3
The traditional view locates the discovery of agriculture in the Middle East before it spread to other parts of the world. There is some evidence that this may have been generally true for Europe as it was repopulated after the Ice Age from the south and east. However, this unidirectional narrative is hardly applicable to the rest of the world and should be seen as a lingering vestige of the biblical world view that all civilization began in the Middle East. We now know that farming emerged independently in many parts of the world and that the Indian Ocean rim had multiple clusters. Thus, the story of the Indian Ocean rim is about the evolution of these clusters and their long-distance interactions from a very early stage.
India’s Early Farmers
To be fair, archaeology had initially supported the traditional view since the earliest evidence of farming in the Indian subcontinent had been found in the extreme north-west. Early excavations in Baluchistan, now part of Pakistan, suggested that this was the first place in the subcontinent to witness agriculture-based settlements. Mehrgarh, in the Bolan valley, is the best documented site and may have been occupied before 6000 BC (i.e. 8000 years ago). The site provides a fascinating view of the transition from hunting to domestication of animals. The early layers are dominated by the bones of wild animals like gazelle, spotted deer, sambar, blackbuck, nilgai, wild ass, and even elephants. Notice that some of these animals would not thrive in the wild in the dry climatic conditions of present-day Baluchistan. In the later layers, the mix gradually switches to cattle, goat and sheep.1
Barley was the earliest crop at the site. Baluchistan would have been within the natural habitat zone of wild barley during that period. So, it is quite possible that at least one variety of barley was domesticated here.2 A bit later we find that the inhabitants of Mehrgarh also began to grow wheat. Remember that Baluchistan is on the Indo-Iranian continuum, and people and trade were passing back and forth. Therefore, it is likely that wheat was acquired from the Middle East at an early stage.
More recent archaeological finds, however, suggest that farming appeared more or less simultaneously in a number of other clusters scattered across northern India. For instance, one concentration of Neolithic farm settlements have been found along the fringes of the Vindhya range in central India, just south of the modern city of Allahabad. As many as forty sites have been identified from the same period as the ones in Baluchistan. Interestingly, the central Indians were eating both wild and domesticated rice. It is currently believed that rice was domesticated in China and it is possible that the crop made its way to India via South East Asia. However, wild rice is still found in the general area, which could mean that it was independently domesticated in India.
We also find animal bones in the central Indian sites including cattle, deer, goats, wild boar and, hold your breath, horses.3 This fits in with what we know of shifting climate zones and the discovery of horses painted on rocks at Bhimbetka, a Stone Age site further to the south in Madhya Pradesh. The point is that Indians appear to have been familiar with horses from a very early age which goes against the common view that the animal was domesticated in Central Asia and came to India in the Iron Age. New Neolithic sites are still being found and excavated, so our knowledge of this period is fluid. As pointed out in the previous chapter, there may have been even older settlements along the coast that are now under the sea, especially off the Gujarat coast.
Interestingly, agriculture appears to have spread to southern India much later. What was the reason behind this? As we have seen, hunter–gatherers were not always impressed by farming. It is possible that climatic conditions in the south allowed them to continue with their existing lifestyle. Equally interesting is the distribution of the farming communities when they did come up. We find them concentrated along the Krishna River and, its tributary, the Tungabhadra. In contrast, there is very little evidence of early farming settlements along the south-east (i.e. Tamil and Andhra) coast. This is puzzling as the area shows evidence of human activity from much earlier periods.4 Perhaps the Neolithic sites were washed away by local rivers and covered with silt. Perhaps the settlements existed near the coast and were submerged by changing coastlines that ate into large tracts of land off the Tamil coast. My own guess is that faced with a choice between big game fishing and watching vegetables grow, the lads made the obvious choice.
The Rivers of Civilization
After a relatively benign period lasting several thousand years, the savannah grasslands that covered the Sahara and Arabia began to dry out again around 4500 BC. Populations that lived in the encroaching desert began to converge along the Nile in Egypt and the Euphrates and Tigris in Mesopotamia, both of which already had established farming communities. The influx led to a sharp increase in population but, unlike the arid period during the Ice Age, the major rivers continued to flow in full force. North-western parts of the Indian subcontinent also became drier while witnessing increased agricultural activity.
The need to manage increased pressure on the land was what possibly led to the emergence of complex political—alongside social and cultural—institutions in all these regions.5 In other words, these hubs of riverine civilizations were born out of adversity and culminated in the creation of the first kingdoms/states. In Mesopotamia, we see the rise of Sumerian city states. By around 3100 BC, Upper and Lower Egypt are unified into a single kingdom with its capital in Memphis.
Meanwhile, the Indian subcontinent witnessed the growth of settlements along two major rivers and their tributaries. One of the rivers is the Indus and the other is now the dry riverbed of the Ghaggar. Satellite photos and ground surveys confirm that the Ghaggar was once a mighty river that emerged from the Himalayas near modern-day Chandigarh, then flowed through Haryana, Rajasthan and Sindh before entering the sea through the Rann of Kutchh in Gujarat. With the Sutlej and the Yamuna as its tributaries, the river initially would have had water flow comparable to that of the Indus.
