Mobility in Roman Britain: Ancient DNA Reveals the Empire's Far-Reaching Connections

Roman Britain was once imagined as a peripheral and culturally homogeneous province on the northwestern edge of the Empire, where a thin Roman administrative and military veneer overlaid a continuous Iron Age British substrate. The ancient DNA evidence accumulated over the past decade has comprehensively revised this picture. The genetic record of Roman-period burials in Britain reveals a province that was demographically integrated with the broader Imperial Mediterranean and continental world. Individual burials at York, Gloucester, London, and other Roman-period sites have yielded ancestries pointing variously to the Levant, the broader eastern Mediterranean, North Africa, the Rhine and Danube provinces, and the Pontic-Caspian steppe. The most-studied case is individual 3DRIF-26 from the Driffield Terrace cemetery in York, one of more than 80 decapitated burials interpreted as gladiators, military convicts, or victims of organized execution. Martiniano and colleagues (2016, Nature Communications) showed that 3DRIF-26 carried Middle Eastern (Levantine) ancestry, with oxygen isotope analysis of his teeth confirming a childhood spent in a hot, arid climate. The Driffield Terrace cemetery as a whole revealed striking demographic heterogeneity, with most individuals showing British or northern European profiles but at least one (3DRIF-26) coming from over 4,000 kilometers away. Subsequent work has extended this picture. Roman Britain was a mobile, cosmopolitan, ethnically integrated province whose burial archaeology records movements of individuals from across the Empire and beyond.

1. Driffield Terrace and the decapitated dead of York

The Driffield Terrace cemetery in York (Roman Eboracum) is one of the most distinctive Roman British burial assemblages known. Excavated between 2004 and 2005 as part of a development project, the cemetery contained more than 80 individuals, most of them young adult males, the majority of whom had been decapitated, with their skulls placed at their feet, between their legs, or on their chests. The decapitation pattern and the demographic profile (young, male, robust) prompted several interpretations.

One interpretation, favored in the popular media coverage of the initial discovery, was that the cemetery was the burial ground of a gladiator school. Eboracum was a major Roman military center, the headquarters of the Sixth Legion from 122 CE onward, and would plausibly have hosted gladiatorial games. The young male decapitated skeletons fit some of the expected pattern for gladiators (combat-related wounds, robust skeletons reflecting professional training, decapitation possibly reflecting execution of defeated combatants or the coup de grace given to mortally wounded ones).

An alternative interpretation, favored by some archaeologists, is that the cemetery was a military execution site, used for the disposal of executed convicted soldiers and military criminals. The decapitation pattern fits Roman military disciplinary practice, and the demographic profile fits the soldier population. A third possibility is a more general elite execution cemetery used for higher-status individuals subjected to capital punishment by the provincial authorities.

The genetic and isotopic evidence has not definitively resolved which interpretation is correct, but it has dramatically enriched the picture of who these individuals were and where they came from.

2. Individual 3DRIF-26 and the Levantine connection

Of the more than 80 individuals at Driffield Terrace, seven were selected for genomic sequencing in Martiniano and colleagues' 2016 Nature Communications study. Six of these showed broadly British-Celtic genetic profiles, consistent with the underlying Iron Age and Romanized British population of Yorkshire and northern England. The seventh, designated 3DRIF-26, was strikingly different. His autosomal ancestry placed him outside the European range entirely. The best matches were with modern populations of the Levant: Lebanese, Syrians, Israelis, Palestinians, and Jordanians.

The genetic finding was reinforced by oxygen isotope analysis of his tooth enamel. Tooth enamel preserves the oxygen isotope signature of the water consumed during childhood, which varies with climate. The oxygen isotope ratio in 3DRIF-26's teeth was consistent with a childhood spent in a hot, arid climate, not the cool maritime climate of northern Britain. The convergence of genetic and isotopic evidence makes 3DRIF-26's identity as a long-distance migrant unambiguous. He had been born and raised in the eastern Mediterranean, probably in the Levant, and had ended up at York several decades later.

Many specific paths can be imagined for how a Levantine boy might have arrived at Eboracum. He could have been the son of a Roman auxiliary soldier from one of the eastern provinces who was eventually transferred to Britain. He could have been a slave from the Levant brought to Britain through the Imperial slave trade. He could have been a free trader, merchant family member, or administrative official's relative. The genetic and isotopic evidence does not specify the path, only the origin. What is clear is that the path existed: an individual born and raised in the Levant ended up buried at York after his decapitation, and his identification places one specific Roman British case study of long-distance Imperial mobility on the firmest possible evidentiary footing.

