is balanced by an equally rainbow-hued range of practices, which we might call an interpracticality, on the model of intertextuality, practices that refer to other practices. Another objection to regarding Hinduism as a monolithic entity is that it is hard to spell out what "they all" believe or do (even if we exclude from "all" people like Shirley MacLaine). There is no single founder or institution to enforce any single construction of the tradition, to rule on what is or is not a Hindu idea or to draw the line when someone finally goes too far and transgresses the unspoken boundaries of reinterpretation. Ideas about all the major issues--vegetarianism, nonviolence, even caste itself--are subjects of a debate, not a dogma. There is no Hindu canon. The books that Euro-Americans privileged (such as the Bhagavad Gita ) were not always so highly regarded by "all Hindus," certainly not before the Euro-Americans began to praise them. Other books have been far more important to certain groups of Hindus but not to others. One answer to this objection is that like other religions--Christianity, Buddhism, Islam--Hinduism encompasses numerous miscellaneous sects. Religions are messy. But intertextuality (as well as interpracticality) argues for the inclusion of this unruly miscellany under the rubric of Hinduism. The fact that later texts and practices often quote earlier ones, right back to the Rig Veda , allows us to call it a single tradition, even though there are many other Hindu texts and practices that have no connection with any Sanskrit text, let alone the Veda. What literary critics call the anxiety of influence13 works in the other direction in India. The individual artist composing a text or performing a ritual can make innovations, but she demonstrates first her knowledge of the traditions of the past and only then her ability to build upon them and even to reverse them. The assumption is that if she thinks she has an original idea, it means that she has forgotten its source. Moreover, some of the people we now call Hindus did, when they wanted to, for more than two millennia, find ways to describe themselves as a group, in contrast with Buddhists or Muslims (or particular subsects of Buddhists or Muslims). They called themselves the people of the Veda, or the people who revere the Brahmins who are the custodians of the Veda, or the people who have four classes and four stages of life ( varna-ashrama-dharma , in contrast with Buddhists). Or they called themselves the Aryas ("nobles"), in contrast with the Dasyus or Dasas ("aliens" or "slaves") or barbarians ( mlecchas ). The texts called the Brahmanas, in the seventh century BCE, define mleccha s as people of unintelligible speech, as does a dharma text of the period, which adds that they also eat cow flesh,14 implying that the Aryas do not. The lawmaker Manu too, in the early centuries CE, treats mleccha as a linguistic term, contrasted with Arya (which he correctly regards as a linguistic term) rather than with Dasyu (an ethnic term); those outside the four classes ( varna s) are aliens (Dasyus), whether they speak barbarian ( mleccha ) languages or Arya languages (10.45). A commentator on Manu, named Medhatithi, glosses mleccha with the Sanskrit word barbara , cognate with the Greek barbaroi ("barbarian," someone who babbles, " barbarbar "). No one ever comments on the religious beliefs or practices of these people. But religious belief and practice are aspects of Hindu identity that both we and they can and do recognize. Caste, the most important of the allegiances by which the people whom we call Hindus do identify themselves most often, is closely regulated by religion. Some people would define a Hindu through exclusion, as someone who doesn't belong to another religion; 15q officials of the British Raj used the term "Hindu" to characterize all things in India (especially cultural and religious elements and features found in the cultures and religions of India) that were "not Muslim, not Christian, not Jewish, or, hence, not Western."16 Taking the opposite tack, the inclusive tack, the Indian Supreme Court, in the Hindu Marriage Act (1955),17 ruled that any reference to Hindus shall be construed as including "any person who is a Buddhist, Jaina or Sikh by religion," as well as "persons professing the Sikh, Jain or Buddhist religion," a blatant appropriation that most Sikhs, Jains, and Buddhists would resent bitterly. r It also defines a Hindu as someone who is not a Muslim, Christian, Parsi, or Jew, but who is (in addition to a Sikh, Buddhist, or Jaina) one of a rather arbitrary selection of people whose marginality made the court nervous: "any person who is a Hindu by religion in any of its forms or developments, including a Virashaiva, a Lingayat or a follower of the Brahmo, Prarthana or Arya Samaj." Significantly, the definition was needed because different religions have different marriage laws; the horror of miscegenation, always lurking in the Brahmin heart of darkness, was exacerbated by the British legacy within the law code. But in addition to the circularity, mutual contradictions, and blatant chauvinism of the "not a Muslim" definition, such paraphrases list only other religions available in India (they seldom specify "not a Navajo, not a Confucian"); otherwise the word "Hindu" might simply have replaced "gentoo" or "heathen." The political problems that arise from this geographical assumption will resurface below when we consider the word, rather than the concept, "Hinduism." In what seems to me to be something like desperation, a number of people have defined Hinduism as the religion of people who cannot or will not define their religion. This view was only somewhat sharpened by Sarvepalli Radhakrishnan (president of India from 1962 to 1967), who defined Hinduism as the belief "that truth was many-sided and different views contained different aspects of truth which no one could fully express," which would, I think, make all Unitarians Hindus, or by the militant nationalist B. G. Tilak (1856-1920), who added helpfully that "recognition of the fact that the means to salvation are diverse; and realization of the truth that the number of gods to be worshipped is large, that indeed is the distinguishing feature of Hindu religion."18 The Supreme Court of India in 1966, and again in 1995, codified and reconfirmed these two nondefinitions of Hinduism. In 1966 the Indian Supreme Court was called upon to define Hinduism because the Satsangis or followers of Swaminarayan (1780-1830) claimed that their temples did not fall under the jurisdiction of certain legislation affecting Hindu temples. They argued that they were not Hindus, in part because they did not worship any of the traditional Hindu gods; they worshiped Swaminarayan, who had declared that he was the Supreme God. The court ruled against them, citing various European definitions of Hinduism and others, including Radhakrishnan's cited above.19 But the Satsangis had brought their case to the court in order to challenge the 1948 Bombay Harijan Temple Entry Act, which guaranteed Harijans (Pariahs, Untouchables) access to every Hindu temple; if the Satsangis were not Hindus, this law would not force them to open their doors to Harijans. Thus the legal ruling that defined Hinduism by its tolerance and inclusivism was actually inspired by the desire of certain Hindus to exclude other Hindus from their temples. THE ZEN DIAGRAM In answer to several of the objections to the word "Hinduism," some scholars have tried to identify a cluster of qualities each of which is important but not essential to Hinduism; not every Hindu will believe in, or do, all of them, but each Hindu will adhere to some combination of them, as a non-Hindu would not. Scholars differ as to the number and nature of those forms,20 and we have seen the attempts of the Indian Supreme Court to come up with an inoffensive cluster, but perhaps we can be a little more specific. The elements from which the clusters are formed might include some combination of belief in the Vedas (which excludes Buddhism and Jainism), karma (which does not exclude Buddhism and Jainism), dharma (religion, law, and justice), a cosmology centered on Mount Meru, devotion (bhakti) to one or more members of an extensive pantheon, the ritual offering ( puja ) of fruit and flowers to a deity, vegetarianism as an ideal (though only between about 25 and 40 percent of Indians are actually vegetarian21), nonviolence, and blood sacrifice (which may or may not be mutually exclusive). This polythetic approach, which owes much to the concept of family resemblance laid out by the philosopher Wittgenstein,22 could be represented by a Venn diagram, a chart made of intersecting circles. It might be grouped into sectors of different colors, one for beliefs or practices that some Hindus shared with Buddhists and Jainas, another largely confined to Hindu texts in Sanskrit, a third more characteristic of popular worship and practice, and so forth. But since there is no single central quality that all Hindus must have, the emptiness in the center, like the still center of a storm, suggests that the figure might better be named a Zen diagram, which is not, as you might think, a Venn diagram with just one ring or one that has an empty ring in the center but one that has no central ring.23 There is therefore no central something to which the peripheral people were peripheral. One person's center is another's periphery;24 all South Asia was just a periphery, for instance, to those Delhi sultans and Mughal emperors who viewed everything from a Central Asian perspective. We may speak of marginalized people in the sense that they have been dispossessed and exploited, but Hinduism has porous margins and is polycentric. The Brahmins had their center, which we will refer to as the Brahmin imaginary, but there were other centers too, alternative centers. The configuration of the clusters of Hinduism's defining characteristics changes through time, through space, and through each individual.25 It is constantly in motion, because it is made of people, also constantly in motion. Among the many advantages of the cluster approach is the fact that it does not endorse any single authoritative or essentialist view of what Hinduism is; it allows them all. Any single version of this polythetic polytheism (which is also a monotheism, a monism, and a pantheism), including this one, is no better than a strobe photograph of a chameleon, a series of frozen images giving a falsely continuous impression of something that is in fact constantly changing. Like the man who proudly displayed a roomful of archery targets, each with an arrow in the bull's-eye, but was forced to confess that he had shot the arrows first and then had drawn the targets around them, we can decide what aspects of Hinduism we want to talk about and find the cluster of qualities in which that aspect is embodied--and, if we wish, call it Hinduism. Or backing off ever so slightly, we can speak of beliefs and practices that many Hindus share, which is what I intend to do. It is often convenient to speak of a Brahmin-oriented quasi-orthodoxy (or orthopraxy--see below), which we might call the Brahmin imaginary or the idealized system of class and life stage ( varna-ashrama-dharma ). But whatever we call this constructed center, it is, like the empty center in the Zen diagram of Hinduisms, simply an imaginary point around which we orient all the actual Hindus who accept or oppose it; it is what Indian logicians call the straw man ( purva paksha ), against whom one argues. The actual beliefs and practices of Hindus--renunciation, devotion, sacrifice, and so many more--are peripheries that the imaginary Brahmin center cannot hold. HINDUS AND HINDUISM BY ANY OTHER NAMES26 If we can agree that there is something out there worth naming, what shall we call it? The main objections to calling it Hinduism or to calling the people in question Hindus are that those were not always the names that Hindus used for themselves or their religion and that they are geographical names. Let us consider these two objections. Most of the people we call Hindus call themselves something else, like Golkonda Vyaparis,27 or, on the rarer occasions when they do regard themselves as a group, refer to themselves not as Hindus but as people with the sorts of definitions that we have just considered (Aryas, people who revere the Veda, who follow the system of class and stage of life, and so forth). Moreover, "Hindu" is not a native word but comes from a word for the "river" ( sindhu ) that Herodotus (in the fifth century BCE28), the Persians (in the fourth century BCE), and the Arabs (after the eighth century CE29) used to refer to everyone who lived beyond the great river of the northwest of the subcontinent, still known locally as the Sindhu and in Europe as the Indus. James Joyce, in his novel Finnegans Wake , in 1939, punned on the word "Hindoo" (as the British used to spell it), joking that it came from the names of two Irishmen, Hin-nessy and Doo-ley: "This is the hindoo Shimar Shin between the dooley boy and the hinnessy."30 Even Joyce knew that the word was not native to India. It was an outsider's name for the people who inhabited the territory around the Indus River, which the Persians called Hindustan,31 as did the Mughal emperor Babur in his memoirs in the sixteenth century CE: "Most of the people in Hindustan are infidels whom the people of India call Hindu. Most Hindus believe in reincarnation."32 It is noteworthy both that Babur singles out reincarnation for the defining belief of Hinduism (one of the circles in our Zen diagram) and that he does not ascribe this belief to all Hindus (implicitly acknowledging their diversity). "Hindu" has, however, been an insider's word too for centuries, and it is the word that most Hindus do use now to refer to themselves. And it is not uncommon for one culture to take from another a word to designate a concept for which the original culture had a concept but not a word. That the word has a geographical basis is, as we have seen, absolutely true. But it is not just the word but the very concept of Hindus and Hinduism that is geographically rooted in history. The textbook of legal code (dharma) attributed to Manu (first century CE) does not use the word "Hindu" but does offer a geographical definition of the people to whom his dharma applies (a definition that, it is worth noting, uses animals to define humans): From the eastern sea to the western sea [the Indian Ocean and the Bay of Bengal], the area in between the two mountains [the Himalayas and the Vindhyas] is what wise men call the Land of the Aryas. Where the black antelope ranges by nature, that should be known as the country fit for sacrifices; and beyond it is the country of the barbarians. The twice-born [the upper classes and particularly Brahmins] should make every effort to settle in these countries [2.23-24]. Much has happened since the time when one could define India as the land where the (deer and the) antelope play from sea to shining sea (eastern to western). The belief that all Hindus (should) live in India may have been strong once, though more honored in the breach than in the observance. The Hindus are, after all, one of the great merchant civilizations of the world, and the diaspora is very old indeed. Even Manu merely expresses the pious hope that the upper classes "make every effort" to stay within the boundary lines. Granted, many Hindus did suffer loss of caste status when they headed west across the Indus (particularly under the British Raj). Nevertheless, Hindus spread first through Southeast Asia and later through the British Empire, and they now live all over the world; there are approximately one and a half million Hindus in the United States, some 0.5 percent of the population. So it has been said for much of Indian history that ideally, all Hindus should live in India. But the corresponding implication, that everyone in India is (or should be) a Hindu, was never true, not true during the millennia of cultures before either the Indus Valley or the Vedas, not true of most of India even after those early settlements of North India, and certainly never true after the rise of Buddhism in the fifth century