Collections: Rome: Decline and Fall? Part III: Things

This is the third and final part (I, II, III) of our series tackling the complicated and still debated question of ‘how bad was the fall of Rome (in the West)?’ In our first part, we looked at the question through the prism of ‘words’ – language, culture, religion and literature. There we found a lot of evidence for continuity, supporting the position advanced by the ‘change and continuity’ school that the collapse of Roman political authority didn’t have catastrophic effects in other parts of society. In the second part, we looked at ‘institutions’ – politics, cities and organized religion. Here the picture was far more complex. On the one hand, the states that emerged in the West were successors of the Western Roman Empire and carried with them some of its traditions, but on the other hand they were much weaker, less effective and more fragmented states, with vastly reduced state capacity. While the institutional Church, a distinctly Roman institution by the early fifth century, survived and even thrived after the collapse of Roman political authority, the cities and vibrant urban culture which had defined not merely Roman but the broader sweep of Mediterranean antiquity collapsed. At the same time, we noted that in the East, there was far more continuity and for far longer.

This week then, we’re going to turn to ‘things’ – economics and demographics (which is also going to include a brief discussion of popular literacy). In my own view, this is the decisive part of the ‘fall of Rome’ question, because these are the areas in which we can get a sense of what the experience of the collapse of Roman authority was like for the vast majority of people in the Roman world who do not write to us, who were not rich or powerful and who are thus very difficult to see historically. After all, even if the collapse of Roman political authority was a neutral or even potentially beneficial experience for the elite stratum at the top of society – and it is not clear that it was, mind you; those elites themselves that write to us certainly did not think so – if it was catastrophically bad for the non-elite population, their experience utterly swamps the elite experience by sheer dint of numbers.

And as those of you who have noticed the trend in how this series is organized may have already guessed, it was catastrophically bad. Buckle up folks, it is all downhill from here.

But first, at the most awkward possible moment – as I noted earlier, this topic was one voted on by the members of the ACOUP Senate, patrons at the Patres et Matres Conscripti level. You too could support this effort on Patreon, but I am just as happy if you share what you are reading here; I rely entirely on word-of-mouth to reach my audience with this public scholarship project. And if you want updates whenever a new post appears, you can click below for email updates or follow me on twitter (@BretDevereaux) for updates as to new posts as well as my occasional ancient history, foreign policy or military history musings.

The Revenge of the Archaeologists

Before we dive into the evidence, I want to speak briefly to the nature of the evidence for these topics. ‘Historian’ is often an odd sort of field because while there is a core discipline and skill set that basically all historians are going to have (focused on reading texts critically and assessing arguments and evidence), beyond this almost all historians end up acquiring other skill sets, often from other fields, depending on what they are investigating. I, for instance, work on military history and so I need to have some mastery of military theory, whereas an intellectual historian might instead have some training in philosophy.

It is thus relevant that over the past half-century or so, it has so happened that effectively all ancient historians have had to develop a strong grasp of archaeological data; we don’t all necessarily learn to do the excavation work, of course (that’s what archaeologists do), but pretty much all ancient historians at this point are going to have to be able to read a site or artifact report as well as have a good theoretical grasp of what kinds of questions archaeology can be used to answer and how it can be used to answer those questions. This happened in ancient history in particular for two reasons: first, archaeology was a field effectively invented to better understand the classical past (which is now of course also used to understand the past in other periods and places) so it has been at work the longest there, but also because the sources for ancient history are so few. As I like to say, the problem for the modern historian is taking a sip of meaning from the fire-hose of evidence they have; but the challenge of an ancient historian is finding water in the desert. Archaeological data was a sudden, working well in that desert and much of the last two decades of ancient history has been built around it. Other fields of history are still processing their much larger quantity of texts; why dig so deep a well when you live next to a running river?

The result, in ancient history, has been what I tend to refer to as ‘the revenge of the archaeologists.’ Not, mind you, revenge on medievalists, but in fact revenge on a very specific ancient historian and classicist, Moses Finley. Moses Finley was, from the 1950s to the 1980s, one of the most prominent classicists and his work touched on many fields, including the study of the ancient economy. Finley, writing in the 1960s was generally skeptical of the ability of archaeology to provide useful answers about the ancient economy (he preferred to understand the question by probing the mentalities1 of the Greek and Roman elite). Archaeology, Finley thought, was frequently over-interpreted and could never give a representative sample anyway; as he quipped in his 1965 article “Technical Innovation and Economic Progress in the Ancient World,” “we are too often victims of that great curse of archaeology, the indestructibility of pots,” a line for which, as far as I can tell, he is still quite unforgiven by some archaeologists.

As if in response, the archaeologists have spent the subsequent almost-six-decades proving again and again the tremendous value of their discipline by, among other things, utterly burying Finley’s The Ancient Economy (1973) under a mountain of archaeological data. It turns out the mentalités of aristocrats who largely hated merchants were not a good barometer of the activities of those merchants.

Via Wikipedia Commons, a collection of Roman amphorae – bulk storage containers used for transporting liquids by ship – displayed at the Market of Trajan in Rome. The durable nature of ceramics mean that while pots like these may shatter, they survive either whole or in fragments to be recovered by us in huge numbers. While Finley critiqued the over-reading of a small number of pot sherds, as the sample size of finds grows – and it has grown massively – the ability to derive useful conclusions from it grows as well.

But you may now guess how this is going to play out in the discussion of Late Antiquity. The ancient historians come to the question ready to think in archaeological terms and ask what archaeological data can do to clear up these questions. Scholars of Late Antiquity trained as medievalists on the other hand, may or may not be well versed in archaeological methods or data (to be clear, some medievalists very much are versed, including prominent voices on the ‘change and continuity’ side of this debate! But it is also very possible to be a ‘pure text’ medievalist in a way that I don’t think I know a ‘pure text’ ancient historian younger than sixty) because their field has not been forced, by dint of the paucity of sources, to revolve so heavily around archaeological data and because the archaeological data on the Middle Ages is not yet as voluminous as that on Classical Antiquity.

As I noted in the first post, beginning in the 1970s, what James O’Donnell calls the ‘reformation in Late Antique studies’ launched a long overdue reassessment of Late Antiquity and the impact of the Fall of Rome – what we’ve called the ‘change and continuity’ argument. I bring up all of this to note that the ‘counter-reformation’ – what we’re calling the ‘decline and fall’ argument – that really emerges beginning in the 90s is in many ways an extension of the ‘revenge of the archaeologists’ in Classical studies (and especially the ancient economy) into the field of Late Antiquity. Indeed some of the scholars are the same (e.g. Willem Jongman) and many of them enter the debate on Late Antiquity as an extension of the debate about the Roman economy (in part demanding that ‘change and continuity’ Late Antique scholars acknowledge things now generally considered ‘proved’ by ancient historians about the earlier Roman economy).

In my own experience, particularly in more informal conversations, the methodological difference that interaction creates between ancient historians – for whom it has long been almost entirely settled that in a ‘fight’ between archaeological evidence and effectively any other kind, the archaeological evidence ‘wins’2 – and medievalists for whom archaeology is a much less central part of their method (in part because their textual sources are more extensive) can lead to situations where the two sides of the debate talk past each other.

But when it comes to questions of demographics, economics and the conditions of life for the sort of people who rarely figure in our sources, archaeological evidence – although it is often incomplete and hard to interpret – offers the possibility of decisive answers to questions that otherwise would have to live entirely within the realm of speculation.3


We can start with the question of population. As we’ve seen, the evidence – largely archaeological evidence, by the by (Liebeschuetz thus fits with many other historians in the ‘decline and fall’ counter-reformation in relying heavily on archaeological data) – suggests that urban centers declined markedly beginning in the fourth century, with that decline accelerating as the empire crumbled. That of course raises the fairly obvious question: where did all of the people go? One possible theory is that the population mostly ruralized, moving out of the city and into the countryside. That might even suggest a positive change, if one accepts the view that ancient cities were mostly ‘consumer’ cities which didn’t produce much value but instead survived off of taxes and rents extracted from the countryside.4 In that view, the decline of cities could simply be a product of the collapse of systems of exploitation as the political order which maintained them weakened.

It’s a plausible theory and the only problem with it is that it doesn’t appear to have actually happened.

Here the key archaeological method is what is called ‘field survey.’ While readers are probably more familiar with the intensive excavation work done at famous sites like Pompeii or Vindolanda, one tool archaeologists have to study the past is to survey large areas, sometimes by air, sometimes by on foot, sometimes with ground penetrating radar, in an effort to map out larger scale settlement patterns in the past than would be possible by labor-intensive single-site excavation work. Dateable remains (pottery most often) allow for archaeologists to get a rough sense of the dates in which sites were inhabited and in some cases building remains and the like can give some sense of what kind of settlement was present. The ‘error-bars’ on some of this data can of course be large, but they offer a tool for tracking long-term changes in land use patterns. On the flip side, these sorts of studies really become valuable only when you have a lot of them to create a robust data-set over a fairly large area that lets you adjust for purely local patterns and distortions. Fortunately in much of the former Western Roman Empire and especially in Roman Italy (where these studies are very important for the study of Roman demography and agriculture) we’ve hit the tipping point where there is enough archaeological data to begin reaching for conclusions.

Now there is an immediate difficulty with using this kind of evidence, which is that for reasons we’ll get to in a moment (though they are reasons that tend to also be bad for the ‘change and continuity’ argument), we have a major confounding variable here: site visibility. Our ability to see a site, archaeologically, is heavily dependent on factors like building material and the quantity of imperishable goods (especially pottery) that people are using. For reasons we’ll get to, compared to, say, second century AD communities, sixth century AD communities tended to build their buildings in far more perishable (and thus less visible) materials (like wood) and also tended to use a lot less imperishable household goods. Consequently, it is substantially harder to see a sixth century village than it is to see a second century villa.

Nevertheless, the decline is so marked and so consistent as to strongly suggest there is something real here. R.P. Duncan Jones (in “Economic Change and the Transition to Late Antiquity” in Swain and Edwards (eds) Approaching Late Antiquity (2006)) assembles some of the site data from around the empire; there is unsurprisingly a lot of regional variation (with some regions, like Syria, actually moving against trend), but in the western Empire (except N. Africa; decline there comes later) the trend is fairly clear, with site numbers declining (often drastically by half or more) beginning in the late third or fourth centuries. Bryan Ward-Perkins in The Fall of Rome notes a field study outside of Rome in which the number of sites declines by three quarters. Site data accumulated like this isn’t often very chronologically precise, so we’re dealing with centuries, not decades, but the clear trend suggests rural population decline, not an urban population ruralizing. To be visible to us in this way, the decline must have been quite severe.

To give a sense of the scale of the decline, here is an abbreviated version of a chart from Bruce Friar’s “Demography” chapter in the second edition of the Cambridge Ancient History, which breaks down the estimated population of the Roman Empire by region and adds the dates when each of those regions got back to their Roman era population:

Region:Population in 164AD (in millions):% Increase from 14 to 164 ADYear When Roman Population Reattained:
Greater Syria4.811.61875
Sardinia & Corsica0.51250
The Maghreb6.585.71825
Iberian Peninsula7.550.01200
Gaul & Germany9.055.21200
Danube Region4.042.21100
Chart from Frier, “Demography” CAH2 XI (2000), 814. Some of these figures would likely see some revision today, mostly downward revisions of growth combined with upward revisions in population reflecting a somewhat (but generally not massively) higher estimated pre-Roman population.
Note that the decline in the East was, as noted last time, both later and generally slower. The reason for the later times to reattain Roman population here in many cases is that the major medieval Islamic population centers were further East (e.g. Baghdad under the Abbasids) placing them outside the traditional bounds of the Roman Empire, but also that the Roman East was much more urbanized and densely populated compared to its land area than the Roman West in the second century (or at any time during the Roman Period) so the ‘population to attain’ bar on the East was much higher. After all, the cities of places like Syria or Egypt were in many cases centuries or even millennia old when the Romans showed up.

Now the long times there to regain the Roman population can be a bit deceiving (and are very approximate). For reasons we’ll get into shortly, population growth from 600 to 900 or so in Europe was very low, so the issue here isn’t that the decline was so steep that it took many centuries to recover from, but rather that the decline was from a high population equilibrium to a low population equilibrium, both of which were, under their own conditions, stable (if that is confusing, don’t worry, we’ll delve more into it in a moment). Second, the apparent gap between places that ‘caught up’ before 1300 and those that ‘caught up’ after it is smaller than it looks, because of course the mid-1300s represent a massive population discontinuity over the entire broader Mediterranean world due to the Black Death such that a lot of those places ‘catching up’ in the 1200s probably fell behind again due to the plague and then caught up again in the 1400s or early 1500s.

But this now raises two related questions: first, why did population decline so sharply and second, what was the impact on quality of life that resulted? The old answer to the first question was of course ‘the barbarians killed everyone’ but as we’ve seen, while the fifth century was a violent time, the violent discontinuities were not that extreme. Surely the violence of the period has something to do with some of this declining population, but as noted, the underlying population (with their language and religion) didn’t much change (and the raw number of ‘barbarians’ coming over the frontier was, in demographic terms, fairly small). Most of those Roman cities decayed, rather than being burned. But if the ‘barbarians’ didn’t kill everyone, what did and why did that somehow have a negative impact on the survivors? The answers to these two questions are actually linked in that they depend on the same evidence, so that is where we will go next.

Living Standards

And we can start with another fairly common theory about this period – ‘perhaps the decline in exploitative cities and population causes life to get better.’ This isn’t as crazy as it seems! The Black Death, which we’ve just mentioned, is an obvious (and of course for any medievalist, readily available) analogy. The Black Death may have killed something like a third of the population of medieval Europe in the mid-1300s. Of course that is very bad! But one of the paradoxes of the Black Death is that in the aftermath of it, living conditions for the survivors clearly improved! The population growth of the previous centuries had meant bringing more marginal, less productive land under cultivation to support that population, which had reduced the per-farmer efficiency of agriculture even as total production grew, which had in turn meant that most farmers lived closer to the subsistence line and thus were poorer, their labor less valued. Killing a third of them thus made the labor of the remaining two-thirds much more valuable. Marginal land fell out of production as farmers focused on the best land, which improved production per-farmer (even as total, aggregate production fell) resulting in higher standards of living for the survivors. This is a classic ‘Malthusian’ interaction and the evidence for the period is robust enough that we can be quite sure it happened.

So it was perfectly reasonable for some Late Antique scholars to suppose that the same thing, in some form, might have happened in Late Antiquity, especially after the collapse of the Roman Empire. After all, the collapse of state power meant that the increasingly heavy burden of Roman taxation will have vanished too, replaced by the less effective and thus less extractive polities that followed. And the cities, with their urban elite extracting rents and – in that ‘consumer city’ model of the economy – providing not much of value in return were sharply reduced in number, reducing the burden on the peasantry to support them.

It is, again, a charming and seductively plausible argument marred only by the regrettable flaw that it doesn’t seem to have happened.

Instead, to jump to the end, what the evidence – again, here mostly archaeological evidence used in a statistics-driven way – suggests is that what we are seeing is that average, per-capita production declined, resulting in a real decline in living standards including nutrition, which resulted in population decline. That population decline was thus the physical expression of a lot of real misery: starvation perhaps, but in most cases more likely heightened infant and maternal mortality as a result of malnourishment. This is, by the by, an example of a simple Malthusian interaction breaking, because the classic Malthusian logic assumes that agricultural production is primarily a function of land – but as it turns out, a lot about how you farm that land, even without major changes in farming technology, can substantially change the efficiency of your agricultural economy and thus the population it could support. In essence, the decline and fall of the Roman Empire caused the carrying capacity of the Mediterranean World, and especially western Europe, to decline, leading to the population declining to follow in step – which is to be clear, an incredibly bloodless way to describe a period of real, sharp human misery.

The evidence for this decline, initially slow in coming, is now quite substantial; Willem Jongman assembles perhaps the most complete set of it in “Gibbon was Right: The Decline and Fall of the Roman Economy” in Crisis and the Roman Empire (2007)5. Jongman considers evidence for coin minting (through atmospheric lead records contained in ice cores), trade (via dated shipwrecks), meat consumption (via bone assemblages) and basic nutrition (via height calculated through femur length in dated human remains), inter alia and finds the same or similar patterns in each indicator. To take the most direct indicator of nutrition effect son people, mean femur length rises over the early Roman Empire, falls slightly in the late second and early third century, rises again but not quite so high over the fourth century, and then utterly collapses in the fifth:

Graph after W.M. Jongman “Gibbon was Right: The Decline and Fall of the Roman Economy” in Crisis and the Roman Empire (2007), 194, graph 7. Measurements are in centimeters.

(Post-publication note: It turns out the evidence for biological health is a lot more complex. For those looking to see the latest on that, check out W.M. Jongman, J.P.A.M. Jacobs, and G.M. Klein Goldewijk, “Health and wealth in the Roman Empire” in Economics and Human Biology (2019). The upshot is that with a larger dataset and a more sophisticated approach to it, they show that in fact the ‘biological standard of living’ in the Roman Empire does seem to have declined as the population grew and then improved as the population declined, the reverse of the pattern of the data above. That said, they note that this is despite all of the other indicators of well-being, including diet indicators, moving the other way, meaning that even as diet improved, height fell. They attribute the shift to a ‘health cost’ mostly due to disease prevalence from the denser population and greater degree of trade, a direct tradeoff for the greater material wealth of the Roman world. One factor they consider but dismiss is the possibility that the data is being shaped by changing burial customs, because elite Romans tended to cremate (which would remove their bones from the sample) but that pre-Roman populations and later Christians, including elites, inhumed. That might mean that our Roman period sample is depressed because the wealthier parts of society have been systematically removed. They dismiss this on the grounds that it would require assuming that the impact of social class was greater than that of period, but I don’t see why it wouldn’t have been given that we’re talking about baseline nutrition impacts and the basic math of subsistence doesn’t change that much over time. I’d assume that a wealthy Roman elite or 9th century noble probably ate more meat than a peasant at any point in pre-modern history, so the possibility that elite burials have been systematically excluded from the same is a serious one and potentially confounds the entire dataset.)

That evidence is in turn largely consistent (with some noise, because data of this sort is always noisy) with our evidence for diet. Jongman presents graphs, for instance, of animal bone deposits, a good indication of the degree to which people were eating meat in a society. Meat is, of course, very high in protein (and also delicious) but it is also expensive – you can feed a lot more people on grain than meat, so the level of meat consumption in a pre-modern society is a good indicator of living standards (especially meat consumption among non-elites). In Italy, Jongman notes, animal bone assemblages become progressively more common from 450 BC until reaching a peak around 50 AD. They then decline a bit (but still well elevated from the pre-Roman norm) to a low in the mid-third century (which you will recall was not a great time for the Roman Empire) before rising again to another peak in the fourth and early fifth century, high but lower than the early imperial peak, before collapsing to almost nothing by 650 AD; the decline from 450 AD to 550 AD is the sharpest change in the data at any point. The data for the provinces also fits expectations – in that graph, assemblages sharply rise beginning in 150 BC, peak in 150 AD, fall down to a lower-but-still-high plateau in the fourth century, and then collapse down to nearly pre-Roman levels by 450.

