News FocusPaleoanthropology

The Fellowship of the Hobbit

See allHide authors and affiliations

Science  10 Aug 2007:
Vol. 317, Issue 5839, pp. 740-742
DOI: 10.1126/science.317.5839.740

Adversaries in a debate with momentous implications—whether a tiny human fossil found in Indonesia is a new species—mulled fresh data on hallowed ground. But the hobbit remains an enigma to many

Cave exploration.

Liang Bua, home of the “hobbit” (inset), swarms with researchers.


FLORES, INDONESIA—Liang Bua cave, as big as a concert hall, with a domed ceiling and a giant stalactite for a chandelier, thrums with the din of a portable generator and excited conversation in many languages. Fifty anthropologists plus a retinue of villagers, police, and curious children mill about in cool air that smells faintly of damp rock and cigarettes. The researchers are on a pilgrimage to ground zero of one of the most contentious debates in human evolution. Here, in 2003, the skull and skeleton of a meter-tall adult woman were unearthed. Ever since, experts have sparred over the “hobbit”: Is it an astonishingly primitive species with a tiny head, dubbed Homo floresiensis, or a diseased member of our species, H. sapiens?

The stakes are high. A new species shakes to the core ideas about the defining role of big brains in our genus and about relations among hominids. The hobbit bones are dated to as recently as 12,000 years ago, so the diminutive hominid must have lingered on Flores for thousands of years while modern humans colonized nearby islands. The tiny human suggests that big brains aren't required for making tools—and, according to a theory proposed by the hobbit's discoverers, may imply that the first hominid migrated out of Africa far earlier than anyone had thought. “Flores is the thorn in the flesh. [It implies] that we have to rethink everything,” says anthropologist Marcia Ponce de León of the University of Zurich in Switzerland.

The visit to Liang Bua cave was the culmination of an unprecedented gathering* of the hobbit's discoverers and their critics, several of whom had battled fiercely in print and on film but had never met in person. For the most part, adversaries were on their best behavior, like feuding relatives gathered for a reunion. Conference organizer Teuku Jacob of Gadjah Mada University in Yogyakarta has argued that the hobbit is a diseased H. sapiens. Two years ago, he borrowed the bones for study over the objections of co-discoverer Mike Morwood of the University of Wollongong, Australia; some of the bones were broken upon return (Science, 25 March 2005, p. 1848). But Jacob was unfailingly smiling and polite to all, and Morwood made a point of speaking to each critic. Still, divisions run deep, among both Indonesians and foreigners.

Field captain.

Co-discoverer Mike Morwood defended his team's view of Homo floresiensis.


At the meeting, funded generously by businessman and philanthropist Hashim Djojohadikusumo, who paid for the travel and five-star accommodations for all presenters, researchers heard much of the latest thinking on crucial aspects of hobbit science and lore, from cave geology to an (unsuccessful) hunt for living “orang pendek,” mythical humanlike creatures that legends say once roamed Flores. The critics, most of whom were in attendance, were unshaken in their belief that the hobbit is a pathological modern human, perhaps one who suffered from microcephaly, a disorder that results in a tiny head, or from a growth hormone insensitivity called Laron syndrome.

Meanwhile, a growing number of those working on the bones, several of whom were not invited or chose not to attend, are convinced that they are dealing with a new phenomenon in human evolution. “This is a different species for sure,” says William Jungers of Stony Brook University in New York, who is working on the skeletal bones but was not invited. “There is no pathology that recapitulates early hominid morphologies and proportions.” Others haven't made up their minds, perhaps in part because the bones themselves were deemed by the Archaeological Institute in Jakarta to be too fragile to send to the conference for viewing. “It was a pity we didn't see the fossils,” said Zhao Lingxia of the Institute of Vertebrate Paleontology and Paleoanthropology in Beijing, as she picked up her baggage after the last trip. “I am not sure what [the hobbit] is.”

Reading the bones

Even for scientists steeped in the mysteries of the hobbit, the seminar offered surprises. One was from archaeologist Carol Lentfer of the University of Queensland in Brisbane, Australia, who analyzed residues and polish on the edges of stone tools found in Liang Bua. Under the microscope, substances such as animal hide, wood, and plant materials leave telltale traces on stone. Because the tools were found near animal bones, especially baby pygmy elephants called Stegodon, researchers had inferred that the little people used the tools to process meat. But to Lentfer's surprise, most of the tools she examined were used for working with woody and fibrous plants, perhaps to craft spear shafts of wood or bamboo or items like traps. “It looks like a tool kit for making other tools,” she said in her talk.

Whatever the tools were for, some experts aren't sure that hobbits made them. James Phillips of the Field Museum in Chicago, Illinois, has argued that long, narrow blades removed sequentially from blade cores are too sophisticated to have been made by anyone other than a member of our species. That means that the hobbit must have been part of a H. sapiens population, he said. Harry Widianto of the Centre for Archaeology in Jakarta also believes that the tool kit belonged to our species. The tools were made with “very high skill,” he said, and resemble artifacts found in other Indonesian caves, including those occupied more recently.

