Here is the twist: the “hobbit” species Homo floresiensis likely did not hunt dwarf elephants or control fire. Instead, analysis of Stegodon florensis insularis bones from Liang Bua cave suggests hobbit humans scavenged what Komodo dragons left behind.

That’s not a vibe check. Researchers counted 54 cut marks on Stegodon bones and nearly twice as many Komodo dragon tooth marks. Even more telling, the Komodo dragon marks concentrated in meaty areas, while the human-made cut marks showed up primarily where there was less meat, aligning with a scavenging strategy rather than active big-game hunting.

Why does this matter beyond satisfying your inner anthropology nerd? Because it strikes at one of the core assumptions about early humans: that tool-assisted big-game hunting and fire use were the big leaps in behavioral sophistication. Homo floresiensis arrived on the Indonesian island of Flores at least 700,000 years ago, and it disappeared around 50,000 years ago as Homo sapiens spread through Southeast Asia. When the species was first discovered in 2003, it earned the nickname “hobbit” for good reason: diminutive size averaging around 3 feet, 6 inches (106 centimeters) tall, plus a small brain, large teeth, and big feet.

Archaeologists have also found stone tools, animal bones with cut marks, and charred bones at the site, which previously supported the idea that Homo floresiensis had sophisticated behaviors similar to those often discussed within the genus Homo. But this new study is trying to answer a narrower, sharper question: do those marks reflect hunting Stegodon meat, or scavenging the leftovers from the island’s main carnivore, Komodo dragons (Varanus komodensis)? The researchers focused on Stegodon florensis insularis, an extinct dwarf species of elephant relative discovered at Liang Bua cave, where both H. floresiensis and stone tools have also been found.

The key step was separating “human tool cuts” from “Komodo tooth damage.” To do that, the team conducted an experiment feeding a goat carcass to a captive Komodo dragon at Zoo Atlanta. They then recovered the goat skeleton and carefully documented the tooth marks, pits, notches, and furrows made by the Komodo dragon’s teeth. The logic is simple and brutally practical: if the tooth pattern is consistent, you can compare modern tooth signatures to the fossil record.

With that baseline, researchers examined ancient Stegodon bones for two competing signatures: cut marks made by H. floresiensis stone tools, and tooth marks made by Komodo dragons. They report that Komodo tooth marks were concentrated in areas with substantial amounts of goat flesh, suggesting Komodo dragons prefer meaty areas. When applied to Stegodon bones, the combined pattern implied “a combination of mostly primary access by Komodo dragons and secondary access by H. floresiensis where both predators consumed Stegodon.” Translation: Komodo dragons did most of the heavy lifting first, and hobbit humans got their share afterward.

The study also targets fire and cooking, because charred bones can be misleading. The researchers found 54 cut marks on Stegodon bones and nearly twice as many Komodo dragon tooth marks, but the smoking gun is what they did not find. They found no evidence on Stegodon bones that the meat had been cooked. They also found no evidence of burning on over 4,000 mouse bones from the site. The team argues that the earlier evidence of charring was likely natural manganese staining.

That absence matters for behavioral interpretation. The researchers say a lack of hunting and fire-making technology suggests Homo floresiensis was not as behaviorally sophisticated as previously thought. For ancestry debates, this doesn’t fully settle the picture, but it reframes it: it raises questions about whether Homo floresiensis branched from a hominin population that did not require dietary strategies like hunting and cooking, as study first author E. Grace Veatch, a paleoanthropologist at the University of Tübingen in Germany, told Live Science in an email.

Veatch also points to the possibility that the ancestor of H. floresiensis diverged from the Homo genus before humans accomplished control of fire and hunting. Island dwarfism is one hypothesis for the origin of hobbits, where limited resources on an island shrink average body size over generations. Another theory is that hobbits descended from an earlier Homo species that was already small-bodied. Still, the ancestry question remains open because very little is known about early hominins in Southeast Asia, such as Homo erectus on Java and other areas of Sunda or Sundaland, a landmass exposed off and on over the past 2.6 million years.

If H. floresiensis really branched from H. erectus, that could imply many evolutionary changes, not just anatomy like reduced body size and brain volume, but also behavioral adaptations. Adam Brumm, an archaeologist at Griffith University in Australia, was not involved in the study and told Live Science that “Flores was clearly a wild card in the story of early human evolution,” including the potential loss of deeply-rooted hominin behaviors like hunting and fire use.

For executives and boardrooms, the second-order lesson is surprisingly applicable even if you never plan to fund a cave excavation. This study leans hard on method, not mythology: taphonomy, the study of what happens to organic remains after death, can reinterpret signals that initially looked like advanced behavior. In business terms, it is a reminder that “evidence” can be distorted by the environment it travels through. For decision-makers tracking emerging narratives in tech, biotech, or any data-heavy domain, the strategy is the same: isolate mechanisms, stress-test assumptions, and beware conclusions built on signals that can have multiple causes.

Homo floresiensis still leaves a lot of unanswered questions about where it fits in the Homo family tree. But for now, the bone marks and the missing cooking evidence directly change the story: hobbit humans appear to have scavenged Stegodon, with Komodo dragons holding the primary access, and with no clear signs that humans cooked the meat or used fire.