News of the Week

Tools Link Indonesian ‘Hobbits’ to Earlier Homo Ancestor

Science  02 Jun 2006:
Vol. 312, Issue 5778, pp. 1293a
DOI: 10.1126/science.312.5778.1293a

The battle of the hobbits is heating up. Two weeks ago, skeptics argued that fossils found on the island of Flores in Indonesia were simply diseased modern humans ( rather than a dwarf species evolved from an early Homo ancestor, as its discoverers had claimed. Now the discovery team fires back. In this week's issue of Nature, they argue that stone tools associated with Homo floresiensis resemble newly discovered tools from a much more ancient nearby site, suggesting cultural continuity over hundreds of thousands of years.

The tool data “establish an independent source of evidence linking late Pleistocene Homo floresiensis with an early Pleistocene progenitor,” says Russell Ciochon of the University of Iowa in Iowa City. But some caution that the tools are so simple that inferences of cultural continuity may not be warranted, and a few skeptics question the dates.

The ancient tools come from Mata Menge, 50 kilometers from the Liang Bua cave on Flores where H. floresiensis bones and tools were found by an Indonesian-Australian team including Michael Morwood of the University of New England (UNE) in Armidale, Australia. Researchers had previously uncovered stone tools at Mata Menge and dated the artifact-bearing layers to between 800,000 and 880,000 years ago using fission-track dating on volcanic tuffs.

In 2004 and 2005, Fachroel Aziz of the Geological Research and Development Centre in Bandung re-excavated Mata Menge and invited Australian colleagues including Morwood and first author Adam Brumm of Australian National University in Canberra. They found a bonanza of artifacts: 507 small, well-shaped pieces made from volcanic cobbles, with a few chert pieces.

The team then compared the Mata Menge tools to the much younger artifacts from the Liang Bua cave, dated from 95,000 to 12,000 years ago—and found a match in both the types of artifacts and the methods used to create them. At both sites, hominids produced elongated flakes by rotating cores and striking downward; they also created “perforators,” pointed tools with retouched edges. “All of the techniques at Mata Menge are also at Liang Bua,” says co-author Mark Moore of UNE. “These are quite common approaches to reducing stone.”

They are also simple approaches. That's in contrast to the team's original publication, which described a few Liang Bua tools as much more sophisticated. That led some researchers to claim that the tools must have been made by modern humans, not a hominid with a brain the size of a grapefruit. But Moore now says that although some elongated flakes resemble “blades” used by modern humans, that may simply be coincidence. Richard Potts of the Smithsonian Institution in Washington, D.C., agrees: “Yes, [the Liang Bua hominids] are making what people have called ‘blades,’ but that doesn't imply that you have to have a certain number of neurons,” he says. Morwood is more emphatic: “Some of our critics have claimed that these Liang Bua artifacts are so sophisticated that they must have been made by modern humans. The [new] evidence shows that the basis of that argument is just plain wrong.”

Toolmaking tradition?

Tools from an ancient site on Flores (top row), including a “perforator” (left column), resemble those found near hobbit bones (bottom).


Morwood adds that the team now considers the hobbits' most likely ancestor to be a small early Homo species, smaller than the classic H. erectus found in nearby Java but perhaps similar to fossils found in Africa and Dmanisi, Georgia.

However, Kathy Schick and Nicholas Toth, knapping experts at Indiana University, Bloomington, caution that the technology is so simple that different kinds of hominids might converge upon it. And James Phillips of the University of Illinois, Chicago, a co-author of the critique published in Science, thinks that the tools may be out of sequence.

Morwood points out that many hominid species were first greeted with skepticism. The type specimen of H. erectus—uncovered in 1891 on Java—was described at the time as a “microcephalic idiot, of an unusually elongated type,” in a review in Nature.

Navigate This Article