When did humans start making art and were Neanderthals artists too?

This is an extract from Our Human Story, our newsletter about the revolution in archaeology. Sign up to receive it in your inbox for free every month.

When we think about art, we tend to think about the relatively recent past. The famous cave art in Lascaux, France, is often said to be around 17,000 years old. A second French cave, Chauvet, has similarly spectacular paintings that have been dated to 30,000 or so years ago – but that date is controversial, with some archaeologists saying the paintings are just too good to be that old. When you consider that our species, Homo sapiens, is probably more than 300,000 years old, and that our genus Homo has been around for more than 2 million years, those few tens of thousands of years are a vanishingly short span of time. Why did humans start painting so late in the day, and why didn’t other hominins like Neanderthals do it?

Well, it’s possible that we actually did make art much earlier than that, and that Neanderthals and other groups such as Denisovans did the same. There are two barriers to demonstrating this: prejudice against the idea that other hominins could express themselves symbolically, and issues with the physical evidence.

In early June, I spent a few days with holidaymakers on a New Scientist Discovery Tour focused on prehistoric rock art in the caves of northern Spain. The centrepiece of the tour was Altamira, the first place where prehistoric European cave art was found – back in 1879, long before Lascaux and Chauvet were discovered.

How old is Europe’s cave art?

As part of my preparation, I tried to find out as much as I could about each of the caves on the itinerary. But I kept having the same problem: figuring out how old the artworks were. In Altamira, that’s because the paintings spanned a long period: one is 36,000 years old, another 22,000, while the artefacts in the cave are as little as 14,000 years old.

But in other cases, the dates for each artwork varied. Take Hornos de la Peña cave, for instance. It has a great many engravings of animals, drawn with striking anatomical accuracy. Spain’s official tourism website says they were created in two phases, one dating back at least 18,000 years and another close to 15,000 years ago. However, a 2014 study lists the dates that have been obtained for the artworks, and they range from around 10,000 years ago to more than 30,000 years ago. Part of the problem is that the cave has been interfered with: it was used as a shelter in the Spanish civil war, and was further altered to enable tourists to visit.

Similarly, El Pindal cave has lots of pictures of horses and bison, with a fish and a mammoth thrown in. Once again, there is no definitive date. The same Spanish tourism website has them at between 13,000 and 18,000 years old, while a book chapter from 2007 identifies El Pindal as one of several caves in the region where the dates of the art are problematic.

I spoke to Alistair Pike at the University of Southampton in the UK, who has studied the age of cave art, to clarify the dates of the paintings. He told me that only “a tiny, tiny proportion” of cave art has been reliably dated.

Some of the reasons for this are good. Until recently, the main method of finding the age of a piece of cave art was radiocarbon dating. This is inherently destructive – you have to scrape off a sample – and understandably the custodians of the caves are slow to give permission. Furthermore, carbon dating only works if there is organic material like charcoal in the art; for engravings, and anything painted solely with minerals, it’s useless.

Unfortunately, there is also a bad reason not to perform carbon dating. “People had assumed that they could tell the age of cave paintings by the style in which it was depicted,” says Pike. Ever since the first prehistoric art was found in the late 1800s, there has been a sense that art should evolve linearly: the oldest pieces should be extremely simple and abstract, with later ones becoming more technically skilled and creative. Hence the scepticism over Chauvet, despite the paintings having been carbon-dated.

This line of thinking was comprehensively exposed in a 2011 study by April Nowell and Genevieve von Petzinger, then both at the University of Victoria in British Columbia, Canada. By asking experts on each cave to explain why they believed the artworks to be certain ages, Nowell and von Petzinger discovered an enormous loop of circular logic. Artworks in separate caves were assumed to be the same age because they looked similar, so cave A’s art was so many thousand years old because it looked like cave B’s art – except that the experts on cave B were basing their own age estimates on cave A.

“It all went round in a really big circle,” says Pike. “It’s one of the most brilliant pieces of work I’ve ever seen.”

Were Neanderthals artists too?

If a lot of the given ages are spurious, our ideas about who made the art are also spurious.

A succession of hominins have lived in western Europe and might theoretically have made the region’s cave art. Modern humans are the most recent inhabitants, having permanently settled in the region around 45,000 years ago after emerging from Africa. Before that, Europe and western Asia were inhabited by Neanderthals for hundreds of thousands of years. And before that, other hominins like Homo antecessor were around.

If all the cave art in western Europe is less than 30,000 years old, it could only have been made by our species. But in the cases where researchers like Pike have managed to get reliable dates, that hasn’t always proved true.

Back in 2012, Pike’s team showed that a red dot on the wall of El Castillo cave in northern Spain was at least 40,800 years old. That was old enough to be borderline: Neanderthals were still around, so they could have made the dot.

The team did this using uranium-thorium dating. This doesn’t find the age of the art itself, but the age of a thin mineral overlaying it. These layers form when water trickles over the cave wall, depositing minerals that gradually build up. The dating technique tells us when the mineral layer formed, giving a minimum age for the art.

In a 2018 follow-up, Pike’s team dated the art in three more Spanish caves. The first was La Pasiega, which is in the same hill as El Castillo. A symbol made of red lines turned out to be at least 64,800 years old. The second was Maltravieso in western Spain, where a hand stencil proved to be at least 66,700 years old, making it the oldest cave art known in the world. Finally, some of the red paint on stalagmites in Ardales cave on Spain’s southern coast turned out to be at least 65,500 years old.

When I mentioned these dates to the holidaymakers in northern Spain, there were audible gasps. They were a knowledgeable and engaged audience, but these results and their transformational significance hadn’t sunk in. If the art is really this old, the most sensible explanation is that Neanderthals made it.

In line with this, Pike points to other sites with evidence of symbolic behaviour by Neanderthals, going way back into prehistory, but which were previously dismissed. In Bruniquel cave in southern France, there is a stone circle made from broken stalagmites that is 175,000 years old. Pigments on shells in Aviones cave in southern Spain are 115,000 years old. There is evidence of Neanderthals collecting ochre, a red pigment often used in cave paintings , at Maastricht-Belvédère in the Netherlands at least 200,000 years ago.

Marshall’s First Law

Pike’s team has struggled to do additional dating in the past few years, partly because of the covid-19 pandemic and partly because “archaeologists who don’t want Neanderthals to have painted have basically banned us from taking samples”. However, he hopes that other groups will have more luck, eventually building up a rigorous timeline of cave art. He suspects art-making may stretch back to the unknown common ancestor we shared with Neanderthals, hundreds of thousands of years ago.

Art may be another example of what in a previous newsletter I self-aggrandisingly called Marshall’s First Law: Never be surprised when something turns out to be older than you thought. A recent study used artificial intelligence to identify hidden evidence of controlled fires at a site in Israel from 1 million years ago – hundreds of thousands of years before evidence for fire use becomes widespread. I will lay odds that painting and symbolic expression will also turn out to be much older, once we start properly looking.

The fundamental difficulty we face is that art is fragile. The stuff in caves has survived because these were very stable environments – especially if the entrance collapsed, keeping people out for millennia. But maybe people were expressing themselves all over the place, as can be seen in the Côa valley in Portugal where there are thousands of open-air engravings from at least 10,000 years ago. It’s just that most outdoor art has long since eroded away. “The landscape was full of symbols,” says Pike, “and very few of them survived.”