If anything, this may understate the problem because the animals here are not all the same either. As Bryan Ward-Perkins notes (op cit), based on archaeological remains, the average height of an iron-age (pre-Roman) cow was 115.5cm. In the Roman period, this rises to 120cm, but in the early Middle Ages, collapses down to 112cm. Now that is of course both bad for the people eating the cows, but it is also bad for what it says about the farmers raising the cows – chances are you are shortchanging the nutrition of your (very valuable) cows because you yourself are experiencing significant food shortfalls. The cows are getting smaller because this society is getting poorer – and not merely the elites are getting poorer; everyone is getting poorer. At this point the evidence is fairly clear on this point: living standards declines across the entire spectrum of Late Roman and early medieval society. Not only were the elites of 600 AD poorer than the elites of 400 AD (and much poorer than the elites of 200AD), so too were the peasants.

But if it wasn’t because the ‘barbarians’ burned everything – because again, they didn’t – how could this happen?

From High Equilibrium to Low Equilibrium

And here it is important to understand how you can have economic growth happen in a system where the basic inputs of production – in this case land and farming technology – haven’t really changed much. I should note of course that farming technology was improving over this period (albeit slowly; a good summary on this here), but the change here isn’t due to farming technology and in any event farming technology doesn’t seem to have declined in the transition to the Middle Ages. The issue is changes in efficiency.

We’ve actually talked about how agricultural systems can operate at multiple different equilibria for a given set of land and technology inputs before. Peasant farmers operating in isolation from effective markets tend to diversity their farming, sacrificing efficiency for a greater margin of safety. But if those same farmers produced for the market, either because they were rich and thus basic survival was less a concern or because the tax man forced them to to pay taxes or simply because markets were reliable and efficient enough to make it a safe enough bet, they could specialize on the crops that were most productive in their area. That would be likely to produce regional centers of production, but of course that can only work if long distance trade in bulk staples is possible and cheap enough so that excess production in one region can be moved to the next.

And because basic capital scarcity, typically in things like draft animals and manure, are factors in all of this, its possible to create a virtuous cycle where these newly more efficient, market oriented-farms benefiting from capital advantage can invest more in things like draft animals and manure (because the agricultural surplus to feed the animals exists), realizing improvements in production. So long as the system works, a society like that could operate at a higher stable equilibrium, with a higher quality of life, and a larger per-capita agricultural surplus which would enable more non-farmers (some of whom might be idle elites but many of whom would be productive workers – like artisans – whose labors might feed back into the living quality of the peasantry).

A system like that, where the gains in per-capita production were the product of trade networks leveraging competitive advantage and the accumulation of agricultural capital (animals, but also potentially mills, olive presses, that sort of thing) could potentially be very fragile. Disruptions to trade or a failure to maintain that agricultural capital through a crisis could lead a society to fall back down to a lower equilibrium. Essentially, because the growth of this system is reliant primarily on institutions rather than technologies, should those institutions decline – as institutions do – the system will decline with them.

In practice that means that the conclusions one draws about how bad the economic decline of the fourth and fifth centuries depend largely on how one understands the economic situation of the earlier Roman imperial period. And so it is perhaps unsurprising that the ‘change and continuity’ argument tends to assume a fairly unimpressive Roman economy. As Peter Brown puts it, the extent of the poverty of sixth and seventh century western Europe “implies, furthermore, that we may have exaggerated the overall wealth and sophistication of western Europe under the Roman empire. It was not as highly developed a society as we might conclude from its surviving monuments…there had never been a commercial unity to be destroyed.”6

This is largely a pre-Revenge of the Archaeologists vision, consistent with Finley’s reaction to earlier writing on the Roman economy (he was reacting against a yet older scholar, M. Rostovtzeff) and his later defenders (a school of thought about the Roman economy called ‘primitivism’ because it holds that the Roman economy was relatively primitive in structure). Indeed, Brown’s description of the Late Roman economy as a “crude but vigorous pump which had ensured the circulation of goods in an otherwise primitive economy” is a direct and colorful description of the model of the Roman economy set out by Keith Hopkins in the 1980s and 1990s, which expressly set out to set out a model of the Roman economy that could accommodate the growth that was, by then, increasingly archaeologically obvious, without wholly abandoning the primitivist model of the economy.7

The problem, of course, is that this is precisely the view of the ancient economy that the Revenge of the Archaeologists has largely overturned. The primitivist view assumes most trade is in luxuries, but in fact we have evidence for high volume trade (mostly by sea, of course) in bulk staples, much of it apparently outside of state action. Seaborne trade, especially during the two first centuries and the second century, turns out to have taken place at a much larger scale (in particular using lots of much larger ships). Most importantly, to judge by shipwreck data (on the assumption that merchant ship design, which changed little over this period, meant that the odds of ships sinking remained relatively constant), the volume of ships on the sea changed massively (the thing to read here is A. Wilson, “Developments in Mediterranean shipping and maritime trade from the Hellenistic period to AD 1000” in Maritime Archaeology and Ancient Trade in the Mediterranean (2011)).

Graph after Fig. 2.5 from A. Wilson, “Developments in Mediterranean shipping and maritime trade from the Hellenistic period to AD 1000” in Maritime Archaeology and Ancient Trade in the Mediterranean (2011).

Now there are some caveats that this graph needs too – as always, archaeological data is tricky. In this case, as Wilson notes, there is a major impact in our graph in the second century when barrels begin to displace amphora for moving liquid goods. Amphorae, being pots, survive to be found on the ocean floor while wooden barrels often decay away. Wilson attempts to control for this by looking at only shipwrecks with stone cargoes; that graph shows shipping remaining steady at first century levels out through roughly 400 AD before dropping off then. We should also note that there is a regional bias in this data; the Western Mediterranean is both shallower (so easier to find wrecks) and sees a lot more divers (both archaeological and recreational). Justin Leidwanger breaks up the Eastern and Western Mediterranean graphs in his book Roman Seas: A Maritime Archaeology of Eastern Mediterranean Economies (2020). What that effort shows is that in the Eastern Mediterranean, the shipwreck numbers decline modestly in the fourth and fifth century, but recover in the sixth before declining more strongly after 650.

What is notable then here in both the Eastern and Western Mediterranean is that the scale of trade seems to follow the territorial integrity of the Roman Empire – where it stayed intact longer in the East, trade continued longer in the East, but the disruption of the loss of Egypt and Syria, while not as extreme as the decline in the West, still has a notable impact.

Meanwhile, evidence for long distance trade based on the movement of the goods themselves or containers has also improved dramatically in the last several decades. As Ward-Perkins notes (op. cit., 97-8) pottery is both very durable but also frequently traceable back to its point of origin either by style, composition or – for bulk storage vessels – tituli picti, stamped or inscribed production labels. So we can, for instance, note that a production center at la Graufesenque, in Aveyron in the south of France, which made ceramic tableware, sent products as far north as Scotland, as far south as Libya, east into Italy and west all the way to the Atlantic coast of Portugal. These sorts of products co-existed with more locally produced versions and imported pottery is not restricted only to elite contexts either; at least some of this was available to regular people.

Via Wikipedia, an amphora (dressel 20 type) with a titulus pictus on it, along with several examples of potters stamps. All were found at Monte Testaccio, a large hill in Rome composed entirely of roughly 53 million discarded shipping amphora (mostly used to ship olive oil) from the first century BC to the third century AD. I have tried in this series to avoid talking too much about Rome itself because Rome was such an unusual place in the Roman Empire, but the scale of trade to Rome was enormous and almost certainly unmatched in Europe or the broader Mediterranean world until the early modern period.

All of that trade demanded new coinage to facilitate it and indeed we can be fairly certain that the money supply expanded dramatically in this period. Silver smelting – to produce coinage (or other silver goods) – releases small amounts of lead into the atmosphere which can be detected and measured when trapped in the ice in places like Greenland. You can thus take the atmospheric lead content as a fairly direct barometer of how many new coins are being minted from freshly mined silver (rather than remelted coins) at any given time). And the graph looks like this:

Via Wikipedia. Again this graph is wrong about one thing – this isn’t World lead production, but rather because the ice cores are from Greenland, likely more strongly reflects the production in the broader Mediterranean world which for our purposes is just fine.

It also seems like the Roman Empire saw heightened amounts and quality of construction. Jongman (op. cit.) notes a study of dated timber remains in western Germany and the clear boom during the principate and then subsequent fall was readily visible in the data (though by 500 or so, things are moving again, but at less than half the level of 100). Meanwhile, the precise reason that it is so much easier to find Roman period sites than early medieval sites in field surveys is because Roman sites both have more pottery but especially are more likely to have buildings constructed in less perishable materials, particularly with ceramic roof tiles rather than something perishable like wood slats or thatch. And we have of course already noted the explosion of expensive public civic building under the early Roman Empire, the slow decay of those buildings in the late Empire and their near-complete disappearance after 450.

It is worth briefly returning to the topic of literacy again in light of all of this because one might reasonable also ask what both the surge in economic well-being meant for literacy and what its collapse meant as well. If the Revenge of the Archaeologists on the Roman economy is in some ways complete or at least mature, it seems like the same process when it comes to literacy is only in its early stages. Nevertheless, the emerging volume of evidence for what Wallace-Hadrill8 refers to as the “trivialization of writing” is beginning to suggest that literacy rates may have been rather higher in the Roman Empire than we normally thought, at least in the urban population. Now we shouldn’t overstate this; we are still at best talking about a fairly small minority of the population and the level of individual proficiency could be very low among the common folk who had no formal training. A good example of this ‘trivilization’ are a selection of graffiti shopping lists from Pompeii discussed here by Kim Bowes; what’s striking about them is that the goods listed are not prestige goods but rather appear to reflect the diet of a not-quite-poor but certainly-not-rich Roman family: mostly bread, but with a bit of ‘splurge’ foods, but nothing exotic. As Bowes notes, scratching these lists into a wall implies two audiences, a writer and a reader and assumes that both could understand what was written there.

Comparing medieval literacy can be difficult, in part because what our sources consider literacy may vary: is a king that can read in French but not in Latin literate in the mind of a clerical writer? Nevertheless, it seems clear that the Western Europe that emerged from the collapse of the Roman Empire was substantially less literate than the Roman world implied by our Pompeii graffiti. Even low literacy estimates for the Roman Empire9 suggest rates somewhere between 10 and 30%, quite a bit higher at the top end of the range than similar estimates for the early Middle Ages, which tend to use 20% as an upper-bound. Far from being ‘trivialized’ by common possession, literacy in the early Middle Ages was to a significantly degree sacralized by its strong association with the Church (much as some of the earliest Bronze Age writing systems seem to have been, for a time, functionally sacred – though it is possible to overstate this and we certainly have Bronze Age writing of non-sacred nature, especially but not exclusively, in the Late Bronze Age). At the same time, it seems fairly clear that literacy among the elite and the urban classes wasn’t stagnant for the whole of the Middle Ages but began rising again, seemingly in lockstep with improving economic conditions and the growth of towns. Cross-period comparisons here are impossible to make with any precision, but if I had to guess, I would suppose that popular literacy by 1200 was as high as it had been in 100 and by 1500 was substantially higher in most of Western Europe (even excepting out the Low Countries and Britain which seem to have been, by the early modern period, somewhat elevated in literacy rates compared to the rest of Europe).

Via Wikipedia, some examples of Roman graffiti from Pompeii, in this case all riffs off of the opening line of the Aeneid. The bottom one, found outside the shop of Fabius Ululitremulus (a fuller) reads, “Of fullers and an owl, I sing, not arms and a man.” It’s a striking use of writing to comment on what was a fairly humble profession.

In short, across a wide range of indicators, the world of 550 was meaningfully poorer than the world of 350 and substantially poorer than the world of 150. People had less access to imported goods, less access to household durables, they built fewer buildings and had less to eat and as a result population fell. We chafe at periods of relative economic stagnation, but this appears by all measures to have been a period of profound and substantial economic decline, a decline which then persisted for generations.

What Happened?

When looking at this sort of data, there is a lot of noise, but a set of basic trends emerge that repeat across all sorts of different archaeological data-sets: a peak of economic activity in the two first centuries that extends into the second century AD, followed by a significant but not catastrophic decline in the third century to a new, apparently stable but lower plateau in the fourth century that is still substantially elevated above the pre- or post-Roman norms, followed by a sharp collapse in the fifth century and a period of apparent depression extending, in the West, from the sixth century to the eighth.

And while it has long been supposed that all of that Roman prosperity was captured by the Roman elite – who to be clear did capture a lot of it and became truly absurdly wealthy in the process as Rome was a terribly unequal society – it is clear from the archaeological evidence we have now that it seems like a fairly broad cut of Roman social classes, reaching down into the peasantry, experienced at least some meaningful elevation in standards of living during this period. Until it didn’t and living standards collapsed down near to – and in some cases below (though in most cases, it seems to me, slightly above) – pre-Roman iron age norms.

In light of that evidence, it seems to me that the insistence that the Roman Empire “had never done as much for its citizens as our exalted image of Rome might lead us to expect” (Brown, op. cit., third edition, in Year of Our Lord 2013) has to be abandoned or at the very least substantially modified.

So what happened? Discussing causes is always a matter of interpretation, but a number of causes for the rise in quality of life under the Roman Empire have been proposed and many of them could also explain the collapse of that improvement. The Roman Empire, quite apart from any intentionally official policy, was good for trade. Rome established what was effectively a customs union (albeit with multiple tariff zones) across the Mediterranean, making it easier to move goods long distances over the sea. The Romans also engaged in piracy suppression and perhaps most importantly, substantially reduced the amount of large scale warfare on the sea. One official policy which probably mattered here was the currency; the pre-Roman Mediterranean was a mess of different currency systems along with societies not yet extensively using coinage. The Romans standardized this, subsuming all of those local currencies into a single Roman currency system and at the same time Roman taxation fairly clearly had the effect of monetizing much of the Roman economy (that is, forcing the adoption of coinage) all of which will have lowered transaction costs.

Those lower transaction costs could, of course, reduce deadweight loss on the economy, but they also enabled a lot of interprovincial trade in bulk staples, enabling a limited (and we must stress that word) degree of agricultural regional specialization. And it is clear that happened! It is at this point clear that regions – like Baetica in Spain – or towns (like Leptiminus in North Africa) specialized in this new economy (in both cases in olives and olive oil). It’s also clear in some areas (this is very clear in Roman Britain, for instance) that small scale production of manufactured goods (e.g. pottery) moved into the cities and occurred on a substantially larger scale, likely experiencing economies of scale in the process. At the same time, the Roman economy, we are learning, was more financially sophisticated (particularly at its higher reaches) and allowed for more mobility not merely of goods but also of labor, though we should be careful here not to get carried away (and we do need to remember that a good portion of that labor mobility, though not, apparently, all of it, was accompanied by a lot of real human suffering through the institution of slavery). At the same time, interconnectedness did allow new technological innovations – watermills, new olive presses, etc. – to spread through the empire more rapidly than they might otherwise have done.

All of these factors, trade, coinage, specialization, etc., were probably working together at the same time to one degree or another; many of them depend on each other. All of that functioned to raise per capita production, allowing for both the quality of life and the population to rise in defiance of the normal Malthusian logic whereby if population rises, quality of life (particularly diet) ought to fall. But this system of improvement was fragile, based on the durability of the Roman political systems which created this imperfectly connected market.

And we should note that nearly all of the blows which brought this system down were self-inflicted by the Romans who for their part never seem to have understood the marvelous thing they had created. The Crisis of the Third Century shattered the political unity of that market and disrupted the limited degree of public peace that created it. Rival emperors both before the crisis but increasingly in it also devalued the currency and extracted supplies directly in kind rather than in cash, leading to weakness in the currency system and a progressive demonitization of the economy. This free-fall was to a degree arrested in the fourth century by Diocletian and Constantine, but the top-heavy, bureaucratic administration they created was probably itself a drag on the economy; Diocletian’s currency meddling and price fixing were disasters and Constantine’s efforts to actively reduce labor mobility to aid in collecting taxes couldn’t have helped. That leads to the lower-but-still-elevated fourth century plateau: elevated by the continued existence of semi-unified10 market and the fact that at least the wars of rival emperors tended to be geographically limited and less destructive than the pre-Roman (or post-Roman) norm of endemic warfare. Finally, there is something to the notion that the state-run systems of extraction and redistribution – taxing the farmers in grain to be shipped to the soldiers – may have continued to encourage a degree of specialization and trade; the “crude but vigorous pump” worked to a degree to elevate living standards over the pre-modern agricultural norm.

But that very cycle of usurpers and civil wars – and the decision of those rival emperors to (in Peter Brown’s phrasing, which I love), “bus in” ‘barbarian’ armies to fight each other – led to the slow but steady disintegration of that united political order, the shift to more and more endemic violence and the final collapse both of that semi-unified market and the “crude but vigorous pump” that had in part replaced it. Living standards thus declined back down to the pre-Roman Iron Age norm while at the same time the carrying capacity of the empire also declined down to that norm, leading to what must have been decades of brutal misery as food ran short, malnourished infants died before their time, cities shrank and the world grew poor.

At the same time, it is possible that both disease and climate change were factors here too. The Antonine Plague (c. 165-180) was lethal and disruptive although if plague was the only issue we’d expect a pattern looking more like the Black Death – population decline but living standard improvement. Climate – a shift to a colder, drier climate – may have also been a more major factor and there’s a fair bit of evidence to suggest that the climate became less favorable beginning in the last third century. That could have pressed down farming yields, worsening the both living standards but also driving Roman authorities to tax harder to sustain military operations in the face of declining production.

This also explains why the experiences of different regions of the empire could end up so different. In the East, the loss of connection to the West wasn’t nearly so catastrophic because the Eastern Mediterranean remained connected (in the somewhat weak, Late Roman way). By contrast, an area that was both only recently urbanized and on the periphery of the empire and thus relied on that interconnectedness to sustain its urban centers – like Britain – was hit much harder by the collapse. It is hard not to notice that a fair number of the strongest voices in the ‘decline and fall’ camp are British and one wonders how much it influences their view that Roman Britain was, above and away the worst hit part of the empire in the immediate aftermath (for more on this, Robin Fleming, Britain After Rome: The Fall and Rise 400 to 1070 (2010))

Now at the same time we need to be clear about what this does not mean: it does not mean that the Roman world was on the verge of beginning that heady, nearly vertical upward trend in human prosperity that launches violently out of the end of the early modern period. The collapse of this Roman system was a real tragedy for the people living at the time and for a considerable time thereafter, but what was elevating this system above the pre-modern agricultural norm was not the unrush of technological change but an unusual period of political unity which enabled greater interconnectedness. Moreover, the disintegration of Roman political authority was hardly a sui generis event; interconnected state systems had collapsed before and would collapse later, some of them quite a bit more completely than the Roman one did.