After decades of debate, it is now accepted by most serious scholars that the Ghaggar is the same river that the earliest Hindu texts refer to as the Saraswati.6 The importance of the river can be gauged from the fact that there are far more settlements clustered around the course of the Saraswati than along the Indus. This is why this civilization is now called the Indus–Saraswati civilization rather than the Indus Valley civilization. It is also known as the Harappan civilization after Harappa, one of its largest cities.
The Harappan civilization went through three phases. The earliest recognizable Harappan site at Bhirrana in Haryana, on the banks of the Saraswati–Ghaggar, has been carbon-dated to 7000 BC.7 This makes it at least as old as the sites in Baluchistan which were once considered the oldest in the subcontinent. The early evolution of the settlements in the area is still being analysed and is not fully understood but the earliest level coincided with a period which enjoyed an increase in monsoon rains. However, from around 5000 BC, the monsoons begin to gradually weaken (although they were still much stronger than today).8 As in Egypt and Mesopotamia, this coincided with densification of settlements along rivers. We know that by 3200 BC, at about the same time that Egypt was being unified, there are a large number of Harappan settlements on both the Saraswati and the Indus basins. This ‘early’ phase lasted till about 2600 BC.
The second phase, often dubbed the ‘mature Harappan period’, lasted from 2600 to 2000 BC. This is the period that saw the rise of major cities like Mohenjodaro, Harappa, Dholavira, Kalibangan and so on. Some of these settlements already existed in the previous phase but they now expanded by an order of magnitude. Recent excavations suggest that the largest of these cities was Rakhigarhi, Haryana, which is also in the Saraswati–Ghaggar basin. After 2000 BC, however, the archaeological evidence shows a steady decline—cities are abandoned, civic management deteriorates and there are signs of economic stress. Some of the settlements struggle on but the ‘late’ Harappan period peters out by 1400 BC. This is a simplified timeline and individual sites would have experienced somewhat different cycles.
The Harappans did not build great monuments like the pyramids but they outmatched their Egyptian and Sumerian peers in terms of population size, the sophistication of their cities and the sheer geographical reach of their civilization. At its height, there were Harappan settlements from Punjab in the north to Gujarat in the south, and from Baluchistan in the west to what is now western Uttar Pradesh in the east. We have even found outposts like Shortughai on the Afghanistan–Tajikistan border and Sutkagen-dor near Pakistan’s border with Iran, not far from modern Gwadar.
The Maritime Hub of Dholavira
Given the maritime orientation of this book, we will focus here on the Harappan sites in the Indian state of Gujarat. To understand the context of the numerous archaeological discoveries in this area, let’s begin with the landscape in which the Harappans built their settlements. First, western India was much wetter than it is today. Not only was monsoon rain stronger, the Rann of Kutchh received fresh water from both the Saraswati and the Indus. The estuary of the Indus was much further east than it is today and one of its major channels flowed into Kutchh. In fact, the Indus used to flow into Kutchh till as recently as the colonial period when a major earthquake in 1819 diverted the river. The fortress of the semi-abandoned town of Lakhpat still stands guard over the channel through which the Indus used to enter the Arabian Sea. Second, the relative sea level during Harappan times was several metres higher than it is today which meant that the Saurashtra peninsula was an island. Thus, ships could comfortably sail through what are now the salt flats and marshes of the Rann of Kutchh and then make their way out to the Gulf of Khambhat.9
Dholavira may look today like it is too far inland to be an effective port but, as shown in the map, it was built on a strategically located island in the third millennium BC. It was accessible by boat from the Arabian Sea to the west as well as the Gulf of Khambhat to the south. Boats from Dholavira would have also been able to sail up the Indus and, at least initially, the Saraswati to the cities that were emerging along their banks. In other words, Dholavira would have served as a very important commercial, and possibly military, node.
There is evidence that by 2600 BC, the Saraswati began to dry up. We do not yet understand the exact factors that caused this, but tectonic shifts in northern India may have caused the Sutlej to shift to the Indus and the Yamuna to the Ganga. There is debate about exactly when the two tributaries shifted but it is reasonably certain that it happened well before the great cities of the mature period were built. This would have deprived the river of two important sources of perennial glacial water. Still, rainfall was quite heavy at first and a rain-fed but diminished Saraswati would have remained a significant river although, as time passed, it was no longer navigable all the way to Dholavira.
Interestingly, the mature period of the Harappan civilization seems to take off at a time that the Saraswati may have already started its decline. We see a sharp increase in villages and sophisticated urban settlements along both the Indus and the Saraswati during this time. It is unclear why the Harappans invested in building so many cities along a dying river. If we do not manage today’s rivers sensibly, it is conceivable that future archaeologists will dig up the remains of twenty-first-century cities and wonder the same thing.
Meanwhile, the urban cluster at Dholavira expanded significantly. The site had a fortified acropolis and a ‘lower town’. At some point the city was expanded to accommodate the growing population and the old lower town became the ‘middle town’ and the expanded area became the new lower town. A wooden signboard has been found near one of the gateways. We do not know what it says as the script has not been deciphered but my guess is that it’s something mundane like: ‘Keep Left for Bullock-Cart Parking’.
