Out of Eden: The Peopling of the World
Page 17
Controversially, human occupation dates as old as 62,000 years ago were also recently published, based on several different methods applied to earth surrounding the skeleton of a gracile Anatomically Modern Human found at Lake Mungo in the Willandra Lakes region of south-east Australia. One of these reports pointed out that, to have been there so long ago, the first Australians must have crossed earlier at a previous low sea level. The low-sea window of opportunity to get to Australia 70,000 to 60,000 years ago suggests that the anatomically modern Lake Mungo people really were among the oldest on that continent, since the previous low sea level was another 80,000 years before, and near the dawn of our species.9
Only one window of opportunity
Whatever the academic claims and counter-claims, there seems now to be a convergence of evidence from a variety of sites that points to a human presence in Australia as far back as 60,000 years ago. This date followed a very deep lowstand of the world’s oceans caused by an increase in water locked up in the northern ice caps, and took Australia’s shores to around 100 metres (330 feet) vertically below today’s levels. Although the date of lowest sea level of that glacial period was 65,000 years ago, the world’s oceans only fell below 100 metres briefly at that point, and then rose rather rapidly.10
The presence of humans on the north-west coast of Australia from 62,000 years ago does not necessarily date their actual arrival from overseas at that time, since the sea-level was at its lowest 3,000 years earlier. There would have been more difficulty in getting to Australia if 62,000 years ago really was the date of arrival. The actual depth of the lowstand makes a big difference to the chances of successfully crossing the sea from Timor to Australia. From a contour map of sea-level depths off the north coast of Australia, we can see that the distance between Australia and Timor increases from 160 km at –100 metres, and 220 km at –40 metres, to 470 km at –20 metres (100 miles at 330 feet, 135 miles at 130 feet, 290 miles at 65 feet). (Figure 4.1.) The difference between 160 and 220 km may seem small but, until 6,000 years ago, 180 km (110 miles) was the absolute limit of inter-island sea crossings anywhere in the world. The oldest instance of this length of crossing, for which there is archaeological evidence, is the colonization of Manus Island, north of New Guinea, by about 20,000 years ago. So, unless there was a single long-drift colonization of Australia, the window of opportunity to get to Australia was at 65,000 years ago – in order to be visible in the Australian archaeological record from the period 55,000–62,000 years ago. In any case, colonization of Australia solely by random accidental drifts is an unlikely scenario in view of the genetic evidence for multiple founding Australian maternal lines. Such a date of arrival is consistent with a 68,000-year-old genetic date estimated for the expansion of Australian populations.11
Figure 4.1 Last leap to Australia. Shows Southeast Asia (Sundaland) and Australia/New Guinea (Sahulland) as they would have looked 70,000 years ago with the low sea stand of –80m. Beachcombing as far as Bali was dry-footed, but after that there were two routes to Sahul involving island-hopping. Up to Timor this was always easier in the south; thereafter the sea-level window was crucial.
