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 notrepresented 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.”³⁹Of 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 Rashomon^au) 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

The assumption of a theocratic elite in the IVC underpins the assertion that the images depicted on pictographic seal inscriptions and terra-cotta figures are divinities and that animism, demonic cults, fertility cults, and the worship of natural forces and mother goddesses flourished in the IVC.⁹⁸But surely it’s possible that the people of the IVC had no religion at all, in the sense of a state cult or an enforced dogma. Is it possible that this was the first secular state, anticipating the European Enlightenment by four thousand years? Could they have been more like protoatheists than protoyogis? After all, there were many people in later Hinduism who had no use for religion, people such as the Charvakasand the Lokayatas (“Materialists”).⁹⁹ (If people are going to argue for religious meanings from hindsight, one might as well also argue against it from hindsight; two can play at that game.) Just as it has been well argued that there is a very good reason why the IVC language remains undeciphered—because the seals may not record any language at all, merely random symbols of ownership¹⁰⁰—so too we may argue that the other symbols are not part of a coherent religious system but equally random artistic creations.

THE END OF THE INDUS

No one knows how the IVC came to an end. Perhaps it simply ran its course, had its day in the sun, and then the sun, as always, set on that empire. Perhaps it was destroyed by drought. Perhaps the Indus River changed its course, or there was an earthquake.¹⁰¹Perhaps massive deforestation degraded the environment. Perhaps the people died of diseases such as severe anemia, as the skeletal remains of what was previously interpreted as a “massacre” suggest.¹⁰² Perhaps it was destroyed by invasions. Last but certainly not least, perhaps it was destroyed by a flood; whatever caused the actual destruction, floods did eventually bury the cities in many layers of Indus mud, which caused both the ground level and the water table to rise by ten meters. Immigrations of new peoples, droughts, deforestation, floods, or alterations in the course of the life-giving river: any of these may have been contributing factors.¹⁰³ Whatever the cause, the result was that “on top of the cities, now consigned to oblivion beneath tons of alluvium, other peoples grazed their goats, sowed their seeds and spun their myths. A great civilization was lost to memory.”¹⁰⁴ But was it in fact lost?

The flood that may have destroyed the Indus cities may have been the inspiration for the myth of the great flood that is described in the Shatapatha Brahmana (c. 800 BCE) and that continues to haunt Hindu mythology to the present day. And it is tempting to argue that some or all of these stories are memories of (if not evidence for) a great flood that destroyed the IVC. But it would be better, I think, to resist that temptation and simply to suggest that present-day scholarly (or nonscholarly) theories of a catastrophic flood at the end of the Indus Valley Civilization were inspired by the myth of the flood, that the scholarly theories themselves are merely the latest variants of the myth of the flood.

TRANSFORMATIONS THROUGH TIME: FAST-FORWARD

Arguments from hindsight pervade the scholarship on the IVC, underpinning correlations between the quasi unicorn and the sage whose mother was an antelope, between the Lord of Beasts on the seal and Shiva, between various images of women and later Hindu goddesses, between the “Great Bath” and the bathing tanks of Hindu temples, between small conical objects that have been interpreted as phallic stones (but may just be pieces used in board games) and the Shiva linga. The obsession with descendants, arguing that the IVC seal can be explained by what we know of Shiva as Lord of Beasts, is the other side of the coin of the obsession with origins, arguing that the figure of Shiva as Lord of Beasts is derived from, and to some extent explained by, the IVC seal. The two approaches scratch each other’s backs. The fascination with the IVC comes in part from the intrinsic appeal of its artifacts but also from a perceived need to find non-Vedic, indeed pre-Vedic sources for most of Hinduism—for Shiva and goddess worship and all the rest of Hinduism that is not attested in the Vedas.

