The Hobbit: Flores, Indonesia
In the last decade the most exciting new addition to the story of human evolution has been the discovery of the remains of diminutive people who lived on the Indonesian island of Flores. It was an event that rocked the world of palaeoanthropology, as well as making news headlines. It was presented as the find of the century, but what really caught public attention was that here was evidence of a different species of human – living at the same time as modern humans. The Hobbits had been on their Indonesian island until as recently as 12,000 years ago. Although we are familiar with this idea in Europe, that there were still Neanderthals knocking around when modern humans arrived on the scene, it’s something that still comes as a bit of a surprise, and even sends a shiver down the spine. We are very used to the idea that we are the only human species on the planet today (although, genetically, there is actually a good argument for chimpanzees and gorillas being included in Homo as well). Some even feel that we are so unlike other animals as to be a special creation. When we start finding other species that challenge our uniqueness, it unsettles that illusion. At the time of the discovery, Chris Stringer said, ‘It’s remarkable, astonishing, sensational, even … It challenges the whole idea of what it is that makes us human.’ The idea that there could have been (and some think still are) other people, not quite human, sharing the planet with us, is somehow spooky. Even spookier are the myths from Flores, of Ebu Gogo: small creatures that inhabited caves, and provoked suspicion and fear among Floresian villagers.
In 1995, Mike Morwood and Doug Hobbs were on the Kimberley coast in north-west Australia, excavating an eighteenth-century site where Indonesian fishermen had boiled up sea cucumbers, ready to sell to the Chinese as a delicacy. But contact between Asia and Australia stretched back much further, way beyond historical records. The first Australians would have arrived from Indonesia. So, the archaeologists started to plot an excavation in Indonesia – to search for the early modern human colonisers of the region and the ancestors of the first Australians.1
The Indonesian island of Flores seemed like a good place to start: ancient stone tools had already been found there, and Mike Morwood could team up with Indonesian palaeontologists and archaeologists who were working there. And there already seemed to be something a bit intriguing, something that ‘didn’t quite fit’ about Flores. Throughout the Pleistocene, Flores was always an island, separated by deepwater sea channels from it nearest neighbours, Bali, Lombok and Sumbawa. And it’s generally thought that modern humans are the only hominins who managed to make sea crossings. But the tools that had previously been discovered on Flores seemed too ancient to have been made by modern humans. Then there was some doubt as to whether they were really stone tools at all.
Digging in 1997, the international team found definite stone tools, embedded in volcanic tuff of the Soa Basin, and got secure dates on them. The tools dated to between 800,000 and 900,000 years ago. This finding was important enough to get into Nature. Morwood suggested that these tools must have been made by Homo erectus, as that was the only hominin known to be around in South-East Asia at that time. But that meant Homo erectus had made a sea crossing. It was a controversial claim.
In the following years work continued in the Soa Basin, but the team also branched out to investigate a couple of cave sites where previous excavations had turned up some, much more recent, evidence of modern humans, including burials and stone artefacts from the last 10,000 years. In April 2001, excavations started at Liang Bua, ‘the cool cave’. A team of local Manggarai people had been taken on as excavators, digging with trowels and bamboo stakes, then swapping them for sledge hammers and chisels to get through layers of hard flowstone. The excavation went deeper than previous digs had done; the sides of the trench were carefully shored up as they went down. Morwood was not satisfied with stopping at layers that looked ‘sterile’, or untouched by any signs of human activity: he wanted to get down to bedrock. And he was rewarded. In deep layers, they found thousands of stone tools, animal bones and teeth. Bert Roberts dated the remains and found them to be between 74,000 and 12,000 years old. A strange, small and rather curved hominin radius was the only apparently human bone found in that first digging season.
In 2003 the digging team turned up what they thought was the skeleton of a Homo erectus child: it was a very exciting find. The skull was thick, with a sloping forehead, which fitted with Homo erectus, and it was very small. The skeleton hadn’t been deliberately buried; the body had somehow ended up in a shallow pool in the cave and become quickly covered over, so that the bones were preserved. The bones were not fossilised and were very mushy; the team used a mixture of UHU glue and acetone nail polish remover to consolidate the fragile bones so that they could be lifted. But when the bones were properly cleaned up it became evident that Liang Bua skeleton number 1 (LB1) wasn’t a child at all, but a tiny adult.
