It's been twenty months now since scientists reported discovering fossils on the Indonesian island of Flores belonging to a three-foot-tall hominid with a brain the size of a chimp that lived recently as 12,000 years ago. Homo floresiensis, as this hominid was dubbed, has inspired two clashing interpretations. Its discoverers declared it a separate species descended from another branch of hominids. In others words, the most recent common ancestor we share with Homo floresiensis lived two or even three million years ago. Skeptics argued that the fossils belonged to human pygmies. The one fossil of a Homo floresiensis brain-case belonged to a female with a rare genetic defect. In brief: healthy hominid versus deformed human. Now comes a third theory. In brief: healthy human. Before I get to this third theory, let me explain why others have argued in favor of the other two. (For longer treatments of both theories, see my archive of Hobbit posts.) The discoverers of Homo floresiensis, a k a the Hobbit, argued that it was a separate species on the basis of a number of traits they claimed set it apart from humans. It had no chin, for example, but it had a very wide pelvis, thick limb bones that were also twisted in some cases, and, of course, that tiny brain. A scan of its brain-case also reveals a distinctive organ which the scientists argue does not resemble any well-studied human with the birth defect that creates small brains (known as microcephaly). The record of fossils also points to a long evolution on Flores, rather than the recent arrival of humans. Humans are generally believed to have arrived in East Asia about 50,000 years ago. But the oldest Homo floresiensis fossil, a fragment of an arm bone, is almost 100,000 years old. Stone tools found with the Hobbit fossils also resemble tools found elsewhere on Flores dating back 840,000 years ago. Skeptics (who have now published three papers and have more in the works) have argued that the people of Flores were human pygmies, and that the small brain case belongs to an individual with a condition known as microcephaly, in which a mutation causes the brain to fail to develop completely. They've also cast doubt on the possibility that a separate small-brained hominid species could have evolved on the island. When mammals evolve to smaller sizes, their brains don't get quite so small, relatively speaking. Compared to other mammals, Homo floresiensis is far too tall at three feet for its brain. And finally, they've argued that a small-brained hominid couldn't have made the stone tools found with the fossils. Now comes a different take, in a dense review published online today in the Journal of Evolutionary Biology. Gary Richards of the University of California at Berkeley argues that scientists who have tried to make sense of Homo floresiensis have not grappled yet with the full potential of the human genome to produce different sorts of bodies. They've been focused too much on this or that individual, when they should be considering the range of variations. Richards takes his own best shot, and comes away with the conclusion that the Hobbit is most likely a population of small-brained pygmy humans. The evidence for the Hobbit being a separate species, Richards argues, loses its punch if you dig deep into the scientific literature on human variation. It was certainly short, but not more than ten or twenty centimeters shorter than the smallest human pygmies known today. Richards considers it eminently reasonable that the same genetic changes that have produced populations of human pygmies--which interfere with the action of growth hormones--could have had a somewhat stronger effect on the Hobbit. The discoverers of Homo floresiensis have put great weight on the twist in the humerus (the upper arm bone), but Richards points out that while it may not be universal in Homo sapiens, it does turn up in some populations of...yes, pygmies. Becoming a pygmy involves more than just becoming short, Richards argues. A person's development is influenced in many ways, perhaps partly because height-related genes also affect other aspects of the body, and perhaps because the human body responds to a short stature by developing differently. The wide pelvis and missing chin are also not that unusual compared to the pelvis and chin of living human pygmies. In fact, pygmies were for a long time thought to retain many "primitive" traits in their anatomy, when in fact those traits had evolved very quickly as they had evolved from taller ancestors. Richards argues that much of the alleged primitiveness in Homo floresiensis could have come about in much the same way. When Homo floresiensis was originally described, some scientists argued that it couldn't have been a human pygmy because human pygmies only stop growing at puberty. Homo floresiensis would have had a very different skeleton if it had grown normally for all that time. But Richards argues that this is something of an urban legend. Early research did suggest that pygmies stopped growing at puberty, but it's now clear that they are significantly shorter from birth. Richards concludes that changes to human growth patterns similar to those in pygmies could have produced Homo floresiensis. Then there is the matter of the brain. More than anything, Richards is critical of the way scientists have looked for one microcephalic brain to compare to Homo floresiensis. He points out that there are many genes that influence the size and structure of the