Cross-societal comparisons then & now

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

At Discover I have a long review up of Empires and Barbarians: The Fall of Rome and the Birth of Europe. I would recommend the book, especially if you enjoyed The Horse, the Wheel, and Language or Empires of the Silk Road. In any case, I want to highlight two points in the author’s argument which I think bear more emphasis.

First, the author argues that there was a mass migration of German (and others, such as the Iranian Alans) across the limes as the Western Roman Empire collapsed in the 5th century. This is in contrast with the model that the barbarian war bands were of trivial size, and most cultural and social evolution occurred through a process of bottom-up emulation. In this model the Ostrogothic elite of Theodoric’s time was a motley ad hoc construction of recent generations whose origins were diverse, German and non-German, without much time depth. Against this model is that of total replacement, which was common in the early 20th century. Empires and Barbarians takes a reasonable middle road; in very few regions was there total replacement, even in what became England where cultural obliteration of the Romano-British heritage was nearly total. But, the author also argues that the core of groups such as the Goths and Vandals were German tribes which had relocated from Central Europe, and whose identities were deep and to a great extent ethnically demarcated. Additionally, not only was this movement of some size, it also included women and children, and so was a classic Völkerwanderung.

But just as there was no total replacement of Romans in northern Gaul, despite the non-trivial influx of Franks into that region, so the migration out of Central Europe did not leave the old Gothic or Vandal heartlands empty. In fact, the majority of those German tribes and clans which identified as Goth or Vandal may have remained in the heartlands. But critically the elites, and in particular the ruling houses and the free warriors and their families decamped. Roughly the top 10-20% of the population.

This is not so surprising. If you read Albion’s Seed: Four British Folkways in America you observe the importance of “forward elements” in British migration to the United States. In Puritan New England there was a proactive attempt to discourage the emigration of the lower classes (as well as the blooded nobility), and a strong bias toward the “upper middle class” (the professions, the gentry, etc.). In the lowland South nobility of England often brought their own hierarchical social system along with them, including their customary retainers. It took some wherewithal to move en masse in an organized fashion. But the author also points out that in the ancient world there was little motive for peasants to move, as there was little difference in quality of life from locale to locale. In a world where productivity gains were marginal and zero-sum economic psychology dominated the motive existed for the rent-seeking elites to move onto greener pastures, not the productive peasantry who were the green pastures no matter where they were resident.

A class dimension to the Völkerwanderung is something that I think might be important, because I recall reading archaeologists noting how robust and tall the Lombards who entered Italy in the 6th and 7th centuries were on average. In the work I had read this was interpreted naturally as an ethnic difference, as the Germans were a larger folk than the local proto-Italians. But if there was this class bias in migration then the size difference has a more natural explanation: the malnourished majority of the German population never emigrated, rather, it was the hale and robust warrior elite who show up in post-Roman states. But there’s a bigger issue here, and that’s the point that pre-modern elites viewed wealth through their own lenses as rentier thugs. In The Fall of Rome: And the End of Civilization Bryan Ward-Perkins emphasizes that the collapse of the Roman Empire did result in a social and economic regress. The data from Britain for example shows that air pollution because of industrial economic activity (broadly construed) did not reach the levels of the Roman period until the 18th century! In Empires and Barbarians the author also agrees that the Roman Empire was wealthier than the barbarian lands to their north and east, and in particular that the German dominated Jastrow and post-Jastrow societies practiced an extensive form of low productivity (per unit of land) agriculture which made their conquest economically a losing proposition for the Roman Empire. And naturally the relatively low per unit economic productivity of the German heartland resulted in fewer rents for its local elite.

And yet set against this we have the arguments in works such as A Farewell to Alms which assume as a default model that the median human in all societies from the emergence of humanity until 1800 was poor. Caught in the “Malthusian trap,” whereby gains in productivity were always swallowed up by population increase. This perspective gains some support from Contours of the World Economy 1-2030 AD. Before 1800 the differences in median wealth across regions is marginal, with the largest gaps being 50%, and 10-20% more typical. From our modern perspective everyone was caught in the Malthusian trap, on average, though some were marginally more well off than others.

