Published in ‘Human settlement’, in D. Denoon (ed) The Cambridge History of the Pacific Islanders. pp.41-50. Cambridge University Press, Cambridge.
Sandra Bowdler
Department of Archaeology
University of Western Australia
The southwest Pacific has not always had the physical configuration it now has. During the Pleistocene era, commonly known as the Ice Age, massive glaciers periodically covered parts of the earth's surface between about 2 million and 12 thousand years ago. This coincides approximately with the time it took for the first human ancestors to appear in Africa, evolve into modern humans (Homo sapiens sapiens) and colonise the major land masses of the world.
Glaciers did not have much direct impact in the southwest Pacific, with the exception of the highest peaks of New Guinea. There were however significant indirect effects. As a result of such wholescale freezing up of the world's water supply, sea levels fell. Thus, some of what are now islands were at these times of glacial maxima part of their neighbouring larger land masses; this applies to what are termed continental islands (such as New Guinea and Tasmania), but not to what are termed oceanic islands (such as New Ireland, the Solomons and the islands of eastern Melanesia and Polynesia). As a result, the southwest Pacific was dominated by a much expanded land mass which we shall refer to as Australasia, and many of the distances between this expanse and the smaller islands were reduced, as were distances between the islands themselves.
Never, however, was Australasia, or "Sahul land"' linked by land to Southeast Asia, "Sunda land". Australasia has been isolated by a water barrier since at least the end of the Cretaceous some 60 million years ago, at the dawn of the age of mammals. As a result, this region's mammals are of the phylogenetically primitive sort known as marsupials and monotremes, as distinct from the more advanced placental mammals (or eutherians) of the so-called Old World (Africa, Asia and Europe). Because of this, if we accept the basic tenets of evolutionary theory and its placement of humans not only amongst the eutherians, but also amongst the primates, we must look for the origin of our species elsewhere. Many of the Aboriginal people of Australia believe they have always inhabited that continent. Evolutionary scientists look towards Southeast Asia for a colonising source, and note that it must have depended on the ability to cross water barriers.
The modern view on human evolution is generally as follows. The ancestors of human beings diverged from the ancestors of the chimpanzees and gorillas relatively recently, perhaps within the last 5 million years. The oldest identifiable human ancestor is probably one of the Australopithecines, dated to c.3.5 million years ago in Africa. The oldest member of the genus Homo, considered by some to be Homo habilis, appeared about 2 million years ago, along with identifiable, patterned stone tools. About 1.6 million years ago, Homo erectus, the species usually considered immediately antecedent to modern humans, appeared, again in Africa. It is however H. erectus who appears to have been the first hominid to spread across the globe, with their remains being long known from China ("Peking man") and Java ("Java man", formerly Pithecanthropus erectus).
The question of the emergence of modern humans is currently controversial. On the one hand, some scholars consider that Homo erectus in different parts of the world evolved into regionally distinctive populations of Homo sapiens; this is the "Regional Continuity" view. Other scholars believe that modern humans evolved only once, in Africa, and spread all over the world, replacing
the more primitive H. erectus populations in other regions. This is, not surprisingly, known as the "Replacement" hypothesis. This replacement is thought to have occurred within the last 100,000 years.
With respect to Asia and the Pacific, adherents to the regional continuity model see two relevant lineages. In China, a line of descent is envisaged leading from early Homo erectus populations represented at the hominid fossil sites Lantian and Yuanmou through the more evolved Zhoukoudian specimens to more sapient forms such as Dali and Maba and eventually modern Asian ("Mongoloid") populations. In Southeast Asia and Australasia, the lineage leads from the Javanese Homo erectus forms of the earliest Pucangan formations, through the younger Trinil then Ngandong fossils to modern Australian Aboriginal ("Australoid") populations. Fossil evidence from Australia is considered by some to support this argument.
With respect to the replacement model, much the same evidence is interpreted differently, with the relatively recent colonisation of Australasia taken to support the movement of fully modern humans out of Africa and across the globe. We will leave this topic for the time being, and return to it after the archaeological, that is to say, cultural rather than biological, evidence has been considered. For all these reasons, it should at least be apparent that the colonisation of the southwest Pacific needs to be considered as part of the general story of Asian prehistory.
