The first footprints

WHEN DID HUMANS FIRST SETTLE IN AUSTRALIA? TIM FLANNERY BOLDLY SETS A DATE.

Is the answer 47,000?

One day, perhaps, Australians will celebrate a different sort of Australia Day - one that commemorates the peopling of our continent rather than its European colonisation. It could be the kind of national day that every Australian could take pride in; for the arrival of the first people on our shore is significant not just to their descendents, but to everyone who seeks a prosperous future on our common island home.

If we are ever to mark such a day in our calendars and in our consciousness, we need to know when those first Australians actually arrived. The first professional archaeological dig in Australia was conducted in 1929, and despite the vast volume of work conducted over the ensuing seventy years the answer has eluded us. Yet ever since radiocarbon dating proved an astonishing antiquity for our species on this continent, researchers have acknowledged that Australia’s colonisation was one of the earliest human conquests; occurring long before the Europeans wrested Europe from the Neanderthals and over 30,000 years before anyone at all lived in the Americas.

Thus the triumph of those first Australians was an important milestone, involving the earliest known use of watercraft over significant distances and the first occupation of a new continent. Perhaps it even evoked the emergence of that pioneering spirit that has characterised our kind ever since.

We will probably never know why those first Australians set out on their journey so long ago. It seems beyond doubt, however, that their voyage was planned and carefully executed. Most likely they departed from the island of Timor, which means that they or their ancestors must have made eight water crossings of considerable difficulty on their journey from mainland Asia.

The culture that gave rise to the first Australians was thus no stranger to sea voyaging, but before these adventurous men and women lay a water crossing of at least 100 kilometres - far greater than any they or their ancestors had broached before. Yet it’s my strong sense that they were not setting off into the wild blue yonder; by observing migrating birds and smoke from fires, almost certainly they knew of the virgin continent that lay just out of view under the horizon. We must not forget, either, that the intellect of these people was every bit as acute as our own and their technology was not entirely rudimentary.

A wide variety of evidence can be employed in trying to determine when these people made their epic journey. We can of course seek to date their campsites and stone tools; but distinctive clues, too, reside in the changes they wrought in Australia’s ecology. This is because every species that establishes itself in a landmass not previously occupied by similar organisms leaves a unique ‘ecological imprint’ in the fossil record.

Accordingly, any future archaeologist interested in dating the European arrival in Australia would be well served by searching for the bones of cattle and sheep as well as those of Europeans. Likewise, researchers interested in the antiquity of the dingo in Australia date the extinction of thylacines and Tasmanian devils on the mainland, because the dingo drove those creatures into oblivion wherever it found them.

The nature of the ecological imprint of the first Australians is still hotly debated, but it is becoming increasingly apparent that humans have left an ecological imprint world-wide, and that the Australian evidence fits the global pattern well. That imprint, best attested to in places like New Zealand and North America, consists of the extinction of large animals (megafauna) and increased evidence of fire. I would argue that these phenomena can be thought of as indelible ecological markers left by the first invaders of new lands.

A body of new evidence relevant to dating the ecological imprint of the first Australians - as well as new dates of human campsites and tools - has convinced me that we can, for the first time, determine the time of arrival of those first people. Not that we will ever know the precise date, or perhaps even the century of that first Australia Day, but we are now at least in a position to establish, within the ‘error bar’ or uncertainty inherent in the dating techniques, the date to the nearest millennium.

A decades-long scientific effort - involving the entire research lifetimes of archaeologists, palaeoecologists, and dating boffins - lies behind the date I’m fixing upon here, and over the last three decades one man has reigned at the centre of this enterprise. He was Rhys Jones, a Welsh-born archaeologist whose vision and research were crucial to the application of new dating techniques in Australian archaeology, and whose work significantly pushed back the earliest date for humans in Australia. Rhys Jones died last September, prematurely, at the height of his research powers, and just before information vital to solving the riddle he devoted his life to studying was published.

In 1962, the year before Rhys Jones arrived in Australia, the received wisdom was that Aboriginal people had been here for just 13,000 years. A decade earlier it was just 5,000 years. Jones changed all that. A milestone came in 1990 when, together with Bert Roberts (an expert in a new dating technique based at the University of Wollongong), he released the finding that people had first occupied two rock-shelters in Arnhem Land as long as 53-60,000 years ago. The announcement electrified the scientific community, for it pushed back the Aboriginal occupation of the continent beyond the ‘radiocarbon barrier’ - the 35-40,000-year limit to the application of traditional radiocarbon dating.