CE. Nowadays there are still enough Muslims in India--15 percent of the population, almost as many Muslims as in Pakistan33--to make India one of the most populous Muslim nations in the world, and Muslim input into Indian culture is far more extensive than the mere numbers would imply. Yet Hindu nationalists have used the geographical implications of the word to equate Hinduism with India and therefore to exclude from the right to thrive in India such people as Muslims and Christians; in 1922, V. D. Savarkar coined the term "Hindutva" to express this equation. But not everyone who uses the word "Hinduism" can be assumed to be in their camp, an assumption that would reduce an intellectual problem to a political problem and a move that we need not make. When we use the word, we can, like Humpty Dumpty, pay it extra, in this case to mean not "the people of India" but the intersecting clusters of Hinduisms outlined above. What's in a name? We might take a page from Prince and call it "the religion formerly known as Hinduism" or "Hinduism après la lettre ." Despite the many strikes against the word "Hinduism," Hinduism by any other name would be just as impossible to categorize, and it is still useful to employ some word for it. We cannot insist that Hindus rethink the name they want to use for their tradition (as they have renamed not only streets in cities but whole cities, like Madras/Chennai, Bombay/Mumbai, and Calcutta/Kolkata), no matter how recent or troubled the name may be.34 "Hinduism" is, in any case, the only poker game in town right now;s it is by far the most immediately recognizable word, or even phrase,35 currently used to describe the Zen diagram of, for want of a better word, Hinduism. In any case, whether or not there really is a Hinduism, there certainly are Hindus. SOURCES OF ALTERNATIVE HINDUISMS Different Hindus not only lived different Hinduisms but privileged different aspects of Hinduism, different qualities among the (non)-defining clusters. Scholars too see the Hindu elephant differently depending upon what part of it they grab (to cite the old Indian parable of the blind men: The one with his hands on the tail imagines that the animal is like a rope; on the side, a wall; on the trunk, a snake). Their politics inevitably colors their ideas of what Hinduism is. In addition to including women's as well as men's voices and Other Ranks as well as Brahmins, Hinduism is composed of local as well as pan-Indian traditions, oral as well as written traditions, vernacular as well as Sanskrit traditions, and nontextual as well as textual sources. The first (often marginalized) elements of each of these pairs tend to reinforce one another, as do the second elements, the dominant elements, but there are important distinctions within each of the two groups. For these contrasting pairs did not translate into polarized groups of people; a single person would often have both halves (as well as non-Hindu traditions) in his or her head; a Brahmin would know the folk traditions, just as, in our world, many people study paleography and then go to church and read Genesis. It is not the case that a puritanical Brahmin studied Manu's dharma texts and a libertine merchant read the Kama-sutra (the textbook of the science of erotics); the same man, of either class, might well read dharma with learned men (pandits) by day and the Kama-sutra with his mistress by night. The elite tendencies of written traditions were exacerbated by the climate. The wet heat and the white ants destroyed any written text within a century or two, particularly since vellum was ruled out by the taboo against using animal substances and palm leaf was far more fragile than vellum. So these written texts by definition belonged to the privileged classes; the written texts that survived had to have been copied over and over again by a scribe patronized by someone with money to spare, and the scribe himself was invariably a male of high caste. Yet oral and written traditions interact throughout Indian history, with oral recitations of written texts and written records of texts recited by people who may or may not have been illiterate. This interaction, which we will note throughout the book, is exemplified by the relationship between writing and reciting in two of the defining texts of Hinduism, the Veda and the Mahabharata .36 The Rig Veda was preserved orally, but it was frozen, every syllable preserved for centuries, through a process of rigorous memorization. There are no variant readings of the Rig Veda , no critical editions or textual apparatus. Just the Rig Veda . So much for the fluidity of orally transmitted texts. Correspondingly, the expected fixity of written texts dissolves when we look at the history of the reception and transmission of the Mahabharata , another enormous Sanskrit text, but one that was preserved both orally and in manuscript. In contrast with the Rig Veda , this text changed constantly; it is so extremely fluid that there is no single Mahabharata; there are hundreds of Mahabharatas , hundreds of different manuscripts and innumerable oral versions. So much for the fixity of written texts. The relationship between Sanskrit and the other languages of India (the vernaculars) further complicates this picture. Sanskrit is the model for most North Indian languages (and the source of much of their grammar and some of their vocabulary), as Tamil is for the Dravidian languages of the south (such as Telugu, Kannada, and Malayalam). The Sanskrit/Tamil distinction therefore overlaps with the North/South distinction, but we certainly cannot simply equate Sanskrit with North and Tamil with South. Many South Indian ideas--like devotion ( bhakti ), to take a case at random--entered Sanskrit literature, not just Tamil literature, through South Indian Brahmins who wrote in Sanskrit in South India. Not only did southern ideas go north, and vice versa, and not only did Tamil flow into Sanskrit and Sanskrit into Tamil, but Tamil went north, and Sanskrit south. A similar mutual interpenetration characterizes textual and nontextual sources. The study of Hinduism in the scholarship of Euro-Americans has been overwhelmingly textual; that's one of the characteristics of what we now call Orientalism, the cluster of attitudes that implicated the first European scholars of India in the European colonization of India. The British used texts as a way of disregarding actual Hindu practices and justifying their own imperial project with textual citations. And the Orientalist orientation to texts is the orientation toward Brahmins (and Sanskrit, and writing). More recently, scholars have begun to pay more attention to ritual, archaeology, art history, epigraphy, the records of foreign visitors, and, in the modern period, ethnography, revealing new aspects of a lived religion that is very differently represented in texts. Coins, for instance, tell a story, for money talks in that sense too. The two sets of sources, textual and nontextual, reveal bits of history to us in different ways, like the lame man riding on the shoulders of the blind man. When it comes to history, you can't trust anyone: The texts lie in one way, while images and archaeology mislead us in other ways. On the one hand, the gods did not fly around in big palaces, as the texts insist that they did, and we cannot know if women really did speak up as Gargi does in the Upanishads, or Draupadi in the Mahabharata . On the other hand, the Indus seal we all once interpreted as an ithyphallic Shiva Pashupati is probably just someone sitting cross-legged as South Asians are inclined to do, with a bulging loincloth knot; well, back to the drawing board. Nontextual sources can provide textualists with an occasional shot in the arm, alerting them to what to look for-- in texts --once they get the idea that they might be there. Texts can do a great deal, with a little help from their nontextual friends. Texts are still useful in a number of ways. First of all, some of those old Brahmin males knew a hell of a lot of great stories. Second, not all texts were written by Brahmins. Woven into the Brahmin texts, as well as standing alongside them, is another great strand of narratives by that extraordinarily prolific writer Anonymous, who was usually not a Brahmin and who should be credited with a great deal of the ancient literature of South Asia. (He--or, just as likely, she--often wrote under the nom de plume of the heavily mythologized authors whom we will soon encounter, people named Vyasa or Valmiki or simply Suta, "the Charioteer Bard.") Third, even those texts that were written by Brahmins were not written (entirely) by Brahmins, nor were all the Brahmins highly literate or elitist; the texts were constantly infused with the contributions of the lower classes and women. Fourth, texts are events too: The Upanishads are part of history as well as imagination. And fifth, texts are also a major source of information about material culture: If we cannot always find the archaeological remains of a plow, we might at least find a text that mentions a plow, just as when we cannot find texts actually written by women, we can at least find references to women, and sympathetic views of their lives, in texts written by men. All these factors greatly expand the caste of characters in ancient Sanskrit texts. I myself am by both temperament and training inclined to texts. I am neither an archaeologist nor an art historian; I am a Sanskritist, indeed a recovering Orientalist, of a generation that framed its study of Sanskrit with Latin and Greek rather than Urdu or Tamil. I've never dug anything up out of the ground or established the date of a sculpture. I've labored all my adult life in the paddy fields of Sanskrit, and since I know ancient India best, I've lingered in the past in this book longer than an anthropologist might have done, and even when dealing with the present, I have focused on elements that resonate with the past, so that the book is driven from the past, back-wheel-powered.37 I have also, for most of the same reasons, inclined more toward written, more precisely ancient Sanskrit traditions than oral and vernacular and contemporary ones. But this book is, when all is said and done, and despite my acknowledgment of the baleful influence of text-oriented scholarship, a defense of the richness of texts as the source of information about the sorts of things that some people nowadays assume you need nontextual sources for: women, the lower classes, the way people actually lived. WOMEN Women are sometimes said to have been excluded from the ancient Indian texts and therefore to have left no trace, history having been written by the winners, the men. But in fact women made significant contributions to the texts, both as the (usually unacknowledged) sources of many ancient as well as contemporary narratives and as the inspiration for many more. Some Hindu women did read and write, forging the crucial links between vernacular languages and Sanskrit. Women were forbidden to study the most ancient sacred text, the Veda, but the wives, whose presence was required at Vedic rituals, both heard and spoke Vedic verses,38 and they may well have had wider access to other Sanskrit texts. Later, in the second or third century CE, the Kama-sutra tells us not only that women had such access but even that they sometimes commissioned such texts to be written (1.3). Women in Sanskrit plays generally speak only dialects (Prakrits), while men speak only Sanskrit, but since the men and women converse together, generally without translators, the women must understand the men's Sanskrit, and the men the women's dialects. Moreover, some women in plays both speak and write Sanskrit, and some men speak in dialects,t trampling on what is left of the convention. It is a basic principle of one school of Indian logic that something can be prohibited only if its occurrence is possible.39 The fact that the texts keep shouting that women should not read Vedic texts suggests that women were quite capable of doing so and probably did so whenever the pandits weren't looking. Women as a group have always been oppressed in India (as elsewhere), but individual women have always succeeded in making their mark despite the obstacles. We can also look for the implied author40 and identify in men's texts the sorts of things that a woman might have said.41 Within the Sanskrit texts, women express views of matters as basic as karma in terms quite different from those of men, and these views become even more prominent when women compose their own tales.42 There is an "ironic" presence of women in the Mahabharata , "perhaps beyond earshot, but definitely heard," and their physical absence may lend a kind of invisible luster to the highly visible women in that text.43 The Kama-sutra , in its instructions to the would-be adulterer, presents a strong protofeminist view of what women have to put up with at the hands of inadequate husbands (5.1). Such texts at least keep women in the picture, however biased a picture that may be, until they do finally get to speak as named authors, much later. Of course, excavating women's voices in male texts must always be qualified by the realization that there may be ventriloquism, misreporting of women, and false consciousness; the male author of the Kama-sutra may have sympathy for women but not true empathy; his interest in their thoughts is exploitative, though no less accurate for all that. But ventriloquism is a two-way street; there is also a ventriloquism of women's voices in male minds. For even when a male Brahmin hand actually held the pen, as was usually the case no matter what the subject matter was, women's ideas may have gotten into his head. We can never know for sure when we are hearing the voices of women in men's texts, but we can often ferret out (to use an animal metaphor) tracks, what the Hindus call "perfumes" ( vasanas ), that women have left in the literature. A hermeneutic of suspicion, questioning the expressed motivations of the author, is therefore required, but it is still worth reading between the lines, even making the texts talk about things they don't want to talk about. Moreover, texts are not our only source of knowledge of this period; women also left marks, perfumes, that we can find in art and archaeology. We can try to resurrect the women actors in Hindu history through a combination of references to them, both unsympathetic (to see what they had to put up with from some men) and sympathetic (to show that other men did treat them humanely), and moments when we can hear women's own voices getting into the texts and, more rarely, discover actual female authorship. FROM DOG COOKERS TO DALITS Brahmins may have had a monopoly on liturgical Sanskrit for the performance of certain public rites, but even then the sacrificer uttered some of the ritual words and performed the domestic rites. And the sacrificer need not have been a Brahmin, a member of the highest class; the other two twice-born social classes--warriors/rulers (Kshatriyas) and, below them, merchants, farmers and herders (Vaishyas)--were also initiated and therefore could be sacrificers. The three upper classes were called twice born because of the second birth through the ritual of initiation (the ancient Indian equivalent of becoming born again), in which a man was born (again) as a fully developed member of the community. u The lowest of the four classes, the servants (Shudras), were excluded from these and many other aspects of religious life, but the exclusion of Shudras doesn't automatically make something "Brahminical." There have been countless terms coined to designate the lowest castes, the dispossessed or underprivileged or marginalized groups, including the tribal peoples. These are the people that Sanskrit texts named by specific castes (Chandala,Chamara, Pulkasa, etc.) or called Low and Excluded (Apasadas) or Born Last (or Worst, Antyajas) or Dog Cookers (Shva-Pakasv), because caste Hindus thought that these people ate dogs, who in turn ate anything and everything, and in Hinduism, you are what you eat. Much later the British called them Untouchables, the Criminal Castes, the Scheduled (they pronounced it SHED-YULED) Castes, Pariahs (a Tamil word that has found