L21 Across the Channel: Tracing Insular Roots in Iron Age Continental Celts

R1b-L21 is one of the most prominent Y-chromosome haplogroups in modern Western Europe, with its highest frequencies in Ireland (approximately 80 percent of Irish men), Scotland, Wales, and Cornwall, and elevated frequencies extending into Brittany, the Iberian Atlantic facade, and parts of western and northern Europe. The haplogroup is the dominant patrilineage of the British Isles, where it traces back to the Bronze Age Bell Beaker arrival of 2450 BCE that brought steppe ancestry to Britain and Ireland. For most of the 2010s, L21 was treated as effectively an "Insular Celtic" lineage, defined by its concentration in the British Isles and presumed to be largely absent from the continental Iron Age Celtic populations of Gaul, the Alpine region, and Central Europe. The growing volume of ancient DNA work, particularly the Fischer et al. 2022 Iron Age Gaul paper, the McColl et al. 2024 Germanic-languages preprint, the Antonio et al. 2024 large-scale European Iron Age analysis, and the Patterson et al. 2022 Bronze Age Britain paper, has now substantially complicated this picture. L21 sub-clades have been identified in Iron Age individuals from France, Italy, the Czech Republic, the Netherlands, Austria, and Croatia, dated from the 8th century BCE through the Roman period. These individuals were buried in contexts associated with the Hallstatt, La Tene, and broader Continental Celtic cultural worlds. The L21 lineage was not insular at all; it was a recoverable component of the Iron Age Celtic genetic landscape across continental Europe, alongside the dominant continental Celtic R1b sub-clades like U152 and the Hallstatt-associated lineages. The implications are substantial: the L21 distribution reflects a much wider Atlantic and Celtic cultural-genetic network than the Insular-only interpretation allowed, with bidirectional movement across the Channel and integration into Continental Celtic societies on the same scale as the Insular Celtic populations.

1. The Insular Celtic association: where L21 reaches its peak

The modern distribution of R1b-L21 has its highest frequencies in the British Isles. Approximately 80 percent of Irish men carry the haplogroup, with similar high frequencies in Scotland and Wales. Brittany in northwestern France carries elevated L21 frequencies (approximately 30 to 40 percent in some areas), reflecting the historical Brittonic migrations from Britain in the post-Roman period. The Iberian Atlantic facade (Galicia, Asturias) carries lower but detectable L21 frequencies, generally interpreted as evidence of Atlantic Bronze Age and Iron Age connections.

The Bronze Age origin of L21 in the British Isles is well established by ancient DNA. The Bell Beaker arrival in Britain around 2450 BCE brought a population that replaced approximately 90 percent of the Neolithic British gene pool over the following few centuries. The Bell Beaker individuals carried R1b-L21 (alongside other R1b sub-clades), and these became the dominant Y-chromosome lineages of the British Bronze Age. The Insular Celtic populations of the Iron Age (Cornish, Welsh, Pictish, Irish, etc.) were demographically continuous with the Bronze Age substrate, with limited continental influence until the Roman, Anglo-Saxon, and Viking periods.

2. The continental L21 samples: who and where

The ancient DNA work of the early 2020s has identified L21 sub-clades in Iron Age individuals across continental Europe. The samples below are organized by date and location.

3. Reading the geographic distribution

The continental L21 samples span a substantial geographic range. The westernmost are in Normandy and Burgundy (UN19, GDF1348, CGG023647); the easternmost in the Czech Republic and the Adriatic (I17143, I26776); the southern in the Po Valley (US3190); the northern in the Lower Rhine and Holland (CGG107746). All but one of the samples date to between approximately 750 BCE and 100 CE (the Iron Age and earliest Roman period), with the Klosterneuburg outlier extending into the 4th century CE.

This is the demographic signature of a continental L21 presence integrated into the broader Iron Age Celtic cultural world. The lineage was not restricted to the Atlantic facade or to the British Isles. It was a normal component of La Tene-period European populations, alongside the more frequent R1b-U152 (the Hallstatt-associated subclade dominant in the western Alpine region) and other R1b lineages.

4. Implications for the "Celtic Atlantic"

The continental L21 distribution reframes the popular concept of the "Celtic Atlantic" or "Atlantic Celtic Bronze Age." Earlier reconstructions (Cunliffe 2001 and 2010, and various popular treatments) suggested an Atlantic-coast network linking Iberian, French Atlantic, British, and Irish populations across the Bronze Age and Iron Age. The L21 distribution confirms that this network was a real demographic reality with a recoverable genetic signal, but extends its reach much further inland: into Central and Eastern Europe, into the Adriatic, and northward into the Lower Rhine. The "Celtic Atlantic" was actually a continent-wide phenomenon connecting populations as far apart as Ireland, Bohemia, the Po Valley, and the Croatian coast.

