In Search of the First Civilizations Read online

  But in the East, and also in the Americas, this split in the psyche of humanity did not take place. There the city and the natural environment remained part of an integrated cosmological structure, which the Chinese for example were able to maintain as a meaningful basis of government right up to the nineteenth century. In the West, the city was an artefact of civilization, a place where laws and institutions sharply distinguished man’s identity from that of primeval nature. And now, under the pressure of sheer necessity in a colonial and imperialist age, the rest of the world is trying to catch up with the revolution of five thousand years ago.

  TWO

  INDIA

  EMPIRE OF THE SPIRIT

  IN 1856 THE British Engineers John and William Brunton were engaged in building the East Indian railway from Karachi to Lahore in what is now Pakistan. John Brunton has a curious niche in the history of archaeology. Only two years previously, while constructing army field hospitals in the Dardanelles during the Crimean War, he had employed troops to dig into the mound of Hisarlik: the first excavation on the site of Troy. Now, empire building still further from home on the eve of the Indian ‘Mutiny’, he had other concerns. The main problems for the railway engineers in the flood plain of the river Indus were how to lay adequate foundations for the track, and where to find ballast. In the south, in Sind, they had hit on an ideal solution: plundering millions of kiln-fired bricks from a great ruined medieval city near Hala. In the north, between Multan and Lahore, the planned line ran close to the mound of another deserted city, known by the name of the village still clinging to its ruins: Harappa. Defended by massive brick fortifications, the citadel stood over the dry bed of an old confluence of the Ravi river; now five miles off, the river was still a great Hindu pilgrimage site in Brunton’s day and would remain so right up to partition in 1947. On top of the citadel a Muslim cemetery, a small brick mosque from Moghul times and the shrine of a local Muslim saint co-existed with an old Hindu shrine to the god Siva with lingam and yoni stones (a hint, this, of the possible antiquity of the site, if the Bruntons had paused to think; for the Muslims had conquered this region in the eighth century AD!) Nevertheless, with that hard-headed gusto typical of Victorian pioneers, the engineers set to work, pulling down the citadel walls to lay the foundations of hundreds of miles of track, over which trains still rumble today.

  During the demolition and excavation, the Bruntons’ workmen turned up numerous antiquities, including steatite seals engraved with an unknown system of writing, and bearing strange figures of humans, trees and animals, especially bulls. These were shown to a visiting British general who, as chance would have it, would become the director of the Archaeological Survey of India when it was set up in 1861. Alexander Cunningham realized immediately that these finds were quite outside the range of Indian antiquities then known to historians. At that time, Indian history only began with the Mauryan Empire of the late fourth century BC, that is, contemporary with the Hellenistic successors of Alexander the Great. Harappa perplexed Cunningham, and he published some of the seals in the 1870s. But it would be half a century before the mystery of the origin of the seals was solved. In fact, they came from one of the key sites in a great Bronze Age urban civilization whose roots went back thousands of years before historical record; a site whose ancient name had been preserved orally by the people who lived and worshipped on the hill top since the city died in the middle of the second millennium BC.

  THE REDISCOVERY OF INDIA’S PREHISTORY

  Until the 1920s, the prehistory of India remained a blank. Unlike China, there was no real historiographical tradition in India until the advent of Islam, so there were no written records detailing the chronology of early dynasties or events in a way useful to historians. There were epic poems, like the Mahabharata and the Ramayana, which harked back to an apparently legendary heroic age, and there were the sacred texts. These sacred books were held by Hindus to have been handed down orally from time immemorial. They had been written down in the Middle Ages in the ‘sacred speech’, Sanskrit, whose demotic form is the ancestor of most of today’s North Indian dialects. Initially, the priestly caste, the Brahmins, who alone memorized and performed the rituals in these texts, were unwilling to let them be seen and copied by non-Brahmins; for them this was sacrilege in itself. (Similarly, Julius Caesar records that the British Druids believed it ‘against their principles to commit their doctrines to writing’.) This taboo has been uniquely long-lasting in India. Even in our own time there have been extraordinary cases such as that of an illiterate Brahmin priest who appeared in Benares and dictated a very long religious work in Sanskrit verse, till then unknown and unrecorded, which on internal evidence of style and language was medieval, or even earlier, and which had been passed down orally through a certain line of priests.