We see a proliferation of settlements in Gujarat during the mature period, in Kutchh as well as the island of Saurashtra. Of these sites, Lothal is one of the best known because of the discovery of a large dockyard which used sluice gates and a spill channel to regulate water levels. Next to the dockyard, there are remains of structures that may have been warehouses and a series of brick platforms where one can imagine stern customs officials inspecting the goods and unscrupulous merchants trying to bribe them.
Lothal is several kilometres away from the sea now and modern-day visitors will be surprised to know that it was once possible to sail from Dholavira to Lothal. In fact, it is quite possible that Lothal was a customs checkpoint for shipping headed north for Dholavira through the Gulf of Khambhat. There may have been a similar checkpoint for ships coming in from the west, perhaps at Bet Dwarka, where a number of Harappan-era anchors have been found.
In addition to internal trade, there is plenty of evidence that the Harappans had strong economic links with the Middle East. The merchant ships from Gujarat made their way along the Makran coast, trading along the way. They may have stopped near Gwadar to visit their outpost at Sutkagen-dor (near what is now the Pakistan–Iran border). A bit further west, they would have interacted with the people of the Jiroft civilization. This civilization has been recently discovered in south-eastern Iran. Although we know very little for sure about its people, archaeologists have found seals like those of the Harappans and signs of close cultural links.10 In particular, several depictions of humped zebu cattle have been found in Bronze Age sites in southern Iran.11 This is interesting because humped cattle have their origins in India and are common in Harappan iconography. In other words, the coastline from Gujarat to southern Iran was still a well-populated continuum with strong economic and cultural links. Were these links a result of Bronze Age trade or did the Jiroft people of southern Iran and the Harappans share a deeper ancestry going back to the Stone Age?
Some of the Harappans sailing to Iran would soon venture across the narrow strait to Oman. Archaeologists digging at Ras al-Junayz, on the eastern most tip of the Arabian peninsula, found that over 20 per cent of objects were of Harappan origin. There are a number of enigmatic beehive ‘tombs’ from this period scattered across Oman. Most of them are in areas that are too arid to sustain a population today but were much wetter in the Bronze Age. The builders of these structures would have almost certainly interacted with visiting Indian merchants. Many of these merchants would then have made their way further into the Persian Gulf and sailed to Bahrain where plenty of Harappan seals, pottery and beads have been found.12
Further west, Harappan-origin artefacts have also been found in ancient Mesopotamian cities like Kish, Nippur and Ur. The records of Akkadian king Sargon I (2334–2279 BC) refer to ships from Dilmun, Magan and Meluhha (i.e. Bahrain, Oman and the ports of Gujarat/Sindh). However, note that the maritime route was not the only link between India and Mesopotamia. There were also land routes that made their way from the northern Harappan cities to Mesopotamia through Afghanistan and central Iran. There was also one that ran through Baluchistan and southern Iran along the Makran coast.
Trade with India had a big influence on the Persian Gulf area. For instance, Harappan weights and measures became the standard across the region. The locals also copied the Harappan seals. This was the beginning of a long commercial and cultural relationship that, despite booms and busts, continues to this day. Till as recently as the 1960s, the Indian rupee was used as legal tender in Oman, Qatar, Bahrain and the UAE! For a while, the Reserve Bank of India even issued a special Gulf rupee for use in these countries. It was only when the Indian rupee sharply devalued in June 1966 that these countries began to issue their own currencies (Bahrain had already made the shift a few years earlier).13
Today the most visible example of these links are the large numbers of Indians who live and work in the Gulf countries. This too has its origins in Harappan times. Mesopotamian inscriptions mention that the Meluhhans were numerous enough to have their own ‘villages’ or exclusive enclaves in and around Sumerian towns. We do not know for sure what these Indians were doing there— they could have been a mix of merchants, artisans and mercenaries—but they seem to have been an important part of the bustling economy of Bronze Age Mesopotamia. We also have a handful of references to individuals. For instance, there is a cylinder seal belonging to a Meluhhan interpreter called Su-ilisu (of course, he may just have known a Harappan language and may not have been an Indian). Amusingly, we also know about a rowdy Meluhhan who was made to pay ten silver coins to someone called Urur as compensation for breaking his tooth in a brawl!14
We have some idea of what the Harappans exported—carnelian beads, weights and measures, different types of wood, pots of ghee (clarified butter) and, most importantly, cotton textiles. The cotton plant was domesticated in India and cotton textiles would remain a major export throughout history. Oddly, we are not sure what the Harappans imported in exchange. Nothing of obvious Persian Gulf origin has ever been found in any Harappan site. Perhaps they imported perishables like dates and wine. Another possibility is that they imported copper from Oman as the remains of several ancient copper mines have been found there. The Harappans had their own copper sources such as Khetri near the Haryana–Rajasthan border, but the Gujarati Harappans may have preferred Omani copper.
