Biology Magazine

First Meat-eating Humans Were Actually Crocodiles

Posted on the 23 November 2017 by Reprieve @EvoAnth

Humans eat more meat than any other primate alive. However, tracking the evolution of this trait is kind of difficult. After all, our ancestors didn't keep recipe books. To get around this issue archaeologists will look for other lines of evidence. Like tool cut marks on animal bones. Except it turns out it might actually be crocodiles that are the cause of this damage.

This revelation might pull back the date for our shift to carnivory by several million years. The earliest evidence we have of meat-eating hominins comes from these bone marks. Some are more than 3 million years old. However, if it turns out crocodiles are the cause then we'd have to look at younger, but more reliable evidence.

Such a shift would really shake up what we thought we knew about both hominin meat eating and ancient tool use.

U wot m8

The real issue here comes from the fact that we're dealing with fossils millions of years old. There's a whole bunch of things that can happen to them in that time. They could be trampled, eroded, buried, eaten, washed away, and more.

This gets even more confusing considering you then have to multiply this by all the different ways those could happen. Take trampling for instance. Our fossil could get in the way of an elephant, or zebra, or any of the other fun stuff you find in Africa. Each will leave a different damage pattern on the bone, making it extra difficult to study.

As such, it's an ongoing effort to test what happens to bones after death. So researchers have been giving bones to lions, hyenas, and more to see what happens when they have their way with them. They then compare these bones to those with tool damage to figure out how to tell them apart. The general trend of this research is that carnivore damage typically leaves a shallower, more U-shaped groove than tool damage.

Enter crocodiles. They've been studied before since they tend to eat bones. This causes damage as they pass through the digestive tract (which is another thing scientists have to keep track of to figure out what happened to fossils). However, comparatively little has been done on the toothmarks they create as they eat.

So researchers have been trying to fill in this gap in our knowledge. And guess what, it turns out they leave behind those same V-shapes we thought only tools made. Uh oh.

Gnawing crocodiles

Armed with this information, some of the people involved in the initial discovery of ancient "tool marked" bones have returned to the scene of the crime. Re-examining their discoveries with the knowledge of how crocodiles damage bones has really shaken their confidence. They now doubt that any of the damage was made by hungry hominins.

Important discoveries we may have to throw out include:

  • The oldest known case of tool damage on possible prey is a 3.39 million-year-old bovid fragment from Dikika, Ethiopia. Re-evaluation reveals that the damage is actually consistent with crocodile teeth. To seal the deal, crocodile teeth were found at the same place.
  • Australopithecus garhi is the pre-human species most closely linked to meat-eating thanks to bovid cut marks found with them at Bouri, Ethiopia. But again, the damage is also consistent with crocodiles.
  • An Australopithecus with tool damage was found at Asa Issie, Ethiopia. Is this the oldest evidence? Actually, a crocodile just ate the unfortunate individual.

Crucially though, they couldn't rule out that hominins were responsible for this damage. Simply that it was impossible to tell the marks apart from those made by crocodiles. In many cases, the fact that the bones were found in deposits made by rivers led them to conclude crocodiles were likely responsible. However, there was nothing about the damage alone that could distinguish between humans and crocodiles.

Based on all this, it looks like we can't confirm predatory human relatives until around 2 million years ago. That's more than a million years younger than the "oldest" evidence: the Dikika bone that a crocodile ate.

Avoiding the crocodiles

At the end of the day, there's one thing I find scary about these crocodiles. It's not their death rolls or creepy stares, but the fact that it was impossible to tell crocodile tooth damage from tool damage apart based on the bones alone. They had to look to context to try and figure it out.

That means there's going to be an element of subjectivity in these assessments. And in future discoveries. Any new finds will get the side eye - perhaps reasonably - simply if they were found near water.

What's more, the people behind this research aren't sure this problem will ever go away. They suggest that, in many situations, we may have to abandon the quest to figure out what caused bone damage from the bones alone.

Although somewhat disappointing, I think this added skepticism is ultimately a good thing. "Groundbreaking" discoveries about tool use and meat eating often only have a bone or two to support them. Often with dubious context. Forcing people to think twice will help improve the quality of research, even though that does make the final conclusion harder to reach.

But of course, extra work keeps me in a job at least.


Coil, R., Tappen, M. and Yezzi‐Woodley, K., 2017. New analytical methods for comparing bone fracture angles: A controlled study of hammerstone and hyena (Crocuta crocuta) long bone breakage. Archaeometry.

Sahle, Y., El Zaatari, S. and White, T.D., 2017. Hominid butchers and biting crocodiles in the African Plio-Pleistocene. Proceedings of the National Academy of Sciences, p.201716317.

Back to Featured Articles on Logo Paperblog