So our knights have dueled and where have we ended up?

Clearly there are some popular notions of this period which have to be rejected outright. The ‘nitwit’ of our duel – the notion that the Middle Ages was some general collapse of ‘progress’ which delayed the course of human development – doesn’t accord with the evidence. The popular perception is that many Roman technologies were lost, but in fact these were fairly few, the most notable being Roman concrete. As noted, what is remarkable about Greek and Roman literature and learning is not how much of it was lost, but how much of it survives to the present. Moreover, the Roman economy was not the durable foundation for a lost early industrial revolution;11 instead it was a delicate clockwork mechanism which could, for a time, haul itself modestly – and only modestly – above pre-modern agrarian norms. When the gears broke, the clockwork stopped and it fell back down.

At the same time, ‘falling back down’ is not the only story of the Middle Ages. I cannot stress this enough, the European Middle Ages were not a stagnant time in Europe or anywhere else. Older scholars, like Rostovtzeff and Gibbon supposed that Europe only reached a Roman level of prosperity in the early modern period, perhaps in the 1500s or 1600s, reasoning from the grandeur of Roman buildings and literature. But a fair look at the economic and demographic history suggests, I think, quite clearly that the ‘crossover’ point is much earlier, well into the Middle Ages. My own rough estimate would be generally around 1100 in most places; state capacity remains lower for longer because the states of Europe were small and fragmented, but one can argue that was a good thing for their long-term development. Moreover, not everything between 476 and 1100 was just ‘recovery’ – some things were new! Speaking of my own expertise, medieval steel-making, especially at its upper end, tends to be quite a bit better than Roman steel-making. Water-mills (which the Romans had) and windmills (which they didn’t) were, by 1100 apparently far more common in Europe than they had been under the Romans.

The collapse of Roman political authority doesn’t represent any sort of clear break in anything we might call ‘Roman civilization’ – on this, Ward-Perkins seems to me to be quite clearly wrong when he terms the fall of the Roman Empire in the West as the ‘Death of a Civilization.’ Latin persisted; Christianity persisted; Roman literature persisted; Roman law persisted; the Roman Empire itself persisted in the East. The claims of Frankish and Gothic kings to be heirs to the legacy of Rome was not an empty one from a cultural standpoint – if Gallo-Romans or Greek-speaking Eastern Romans could be heirs of Rome, why not Latin-speaking Franko-Romans?

At the same time, what we might call the ‘strong’ form of the ‘change and continuity’ position – that essentially nothing of value was lost as the Roman Empire crumbled in the west – doesn’t seem sustainable in light of the evidence either. One understands the impulse for medievalists in particular to defend their field against the nitwit position above by taking this strong version. But it is quite clear that, on the one hand, the collapse of the Roman Empire in the West represented a substantial decline in state capacity, borne out by smaller churches, fewer public buildings, smaller armies, and smaller, less centralized, more fragmented polities and a greater degree of endemic warfare (although those wars were often on a smaller scale). Resource mobilizations that were casual accomplishments for the Romans – the construction of the infrastructure of Roman cities essentially ex nihilo through much of Gaul and Spain – would be flatly out of reach for western European states until the High Middle Ages.

At the same time, it seems fairly clear from the evidence that the collapse of Roman connectedness took a slow economic decline and turned it into a collapse. As I’ve said, what you see here depends on where you look and what you think is the most important; for me – as I’ve noted before – my focus is drawn to the living conditions of the people in a society. From that perspective, the fall of Rome was an unmitigated disaster, a clear (but not total) break with the economic patterns of antiquity which had enabled a measure of prosperity in the Mediterranean world. The world that emerged in the sixth century was one that was substantially poorer, its population brought back in line with its reduced production by decades of grinding misery and shortage.

The ‘Fall of Rome’ has of course become a touchstone in political discourse and that has produced responses to this question which are politically freighted. Peter Brown in the preface to the revised third edition of The Rise of Western Christendom (2013), notes this expressly and it is hard not to feel at least a bit that the desire to push back on, as he puts it, “conventional images of the fall of Rome from which, as we have seen, extremist politicians and demagogues in contemporary Europe had conjured up a toxic discourse based on prejudice and fear” (xxxi). But I don’t think the way to push back on that notion is to play down the economic complexity and success of the Roman Empire – even the Late Roman Empire – in increasing defiance of the fairly clear archaeological evidence so as to pretend that nothing much of value was lost when the empire disappeared.

Instead, I think the stronger point here (and one Peter Brown – lest anyone think I think his work is without merit, which is far from the truth – and also many others make well) is that the collapse of the Roman Empire in the West – while it was a catastrophe for those people living at the time – was less a product of ‘hordes of barbarians’ coming over the frontier (who again, were mostly invited in by Roman leaders looking for advantage in their endless struggles with each other) and instead a product of actors within the political system, within the empire, tearing it apart out of the pursuit of their own interests, deceived by the assumption that something so old could never simply vanish…until it did. The consequences of their decisions and of their failure to recognize the fragility of the clockwork machine that suspended them above the poverty to come (and that it was already damaged) were great and terrible.

If you will permit me an extended metaphor, Rome wasn’t so much demolished by invaders as it was burned down by Roman arsonists who set fire to their own house – and they had been setting those fires since at least 235, long before Adrianople. The emperors of the fourth century (particularly Diocletian and Constantine) may have put out some of the fires by collapsing a wing of the house to smother them, but this can hardly be regarded as improvement, not the least because neither of them did anything to deal with the arsonists (one of which, Constantine, at least, must be reckoned). The emperors of the late fourth and fifth centuries then proceeded to invite people into the house, promising its shelter, if only they would help them light one more fire – and then when the house was burned down and everyone was left on the cold ground, they tried to shift the blame onto the very guests they had invited.

The collapse of the Roman Empire in the West is a complex sequence of events and one that often resists easy answers, but it is a useful one to think about, particularly as we now sit atop our own fragile clockwork economic mechanism, suspended not a few feet but many miles above the grinding poverty of pre-industrial life and often with our own arsonists, who are convinced that the system is durable and stable because they cannot imagine it ever vanishing.

Until it does.

  1. A technical term from the Annales school: mentalités, the feelings and worldview people possess
  2. That is, it is possible for archaeological evidence, like say, the distribution of trade goods, to falsify a textual report, but generally not the other way around. If your textual source says that no one buys imported Gaulish wines, but you in fact find lots of wine amphorae from Gaul, your textual source is wrong and the amphorae are right.
  3. Of course the attentive reader will note here that this very statement puts me on a ‘side’ of how this evidence gets treated. Like most ancient historians, I tend to hold that archaeological data – particularly when you have a big, well-dated sample – can settle some debates decisively, proving one side right and one side wrong largely beyond questions of interpretation. In the past I have found medievalist colleagues skeptical of this approach and the conclusions it, at this point, practically necessitates, though I have not yet seen a systematic rebuttal of the method.
  4. This view of the pure ‘consumer city’ is at best under siege in the study of the ancient economy for reasons discussed below. For a direct assault on it, see N. Morley, Metropolis and Hinterland (1996)
  5. Frustratingly in a volume that I worry many scholars of Late Antiquity may not have read because it is mostly a volume about the third century crisis in a series – Brill’s Impact of Empire – that is generally of more interest to scholars studying the early and high Roman Empire than to those studying Late Antiquity.
  6. The Rise of Western Christendom (2003, 2nd ed.), p. 12). The preface of the third edition (2013) has a more extended discussion on this point, citing some more recent scholarship, but still concludes “the Roman empire had never done as much for its citizens as our exalted image of Rome might lead us to expect. It did not protect the Roman populations against barbarians…rather…it created the image of a barbarian threat so as to justify its fiscal demands” (xxiv). As a description of the late Roman Empire, there are shades of arguability here, but as a general description of the empire the Romans had beginning in the Middle Republic – implied by that word ‘never’ there – this statement seems difficult to sustain, unless one assumes a very ‘exalted image’ indeed.
  7. There are two articles of note here, K. Hopkins, “Taxes and trade in the Roman empire,” JRS 70 (1980): 101-125 and K. Hopkins, “Rome, taxes, rents and trade,” Kodai 6/7 (1995/6): 41-75.
  8. in “Scratching the surface: a case study of domestic graffiti at Pompeii” in L’écriture dans la maison romaine (2011)
  9. e.g. W.V. Harris in Ancient Literacy (1989). On the author, please note.
  10. Emphasis on semi – we must in all of these things not get carried away. The Roman Empire was never the EU.
  11. To expand on this point for a moment, many of the precursor technologies are simply not in place for anything like an industrial revolution at any point in Roman history. Roman crop rotations may, in some cases, have been closer to the late medieval three-field system than the early medieval two-field system (on this see K.D. White, “Fallowing, Crop Rotation, and Crop Yields in Roman Times” Agricultural History 44.3 (1970)), but what they weren’t was the four-course system of the early modern period. They also didn’t have key New World crops, of course. The Romans also didn’t have cheaper blister or crucible steel, nor did they make heavy use of mineral coal. For textiles – a crucial part of the industrial revolution – the Romans entirely lacked the key medieval precursor technology, the spinning wheel (industrialized into the spinning jenny), so even if you gave the Romans a 19th century steam engine they couldn’t use it to spin thread – one of the critical early uses that steam engines were put to which justified further development of more efficient steam engines. Moreover, a lot of what was motivating both the desire to machine out cylinders (for pistons or for cannon) and delve deeper into chemistry was the security interest in gunpowder, which of course was centuries away from being invented in China. And while the Roman Empire wasn’t technologically stagnant, these key precursor technologies were mostly not the direction the Romans were moving – I wouldn’t expect them to have arrived at any of these given a few more decades or even a couple of centuries at their prosperous height. No early industrial revolution was lost.

288 thoughts on “Collections: Rome: Decline and Fall? Part III: Things

  1. Bravo! I don’t actually have any questions immediately occurring to me, but I at least wanted to express congratulations for the latest article set.

  2. Fascinating! I love when everybody’s partly right but distorting their argument a bit much for political reasons

    1. If you had a magic wand, how would you change the Roman political system to get political stability, prosperity, etc

      1. If I had a magic wand I’d make sure Augustus’ grandsons and adopted heirs survived. It’s not that I believe Gaius and Lucius would be such great rulers but it would have established a peaceful direct succession. A steady supply of sane, reasonably effective heirs would have done a lot for stability at the center of the empire. Mind you chaos in Rome seems to have mostly impacted the elites there. The provincial administrations ground on regardless.

        1. I see problems. First, the problem of all hereditary monarchies: incompetent or evil heirs. Note that the end of the “Five Good Emperors” came with an hereditary heir. Second, the traditional Roman ideology was hostile to monarchy, which is why Augustus pretended not to be such, and emperors for centuries avoided the traditional trappings of kingship (including the title).

          What was needed was a strong norm against inherited rule and in favor of appointment by each emperor, by and with the consent of the senate, of a respected and competent successor. I have no idea how to instill such a norm, to be sure, but it wouldn’t be out of reach for the culture and period.

          1. Letting the emperor appoint his successor would undermine non-hereditary rule. Better to have the senate choose the new emperor after the old one dies.

          2. The problem is that by then the successor really was whoever the army would accept as the successor. Occasionally an influential coalition of senators could effect that, but more often it was the person the army already obeyed- the previous emperor. To remedy that you would need either a return to the Republic, or at a minimum a Senate that fully controlled the power of the purse, without which neither the Emperor nor the Army could operate.

          3. A fixed hereditary succession often solves that problem by developing shadow centres of power (shoguns, prime ministers, regents). But Rome had a phobia about kingship, and then later the army and elite had a strong preference for some form of hereditary succession (it was, after all, the family norm). As the army was the real power (in keeping with Rome’s origin as a soldier-republic), the trick would be to find some way of peaceful selection of the most competent general. A vote by the legions, perhaps?

  3. > if plague was the only issue we’d expect a pattern looking more like the Black Death – population decline but living standard improvement

    Is this so, though? I remember reading that when the Black Death hit, economy in Europe was still notably not very specialised – most people were basically just peasants doing little more than subsistence farming. If everyone makes food for themselves, when one third of everyone dies, there’s also at worst one third less food, and it stops there. Conversely, if today Thanos snapped his fingers and killed even just one third of our population, all sorts of supply lines and complex systems would collapse, snowballing into a lot more death and misery. COVID wasn’t quite bad enough to be a civilisation-threatening disease, but most people probably underestimate how close it might have come – had it been as infectious as Omicron, and somewhat more deadly, maybe more across the age range, it could have been a serious risk in that sense. We changed a lot even since just 1918 and the Spanish Flu. So in that respect, given how fragile and important the Roman economy was to maintain that sort of population, couldn’t even simply this one blow have significantly pushed it past the edge? Or at least, does it make sense to compare it directly with the Black Death when the latter affected a much more decentralised, and thus sturdier and more stable, world?

    1. I wonder about the Black Death = Everyone becomes better off narrative as well. Is there any evidence of a plague “helping out” a civilization in other parts of history? It certainly didn’t make things better for the Native Americans when they got ravaged by European diseases.

      I guess it’s possible that the Black Death hit in some sort of quirk of history where something positive came out of it due to some unique historical conditions, but this seems to go against what we know about economics and even common sense.

      1. The Native Americans also had to deal with the Europeans invading at the same time, and were struck down by European diseases at a rate far higher than Europeans experienced with the Black Death.

        1. Yes, estimates of North American Indian population decline, substantially all of it attributable to disease, run as high as 90%, not the 30% generally ascribed to the Black Death.

          1. The collapse of the Mound Builders makes the collapse of Rome look like a joke. In places, their descendants didn’t even realize the mounds were not natural formations.

          2. There’s been a bit of a pushback to that, in general it seems that where disease alone was an issue you saw substantially lower death tolls, but disease combined with general social disruption saw the extreme numbers.

            Most people probably *died* by disease, but they were vulnerable to diseases partially because of social and economic disruption.

          3. First of all, these 90% losses run over a period of – depending on the region – 60- 80 years, not half a decade, as the black death did.

            The loss to the initial epidemic, over a handful of years, was indeed around 30% on average (some regional variety).

            Secondly, these 90% aren’t losses to diseases, but losses to all causes. Spain campaigned continuously for around 80 years before the initially conquest period stopped. The labour conditions imposed by the Iberians did, of course, also have an impact. When Spanish chroniclers write about how Indios could keep working 48 hours non-stop in the mines, they’re referring to forced labour every native family was required to provide. These people weren’t expected to ever return for a reason.

            That such labour conditions aren’t great for one’s physical resilence to infection shouldn’t be shocking news, either. And we count Anne Frank as being killed by the Nazis, not an accidental death to typhus for a reason. No different for the natives exploited in, well, lethal fashion and – in North America at least – with the express purpose of their extinction.

            Furthermore, in regions where the forced labour regimes weren’t introduced (jesuit reductions in and around modern Paraguay; areas that successfully resisted, such as Araucania), the population drop wasn’t anywhere near as severe. This can also be observed in North America, where the Cherokee held at circa a 50% population loss until Jackson’s shenanigans.

            Populations in the general area of the Amazon (outside the sphere of European control), far from the popular narrative of a single Spanish expedition killing them all by accident, managed to retain agricultural societies at the village (not city, as popscience has recently started to claim. The actual study only mentions villages) level until the late 19th, early 20th century rubber rush saw European penetration and local populations being mass-enslaved and dropping by 90%. The Kongo Free State happened twice.

            Conversely, a lot of the alleged victims of the epidemics were already in decline or gone by the time of Columbus. The mount builders were collapsing, and the villages the Spaniards found in the early 16th century only somewhat larger (certainly nowhere near an order of magnitude) than what the French fought in Louisiana and the British on the east coast, which incidentally exported, rather than imported slaves during the 17th century.

            Epidemics without the simultaneous European invasion would indeed have looked a lot more like the Black Death.

            Amazonian populations:
            General (alas, popscience-y) pushback against the epidemics all thing: Can’t link for some reason (phone acting up?) But ‘Beyond Germs’, Amazon & Google Books.

          4. Ok, but the original question was whether it was plausible that a 30% death toll in a settled agricultural society might leave the remaining inhabitants better off. The Amerindian experience, featuring war, enslavement, and much higher death tolls, sheds little light on that question.

        1. As a former server, I should warn you that the non-fastfood section of US restaurant industry is *entirely* structured around making real wage figures undiscoverable.

      2. As someone who has worked a bit on the Black Death – the answer is a bit more complicated (and quite regionalized) but is more true than not.

        In England for example, the massive fatalities ended up knitting together agricultural lands that had been split by inheritances to small landholders. In essence, instead of inheriting 1/5 of a parcel of land, as a small landholder or freeholder you would get the other fifths back as your siblings passed away, giving you potentially enough land to move above subsistence or near subsistence.

        This was also (grimly might I add) reflected in the distribution of alms – evidence suggests that relatively large (I don’t have the data in front of me so I don’t have hard numbers sadly) numbers of people within communities were partially or even fully reliant on alms and charity to survive. A decline in the population means that those who survived’s situation ends up improving as a similar amount of resources are now being distributed amongst a smaller population.

        Furthermore (again, in England), wages or wage-equivalents did rise because both specialized and non-specialized labour became in short supply. Hence the creation in 1351 of the Statute of Labourers, which set a maximum wage at pre-plague rates (this also contributed heavily to the Great Revolt of 1381). While inflation and other economic woes were interelated with this, the core concern was to do with the fact that labourers bargaining for their power and driving their wages up threatened the economic power of the elite, who pushed Edward to legislating against them.

        I would disagree that this has anything to do with any sort of Malthusian logic per se however, its much more to do with the quirks of the economic structure of medieval England. If subdivision of land through inheritance had not been so extreme (I recall an example where legal records indicate that a man inherited roughly a few dozen square feet of land), then there would not have been a marked improvement in living conditions amongst small landholders for example.

        This however was not universal – its more outside my subject area but in Mameluke Sultanate for example, there were devastating economic damages that weren’t recovered from for (I believe) centuries because the economic, urban and demographic situation was simply markedly different.

      3. The native americans had the worst biological situation. They had to face repeated zoonotic viruses without having DNA memory of even the distant relative of those viruses. Each tribe lost some 40-70% of its members in about 50-80 years due to a succession of illnesses: measles, smallpox, influenza, and anything else. The survivors faced a new deadly illnes at about 15-20 years.

    2. This fit’s well i a thought I had, while reading this. What if economic output did not decline because the Empire collapesed in the West, but the Empire collapsed in the West, because economic output declined. At first the plague, the civil wars, and some babarian incursions damaged the econimic clockwork, as our host put it, a little. So the empire declined, wich caused more damage to the economy, etc. Until there wasn’t the “carrying capacity” for a state in South West Europe, let alone an Empire

      1. If you are talking about the Antonine plague (2nd century), your model does not seem to account for the substantial recovery in the 4th century, which in some places, like Italy, persisted well into the 5th. If you are talking about the Justinianic plague (6th century), the West had already seen most of the population decline by then.

