The Barbarians Did It
Working out the reasons for the fall of the Western Roman Empire in the C5th is a hardy perennial. As deathbeast correctly pointed out in a comment on a recent post of mine: The problem with the debate about the fall of the Western Roman empire (or 'transformation' in more postmodern terms) is that you get a real lack of information and evidence. Which creates a large vacuum. Into which folk project their own ideologies, pouring into the historiography like barbarians over the frontier.
Analysis of the fall of the Western Empire has to pass various tests, starting with fitting the available evidence (which archaeology has filled out to a considerable degree). It also has to be compatible with the continuation of the Eastern Empire. It has to identify what was different about the fatal crisis of the C5th compared to the (survived) crisis of the C3rd. And it also has to explain why the Western Empire succumbed in the C5th when the (much smaller) Roman Republic survived the threat of Hannibal in the C4th BC.
Blaming Christianity a la Gibbon fails the ‘Eastern Empire survived’ test. As does blaming the reforms of Diocletian or of Constantine the Great or resource depletion (whether silver mines or forests). Citing climate change has to explain why the Western Empire suffered worse than its opponents. Making the battle of Adrianople some sort of marker has to explain why the loss of the major field army of the Eastern empire led to the collapse of the Western empire decades later.
Peter Heather teachers at Worcester College, University of Oxford. He has written a very thoughtful and informative account of the fall of Rome, The Fall of the Roman Empire: A New History of Rome and the Barbarians which manages to pass these tests.
Like Ward-Perkins, Heather zeroes in on the superior geographical position of the Eastern Empire to explain its survival. While its Balkan provinces were ravaged by barbarians, notably by both the Gothic incursion of the 370s which led to the disaster at Adrianople and by the Huns in the 420s and 430s, the land walls of Constantinople and the water barrier of the Bosporus meant that most of its tax-producing provinces were not subject to barbarian incursions. This explanation gains added weight when one considers the disaster of the Arab explosion of the C7th from a much less easily defended frontier.
If geography is crucial, then simple internal decay is not. Heather argues strongly and, when conjoined with the archaeological evidence that Ward-Perkins assembles some of which Heather also refers to, convincingly that internal economic decay was not a noted feature of the Roman system. He contrasts the Western Empire with the Carolingian Empire as one that did fall apart for internal reasons.
What is particularly attractive and informative about Heather’s analysis is careful consideration of what was going on outside Rome’s borders. Again, the increase in archaeological evidence provides a much better basis for historical understanding than has existed previously.
The first process was a lessening of Roman economic primacy. Interaction with the Mediterranean economic world seems to have stimulated increased economic activity beyond Rome’s imperial borders. The technological sophistication and population levels of the various barbarian tribes seems to have been definitely increasing, particularly by the C4th.
This process also saw a lessening of Roman technological superiority. Something that made the Hunnic incursions from the 420s onwards such a shock is that the Huns were, from the start, able to take walled cities and forts, something previous barbarian incursions, such as that of the Goths in the 370s, had not been able to do. Barbarians, such as the Vandals under Geiseric, who could take walled cities were much more of a permanent threat than those who could only ravage the countryside.
The Roman Empire was also the Sole Big Target. It was by far the greatest and most extensive concentration of wealth around. This made it a perennial target of the border-raiding that empires always suffer. But if larger groupings could be put together, then the Empire was the desirable happy hunting ground in a much more extensive and permanent way.
There were two features that particularly encouraged such combining beyond the increased chance to get hold of Roman wealth. One was the eruption of the Huns onto the Caucasus steppes and then the Hungarian plain. The other was a reaction to long-standing Roman habits of punitive retaliation and playing groups off against each other.
Heather sees the rise of the Sassanid Empire, replacing the more loose and quiescent Parthian Empire, as crucial to causing crisis of the C3rd. What the crisis showed was that “Romanisation” had created the need for the spoils of patronage across the entire empire if usurpations were to be minimised. This, combined with the slowness of communications and the need to deal with a rival Persian ‘superpower’, is what led to the division of the Empire into Western and Eastern halves. Heather is particularly informative about how the Roman system functioned. How social systems work is obviously of great interest to him, there are some very perspicacious and informative discussion of the problems of control in a pre-industrial agrarian empire, the nature of landowner politics, the nature of nomad pastoralist society and politics.