When we turn to a consideration of the archaeological record, some technical matters should be borne in mind. Firstly, we should be aware of the limitations of our dating techniques. Radiocarbon dating has revolutionised our grasp of the past, and has been the main tool in throwing a chronological grid over our data. It is however only able to cover the last 40,000, or perhaps 50,000, years. The new technique of thermoluminescence dating as applied to sediments promises to expand that frame, but is still controversial in the interpretation of its results so far.
With respect to dating the human presence, we further need to be absolutely clear as to what is providing the date, and what the event or activity is which we wish to know the age of: these might not be the same, and here also lie problems. Radiocarbon dating dates the time of death of the organism providing our sample. If we date a sample of charcoal, we must be sure that either it derives from the human burning of the original wood, or that that burning was contemporaneous with some human activity. We thus need to be sure of the firm association between the dated sample and the human presence; this generally calls for an informed archaeological assessment of site integrity and such matters.
The question of the human colonisation of a hitherto unoccupied region demands some assessment of "the earliest", but this is fraught with archaeological difficulty. Not only do we need to sift the evidence carefully with respect to the foregoing considerations, but we are also hampered by the large statistical standard deviations inevitably attached to our dates, and we can never be sure of all the ways in which the vagaries of time and environment have differentially spared or destroyed our original evidence. One of the most obvious considerations which will have biassed our record is the effects of the post glacial sea rise.
It can safely be asserted that humans had reached Australasia 40,000 years ago, establishing a continuous presence which continues up to the present. There is argument amongst archaeologists as to whether an earlier presence can be established. Thermoluminescence dating of sediments in the Arnhem Land (Northern Territory, Australia) rockshelter site of Malakunanja II shows them to have been deposited between 50 and 60,000 years ago. There is not clear evidence however that those sediments are associated firmly with evidence of a human presence. It is also possible that thermoluminescence time is not the same as radiocarbon time[1]. Other archaeological sites in Arnhem Land show a human presence no older than 24,000 years.
Elsewhere in Australasia, several archaeological sites are dated to between 40,000 and 35,000 years. The Huon Peninsula on the north coast of New Guinea has evidence of human occupation said to date to 40,000 years ago, but there is some dispute about this[2]. Also on the north coast of New Guinea, occupation of the Lachitu rockshelter seems firmly dated to 35,000 years ago.
The Australian mainland (as it is today) shows human occupation of a similar age in the north, at Carpenters Gap 1 in the Kimberley (39,000 BP)[3] and Nurrabulgin Cave in Cape York Peninsula (37,000 BP); in the southwest, at Upper Swan on the outskirts of metropolitan Perth (38,000 BP); in the southeast, at Lake Mungo and associated sites in the Murray-Darling basin (37,000 BP); and also in Tasmania, at the cave site Warreen (35,000 BP)[4].
These may or may not represent the earliest occupations of these areas, given the problems mentioned above, and noting particularly that radiocarbon dating does not cover periods much older than 40,000 BP. It may however be noted that dates of this sort of age occur only at the bottom of deep deposits, not on top (as in many sites in Europe, Africa and the Middle East), which would indicate a long undated record of occupation.
If the dates above are accepted as evidence of earliest occupation, does this indicate a rapid colonising process? This is unfortunately inestimable. The standard deviations of radiocarbon dates indicate the range of time which such dates may actually represent. The proper radiocarbon date for Lachitu is 35,360±1400, which indicates that the actual date falls somewhere within the range 36,760 and 34,200, or, to be 98% sure, between 39,160 and 32,800. The date thus embodies some 7000 years of uncertainty. All we can really say is that in terms of archaeological time, colonisation of the new continent was extremely rapid.
When we look, as we naturally do, to Southeast Asia for the antecedents of this colonising push, we are faced with a puzzle. The famous "Java Man" fossils were discovered here a century ago, and now take their place as one of the furthest flung examples of Homo erectus, believed to originate in Africa at least 1.6 million years ago. The dating of the Javanese specimens is controversial. Most would agree with a date of a little under 1 million years for the oldest examples, although radiometric dates of nearly 2 million years BP have been obtained. Even more problematic are the Ngandong or Solo forms, considered to be a more evolved form of Homo erectus, or perhaps even an archaic Homo sapiens; they are considered to date from anything between 600,000 to 60,000 years ago, with a general view that between 300,000 and 100,000 years is probably most likely.
