Underworld: The Mysterious Origins of Civilization

by Graham Hancock

  The world’s greatest relief is a locus of enormous geodynamic energy consisting of a complex interplay between tectonics and glacial and fluvial erosion associated with widespread and frequently catastrophic mass wasting. One obvious product of such a situation is the problem of reliably discriminating between diamictons deposited by glacier ice and those laid down by other processes. The two suites of processes are frequently intimately related, posing a recurrent challenge to those attempting to establish limits of past glaciations.64

  ‘Diamicton’ is a general term used to describe a mixture of sand, clay, silt or gravel that is laid down by various geological processes – notably the forces of flowing rivers, or moving glaciers, or lakes draining catastrophically. Derbyshire’s point is that where ongoing geological activity results in a continuous mixing up and redepostion of the materials being studied – as is most definitely the case in the Himalayas – then there is obviously going to be uncertainty over the extent of glaciation in the region at any particular moment in the past.

  The range of the uncertainty surrounding the extent of the ice-cap at the LGM is, however, surprisingly large – since there is all the difference in the world between ‘an ice-cap of continental scale’ on the one hand, and a regional ‘Alpine glaciation’ on the other. Moreover, this uncertainty seems even greater when it comes to immediately post-glacial events. Indeed, although a great deal is known about the cataclysmic meltdown of other ice-caps in this period, I was surprised to discover that the literature has relatively little to say about what happened in the Himalayas after the LGM.65

  Before and after

  Scientists have been able to pick up traces of at least one cataclysmic melting event that took place in the area before the LGM. It is another measure of the uncertainty of the data available for study that the date-range offered for this flood is very wide – it could have happened any time between 28,000 years ago and 43,000 years ago.66 Fortunately, its imprint on the landscape has not been as badly obliterated and jumbled as those of earlier and later floods and geologists have narrowed its location to the Upper Chandra valley in the Lahul Himalaya. Using landforms and sediment data, Peter Coxon, Lewis Owen and Wishart Mitchell, writing in the Journal of Quaternary Science, conclude that former glacial Lake Batal – which had backed up the Chandra valley for about 14 kilometres – suddenly burst through its ice dam. When it did so it released almost one and a half cubic kilometres of water into the valley in less than a day: ‘This cataclysmic flood was responsible for major resedimentation and landscape modification within the Chandra valley.’67

  Further striking but unfortunately undatable evidence of colossal ancient outburst floods is provided by the presence of numbers of large boulders scattered across the Potwar plateau – so-called ‘Punjab erratics’ – which geologists now believe were ‘carried down the Indus valley by catastrophic flooding, probably in iceberg rafts’68 The traces of violent outburst floods long after the post-glacial meltdown was over, have also been widely recognized and there are a number of eye-witness accounts. In 1959, for example, there was

  a sudden outburst from an ice-dammed lake in the Shimsal valley which caused a flood wave of approximately 30 metres to be produced, destroying the village of Pasu at the confluence with the Hunza River, 40 kilometres down-valley.69

  Similarly, when a moraine-dammed glacial lake in the Khumbu area of eastern Nepal called Dig Tsho burst on 4 August 1985, the consequences for the region were catastrophic:

  The destruction of a newly-built hydroelectric power plant, 14 bridges, about 30 houses, and many hectares of valuable arable land, as well as a heavily damaged trail network resulted from 5 million cubic metres of water plummeting down the Bhote Kosi and Dudh Kosi valleys. The breaching of the moraine was triggered by wave action following an ice avalanche of 150,000 cubic metres into the lake. The surge had a peak discharge of 1600 cubic metres per second; 3 million cubic metres of debris were moved within a distance of less than 40 kilometres.70

  The most spectacular event, however, was undoubtedly the great Indus flood of 1841 – a deluge of near biblical proportions which, like the return of the waters of the Red Sea after the Hebrews had passed safely through to the other side, destroyed a vast army.