its way into English), the Depressed Classes, and Outcastes. Gandhi called them Harijans ("the People of God"). The members of these castes (beginning in the 1930s and 1940s and continuing now) called themselves Dalits (using the Marathi/Hindi word for "oppressed" or "broken" to translate the British "Depressed"). B. R. Ambedkar (in the 1950s), himself a Dalit, tried, with partial success, to convert some of them to Buddhism. Postcolonial scholars call them (and other low castes) Subalterns. Another important group of oppressed peoples is constituted by the Adivasis ("original inhabitants"), the so-called tribal peoples of India, on the margins both geographically and ideologically, sometimes constituting a low caste (such as the Nishadas), sometimes remaining outside the caste system altogether. It is important to distinguish among Dalits and Adivasis and Shudras, all of whom have very different relationships with upper-caste Hindus, though many Sanskrit texts confuse them. So too, the Backward Castes, a sneering name that the British once gave to the excluded castes in general, are now regarded as castes separate from, and occasionally in conflict with, certain other Dalit castes; the Glossary of Human Rights Watch defines Backward Castes as "those whose ritual rank and occupational status are above 'untouchables' but who themselves remain socially and economically depressed. Also referred to as Other Backward Classes (OBCs) or Shudras," though in actual practice the OBCs often distinguish themselves from both Dalits and Shudras. All these groups are alike only in being treated very badly by the upper castes; precisely how they are treated, and what they do about it, differs greatly from group to group. All in all, when we refer to all the disenfranchised castes below the three upper classes known as twice born, it is convenient to designate them by the catchall term of Pariahs (a Tamil word--for the caste that beat leather-topped drums--that finds its way into English) up until the twentieth century and then to call them Dalits. But whatever we choose to call them, the excluded castes play an important role in the history of the Hindus. Thanks to the Subaltern studies movement, there is a lot more available light for Dalits, particularly in the modern period (from the time of the British); this book aims to contribute to that movement by including more information about Dalits in the ancient period. There have been protests against the mistreatment of the lower castes from a very early age in India, though such protests generally took the form of renouncing caste society and forming an alternative society in which caste was ignored; no actual reforms took place until the nineteenth century and then with only limited success. Much of what I have said about women also applies to Pariahs, and vice versa; Brahmin ventriloquism functions similarly to male ventriloquism, and the lower castes, like women, leave their "perfumes" in upper-caste literature. The positive attitudes to Pariahs in such texts represent a beginning, a prelude to reform; they change the world, even if only by imagining a world in which people treated women or Pariahs better. The Brahmins did produce a great literature, after all, but they did not compose it in a vacuum. They did not have complete authority or control the minds of everyone in India. They drew upon, on the one hand, the people who ran the country, political actors (generally Brahmins and kings, but also merchants) and, on the other hand, the nonliterate classes. Because of the presence of oral and folk traditions in Sanskrit texts, as well as non-Hindu traditions such as Buddhism and Jainism, Dog Cookers do speak,w not always in voices recorded on a page but in signs that we can read if we try. For the ancient period, it's often harder to find out who had a plow than to find out, from inscriptions, who endowed what temple. Some people today argue that the Brahmins erased much of the low-caste contribution to Indian culture--erased even their presence in it at all. Certainly the Sanskrit texts stated that the lower castes would pollute any sacred text that they spoke or read, as a bag made of the skin of a dog pollutes milk put into it.44 But this probably applied only to a limited corpus of texts, Vedic texts, rather than Sanskrit in general. The fact that a sage is punished for teaching the Vedas to the horse-headed gods called the Ashvins, who associate with the class of farmers and herdmen, should alert us to the possibility that teaching the Vedas to the wrong sorts of people might also be a rule honored at least sometimes in the breach as well as in the observance. And we can, as with women's voices, ferret out voices of many castes in the ancient texts, and once we have access to the oral and folk traditions, we can begin to write the alternative narrative with more confidence. ANIMALS: HORSES, DOGS, AND COWS AS POWER, POLLUTION, AND PURITY Animals--primarily not only dogs, horses, and cows, but also monkeys, snakes, elephants, tigers, lions, cats, and herons--play important roles in the Hindu religious imaginary, both as actual living creatures and as the key to important shifts in attitudes to different social classes. Yogic postures ( asanas ) and sexual positions, as well as theological schools, are named after animals. Gods become incarnate as animals and have animal vehicles in the human world. The process works in opposite directions at once. On the one hand, the observation of the local fauna provides images with which people may think of their gods; whether or not people get the gods that they deserve, they tend to get the gods (and demons) that their animals deserve--gods inspired by the perceived qualities of the animals. On the other hand, the ideas that people have about the nature of the gods, and of the world, and of themselves will lead them to project onto animals certain anthropomorphic features that may seem entirely erroneous to someone from another culture observing the same animal. And knowing what animals, real live animals, actually appeared in the material culture at a particular time and place helps us place aspects of that culture geographically and sometimes chronologically. Thus animals appear both as objects, in texts about the control of violence against living creatures (killing, eating), and as subjects, in texts where they symbolize people of different classes. Clearly the two--the animals of the terrain and the animals of the mind--are intimately connected, and both are essential to our understanding of Hinduism. If the motto of Watergate was "Follow the money," the motto of the history of Hinduism could well be "Follow the monkey." Or, more often, "Follow the horse." Three animals--horses, dogs, and cows--are particularly charismatic players in the drama of Hinduism. The mythological texts use them to symbolize power, pollution, and purity, respectively, and link them to three classes of classical Hindu society: Kshatriyas or rulers, particularly foreign rulers (horses), the lower classes (dogs), and Brahmins (cows).x Horses and dogs function in our narrative as marginalized groups on both ends of the social spectrum (foreigners and Pariahs), y while cows are the focus of the ongoing debate about vegetarianism. These three animals pair up first with one and then with another in a complex symbolic dance. Horses and cows provide mirror images of each other's genders. The cow (f.) is the defining gender for the bovine species and the symbol of the good human female (maternal, docile); the negative contrast is provided not by bulls and steers, who have a rather ambivalent status (Shiva's bull, Nandi, is generally docile and benign), but by male buffalo, who have taken over this spot in the paradigm and symbolize evil in both myth and ritual, as well as being often associated with Pariahs. By contrast, the stallion (m.) is the defining gender for equines, mares generally being the symbol of the evil female (oversexed, violent, and Fatally Attractive).45 Cows and horses can also represent religious contrasts; the Hindu cow and the Muslim horse often appear together on chromolithographs. Because horses are not native to India and do not thrive there, they must constantly be imported, generally from western and Central Asia.46 The reasons for this still prevail: climate and pasture.47 The violent contrast between the hot season and the monsoon makes the soil ricochet between swampy in one season and hard, parched, and cracked in another. The grazing season lasts only from September to May, and even then the grasses are spare and not good for fodder. Moreover, since the best soil is mostly reserved for the cultivation of grains and vegetables to feed a large and largely vegetarian population, there is relatively little room for horses even in those places where more nutritious fodder grasses are found (such as the eastern extensions of the arid zone in the north and northwest of India, particularly in Rajasthan, where horses have in fact been bred successfully for centuries). There is therefore no extensive pasturage, and horses are stabled as soon as they are weaned, unable to exercise or develop strength and fitness. Here, as elsewhere, wherever conditions are poor for breeding, "a regular injection of suitable horses is vital for the upkeep and improvement of the breed," to keep it from degenerating.48 It is therefore part of the very structure of history that India has always had to import horses,49 which became prized animals, used only in elite royal or military circles. And so the horse is always the foreigner in India, the invader and conqueror, and the history of the horse in India is the history of those who came to India and took power. There is still a Hindi saying that might be translated, "Stay away from the fore of an officer and the aft of a horse" or "Don't get in front of an officer or behind a horse." It dates from a time when petty officials, especially police, revenue collectors, and record keepers, were mounted and everyone else was not. These horsemen were high-handed (" . . . on your high horse") and cruel, people whom it was as wise to avoid as it was to keep out of the range of those back hooves. The horse stands as the symbol of the power and aristocracy of the Kshatriyas, the royal warrior class; the horse is the key to major disputes, from the wager about the color of a horse's tail made by the mother of snakes and the mother of birds in the Mahabharata (1.17-23) (an early instance of gambling on horses), to heated arguments by contemporary historians about the seemingly trivial question of whether Aryan horses galloped or ambled into the Indus Valley or the Punjab, more than three thousand years ago. Horses continued to be idealized in religion and art, in stark contrast with the broken-down nags that one more often actually encounters in the streets of Indian cities. Under the influence of the Arab and Turkish preference for mares over stallions, the Hindu bias in favor of stallions and against mares gave way to an entire Hindu epic literature that idealized not stallions but mares. Finally, horses are also metaphors for the senses that must be harnessed, yoked through some sort of spiritual and physical discipline such as yoga (a word whose basic meaning is "to yoke," as in "to yoke horses to a chariot"). The cow's purity is fiercely protected by Brahmins and is at the heart of often hotly contested attitudes to food in the history of Hinduism. In the Vedic period, people ate cattle (usually bulls or bullocks or castrated bulls), as they ate all other male sacrificial animals (with the exception of the horse, which was not eaten). But though the Vedic people also occasionally ate cows (the female of the species), cows soon became, for most Hindus, cultural symbols of non-violence and generosity, through the natural metaphor of milking; unlike most animals (but like other lactating female mammals--mares, female camels, buffalo cows, nanny goats), cows can feed you without dying. Cows therefore are, from the earliest texts to the present moment, the object of heated debates about vegetarianism. At the other end of the animal spectrum are dogs. For caste-minded Hindus, dogs (not significantly gendered like horses and cows) are as unclean as pigs are to Orthodox Jews and Muslims, therefore symbols of the oppressed lowest castes, of the people at the very bottom of human society, indeed outside it, the sorts of people that we call underdogs and Sanskrit authors sometimes called dog cookers.z Dogs are also associated with the Adivasis, the so-called tribal peoples of India. Animal keepers, leatherworkers, people who touch human waste are often referred to as pigs and dogs.aa The ancient Indian textbook of political science, the Artha-shastra , even suspects dogs of espionage; the author warns the king not to discuss secrets when dogs or mynah birds are present (1.15.4). The mynah bird of course could talk, but the dog? Would he reveal secrets by wagging his tail? (He might recognize a secret agent and blow his cover by not barking in the night.)ab But texts covertly critical of the caste system reverse the symbolism and speak of breaking the rules for dogs, treating them as if they were not impure. The dog who doesn't bark is about a silence that speaks; it is a good metaphor for the Pariah voice, the dog's voice, that we can sometimes hear only when it does not speak. The shifting tracks of these animals form a trail of continuity within the diversity of alternative Hinduisms. PLURALISM AND TOLERANCE50 The proliferation of polythetic polytheisms may pose problems for the definition of Hinduism, but they are its glory as a cultural phenomenon. Pluralism and diversity are deeply ingrained in polylithic Hinduism, the Ellis Island of religions; the lines between different beliefs and practices are permeable membranes. Not only can we see the Hindu traditions as divided among themselves on many central issues throughout history, but we can see what the arguments were on each point, often far more than two views on major questions. The texts wrestle with competing truths, rather than offer pat answers. One sort of pluralism that has always prevailed in India is what I would call eclectic pluralism, or internal or individual pluralism, a kind of cognitive dissonance, 51 in which one person holds a toolbox of different beliefs more or less simultaneously, drawing upon one on one occasion, another on another.52 Multiple narratives coexist peacefully, sometimes in one open mind and sometimes in a group of people whose minds may be, individually, relatively closed.ac A pivotal example of such individual pluralism can be found in the law text of Manu , which argues, within a single chapter, passionately against and then firmly for the eating of meat (5.26-56). Or as E. M. Forster once put it, "Every Indian hole has at least two exits."53 When it comes to ritual too, an individual Hindu may worship several different gods on different occasions, to satisfy different needs, on different festival days, in fellowship with different members of the family (a bride will often bring into the home a religion different from that of her husband's), or as a matter of choice as new gods are encountered. The compound structure of Sanskrit and the fact that most words have several meanings (it used to be said that every Sanskrit word means itself, its opposite, a name of god, and a position in sexual intercoursead) enabled poets to construct long poems that told two entirely different stories at the same time and shorter poems that had multiple meanings, depending on how you divided up the compounds and chose among the various connotations of each word. This poetry, rich in metaphors, could itself stand as a metaphor for the Hindu approach to multivalence. Eclectic pluralism between religions is more cautious, but it has allowed many an individual, such as a Hindu who worships at a Sufi shrine, to embrace one tradition in such a way as to make possible, if not full engagement with other faiths, at least full appreciation and even admiration of their wisdom and power.54 The sorts of permeable membranes that marked one sort of Hinduism from another also marked Hinduism from other religions; the dialogues