One aspect of the puzzle is that there is no cultural evidence whatsoever associated with these fossils, nor is there any such evidence anywhere in this region, not only Java but throughout Southeast Asia (i.e. south of China) which can be confidently dated to that period of time. In fact, there is no archaeological evidence from Southeast Asia which is older than that found in Australia, i.e. 40,000 years. This evidence is remarkably similar to that from Australasia in its distribution. In peninsular Thailand, human occupation of Lang Rong Rien cave has been dated to c.37,000 BP; in northern Vietnam, the oldest of a series of rockshelters is dated to c.33,200; and in peninsular Malaysia, in Perak, the Kota Tampan stone workshop site is dated to c.31,000 BP. Similar dates are found in island Southeast Asia: the Niah Cave in Sarawak, in the north of the island of Borneo, is dated to c.40,000 BP; the rockshelters Leang Burung 2 in southern Sulawesi is dated to c.31,000; and Tabon Cave on the Philippine island of Palawan has human occupation by 30,000 years ago[5].
This evidence, taken overall, suggests two things. Firstly, we may entertain the hypothesis that, regardless of the older fossil material, Southeast Asia was colonised by modern humans at the same time as Australasia. Secondly, this colonising event was, in archaeological time, extremely rapid. It leaves open the issue of whence came the colonists; at the present time, the most likely source appears to be China.
The story envisaged here is that modern humans swept out of southern China to find their way with archaeological rapidity to the mainland and islands of Southeast Asia, the continent of Australia and at least part of the island world of the Pacific. In real time, this process could have occurred in anything between 50 and 5000 years; we currently have no way to assess this. There is also little to help us explain why this occurred at this time, and no earlier. It has been said that it is due to the uniquely developed capabilities of Homo sapiens sapiens, fully modern humans, but this begs as many questions perhaps as it answers. Obviously, the new immigrants needed the adaptive skills to cope with a variety of new environments. Another essential prerequisite was of course adequate watercraft and navigational skills.
During the Pleistocene, as we have seen, there were times of lowered sea level, and hence of expanded land masses. To get to Australasia however there were always water barriers to be crossed. To reach Vietnam, Thailand, peninsular Malaysia and indeed Java, Palawan and Borneo did not require water crossings at times of low sea level. To reach Sulawesi did however necessitate watercraft, with a crossing of perhaps 50 km. To reach the expanded land mass of Australasia, several crossings were needed, of a maximum of 100 km. The new colonists did not however stop there. Archaeological evidence showed they went beyond the Australasian land mass, to the oceanic islands of New Ireland in eastern Papua New Guinea, by 33,000 years ago, and to Buka, the northernmost in the Solomon Islands chain, by 28,000 years ago.
These feats of voyaging suggest many questions. What kind of watercraft were used, what routes were followed, were such voyages accidental or deliberate? To some extent, these arise from a continued disparaging and perhaps Eurocentric view of the capabilities of our forbears. In Europe, it is still a matter of debate whether pre-agricultural humans had the ability to voyage to Cyprus or Sardinia, Mediterranean islands requiring maximum water crossings of less than 50 km.
We have no idea what kind of watercraft may have been used at this early time. No archaeological evidence, direct or indirect, exists to shed light on this question. The sea-going canoes of the Pacific, outriggers and dugouts, known from the historical period are of unknown antiquity. It is however usually assumed that they date from relatively recent times, consistent with the Austronesian expansion [see Spriggs, this volume?]. In Australia, it has usually been assumed that that expansion made little or no impact[6]. It might be thought therefore that Australian watercraft of the ethnographic present reflect a Pleistocene inheritance. We need to discount the dugouts of the north coast and the outriggers of the northeast, as these are thought to reflect the recent influences of Macassans and Papuans[7].
We are thus left with a variety of bark canoes and log rafts which were made and used by Aboriginal people in different parts of Australia. They do not generally seem capable of long ocean voyages, and indeed there is some archaeological evidence to suggest they are none of them of any great antiquity. A survey of archaeological and ethnographic evidence shows that offshore islands visited by Aboriginal people by sea during the Holocene with such craft involved water crossings of no more than 25 km, with most crossings being less than 10km. Furthermore, the earliest dates for such crossings are, with perhaps two or three exceptions of 26 dated instances, only within the last 4000 years[8]. This suggests therefore that the maritime technology observed in recent times in Aboriginal Australia was not that used to colonise the Pacific originally.
