Australian Megafauna: Aboriginal Evidence

What megafauna were, and who lived alongside them

During the Pleistocene, Australia was home to a range of giant animals, its megafauna, and Aboriginal Peoples were already living here while many of these animals still walked the land.

Megafauna are large-bodied animals. Australia's extinct megafauna included Diprotodon optatum (the largest marsupial ever known, a relative of the wombat estimated at around 2.7 tonnes), Genyornis newtoni (a flightless bird around 2 m tall and roughly 230 kg), Thylacoleo carnifex (the marsupial "lion", a predator with an enormous slicing premolar and a hooked thumb claw), the short-faced kangaroo Procoptodon goliah, and the giant goanna Varanus priscus. Most of these species disappear from the record by roughly 40,000 years ago.

Critically, Aboriginal Peoples were already here. Excavation of the Madjedbebe rock shelter on Mirarr Country in the Northern Territory places human occupation at about 65,000 years ago, dated by optically stimulated luminescence on the sediments and supported by more than 10,000 stone artefacts (Clarkson and colleagues, Nature, 2017). That means people and megafauna shared Country for many thousands of years, which is exactly what makes Aboriginal records a direct source of evidence: the people who made them could have seen these animals alive.

At Cuddie Springs on Ngiyampaa Country in north-western New South Wales, stone artefacts and megafauna bone have been recovered from the same sediment layers, which has been read as direct evidence of overlap (Field and Dodson, 1999). The dating of those layers is contested (Grun and colleagues, 2010), and that disagreement is worth noticing rather than hiding: a single site rarely settles a question in palaeontology.

Pause, copy the highlighted megafauna examples into your book before moving on.

Two forms of Aboriginal palaeontological evidence

We just established the overlap between people and megafauna. That raises a question: what form does the Aboriginal record actually take? This card answers it → two main kinds: images on rock, and knowledge carried in oral traditions.

Aboriginal evidence for megafauna comes mainly through rock art that depicts the animals and oral traditions that describe creatures matching the extinct species.

Rock art, case study 1: a marsupial lion in the Kimberley. A painted figure recorded in the north Kimberley of Western Australia shows a striped, thick-tailed animal with a heavy forelimb and a prominent clawed first digit. Kim Akerman identified it as Thylacoleo carnifex, arguing that the striping, the tail and above all the enlarged thumb claw match the skeleton and match no living Australian animal (Akerman, The Beagle, 1998; Akerman and Willing, Antiquity, 2009). If the identification holds, the artist had seen the animal, because that anatomy is not something that could be reconstructed from bones alone.

Rock art, case study 2: a large flightless bird in Arnhem Land. A painting recorded in western Arnhem Land shows a heavily built bird with a deep bill, which the recording team suggested might depict Genyornis newtoni (Gunn, Douglas and Whear, Australian Archaeology, 2011). Other specialists have argued the figure is a stylised emu or an Australian bustard rather than Genyornis (Welch, Rock Art Research, 2016), and a later morphometric re-test compared the painting against the emu and the magpie goose. Both cases are interpretations, and specialists disagree. That is normal science, not a weakness: an identification is a claim to be tested against anatomy, dating and independent evidence.

Oral tradition, case study 3: the mihirung paringmal. Tjapwurrung traditions from western Victoria, recorded in the nineteenth century, describe a giant emu-like bird hunted by people. The Tjapwurrung name for it, mihirung paringmal, was later adopted by palaeontologists as the common name for the whole family of extinct giant birds, the dromornithids or "mihirungs" (Rich, The Dromornithidae, 1979).

Why an oral tradition can be trusted over such a distance in time. Gunditjmara and Tjapwurrung traditions describe volcanic eruption in western Victoria. Argon dating of the Budj Bim and Tower Hill eruptions places them at roughly 37,000 years ago, and a stone tool recovered beneath the Tower Hill ash shows people were present before the eruption (Matchan and colleagues, Geology, 2020). If a tradition can carry an accurate account of a specific event across tens of thousands of years, the same rigour applies to what those traditions say about the animals of that time.

Two forms of Aboriginal palaeontological evidence: (1) rock art depicting large animals matching megafauna, such as the Kimberley figure identified as Thylacoleo carnifex and an Arnhem Land bird suggested to be Genyornis newtoni, and (2) oral traditions describing such animals, such as the Tjapwurrung mihirung paringmal. Both indicate that people lived alongside, and observed, these animals. Identifications in rock art are interpretations and are debated, which is why they are weighed alongside dated fossil evidence rather than on their own.

Pause, copy the two forms of evidence and an example of each into your book.

