… last week we noted how the collapse of the Roman Empire in the West did not destroy the Roman cultural sphere so much as accelerate its transformation (albeit into a collection of fragmented fusion cultures which were part “Roman” mixed with other things), it did bring an end to the Roman state in the west (but not the east) and an end to Roman governance. But here too, we have to be careful in defining what that governance meant, because the Roman Empire of August, 378 AD was not the Roman Empire of August, 14 AD. This is a point that is going to come up again and again because how one views the decline of the fifth and sixth centuries depends in part on what the benchmark is: are we comparing it to the empire of Hadrian (r. 117-138) or the empire of Valentinian (r. 364-375)? Because most students are generally more familiar with the former (because it tends to be get focused on in teaching), there is a tendency to compare 476 directly with Rome under the Nervan-Antonines (96-192) without taking into account the events of the third and early fourth century.

Roman rule as effectively codified under the first emperor, Augustus (r. 31BC – 14AD) was relatively limited and indirect, not because the Romans believed in something called “limited government” but because the aims of the Roman state were very limited (secure territory, collect taxes) and the administrative apparatus for doing those things was also very limited. The whole of the central Roman bureaucracy in the first century probably consisted of just a few hundred senatorial and equestrian officials (supported, of course, by the army and also several thousand enslaved workers employed either by the state directly or in the households of those officials) – this for an empire of around 50 million people. Instead, day to day affairs in the provinces – public works, the administration of justice, the regulation of local markets, etc. – were handled by local governments, typically centered in cities (we’ll come back to them in a moment). Where there were no cities, the Romans tended to make new ones for this purpose. Roman officials could then interact with the city elites (they preferred oligarchic city governments because they were easier to control) and so avoid having to interact directly with the populace in a more granular way unless there was a crisis.

By contrast, the Roman governance system that emerges during the reigns of Diocletian (r. 284-305) and Constantine (r. 306-337) was centralized and direct. The process of centralizing governance had been going on for some time, really since the beginning of the empire, albeit slowly. The Constitutio Antoniniana (212), which extended Roman citizenship to all free persons in the empire, in turn had the effect of wiping out all of the local law codes and instead extending Roman law to cover everyone and so doubtless accelerated the process.

During the Crisis of the Third Century (235-284) this trend accelerates substantially; the sources for this period are relatively poor, making it hard to see this process clearly. Nevertheless, the chaotic security situation led Roman generals and usurpers to make much greater demands of whatever local communities were in their reach, while at the same time once in power, emperors sought to draw a clearer distinction between their power and that of their subordinates in an effort to “coup proof” their regimes. That new form of Roman rule was both completed and then codified by Diocletian (r. 284-305): the emperor was set visually apart, ruling from palaces in special regalia and wearing crowns, while at the same time the provinces were reorganized into smaller units that could be ruled much more directly.

Diocletian intervened in the daily life of the empire in a way that emperors before largely had not. When his plan to reform the Roman currency failed, sparking hyper-inflation (whoops!), Diocletian responded with his Edict on Maximum Prices, an effort to fix the prices of many goods empire wide. Now previous emperors were not averse to price fixing, mind you, but such efforts had almost always been restricted to staple goods (mostly wheat) in Rome itself or in Italy (typically in response to food shortages). Diocletian attempted to enforce religious unity by persecuting Christians; his successors by the end of the century would be attempting to enforce religious unity by persecuting non-Christians. Whereas before taxes had been assessed on communities, Diocletian planned a tax system based on assessments of individual landholders based on a regular census; when actually performing a regular census proved difficult, Constantine responded by mandating that coloni – the tenant farmers and sharecroppers of the empire – must stay on the land they had been farming so that their landlords would be able to pay the taxes, casually abrogating a traditional freedom of Roman citizens for millions of farmers out of administrative convenience. Of course all of this centralized direction demanded bureaucrats and the bureaucracy during this period swelled to probably around 35,000 officials (compared to the few hundred under Augustus!).

All of this matters here because it is this kind of government – centralized, bureaucratic, religiously framed and interventionist, which the new rulers of the fifth century break-away kingdoms will attempt to emulate. They will mostly fail, leading to a precipitous decline in state capacity. This process worked differently in different areas: in Britain, the Roman government largely withered away from neglect and was effectively gone before the arrival of the Saxons and Angles, a point made quite well by Robin Flemming in the first chapter of Britain after Rome (2010), while in Spain, Gaul, Italy and even to an extent North Africa, the new “barbarian” rulers attempted to maintain Roman systems of rule.