The G25 coordinates below place the Iron Age British substrate (Durotrigian samples, England_IA, England_LIA), the Anglo-Saxon overlay (England_Anglo-Saxon), the Sarmatian populations documented in Roman British military history (multiple individuals from the Southern Urals and Kazakhstan), and Levantine reference samples for the 3DRIF-26 individual's ancestry. The 3DRIF-26 G25 coordinates themselves are not yet in the existing panel.

Durotrigian:WBK01,0.127482,0.136081,0.067882,0.046512,0.0397,0.019243,-0.000235,0.006461,0.009408,0.001093,-0.001299,0.002398,-0.014717,-0.023671,0.021987,0.013126,-0.009388,0.00228,-0.001131,0.006003,0.012603,-0.006183,0.003328,0.01217,0.001197
Durotrigian:WBK02,0.135449,0.147252,0.058077,0.028101,0.050779,0.01004,-0.002115,-0.003231,0.02127,0.024784,-0.010555,0.014987,-0.015609,-0.016652,0.007465,0.005171,-0.000522,0.002914,0.003268,-0.002376,0.002246,0.006677,0.001602,0.000723,-0.001557
Durotrigian:WBK03,0.12862,0.144205,0.056945,0.047158,0.041238,0.011156,0.00141,0.002077,0.001023,0.015308,-0.008119,0.003597,-0.019921,-0.010459,0.017236,0.001193,-0.007693,-0.003167,0.004274,-0.002751,0.00549,-0.000989,0.000493,0.003374,-0.002515
Durotrigian:WBK04,0.124067,0.133034,0.056191,0.044897,0.02462,0.016455,0.004935,0.003231,0.006749,0.007836,0.004222,0.005095,-0.013231,-0.016377,0.016151,0.016176,0.008475,-0.002787,-0.008799,0.001251,-0.000499,0.004946,-0.003574,0.008435,0.017962
England_IA,0.1215,0.1533,0.0310,0.0020,0.0220,-0.0045,-0.0050,0.0020,0.0030,0.0080,-0.0025,0.0010,-0.0075,0.0015,-0.0050,-0.0010,-0.0075,0.0015,0.0060,-0.0005,0.0005,-0.0045,0.0020,0.0095,0.0005
England_LIA,0.1278413,0.1347448,0.0603195,0.048331,0.0393271,0.0164399,0.0048114,0.0058904,0.0064586,0.0043066,-0.0037351,0.0049851,-0.0129882,-0.0130815,0.0205865,0.0087997,-0.0015028,0.0022937,0.0023882,0.0067664,0.0040456,0.0037291,-0.0016476,0.0061706,-0.0022185
Scotland_IA,0.1195,0.1548,0.0285,0.0005,0.0210,-0.0060,-0.0060,0.0018,0.0035,0.0095,-0.0020,0.0008,-0.0080,0.0020,-0.0055,-0.0008,-0.0085,0.0012,0.0055,-0.0008,0.0003,-0.0040,0.0022,0.0100,0.0008
England_Anglo-Saxon,0.1165,0.1445,0.0375,0.0155,0.0290,0.0050,-0.0055,0.0035,0.0005,-0.0045,-0.0010,0.0005,-0.0065,-0.0025,0.0020,-0.0015,-0.0105,0.0030,0.0080,0.0010,0.0015,-0.0060,0.0028,0.0120,0.0005
England_EarlyMedieval_Saxon,0.1299946,0.1329385,0.0658823,0.0573538,0.0401816,0.019917,0.0043808,0.0054903,0.0036389,-0.0034728,-0.0053618,0.0048128,-0.0099658,-0.0097504,0.0246781,0.0090236,-0.006497,0.0031625,0.00462,0.0042992,0.0069548,0.0025477,-0.0005884,0.0154329,-0.0016833
Russia_EarlySarmatian_SouthernUrals.SG:LS-13_noUDG.SG,0.108132,0.041637,0.041106,0.080104,-0.01908,0.030399,0.00846,0.013153,-0.023316,-0.045923,-0.010718,-0.004946,0.005352,-0.014588,0.032301,0.002254,-0.029728,0.007601,-0.010433,-0.010255,-0.015223,-0.002597,-0.005176,-0.002169,-0.002275
Russia_MiddleSarmatian_SouthernUrals.SG:MJ-44_noUDG.SG,0.112685,0.049761,0.031678,0.062662,-0.030159,0.028447,-0.000235,-0.000923,-0.028224,-0.031891,-0.006171,0.003297,-0.000297,-0.015964,0.020222,0.015115,-0.008605,-0.009248,-0.016089,0.008504,-0.009109,-0.004699,0.006532,0.00735,-0.002994
Kazakhstan_Sarmatian_IA:AIG002,0.101303,0.053823,0.047894,0.053941,-0.026466,0.035698,-0.007755,-0.002077,-0.029247,-0.024602,-0.006983,0.001948,0.000892,-0.018854,0.015879,0.014717,0.0103,0.022931,0.001257,0.009004,-0.005865,0.014591,0.002095,0.000241,-0.006945
Kazakhstan_Sarmatian_IA:AIG006,0.106994,0.05687,0.03017,0.077843,-0.018157,0.029562,-0.00141,-0.001154,-0.02352,-0.03098,-0.005684,-0.016335,-0.001487,-0.015138,0.018594,0.004375,-0.011083,0.001014,-0.003017,-0.016508,-0.002371,-0.003215,-0.013064,0.00494,0.005508
Levant_South,0.085367,0.148674,-0.057398,-0.092508,-0.009602,-0.035029,-0.002209,-0.008169,0.016321,0.008237,0.009159,-0.009921,0.020485,0.011863,-0.005537,0.002121,-0.011239,0.001014,0.002112,-0.003652,-0.000524,0.003116,0.000567,-0.00294,0.004646
Lebanon_IA3.SG,0.0879281,0.15106,-0.0490729,-0.0918128,-0.0102324,-0.0320375,-0.0045531,-0.009288,0.0108651,0.0133488,0.0070232,-0.0040839,0.0111311,0.0019781,-0.0098059,0.0023701,0.0016461,-0.0003326,0.0031268,0.0003438,0.0026984,0.0054406,-0.0038975,-0.0026359,-0.0047451
Levant_PPNB,0.072847,0.1639064,-0.0316026,-0.1361132,0.0332986,-0.0645352,-0.0134426,-0.0147684,0.0741604,0.03601,0.0188046,-0.0150764,0.035738,0.0025596,-0.0217696,0.006099,0.0098048,-0.0013176,-0.0047264,0.0188088,-0.001797,0.0071472,0.0008872,-0.0056874,-0.0037602