  Sanskrit began to be known in the West in the seventeenth century, beginning with Roberto de Nobilis and his fascinating attempt to synthesize Hindu and Christian metaphysics in Madurai after 1605. In the eighteenth century, at the same time as the Europeans were becoming heavily embroiled in commercial and military ventures in India, some scholars of the Enlightenment began to be interested in the gods and rituals of India, and to speculate that Hinduism had perhaps been the primeval religion, earlier than any of the monotheisms of the Near East. It was in this intellectual climate in 1786, that a British judge and polymath, Sir William Jones, announced to the Asiatick Society in Calcutta his famous discovery that Sanskrit was related to Latin and Greek through some parent language:

  The Sanskrit language, whatever its antiquity, is more perfect than the Greek, more copious than the Latin and more exquisitely refined than either; yet bringing to both of them a stronger affinity than could have been produced by accident; so strong that no philologer could examine all three without believing them to have sprung from some common source, which perhaps no longer exists.

  Among the striking parallels Jones made was the identity of the word for father in Greek (pater), Latin (pater) and in Sanskrit (pitar), to which the Germanic fadar and modern English father were obviously related. Jones himself (whose languages included Welsh and Persian) suggested that the Celtic and Germanic languages were another stem of this ‘Indo-European’ tree. Remarkable correspondences were subsequently noted in more obscure European tongues: for example, the Sanskrit word for horse, asva, is closely matched in Lithuanian by aszwa. But if the languages had a common root, did they have a common geographical origin?

  Since Jones’ day the question of the ‘common source’ of the Indo-European languages has become both clearer and more complex, and is still hotly disputed by philologists and archaeologists. The majority, however, believe that Sanskrit was not indigenous to India, but originated in a wider Indo-European heartland in south Russia or eastern Europe, from where Sanskrit-speaking peoples migrated into north-west India during the second millennium BC. The track of their migration, indeed, can be recovered. In the fourteenth century BC a dynasty of Indo-European speakers was established in the kingdom of the Mitanni in Upper Mesopotamia. The names of some of their gods are recorded on contemporary clay tablets in forms virtually identical to the Sanskrit: Mitra, Varuna and Indra. Likewise their terminology for chariotry – the classic Indo-European form of warfare – is exactly matched in Sanskrit. At just the same time, around 1400 BC, in north-eastern Iran, on the banks of the Oxus, the prophet Zoroaster had his great revelation, recorded in the Gathas in archaic Persian which has many affinities with early Sanskrit. At that time then, these three branches of Indo-European had not long diverged, though no doubt Sanskrit speakers were by then well established in the area of the North-West frontier and the Hindu Kush.

  So the Sanskrit-speaking Indo-Europeans came into India from the north-west, establishing themselves around the Kabul river and down into the Punjab, presumably over a long period in the second millennium BC. And in fact the oldest Sanskrit texts enable us to add to this picture. The Rig Veda is a series of 1028 hymns and chants which were probab
ly composed over several centuries, between 1500 and 900 BC, though the earliest could just possibly be earlier still. The Rig Veda is still sung in Hindu temples today – indeed it is a key part of the coming-of-age ritual of any young Brahmin – and though only committed to writing in the Middle Ages, it has preserved with astonishing accuracy the linguistic forms of the Bronze Age. The earlier verses describe the Sanskrit speakers settled in north-west India, with no hint of migration from outside, so they had clearly been there for some time. They call themselves Aryans, literally ‘noble ones’ (an Indo-European word which survives in the names of Iran and Eire). Their lands stretch from the Kabul river to the Oxus, along the North-West frontier. Later hymns in the series suggest that by then they had spread into the ‘Seven Rivers’, today’s Punjab. Only in the last of the series is the scene set in the historic heartland of Indian civilization, the valleys of the Ganges and the Jumna. So the Rig Veda implies a sequence spanning many centuries, with the expansion of a Sanskrit-speaking aristocracy south-eastwards from the Oxus to the Ganges. But it also shows that the Aryans were not the only people in north-west India in the late Bronze Age, for it often speaks contemptuously of Dasas, Dasyus, or Panias, meaning ‘dark-skinned ones’ or ‘slaves’ (recalling the way the early English named their British neighbours weallas – Welsh – literally ‘slaves’). Evidently the Dasas were an earlier indigenous people; they were ‘rich in cattle’, they worked gold, they had forts and towns, but were distinguished from the Aryans by speech, looks, and by the colour of their skin. To the evident distaste of the Aryans, they were also sisna-devi, ‘phallus worshippers.’ The Rig Veda implies a long period of interaction between the Aryans and the Dasas, at times peaceful, at times warlike, with the destruction of towns and forts. But who were these earlier people?

