Homo floresiensis: The 'Hobbit' Humans of Flores, Indonesia

Homo floresiensis, nicknamed the “hobbit”, was a species of tiny human that lived on the Indonesian island of Flores. Its remains were discovered in 2003 in Liang Bua cave by an Indonesian and Australian team and announced in October 2004. Adults stood about one metre tall and had brains roughly the size of a chimpanzee’s, yet they made stone tools and survived on the island until around 50,000 years ago. The find upended assumptions about human evolution and remains one of the most debated discoveries in palaeoanthropology.

Background: Flores and Liang Bua

Flores is a long volcanic island in the Lesser Sunda chain, east of Java and Bali. It lies beyond the Wallace Line, the deep-water boundary that separates the animals of Asia from those of Australia and New Guinea. Even when Ice Age sea levels were more than 100 metres lower than today, Flores remained an island, so any land animal reaching it had to cross the sea.

Liang Bua, meaning “cool cave” in the local Manggarai language, is a large limestone cave in the hills of western Flores, about 14 kilometres north of the town of Ruteng. Indonesian archaeologists, notably Raden Pandji Soejono of the National Research Centre for Archaeology, excavated its upper layers from the 1970s, finding burials of recent modern humans.

In 2001 Soejono and the Australian archaeologist Mike Morwood of the University of New England launched a joint project to dig deeper. Morwood’s earlier work in the So’a Basin of central Flores had shown that stone tools there were roughly 800,000 years old, implying that early humans had crossed the sea long ago. The team hoped Liang Bua would reveal who those people were.

The discovery of LB1

In September 2003, at a depth of about six metres, the team uncovered a small skeleton. Catalogued as LB1, it included a nearly complete skull and jaw, much of the legs, parts of the pelvis, hands and feet. The bones were not mineralised like typical fossils; they were damp and very fragile, and had to be dried and consolidated before they could be lifted.

At first the team thought it was a child. Worn teeth and closed skull sutures showed it was an adult, probably female. Its height was estimated at about 1.06 metres and its weight at roughly 25 to 30 kilograms. Its brain volume was initially measured at about 380 cubic centimetres; later scans gave figures of around 420 cubic centimetres. That is similar to a chimpanzee or an early australopith, and about a third the size of an average modern human brain.

On 28 October 2004 the team, with the physical anthropologist Peter Brown describing the skeleton, published it in Nature as a new species, Homo floresiensis, “human from Flores”. Further excavations recovered bones and teeth from several more individuals, with estimates of up to about 15, including a second jaw. They confirmed that LB1’s small size was typical of the population.

Anatomy: a strange mix

Homo floresiensis combined features rarely seen together. Its skull had a low, long braincase and thick brow, like Homo erectus, but in miniature. It had no chin. Its arms were relatively long and its legs short.

The feet were remarkably long for its height, about 20 centimetres in LB1, with a flat arch and short big toe, suggesting it walked upright but perhaps less efficiently over long distances than modern humans. The wrist bones, studied in 2007, resembled those of African apes and early hominins rather than those of Neanderthals and modern humans. The shoulder too had primitive features.

Despite its small brain, the species made stone tools, mostly simple flakes struck from cores, similar in technique to much older tools from the So’a Basin.

Dating: from 18,000 to 60,000 years

The first studies dated the Liang Bua remains to as recently as 18,000 years ago, with the hominin remains thought to span about 38,000 to 18,000 years ago and the species’ stone tools to persist until about 12,000 years ago. This suggested that a second human species had survived until the end of the Ice Age, long after modern humans reached Australia.

Continued excavation showed that the cave’s layers were more complex than first thought. An ancient erosion surface meant that older sediments sat beside younger ones. In 2016 a team led by Thomas Sutikna published a revised chronology using several dating methods: the Homo floresiensis skeletal remains are about 100,000 to 60,000 years old, and stone tools attributed to the species range from about 190,000 to 50,000 years ago. The very young dates were withdrawn.

Also in 2016, researchers led by Gerrit van den Bergh reported a jaw fragment and teeth from Mata Menge in the So’a Basin, dated to about 700,000 years ago. They closely resemble Homo floresiensis and are, if anything, slightly smaller. A small arm bone from the same site, published in 2024, pointed to individuals even shorter than LB1. Stone tools at Wolo Sege in the same basin are about one million years old.

