New Analysis Places 'Hobbit' on Unexpected Limb of the Human Family Tree

A bust in the David H. Koch Hall of Human Origins at the Smithsonian Natural History Museum recreates what a living Homo floresiensis may have looked like. Ryan Somma/Flickr

Once upon a time, there wasn't just one hominin game in town like there is today. Nobody knows how many human species have lived on this planet, but it's generally accepted that we're one of about 20 human species to have called Earth home. When these human species lived and who begat whom, evolutionarily speaking, is constantly being studied and debated, so it's exciting when paleoanthropology nerds get a new piece of meat to chew on.

A new study published in the Journal of Human Evolution finds that Homo floresiensis, a teensy little hominin species first discovered only a little more than a decade ago in a cave on the indonesian island of Flores, probably doesn't fit into the human family tree the way we thought. For the past few years, the popular view of H. floresiensis — or "hobbit" as it's popularly known, because the mature adults were only about the size of modern kindergarteners — were thought to have descended from Homo erectus, another human species found in Asia. One popular idea was that H. erectus found its way to the Indonesian islands, and over time yielded to "island dwarfism," an evolutionary process in which individuals become diminutive (and along with their overall size, their brains also shrink) as a result of isolation, scarce resources and few predators. But a comprehensive study of the bones of H. floresiensis finds that not only is the species probably older than H. erectus, but it inhabits a completely different limb of our evolutionary tree.


"Our research shows that Homo floresiensis was not derived from, or did not descend from, Homo erectus," says lead author Dr. Debbie Argue of the Australian National University School of Archaeology & Anthropology, via email. "H. erectus, although not modern by any means, is more modern than H. floresiensis. H. erectus is most closely related to H. ergaster, known from Africa and is dated to 1.5 - 1.8 million years ago."

What Argue and her co-authors are saying is that, based on their analysis of 133 data points from the skull, jaws, teeth, arms, legs and shoulder bones of H. floresiensis, it seems this human species we assumed was rather young — the bones that were found on Flores suggest hobbits existed in Indonesia between around 60,000 and 100,000 years ago — is most likely a close relative of one of the most ancient human species. That'd be Homo habilis, which lived half a world away in Africa between 1.75 million years ago to 1 million years ago.

Full Width
The cave on the Indonesian island Flores in which Homo floresiensis fossils were discovered in 2003.
Bryn Pinzgauer/Flickr

"We can be 99 percent sure it's not related to Homo erectus and nearly 100 percent chance it isn't a malformed Homo sapiens," said co-author Dr. Mike Lee of Flinders University, who did the statistical modeling for the study, in a press release.

How these ancient hominins would have gotten to Indonesia from Africa is, of course, a mystery. The continents during this period were in the same position as now, but since Indonesia is part of a highly tectonic area, it's possible there was uplift of land that could have formed a land bridge between Flores and the adjacent islands. Certainly, stranger things have happened, but not everybody thinks we can be "99 percent sure" H. floresiensis is not descended from H. erectus.

"This study is purely based on differences in morphological characters between fossil specimens, with each character weighted equally, and with disregard of any functional aspects of every character," says Dr. Gerrit van den Bergh of the University of Wollongong in New South Wales, one of the authors of the 2016 study published in Nature that supports the idea that H. floresiensis descended from H. erectus and was made small by insular dwarfism. "The problem is, we actually don't understand the function of many morphological characters.

According to van den Bergh, the only way we will ever know which hypothesis is correct, is by finding fossils of the founder population of the island of Flores. We know there were humans on Flores by 1 million years ago because they left stone artifacts that have been dated. These toolmakers were small, and stayed small between 700,000 and 50,000 years ago.

"So if we can find hominin fossils from about 1 million years ago on Flores, we may get stronger evidence about the ancestry. But until then the discussion about the ancestry of H. floresiensis will not be easily resolved," says van den Bergh.

Full Width
A reconstructed Homo floresiensis skull.
Stuart Hay, ANU

But, according to Argue, we may find more hobbit fossils on Flores:

"Another exciting thing is that we only know H. floresiensis from one cave on Flores. There are hundreds of caves on Flores," she says, "which means there is a huge potential to do archaeological excavations and search for more remains so that we can find out more about this species, including how long it lived there and when it finally became extinct."