How to eat an elephant: fossil find in Tanzania shows oldest signs of butchering these giant mammals

Imagine a creature nearly twice the size of a modern African elephant (which can weigh up to 6,000kg. This was Elephas (Paleoxodon) recki, a prehistoric titan that roamed the landscape of what is now Tanzania nearly two million years ago. Now, imagine a group of our ancestors standing over its carcass, then butchering it and eating it.

For decades, archaeologists have debated when the hominin ancestors of humans first started eating megafauna – animals weighing more than 1,000kg.

In a new study, our team of archaeologists studying the evolution of the earliest humans in Africa has identified one of the earliest cases of elephant butchery.

This was at Olduvai Gorge in Tanzania, a site famous for containing some of the oldest and best preserved remains of our human ancestors. Dating back to 1.80 million years ago, this discovery at the site known as EAK reveals that our ancestors were engaging with megafauna substantially earlier than previously thought (about 1.5 million years ago was the previous estimate at Olduvai), and in a more sophisticated way.

This finding suggests that hominins (most likely, Homo erectus) may have been living in large social groups at this period, probably because their brains were developing and demanding higher-calorie diets rich in fatty acids.

‘Smoking guns’

Part of the reason our ancient diet has been debated is that it is not easy to find evidence of how much animal food early humans were eating and how they were acquiring it.

In traditional archaeology, the “smoking gun” for butchery (cutting up carcasses) is a cut mark left on a bone by a stone tool. However, when dealing with big animals like elephants, these marks are difficult to find. An elephant’s skin is several centimetres thick, and its muscle mass is so vast that a butcher’s tool might never touch the bone. Furthermore, millions of years of burial can weather the bone surface, erasing any subtle traces. And if a bone is deposited in an abrasive sediment, trampling by other animals may generate marks on bones that look like cut marks.

At the EAK site, we found the partial skeleton of a single Elephas recki individual in the same place as Oldowan stone tools. But to prove that this wasn’t just a natural death or the work of scavengers, we couldn’t rely on bone marks. Instead, we turned to a new kind of detective work: spatial taphonomy. This is the study of how stone artefacts and bones occur spatially on the same site. We also turned to more direct evidence: bones from those fossilised elephants that had been splintered while they were fresh (“green breaks”).

The geometry of a carcass

To solve this 1.8-million-year-old mystery, we analysed the way the bones were scattered across the site. Every agent that interacts with a carcass – whether it’s a pride of lions, a group of hyenas, or a band of humans – leaves a unique “spatial fingerprint”. Lions and hyenas tend to drag bones away, scattering them in predictable patterns based on their weight and the amount of attached meat. Natural deaths, like an elephant dying in a swamp, result in a different, more localised skeletal “collapse”.

By using advanced spatial statistics, and later comparing the EAK site to several modern elephant carcasses that we studied in Botswana (not yet published), we found that the spatial configuration at EAK was unique. The clustering of the bones and the density of the stone tools among them did not match the “random” or “scavenger-driven” models. Instead, it reflected a focused, high-intensity processing event. The spatial signature was a match for hominin butchery, which has also been documented at Olduvai sites that are half a million years younger.

This was confirmed by the presence of green-broken long bones not just at EAK, but in several locations in the landscape where other elephant and hippopotamus carcasses were butchered. Today, only humans can break elephant long bone shafts; not even spotted hyenas, which have very powerful jaws, can do it.

Glimpses of this behaviour can be detected at other sites too. For example, a cut-marked bone fragment of a large animal (probably a hippopotamus) was documented at El-Kherba (Algeria) dated to 1.78 million years ago.

This intensive and repeated discovery of multiple elephant and hippopotamus carcasses butchered at different landscape locations indicates that humans were butchering the remains of large animals, whether hunted or scavenged.

Why does an elephant meal matter?

This discovery isn’t just about a prehistoric menu; it’s about the evolution of the human brain and social structure. There is a long-standing theory in paleoanthropology called the “expensive tissue hypothesis”. It suggests that as our ancestors’ brains grew larger, they required a massive increase in high-quality calories, specifically fat and protein. Large mammals like elephants are essentially giant “packages” of these calories. Processing even a single elephant provides a caloric windfall that could sustain a group for weeks.

Butchering an elephant is a monumental task, however. It requires sharp stone tools and, most importantly, social cooperation. Our ancestors had to work together to defend the carcass from predators like sabre-toothed cats and giant hyenas, while others worked to extract the meat and marrow.

This suggests that even 1.8 million years ago, our ancestors already possessed a level of social organisation and environmental awareness that was truly “human”.

The discovery also has another dimension. Humans at that time, like modern carnivores, consumed animals whose size was related to their own group size. Small prides of lions eat wildebeests; larger prides eat buffalo and in some places even juvenile elephants. The evidence that those early humans were exploiting large animals comes in parallel with evidence that they were living in much larger sites than before, probably reflecting bigger group sizes.

Why early humans started living in large groups at that time remains to be explained, but this indicates that they certainly needed more food.

A shift in the ecosystem

The EAK site also tells us about the environment. By analysing the tiny fossils of plants and microscopic animals found in the same soil layers, we reconstructed a landscape that was transitioning from a lush, wooded lake margin to a more open, grassy savanna. Our ancestors were already eating smaller game. There is evidence that two million years ago, they were hunting small and medium-sized animals (like gazelles and waterbucks). A little earlier, they began using technology (stone tools) to bypass their biological limitations.

Read more:
Large mammals shaped the evolution of humans: here’s why it happened in Africa

The evidence from Olduvai Gorge shows that our ancestors were remarkably adaptable, capable of thriving in changing climates by developing new behaviours.

As we look at the spatial layout of these ancient remains, we aren’t just looking at the bones of an extinct elephant. We are looking at the traces of a pivotal moment in our own history – when a small group of hominins looked at a giant and saw not just a threat, but a key to their survival.