Over a million years ago, ancient human ancestors sat down to shave flakes off bones, producing a tool with a carefully created sharp edge.
According to Jackson Njau, an archaeologist at Indiana University in the US, this is not easy to do.
When he gets his students to try it, "they nearly complete the tool, and then they end up breaking it", he said.
"It's very, very hard."
Dr Njau and his colleagues have unearthed 27 ancient crafted bone tools, mostly from the legs of hippopotamuses and elephants in the Olduvai Gorge in Tanzania.
Their discovery, published today in Nature, pushes back the age of a large bone tool kit by more than a million years.
And that may change the way scientists think about when ancient humans started using crafted bone instead of stone as tools.
A wealth of fossils
The Olduvai Gorge has been a treasure trove of ancient human artefacts since the 50s.
Dr Njau and his colleagues have been investigating the dig site for almost 10 years.
The site, called the T69 Complex faunal assemblage, had been visited and excavated seven times between 2015 and 2022.
The team found a wealth of ancient material, including 9,000 animal fossils and 13,000 unidentified bone fragments.
It was in these deeper layers, dated to about 1.5 million years ago, that the team discovered that some of the unearthed bones had small scrapes to sharpen them up as tools.
Although the paper doesn't speculate on which species of human would have done this, Dr Njau suggests that it might have been Homo erectus, an ancient human with a large brain and among the first to leave Africa.
Sixteen out of the 27 bone tools could be linked to an animal species.
Eight were from elephants, six from hippos and two from a cloven-hoofed, ruminant mammal known as a bovid.
Almost all of them were made from the leg bones of these animals, which suggested they were likely scavenged. Dr Njau said.
Bones hard to preserve
The team had not expected to find bone tools.
Until now, Olduvai Gorge and the surrounding area has been known for its extensive stone tools and ancient human fossils.
But it wasn't an area known for bone tools.
[bone tool location]Michelle Langley, an archaeologist at Griffith University specialising in artefacts on hard animal materials and who was not involved in the research, said bones were harder to find.
"We have a lot of stone tools found from this period and earlier, because they're indestructible," she said.
But when it comes to bones, "the key problem is preservation".
Dr Langley notes that a million years is a long time for organic objects like bones to stay intact.
"If it's left on the surface, it can be eaten, pulled apart, weathered, attacked by insects … it doesn't last very long," she said.
But even for the bone tools that have lasted, Dr Njau said there could be an expertise problem in identifying them.
Mid-shaft bone fragments, like the tools found at the site, are commonly not valued highly by researchers, as they are hard to identify as a particular species.
And the scientists that are trained to be able to spot the slight markings on stone which designate it as a hand axe or other tool don't usually look at bones.
But with the discovery of this cache of intentionally designed bone tools, Dr Njau and Dr Langley think archaeologists are just scratching the surface of the timing of bone tool usage.
More bones to find?
While stone tools are common in the Lower Pleistocene around 1.5 million years ago, it was thought that bone didn't become the tool of choice until the Eurasian Middle Pleistocene around 500,000 years ago.
Occasional discoveries of bone tools had been found between 2.4 million and 800,000 years ago, but a collection as large as the new discovery shows that this wasn't a one-time event.
"They're showing that they were shaped — deliberately shaped — rather than just picking up a bone that might fortuitously be pointy," Dr Langley said.
And now researchers know to look for them, "we might see a series of these things starting to pop up", Dr Langley said.
"People could go back to faunal assemblages that were dug up 50 years ago and find [bone tools] that we weren't expecting to be there."
Dr Njau agrees, and suggests he might even investigate some of these unidentified bones in the lab collection when he goes on sabbatical.
"There will be more interest in checking the stuff that we already have in the labs and the museums," he said.
Dr Langley takes this a step further, suggesting there might be even older bone tools that we just haven't found due to poor preservation.
"Generally archaeologists believe that stone came first because we've got stone tools back to 3.3 million," she said.
"That's what we traditionally thought. It might not necessarily be true. It might just be owing to preservation."
