Romans, Barbarians and Provincials: Social Boundaries and Class Conflict in Late Roman Gaul

Leslie Dodd (Classics: University of Glasgow)

This paper is dedicated to Nathan Maxwell Harlan (b. 8.5.04) and Charlotte Chance Gemmill (b. 25.7.04)

1	The Romans traditionally characterised their identity in very simple, very stark terms. Romans were defined by their romanitas (Roman culture) which included the use of Latin, regard for classical Latin literature, adherence to Roman law and ancestral mores and even the custom of having three names. Everyone else - everyone who was not a Roman and did not share in this culture - was a barbarian (a word which could, but need not always, be pejorative).[1] All the disparate peoples living beyond Rome's frontiers were conceptualised by Romans in terms of their foreignness and their cultural distance from the civilised ideal of romanitas.[2] By the same measure, all those who lived within the frontiers of Rome's empire were, theoretically, united by their common participation in Roman civilisation and culture.
2	In this paper, I will argue that this conception of Roman identity can be refined. The Gallo-Romans, far from being a homogenous group, can be shown to have been divided by boundaries of class, wealth and social status. Their relations with barbarian incomers were often shaped by internal stresses and conflicts and by the need of an increasingly insecure Roman elite to control and dominate a restive peasantry.
3	In 418 or 419, the Visigoths were recalled from Spain, where they had been campaigning on Rome's behalf against the marauding Vandals, Alans and Sueves, and settled in the Roman province of Aquitanica Secunda (a region on the Atlantic coast of Gaul, bounded roughly by the valleys of the Loire to the north and of the Garonne to the south). The plantation of barbarians in Roman provinces was itself nothing new. It had many precedents and has been described as "a very ancient feature of imperial policy".[3] MacMullen and de Ste. Croix have each identified a number of barbarian groups and tribes residing, at various times, within the Roman Empire in northern Gaul, reinforcing the idea that there was nothing innately exceptional in allowing barbarians to settle inside the empire.[4] For the Roman state, these barbarian settlements offered a new source of manpower and a manageable system for the entry of Germanic barbarians into the empire.[5] The fundamental difference between earlier settlements and that of the Goths (and, later, the Burgundians) in the fifth century is in scale. Never before had so many barbarians been settled together in a Roman province, with the approval of the Roman government, effectively making the region into a Romano-Gothic condominium. Yet, the Roman elite had their own reasons for seeking to involve barbarians in provincial life.
4	Society in the later Roman Empire had become polarised. The empire's ruling class was a senatorial aristocracy, enlarged by the absorption, during the third and fourth centuries, of the equestrian order. Many of these aristocrats were the descendants of Romanised tribal chieftains and dynasts who had adopted the culture of their conqueror and had consequently been accepted into the Roman ruling class; a shared culture and mindset, rather than ethnicity, were the defining characteristics of the Roman senatorial nobility. Within this class, internal gradations distinguished those whose ancestry was noble from those whose title and rank were a consequence of imperial favour or of having held office in the civil service. A range of titles were held from the clarissimus ("most distinguished"), the basic title of aristocracy (and one held hereditarily by the sons of all senators), to the spectabilis ("worthy") and, highest of all, the illustris.[6] ("illustrious"). Land ownership, which, in an agrarian society like that of the empire, formed the basis of all economic life, was concentrated in the hands of these aristocrats as, needless to say, was political power and influence.[7] Gallic aristocrats had traditionally shunned participation in wider imperial politics in favour of localising their political power in their own region. It is hardly surprising, then, that their efforts to control and lead are manifested entirely at the provincial or regional level.[8]
5	By contrast, the situation for the non-aristocratic population of rural Gaul appears to have been particularly hard. Ever greater burdens were placed upon the lower classes both by the state and by landowners.[9] The ancient boundary between the slave and the free man was eroded by the institution of the colonate which tied a tenant, and his descendants, to the land he farmed and allowed him (or, more correctly, his labour) to be sold along with the land. Indeed, the Codex Justinianus could say of coloni "let them be considered slaves of the very soil to which they are born and they shall not have the right of going where they please or of changing their place, but let their landlord exercise his right with the attention of a patron and the power of a master".