Did the Harappans Compose the Vedas?
Despite the abundance of archaeological remains, we know little about the Harappans themselves. We are not sure what languages they spoke, what gods they worshipped, and their script remains stubbornly undecipherable. We do not even know if it was a unified empire or a network of independent city states that shared a civilization. These matters remain hot topics of acrimonious debate among academic historians.
The most obvious place to look for clues is the Rig Veda. It is the oldest and among the most sacred of Hindu scriptures. Composed in a very archaic form of Sanskrit, the Rig Veda was compiled over several generations by rishis and poet-philosophers, and remains in active use today. It contains ten ‘books’ of hymns and chants in praise of the gods. Although the text is mostly concerned with religious practice and philosophy, one can discern some things about the social and geographical context of those who composed the hymns.
The traditional view was that the Rig Veda was composed by so-called Aryans who came to India from Central Asia around 1500 BC. The problem is that the date is entirely arbitrary and there is no archaeological or genetic sign of a large-scale invasion/migration. The Rig Veda itself mentions no invasion/migration and suggests no knowledge of Central Asia. It is possible that its composers were aware of Central Asia and southern India, but the text does not mention it. Instead, its geographical horizons are mostly concerned with an area that it calls the Sapta Sindhu or Land of the Seven Rivers.
As discussed more fully in my previous book, Land of the Seven Rivers: A Brief History of India’s Geography, the Sapta Sindhu was a relatively small area covering the modern-day state of Haryana and a few adjoining parts of Punjab and Rajasthan. This was the original homeland of the Bharata tribe that, according to the Rig Veda, defeated an alliance of ten tribes on the banks of the Ravi in Punjab. They then expanded their empire to the east by defeating a chieftain along the Yamuna. Thus, the Bharatas created the first known empire in the subcontinent and gave Indians the name by which they still call themselves. The text also suggests knowledge of a wider area including the Himalayas in the north, the Ganga on the east, the sea to the south and the rivers of what is now Pakistan’s North-West Frontier Province. This roughly coincides with the geographical footprint of the Harappan civilization. So we have the remains of a large civilization and an extensive body of literature occupying the same geographical space during the Bronze Age. Surely it is not unreasonable to say that they belonged to the same people.
Given the genetic evidence discussed in the previous chapter, it would appear that most of the Harappan people were derived from the ANI gene pool who had coalesced in northern India around the end of the last Ice Age. However, given the large geographical spread and extensive trade links, it is likely that the civilization was multi-ethnic. The Rig Vedic people were part of the broader Harappan milieu. The archaeological evidence suggests this as well. For instance, sacred fire altars have been identified at a number of places like Kalibangan whereas no large religious building has ever been found.15 This is exactly what one should expect since we know that Vedic-era Hindus conducted elaborate fire sacrifices but did not build temples or worship idols.
There is an additional piece of evidence that one needs to consider. The Rig Veda repeatedly mentions the Saraswati River as the greatest of rivers.16 It is clearly the most important geographical feature of the Vedic terrain. Forty-five hymns are dedicated to the river while the Ganga is barely mentioned twice. One of the hymns clearly places the river between the Yamuna and Sutlej—exactly where the dry riverbed of the Ghaggar is located. Importantly, the hymns describe a river in full flow and, unlike later texts, there is no mention of the river drying up. This would suggest that the text was certainly written before 2000 BC and most likely before 2600 BC—which would imply that we may be dealing with an early Harappan text. I am aware that some scholars will disagree but this is the simplest explanation from the available evidence. All other accounts inevitably require complicated explanations and implausible contortions. New information about this period is still flowing in and a clearer picture should emerge over the next decade.
The Churning of Genes
The great Harappan cities flourished till around 2000 BC when we see a sudden deterioration in economic and social conditions. It has now been confirmed by a series of studies that this was due to another shift in climate that also seems to have affected other Bronze Age civilizations. A study of an old lakebed in Haryana by scientists from Cambridge University found conclusive evidence that the summer monsoon abruptly became weaker 4100 years ago in north-west India.17
Note that this change in weather patterns in India was not an isolated incident as it coincides with similar shifts in Egypt and Mesopotamia where it caused great economic and political disruption. Iran’s Jiroft civilization died out and was forgotten. Egypt’s Old Kingdom collapsed and the country went back to being divided into Upper Egypt and Lower Egypt. In Mesopotamia, the Akkadian empire also collapsed around 2100 BC. A lamentation called ‘The Curse of Akkad’ describes what happened:18
For the first time since cities were founded,
The great farmlands produced no grain,
The rivers produced no fish,
The irrigated orchards produced neither syrup nor wine,
The gathered clouds did not rain, the masgurum did not grow.
At that time, one shekel’s worth of oil was only one-half quart,
One shekel’s worth of grain was only one-half quart. . . .
These sold at such prices in the markets of all the cities!
He who slept on the roof, died on the roof,
He who slept in the house, had no burial,
People were flailing at themselves from hunger.