        1. I was talking about the Antonine plague. Sure there was a substantial recovery, but as our host mentioned several times, there is evidence that it recovered to a lower standard than before. Which would have played into the political strife in the next centruy. The begin of the crisis of the third centruy was within living memory of the plague. After which West Rome again recovered but to a lower standard etc.
          But I’m happy to hear arguments to the contrary. It was just a thought I had while reading.

          1. Part of what makes it hard to tell is that the end of the plague corresponds almost exactly to the death of Marcus Aurelius and the accession of Commodus. So far as I’m aware, Commodus is pretty much universally viewed as a bad Emperor, without much in the way of revisionist appraisal that Nero and even Caligula have seen.

            This ended a nearly 100-year-run of emperors generally regarded as conscientious and able and for the next fifty years Rome saw a series of emperors who were weak, young, short-lived, uninterested, downright malevolent, or a combination of the above – with only one clear exception – before a military coup in 235 triggers the official start of the Crisis period.

            It may be that part of the reason emperors like Commodus and Caracalla (who must surely be counted among the worst ever, if one follows the conventional perception) have such a negative reputation is that they presided over a period of difficulty principally caused by issues beyond their control, in the same way that Justinian will always be able to play his eponymous plague as a trump card in assessing his achievements. In which case, they may be judged unfairly by history, and blamed for a decline in living standards that happened on their watch despite its not being their fault.

            But equally, it might be that they inherited a more difficult situation than normal but that they either failed to arrest that or indeed made that worse through inattention and/or incompetence.

            It will take someone more knowledgeable about the period than me to give a pointer as to which take is correct, although my own inclination tends towards the latter opinion.

          2. @TOm

            which reminds me that, not exactly the most well documented historical work ever… but that was essentially the thesis of the 1964 movie “The Fall of the Roman Empire” (which then served as a template for the much more recent and famous “The Gladiator”). It portrays a Roman Empire that’s more open and future facing under Marcus Aurelius, save everything going to the dogs and falling into further corruption and selfishness with Commodus. It ends with the protagonist not dying like in the remake, but simply leaving in disgust after witnessing the spectacle of senators trying to buy his loyalty as soon as Commodus was dead, in a scene that basically foreshadows this corruption and shortsightedness being ultimately the end of the Empire.

            It was very blatantly a political movie, and one that was trying to talk about 20th century problems, not 2nd century ones. But it’s still interesting to think about to me right now because – besides being a pretty enjoyable movie in its own right – it definitely seems to espouse Brett’s view of “arsonists” taking down the system from the inside by not appreciating both what made it tick and how far it could be stretched before breaking.

      2. Bit of both. But the empire was primarily a military institution – the army chose the emperors and was the dominant thing in it. The million of more soldiers plus dependants were a significant fraction of the entire population. The army presence sustained the towns and was the reason the empire pumped wealth and materials from the Mediterranean to the Rhine, Danube and to Britain. Trade followed the vexilla. So when the state declined, trade and towns went with it.

        1. Florin Curta suggests that over the 6th century, the whole of the Balkans north of Constantinople lost its farming population. Towns and forts were sustained through annona (salaries for soldiers paid in kind). The empire was militarily successful, campaigning across the Danube at the end of the century, but when the army had to be called in to the capital and to the eastern border in 602, the entire population of the Northern Balkans collapsed. It had been literally kept alive through the pump of the taxation and army system.

      3. This seems true to me also. Marcus Aurelius was ending a balanced reignship, with several good emperors behind him. Still…. he was trying really hard to defeat some barbarians in Pannonia while Tiberius or Traian would have wiped them out quite rapidly. The coins were already deep diving with reduced content of gold. We know that Vespasian and Trajan had increased the gold content in coins after acquiring gold in Jerusalem and Dacia.
        So the economic situation and army efficieny were degrading even before the crisis of the third century. It took some time to reach breaking point but the Roman institutions saved the state when economy was already dying.
        Diocletion usedthose instituions to create a new kind of Soviet style economy (qin and Han China had managed with it for a few hundreds of years) but the system added an even bigger pressure to the economy. This economic system allowed the Western Roman Empire to susrvive the loss of all its provinces until the fifth century.

    3. The Roman Empire had a lot of economic specialization by pre-industrial standards, but they were still mostly farmers. Our own economy is far more complicated.

        1. I don’t know about the likelihood of collapse, but if it happens I don’t think we could handle it as well as the Romans did. They were mostly farmers before and after collapse, so they at least had the job skills they needed for their new circumstances.

      1. Yes, but they were farmers relying on a complex import/export network. Exactly the sort of thing that gets easily disrupted by a shock and then snowballs into famine.

    4. > I remember reading that when the Black Death hit, economy in Europe was still notably not very specialised – most people were basically just peasants doing little more than subsistence farming. If everyone makes food for themselves, when one third of everyone dies, there’s also at worst one third less food, and it stops there.

      And that’s the point OGH makes. The Roman economy was complex and specialized and interconnected, albeit in a rather unsophisticated and haphazard manner with little planning from the folks at top. That makes it markedly different from the effect that the Black Death had on Europe.

      Moreover (and this is more OGH’s post in this section if I take the meaning properly) it means that you can’t merely cut-and-past the effect the BD had on medieval Europe — lowered population but rising living standards — onto Rome, as some have attempted to do with the fall of the Western Empire.

      1. That was my point too. The Roman economy had some aspects of “globalization” that the medieval one didn’t. So there was more to break, hence why I was wondering if it could not have been sent into this downward spiral originally by the blow inflicted by that first major epidemic.

    5. The Antonine plague was followed by the less documented cyprian plague a century later. That may well have killed a similar percentage of the population. And then two centuries after that by Justinian’s (now confirmed as bubonic) plague.

      So it was a multi century one-two- three punch.

      It also happened in an environment of relatively densely populated cities as opposed to the rurality of Europe during the Black Death. So the diseases would spread faster and be more disruptive to the more complex interconnected societies.

  4. So, less Barbarians Storming the border, more steadily increasing background noise of violence as things fall apart in other ways

  5. A little off topic: To what extent is the modern system a fragile clockwork system as opposed to something more robust and resistant to catastrophe? “modern” society seems to have survived many catastrophies (WW1, WW2, extreme famines in russia, China …) with hardly any global change in the level of comfort.

    Knowledge seems more broadly diffused in the system and it is hard to imagine a total collapse with a complete regress to pre-agrarian society at this point (without something extremely catastrophic like a few decades of global warming).

    1. As COVID has shown, our global industrial economy is very fragile – witness the supply chain problems which have dogged us for the last year and change, simply due to the side-effects of a pandemic, and consider what would happen if a WW2-scale conflict occurred today. The massive global industrial supply chain as it is today simply didn’t exist on this scale and this level of interconnectedness like it did in WW2 – indeed, it is a product of the post-WW2 world order.

      Consider: just as a Roman was more likely to eat grain shipped from Egypt than from their backyard in Italy, you are more likely to use products made in South America or East Asia than the United States.

      Also your “without global warming” statement is doing a lot of lifting, since it’s almost certain that natural disasters exacerbated by unchecked global warming will have a detrimental effect on most of the world. Famines, for one thing, the mass migration of climate refugees for another, to say nothing of political instability.

      I agree that it’s unlikely we will fall back to a premodern standard of living, many technologies are just too entrenched and common to be lost and while I don’t foresee a rapid decline in global population and living standards, a gradual one over the next few centuries is almost certain at this point unless massive efforts are undertaken to mitigate global climate change.

      And while we won’t all go back to being medieval peasants, the gulf is so large that things can get a LOT worse before we reach that point.

      1. I think we could just as easily make an argument that Covid has shown how robust the global industrial economy is, given how many things were closed for so long, and given how many places where so many things are made are still employing complete lockdowns to control outbreaks. It could have been a lot better for sure, but I’m not sure everything that’s happened tells a convincing story of fragility.

        1. Absolutely. We saw beef slaughter capacity drop by more than a quarter virtually overnight early in the pandemic. Pork slaughter capacity dropped by a fifth. For most societies, such drops would’ve been catastrophic, but meat remained center of the plate for most people in the United States. Yeah, you couldn’t always find the cuts you wanted and prices grew, but there was usually some type of meat available in the stores. I would’ve never predicted that before it happened.

      2. There’s now the question of how badly a large geomagnetic storm would affect us today, when pre-WW2 it might have scarcely had any effect beyond disrupting international telegraph communications.

    2. The world wars weren’t this kind of threat, though, at least in the US. We lost a lot of people by numbers but not really by percentage, and our infrastructure was not seriously damaged. WWII did a lot of damage to infrastructure in Europe, and recovery there was slow, but also aided by the powers who emerged most successfully from it as part of their cold war jockeying for power.

      Our current system surely isn’t fragile in exactly the same ways, but it has if anything more prerequisites. For many people they eat almost nothing produced locally; certainly very few modern regions are prepared to feed themselves year-round. And all of the technology that enables this is fundamentally industrial, which relies on economies of scale and therefore is extremely reliant on transportation and population to be viable. And all of that relies on institutions insofar as it relies on the ability to move things safely and cheaply over very long distances, which is a function not only of technology, in which we obviously have an edge, but institutions and norms. We’re already seeing some strain on international shipping – my industry (tabletop games) has been badly hurt by increased shipping costs from China, for example.

      My suspicion is that we’re at least somewhat more resilient than Rome was, but we have a very long way to fall, and knowledge isn’t the big issue – it’s that everything relies on everything else. If it were to become significantly more difficult to move things around, for example, we would have a great deal of difficulty adapting our system to that, because everything about modern industrial society relies on very easy movement of goods, both to leverage economies of scale, and also because technology requires extremely rare components. Similarly, if luxury markets were to collapse (due to either increased costs of staple goods or due to concentration of wealth) it’d do huge damage to the whole house of cards, and so forth.

      In short, knowing how to build a factory doesn’t do you any good if you don’t have a road, rail, and shipping network to connect it to, if you don’t have a large number of local workers to staff it, and if you don’t have ready markets for the goods. Our power is so much in scale that most of our knowledge is useless absent it. That power makes us more resilient, but if things were to go seriously wrong, it means the fall would be quite far and harsh.

      1. But we know how to build not just factories, but road, rail, and shipping networks. German infrastructure was devastated during World War II, but the survivors didn’t scavenge stones from the ruins while turning the former autobahns into cart paths. They rebuilt within an instant of archeological time.

    3. Parts of the system are incredibly fragile. International trade did not regain 1914 levels as a share of GDP until the 1980s. The globalized economy was shattered, then got worse with protectionism during the Great Depression, and recovery was very slow with WWII, decolonization, and the Cold War.

      Before WWI, the world was growing exponentially more connected. For instance there was a steel alloy that couldn’t be made during WWI (don’t remember the source or name anymore). The necessary ore was mined in British Burma, shipped to Germany where it was processed, and then the actual steel was made in Britain. Trade was equal to 54% of British GDP and 40% of German. Enormous foreign investment, with the majority of it outside Europe for most of the Great powers, with the Americas (not just US) getting an especially large share. Most of Africa’s railroads were originally built shortly before WWI, since after WWI the European powers had less cash sloshing around and since independence a lot of the countries have been poor and unstable.

      Part of the reason the US did well (besides lesser involvement in the war) is that trade was a much smaller share of its GDP, so the collapse of the global economy gave it a huge leg up comparatively.

      Technological advances raising productivity masked much of the disruption (which once again was less for US). However, there are some suggestions that technological progress may be losing momentum, takes ever increasing people in scientific community to turn out advances at the same rate, and even that may be starting to not be enough. Basically we grabbed a lot of the low-hanging fruit, so a similar globalization collapse might be much more noticeable.

    4. Good point but not everyone survived the world wars equally. Maybe the modern “system” is more resilient because it has permeated the globe. Modern world powers are not so resilient. Europe has been in more or less permanent decline since WWI has it not? Hard to see this trend reversing soon. The world wars might have been roughly equivalent to the Crisis of the Third Century. Let’s see how we’re doing 100 years from now.

      1. Europe in decline since WW1?

        Relative to the rest of the world? Sure.

        In any absolute sense? No. Europe is a LOT wealthier and technologically more advanced than it was a century ago.

  6. Great conclusion to the series! It is interesting that the decline in the economy seems to have started already in about 150 AD, do you think this was entirely due to natural phenomena (climate change and the Antonine plague) or did the Antonines and Severans mismanage things compared to their predecessors? I also wonder what policies of Constantine were so bad as to label him an “arsonist” of the Roman system

    1. I could be mistaken, but I think Prof. Devereaux was using the term ‘arsonist’ to metaphorically describe anyone who was putting their own personal good first and worrying about potential damage to the overall edifice of the Roman empire second. Since Constantine rather enthusiastically participated in the civil wars, he gets to join the arsonist club.

  7. Damn this was an enjoyable read- all three articles were really interesting and well argued.

    I’m reminded of Prof. Lendon’s final lecture in his Roman History course I took two decades ago- the fall of Rome (in the west) was best for the trees, because they got to grow and expand after being held back and chopped down for centuries, they could finally advance and take back their rightful place. It was bad for essentially all the humans, but the the trees were the only ones who prospered.

    1. In Dirt: The Erosion of Civilisations (2016), soil scientist David Montgomery argues that soil erosion caused by land-clearing was a major factor. Hillsides brought under cultivation allowed soil to wash down into valleys, forming malarial marshes, limiting pasture and so on. Western Europe has been largely deforested three times – in the neolithic, in Roman imperial times and from the High Middle Ages on. Fits with Joseph Tainter’s argument that as complexity pushes on the natural resource base, resilience to shocks decreases. Late Rome could not recover from events which the Republic had powered through. Side note: mid-Republic imagery of the countryside is a cornucopia – a horn of plenty. Not seen that way by the mid Empire.

  8. Thanks for making this series. I really enjoyed all the arguments laid out side by side, and your integration of the economic, cultural and political aspects.

    I think I would be less eager to completely dismiss the “barbarians” part of the equation, though. While you explain that the number of external invaders isn’t high enough to explain most of the economic decline, the collapse of rule of law (as well as increasing instability, etc.) would mean that roaming bands of displaced peoples within the Roman empire would form semi-nomadic groups of “barbarians” and mercenaries, and the resulting violence and banditry would certainly contribute to the instability and population decline.

    Indeed, it is in this period that the chaotic incentives that crystallized into feudal Europe became real. The same local lords, the same incentives for militarization and raiding of neighboring villages, but before either the construction of the network of castles and walled cities to protect local populations from marauders or the establishment of the stabilizing feudal hierarchy. (From what I understand, both of these systems had started forming during the crisis of the third century, but were far from having the universal coverage they attained by the Carolingian age.) In addition to this, even after the feudal system is in place, the lack of Roman border integrity means that we start seeing increased penetration of barbarians in the sense of “professional raiders” (the Vikings, the Magyars, the Mongols) from both within and without the empire, which probably contributed a constant low level of instability pretty much everywhere.

    1. While some barbarians did immense damage (eg the Huns destroyed Aquileia and many of the cities along the Danube), others simply displaced some local landowners and settled in as the new elite (eg Goths in Spain, France and Italy, Vandals in Africa. They were relatively cohesive groups negotiating a share of power. Castles and the ‘feudal’ order were late Carolingian responses to the breakdown on central governance under the stress of Magyar, Viking and Saracen raids. These 9th/10 century attacks genuinely were looting expeditions, and extremely destructive (much of the loot was people, on-sold in the East).

      1. There were a large series of barbarian invasions and raids in the later half of the 3rd century that did not lead to the political breakup of the empire, but did in fact result in the pillaging of many cities and the destruction of several. That damage certainly contributed to the population decline by direct damage and also by damaging the “clockwork” economy of the Mediterranean basin at the time.

  9. How hard was it to resist temptation and never mention “Marxism” and “Stages of History” when discussing Finley and the Primitivist School’s approach? 😉

    1. Pretty hard? Finley was a committed Marxist, to the point that the reason he was at Cambridge was that he had to leave the United States (he was born in New York) during the Red Scare.

      That said, even Finley recognizes problems with the traditional Marxist stages of history at least as they come to economic systems, though its not hard to also see him attempting to restore some sense of its general pattern. In any event, a full discussion of Finley would doubtless be its own post.

      Amusing fact while we’re here – Moses Finley, teaching at Cambridge taught Richard Talbert who at UNC taught me, so I am funnily enough in the Finley academic lineage.

      1. Funny. G.E.M. de Ste. Croix (another Marxist) taught Victor Bers (I mean, Bers took one or more courses from him), and Bers taught me (in the same sense), so here I am. Also, Alois Podhajsky taught horsemanship to Lyn Preovolos (in the same limited sense), and I trained with her. I’m afraid neither my classical learning nor my horsemanship would greatly please my progenitors.

  10. “its possible to create a virtuous cycle where these newly more efficient”

    I assume that’s supposed to be a “vicious cycle,” unless you’re one of those “poverty is virtuous” guys.

    1. I thought so too, so I looked it up and a virtuous cycle is a neologism that is the opposite of a vicious cycle. In that it has a positive feedback loop that is good for people, rather than bad.

    2. I think you’re missing the context of that sentence and paragraph. He is talking about how excess capital can create more efficient farms, leading to more capital spent on farms, further increasing effiency. It is the loss of this virtuous cycle that creates the vicious cycle that is a factor in the decline/collapse.

      “And because basic capital scarcity, typically in things like draft animals and manure, are factors in all of this, its possible to create a virtuous cycle where these newly more efficient, market oriented-farms benefiting from capital advantage can invest more in things like draft animals and manure (because the agricultural surplus to feed the animals exists), realizing improvements in production.”

      1. And there’s the further context that the improvement doesn’t rely on new technology, just on the existence of widespread trade and a more-or-less peaceful empire.

  11. “on the assumption that merchant ship design, which changed little over this period, meant that the odds of ships sinking remained relatively constant”

    This assumption strikes me as a pretty fragile one. Industrial accidents aren’t just a technical problem, but an economic one. The risk tolerance for both owners and workers varies greatly with the surrounding economic context. In an environment of general prosperity, for instance, you might see ship owners willing to risk more frequent wrecks because the returns on trade are high enough to make it worthwhile and their business has enough volume to sustain an occasional loss, but workers unwilling to accept this relatively high risk of death because they have other, safer jobs available. This, in turn, leads to a complex cost-benefit calculation for the shipping industry: does it cost more to invest in safety, or to pay higher wages?

    It seems unlikely, to me, that these factors were all constant (or even “relatively constant”) across centuries when, as shown above, everything else about the broader economic situation in the Mediterranean changed dramatically.