It has been suggested that the original voyages of Pacific discovery were accomplished with bamboo craft, perhaps rafts[9]. Since extensive stands of large bamboo probably did not occur indigenously in Australia, this could explain the lack of survival of this technology into recent times.
The possible migration routes out of Southeast Asia into Australasia have been canvassed in detail by Birdsell and more recently by Irwin[10]. These routes are based on the assumption that routes involving the shortest possible crossings will have been the most likely, and perhaps the most favoured; they are thus essentially variations on island-hopping. Birdsell suggested there were two main likely routes, one leading from Java through Timor to northern Australia, the other from Sulawesi through Halmahera to West Irian. He further assumes that such voyages would have been easier, and thus more likely to have been successful, during times of lowered sea level, when distances between islands would be lessened.
Irwin suggests that this is not a necessary assumption. He argues that "to make fine distinctions between the different distances is to miss the point that they were probably all short enough for the risks to remain much the same ... a boat that is seaworthy enough to cross 10 nautical miles can probably cross 100 or more, provided it is not of a type that becomes waterlogged and provided the weather remains the same"[11].
Irwin also addresses the issue of accidental versus deliberate voyaging. This has long been a matter of debate, with respect to the more recent Holocene colonisation of the far Pacific, as well as the earlier period under discussion here. It would now be generally agreed that, as a number of computer simulations have shown, any minimalist scenario involving the least possible number of people would have been demographically doomed to extinction. It seems that intentional voyaging is much more likely to have been the order of the day. As to the motives of such expeditions, we can of course only guess. Various ideas have been offered up, including population pressures in the hypothetical source area(s), and environmental episodes driving people in search of alternative resources.
The nature of the archaeological record does not lend any great weight to these kinds of suggestion. The patterning of dates in geographical space may (with the caveats adumbrated above) be interpreted as representing an extremely speedy colonisation of the region. A theoretical colonisation model, based on expansion under duress, would posit a noticeable "gradient of antiquity"[12], with oldest dates in mainland Southeast Asia, which would have included the modern islands of Java and Borneo in the Pleistocene. Younger dates would be expected in Sulawesi and the other non-continental islands, and younger still dates in continental Melanesia. We would then expect to find dates in northern Australia for early colonisation either contemporary with, or younger than, those in New Guinea, with successively younger dates in southern Australia and oceanic Melanesia respectively. This is not the case; the evidence does not suggest a gradual settling and "filling up" of new islands and ecological zones. There is no support here for a model of gradual population buildups with ensuing pressure on resources forcing migrations onwards into new regions.
Some other motivation must have driven early humans to make deliberate and purposeful voyages of discovery. This may lay in the realm of unguessed at ideologies, social upheavals, or some aspect of economic endeavour beyond the exigencies of subsistence. Most archaeologists would agree that the people we are considering at this time were hunter-gatherers, dependent on wild plant and animal resources. While however it used to be thought that the hunter-gatherer (or "forager") way of life was, economically speaking, a somewhat random and difficult one, it has been recognised for some time now that such people had extremely sophisticated and systematic economic strategies, passing beyond mere subsistence. It is possible that such strategies are indeed the hallmark of modern humans, and may explain their extraordinary adaptive success[13].
Given their evident maritime abilities, it is logical to suppose that the early Pacific colonisers were coastally adapted, and were probably fishers and shellfish gatherers as much as hunters and plant gatherers[14]. Many early sites do not preserve the kinds of organic remains which might enable us to determine this, but it is supported by such evidence as is available. In the oldest levels of the Mandu Mandu Creek Rockshelter (Northwest Cape, Australia) and Matenkupkum (New Ireland), both about 33,000 years old and both located near their respective Pleistocene coastlines, the remains of fish bones and shellfish are preserved. At Lake Mungo and other Willandra Lakes sites, there is considerable evidence for freshwater fish and shellfish being part of the diet from 37,000 years ago.
Other sites of similar age are however located at considerable distances from the then coastline, and it is evident that the early immigrants had the ability to exploit a wide range of resources in unfamiliar terrain, including what must have been previously unknown, not to say strange, resources. The range of ecologies brought into the economic sphere by 35,000 years ago is indeed remarkable. What we must assume to have been tropical dwellers accustomed to the products of the Asian forests and savannahs are by this time seasonally targetting wallabies on the edge of the Tasmanian glaciated highlands and collecting emu eggs on the edge of the Western Australian desert.
