The Fall of Roman Britain and the myth of the Anglo-Saxon Conquest

I’ve been thinking recently about post-Roman Britain and the accepted version of how the Angles, Saxons and Jutes, led by warrior leaders, displaced the native British and created the Anglo-Saxon kingdoms that came to make modern England.

I think that there is a real problem with this narrative, and that problem is really rooted in the decline in urbanism and end of a money-based economy in the decades after the year 400.

One of the problems is a shortage of contemporary written sources. Most of the written sources come from the 6th century and later, such as Gildas’ De Excidio et Conquestu Britanniae (On the Ruin and Conquest of Britain), a text that is not a history but is more of a polemic against the British rulers and clergy of his own age.

What we do know about the beginning of the 5th century is that Britain effective ceased to be a part of the Western Empire of Honorius and gradually separated into distinct “kingdoms”. The timeline for this begins at the start of the century when Stilicho, the Magister Militum of the West, appears to have withdrawn troops from Britain to support his campaigns against Alaric.

Within a few years, Roman forces that remained in Britain revolted and raised a number of pretenders to the purple. The last of these was Flavius Claudius Constantinus, who, in 407 declared himself to be Western Emperor and is known in British history as Constantine III.

It is widely assumed that he took the bulk of the Roman forces remaining in Britain across the Channel and his generals fought a number of engagements in Gaul and along the Rhine against both barbarian opponents and agaist Stilicho’s Roman forces.

His armies continued to campaign in Gaul, Hispania and in Italy, with Constantine setting up a capital at Arles. Eventually he was recognised as co-emperor by Honorius in 409 and neither he nor his army ever returned to Britain. He was finally defeated by the future emperor Constantius III in 411 and beheaded.

Britain was therefore left denuded of front-line troops (comitatenses), although it seems possible that various units of limitanei (border guard units) and Germanic foederati remained in the provinces. Certainly there seems to have been forces available to repel barbarian incursions around 408-9.

One of the main arguments in favour of a break with Rome is the end of a monetary economy in the first two decades of the 5th century. However, this may be misleading. The empire was never a fully-monetarised economy and coinage was mainly used to pay the troops, although coins were in circulation to some extent.

Therefore, it seems to me that once there were no regular comitatenses-type troops to pay, the need for coinage declined. Border militias and foederati could have been supported by land grants and by payment in kind, rather than with cash.

In this period, it also seems that the local Romano-British magnates expelled the imperial administrators. It is possible that these administrators were most closely identified with tax collection and with the military administration. With no troops to pay, it is entirely possible that the local landowners saw no need for any tax collectors. In the Dominate period, tax collection was mainly for the purpose of paying for the armies.

Of course, the administration may also have seemed like a needless or corrupt luxury and one that served no real purpose in Britain once the military was reduced to a shadow of what it had been previously.

We also see a decline in urbanism in this period and, in the traditional narratives, this has been used as a way of showing the decline of Roman life in the face of barbarian invasion and conquest. However, there is another possible explanation. Britain was a society where the local Roman “aristocracy” was based around a sophisticated villa society. Villas were not just country houses, they were powerful economic hubs, the centres of networks of agricultural and manufacturing activities.

Towns, on the other hand, were not really essential to the economy and, in the Dominate, were part of the imperial structure and administration. One needs to remember that even at the peak of the Roman empire, probably no more than 20% of the population lived in urban centres, a figure that may have been as low as 10% in the West.

In Britain we see a change of use of many urban public buildings in the 5th century, a decline in urban populations and rise in a localised economy, with new unofficial populated centres developing near to villas and at important centres of local manufacture rather than in the older public towns. This may reflect a change towards internal trade rather than international, as well as population shifts away from declining towns.

We also see some urban centres becoming used for small-scale manufacturing and agriculture, before gradually being abandoned.

Therefore, a decline in urbanism may represent a changed emphasis inside British life rather than a decline brought on by external pressures. It would certainly seem that many urban centres were abandoned before there was any significant influx of Germanic peoples across the North Sea.

However, we do see a genuine decline in the traditional fabric of Roman life as the 5th century progresses, including a decline in the villa system. What are we seeing here?

It seems to me that we see a number of factors at play.

