Ancient DNA | Neolithic Europe | The megaliths

Who built the megaliths?

Thirty-five thousand standing stones and stone tombs line the Atlantic coast of Europe, from Portugal to Sweden. For a century, no one knew whether they were built by one people or many. Radiocarbon dating now points to a single origin in Brittany, and ancient DNA reveals who the builders were and how they lived.

about 4500 BCwhen the first megaliths were raised in Brittany, nearly two thousand years before the first Egyptian pyramid
By seathe idea spread along the Atlantic and Mediterranean coasts in three waves, not overland
Patrilineal kinthe people buried in the tombs were related through the male line, over ten generations

Hero image: AI generated illustration of Neolithic people raising a great stone at a megalithic tomb on the Atlantic coast of Europe, about 3500 BC, with a passage grave and standing stones under a dramatic sky. Clothing, tools and the scene are approximations, and it is not a reconstruction of any documented monument or of the individuals whose genomes are discussed here.

They are among the oldest monuments in Europe. The great dolmens, passage tombs and standing stones of Atlantic Europe range from the alignments of Carnac to the chambers of Newgrange and the circle of Stonehenge, and the oldest of them were already nearly two thousand years old when the first Egyptian pyramid was built. There are about 35,000 of them still standing, strung along the western and southern coasts of the continent. And for most of the last century, two basic questions about them went unanswered: did one people invent them, or many, and who were those people?

Two strands of science have now answered both. Radiocarbon dating has pinned down where and when the tradition began, and ancient DNA has revealed who the builders were and how their societies worked.

Seven thousand years of standing stonesKey dates in the European megalithic traditionabout 5000 BCThe first monumental gravesLong earthen mounds, the Passy tombs, appear in the Paris basin.The idea of the monumental grave is born.about 4500 BCThe first megaliths in BrittanyThe first stone chambers and standing stones rise around Carnac,the earliest in the radiocarbon record.4300 to 3500 BCThe great spreadBy sea, the megalithic idea reaches Iberia, Britain, Ireland andthe Mediterranean within a few centuries.about 3200 BCNewgrangeThe great passage tomb of the Boyne valley in Ireland is built,aligned to the midwinter sunrise.3000 to 2500 BCOrkney and StonehengeThe monuments reach their peak in the north; the great stonesof Stonehenge are raised around 2500 BC.2019The maritime mapSchulz Paulsson argues for a single origin in northwest France,spread by sea in three waves.Dates are approximate, from the radiocarbon chronologies in the references.
Key dates in the European megalithic tradition, from the first monumental graves to the maritime-diffusion study. Dates are approximate, from the radiocarbon chronologies cited.

A single origin, spread by sea

The old debate had two sides. The nineteenth-century view held that megaliths came from a single source, somewhere in the Near East, and spread outward. The late twentieth-century view, once radiocarbon dating arrived, swung the other way: the monuments had sprung up independently in several regions, a case of many peoples hitting on the same idea.

In 2019, Bettina Schulz Paulsson made the strongest case yet. She assembled and modelled 2,410 radiocarbon dates from megalithic, earlier monumental and contemporary sites across Europe. The earliest megaliths, she found, appear in northwest France, above all around Carnac in Brittany, around 4500 BC, preceded in the region by monumental graves without large stones. Within two or three centuries the idea appears in southern France, the Mediterranean and on the Atlantic coast of Iberia, and in later waves in Britain, Ireland and Scandinavia: three successive phases in all.

That was the surprise. In her reading, the megaliths did not spread overland with a slow expansion of farmers; they travelled by boat. To spread a complex architectural and religious idea that quickly along thousands of kilometres of coast, she argues, these societies must have had seafaring skills far beyond what anyone had credited to Stone Age Europe, pushing advanced navigation back by about two thousand years. The Stonehenge archaeologist Michael Parker Pearson called it a demonstration that Brittany was the origin of the European megalithic phenomenon. Not every archaeologist accepts a single origin, and some still see room for several centres, but her chronology has become the reference point of the debate.

The builders were farmers

If the idea came from Brittany, did the people? Here ancient DNA takes over. We modelled Neolithic genomes from five regions in Global25, as a mixture of the two great ancestral components of Neolithic Europe: the Anatolian farmers who had brought agriculture into the continent, and the western hunter-gatherers who were there before them. One gap should be stated plainly: there are no genomes from the Breton megaliths themselves in Global25. Our French sample comes mostly from the Neolithic cemetery of Gurgy in Burgundy, a broadly contemporary farming population of northern France.

The builders were farmers, everywhereNeolithic people by region, modelled as Anatolian farmer plus western hunter-gatherer ancestry (our analysis)0%25%50%75%100%fitFrance (Paris basin)99 genomes841643Iberia11 genomes762446Britain79 genomes792141Ireland29 genomes792139Scandinavia3 genomes732729Anatolian (early farmer) ancestryWestern hunter-gatherer ancestryMostly early-farmer ancestry everywhere. Fits near 40 show that a two-source model is a simplification. No Breton megalith genomes are available.
Neolithic people from five regions modelled as Anatolian early-farmer plus western hunter-gatherer ancestry in Global25 (our analysis). The French sample comes mostly from the Gurgy cemetery in Burgundy; no Breton megalith genomes are available. Farming ancestry dominates everywhere, with a hunter-gatherer minority rising toward the north and west.