North-west India probably experienced something similar as monsoon rains failed. The Saraswati, already in decline, seems to have stopped flowing altogether. Post-Vedic texts tell us of how the river ‘disappeared’ underground. There is evidence that the Harappans tried to adapt to drier conditions by switching from high-yield wheat and barley to drought-resistant millets.19 The problem was that the new crops had yields that could only support small rural communities but not large urban centres. The great Harappan cities were no longer tenable and were abandoned one by one as people migrated in search of water. Some groups would have shifted east into the Himalayan foothills and into the Gangetic plains. Those from Gujarat seem to have drifted south to the Narmada and Tapti valleys. Others would have drifted further afield.
These migrations show up clearly in the genetic records. The ANI and ASI populations suddenly go through a period of rapid mixing from around 2100 BC onward.20 It is possible that this mixing was quickened by some ASI groups moving north at the same time that the ANI were moving south and east. The mixing of these two genetic pools is responsible for the bulk of India’s present-day population. Of course, India is a large and diverse country and there are many groups that do not fit into this simple ANI–ASI framework but the coming together of these two populations is a very important event in the history of India and effectively triggers Indian civilization as we know it.
Genetic markers also suggest that this mixing went on for more than two thousand years, so much so that there are no ‘pure’ ANI and ASI any more. As a recent study put it: ‘The most remarkable aspect of the ANI–ASI mixture is how pervasive it was, in the sense that it has left its mark in nearly every group in India. It has affected not just traditionally upper-caste groups, but also traditionally lower- caste and isolated tribes, all of whom are united in their history of mixture in the past few thousand years.’21 In other words, after all this blending, the majority of Indians are most closely related to each other irrespective of their ancient origins. Sorry if this scientific finding offends any ‘pure race’ advocates!
Then, around AD 100 (a new study suggests AD 50022) the mixing abruptly stopped as different castes and tribes became strictly endogamous. The reasons for this are hazy, but castes do seem to be quite fluid in the oldest Indian texts and become much more rigid in later writings (although endogamous, the relative positions of most castes continued to be fluid into modern times). A fuller discussion on caste is beyond the scope of this book, but genetics has broadly confirmed the assertion by Dr Bhim Rao Ambedkar that the Indian caste system was a case of ‘superimposition of endogamy on exogamy’.23
The migration of Harappans to the east and south would have spread their technologies and culture to these areas. Mainstream historians seem to assume that this implies a unidirectional flow of cultural and technological influence from the north-west of India to the east and south and then onward to South East Asia. This assumption is deep-rooted in Indian history writing but ends up giving a completely wrong impression. The reality is that there were already established populations and possibly even cities in the areas where the Harappans settled (a preliminary study hints that Varanasi may be as old as the Harappan cities).24 Thus, plenty of influence flowed in reverse and Indian civilization is the result of a messy process where people, ideas and influences flowed in multiple directions—a bubbling mix and not a steam engine running on fixed rails. The people living in southern India may not have built sophisticated cities in the Bronze Age but it is they who initiated the Iron Age.
While northern Indians were building the great Harappan cities, southern Indians had continued to live in Neolithic settlements or as hunter–gatherer communities. The Bronze Age largely bypassed southern India, perhaps due to the paucity of copper ores in the region. Then, late in the third millennium BC, they did something amazing—they invented iron technology! The traditional view is that iron was introduced to India by invaders from Central Asia. Archaeological finds over the last two decades suggest instead that India was the likeliest place where iron was first mass produced.
Initially, the evidence pointed to the earliest use of the metal around the eighteenth century BC in the middle Gangetic plains but now it appears that it originated much further south and at a much earlier period. Students of the University of Hyderabad made a startling discovery in 2014–15 while doing excavations in their campus.25 They found a number of iron artefacts, including weapons, that dated from around 2400–1800 BC. This is arguably the oldest systematic use of iron in the world. Far from being a military advantage exercised by Central Asian marauders, it appears that iron weaponry would have been an indigenous technological advantage.
The Indo-Iranians
So, if there was no ‘invasion’ from Central Asia, what explains the close cultural links between Vedic Indians and ancient Iranians? For instance, the oldest texts of Zoroastrianism, the religion of ancient Persia, are composed in a language very similar to Vedic Sanskrit. To this day, the Zoroastrians follow rituals and customs such as the fire sacrifice and the sacred thread that closely resemble Vedic tradition. As discussed in Land of the Seven Rivers, one possibility is that this similarity is derived from some Indian tribes that may have migrated west. Unlike the Vedic texts, ancient Iranians do refer to an ‘original homeland’. They also seemed to have knowledge of the Saraswati River and of Sapta Sindhu. Perhaps the Persians are descendants of the Parsu tribe who were part of the ten-tribe alliance defeated by the Bharatas.
Another possibility is that the cultural links merely reflect the fact that the Iranians and north Indians were part of the same continuum until the Bronze Age. Given their geographical proximity, why does one need Central Asians to facilitate interaction? The location of the Jiroft civilization is particularly intriguing in this context because its sites are in south-east Iran and very close to the western most Harappan sites. Given the evidence of trade and use of zebu cattle, they may have been part of an ethnic continuum extending from north-west India. After 2000 BC, as eastern Iran and Baluchistan became increasingly hot and arid, the Jiroft people would have moved west towards Fars province. The ancient name of Fars is ‘Parsa’ and it is here that the Persians emerged as an identifiable people. Again notice that Fars is in southern Iran and not in the north as would be expected if they were migrants from Central Asia.
