TRAC 1991: When (and What) was the End of Roman Britain, by Kurt Hunter–Mann

Download file in PDF format: TRAC 1991: When (and What) was the End of Roman Britain (pp. 67–78)

TRAC 1991:
By Kurt Hunter-Mann

A failure to distinguish archaeology from historiography has led many archaeologists to become over-reliant on the sub-division of British history into periods such as Roman and Anglo-Saxon. Furthermore, the unquestioning use of such normative racial/cultural terms as ‘Roman’ obscures the view of socio-economic change during the Roman period. This paper discusses the end of the Roman Period as part of a continuum of social, political and economic change that refers back at least to the Iron Age and greatly influenced the Anglo-Saxon period. Marxist theory is used to suggest why British society changed so much during the fifth century. The model employed involves three basic revisions to the Marxist explanation of diversity and change: the preeminence of the societal superstructure; greater interaction within and between social classes; and the potential of any individual to promote change.


The study of Roman Britain is dominated by dates. Historical events provide convenient chronological markers – not just AD 43 and 410, but also 55 BC, AD 61 and 367, for example. Some of the more fiercely debated topics also revolve around dates: just when exactly did Agricola subjugate Scotland, and when was the Antonine Wall manned? I do not wish to suggest that historical events are not important to archaeologists, but neither should they dictate the archaeological agenda. Whereas Roman Britain is seen as ending in 410 (if not 402), studies of the Anglo-Saxon period prefer to use a date of ‘circa 450’ for its commencement. Consequently, relatively little attention has been paid to the interface between the two periods, to the detriment of both. For example, Simon Esmonde Cleary recently suggested that ‘there is a discernible post-Roman but nonSaxon interlude over most of the country of half a century at least’ (1989, xi). Such historicism is self-perpetuating in that it confines evidence of romanitas to the period before circa 400, implying that society after 400 was no longer Roman; this view is normative in the extreme.

An Alternative Perspective
This paper is based on the Marxist assumption that the primary causes of social change are the contradictions between the forces of production and the relations of production; such contradictions result in conflict between those classes controlling production and the classes exploited in this process. Marxist-inspired Western archaeologists have developed a spectrum of views on the relative importance of economic and non-economic phenomena, such as ideology, in dictating the nature of the socio-political superstructure; there is also disagreement among neo-Marxists as to the dominance or reciprocity of the economic base and the socio-political superstructure (Trigger 1989, 340-1). In order to avoid over-emphasis on the economy, I prefer to see social change as the result of contradictions between the forces and relations of society, rather than of production; this means that non-economic factors, such as ideology, can have a direct influence on society.

One problem with Marxist theory is its division of society into classes (traditionally, the exploiters and the exploited), as this tends to reduce the possibilty of conflict within classes or consensus between classes. The potential for reciprocity between the socio-political superstructure and the relations of society has to be considered. This would allow sub-classes to contribute to the relations of society, whether or not they belonged to classes controlling the means of society; such a process, operating across classes rather than upwards between classes, can be regarded as lateral relations of society.

The elaboration of the class system acknowledges the potential of the individual to contribute to the relations of society, something not easily accommodated in traditional Marxist theory. Formal economic models, for instance, include an error term because it is recognised that any individual can make an impact on the group norm (Anderson 1991, 13-14).

Unlike modern capitalist imperialism (conscious, systematic economic exploitation of the periphery by the core), Roman imperialism was largely the product of social struggle within the Roman elite (Millett 1990, 6-8). Such socially motivated action, operating within classes (often for personal prestige), incorporates the three refinements to Marxist theory outlined above.

There was a degree of economic exploitation of the Roman imperial periphery, but in the case of Britain this took the form of an opportunity for the developing provinces (such as those in Gaul) to enter a new market; the process involved was casual, lateral interaction between provinces – not planned, hierarchical exploitation by a centralised imperial economy.

