For decades, the history of dog domestication has been one of science’s most stubborn evolutionary arguments. Were dogs domesticated once or more than once? Did they emerge in Europe, Central Asia, the Near East, Siberia, or somewhere still unsampled? Were the earliest “dogs” truly dogs, or wolves that happened to live unusually close to people? New paired studies in Nature do not settle every part of that argument. But they do something nearly as important: they move the genetic record of dogs thousands of years deeper into the Ice Age and show that by roughly 15,800 to 14,200 years ago, dogs were already real, widespread, and entangled with human societies across western Eurasia.[1][2]

That matters because the field has long been haunted by a mismatch between bones and genes. Archaeologists have pointed to dog-like canids in Upper Palaeolithic sites for years, sometimes tens of thousands of years old. Geneticists, meanwhile, kept finding that many of the most tantalizing candidates were not dogs at all, but extinct wolves. The result was a scientific whiplash: bold claims for very early dogs, followed by genomic demotions. The new papers narrow that gap by showing that some of the old archaeological hunches were right after all — just not always in the places researchers first thought.[1][2]

One of the new papers, led by William Marsh and colleagues, identifies dogs from Pınarbaşı in Türkiye dating to about 15,800 years ago and from Gough’s Cave in Britain dating to about 14,300 years ago, along with Mesolithic dogs from Serbia. The authors conclude that a “genetically homogeneous dog population” was already spread across Europe and Anatolia during the Late Upper Palaeolithic.[3] That phrase is deceptively compact. It means that long before farming, before herded cattle and sheep remade Eurasia, people at the western and eastern edges of this region were living with dogs that were recognizably related to one another. This was not a tiny, isolated experiment in taming wolves. It was a population.[2]

---

---

The second paper, by Anders Bergström and colleagues, pushes hard on Europe specifically. Using genome-wide capture on 216 canid remains, the team confirms a 14,200-year-old dog from Kesslerloch in Switzerland and shows that it already shared ancestry with later dogs across the world. That finding undermines a once-popular idea that Europe’s early dogs represented a wholly separate domestication that later vanished.[4] Instead, the Kesslerloch dog looks connected to the broader domestic dog lineage. The paper also shows that by 14,200 years ago, dog populations were already structured: the Kesslerloch dog was more closely related to later European dogs than to Asian dogs. In other words, dog diversification was already underway deep in the Late Glacial period.[4]

A useful line from Nature’s accompanying News & Views captures the significance neatly: these studies are “getting closer in time to the origin of dogs than ever before.”[5] That sounds incremental. It is not. Ancient DNA has repeatedly shown that the earliest stages of domestication are where assumptions go to die. The ability to distinguish true dogs from wolves at these dates depends not on one conspicuous skull or one burial, but on genome-wide evidence.

That challenge has shaped the entire modern history of canine-origins research. In 1997, an influential mitochondrial DNA study suggested dogs might have much deeper and more ancient roots than anyone expected, even implying an origin more than 100,000 years ago.[6] Later work complicated that picture. In 2015, a major global village-dog study argued for Central Asia as the likeliest origin region, emphasizing the importance of free-ranging dogs rather than modern breeds.[7] In 2016, another high-profile study proposed a dual origin, suggesting that dogs may have been domesticated independently in eastern and western Eurasia, with later replacement events blurring the picture.[8] By 2020, a large ancient-DNA study showed that by 11,000 years ago at least five major dog ancestry lineages already existed, proving that dogs had a deep pre-agricultural history even if their exact birthplace remained elusive.[9] Then, in 2022, a Nature study of ancient wolves argued that dogs derived ancestry from at least two wolf populations, further reframing the debate away from a single simple founding event.[10]

The new papers do not erase that complicated history. They sharpen it. Marsh and colleagues suggest that dogs spread across western Eurasia alongside interactions among genetically and culturally distinct hunter-gatherer groups — Magdalenian, Epigravettian, and Anatolian hunter-gatherers.[2] In effect, dogs may have moved through human social networks faster, or at least differently, than people themselves. Bergström and colleagues arrive at a complementary conclusion from another angle: Europe’s early dogs were not genetically sealed off from the wider dog story, but neither were they blank slates later overwritten by incoming farmers.[4]

That last point may be one of the most important. In human ancient-DNA studies, the Neolithic transition in Europe is often told as a story of massive migration from Southwest Asia. The dog story now looks more mixed. Bergström’s team finds a Neolithic influx of Southwest Asian ancestry into European dogs, but smaller than the corresponding human turnover.[4] Local Mesolithic dogs, in other words, appear to have contributed substantially to the dogs of Neolithic Europe and probably to modern European dogs as well. Marsh’s paper reaches a related conclusion from a different dataset, arguing that ancestry components established by at least 10,900 years ago persisted into later European dogs and even modern breeds.[2]

The implication is striking: Europe’s dogs may have been more continuous than Europe’s people.