The bidirectional character of the cross-Channel and east-west movements is now also visible. The conventional model of Iron Age population movement is from the continent to Britain (the "Celtic migrations" of antiquarian historiography, recasted in 2022-2024 as the Patterson et al. Late Bronze Age migration and the broader Iron Age Atlantic exchange). But L21 individuals in continental Iron Age contexts indicate movement in the opposite direction as well: British and Irish individuals integrated into Continental Celtic communities, particularly during the La Tene period when elite networks, mercenary recruitment, and trade routes connected the entire Celtic world.

5. The Gaulish samples and their integration in La Tene Europe

Among the Gaulish samples, sites like Bucy-le-Long (Aisne), Barbuise (Aube), Pech Maho (in southern France, not L21 but in the broader sample of Iron Age Gaul), and Urville-Nacqueville (Manche) stand out for their direct L21 representation. These individuals lived at the heart of La Tene Europe, where elite networks, warrior aristocracies, and long-distance exchanges flourished. Bucy-le-Long was a major La Tene cemetery with rich grave goods and clear elite burials. Barbuise was integrated into the eastern French La Tene cultural province. Urville-Nacqueville on the Cotentin peninsula was particularly connected to British La Tene communities, with material culture parallels in Iron Age southern Britain. The L21 ancestry of these individuals is not a sign of isolation: it is a sign of full integration in the broader Atlantic and Continental Celtic identity, with people, lineages, and ideas flowing across the Channel and across the continent throughout the Iron Age.

6. Conclusion: a wider Celtic ocean

The new genetic evidence challenges the older dichotomy between Insular and Continental Celts. R1b-L21 was not a relic of isolated island populations but a living strand of the genetic fabric of Iron Age Europe. It is documented from Burgundy to Bohemia, from the Lower Rhine to the Po Valley, and from Normandy to the Dalmatian coast. It points to a long history of connectivity, mobility, and exchange linking Britain, Ireland, and mainland Europe in a shared Atlantic story.

The implications extend beyond the specific Y-chromosome haplogroup. They suggest that the broader Iron Age Celtic world should be understood as a continent-wide demographic and cultural phenomenon, not a series of regional cultures separated by the Channel or by the Alps. The Hallstatt and La Tene cultural horizons reach from Ireland to the Adriatic, and the recoverable Y-chromosome and autosomal evidence confirms that the genetic substrate underlying these cultural horizons was correspondingly broad.

7. References

1. Fischer, C. E., Pemonge, M. H., Ducoussou, I., Arzelier, A., Rivollat, M., Santos, F., et al. (2022). Origin and mobility of Iron Age Gaulish groups in present-day France revealed through archaeogenomics. iScience, 25(4), 104094. DOI: 10.1016/j.isci.2022.104094 Iron Age Gaul
2. McColl, H., Vinner, L., Damgaard, P. de B., Sjogren, K. G., Allentoft, M. E., et al. (2024). Steppe ancestry in western Eurasia and the spread of the Germanic languages. bioRxiv preprint (manuscript under peer review). Germanic origins L21 continental
3. Antonio, M. L., Weiss, C. L., Gao, Z., Sawyer, S., Oberreiter, V., Bandt, R., et al. (2024). Stable population structure in Europe since the Iron Age, despite high mobility. Nature, 625(7994), 320-326. European Iron Age
4. Patterson, N., Isakov, M., Booth, T., Buster, L., Fischer, C. E., Olalde, I., et al. (2022). Large-scale migration into Britain during the Middle to Late Bronze Age. Nature, 601(7894), 588-594. DOI: 10.1038/s41586-021-04287-4 Bronze Age Britain
5. Laffranchi, Z., Coelho, C., Delgado-Raack, S., Carrillo, J. M., Garcia-Mardones, M., et al. (2024). Until death do us part: A multidisciplinary study on human-animal co-burials from the Late Iron Age necropolis of Seminario Vescovile in Verona (Northern Italy, 3rd century BCE). PLoS ONE, 19(1), e0293434. Cenomani Celts
6. Olalde, I., Carrion, P., Mikic, I., Rohrlach, A. B., Posth, C., Lazaridis, I., et al. (2023). A genetic history of the Balkans from Roman frontier to Slavic migrations. Cell, 186(25), 5472-5485. Balkans
7. Olalde, I., Brace, S., Allentoft, M. E., Armit, I., Kristiansen, K., Booth, T., et al. (2018). The Beaker phenomenon and the genomic transformation of northwest Europe. Nature, 555(7695), 190-196. DOI: 10.1038/nature25738 Bell Beaker
8. Cunliffe, B. (2010). Druids: A Very Short Introduction. Oxford University Press. Standard introductory synthesis of Celtic archaeology and historiography. Celtic context
9. Davidski, A. (ongoing). Global25 PCA modern and ancient population averages. eurogenes.blogspot.com G25 panel