Ancient DNA | Britain | Wales and the Britons

Are the Welsh the last true Britons?

The Welsh are often described as the heirs of the ancient Britons, pushed into the western hills by the Anglo-Saxons. Ancient DNA confirms part of the story, and overturns another: the Britons themselves were born of centuries of exchange with France, and of a large migration from across the Channel some 3,000 years ago. We tested it in Global25.

About halfof the ancestry of Iron Age people in England and Wales came from Bronze Age migrants related to the people of France, according to the largest ancient DNA study of Britain
1000 to 875 BCthe period when that migration reshaped southern Britain, after centuries of trade across the Channel
Closestthe Welsh are the living population nearest to the Iron Age Britons in our analysis, but only by a narrow margin

Hero image: AI generated illustration of newcomers from across the Channel landing a plank boat on a pebble beach in southern Britain around 1000 BC, with cattle, sheep and baskets of bronze axes, greeted by local people below a palisaded village of round houses. Clothing, the boat and the scene are approximations, and it is not a reconstruction of a documented event or of the individuals whose genomes are discussed here.

The story is told in schoolbooks, pubs and rugby stadiums. Before the English, Britain belonged to the Britons, a Celtic people who spoke the language that became Welsh. When the Angles and Saxons arrived in the fifth century, the Britons were killed or driven west, into the mountains of Wales and the peninsula of Cornwall, or across the sea to Brittany. The Welsh, in this story, are the last true Britons, and the English are newcomers.

Even the names carry the story. The Welsh call themselves Cymry, from an old Brittonic word meaning "fellow countrymen". The English word Welsh comes from the Old English wealh, a foreigner, and in particular a Briton or a Roman provincial; the same Germanic root, by most accounts, goes back to the Volcae, a tribe of Gaul. Ancient DNA can now test the story, and it shows that the deeper history of the Britons is a story of exchange with France.

Who were the ancient Britons?Key dates, from the Bronze Age to the genomesabout 2450 BCThe Beaker peopleNewcomers from the continent replace about 90% ofthe ancestry of Britain's Neolithic farmers.about 1600 to 1400 BCThe Great Orme copper boomA mine in north Wales supplies copper from Brittanyto the Baltic: Britain is plugged into Europe.about 1000 to 875 BCMigrants from FranceAfter centuries of contact, newcomers related topeople of France reshape southern Britain.800 BC to AD 43Iron Age BritonsHillforts, Celtic languages and trade with Gaul;Britain then follows its own genetic path.AD 43 to 410Roman BritainRome rules most of Britain, including Wales,for nearly four centuries.5th to 6th centurySaxons arrive, Britons leaveAnglo-Saxon migration into eastern England; Britonscross to Armorica, which becomes Brittany.2015 and 2022The genomesThe People of the British Isles map; the Bronze Agemigration and the Anglo-Saxon migration papers.Dates are approximate.
Key dates in the history of the Britons, from the Beaker migration to the genomes. Dates are approximate.

An island that was never cut off

The idea of a timeless British people was tested first. In the 2000s, some geneticists argued from present-day DNA that most of the ancestry of the British Isles went back to hunters who came from Iberia after the Ice Age. Ancient genomes overturned that view. Farming arrived around 4000 BC with migrants from the continent, and around 2450 BC the people of the Beaker culture, of largely steppe ancestry, replaced about 90% of the ancestry of Britain's Neolithic farmers within a few centuries. The builders of Stonehenge contributed little to later Britons.

The Bronze Age that followed was an age of the sea. The rich Early Bronze Age graves of Wessex, around Stonehenge, have close parallels, in their gold work and bronze daggers, in the princely burial mounds of Brittany. In north Wales, the copper mine of the Great Orme, near Llandudno, boomed between about 1600 and 1400 BC, and analysis of the metal shows that its copper reached France, the Netherlands and Sweden. The Dover boat, a sewn-plank vessel built around 1550 BC and found in 1992, was large enough to carry people, animals and goods across the Channel, and off Langdon Bay, near Dover, divers found more than 350 bronze objects from a lost cargo, most of them of types made in France. Britain, Ireland, Brittany and Iberia shared one Atlantic world of metal, styles and probably ideas.