brain, and there are thus many kinds of birth defects. It might seem as if microcephaly would leave a person essentially brain dead, but that's not the case. In some cases, it causes only mild retardation, leaving people able learn basic skills and with good motor control. Microcephaly is rare in the United States and Europe, with estimates ranging from 1 in 30,000 to 1 in 250,000 birth. But in communities with a lot of inbreeding, the rate can be far higher--45% in one Canadian community. From this review, Richards offers up this scenario: tall, large-brained humans come to Flores. They settle there and become isolated from other people. As on many islands, they acquire mutations that make them very short. These mutations also lead to other changes in their skeletons. Other mutations also crop up in the people of Flores, including at least one that alters their brains. Because the population is small and inbred, these mutations become common. Not all of the Floresians may have been quite so small-brained, but a significant fraction of them were. The effect of these mutations on the mental abilities of Floresians wasn't catastrophic. They were still able to make stone tools and find food and survive. In fact, having a small brain may have actually been an advantage in some ways. Our brains demand 20% of the calories in our food. A smaller brain would demand less food--a potential boon on an island with limited resources. "I argue that these remains do not represent diseased, pathological, or aberrant individuals," Richards concludes. Just a remarkable population of humans. I have heard this argument before. When I interviewed Tim White, also of Berkeley, for a 2005 Discover story, he floated the idea of Hobbits as healthy humans. (I ended up leaving it out.) Richards is the first scientist I know of who has presented this work in a journal. I'm curious to hear what the other scientists in the debate think about this third way. (I've sent out requests for responses and will post them as they come in.) To my mind there are a couple important issues that--while not scenario-killers--must be addressed one way or another. One is the age of Homo floresiensis fossils. The oldest fossils are nearly 100,000 years old. If they were actually Homo sapiens, it would be quite a big deal for them to be on a remote Indonesian island by then. Consider the evidence of the spread of Homo sapiens offered last week in a review published in the Proceedings of the National Academy of Sciences by Paul Mellars of Cambridge. He argues that our species evolved in Africa about 200,000 years ago. They expanded briefly into the Middle East 110,000 years ago but seem to have retreated back to Africa 90,000 years ago. It's not until 50,000 years ago that scientists find strong evidence of modern humans outside of Africa. Instead of claiming that humans got to Asia over 100,000 years ago, Richards questions the oldest Homo floresiensis fossils. They consist of part of a lower arm bone (the radius) and some teeth. Richards claims that these bones are not distinctive enough to be definitely considered Homo floresiensis, leaving a population of humans that lived between 18,000 and 12,000 years ago. He argues that humans could have arrived 50,000 years ago and then evolved into the Hobbit form by the time the fossils were laid down. This might be true, but it leaves open just who left that arm bone and those teeth almost 100,000 years ago? Not to mention those 840,000 year old tools. The other question that comes to my mind is why other humans who have settled on islands haven't gone all hobbity too. The people of the Andaman Islands, for example, have markers in their DNA suggesting they arrived on the islands a long time ago--perhaps during the initial spread of humans out of Africa. Some became pygmies, but their brains did not shrink to a third the size of their ancestors. Still, I am curious to see the reaction this paper gets from other scientists studying Homo floresiensis. If Richards is right, you have to wonder just how malleable our species is. If we can become Hobbits, what else lies in our future? Update:, 6/25/06 10:30 pm: Peter Brown, from the discovery team, disputes Richards thusly--
An interesting paper but not supported by the skeletal data from Flores. There are a number of early Holocene skeletons, with good archaeological contexts, from various caves on the island. None of these have the skeletal and dental features found in H. floresiensis. They have all been described in a thesis out of Leiden University, and a variety of other places. There were also a number of mesolithic human skeletons recovered from Liang Bua cave two decades ago (now in Tekeu Jacob's laboratory and some in Jakarta) which are also just modern human skeletons. They don't support the claims made by Teuku Jacob (which is one reason they have never been presented to the public), or the claims in the latest paper either. In other words the skeletal evidence does not, and never has, supported the claims for microcephaly. There are no syndromes which reproduce the combination of traits in H. floresiensis. If there was skeletal evidence would have been displayed by the sceptics long ago. The author of the latest paper has not seen any of the primary evidence, most importantly the skeletal evidence from Flores which he argues supports his story. The articles he quotes (eg, Jacob) do not provide the evidence either.
Update, 6/22/06 8:20 am: John Hawks offers his take: deformed human