But every shred of hobbit evidence has conflicting interpretations. Morwood and two colleagues—hobbit excavator Thomas Sutikna of the Centre for Archaeology and archaeologist Mark Moore of the University of New England in Armidale, Australia, who did his Ph.D. on the artifacts—don't agree that they are sophisticated. Moore, who was not invited to the meeting, told Science that although “blades”—flakes twice as long as they are wide—are found at Liang Bua, there is no evidence that they were flaked deliberately, as seen in classic H. sapiens tools. “These are simple stone artifacts,” agrees Sutikna.

When it was Morwood's turn at the podium, he emphasized the evidence against the theory that the hobbit was a modern human. His voice crackling with intensity, Morwood reminded colleagues that the famous skull and skeleton known as LB1 is not alone. To date, the team has found an additional jawbone plus various bones of the leg, arm, and shoulder, all petite, from different layers. “We have a minimum of 12 individuals … going back to 95,000 years ago,” he said. “That's twice the accepted date for H. sapiens in Southeast Asia and Australia. Twice the accepted date, okay?” He adds that the hobbit layers show no traces of pigments, ornaments, or formal burial—all signs of H. sapiens that are found in the cave's upper levels.

Morwood also underscored the similarities among the hobbit bones. “The radius and leg bones in the deeper deposits have the same unique characters as seen in the higher levels,” he said. Because it's highly unlikely that only diseased individuals died in the cave over thousands of years, the additional specimens rule out pathology, he said. But critics have argued that, except for the tiny brain, some hobbit traits are also seen in living Southeast Asians and so aren't signs of either pathology or a new species (Science, 25 August 2006, p. 1028).

Not the brainiest

For others, no trove of skeletal bones can compensate for the puzzle of that puny brain. Short stature alone does not mean that it is a distinct species, as small stature is known from living pygmies, including those in the village of Rampasasa near Liang Bua, say Jacob and others. But pygmies have brains nearly the same size as those of other modern humans. The minute brain that would have fit inside LB1's skull was only 400 cubic centimeters, compared to the roughly 1350 cubic centimeters seen on average in living humans (see graph). “Flores falls outside the range of anything I have seen before,” said Robert D. Martin, a paleoanthropologist at the Field Museum.


The hobbit's tiny brain (in red, top) stands out when plotted with those of other hominids; discoverers say it evolved on Flores thanks to currents that isolated the island (bottom).


Martin and other presenters suggested that LB1 was diseased, perhaps suffering from microcephaly, Laron syndrome, or both. The Laron's hypothesis debuted in the hobbit debate on 27 June, in a paper published online in the American Journal of Physical Anthropology by a trio from Tel-Aviv University who were not at the meeting: Israel Hershkovitz, Liora Kornreich, and Zvi Laron, discoverer of the disease. Their paper lists Laron's symptoms—although with few measurements—that are also seen in hobbits, including pillars of bone near the nose, a short clavicle, a curved tibia, and an upper arm bone whose top end is not twisted. Some of these traits have been considered signs of a primitive ancestry for H. floresiensis. “All [Laron's] patients share a battery of traits, which they also share with Homo floresiensis,” Hershkovitz and Laron told Science.

At the meeting, Dean Falk of Florida State University in Tallahassee, who has concluded from computed tomography (CT) scans of the skulls of LB1 and microcephalics that the hobbit is a new species, tackled the Laron's hypothesis head-on. Hershkovitz and colleagues note that many Laron's patients also lack the sinuses of a normal human head. And although in most people the texture of the mastoid process—the bony bump behind the ear—is spongy and air-filled, in Laron's patients this bone is dense. CT scans of LB1's skull show that it has normal sinuses and a porous mastoid process, Falk said. “We don't think LB1 comes close to looking like their description of Laron's,” she said firmly. Hershkovitz responds that some Laron's patients do have normal sinuses, and so their presence does not disprove the hypothesis.

Island living

If the hobbit is a new species, who were its ancestors? Morwood's team once postulated that H. floresiensis evolved from H. erectus, the first human ancestor known to have left Africa. Dozens of H. erectus specimens have been unearthed on the nearby island of Java. But the team now argues for a more radical idea of the hobbit's origins: a “pre-erectus ancestor”—a small-bodied, small-brained, primitive hominid, which shrank further once on Flores. In her talk analyzing the skull and jawbones, Debbie Argue of Australian National University in Canberra proposed that the hobbit shared an ancestor with 2-million-year-old H. habilis (see p. 733).