We also have to put evidence of occupation 60,000 years ago into another context: there is evidence of shell middens from some of the earliest Australian sites. Presumably, then, the first people in north-west Australia were still beachcombing when they arrived. If they were, we would not expect to find any dated evidence of their initial presence on the seashore, since the Timor Sea has now advanced up to 500 km (300 miles), covering those first beach habitations. The first datable evidence in north-west Australia following a lowstand entry 65,000 years ago would naturally be from above the next highstand. This highstand occurred between 50,000 and 60,000 years ago and took the sea level to only 40 metres (131 feet) below today’s, and the beach nearer to its present location. A beach close to the Arnhem Land rock shelters 50,000–60,000 years ago is consistent with projected sea levels and with geologist Richard Roberts’ dates of 50,000–60,000 years ago for the occupation of those sites. With this sea-dependent lag period between colonization and archaeological evidence of occupation, the next available lowstand 47,000 years ago would clearly have been too late for sea-crossing and an archaeologically visible presence any earlier than 45,000 years ago.12
Although Australia’s neighbour, the giant island of New Guinea, appears to pose an identical seafaring problem for would-be colonists, there are significant differences. There is a distinct possibility that New Guinea was colonized even earlier than Australia. First of all, there are two stepping-stone routes across the islands of eastern Indonesia (see Figure 4.1). As we have seen, the direct route to Australia from Timor requires a very low sea level for any chance of survival across the long last leg, across the Timor Sea. The natural route to New Guinea via Sulawesi and the Moluccas, however, has shorter inter-island distances and is virtually unaffected by changing sea levels. From each island along the way the next was visible. The islands have steep shorelines, and the voyage to New Guinea passed across very little continental shelf. As a result, sea-level changes made little difference to inter-island distances, and it was an equally easy (or difficult) trip at any time after our ancestors arrived in Borneo overland.13
One could ask whether there is any evidence that the Australians did not take this alternative but slightly longer northern route, since Australia was actually connected to New Guinea throughout this period. The genetic picture tells us that there are no shared clans between Australians and New Guineans, and that New Guineans are as genetically distant from Australians as they are from any other non-African peoples. Not only that, but estimates of the times of expansion for indigenous New Guinea mtDNA clans are around 77,000 years, much older than for Australians. This suggests that New Guinea may have been colonized before Australia, and also that Australian clans were not descended from New Guinean ones; in other words, the first Australian colonists may have found it easier to follow an unoccupied island trail, albeit the more challenging one.14
This tempting genetic story, which is consistent with the extreme antiquity and independence of the colonization of Australia and New Guinea, is not matched by much dated archaeological evidence from the islands of eastern Indonesia or New Guinea. Until recently there was no evidence for human presence along either of these two island trails across eastern Indonesia much more than 40,000 let alone 65,000 years ago. Exciting new evidence from Flores, one of the largest islands on the southern route, now could change all that. A deep (7 metres, 23 feet) pit excavated in Liang Bua cave on Flores has revealed a continuous archaeological record spanning 840,000 years. Liang Bua could play a significant role in providing cultural and palaeontological evidence for much of that period, including the date of arrival of modern humans. Two human species occupied the cave over that period. The first human occupants were probably not modern. Estimated dates for modern human occupation are still under review, but may be the best evidence yet for dating the early arrival of modern humans in East Asia.15
Now, if, following the window-of-opportunity argument, the actual colonization of Australia was 65,000 years ago, and Flores and even New Guinea were colonized 75,000 years ago, we have to start asking how long before that would modern humans have had to leave Africa in order to trek all the way to the South Seas. Short of a guess, that question is remarkably difficult to answer. For a start, we cannot tell how long it took our beachcombers to literally eat their way right down to Timor or Sulawesi unless they left some evidence as they went. As is clear from the sea-level record, however, the beaches our ancestors combed over 60,000–85,000 years ago are now well beneath the sea, so that we should not expect to find much of the evidence – unless we go diving.
Oldest relics of the beachcombing trail?