  The first step was an earthquake in late 1840 or early 1841. The earthquake caused the collapse of the Lichar Spur, part of the flank of Nanga Parbat, which blocked the Indus valley to a depth of 300 metres, strangled the downstream flow of the Indus to a trickle for six months and caused a lake 60 kilometres long and 300 metres deep to back up behind it. When the blockage was breached in June 1841 a gigantic flood wave was released. The wave raced downstream along the (by then almost dry) course of the Indus at a terrifying pace and fell upon a Sikh army that was camped on the Chach plain near Attock, 400 kilometres downstream.71 Eye-witnesses later reported that:

  A wall of mud, many tens of metres high, rushed down the watercourses. Those people not fast enough to reach the high ground, numbering several thousand troops and camp followers, were lost. Trees were uprooted, buildings destroyed, artillery guns scattered, and farmland washed away. Large areas of the Vale of Peshawar were flooded as the various tributaries banked up against the Indus floodwaters.72

  Today there is increasing awareness of the dangers posed by outburst floods specifically related to glaciation. It has been pointed out, for example, that more than thirty glaciers in the Karakoram mountains are presently in a position to ‘form substantial dams on the Upper Indus and Yarkand river systems. Many more interfere with the flow of rivers in a potentially dangerous way.’73 According to Kenneth Hewitt of Wilfred Laurier University, Canada:

  A particularly large and dangerous dam occurs where a glacier enters and blocks a major river valley of which it is a tributary … In one region of the world, … the Karakoram Himalaya and neighbouring ranges, there has been a substantial number of these main valley glacier lakes in modem times. Outbursts from a series of dams … between 1926 and 1932 brought devastating floods along more than 1200 kilometres of the Indus. Some even larger landslide dams and outburst floods occurred here in the nineteenth century and an exceptional concentration of surging glaciers has been found. Some of the latter have formed main valley ice dams … Thirty-five destructive outburst floods have been recorded in the past two hundred years.74

  Stocktaking

  There are a few details that are worth holding on to.

  The Equilibrium Line Altitude of glaciation in the Himalayas at the LGM was about three-quarters of a kilometre or more lower than it is today.

  The ice-cap at the LGM was much more extensive than it is today – although there is no agreement over exactly how much more extensive.

  There have been catastrophic outburst floods from the Karakorams and the Himalayas in the past, floods that reshaped landscapes, floods that carried icebergs full of huge impacted rocks all the way down to the Potwar plateau.

  Such outbursts continue to occur and even in the much reduced conditions of today’s glacial cover they can produce floodwaves 30 metres high capable of smashing whole villages to smithereens and destroying armies.

  The region is uniquely plagued by the particularly dangerous and rare phenomenon of its main river valleys being dammed by gigantic landslides or by the encroachment of glaciers – a sure recipe for catastrophic outburst flooding.

  Paradoxically, despite the evidence for catastrophic outburst floods before the Last Glacial Maximum, as well as in much more recent times, the literature pays scant attention to the issue of outburst flooding in the Himalayas during the 10,000 years after the LGM.75

  But this shouldn’t prevent us from asking a few common-sense questions:

  If main river valleys are threatened by glaciers today, and if even a giant river like the Indus can be blocked for six months, then isn’t there every probability that the threat would have been much bigger and much worse under LGM conditions?

  Is it unreasonable to speculate – as the Rig Veda has be
en telling us all along – that there could have been a time, within the memory of man, when some of the great rivers of north India were indeed choked off, most likely by giant glaciers entering and blocking their main valleys up in the Karakoram and Himalayan ranges? If so, then those glacial dams would eventually have burst asunder and the rivers chained up within them would have been set free once again …

  Last but not least, is it so far-fetched to wonder if such a sequence of events might have inspired the great Vedic myth of Indra’s slaying of Vrtra with its specific symbolism of the freeing of the Seven Rivers?

  Probably no more far-fetched than the more orthodox ‘cloud-demon’ and ‘drought demon’ ideas, but hardly foolproof as a theory. For example, there’s the absence of evidence of flooding in the Himalayas after the LGM – but that means very little given the state of the geological record (and the level of disagreement amongst geologists on the actual extent of the maximum glaciation).