The migration from France

Trade, however, is not migration, and archaeology alone could not tell whether people were moving along these routes. In 2022, Nick Patterson, Ian Armit, David Reich and colleagues published 793 new genomes from Britain and the continent, the largest study of its kind. They found that the share of early European farmer ancestry, a marker that was higher in France than in Britain, rose in southern Britain through the Middle and Late Bronze Age. The change was strongest between about 1000 and 875 BC, and it was carried by newcomers genetically similar to the ancient people of France. By the Iron Age, these migrants and their descendants accounted for about half of the ancestry of people in England and Wales. The change was much weaker in Scotland.

We repeated the test in Global25, modelling every ancient British genome as a mixture of three classic sources: western hunter-gatherers, Anatolian farmers and steppe herders.

The Bronze Age shift from FranceEarly European farmer ancestry in Britain, Bronze Age to today (three-source model, our analysis)25%30%35%40%45%50%55%2500 BC2000 BC1500 BC1000 BC500 BCAD 1migrationBronze and Iron Age FranceEnglandScotlandTodayBretons 42%Cornish 40%Welsh 40%English 39%Scots 38%Irish 37%EnglandScotlandWales (single genomes)FranceEngland: 13 to 88 genomes per period. Sources: western hunter-gatherer (Loschbour), Anatolian farmer (Barcin),steppe (Yamnaya). Fit distances 30 to 49.
Early European farmer ancestry modelled with western hunter-gatherer (Loschbour), Anatolian farmer (Barcin) and Yamnaya steppe sources (our analysis). England and Scotland: period averages; triangles: single genomes from Wales; squares: averages from France. Published values (Patterson 2022) rise from 31% in the Early Bronze Age to about 36% in the Late Bronze Age, close to ours.

The rise is clear. In England, the farmer share climbs from 29% among the Beaker people and 31% in the Early Bronze Age to 34% in the Middle Bronze Age, 37% in the Late Bronze Age and 39.5% in the Early Iron Age, very close to the published values. In Scotland the curve stays flatter, around 33 to 35%. The genomes from Wales are single individuals, but they follow the English pattern: 27 to 33% in the Bronze Age, 33 to 38% in the Iron Age. The people of Bronze and Iron Age France, at about 50%, sit above everyone, as the source of such a shift should. And the Welsh of today, at 39.6%, sit exactly at the level of Iron Age England.

We then measured how much of a Bronze Age French source each later British group needs, on top of the Early Bronze Age people of England.

How much came from France?Share of a Bronze Age French source in later Britons, against an Early Bronze Age English baseline (our analysis)0%10%20%30%40%50%England, Middle Bronze Age12%England, Late Bronze Age24%England, Early Iron Age31%England, Middle Iron Age30%England, Late Iron Age28%Wales, Middle Iron Age (1 genome)28%Scotland, Middle Bronze Age14%Scotland, Late Bronze Age12%Scotland, Iron Age9%Other French sources give higher values (up to 44% in England). Published estimate: about half.
Share of a Middle Bronze Age French source in later British groups modelled as Early Bronze Age England plus France (our analysis). With other French Bronze and Iron Age sources, the Iron Age values rise to 38 to 44%. Patterson and colleagues estimate about half.

The French share rises from 12% in the Middle Bronze Age to 24% in the Late Bronze Age and 31% in the Early Iron Age of England, then stays near 30%. The one Middle Iron Age genome from Wales gives 28%, in line with England, while Scotland stays at 9 to 14%. Our estimates are lower than the published "about half", and they depend on the French source: with other French Bronze and Iron Age groups the Iron Age values reach 38 to 44%. The real migrants have not been sampled, and Global25 can only use proxies. The direction, the timing and the contrast with Scotland are robust; for the exact proportion, the formal analysis of Patterson and colleagues is the reference.

Did the migrants bring the Celtic languages?

Patterson and colleagues proposed that this migration was a plausible vector for the spread of early Celtic languages into Britain. Their argument is one of timing: the Celtic languages are too similar to one another to have separated as early as the Beaker migration, 4,500 years ago, and almost no migration is visible later in the Iron Age. A migration from France around 1000 BC fits the linguistic dates better. It is a hypothesis, not a proof, and it remains debated by linguists and archaeologists who favour an earlier spread along the Atlantic coast. The same study also noticed a biological change: the genetic variant that allows adults to digest milk spread much faster in Britain than in central Europe, reaching a frequency of about 50% by the Iron Age.

