Although, as the authors of this new Nature article note, there is some evidence of human fire use in Africa going back 1.6 million years, they don’t consider the evidence definitive because “the evidence for early fire use is limited and often ambiguous, typically consisting of associations between heated materials and stone tools.” They also note that there is more direct evidence but it’s quite recent:
. . . . direct evidence of fire-making by pre-Homo sapiens hominins has, until recently, been limited to a few dozen handaxes from several French Neanderthal sites, dating to around 50 ka, that exhibit use-wear traces consistent with experimental tools that were struck with pyrite to create sparks.
In this paper the authors investigate a site in Sussex, dated about 400,000 years ago, that has several lines of evidence suggesting regular use of fire, and controlled use, since there were materials like pyrite that could be used to strike sparks. Note that the paper considers this the earliest evidence for making fire, not simply using fire. The authors consider their work to provide pretty definitive evidence of fire-making and fire use in H. sapiens. (Note that we are the only species to use fire.)
Click the headline below to read the article, or you can find the pdf here.
The evidence came from an unused clay pit in the Breckland area of Suffolk, with deposits of clay and silt as well as human artifacts like hand axes. The evidence for persistent fire use at this site (the authors suggest at least two groups of humans, and comes from five observations and experiments. I’ve put them below under the letters.
a.) Red clayey silt (RCS) in the layers, silt that seems to have required prolonged heating to form. Here’s what it looks like. The unexcavated section is in the top photo, and the bottom is the partly excavated area which is an enlargement of the box in (a). I’ve put a red arrow in (a) at the RCS layer thought to reflect heating of the sediments by the presence of “hearths”: areas where cooking or other uses of fire regularly took place. The layer is more obvious in the bottom photo:
The authors say that the red layer reflecs heating or sediments containing iron:
The reddening is attributable to the formation of haematite—a mineral produced through heating of iron-rich sediments. Its distribution is homogeneous and not associated with particular microfacies or voids, indicating that it was preserved in situ.
b.) Experimental heating of the non-red sediments. The authors showed that the magnetic properties of material in the RCS differ markedly from unheated “control” samples of material taken from the lower layer (“YBCS” in second photo above). But by heating the YCBS layer extensively, it assumed some of the magnetic properties of the RCS, suggesting that the RCS involved heating of clays by fire. As they say (bolding is mine):
Three samples were taken from the RCS and two from the adjacent YBCS, which served as unheated control samples. The magnetic properties of the RCS (Supplementary Information, section 5) differ markedly from those of the unheated control samples, exhibiting elevated levels of secondary fine-grained ferrimagnetic and superparamagnetic minerals of pyrogenic origin, unlike the control samples. To assess whether these characteristics could result from heating, a series of experiments of single and multiple heating events of varying durations, was conducted. The aim was to determine whether the reddening could have arisen from one or multiple heating events, as repeated, localized burning is more typical of human than natural fire events (S.H. et al. manuscript in preparation).
The closest experimental analogue in terms of the minerology and grain size distribution, was observed after 12 or more heating events, each lasting 4 h at temperatures of 400 °C or 600 °C. Although the archaeological samples exhibit substantially lower magnetic susceptibility values, this may result from post-depositional mixing with unheated illuviated clay. Overall, the experiments indicate that the magnetic properties of the RCS result from an indeterminate number of short-duration heating events, consistent with repeated human use (Fig. 3).
Note that prolonged heating—nearly 50 hours of heating at 400-600 degrees C, was required to approximate the magnetic properties of the presumed fire-use layer. This suggests also that the heating did not reflect wildfires, but repeated, localized, and intentional burning.
c.) Infrared spectroscopy of heated control samples changed in infrared absorbtion spectra of the “control” samples, making it closer to that of the presumed hearth layer of RCS.
d.) The area contained four handaxes that showed marks of heat-shattering. Here is a picture of a handaxe with “closeup of fractured surface caused by fire.”:
Presumably this is based on experiments using recently made handaxes, with some treated by fire and then compared to unheated controls.
e.) Fragments of pyrite were found in the heated area, and pyrite is used with flint to produce fire (before that, people presumably had to get fire from lightning burns and somehow preserve it). Moreover, pyrite was not found in this locality; the nearest accessible mineral was about 15 km away, suggesting that people picked it up and brought it to the site to strike against flint (flint was also found in the area). As the authors note:
The occurrence of pyrite at Barnham warrants further consideration. Pyrite is a naturally occurring iron sulfide mineral that can be struck against flint to produce sparks to ignite tinder. Its use for this purpose is well documented in ethnographic accounts worldwide. Pyrite has been recovered from European archaeological sites dating from the late Middle Palaeolithic to the historic periods, occasionally bearing wear traces consistent with use for fire-making and, in some cases, found in association with flint striking tools.