were both intrareligious and interreligious. Hinduism interacted creatively with, first, Buddhism and Jainism, then Judaism and Christianity, then Islam and Sikhism, as well as with tribal religions and other imports (such as Zoroastrianism). The interactions were sometimes conscious and sometimes unconscious, sometimes appreciative borrowings and sometimes violent but productive antagonisms (as we will see, for instance, in the sometimes positive and sometimes negative attitudes toward the story of Vishnu's incarnation as the Buddha). In Rohinton Mistry's novel Such a Long Journey , there is a wall in Bombay/Mumbai that the neighborhood men persist in peeing and defecating against, creating a stench and a nuisance of flies. The protagonist of the novel hires an artist to paint images of all the religions of the world on the wall, a multireligious polytheistic dialogue of gods and mosques (respecting the Muslim rule against representing figures), so that no one, of any religion, will foul the wall.55 (It works, for a while, until the city knocks down the wall to widen the road.) This seems to me to be a fine metaphor for both the hopes and the frailty of interreligious dialogue in India. Hindus, Jainas, and Buddhists all told their own versions of some of the same stories. Hindus and Buddhists (and others) in the early period shared ideas so freely that it is impossible to say whether some of the central tenets of each tradition came from one or the other; often two Hindu versions of the same story, composed in different centuries, have less in common than do a Hindu and a Buddhist version of the same story. The stories change to fit different historical contexts, and often one can date one telling later than another (the language is different, it mentions a later king, and so forth), but where it comes from, and when, nobody knows. Many of the same religious images too were used by Buddhists and Jainas as well as Hindus.56 To this day Hindus and Christians, or Hindus and Muslims, often worship the same figure under different names; Satya Pir, for instance, is a Muslim holy man ( pir ) who had come, by the eighteenth century, to be identified with a form of the Hindu god Vishnu (Satya Narayana).57 The great Indian poet and saint Kabir, who self-consciously rejected both Hinduism and Islam, nevertheless built his own religious world out of what he would have regarded as the ruins of Hinduism and Islam, as did many of the great Sufi saints, at whose shrines many Hindus continue to worship. Building a shrine on the site where a shrine of another tradition used to stand is thus both a metaphor of appreciation and an act of appropriation in India, unhindered by any anxiety of influence. This open-mindedness was supported by the tendency of Hindus to be more orthoprax than orthodox. That is, most Hindus have not cared about straight opinions (ortho-doxy) nearly so much as they care about straight behavior (ortho-praxy). Although there is a very wide variety of codes of action, each community has a pretty clear sense of what should and shouldn't be done, and some things were Simply Not Done. People have been killed in India because they did or did not sacrifice animals, or had sex with the wrong women, or disregarded the Vedas, or even made use of the wrong sacred texts, but no one was impaled (the Hindu equivalent of burning at the stake) for saying that god was like this rather than like that. Each sect acknowledged the existence of gods other than their god(s), suitable for others to worship, though they might not care to worship them themselves. Hindus might therefore best be called polydox.58 Yet renouncers, certain monists, and some of the bhakti sects tended to be more orthodox than orthoprax, and those movements that challenge Brahmins, the Veda, and the values of class and caste are generally called heterodox, or even heretical ( pashanda or pakhanda ).59 The Hindu concept of heresy was thus applied to some people within the Hindu fold, though more often to Buddhists and Jainas. HYBRIDITY AND MULTIPLICITY The "solitarist" approach to human identity sees human beings as members of exactly one group, in contrast with the multiple view that sees individuals as belonging to several different groups at once. Visualize our friend the intra-Hinduism Venn/Zen diagram, now in an interreligious guise. The multiple view is both more appropriate and more helpful for people caught up in the confrontation of communities, such as Hindu and Muslim in India.60 People sometimes make a further distinction between multiplicity and hybridity. Multiplicity implies a combination in which the contributing elements are theoretically unchanged even when mixed. Hinduism in this sense of multiplicity is perceived to have elements that a Muslim would recognize as Muslim, a Buddhist would recognize as Buddhist, and so forth. An example of religious multiplicity in an individual: On Sunday you go to church and attend a basic Catholic mass much as you would experience it in many (though certainly not all) churches in another city or another country, mutatis mutandis, and on Tuesday you go to a Hindu temple and assist at a ceremony of killing a goat, much as you would experience it in many (though certainly not all) Hindu temples in another city or another country, mutatis mutandis. Hybridity, by contrast, implies fusion. An example of religious hybridity in an individual: On Monday you attend the same sort of basic Catholic mass, but in place of the Eucharist you kill a goat, or you attend the same sort of basic Hindu puja but the goddess to whom you pray is Mary, the mother of Jesus, with all her epithets and physical characteristics. The Oxford English Dictionary defines "hybrid" as "anything derived from heterogeneous sources, or composed of different or incongruous elements," which, when applied to a community, leaves conveniently open the question of whether those elements remain unchanged. The OED definition applies to individuals rather than communities: "the offspring of two animals or plants of different species, or (less strictly) varieties." Both hybridity and multiplicity can be applied to both communities and individuals. The trouble with both multiplicity and hybridity (as well as syncretism) lies in the assumption that the combinatory elements are separate essences that exist in a pure form before the mix takes place and that the combination either does (for hybrids) or does not (for multiplicities) change them in some way. But there are seldom any pure categories in any human situation, certainly not by the moment when history first catches up with them. Long before 2000 BCE, the Indus Valley Civilization was already a mix of cultures, as was Vedic culture at that time, and eventually the two mixes mixed together, and mixed with other mixes. Hybridity defies binary oppositions and understands reality as a fluid rather than a series of solid, separate boxes. Hyphens can be read as multiple or as hybrid. The hybrid, hyphenated word "Anglo-Indian" confusingly denotes two opposite sorts of people: The OED defines "Anglo-Indian" as "a person of British birth resident, or once resident, in India," or "a Eurasian of India,"which is to say either a privileged Englishman ruling "Inja" or a hybrid, an underprivileged person whom the British regarded as the lowest of all castes, a mixed breed. Hybridity, traditionally, has had the additional disadvantage of carrying a largely negative attitude to the mixing of categories, an attitude that we now regard as reactionary. Thus the hybrid has been despised as a hodgepodge, a mix in which both (or all) of the contributing elements are modified; the OED adds, gratuitously, to its definition the phrase, "a half-breed, cross-breed, or mongrel," the racist overtones of its definition echoing the Hindu fear of the mixture of social classes ( varna-samkara ). But nowadays both postcolonial and postmodern thinkers prefer hybrids, define "hybrid" more positively, and indeed argue that we all are hybrids,61 all always mixed and mixing.62 The Parsis ("Persians"--i.e., Zoroastrians) in several communities in India tell a positive story about social hybridity. They say that when the Parsis landed in India, the local Hindu raja sent them a full glass of milk, suggesting that the town was full. The Parsi leader added sugar and returned the glass, indicating that his people could mix among the Arabs and Hindus like sugar in milk, sweetening it but not overrunning it.63 The metaphor of sugar in milkae suggests the extreme ideal of communal integration, in which individuals change the community by melting into it, flavoring it as a whole with their qualities (Zoroastrianism, or sweetness). The Parsis did not in fact dissolve into Islam and Hinduism; they remained Parsis and indeed were often caught in the crossfire during the riots that followed the Partition of India and Pakistan in 1947. This seems to me the more accurate way to view such cultural mixes: as a suspension of discrete particles rather than a melting pot. Despite their shortcomings, the concepts of hybridity and multiplicity are useful, if used with care. The phenomenon is basically the same in either case; it's just a matter of points of view, and it doesn't really matter whether you call it multiple or hybrid (or even syncretic). What does matter is how you evaluate the fused mix. Whatever word you use for it, I think it applies to Hinduism, and I think it is a Good Thing.af I once (in a very different context) characterized Hindu mythology as a pendulum of extremes that are never resolved and that are also constantly in motion: "By refusing to modify its component elements in order to force them into a synthesis, Indian mythology celebrates the idea that the universe is boundlessly various, that everything occurs simultaneously, that all possibilities may exist without excluding each other . . . [that] untrammeled variety and contradiction are ethically and metaphysically necessary."64 Keeping both extreme swings of the pendulum in mind simultaneously means realizing that an individual actor in the drama of the history of the Hindus may regard herself as a fused hybrid of Muslim and Hindu or as a fused multiple, fully Muslim in some ways and fully Hindu in others, as many Indians have been, throughout history. In either case, there would be no perfectly pure category of Muslim or Hindu anywhere along the line of fusion. Such a person might worship at a Hindu temple on certain days and at a Sufi shrine on others, might read both the Upanishads and the Qu'ran for spiritual guidance, and would celebrate both the great Muslim holy days and the great Hindu festivals. I would therefore argue for the recognition of the simultaneous presence of a number of pairs of opposites, throughout the history of the Hindus, the both/and view of community. The historiographic pendulum of reconciliation, never resting at the swing either to one side or the other, forces us to acknowledge the existence, perhaps even the authenticity, of the two extremes of various ideas, and also their falseness, as well as the fact that there is no pure moment at either end of the swing, and leave it at that. With apologies to Buddhism, there is no middle way here. Or rather, the middle way has got to take its place alongside all the other, more extreme ways in the Zen diagram. CHAPTER 2 TIME AND SPACE IN INDIA 50 Million to 50,000 BCE THE BIRTH OF INDIA The Ganges, though flowing from the foot of Vishnu and through Siva's hair, is not an ancient stream. Geology, looking further than religion, knows of a time when neither the river nor the Himalayas that nourished it existed, and an ocean flowed over the holy places of Hindustan. The mountains rose, their debris silted up the ocean, the gods took their seats on them and contrived the river, and the India we call immemorial came into being. --E. M. Forster, A Passage to India (1924)1 TIME ORIGINS: OUT OF AFRICA To begin at the beginning: Once upon a time, about 50 million years ago,ag a triangular plate of land, moving fast (for a continent), broke off from Madagascar (a large island lying off the southeastern coast of Africa) and, "adrift on the earth's mantle,"2 sailed across the Indian Ocean and smashed into the belly of Central Asia with such force that it squeezed the earth five miles up into the skies to form the Himalayan range and fused with Central Asia to become the Indian subcontinent.3 Or so the people who study plate tectonics nowadays tell us, and who am I to challenge them?ah Not just land but people came to India from Africa, much later; the winds that bring the monsoon rains to India each year also brought the first humans to peninsular India by sea from East Africa in around 50,000 BCE.4 And so from the very start India was a place made up of land and people from somewhere else. So much for "immemorial." Even the ancient "Aryans" probably came, ultimately,5 from Africa. India itself is an import, or if you prefer, Africa outsourced India. This prehistoric episode will serve us simultaneously as a metaphor for the way that Hinduism through the ages constantly absorbed immigrant people and ideas and as the first historical instance of such an actual immigration. (It can also be read as an unconscious satire on histories that insist on tracing everything back to ultimate origins, as can the E. M. Forster passage cited at the start.) The narratives that Hindus have constructed about that stage and those actors, narratives about space and time, form the main substance of this chapter. The flood myth, in particular, is about both space (continents sinking) and time (periodic floods marking the aeons). Often unexpressed, always assumed, these narratives are the structures on which all other narratives about history are built. We will then briefly explore the natural features of India--rivers and mountains--that serve not only as the stage on which the drama of history unfolds but as several of the main actors in that drama, for Ganga (the Ganges) and Himalaya appear in the narratives as the wife and father-in-law of the god Shiva, respectively. GONDWANALAND AND LEMURIA Francis Bacon was the first to notice, from maps of Africa and the New World first available in 1620, that the coastlines of western Africa and eastern South America matched rather neatly. Scientists in the nineteenth century hypothesized that Antarctica, Australia, Africa, Madagascar, South America, Arabia, and India all were connected in the form of a single vast supercontinent to which an Austrian geologist gave the name of Gondwana or Gondwanaland. (He named it after the region of central India called Gondwana--which means "the forest of the Gonds," the Gonds being tribal people of central India--comprising portions of the present states of Madhya Pradesh, Maharashtra, and Andhra Pradesh,6 the latter a region famous for its enormous rocks, the oldest on the planet.) Scientists then suggested that what were later called continental shiftsai began about 167 million years ago (in the mid- to late Jurassic period), causing the eastern part of the continent of Gondwana to separate from Africa and, after a while (about 120 million years ago, in the early Cretaceous period), to move northward. It broke into two pieces. One piece was Madagascar, and the other was the microcontinent that eventually erupted into the Deccan plateau and crashed into Central Asia.aj Australian Indologists joke that the Deccan is really part of Australia.7 The Gondwanaland story takes us to the farthest limit, the reduction to the absurd, of the many searches for origins that have plagued the historiography of India from the beginning (there, I'm doing it myself, searching for the origins of the myth of origins). Both nineteenth-century scholarship and twenty-first-century politics have taken a preternatural interest in origins. Nineteenth-century scholars who searched for the ur-text (the "original text," as German scholarship defined it), the ur-ruins, the ur-language carried political stings in their tales: "We got there first," "It's ours" (ignoring the history of all the intervening centuries that followed and other legitimate claims). They viewed the moment of origins as if there were a kind of magic Rosetta stone, with the past