    1. This is a topic where I rely on experts and the experts in ancient Mediterranean shipping generally suggest that ship design, especially cargo ship design, changed only relatively slowly.

      1. There’s a lot of safety stuff that wouldn’t show up in design, though. If you hire a larger crew, travel less each day, and favor longer safer routes over shorter but stormier ones, your voyage will be safer but more expensive even in the same ship. Time of year would matter, too; my understanding is that the Mediterranean has most storms in winter and fewest in summer, so this balance of incentives would also affect decisions about when it is safe to sail.

        Sadly, it seems likely to me (an amateur) that this is a question for which we lack evidence. Non-physical factors by definition wouldn’t show up in the archaeological record, and the subject matter is not one that elite Roman sources seem likely to write about. Even if there was a lot of variation in the frequency of shipwrecks, I’m not sure how we’d detect it now.

        1. As long as the trend of safety vs. economic activity is consistent (no effect, more activity equals less safe, or more activity = more safe), the estimation should still be good. Archaeologists will still have to proportionally turn shipwrecks into trade, which presumably involves comparing to some other data, so changes in shipwrecks per shipping will just get incorporated into the estimation as other factors are.

        2. We can assume that merchants don’t want to lose their investments and sailors don’t want to die.

          An evil capitalist willing to risk the lives of sailors for money still cares about the cargo, and a shipwreck means losing all your profits. A ship itself is a big expensive investment in a pre-industrial economy. While the sea has plenty of surprises, there’s also enough predictability about climate and seasons for sailors to figure out the best times and routes from A to B.

          So the proportion of ships wrecked, for a fixed / slowly changing ship technology and known environmental conditions, is going to be as low as possible, especially over a timescale of decades or centuries.

          This can (I assume) also be cross-checked against other archeological data. If the shipwreck data was misleading and in fact the shipping rate in the 5th century CE was as high as it was in the 1st C, where did all the stuff they were shipping go? If the pottery, metalwork, timber, … isn’t showing up in the 5th C rubbish dumps that would seem a good indicator that nobody was shipping it.

          1. The economics of labor mortality are a bit more complicated than that. Here in the present, some jobs are objectively more dangerous than others, but people still do them. Generally this is because they pay more than others jobs with comparable skill/education requirements. If we made those jobs safer, but did so at the cost of worker productivity and thus lower wages, that is not necessarily a bargain the affected workers would accept- after all, if they valued maximizing safety at any cost, they wouldn’t be loggers/roofers/fishermen/etc. in the first place.

            There’s no reason to assume that the risk of death (or capital loss, which follows a similar cost/benefit analysis for business owners) is “as low as possible” in any industry, ancient or modern, because there’s a tradeoff between risk and reward. The optimal level of risk is a fairly complicated matter that varies with temporary market conditions. So it seems to me (speaking as a non-expert who you shouldn’t trust about this stuff) that the shipwreck/trade correlation likely has some pretty big error bars, because we don’t have any way of factoring in the non-physical changes in risk over time.

            (Also, pedantic aside, but “as low as possible” is zero for stuff like long-distance trade. The downsides of producing everything locally would be enormous, but it would mean no shipwrecks.)

            At any rate, I don’t seriously doubt the broad professional consensus about Roman economic history. As you noted, there’s a bunch of other indices that seem to match up pretty well with the shipwreck stuff, and it would be pretty unlucky for us if they *all* had faulty assumptions baked in that happened to produce the same incorrect conclusions.

        3. Interesting theory. Has anyone researched whether there is indeed a correlation between the level of economic activity and the prevalence of shipwrecks? I would think that there was enough evidence from the 19th and 20th centuries to analyze this question. A priori, I could see things going either way: maybe poverty forces shipowners and captains to drive their ships and crews harder, cut corners on safety, etc. Or maybe it’s the opposite, as suggested.

          1. It would be very hard to settle this question for the 19th and 20th centuries because advances in ship design, weather prediction, and radio communications have been increasing the overall safety of ships really fast. The probability of a merchant ship being lost in an accident at sea in 1825, 1875, 1925, and 1975 aren’t really comparable to one another, so “how many ships sank” breaks down as an indicator of “how many ships were sailing.”

      2. Once the Republic set the trade routes free from piracy and warfare and people could concentrate on ships as a means of transport instead of a necessary platform of self-defense, the design of trading vessels largely standardized on a deep-draft, round-bellied trading vessel, with one square sail with enough flexibility to be set as a fore-and-aft lug sail. This enabled a trader to be manned with a smaller crew, and freed them from the necessity of sticking close to the coast so as to beach overnight.
        Ship design in Europe changed quite dramatically once the Vikings had emerged about three hundred years after the Western Roman Empire ended, as they had mastered the art of building strong, flexible ships with nail clenching – hammer a nail through a plank and through a beam, then hammer the end of the nail over so it locks itself tight. Then the rest of Northern Europe took the technique on, with the hulc and the cog, while the Southern Europeans were making do with a set of techniques that were unchanged since Roman times, and when the two got together, we got ocean-crossing vessels such as the caravel and the galleon.

        1. Sorry, s/beam/rib. My bad. I got sidetracked, thinking of carpentry. Mortise and tenon was the method used in the Mediterranean until the Northern Europeans outcompeted them with iron nails sometime around the time of the Hanseatic League.

    2. It’s an interesting topic you’ve raised. Speaking also as someone who isn’t an expert and you shouldn’t trust, it seems to me that the economic context for shipping is very different to that on land. There aren’t the same opportunities for marginal improvements, the risks are more evenly distributed, and the labour force is skilled and (literally) mobile.

      Shipping in the Med was sailing not rowing for the same reason as everywhere else: the wind is a wonderful energy source that isn’t human or animal based. A small crew, low double digits, can move hundreds of tons over hundreds of kilometres.

      So the cost of the crew is peanuts compared to the value of the cargo. (One of Shakespeare’s plays starts with a principal character facing financial ruin from just one shipwreck.) The profit you can gain by squeezing every last penny (denarius?) by cutting down on the crew or conditions is likewise nothing compared to the massive financial loss it might cause.

      In ancient and medieval times nobody is running a just in time economy, and insurance is harder to get or claim. Merchants get paid for delivering cargo, but because it’s wind and weather based with a lot of variability almost nobody cares about – or pays for – early delivery. (I’m sure there were occasional exceptions, but a very small part of overall trade.) So again, pressuring the crew into accepting risks to shave a day off the journey time, or even to avoid losing a day or two, has no financial gain, but a small but probably accountable increased risk of massive financial loss.

      An overseer on a factory or farm can impose harsh working conditions that result in loss of health or life for the workers without personal consequences. At sea, a risk taking overseer (presumably the captain) is risking their own life as well as the crew. Most people become much less interested profit margins under such circumstances. A captain and crew can be pressured or paid to take risks, but it has to be more of a collective agreement on their part rather than a unilateral imposition from above.

      A merchant on land isn’t at personal risk and can order a ship to sail under dangerous conditions. But once the ship has left harbour they can’t do anything if the captain decides “nope, changed my mind”. Before radio they can’t even yell at them.

      And since ships can operate almost everywhere in the Med, especially under the Pax Romana, I’d expect to see a preference for employers who are less likely to get them killed. You are right that “as low as possible” could only be achieved by not having ships at all. Make that “level of risk considered acceptable by the majority of captains and crew”.

      1. All that is nice (free market) theory. In practice, too-small crews, lack of maintenance and other high-risk behaviour is very thoroughly documented as quite often standard practice for early modern and later shipping. Like a lot of risky trades, it comes back to the social fact that sailoring is mostly a career dictated by one’s surround, not one’s individual choice. Grow up in a fishing village – go fishing. Grow up in a lumber town – go lumbering. Ditto mining and many others. The maritime trades were a very distinct social world, with their own dress, customs, places and so on. Like any social ecosystem, it had its best performers, its laggards and its parasites. Also, even landside Roman social relations were very coercive by any modern standard.

        1. Early modern shipping operates under similar enough conditions to the Roman Empire to be comparable?
          Does the social structure around shipping counter economic pressure, amplify it, neutral?
          Roman era, or other era, shipping would be subject to the same kind of economic pressures that lowered working conditions and increased risks on land?
          Happy to learn I’m wrong, can you suggest any good sources for the interested amateur?

        2. Mindstalko’s right.

          If the massive increase in shipwrecks during the Roman Empire were being caused by, say, economic desperation leading more ships to sail in unsafe conditions, then you’d need a deeply counterintuitive underlying situation. The Roman Empire’s (shipwreck-measured) economic desperation would have had to have peaked exactly when all other indicators suggested a Golden Age.

          You’d need captains to have felt much more secure about sailing only during safe conditions and with full crews in the 4th century BC (when piracy and internecine warfare were rampant in the Mediterranean) than in the 2nd century AD (when they weren’t). More confident about not needing to hurry from port to port to keep up their profits during the 5th century AD (when long distance trade in staples and luxuries alike was falling apart) than during the 1st century AD (when it was thriving).

          It just doesn’t make sense. We can reasonably suppose that yes, Roman-era shipwrecks were often caused by Roman shipowners or captains doing something unwise in pursuit of profit… But that doesn’t mean that we should expect to see a breakdown in the general trend of “more shipwrecks in times when there was more shipping.”

          1. To build on this bit of logic…

            Most markets work on some form of risk to profit function. More risk generates more profit. There is also a opportunity cost function, what other risks can I take to gain profit?

            So, in 150 AD there are many prosperous shore side trade dependent communities and in 550 AD many fewer prosperous shore side trade dependent communities. To the extent that scarcity increases price, and risk increases with profit…

            It’s likely the rate of shipwrecks is increasing while the total number of shipwrecks is decreasing. Which would be an even bigger collapse in trade then if someone assumed a constant rate of disaster.

        3. I think we can add a lack of regulations for civilian ship mantainance. Moden states impose quite tight safety regulations on any aspect of life and industry which greatly reduces the risks.

  12. I wonder if discontinuity helps explain the increased availability of archeological data for the classical period over the medieval. That is, it’s easier to excavate classical sites because we’re less likely to still be living in them whereas medieval sites are more likely to have been continuously occupied since the medieval period. It’s easier to dig up an abandoned ruin than to excavate beneath the basement of a shopping mall.

    Regarding the dream of a lost Industrial Revolution I think David Drake has an alternate history series where future knowledge is introduced to Justinian’s Byzantium and so assesses how Roman technology could be a precursor to industrialization. It could be argued they’d do better than expected, especially since they could employ ridiculously large cottage industries to fill in gaps, but still obviously missing key technologies and innovations.

    1. > It’s easier to dig up an abandoned ruin than to excavate beneath the basement of a shopping mall.

      In general, for sure. But anecdotally, my cousin is a rescue archaeologist, and he found a street under the site of a new shopping mall that doubled previous estimates of the Roman size of our home town

      1. I do wonder, less about the ease of excavation, but about the likelihood. In order to excavate a greenfield archaeological site, you need funding to pay people to dig it all up. I don’t know a great deal about the laws around development in areas of historical interest outside of the UK, but if I wanted to dig a hole (which I do) in the garden of my house (C16 timber framed cottage, registered as a building of historic interest), I need to consult “archaeology”. I expect there’s some form of payment involved on my behalf, and I’ll be funding (or, rather, doing) the digging myself.

        Higher likelihood of a chance find from someone already digging stuff up, and more of a legal framework to provide funding for further excavation.

    2. I believe that’s Flint & Drake together, co-authoring. If it is that series, there’s two incursions from the future; the Romans get foresight and some advice on how to incorporate technologies in use by other peer states, while the Instant Industrial Revolution happens in central India.

      1. Nitpick: that Industrial Revolution wasn’t instant; it just happened in the backstory. Also, to a large extent the Indian state benefiting from incursions from the future was also relying on huge cottage industries.

        Thus, for instance, their gunpowder armies relied heavily on grenades and rockets, which can be mass-produced with Bronze/Iron Age technology as long as you know how to make gunpowder in the first place. And not so heavily on cannons or handguns, which require Early Modern levels of metallurgy that even a superhuman time-traveling AI cannot impart to more than a handful of workers.

        Honestly, the realism or lack thereof of how fast the metallurgy to build cannons and muskets advanced within the series is a weak point in the Belisarius series, now that I think about it.

        1. On the one hand, it’s a necessary weasel to be able to tell the story they want to tell.
          On the other, a fair amount of page space is dedicated to “One Side Gets Impractically Impressive Siege Gun Designs From Their Time Traveler, The Other Side Gets Stirrups” (which is almost certainly a deliberate reference to the WWII attitude that Neal Stephenson described as “worshipping Ares vs worshipping Athena”

      2. Drake’s outline, Flint’s writing, Drake’s polish. The characters, plot, and structure are very Drake, while the “set dressing” (in particular most of the dialog) is very Flint. Drake is probably my second or third favorite author, and I’m a pretty big fan of Flint as well.
        This series is comfort fiction for me.
        Drake did a lot of research, but was also perfectly willing to change details in the name of a better story

  13. And here it is important to understand how you can have economic growth happen in a system where the basic inputs of production – in this case land and farming technology – haven’t really changed much.

    Something like this post is a source of fun in modern economics: You can measure things like capital investment. how educated the population is, amounts of various resources (with the usual conditions on measuring stuff like this: education is something like ‘years of schooling” rather than an exact measure of schooling + training quality), and it turns out that the biggest sources of growth end up being hard to define things like technology and institutions. Which makes figuring out how to grow an economy tricky, since getting the institutions right is itself a mess, as this post clearly points out.

    Given how messy a lot of this data is (even modern economics often can get messy, as I’ve noticed in some attempted research ideas that had unusable/nonsense results that can’t be interpreted any particular way) it is a heck of a result for Roman political history to show up so cleanly.

  14. Thank you for another great post ! I have shared and will continue to share this series with people.

    There is one thing that I was hoping to see: Estimates of the population of various regions of the Mediterranean / Western Europe in 150, 350, 550, and 850.

    Friar’s chart is similar, but not the same thing. It estimates how long lasting the population declines were, but not how deep they were.

    Do you know of somewhere which has this sort of comparison?

  15. I am curious what, if anything, graffiti tells us about the decline of Greek as a language in Western Europe. I went to graduate school for philosophy so this looms rather large in my understanding of the fall of Rome—especially because (as I understand it) assumed bilingualism meant the Romans weren’t big on translating Greek texts into Latin, ultimately causing many ancient Greek texts to be forgotten in the West until roughly the time of Aquinas (or much later in some cases).

    1. It is not only the question of not knowing Greek. It was, very much, a question of deliberately not being interested. The Western Church built heavily on the philosophical tradition of Stoic and Neo-Platonist philosophers, whose works were Christianised and Latinised by Ambrose, Augustine and Boethius, inter alia. Aristotle, on the other hand, was not in the fashion even among non-Christian Western Romans of the late empire. There was no demand for his works, so they were not translated.

      So, from the viewpoint of a sixth-century librarian who wonders at what book should Brother Alembertus be copying next, Aristotle is a completely marginal author, a barren sidepath in the great flow of philosophy which goes from Plato to Zeno, Cicero, Plotinus and Marcus Aurelius, then crossing over into Christianity.

      1. Both knowledge of Greek and direct knowledge of Plato were lost in medieval Europe. (Not only did no one know Greek, but there were very few Latin translations of Plato available.) So Brother Alembertus wouldn’t be copying either Aristotle or Plato, because no one could read either of them. Indirect knowledge of Platonism survived, and remained influential, through Augustine and other Latin writers.

        1. Some of Plato and Aristotle’s texts were known throughout the period, IIRC, but often fairly (today) obscure works. (and of course, a lot of both have *remained* lost. Eg, bits of Timaeus and a few bits of aristotles on logic were known.

  16. I wonder why the 5th century conflicts were so devastating, when Rome had gone through many civil wars before without the repercussions being so severe.

    For example, I wonder what makes the 5th century conflicts different from the civil wars of the late Roman Republic, in terms of destructiveness to political unity. In the World Lead Production chart, production actually _increases_ from 50 BCE to 1 CE, whereas in the 5th century it declines quite a bit.

    I guess another question I have is “why then” – why did Rome collapse from internal conflict in the 5th century specifically, rather than during the 3rd century, or the 7th century, or at some other time? Was it due to some external change (e.g. the availability of “barbarians” to recruit)? Was it due to some internal change (e.g. some change in succession norms or how the state was organized which made conflicts more destructive)? Or was there a progressive decay from the early Empire onwards, and the 5th century was simply the point where the decay became severe enough to cause collapse? (And if this progressive decay did happen, is there a way to measure how “decayed” things were at any given point in time?)

    1. One thing that I notice in the civil wars of the 5th century is that the instigators often lack imperial ambitions. Even when they have the chance to become emperor, they normally prefer to place a puppet on the throne. Especially the “barbarian” leaders are often happy to gain control over a particular region and then make peace for a time. Even in the instances when this had happened before (the Gallic and Palmyran Empires in the 3rd century, the tetrarchy, various divisions during the 4th century), there was always someone who wanted to be the sole emperor and ended up reuniting the empire. But after Theodosius it seems this stops. Note that there aren’t any direct wars between the eastern and western emperor either.

      It’s an interesting paradox that the civil wars are smaller in scale, but end up more destructive in the long run.

      But why the leaders in these civil wars seem to lack imperial ambitions I don’t know.

      1. I’m tempted to run this question backwards: Why did previous rebels always march on Rome and try to take the imperial office, rather than sitting back and declaring independence? I suppose if they had not done so, they would have been cut off from recruits and resupply from Rome, and faced attack by loyalist elements of the army under Rome’s control

        So a rebel could either seize the throne, or face destruction by its occupant. Once the provinces are well fortified, and have their own locally-raised armies, that becomes a lot harder for whoever the Emperor is.

    2. Guy Halsall in “Barbarian Migrations and the Roman West” makes the point that the early Roman Empire was an absolute exception in the history of Europe. Never before and never after were the Seine, Rhine and Thames basins united with the Mediterranean (down to Andalusia and Sicily) in a single state, because the centrifugal forces are just too great. The reason it worked in the early Empire was the exceptional civilizational gradient towards and attraction of Rome. Once conquered, the provincial elites fell over one another in their desire to rise in the ranks, to embed their home towns more deeply into Roman cultural and commercial networks, ultimately to become Roman senators. From the 3rd century onwards, the normal centrifugal forces came to the fore and had to be constantly appeased by at least one Caesar residing most of the time in Gaul (Treveris). As soon as the imperial presence vanished in 380 CE, Northern Gaul and Britain deurbanized.

      1. Europe is geographically no harder to integrate than say China. The difference is that as China expanded southwards it was able to assimilate the locals, and establish the ideological norm of political unity. Thereafter, Chinese history is one of unity, breakdown and war leading back to unity (none of the contenders ever thought of themselves as other than temporarily embarrassed sole rulers). By contrast, while 4th century Romanised barbarians (like Stilicho) saw themselves as custodians of a tradition of unity, the 5th century ones (Clovis, Theodoric et al) drew on the ‘barbarian’ as well as Roman ideology of ruling, – they were tribal kings more than imperial claimants. When in the 9th and 10th centuries the European centre of gravity shifts beyond the Rhine, imperial secular unity is so vague a concept as to have very little force (religious unity is a different matter).