[10] Like slaves, coloni were disbarred from suing their landlords and were forbidden from alienating their property without the landlord's permission. This condition was not slavery, in the strictest sense of that word, but it was still far from the ancient Roman ideal of the free peasant farmer. Moreover, the Codex Theodosianus allowed for coloni to be punished like slaves: "It is appropriate that those coloni who consider flight be bound by chains in the manner of a slave, so that they are compelled to perform duties which are appropriate to a free man while being punished like a slave".[11] It had been imperial policy since Diocletian's time to tie people to their land, restricting their freedom of movement (both geographically and socially) in order to render tenants more pliable to their landlords and ensure a favourable labour supply while, simultaneously, allowing landlords to extract taxes from the poor on behalf of the central government.[12] The colonate was merely one facet of that wider policy.
6	Not all coloni were necessarily full-time farmers. People with a trade or a craft to practise might have signed over their labour for a specified part of the year.[13] Nevertheless, they and their descendants would be bound to provide labour on a specified piece of land in perpetuity and, while they would retain the right to undertake their craft, they would continue to be tied to the area and unable to leave.
7	In a similar vein, not all full-time farmers were necessarily coloni. Small freeholders continued to farm their own land but, in the uncertain world of late antique Gaul, beset by bandits, barbarian raiders and rapacious landowners, many smallholders would find themselves compelled to enter the colonate, handing their lands and labour over to a magnate, in return for protection.[14]
8	Even for those who avoided the worst and retained their personal freedom and property, the pressures and demands of the state could still bring grave difficulties. Salvian, in his De gubernatione Dei ("On the governance of God"), tells us of smallholders who fled their farms to escape the tax-collector.[15] Others simply signed themselves over to the care of powerful lords, voluntarily entering the colonate.[16] Their situation is compared to that of Odysseus' sailors because, just as Circe turned men into pigs, so the rich now turn free men into slaves.[17]
9	Such troubles as these were exacerbated by the parlous economic conditions of early fifth century Gaul. The anonymous Carmen de providentia Dei ("Poem on the providence of God") is virtually a catalogue of Gaul's economic distresses while many of our other extant sources paint an equally bleak picture of economic deterioration, decline and, in places, devastation.[18]
10	Inevitably, the rural poor reacted to their condition. Some Romans defected to the barbarians rather than continue to live under the oppressive imperial government. As early as 362, Mamertinus noted that some people found life under the barbarians preferable - "now the barbarians are longed for, so that the lot of captives is preferred by the wretched".[19] Likewise, Faustus of Riez writes of captives defecting to their captors.[20] Salvian tells us that poor Romans acted as allies of the barbarians.[21] Orosius describes Romans seeking liberty amongst the barbarians and, in a similar remark, Salvian sees Romans seeking "humanity" amongst the barbarians because they cannot abide the barbarous inhumanity of the Romans.[22] In a reversal of traditional ideology, some Roman citizens find imperial rule so harsh that their only solution is to flee beyond the reach of Roman law and, in effect, to stop being Roman.[23] Others, though, chose to rise in rebellion against the state and the landowning elite.
11	Early in the fifth century, rebellious armed bands - known collectively as Bagaudae (from the Celtic baga, meaning "war", and the suffix -aud) - emerged in Gaul, forming one of the most characteristic features of the region and period.[24] These groups were of varied size and composition; some doubtless consisted of local bandits or poachers while others were made up of disaffected farm workers, slaves and peasants; in isolated highland areas, some clans or tribes may have augmented their income by banding together and extorting money from travellers. For example, in 408, one such group of highland Bagaudae waylaid the Visigothic freebooter Sarus as he crossed the Alps and compelled him to surrender all the loot his band had taken in Italy.[25] The motives of the Bagaudae no doubt varied from group to group with some seeking nothing more than plunder while others seem to have aimed at something like revolution. In all cases, however, we can say that the mere existence of armed bands was an implicit challenge to the authority of the state and the security of the landed classes, as well as being a comment on the absence of order in early fifth century Gaul.