The pathology controversy

The discovery provoked fierce disagreement. The Indonesian palaeoanthropologist Teuku Jacob of Gadjah Mada University argued that LB1 was a modern human with microcephaly, a condition causing an abnormally small head. In late 2004 he took most of the bones to his laboratory in Yogyakarta, and when they were returned in early 2005 some, including the pelvis and a jaw, had been damaged.

Other researchers proposed Laron syndrome, a growth-hormone disorder, endemic cretinism caused by iodine deficiency, or Down syndrome. Each hypothesis attracted rebuttals. The main arguments against a disease explanation are that several individuals share the same anatomy, that the wrist, shoulder and foot are primitive rather than abnormal, and that the Mata Menge fossils show small-bodied humans on Flores some 600,000 years earlier. Most specialists now accept Homo floresiensis as a genuine species.

Origins and island dwarfism

How did such a small human arise? One explanation is insular dwarfism: on islands with limited food and few large predators, big animals often evolve to smaller sizes over many generations. Flores itself offers examples, including a dwarf form of the elephant relative Stegodon. On this view, Homo floresiensis descended from Homo erectus, perhaps from a population like Java Man, which shrank after reaching the island. The Mata Menge fossils show this could have happened within about 300,000 years.

A second view notes that features of the wrist, feet and jaw are more primitive than in Homo erectus. Some anatomical analyses, such as one published in 2017, place Homo floresiensis closer to early species like Homo habilis, implying that a small-bodied hominin left Africa earlier than currently documented. The question remains unresolved, and no DNA has been recovered from the bones.

The discovery in 2019 of Homo luzonensis in the Philippines, another small island species, strengthened the idea that isolated islands in South-East Asia produced distinctive human lineages.

Life on a strange island

Homo floresiensis shared Flores with an unusual set of animals. These included the dwarf Stegodon, giant rats, Komodo dragons, and a giant marabou stork that stood about 1.8 metres tall. Bones of young Stegodon with cut marks at Liang Bua suggest the hobbits hunted or scavenged them. Vultures and storks would also have competed for carcasses.

The species vanished from Liang Bua around 50,000 years ago, at about the same time as the dwarf Stegodon, the stork and vultures. Evidence of modern humans in the region appears shortly afterwards, and layers of volcanic ash in the cave record eruptions around the same period. Whether newcomers, climate change, volcanic activity or a combination of these caused the extinction is not known, and researchers continue to search for younger traces of the species elsewhere on the island.

Myths and facts

• Myth: Homo floresiensis lived until 12,000 years ago. Fact: Those early dates were revised in 2016; the remains are about 100,000 to 60,000 years old.
• Myth: Hobbit genes survive in local pygmy people. Fact: A 2018 genetic study of the short-statured Rampasasa people living near Liang Bua found no trace of Homo floresiensis ancestry; their small stature evolved separately.
• Myth: The Ebu Gogo folk tales prove the hobbits survived into recent times. Fact: Local legends of small wild people are intriguing but there is no evidence linking them to the species.

Timeline

• c. 1 million years ago: Stone tools at Wolo Sege show humans on Flores.
• c. 700,000 years ago: Small hominins at Mata Menge, probably ancestors of Homo floresiensis.
• c. 190,000-50,000 years ago: Stone tools attributed to Homo floresiensis at Liang Bua.
• c. 100,000-60,000 years ago: Age of the Homo floresiensis skeletons at Liang Bua.
• 2001: The Indonesian-Australian excavation at Liang Bua begins.
• 2003, September: The LB1 skeleton is found.
• 2004, 28 October: Homo floresiensis is published in Nature.
• 2016: Revised dating of Liang Bua and publication of the Mata Menge fossils.
• 2024: A tiny arm bone from Mata Menge indicates even smaller early hominins.

Why it matters

Homo floresiensis showed that human evolution was not a straight march towards bigger brains and bodies. A small-brained species made tools, crossed the sea and survived on an island for hundreds of thousands of years, overlapping with Neanderthals, Denisovans and early modern humans. Its story is still unfolding, and it remains a reminder of how diverse the human family once was.