Until about 60,000 years ago, diminutive hominin cousins, Homo floresiensis (affectionately nicknamed Hobbits for obvious reasons), shared the island of Flores with Komodo dragons, pygmy elephants, and giant rats.
Based on the presence of hominin and pygmy elephant bones in the same layers of cave sediment, it originally looked like the Hobbits had hunted and butchered dwarf elephants—an impressive feat for such a tiny hominin. But according to University of Tübingen anthropologist Elizabeth Veatch and her colleagues, it was the Komodo dragons that were the hunters, while the Hobbits only showed up to scavenge what was left.
If Veatch and her colleagues are right, their findings may challenge some of the assumptions we’ve made about Homo floresiensis—and about which hominin species was the first to venture into the wider world beyond Africa.
These small hominins weren’t big-game hunters
Extinct pygmy elephant bones unearthed at Liang Bua (the cave site that also seems to have sheltered Homo floresiensis) are covered in marks from Komodo dragon teeth, as well as cut marks from stone tools. Based on these bones, we know that Hobbits and the ancient ancestors of today’s Komodo dragons shared a taste for the same type of meat: pygmy relatives of modern elephants, called Stegodon. At least three species of Stegodon lived on Flores, ranging from 1.25 to almost 2 meters tall and weighing anywhere from 500 kilograms to 1.5 tons.
To better understand the Stegodon bones and how they got to Liang Bua, Veatch and her colleagues started by feeding a nearly whole goat carcass to a Komodo dragon (as one does). The Komodo dragon at Zoo Atlanta had its best day ever, and the researchers compared what resulted to the Stegodon bones from Liang Bua.
The Komodo dragon has serrated teeth and a habit of gripping prey and then shaking its head side to side to rip the flesh away from the bone. This left distinctive marks on the bones, marks that were usually shallower, shorter, and wider than cut marks from stone tools. Veatch and her colleagues also noticed that the zoo’s Komodo dragon went straight for the meatiest parts of the body, which happened to be the same areas where archaeologists found tooth marks on the Stegodon bones at Liang Bua: parts like the limbs and the surprisingly fat-rich feet, as well as the ribs.
Stone tool marks, on the other hand, showed up on the less desirable parts. The pattern didn’t match what you would expect if hominins had first dibs on an elephant they’d just killed, but it did match what you’d expect if they were scrounging for leftovers after the Komodo dragons had eaten their fill. Veatch and her colleagues also found no evidence of fire in the Homo floresiensis layers of the site, which means they probably ate their leftover elephant bits raw. (At least one source claims that elephant meat can be tough to chew if it’s cooked over an open fire, so maybe the Hobbits were onto something.)
That challenges the earlier idea that Homo floresiensis organized and equipped themselves well enough to bring down something as large as a Stegodon. And it may add an interesting angle to an ongoing debate about where Homo floresiensis came from and which hominins were the first to migrate out of Africa. That debate has implications not only for understanding the Hobbits, but for how we make sense of 2-million-year-old stone tools at sites in China.
The Hobbits and “Out of Africa”
The most widely accepted origin story for the Hobbits (and some similarly short-statured cousins on a relatively nearby island) is that they’re descendants of a species called Homo erectus, which first appears in the fossil record around 1.9 million years ago in Africa. Within a few hundred thousand years, Homo erectus fossils show up everywhere: the Levant, Georgia, China, and Indonesia. In Indonesia, it may have been scattered among the islands in isolated pockets that eventually evolved into separate species, like Homo floresiensis and Homo luzonensis. Or at least that’s one hypothesis.
Homo erectus makes sense as the ancestor of the Hobbits, mostly because we typically think of Homo erectus as the first of our hominin ancestors to expand beyond Africa. The oldest hominin bones found anywhere outside Africa belong to Homo erectus: five skulls and hundreds of other bones, all dating between 1.77 million and 1.85 million years from Dmanisi Cave in Georgia. But stone tools tell a different story.
Stone tools from two sites in China seem to be older than Homo erectus. At Shangchen, a site on the southern edge of China’s Loess Plateau, archaeologists unearthed stone tools from a 2.1-million-year-old layer of sediment. And at the Xihoudu site in northern China, stone tools date to 2.43 million years ago. So either Homo erectus is older than we thought, or some other hominin species got there first. If that’s the case, then an even older member of our genus, like Homo habilis or Homo rudolfensis—species that anthropologists previously surmised weren’t adaptable enough to gain footholds in so many different parts of the world—may be the real ancestor of the Hobbits.
Veatch and her colleagues’ study, along with several previous studies of Homo floresiensis‘ anatomy and behavior, may lend some support to that idea.
