91av

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

Among the many other human species that once inhabited Earth, the Neanderthals are the most famous. They lived until relatively recently and in many ways, they were like us.

Just in the past few months, we’ve seen tentative evidence of them treating wounds using tar with antibiotic properties made from birch bark. An ancient yellow crayon, made of ochre, gave us a hint of their artistic practices. A well-preserved skull suggested that their noses weren’t adapted for cold climates, as many had thought. Elephant bones from Germany show signs of having been butchered by Neanderthals. There is even suggestive evidence of Neanderthals crossing wide expanses of water.

All in all, we have a rich picture of Neanderthals’ lives, and that picture gets fleshed out more every year. But what we don’t have is a good account of their origins. In fact, how they evolved and who their ancestors were is one of the biggest mysteries in human evolution.

That’s not to say there aren’t plausible human family trees out there, but each of them throws up questions we can’t answer with the evidence we have, whether it’s from bones or from genes.

There are also some decidedly non-obvious explanations for how Neanderthals came to be. One, put forward as a tentative hypothesis, is that they are the product of an ancient period of interbreeding that occurred hundreds of thousands of years ago. If it’s true – and I have to emphasise that it is only a suggestion – it would require the biggest rewrite to that family tree to date, which could change how we see the role of our own species in human evolution.

A tangled history

Let’s lay out what we actually know. Neanderthals lived in Europe and Asia for several hundred thousand years, but exactly how long is a tad uncertain.

The oldest fossils are from a cave called Sima de los Huesos in northern Spain and are around 430,000 years old. They don’t have all the features of later Neanderthals, so they have sometimes been interpreted as ancestors or close relatives of Neanderthals. However, in 2016 geneticists sequenced their DNA: it proved to be very Neanderthal-like.

Of course, it’s unlikely that those remains are the very first Neanderthals. Genetics suggests that the group was around for tens of thousands, or perhaps even hundreds of thousands, of years earlier.

The Neanderthals lived through all kinds of upheavals, including multiple glacial periods, until they finally disappeared around 40,000 years ago. Genetics and archaeology reveal a complicated story. Towards the end, a cold climatic episode seems to have forced them into a refuge in southern France, wiping out much of their genetic diversity. The last reliably dated Neanderthals lived in southern Europe, mostly in what is now Spain.

To trace the Neanderthals’ origins, researchers have tried to link them to other known hominins. One key group is the Denisovans, who lived in east Asia around the same time as the Neanderthals. Genetics indicates the two groups were closely related, having evolved from an unknown common ancestor, sometimes dubbed the Neandersovans.

Genetics also shows Neanderthal DNA is more similar to Denisovan DNA than it is to modern human DNA. The implication is that there was an unknown shared ancestor, called Ancestor X. At some point, the story goes, that mystery population split into the ancestors of modern humans and the first Neandersovans, some of whom gave rise to the Neanderthals.

I should emphasise that this story is largely based on genetic evidence. If we accept it, we should expect to find two additional kinds of fossils: Neandersovans and Ancestor X. The problem is, we haven’t found either.

What does that mean? It could be that we just haven’t found, or correctly identified, the key remains. Or it could be that the genetic evidence is in some way misleading us, and that the whole story is wrong in some way.

To see what I mean, let’s look at the fossils we do have, and see if any of them could be the ancestors of Neanderthals.

Candidate ancestors

The earliest known hominins, between 7 million and 2 million years ago, are known only from Africa. It seems our ancestors remained on that continent for several million years.

The hominins that lived outside Africa, and therefore seem like plausible ancestors for the Neanderthals, are fairly limited. There are basically three candidates.

Let’s start with Homo erectus, the first hominin known to have ventured into Europe and Asia. They were present in east Africa by 2 million years ago, and by 1.8 million years ago some of them were living in what’s now Georgia, where Europe meets Asia. Some of them wandered east from there, and their descendants ended up all the way over in Java, Indonesia.

Based on the timing, it may seem obvious that H. erectus were the ancestors of Neanderthals. However, there’s an issue: despite decades of searching, nobody has found H. erectus in Europe. The closest anyone has come is some face bones, part of the cheek and upper jaw, found in a cave in northern Spain and described in 2025. The remains are between 1.1 million and 1.4 million years old, which would fit the timeline, but they are so fragmentary that they can’t be confidently identified. The researchers have called them Homo aff. erectus, which is taxonomy-speak for “we think they might be this but we’re not remotely sure”.

So, H. erectus might be distant ancestors of the Neanderthals, but at present we can’t draw a firm link between the two. There’s a long span of time between the one possible European H. erectus and the oldest known Neanderthals.

Next, there’s Homo antecessor, which lived in northern Spain. That’s the right general region, but what about the time? There’s been some back and forth about the dating of the remains, with the current best estimate being . In 2020, researchers extracted proteins from an H. antecessor tooth and found that it was closely related to Ancestor X. The timeframe therefore looks reasonable, and there’s molecular evidence.

The trouble is that H. antecessor are known from only one site: in Spain. While the site held remains from , we don’t know how widespread or long-lived the species was. Mark it as promising but unproven.