12	Amongst modern historians, the most dogmatic view of the Bagaudae comes from the late E. A. Thompson who saw them in the narrowest of terms as servile revolutionaries and, conceivably, agents of a pan-Occidental movement aiming at the liberation of the slaves and overthrow of the ruling classes.[26] Raymond Van Dam presents an opposing, but equally narrow, interpretation in which Bagaudic rebellion is taken as evidence that peasants rallied to local patrons and leaders in order to protest the state's inability to protect them.[27] Van Dam's Bagaudae rise not to challenge the established order but to reinforce it, "to remain Roman" and because they wanted their leaders "whether landlords or emperors... to function better".[28] Neither view is wholly correct nor wholly supported by the sources. Thompson's argument fails because of its reactionary dogmatism but, to my mind, similar problems are found with Van Dam. The extant sources give us perfectly clear accounts of a number of rebellions and uprisings which were revolutionary in character and sought to overthrow the established social order. In the first few decades of the fifth century, the most dramatic of these seems to have been in Armorica (modern Brittany).
13	In 406, a large force of Alans, Sueves and Vandals had crossed the frozen Rhine and entered Gaul. In the same year, a new usurper - Constantine - was proclaimed emperor by the British legions. He soon crossed over to Gaul and established his headquarters at Arles, in Provence. Italy, at this time, was facing troubles of her own. Radagaesus had invaded the peninsula from the north in 405 while Alaric and his Visigoths were just about to enter Italy from Illyricum. Amidst such chaos, the Roman state was no longer in a position to protect the citizens of the western empire. Provincials were left to defend themselves as best they could and, in Armorica, local people rose in arms and liberated their towns from the barbarians. The other Gallic provinces did likewise.[29] Thus far, Van Dam's thesis of peasants rallying to their overlords is clearly defensible (although, I think, the idea that they strove to defend their romanitas is much less convincing). Zosimus, however, also tells us that after expelling the barbarians, the Armorican provincials "drove out the Roman magistrates and set up their own government".[30]
14	It is reasonable to suppose that the general mobilisation against the invaders morphed into a servile rebellion, possibly because the Armoricans realised that the Roman state was no longer able to maintain the dominance of the local potentes - landowners and aristocrats - over the rural non-elite population. Certainly, we can unmistakably see evidence of revolution in the words of Rutilius Namatianus, in his poem De reditu suo ("On his return"), when he describes the subjugation of Armorica by the Roman leader Exsuperantius in 417: "Father Exsuperantius now teaches the shores of Armorica to love the return of peace; he restores the laws and brings back liberty and does not allow men to be the slaves of their own servants".[31] Van Dam dismisses the sources on grounds of "cultural prejudice" that leads them to view any rising - even one undertaken on behalf of the Roman state and the local elite - as necessarily being an attack on the state.[32] I think, though, that we can safely take the sources at face value and assume that they are accurately and sincerely describing actual events (viz., a servile rebellion against the ruling class).
15	Indeed, the Armorican revolt was merely one of a number of Bagaudic risings which, whether we characterise them as "servile" or not, were aimed at disrupting or overthrowing the established social order - the Roman elite and the governmental apparatus which maintained it. At Bazas, south of Bordeaux, a slave revolt broke out that was aided by a number of free men.[33] The Chronica Gallica ad CCCCLII recounts a rebellion in Gallia Ulterior, in 435, led by Tibatto and tells how "almost all the slaves of Gaul conspired in a Bagaudic movement".[34] Elsewhere in the same chronicle, we find a (Greek?) doctor, Eudoxius, who is not obviously a slave but who had, nevertheless, Bagaudic sympathies and defected to the Huns.[35]
16	While one will certainly wish to interpret the available evidence more critically than Thompson, for whom every appearance of the word Bagaudae is evidence of a slave revolt, one cannot avoid the conclusion that many of the armed insurrections in fifth century Gaul are, in fact, attempts at revolution. There was not necessarily a "class war", in the classical Marxist sense, but plainly quite a number of Gallo-Roman provincials, not all of whom were slaves or, even, peasants, felt so alienated by governmental and aristocratic oppression that armed rebellion became an attractive option; with the barbarian invasions and attempted usurpations by the legions, the Roman state was so overwhelmed that this attractive option became fully viable.
17	By 419, however, imperial political authority was being re-established over Gaul. The Gallic Council of the Seven Provinces was reconstituted at Arles in 418, the campaign against the Armorican rebels was successfully concluded and the Visigoths were settled in Aquitanica Secunda. On this last point, we can see that the military power of the Roman state, having dealt with and recovered from its earlier distractions, was now once more ready to be turned in defence of aristocratic power and against servile dissent. Wood rightly interprets these developments as evidence of a general restoration of order in Gaul and, like Thompson, considers the Visigothic settlement to be closely connected to the Bagaudic troubles in Armorica.[36]