The provincial administrative system suggests a desire on behalf of the imperial authorities to avoid an excessively centralised system of government. If there was an imperial strategy, it was the use of Romanisation to encourage its subjects’ participation in each provincial socio-economic system. Romanisation is generally seen as a gravitational form of cultural assimilation, with the aristocracy being the first native class to adopt aspects of Roman culture in order to be identified with imperial authority, thus maintaining their status within native society (ibid., 68); the lower levels of native society followed suit later for both social and economic reasons.

The question relevant to this paper is how ‘Roman’ the Romano-Britons had become by circa 400. The two main infrastructural elements of Romanisation, the towns and the army, have long been the focus of archaeological investigation in this country, yet they encompass probably less than ten per cent of the population of Roman Britain. The more common agricultural establishments and the small towns, displaying less evidence of romanitas, are severely under-represented in the archaeological record investigated thus far. A consequence of this is an over-estimation of the degree of Romanitas in Romano-British society. The pragmatic nature of Roman imperialism meant that suitable elements of the LPRIA (late pre-Roman Iron Age) socio-economic system were accommodated in that of the Roman province. The maintenance of the native aristocracy within the provincial system probably involved the continuation of the LPRIA socially-embedded economy. As far as the bulk of the land-based native population were concerned, isolated as they were from much that was ‘Roman’ about Romano-British society, there was little inherent need for widespread adoption of Roman cultural attributes.

Towns are considered to have played an important role in the process of Romanisation, acting as foci of the imperial socio-economic system. Yet there is evidence of a decline of the urban functions within towns, beginning as early as the second century (Reece 1980). As the need or desire of the imperial government to maintain towns as infrastructural elements of Romanisation diminished, so the level of the government’s investment in towns decreased. Consequently, the towns had to rely on their functional roles as social, economic and administrative central places within the local socio-economic system. In addition, increased investment in both rural society and economy – which is evident in the increasing numbers of villas and rural temples, and the location of later Roman pottery industries away from the larger towns (Millett 1990, 167-8 and fig. 52) – suggests that towns were possibly centres of trade, but not of industry. The documented troubles of the third century, with inflation, and disruption of trade with Gaul, probably saw a decline of trade via the towns. Although towns retained some social and trading importance, they acted primarily as centres of administration and, where walled, defence.

Although the army may have contributed to the Romanisation of Britain, its role in this process has been exaggerated. The military never totalled more than 50,000, whereas the civilian population numbered several millions, the vast majority of whom had no regularised direct contact with the army. Furthermore, a large proportion of the army consisted of auxiliaries, who would themselves have been Romanised only to a degree at first. As time passed, the army became less a beacon of romanitas and more a military wing of Romano-British society, hardly distinguishable from the civilian population. Consequently, the socio-economic influence of the army is considered to have been limited and largely indirect.

The concept of Romanisation filtering down from the native aristocracy to the rest of the population has merit, but requires qualification. Romanisation was a protracted process which occurred at varying, inconsistent rates in different places; the adoption of the villa house at the expense of the round-house may be a case in point. LPRIA society differed greatly across Britain, and the reaction to Romanisation would have varied accordingly (ibid., 1lff.). Clearly, Romanisation was not a normative process; the rate of assimilation (and the meaning it had to its subjects) would have varied from tribe to tribe, and within these groups between communities, families and even individuals.

In short, there was no classical Roman society in fourth century Britain; indeed, there probably never had been as far as the majority of the population were concerned. The socio-economic system displayed signs of reversion to the socially-embedded, land-based LPRIA system that had been maintained in the Roman period to form the foundation to the ‘Roman’superstructure of Romano-British society. Accordingly, the role and nature of the infrastructural elements (notably the towns) differed greatly from the early Roman period. Finally, it is unwise to envisage a single RomanoBritish society; there was a multitude of perpectives of social identity, held by individuals and groups.

Keeping this view of later Romano-British society within a continuum of socio-economic change, we may now be able to look for evidence of change in the fifth century and try to determine what it represents.