Peter Brown, Professor of Palaeoanthropology at the University of New England, flew out with Mike Morwood to Jakarta to examine the tiny skeleton. Using mustard seeds, which he poured into the skull, Peter Brown found the braincase was astonishingly small: just 380ml. Anything in the genus Homo is expected to have an adult brain size of at least 600ml, based on previous fossils, and modern humans have brains anywhere between 1000 and 2000ml. Big brains are a fundamental characteristic of humans. Small brains can be caused by pathological conditions, such as microcephaly, but even then it’s unusual to end up with a brain volume of less than 600ml. The oldest known hominin fossils outside Africa, from Dmanisi in Georgia, dating to 1.8 million years ago, are small-bodied and small-brained (with a stature of 1.4m and brain size of 600ml), but are still nowhere near as small as LB1.
Peter Brown did not think the skeleton was pathological, neither did he think it was Homo erectus. In fact, he originally wanted to give it a brand new genus and species name: Sundanthropus tegakensis. LB1 was something very strange indeed: a tiny hominin that Peter Brown thought looked even closer to the ancient African australopithecines than to any member of the Homo genus. Morwood, though, in spite of the tiny brain size of LB1, thought there were enough traits in the skeleton to label her Homo.1 And then there was the behaviour: tool-making is not meant to pre-date Homo. In 2004, the find was published in Nature: ‘a new small-bodied hominin’, named Homo floresiensis.2 Morwood and his colleagues suggested that this species had derived from an ancestral population of Homo erectus, which had become isolated on Flores, with the whole population undergoing ‘endemic dwarfing’.
The publication produced a palaeoanthropological storm and made newspaper headlines all over the world. The main debate centred on whether this really was a new species, or whether it was a pathological, modern human. Indonesian palaeoanthropologist Teuku Jakob argued that LB1 was just a microcephalic, pygmy modern human,3 but further excavations at Liang Bua produced partial skeletons of another twelve individuals, so any idea that LB1 was a pathological one-off could be discounted. Even with a reassessment of LB1’s cranial capacity as slightly larger, around 417ml, the shape of the braincase is quite unlike that of a microcephalic.4 The debate has rumbled on, though, with a recent paper in the Proceedings of the Royal Society claiming that the diminutive nature of the Flores specimens could have been caused by a congenital disease, with a defective thyroid gland producing dwarfism and small brain size.5 But the researchers making this latest claim didn’t even look at the real skeletal remains; they made their diagnosis from photographs, which seems decidedly dodgy. They claimed that the pituitary fossa (the cradle of bone inside the skull in which the pituitary gland sits) looked larger than normal in the photographs: a sign of congenital hypothyroidism. But other researchers who have looked at CT scans of the skull have refuted this claim.
The pathological explanations for Flores tend to assume that the small skull is the fundamental characteristic that needs explaining. What they don’t tackle is why the whole skull looks so remarkably different from modern humans, or why the limb proportions
of the skeleton might be so different. An Australian team of researchers, led by Debbie Argue, compared LB1 with pygmy modern humans and other hominin species, and agreed with Morwood that LB1 was not microcephalic, and that it was different enough from both modern humans and Homo erectus to warrant its own, new species name. However, they also argued against Homo floresiensis having developed from Homo erectus, as it possessed features which suggested it was evolutionarily somewhere between the australopithecines and early Homo.6 Careful analysis of the wrist bones and shoulder of LB1 have also shown them to be primitive and unlike those of modern humans.7,8
So I was very excited indeed by the prospect of looking at the original bones for myself. On a fairly grey March day in Jakarta, I met Tony Djubiantono, Director of the National Research Centre of Archaeology. He led me down a corridor in the Centre and into a room where there was desk, a sofa and a safe. He took Tupperware boxes, containing the Flores bones, out of the safe and carried them into a larger store room where there was a table ready for me to lay out the skeleton of LB1.
