Show Summary Details

Page of

Printed from Oxford Classical Dictionary. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 25 September 2022



  • Peter Heather


  • Ancient Geography

Currently a highly problematic topic, where argument focuses on the correct degree of necessary reaction to past overuse of an extremely simple migration model. The problem has its roots in the development of archaeology as an academic discipline. As the available data grew exponentially and its study became more scientific in later 19th-cent. Europe, it became apparent that remains of the same time period—sometimes—displayed clearly bound regional distribution patterns. This emerged at exactly the same time that rampant nationalism was at its height, and, in this cultural context, it seemed natural to suppose that these regional distribution patterns—‘cultures’—reflected the territories of ancient ‘peoples’: closed, endogamous population groupings, each of whom had their own specific material and non-material cultural peculiarities. And if you supposed that each ancient ‘people’ had its own material culture, then it was only logical to argue that any significant disjuncture in the observable development of one of these cultures was caused by the arrival of a new ‘people’, with a limited number of cases described by ancient sources (such as Caesar on the Helvetii or Ammianus on the Goths) being taken to provide broader support for the idea that ancient peoples tended to move en masse. Up to the early 1960s, consequently, the ancient history of Europe down to the end of the first millennium ce was seen as moving through a series of epochs marked off from one another by incidences of mass migration, where a new population group largely replaced its predecessor. Within Classics, this migration model was applied particularly to its fringes: to the upheavals of the late bronze age, on the one hand, and the so-called Volkerwanderungen of the late Roman period on the other.

In an important sense, however, it did not actually explain that much, since it was never explored in any detail how and why such mass population movements occurred, and they had few parallels in more modern contexts. Equally important, from the 1960s onwards, archaeologists demonstrated that patterns of material culture can change—even dramatically—for a variety of reasons, whether material (as human populations evolve new technologies to cope with their environments), or non-material (since new ideas and mental structures can profoundly affect patterns of material culture). At the same time, social science was unpicking the nationalist assumption that ancient peoples would have had an unchanging group identity, expressed in material form, which further undermined any straightforward link between changing patterns of remains and the arrival of new ‘peoples’. The cumulative effect of these intellectual revisions has been massively to downgrade the perceived importance of migration as a cause of major change in the ancient past. For many Anglophone archaeologists in particular, thinking about migration is associated with a more primitive era in the discipline's history, and this has rubbed off upon historians. Some would now even argue that those ancient sources which seem to describe migratory phenomena uncomfortably close to the old mass migration model must have been infected with a ‘migration topos’—the preconceived notion among the Mediterranean intelligentsia that ‘barbarians’ always moved en masse with their families—and should therefore be discounted.

But this final twist of revisionist argument has been advocated by just a few, and is itself problematic. Ammianus, for instance, was demonstrably capable of describing a whole range of barbarian groups on the move on Roman soil, and refers to massed groups, mixed in age and gender, in only a few specific instances. In the case of the Goths, Ammianus actually describes a complex process, where even originally large groups on the move were subject to political reformation as they moved, a pattern much repeated in the so-called Völkerwanderungen, where Ostrogoths, Visigoths, and Vandals were all new confederations formed on the march. There is also no sign of any large-scale replacement of indigenous populations in the areas affected by these groups. Looked at more generally, good quality historical sources for the first millennium throw up not only occasional moments of large-group migration, but other types of cumulatively significant migratory phenomena, such as the flows of expansion in the Viking period, which increased in scale through the 9th cent. until they encompassed groups of up to 10,000 warriors in the ‘Great Army’ period. It remains the case that other factors entirely caused many of the transformations observable in both archaeological remains and historical sources. Nonetheless, part of the necessary response to previous overuse of a very simple model must take the form of developing more complex and varied ones, based on a comparative study of the best evidence, aiming to generate a more profound understanding of the forms and causes of ancient migration. Simply to reject large-scale migration altogether as a cause of significant change, fails to do full justice either to the evidence, or to modern experience.


  • C. Renfrew and P. Bahn, Archaeology: Theories, Methods and Practice, 4th edn. (2004).
  • G. Halsall, Barbarian Migrations and the Roman West, 376-568 (2007).
  • P. Heather, Empires and Barbarians: Migration, Development, and the Birth of Europe (2009).