A rapid decline in coin circulation has been cited as an indication of the rapid collapse of the economy circa 400. However, it may be that the presence of Roman coinage in Britain largely represents central government expenditure, at first to pay the army, and later as remuneration for civil work and offices (Reece 1987, 125 and fig. 7.4). Copies of coins were often made during the fourth century, but arguably because the diocesan government had to produce coinage for its own administrative purposes. Copying tended to occur when silvered bronze coins were in circulation, perhaps because it was the intrinsic value of the metal that mattered, rather than the nominal cash value. This suggests that the bronze coins were largely the product of the administration’s use of coins, with site finds representing the discarding of intrinsically relatively worthless, unsilvered coins, rather than loss as a result of circulation within a cash economy (ibid., 43-44). The increasing vigour of the Romano-British pottery industries during the later Roman period suggests that pottery was supplied on a more regional basis. Relatively little fourth-century Romano-British pottery was exported out of the diocese, and the level of pottery imports from the continent was much lower than that of the earlier Roman period (Millett 1990, 161 and fig. 65); a decrease in such long-distance trade probably reduced the need for large-scale cash transactions.

The use of coinage may not have ended so abruptly anyway. Coins issued after 402 do occur, albeit uncommonly, on British sites; moreover, the lack of post – 402 bronze issues from the continent may have been partly the product of an abundance of issues from the preceding period (388-402), a pattern of coin usage evident throughout the fourth century. Assuming a use half-life of twenty years, approximately a quarter of the coins issued during 388-402 could still have been in circulation by circa 430. Such a diminishing coin supply could have facilitated a gradual change to a cashless economy. Turning this argument on its head, there is no reason why the Romano-British could not have minted their own coins for use in a cash economy after 402. The abrupt cessation of the importation of imperial coinage after 402 is more in keeping with the end of imperial administrative and military involvement in Britain circa 410. The evidence of the coinage is therefore inconclusive, and must be considered in its socio-economic context.

Another widely accepted sign of the collapse of the Roman economy circa 400 is the apparent decline of pottery production. Yet the fourth century saw the emergence of manufacturers of both coarse and fine wares, and hand-made wares (ibid., 224 and fig. 95). It is possible that the hand-made wares represent a phase of pottery production operating at a lower technological level after either the demand or the ability to supply wheel-thrown fine wares had ceased. Such regionalised industries, not widely distributed far from the centres of production and so less likely to be influenced by imported forms, may have developed a degree of typological conservatism. It is possible that such pottery production could have continued well into the fifth century, involving forms not readily distinguished from those of the early fifth century.

Whereas the archaeological evidence is inconclusive, there is documentary evidence pointing to a degree of continuity of Romano-British society beyond 400. A council appears to have been governing the diocese in the middle of the fifth century (Gildas, 23.1), and at least one dynasty of the Romano-British aristocracy maintained their political eminence throughout the fifth century and into the sixth in central southern Britain, for example (ibid., 25.3). Documentary sources in fifth and sixth century Britain do not perceive any abrupt end to Roman Britain. They do identify the conflict between British and Germanic elites during the third quarter of the fifth century as a most serious disruption, if not a catastrophe, and it is perhaps at this point that the society and economy was forced to change markedly from that of the fourth century. The use of central places with a primary defensive function, particularly evident in the re-use of Iron Age hillforts (Burrow 1981), indicates a fragmentation of society into chiefdoms, albeit perhaps grouped together into petty kingdoms. Some of these kingdoms in the east were controlled by immigrant elites, but native kingdoms survived in the West for centuries after AD 400.

The Social Fabric of Later Roman Britain
The differences between fourth and sixth century British society and economy do suggest great change in the intervening years. In addition, the ultimate dominance of Germanic authority and language also requires explanation. It may be useful to look at the various social, economic and political factors that influenced the nature of the diocesan society in the early fifth century; it is these factors that dictated the development of a subsistence economy, through an exchange economy, and ultimately (perhaps) to a market economy. Economic stability was directly related to political stability: the Roman imperial system provided that by preventing internal conflict as well as rivalry with neighbouring political groups. The fifth century Romano-British diocesan authorities, on the other hand, had to maintain this control after circa 410.