He sees Diocletian’s massive expansion of the bureaucracy as more a replacement of political forms (primarily local city politics) than some great burden.
The rise of a Hunnic superpower had a range of effects. First, it drove the Goths into their 370s incursion. Valens’ military disaster at Adrianople was, for example, not a significantly greater defeat than the destruction of three legions at the Teutoburger Wald. The more serious effect was that it was one of series of defeats on long-standing Roman territory that led to a far less complete victory than was the norm in that the Goths retained independent existence within the Empire. A sign that the Roman advantage over the barbarians was declining.
The Huns seem to have been similarly responsible for encouraging the various incursions in the first decade of the C5th which led to the establishment of various unassimilated barbarian groups within the Empire.
The various direct Hunnic incursions from the 420s to 440s caused further damage and, of particular importance, frustrated a serious attempt to reconquer North Africa-a key source of revenue for the Western Empire-from the Vandals. The Hunnic Empire also concentrated various barbarian groups close to the Empire and, when the Hunnic Empire collapsed, had created a region where habits of combining together for common military action had been established. Habits that various groups then engaged in to seize the wealth of the Empire.
In the end, it was a case of one damn thing after another. A particularly successful Roman general, a Flavius Constantius or an Aetius, could stabilise things for a while, but there was always something else coming along.
Once a final joint Western-Eastern attempt to retake North Africa had failed in 468, there was no longer the tax base to support a Roman Empire in the West and even the pretence of such a thing came to be seen to serve no purpose.
I have a couple of quibbles. First, although Heather admits that the level of social collapse in Britain was precipitous, he seems a little too sanguine about how much continuity there was in the rest of the Empire, though his point about military prowess replacing literacy as the basis for political advancement is well-taken. The barbarian kingdoms clearly did want the goodies of the more successful society they were occupying. This is a common feature of newcomers drawn to such a society. But retaining said goodies is only possible if the newcomers replicate and reinforce the patterns of behaviour which produce those goodies. If, on the contrary, they continue the patterns of behaviour which produced the less successful societies they are coming from, they will undermine (even destroy) the benefits they are after. Which, despite the initial adoption of Roman government forms, is clearly what happened if less precipitously than occurred in Britain.
Second, Heather’s comment at the end about the barbarians becoming more effective as a result of ‘unbounded Roman aggression’ (imperialism as its own rewards) seems a little too precious. That imperialism may have a tendency to create more effective opponents is a striking idea. But Roman punitive expeditions, no matter how nasty, were rarely, if ever, gratuitous. The consequences of failing to maintain dominance over the barbarians were sufficiently horrific to suggest the Romans may have had a point in such policies.
On the other hand, though he does not consider the comparison directly, there is more than enough in Heather’s analysis to pass the test of why the larger Western Empire was not able to survive its terminal crisis when the, much smaller, Roman Republic was able to survive and, eventually, triumph over Hannibal.
Maintaining Roman rule over Western Europe for 450 years is a hugely impressive achievement. But it is also clear enough that, in a crisis situation, the late Empire had less comparative ability to mobilise resources and contain political rivalry than did the Roman Republic at the time of Hannibal.
One of the damn things that kept happening was murderous political rivalry within the Roman system, up to the point of civil war. Civil war was a periodic feature of the Roman system for centuries, right back to Marius and Sulla. But continuing the habit while repeated barbarian incursions and occupations were occurring was not conducive to survival. The imperial system so concentrated political power and its benefits that there was no forum for managing tensions except fighting it out until some strongman emerged victorious.
The other was the military implications of the specialisation Ward-Perkins draws attention to. Military skills were essentially concentrated among paid professionals. If local field forces were defeated and, over the longer term, the tax base to support the paid professionals was undermined, the Roman citizens were extremely vulnerable. This was quite a different social contract than that which had sustained the Roman Republic. There citizenship and military prowess went hand-in-hand- the “deal” of citizenship was military service in return for legal and political rights. As a result, the Roman Republic could, within its much smaller population, draw upon a proportionately much larger pool of motivated and skilled soldiers. In a prolonged crisis, this-along with the ability to contain and manage political tensions-made the (much smaller) Roman Republic of the C4th BC much tougher than the much larger Western Empire of the C5th. Which is not to deny that Western Empire still showed considerable powers of resilience. Just not enough to deal with one damn thing after another.