18	The precise details and mechanics of the Visigothic settlement remain controversial and I will not attempt, in this paper, to offer any comment on the subject. It will be enough to say that the traditional view (viz., that Visigothic settlers received parcels of land from the Roman aristocracy) was challenged by Walter Goffart who argued that no land changed hands but, instead, barbarian settlers received grants of tax revenue.[37] To my mind, Goffart's argument has been rendered obsolete by the work of Hagith Sivan and Samuel Barnish who have each argued convincingly in favour of the traditional view.[38] For the purposes of this argument, I will assume that the traditional view is correct, that a land division took place and that Visigothic settlers in Aquitaine received shares of the estates of Roman aristocrats. The land divided in this way would then become the possession of the Visigoths for the duration of their foedus (treaty of alliance) with Rome.
19	The arrival of barbarian federate troops could conceivably have provided enormous security for the elite of the region, provided their reliability could be guaranteed. The Goths would have defended the region south of the Loire in the event of another Armorican uprising (and it is telling that they are settled in this region only a year or so after Exsuperantius' reconquest of Armorica). Even more importantly, the presence of a Visigothic garrison in Aquitanica Secunda would have discouraged the local peasants from attempting a rebellion of their own.
20	The choice of the Visigoths for this rôle is itself interesting as they had, in the quite recent past, made common cause with Roman slaves and others who were hostile to the established Roman order. In 408, they had been joined by many slaves who had fled Roman captivity.[39] In all likelihood, some of the "Gothic" warriors who were settled in Aquitaine were Roman deserters or defectors of the kind we find mentioned in Salvian and Orosius. It would be foolhardy, then, to expect great fidelity to Rome from these people unless their material interests could be wedded to those of the Roman elite. By sharing their estates, the elites could achieve just such a bonding of mutual self-interest. The Goths were turned into "gentlemen-farmers" whose material interests, like those of the Roman elite, were threatened by peasant uprisings and who benefited from the preservation of the status quo.
21	The economic interests of the Visigoths and those of the potentially rebellious Gallo-Roman provincials were set on divergent courses. After the division of land, the Visigoths can profit only from the defence and preservation of the existing social order which, in its turn, is dependent on elite ownership of most of the land and the strict control of the non-elite rural population and its labour. At a time when the Roman aristocrats are feeling increasingly isolated from the state and when the state is itself dependent on barbarian mercenaries and allies to do its fighting, the magnates of Aquitaine were able to utilise the Visigoths as agents to enforce aristocratic authority over the bulk of the Gallo-Roman population.[40]
22	In an analogous situation, Amory has argued - correctly, in my opinion - that the later Burgundian settlement of Sapaudia saw the Burgundian warriors occupying the positions of nobiles and mediocres.[41] That is to say, the highest Burgundian chieftain would have been equivalent in social position to the senatorial aristocracy; the lowest foot soldier, while not ennobled, would still have outranked the Gallo-Roman peasants. We can anticipate that the situation in Visigothic Aquitaine would have been very similar or even identical. In order to control Gallo-Romans of low status, the Gallo-Roman elite is willing, for all practical purposes, to promote barbarian incomers to the highest aristocratic and social ranks and to share its wealth and privilege with them.
23	The great boundary in fifth century Gaul is not that between Roman and barbarian but between aristocrat and provincial, between elite and non-elite. The perceived threat in Gaul was not so much from the invading barbarian horde as from the exploited rural masses. For both the elites who wish to perpetuate a system of control and exploitation and for the provincials who wish to escape that system, the barbarians present an opportunity. Wretched peasants flee to the barbarians to escape Roman injustice while Roman nobles see, in the strength of barbarian arms, a means of underpinning their hold.
24	Ultimately, though, the position of the Roman aristocracy was dependent on the health of the Roman state. Despite the utilisation of barbarian arms, the erosion of imperial authority in the western empire continued apace throughout the fifth century until, eventually, a point would be reached where Romans could be sidelined entirely. The Visigothic king Euric who could ostentatiously show that political supremacy in Aquitaine lay squarely with the Visigoths.[42] The Germanic settlers would come to displace the Roman aristocracy and, instead of acting as agents of Roman power, would come to wield that power for themselves.