The papers are also important because they restore archaeology to the conversation without letting morphology run wild. At Gough’s Cave, dog remains received postmortem treatment similar to that seen in human remains. At Pınarbaşı, dogs were buried in the same area as humans, and isotope work suggests they may have been provisioned, directly or indirectly, by people.[2] These are not just evolutionary datapoints. They are signs of relationship. The earliest confirmed dogs were not merely tolerated scavengers haunting the edge of campfires. They were already socially and symbolically entangled with humans.

Still, caution remains essential. Neither paper identifies the exact place where wolves first became dogs. Neither can say with confidence whether domestication happened once in a narrow sense or through a more diffuse process involving multiple wolf populations, repeated contact, and later mixing.[4][10] The authors themselves are careful on that point. Bergström and colleagues explicitly note that their findings do not resolve where, or how many times, dogs were domesticated.[4] Marsh and colleagues likewise frame their conclusions around spread, exchange, and ancestry rather than a final answer to origins.[2]

That restraint is part of what makes the studies convincing. In a field with a long history of grand claims, these papers advance by being precise. They show that by the Late Upper Palaeolithic, dogs were already present from Anatolia to Britain. They show that some supposed early dogs were indeed wolves, vindicating genomic skepticism. They show that some ancient European dogs belonged firmly within the lineage of domestic dogs living today. And they show that canine prehistory, like human prehistory, was shaped by movement, mixture, and survival.

The oldest chapters of dog evolution are still incomplete. But they are no longer quite so dark. The story of how wolves became dogs remains unresolved. The story of what dogs were by the end of the Ice Age is now much clearer: they were already traveling with us, living with us, and beginning to diversify into the lineages that would eventually circle the globe.[2][4][5]

Endnotes

[1] William A. Marsh et al., Dogs were widely distributed across western Eurasia during the Palaeolithic (Nature, 2026), uploaded study.

[2] Anders Bergström et al., Genomic history of early dogs in Europe (Nature, 2026), uploaded study.

[3] Marsh et al. describe a “genetically homogeneous dog population” already spread across Europe and Anatolia in the Late Upper Palaeolithic. Uploaded study.

[4] Bergström et al. conclude that the 14,200-year-old Kesslerloch dog shared ancestry with later worldwide dogs and that Neolithic Southwest Asian ancestry entered Europe’s dogs at a smaller scale than it did Europe’s humans. Uploaded study.

[5] Lauren M. Hennelly and Mikkel-Holger S. Sinding, Dogs have deep genetic roots in ice-age Europe, Nature News & Views (2026). The piece says the new studies are “getting closer in time to the origin of dogs than ever before.” (Nature)

[6] C. Vilà et al., Multiple and ancient origins of the domestic dog (Science, 1997), summarized in PubMed. (PubMed)

[7] L. M. Shannon et al., Genetic structure in village dogs reveals a Central Asian domestication origin (PNAS, 2015). (PNAS)

[8] L. A. F. Frantz et al., Genomic and archaeological evidence suggest a dual origin of domestic dogs (Science, 2016), summarized in PubMed. (PubMed)

[9] Anders Bergström et al., Origins and genetic legacy of prehistoric dogs (Science, 2020). By 11,000 years ago, at least five major dog ancestry lineages had already diversified. (PubMed)

[10] Anders Bergström et al., Grey wolf genomic history reveals a dual ancestry of dogs (Nature, 2022). Dogs derive ancestry from at least two wolf populations. (Nature)

[11] L. R. Botigué et al., Ancient European dog genomes reveal continuity since the Early Neolithic (Nature Communications, 2017). (Nature)

[12] Tatiana R. Feuerborn et al., Modern Siberian dog ancestry was shaped by several thousand years of Eurasian-wide trade and human dispersal (PNAS, 2021). (PNAS)

---

---