The Great Divide Read online

  Around 4200 BP all of the ancient civilisations in Mesopotamia – even the Akkadian – either contracted markedly or collapsed entirely and excavations confirm that this collapse was triggered, at least partly, by drought. These studies have shown, say Peter Clift and Alan Plumb, that, at around 4200 BP, there was a marked rise in the levels of eolian dust (silt-sized dust particles, deposited on the surface of the Earth by wind, and a common way of dating sites) in the Gulf of Oman, which geochemical and mineralogical analysis shows came from the west, from Sumeria and Akkadia. Carbon-14 analysis likewise shows that this excessively dry spell, with strong dust storms and low water levels in the Dead Sea, lasted for some 300 years, accounting for the societal collapse.9 The overall picture, therefore, is that, in this region, at around 5000 BP, city-states formed to cope with the increasing desiccation – cities being where elaborate collaborative irrigation systems were formed – but then, at ~4200 BP, the desiccation had reached a point where even city-states could not cope with the strains this imposed, as a result of which the society collapsed, and the population dispersed to survive as best it could.

  A complicating factor, but still weather-related, is that northern Mesopotamia was a rain-fed society, which tended to mean that administrative areas, laid out like a patchwork quilt, developed earlier and were equal (and therefore more competitive and warlike), whereas southern Mesopotamia developed later, being irrigated by rivers and therefore was geographically more linear, the irrigation systems implying a more hierarchical (and stable) administration. This latter arrangement tended to cope better with the later desiccation.10

  This is a pattern paralleled in Egypt and the Indus Valley, where – in the second case – the earliest human settlements on the Indian subcontinent are to be found in what is now western Pakistan and eastern Afghanistan.11 These small-scale farming communities are known as the Mehrgarh Culture, so called after a site near modern Quetta in Pakistan. The earliest Mehrgarh remains have been dated to 9000 BP which, though they show no evidence of pottery, do appear to have been in contact with the Fertile Crescent, with the earliest farming being developed by semi-nomadic people growing wheat and barley and herding sheep, goats and cattle.12 Judging by the development of pottery, and its elaboration, this culture was flourishing at 7500 BP but, towards 5500 BP, less and less material was being buried alongside the inhabitants, indicating a society in decline, and suggesting that the weakening of the summer monsoon was putting stress on the communities, which seem to have been largely abandoned between 4600 and 4000 BP. This decline corresponded with the rise of the Indus Valley civilisation on the flood plains of the Indus/Sarasvati Rivers to the east, with the inhabitants of the Mehrgarh culture migrating because their own homeland became more arid as the summer monsoon continued to weaken.

  The Harappan (and Mohenjo-Daro) civilisation that grew out of these displacements was very advanced for its time, with a brilliantly developed water, sanitation and waste disposal system.13 Water was obtained from wells, houses had rooms set aside for bathing, and covered drains ran down the main streets – there was at this time little apparent problem with water. It was this culture which developed the Indus script (more or less simultaneously with other scripts in Mesopotamia and Egypt), and by 4600 BP, Harappa-Mohenjo-Daro was a complex civilisation. But its brilliance, the surplus it generated, enabling technological achievement – including the plough – and its non-food-producing artists, scribes and artisans, concealed the environmental stresses with which the brilliant organisation was designed to cope and, again around 4200 BP, the civilisation underwent a profound change, with the population moving southward into a post-urban phase of smaller settlements. Studies of the sedimentation in and around Lake Lunkaransar, for example, show that, at about 4230 BP, the lake was only temporarily filled with water, indicating a ‘rapid weakening’ of the summer monsoon. Analysis of the plankton in the mouth of the Indus River shows that its water discharge slowed rapidly at 4200 BP, for the same, monsoon-related reason.14

  Which brings us back to the Sarasvati River. Peter Clift and Alan Plumb remind us that this river is mentioned in the Rig Veda no fewer than seventy-two times, and was comparable to the Indus in size, but does not exist today. However, the Rig Veda, written down at about 5500 BP, describes it as being a major river and, as was alluded to earlier, nearly 2,000 of the 2,600 Harappan sites that have been discovered lie along the palaeo-channel of the Sarasvati. This major change in riverine layout may well have been related to the weakening of the monsoon.