3. The broader Roman British genetic picture

The Driffield Terrace case is the most-studied but not the only Roman British example of individual diversity. Subsequent and parallel studies have identified additional Roman-period individuals with continental European, North African, eastern Mediterranean, and other distant ancestries at various British sites.

At Roman London (Londinium), the Walbrook stream area has yielded individual burials with diverse ancestries, including continental European profiles distinct from the local British substrate. Roman Gloucester (Glevum) has produced individuals with broadly Mediterranean profiles. Roman Colchester (Camulodunum), the original Roman colonial capital, has yielded skeletons with continental European affinities reflecting the legionary and veteran-settler population. At York itself, beyond the Driffield Terrace cemetery, individual burials have shown additional non-British ancestries.

The Patterson et al. 2022 (Nature) study, while focused primarily on the Middle to Late Bronze Age continental gene flow into Britain, included Iron Age and Roman-period comparisons that confirmed the broader pattern. The Iron Age British population (sampled across multiple sites) showed strong continuity from the Bronze Age Bell Beaker substrate with modest continental input. The Roman period added individual long-distance migrants to this substrate, including individuals with eastern Mediterranean and other distant ancestries, but did not transform the underlying gene pool.

4. The Sarmatian connection

A particularly intriguing finding from the broader Roman British genetic literature is the identification of an individual with steppe-related ancestry pointing toward the Sarmatian populations of the Pontic-Caspian zone. Sarmatian auxiliaries are documented in the Roman army during the late 2nd century CE, with Marcus Aurelius transferring approximately 5,500 Sarmatian cavalry to Britain as part of the settlement following the Marcomannic Wars in 175 CE. Roman military records and inscriptions from Britain include references to Sarmatian units at Ribchester and other northern British sites.

The identification of an individual with Sarmatian-related steppe ancestry in the Roman British genetic record provides direct biological confirmation of this documented Roman military movement. The Sarmatian population of the western steppe at this period was an Iranian-speaking nomadic confederation closely related to the Scythian populations that had preceded them in the same broad zone. A Sarmatian cavalry soldier or his descendant ending up buried in Britain would represent another specific case of long-distance Imperial mobility, fitting alongside 3DRIF-26 and the various other long-distance migrants as elements of the broader Roman British demographic mosaic.

The specific interpretation of any one such finding requires caution. Ancient DNA can identify ancestral affinity but cannot definitively distinguish between a Sarmatian-descended Roman soldier, a slave from the steppe region, or a free migrant from the Black Sea provinces with steppe-related ancestry. The general pattern, however, is clear: Roman Britain was integrated into the broader demographic flow of the Empire, and individuals from across this vast geographic range can be detected in its burial record.