The final candidate is Homo heidelbergensis. For a while, they were the leading candidate for Ancestor X, because their skulls looked quite similar to those of Neanderthals and modern humans, and they were known from Europe and Africa between 300,000 and 500,000 years ago.

However, they now look much less plausible. First, most of the fossils originally thought to be H. heidelbergensis have been reassigned to other species upon further analysis, so nowadays there are only a few left, all from Europe. This means our knowledge of the species is patchy. And crucially, the fossils left in the group seem to be too recent to be Ancestor X.

That gives us three candidates, all with pluses and minuses. A key problem here is that we don’t have any preserved DNA from H. erectus, H. heidelbergensis or H. antecessor: the fossils are either very old or found in hot and wet tropical places where any genetic material they held would have degraded by now. That means we can’t be confident of the relationships between them, or their relationships with the Neanderthals.

We have the fossils, of course, but many of them are partial or damaged. Based on their ages, it would be tempting to suggest that H. erectus gave rise to H. antecessor and H. heidelbergensis in Europe, and that one of them – maybe H. antecessor – is Ancestor X. But we just don’t know.

Anyway, there’s a glaring issue: this story doesn’t make a lot of sense.

Who moved where?

The problem, as I see it, is that the genetics and the archaeology seem to be telling different stories.

Genetics tells us that the ancestors of modern humans split from the Neandersovans (the ancestors of Neanderthals and Denisovans) . That gives us a timeframe for Ancestor X.

It also points at where Ancestor X should have lived. The earliest Neanderthals were in Europe, all the Denisovans were in east Asia and the earliest modern humans were in Africa. The simplest possible story is that there was a source population that split three ways, all heading off in different directions to form these populations.

This would point to somewhere in western Asia, in the general vicinity of the eastern Mediterranean, Levant, Middle East, Caucasus or Ukraine. I am being deliberately, obstreperously vague about exactly where, because this is a crude line of reasoning and it would be silly to get specific. Just imagine a wide, fuzzy circle over that whole area.

Some recent archaeological finds might actually help to make sense of this. A skull from Yunxian in China seems to be an early Denisovan, and it’s 940,000 to 1.1 million years old. This points to an earlier Ancestor X, in which case, H. erectus might come back into the frame. Likewise, there are some hominin remains from a cave called Grotte à Hominidés in Morocco, dating to about 773,000 years ago, which seem to have the right features for Ancestor X. Morocco isn’t western Asia, but it is at least northern Africa, so not too ridiculously distant from our theoretical Ancestor X habitat.

But there’s one last twist, which is the absolute weirdness of Neanderthal genetics. Famously, Neanderthals and modern humans interbred. The main interbreeding seems to have been between 50,000 and 43,000 years ago, and may have occurred throughout Europe and west Asia. But there were also earlier bouts of interbreeding, which are harder to pin down.

This means people with non-African ancestry today all carry some Neanderthal DNA. When we look at the genomes of ancient Neanderthals, we can also see traces of modern human DNA from interbreeding. But weird things happened. In modern humans, Neanderthal DNA has been virtually expunged from our X chromosomes. Meanwhile, Neanderthals lost their original Y chromosomes, replacing them with . The same thing happened to their mitochondrial DNA, which is inherited from only the mother.

Hence a striking new hypothesis of Neanderthal origins, set out by geneticist David Reich at Harvard University. Reich explained his idea in released on 13 March.

Let me say up front: treat this idea as tentative. The paper’s title starts with the word “hypothesis”. Reich declined to be interviewed, saying that he had only put the paper out to garner comments from colleagues, and he acknowledged that those comments might contain arguments that knock down his idea.

Suitably braced? Off we go.

Reich suggests that the origin of Neanderthals lies in an early migration out of Africa by modern humans. The oldest examples of our species are about 300,000 years old, from Morocco. Reich proposes that some of them wandered into Europe and interbred with the as yet unidentified local hominins, sometime between 400,000 and 250,000 years ago.

The resulting hybrids lost most of their modern human DNA, but they did keep their modern human Y chromosomes and mitochondrial DNA. These hybrids were the Neanderthals. This is a wholesale reinterpretation of the genetic evidence, which would mean those earlier episodes of interbreeding that messed with the Neanderthal Y chromosome and mitochondrial DNA weren’t a minor detail: they were the origin of Neanderthals.

This scenario neatly explains a curious archaeological finding. There are distinctive stone artefacts called Levallois tools, which were used in Africa from at least 400,000 years ago (presumably by modern humans), but also in Europe and the Middle East (presumably by Neanderthals) between 480,000 and 300,000 years ago. We might imagine that modern humans and Neanderthals independently invented Levallois tools, but it’s a little neater to suggest that some modern humans took them with them when they left Africa.

Maybe Reich’s idea will fall down on closer examination. I find myself wondering about the timing: given how old Ancestor X seems to have been, the crucial period of interbreeding seems a bit recent.

Nevertheless, it seems like a brilliant way to reframe the question. The last couple of decades have shown us that our species is a hybrid one, formed from multiple populations in Africa plus interbreeding with Neanderthals and Denisovans. Once you accept that, the next logical step is to imagine that other hominin groups are also the product of interbreeding.

Also, it brings the Neanderthals even closer to us. In this scenario, they are literally descended from the earliest members of our species.