on one side and the present on the other, enabling them to do a simple one-to-one translation from the past into the present ever after. But even if they could know the ur-past, and they could not (both because logically there is no ultimate beginning for any chain of events and because the data for the earliest periods are at best incomplete and at worst entirely inaccessible), it would hardly provide a charter for the present. Other scientists in the colonial period agreed about the ancient supercontinent but imagined its disintegration as taking place in the opposite way, not when land (proto-India) broke off from land (Australia/Africa) and moved through water (the Indian Ocean) to join up with other land (Central Asia), but rather when water (the Indian Ocean) moved in over land (a stationary supercontinent like Gondwanaland) that was henceforth lost under the waves, like Atlantis. According to this story, water eventually submerged (under what is now called the Indian Ocean) the land that had extended from the present Australia through Madagascar to the present South India. In 1864 a geologist named that supercontinent Lemuria, because he used the theory to account for the fact that living lemurs were found, in the nineteenth century, only in Madagascar and the surrounding islands, and fossil lemurs were found from Pakistan to Malaya, but no lemurs, living or dead, were found in Africa or the Middle East (areas that would never have been connected to Lemuria as Madagascar and Pakistan presumably once were).8 Animals, as usual, here define human boundaries, and the myths about those boundaries, as usual, proliferated. In 1876, Ernst Haeckel, a German biologist of a Darwinian persuasion, published his History of Creation , claiming that the lost continent of Lemuria was the cradle of humankind; in 1885 a British historian argued that the Dravidian languages had been brought to India when the ancestors of the Dravidians came from Lemuria;9 in 1886 a teenager in California "channeled" voices that suggested that the survivors of Lemuria were living in tunnels under Mount Shasta in California;10 and in 1888, in The Secret Doctrine, Madame Blavatsky claimed that certain Indian holy men had shown her a secret book about Lemuria. This myth nurtured among the colonial powers was then taken up, in the 1890s, by Tamil speakers on the southern tip of South India, who began to regard Lemuria as a lost ancestral home from which they all were exiled in India or to argue that the extant India, or Tamil Nadu, or just the southernmost tip of India, Kanya Kumari (Cape Comorin), was all that was left of Lemuria; or that when Lemuria sank, Tamilians dispersed to found the civilizations of Mesopotamia, Egypt, China, the Americas, Europe, and, in particular, the Indus Valley.11 Nowadays some Tamil separatists want to reverse the process, to detach Tamil Nadu from the rest of India, not, presumably, physically, to float back over the Indian Ocean like Gondwanaland in reverse, but politically, in order to recapture the glory of their lost Lemurian past. The passage from E. M. Forster cited above, "The Birth of India," begins with the Himalayas rising up out of the ocean, Gondwanaland fashion, but then, as it continues, it slips into the other variant, the story of the submersion of Lemuria, and regards that submersion as preceding the Gondwanaland episode, pushing back the origins even farther: "But India is really far older. In the days of the prehistoric ocean the southern part of the peninsula already existed, and the high places of Dravidia have been land since land began, and have seen on the one side the sinking of a continent that joined them to Africa, and on the other the upheaval of the Himalayas from a sea."12 Forster concludes his passage with a third aspect of the myth, its periodicity or cyclicity, its prediction that the flooding of South India will happen again and again: "As Himalayan India rose, this India, the primal, has been depressed, and is slowly re-entering the curve of the earth. It may be that in aeons to come an ocean will flow here too, and cover the sun-born rocks with slime." So that's how it all began. Or maybe it didn't. Forster of course has the carte blanche of a novelist, but even the plate tectonics people may be building sand castles, for the plate tectonics theory is after all a speculation, albeit a scientific speculation based on good evidence. Whether or not a subcontinent once shook the dust of Africa off its heels and fused onto Asia, the story of Gondwanaland reminds us that even after the Vedic people had strutted around the Ganges Valley for a few centuries, all they had done was add a bit more to what was already a very rich mix. The multiplicity characteristic of Hinduism results in part from a kind of fusion--a little bit of Ravi Shankar in the night, a Beatle or two--that has been going on for millions of years, as has globalization of a different sort from that which the word generally denotes. The pieces of the great mosaic of Hinduism were put in place, one by one, by the many peoples who bequeathed to India something of themselves, planting a little piece of England, or Samarkand, or Africa, in the Punjab or the Deccan. APRÈS MOI, LE DÉLUGE Hinduism is so deeply embedded in the land of its birth that we cannot begin to understand its history without understanding something of its geography and in particular the history of representations of its geography. The central trope for both time and space in India is the great flood. The myth of the flood is told and retold in a number of variants, some of which argue for the loss of a great ancient civilization or a fabulous shrine. The telling of a myth of such a flood, building upon a basic story well known throughout India, allows a number of different places to imagine a glorious lost past of which they can still be proud today. The myth of the flooding of Lemuria, or Dravidia, builds on the traditions of other floods. There is archaeological evidence for the flooding of the Indus Valley cities by the Indus River (c. 2000 BCE), as well as for that of the city of Hastinapura by the Ganges, in about 800 BCE.13 There is also textual evidence (in the Mahabharata ) for the flooding of the city of Dvaraka, the city of Krishna, at the westernmost tip of Gujarat, by the Western Ocean (that is, the Arabian Sea),14 in around 950 BCE. (Sources differ; some say 3102 or 1400 BCE.)15 The appendix to the Mahabharata also tells of the emergence of Dvaraka from the ocean in the first place. When Krishna chose Dvaraka as the site for his city, he asked the ocean to withdraw from the shore for twelve leagues to give space for the city; the ocean complied.16 Since the sea had yielded the land, against nature (like the Netherlands), it would be only fair for it to reclaim it again in the end. Later texts tell of a different sort of bargain: Krishna in a dream told a king to build a temple to him as Jagannatha in Puri, but the ocean kept sweeping the temple away. The great saint Kabir stopped the ocean, which took the form of a Brahmin and asked Kabir for permission to destroy the temple; Kabir refused but let him destroy the temple at Dvaraka in Gujarat. And so he did.17 Even so, some texts insist that the temple of Krishna in Dvaraka was not flooded; the sea was not able to cover it, "even to the present day,"18 and the temple, able to wash away all evils, remains there,19 just as in the periodic flooding of the universe of doomsday, something always survives. (The physical location of the shrine of Dvaraka, at the very westernmost shore of India, where the sun dies every evening, may have inspired the idea that the town was the sacred gate to the world of the dead.20) In direct contradiction of the Mahabharata 's statement that the entire city was destroyed, these later texts insist that it is still there. Dvaraka is said to exist today in Gujarat, and archaeologists and divers have published reports on what they claim to be its remains.21 We may also see here the patterns of the myths of both Lemuria (the ocean submerging Dvaraka) and Gondwana (Dvaraka emerging from the ocean to join onto Gujarat). Other myths too follow in the wake of this one, such as the story that the ocean (called sagara ) was first formed when the sixty thousand sons of a king named Sagara dug into the earth to find the lost sacrificial horse of their father, who was performing a horse sacrifice.22 (Some versions say that Indra, the king of the gods, stole the horse.)23 A sage burned the princes to ashes, and years later Bhagiratha, the great-grandson of Sagara, persuaded the Ganges, which existed at that time only in the form of the Milky Way in heaven, to descend to earth in order to flow over the ashes of his grandfathers and thus purify them so that they could enter heaven; he also persuaded the god Shiva to let the heavenly Ganges River land first on his head and meander through his matted hair before flowing down to the earth, in order to prevent her from shattering the earth by a direct fall out of the sky.24 According to another text, when Sagara performed the horse sacrifice, the oceans began to overflow and cover all the land with water. The gods asked the great ascetic Parashurama to intercede; he appealed to Varuna (the Vedic god of the waters), who threw the sacrificial vessel far away, causing the waters to recede and thereby creating the western kingdom of Shurparaka.25 (In a different subtext of this version, when Parashurama was banished from the earth and needed land to live on, Varuna told him to throw his ax as far into the ocean as he could; the water receded up to Gokarna, the place where his ax finally fell, thus creating the land of Kerala.26) There are other legends of submerged cities or submerged lands or land-masses. ak27 The cities where the first two assemblies that created Tamil literature were held are said to have been destroyed by the sea.28 In the seventeenth century, people claimed to be able still to see the tops of a submerged city, temples and all, off the coast of Calicut.29 For centuries there were said to be seven pagodas submerged off the coast of Mamallipuram, near Madras, and on December 26, 2004, when the great tsunami struck, as the waves first receded about five hundred meters into the sea, Frontline (an Indian news Web site) reported that tourists saw a row of rocks on the north side of the Shore Temple and that behind the Shore Temple in the east, architectural remains of a temple were revealed. "When the waves subsided, these were submerged in the sea again."30 Archaeologists denied that there could be any submerged temples there.31 Our knowledge of the long history of the imaginative myth of the submerged Hindu temple inclines us to side with the more skeptical archaeologists. Behind all these traditions may lie the story of another great flood, first recorded in the Shatapatha Brahmana (c. 800 BCE), around the same time as one of the proposed dates for the Mahabharata flood, a myth that has also been linked to Noah's Ark in Genesis32 as well as to stories of the flood that submerged the Sumerian city of Shuruppak and is described in the Gilgamesh epic. Indeed flood myths are found in most of the mythologies of the world: Africa, the Near East, Australia, South Seas, Scandinavia, the Americas, China, Greece. They are widespread because floods are widespread, especially along the great rivers that nurture early civilizations (and even more widespread in the lands watered by the monsoons). There are significant variants: Some cultures give one reason for the flood, some other reasons, some none; sometimes one person survives, sometimes several, sometimes many (seldom none--or who could tell the story?--though the creator sometimes starts from scratch again); some survive in boats, some by other means.33 In the oldest extant Indian variant, in the Brahmanas, Manu, the first human being, the Indian Adam, finds a tiny fish who asks him to save him from the big fish who will otherwise eat him. This is an early expression of concern about animals being eaten, in this case by other animals; "fish eat fish," what we call "dog eat dog," is the Indian term for anarchy. The fish promises, in return for Manu's help, to save Manu from a great flood that is to come. Manu protects the fish until he is so big that he is "beyond destruction" and then builds a ship (the fish tells him how to do it); the fish pulls the ship to a mountain, and when the floodwaters subside, Manu keeps following them down. The text ends: "The flood swept away all other creatures, and Manu alone remained here."34 The theme of "helpful animals" who requite human kindness (think of Androcles and the lion) teaches two morals: A good deed is rewarded, and be kind to (perhaps do not eat?) animals. Centuries later a new element is introduced into the story of the flood, one so important and complex that we must pause for a moment to consider it: the idea that time is both linear and cyclical. The four Ages of time, or Yugas, are a series named after the four throws of the dice. Confusingly, the number of the Age increases as the numbers of the dice, the quality of life, and the length of the Age decrease: The first Age, the Krita Yuga ("Winning Age") or the Satya Yuga ("Age of Truth"), what the Greeks called the Golden Age (for the four Ages of time, or Yugas, formed a quartet in ancient Greece too), is the winning throw of four, a time of happiness, when humans are virtuous and live for a long time. The second Age, the Treta Yuga ("Age of the Trey"), is the throw of three; things are not quite so perfect. In the third Age, the Dvapara Yuga ("Age of the Deuce"), the throw of two, things fall apart. And the Kali Age is the dice throw of snake eyes, the present Age, the Iron Age, the Losing Age, the time when people are no damn good and die young, and barbarians invade India, the time when all bets are off. This fourth Age was always, from the start, entirely different from the first three in one essential respect: Unlike the other Ages, it is now, it is real. The four Ages are also often analogized to the four legs of dharma visualized as a cow who stands on four legs in the Winning Age, then becomes three-legged, two-legged, and totters on one leg in the Losing Age. But time in India is not only linear, as in Greece (for the ages steadily decline), but cyclical, unlike Greece (for the end circles back to the beginning again). The cosmos is reborn over and over again, as each successive Kali Age ends in a doomsday fire and a flood that destroys the cosmos but is then transformed into the primeval flood out of which the cosmos is re-created, undergoing a sea change in a new cosmogony.al The idea of circular cosmic time is in part the result of Indian ideas about reincarnation, the circularity of the individual soul. The ending precedes the beginning, but the end and the beginning were always there from the start, before the beginning and after the end, to paraphrase T. S. Eliot.am In later retellings of the story of the flood, therefore--to return at last to our story--the fish saves Manu from the doomsday flood that comes at the end of the Kali Age, the final dissolution ( pralaya ): THE FISH AND THE FLOOD Manu won from the god Brahma, the creator, the promise that he would be able to protect all creatures, moving and still, when the dissolution took place. One day, he found a little fish and saved it until it grew so big that it terrified him, whereupon he realized that it must be Vishnu. The fish said, "Bravo! You have recognized me. Soon the whole earth will be flooded. The gods have made this boat for you to save the great living souls; bring all the living creatures into the boat, and you will survive the dissolution and be king at the beginning of the Winning Age. At the end of the Kali Age, the mare who lives at the bottom of the ocean will open her mouth and a poisonous fire will burst out of her, coming up out of hell; it will burn the whole universe, gods and constellations and all. And then the seven clouds of doomsday will flood the earth until everything is a single ocean. You alone will survive, together with the sun and moon, several gods, and the great religious texts and sciences." And so it happened, and the fish came and saved Manu.35 In this text, Manu saves not himself alone but all creatures, and this time the gods, instead of Manu, build the boat. This variant also gives us a much more detailed, and hence more reassuring, image of what is to follow the flood; a new world is born out of the old one. These stories suggest that floods are both inevitable and survivable; this is what happens to the world, yet the world goes on. More significantly, the myth is now part of the great story of the cycle of time, involving fire as well as water, so that the flood now appears more as a solution than as a problem: It puts out the mare fire that is always on the verge of destroying us. For a mare roams at the bottom of the ocean; the flames that shoot out of her mouth are simultaneously bridled by and bridling the waters of the ocean,36 like uranium undergoing constant fission, controlled by lithium rods. In several of the myths of her origin, the fire is said to result from the combined fires of sexual desire and the fire of the ascetic repression of sexual desire,37 or from the fury of the god Shiva when he is excluded from the sacrifice. 