  Much the same picture emerges in China – more humid conditions in the early Holocene, followed by drawn-out drying after ~8000 BP. Stone tools found in China and dated to ~11000 BP, around the end of the Younger Dryas, show a wide variability, suggesting a transition by early man to broad-spectrum foraging and seed-processing by hunter-gatherers. This developed into agriculture, with the evolution of permanent settlements associated (at least in time) with the drying of the climate and the weakening of the monsoon. The earliest human settlements in China are found in the Yellow River Valley between the modern cities of Xian and Lanzhou, where cores drilled into the earth show swamp and river sediments that contained abundant aquatic mollusc shells, indicating wet conditions between ~8000 and ~6000 BP, followed by drying at ~6000–5000 BP.

  The oldest cultures known in the Yellow River area are the Dadiwan (7800–7350 BP), followed by the Yangshao (6800–4900 BP, better known perhaps after Banpo, its best-excavated site) and the Dawenkou (6100–4600 BP). In most cases, settlements were far more numerous after 6000 BP, indicating a weakening of the monsoon and a drying of the climate which drove populations to work together in larger groupings to maximise agricultural production.15

  The evidence is also plentiful that the Yangshao people cultivated wheat, millet and rice, using the plough, from 6000 BP. They kept domestic animals but also hunted and fished, and had specialised, highly polished stone tools. They may have cultivated the silkworm and had painted pottery. They were succeeded, around 5300 BP, by the Majiayao in the upper Yellow River region, which developed copper and bronze technology. In turn, they were followed by the Qijia culture at 4400–4000 BP and the Longshan culture on the Shandung Peninsula.16

  Clift and Plumb observe that the bone and dental remains in these cultures show that there was an onset of a cooler and drier climate at around 5900 BP, and that this marked a decline of more egalitarian societies, such as the Yangshao, to be replaced by the more stratified, chiefdom-like society of the Longshan, who employed pottery wheels. The Longshan culture also marked a transition to city life, with earth walls and defensive moats, and rice cultivation. This, they say, was an adaptation to cope with a weaker monsoon.

  Like the Qijia culture, the Longshan died out at about 4000 BP, accompanied by the disappearance of high-quality pottery, and all over the Yellow River Valley the density of settlements after 4000 BP decreased, with conditions becoming less hospitable.17

  The picture which emerges, therefore, is fairly consistent. There was, across the Old World – or those civilisations that we know most about, in Arabia, northern Africa, in India – a slow weakening of the monsoon after ~8000 BP, when the last great flood occurred. The great glaciers of the polar arctic, melting in the north Atlantic, would have evaporated, producing increased snowfall over the Himalayas and the Tibetan plateau, which in turn consumed more and more of the sun’s energy, weakening evaporation in the south-west Pacific, sapping the strength of the monsoon. This provoked people to live closer together, in urban contexts.

  Then there was a further rapid shift to even drier conditions after ~4200 BP, which appears to be slightly before the Chinese cultures collapsed. This may be due to faulty dating, or may indicate that these cultures were able to hang on for a while – for as long as two- to three-hundred years – before finally succumbing. This would also fit with the emergence of the oldest known imperial dynasty to control large areas of China – the Xia dynasty – which dates from ~4000 BP. Indeed, say Cl
ift and Plumb, maybe this is what ‘dynasty’ implies – the ability to cope well with a deteriorating climate with greater centralised government, spread over larger areas, more varied and therefore better able to sustain food production, after other, smaller entities have succumbed to severe droughts and floods.18

  THE EL NIÑO TIMETABLE

  On the other side of the Pacific, in South America, there was no monsoon but there was El Niño, in some ways the other ‘end’ of the world’s weather system. The climate of the Pacific coast is of interest to us in two ways. In normal times, in the past, before say 5800 BP, archaeological evidence shows that the El Niño phenomenon was very rare, occurring much less than it does today or in historical times.* We know this from the analysis of fish and shellfish remains in the middens of Peruvian and Chilean archaeological sites, which contain bones and mollusc shells of species that prefer colder waters than are brought about by El Niño events.