If the hypothesis is right, the language that would become Welsh was itself, in part, a gift of the people who crossed from France.

The Iron Age: Gaul next door

After 875 BC the curve flattens. Through the Iron Age, the ancestry of southern Britain barely changes, and Patterson and colleagues found little sign of new migration. Contact, however, did not stop. Julius Caesar, writing in the 50s BC, reported that the coast of Britain was held by Belgae who had crossed from Gaul, that a king of the Suessiones had ruled over part of Britain as well as northern Gaul, that the Britons had sent help to the Veneti of Armorica against Rome, and that Gauls who wanted to study the druidic teaching went to Britain to learn it. At Hengistbury Head, in Dorset, a port of the first century BC received wine from Italy and coins from Armorica. In East Yorkshire, a people named Parisii, like the tribe on the Seine, buried their dead with chariots.

The genomes show that these intense contacts left no large mark on ancestry: in our analysis, the farmer share stays between 38 and 40% from the Early to the Late Iron Age. Britain and Gaul spoke related languages, traded and fought together, but after the great Bronze Age migration, the Britons followed their own genetic path.

Then came the Saxons

The next great change came after Rome. In 2022, Joscha Gretzinger, Stephan Schiffels and colleagues showed that people buried in eastern England in the early medieval period carried up to 76% ancestry from the continental North Sea coast, the homeland of the Angles and Saxons: a mass migration, not the arrival of a small elite. The people of England today still carry a large part of that ancestry, highest in the east, diluted by older British ancestry and by a later input resembling the Iron Age people of France.

Wales was a different story. The 2015 map of the People of the British Isles, built from more than 2,000 people whose grandparents were born in the same rural areas, found that the Anglo-Saxon contribution was substantial in central and southern England but less than half, that Wales was little affected, and that there was no single "Celtic" population: the people of north and south Wales, of Cornwall and of Scotland form distinct groups. The same map had already detected, in all British populations, a contribution from people similar to the present-day French, arriving after the first settlers but before Roman times. Ancient DNA has since given that signal a date.

Here Global25 reaches its limits, and we prefer to say so. The continental North Sea people who came with the Anglo-Saxons sit only 19 to 26 units from the Iron Age Britons in Global25, too close to be separated reliably. When we tried to measure the Anglo-Saxon share of present-day populations, the model gave the Irish more of it than the Welsh, which is not plausible. We therefore rely on the published estimates for that question.

Who looks most like the Iron Age Britons?

What Global25 can do well is measure distance. We compared the 60 genomes of Middle and Late Iron Age England with present-day populations.

Who looks most like the Iron Age Britons?Global25 distance from Middle and Late Iron Age England (60 genomes) to present-day populations (our analysis)0.03.06.09.012.015.018.0Welsh9.9Bretons11.7English12.0Orcadians12.3Cornish12.7Scots12.9Irish14.9Dutch15.3Distances x 1000. Everyone in the British Isles is close; the Welsh are closest.
Global25 distance from the average of 60 Middle and Late Iron Age genomes from England to present-day populations (our analysis). Against the average of all Iron Age British genomes, the Welsh are also closest, at 15.8.

The Welsh come first, at 9.9 units, and they also come first against the average of all Iron Age Britain, at 15.8. That is a real result. But the margin is small: the Bretons, the English, the Orcadians, the Cornish and the Scots all sit between 11.7 and 12.9 units, and even the Dutch are only 15.3 away. Everyone in Britain today is close to the Iron Age Britons, and the English are not far behind the Welsh.

The second place of the Bretons is not a coincidence. In the fifth and sixth centuries, while the Saxons settled in the east of Britain, Britons crossed the Channel in the other direction and settled in Armorica, which took their name and became Brittany. Their language, Breton, is the closest relative of Cornish. Bretons today carry a farmer share of 41.5%, a little above that of the Iron Age Britons, and their closeness to them reflects both that migration and the much older kinship between the two sides of the Channel.

So, are the Welsh the last true Britons?

In part, yes. The Welsh are the living population closest to the people of Iron Age Britain, their language descends directly from the Brittonic language of the Iron Age, and the Anglo-Saxon migration touched Wales much less than England. In that sense, the old story holds up.