The answer is consistent from Portugal to Sweden. Wherever we have their genomes, the megalith builders were, overwhelmingly, descendants of the first farmers. They carry a majority of Anatolian farming ancestry, with a minority of hunter-gatherer ancestry that rises toward the north and west, from about 16% in France to 27% on Gotland. The fit distances, around 40, show that a two-source model is a simplification, but the pattern is clear. The monuments were not built by the old hunting peoples of Europe, nor by a single migrating tribe, but by the settled farming societies of the Atlantic facade, who shared a common Neolithic origin and, on the radiocarbon evidence, a common idea carried between them by sea.

The living peoples who most resemble them confirm it, with a caveat: nobody alive is genuinely close. The nearest present-day population is the Sardinians, at 53 units, followed by the Basques, then the peoples of southern Iberia. The north of Europe is far away, and so, strikingly, are today's Bretons (150): the Bronze Age migrations from the steppe, which reached Britain with the Beaker people around 2500 BC, transformed the population of the very regions where the megaliths stand. We told that part of the story in our article on Stonehenge and the Beaker people.

The last of the first farmersGlobal25 distance from the megalith builders (89 genomes) to present-day Europeans (our analysis)04590135180Sardinians53Basques (Soule)88Andalusians104Galicians108Portuguese108French, Occitanie120Tuscans126Bretons150Germans154English158Irish166Swedes177Distances x 1000. Nobody alive is close; Sardinians come nearest. Blue: the north, transformed by Bronze Age steppe migrations.
Global25 distance from the average of 89 genomes from megalithic tombs in Britain, Ireland, Portugal and Sweden to present-day Europeans (our analysis). Sardinians come nearest; the north of Europe, Brittany included, was reshaped by Bronze Age steppe migrations.

Tombs for a kindred

Ancient DNA can do more than trace ancestry; it can reconstruct the families buried inside the tombs. In 2019, Federico Sanchez-Quinto, Mattias Jakobsson and colleagues sequenced 24 people from five megalithic tombs in Ireland, Scotland and the Baltic island of Gotland, whose remains date between about 3800 and 2600 BC. What they found turned the tombs from monuments into family histories.

The people buried together were close relatives. According to the team, the kin relations can be traced across more than ten generations, and in one remarkable case a likely parent and child were buried in two different tombs, at Carrowmore and Primrose Grange, two kilometres apart. These were not random samples of a community. They were the graves of particular kindreds.

And the kindreds were reckoned through the male line. The tombs of the British Isles held more men than women, the same Y-chromosome lineages recurred over time, and male genetic diversity was sharply reduced, the signature of a society where descent passed from father to son. Women were not excluded: three of the six kin relationships identified involved women. But the megaliths were, above all, the monuments of patrilineal families who used them for centuries. The Ansarve tomb on Gotland was in use for over 700 years by a group genetically distinct from the hunter-gatherers who dominated the island at the time, holding onto both their farming ancestry and their burial tradition.

A later study made the pattern even clearer. At Hazleton North, a long cairn in the Cotswolds built around 3700 BC, Chris Fowler, Inigo Olalde and colleagues reconstructed a family tree of five generations: most of the people buried there descended from one man who had children with four women.

The elite angle. In the greatest of the Irish passage tombs, Newgrange, the genome of an adult man whose remains lay in the inner chamber showed that his parents were first-degree relatives (Cassidy et al. 2020). Such unions are almost universally taboo; they are tolerated, in the historical record, only for ruling dynasties who saw themselves as set apart. It is a hint that some of these societies had a hereditary elite to match their monumental tombs.

A connected Stone Age world

Put the two sciences together and a new picture of Neolithic Europe emerges. Its coasts were not a string of isolated farming villages but a connected world, linked by sea, sharing a powerful religious idea about how to honour and house the dead. That idea was, on current evidence, born on the shore of Brittany and carried by boat along the coasts of the continent, where settled farming kindreds, organised around the male line, raised stones that are still standing more than five thousand years later. We can no longer see Stonehenge or Carnac as the work of a mysterious lost race. They were built by the descendants of Europe's first farmers, whose closest living relatives today are the Sardinians.