        1. Harder to integrate? Maybe not. Harder to conquer? Absolutely, yes. The problem is rivers and transport.

          OGH has made the case and provided the evidence time and again that rivers were the highways and arteries of the pre-industrial world, so I’m going to proceed with that point as given. China, from Beijing to Hong Kong, has two river systems, the Yangtze and the Yellow. You can embark an army in Lanzhou, follow the rivers and coast down to Hong Kong, and then sail up to Guangzhong or near Chengdu without ever leaving the water. All told the trip would be ~4.5kms.

          By contrast, to go from London on the Thames to Rome would cover ~5.5kms — and that’s only going from the mouth of one river to the mouth of the next! Yes, those river mouths are where the largest, most important cities are likely to be placed, but the fact is that China’s great river systems all fed out the same two kilometers of coastline. Europe’s river systems are spread out along its entire exterior!

          1. Except that pretty much all campaigns for unity start in north China – with its base of peasant farmer-soldiers – and go south to the middle Yangtze, overland, then mop up along the coast. Or take the Middle East – mostly united from cores in either Mesopotamia (Assyria/Seleucids/Abbasids) or Iran (Achaemenids/Parthians/Sassanids/Timurids). The Carolingians nearly managed to re-unite western Europe, but the ideological base was not there.

        2. I see a very far-reaching homology between the “kings of the Franks, Visigoths, Vandals and Ostrogoths” (all of whom, including Clovis, had been Roman officers at some point), and the “Northern Barbarian” rulers of the Yellow River basin in the 4-6th century CE. In both cases, the ruling elite maintained some “barbarian” identity in contrast with the “civilized” conquered. In both cases, the “barbarian” elite also intermarried with “civilized” aristocrats and became bilingual and bicultural. In both cases, the “civilized” and “barbarian” parts of the population converted to the same faith (Christianity and Buddhism). In both cases, languages derived from the former empire finally prevailed over the foreign languages.

          I see the point of divergence where China was re-united at the end of the 6th century, after about 400 years of disunity, while the rulers of Constantinople
          never managed that feat, nor did the Umayyad caliphs.

          Maybe the root reason is that Han Chinese culture had the equivalent of the summed prestige of Egyptian, Mesopotamian, Greek and Roman culture. The Umayyads did not adopt Greek nor Latin.

          1. They did, to a reasonably large extent at least among elites (and more into the Abbasid) adopt *persian* though. Various forms of persian remained a prestige language (and even outside of the core iranian sphere, eventually)

  17. For teaching purposes, would love a sequel/follow up post talking about the redevelopment of Mediterranean trade in the eighth century (reviewing/summarizing the Pirenne thesis and contemporary responses to it, e.g. McCormick’s Origins of the European Economy).

      1. Then again, could be interesting to get a take on that scholarship from someone a bit outside the period and without a dog in the fight so to speak.

        1. I can’t speak for Prof. Devereux, but I would hate to publish something where it might be thought that I was an expert (e.g., some area of law very far from my practice), but knew so little that I would be capable of some major howler. It would hurt my professional reputation and credibility in a way that writing something stupid about, say, quantum mechanics would not.

  18. ” It’s a striking use of writing to comment on what was a fairly humble profession.”

    It’s also striking that the opening line of Virgil’s epic had become a pop-culture meme– which in turn tells us something about the Roman lower-middle-class’ level of education.

    1. Yes, it seems like the Aeneid very quickly became a pop-cultural icon almost like its predecessors in the Greek world, I guess imperial patronage helps a lot. As you say it is also interesting to see what literature was common in different social classes. I am reminded of Ammianus Marcellinus’ scornful comments on haughty and decadent nobles who only read Juvenal and Marius Maximus

    2. I am reminded of a philosophy professor who spent summers in Italy and spoke of discussing Dante with the local butcher, who had been made to memorize swaths of the Divine Comedy in school as a punishment for misbehavior.

      1. Without it having been punishment, I can probably still recite from memory the opening of the Divine Comedy, a handful of other verses here and there, and some poems by Giosué Carducci, Giovanni Pascoli, and Giacomo Leopardi. Italian school used to be very mnemonic, I think now it got a bit better. Not just for the letters either: I had a chemistry professor who once made us learn the whole periodic table, and a geography one who expected us to be able to name all the world’s countries on a blank map.

        By the way I don’t think the Dante thing is a new phenomenon either. I remember reading at some point that people in Tuscany could already recite parts of the Divine Comedy less than 100 years since it was written. It was just really popular, almost instantly has been.

  19. Interesting example of how a connected system can be fragile is the global economy after WWI. International trade did not regain 1914 levels as a share of GDP until the 1980s. The globalized economy was shattered, then got worse with protectionism during the Great Depression, and recovery was very slow with WWII, decolonization, and the Cold War.

    Before WWI, the world was growing exponentially more connected. For instance there was a steel alloy that couldn’t be made during WWI (don’t remember the source or name anymore). The necessary ore was mined in British Burma, shipped to Germany where it was processed, and then the actual steel was made in Britain. Trade was equal to 54% of British GDP and 40% of German. Enormous foreign investment, with the majority of it outside Europe for most of the Great powers, with the Americas (not just US) getting an especially large share. Most of Africa’s railroads were originally built shortly before WWI, since after WWI the European powers had less cash sloshing around and since independence a lot of the countries have been poor and unstable.

    Part of the reason the US did well (besides lesser involvement in the war) is that trade was a much smaller share of its GDP, so the collapse of the global economy gave it a huge leg up comparatively.

    Technological advances raising productivity masked much of the disruption (which once again was less for US). However, there are some suggestions that technological progress may be losing momentum, takes ever increasing people in scientific community to turn out advances at the same rate, and even that may be starting to not be enough. Basically we grabbed a lot of the low-hanging fruit, so a similar globalization collapse might be much more noticeable.

  20. This jumped out at me near the end — “and instead a product of actors within the political system, within the empire, tearing it apart out of the pursuit of their own interests, deceived by the assumption that something so old could never simply vanish…until it did.” Could anything be a better allegory of the current political situation in the USA? And the disinclination to see it or admit it by those convinced “it can’t happen here”?

    Unrelated note: I can see one reason for cities in the later Roman era to become depopulated and then decay — those “barbarian invasions”. Even small numbers of hungry, ferocious invaders would wreck havoc on large settlements. Once the borders became porous and outsiders began to enter the empire (by invitation or violently), where would they go in search of ample food supplies and portable wealth? Target the cities! Records do indicate sackings and lootings in this period, and it wouldn’t take too many such episodes to send a lot of city-dwellers packing off to the countryside for refuge. Shoot, you see this still happening today in war zones.

    1. Modern war tends to force people into cities. Baghdad, Damascus, Aleppo (and previously Saigon) all grew during the conflict.

      1. Yes, and besides, Dr. Devereux spent several paragraphs deconstructing the idea that people ran off in mass to the countryside…

        1. One thing to remember is that before modern medicine and sanitation, cities were population sinks. People had children and raised them in cities, but never enough to keep up with the death rate. There were always people moving into the cities from rural areas.

          When times get hard in the cities, obviously the death rate will increase, and fewer people will move there from the countryside.

          1. And modern medicine means at least the 19th century if not the 20th. Given that London got as big as Rome in the 18th century (I think) that’s horrifying. Now I want a monograph comparing urban mortality among the various great cities of the world across time.

      2. The urbanized enviroment probably presents the only persistently artificial ‘fortress’ enviroment that remains difficult for a modern system army. See: Battle of Grozny. If I was a civilian in a modern war zone, I’d want to be in a city too.

  21. “demonitization”

    I guess that’s a typo, but still, I can’t help but imagine some sort of demonically boosted economy.

    1. Should be “demonetization” (which still looks devilish). Meaning de-monetary-ization; becoming less monetary.

    2. The problem with a demonically boosted economy is that every seven years, all the rich and powerful people die and are carried off to Hell when their soul contracts expire.

    3. Presumably demonitization means canceling prior warnings. Sort of like politicians are doing now with COVID.

  22. I started out getting interested in ancient warfare but soon discovered an interest in ancient, specifically, Roman economy – how to pay for all that warfare. This has been a fantastic post.

  23. Has anyone attempted to quantify the number of Roman soldiers killed by Romans during the endless civil wars versus the number killed by external forces, and, in particular, by “barbarians?”

  24. I think you are a bit too light on the disease explanation, and hit on the wrong plague.

    Specifically there is some evidence that the Plague of Justinian in the mid 6th century was black death level. Indeed it’s the same type of bacteria that caused it. Records of its effects outside of eastern roman territories are a bit spotty, but it killed around 20 percent of the population of several regions in the surviving roman empire, and I don’t imagine it would kill fantastically less everywhere else it spread.

    This touches on a light issue of archaeology in that it is hard to date trends so specifically. You can see that there was a decline in population by the 6th century, but is that just collapse, or is that a function of Europe (and the middle east’s) first brush with the black death? What is the more prominent reason?

    1. In Britain and Gaul, cities already became much smaller in the 3rd and 4th centuries, well before the Justinianic plague (see part II of this series). There is no doubt about the dating of the new, smaller city walls and about the diminished circulation of coins in Britain, which are easily dated.

  25. Fascinating, and for the first time I understand why the reduction in long-distance trade led to such precipitous population decline!

    Dr. Devereux, when you say Britain was the worst hit part of the Empire, would you agree in restricting that to “worst hit part of the Western Empire”, or “worst hit part of the Empire in the 5th century”? As a non-historian, I am impressed with Florin Curta’s work, most recently in “The Long 6th Century in Eastern Europe”. He seems to make a strong argument that the Peloponnesos until the second half of the 6th century was highly urbanized and prosperous; that the northern Balkan peninsula already in the 6th century was not feeding itself and was kept alive only through annona; that almost the entire Balkan peninsula, including almost all of modern Greece, saw an enormous depopulation and the complete loss of literacy in the 7th century; and that some areas in the centre of the peninsula are archaeologically void of human traces in the middle of the 7th century.

    The reasons Curta favors are similar to what you argue here: breakdown of long-distance trade, deurbanization, finally loss even of rural population. At the same time, this implies that the (Eastern) Roman Empire retreated from the Balkan peninsula instead of being overwhelmed by a flood of (Slavic) barbarians.

    As I said, I am impressed by this case, but I would like to hear the opinion of professional historians of the epoch.

  26. Two questions and two nitpicks


    1. What’s causing the steep decline in shipwrecks right after the early empire peak? Squinting it looks like the end of the Julio-Claudians. Was that so disruptive?

    2. Rome couldn’t have experienced an industrial take-off due to lack of tech but what sort of things COULD it have achieving if the economic peak was sustained longer? That’s alternative history speculation of course, but what sort of things would’ve been within the reach of a wanked Roman Empire?


    1. Like some other people I think a bit more emphasis on disease and climate would’ve been useful here.

    2. Also after the Roman system broke down things kept on getting worse for quite some time before hitting a nadir. These articles mostly focus on the immediate fall of Rome instead of that later period of Late Antiquity.

    1. 1. I think this may be the “barrel” point Dr Devereaux mentions in the text, although that’s not entirely clear. What I understood from that (on my second reading) was that the shift to a wooden medium of transport around that time for liquid goods means that shipwrecks thereafter are much worse-preserved in general, and gives the appearance of a steep decline whereas if that is corrected for, that peak is more of a plateau, and the actual drop is a while later.

      2. A wanked Roman Empire?!

      1. Oh, sorry. “Wank” is alternate history terminology for overpowering your favorite country because of how mastabatory those stories often are.

  27. “what was elevating this system above the pre-modern agricultural norm was not the unrush of technological change”

    Onrush? Maybe?

  28. Oh there’s graffiti riffing off the first line of the Aeneid. Oh, it was famous then too, that’s why we know it now.

    1. Maybe a commercial shopping list. You send what ever worker is free that day, to the market every week. So instead of remebering the list, and telling it to the worker every day, you just say: “Get what we always get, there is a list on the wall.”

      1. When FB first arrived, I was very disappointed that someone had beaten me to my “virtual dorm bathroom toilet-stall wall” idea. I could never get the name to flow right, you see…

      1. I was just thinking that. A list of your shopping staples — how much grain, wool, olive oil, wine, etc did your household consume of a regular basis? Write that on a wall and you or your servants can check it every time you go out to market.

  29. The “fragile clockwork” warning about the possibility of decline in our own future is certainly valid, as seen not only from the breakup of the Western Roman Empire, but also the end of Mycenaean civilization, the end of the Han and the Tang dynasties, of Heian Japan, and other examples.

    Alongside this political warning, it seems fairly evident that the “heady, nearly vertical upward trend in human prosperity” of the 18th to 20th centuries depended on cheap energy, which is now running out. For this biophysical aspect, the Roman Empire seems to offer little parallels, except maybe the loss of soil fertility that I have vaguely seen proposed sometimes.

      1. The only wrong word in that sentence is “just”.

        Aside from the fact that nuclear power plants are expensive to set up (and less profitable than fossil fuels), there is considerable social pressure against nuclear energy in the modern world. Not all of it is unjustified—we have solutions for the problems associated with nuclear power, but that doesn’t make the problems go away—and even if it was pure fantasy, getting people to support something they’re currently opposed to is not the kind of thing you can sweep away with “just”.

        If you’ll allow me to butcher Clausewitz, even if green energy is simple, the simplest thing is very hard.

        1. The nature of the obstacles is moot. Besides the lights going out is a good argument. As with rising crime, people will demand solutions.

          1. My point wasn’t the nature of the obstacles, it was their existence. Because they exist.

            As to your second point, I have three objections.
            1. As with rising crime, the solutions people demand (or that get implemented in response to their demands) may not be effective ones—in fact, they might serve to make the problem worse! (Have you looked at how often people jailed for minor offenses go on to commit violent crime afterwards? That’s what “tough on crime” policies get you!)
            2. If power grids are only improved after people start losing power, I’d count that as a failure. After all, those power grid failures would lead to a lot of human suffering and even death. We know this for sure because…
            3. …the scenario you’re describing isn’t a distant dystopian future—it’s happening now. Remember how power grids in Texas failed last winter, leading to a lot of people freezing to death and a lot more suffering in various ways? It’s happening again. People demanded solutions, and yet here we are.

          2. 1. Have you ever looked at how many people commit crimes after being treated leniently?

            2. It doesn’t matter whether you count it as a failure. It matter whether the kooks still get to demand that only their sacred methods be used to generate power.

            3. Yes, the government insisted that the wind turbines could not use gas-powered heaters to heat them up, and had to use electric ones, thus producing a vicious cycle. That was the kooks, not the people in general.

        2. Drink!

          Nukes also have issues with fluctuating demand so can need to be supplemented with more responsive forms of generation which is usually fossil fuels. Though great honking batteries would help.

      2. Not to derail the overall subject, but note that I specified cheap energy. Private capital doesn’t invest in nuclear power because it’s a money-losing proposition. Note also that I am not arguing any other source of energy is cheaper than nuclear energy, because it doesn’t matter. What matters is if an energy source is as cheap to produce as coal was in the 19th, and oil was in the middle of the 20th Century.

        1. I believe it’s only money losing due to the amount of overbuilding required. Which may be necessary for a reasonable level of safety but the political consequences of even a non-hazardous accident are so severe that nukes are giant money sinks despite how cheap they are to produce power. Even if oil and gas only had to pay an amount proportional to the severity of accidents, they wouldn’t make money either.

          I believe the safety standard for nukes is a million reactor years between critical accidents and the systems work long enough to prevent a meltdown regardless.

          1. Coal, solar, and wind are much the same. Nuclear power is extremely safe as long as you don’t do dumb stuff like what the Soviets did that led to the Chernobyl disaster. Three Mile Island and Fukushima both had virtually negligible effects, and even the Hanford site wasn’t catastrophic except for the immediate area.

            Frankly, I would much rather have a nuclear reactor in my backyard providing my power than a wind farm.

    1. I’m not sure I’d call it fragile clockwork so much as brittle clockwork. Complex societies—especially ones that are complex because their institutions and trade networks extend across wide areas with diverse peoples—tend to be more resilient to small shocks that could devastate simpler or more localized societies. But when they fail, they fail hard.

      This is only one of the reasons I think that people who look at “fragile clockwork” style arguments and come to the conclusion that we should have simpler, traditional societies are wrong. Complex societies let us make do a lot of things you can’t in a smaller, simpler society, and the risks of losing all that seem smaller than the risks of not having it at all.

      But at the same time, that doesn’t mean we can’t make the clockwork less brittle. One obvious point in the modern world is the just-in-time supply chain; it reduces the amount of unused raw materials or capital in the system during normal times, improving production efficiency, but also makes the system as a whole less elastic to events which cause sudden spikes in demand.
      To pick a random example, having just enough hospital capacity to satisfy typical demand is well and good until you have a pandemic. At that point, everyone who doesn’t run a for-profit hospital is gonna wish that hospitals were a little less efficient…

      1. Markets are great at optimizing for efficiency, not so good at resilience. Key role for government, there… if you can get a government that believes in long-term resilience.

        1. This is why the ‘run government like a business’ is such a poor slogan.

          The costs to governmental failure is usually measured in the 100s to 1,000,000s of lives, so government is MUCH less accepting of failure. Secondly, in government revenues and demand are inversely correlated which is literally impossible in a market oriented business.

      2. What are some real life examples of “fail[ing] hard”? I don’t see it happening, either now on account of COVID or in industrial era history. I mean, right now I can’t get the new tablecloth I want because it’s out of stock until April, due I guess to supply chain problems, but that seems like failing pretty easy. And people had it bad in Japan and Germany in 1946 (even worse than my tablecloth problem!), but their cities didn’t depopulate and fail to recover for centuries. In fact their economies don’t seem to have been “brittle” at all.

        1. Most of the key technologies of the 40s were very resilient. Railroads, for instance, are very tough (things break and get fixed all the time). Ditto steel mills and such. There’s an enormous amount of skilled hand-labour, and so long as the skills are there, the material can be re-built quite quickly. If you want to look at a modern failure point, think of how few chip fabs there are, and how chips are a key component in everything. A few dozen plants, each taking a decade to build.

    2. “Alongside this political warning, it seems fairly evident that the “heady, nearly vertical upward trend in human prosperity” of the 18th to 20th centuries depended on cheap energy, which is now running out.”

      I dunno; apparently within the last year or two we’ve reached a tipping point where it’s now cheaper to build new solar plants than it is oil or coal plants. So even as stores of those run out it might not be too hard to transition. And with the milestone in nuclear fusion research announced just this week, it’s feasible that within out lifetimes we could see the onset of the cheapest (by far!) power source humanity has ever known, or ever will.