But the genomes add three corrections. First, the English are Britons too: a large part of their ancestry also goes back to the Iron Age population, and in our analysis they sit only two units further from it than the Welsh. Second, the Bretons are the cousins the story forgets, descended in part from Britons who crossed the sea. Third, and most surprising, the ancient Britons were not an untouched island people. Their ancestry was formed by the Beaker migration, then by centuries of trade with France and by a migration from across the Channel around 1000 BC that contributed, by the published estimate, about half of their ancestry. If the Welsh are the last true Britons, the true Britons were already, in good part, the children of France.

Some limits should be kept in mind. The ancient genomes from Wales are few and mostly single individuals. The present-day Welsh in Global25 are a small sample, and the People of the British Isles project showed that north and south Wales differ. The source of the Bronze Age migrants has not been sampled directly. And genes do not carry language: the link between the migration from France and the Celtic languages remains a hypothesis.

Key points

  • Britain was connected to France throughout the Bronze Age: Wessex and Brittany, the Great Orme copper trade, the Dover boat and cargoes of French bronzes.
  • Between about 1000 and 875 BC, migrants related to the people of France reshaped southern Britain; by the Iron Age they accounted for about half of the ancestry of England and Wales, much less in Scotland.
  • In our Global25 analysis, the farmer share of England rises from 31% in the Early Bronze Age to 39.5% in the Early Iron Age; the Welsh today stand at 39.6%.
  • Our French Bronze Age share reaches about 30% in Iron Age England and 28% in one Iron Age genome from Wales, depending on the source used; the published estimate is higher.
  • Iron Age contacts with Gaul were intense but left little new ancestry; the Anglo-Saxon migration later reshaped England and touched Wales much less.
  • The Welsh are the closest living population to the Iron Age Britons, but the Bretons, English, Orcadians, Cornish and Scots are only slightly further.

Global25 coordinates used in this article

The model sources, the ancient British and French averages and a few present-day populations. They can be pasted directly into the Vahaduo tool.