I found Heather’s analysis extremely informative and generally persuasive. It is, however, best read in conjunction with Ward-Perkins’ study.
Analysis of the fall of the Western Empire has to pass various tests, starting with fitting the available evidence (which archaeology has filled out to a considerable degree). It also has to be compatible with the continuation of the Eastern Empire. It has to identify what was different about the fatal crisis of the C5th compared to the (survived) crisis of the C3rd. And it also has to explain why the Western Empire succumbed in the C5th when the (much smaller) Roman Republic survived the threat of Hannibal in the C4th BC.
Blaming Christianity a la Gibbon fails the ‘Eastern Empire survived’ test. As does blaming the reforms of Diocletian or of Constantine the Great or resource depletion (whether silver mines or forests). Citing climate change has to explain why the Western Empire suffered worse than its opponents. Making the battle of Adrianople some sort of marker has to explain why the loss of the major field army of the Eastern empire led to the collapse of the Western empire decades later.
Peter Heather teachers at Worcester College, University of Oxford. He has written a very thoughtful and informative account of the fall of Rome, The Fall of the Roman Empire: A New History of Rome and the Barbarians which manages to pass these tests.
Like Ward-Perkins, Heather zeroes in on the superior geographical position of the Eastern Empire to explain its survival. While its Balkan provinces were ravaged by barbarians, notably by both the Gothic incursion of the 370s which led to the disaster at Adrianople and by the Huns in the 420s and 430s, the land walls of Constantinople and the water barrier of the Bosporus meant that most of its tax-producing provinces were not subject to barbarian incursions. This explanation gains added weight when one considers the disaster of the Arab explosion of the C7th from a much less easily defended frontier.
If geography is crucial, then simple internal decay is not. Heather argues strongly and, when conjoined with the archaeological evidence that Ward-Perkins assembles some of which Heather also refers to, convincingly that internal economic decay was not a noted feature of the Roman system. He contrasts the Western Empire with the Carolingian Empire as one that did fall apart for internal reasons.
What is particularly attractive and informative about Heather’s analysis is careful consideration of what was going on outside Rome’s borders. Again, the increase in archaeological evidence provides a much better basis for historical understanding than has existed previously.
The first process was a lessening of Roman economic primacy. Interaction with the Mediterranean economic world seems to have stimulated increased economic activity beyond Rome’s imperial borders. The technological sophistication and population levels of the various barbarian tribes seems to have been definitely increasing, particularly by the C4th.
This process also saw a lessening of Roman technological superiority. Something that made the Hunnic incursions from the 420s onwards such a shock is that the Huns were, from the start, able to take walled cities and forts, something previous barbarian incursions, such as that of the Goths in the 370s, had not been able to do. Barbarians, such as the Vandals under Geiseric, who could take walled cities were much more of a permanent threat than those who could only ravage the countryside.
The Roman Empire was also the Sole Big Target. It was by far the greatest and most extensive concentration of wealth around. This made it a perennial target of the border-raiding that empires always suffer. But if larger groupings could be put together, then the Empire was the desirable happy hunting ground in a much more extensive and permanent way.
There were two features that particularly encouraged such combining beyond the increased chance to get hold of Roman wealth. One was the eruption of the Huns onto the Caucasus steppes and then the Hungarian plain. The other was a reaction to long-standing Roman habits of punitive retaliation and playing groups off against each other.
Heather sees the rise of the Sassanid Empire, replacing the more loose and quiescent Parthian Empire, as crucial to causing crisis of the C3rd. What the crisis showed was that “Romanisation” had created the need for the spoils of patronage across the entire empire if usurpations were to be minimised. This, combined with the slowness of communications and the need to deal with a rival Persian ‘superpower’, is what led to the division of the Empire into Western and Eastern halves. Heather is particularly informative about how the Roman system functioned. How social systems work is obviously of great interest to him, there are some very perspicacious and informative discussion of the problems of control in a pre-industrial agrarian empire, the nature of landowner politics, the nature of nomad pastoralist society and politics.
He sees Diocletian’s massive expansion of the bureaucracy as more a replacement of political forms (primarily local city politics) than some great burden.
