The “hobbit” had neighbors. Back in 2004, researchers announced the discovery of this tiny, ancient human, which apparently hunted dwarf elephants with stone tools on the Indonesian island of Flores 18,000 years ago. Its discoverers called the 1-meter-tall creature Homo floresiensis, but skeptics wondered whether it was just a stunted modern human. In the years since, researchers have debunked many of the “sick hobbit” hypotheses. Yet scientists have continued to wonder where the species came from.
Now, an international team originally led by the hobbit discoverer reports stone tools, dated to 118,000 to 194,000 years ago, from another Indonesian island, Sulawesi, likely made by another archaic human—or possibly by other hobbits. “It shows that on another island we have evidence of a second archaic early human,” says paleoanthropologist Russell Ciochon of the University of Iowa in Iowa City, who was not involved with the work. The discovery makes the original hobbit claim appear more plausible, he says, by suggesting that human ancestors may have island-hopped more often than had been thought.
After international debate over the hobbit’s origins, co-discoverer Michael Morwood—then an archaeologist at the University of Wollongong (UOW) in Australia—set out to search other islands from which the tiny humans may have come. Java—more than 800 kilometers west of Flores but with a chain of islands in between—was already known to be the ancient home of the human ancestor H. erectus, a globe-trotting species that dates as far back as 1.7 million years ago. But Morwood instead set out for Sulawesi, 400 kilometers to the north, because powerful ocean currents sweep southward from this island toward Flores. Researchers had already found some simple stone tools on Sulawesi, but they couldn’t date the artifacts because they were found on the ground rather than buried with datable minerals.
Morwood assembled a team including paleontologist Gerrit van den Bergh, now of UOW. In 2009 they excavated a promising site, where they found tools down to a depth of about 4 meters. The simple stone flakes and choppers were clearly shaped by humans who knew how to hit rock to produce sharp flakes, says archaeologist and team member Mark Moore of the University of New England in Armidale, Australia. They were probably used to butcher animals such as the pigs and dwarf elephants whose remains were found nearby.
Morwood died in 2013, but team members continued the work. Using a method that estimates when certain mineral grains were last exposed to light, they dated sediment near some of the tools to 118,000 to 194,000 years ago, the team reports in Nature today. Other tools couldn’t be dated, but are from deeper layers and so must be older. The oldest H. floresiensis remains date back to 95,000 years ago, so toolmakers on Sulawesi were striking rocks not long before hobbits were doing the same on Flores.
But just who made the tools? The paper lays out the options: very early members of our own species, H. floresiensis itself, H. erectus, or the mysterious humans called Denisovans, known from a handful of fossils in Siberia in Russia, and by their genetic legacy in living people, including today’s Melanesians, near Sulawesi.
When pushed, Van den Bergh says he’d bet on H. erectus as the toolmaker, because remains of that species are found elsewhere in Southeast Asia as recently as 150,000 years ago. Ciochon agrees, and notes that the finds offer indirect support for a leading theory of the hobbit’s origin: that members of H. erectus washed up on Flores—perhaps rafting from Sulawesi—and then evolved into a dwarf form, as many other species have done on islands. Because islands are small and so have fewer resources for food, large animals often evolve into smaller forms after generations of being marooned; that’s why dwarf elephants have been found on both Flores and Sulawesi, for example. Stone tools on Flores date all the way back to 1 million years ago, also supporting the idea that H. erectus could have landed on the island early and evolved into the diminutive H. floresiensis over time.
Yet how could archaic humans have crossed hundreds of kilometers of open ocean to Sulawesi or Flores in the first place, especially if they lacked the smarts and technology of our own species? Van den Bergh suspects tsunamis, which are common in the region, may have washed them from island to island. He notes that in the Indian Ocean tsunami of 2003, people survived 7 days in the open ocean by hanging onto a floating tree. “Considering that big tsunamis strike the area perhaps once every hundred years, combined with deep geological time, I think that the chance of undeliberate sea crossings by early humans are significant,” he says. In any case, he says, the Sulawesi tools show that archaic humans were marooned on islands more often than anyone had thought.
Given the unanswered questions, researchers are itching to do more excavation on Sulawesi and other islands in hopes of connecting the dots between these discoveries. “The reason why paleoanthropology is so exciting is that occasionally it throws up something wholly unexpected that we can’t explain,” says archaeologist Robin Dennell of the University of Sheffield in the United Kingdom. “This is one of those.”