Loschbour_WHG,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
Barcin_Anatolian_farmer,0.112685,0.182795,0.010936,-0.100130,0.055087,-0.046854,-0.002350,-0.002769,0.046836,0.081642,0.008607,0.010191,-0.015758,0.006193,-0.040173,-0.020021,0.003520,0.000507,0.011816,-0.014132,-0.008735,0.006306,-0.010969,-0.004940,-0.004670
Yamnaya_Samara,0.125838,0.089254,0.042908,0.115456,-0.027868,0.044685,0.004491,-0.002949,-0.054858,-0.072996,0.001858,0.000350,-0.001652,-0.023610,0.037263,0.015734,0.000000,-0.001478,-0.001704,0.012506,-0.003120,0.001374,0.011229,0.018436,-0.004524
England_EBA,0.125402,0.126416,0.061471,0.063742,0.028249,0.020407,0.004838,0.003963,-0.000310,-0.008163,-0.004273,0.004801,-0.011375,-0.015912,0.025146,0.008413,-0.005310,0.002031,0.001976,0.005481,0.004277,0.003914,-0.000289,0.005318,0.000165
England_LBA,0.126431,0.132175,0.061761,0.047556,0.036457,0.017291,0.001139,0.004065,0.003147,0.002607,-0.003498,0.006248,-0.014717,-0.016356,0.020713,0.008251,-0.003159,0.002105,0.001306,0.004242,0.008264,0.005916,-0.003783,0.007944,-0.000617
England_EIA,0.127197,0.134521,0.061606,0.045243,0.039996,0.015907,0.002744,0.004459,0.007816,0.009717,-0.002877,0.005465,-0.015668,-0.014028,0.020557,0.003940,-0.006556,0.000733,0.000323,0.001434,0.007010,0.002932,-0.002016,0.002380,0.001535
England_MIA_LIA,0.129779,0.136413,0.061080,0.046330,0.040824,0.015618,0.002547,0.004296,0.007408,0.007442,-0.004195,0.006657,-0.014601,-0.013960,0.021424,0.006352,-0.008610,0.002043,0.001879,0.003700,0.005347,0.003303,-0.002178,0.005328,-0.001293
France_MBA,0.133173,0.142174,0.051666,0.016150,0.046470,-0.001394,-0.008930,-0.000462,0.022907,0.031891,-0.000162,0.007044,-0.019177,-0.022432,-0.000136,0.017634,0.014864,-0.002154,0.003142,-0.007003,0.006364,-0.001731,-0.010230,-0.000723,0.000479
Welsh_today,0.131978,0.138163,0.060132,0.043024,0.040592,0.016622,0.003314,0.006392,0.006279,0.004337,-0.004230,0.005912,-0.012911,-0.010975,0.020596,0.005708,-0.008703,0.002356,0.003086,0.002007,0.004985,0.003852,-0.000012,0.011224,0.000808
English_today,0.131855,0.137043,0.061788,0.044013,0.039230,0.016748,0.004985,0.005745,0.005264,0.005678,-0.004769,0.005616,-0.012589,-0.010351,0.020615,0.003552,-0.010369,0.004094,0.003685,0.002962,0.005937,0.003404,-0.003269,0.013794,0.000031
Cornish_today,0.130197,0.139752,0.061355,0.043183,0.041901,0.016412,0.003435,0.006390,0.006482,0.006967,-0.004634,0.006479,-0.013745,-0.010248,0.020755,0.003172,-0.012527,0.002115,0.003761,0.003242,0.004809,0.003871,-0.000758,0.014367,0.000691
Breton_today,0.131466,0.138340,0.057850,0.039285,0.039169,0.015813,0.003155,0.004211,0.008186,0.009490,-0.005131,0.005590,-0.014368,-0.013102,0.018343,0.004399,-0.007302,0.001280,-0.000462,0.001773,0.004467,0.001787,-0.001254,0.010547,0.000659
Scottish_today,0.131466,0.134232,0.062265,0.047043,0.039106,0.017281,0.003550,0.005176,0.003572,0.002825,-0.005614,0.005047,-0.011680,-0.011973,0.023024,0.003125,-0.011255,0.002774,0.002550,0.001452,0.004225,0.003308,-0.000656,0.013556,-0.000770
Irish_today,0.133361,0.134098,0.061169,0.048864,0.037788,0.019352,0.003293,0.004713,0.003568,0.002927,-0.006971,0.005840,-0.014189,-0.014076,0.025935,0.005215,-0.011145,0.001894,0.000597,0.001776,0.005114,0.001279,0.000293,0.014407,0.000661
Orcadian_today,0.130669,0.134253,0.062979,0.048450,0.038715,0.017542,0.001786,0.003923,0.005031,0.001786,-0.006837,0.004811,-0.011908,-0.010831,0.022353,0.003036,-0.012386,0.003471,0.002677,0.000300,0.004954,0.002250,-0.002157,0.010038,-0.000599

References

  1. Ancient DNA Patterson N, Isakov M, Booth T, Buster L, Fischer CE, Olalde I, et al., Armit I, Reich D (2022). Large-scale migration into Britain during the Middle to Late Bronze Age. Nature 601: 588 to 594. doi.org/10.1038/s41586-021-04287-4
  2. Ancient DNA Gretzinger J, Sayer D, Justeau P, Altena E, Pala M, et al., Schiffels S (2022). The Anglo-Saxon migration and the formation of the early English gene pool. Nature 610: 112 to 119. doi.org/10.1038/s41586-022-05247-2
  3. Ancient DNA Olalde I, Brace S, Allentoft ME, Armit I, Kristiansen K, et al., Reich D (2018). The Beaker phenomenon and the genomic transformation of northwest Europe. Nature 555: 190 to 196. doi.org/10.1038/nature25738
  4. Genomics Leslie S, Winney B, Hellenthal G, Davison D, Boumertit A, et al., Bodmer W, Myers S (2015). The fine-scale genetic structure of the British population. Nature 519: 309 to 314. doi.org/10.1038/nature14230
  5. Archaeology Williams RA, Le Carlier de Veslud C (2019). Boom and bust in Bronze Age Britain: major copper production from the Great Orme mine and European trade, c. 1600 to 1400 BC. Antiquity 93(371). doi.org/10.15184/aqy.2019.130
  6. Historical source Julius Caesar, Commentaries on the Gallic War, books 2 to 6 (written in the 50s BC).
  7. Data Global25 coordinates by Davidski (Eurogenes), and the Moriopoulos Collection 2026 of averages. Admixture models and figures: ExploreYourDNA.