The Persians were not the only people with Vedic links in the Middle East. A military elite called the Mitanni migrated from the east into northern Iraq in the middle of the second millennium BC and came to dominate the region. In 1380 BC they entered into a treaty with the Hittites. The agreement was solemnized in the name of Vedic gods Indra, Varuna, Mitra and Nasatya. Perhaps it is no coincidence that the arrival of the Mitanni in the region also witnessed the introduction of a technology of Indian origin—iron. It is noteworthy that this is five centuries after the earliest mass production of iron took place in southern India.
Who were the Mitanni? Were they related to warrior tribes that, according to ancient Indian texts, migrated to the north-west and became ‘mlechhas’ or barbarians? Prior to the arrival of the Mitanni, iron was treated as a precious metal in the Middle East. Tiny quantities of meteorite iron were used to make prestige items in Anatolia (now Turkey) but large-scale extraction from ore was not known.26 It is likely that the Mitanni used iron to carve out their empire in northern Iraq. However, their Hittite rivals soon mastered the technology and the two fought several wars for control of ore-rich areas.27 The wars ended with the Hittites winning and, with the rise of the Assyrians, the Mitanni found themselves crushed between two powers in the thirteenth century BC. Nonetheless, the presence of a Vedic-related people using an Indian technology does suggest that some Indian tribes migrated out of India in the second millennium BC.
Let it be clear that I am not attempting to replace a unidirectional inward migration into India with a unidirectional outward migration. We are dealing with very long periods during which human and cultural movements would have gone in different directions for different reasons at different points of time. It appears that broadly the same genetic and/or cultural pool was sloshing back and forth between northern India and Iran since the Stone Age. At the end of the Ice Age, one branch colonized Central Asia from where a sub-branch migrated to Eastern Europe. Those who remained in Asia would have kept interacting. This sloshing may explain why the mixing of genes in India seems to happen in layers between the same groups.28 Note that such non-linear movements are echoed in the later oral histories of many communities. The Gouda Saraswat Brahmins, for instance, claim to have migrated from the banks of the Saraswati River to Bengal and then later to the western coast of India. Long dismissed as myths, one wonders if these oral histories contain a memory of real population movements.
Interestingly, the Vedic–Mitanni god Mitra would remain a popular deity in the Middle East and, centuries later, would witness a major revival in the Roman empire (where he would be known as the solar god Mithras). The cult of Mithras would become very widespread in the late Roman period and, for a while, would provide serious competition to early Christianity. The pagan Romans used to celebrate a big festival called Saturnalia that went on for a week from 17 December. At the end of the festival, on the 25 December, the Mithras cult would celebrate the feast of Sol Invictus or Unconquered Sun. Many scholars believe that when the Christians came to dominate the Roman empire, they simply took over the popular pagan festival (after all, the actual birth date of Jesus Christ is not known).29
Mind you, not everyone agreed with this choice and the Orthodox Church still celebrates Christmas on 7 January. The Puritans would later disapprove of the unseemly heathen celebrations that clung to the festival and would try to ban Christmas in North America and Britain in the seventeenth and eighteenth centuries; they obviously thought that Merrie Olde England was a bit too merry.30 Nevertheless, the 25 December holiday has survived as a day of festivity for most Christians and even non-Christians. Thus, one of the unintended consequences of early Iron Age migrations seems to be that the world has come to celebrate the birthday of an ancient god from Haryana!
The Ark of the Covenant
The collapse of the major Bronze Age cities around 2000 BC affected the thriving trade route between the Persian Gulf and India’s western coast. By chance, the personal correspondence of a merchant in Ur, now Iraq, has survived in the ruins of his house and provides a direct view of how business networks may have begun to break down. The merchant, a certain Ea-Nasir, seems to have imported copper ingots from Magan (i.e. Oman) via Dilmun (i.e. Bahrain) around 1900 BC. Angry letters from his customers and creditors make for amusing reading after four thousand years although those who wrote the letters were clearly not amused. Here is one example from a business associate called Nanni:31
Now when you had come, you spoke saying this, ‘I will give good ingots to Gimil-Sin.’ This you said to me when you had come. You have offered bad ingots to my messenger saying, ‘If you will take it take it, if you will not take it go away.’ Who am I that you are treating me in this manner, treating me with such contempt, and that between gentlemen such as we are.