We may also be seeing complex economic systems at work by this time. Evidence from New Ireland suggests that seafaring humans may have been introducing mammals into new environments during the Pleistocene. Intriguing evidence from the Kilu site on Buka Island (north Solomon Islands) shows the presence on starch grains on stone tools dated to c.28,000 years ago; the starch has been identified as deriving from a species of taro. It is still open to argument as to whether taro might be indigenous to Buka, or whether indeed it had to have been brought by humans. These areas need further work and replication, but they do hint at the deliberate human dispersal of both plants and animals during this early colonising period. This may in turn be open to several interpretations. On the one hand, we may be dealing with extremely sophisticated environmental managers, which is not necessarily to call them "agriculturalists". On the other hand, such dispersals may have been the by-product of providing sufficient provisions for extended sea voyages.
There are also some early suggestions of the movement of hard goods amongst early colonists over considerable distances. In Australia, rockshelter sites on what is now the coast of the Kimberley region, contain pieces of baler shell dated to c.28,000 BP; at that time, the coast would have been over 50km away. Later in the sequence, pearlshell as well as baler is found in levels dated to c.18,000 BP, when the sea was some 200 km away. Further to the south, a site at Shark Bay has also produced baler shell dated to c.30,000 BP, and here also the coast would then have been some 100km to the west. In several areas of Australia in the ethnographic present, both pearlshell and baler shell are known to have been significant items in long distance trading patterns. Pearlshell was a generally of ritual significance; baler shell was also sometimes considered in this way, but also had a pragmatic function, in often being used as a receptacle. We cannot of course conclude that these items held such significance 30 to 18,000 years ago, nor can we be sure that they imply trading patterns similar to those of recent times. They do however show that the early colonists who had begun to exploit the interior resources of the new continent were in some way maintaining links with the coast.
In southwest Tasmania, the Darwin meteorite impact crater contains small seams of glass produced by the collision of the meteor with local rocks. Small pieces of Darwin glass in the form of flakes and tools has been identified in dated archaeological deposits up to 28,000 BP. The distance of the sites from the Darwin crater, in terms of the actual routes humans would need to take in this terrain, varies from 25 to over 100km.
Natural glass occurs also on several Melanesian islands, in the form of volcanic obsidian. This has been known to be an important trade item in recent times, and archaeological research keeps extending its significance back in time. New Ireland sites contain obsidian demonstrated to derive from New Britain sources in deposits dated to 20,000 BP. The distance from the source is some 350km in a straight line, involving a minimum sea crossing of 30 km. Small but consistent amounts of this material were deposited over a period of some 2000 years.
All these instances suggest that the early colonists not only made rapid adaptations to new and unfamiliar environments, but also quickly established local regional networks which allowed them to maximise the use of new resources. Again, this is far from the idea of desperate drift voyages, accidentally happening into new lands, or even of deliberate but desperate explorers driven by the vagaries of environmental change.
Even if we cannot identify with any certainty the driving forces behind these pioneering exploits, we should at least be able to trace their origin. This however is also more difficult than might be expected. There are two kinds of evidence we might turn to, the cultural and the biological. In the first instance, we would expect that the colonists would bring with them some form of identifiable "baggage" which should be traceable to its source. On the other hand, it has always been thought that the physical morphology of the newcomers should also indicate their relationship to possible source populations. In both cases, the evidence is both scanty and subject to differing interpretations.
At most of the known early sites in the region, the only clearly identifiable cultural evidence comes in the form of stone artefacts. These present many problems for the archaeologist, some intrinsic to the nature of such evidence, some compounded again by predetermined ideas based on Old World experience. Stone artefacts are known from the earliest sites in Africa at which human ancestors have been identified at some 2 million years ago, and have indeed been taken as indicative of early human status, that is, of the genus Homo. These early assemblages make up what is known as the Oldowan industry (after Olduvai Gorge). Examples of the Oldowan industry occurs in many sites as a patterned, recurrent set of types, which include pebble choppers and smaller tools, called flake tools and scrapers.
