Romans and Barbarians

If you can bear not just Romans but Christian Romans, late antiquity is a fascinating time. The period from when the Roman empire fell into civil wars in the third century CE, and the remainder of the empire drew inwards under pressure from Arabs and Slavs and angry theologians was a time of rapid changes that we know just enough about to argue about. Some of the biggest questions are about how to think about interactions between Romans and barbarians. This has been discussed so intensively by very clever people with very similar backgrounds that debates sometimes get dogmatic and people have a hard time listening to new perspectives.

In his brief period of experimenting on posting on other people’s sites, Canadian historian and essayist Phil Paine had a discussion with medievalist Jonathan Jarett. For my post in October I would like to share his words, and the comparative evidence that he uses.

North America has many well-documented cases of tribal migrations within historic times, in which family groups of hundreds, and sometimes thousands, moved considerable distances with the specific intention of setting themselves up in a new locale. Sometimes this involved making war against existing occupants of a place. Sometimes they were compelled to do so by defeat at the hands of another tribe. At other times it involved deal-making or confederation. It is not known what prompted the entire Mandan Nation, for instance, to migrate a thousand miles from the Midwest to the Upper Missouri country, but they were joined there by the Hidatsa, who were migrating from the equally distant Gulf coast, and they established themselves as allied farmer-traders in a region that had known no agriculture. Some of the locals joined them, some among them split off to become plains warriors. Western Canada witnessed many large scale migrations of people that are traceable over a period of three centuries.

We cannot assume automatically that things worked the same way in the Europe of late antiquity, but drawing analogies from native North America seems to me a valid way of discussing what is likely, unlikely, possible, or impossible.

Many countries in Latin America, equipped with modern armies and technology, are unable to prevent tribal peoples from migrating to the edges of their cities and setting themselves up in favelas or bidonvilles, retaining their own languages and customs without much difficulty. Often this takes the form of “chain migration”, where small groups make a foothold, and then whole villages follow them. The national authorities often send in police, or even the army, to stop such incursions, only to find themselves faced with well-organized and effective opposition.

Despite the Roman Empire’s urbanization and fairly impressive technology, the various tribal peoples on the periphery of the empire could often put together fighting forces that had a good chance of defeating a legion. The difference in military technology was not great — it was money and large-scale co-ordination that kept them out. If that co-ordination and financing was absent, then what was to stop any enterprising, and reasonably aggressive group from simply walking in and carving out a little space for themselves? Especially if they found depopulated areas, or plantations farmed by slaves or aged coloni, or areas in which the local elite saw no percentage in defending the Empire? The fact that the Romans had aqueducts and hypocausts, and the invaders did not, doesn’t seem to weigh much in the equation. Nor does any greater degree of “social complexity” the Empire may have had. The invaders didn’t have to be complex, they just had to fight well. In such a situation, the invading “horde” need not be especially large to put it’s stamp on a region. A well-organized empire could rush disciplined troops to stop isolated incursions — but what about when there were twenty incursions occurring simultaneously, in different locations? The logistic problems pile up quickly. Whatever differences in social complexity existed might work as much in the barbarians favour as against it, just as the crude tribal organization of Afghanistan’s Pathans has proven to be militarily effective against British, Soviet, and American global empires.

The Rajputs of India were little more than a small military caste with an associated ethnic group, from the marginal lands of the Thar desert. Their associated peasantry migrated with them in some conquests, but not all. The states they attacked far surpassed them in organization, wealth, and technology. Yet Rajputs ruled more than four hundred of the estimated six hundred princely states at the time of India’s independence.

Phil Paine, comment to https://tenthmedieval.wordpress.com/2009/02/23/two-seminars-two-cities-part-1-seminary-xl-with-peter-heather/

Roman historians developed a long theoretical and philological argument against barbarian migrations. They used a few historical models of the sort cited by Eric Hobsbawm to argue that what the Romans call “the Ostrogoths” was probably a multiethnic army that picked up soldiers, medics, and camp followers wherever it went, and was united at least as much by their shared hardships as shared language, customs, or ancestry. This works well in some places, in other places the new research on archaeogenetics has challenged some things. But I have never seen any of these historians cite parallels from Turtle Island north of Mexico. It seems generally agreed that the Mexica migrated to their current homeland from far in the north, because that was their tradition and their language is related to languages spoken far to the north but not languages near Mexico City.

Historians have trouble with comparative evidence because our methods require studying the unique evidence from a specific place and time, and there is just so much history. The case that comes to my mind is probably not one that you know. The structure of our departments also discourages breadth: you keep a job at a research university by publishing many things on a narrow specialty. Historians tend to be lone wolves who resist being told what or how to write, but if 100 historians write case studies of 100 societies, those studies are much more useful if they follow the same format. Informal chats about comparative evidence often break down because not everyone in the discussion is a trained scholar who knows every case well. Keepers of bookandswordblog lore will remember my chat with S.M. Stirling where I made some mistakes because I was going outside my specialty and working quickly without time to fact-check.

I admire projects like the World History Association where Roman historian Morgan Lemmer-Webber works, or the database of Religious History at UBC, or Robert Rollinger’s project on empires in world history. Broad comparative studies are hard to do well, and neither universities not the market reward them, but without them people can believe that something is a law when its just true some of the time, or believe that something about their favorite society is special when its actually typical.

Further Reading

Pereltsavig and Lewis’ book on the Indo-European controversy covers some similar issues

The rich world that produced the open web is dying, and a new poorer world is struggling to be born. I would like to keep blogging monthly as the lamps go out one by one. If you can, please support this site by sharing pages, donating, or talking about what you read at your next coffee with friends.

PS. I like Guy Halsall’s idea that the most organized Anglo-Saxons in Britain may have been soldiers from the hill zone that runs north-east to south-west through England who decided that if nobody was paying them any more, maybe they were not Romans either (Caesar’s soldiers had loved playing barbarian for centuries, and by the third century, some European barbarians had reorganized their egalitarian societies so they could support men a lot like Roman soldiers, both in the sense that they had steel swords and silver-plated helmets, and that they bossed around their neighbours and beat or stomped on anyone who was not humble enough).

(written summer to October 2025, scheduled 10 October 2025)