How Aboriginal evidence complements the fossil record

We just described the two forms of Aboriginal evidence. That raises a question: how do they fit with the fossil bones scientists dig up? This card answers it → they complement each other, and together they are stronger than either alone.

No single source gives the whole picture; combining Aboriginal knowledge with the fossil record produces a fuller and more reliable account of the megafauna.

Fossil bones give the animal's size, skeleton and, through dating methods, an estimate of when it lived. Aboriginal rock art can show external appearance, posture and details soft tissue does not preserve. Aboriginal oral traditions can preserve information about behaviour, habitat and the relationship between the animal and Country.

When these sources agree, the bones, the art and the oral record all pointing to the same animal, the conclusion that the megafauna existed, and that people lived alongside them, becomes very strong. This is a clear example of why scientists value multiple, independent lines of evidence, and why Aboriginal and Torres Strait Islander knowledge is an important part of Australian science.

A worked example of a claim being tested. Miller and colleagues dated large eggshell from arid-zone sites and showed that it disappears from the record about 50,000 years ago (Science, 1999). Burnt fragments of that eggshell, found in tight clusters with scorch patterns graded across single pieces, were later argued to be the remains of human cooking fires, and so evidence that people had harvested Genyornis eggs (Miller and colleagues, Nature Communications, 2016). That claim was then challenged: other researchers argued the eggshell was too small and too thin for Genyornis and instead belonged to a giant extinct megapode (Grellet-Tinner and colleagues, 2016). The dispute was settled by a method neither side had used, sequencing the ancient proteins preserved in the shell, which matched the eggshell to Genyornis newtoni (Demarchi and colleagues, PNAS, 2022).

Read that sequence carefully, because it is how palaeontology actually works: a claim, a challenge on anatomical grounds, then an independent method that resolves it. It is also why no line of evidence, Aboriginal or Western, is asked to carry a conclusion alone.

Fossil bones, Aboriginal rock art and Aboriginal oral traditions each contribute different information (size/dating; appearance; behaviour/habitat). When independent sources agree, the conclusion is far stronger. Aboriginal knowledge is valid, complementary palaeontological evidence, not a replacement for, nor inferior to, the fossil record.

Add the "different sources, different contributions" summary to your notes before the check below.

Every case study above is drawn from published work, so that you can check it rather than take it on trust. This is also how you should treat any claim you meet about the deep past.

• Clarkson, C. and colleagues (2017). Human occupation of northern Australia by 65,000 years ago. Nature 547, 306 to 310. Madjedbebe, Mirarr Country.
• Field, J. and Dodson, J. (1999). Late Pleistocene megafauna and archaeology from Cuddie Springs, south-eastern Australia. Proceedings of the Prehistoric Society 65, 275 to 301. The dating of the deposit is disputed by Grun, R. and colleagues (2010), Quaternary Science Reviews 29 (5 to 6), 596 to 610.
• Akerman, K. (1998). A rock painting, possibly of the now extinct marsupial Thylacoleo, from the north Kimberley. The Beagle: Records of the Museums and Art Galleries of the Northern Territory 14, 117 to 121; and Akerman, K. and Willing, T. (2009). An ancient rock painting of a marsupial lion, Thylacoleo carnifex, from the Kimberley. Antiquity 83 (319).
• Gunn, R. G., Douglas, L. C. and Whear, R. L. (2011). What bird is that? Identifying a probable painting of Genyornis newtoni in western Arnhem Land. Australian Archaeology 73, 1 to 12. The identification has been questioned by other researchers.
• Rich, P. V. (1979). The Dromornithidae. Bureau of Mineral Resources Bulletin 184. Records the Tjapwurrung term mihirung paringmal and its adoption as the common name of the family.
• Matchan, E. L., Phillips, D., Jourdan, F. and Oostingh, K. (2020). Early human occupation of southeastern Australia: new insights from 40Ar/39Ar dating of young volcanoes. Geology 48 (4), 390 to 394.
• Miller, G. H. and colleagues (1999). Pleistocene extinction of Genyornis newtoni. Science 283, 205 to 208, which dated the eggshell and showed its disappearance about 50,000 years ago. The burnt-eggshell argument for human harvesting is Miller, G. H. and colleagues (2016). Human predation contributed to the extinction of the Australian megafaunal bird Genyornis newtoni about 47 ka. Nature Communications 7, 10496. Challenged by Grellet-Tinner, G. and colleagues (2016), and resolved by Demarchi, B. and colleagues (2022), Ancient proteins resolve controversy over the identity of Genyornis eggshell, PNAS 119 (43).