--------------------------------------------------------------------------------

[1] P. J. Geary, 'Barbarians and ethnicity', in Interpreting Late Antiquity, ed. by G. W. Bowersock, P. Brown and O. Grabar (Cambridge, Mass.: Harvard University Press, 2001), pp. 107-29 (pp. 107-109).

[2] P. J. Geary, Before France and Germany: The Creation and Transformation of the Merovingian World (Oxford: Oxford University Press, 1988), pp. 39-42.

[3] S. J. B. Barnish, 'Taxation, Land and Barbarian Settlement in the Western Empire', PBSR, 54 (1986), pp. 170-95 (p.174).

[4] R. MacMullen, 'Barbarian enclaves in the northern Roman empire', L'Antiquité Classique, 32 (1963), pp. 552-61; G. E. M. de Ste. Croix, The Class Struggle in the Ancient Greek World (London: Duckworth, 1981), pp. 509ff.

[5] G. Wirth, 'Rome and its Germanic Partners in the Fourth Century', in Kingdoms of the Empire: The Integration of Barbarians in Late Antiquity, ed. by W. Pohl (Leiden: Brill, 1997), pp. 13-55 (p. 31).

[6] A. Marcone, 'Late Roman Social Relations', in The Cambridge Ancient History XIII: The Late Empire, A.D. 337-425, ed. by Av. Cameron and P. Garnsey (Cambridge: Cambridge University Press, 1998), pp. 338-70 (p. 339); A. H. M. Jones, The Later Roman Empire, 2 vols. (Baltimore: Johns Hopkins University Press, 1964 repr. 1986), pp. 527-9; T. D. Barnes, 'Who were the Nobility in the Roman Empire?', Phoenix, 18 (1974), pp. 444-49.

[7] Marcone, CAH XIII, p. 357; M. T. W. Arnheim, The Senatorial Aristocracy in the Later Roman Empire (Oxford: Clarendon, 1972), p. 143.

[8] R. Van Dam, Leadership and Community in Late Antique Gaul (Berkeley: University of California Press, 1985), pp. 9-56; R. W. Mathisen, Roman Aristocrats in Barbarian Gaul: Strategies for Survival in an Age of Transition (Austin, Tex.: University of Texas Press, 1993), pp. 17-24.

[9] Av. Cameron, The Later Roman Empire (London: Fontana, 1993), p. 110.

[10] Codex Justinianus xi. 52. 1, 393, servi tamen terrae ipsius cui nati sunt aestimentur nec recedendi quo velint aut permutandi loca habeant facultatem, sed possessor eorum iure utatur et patroni sollicitudine et domini potestate.

[11] Codex Theodosianus, v. 17. 1, ipsos etiam colonos qui fugam meditantur in servilem condicionem ferro ligari convenient ut official quae liberis congruent merito servilis condemnationis compellantur implere.