This is thus an odd point where the “change and continuity” and “decline and fall” camps can both be right at the same time. There is continuity here, as new kings mostly established regimes that used the visual language, court procedure and to a degree legal and bureaucratic frameworks of Late Roman imperial rule. On the other hand, those new kingdoms fairly clearly lacked the resources, even with respect to their smaller territories, to engage in the kind of state activity that the Late Roman state had, for instance, towards the end of the fourth century. Instead, central administration largely failed in the West, with the countryside gradually becoming subject to local rural magnates (who might then be attached to the king) rather than civic or central government.

The problem rulers faced was two-fold: first that the Late Roman system, in contrast to its earlier form, demanded a large, literate bureaucracy, but the economic decline of the fifth century (which we’ll get to next time) came with a marked decline in literacy, which in turn meant that the supply of literate elites to staff those positions was itself shrinking (while at the same time secular rulers found themselves competing with the institutional Church for those very same literate elites). Second – and we’ll deal with this in more depth in just a moment – Roman rule had worked through cities, but all over the Roman Empire (but most especially in the West), cities were in decline and the population was both shrinking and ruralizing.

That decline in state capacity is visible in a number of different contexts. Bryan Ward-Perkins (Rome and the End of Civilization (2005), 148ff) notes for instance a sharp decline in the size of Churches, which for Christian rulers (both the post-Constantine emperors and the new “barbarian” kings) were major state building projects meant to display either royal or local noble wealth and power; Church size really only reaches Late Roman equivalent in the west (an important caveat here, to be sure) in the ninth century. In this kind of context it is hard to say that Visigothic or Merovingian rulers are actually just doing a different form of rulership because they’re fairly clearly not – they just don’t have the resources to throw at expensive building projects, even when you adjust for their smaller realms.

Nor is it merely building projects. Under Constantine, the Romans had maintained a professional army of around 400,000 troops. Much of the success of the Roman Empire had been its ability to provide “public peace” within its borders (at least by the relatively low standards of the ancient world). While the third century had seen quite a lot of civil war and the in the fourth century the Roman frontiers were cracking, for much of the empire the legions continued to do their job: war remained something that happened far away. This was a substantial change from the pre-Roman norm where war was a regular occurrence basically everywhere.

The kingdoms that emerged from the collapse of Roman rule proved incapable of either maintaining a meaningful professional army or provisioning much of that public peace (though of course the Roman state in the west had also proved incapable of doing this during the fifth century). Instead those kingdoms increasingly relied on armies led by (frequently mounted) warrior-aristocrats, composed of a general levy of the landholding population. We’ve actually discussed some of the later forms of this system – the Anglo-Saxon fyrd and the Carolingian levy system – already; those systems are useful reference points because they’re quite a bit better attested in our evidence and reflect many of the general principles of how we suppose earlier armies to have been organized.

The shift to a militia army isn’t necessarily a step backwards – the army of the Middle Roman Republic had also been a landholder’s militia – except that in this case it also marked a substantial decrease in scale. Major Merovingian armies – like the one that fought at Tours in 732 – tended to be around 10,000-20,000 men (mostly amateurs), compared to Late Roman field armies frequently around 40,000 professional soldiers or the astounding mobilizations of the Roman Republic (putting around 225,000 – that is not a typo – citizen-soldiers in the field in 214BC, for instance). Compared to the armies of the Hellenistic Period (323-31BC) or the Roman Empire, the ability of the post-Roman kingdoms to mobilize force was surprisingly limited and the armies they fielded also declined noticeably in sophistication, especially when it came to siege warfare (which of course also required highly trained, often literate engineers and experts).

That said, it cannot be argued that the decline of “public peace” had merely begun in the fifth century. One useful barometer of the civilian sense of security is the construction of city walls well within the empire: for the first two centuries, many Roman cities were left unwalled. But fresh wall construction within the Empire in places like Northern Spain or Southern France begins in earnest in the third century (presumably in response to the Crisis) and then intensifies through the fifth century, suggesting that rather than a sudden collapse of security, there had been a steady but significant decline (though again this would thus place the nadir of security somewhere in the early Middle Ages), partially abated in the fourth century but then resumed with a vengeance in the fifth.

Consequently the political story in the West is one of an effort to maintain some of the institutions of Roman governance which largely fails, leading to the progressive fragmentation and localization of power. Precisely because the late Roman system was so top-heavy and centralized, the collapse of central Roman rule mortally wounded it and left the successor states of Rome with much more limited resources and administration to try to achieve their aims.

Bret Devereaux, “Collections: Rome: Decline and Fall? Part II: Institutions”, A Collection of Unmitigated Pedantry, 2022-01-28.