I unpacked the bones in silence, watched by Tony and by a television camera. I was quite taken aback. The bones were absolutely tiny.
I systematically laid out the skeleton: skull first, at one end of the table, followed by fragments of vertebrae, and then arm bones, hands, pelvis, leg bones and feet, just as I would do with any archaeological skeleton in the bone lab in Bristol.
It was a very, very strange little skeleton. There was no question that LB1 was an adult: all the epiphyses (the knobbly ends of the long bones) had fused to their shafts, and there was a full set of adult teeth. But this was an incredibly small adult. It didn’t look in any way pathological, and, anyway, it would have to be some weird science fiction disease to produce the mixture of traits in LB1: a dreadful affliction that caused some kind of time warp and pushed bits of the human body backwards through millions of years of evolution.
Some parts of LB1 looked quite modern human: the teeth in particular were set in a parabolic curve in the jaws, and looked quite like modern human teeth in shape and size. But there were things about it that were distinctly unhuman: the incredibly tiny skull, which had its widest point low down around the ears; the way the middle of the mandible was rounded off, chinless, and thickened on the inside; and the general thickness of the mandible and the wide ramus – the part of the jaw that swoops up to the jaw joint. The limb bones were chunky for their size, and the pelvic bones were a weird shape that reminded me of the flared pelves of australopithecines. It also had longish arms and shortish legs: these were not modern human proportions, but more like early Homo, or even australopithecines – fossils of which have only ever been found in Africa.
All my experience of early hominin skeletons up to that point was from photographs and casts. But, to me, this little skeleton looked a lot like the gracile australopithecines (Australopithecus afarensis and africanus) – who lived in Africa between two and a half and four million years ago. So what on earth was a hominin like this doing in Indonesia?
It would certainly be ‘easier’, and fit better with current theories, if the Hobbits were a pathological modern human population: an unfortunate group of people who were all congenital microcephalic dwarfs. But it seems unlikely, from both the shape of the skeletons and the date of the remains. While the LB1 skeleton is dated at around 18,000 years old, dates for other skeletal remains of Homo floresiensis and stone tools go back as far as about 95,000 years ago.9 This creates problems for the scientists arguing that the Hobbits are pathological modern humans, as this is twice as old as any evidence of Homo sapiens in South-East Asia. So the Hobbit presents some significant challenges to the accepted thinking about patterns of dispersal and migration, and about the range of variation in the anatomy of hominins. Debbie Argue contends that the species must have been intermediate between australopithecines and Homo at the point it left Africa, and that it must therefore have emerged out of Africa prior to two million years ago: another – even earlier – Out of Africa. You can see why the palaeoanthropological world is reeling from the blow dealt by this Indonesian cave man. While Mike Morwood seems convinced that it is Homo, he also knows that the implications for hominin dispersals are profound. In his book about the discovery of the Hobbit, he even raises a question about whether Homo could possibly have arisen in Asia. Shocking stuff indeed.
As well as the skeletons from Flores, the stone tools throw a bit of a spanner in the works – this time for neat theories about ‘who made what’ in South-East Asia. Previous explanations have attributed large ‘core tools’ to Homo erectus and small-sized ‘flake tool’ assemblages to Homo sapiens. But it seems that Flores Man was doing something that brought those two ‘types’ of tool together as stages in a production line. Large cobbles suitable for making flake tools are cumbersome to carry around, so the hominins struck off big flakes from these cores and left them in the landscape. Then they took the flakes off with them and made smaller flakes to use as tools. This explains why the two different ‘assemblages’ occur in different places. But they actually appear to be part of a single process, and not two culturally different ways of making tools, used by different hominins. If that was happening on Flores, it begs questions about attributing particular tools to particular species anywhere in Asia.9 Again, it makes me glad that I spend my time in the lab looking at bones, not stones!
So, what about the question that everyone wants the answer to: did humans and Hobbits ever meet? I interviewed two Floresian men, Gregorius Buü Wea and Anselmus La Li Wea, who told me their folk stories of small people who lived in caves in the hills: Ebu Gogo. Gregorius and Anselmus said that these human-like creatures used to be attracted to feasts, but would stay very much at the edges, not engaging with humans. But they did seem to aggravate the humans as well.
