The Roman Empire’s Cosmopolitan Frontier
Genetic analysis reveals a culture enriched from both sides of the Danube.
At its peak, the Roman Empire stretched from North Africa to northern Britain and was home to perhaps a quarter of the world’s population. The early Empire’s well-documented center in Rome, a vibrant hub of trade and communication, drew an ethnically diverse population. Much less is known about life on the Roman frontier in the Balkans, where the river Danube marked the Empire’s edge, and a bristling military presence kept “barbarians” from the north at bay. Now an analysis of the unique ancestries of more than 130 individuals buried in the Balkans region during the first millennium C.E. (Common Era) has enabled a team of historians, archaeologists, and geneticists to see that the region was “as cosmopolitan in many respects as the megacity of Rome,” says Goelet professor of medieval history Michael McCormick, a coauthor of the study released today. The patterns of change revealed by the archaeogenetic data—pioneering new techniques of discovery advanced by scholars in the Initiative for the Science of the Human Past at Harvard—enrich historical understanding of the Empire’s collapse in the sixth century, and document the rise of successor populations, whose descendants still inhabit the region more than a thousand years later.
Among the stunning discoveries: a young man about 15 to 18 years old of East African descent, far from his homeland, buried honorably in the second or third century with an eagle-decorated oil lamp to light his way into the underworld; people of mixed central/northern European and Sarmatian-Scythian ancestry (perhaps Goths) buried as Romans; and a woman of Eastern European (Slavic) origin, interred with care, who lived in the Balkans centuries before the Slavic migrations that reshaped the region’s demographics after the Empire’s collapse.
Striking, too, was absence of Italians, says senior author David Reich, whose pioneering studies of ancient DNA are now being applied to the study of civilizations with scattered written records. The researchers instead detected significant ancestry from Anatolia, a Greek-speaking region now within Turkey. This discovery echoed another published in 2022: that essentially “there were no Italians,” he says, “even in Rome itself.” In that massive study, Reich and 205 colleagues found that the ancestry of the population of Rome during the Imperial period (from 1 to about 550 C.E.) was drawn principally from Anatolia. “Why are there no Romans in Rome?” asks Reich. “Why are there no Romans in the Balkans?” The genetic data so far point to a startling conclusion: that the population of Iron Age Italians (circa 1,300 to 100 B.C.E.) essentially disappears at some point.
“It’s a new enigma,” says McCormick, among many “completely new facts” uncovered by the genetic analysis. One is the evident internal mobility of populations within the Empire. The burials include two men with ancestry from North Africa, “one of the Empire’s richest provinces, which produced grain, oil, some wine, and the red slipware that became the tableware of the middle classes of the Roman Empire from about 200 C.E. on.” Another is the evidence of mobility across the frontier, attested by the presence of migrants from north of the Danube that appear during the late Imperial period between 250 and 550 C.E.
“The Roman frontier was not an ‘iron curtain’ or wall,” explains McCormick, “it was a giant communications corridor studded with fortresses and towns, which in this case ran from the North Atlantic, all along the Rhine, then over to the Danube, and then from the Danube to the Black Sea.” To move things by land at that time, he says, cost 10 times what it cost to move them by water. “And so the Romans liked to make the frontier on the river, not because they were trying to keep people out, but so that they could move men and materiel up and down very quickly, easily, and cheaply,” he says. The frontier, the study now reveals, was also “a magnet for immigrants.”
Several of the sequenced individuals possessed mixed ancestry from Northern Europe and the Pontic-Kazakh steppe—an admixture of central/northern European and Sarmatian-Scythian migrant populations that probably mixed north of the Black Sea before entering the Empire. “And one of the most moving individuals,” says McCormick, “is a woman who has genetic ancestry that will be identified in future centuries as Eastern European, that is to say, a member of a Slavic population.” She lived on the frontier at least 300 years before any other known Slavs. McCormick describes this as “a kind of percolation and advance movement of what would become one of the great migrations of late antiquity, the migrations toward the Mediterranean and toward the Atlantic of the Slavic populations” that occurred after the Empire’s collapse.
