Discovering new chapters in our story

How researchers made the discovery of a lifetime

The story of the Hobbits, a new species of tiny people that would rewrite history books, capture imaginations around the world

It's hard to tell how large Liang Bua cave is in photos, until you spot the people dwarfed in silhouette by the twenty-metre-wide cave mouth. The opening is partially shielded by a stand of coffee trees, hiding it from the nearby road. Inside, hundreds of stalactites form an inverted cathedral above the workers' heads, and the rocky shelves and the mud: a Sagrada Familia in reverse, its spires pointing down into the open pits of an archaeological dig. This is the cave and those were the pits where, six metres down, a worker's trowel would unearth the skull of the most important and controversial palaeoanthropological finds this century: the Hobbit of Flores.

Flores is a 350km-long island in the middle of the Indonesian archipelago, between Java and Timor. Liang Bua is ten kilometres north of the small town of Ruteng, towards the western end of the mountainous island. Flores' rich earth had already yielded many ancient stone tools, but until this century, no fossil evidence of these early toolmakers had surfaced.

In 2003, archaeologist Professor Mike Morwood was jointly leading an expedition on Flores that he hoped would finally yield the evidence they were missing. The dig had its share of problems: one of their attempts at a deep shaft had started to collapse, so they tried another site closer to the cave wall. By August the dig was winding down and most of the team, including Morwood, had already left the site to attend other business. They had found a number of fauna fossils, including elephant-like pygmy Stegodons-but no early humans.

The surprising find

Even as he departed, Morwood was joking drily about making such a discovery, he reveals in his 2007 book A New Human.

"I pointed to a concentration of Stegodon bones and stone artefacts … and jokingly asked [work supervisor] Wahyu, 'When are you going to find us a premodern hominid skull to go with those?'".

Days after that quip, a worker named Benyamin Tarus was digging through a supposedly 'sterile' layer of mud and struck the find of a lifetime. It was a skull, with features that strongly suggested it was exactly what Morwood was referring to: a premodern hominin. The find shocked the team.

The subsequent excavation of Liang Bua would reveal a significant partial skeleton, including articulated limbs and a near-complete foot. The skull had a sloping forehead and other primitive features, and the skeleton suggested an overall height of just over a metre. This initial find was catalogued LB1, for Liang Bua 1. Other remains were found, including another jawbone, but nothing so complete as LB1. Initially the team supposed they had found a child.

Further investigation showed that the skeleton was in fact an adult woman aged around thirty years at the time of her death. This made her tiny, significantly shorter than any modern pygmy population. The fine details made her even stranger. Her skull and mandible had an odd mix of the ancient and the modern: a negative chin that resembled that of Homo ergaster, but a relatively flat face without the pronounced prognathism of many early humans. Her feet were huge and her arms were very long, compared to her relatively short legs.

The skull hid one further secret that none of the team were expecting-a tiny brain, with a volume small enough that it compared to some chimpanzees.

So what, exactly, had they found in the cave?

A new species

The body of evidence suggested most strongly that the skeleton from Liang Bua was, in fact, a new type of human, Homo floresiensis. Its relationship to our own species Homo sapiens, (or 'wise man') was unknown. Early dating efforts led by UOW dating expert Professor Richard 'Bert' Roberts, indicated the skeleton was only 18,000 years old. This meant it would have lived in the same region as our own species for tens of thousands of years (though this age was recently revised to 50,000 years by Roberts, in light of new evidence collected by UOW archaeologist Thomas Sutikna, and published in Nature in April).

The team knew they had to publish in a peer-reviewed journal, and submitted two papers to Nature. The embargo period until publication in October 2004 was a tense wait, wondering how the scientific community would receive the finds, considering the implications of identifying a new species of primitive human. The resulting scientific and media storm triggered a feud between the Australian - Indonesian research team and an unconnected palaeoanthropology lab over custody of the LB1 remains.

"What [proposing a new species] implied was bucking not one, but two, of paleoanthropology's most basic premises," Morwood later wrote. "That the genus Homo originated in Africa, and that an early type of Homo erectus was the first hominid to leave Africa about 1.8 million years ago."

He was referring to what is commonly called the 'Out of Africa' theory of human evolution, in which modern humans evolved in Africa and left in a single wave of migration that subsequently spread people across the world. As they dispersed, they out-competed, displaced or killed the more primitive hominins they encountered. A popular presentation of this idea has been of modern humans beating out the Neanderthals for dominance in Europe-a strict 'replacement' model-though recent evidence tells a more complicated story including interbreeding.

The chief rival for 'Out of Africa' is the 'multiregional' theory, in which modern humans didn't just come from Africa, they evolved in parallel from different regional populations of Homo erectus or another predecessor species.

The Hobbit threw a spanner in the works of both theories. If Homo floresiensis joined the human family, it would seriously disrupt the accepted timeline for human dispersal out of Africa. The primitive features in her skeleton would also deliver a serious blow to the multiregional theory, because they suggested a link to a more ancient species than Homo erectus of Southeast Asia-maybe as far back as African Australopithecus.

The papers ran on the cover of Nature in October 2004 and LB1 was an instant hit with the popular media. It put her in the public eye at the same time as Peter Jackson's Lord of the Rings films were at the height of their popularity, so LB1 quickly got a nickname that would stick in a way that Homo floresiensis never would: The Hobbit of Flores.

The shoe fit, at any rate. LB1 was a tiny person with large feet who lived in a hole in the side of a hill. That Morwood was a confessed fan of JRR Tolkien didn't hurt.

