The Lost River: On The Trail of Saraswati

by Michel Danino

  If one were to create an iconography for this whole symbolism, it would be hard to think of a better one than the unicorn. Moreover, the Rig Veda compares the Maruts (a family of violent gods) to bulls whose horn is ‘uplifted unto the highest’,49 just as we see on the seals.‡ Even the invariably double or S-shaped curve of the unicorn’s horn, executed so carefully that it must have a precise meaning (the bulls’ horns, by contrast, have a single curve), could be explained within the same Vedic imagery, as a means to represent in a compact space the ‘hundred joints’ of Indra’s thunderbolt.50

  That the unicorn actually stood for Indra in the eyes of Harappans cannot be proved, but its affinity with the Vedic concepts of a mighty bull with a sharpened horn certainly calls for attention.

  FIRE WORSHIP

  Two more elements of Harappan culture throw important bridges across the ‘Vedic Night’. The first is fire worship, evidenced by a few altars that have come to light in Harappan cities.

  We have already seen (Fig. 7.2) the peculiar apsidal altar on Banawali’s acropolis, itself enclosed within an apsidal structure. When it was uncovered towards the end of the partial excavations at this site, R.S. Bisht noted that the altar ‘was full of fine loose ash’;51 a large jar was found in the structure. Seven years ago, Banawali’s altar came on the screen as I showed slides of the Indus-Sarasvatī civilization to venerable Vedic scholars assembled in a field at Pañjal, in Kerala.52 Some of them, masters of one or several of the Vedas, interrupted me to point out that the semi-circular shape is one of the three basic shapes of fire altars (called dakshināgni), with the other two being the square (āhavanīya) and the circle (gārhapatya); the three could be seen on the grounds of the conference, and are still in use in various ceremonies.53 (They are visible inside the enclosure to the west of the mahāvedi, see Fig. 9.4, Nos 2, 3 and 4). While those Vedic scholars of Kerala recognized in the Banawali structure a fire altar, Bisht reached a similar conviction independently: to him, this clearly non-utilitarian structure enclosing an altar with layers of ashes could only be a temple dedicated to fire rituals.

  Further down the Sarasvatī, at Kalibangan, we already visited a series of seven altars located atop a brick platform on the southern side of the acropolis (p. 160). Sceptics suggested that this could be a community kitchen, with the central steles acting as base for the pots,54 but this is unlikely for several reasons. One, generally kitchen ovens or hearths used for cooking had a lateral opening for the insertion of firewood, but that is not the case for Kalibangan’s seven altars. Two, the central steles, with diameters of 10 to 15 cm, are too slender to support pots. Three, here again the altars contained ashes (with some charcoal), but no bone remains55 as one would expect from domestic hearths, since we know that Harappans were non-vegetarian. The scales are thus clearly tipped in favour of some fire ritual.

  Such altars were also found in some individual houses of Kalibangan’s lower city; further east, outside the fortified lower city, a small but badly eroded mound produced a unique complex of altars—neither apsidal nor rectangular, but more or less circular this time: a large altar, over 2.5 m in diameter, was irregularly surrounded by five small ones, all of them containing ash and terracotta cakes. Noting the presence of altars on Kalibangan’s three mounds, Raymond Allchin remarked, ‘These three contexts suggest that fire rituals formed a part of the religious life of the town, at a civic, domestic and popular level.’56

  Moving to Saurashtra, Lothal yielded similar structures not only of an oval shape, but also a few square ones too. Here also, the excavator S.R. Rao concluded that the latter could not have been kitchen hearths: among other reasons, they are much too large (about 2.7 m square) and some were located right on the street.57 One of them (Fig. 10.15) showed depressions in the top row of bricks, probably designed to hold vessels in place, and a post hole in a corner. But it also offered three strong clues to the Vedic fire ritual. A fine painted jar of terracotta with a broad mouth was placed against the altar; a few steps away, a terracotta ladle58 was found, with blackened edges and bottom pointing to contact with fire. It would have been the right instrument for making offerings to the fire of oil, milk and clarified butter drawn from the jar. (Incidentally, both sacrificial jar and ladle figure in the Vedas in association with the constant theme of libation.)