  Such is the tale revealed by the earliest Sanskrit source. And such was the state of knowledge until 1921 when the first scientific excavations began in Sind and Punjab, in the lands of the ‘Seven Rivers’ of the Rig Veda. Until that point it was generally believed that Indian classical culture, and the Hindu religion, were Aryan in origin: a belief no doubt strongly coloured by the pervasive influence of Western theories about the racial and cultural superiority of the Indo-Europeans. Until then it was simply unbelievable that the Aryan civilization had been preceded by a much earlier one, contemporary with those of ancient Iraq and Egypt. But this is precisely what was discovered in 1921–2, pushing back the frontiers of India’s history by several thousand years, and making it the oldest known living civilization.

  THE LOST CITIES OF THE INDUS

  In 1920 excavations started at Harappa under an Indian archaeologist, Daya Ram Sahni, pursuing the clues left by Alexander Cunningham. Despite the damage caused by the Bruntons’ depredations, the very first trial diggings produced more seals like those picked up in the 1850s. The upper levels of the site showed occupation during the Mauryan Empire (fourth to third century BC). The deeper levels were obviously prehistoric, but from how far back? The answer came with dramatic swiftness.

  The next year, R. D. Banerjee began an excavation at a huge site to the south in Sind: Mohenjo-Daro, the ‘Mound of the Dead’. Crowned by a brick stupa from the first century AD, Mohenjo-Daro was apparently a Buddhist ruin, and Buddhist remains were what Banerjee was expecting to find. But almost at once he found evidence of the same prehistoric culture uncovered the previous year at Harappa. The citadel stood on a tract of land called locally ‘the Island,’ and had been eroded by huge flood channels of the Indus which had gouged a path between it and the eastern suburbs of the ancient city which extended for about a square mile. The finds were among the most spectacular from the ancient world.

  About 450 yards long, and probably fortified, the citadel contained an impressive series of ceremonial or governmental buildings including a large columned hall, a huge granary and, most striking of all, a colonnaded tank with a brick-lined bath 40 feet long, 23 feet broad and 8 feet deep which strongly recalled the ritual bathing tanks still seen all over India. Connected to the bath was a large ‘college’ (as the excavators called it). The temple (if there was one) presumably lay under the Buddhist stupa on the summit of the hill, but this was not removed. In the suburbs were wide streets with grand mansions; there were wells and public latrines for every block; sewers large enough to walk in: this was in some respects a culture more advanced than Egypt or Mesopotamia. And the discovery of seals from Iraq at Mohenjo-Daro dated its heyday to the same period as the flowering of the cities of Sumer – the mid-third millennium BC.

  Some aspects of Mohenjo-Daro were puzzling, especially its redbrick uniformity. ‘Anyone walking through it for the first time,’ wrote the excavator, ‘might fancy himself surrounded by the ruins of some present-day working town in Lancashire.’ There was barely any sign of exterior decoration, little figural sculpture and the grid plan of the streets was monotonous and regular. ‘Stark utilitarianism,’ said the final dig report – in short, everything modern India is not! And yet there were also clear signs of a connection with later Indian life. On the seals were sacred trees and animals, including the cow and the humpbacked bull still to be seen nosing inviolate through any crowded Indian bazaar today. Clay models of the mother goddess were like those still made throughout India. A toy bullock cart matched those seen anywhere on the roads of the Punjab today. Most striking of all was a seal depicting a divine figure seated lotus fashion on a deer throne in a yogic posture with bangled arms and a horned head-dress, and surrounded by wild animals. The excavators also believed the figure to be three-faced and ithyphallic, but the seal is so small that these details are not quite certain. All these attributes, though, are very close to those of the great god Shiva in medieval and modern Hinduism, specifically in his role of ‘Lord of animals’. Recently there have been attempts to discount the connection of this remarkable image with the later Shiva, but they are unconvincing. No iconography stays still over several millennia; nor do symbols necessarily always retain exactly the same meaning, but that this image is in some sense a ‘proto-Shiva’ – perhaps in his incarnation as buffalo-god – seems undeniable.