Despite Jones and Roberts’ findings, when writing my ecological history of Australasia, The Future Eaters, in 1994, I had to concede defeat when it came to specifying a date for the first human arrival on the continent. It wasn’t that Jones and Roberts’ findings seemed unreasonable or faulty to me; rather it was because scientific results need to be replicated before an argument is clinched, and their dates stood alone. Moreover, establishing the nature of the human ecological footprint was very confusing, as was establishing just when the changes occurred.

I decided to concentrate on understanding the human ecological footprint, and in 1999 Bert Roberts and I teamed up to try to determine the time of extinction of Australia’s megafauna. These great creatures comprised at least 60 species of giant marsupials, reptiles and birds, the largest of which were the size of white rhinos. Their remains are widespread, obvious and reasonably common throughout the continent. I felt that if we could establish just when they went extinct, we could determine if the extinction fitted the global pattern where extinction follows human arrival, and also tell whether the event was synchronous continent wide or not.

Megafaunal extinction, however, was only part of the then putative ecological imprint. Palaeobotanists had detected evidence that they interpreted as a dramatic, human-caused change in Australia’s vegetation and fire history in sediments laid down in lakes and the ocean. It revealed a huge increase in charcoal, along with the abrupt extinction of fire-sensitive plant species and their replacement with eucalypts and other fire-lovers.

While both the extinction of Australia’s megafauna and the change in the botanical record were clear and widely recognised, dating them proved to be exceedingly difficult. When I wrote The Future Eaters we had no clear evidence at all of when the megafauna became extinct, and it was clear that new techniques would be needed before the paradox could be resolved. They were surprisingly quick to appear, and today many different dating methods are available to help answer the question. Three are of special importance: so-called A-BOX radiocarbon dating, Amino Acid Racemisation dating (known as AAR), and Optically Stimulated Luminescence dating (OSL).

AAR dating was the first to pay dividends, and in 1999 it provided a breakthrough. The technique is based on the fact that all amino acids (the building blocks of living things) are ‘left-handed’ or fold to the left, but that over time some amino-acids ‘flip’ to become right-handed. The rate at which this happens is known, so by measuring how many amino acids in a piece of organic matter have ‘flipped’, we can estimate its age.

The trouble with this technique is that it can be applied only where amino acids have been locked away from the outer environment. Bird’s eggshell is the perfect material to date using AAR, because proteins get trapped inside the calcium crystals that the bird secretes as it lays down the shell; sealing them forever from the outside world. Central Australia brims with fossil eggshells, some of which were laid by a giant extinct bird known as Genyornis, which looked somewhat like an enormous emu (but was actually related to waterfowl). In 1999 AAR showed that those giant birds abruptly stopped laying eggs 50,000 years ago - give or take 5,000 years. Some level of uncertainty is inherent in all dating, but the ‘error bar’ associated with these AAR findings - 10,000 years - is relatively large.

This finding was supported and refined in 2001, when the study initiated by Bert Roberts and myself was published. It used OSL dating to establish that many other kinds Australian megafauna died out abruptly, continent-wide, at 46,400 years ago - give or take a few thousand years.

OSL dating is an amazing technique, and over the past decade Bert Roberts has used it to transform our understanding of Australian prehistory. The technique is based on the discovery that when grains of sand are buried in the ground, they trap and accumulate electrons in their crystal lattice. Those electrons come from background radiation in surrounding rocks. When the sand grain is exposed to light, the electrons escape from the traps. By measuring the local background radiation and the number of trapped electrons, researchers can tell how long ago the grain was buried.

We dated grains covering the skeletons of giant marsupials from 27 locations across Australia to establish how long ago the creatures died. The most recent megafaunal sites we found were from the southwest of Western Australia and southeastern Queensland, strongly suggesting that the event was simultaneous across the continent. Furthermore, the bones or eight species of megafauna were recovered from those sites, suggesting that a considerable proportion of the fauna, at least, died out at once.

A research article announcing a breakthrough in dating the vegetation and fire was also published in 2001. It relied on A-BOX dating, which dramatically improves the reach and accuracy of radiocarbon dating by removing contaminants, thereby allowing reliable dates up to 60,000 years to be obtained. A brilliant young researcher from Queen’s University in Belfast, Dr Chris Turney, used the technique to reveal that the human-caused burning and vegetation change in Queensland occurred 47,000 years ago, give or take a couple of thousand years.