  The third clue emerges from a comparison with the Shulbasūtras, ancient technical texts that I briefly invoked in relation to the proportions of the mahāvedi; they give elaborate geometrical instructions on the construction of altars of various shapes—falcon, tortoise, wheel, or a simple square. The last has, on one side, a small platform jutting out, called the ‘handle’ of the altar (Fig. 10.16). Lothal’s structure is about half the size of the Shulbasūtras’ square altar, but surprisingly, on one side, it has a small platform which looks very much like the ‘handle’. Moreover, the length of the platform is one-fourth of the altar’s side—exactly as with the Shulbasūtras’ square altar.59 The ‘coincidence’ is intriguing, to say the least, and while the Shulbasūtras belong to a later age, some of the concepts they articulate appear to have originated in the Harappan tradition. At any rate, if Lothal’s structure, located on one of the town’s main streets, with its platform, jar and ladle, was not a fire altar, no better explanation has been proposed so far.

  Such fire structures have been reported from other Indian sites, such as Nageshwar, Vagad (both in Gujarat) and Rakhigarhi,60 but not from Mohenjo-daro or Harappa. (At Mohenjo-daro, archaeologists have occasionally identified one or another building with a temple,61 but there are no clues as to the exact type of worship conducted in them.) It may be that fire worship was only practised in the Sarasvatī Valley (Banawali, Rakhigarhi, Kalibangan) and in Gujarat. Conversely, mother-goddess figurines are common in Mohenjo-daro and Harappa but rare in the Sarasvatī region. Pending more extensive excavations, it is legitimate to assume a certain regionalization of the Harappan religion: it was not uniform but allowed for ‘diversity of practice’,62 as Possehl puts it.

  The presence of fire worship in the Sarasvatī region and in Gujarat has generated a good deal of interest as well as disbelief: not all scholars have accepted it as readily as Allchin, Lal, Rao or Joshi. The reason for such scepticism is plain enough: in the Indian context, fire worship conjures up Vedic culture.

  YOGA

  So does another Harappan practice, which has caused a lot of scholarly ink to flow. The impressive ‘proto-Shiva’ is seated in such a way that he rests on his toes with the two heels in contact; several more seals and tablets represent a figure—very likely the same ‘proto-Shiva’—seated in the same manner on a low platform or pedestal. This posture has sometimes been wrongly identified as padmāsana, for in the lotus posture, the legs are crossed, which is not the case here. At least two scholars—T. McEvilley63 and Yan Y. Dhyansky64—went into minute detail to establish that this posture is the classical mūlabandhāsana, a difficult āsana whose function is to help awaken the kundalinī.

  Besides, at Mohenjo-daro, Harappa and Lothal, excavators found figurines in different postures recalling other āsanas.65 A striking example from Harappa (Fig. 10.17) shows a man seated cross-legged with his hands joined in the traditional Indian namaste: an average Indian would be hard put to decide whether it was made 4500 years ago or yesterday!

  Even more arresting is the famous ‘priest-king’ (Fig. 10.18). At the time of its discovery at Mohenjo-daro, it was often speculated that, as in Mesopotamia, the city must have been governed by a class of ‘priest-kings’; with no evidence in support of this hypothesis, it was gradually abandoned, but the label stuck to the statuette. This impressive personage has also been compared to figurines from Bactria on stylistic grounds,66 but even if one admits some likeness, there is a fundamental difference : as early as in 1929, the Indian archaeologist Ramaprasad Chanda pointed out that the figure’s ‘half-shut eyes looking fixedly at the tip of the nose’67 signified a distinctly Indian trait not to be found anywhere else, which suggested spec
ific techniques of meditation.