5. What this tells us about Roman demographics

The Roman British genetic findings illustrate a broader principle that the genetic study of other Roman-period provinces has independently confirmed. The Roman Empire operated through individual mobility of soldiers, slaves, traders, and administrators rather than mass migration. The underlying population of any given province (Britain, Gaul, Hispania, Anatolia, the Balkans, North Africa) remained largely continuous with the pre-Roman substrate, with a layer of individual long-distance migrants who often did not contribute substantial descendants to the long-term local gene pool.

This is consistent with the demographic structure of the Empire as a whole. Roman armies were posted across the Empire, with units recruited in one province often serving in distant provinces. Slaves were transported across the Mediterranean from one region to another. Administrative officials and their families circulated throughout the Imperial system. Traders and their families settled in commercial centers across the Empire. The result was a network of individual movement that left detectable traces in burial archaeology and ancient DNA but did not transform the underlying population structure of any given province.

The Imperial Roman city of Rome itself was the major exception, where the cumulative effect of slave imports, administrative migration, and commercial flow did substantially transform the demographic profile (as documented by Antonio et al. 2019). The provinces, however, including Britain, generally remained demographically continuous with their pre-Roman substrate, with the Roman period adding individual mobility as a detectable but not transformative layer.

6. The genetic history of Roman Britain in five phases

7. Myth and reality of Roman Britain

8. The wider significance

The Roman British case has become an exemplar for how ancient DNA and isotopic analysis can recover the lives of individual migrants in the deep past. The story of 3DRIF-26, in particular, has captured popular and scholarly imagination as a concrete example of long-distance Imperial Roman mobility. A boy born in the Levant in the second or third century CE, growing up under a hot, arid climate that left its signature in his teeth, somehow ending up at York more than 4,000 kilometers away and being decapitated and buried alongside dozens of other men in a cemetery on the edge of the Empire. The path he took, the circumstances of his arrival in Britain, the reasons for his death, are all unknown. But the basic facts of his origin, his journey, and his death are now documented with the clarity that only converging biological evidence can provide.

This kind of individual-level reconstruction was simply not possible before the application of ancient DNA and isotopic analysis to archaeological skeletons. Earlier methods could identify burials as Roman-period, could classify them by sex and age, could document grave goods, but could not establish where the individual had been born. The new methods open a window on Imperial Roman demographics at the resolution of the individual, complementing the population-level findings of studies like Antonio et al. 2019 and Patterson et al. 2022. Roman Britain, viewed through this window, is revealed as a province more fully integrated into the demographic life of the Empire than the older historiography had recognized, even while its underlying population structure remained continuous with the Iron Age substrate.

9. References

1. Martiniano, R., Caffell, A., Holst, M., Hunter-Mann, K., Montgomery, J., Muldner, G., et al. (2016). Genomic signals of migration and continuity in Britain before the Anglo-Saxons. Nature Communications, 7, 10326. DOI: 10.1038/ncomms10326 Driffield Terrace aDNA
2. Schiffels, S., Haak, W., Paajanen, P., Llamas, B., Popescu, E., Loe, L., et al. (2016). Iron Age and Anglo-Saxon genomes from East England reveal British migration history. Nature Communications, 7, 10408. DOI: 10.1038/ncomms10408 East England
3. 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
4. Gretzinger, J., Sayer, D., Justeau, P., Altena, E., Pala, M., Dulias, K., et al. (2022). The Anglo-Saxon migration and the formation of the early English gene pool. Nature, 610(7930), 112-119. DOI: 10.1038/s41586-022-05247-2 Anglo-Saxon
5. Antonio, M. L., Gao, Z., Moots, H. M., Lucci, M., Candilio, F., Sawyer, S., et al. (2019). Ancient Rome: A genetic crossroads of Europe and the Mediterranean. Science, 366(6466), 708-714. DOI: 10.1126/science.aay6826 Imperial Rome
6. Mattingly, D. (2006). An Imperial Possession: Britain in the Roman Empire, 54 BC-AD 409. Penguin. Standard modern historical synthesis on Roman Britain. Roman Britain
7. Hunter-Mann, K. (2006). Romans lose their heads: An unusual cemetery at The Mount, York. The Archaeology of York Web Series, 6. Original excavation report on Driffield Terrace. Driffield Terrace
8. Eckardt, H., Muldner, G., Lewis, M. (2014). People on the move in Roman Britain. World Archaeology, 46(4), 534-550. DOI: 10.1080/00438243.2014.931821 Isotopic analysis of Roman British mobility. Isotopes
9. Sulimirski, T. (1970). The Sarmatians. Praeger. Standard work on the Sarmatian populations and their integration into the Roman military system. Sarmatians
10. Davidski, A. (ongoing). Global25 PCA modern population averages. Eurogenes Blog. eurogenes.blogspot.com G25 panel