      1. Note that I am not arguing any source of energy is cheaper than any other one right now, because it doesn’t matter. What matters is if an energy source now is as cheap to produce as coal was in the 19th, and oil was in the middle of the 20th Century. With regard to fusion energy, it was a mere few decades from fruition back in the 1960s, and it continues a mere few decades from fruition.

        However, I should probably stop arguing this point because it distracts from Dr. Deveraux’s magnificent series of posts.

        1. The coal and oil plants are also heavily subsidized, if only by being allowed to spew their pollution into the air. The economics would be rather different if power plants — or cars! — had to capture all their CO2.

          1. That the evils of carbon dioxide have never manifested in the past as the scientists claimed as an additional evil.

          2. If that something demonstrably causes harm that is then shouldered by third parties, then yes, not taxing it is a subsidy and alters the correct pricing of all goods and services that depend on it, causing major market inefficiencies.

          3. The reason why your scientists are predicting harm half a century off is that they have learned that they will live twenty years and be shown up as liars

            So even accepting your premise, you are lying. And also revealing the deep evil: you think that you have a right to take the fruit of other people’s labor as long as you get some academic to claim they are doing harm. And, as we all know, they will make such claims regardless of truth.

            Forcing other people to labor for you is slavery. By definition.

          4. I don’t think that’s a valid comparison. Solar is not only being _directly_ subsidized by tax write-offs but indirectly in that solar cell factories don’t have to run entirely on electricity generated by solar. Whereas CO2 emissions weren’t even recognized as an externality until the latter 20th century; until then the atmosphere was presumed to be effectively infinite relative to industrial emissions. Indeed it was a major battle just to get coal burners to stop emitting black carbon soot, ash and sulfates.

          5. I didn’t even *mention* taxation, Mary. That’s on you. I suggested they have to capture their CO2 (and other pollution, I didn’t say.) Like being forced to pick up your dog’s poop, or is that also evil?

            The idea that you have the right to pollute without consequence is a much better candidate for pure evil, the attitude of orcs and Saruman. “I fouled the water and put filth in the air, but stopping me would be tyranny!”

          6. But again, for the longest time CO2 wasn’t even viewed as a pollutant. Like asbestos, no one saw anything bad or wrong about it, and it’s only in 20/20 hindsight that we know it was a problem. It isn’t like evil polluters were being given a pass on destroying the environment, because no one saw it as a global problem. Even severe pollution was seen as a purely local problem, like living downwind of a pig farm.

          7. False analogy. You would have to clean up after your pet’s breathing.

            This shows you are using the term “pollution” dishonestly, to tyrannize and claim moral superiority at once.

            And forcing people to pay money to your cronies, which is what your demand amounts to, is a trivial difference.

          8. The minor problem with comparing powerplants to breathing is that you could breathe for a million years and not produce as much CO2 as a coal power plant does in a day.

            “The dose makes the poison” is just as true for the environment as it is for people.

          9. The much more fundamental problem with comparing power plants to breathing is that breathing is carbon-neutral, apart from modern fertilizer: live plants fix CO2 from the air, we eat the plants and return the carbon. Power plants burn fossil fuels, emitting carbon that has been sequestered for 300 million years. To equate the two processes is utterly absurd, and not to be taken seriously.

          10. The carbon you breathe out might have been sequestered without you.

            It’s more likely than that the scientists who have been wrong in every prediction about global warming, and have stopped making predictions in a time frame they will live to see, are right.

          11. The claim that scientists “have been wrong in every prediction about global warming” is flatly false. We are seeing disastrous effects already, and in some cases the speed of the change has exceeded even the pessimistic predictions. Last year saw a litany of climate disasters, including the heat dome over the Pacific Northwest that killed hundreds of people and produced the highest temperature ever recorded above 45˚ latitude.

          12. Oh, shock. Because weather disasters have never in the history of man occurred before.

            No. Serious analysis across time does not bear your claim out.

          13. Define “serious analysis”. Is it not serious analysis when one points out that the ten hottest years on record were all within the past twelve, or that rain in Greenland and tides washing into Miami streets are not normal? When the NOAA says “climate change is also supercharging the increasing frequency and intensity of certain types of extreme weather that lead to billion-dollar disasters—most notably the rise in vulnerability to drought, lengthening wildfire seasons in the Western states, and the potential for extremely heavy rainfall becoming more common in the eastern states”? When Michael Wehner, a specialist in climate attribution at the Lawrence Berkeley National Laboratory, says “The fingerprint of climate change on recent extreme weather is quite clear. But even I am surprised by the number and scale of weather disasters in 2021”?

          14. Given the length of the records and their reliabilty, the assertion about the ten years is more dubious than you realize.

            Also that there have been no times in which weather events have not been “normal.” And that the experts have lied so frequently that they are not evidence.

          15. It’s sure convenient to argue this way. Just spout random unsubstantiated claims and act smug. Who cares if several years in a row are found to be the hottest in over a century of records, if we got completely unheard of temperature in summer, floods, storms and hurricanes of increasing violence. It’s always rained some! Therefore that proves there is no problem!

            True science.

          16. You claimed climate scientists have a history of lying, without substantiating the accusation. That’s not just “questioning the claims”. The evidence that human-caused global warming is a major crisis has been piling up for 30 years, and some of the worst predictions have been borne out. If you can’t come up with better arguments than handwaving and well-poisoning, just don’t bother.

          17. You lie.

            I explicitly pointed out that they kept claiming disaster would come with twenty years over more than thirty, and instead of admitting error, shifted to fifty years.

            Falsely claiming you are reliable when you aren’t is a lie.

            Sane souls notice that they are pushing the same old leftist demand to micromanage everyone’s life that they push as a cure for everything.

          18. You are the one making stuff up. You claim scientists “predicted disaster” and then moved the goalposts. Who did this? When? Names, dates, references? If your argument is just “I vaguely heard in the generalist media some extremely doomerist click bait predictions that didn’t turn out to be true, thus all climate science is bunk”, well, take that with the media. They often don’t understand much about science and make sensationalist claims on all topics.

            But actual climate science predictions and models, the mainstream stuff, not the fringe, are turning out to be pretty correct – optimistic, if anything. They also generally come in a range of scenarios because obviously every serious scientist knows that there’s not ONE certain prediction to make, given the existence of unknowns in the data, and so provides a set of possible ways things can pan out depending on circumstances. Seems to me like if anything it’s the other side that tends to be extremely invested in denying the existence of ANY problem that would require collective, coordinate action rather than admitting that pure individualism and the emergent properties of free market aren’t tools enough to solve literally any crisis.

          19. Names, dates, references?

            Which you yourself are completely omitting?

            It is this sort of bad faith argument that has heavily contributed to the complete lack of trust in your side.

      2. I’m not sure how cheap working fusion will be to actually implement; a proof of concept plant that costs multiple billions of Euros isn’t a good sign. The main thing about fusion is that it would be inexhaustible for the foreseeable future: deuterium is recoverable in amounts that 1000x our present energy usage wouldn’t greatly deplete for a million years or longer (the necessary lithium for breeding tritium might be a bottleneck, but more advanced designs might eventually dispense with it).

  30. Thanks for this great series! One question that this article reminded me of: Why are inventions invented when they are and not earlier? E. g. why did the Romans not have windmills? I assume it is a combination of other technical advances, similar to how (from my limited understanding) many inventions during the industrial revolution were only possible due to the increased steel production which was possible due to the warfare that needed it which was possible to to hire state power.
    What were these stories for the core inventions of human history? E. g.
    – the compass
    – taming of the horse
    – gun powder
    – printing press

    1. The Romans had a design for a functioning steam engine availible to them and the Phoenecians had movable type printing, they just dient need either. The Romans had a massive population of laborers and a huge number of slaves which meant the steam engine’s labor saving capabilities were not as important. At the same time, the Romans had not developed charcoal yet and did not have as.large a coal mining industry as, say, early modern england.

      By comparison, england was a country with a large population surplus but no slaves, with an economy largely built on the sale of textiles, and a large domestice coal industry. There was also a functionally infinite market for more cheap textiles, so any increase in productivity would be rewarded. The conditions in England were perfect for industrialization, those conditions did not exist in Rome (nor even in most of France until much later).

      1. If by functioning steam engine you mean the aeolipile, that was a gimmick or toy that could never have been a working industrial steam engine. It’s comparable to thinking that a radiometer ( could be a practical or economical way of harvesting solar energy.

      2. I got interested in the claim that the Phoenicians invented movable type printing, but couldn’t find any source. Do you have a reference?

      3. “The Romans had not developed charcoal yet?” Didn’t Dr. Devereaux do a whole series of posts that extensively discussed Roman iron working, and iron working in general, in which he pointed out that you must use charcoal or mineral coal to smelt iron ore, because a wood fire doesn’t get hot enough? I’m pretty sure you’re wrong about the Romans not having charcoal, though I don’t see why it’s relevant to what you were talking about.

        As to coal mining, the Romans had deep mines. Quite a few of them. A reliable engine for pumping water out of mines or something similar would have been desirable to them, because endlessly working slaves to death doing it (or having oxen turning capstans) presents a lot of cost problems in its own right. Enslaved labor isn’t free; the employers have to pay to get ahold of the slaves, pay to keep them in chains, and feed them for as long as they live in order to get any work out of them.

        What was missing in ancient Rome wasn’t the economic incentive to build a steam engine. It was the metallurgy to build large, reliable pieces of metallic machinery that could hold steam pressure and extract mechanical work from it. Something Early Modern Europe had, not because of a lack of slave labor, but because of cannons.

        1. Perhaps what he was thinking of was that _coking_ hadn’t been invented yet: the process of converting raw coal into a refined furnace fuel. Without this mass production of iron by blast furnaces isn’t practical.

    2. My guess is that a major root of most progress has been some variation on writing: the ability to transmit information remotely across space and time. At least in modern times the history of invention is less about lone geniuses and more about several different people tinkering with an idea. The first few iterations fail or are not practical, but someone proposes an improvement which gives someone else an idea for a further improvement, and so on. All made possible by long-distance collaboration or the records of previous experimenters.

      And for that you have to have something to write on: The Egyptians invented papyrus scrolls while the Mesopotamians carried inscribed clay tablets as far as it could go. The Chinese developed paper, which almost certainly was responsible for the precociousness of Chinese civilization. It made possible a long recorded alchemical tradition that eventually produced gunpowder. There were lots of precursors to the printing press but it seems that Johannes Gutenberg was the first to combine several different elements into a single winning invention- which was able to be as pivotal as it was because scarce and expensive parchment was replaced by paper introduced from the East.

  31. The attribution of Roman Population to increased trade which broke down due to incessant civil war makes sense. European population and complexity seems to rise quicker in the 12th and 13th centuries, potentially linked to the increase of trade following the crusades. The establishment of Venetian, Pisan and Genoese merchant quarters in the holy land likely acted as a pump powering trade between the middle east and Europe.

    The Byzantines and Abbasids although having naval superiority at times didn’t really branch trade to europe, muslim corsairs especially performing raids against southern europe. But the Merchant cities reconnected them in trade despite heavy piracy against each other.

  32. This definitely answers the questions I had last week.

    It seems to me the urbanization of the Roman empire (and subsequent loss of it, even in much of the East) shows that cities as centers of trade, manufacturing, and knowledge had a synergistic effect even before industrialization. Now, of course, we are urbanized to an extreme and perhaps unsustainable level, but from the evidence it sure seems cities aren’t just a drag on some idealized rural society.

    1. Well from a government perspective, modern cities produce wealth and rural communities consume wealth. The cost to provide services – government or civilian – is much lower to 400 people living in a single building then 400 people spread over a square mile. But, the revenue for those services is generally the same.

      I’d imagine that even in ancient world this advantage to scale between cost of services and price extracted for those services ment that cities drove trade and almost all comparative advantage.

        1. This is true but only 3% of the population, at least in the US, are actively farming. Many people live in rural areas for lifestyle reasons.

          And if you discuss this with government folks, it’s not really something they miss. Half of government is just trying to get service to rural areas: fire, med, school, transportation, and police and the, especially, the whole suite of civilian services to rural areas in a society politically opposed to government ownership of business.

          My state just opened the first new medical school in the region in probably close to 100 years in the hopes that a med school in a midsized city at the center of an agricultural region will encourage new doctors to work in rural hospitals and practices.

      1. I’m not aware that per capita government expenditures are lower in cities. An urban government needs to provide more parks (rural residents have de facto parks all around them), more policing (in an anonymous urban environment, neighbors can’t police each other), more public transportation (it isn’t feasible for every city resident to do all or most of his travel in a private car), etc. That’s why taxes are much higher in New York City than in Maine or Montana.

        1. Rural areas do get by with providing less services, but I think you miss a little on how expensive providing say transportation is to rural areas. Roads deteriorate regardless of whether they are being used constantly and it’s much easier to extract the funds to maintain roads when it carries 50 cars an hour rather than 50 cars a month. Most states build their transportation budgets around “donor” counties, the one to two counties per state with populations excess of 350k, and “reciever” counties, all the other counties where local taxation is insufficient to cover routine maintenance and repair.

          Probably the largest part of the taxation difference between New York and Montana is radically different views on the role of government in the economy. The highest cost to the government is exurbs where the citizens demand the full suite of urban amenities while disconnected from the urban core.

          My home state of Washington has 5 donor counties centered around the cities of Seattle, Spokane, Everett, Tacoma, and Vancouver, and 34 reciever counties. And of those five donor counties only King County around Seattle produces enough revenue to pay for stateside transportation. The other 4 are pretty close to break even.

  33. I’ve seen a theory put forward that the romans also suffered from an energy crisis, as over the centuries they deforested the mediterranean basin for fuel. Devereaux in his series on metalworking mentions how elba was a site for ironworking. And after they chopped down every last tree on the island, they had to ship it in. While it might be a bit of a stretch, I think you can draw a parallel between the roman empire running out of wood (and other easily accessible resources close to the earth’s surface) and the modern era where we run out of oil and we have to dig deeper for new resources.

    Sadly I cant find a source of it. searching for Roman empire and energy crisis mostly turns up other crises. Although I did find one short radio article on energy ( Although they don’t talk about running out of wood, but rather running out of slaves. As they stopped expanding the inflow of slaves stagnated, and thus the cost of energy (in this case muscle power of slaves) rose.

  34. Reading the conclusion, I can’t help thinking about that Midnight Oil song: “our beds are burniiiing…”

    Thanks for this series, a pleasure to read as always!

  35. Maybe Western Europe is just as ungovernable as Afghanistan (certainly it is the graveyard of empires), and the Romans were eventually tired of ruling it (certainly climate, plague helped) and left.

  36. The Wikipedia article on the Antonine Plague (~170 AD) contradicts some of the claims here about Romans being better fed than pre-Romans or post-Romans:

    There are references there as well which show that skeleton height dropped in Roman times, and recovered quickly following the collapse of the Western empire.

  37. So basically, your thesis is that the collapse of trade is what caused crushing poverty. Which makes sense, really. We see this pattern over and over in history. That’s probably what caused the Bronze Age collapse, and it’s definitely what kicked off the Great Depression.

    (That one is multi-causal when you look into the details of why it was so long and so deep, but the timeline of the stock market crash that kicked it off is intimately tied up with the progression of the Smoot-Hawley tariff through Congressional votes – see for a timeline here. And yes, I know that’s a very ideological source, but the dates and stats are accurate.)

    We also see the pain inflicted by autarky in the modern world. As a rule, states that try to sever themselves from the global trade network, or who are forcibly severed by sanctions, get substantially poorer as a result. And of course, the flip side (trading-heavy nations being very rich and prosperous) is scattered all over human history – whether it’s the Athenians 2500 years ago, the Dutch 250 years ago, or Singapore today, high levels of trade is strongly associated with prosperity. And that’s for the nation as a whole, not just for a handful of traders.

    So yeah, this theory matches what I know of economics and economic history very well. Good to see that it also seems to be roughly the median explanation in historical literature, if I’m reading you right.

    1. ” The collapse of trade caused crushing poverty” – yes, to the beneficiaries of the trade. Perhaps not so much to the losers. Looking in toto, trade itself caused crushing poverty (and worse) from the C17 to C20 – in Africa, the East Indies, India and much of the Americas, as people were coerced into deeply exploitative relationships. Much of the gdp growth in Europe and north America pre-1914 was looting – of Indian and Chinese labour, Australian pastures, American prairies and forests. In the Roman case, the main commodity the barbarians supplied up to the 3rd century or so was slaves (as one historian noted, there’s a reason the modern German word for a merchant derives from the specific Latin word for a slave-trader). This is not to argue for autarky, but to note that ‘trade’ is often not the simple mutual gain proposed in economic models.

      1. What German word is that? The word for merchant that I know is Kaufmann, which Wiktionary says is from Latin caupo, meaning a tradesman, shopkeeper, etc.; no mention of slavery.

        1. In Peter Heather’s Empires and Barbarians. P 75 of the Pan edition: “Germanic languages have as one of their basic word-stems for trade and merchants a series of terms derived from the Latin ‘mango’. But in Latin ‘mango’ means not a merchant in general, but very precisely a slave-trader.” My recollection was a little at fault.

      2. It’s important to remember that, with the exception of the African slave trade, most of the harm that Europeans did to other societies was by holding them economically and technologically stagnant while Europe industrilized.

        Most of Europes colonies were a money losing proposition for the metropolitan core. The Europeans were just mentally locked into an approach of expansionist governance from the feudal system which they never re-assessed.

        Before the opinum wars the East India Company gained more profit from a single trading factory in Canton than from all of Bengal and India. Basically a single building housing merchants in China was more valuable financially then all of conquered India.

        1. Long run harm. The conquering and attempting to succesfully exploit was incredibly violent.

          The difference I’m trying to draw out is absolute wealth v. relative wealth. Most of Europe’s absolute wealth came from internal industrial trade, the relative wealth difference came about because Europe prohibited, by force of arms, the rest of the world from industrilizing.

          1. I’d be cautious about one thing here: it is not clear to me the rest of the world would have industrialized rapidly even without the disruptions of European colonialism. The success story here is presumably Japan, but Japan was hardly caught up even by the 1930s in terms of industrial output. Meanwhile, industrialization pre-WWII in places like China and Ethiopia which had remained relatively independent wasn’t particularly fast. Likewise, the Ottoman Empire pre-1920 had industrialized only very, very slowly.

            I think there’s a strong argument to be made about the role of colonialism in providing the resources and wealth necessary for industrialization in Europe (but then one can hardly argue that Qing China and the Ottoman Empire were not also the beneficiaries of their own imperial systems), but ‘Europe prevented the rest of the world from industrializing by armed force’ is a stretch. Pre-industrial social systems were surprisingly stable, even stubbornly resistant to change, even after being exposed to competition with industrial production.