When asked by colleagues, Morwood referred to unpublished work presented at recent meetings that unites H. floresiensis with early Homo or with the even more ancient australopithecines. He ticked off a few key features: an odd shoulder joint (Science, 19 May 2006, p. 983), a wrist like that of an ape (Science, 6 April, p. 34), and primitive feet. Jungers agrees that the hobbit “has australopithecine limb proportions and australopithecine/ape wrists and tarsals.” Given these primitive traits, Morwood argues for a very ancient ancestor. “I believe now that they split from us 2 to 3 million years ago,” he said. That would imply that an australopithecine or very early Homo, rather than H. erectus, was the first hominid to leave Africa.

Skeptics are unimpressed. For starters, they point out that there is no evidence of the hobbit lineage for those millions of years. “If you go back that far, where are all the intermediates?” asks Martin.

Other scientists, including those convinced that the hobbit is a new species, think it may be premature to eliminate H. erectus as a possible ancestor. To date, the most complete H. erectus skeleton published is much larger than H. floresiensis, but there are other, smaller skulls of the species, particularly from the 1.7-million-year-old site of Dmanisi, Georgia.

And evolving into a smaller form on an island is a common phenomenon in other mammals. In his talk, paleontologist John de Vos of the National Museum of Natural History in Leiden, the Netherlands, gave a whirlwind tour of such island dwarfing. On islands from the Mediterranean to Southeast Asia, some large species such as elephants and hippos shrank while others such as rats and hedgehogs evolved into giant forms, depending on the available ecological niches. “On islands, we get relict lineages,” de Vos added—lineages that hark back to primitive ancestors. In his view, animal and hominid bones from Flores fit this pattern.

Dwarf island forms are also often pedomorphic, in that they retain childlike traits into adulthood, de Vos said. Thus on islands, adult elephants, hippos, and deer retain short snouts and short legs. This could be what happened to a H. erectus-like ancestor on Flores, he speculated, for many of H. floresiensis's peculiarities appear to be pedomorphic: the lack of twisting at the top of the arm and leg bone, a flat face, and short legs. Christoph Zollikofer of the University of Zurich, who works with Ponce de León on the H. erectus fossils from Dmanisi, compared them to the Flores bones and independently came to the same idea: “Homo floresiensis can be understood as a pedomorphic, dwarf erectus,” Zollikofer said. Some degree of island dwarfing makes sense, even if starting from a smaller ancestor, said Morwood.

Yet skeptics aren't swayed by the dwarfing evidence. Martin argues that mammalian brains are unlikely to shrink to the same degree as their bodies during dwarfing. “Weird things happen on islands, but not that weird,” he said. He and de Vos are considering a joint project to probe dwarfing and brain size.

The controversy may rage until more fossils emerge—and new specimens may not all come from Liang Bua. Given ocean current patterns, Morwood thinks that H. floresiensis might have originally been swept to Flores from Sulawesi, and he's planning to dig there to find out. On many Southeast Asian islands, says de Vos, excavations stopped at layers dated to the beginning of our current Holocene Epoch, about 11,000 years ago. Morwood found the hobbit “because he dug deep,” says de Vos. “The message for the future is, dig a very deep hole.” Eusebio Dizon of the National Museum of the Philippines in Manila, for one, is eager to excavate. “In the Philippines, we have the right fauna: Stegodon, giant rats, and turtles,” he said. “It is perhaps just a matter of time until such species [as the hobbit] appear.”

Researchers on both sides are also pinning their hopes on DNA. Genetic data could establish or rebut the existence of a new species, test the Laron's hypothesis, and perhaps even identify a mechanism of dwarfing. Previous attempts at retrieving DNA from hobbit bones failed, but clean sampling will be done immediately at the site if more bones are found, Morwood says. “The issues won't be resolved here at the meeting,” he adds.

Nor were they resolved on the brief visit to Liang Bua cave, where Morwood and Sutikna recounted this year's excavations to a handful of colleagues, pointing to two pits the size of telephone booths, shored up with wood and with stratigraphic layers neatly revealed on the sides. In one excavation, the team had just struck a layer of a whitish volcanic tuff dated to 12,000 years ago that they say runs through the cave. Yet skeptics such as the Field Museum's Phillips say cave stratigraphy is notoriously hard to decipher, and they aren't ready to accept the dates.

Meeting of the minds.

Excavator Thomas Sutikna (bottom, left) showed off artifacts in Liang Bua; conference organizer Teuku Jacob (top) did not visit the cave.


Later, after the visiting scientists troop out of the cave, Sutikna and other excavators welcome one more group: a priest and a few dozen of his flock who have come to sing and pray for the excavation's success. After the last hymn, the worshippers file out, and Sutikna and the crew clean up. The next day they will dig deeper, continuing their quest for more tiny bones with big implications.

  • *International Seminar on Southeast Asian Paleoanthropology, 22-29 July, Yogyakarta, Indonesia.

View Abstract

Navigate This Article