  After ~5,800 years ago, however, two things happened. In the first place, and for almost 3,000 years, El Niños began to occur more often than before but still relatively infrequently, perhaps only once or twice a century. Second, the coastal cultures began living in large villages and building temples. These infrequent El Niño events would have brought with them widespread devastation – the plentiful anchovies in coastal waters would have disappeared, along with the seabirds and sea mammals which fed on them, tidal waves would have flooded inshore, terrible winds would have devastated buildings and trees. With such catastrophes occurring, say, every other generation, or every three generations, they would be fresh in people’s memories and very much part of the folklore and myths of these early peoples. Is that why they huddled together in villages at this time, and constructed temples, so they could appeal to their gods to protect them from these devastations? (More will be made of this later on.)

  This situation persisted for nearly 3,000 years. Then, according to the mollusc and other evidence, at about 3,000 years ago there was another change. Now the El Niños started to occur even more frequently, closer to the modern rate. This rate is itself interesting. Currently, El Niño events occur every three to seven years but in the earlier part of the twentieth century they occurred every seven to fifteen years. The question naturally arises therefore as to whether El Niño has been getting progressively more common as the years have passed since 3,000 years ago, and if so why? Is there a link here between the weakening monsoon over the past 8,000 years and the increase in frequency of El Niño? Or is that the wrong way round? We know that there is a link – albeit imperfectly understood – between El Niño events and drought in India.20

  We shall come back to this matter time and again but for now the important point is that, at 3,000 years ago, when El Niño started to become much more common, the first civilisations of Peru collapsed and it was several centuries before urban structures reappeared, and even then under very different political systems.

  Here, therefore, we see both similarities and differences between the Old World and the New. Urban cultures – primitive civilisations – appeared and disappeared in both hemispheres, most likely as a result of grand-scale changes in the weather, but the climatic phenomena were different and the time-scale, though it may have been linked, was also different. In the Old World the weakening monsoon at first caused communities to form, to develop irrigation techniques to cope with desiccation. But then, later, desiccation got so bad that not even city-states could cope with the situation. In the New World, the more frequent onset of enso events was an important factor in causing villages and temples to be built, to deal with the phenomenon by communal worship, but then got so frequent that the villages and temples could not cope with that situation. The perturbation became so common that people concluded it was not worth the trouble to rebuild after an enso-caused devastation. This divergence will be echoed in later developments.*

  THE UN-PACIFIC PACIFIC

  There were two other geographical factors that affected the New World civilisations disproportionately. As is now well known, the Earth’s surface consists of a number of tectonic plates, vast slabs on which the continents gradually slide over the mantle below the crust at a rate of a few centimetres a year. Where these plates meet we find almost all the Earth’s volcanoes and this is also where the majority of earthquakes occur, as one plate jostles another. About 75 per cent of the Earth’s seismic energy is released along the boundaries of the Pacific Plate and another 23 per cent comes from a zone extending eastwards from the Mediterranean. The rest of the world accounts for just 2 per cent.23 Map 8 shows the layout of these plates, plus the range of the active volcanoes and the spread of recent important earthquakes.

  Collectively, this shows that, for the most part, the active volcanoes line the edge of the Pacific (the so-called ‘Ring of Fire’), and this does not include underwater volcanoes, which are also disproportionately common around the Pacific rim. It is known that earthquakes and volcanic activity are related to tidal activity, in itself related to the gravitational effects exerted by the moon, and that they occur on a roughly regular cycle of just over four years, occurring mainly in (the northern) winter.24 There is also a relationship between volcanic activity and El Niño, in that underwater volcanoes, when they erupt, release heat into the ocean in massive amounts and this can trigger an enso event. As an indication of the level of activity, there were 56 confirmed underwater eruptions in 2001–2002.