        2. Really? The Portuguese, Dutch, English and French fought vigorously for 200 years to control trades that were of no profit to them? The opium that was shipped to China was grown in Bengal; the single largest global trade in the C18 was in Indian textiles; from the C14 to the C17 spices routinely returned trade profits over 100% (the Dutch enforced spice cultivation at the expense of subsistence across the East Indies). The most profitable trade in the C18 was sugar, hence vicious fights over small Caribbean islands, but also over the associated African slave trade (labour) and Indian cloth (exchanged for slaves).

          1. There is a distinction here between “makes money for the metropolitan core” and “makes money for the elites who gets jobs administering them”.

            Generally the argument tends to be that the West Indies and the Spice islands were profitable (though the latter only briefly) while african colonies (as well as India itself) was either a wash or not profitable (in terms of state revenue, *individuals* made bank from all of them) What unites them is the fact that they tended to be relatively easy to administrate and defend, and have huge returns. (You can also tell how relatively unwilling european powers were to expend resources on colonies, when you compare to how mcuh they spend on european wars)

      3. This significantly overstates the case. Practically every economic study ever done indicates that when Europeans didn’t wipe out the local societies, said societies generally stagnated, or improved at a lower rate than they had been. Actual regression was an extreme rarity.

        1. Paul Kennedy in “The Rise and Fall of the Great Powers” claims that India de-industrialized on an absolute level: local cloth manufacture was mostly driven out of business by cheaper and better machine-made cloth from Britain.

          1. But India was not, in 1945, notably poorer than China. The real harm the British did to India was to fill the heads of its indigenous elite with socialist and social democrat ideas utterly inappropriate for a developing economy, so that (for instance) India has labor protections similar to those of Western countries, but their primary effect is to inhibit the expansion of businesses above the size of a family firm.

          2. India’s lead was in dying and printing as much as weaving. Britain set out to displace India in these, using control of trade, tariffs and an active industrial policy. Machine weaving was the icing on the cake.

      4. Oh, there was a ton of exploitation through history – imperial expansion, in any era, tends to be grossly exploitative, whether it’s the Roman Empire, British Empire, or Putin’s hopefully-not-actually-pending invasion.

        But it isn’t generally the trade that exploits people. It’s the conquest and subjugation. For example, it’s not trading slaves that exploits people, it’s the fact that they got enslaved in the first place that exploits them.

        1. You have to look at the whole chain. Why did coastal African states like Ashanti come to depend on the slave trade? Because there was a major market overseas and the export paid for cloth and metals and weapons. Absent the market, slavery would have stayed small-scale and local. Again, why did the Dutch coerce Indonesian and Sri Lankan peasants into cultivating spices – at the expense of their diet and health? Because there was a huge profit to be made trading spices to Europe. Ditto Bengali weavers, Congolese farmers and many, many others. It’s the trade opportunity that drives the slaving.

  38. The result, in ancient history, has been what I tend to refer to as ‘the revenge of the archaeologists.’ Not, mind you, revenge on medievalists, but in fact revenge on a very specific ancient historian and classicist, Moses Finley.

    That kind of sentence promises a story that is either fun or infuriating (but hopefully with a cathartic conclusion). And in this case, it sounds like the “dunking on a posh a-hole” kind of fun.

    This is, by the by, an example of a simple Malthusian interaction breaking, because the classic Malthusian logic assumes that agricultural production is primarily a function of land…

    The version I’d heard proposed that it was a function of land and technology, and that technology increased productivity only linearly (compared to the exponential growth of human populations under ideal conditions).
    Now, I don’t know how close that “productivity per unit land increases linearly” claim stands up to scrutiny, but it still ignores some important factors. It doesn’t matter if you have a factory that can produce nigh-infinite quantities of artificial fertilizer, if the trade between your area and the places supplying raw materials for that factory breaks down.

    A system like that…could potentially be very fragile.

    I’d call it “brittle” before “fragile”. An economic system where farmers are producing more surplus per capita, and where buying food from distant areas is normal, would be more resilient against many events which could cause devastating famine in more isolated, self-sufficient communities. Of course, when something hits the system hard enough to break it, it breaks hard…and the metaphorical arsonists within the system can hit quite hard indeed.

    To me, “fragile” just means “easy to break,” while “brittle” means more precisely encapsulates that idea of “rigid but unyielding,” of breaking completely or not at all.

    …and often with our own arsonists, who are convinced that the system is durable and stable because they cannot imagine it ever vanishing.

    “It is easier to imagine the end of the world than the end of capitalis[tic-democratic world order].” And this belief could easily cause both…
    And the use of an arsonist is a particularly potent metaphor for anyone who watches PhilosophyTube, of course.

  39. ” Roman crop rotations may, in some cases, have been closer to the late medieval three-field system than the early medieval two-field system”

    One thing that I remember from my historical demography days is that the notin of a uniform movement from one to the other seems a bit flawed, that is to say, people often moved from three-field systems to two-field systems depending on local circumstances. (like access to fertilizer)

  40. It’s interesting to see how views of the Fall of the Western Empire have changed in recent decades. The traditional view would seem to be that the Roman political order disintegrated in the West, and that the result was an unmitigated disaster for its inhabitants, even if some elements of the later Empire, like Christianity or the Latin language, survived in some form.

    But now, after the work of the Change and Continuity people, we see that the Roman political order disintegrated in the West, and that the result was an unmitigated disaster for its inhabitants, even if some elements of the later Empire, like Christianity or the Latin language, survived in some form.

    That is not a snarky as it sounds. Any discipline attempting to discover the truth about something should steadily produce more accurate representations of it so the result, over time, should be better-focussed representations, not wildly different ones. A discipline with views that change out of recognition every generation or so is probably very little constrained by reality. If History is any good at its job, the past shouldn’t change very much. So I find this rather reassuring.

    1. I think you are right in what you wrote. What may have changed is the understanding of *why* the Roman political order disintegrated. The old answers were barbarians, Christianity (GIbbon!) or lack of virtue. The new answers are less clear and would maybe lie in some combination of “at some point everything disintegrates”, the catchup of the non-Mediterranean parts of Europe (Halsall) and economic mismanagement.

      1. I suppose how things happened will always be less clear than what things happened. And maybe it shouldn’t need much explaining: If the Roman Empire were a very unusual society in it’s day and age, we should not be surprised if it was replaced by more normal ones.

      2. Don’t forget Civil War. If people weren’t constantly rising up in rebellion, the story is probably different.

  41. For Bret, or anyone: Are there revised estimations of total population in the regions discussed over time? I know there were the caveats briefly discussed but I would be extremely interested in seeing how population declined and recovered.

      1. Clearly, the local magnates, whether that is landlords or officials. Whether the Roman state ordered the executions or just let them happen is the kind of question archaeology can’t easily answer.

        1. Almost no society in history has been without violence, and certainly no large or economically sophisticated society has been. But there’s a difference between “killed by a bandit/invading army” and “killed by the magistrate because they broke the law.” One is chaos and random, and the other is done to avoid chaos and create a stable way of life. And chaos almost universally leads to more death and suffering.

    1. “The recent discovery of a Late Roman cemetery where a noticeable share of the deceased had been beheaded also challenges the idea that Rome brought peace ”

      I’m not so sure about that argument: It sounds more like the graveyard for an execution ground. Executions in 19th century continental Europe were typically by the guillotine: presumably many countries there have a graveyard somewhere full of decapitated criminals. If someone were to dig up such a graveyard in, for example, Holland, would that challenge the idea that Holland was pretty peaceful?

    2. Death by beheading is almost always judicial, although the ‘court’ may be ad hoc or even itself illegal. Conflict deaths are messy, and it’s incredibly rare for even a trained fighter to hit someone hard enough to dismember them without getting themselves stabbed in turn.

  42. I don’t have much to say about the article as a whole (other than, another great read as usual), but in the spirit of unmitigated pedantry and constructive criticism, I want to point out that that plot of femur lengths is an excellent case study in how NOT to present data.

    The most glaring omission, of course, is the lack of any sort of uncertainties. (The very first thing drilled into my head by my academic advisor.) A plot without error bars is worse than no plot at all because it can hide all kinds of important information about the spread of the data (not even maliciously or on purpose, just because). Not showing uncertainties doesn’t make them go away, it just prevents viewers from making a fully-informed decision about the data presented. What’s the variance in each of those bins? We don’t know. How many data points went into each bin? We don’t know. What’s the uncertainty on each individual measurement? We don’t know. The whole plot only covers a range of about 2 cm, so knowing the spread of the points is critical to see if there’s a real trend or not—if inter-individual variance is on the order of 2 cm already, then it’s a lot harder to claim that there’s a trend there. It’s entirely possible that with all that information presented, the apparent pattern might fade into statistical noise. It’s ALSO entirely possible that with all that information presented the pattern would be reinforced and our faith in its conclusions strengthened; I’m not accusing the plot’s author of duplicity, merely poor statistics.

    There’s also a distinct lack of discussion of any possible sources of systematic error that might affect such a plot in the original paper. (I wouldn’t require such a discussion in a popular article like this one, of course, but it bugged me enough to look up the source.) I’m an astronomer, not an anatomist or historian (both of whom I’m sure could easily come up with many more potential sources of systematic error), but my immediate thought upon seeing that plot is wondering how the author has accounted for human sexual dimorphism. Simply measuring different ratios of male/female femurs in each date bin could conceivably reproduce the pattern seen. Again, maybe this has been accounted for, or maybe it can’t be accounted for if the sex of each individual isn’t known—but that’s important information for a reader trying to interpret the plot! Ideally, I’d like to see a version of this plot where all the data points in each bin are plotted (with their individual error bars), broken out into male/female/unknown, with the means for each category plotted to see if they agree with each other. That’d be vastly more useful in assessing the likelihood of what this plot seems to be saying.* (Also, always label your axes, with units, even when you think the context is clear; it costs you nothing and your viewers will appreciate it. [Thanks to Dr. Devereaux for providing them in this case!]) I did notice in a brief skim of a recent paper by the same author that results were being broken out by sex and analyzed in more detail, so there’s always hope for improvement. 🙂

    To be 100% clear, I’m not necessarily arguing against the conclusions of this plot; it may very well be that, beneath all the obfuscating plotting choices, the plot is indeed showing something real (though almost by accident, as it were). There are multiple lines of evidence that all seem to show similar things about conditions during this period of history, so I personally find it likely that the plot is showing a real effect; I’d simply argue that we can’t actually draw any firm conclusions from it in isolation in its current form.

    *To give a personal example of the kind of analysis I’m talking about: for my first academic paper, I literally wrote an entire second paper (~50% longer than the first) dedicated to the analysis of over a dozen possible sources of systematic error in my methodology (with nearly as many additional plots), all to justify the (single) plot of the main results of the first paper. (And then in my thesis defense, one of my examiners casually mentioned a few more sources that neither I nor my advisors had thought of it four years—reality is messy!)

    1. I am an ancient historian, and all the surveys I have seen show that health improved measurably in the early middle ages/post-imperial Europe. Deceased were taller and had fewer parasites and traces of chronic illness in their bones. In Britain, Roman peasants are very hard to find in the archaeological record- presumably they were not using the nice pots or living under tiled roofs.

      Maybe Bret Devereaux has seen evidence I have not (and not just someone with a minority view) but the bit about femurs above (which I missed when composing my first comment) is contradicted by the state of the question as I and the specialists I have read understand it.

      1. There is one potential difference here (and something that also shows up a bit with the “healthy hunter-gatherer” thing) in that it might simply be trading high morbiidty for high mortality: IE: People dying while they are in good shape, rather than after prolonged illness. If people die early they’d expect to have fewer illnesses and such, for instance.

        1. This was my thought. Under a more comfortable set of living circumstances, those who are less fit (in the evolutionary sense) have more opportunity to survive, and for longer.

          An interesting case of the data being somewhat counter intuitive at first glance is in the recent (past few years) reconfiguration of stroke services in the NHS which has significantly improved the health outcomes of people with strokes. However, what the data is showing is that more people are having strokes than before (because where previously a stroke may have killed someone, they’re now surviving for long enough to have another one).

        1. And in fact, the paper I cited in “The Key Question in the Fall of the Roman Empire” is by Jongman (and a list of co-authors) but it is newer, described the height and health of human beings not cattle, and fits everything else I have heard from talking to specialists, hearing their lectures, and reading bits and pieces of their popular books. I could always be wrong since its not my field, but Dr. Devereaux takes for granted that health decreased in a time when every other scholar I know says it increased.

          1. It may be that relative health increased while absolute health did not. My understanding is that until modern medicine and especially modern sanitation, cities generally had signifigantly less healthy population that rural areas. One of the effects illustrated is that the urban areas collapse and the population becomes more rural.

            The collapse could be bad everywhere in the ‘starvation and death is bad sense’ while leaving an overall more healthy population in the ‘rural communities than can deal with thier poop are healthier than urban communities that produce mass quanities of poop.’

      2. I’ve now read the “Physical wellbeing in the Roman world” version 2 paper by Scheidel referenced in your blog post.. (Because it is online and free – I can’t afford the others.)

        The paper goes into much finer detail on very specific aspects, as would be expected from an academic paper rather than a blog post for a general audience, but I’m not seeing an elephant in the room or direct contradiction. From the conclusion:

        “It appears that the imperial economy did not generally enhance biological living standards. Physical wellbeing was unevenly distributed, with more benefits accruing to peripheral areas than to the core.”

        My understanding is that most of western Europe was peripheral to the Roman Empire? So the fall of the Western Roman Empire causing a reduction in physical wellbeing is what would be expected?

        This is probably not the right place to discuss someone else’s history blog. For those interested check out Sean’s blog and other writings:

        I will add that there seems to be a *third* school of thought, that the decline and fall of Rome was A Good Thing. Bret, perhaps a topic for part IV?

        1. Apologies, I haven’t read the source. Do they mention the confounding variable of elite Imperial Roman burials tending to be cremations rather than burials which excludes the healthiest proportion of the population from the analysis?

          It seems like a variable that’s controllable for. Say, excluding the elite post-Roman burials from the data set, or establishing the difference in elite-to-non-elite height difference in what data we have available (either post-Roman or Roman) and applying that to the Roman sample.

          1. As Bret often reminds us and this paper agrees, elites usually have many more details of their life written down. So average life expectancy can be estimated fairly well, even without the remains. (According to the paper, Roman elites did not on average live longer.)

    2. Being a layman, I find myself wondering just what the population is whose average stature is being studied here. I gather city dwellers tended to be shorter than country dwellers. So if urban population numbers fell further, as seems to be the case, average height of the whole population would be expected to go up, and might do so even if the average height in both city and country declined.

  43. It occurs to me that the Indestructibility of Pots means that, 2000 years from now, the only surviving Far Side comics will likely be those on coffee mugs.

    It is deeply… interesting… to consider how important coffee mugs will be for far-future archaeologists trying to understand our era.

    1. Crazy theories about coffee mugs and various small metal objects are running gags in the video game “Horizon: Zero Dawn.”

    2. “It would seem that the core faith shared by the suburban population of early C21 focussed upon a trinity of household spirits, representing vitality (Live), passion (Love), and humour (Laugh). The names of these spirits recur with predictable regularity on both ceramic drinking vessels, indicating some form of imbibery cult and the sacred nature of the coffee bean, and metalwork ornamentation displayed on prominent walls of the household.”

      1. Honestly, truth in disguise. You’d probably learn more about the mass culture, mindset and spirituality of Western countries today from vapid self-help quotes than from the holy texts of Christianity that is still nominally the majority religion.

  44. One of the major problems I think is that of separating cause from effect, did the economy decline because the roman empire did or did the roman empire decline because the economy did? (or of course, both)

  45. On literacy, I recently took a 100-level “History of Technology” course, in which the professor argued that literacy was very high among all classes in the Roman Empire. In particular, he talked about the Vindolanda letters, saying that there were letters written by slaves to other slaves. So I guess my impression was that literacy rates in the Roman Empire were quite a bit higher than what you presented here.

  46. Thanks very much for this fascinating series. I’m inclined to give more credence than you do to Peter Heather’s analysis in “The Fall of the Roman Empire” – specifically, that the long-term interaction between the Empire and its “barbarian” neighbours, transforming the latter into larger units and far more formidable fighting forces, was crucial to the Western Empire’s disintegration. After all, Roman competitors for power placing their own interests before that of the Empire was hardly new in the 5th century, or even the 3rd, and extensive civil wars resulted from this in the 1st century BCE (from Marius vs Sulla to Octavian vs Anthony), the 1st century CE (the “Year of Four Emperors”, and the late 2nd century CE (the “Year of Five Emperors”). In none of these cases, IIRC, did the combatants try to recruit “barbarians” to do the fighting. Why? The obvious answer is that there were no barbarians worth recruiting, because they were not organised into sufficiently large, disciplined and trained forces. Heather extends this analysis to the spread of state-level organization into northern and eastern Europe during the 1st millennium CE in “Empires and Barbarians”, and I wonder if it could not also be applied to the rise of the first Islamic Empire and the concomitant complete destruction of the Sassanid and partial destruction of the East Roman Empire – Arab allies/mercenaries having been extensively employed by both sides in the Roman/Sassanid wars. The tendency of empires to bring about transformations in their neighbours can also be seen in the 19th century history of Japan, and the 20th century history of China – and even in the effect of European colonization of North America on native communities (as shown in Alan Taylor’s “American Colonies”).

    1. “the long-term interaction between the Empire and its “barbarian” neighbours, transforming the latter into larger units and far more formidable fighting forces, was crucial to the Western Empire’s disintegration. After all, Roman competitors for power placing their own interests before that of the Empire was hardly new in the 5th century, or even the 3rd, and extensive civil wars resulted from this in the 1st century BCE (from Marius vs Sulla to Octavian vs Anthony), the 1st century CE (the “Year of Four Emperors”, and the late 2nd century CE (the “Year of Five Emperors”). In none of these cases, IIRC, did the combatants try to recruit “barbarians” to do the fighting. Why? The obvious answer is that there were no barbarians worth recruiting, because they were not organised into sufficiently large, disciplined and trained forces. ”

      They are slightly reported as trying.
      Sertorius tried to civilize and organize the Spanish barbarians.
      Catilina was accused of plotting with Gallians to invade Rome.
      Julius Caesar recruited units of German mercenaries – including his cavalry bodyguard.
      And while Egypt was expressly civilized, it was alien for Romans.

      But now look at the limitations.

      Sertorius ended up losing. Catiline´s plot came completely to naught and he died at head of Romans alone.
      Caesar won, but most of his supporters were his Roman legionaries – his German mercenaries were too few in number to be important political players.
      And Egypt, as stated, was civilized.

      The barbarians could not conquer Rome in 1st century BC even with support of a faction of Romans because they lacked the organized force to do so – even if they won, their Roman allies would have been the winners. And Cleopatra, Mithridates or Hannibal, had they won, probably could have run an empire as complex as Rome.

      The barbarians of 5th…6th century AD were sufficiently organized to take Rome down – and yet not organized enough to keep Rome running.

Leave a Reply