Scientists Discover 3.3-million-year-old Stone Tools, Predating Big-brained Humans by 500,000 Years

Who made the primitive tools found in Lomekwi, Kenya isn't clear, yet they had more dexterity and planning than apes seem capable of producing

West Turkana Archaeological Project

Crude but unmistakable stone tools dating back 3.3 million years have been found in Kenya, well before modern humans were a gleam in some ape's eye.

Who made the tools, of which 149 were found, is anybody's guess. The conventional wisdom has been that early humans began making such accessories only when pressed by environmental change to adapt to the spreading African savannahs and dwindling woodlands. But first of all, the beings who made the tools found in Lomekwi, Kenya lived in a shrubby, woody environment, the scientists demonstrate.

Secondly, who says the makers were ancestral to our genus, the genus Homo?

Previously, the oldest-known tools were 2.6 million years old. Those postdate the oldest-known fossils associated with human-lineage hominins, which go back to 2.8 million years. ("Hominins" include modern humans, Homo sapiens, and our direct predecessors, whoever they may have been. The definitive hominin characteristic is walking on two legs, known as “bipedality”.)

Not how apes do it

West Turkana Archaeological Project

Other animals have also been known to use tools, and even to perfect them. Just this week wild bearded capuchins in Brazil (which last shared lineage with humans 35 million years ago) were observed cleverly selecting appropriate stones to crack nuts. They chose stones carefully and modulated their strikes: As the nut began to crack, they restrained their blows lest they crush the soft internal fruit into unappetizing oblivion. Ergo, they could plan. Rooks have been observed making tools cleverly, in captivity at least, and so have a host of other animals.

However, Christopher Lepre, of the Lamont-Doherty Earth Observatory and Rutgers University, the paper’s co-author who dated the artifacts, argues the case that the maker was either our ancestor, or closely related to it.

"The type of deliberate stone knapping that we see in the archaeological record is often assumed to be only the work of our lineage because there's planning, and the rocks have signature fracture patterns," he explains. (Knapping is banging one rock against another together to produce sharp-edged stone flakes.)

Experiments have been done in which captive primates make stone tools. The traditional way early hominins made stone flakes is to take one rock in hand and use the other as a hammer, knocking off flakes from the rock held in the first hand. But the apes' method was to take the rock and throw it against the floor, where it would shatter. "That produces different fracture patterns that we don't see in the archaeological record," Lepre sums up.

That does not prove that the maker of the Kenyan stones were direct predecessors of humans. They could have been some lost lineage. 

One possibility for the maker is Kenyanthropus platytops: a skull from this species was found in 1999 about a kilometer from the tool site, and a tooth was found even closer. But it isn't known whether K. platyops is a human predecessor: this hominin predates the earliest known Homo species by a half a million years. (On March 4, a jawbone found in Ethiopia pushed the fossil record for the genus Homo to 2.8 million years ago.)

Australopithecus afarensis also existed in the area at the same time and there could have been any number of other species that we don't even know about.

Also, fossilization is hit-or-miss. There could have been a burgeoning community of unknown hominins living there who made the tools, whose remains are lost.

The most primitive tools found yet

The previously "oldest" tools dating from about 2.6 million years ago, found at Gona in Ethiopia, were more advanced than the set now found in Kenya, Lepre tells Haaretz. That Oldowan "tool kit" – which had been considered the most ancient until now - had a number of different types of tools – choppers and bashers and flakes; the tools now found in Kenya do not seem to have reached that level of sophistication.

"It's unclear that if there's an assemblage that we see," Lepre says. Also, the Kenyan tools, which are made of hard lava rocks, are massive, averaging 3.1 kilos.

With "younger" tool kits, the hominins would take one rock in one hand, and had the manual dexterity to manipulate and strike carefully, producing precisely crafted tools. That doesn't seem to be the case in Lomekwi, not least because the rocks were too big and heavy.

"They did a lot of bashing. They put one stone on the ground and then took another stone held in two hands and bashed it on the stone on the ground," Lepre surmises.

Possibly hominin tool use is even earlier. Animal bones marked with slashes and other cut marks dating from 3.39 million years ago were discovered in Ethiopia: it seems someone used stones to trim flesh from bone and perhaps crush bones to get at the marrow inside. No tools were found at the site. (The only hominin fossil remains found in the area from that time are Australopithecus afarensis.)

Dating the site

The site was dated with the help of a layer of volcanic ash below the tool site, which matched ash elsewhere that had been dated to about 3.1-3.3 million years ago.

"We used the record of earth's magnetic field preserved in the sedimentary rocks," Lepre explains. "Earth's magnetic field changes from normal polarity (i.e., like today) to reverse polarity throughout geological time. We matched the series of magnetic polarity reversals from the archaeological-bearing rocks to a known inventory of dated magnetic polarity reversals."

The one thing we know for sure is that stone tool use became quite the trend among hominins, but the find in Kenya is no "factory" like the extraordinary discovery in the Messak Settafet escarpment in Libya , where stone tools litter the landscape. That too is associated with hominins who preceded Homo sapiens, but it's much younger than the new discoveries. "There the human derivation is much clearer," says Lepre.