Although new empires would later be re-established in Mesopotamia, it would not resurrect the fortunes of Persian Gulf copper merchants as Cyprus would replace Oman as the more important source. The transition to the Iron Age would further reduce the importance of Omani copper. Thus, Oman seems to have become a backwater. Nonetheless, some trade must have continued to flow to the Indian Ocean. We know this due to a most unusual artefact—the discovery of black peppercorns stuffed up the nostrils of the mummy of Pharaoh Ramses II. While we do not know if this condemned the pharaoh to frequent sneezing in the afterlife, it shows that the Indian Ocean trade networks of the thirteenth century BC were capable of transporting pepper from its origin in south-western India to Egypt. Since it is unlikely that sailors of this period could directly cross the Arabian Sea, the spice was probably shipped up the coast from Kerala to Gujarat, and then along the Makran coast to Oman. From here, there were two possibilities. It could have continued along the old Persian Gulf route to Mesopotamia and then overland to Egypt. However, it is just as possible that pepper made its way south from Oman to Yemen and Eritrea, a region that the ancient Egyptians called the Land of Punt.
We know that the ancient Egyptians sailed down the Red Sea on frequent trading expeditions to the Land of Punt. One of the expeditions is depicted on a panel at the memorial temple of Queen Hatshepsut who ruled Egypt between 1473 and 1458 BC. Hatshepsut is a fascinating character—a female pharaoh who dared to rule in her own name. There are other female rulers in Egyptian history but she overshadows them in terms of the scale of her building projects, her military victories and, of course, her ambitious maritime expedition to the Land of Punt.
The temple panels depict large galleys with sails, oars and stern-side paddles for steering. They probably embarked from Wadi Gawasis on Egypt’s Red Sea coast where archaeologists have found coils of rope and cedar planks from that era.32 From here they would have sailed down the Red Sea to what is now Yemen and Eritrea. The expedition would return with gold, ivory, different kinds of wood, exotic animals and, most importantly, frankincense (also perhaps Indian black pepper for stuffing up some mummy’s nostrils).
Frankincense was a very valuable product of ancient Yemen and Oman. It is the dried resin of a thorny tree and gives off a pleasant smell when burned. It was widely used in religious ceremonies and is one of the gifts that the ‘three wise men’ are said to have brought for the baby Jesus. Even today, anyone visiting the Yemen–Oman coast will find it commonly sold in traditional markets and the smell of burning frankincense pervades shops, restaurants and homes.
So, who were the people who lived in the Land of Punt? Today we think of the Yemeni as being Arab but till the advent of Islam, south-eastern Arabia was home to a culture that was quite distinct from that of the rest of the peninsula. Its mountains and coasts were home to a number of related but constantly feuding tribes such as the Himyar, the Hadramawt and the Sabeans. Archaeologists have found as many as ten thousand inscriptions describing the lives, feuds, treaties and rulers of these tribes.
The Sabeans were one of the most powerful tribes. Around the eighth century BC, they created an empire that extended from Yemen into Eritrea and parts of Ethiopia. The Sabeans may have introduced wheat and barley to Ethiopia although a local cereal called ‘teff’ continues to be popular to this day. They also introduced their script which was originally written left-to-right and right-to-left on alternate lines; not such a silly idea if you think about it. The Sabean script would evolve into the Ge’ez script of the kingdom of Aksum and survives as the modern Ethiopian script (although it is now written exclusively left-to-right).
These early interactions between the Yemenis and the Egyptians would later extend to neighbouring lands and probably gave rise to the legend about King Solomon and Queen Sheba (i.e. Saba). According to the Ethiopian version of the legend, when Queen Sheba returned after meeting Solomon in Jerusalem, she gave birth to a child named David. This child would grow up to be King Menelik I, founder of the Solomonic dynasty of Aksum. The legend also says that when David was a young man, he visited his father Solomon’s court. At the time of departure, he stole the Ark of the Covenant and brought it back to Aksum. The Ark is supposed to be a wooden box covered in gold containing the two tablets received by Moses on which the Ten Commandments were inscribed by God himself.
Although there is no historical evidence supporting this story about Solomon, Sheba, Menelik and the stealing of the Ark, it has been part of the national founding myth of Ethiopia since the medieval ‘Solomonic’ dynasty came to power in AD 1262. The medieval dynasty promoted the story to give itself a biblical lineage and it would provide legitimacy to the royals till their rule ended in 1974 with Haile Selassie, the last emperor of Ethiopia. Nonetheless, the Ethiopian Orthodox Church claims that it is still in possession of the original Ark and replicas, known as tabots, continue to play an important role in religious festivals across the country.33
The Sundaland Diaspora
With all this migrating and churning going on in the western Indian Ocean rim, one adventurous band of Indians decided to be different and head east. They seem to have got into their boats somewhere on the country’s eastern shore and sailed along the coast, past Sumatra and Java and eventually ended up in Australia! Recent genetic studies show that a bit more than 4000 years ago, a band of Indians turned up in Australia and contributed their DNA to the aborigines. This finding confounds the earlier belief that there were no new arrivals to the island continent between the initial migration of Melanesians 45,000 years ago and the arrival of the Europeans. Moreover, the new immigrants brought along their pets, the ancestors of the dingo dog.34 This is possibly why the dingo looks suspiciously like the stray dogs one sees all over India.