The rise of a Hunnic superpower had a range of effects. First, it drove the Goths into their 370s incursion. Valens’ military disaster at Adrianople was, for example, not a significantly greater defeat than the destruction of three legions at the Teutoburger Wald. The more serious effect was that it was one of series of defeats on long-standing Roman territory that led to a far less complete victory than was the norm in that the Goths retained independent existence within the Empire. A sign that the Roman advantage over the barbarians was declining.
The Huns seem to have been similarly responsible for encouraging the various incursions in the first decade of the C5th which led to the establishment of various unassimilated barbarian groups within the Empire.
The various direct Hunnic incursions from the 420s to 440s caused further damage and, of particular importance, frustrated a serious attempt to reconquer North Africa-a key source of revenue for the Western Empire-from the Vandals. The Hunnic Empire also concentrated various barbarian groups close to the Empire and, when the Hunnic Empire collapsed, had created a region where habits of combining together for common military action had been established. Habits that various groups then engaged in to seize the wealth of the Empire.
In the end, it was a case of one damn thing after another. A particularly successful Roman general, a Flavius Constantius or an Aetius, could stabilise things for a while, but there was always something else coming along.
Once a final joint Western-Eastern attempt to retake North Africa had failed in 468, there was no longer the tax base to support a Roman Empire in the West and even the pretence of such a thing came to be seen to serve no purpose.
I have a couple of quibbles. First, although Heather admits that the level of social collapse in Britain was precipitous, he seems a little too sanguine about how much continuity there was in the rest of the Empire, though his point about military prowess replacing literacy as the basis for political advancement is well-taken. The barbarian kingdoms clearly did want the goodies of the more successful society they were occupying. This is a common feature of newcomers drawn to such a society. But retaining said goodies is only possible if the newcomers replicate and reinforce the patterns of behaviour which produce those goodies. If, on the contrary, they continue the patterns of behaviour which produced the less successful societies they are coming from, they will undermine (even destroy) the benefits they are after. Which, despite the initial adoption of Roman government forms, is clearly what happened if less precipitously than occurred in Britain.
Second, Heather’s comment at the end about the barbarians becoming more effective as a result of ‘unbounded Roman aggression’ (imperialism as its own rewards) seems a little too precious. That imperialism may have a tendency to create more effective opponents is a striking idea. But Roman punitive expeditions, no matter how nasty, were rarely, if ever, gratuitous. The consequences of failing to maintain dominance over the barbarians were sufficiently horrific to suggest the Romans may have had a point in such policies.
On the other hand, though he does not consider the comparison directly, there is more than enough in Heather’s analysis to pass the test of why the larger Western Empire was not able to survive its terminal crisis when the, much smaller, Roman Republic was able to survive and, eventually, triumph over Hannibal.
Maintaining Roman rule over Western Europe for 450 years is a hugely impressive achievement. But it is also clear enough that, in a crisis situation, the late Empire had less comparative ability to mobilise resources and contain political rivalry than did the Roman Republic at the time of Hannibal.
One of the damn things that kept happening was murderous political rivalry within the Roman system, up to the point of civil war. Civil war was a periodic feature of the Roman system for centuries, right back to Marius and Sulla. But continuing the habit while repeated barbarian incursions and occupations were occurring was not conducive to survival. The imperial system so concentrated political power and its benefits that there was no forum for managing tensions except fighting it out until some strongman emerged victorious.
The other was the military implications of the specialisation Ward-Perkins draws attention to. Military skills were essentially concentrated among paid professionals. If local field forces were defeated and, over the longer term, the tax base to support the paid professionals was undermined, the Roman citizens were extremely vulnerable. This was quite a different social contract than that which had sustained the Roman Republic. There citizenship and military prowess went hand-in-hand- the “deal” of citizenship was military service in return for legal and political rights. As a result, the Roman Republic could, within its much smaller population, draw upon a proportionately much larger pool of motivated and skilled soldiers. In a prolonged crisis, this-along with the ability to contain and manage political tensions-made the (much smaller) Roman Republic of the C4th BC much tougher than the much larger Western Empire of the C5th. Which is not to deny that Western Empire still showed considerable powers of resilience. Just not enough to deal with one damn thing after another.
I found Heather’s analysis extremely informative and generally persuasive. It is, however, best read in conjunction with Ward-Perkins’ study.