On the other hand, it is always tempting to look for the keys to the IVC where there is available light in later Hinduism, to let Hindu phenomena, which have the context of texts to explain them, illuminate the darkness that surrounds many early Indus images and objects that lack such verbal commentaries. But too often scholars read the Indus images like the pictures in the puzzle books of my youth: How many (Hindu) deer can you see hiding in this (Indus) forest? Throughout this chapter, and indeed throughout this book, I have poured into my ears the wax of pedantic caution, in an attempt (sometimes in vain) to resist the siren song of hindsight. For fig trees, horns, bulls, phalluses, and buxom women do play a central role in later Hinduism, and such images may have been important to Hindus in part because they never lost some of the power they had had in the Indus period. Although these images certainly also occur in many other cultures, the hypothesis that Hinduism inherited them from the Indus seems a more efficient explanation than coincidence or independent origination. Nor is there any likely source (with the possible, but by no means established, exception of Elam) from which the two cultures could have borrowed the same images. It is probable that the forms survived; the mistake is in assuming that the function follows the form. The inhabitants of both Mohenjo-Daro and modern Mumbai had bulls, but they surely had very different ideas about bulls.

It is useful to distinguish hindsight from fast-forwarding. Hindsight often misreads an earlier phenomenon by assuming that it meant then the same thing that it meant later, reading the past through the present, forgetting that we cannot simply lay the present over the past like a plastic map overlay. The false Orientalist assumptions that India was timeless and that the classical texts of the Brahmins described an existing society led to the equally false assumption that the village and caste organization of colonial or even contemporary India was a guide to their historical past.¹⁰⁵

But at times the atavisms, the modern traces of ancient phenomena, are so striking that it would be perverse to ignore them, and from time to time I have fast-forwarded to note them. We should not impose the meaning of the later icons upon the earlier images, but once we have explored the meaning of the Indus representations within the constrictions of their own limited context, we can go on to speculate on how they may have contributed to the evolution of a later iconography that they sometimes superficially resemble.¹⁰⁶

For the resemblances between some aspects of the IVC and later Hinduism are simply too stunning to ignore. As the Late Harappan culture declined, its survivors must have carried some of it into the Ganges-Yamuna basin. There are links between archaeological records among the communities of the third millennium BCE, which used only stone and bronze, and the people of the Gangetic plain and the Deccan in around 1000 BCE, who developed the use of iron. At this time, or even a few centuries earlier (in 1500 BCE), the process of urbanization moved gradually south from the Indus cities to the site of Kaushambi, near modern Allahabad in the Gangetic plain, and to the surrounding villages. The material culture does not show continuities; the use of bricks of standard sizes, the geometrical grids, the seals, the sewers, the large urban plan, none of this is preserved.¹⁰⁷ Above all, the technique of administration was lost; not for many centuries would anyone know how to govern such a large community in India. But someone succeeded in preserving on the journey south and east some of the cultural patterns nurtured in the Indus cities, for some of these patterns lived on long after the cities themselves were gone.¹⁰⁸ The Indus civilization may not have simply gone out like the flame of a candle or, at least, not before lighting another candle.

We can see the possible survival, in transformation, of a number of images. The Harappan motif of the fig (pipal)—as a leaf decoration on pottery and as a tree on seals—reappears in the imagery of some later religious sects.¹⁰⁹ There is a conch shell, etched in vermilion, that may well have been used as a libation vessel, just as conch shells, etched in vermilion, are used in Hinduism today. Not only individual images but also aspects of the art forms—especially the so-called animal style, stylized and rounded, with just a few meticulous and suggestive details—seem to have survived. Some of the stylized depictions of the animals on the seals bear a striking resemblance to the depictions of the same animals two thousand years later (and magnified many hundredfold) on the capital plinths of the pillars of Ashoka.¹¹⁰ These patterns, and the rough outlines of other images that we have considered, perhaps even the stone lingas and the voluptuous women, may have gradually merged with the culture of the people of the Veda.

Horse on the Ashokan Column at Sarnath.