The Indians were not the only people on the move in the region. As mentioned in the previous chapter, the coastlines of South East Asia witnessed major changes as Sundaland was inundated by the post-Ice Age floods. Recent genetic studies confirm that the region’s current population landscape is heavily influenced by human migrations following the floods.35 These South East Asian migrations involved two major ethnic groups—the Austronesians and the Austroasiatic. Someone with a dark sense of humour must have given them such similar names in order to purposely confuse future generations of researchers. To make it less confusing, let’s call them AN and AA respectively. The AN included the ancestors of the Malays, Indonesians, Filipinos, Bruneians, Timorese and significant minorities in neighbouring countries. It also includes Taiwanese aborigines and the Polynesians spread across the Pacific. As one can see, they had a strong maritime culture.
It was once thought that this group originated in Taiwan but it now appears that they lived along the eastern coast of Sundaland and were forced by the floods to search for new homes. The outrigger canoe was an important part of their maritime culture. It is a simple design but clearly very effective as it allowed the AN to colonize most of the islands in South East Asia during the Neolithic period. Some of these islands may have had existing Melanesian populations who seem to have been squeezed into a smaller area in and around New Guinea and Fiji. A few centuries later, the eastern Polynesian branch of ANs would set out to colonize a swathe of islands across the Pacific—from New Zealand to Hawaii and Easter Island. Similarly, the western branch would sail across the Indian Ocean and settle in Madagascar. Thus the AN came to colonize a large swathe of the planet from Madagascar to Hawaii!
The speakers of AA languages were the other important ethnic group of South East Asia. They include the Vietnamese, Khmer (i.e. Cambodian) and the Mon in Myanmar and Thailand. Unlike their Malay and Polynesian cousins, however, this group seems to have preferred to migrate over the land rather than over sea. At some point, small groups of AAs drifted into north-east India. The descendants of these migrations are the Munda-speaking tribes, such as the Santhals, who are scattered all over eastern and central India. A somewhat later wave survives today as the Khasis of the state of Meghalaya. Thus it came to be that India’s population mix includes people who speak languages related to Vietnamese and Khmer!36
The folk tales and legends of South East Asia recall the Great Flood. For instance, the Laotian founding myth of Khun Borom tells us that the gods once became angry with the sinful and arrogant behaviour of humans and caused a flood that washed away all mortals. After the deluge they sent a buffalo that died and from its nostril grew a creeper that bore giant gourds. When the ‘khun’ (i.e. lords of heaven) cut open the gourds, a new generation of humans emerged from them with different gourds giving birth to different ethnic groups.37
The Laotian story is quite different from that of Noah or Manu, but it too remembers an earlier way of life that was destroyed by a huge flood and of how civilization had to be re-established. Similarly, the oral traditions of Australian aborigines also speak of swathes of coastline that were flooded. Till just a decade ago, it was common for scholars to dismiss indigenous oral histories as mere fantasy but latest research shows that they often contain folk memories of real events.38
The Daughters of Chitrangada
Matrilineal customs appear to have been an important feature of the AA-speaking groups migrating across South East Asia and into India’s north-east. The Khasis of Meghalaya, for instance, remain matrilineal to this day. Traces of matrilineal customs seem to have been imbibed even by neighbouring communities that may never have been matrilineal. For instance, in Assamese Hindu weddings, the ‘sindoor’ (red vermilion) is applied to the forehead of the bride by the mother-in-law at the ‘jurun’ ceremony that precedes the wedding. The act of applying sindoor is a key part of Hindu marriage rituals and is usually the prerogative of the husband. The performance of this rite by the groom’s mother symbolizes the women of the family accepting a new member—a very matrilineal view of a wedding.
So why were the AAs matrilineal? The answer to this riddle is found in the study of the genetics of the AA groups. It appears that the Indian branch is the result of almost exclusively male migrations.39 At the risk of oversimplifying, one could say that groups like the Santhals and Khasis are the result of male migrants from South East Asia marrying local women. This fits in with the hypothesis offered in the previous chapter of matrilineal customs emerging in South East Asia due to a Neolithic male population that was relatively more mobile than the female population. We do not know exactly what drove these migrations but this movement was not quite one of conquest since the incoming males seem to have accepted the property rights of the local women.
It is fascinating that the Iron Age epic Mahabharata hints at the matrilineal streak in India’s north-east. It tells us how the exiled prince Arjun visited the kingdom of Manipur. There he met the warrior princess Chitrangada and married her. However, note that the marriage took place on the explicit condition that Chitrangada would not have to follow Arjun back home as she and her children were heirs to the throne. Again notice the easy acceptance of a male outsider combined with the rootedness of the local female. The story does not end here. Ulupi, the queen of a neighbouring Naga tribe,40 also falls in love with Arjun and kidnaps him. The epic then tells us of how Arjun is eventually restored to Chitrangada.
We are not concerned here about the historical veracity of this story. What is interesting for our purposes is the portrayal of two strong female characters and a social context that is different from that of the Gangetic heartland. It appears that the Iron Age composers of the Mahabharata, based in the north-west of India, were aware that the status of women was different in the north-east. Manipur remains home to formidable women including Mary Kom, five-time world boxing champion. The daughters of Chitrangada are alive and well.