  More seriously there is the other ‘face’ of the Vrtra myth – the clear association that some of the hymns make between the presence of the Dragon and the withholding of rain on the one hand, and between the slaying of the Dragon by Indra and the return of the rain on the other.

  How is that to be explained if Vrtra is a symbol for glaciation?

  The dry and the wet

  Sediments in ocean-bottom cores taken in the Arabian Sea off the south-west coast of India contain pollen traces that tell us about the types of vegetation that grew on the subcontinent at different periods going back to the Last Glacial Maximum – and since vegetation cover is determined by climate, reliable deductions can be made from these pollen records about India’s climate in past epochs.

  The Arabian Sea cores demonstrate that there was a period of extreme cold and aridity in India between 25,500 years ago and 21,500 years ago.76 This period is described by Elise Van Campo of the Université des Sciences et Techniques du Languedoc as ‘the LGM interval’77 and coincides exactly with other indications from around the world of the duration of LGM conditions (i.e., the Last Glacial Maximum was not a peak reached for a very short time, but rather a plateau of extreme glaciation that was sustained, in India at any rate, for 4000 years). When warming did set in, it set in quickly and between 21,500 years ago and 13,000 years ago the Indian climate did a 180-degree flip from cold and arid to warm and wet:

  The major fluctuations of the Indian monsoon climate are characterized by two extreme periods, a very arid period around [25,500 to 21,500 years ago] and a very humid period culminating at [13,000 years ago] … The climate conditions of the LGM interval were greatly different from modern conditions. The southwest monsoon, which produces a strong asymmetry between the western and the eastern coasts of the Arabian Sea, was considerably reduced and arid conditions were very similar on both sides) …78 (Carbon-14 dates in original text replaced with approximate equivalents in calendar years.)

  What this would have meant in the Himalayas between 25,500 and 21,500 years ago was 4000 years of deep freeze as the ice tightened its grip on the valleys and the headwaters of the rivers in the mountains.

  Then at the peak of the LGM interval, some time soon after 21,500 years ago, the phase of warm, wet climate in India abruptly kicked in. Back to the Arabian Sea cores, which demonstrate:

  an increase of monsoonal rainfall as early as about [19,700 years ago] at 10 degrees north and at [18,500 years ago] at 15 degrees north. This period … culminates synchronously at [13,500 years ago] at 10 degrees and at 15 degrees north and is considered as the period of the greatest abundance of monsoonal rains.79

  Worldwide, we know that the period of 14,000 to 13,000 years ago, which coincides with the peak of abundant monsoonal rains over India, was marked by violent oceanic flooding – in fact, the first of the three great episodes of global superfloods that dominated the meltdown of the Ice Age. The flooding was fed not merely by rain but by the cataclysmic synchronous collapse of large ice-masses on several different continents and by gigantic inundations of meltwater pouring down river systems into the oceans.80

  If this was happening in other glaciated regions such as North America and northern Europe between 14,000 and 13,000 years ago, then things are unlikely to have been very different in the Himalayas, and it seems safe to assume that there must have been episodes of exceptionally powerful outburst flooding and that all the great rivers from the Indus to the Ganges would at that time have been in full flow.

  So is 14,000 to 13,000 years ago a candidate epoch for the events recounted in the Rig Veda as the slaying of Vrtra and the freeing of the Seven Rivers?

  The answer has to be no – simply because the previous 7000 years had witnessed a continuous worldwide increase in temperature and because 14,000 to 13,000 years ago was the peak and the climax of this long, humid warming phase in India. As such, it is most unlikely that the glaciers in the Karakorams and the Himalayas would have been surging or advancing so as to block or ‘enchain’ rivers in the way that the Rig seems to describe. On the contrary, everything suggests that the flow of the rivers should have been uninterrupted from the end of the cold, dry LGM interval 21,000 years ago until the clear end of the humid phase that shows up in the cores at around 13,000 years ago.