  We can be more specific than this, however. In 1981 the Smithsonian Institution in Washington DC published a directory, gazetteer and chronology of volcanism during the last 10,000 years, under the lead editorship of Tom Simkin. This identified 5,564 eruptions from 1,343 volcanoes all over the world.25 Volcano lists were not new. The first volume of the Catalog of Active Volcanoes of the World (cavw) was published in 1951 and updated via the Bulletin of Volcano Eruptions since 1960, but the Smithsonian publication had 400 more eruptions than any previous publication. That chronology tells us that in the years 1400–1500 there were only three recorded eruptions in Meso- and South America, two in Guatemala and one in Peru. In the years 1500– 1600, on the other hand, 139 eruptions are reported for exactly the same area. Geological phenomena being the slow-moving entities that they are, this startling disparity cannot reflect a real sudden upsurge in seismic activity of some 4,600 per cent. Instead, the arrival of the Europeans must account for the greater reporting of eruptions. This is reinforced by the records in Italy. That country does not have an especially large number of volcanoes (18 compared with 35 in Mexico and 57 in north Chile and Bolivia) yet it has by far the greatest number of reported eruptions, pre-1492.

  Beyond the Ring of Fire, two areas of less dramatic but still extensive activity are the Rift Valley of north-east Africa and the north and east coasts of the Mediterranean. Central and northern Europe, Russia and mainland Asia (China and India), together with Australia and the eastern parts of the Americas are relatively free of earthquake activity.

  A closer look at the so-called Ring of Fire shows further that the volcanoes of the western rim lie not along the Asian mainland but on the peninsulas and offshore islands: Kamchatka, Japan, the Philippines, Indonesia, Sumatra, the Sundas and New Guinea. This means that the only areas where volcanoes occur on continental mainlands, and where early civilisations formed, are around the north and east Mediterranean and in Meso- and South America.

  If we accept the slow pace of geological change, and that records for Meso- and South America were fuller – and therefore more accurate and reliable – after the Europeans arrived, then we may say that the sixteenth-century figure – 139 eruptions – was more typical. This translates into one eruption every thirty-seven weeks. Put another way, Popacatépetl (whose name in Nahuatl means ‘Smoking Mountain’), which is 43 miles south-east of where Tenochtitlán, the Aztec capital was, and was easily visible, erupted in 1519,in 1521 and 1523, while Santa Ana in El Salvador erupted in 1520 and 1524, and volcanoes in Nicaragua erupted in 1523 and 1524. In fact, it was not
until 1554 that a year passed without a volcanic eruption in Meso- or South America. Only Italy could keep pace but that was thanks to one volcano, Etna, which erupted 39 times between 1500 and 1541 but then quietened down for nearly a decade.

  Undoubtedly, then, what we now call Latin America, at the time of the Conquest, had – and had had, in the previous centuries – far more active volcanoes than anywhere else on the world’s mainland continents (as opposed to offshore islands) where high civilisations developed. It may also be worth pointing out that earthquakes and eruptions are more serious for sedentary peoples, who can’t move away. South America, with its many small, isolated language groups, dependent on potatoes, manioc and maize, would have been especially vulnerable to quakes and eruptions.

  Even extinct volcanoes can have dangers, collapsing inwards and causing landslides which, near the sea, can provoke tsunamis. Evidence has also emerged recently that, at the end of the Ice Age, thinning glaciers on volcanoes could destabilise vast chunks of their summit cones, triggering ‘mega-landslides’. One is now known to have occurred around 11,000 years ago on Planchón-Peteroa, a glaciated volcano in Chile, when around one-third of the volcanic cone collapsed, according to Daniel Tormey, of entrix, an environmental consultancy based in Los Angeles. On this occasion ten billion cubic metres of rock crashed down and smothered 370 square kilometres of land, travelling 95 kilometres in total.