  In other words, our ‘priest-king’ is deep in contemplation—an attitude not known to be attached to kings or priests, especially of the Mesopotamian sort: as Wheeler noted, ‘the fixed Mesopotamian stare is very different from the contemplative expression of the Indus faces’.68 Though quite unlike the powerful god with a tricorn head-dress, our personage is part of the same cultural complex of yoga, if we take the word in its central sense of inner exploration and mastery. Dilip Chakrabarti refers to him as a ‘Shramana’ (ascetic) or a ‘sacred person’,69 certainly a better description than that of a priest or a king.

  It is often claimed that there is no notion of yoga in the Rig Veda, and that the concept of a ‘Harappan yoga’, if admitted, can still be pre-Vedic. But while the Rig Veda does not use the term ‘yoga’, it asks for our thoughts to be harnessed or yoked to a higher consciousness:

  The illumined yoke their mind and they yoke their thoughts to the illumined godhead, to the vast, to the luminous in consciousness.70

  In many hymns (as, later, in the Bhagavad Gītā), the metaphor of the well-yoked chariot transparently stands for mastery over the instrument: horses are ‘yoked by thought’71 or ‘yoked by prayer’72 to the chariot.§ Central to the Rig Veda’s quest for immortality is the effort to release Agni, the divine fire ‘who dwells in creatures, in whom all creatures dwell, [but who] is hidden within mortals by hostile powers’.73 There would be much more to say on the spiritual pursuit of the Rig Vedic rishis, but while it is couched in a language very different from that of the Upanishads, its goal and methods are the same.74

  If ‘Yoga was present in India five thousand years ago’,75 as Dhyansky concludes in his paper entitled ‘The Indus Valley Origin of a Yoga Practice’, there is no need to invoke some unknown and unverifiable ‘pre-Vedic’ tradition.

  LIFE AND DEATH

  Harappan funerary practices were of several types. Some of the dead were buried in earthen graves often lined with bricks, their heads generally to the north, the direction used today during cremation (since the feet should point southward, towards Yama’s realm). Some were buried in wooden coffins, but that seems to be the exception rather than the rule; whether they were rulers, high officials or high priests we may never know, although the last is not unlikely in the case of a few skeletons wearing necklaces of beads and various amulets. Some burials were symbolic ones containing grave offerings with no bones. Other modes of burial, in big urns devoid of bones, may have involved cremation, and since the overall number of graves only accounts for a fraction of the population, cremation is often believed to have been practised by the lower strata of the society at least.

  Even if the graves were reserved for an elite, they were quite plain in comparison with the sumptuous royal tombs of ancient Egypt: nothing marked them out in the landscape, and an assortment of pots of various sizes and shapes was all that accompanied the dead, along with a few personal objects (beads, rings, bangles or copper mirrors)—there was no gold, no frescoes, no decorated sarcophagus, nothing to cause awe in the grave’s discoverer. Several archaeologists have pointed out that this restraint may hold a message: Harappans did respect their dead, but did not believe in glorifying them. To ancient Egyptians, there was eternal life after death; pharaohs and members of the social elite therefore had to be suitably ‘equipped’ with every possible luxury. In contrast, if Harappans did believe in some afterlife, as the grave offerings show, they preferred to keep their wealth in circulation, favouring life over death—that is, indeed, a characteristic Indian attitude.

  Another approach comes from D.D. Kosambi, who drew attention to a funerary urn at Harappa (the Cemetery H phase) : on it is painted a peacock with a circular body, inside which can be seen a recumbent human figure (Fig. 10.19). Kosambi found here a parallel to the Mahābhārata’s reference to the dead ‘having been eaten by birds and insects of various sorts, but specifically by peacocks’.76 I may add that a Rig Vedic hymn, which speaks of ‘a glory beyond this realm’ to which we pass through old age, mentions ‘man-consuming birds’77 in the very next verse.

  A bird was also sculpted on the lid of a terracotta pot in one of the graves at the recently discovered Harappan site of Sanauli, a few kilometres east of the Yamuna. This unique site turned out to be an extensive burial ground: 116 graves have so far been excavated, and the full extent of the cemetery is thought to be some four hectares, prompting the excavators to call it a ‘necropolis’.78 They felt tempted to correlate some of the unique artefacts and burial modes that came to light in Vedic literature, but more research is needed to put this on a sound footing. Meanwhile, we should note the discovery of an oblong trough of clay: its vitrified inner walls and the presence of ash and charred human bones point to cremation, perhaps the first direct evidence of the practice.