Collapse of the Roman Empire
Although the genetic record does not help explain what led to the demise of Roman civilization, the history, informed by other sources, is certainly reflected in the changing demographics of the Balkans during the first millennium CE. As coauthor Kyle Harper, Ph.D. ’07, a professor of classics and letters at the University of Oklahoma, and author of The Fate of Rome: Climate, Disease, and the End of an Empire explains, in 536 a major volcanic eruption or eruptions led to a visible dimming of the sun described in written sources as lasting more than a year: “Ice core evidence documents sulfate peaks at precisely this layer of 536 that are consistent with a northern hemispheric volcano or volcanoes, which induce short-term climate cooling for a period of two to three years.” Tree ring records provide corroboration of a major climate disruption. “Then, approximately four years later,” Harper continues, “around 540, there is another volcanic eruption. This one seems to be tropical, because the sulfate peaks are recorded at both the north and the south pole, which is usually indicative of volcanic eruption in the tropics. This is a massive volcano, even bigger than 536. And then there’s another one a few years later, but the concatenation and magnitude of these repeated episodes of volcanic eruption induced short-term cooling that probably caused harvest failure, famine, suffering, and may have contributed to the onset of the bubonic plague.”
The result is that the decade of the late 530s and 540s is one of the coldest in the last 2000 years, and marks the beginning of what is known as the Late Antique Little Ice Age. The cold persists for a century and half. The volcanism didn’t continue, Harper explains, “But it was strong enough to force climate change,” perhaps combined with other mechanisms.
Among the burials sampled for DNA, there are very few from this period of upheaval, the moment when the Empire began to fall apart. “It is a little bit provocative,” says Harper, “that there are so few samples from the sixth century. It doesn’t mean that there was a significant population collapse, but it’s certainly consistent with that. And this is a period of upheaval, wars, but also environmental crisis, major climate change, and pandemic disease.” The plague of Justinian, which some scholars estimate to have killed as much as 40 to 60 percent of the Empire’s population, may have affected the sampling. “But the sampling strategy isn’t targeted enough,” he says, to make strong inferences.
By the seventh century, evidence of the Slavic migrations (into what may have been a relative demographic vacuum) appears in the genetic record, and has left its mark on present-day populations from the Balkans stretching to the Aegean Sea. Says Reich, “I think that a really valuable thing to do would be to fill in precisely that period of time, that moment of earliest arrival of Slavic speakers in the region. We haven’t captured that with the current genetic data.”
McCormick notes that comparisons are being drawn between what happened on the Balkan frontier and what took place in the Empire’s northern reaches in Britain at the same time: a very large-scale migration of people of Northwest European ancestry (Anglo Saxons) into this territory in the late fifth and sixth century. “The two movements appear comparable in relative scale,” he says, “and also because of the linguistic component, and the future demographic implications.”
“A New Kind of Roman History”
“This is the beginning of a new kind of Roman history,” says McCormick, “in which we can go beyond the rich but very uneven, written record, and hear from the people who had no voice in the historical record until this time. And that’s extraordinarily moving, and also laden with information.”
Beyond the historical implications, the study is remarkable for its direct engagement with historians and archaeologists. More than 70 scholars collaborated on the research. Among them: Reich; first author Iñigo Olalde, an associate of Harvard’s department of human evolutionary biology who is a population geneticist at the University of the Basque Country; McCormick; and Harper. Reich and Olalde worked closely with Pablo Carrión and Carles Lalueza-Fox of CSIC-Universitat Pompeu Fabra, Ilija Mikic of the Archaeological Institute of Serbia, Miodrag Grbic of the University of Belgrade, and many others to assemble samples for genetic analysis. McCormick and Harper assembled a 50-page archaeological overview that describes the geography, grave goods, and other salient details of each burial—an effort that Reich describes as “a huge, central contribution to the final output.”
Says Harper, “I’m proud to be part of a team that has achieved a very, very high degree of interdisciplinary collaboration. A broad group of people from different countries, speaking different languages, working in the different traditions of archaeology and history, came together to generate a study that provides rich historical and archaeological context…to the genetic results. The archaeological supplement to this paper…is full of hard-won insights that are the product of a tremendous amount of expertise…in history and archaeology.”
Adds Reich, “I think our Balkans paper is particularly significant as a case example of co-equal collaboration between historians and archaeologists and geneticists.” McCormick describes the work as “the fruit of the extraordinary intellectual ecosystem at Harvard that the Initiative for the Science of the Human Past has fostered in path-breaking interdisciplinary investigations.” “By integrating advanced humanistic, historical, and archaeological knowledge” with scientific discoveries, he said, “these findings create new knowledge” that transcends any single discipline.