It’s the archaeological sites themselves - containing evidence of a human presence - that have proved more difficult to reconcile, perhaps in pert because of the intense rivalry between archaeologists. In 1996 researchers announced that humans had lived in the Kimberley from as long as 116,000 years ago. This date, based on an earlier trapped-electron technique, has now been discredited, and we know how the mistake was made. Other older dates are not as easy to resolve. Among the most stubborn are the ones obtained by Rhys Jones from those excavations in Arnhem Land, indicating a human presence in Australia between 53 and 60,000 years ago. They still stand alone, while the oldest dates from elsewhere in Australia are at 46-48 or perhaps 50,000 years ago.

The trouble is that while the Arnhem Land dates look to be very reliable estimates of when the sand grains were buried, they may not be indicative of when the stone tools were actually made. Tools can move about in sand, and just a few that have slipped to a lower level can become associated with older sand. Increasingly my sense is that in fact this has happened in the Arnhem Land sites. If the dates are correct, we must accept that humans resided in Arnhem Land for 10,000 years before they moved south to occupy the rest of Australia. Given the curiosity and mobility of our species, I find this more difficult to accept than an anomaly in the dating.

As is clear from all of this, the dates from all three lines of evidence - extinction, vegetation and fire change, and archaeology - is beginning to point towards 47,000 years ago as the most likely millennium for the peopling of Australia. All dating techniques produce dates with a degree of uncertainly, which means that we must allow a few thousand years either side of this millennium as less likely but plausible times. It is also true that the data is still partial. We do not have, for example, firm dates for megafaunal extinction from northern Australia, or firm dates for the vegetation change outside Queensland. But there is, I feel, now sufficient information to have some confidence that the forty-seventh millennium was a signal one for both Australia and our species.

The scenario for human occupation that I first suggested in The Future Eaters - that human arrival was followed by extinction, then vegetation and fire change, sits comfortably with the new data, but it is now clear that we may never possess the requisite level of accuracy in dating to be able to test this. Other intriguing tests, however, have been identified. It seems that Tasmania was cut off from the mainland 47,000 years ago. Did its megafauna and vegetation change then, or later when people arrived? And what of New Guinea - did its megafauna of seven species of giant wallabies and wombat-like creatures die out when people arrived or at some other time?

Some prehistorians will try to test these ideas by pointing to a date or site that is discordant with the hypothesis. Yet reference to a single site can only be a partial test, for dating is still an imprecise science and it is only where we have several lines of evidence reinforcing a scenario that we can have confidence in the findings.

Other prehistorians are certain to criticise me on the grounds that I have made the call too early - before the wrinkles have been ironed out. Perhaps even the great Rhys Jones himself would have objected here. Whatever the case I’m absolutely certain that he would have defended his Arnhem Land dates with further research. And together we would have shared the delight if he had proved me wrong because, as we know, that’s how science progresses. Equally certain is that Rhys Jones would have been elated - even astonished - at what his young successors have achieved, over such a brief time, in elucidating this continent’s history. Yet for all their achievements, we are still clearly only at the beginning.

Tim Flannery is an acclaimed scientist, explorer and writer. In January 2002 he gave the annual Australia Day address to the nation. This article was first published in the Melbourne Age (5/10/2002). It is re-published here with the author’s permission.

Discussion questions

1. What is the major thrust of Flannery’s article? (What is the connection between advances in dating techniques, the arrival of the first humans on Australia’s shores and the disappearance of the megafauna?)
2. Based on Flannery’s article, what can we glean about the megafauna? What else do we know (from other sources) about the megafauna?
3. How many different types of scientific dating does Flannery mention, and what are their limitations?
4. What is an "ecological footprint"? Give some examples of your own personal ecological footprint.
5. According to Flannery, the evidence strongly suggests that megafaunal extinction "was simultaneous across the continent." What might be some of the possible causes of such an event?
6. How does this discussion about possible causes of megafaunal extinction inform our understanding about "human nature" and human relationships with nature?

Other references

Attenborough, D. (narrator) State of the planet, ABC Video (2000)

Flannery, T. The future eaters, Reed Books (1994)

Flood, J. Archaeology of the Dreamtime, Angus & Robertson (1999)

Back to top