  VEDIC VS HARAPPAN

  There are other, subtler threads connecting the two civilizations of early India, or bridging the Harappan and Vedic cultures. We noted earlier the community-based distribution of power (and, probably, functions) in Harappan society, a culture of ‘unity in diversity’, and the absence of glorified rulers—traits equally visible in the early stages of the Ganges civilization. On an ethical level, the four classical goals of life—dharma (the cosmic and individual order), artha (wealth), kāma (pleasures) and moksha (liberation)—constituted the cultural foundation of historical India; referring to this four-fold order, Kenoyer judiciously observes that the Indus rulers appear to have observed the first three concepts by promoting trade and wealth ‘without the extensive use of military coercion’.79 Indeed, with the just noted Harappan tradition of yoga and meditation, we are entitled to make out the presence of the fourth concept too—moksha!

  If we return to Dholavira’s town planning (p. 198), we can discern two important Vedic concepts at play. The first is the addition of unity, visible in the ratios (5:4 is nothing but ¼ plus one unit; 9:4 is the same plus another unit), and crucial to the calculations of fire altars in the Shulbasūtras, for instance.80 This addition to the unit of a fraction of itself represents, at a deeper level, a process of expansion, of auspicious increase symbolizing or inviting prosperity, and it becomes a standard method of generating auspicious proportions in classical Hindu architecture.81 The second Vedic concept is that of recursion or repetition of a motif, also visible in classical architecture as a way of repeating the initial unity and of growing from it.82 (In temples, shikharas of increasing height build up towards the towering last one.) At Dholavira, the ratio 5:4 found in the castle’s proportions is repeated in those of the lower town; again, the ratio 9:4, between the lengths of the ‘castle’ and the middle town, is repeated between the lengths of the middle and the lower towns. In view of the very low margins of error, such a double repetition can only have been deliberately designed.

  Many more meeting points between Harappan and Vedic cultures have been proposed right from the 1920s onwards.83 But scholars have proposed widely divergent explanations to account for them.

  At one end of the spectrum, Asko Parpola, a Finnish Sanskritist, has been a strong advocate of the theory that Harappans were Dravidian speakers (a theory first propounded by Marshall). His well-known work on the Indus script is essentially based on Dravidian readings; if it has not gained acceptance, it is mostly because it could make no headway beyond a few seals. But it is fascinating to note how, in order to explain Harappan motifs, Parpola takes constant recourse to Vedic and classical Hindu, Buddhist and Jain concepts and themes. For example, he parallels Kalibangan’s row of seven fire altars with the seven altars within the mahāvedi of the Vedic Soma sacrifice (Fig. 9.4).84 He gives a fairly central place to the ‘Dravidian’ god Murugan, but correlates him with Harappan symbols through the Vedic god Rudra and other elements of Sanskrit literature. His reading of the iconography of the ‘Divine Adoration’ seal (Fig. 10.13) draws on the Brāhmanas, Purānas, the two epics and a traditional sacrifice to Durga; it weaves an entire symbolism around the constellation
of Rohini (which holds an important place in early Vedic literature) and sees in the seven standing figures at the bottom of the seals the ‘seven mothers’ or else the seven wives of the rishis and, therefore, the Pleiades (Krittikā). None of this is remotely ‘Dravidian’!

  We saw earlier how I. Mahadevan, who has also proposed Dravidian readings of a few seals, interpreted the ritual stand placed in front of the unicorn as a sacred filter for Soma—the extraction, filtering and consumption of which form a core theme of the Rig Veda.

  How do such scholars reconcile the apparent contradiction between a supposedly Dravidian-speaking population and important Vedic cultural themes? They do so simply by assuming that all these components of Harappan religion and mythology were somehow ‘borrowed’ from the Late Harappans by the freshly arrived Aryans, who then integrated them in their Vedic literature.