Adopting the revised Marxist theory propounded earlier, the forces of society were the result of interaction between the means and the organisation of society. The ‘means’ in this case refer not only to subsistence, but also to any factor required in the maintenance of the existing standard of living. In later Roman Britain, technological development evident in the rural economy – such as corn-driers/malting kilns – imply a relatively vigorous land-based economy. The construction of walls around some towns in the later Roman period is thought to be the result of government investment in these administrative nodes.

Of course, such investment assumes the success of subsistence farming, to provide the basic means of life. The demographic stability required to maintain a subsistence economy was subject to the determinist factors of climate and disease. Gildas (22.2) refers to two famines and a plague in the earlier fifth century: ‘a deadly plague … in a short period laid low so many people . . . that the living could not bury all the dead.’ There had been a climatic deterioration in the later Roman period, and regular pandemic diseases appear to have been encouraged by empire-wide trade and travel. It is likely that there was a significant fall in population, which could have been one reason for the settlement of joederati on under-populated land. Consequently, determinist factors may have weakened Romano-British society. On the other hand, such problems were not unique to fifth-<:entury Britain; there had been epidemics and pandemics during the Roman period. Climate and disease should be seen only as indirect causes of political instability. In order to maintain the socio-economic system the workforce had to be organised. One element of this organisation was the military, needed to prevent the loss of resources to neighbouring powers through either raiding or annexation. A sign of increasing regimentation of rural society was the growth of the colonate system, a result of the desire to ensure not only subsistence production, but also the generation of wealth for the landowners. The coloni were tied to estates on social and legal grounds. The government had to pay for such elements of the infrastructure as the army, the bureaucracy and town defences. This was simply obtained from the inhabitants of the empire by either taxation or service. The curial classes, for instance, were expected to support their local municipalities. But while the forces of society were attempting to maintain the status quo, and extract an economic surplus to benefit the exploiting groups, there was an inevitable desire on the part of the exploited to overcome such inequality. We have already seen that there was a tension between the process of Romanisation and the need or desire of the British to accept cultural assimilation. Even in the fourth century, it is not clear to what degree Romano-Britons saw themselves as Roman, rather than British with Roman influences, or even as tribal, with only superficial diocesan and imperial loyalties. Romano-British society, compared to that of the LPRIA, was cosmopolitan. Initially, foreigners dominated the administration and economy. Imperial exploitation of Britain, such as to pay for defence - a large fraction of the economic surplus - was likely to be merely tolerated, and only then when some benefit to the contributors was evident (the Western Empire's military forces were relatively more costly than those of the wealthier Eastern Empire). The failure of the imperial system to both provide services and maintain its political control would have encouraged the manifestation of hitherto suppressed socio-political discordances within the province during the fifth century. Even during the fourth century, it appears that both the coloni and the curial classes were attempting to evade either the provision of services or the payment of taxes, or both; hence the government's attempts to enforce services and taxation by law. Another source of tension was the exploitation of the poor by the wealthy. The land-owners tried to avoid supporting their local towns, investing resources into their rural estates instead. Landholding was increasingly concentrated in fewer and fewer hands. Consequently, the coloni found themselves not only having to shoulder a proportionally greater share of the tax burden, but also working as mere tenants, unable to take their labour elsewhere. The Germanic immigrations, notably during the fifth century, gave scope for further racial conflict. Martin Millett (1990, 219ff.) has suggested that political authority became personal rather than constitutional during the fifth century. Consequently, whereas in Gaul the political framework could accommodate immigrant groups by, for example, the legalised transfer of lands, the personalised