In Africa, the emergence of the immediate pre-human ancestor, Homo erectus, is indicated by a new range of tool types typified by large bifacial core tool called a "hand axe", which is known in general as the Acheulian industry. As the name might indicate, the Acheulian is also known from Europe, and also the Middle East and parts of India. It is not however known from east Asia. Assemblages from China, which can be assumed to be contemporary with Acheulian sites to the west, are characterised by a range of flake tools, with some pebble choppers. As discussed above, the Homo erectus fossils of Java have no associated, nor identifiably contemporary, cultural manifestations at all. In Europe, the Acheulian industries are succeeded by the Mousterian industries associated with Neanderthal humans (Homo sapiens neanderthalensis), in turn succeeded by the Upper Palaeolithic industries associated with fully modern humans (Homo sapiens sapiens). All these later assemblages are assumed to demonstrate ever-increasing complexity and refinement. In China on the other hand the stone tool assemblages continue throughout the Pleistocene with few observable changes.
The traditional interpretation of these differences has been, perhaps predictably, directional and indeed judgemental. East Asian stone tool industries have been considered to be lacking in progressive change through time compared with the African and particularly European sequences. While it is not usually stated very clearly, there is little doubt that it is considered that the Asian industries are simply a continuation of an original Oldowan type of early human industry, stagnating in an evolutionary cul-de-sac.
These traditional views are not entirely motivated by blind prejudice, of course. We must recognise in fact the extremely intractable nature of stone artefacts as an informational medium. Stone is an inflexible medium, unlike for instance pottery, which is plastic in the extreme, and which also lends itself to a variety of pragmatically inessential decorations, thus making it an excellent medium for identifying non-functional traits, which might be regarded as stylistic only, and hence capable of reflecting cultural identification. Stone artefacts are produced by percussion, technically known as knapping or flaking, but basically meaning that rocks are struck in some way to detach and reduce, which means a comparatively limited repertoire of forms can be produced. Archaeologists are faced with a number of problems in dealing with stone tools. Given their ubiquity in pre-agricultural sites, we are sadly limited in our knowledge of how they were used. This compounds the problem of identifying what characteristics are essentially functional, and what might be considered to be stylistic, and hence culturally determined. In recent years, our understanding of the technical processes involved in producing stone tools has grown considerably, and residue analysis is beginning to lead to a better understanding of stone tools functions. Stylistic studies have however fallen into a certain deal of disfavour, largely because of the problems described.
Observations on the earliest Pacific stone industries are hampered by all these difficulties, as well as the fact that much of the new material awaits detailed publication. Comparative studies are in train, but few results are as yet available. Some general comments can be made, but these must be tempered with the understanding that research is in an early stage. The stone tool industries of early Australian sites have been grouped under the title of the Australian Core Tool and Scraper Tradition. This particular entity was based on a consideration of artefacts largely collected from the surface of the Lake Mungo site identified in 1970. There has since been considerable dissatisfaction at this concept, largely as it is felt to imply a unity which may not exist, and perhaps mask a perceived variety in artefact assemblages from Australasian Pleistocene sites. Pending more detailed statistical studies, it can be observed that the artefact assemblages from early Australasian sites do share many similarities. Artefacts from sites including those in the Kimberley, Shark Bay, Cape York, Tasmania and New Ireland which are 30,000 years old or older appear as part of a common tradition, albeit one which from a European perspective would be characterised as amorphous, or (being slightly more polite) "ad hoc". The technical procedures involved would seem to have been identical in all cases. Some types are recurrent, namely small steep edge scrapers and thumbnail scrapers.
Similar attributes characterise the stone artefact assemblages from Southeast Asian sites of similar age. As argued above, the complete lack of earlier cultural evidence from Southeast Asia implies a colonisation by modern humans contemporaneous with that of Australasia, c.40,000 years ago. The nature of the cultural, that is, artefactual, evidence, only suggests one possible source for this colonisation: China. Chinese Pleistocene artefact assemblages do not appear to change significantly through time, and the very earliest Chinese assemblages, which are up to c. 1 million years old, bear a remarkable resemblance to those of Australasia, as do those from China of more recent antiquity, up to c.20,000 BP.
The biological evidence presents problems of a different order. Human remains of Pleistocene age are not common in any part of the world, and Australasia is no exception. No examples are currently known from Melanesia and there are very few from Southeast Asia apart from the Javanese Homo erectus examples. The best described example is the skull from the Niah Cave (Sarawak), usually said to date from about 40,000 years ago; this dating has however been called into question, as it may be a much younger intrusive burial.