  So, like Iraq and Egypt, India had indeed produced a great Bronze Age civilization. And like them it had been literate. Its relation to later history, though, remained problematic. The biggest stumbling block was the writing itself; the key to the identification of the Indus people. Examples of the script had survived engraved on hundreds of seals. But none of the inscriptions was more than a handful of characters, which made decipherment extremely difficult. Indeed, the script remains undeciphered today: one of archaeology’s greatest mysteries.

  THE INDUS CIVILIZATION

  In the last few decades a much fuller picture has emerged of the civilization of Mohenjo-Daro and Harappa. More than a thousand settlements are now known, extending from Afghanistan to Delhi, and from the Himalayan foothills down to Bombay: an area the size of Western Europe, and much larger than either early Iraq or Egypt. Most remarkable, though, is that over this huge area there seems to have been a unity of culture, art, script and technology (and even of weights and measures). The largest city, Mohenjo-Daro, is now thought probably to have reached eighty thousand in population, with five million for the entire state – if state it was. Like Iraq and Egypt its civilization depended on river irrigation. The Indus floods every year, inundating the plain, and on the alluvium they grew wheat, barley, rice and cotton, which seems to have been cultivated in Sind for the first time in history (hence its name in Greek – sindon). The Indus civilization also depended on long-distance trade, shipping cotton, hardwoods, ivory and precious stones to Sumerian cities like Ur, where there was a colony of Indus merchants in the third millennium BC. Indus seals found in Sumer often carry the marks of the bales of merchandise to which they were affixed. This trade with the Persian Gulf region is one of the oldest international connections in Indian culture. A Greek merchant’s manual from around 60 AD describes sandalwood, ebony, cotton, spices and pearls c
oming to the Gulf from Gujerat. This same produce was the mainstay of the trade between Basra and India in the Ottoman and British periods, and the Indian trade with Iraq remains unbroken today.

  THE DEEPER ROOTS: NEW EXCAVATIONS IN BALUCHISTAN

  Our understanding of the origins of the Indus cities has been transformed in the last thirty years by excavations at Mehrgarh near Quetta in north-west Pakistan. Here, the story of settled continuous occupation in the Indus region has been taken back to the seventh millennium BC, four thousand years before the flowering of the Harappan age; back into the same period when agricultural communities were forming across the Near East from Palestine to Iran. As late as the 1970s there was no evidence of agriculture in India much before 3000 BC, underlining what a revolution these finds have brought. These discoveries reveal the great antiquity of not only the farming economy in the Indus valley, but also of craft specialization (including steatite cutting), and of long-distance trade (in turquoise and lapis). In the fifth millennium BC builders at Mehrgarh used the long planoconvex brick we find later in the Indus cities; and cotton was already cultivated. In the fourth millennium BC the little town was part of a wide common cultural zone extending into Iran. In about 2500 BC it was abandoned, superseded by a large town five miles away at Naushahro which had massive brick fortifications and impressive buildings including a possible temple. By then we are into the Harappan age proper.

  These new discoveries make it absolutely certain that the Indus civilization was an Indian phenomenon, growing out of the native prehistoric cultures of its region, and not, as was thought in some quarters, stimulated by diffusion of cultural ideas from Iraq. But who were the people of the Indus civilization, if they were not Aryan Indo-Europeans? Were they the Dasas of the Rig Veda? A new picture has begun to emerge, only since the 1980s. The crucial discovery has been the proving beyond any doubt, that the ancient pre-Aryan language of Iran, Elamite, is cognate with the ancient Dravidian languages still spoken in Southern India, best known of which is Tamil. These languages descend from a prehistoric speech spoken in Iran and north-western India, and doubtless in the early villages like Mehrgarh: indeed a pocket of a related language, Brahui, is still spoken today in a small area of West Pakistan on the Iranian border. The original proto-language split up around 5000 BC, at a guess, after the invention of agriculture, to judge by its common terminology in Elamite and Dravidian. The Elamite branch was spoken in the early urban societies in Iran contemporary with Sumer around 3000 BC, and ceased to be spoken in the tenth century AD. Dravidian is still spoken now by over 200 million South Indians.