But it isn’t the average difference which matters, it is the aggregate whereby you calculate total wealth in a political order. By that measure, even if the Roman peasant was no more well off (or even less well off!) than the average peasant on the Baltic, the high population density of the Roman political order was extremely beneficial to any rentier elite seeking to capture or extract surplus productivity from these teeming masses. In a pre-modern political order poverty may have been a permanent feature of the lives of most, but the configuration and implementation of subsistence and the distribution and flow of goods above subsistence were of the essence. It is much easier for men with swords to steal from densely settled agriculturalists than nomads or slash & burn cultivators. The Roman peasant may not have been wealthier than the German peasant, but the Roman aristocrat of the 4th century lived a life of glamorous opulence in relation to the German warlord. Similarly, the Chinese peasant may not have been wealthier than the cultivators who lived beyond the frontier in 16th century Manchuria, but the Manchu dynasty fell into orders of magnitude more wealth after they toppled the Ming because they captured the much richer flow of rents.

The narrative of Empires and Barbarians is much denser than the above, and the analytical framework more sophisticated. But I think it is critical to emphasize why ancient barbarian elites were so keen on conquering civilized states, and why there seems to have been less mass migration of the peasantry. In the modern world when we think of differences between societies in regards to wealth, complexity or glory, we consider the median man on the street. This would tell us little for most of human history, rather, we would have to focus on the top 10% to truly get a sense of the difference, and in particular the top 1%. To a great extent civilization has been a racket which operates to the benefit of the tiny elite by making rent-seeking much more efficient.


  1. German dominated Jastrow and post-Jastrow societies practiced an extensive form of low productivity (per unit of land) agriculture which made their conquest economically a losing proposition for the Roman Empire.

    That’s a pretty general rule. “Barbarian” areas are areas which are more expensive to conquer and control than the taxes collected on the area can pay for. In the case of the steppe, it’s pretty fundamental, since the steppe is just not productive of anything except horses which in turn were primarily military in use at that time, so steppe dwellers balanced the profit of horse sales to e.g. the Chinese military against the profits of plunder, conquest, and tribute extortion. .

    In Germany there had to be another factor though, since Germany was not inherently unproductive the way the steppe was. Perhaps it was just that Rome was too overextended to be able to conquer and exploit Germany, which would have required a military expenditure plus startup investment for more productive agriculture plus farm personnel to develop the agriculture.

    I’ve also thought that in some cases warfare in the barbarian areas, which tended to be lacking a central power enforcing order, had the effect of keeping population down and avoiding Malthusian stunting as found in genuinely overpopulated areas. This might coexist with your theory that only the well-fed elite migrated, or it might not.

    In any case, the explanation of invasions by “overpopulation” is wrongheaded, because invador have to be well fed and tend to come from thinly-populated areas.

  2. totally agree on the last point.

    re: farming practices, it looks like northern europe switched to three-field and better ploughs (harnessing horse power) after 500. i think some of this is ecological, farming spread from the southwest, and pre-agricultural lifestyles persisted down to the historical period in the fennic borderlands because mid eastern grains were so ill-equipped to flourish up north. it just took a while for northern europeans to optimize agriculture to their ecology.

  3. What do you think of James Scott‘s argument that the culture of certain highland/hillbilly peoples (even including illiteracy) is a strategy to avoid being under the rule of a state?

  4. What do you think of James Scott’s argument that the culture of certain highland/hillbilly peoples (even including illiteracy) is a strategy to avoid being under the rule of a state?

    i agree with this model. probably would make a more general argument about centralizing powers, of whatever form. in the book referred to above the author notes that a particular type of small-scale low-complexity slavic culture seems to have spread mostly via the east-central european uplands. territory that’s hard to control and is also of lower productivity.

  5. Similar mountain antistate patterns can be found in the Atlas mountains of N. Africa, the Balkans, the Caucasus, probably Afghanistan and parts of SE Asia. Tibet as a whole used to fit that pattern; probably it was a non-state society up until the very recent Chinese takeover.

    The steppe nomad antistate pattern is different, and steppe nomads could also be state-formers. Basically the steppe was organized on a large scale only for military attack or defense.

    I always recommend Black-Michaud’s “Cohesive Force” when this stuff comes up. He describes the organization of violence and the structuring of society by feud and vendetta in various parts of the world.

    Iceland during the saga era was another non-state society. Even in the 19th century South China was only loosely under state control, and it wasn’t just the Taipings.

  6. To develop the “overpopulation” angle.

    What the Vikings, Goths, Mongols, et al were overpopulated with was leaders or elites. Their societies were too small and unproductive to be able to maintain many positions of wealth and authority, but (in part because of their loose political structure) they had a large number of aspirants for high positions. One thing this led to was a high level of internal violence. Another was a high level of external aggression. (Many empires were founded and/or expanded by exiles, or by refugees fleeing their victorious enemies back home.)