“Evidence for behavioral complexity in Homo floresiensis, including complex tool and fire use, have weakened considerably over time,” wrote Veatch and her colleagues. “The evidence to date suggests that Homo floresiensis did not engage in a behavioral repertoire as diverse or as flexible as in modern humans or Neanderthals, possibly due to an ancestry in which large game hunting and controlled use of fire did not evolve.”
In other words, hominins descended from Homo erectus should have at least the same skills and cognitive abilities, which would include things like using fire and, potentially, organized hunting parties. Without those abilities, it’s easier to see Homo floresiensis as potentially descended from some earlier, less brainy species—still tool-users, but not fire-using big-game hunters—like Homo habilis or Homo rudolfensis.
Of course, we still don’t know the answer, and the evidence is complicated.
Evidence is clear; interpretation is tricky
Veatch and her colleagues’ recent study isn’t the first to challenge the idea that Homo floresiensis hunted big game and used fire, or that they descended from Homo erectus. Some of Homo floresiensis’ features, like the shapes of bones in its feet and the angle of its upper arms, suggest that it may be more closely related to Australopiths like Lucy than to Homo erectus.
On the other hand, despite having a smaller brain than Homo erectus (even relative to its body size), the contours of the inside of Homo floresiensis‘s skull suggest that its prefrontal cortex—an area of the brain associated with cognitive feats like planning and executive function—may have been fairly similar in size and structure to ours.
And it’s hard to say whether hunting Stegodon is a good measure of the Hobbits’ intellectual prowess anyway. Wielding projectile weapons, strategizing group hunts, and bringing down big prey all point to well-developed cognitive abilities (exactly the kind suggested by that hefty prefrontal cortex, in fact), and so does using fire to cook the meat afterward. However, hunting pygmy elephants may simply not have been worth the effort, even for a species that was otherwise capable of doing it.
When you calculate the cost, in time and energy, of killing, butchering, and transporting a carcass (which Veatch and her colleagues did, based on ethnographic data from people who live near modern elephants), elephants just aren’t worth the effort, let alone the risk.
“Although Homo floresiensis could gain a large total calorie return by successfully hunting Stegodon, the costs involved could potentially outweigh any social and/or caloric advantages,” Veatch and her colleagues suggest. Giant rats offered a much better return on investment; they rank ninth on Veatch and her colleagues’ list of Flores most appealing prey species, with Stegodon clocking in at 17th. And that’s probably why giant rat remains are so abundant at Liang Bua, in the layers of the cave associated with Homo floresiensis and the later Homo sapiens layers.
Even though it’s now clear that the Hobbits were not big-game hunters, and it doesn’t look like they used fire at Liang Bua, it’s much less clear what those facts tell us about the Hobbits’ brains or which branch of our family tree they actually belong on. As always, we need more evidence to fill in the details.
The questions we can answer
We know you’re wondering: Komodo dragons are venomous, so wouldn’t eating scraps scavenged from their kills have poisoned the Hobbits? The proteins in Komodo dragon venom are too large to pass through modern humans’ stomach linings and would probably have been broken down by the Hobbits’ digestive enzymes, according to Veatch and her colleagues. Please do not test this on yourself (but if you do, email us).
Scavengers would probably have been cautious about approaching kill sites, though. At around 3 meters long and weighing in at 70 kilograms, a Komodo dragon isn’t a beast you would want to encounter in a dark alley or even a reasonably well-lit Indonesian jungle. Now picture yourself as a Homo floresiensis, standing just over a meter tall. The Komodo dragon is three times your size, and has bony-armored skin, serrated teeth (the better to rip your flesh with), and a keen sense of smell (the better to track wounded prey with).
We don’t know whether Komodo dragons snacked on Hobbits back in the day, but there’s no evidence that they did, and Komodo dragons today mostly avoid humans, usually attacking only when they can’t flee.
Of the three species involved here (Homo floresiensis, Stegodon, and Komodo dragons), only the Komodo dragons survive today. Make of that what you will.
Science Advances, 2026 DOI: 10.1126/sciadv.aeb7219 (About DOIs).
![Security forces take strict security measures following two consecutive explosions that occurred near the Ministry of Tourism building, injuring 18 people including four security personnel, in Damascus, Syria on July 7, 2026. [Hişam Hac Ömer - Anadolu Agency]](https://i0.wp.com/www.middleeastmonitor.com/wp-content/uploads/2026/07/AA-20260707-41902968-41902961-EIGHTEEN_INJURED_IN_DOUBLE_EXPLOSIONS_NEAR_TOURISM_MINISTRY_IN_SYRIAN_CAPITAL_DAMASCUS-scaled-e1783624816272.jpg?fit=920%2C613&ssl=1)