Firstly, we do see a change in the structure of society and a gradual fragmentation, with local magnates becoming powerful leaders in their own right, possibly as “protectors” of the peasantry, with their own militias acting as a defence against other magnates’ territorial ambitions and as a means of extending their own area of control.

Secondly, although there was still a sense of the people of Britain somehow still being “Roman”, there was also a gradual change in how people saw themselves. Throughout the 5th century old, pre-Roman sites seem to have been re-populated, hill-forts dating back to the Iron Age become centres of local power and we see Irish and Germanic people begin to arrive in Britain.

Now, we need to look at these Germanic arrivals. The striking thing about this is that in the early period we do not see, from archaeology, a militarised Germanic society with much in the way of high status sites, as one might expect from the standard narrative of conquest by war leaders and their warrior elites.

What we do see is a lot of small settlements and subsistence farming. This suggests to me that the traditional narrative is flawed. If there is an absence of archaeological evidence which shows an absence of high status communities in the 5th and early 6th centuries, sites that might indicate the presence of “warrior” elites, that suggests to me that there might have been an absence of warrior elites in that period too. It is not until the later 6th and 7th centuries that high status elites emerge in the burgeoning “English” areas of Britain

My assessment here is that the commonly accepted narrative of warbands Angles, Saxons and Jutes arriving under elite leaders, bent on conquest, is wrong and that these elites only emerged from agrarian settlements once immigration had established itself. Furthermore, I would suggest that the actual identities of the “Angles, Saxons and Jutes” are a process of deliberate creation by newly emergent elites during the 7th century, and that these ethnic identities were manufactured in along with the claimed ancestries that the various new Anglo-Saxon elites constructed to claim traditional rights of leadership.

There is a lot more evidence for post-Roman British elite centres in the 6th century, in the western part of the country than “English” ones and it seems possible that the emergent “English” elites were consciously influenced by these post-Roman British elites, together with continental ones like the Franks, when developing their own ruling elite families.

There is also a point to consider about the Christianisation of the new English elites. It may well be that they adopted Christianity, not from faith, but because it validated their rule in line with what was happening elsewhere in post-Roman Western Europe. In other words, to be a king in the late-7th and 8th centuries, one had to be a Christian.

However, we are getting ahead of ourselves.

To address the question of why these Germanic peoples made the journey across the North Sea and Channel, we need to look at conditions elsewhere.

5th and 6th century western Europe was a region of massive cultural and political change. There were large-scale movements of peoples and the replacement of Roman provinces with Romano-Germanic successor kingdoms, a process that would continue into the 7th century. In addition there was regular warfare, both between different Roman factions and also between Romans and barbarians.

It is entirely possible that people from the northern fringes of Gaul, Germania and what is now Denmark were actually fleeing from the upheavals, uncertainty and warfare and were not conquerors. Their reasons for coming to Britain were more akin to those of modern economic migrants and refugees rather than anything else.

There would already have been trade connections and other contacts across the sea, as well as numbers of Germanic-speaking people in Britain, serving as Roman auxiliary troops and foederati. It is possible that other groups were encouraged to move to Britain by local magnates to replace the military forces withdrawn at the beginning of the 5th century. The legend of Vortigern and Hengist and Horsa may well be an echo of such practices.

So, to summarise;

Specifically in the case of southern and eastern post-Roman Britain, we should consider factors such as a collapse of central authority in the early part of the 5th century, the lack of any kind of military capability that could limit immigration, the fragmentation of the provinces following the decline of centres of power and the subsequent reversion to a subsistence level of culture.

The traditional model blames these things on incoming barbarian warrior bands, but the abandonment of urban centres, the end of organised large-scale manufacturing and agriculture and the cessation of a money economy predates any such putative invasion.

The triggers for the collapse of Roman Britain were the withdrawal of the military and the expulsion of Roman administrators during the time of Constantine III and his abortive attempt at gaining the western imperium.

We should remember that the model of urbanism in Late Roman Europe, especially in the West, was a changed thing from the urban culture of the earlier Principate. The independence of cities, with local senatorial elites declined and cities became administrative and military centres for the more pervasive Dominate administration, with the old senatorial elites retiring to their country estates.