38 The submarine mare is (to continue the nuclear metaphor) like a deadly atomic U-boat cruising the deep, dark waters of the unconscious. It should not go unnoticed that the mare is a female, the symbol of all that threatens male control over the internal fires of restrained passions that are always in danger of breaking out in disastrous ways. This delicate balance, this hair-trigger suspension, is disturbed at the end of the Kali Age, the moment of doomsday, when the mare gallops out of the ocean and sets the world on fire, and the newly unchecked ocean leaves its bed and floods the ashes of the universe, which then lie dormant until the next period of creation.39 And then, like the ashes of Sagara's sons, the ashes of the entire universe are revived as it is reborn. A remnant or seed, a small group of good people, is saved by a fish (usually identified as one of the several incarnations of the great god Vishnu), who pulls a boat to a mountain, where they survive to repeople the universe.40 (The mountain, the Hindu equivalent of Ararat, is identified with numerous sites throughout India.an) The myth expresses the barely controlled tendency of the universe to autodestruct (and perhaps a kind of prescientific theory of global warming: When it gets hotter, the ice caps will melt, and there will be a flood41). Recent attempts to excavate both Lemuria and the submerged city of Dvaraka,42 correlated with recent oceanographic work carried out in 1998-99 around the Kerguelen plateau in the southernmost reaches of the Indian Ocean, have rekindled speculations about a "lost continent" in the Indian Ocean. What surprised the excavators most was not the enormous plateau that they found in the middle of the Indian Ocean but signs that "near the end of the plateau growth, there is strong evidence of highly explosive eruptions."43 The volcanic activity of the submarine mare, perhaps? SPACE MAPS So much for origins. Whatever fused with India had to make its peace with what was already there, its unique climate, fauna, and, eventually, culture. The land and its people transformed all who came to them; they did not simply passively receive the British, or the Mughals before them, let alone the people of the Veda, or that migratory bit of Africa. Geographical and Mythological Map of India. Ancient Indian cosmology imagined a flat earth consisting of seven concentric continents, the central one surrounded by the salty ocean and each of the other roughly circular continents surrounded by oceans of other liquids: treacle (molasses), wine, ghee (clarified butter), milk, curds, and freshwater. (This prompted one nineteenth-century Englishman's notorious tirade against "geography made up of seas of treacle and seas of butter."44) In the center of the central mainland (the "Plum-tree Continent" or Jambu-Dvipa) stands the cosmic mountain Meru, from which four subcontinents radiate out to the east, west, north, and south, like the petals of a lotus; the southernmost petal of this mainland is Bharata-varsha, the ancient Sanskrit name for India. If you bisect the lotus horizontally, you see India as a kite-shaped landmass with mountains in the north and (salt) oceans on all other sides, much as it appears on any Rand McNally map.ao The watery world under the earth, which the cobra people (Nagas) inhabit, is also there, in the water table that we encounter every time we sink a well anywhere in the world. Cosmography and cartography overlie each other, as do the rabbit and the man in the moon, myth and history. It has been rightly remarked that texts are just maps, and map is not territory;45 but when the maps are big enough, they become territories of their own. There is a shared core underlying both maps and territories, from which myths and political narratives spread out in different directions. There was a flood, and now there is a politically useful myth about it; there is an arrangement of water and land, and there is a politically inspired diagram of it (for different countries draw the borders of Kashmir, to take a case at random, in very different ways). The map of the Plum-tree Continent is to a Rand McNally map as the flood myth is to the geological record. The natural layout of water and land serves as the basis of the myth of a flood and the diagram of the cosmos, which in turn support the construction of a politically useful chart of time and space. TERRITORY: MOUNTAINS, RIVERS, MONSOON Many people have imagined the Himalayan Mountains as posing an impregnable barrier, but this image of Inaccessible India is simply a part of the Unchanging India package. India functioned, throughout history, less like Shangri-la than like Heathrow or O'Hare. The Himalayas are indeed high, and no one ever strolled casually across them, but they did not keep people out of India. Alexander the Great managed to get into India over the Himalayan Mountains (probably through the Khyber Pass), horses, mules, camels, and all, and many others followed. Not without reason was the Hindu imagination haunted by the specter of invasion, expressed in the persistent myth of the degenerate Kali Age, a nightmare of barbarian penetration. The physical boundaries of India were as porous as those between its internal belief systems. Silk came from China across the Central Asian silk route (the word for "silk" in Sanskrit is china ), and just about everyone in the ancient world--beginning with traders from Mesopotamia, Crete, Rome, and Arabia--washed up sooner or later on some coast of India. So too the Vindhya Mountains form the barrier between North and South India, but the stories about the Vindhyas tell how that barrier was breached, not how it kept people apart: When the Vindhyas began to grow so tall that even the sun had to go around them (just as it circumambulates Mount Meru), the great sage Agastya asked them to bow their heads before him so that he could cross from North to South India (bringing Sanskrit and the Vedas to the Dravidian lands) and to remain that way until he returned; the Vindhyas agreed to this, and since Agastya never returned from South India, where he established the Tamil language, the Vindhyas remain conveniently low. In South Asia, history flows with the rivers. Three great river systems divide the northern subcontinent: first the Indus ("the River") in Pakistan, with its five tributaries, the rivers of the Punjab (Jhelum, Chenab, Ravi, Beas, and Sutlej) that give the Punjab its name ("Five Waters"); then the Doab or "Two Rivers," the Ganges (Ganga, "Going to Earth" [from heaven, where she is the Milky Way]) and the Yamuna ("Twin Sister," now Jumna) in North India; and then the Brahmaputra ("Son of Brahma") in Bangladesh. All three rivers originate in a single region of southwestern Tibet, their sources so close that they may once have belonged to a single icy lake that was shattered when the piece of Africa that crashed into Central Asia drove off the waters in diverse directions. 46 The Indus flows eighteen hundred miles before it empties into the Arabian sea. The Narmada ("Jester"), the great river that, like the Vindhya Mountains, divides the north from the south, has inspired an extensive mythology that balances that of the Ganges in the north. What is the relationship between climate and culture in India? Is there some causal link between, on the one hand, "the ambivalent natural environment, where lush harvests coexist with barren soil, drought with flood, feast with famine," and, on the other, the fact that Hindu logicians were the first to posit the coexistence of the elements of contradiction?47 Other countries have "ambivalent natural environments" too; farmers the world over are at the mercy of the elements. But the violence and uncertainty of the monsoon create an ever-present psychological factor that may well be related to Hindu ideas about the capriciousness and violence of fate and the gods. CONCLUSION: CON-FUSION What does the geology of the formation of India tell us about the formation of Hinduism? The answer is suggested by a story that A. K. Ramanujan retold, from Tamil sources: THE BRAHMIN HEAD AND THE PARIAH BODY A sage's wife, Mariamma, was sentenced by her husband to death. At the moment of execution she embraced a Pariah woman, Ellamma, for her sympathy. In the fray both the Pariah woman and the Brahmin lost their heads. Later the husband relented, granted them pardon, and restored their heads by his spiritual powers. But the heads were transposed by mistake. To Mariamma (with a Brahmin head and Pariah body) goats and cocks but not buffalo were sacrificed; to Ellamma (Pariah head and Brahmin body) buffalo instead of goats and cocks.48 This text is itself an example of what it tells about: It mixes together the story of Mariamma from two different Indian geographical and linguistic traditions, North Indian Sanskrit literature, where she is called Renuka, and South Indian Tamil oral folktales about the origins of two South Indian goddesses.49 This sort of juxtaposition, in various forms, is widespread in both myth and history, beginning with the piece of Africa stuck onto Central Asia, like a head upon a body, and continuing through all the ideas of women and low castes that get into the heads of Brahmin males. It can stand as a metaphor for all the fusions that make up the rich mix of Hinduism. The mixing together of various human streams is so basic to the history of Hinduism that the Brahmins could not stop trying, and failing, to prevent it, even as their fear of the powers of the senses to invade the rational control center made them try, also in vain, to control addiction through asceticism. Their ultimate terror was the "confusion" of classes, the miscegenation brought on by the Kali Age. They visualized the mixing of classes as a form of impurity, which should not surprise anyone who has read the British anthropologist Mary Douglas's explanation of the ways that throughout the world, "category errors"--things that do not fall entirely into one class or another--are characterized as dirt and as danger.50 Brahmins regarded the woman with the Brahmin head and Pariah body--and her twin and partner, with the Pariah head on the Brahmin body--as monstrosities, a double hodgepodge. But from the standpoint of a non-Brahmin, or a scholar of Hinduism, this rich hybrid or multiple mix is precisely what makes Hinduism the cultural masterpiece that it is. Such a conflation is not a monstrosity, nor is it a mistake--or if it is, it is a felix culpa . The transpositions result in two goddesses (read: many Hinduisms), each of whom is far more interesting than the straightforward realignment would have been. Whatever its disparate sources, the resulting creature has an integrity that we must respect, rather like that of my favorite mythical beast, created by Woody Allen, the Great Roe, who had the head of a lion and the body of a lion, but not the same lion .51 The question to ask is not where the disparate elements originated but why they were put together and why kept together. The political implications of regarding Hinduism as either a hodgepodge or, on the other hand, culturally homogeneous or even monolithic are equally distorting; it is always more useful, if a bit trickier, to acknowledge simultaneously the variety of the sources and the power of the integrations. A Hinduism with a Pariah body and a Brahmin head--or, if you prefer, a Pariah head and a Brahmin body--was re-created again and again throughout India history, and these multiple integrities are what this book is about. CHAPTER 3 CIVILIZATION IN THE INDUS VALLEY 50,000 to 1500 BCE CHRONOLOGY (ALL DATES BCE) c. 50,000 Stone Age cultures arise c. 30,000 Bhimbetka cave paintings are made c. 6500 Agriculture begins c. 3000 Pastoral nomad societies emerge c. 2500 Urban societies emerge along the Indus River c. 2200-2000 Harappa is at its height c. 2000-1500 Indus civilization declines "Pashupati" Seal (Seal 420). In place of an opening epigram, we begin with an image, whose meaning is much disputed, for one of the many challenges of interpreting the Indus Valley Civilization (IVC) lies in deciphering pictures for which we do not know the words. The second challenge is trying to decide what, if anything, of the IVC survives in later Hinduism. For the IVC is older than the oldest extant Hindu texts, the Vedas, and its material remains include many images that may be the earliest-known examples of important Hindu icons that only (re)surface much later. EARLY HISTORY: BHIMBETKA CAVE PAINTINGS Much of what we now call Hinduism may have had roots in cultures that thrived in South Asia long before the creation of textual evidence that we can decipher with any confidence. Remarkable cave paintings have been preserved from Mesolithic sites dating from c. 30,000 BCE in Bhimbetka, near present-day Bhopal, in the Vindhya Mountains in the province of Madhya Pradesh.1 They represent a number of animals that have been identified as deer, boars, elephants, leopards, tigers, panthers, rhinoceroses, antelope, fish, frogs, lizards, squirrels, and birds. One painting seems to depict a man walking a dog on a leash. The animals represented probably existed there (it would be hard for someone who had never seen an elephant to draw a picture of an elephant), though there may be false positives (an artist could have copied someone else's picture of an elephant, and the existence of images of a creature half bull and half man certainly does not prove that such tauranthropoi actually existed). On the other hand, animals that are not represented may well also have existed there (the Bhimbetkanese may have had snakes even though they did not make any paintings of snakes); the missing animals may simply have failed to capture the artist's imagination. False negatives in this realm are even more likely than false positives. Several of the animals in the paintings have horns, like gazelles, and one painting shows people dancing with what may be a unicorn with a close-clipped mane.2 This possible unicorn continues to tease art historians when it reappears in the IVC. THE INDUS VALLEY MATERIAL CULTURE There were other early settlements in India, notably the culture of Baluchistan, in the westernmost part of what is now Pakistan, dating to before 6000 BCE. But from about 2300 BCE the first urbanization took place, as great cities arose in the valley of the Indus River, 150 miles south of Baluchistan, also in Pakistan. The material remains of this culture, which we call the Indus Valley Civilization or the Harappan Civilization (named after Harappa, one of the two great cities on the Indus, the other being Mohenjo-Daro), present a tantalizing treasure chest of often enigmatic images that hover just beyond our reach, taunting us with what might well be the keys to the roots of Hinduism. The Indus Valley plain, much like the valleys of the Nile and the Tigris-Euphrates, cradles of Neolithic civilizations, is a semiarid, river-watered region; the "semi" means that on the one hand, the relatively sparse vegetation, not so rich as that of the effluvial plain of the Ganges, for instance, required no iron tools to clear and settle