Key points

  • A model of 2,410 radiocarbon dates points to a single origin of European megaliths in northwest France around 4500 BC (Schulz Paulsson 2019), although some archaeologists still argue for several centres.
  • The idea spread by sea along the Atlantic and Mediterranean coasts in three waves, implying advanced Stone Age seafaring.
  • In Global25, Neolithic people from France and Iberia to Scandinavia are mostly of Anatolian early-farmer ancestry, with a minority of hunter-gatherer ancestry rising toward the north and west; no Breton megalith genomes are available yet.
  • No living population is close to them; Sardinians come nearest, then Basques, while northern Europeans, Bretons included, were transformed by later steppe migrations.
  • The people buried in the tombs were close kin reckoned through the male line, with a male bias and reduced Y-chromosome diversity (Sanchez-Quinto 2019); at Hazleton North, most descended from one man and four women (Fowler 2022).
  • At Newgrange, an elite individual was the child of a first-degree union, a hint of hereditary ruling families.

Global25 coordinates used in this article

The regional megalith-builder averages and the two ancestral sources. They can be pasted directly into the Vahaduo tool.

France_Neolithic_Paris_basin,0.125447,0.173481,0.043007,-0.051850,0.080142,-0.029900,-0.004961,-0.002088,0.061027,0.094606,0.001535,0.015547,-0.032073,-0.013955,-0.015652,0.000222,0.014195,0.003252,0.005010,-0.006238,0.003418,0.003637,-0.011979,-0.024122,-0.000233
Portugal_Megalithic_Neolithic,0.123136,0.162392,0.052523,-0.028718,0.084939,-0.022869,-0.006324,0.001468,0.064983,0.091151,0.000650,0.015586,-0.033678,-0.017891,-0.005244,0.004375,0.016831,0.003962,-0.000766,0.000307,0.014883,0.004002,-0.017423,-0.046513,0.000327
Britain_Neolithic,0.127453,0.172190,0.053341,-0.037538,0.084997,-0.025227,-0.002951,0.001776,0.063542,0.088894,-0.000214,0.012845,-0.027455,-0.012621,-0.011514,0.003624,0.011667,0.003310,0.003699,-0.002873,0.010003,0.004965,-0.013952,-0.029347,0.000488
Ireland_Megalithic,0.126383,0.170574,0.052303,-0.037368,0.084790,-0.023773,-0.003614,0.002156,0.061646,0.087423,-0.001484,0.013142,-0.029353,-0.011589,-0.009936,0.002464,0.013029,0.004334,0.002947,-0.001264,0.010012,0.005070,-0.014832,-0.031587,0.000925
Sweden_Ansarve_Megalithic,0.129379,0.164854,0.063608,-0.017119,0.088119,-0.015153,0.000705,0.005000,0.053381,0.067184,0.002598,0.006694,-0.011893,-0.000734,-0.007691,0.010342,0.005476,0.005659,0.005992,0.001459,0.013476,0.007295,-0.012612,-0.033458,0.000120
Anatolia_Neolithic_farmer_source,0.117902,0.180087,0.003426,-0.101059,0.051240,-0.047969,-0.003799,-0.006846,0.036167,0.080678,0.008261,0.011309,-0.024164,0.000579,-0.042712,-0.010370,0.022556,0.001388,0.013649,-0.010448,-0.014261,0.005693,-0.004904,-0.003750,-0.004436
WHG_Loschbour_source,0.130897,0.109677,0.203645,0.198000,0.162492,0.059125,0.015041,0.038075,0.100217,0.016219,-0.015427,-0.017235,0.019921,-0.001239,0.061346,0.070670,0.002608,0.007348,-0.008925,0.065406,0.117543,0.010387,-0.049422,-0.173639,0.019519
Sardinian_today,0.121687,0.167285,0.028490,-0.050652,0.060151,-0.022134,-0.003952,0.002496,0.041574,0.077351,-0.000059,0.016649,-0.028664,-0.012974,-0.013572,-0.003001,0.011403,-0.001347,0.001680,-0.012995,-0.002121,-0.001102,-0.010084,-0.021383,0.000337

References

  1. Archaeology Schulz Paulsson B (2019). Radiocarbon dates and Bayesian modeling support maritime diffusion model for megaliths in Europe. PNAS 116(9): 3460 to 3465. doi.org/10.1073/pnas.1813268116
  2. Ancient DNA Sanchez-Quinto F, Malmstrom H, Fraser M, Girdland-Flink L, Svensson EM, et al., Stora J, Jakobsson M (2019). Megalithic tombs in western and northern Neolithic Europe were linked to a kindred society. PNAS 116(19): 9469 to 9474. doi.org/10.1073/pnas.1818037116
  3. Ancient DNA Cassidy LM, Maolduin RO, Kador T, Lynch A, Jones C, et al., Bradley DG (2020). A dynastic elite in monumental Neolithic society. Nature 582: 384 to 388. doi.org/10.1038/s41586-020-2378-6
  4. Ancient DNA Fowler C, Olalde I, Cummings V, Armit I, Buster L, et al., Reich D (2022). A high-resolution picture of kinship practices in an Early Neolithic tomb. Nature 601: 584 to 587. doi.org/10.1038/s41586-021-04241-4
  5. Data Global25 coordinates by Davidski (Eurogenes), and the Moriopoulos Collection 2026 of averages. Admixture models and figures: ExploreYourDNA.