‘They stole the crops that belonged to the peasants. Crops like cassava and fruit were always stolen by Ebu Gogo.’
‘Were they at all threatening?’ I asked.
‘Not really,’ came the reply. ‘But they did sometimes steal children as well.’
That sounded fairly threatening to me. ‘What did they want with the children?’ I asked.
‘They just liked children,’ said the Floresians.
I asked them what Ebu Gogo looked like.
‘They were hairy all over. Their faces were like those of monkeys. And they were short – less than one and a half metres tall.’ This all seemed very credible. But there were some more bizarre characteristics as well.
‘They had broad chests, with kangaroo-style pockets which they used to keep their stolen goods in. And the female ones had big, long breasts, right down to their knees.’
Very interesting. Well, apparently the people of Flores eventually grew tired of putting up with these strange little people who threatened their food supply and their children, and they plotted their downfall. The story went that a group of people went up to the caves where Ebu Gogo lived, and offered them rattan mats. The small creatures accepted one mat after another, and the humans set fire to the last mat before handing it up to the cave. All the mats inside the cave caught fire and the Ebu Gogo were destroyed. It was a nasty little story, but reminiscent of what happens everywhere around the world where different groups of humans take against each other.
Of course the folk stories don’t prove anything. So what about the archaeological evidence? There were certainly modern humans around in the vicinity at that time. Evidence of modern humans has been found in nearby Timor, going back to more than 42,000 years ago, from excavations at the Jerimalai shelter.10 But modern humans did not seem to reach Flores until much later, so there is no archaeological or genuine historical evidence to suggest that humans and Hobbits ever met. Although Hobbits were around on Flores until at least 12,000 years ago,11 there is no overlap of the archaeological evidence of Homo floresiensis and later evidence of Homo sapiens on the island of Flores itself.
I asked the men from Flores if they thought any Ebu Go
go could have escaped and still be alive today.
‘Maybe,’ they replied.
A Stone Age Voyage: Lombok to Sumbawa, Indonesia
Even before the Out of Africa theory became widely accepted, archaeologists had suggested that early moderns would have spread across the globe most easily by following the coast. It made sense from an ecological point of view: there are rich pickings for hunter-gatherers along the seashore, and genetic evidence has provided strong support for this theory. But there was a point at which the coast ran out. Australia was always separate from South-East Asia, and separate by a long way.
Analysis of mitochondrial DNA suggests that modern humans emerged from Africa some time after 85,000 years ago, then spread ‘rapidly’ (in prehistoric terms) eastwards.1,2,3 Studies of mtDNA in Aboriginal Australian populations suggest that the founder population arrived some time between 40,000 and 70,000 years ago.3,4,5 Some archaeologists still seem to prefer a conservative date for colonisation of between 40,000 and 50,000 years ago,6 but the consensus appears to be – based on the genetics but also on the oldest dated sites in Australia – that the colonisation of Australia happened some time between 50,000 and 60,000 years ago.7
At this time the environment in South-East Asia – rich with reefs, mangroves and estuaries – would have been ripe for exploitation by beachcombers with simple watercraft.8 The sea level was about 40m lower than today: Borneo, Sumatra, Java and Bali would have been part of the mainland, forming the ‘Sunda Shelf’. Early colonisers may have been able to move down through the middle of this land bridge, rather than sticking to its edges. Because running right through the centre of Sundaland would have been a long, grassy plain, a ‘savannah corridor’.9 To the south-east, New Guinea, Australia and Tasmania formed the great landmass of Sahul. In between lay the islands of the Wallacean archipelago, corresponding to the modern-day Indonesian Nusa Tenggara islands, principally, Lombok, Sumbawa, Flores, Sumba and Timor. To South-East Asian animals these islands represented a major barrier to diffusion, and modern humans are the only large mammals who made it across.6 So, although there was much more above water than there is today, colonisation of the islands of Wallacea and Sahul itself would have required sea crossings.
The Incredible Human Journey Page 14