politics in Britain encouraged a more personal and inconsistent response to such immigrations into the local socio-economic systems, which were increasingly socially-embedded; in particular, there may have been some reluctance to accommodate the immigrant elites within the political superstructure. This may have resulted in tension developing against such constraints. Religion is another factor that should be considered, perhaps as a lateral tension. The success of the Romanisation of Britain has already been questioned above, and the degree of acceptance of the Roman pantheon by the British is a part of that debate. In view of this, it is likely that the adoption of Christianity as the official imperial religion in the fourth century had little immediate effect, as is indicated by an increase in the number of pagan temples in use (ibid., fig. 83). If Christianity spread as a form of assimilation, it may have been accepted more readily by those officials and aristocracies most dependent on the imperial government, but this means that Christianity had yet to reach the lower orders. Faced with struggles for political control during the fifth century, one wonders how a pagan Briton, perhaps ambivalent to romanitas, would have viewed the relative merits of being subject to a Romanised Christian native, or a pagan German. Germanic people had long resided in Britain, and the settlement of foederati was an established policy. By the middle of the fifth century, there may have been a considerable Germano-British population, probably considerably larger than the number of first and even second generation immigrants, integrated into the lower orders at least. It may be significant that Cerdic, claimed in the Anglo-Saxon Chronicle as the founder of Wessex, is a British name. Overall these social factors indicate a fragile Romano-British society, divided both hierarchically and laterally. This echoes Michael Grant's (1990) view of internal weakness within the Roman Empire as a whole. It is simplistic and normative to regard the upheavals of the later fifth century as a struggle between two distinct groups (Arnold 1984, 122ff.) - (perhaps) ten thousand Angles, Saxons and Jutes, against three million Britons; it was much more complicated than that. Even if society was stable internally, overall political stability was also dependent on external political conditions. Hostile neighbours attempting to increase territories, wealth and influence could raid or annexe weaker rivals. In the case of late Roman Britain, the diocese was subjected to raids rather than annexations, and indeed these had occurred in the earlier Roman period. In the fifth century Germanic settlements in the east of the diocese could be regarded as an external threat, but it has already been suggested that this may have been more of an internal, socio-political problem. However, many parts of the country, particularly the coastal regions, seem to have suffered from an intensification of raiding (and limited invasion). An inability to counter such threats on a diocesan scale would surely have led to a fragmentation of authority. Although the barbarian invasions were an important factor in the fall of the Western Roman Empire (Ferrill 1986, 164), Britain was lost by the empire for political rather than military reasons. POLITICS AND THE ECONOMY, A LONG-TERM VIEW The Roman Empire satisfied the requirement for economic stability by imposing political stability on Britain. The Empire also facilitated the development of long-distance trade links, which would promote an exchange economy. The exchange economy, with close interaction and greater competition throughout the system, may be evident in more rapid technological development and typological innovation. In later Roman Britain technological development occurred within the rural economy, presumably as a result of the increased investment in the countryside at the expense of the towns. However, the decline of urban functions, and typological conservatism evident in the pottery, for example, betrays a decline in aspects of the economy, particularly long-distance trade. It can be argued that there is a correlation between the magnitude of political authority and the scale of the economy in southern Britain during the first millennium AD. The imposition of the Roman imperial system clearly had an impact, but arguably only in terms of (initially) accelerating, and in particular providing a stable political framework for, the development of the LPRIA socio-economic system. The political fragmentation of the later fifth century presumably caused a major economic recession that continued through to the seventh century. However, an exchange economy continued in the Romano-British kingdoms of western Britain, involving some long-distance trade with Gaul and