    The other factor that has to be explained is why these people won. One answer is that they often didn’t. Many barbarian peoples just disintegrated after unsuccessful attacks and have been simply forgotten. But for them to win as often as they did, they needed to have a considerable surplus of big strong men, which means that at least the top segment of the population had to have been well fed. (I have no idea whether the lower orders in these societies suffered stunting from malnutirtion. In Mongol society I doubt it, all Mongols between 15 and 50 were in the army.)

    And again, maybe thinning by violence prevented Malthusian stunting.

    Another equation to consider: militarized peoples minimize the down side of war and emphasize the glory, but everyone knows deep down that being hacked to death is a very bad thing. The combination of the promise of wealth and power and the absence of any alternative to war creates men dedicated to warfare, but there have to be some kind of equation where the lowered realistic hope of plunder, and the increased possibility of a tolerable non-military alternative leads peoples to abandon aggression.

  7. John said:
    they needed to have a considerable surplus of big strong men, which means that at least the top segment of the population had to have been well fed.

    I remember reading somewhere that the Europeans considered the Mongols to be larger and more physically powerful than them. This may be because the Mongols lived a tough physical lifestyle and subsisted on raw meat – stored under their saddle – mostly. Of course it could also be ancient propaganda for why they consistently lost most battles to them.

  8. The defeated usually overestimated the numbers in the Mongol armies, partly as an excuse and partly because Mongols had systematic ways of making their armies seem bigger.

    Mongols bodies were specialized to archery and horseback riding, not necessarily genetically but certainly by lifelong exercise patterns. They didn’t walk if they could help it but had very powerful upper bodies, according to report.

  9. I think that the economics of conquering Germany may have not worked out for Rome in part because the kinds of foods that were produced there were too perishable. I suppose barley could have been exacted as tribute, but I don’t think they could grow much wheat (which keeps better), and the Alps would stop anyone who wanted to do a cattle drive to Rome :-). And dairy products wouldn’t have been too portable either.

  10. This was essentially David Ricardo’s viewpoint.

    As we say in England, the upper classes live on the “fat of the land”.

    It’s a point of view with many problems though.

  11. “in very few regions was there total replacement, even in what became England where cultural obliteration of the Romano-British heritage was nearly total.”

    In his book “Our Magnificent Bastard Tongue”, John McWhorter makes the same argument on linguistic grounds. The Anglo-Saxon invaders probably did not replace the local Celts, who may very well have even remained the majority. They had to adopt English, though, first as a second language. Eventually, English became their first language, but not before it got mangled in their Celtic mouth, acquiring quirks like, for example, the weird use of auxiliary verbs, which is unique among Germanic languages.

  12. “little motive for peasants to move, as there was little difference in quality of life from locale to locale”

    I disagree. A peasant living under a stable landowner might grumble at the tax he has to pay; but at least he can predict the amount, and to whom he must pay it. That peasant can plan ahead, like he plans for the harvest and for the winter season.

    But a peasant who is stuck in a war-zone does not know who is going to ride up to his doorstep the next month. He cannot even be sure that this rider will respect his religion or speak his language.

    The Latin-speaking peasants in the Balkans seem to have hunkered down into defensible valleys (the Romanians) or else fled the outposts entirely (ending up in Dalmatia, or even Italy). As for the German-speaking peasants, I don’t think they’d have had any choice but to follow their landowners wherever they went. When the whole world has gone Mad Max, safety becomes the most precious commodity.

  13. Any plans to set up the old open thread here on the new carrier? Here’s an off-topic find.

    The renowned fieldwork evolutionist, Moller, has a new paper on the behavioral psychometrics, loosely speaking, of barn swallows. He says it is an almost completely novel kind of work. I read only the abstract:

    [...] Brain mass of the barn swallow Hirundo rustica was strongly predicted by external head volume, explaining 99.5% of the variance, allowing for repeatable estimates of head volume as a reflection of brain size. [...] In a 2 years study of 501 individual adult barn swallows, I showed that head volume differed between sexes and age classes, with yearlings having smaller and more variable heads than older individuals, and females having smaller and more variable heads than males. Large head volume was not a consequence of large body size, which was a poor predictor of head volume. Birds with large heads arrived early from spring migration, independent of sex and age, indicating that migratory performance may have an important cognitive component. Head volume significantly predicted capture date and recapture probability, suggesting that head volume is related to learning ability, although morphological traits such as wing length, aspect ratio and wing area were unimportant predictors. Intensity of defence of offspring increased with head volume in females, but not in males. Barn swallows with large heads aggregated in large colonies, suggesting that individuals with large heads were more common in socially complex environments. [...]

Leave a Reply