There are several collections of human remains from Australia which have been assigned a Pleistocene age. Many of these are not in fact well dated. The best known and studied are the two Lake Mungo burials dated to between 26 and 32,000 years BP, and the large Kow Swamp collection, dated to between 14,000 and 9000 years BP. All specimens are ascribed to the fully modern human species, Homo sapiens sapiens. The Mungo examples, and some others, are said to be extremely gracile (light boned), and the Kow Swamp and other examples also from the Murray Valley are said to be extremely robust (heavily boned). One interpretation of these differences is that the gracile specimens derive from the Chinese Mongoloid lineage, and that the Kow Swamp group derives from the Javanese Australoid lineage. This raises many problems, not least of which is the implication that the Kow Swamp group is considerably more archaic than the other group, and yet its antiquity is less. In this view, at least two separate colonising events need to be envisaged, one from Java and one direct from China.
An alternative interpretation is that the two morphologies do not represent such extreme differences, but that both fall within the range of variation of modern Aboriginal populations. Furthermore, it has been suggested that the characteristics of the Kow Swamp group have developed since late Pleistocene times only. In this view, only one colonising source population is required.
If we accept the overall view favoured here, one further problem remains. This view is that Australia and the southwest Pacific region were colonised some time within the last 40,000 (or perhaps 50,000) years by modern humans, Homo sapiens sapiens, who emerged from China and moved with extreme (archaeological) rapidity across Southeast Asia and Australasia. The Homo erectus populations of Java are not considered to have contributed to this colonising population, either culturally or biologically, and it is likely that they had in fact become extinct by this time.
This does however leave unexplained the origin of the physical characteristics thought to distinguish the populations of the region, those which define the "Mongoloids" of eastern and Southeast Asia and also Polynesia, and those which characterise the "Australoids" of Australia and much of Melanesia. To put this simply, it would appear to mean that Australian Aboriginal people of recent times, Melanesians, Polynesians, Southeast Asians and the Han Chinese are all descended from a common ancestral group within the last 40,000 years. This does not satisfy those who see these differentiating characteristics present in different Homo erectus populations, and consider they need more time than 40,000 years to have become so differentiated. It seems however that, firstly, most of the characteristics used to define these groups are as much soft tissue differences as anything else, and thus have not been preserved from any great antiquity. There is furthermore discussion as to how significant such differences are adaptively, and how rapidly they may change under changing environmental circumstances. We really do not know how long it takes such characteristics to become distinctive.
Some 40,000 years ago, fully modern humans embarked on a series of voyages which brought them out of China, across Southeast Asia and into the new lands of Australia and Melanesia. It is likely that archaeological research will show in due course that this colonising movement reached as far as the southern end of the Solomon islands chain, but probably not beyond. Adaptations to an extremely wide range of environments appear to have been rapid, with the sophisticated economic networks set up with remarkable rapidity. The archaeological record shows continuity from the earliest times, up until the end of the Pleistocene. This is not to say however that no further voyages took place. It is interesting to speculate as to whether there were return voyages or not; at this time it is not possible to comment on the basis of the archaeological record.
In Australia, there is evidence to suggest that certain dislocations in population took place during the height of the last glaciation c.18,000 years ago, with increased aridity making previously comfortable environments marginal for human existence. It is not clear whether such dislocation, or other effects, took place in Melanesia. It is clear however that any such problems were overcome insofar as populations survived and continued to expand and respond to new challenges.
BIBLIOGRAPHIC ESSAY
A useful volume for the Pleistocene archaeology of Australia, New Guinea and Island Melanesia is Sahul in Review, edited by M. A. Smith, M. Spriggs and B. Fankhauser, Canberra 1993 (Occasional Papers in Prehistory No. 24). This consists of papers given at the Annual Conference of the Australian Archaeological Association in 1991, some of which are review articles, while others contain useful archaeological data. It also includes a bibliography of Pleistocene archaeological sites in the region, compiled by M. A. Smith and N. D. Sharp. Somewhat older but still useful reviews are those of Jim Allen ('When did humans first colonise Australia?', Search 20 (1989) and myself ('Homo sapiens in Southeast Asia and the Antipodes: archaeological vs biological interpretations', In T. Akazawa et al, eds, The Evolution and Dispersal of Modern Humans in Asia, Tokyo,1992). The latter essay also contains a review of, and references to, the Asian evidence referred to here.