Here are some fragments of pyrite; caption is from paper:
e.) The heated sites were located in areas amenable to prolonged fire use. This is weak evidence, but I present it nevertheless. From the authors:
Notably, all three sites occupy marginal locations, away from the main river valleys and associated with small ponds or springs. In the absence of caves, these locations probably provided safer, more sheltered environments for domestic activities. Taken together, these findings present a strong case for controlled fire use across the Breckland region during MIS 11.
The upshot: We often forget that any meat eaten by people before the advent of cooking would have to be raw, and raw meat is tough and, at least to us, somewhat unpalatable. (I do like a very rare steak, as well as steak tartare, though.) But our ancestors didn’t grind up meat, though they may have pounded it to make a kind of raw Pleistocene schnitzel. By making meat more palatable, cooking would promote eating more of it, and that itself could change the selective pressures on humans, giving them the extra nutrients they’d need if they were to evolve big brains (brains use a lot of energy!). This is one (disputed) theory for a rapid increase of human brain size that lasted between 800,000 and 200,000 years ago, though brain size was also getting bigger, albeit at a slower pace, before then. Cooking has also been suggested to have changed human social behavior (and perhaps social evolution), with pair bonding and mutual aid increasing as a way to gather, store, and protect food that needed to be cooked. And more complex social behavior could itself have promoted the evolution of larger brains to figure out how to regulate and get along in your small social group.
These theories, while suggestive, really should be downgraded to “hypotheses,” since there isn’t much evidence to support them—only correlation and speculation. However, they are interesting to contemplate, even if we never can get strong evidence for them. At the end of the paper, the authors do seem to sign onto some of these, but not strongly.
The kernel of this paper is the several lines of evidence that do, to my mind, support the idea that humans were making and using fire at least 400,000 year ago. Here’s what the authors say about the advantages, evolutionary and otherwise, of controlling fire:
The advantage of fire-making lies in its predictability, which facilitated better planning of seasonal routines, the establishment of domestic sites in preferred locations and increased structuring of the landscape through enculturation. Year-round access to fire would have provided an enhanced communal focus, potentially as a catalyst for social evolution. It would have enabled routine cooking, could have expanded the consumption of roots, tubers and meat, reduced energy required for digestion and increased protein intake. These dietary improvements may have contributed to increase in brain size, enhanced cognition and the development of more complex social relationships, as articulated in the Social Brain Hypothesis. Moreover, controlled fire use was instrumental in advancing other technologies, such as the production of glues for hafting. The widespread appearance of Levallois points from Africa to Eurasia by MIS 7 (243–191 ka), often interpreted as spear-tips, provides strong evidence of effective hafting. This interpretation is supported by use-wear evidence and the identification of heat-synthesized birch bark tar as a stone tool adhesive.
I haven’t finished the post yet, but I would like to bring attention to another species that uses and manages fire: the Black Kite and relatives from Australia. Their common name is “Firehawk”
https://www.facebook.com/bbcearth/videos/420626738710256/
Starting a fire from ordinary ambient-temperature materials, however, is unique to humans, as far as I know. (I suspect chimps and some other animals could do it if taught, but that is not fair.)
I remember reading a description of people who were studying apes sitting around campfires in the jungle. Every night the apes they were studying would slowly gather around the perimeter, just watching the fire along with them.
Imagine if they had felt comfortable enough to sit together with humans around a fire.
Well, I’ll be. Black kites actually SPREAD fires. That’s new to me, thanks.
I’ve seen a video of a Bonobo using matches to start a fire. And to me it looked like it knew exactly how to strike the match.
The Bonobo was probably just aping something he saw.
I’ll see myself out….
I enjoyed learning (from the fire-shattered rocks and glue-making) that early people were using fire for industrial purposes beyond the obvious heating of the living space and the later, not-so-obvious cooking. It shows curiosity: what happens when we stick this substance into a really hot, prolonged fire? And the effort to bring pyrite and flint together gets trade started, and conflict over immovable deposits of both, not just following game. Territory comes to have value for its own sake.
Can someone fill me in on what species were likely to be living in Britain 400,000 years ago? Isn’t this somewhat early for fully modern H. sapiens, so would it have been heidelbergensis or Neandertals or just an archaic H. sapiens?
Would that have been Cheddar Man? He invented cheese, and probably resembled Wallace in Wallace and Gromit, but with better, home-made clothing.
Neanderthals are, to me, a subspecies of H. sapiens, as they bred with them and produced fertile genes (I have some Neanderthal genes). So no, they might not have been fully modern H. sapiens, but if they contacted them, they could have either taught them about fire or learned about it.
Ten to one using flint and pyrite was a one-time discovery, but I’m just guessing.
Neanderthals, like Jerry says. Maybe heidelbergensis.
My view is that the cognitive development of hominins began when they discovered moules marinieres, around 160,000 years ago, as described at: https://www.scientificamerican.com/article/earliest-known-seafood-di/ [This appeared a little before Sci Am’s evolution into a Progressive outlet, after which articles would no doubt have traced a connection between seafood and eugenics.]