the Mediterranean which included imports of pottery (Thomas 1981), perhaps associated with items such as wine and olive oil (Thomas 1990), and exports of raw materials, including salt (Hurst 1990). Such trade is comparable with that of the LPRIA, and it also has similarities with British long-distance trade links of the fourth and earlier fifth centuries; this may represent the continuation of sociallyembedded long-distance trade from the LPRIA through the Later Roman period. From the seventh century onwards, the increasingly stable sociopolitical situation in eastern England facilitated an expansion of the economy. This coincided with similar developments on the continent, so that the reciprocated trade encouraged an exchange, and finally a market, economy. One wonders if this political and economic consolidation would not have occurred sooner without the imposition of the Roman imperial socio-economic system. CONCLUSION Due to a bias in the settlement hierarchy sample taken thus far, the influence of romanitas has been over-estimated; there is reason to doubt that Romanisation permeated through to all levels of LPRIA society. The Romano-British socio-economic system was essentially the LPRIA system with 'Roman' embellishments. Once the initial phase of imperial government investment had ended, the influence of the embellishments (such as towns) declined in favour of a more socially-embedded, rural economy that is best explained as a development of the LPRIA society and economy. Consequently, the vision of a fully-Romanised, town-based socio-economic system suddenly collapsing at the end of the fourth century is illusory. Instead, with a society in flux, and a steadily changing economy, it is very difficult to identify from the archaeological record any major changes before the later fifth century. It is then that diocese-wide authority ends, and the Romano-British economy ends with it. Yet even then, there was no clean break with Roman Britain. Within the fragmented society and economy, the forces and relations of society would have changed, but persisted; and they would have had a great influence on the nature of the succeeding socio-economic system. Romanists have tended to examine Romano-British society in isolation, only paying lip-service to the need to see it as part of a continuum of social and economic change (Anderson 1991). Anglo-Saxon scholars are just as guilty of this tunnel vision, preferring to concentrate on the origins of the Germanic immigrants rather than their interaction with the native population (Arnold 1984, 5). Archaeologists should be as concerned with material culture as with historiography; the compartmentalisation of time into historical periods should be seen as a convenience, intended to facilitate the archaeological debate, not to dictate it. Bibliography Anderson, J. L. 1991 Explaining Long-Term Economic Change. Basingstoke; Macmillan. Arnold, C. 1984 Roman Britain to Saxon England. Beckenham; Croom Helm. Burrow, I. 1981 'Hill-forts after the Iron Age: the relevance of surface fieldwork', in Guilbert (ed.), 122-49. Esmonde Cleary, A. S. 1989 The Ending of Roman Britain. London; Batsford. Ferrill, A. 1986 The Fall of the Roman Empire: the military explanation. London; Thames and Hudson. Gildas see Winterbottom 1978. Grant, M. 1990 The Fall of the Roman Empire 2nd edition. London; Weidenfeld and Nicolson. Guilbert, G. (ed.) 1981 Hill-Fort Studies. Leicester; Leicester University Press. Hurst, D. 1991 'Major Saxon discoveries at Droitwich - excavations at the Upwich brine pit' Current Archaeology 126, 252-55. Millett, M. 1990 The Romanisation of Britain. An essay in archaeological interpretation. Cambridge; Cambridge University Press. Reece, R. 1980 'Town and country: the end of Roman Britain' World Archaeology 12,77-92. Reece, R. 1987 Coinage in Roman Britain. London; Seaby. Thomas, C. 1981 A Provisional List of Imported Pottery in Post-Roman Western Britain and Ireland. Redruth; Institute of Cornish Studies. Thomas, C. 1990 '"Gallici nautae de Galliarum provinciis" - a sixth/ seventh century trade with Gaul, reconsidered' Medieval Archaeology 34, 1-26. Trigger, B. G. 1989 A History of Archaeological Thought . Cambrdige; Cambridge University Press. Winterbottom, M. (ed. and trans.) 1978 Cildas' The Ruin of Britain and Other Works. Chichester; Phillimore. [/toggle]

Add a comment

TRAC is the organisation behind the annual Theoretical Roman Archaeology Conference, TRAC Workshops, TRAC Proceedings (1991-2016), the TRAC Themes in Roman Archaeology book series, and the Theoretical Roman Archaeology Journal (TRAJ). This site is maintained by the TRAC Standing Committee. Contact us for further details or inquiries.

Follow us: Facebook | Google+ | Twitter

Copyright © 1991–2019 TRAC and individual authors. All rights reserved.