while, on the other hand, the silt from the river floodings provided sufficient natural fertilizer to create the surplus that makes civilization possible.3 The river was also a channel of trade. Here's another origin story. In 1856 an English general named Alexander Cunningham, later director general of the archaeological survey of northern India, visited Harappa, where an English engineer named William Brunton was gathering bricks (including what he recognized as bricks from the IVC) as ballast for a railway he was building between Multan and Lahore. Cunningham took note of the site but did nothing about it, and the trains still run on that route, on the main line from Peshawar, on top of a hundred miles of third-millennium BCE bricks. Only after 1917, when an Indian archaeologist found an ancient knife at a place named, significantly, Mound of the Dead (Mohenjo-Daro), and excavations carried out there revealed artifacts identical with those that had been at Harappa, did this civilization begin to be appreciated. Among the treasures that they found were carved stones, flat, rectangular sections of soapstone about the size of a postage stamp, which were used as stamps or seals, as well as sealings (impressions) of such stamps. The civilization of the Indus Valley extends over more than a thousand sites, stretching over 750,000 square miles, where as many as forty thousand people once lived.4 Four hundred miles separate the two biggest cities, from Harappa on the Ravi tributary in the north (one of the five rivers of the Punjab) down to Mohenjo-Daro (in the Larkana Valley in Sindh) and on down to the port of Lothal in the delta on the sea. Yet the Indus cities were stunningly uniform and remarkably stable over this wide range, changing little over a millennium, until they begin to crumble near the end. They had trade contacts with Crete, Sumer, and other Mesopotamian cultures, perhaps even Egypt.5 There are Harappan sites in Oman (on the Arabian Peninsula), and Indus seals show up in Mesopotamia. There was direct contact with Iran, particularly just before the end, a period from which archaeologists have found a very late Indus seal with Indus motifs on one side and Iranian on the other, together with many seals reflecting Central Asian influences.6 Some Indus images bear a striking resemblance to images from Elam, a part of ancient Iran that was closely linked to adjacent Mesopotamian urban societies.7 Trade with Central Asia continued in the Indus area even after the demise of the Indus Valley Civilization. In a sense, the Hindu diaspora began now, well before 2000 BCE. Archaeological evidence suggests that the use of cubical dice began in South Asia and indeed in the IVC.8 Sir John Hubert Marshall, the director general of the Indian archaeological survey from 1902 to 1931, found many cubical terra-cotta dice, with one to six dots, at Mohenjo-Daro,9 and a number of other dice have been identified since then from Harappa and elsewhere, including several of stone (agate, limestone, faience, etc.).10 This is a fact of great significance in light of the importance of gambling in later Indian civilization, from 1200 BCE. They had gold, copper, and lead, and they imported bronze, silver, and tin (as well as lapis lazuli and soapstone), but they had no iron; their weapons were made of copper and bronze. There was a huge wheat and barley storage system, and there were household and public drainage works superior to those in parts of the world today, including much of India. Most of the buildings are constructed out of bricks (both sun baked and kiln fired) of remarkably consistent size throughout the extended culture; equally unvarying stone cubes were used to measure weights. The roads too did not just evolve out of deer tracks but were carefully laid out all in the same proportion (streets twice as wide as lanes, avenues twice as wide as streets) and arranged on a grid (north-south or east-west), like the "pink city" of Jaipur in Rajasthan that Maharajah Jai Singh designed and built in the eighteenth century CE. All this uniformity of material culture across hundreds of miles and a great many centuries implies considerable control and planning11 and suggests, to some scholars, a threat of authoritarian or even totalitarian government. Some speak of the "affluent private residences with bathrooms served by a drainage system," while "the poor, however, lived huddled in slums, the inevitable underclass in a hierarchical system,"12 and others have seen in the tiny identical houses (protohousing projects? ghettos?) and in the massive government structure, regulating every single brick, an "obsessive uniformity."13 There is evidence that different professions worked out of distinct areas of the cities, suggesting the existence of something like protocastes.14 Some scholars have taken the visible signs of an overarching hand of authority and urban planning as evidence of "urbanity, sophistication, well-being, ordered existence." 15 One might also see, in the tiny scale of the seals and the figurines, and in the children's toys, a delicate civilization, whose artwork is fine in both senses--beautiful and small. PICTURES AND SYMBOLS: THE SEALS AND THE SCRIPT The civilization of the Indus is not silent, but we are deaf. We cannot hear their words but can see their images.ap Most of the seals, which are found throughout the Indus Valley Civilization, are engraved with a group of signs in the Indus script, or a drawing or design, or a combination of these.16 There are well over two thousand inscriptions, using about four hundred graphemes, and many people have claimed to have deciphered them, often demonstrating truly fantastic flights of imagination, but no one has definitively cracked the code.17 The individual messages are too short for a computer to decode, and since each seal had a distinctive combination of symbols, there are too few examples of each sequence to provide a sufficient linguistic context. The symbols that accompany a given image vary from seal to seal, so that it's not possible to derive the meaning of the words from the meaning of the images. Many people have speculated that it is an Indo-European language, or a Dravidian language, or a Munda or "Austro-Asiatic" language18 (supported by the plate tectonics narrative), or not a language at all.aq19 The seals may well have been nothing but devices to mark property in the manner of a signet ring, a stamp of ownership, rather like a bar code,20 probably made for merchants who used them to brand their wares, signifying nothing but "This is mine." Perhaps the writing is a form of ancient shorthand. Because they present a vivid, highly evocative set of visual symbols, but no text, these images have functioned, for scholars, like Rorschach shapes onto which each interpreter projects his or her own vision of what the hypothetical text should be and should say.ar The ambiguity and subjectivity of the interpretation of visual images are yet another aspect of the shadow on the moon that is, for some, a rabbit, and for others, a man. But the images on the seals do make a more general statement that we can decipher, particularly in the realm of flora and fauna. The vast majority of Indus signs can be directly or indirectly related to farming: Typical signs include seeds, fruits, sprouts, grain plants, pulses, trees, farm instruments (hoes, primitive plows, mortars and pestles, rakes, harvesting instruments, etc.), seasonal/celestial or astral signs, and even at times anthropomorphized plowed fields. The images, as well as other archaeological remains, tell us that the winter Indus crop was barley and wheat; the spring crop, peas and lentils; and the summer and the monsoon crops, millets, melons, dates, and fiber plants.21 They also probably grew rice.22 They spun, wove, and dyed cotton, probably for the first time on the planet Earth, and may also have been the first to use wheeled transport.23 They ate meat and fish.24 INDUS ANIMALS Animals, both wild and tame, dominate the representations from the IVC, both on the seals, where they seem to have been drawn from nature, and on figurines, paintings on pottery, and children's toys. These images tell us that tigers, elephants, and one-horned rhinoceroses, as well as buffalo, antelope, and crocodiles, inhabited the forests of this now almost desert region, which then had riverine long grass and open forest country, the natural habitat of tigers and rhinoceroses.25 (A rhinoceros, a buffalo, and an elephant, all on wheels, were found in a later site in northern Maharashtra, perhaps connected with Harappa.)26 There are also animal figurines of turtles, hares, monkeys, and birds, and there is a pottery model, 2.9 inches long, of an animal with a long, bushy tail, perhaps a squirrel or a mongoose.27 But it is the representations of domesticated animals, as well as the archaeological remains of such animals, that tell us most about the culture of the IVC, in particular about the much-disputed question of its relationship (or lack of relationship) with later Indian cultures such as that of the Vedic peoples. Millennia before the IVC, people in South Asia had hunted a number of animals that later, in the IVC, they bred and domesticated (and sometimes continued to hunt). Before the IVC, they had also domesticated two distinct species of cattle--the humped zebu ( Bos indicus ), with its heavy dewlaps, and a humpless relation of the Bos primigenius of West Asia.28 Zebu and water buffalo ( Bubalus ) were used as draft animals, and elephants (domesticated, more or less) were used for clearing and building.29 Elephants are not native to the lands found west of central India, but they might have been imported into the Indus Valley.30 They had dogs (which may already have been domesticated at Bhimbetka). Marshall, who participated in the first excavations of the site, commented on them at length: As would be expected, the dog is common, but all the figures but one are roughly modeled and evidently made by children. That this animal was a pet as well as a guard is proved by some of the figures being provided with collars. We have found a very mutilated figure of a dog with a collar, fastened by a cord to a post, which suggests that house animals were sometimes too fierce to be allowed at large. The one well-made exception . . . almost resembles the English mastiff of to-day.31 He also noted a figure of a dog with its tongue hanging out, "a detail seldom shown in a pottery model."32 The particular breeds of dogs depicted in small statues at the IVC include pariah dogs and, surprisingly, dachshunds.33 They had also domesticated camels, sheep, pigs, goats, and chickens. This may have been the first domestication of fowl,as a major contribution to world civilization.34 Apparently they did not have, or at least think it worthwhile to depict, cats. On seals and pottery, and depicted as figurines, the favorite subject is male animals--most frequently bulls with pendulous dewlaps and big pizzles. There are also short-horned bulls,35 but in general they went in for horned males: bulls, water buffalo, rams, and others. One scene even depicts a tiger with horns.36 By contrast, they do not seem to have found female animals very interesting, and significantly, no figurines of cows have been found.37 Marshall even comments on this absence, the cow that does not moo in the night, as it were: "The cow, even if it was regarded as sacred, was for some reason, at present unexplained, not represented in plastic form or carved in stone."38 Of course they must have had cows, or they couldn't very well have had bulls (and indeed there is material evidence of cows in the IVC), but the art-historical record tells us that the Indus artists did not use cows as cultural symbols, and why should we assume, with Marshall, that they were sacred? Why, in fact, do archaeologists reach for the word "sacred" every time they find something for which they cannot determine a practical use? (This is a question to which we will return.) The seals depict animals that have been characterized as being "noted for their physical and sexual prowess--bulls, rhinoceroses, elephants, and tigers--or, as is true of snakes and crocodiles . . . widely regarded as symbols of sexuality, fertility, or longevity."39 Of course we don't really know how good crocodiles are in bed; our culture thinks of them (or at least their tears and smiles) as symbols of hypocrisy; why should they be symbols of sexuality, and to whom (other than, presumably, other crocodiles)? It has even been suggested that "the present untouchability of dogs could originate from their being sacred [in the IVC] and thus untouchable."40 The equation of sacrality and untouchability is as unjustified as the assumption that the attitude to dogs did not change in four thousand years. THE UNICORN (AND OTHER POSSIBLY MYTHICAL BEASTS) This question of symbolic valence becomes more blatant in the case of more fantastic animals, like unicorns. The most commonly represented Indus animal, depicted on 1,156 seals and sealings out of a total of 1,755 found at Mature Harappan sites (that is, on 60 percent of all seals and sealings), is "a stocky creature unknown to zoology, with the body of a bull and the head of a zebra, from which head a single horn curls majestically upwards and then forwards."41 What is this animal? Is it just a two-horned animal viewed from the side or a kind of gazelle with a horn on its nose? Is it a horse with a horn? (It doesn't have the proportions of a horse.) Or a stylized rhinoceros?42 Or is it, by analogy with its European cousin, a mythical beast? The quasi unicorn always (like other Indus animals sometimes) has a manger in front of him. The manger is sometimes said to have "religious or cultic significance," since one seal shows an image of a unicorn being carried in procession alongside such a manger.43 Often the manger is called sacred (presumably on the basis of the sacrality of mangers in Christianity). The unicorn lands us on the horn of a dilemma: Are the animals represented in the art of the IVC religious symbols? Though many Indus animal figurines are simply children's toys, with little wheels on them, scholars persist in investing them with religious meanings. Some of the fossil record too has been invoked as religious testimony. The excavation, in 1929, of twenty severed human skulls "tightly packed together," along with what the excavator interpreted as ritual vessels and the bones of sacrificed animals, has been taken as evidence that human heads were presented to a sacred tree,44 a scenario reminiscent of the novel The Day of the Triffids or the film The Little Shop of Horrors ("Feed me! Feed me!" cried the carnivorous plant). And why should an archaeologist have identified the image of a dog threatening a man with long, wavy hair as the hound of Yama, the god of the dead,45 simply because the dog appears on the burial urns at Harappa? Why can't it just be a dog faithful in death as in life? Unicorn Seal from Harappa. And why are the two figures in front of a pair of cobras "a pair of worshipers"? 