As Roman central authority declined, cities and public towns became much more associated with the maintenance and support of luxury villa complexes, rather than being population and administrative centres in their own right. In many towns, public spaces declined, to be replaced with urban farms, markets and manufactories of various kinds.

As the villa system collapsed, the public towns themselves served little or no purpose.

At the same time the public centres changed their roles, other settlements, towns and villages sprang up in various places, but without the formal public spaces, temples, basilicas, forums etc of the public towns.

These relatively egalitarian new establishments seem to have developed to act as a focus for local trade and manufacture, rather than as any kind of centre of government.

As the role of old urban centres of the Roman provinces declined, people slowly abandoned them.

4 thoughts on “The Fall of Roman Britain and the myth of the Anglo-Saxon Conquest

  1. Brilliant, I have been thinking along similar lines for quite a while – following writers such as Nick Higham et al in Britons in Anglo-Saxon England, Francis Pryor (Britain A.D.) and Michael Jones (The End of Roman Britain). I am currently writing a (fictional) history of Carausius, but may follow this with a novel from the 6th century, based on some of these ideas. If interested, contact my by email ( or on Facebook.

  2. The Dna doesn’t support an Anglo-Saxon invasion! I even have my doubts about the English language, burgh is a word that appears to be “Celtic” in origin not Teutonic .

    • Well, there isn’t a gene for “invader” so it isn’t appropriate to cite DNA evidence for an invasion or for an absence of invasion. What is interesting is that the recent DNA mapping of the UK showed up that around 20-30% of the DNA of modern British people living it what were traditionally seen as areas of AS settlement is Germanic in origin. That suggests that the Germanic incomers did not wipe out or ethnically cleanse the local populations but mixed in with them. Similarly, in areas where the Danes settled, we see about 10-11% of Danish origin genes. Again, a sign of mixing, not replacement. The real problem for the invasion model is that it is implausible.

      The points that really stick out for me are how the DNA survey illuminates a few things that have long been massive points of argument. Firstly, and this is something that interests me hugely, the evidence demonstrates that the Romano-British inhabitants of what came to be England were not wiped out or ethnically-cleansed by the incoming Germanic speakers we call the Anglo-Saxons. It shows that what happened was a mixing of populations and the absorption of the incomers into existing populations. It also demonstrates that the culture of the incomers, together with their language became dominant, with the Romano-British in those areas where Germanic settlement was strongest adapting their language, dress and other things. This suggests to me several things; that the incomers effected a larger element of societal change than just elite transfer, that after the end of Roman rule, the locals may well have improvised a new way of living that they wore lightly and which never became ingrained and which they gave up when something more robust came along and also that what emerged from the 5th and 6th centuries was basically a synthetic new culture, which eventually became English. This is far from the traditional narrative of the emergence of an “Anglo-Saxon” English race, so beloved of 19th century English historians.

      Secondly, it is also interesting that there seems to be a problem with the traditional narrative of a “Celtic” people on the fringes of the new English world. This again doesn’t really surprise me. The idea of a “Celtic nation” is very much an invention of the late 18th and 19th centuries. There is undoubtedly a linguistic and cultural Celtic identity, it would be nonsensical to deny this, but the evidence suggests that Celtic language, technology and artefacts arrived here via diffusion rather than in waves of migration and ethnic replacement. This is something I have long considered to be the most likely explanation anyway.

      The third thing that really struck me was the close similarity between the southern population of Scotland and northern English. Surely this reflects the spread of the Northumbrian kingdom, itself a merger of the earlier Bernicia and Deira? Bernicia itself is interesting because the name is Cymric not Germanic, although the kingdom was a Saxon one (which covered the modern North-east of England and South-eastern Scotland). Again, I expect that we are seeing a mixed Romano-British and Germanic population in this region.

      Anyway, I find this all fascinating, because it shows how national mythologies arise from origins which are far more complex than the narratives that they spawn and which often bear little resemblance to any objective reality.

      Finally, it is interesting that we see little evidence for the Normans in the genetic survey. I suspect that this is because the survey was aimed at mapping the general population, rather than the aristocracy. The actual number of Normans who settled in England, Ireland and Scotland was tiny, it was a real example of elite transfer and I would say that this is the reason why we see little DNA overlap with the general population of Normandy.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s