[12] G. Ostrogorsky, History of the Byzantine State, (Oxford: Blackwell, 1993), pp. 40-41; Geary (1988), pp. 35-36.

[13] A. J. B. Sirks, 'The Farmer, the Landlord and the Law in the Fifth Century', in Law, Society, and Authority in Late Antiquity, ed. by R. W. Mathisen (Oxford, 2001), pp. 256-71 (p. 260).

[14] Arnheim, p. 144.

[15] Salvian, De gubernatione Dei, 5.8, 5.24-28.

[16] Salvian, De gub. Dei, 5.38.

[17] Salvian, De gub. Dei, 5.45.

[18] Rutilius Namatianus, De reditu suo, 1.27; Sidonius Apollinaris, Epistulae 8.9.2; Paulinus of Pella, Eucharisticon, 285-90; Carmen de providentia Dei, passim.

[19] Claudius Mamertinus, Panegyrici latini, 11/3.4, ut jam barbari desiderarentur, ut praeoptaretur a miseris fortuna captorum.

[20] Faustus of Riez, De gratia Dei, 1.16.

[21] Salvian, De gub. Dei, 5.8.

[22] Orosius, 7.41.7; Salvian, De gub. Dei, 5.21.3.

[23] Mathisen, pp. 67-76.

[24] C. E. Minor, '"Bagaudae" or "Bacaudae"?', Traditio, 31 (1975), pp. 318-22.

[25] Zosimus, 6.2.3-5.

[26] E. A. Thompson, Romans and Barbarians: The Decline of the Western Empire (London: University of Wisconsin Press, 1982; repr. 2002), pp. 34-35.

[27] Van Dam, pp. 41, 46, 48, 53.

[28] Van Dam, p. 41.

[29] Zosimus, 6.5.3.

[30] Zosimus, 6.5.3, ??????????? ??? ???? ??????? ????????, ??????? ?? ???' ???????? ????????? ??????????.

[31] Rutilius Namatiantus, De reditu suo, 1.213-216, Cuius Armoricas pater Exsuperantius ora/ nunc postliminium pacis amare docet;/ leges restituit libertatemque reducit/ et servos famulis non sinit esse suis.

[32] Van Dam, p. 40.

[33] Paul. Pell. Euch., 340.

[34] Chron. 452 s.a. 435, Gallia ulterior Tibattonem principem rebellionis secuta a Romana societate discessit; a quo tracto initio omnia paene Galliarum servitia in Bacaudam conspiravere.

[35] Chron. 452 s.a. 448, Eudoxius arte medicus pravi sed exercitati ingenii in Bacauda id temporis mota delatus ad Chunos confugit.

[36] I. N. Wood, 'The Barbarian Invasions and First Settlements', in The Cambridge Ancient History XIII: The Late Empire, A.D. 337-425, ed. by Av. Cameron and P. Garnsey (Cambridge: Cambridge University Press, 1998), pp. 516-37 (p. 532); Thompson, p. 50.

[37] W. Goffart, Barbarians and Romans A.D. 418-584: The Techniques of Accommodation (Princeton: Princeton University Press, 1980), passim.

[38] Barnish, pp. 170-195; H. Sivan, 'On foederati, hospitalitas and the Settlement of the Goths in 418 AD', in American Journal of Philology, 108 (1987), pp. 759-72.

[39] Jones, p. 196.

[40] Mathisen, pp. 18-24; Wood, CAH XIII, p. 534.

[41] P. Amory, 'Names, Ethnic Identity and Community in Fifth- and Sixth-century Burgundy', Viator, 25 (1994), pp. 1-30 (p. 4).

[42] J. Harries, 'Legal Culture and Identity in the Fifth-century West', in Ethnicity and Culture in Late Antiquity, ed. by S. Mitchell and G. Greatrex (London: Duckworth, 2000), pp. 45-58 (pp. 48-50).

eSharp issue: autumn 2004. © Leslie Dodd 2004. All rights reserved. ISSN 1742-4542.