There is also the problem that if our forefathers left Africa over 70,000 years ago, they could initially have been making and using old-fashioned tools that were largely similar to the ones used by archaic Homo sapiens, who m
ay have left Africa at least 100,000 years earlier (see the Prologue). Such Middle Palaeolithic tools are abundant in India from as long ago as 160,000 years, and are therefore not very helpful as markers for the arrival of Anatomically Modern Humans. Middle Palaeolithic tools also abound in the sands of the southern coast of the Arabian Peninsula, but they have not been dated with much certainty. The Yemen, my suggestion for the landing point of the single exodus, has not received much recent archaeological attention, but stone tools similar to those of the African late Middle Stone Age have been found there.16
The early Australian archaeological dates and the suggestion of even older colonization of New Guinea have been reinforced by an extraordinary reappraisal of the Kota Tampan Palaeolithic culture found in Lenggong Valley, in Perak on the Malay Peninsula, two-thirds of the way from Africa to Australia. This culture first identified by the find of large, curious, and rather crude pebble tools, fashioned on one side only, was thought by archaeologists in the 1960s to be the work of an earlier human species. When the geological layers surrounding the tools were reassessed, it became clear they were more recent. Wider interest was sparked in 1975, when Tom Harrison, the colourful curator of the Sarawak Museum, tried to relate the tools to the great eruption of the Sumatran volcano Toba. His suggestion, based on examination of the volcanic ash that surrounded the tools and the site, set off a hare which has only recently slowed down enough for others to get a good view.17
No one has done more research into Kota Tampan and the Lenggong Valley culture than archaeologist Zuraina Majid, of the Universiti Sains Malaysia in Penang. Her extensive work at a number of sites in the Lenggong Valley suggests that a local pebble-tool culture may have existed from the days of that great Toba volcanic eruption right up until 7,000 or even only 4,000 years ago.18 If that is so, it may provide the answer to one of the most nagging questions about the unifacial oval pebble tools: who made them? On the face of it, these are by no stretch of the imagination sophisticated tools. Better-looking tools were made long before in Africa and Europe by archaic humans, so why should anyone think that the unifacial pebbles encased in volcanic ash had been made by modern humans living at the time of Toba?
Two of the highest authorities on the Southeast Asian Palaeolithic, Australian archaeologists Peter Bellwood and Sandra Bowdler, agree with Zuraina Majid and Tom Harrison to the extent that these tools were most likely made by Anatomically Modern Humans. For a start, the dates for most of the pebble tools found in the Lenggong Valley are too recent for them to have been made by anyone else. Second, no pre-modern humans have ever been found in the Malay Peninsula, let alone the Lenggong Valley. Zuraina’s trump card in this respect is the much publicized finding by her team of ‘Perak Man’ in the Gunung Runtuh cave in the Lenggong Valley in 1990. Surrounded by the same class of pebble tools, this complete skeleton of a modern human was described by experts as having Australo-Melanesian characteristics. He was about 10,000 years old. This clear recent association of pebble tools with modern humans undermines the argument that the Kota Tampan pebble tools were too crude to be the work of moderns. The same locality also provides a continuity link with the older tools, which Zuraina argues is supported by technical comparisons. So, for the moment at least, Perak Man is the best local evidence that the older pebble tools encased in ash were made by the same (modern) human species.19
Another venerable expert on the archaeology of Southeast Asia is Richard Shutler. He makes the more general point that these kinds of tools were first brought to Island Southeast Asia by Homo sapiens about 70,000 years ago. (Island Southeast Asia, or ISEA – as opposed to Mainland Southeast Asia – means all the islands of Indonesia and the Philippines, excluding West New Guinea but including all of Borneo.) Shulter cautions against the view that such tools reflect cultural backwardness, agreeing with others that the quality of the available raw material determined what could be used for tools, and that for more sophisticated implements such as knives, bamboo was more likely to have been used.20
So how old was that Kota Tampan ash? When it was first dated, several decades ago, the result came out at 31,000 years old. This date for ash from the Toba volcano has always worried geologists, and even archaeologists such as Peter Bellwood.21 The trouble is that Toba did not undergo a massive explosion at that time. Toba’s last big bang, the largest explosion in the world in the past 2 million years, came much earlier, 71,000–74,000 years ago. More recently several geologists, including the one who did the original dating, have agreed that the ash surrounding the tools was indeed 74,000 years old. The dating is critical. If the Kota Tampan pebble tools were made by modern humans, they would be the oldest precisely dated evidence for modern humans outside Africa. It therefore looks as though the ancestors of the Australians could well have left Africa and arrived in Malaysia on their beachcombing trail before the great Toba explosion.