Increased brain size, and a steady improvement in implements began earlier, with Homo erectus.
Yes indeed. A very interesting piece of the puzzle seems to have been the result of a single point mutation that occurred about 1 million years after our lineage split from the chimpanzee lineage.
This article, Scientists Make Bigger Monkey Brains With The Help Of A Human Gene, relates research of the human specific ARHGAP11B protein-encoding gene. Here is a link (also in the article) to the paper about verifying the function of this gene, Human-specific ARHGAP11B increases size and folding of primate neocortex in the fetal marmoset.
To summarize, researchers identified a specific gene unique to humans that causes an increase in size and folding of the neocortex. They verified the function of this specific gene by inserting it into fetal marmosets which, according to the paper, . . .
And, perfect plot point for a science fiction story about aliens uplifting early humans, this human specific gene appears to be the result of an SPM, as described in this paper, A single splice site mutation in human-specific ARHGAP11B causes basal progenitor amplification. From the article describing the research (1st link above) . . .
The timing of this mutation is also very interesting. From the paper . . .
Didn’t H. naledi use fire to light their way through the Rising Star cave? Maybe I’m assuming they did.
Just looked it up and according to Lee Berger there is a lot of evidence that they did. Interesting.
To my mind, none of the pieces of evidence is definitive on its own, but taken together, the consilience of evidence is pretty convincing. Plate tectonics gained rapid and wide acceptance for the same reason—consilience of evidence—as did evolution. Darwin’s great book brought together anatomy, geographical distribution, embryology, use and disuse, evidence from domestication, etc., to convince the world of evolution. This is a smaller discovery—it’s just the discovery of fire, no big deal*—but the contours of the evidence are the same.
By the way, cats use fire. Our cats always snuggled up next to the fireplace when it contained a fire.
Interesting study showing humanity on the pathway toward McDonald’s hamburgers and Kentucky Fried Chicken.
*kidding
Here’s the Friday : Hili Dialogue in which (if I follow) PCC(E) first reported this Nature paper :
https://whyevolutionistrue.com/2025/12/12/friday-hili-dialogue-554/
Such an expansive, stimulating topic and commentary.
A thought :
A campfire (say) is more readily started by use of ever-finer materials such as (say) dryer lint – which is composed of cotton (etc.) from clothing. This makes a very big difference for getting the fire started. I imagine lint is at least as old as clothing – even older, as mice can build nests with such materials! But I did not look deeper into this practical technique – among the numerous ideas that occur upon reading the excellent treatment here. … and … gee, I’d have to check again now if using one’s palms to spin a stick pointed into a piece of wood is a myth, practical, or something else.
The geologic factor is intriguing to me, e.g. which areas of land might not have flint or pyrite, etc., or which similar minerals etc. that might produce sparks. I know pure magnesium shavings work great as starters – even in wet conditions – but AFAIK magnesium must be purified, and isn’t “Naturally occurring” … another idea I have to look up!
It’s pretty well accepted that Homo erectus used fire, with evidence at encampments over a million years old with signs of fire use. Of course making fire is another thing entirely. The bit about using cooking to increase nutrients and thereby allow selection for larger brains is controversial, but I don’t know of serious problems with the idea.
Thanks for the descriptions. I saw the article and news articles when the paper came out. It immediately made me think of Richard Wrangham’s book Catching Fire, which is about the use of fire and how cooking accelerated humans’ development. An interesting book: https://www.amazon.com/Catching-Fire-Cooking-Made-Human/dp/0465013627
I wonder to what extent fire was preserved (as in the movie Quest For Fire) after it was acquired from a natural source? Maybe there is a long history of regular fire use preceding the invention of fire making – because making wasn’t critical to fire having.
I ate all my meat raw for many years, and I greatly prefer it to cooked meat. I don’t really understand the importance of cooking meat, except to kill parasites.
Just out of curiosity, how did you get started doing that, since it’s quite counter-cultural? I’ll grant that I like steaks rare, I like steak tartare, I like tuna just seared on the outside and raw inside. So I can see where you are coming from but those are all things that are easy to learn. Eating all your meat raw is pushing the envelope. Interesting. (and interested.)
My grandfather (who lived to be 100) believed that cooking things altered them in ways that were not healthy, and he convinced me (I was probably 15). So I cooked nothing, and only ate unpasteurized milk and cheese, raw veggies, etc. And when I went to college I didn’t have a stove, or money to buy cooked meals, so it became a practical advantage. After a while I gave it up, but I still dislike most cooked beef, and avoid it if possible. Cooked chicken doesn’t bother me now, and I never ate it raw either.
There’s a wonderful film about two early hominid groups titled “Quest for Fire” that shows a fictional transition from incidentally finding fire to actively making and controlling fire (1981 dir: Jean-Jacques Annaud). It demonstrates some of the points the article makes related to increased social complexity and skill development. The filmmakers developed a “language” the actors used to show the difficulty early disparate groups might’ve had. Plus it’s just a sweet story.