  Moreover, the Vedic myth portrays the slaying of Vrtra as being followed by the release of the waters – both rivers and rain. This is very clear and, in a way, the point of the whole thing. But that was not what happened.

  A Dragon called the Younger Dryas

  What happened, at around 13,000 years ago, was that the long period of uninterrupted warming that the world had just passed through (and that had greatly intensified, according to some studies, between 15,000 years ago and 13,000 years ago)81 was instantly brought to a halt – all at once, everywhere – by a global cold event known to palaeoclimatologists as the ‘Younger Dryas’ or ‘Dryas III’.82 In many ways mysterious and unexplained, this was an almost unbelievably fast climatic reversion – from conditions that are calculated to have been warmer and wetter than today’s 13,000 years ago,83 to conditions that were colder and drier than those at the Last Glacial Maximum, not much more than a thousand years later.84

  From that moment, around 12,800 years ago, it was as though an enchantment of ice had gripped the earth. In many areas that had been approaching terminal meltdown full glacial conditions were restored with breathtaking rapidity and all the gains that had been made since the LGM were simply stripped away:

  Temperatures … fell back on the order of 8–15 degrees centigrade … with half this brutal decline possibly occurring within decades. The Polar Front in the North Atlantic redescended to the level of Cabo Finisterre in northwest Spain and glaciers readvanced in the high mountain chains. With respect to temperature the setback to full glacial conditions was nearly complete …85

  For human populations at the time, in many except the most accidentally favoured parts of the world, the sudden and inexplicable plunge into severe cold and aridity must have been devastating. And in the Karakoram-Himalayan region, as in other glaciated areas, it is very likely that it was accompanied by a significant readvance of the ice-cap that previously had been in recession for some 7000 years.

  Is it possible that that this hypothetical readvance of the Himalayan ice-cap between 12,800 years ago and 11,40086 years ago could be the event personified in the Rig Veda as Vrtra the Dragon, the enchanter, the great magician, ‘who barred up the waters’?

  Since the slaying of Vrtra resulted in the release of the waters to flow to the sea, it obviously made sense to find out if there was evidence of sudden large-scale meltwater floods off the mountains shortly after 11,400 years ago when the ‘climate switched back to warm, moist Holocene conditions, over only a few decades’.87

  Salt and freshwater

  I did find evidence of floods. It was in another set of cores taken off the Indian coast. According to a report in Nature by a team of Australian scientists:

  Microfossil, sediment and oxygen-isotope stu
dies of deep-sea cores from the Bay of Bengal and northern Arabian Sea have revealed strong contrasts between high late Pleistocene and low early Holocene salinity values, indicative of major changes in runoff from the large rivers of southern Asia.88

  Some definitions: salinity values measure the ‘saltiness’ of the sea, so ‘high salinity values’ mean a saltier sea and low salinity values mean a less salty sea – i.e, a sea with more freshwater in it. The Pleistocene-Holocene boundary is set, arbitrarily, at 12,000 years before the present. ‘Late Pleistocene’ is loose language but generally means the few thousand years before 12,000 years ago. ‘Early Holocene’ is loose language too but generally means anywhere between 12,000 years ago and 10,000 years ago.

  Why were India’s seas so salty just before 12,000 years ago? The most likely explanation is that the flow of the great rivers draining the Karakoram-Himalayan region had virtually ceased because of the advance of glaciers into their main valleys during Dryas III – pretty much as the Rig Veda tells us (‘Ahi who besieged the waters … the insatiate one, extended, hard to waken, who slumbered in perpetual sleep’). Likewise, the explanation for the low salinity values that suddenly appear soon after 10,000 years ago is a sudden gigantic inrush of freshwater to the Arabian Sea and the Bay of Bengal on a scale that could have been caused by the breaching of ice dams in the Himalayas, the freeing of rivers pent up behind them, and the flushing out of parts of the ice-cap. (The Dragon stretched against the seven prone rivers, where no joint was, thou rentest with thy thunder.’ ‘Like lowing kine in rapid flow descending, the waters glided downward to the ocean.’)