Most of the primary evidence is to be found in specialised journals and monographs, but much of it, including material in theses, has not yet been published in detail. Relevant journals include Australia Archaeology, Archaeology in Oceania and Australian Aboriginal Studies, and the English journal Antiquity has recently included several relevant papers. The Department of Prehistory of the Research School of Pacific Studies at the Australian National University in Canberra publishes importatant monographs in its Terra Australis series, as well as relevant volumes (such as Sahul in Review) in its Occasional Papers in Prehistory.
There is a vast and ever growing literature on the emergence of anatomically modern humans, Homo sapiens sapiens. The main opposing viewpoints are represented by M. Wolpoff (regional continuity) and C. B. Stringer (replacement), both of whom have papers in the following collections: Paul Mellars and Chris Stringer, eds, The Human Revolution: Behavioural and Biological Perspectives on the Origins of Modern Humans, Edinburgh, 1989; Günter Bräuer and Fred Smith, eds, Continuity or Replacement: Controversies in Homo sapiens Evolution, Rotterdam, 1992.
Also of relevance is Erik Trinkaus, ed, The Emergence of Modern Humans: Biocultural Adaptations in the Later Pleistocene, 1989.
--------------------------------------------------------------------------------
[1] For this debate see R. G. Roberts et al., 'Thermoluminescence dating of a 50,000-year-old human occupation site in northern Australia', Nature 345, (1990); 'Early dates at Malakunanja II: a reply to Bowdler', Australian Archaeology 31 (1990); 'Stratigraphy and statistics at Malakunanja II: reply to Hiscock' Archaeology in Oceania 25 (1990); Peter Hiscock, 'How old are the artefacts in Malakununja II?', Archaeology in Oceania 25 (1990); Sandra Bowdler '50,000 year-old site in Australia - is it really that old?', Australian Archaeology 31 (1990); 'Some sort of dates at Malakunanja II: a reply to Roberts et al.', Australian Archaeology 32 (1991).
[2] A discussion of early dates in Australasia may be found in Jim Allen, 'When did humans first colonise Australia?', Search 20 (1989) and in Sandra Bowdler 'Homo sapiens in Southeast Asia and the Antipodes: archaeological vs biological interpretations', in T. Akazawa et al., eds, The Evolution and Dispersal of Modern Humans in Asia, Tokyo, 1992; 'Sunda and Sahul: a 30KYR BP culture area?', In M. A. Smith et al., eds, Sahul in Review: Pleistocene Archaeology in Australia, New Guinea and Island Melanesia, Occasional Papers in Prehistory no. 24, Canberra, 1993.
[3] I thank Dr Susan O'Connor for this information.
[4] See footnote 2.
[5] This evidence is reviewed in Bowdler, 'Homo sapiens in Southeast Asia and the Antipodes'.
[6] For suggestions otherwise, see Sandra Bowdler 'Views of the past in Australian Prehistory', in M. Spriggs and A. Thorne, eds, A Community of Cultures: Papers for Jack Golson, Canberra, 1993, and M. J. Rowland, 'The distribution of Aboriginal watercraft on the east coast of Queensland: implications for culture contact', Australian Aboriginal Studies, no.2 (1987).
[7] D. S. Davidson, 'The chronology of Australian watercraft', Journal of the Polynesian Society 44 (1935), and Rowland, 'The distribution of Aboriginal watercraft'.
[8] Sandra Bowdler, 'Offshore Islands and Maritime Explorations in Australian Prehistory', to be published in World Archaeology.
[9] J. B. Birdsell, 'The recalibration of a paradigm for the first peopling of Greater Australia', In J. Allen et al., eds, Sunda and Sahul: Prehistoric Studies in Southeast Asia, Melanesia and Australia, London, 1977.
[10] Geoffrey Irwin, The Prehistoric Exploration and Colonisation of the Pacific, Cambridge, 1992.
[11] Irwin, Prehistoric Colonisation 27-8.
[12] A concept expressed by Rhys Jones in 'The geographical background to the arrival of man in Australia and Tasmania', Archaeology and Physical Anthropology in Oceania 3 (1968).
[13] An idea explored further in my 'Peopling Australasia: the "coastal colonisation" hypothesis reconsidered', in P. Mellars, ed, The Emergence of Modern Humans: an Archaeological Perspective, vol.2, Edinburgh, 1990.
[14] Originally proposed in my 'The coastal colonisation of Australia', In J. Allen et al., eds, Sunda and Sahul.
|