Perhaps more important than the precision of the dating, this connection between stone tools and ashes in Malaysia puts the first Indians and Pakistanis in the direct path of the greatest natural calamity to befall any humans, ever. The Toba explosion was that disaster, the biggest bang in 2 million years. Carried by the wind, the plume of ash from the volcano fanned out to the north-west and covered the whole of the Indian subcontinent (see Figure 4.3). Even today, a metres-thick ash layer is found throughout the region, and is associated in two Indian locations with Middle and Upper Palaeolithic tools.22 An important prediction of this conjunction of tools and ash is that a deep and wide genetically sterile furrow would have split East from West; India would eventually recover by recolonization from either side. Such a furrow does exist in the genetic map of Asia, as we shall see.
In spite of the proximity of Toba to Perak, the Toba ash plume only grazed the Malay Peninsula. The occupants of Kota Tampan were the unlucky ones – others on the peninsula escaped. Some argue, on the basis of comparing skull morphologies, that the Semang aboriginal ‘Negrito’ hunter-gatherers, who still live in the same part of the dense northern Malaysian rainforest, are descendants of people like Perak Man. The continuity of the Kota Tampan culture as argued by Zuraina Majid provides a link back to the 74,000-year-old tools in the Toba ash. In the next chapter we shall look at the physical evidence that links the Semang to the first modern human explorers along the coast of the Indian Ocean.23
For a film documentary, The Real Eve (Out of Eden in the UK), with which this book is associated, Discovery Channel helped to fund a genetic survey of the aboriginal groups of the Malay Peninsula which I conducted in collaboration with English geneticist Martin Richards and some Malaysian scientists (see Plate 16). This survey was part of a much larger ongoing study of East Asian genetics. The mtDNA results were very exciting: three-quarters of the Semang group (i.e. the ‘Negrito’ types) have their own unique genetic Manju and Nasreen lines with very little admixture from elsewhere, which is consistent with the view that their ancestors may have arrived with the first beachcombers. Their two unique lines trace straight back to the Manju and Nasreen roots (the first two daughters of Eve outside Africa – see Chapter 1). Their Manju line is not shared with anyone else in Southeast Asia or East Asia (or anywhere else) and, although it has suffered severe genetic drift through recent population decline, retains sufficient diversity to indicate an approximate age of 60,000 years. Their unique group on the Nasreen side, R21, comes from Rohani, Nasreen’s genetic daughter. This lack of any specific connection with any other Eurasian population is consistent with the idea that after arriving here so long ago, they have remained genetically isolated in the jungles of the Malay Peninsula.24
Another piece of evidence from the region may help place Anatomically Modern Humans in the Far East over 70,000 years ago. This is the famous southern Chinese Liujiang skeleton. Consisting of a well-preserved skull (see Plate 19) but few other bones, Liujiang was discovered in a small cave at Tongtianyan in the Guangxi Zhuang Autonomous Region in 1958 by people collecting fertilizer. There is no doubt that this person was anato
mically modern, but from the start there has been controversy over its age. A uranium date of 67,000 years was reported, but has been questioned on the basis of its exact location in relation to dated geological strata. In December 2002, a Chinese group headed by geologist Shen Guanjun reported their reinvestigation of the stratigraphy of the cave and dating of the skull (extending to several neighbouring caves) and claim it should be placed in a time bracket between 70,000 and 130,000, and not less than 68,000, years ago. The skull was found in a so-called intrusive breccia, a secondary flow of debris containing jumbled material of different ages. From their paper in the prestigious Journal of Human Evolution, the lower date bracket of 68,000 years seems solid, since it comes from multiple date estimates of the flowstone above and covering the breccia. (A flowstone forms when flowing water deposits calcite down a wall or across a floor.) Their preferred dating of 111,000–139,000 years ago based on unstratified fragments of flowstone and calcite within the breccia seems to me more speculative. Whatever the outcome, the safer lower bracket for the presence of an anatomically modern human in the Far East additionally tends to support the early exodus across the Red Sea before the Toba event of 71,000–74,000 years ago. This is because, to be in China by 68,000 years ago, their ancestors would still have had to expand across a quarter of the globe.25