46 Why not just two, probably nervous blokes? Yet the rest of the scene does indeed suggest something other than common or garden-variety snake charming. The couple with the cobras is kneeling beside a seated figure; another human figure holds back two rearing tigers; a monster half bull and half man attacks a horned tiger. This is not a snapshot of everyday life in the IVC. Scenes and figures such as these may give us glimpses of rituals, of episodes from myth and story, yet we have "nothing to which we can refer these isolated glimpses to give them substance."47 Other seals too seem to be telling a story that we cannot quite make out. One scene depicts what has been called "a three-horned deity" (but may just be a guy, or for that matter a gal, in a three-horned hat) apparently emerging from the middle of a tree, while another figure outside the tree is bent "in suppliant posture" with arms raised; a bull stands behind it, and seven girls below them.48 (This is one of a small number of scenes that occur on seals found in four different cities: Harappa, Kalibangan, Mohenjo-Daro, and Chanhujo-Daro). Another seal depicts a similar scene, this one involving a fig (pipal) tree:at A nude figure with flowing hair and "a horned headdress" (or his own horns?) stands between the upright branches of a pipal tree; another figure, much like the first but seen from the side, kneels at the base of the tree; a huge goat towers over him from behind. On yet another seal a figure squats among a group of animals on his left, while on his right a tiger is looking upward at a tree in which a man is seated.49 Something is certainly going on here, but what? A folktale, perhaps? A ritual? These wordless scenes remind me of those contests that magazines run, inviting readers to supply the caption for a cartoon. But a lot more is at stake here than a cartoon. GENDERED FIGURES THE LORD OF BEASTS Marshall began it all, in 1931, in his magisterial three-volume publication, Mohenjo-Daro and the Indus Civilization , which devotes five pages of a long chapter entitled "Religion" to seal 420: "There appears at Mohenjo-Daro a male god, who is recognizable at once as a prototype of the historic Siva. . . . The lower limbs are bare and the phallus ( urdhvamedhra ) seemingly exposed, but it is possible that what appears to be the phallus is in reality the end of the waistband."50 ( Urdhvamedhra ["upward phallus"] is a Sanskrit term, like the Greek-based English euphemism "ithyphallic," for an erect penis.) The urdhvamedhra -or-is-it-perhaps-just-his-waistband-or-the-knot-in-his-dhoti? has come to rival the Vedantic snake-or-is-it-perhaps-just-a-rope? as a trope for the power of illusion and imagination. The image suggested to Marshall an early form of the Hindu god Shiva, and Marshall's suggestion was taken up by several generations of scholars. This was to have far-reaching ramifications, for if this is an image of Shiva, then an important aspect of Hinduism can be dated back far earlier than the earliest texts (the Vedas). Much was made of this tiny bit of soapstone (remember, the whole seal is barely an inch high); the millimeter of the putative erection on this seal has, like the optional inch of Cleopatra's nose, caused a great deal of historical fuss. Scholars have connected the "big-nosed gentleman . . . who sits in the lotus position with an erect penis, an air of abstraction and an audience of animals"51 with well-known images of the ithyphallic Shiva.52 The discovery at Indus sites of a number of polished, oblong stones, mostly small but ranging up to two feet in height, and probably used to grind grain, has led some scholars53 to identify these stones as replicas of the erect phallus (linga) of Shiva and the vagina (yoni) of his consort, and to link these stones with "the later aniconic representations" of Shiva in the form of the linga.54 Other scholars have suggested that "the Vedic criticism of 'those who worship the phallus'" may refer to this "early Indus cult."55 There are so many assumptions here that it makes your head spin: that the Indus had a "cult" (a rather pejorative word for a religious sect), that the people of the Veda knew about it, that they disapproved of it instead of assimilating it to their own worship of the phallic Indra--no lawyer would go into court with this sort of evidence. These all are arguments from hindsight. Marshall identified the figure as Shiva because (1) the Indus figure is seated on a low stool with knees pointed to the sides, feet together at his groin, and arms resting on his knees, a posture that many have identified as yogic (though it is the way that South Asians often sit), and Shiva is the god of yogis; (2) the Indus figure wears a horned headdress (or has horns), perhaps a buffalo mask as well as buffalo horns,56 just as Shiva wears the horned moon, or a trident, in his hair; [3] in two examples of this scene the Indus figure has faces (or masks?) on the sides as well as the front of his head, while Shiva is often "Five-faced" ( Pancha-mukha ); (4) the figure is flanked by an elephant, a rhinoceros, and a water buffalo; smaller horned animals--antelope or goats--huddle beneath his stool, and he wears a tiger's skin on his torso, while Shiva is called the Lord of Beasts, Pashupati, and wears an animal skin, sometimes of a tiger, sometimes of an elephant;57 and (5) both figures are ithyphallic. I bought into the identification with Shiva in 1973,58 as most scholars have continued to do right up to the present day. Yet many other candidates have also been pushed forward,59 another good example of the Rorschach (or Rashomonau) phenomenon that produced such rich fantasies about the decipherment of the script. A list of just a few of the figures with which the so-called Lord of Beasts has been identified, a list that the reader should not take seriously but merely skim over to see how creatively scholars can run amok, might run like this (in more or less chronological order): 1. A goddess, on whom the bulge previously identified as an "erect phallus" is nothing but a girdle worn by female IVC figurines.60 2. Mahisha, the buffalo demon killed, in later mythology, by the goddess Durga, who is often represented as a riding on a tiger61 (or a lion). 3. Indra, the Vedic king of the gods,62 a conclusion supported by taking the first syllables of the Sanskrit words for three of the animals (eliminating the tiger, because it was much larger than the other animals, and the deer, because they are seated apart from the others, and repeating the first syllable of the word for "man," because he was twice as important as the others), so that they spell out ma-kha-na-sha-na , an epithet of Indra (though also of Shiva), "destroying the sacrifice."av 4. Rudra, a Vedic prototype of Shiva, surrounded by animals who are incarnations of the Maruts, the storm gods who serve Indra and Rudra.63 5. Agni. The pictograms are read to mean "burning in three ways" and so to identify the figure as Agni, the god of fire, who has three forms.64 6. A chief named Anil, who ruled over the clans whose totems were the animals on the seal.65 7. A "seated" bull.66 8. A sage (named Rishyashringa ["Antelope-horned"]) who had a single antelope horn growing out of his forehead (his mother was a white-footed antelope; it's a long story); he appears in the earliest layers of Hindu and Buddhist mythology.67 9. Part of "a bull cult, to which numerous other representations of bulls lend substance."68 10. A yogic posture,69 even if the link with Shiva is tenuous.70 Most, but not all, of these fantasies assume that the image is a representation of either a priest or a god, more likely a god.71 In each case, the interpretation was inspired or constrained by the particular historical circumstances and agendas of the interpreter, but I'd love to know what the scholars who came up with these ideas were smoking. There is, in fact, a general resemblance between this image and later Hindu images of Shiva. The Indus people may well have created a symbolism of the divine phallus, or a horned god, or both. But even if this is so, it does not mean that the Indus images are the source of the Hindu images. We must keep this caution in mind now when we consider the images of women in the IVC. MOTHERS AND MOTHER GODDESSES The widespread depiction of women in the IVC artifacts suggests that they were highly valued. In contrast with the predilection for macho animals (includingmen) on the seals, the many terra-cotta figurines are mostly women, some wearing a wide girdle, a necklace, and an elaborate headdress. They are "Pop-eyed, bat-eared, belted and sometimes mini-skirted."72 Some of them seem to be pregnant, or to hold, on their breasts or hips, small lumps that might be infants, "evidence perhaps that they expressed a concern for fecundity," a reasonable assumption;73 they may have been symbols of fecundity in a "loosely structured household cult."74 But why assume any cult at all? Why need they symbolize fertility? Or even if they do, why should fertility have to be ritual? (I must confess to having fallen for this too more than a quarter of a century ago: "[S]trong evidence of a cult of the Mother has been unearthed at the pre-Vedic civilization of the Indus Valley [c. 2000 B.C.]."75 Live and learn.) But not every image is symbolic; not every woman is a goddess. The "prominent and clumsily applied breasts" of these figures have been taken as evidence that they were "fertility symbols,"76 but they may have been valued simply for what P. G. Wodehouse's Bertie Wooster used to refer to as a "wonderful profile." Big breasts are as useful to courtesans as to goddesses. Are the buxom centerfolds of Playboy magazine fertility symbols, or the voluptuous women that Rubens loved to paint? One seal shows a woman, upside down, with a child (or is it a scorpion?) coming out of (or into?) her, between her spread thighs.77 This has been taken to refer to "a possible Mother Earth myth,"78 but what was the myth, and is the upside-down woman a goddess, let alone an earth goddess? Why is she not simply a woman giving birth? Scholars have seen connections between the alleged Lord of Beasts and a goddess, particularly the Hindu goddess who rides on a lion; some (casually conflating lions and tigers) connect the tiger on the seals with Hindu goddesses of a later period or with goddesses of ancient Egypt, the Aegean, Asia Minor, and the whole of West Asia, who were thought to consort with lions, leopards, or panthers.79 The assumption that the figures of women found at the Indus sites are goddesses is then used to support the argument that the goddesses in later Hinduism--or the minor Vedic goddesses, Yakshinis and Apsarases, associated with trees and water80--may be traced back to this early period.81 Hindsight speculations about fertility sects associated with female figurines, the bull, the horned deity, and trees like the sacred fig (pipal) are tempting. The seal with the person emerging from the middle of a fig tree may or may not prefigure the later Indian iconography of fig trees and banyan trees.82 But it is going too far to interpret something so straightforward as a grave containing a male and female skeleton as "possibly the first indication of the well-known Hindu custom of sati" (live widows burning themselves to death on their dead husbands' cremation pyres or entombing themselves in their husband's graves).83 The couple may simply have been buried side by side, whenever they died. Some of the figures of well-endowed women are "curiously headless," and in some cases of actual adult burial the feet had been deliberately cut off, a fascinating correspondence, perhaps joined in a Procrustean syndrome. These headless female figures84 may foreshadow the headless goddesses who people later Hindu mythology, such as the Brahmin woman who exchanged heads with the Pariah woman. (Or is it just that the neck is the thinnest part of such figures and most likely to break?) The prevalence of images of women may well indicate "a greater social presence of the female than in later times, which may also have been a generally more assertive presence."85 One tiny (ten-centimeter) bronze image supports the hope that some Indus women did in fact have an "assertive presence" and that is the so-called dancing girl of Mohenjo-Daro, in whom Marshall saw a "youthful impudence." John Keay describes her wonderfully well: Naked save for a chunky necklace and an assortment of bangles, this minuscule statuette is not of the usual Indian sex symbol, full of breast and wide of hip, but of a slender nymphet happily flaunting her puberty with delightful insouciance. Her pose is studiously casual, one spindly arm bent with the hand resting on a déhanché hip, the other dangling so as to brush a slightly raised knee. Slim and attenuated, the legs are slightly parted, and one foot--both are now missing--must have been pointed. . . her head is thrown back as if challenging a suitor, and her hair is somehow dressed into a heavy plaited chignon of perilous but intentionally dramatic construction. Decidedly, she wants to be admired; and she might be gratified to know that, four thousand years later, she still is.86 Others too admired her "gaunt and boyish femininity," her provocative "foot-less stance, haughty head, and petulantly poised arms,"87 and found "something endearing" in "the artless pose of an awkward adolescent."88 She is said to have "proto-Australoid" features that are also attested in skeletons in the Indus Valley.89 This native girl mocks us, perhaps for our clumsy and arrogant attempts to figure out what she, and her compatriots in bronze and clay and soapstone, "mean." IS INDUS RELIGION A MYTH? The larger archaeological remains are equally ambiguous. Consider the very large swimming pool or bathing tank or public water tank in the citadel at Mohenjo-Daro, approximately forty feet by twenty-three feet and eight feet deep. There are wide steps leading down to it at each end and colonnaded buildings with small rooms around it. From this some have concluded that it was the site of a "Great Bath" where ritual bathing took place as part of a state religion.90 But all that this structure tells us is that the IVC people liked to bathe, just to get clean or to cool off on hot days or to splash about, same as we do. Cleanliness is next to godliness, but not synonymous with it. The great attention paid to the sewage system in the IVC suggests a hard-headed approach to hygiene (unless, of course, one wants to view the sewers as sacred underground chambers). Why does the bath have to be a ritual bath?aw Bronze Dancing Girl from Mohenjo-Daro. The answer is simple enough: because the so-called Great Bath resembles the ritual bathing tanks of Hindu temples that began to appear in the subcontinent in the first few centuries CE91 and because such a tank reflects a concern with ritual purification through water, an important idea in Hinduism.92 Four thousand years later, indeed, every temple has its tank. Therefore, the argument goes, the tank must have served the same function in the IVC. Similarly, the so-called College of Priests in Mohenjo-Daro has been taken as evidence for the existence of a widespread priesthood.93 Well, it's a big building, true, but why couldn't it be a dorm, or a hotel, or a hospital, or even a brothel? Works of art such as the images on the seals and other artifacts provide abundant evidence of imaginative art, perhaps mythological but not necessarily ritual. They may have been purely decorative, or they may illustrate narratives of some sort or convey some sort of symbolic meaning, probably more than one, as symbols often do. But did they necessarily express the symbols of an organized religion? There are no recognizable religious buildings or elaborate burials in the Indus cities ("Clearly, they did not expect huge demands on the dead in the after-life"94), no signs of ancestral rituals or "magnificent icons" or any "specially decorated structures." The conclusion is clear enough: "If there were temples they are difficult to identify. . . . The cities may not therefore have been the focus of religious worship."95 Yet the same fact--that no great temple or center of worship has been found as yet at Mohenjo-Daro--has inspired a very different conclusion: One place where such a structure might have been situated, just east of the bath, has not been excavated because a Buddhist stupa (reliquary mound) stands there, and permission has never been granted to move it.ax96 The stupa is indeed a strong hint that the structure underneath it might have been religious, for Buddhism shares with other religions (including, notably, Hinduism and Islam) the habit of sacred recycling, putting one religious building on the site hallowed by another, the funeral baked meats served cold for the wedding breakfast that follows. And one might argue that it would be odd if given the great regulation and standardization of everything else in public life, the governing powers did not also regulate belief. But all speculations about the role of religion in the lives of the IVC people rest on doubtful retrospective hindsight from Hindu practices many centuries later.97 Excerpted from Ebk Hindus: An Alternative History by Doniger All rights reserved by the original copyright owners. Excerpts are provided for display purposes only and may not be reproduced, reprinted or distributed without the written permission of the publisher.