The Hobbit skull on glass. 

Fame and feuds

The find sparked a wave of international curiosity in both the popular and scientific press-and a long period of unusually heated debate and duelling publications. For more than a decade, successive papers presenting more evidence in support of the Hobbit as a valid new species were quickly met with counter-claims. The intensity and acrimony of these arguments may owe a lot to the stakes. If the Hobbit was a new species, the scientific community would have to reach a new consensus on how humans dispersed around the planet. If she was a sick modern human, she was simply a local curiosity.

In a sense, the two camps were arguing the ground rules for a discussion on what it means to be human.

The most common thread of arguments against the Hobbit being a new species was that she was a modern human, Homo sapiens, deformed and stunted by an illness. The specific theories changed over time: her stature was attributed variously to microcephaly, cretinism induced by an iodine deficiency, or Down's syndrome. In amongst these theories of a pathological Hobbit were more extraordinary claims, including one that LB1 showed evidence of modern dental work.

The claims were each refuted with due diligence, but the criticisms seemed endless. In retrospect, the trail over arguments over the Hobbit resembles a game of academic whack-a-mole, and the Liang Bua team was unable to provide the 'silver bullet' needed to silence their critics.

Morwood, who had by this point moved to join the Centre for Archaeological Science at UOW returned to Flores in 2007 leading a team including UOW's Dr Gert van den Bergh, an experienced palaeontologist and sedimentologist, to resume his search at another site, Mata Menge, in the nearby So'a Basin of central Flores.

The Basin had been a site of archaeological interest for decades. It was first excavated in 1963 by Catholic priest and local amateur archaeologist Father Theodor Verhoeven, who found Stegodon fossils together with stone tools and theorised that Homo erectus and Stegodons had both migrated from Java, maybe as early as 700,000 years ago. It seemed a good candidate for older hominin remains, since Liang Bua had only been occupied by hominins from 190,000 years ago.

Morwood and his team found a wealth of stone tools at Mata Menge, dated to around 800,000 years ago, and they started to develop a predictive model of where to find hominin fossils in the Basin-but the bones eluded them.

"A pity," Morwood wrote of the disappointment. "The whole point of our two-month-long excavation at Mata Menge was to get evidence for the type of hominids that initially colonised the island, and were presumably ancestral to Homo floresiensis."

The teeth recovered from Mata Menge in the So'a Basin. Photo: Kinez Riza

Missing pieces

For Morwood, this final evidence would remain elusive; he died in 2013 after a period of illness. Tributes came from all over the world, noting his contributions to the field. In later dedications, there was a distinct sense of unfinished business; that he had a stake in the investigation that wouldn't be satisfied by an unanswered question.

Dr Gert van den Bergh and UOW colleague Dr Adam Brumm (now at Griffith University) continued working at Mata Menge. Dr van den Bergh was no newcomer to the So'a Basin, having done work there over the past 20 years. His knowledge of sedimentology and the data he'd collected from previous expeditions with Morwood had given him a firm idea of where to look-how deep, specifically.

In late 2013, the team started to target a layer 10 metres closer to the surface and it was there, in the final weeks of the 2014 excavation season, that they found their silver bullet. A hominin lower right jaw fragment and six teeth from at least one adult and two children were found in an ancient riverbed.

 The partial jawbone recovered from Mata Menge. Photo: Kinez Riza

The findings, published as two papers in the 9 June 2016 issue of Nature, detail the anatomy of these remains and show that the finds pre-date Homo floresiensis at Liang Bua by more than half a million years.

"Remarkably, these fossils, which include two milk teeth from children, are at least 700,000 years old," Dr van den Bergh said.

"This find has important implications for our understanding of early human dispersal and evolution in the region and quashes once and for all any doubters that believe Homo floresiensis was merely a sick modern human."

Dr Yousuke Kaifu, from Tokyo's National Museum of Nature and Science, compared the fossils with a large dataset of modern and fossil hominins to identify them.

"All the fossils are indisputably hominin and they appear to be remarkably similar to those of Homo floresiensis," he said.

"The morphology of the fossil teeth also suggests that this human lineage represents a dwarfed descendant of early Homo erectus that somehow got marooned on the island of Flores. What is truly unexpected is that the size of the finds indicates that Homo floresiensis had already obtained its small size by at least 700,000 years ago."

A new interpretation of human

These fossils from Mata Menge may not seem as dramatic to the casual viewer as LB1 did. The Mata Menge hominin is a few centimetres of a yellowed jaw bone and some scattered teeth. No evocative skull peering into a camera, no easy mirror for questions about our humanity. However, its impact on Hobbit ancestry will change our understanding of how humans evolved, and the extraordinary diversity of early hominins. This is a change deep inside the story we tell about ourselves.

In a sense, the Hobbit posed a stern challenge to accepted wisdom and invited the international community to ask What if? Mata Menge demands we answer What now?

Fifty-three years after his first excavation, Father Verhoeven had been proven right. Thirteen years after LB1 saw the light of day, Morwood's theory was also proven correct. That neither lived to see their grand visions vindicated seems cruel, though both were men whose passion took them to worlds that only exist in hindsight; they were always too late for the moment itself. Perhaps they would understand.

In Morwood's words: "so many pieces of the jigsaw are missing that paleoanthropology should be in a continual state of flux as new evidence is unearthed."

Van den Bergh concurs.

"I think Mike would have quite enjoyed the fact that the field of palaeoanthropology is poised for another major shakeup."