Adventus, patrocinium, and the urban landscape in late Roman Britain

Alex Woolf

In 1980 Richard Reece published his paper “Town and country:
the end of Roman Britain” in which he claimed “in cruder terms, that the
towns of Roman Britain had gone by 350” (Reece 1980, 77). Reece outlined a model that pictured organised civic life disintegrating in the third
century and a more limited use of the town space by “Romanised”
people petering out in the final quarter of the fourth century, if not before: fifth century occupation like that uncovered by Barker at Wroxeter
or the Grubenhtiuser from Canterbury was not of Roman character and
therefore indicates reuse rather than continuity (op. cit, 84). The buildings of clearly “Roman” style present in the archaeological record of the
fourth century at some towns are, Reece asserted, merely villas or farms,
almost coincidentally situated within the redundant walls. Some administrative functions are admitted to be a possibility in a handful of cases
but these are imagined to have been closer to early medieval royal
structures in fonn than to those of a civic and bureaucratic classical state.
More recently Simon Esmonde Cleary (1989, esp. 64-85 and 131-161)
has summarised the opposing views. Whilst admitting that examples of 34 A. D. WOOLF
public buildings going out of use can be found, at sites such as Wroxeter
and Silchester, he asserts that they continue in primary use through the
fourth century at at least as many other civic centres, if not at more (ibid,
71-72). Esmonde Cleary also points out that reuse of civic masonry in late
defences is rare in Britain compared to Gaul, countering Reece’s claim
that French archaeologists have correctly identified a situation in the
north of their country which also prevailed, unrecognised by modern
archaeologists, in Britain (Reece 1980, 80). He does, however, concede
that, apart from defensive works, no new civic building can be detected
in the fourth century. His survey also indicates that there may well have
been fewer “commercial” buildings, that is to say the well-known strip
buildings, in fourth century towns than in those of the second century.
The 1/ day to day production and/ or distribution of everyday goods such
as leather-work, metalwork, pottery, textiles and woodwork” (Esmonde
Geary 1989, 75) are in evidence for the later period, but such activities
are also in evidence from many rural sites and so may be ancillary to,
rather the than cause of, residence. Esmonde Cleary’s main argument is
that relatively sophisticated town houses not only continue to exist in the
fourth century but may well increase both in numbers and in importance.
Finally he points to large sub-urban cemeteries as a sign of high urban
population.
Esmonde Cleary agrees with Reece in saying that whatever is happening in the towns in the fifth century they are not, by and large, Roman.
Indeed, as I understand it, the plea of his book is for archaeologists and
historians alike to recognise that there is no transition between Roman
Britain and Anglo-Saxon England; another, intermediary, society of
independent Britons emerges and lasts for at least fifty years in the east of
the country and for far longer over most of it The disagreement comes,
however, in the speed of transition. Reece claims that Britain was already
becoming medieval by the late third century while Esmonde Cleary sees
a fairly rapid disappearance of romanitas in the decades following 380. He
does not claim that time stood still between the growth of RomanoBritish urbanism in the second century and the dread year of 410, but he
sees the changes as relatively minor developments within an essentially
antique milieu.
Given Reece’s self-confessed penchant for controversy it might be
argued that the differences between their stances are not so very different, but it is my intention to argue that whilst quite radical changes did
take place through the duration of Romano-British history, changing the
function of the civic centre fundamentally, the transition identified in the lEE URBAN LANDSCAPE IN LATE ROMAN BRITAIN 35
archaeological record by Esmonde Cleary is indeed rapid, and is a result
of the function of the city in the late empire.
SOCIAL DEVELOPMENT IN CMC BRITAIN: AN ALTERNATIVE MODEL
In order to understand the changing role of the civic centre in RomanoBritish society it is necessary to produce a brief narrative of evolutionary
trends in the early first millennium in Britain. With deference to Reece
(Reece 1988), what follows is “my Roman Britain”.
The policy followed by the Romans for establishing an administrative
framework for their growing empire, in the western provinces at least,
was to offer the various polities they subdued and allied to themselves a
republican constitution, modelled more or less closely on their own (see
Gonzalez 1986, for Lex Irnicana, the only western civic code extant). Thus
the early empire was technically a league of Rome and a vast number of
allied city states. In order to conform to this model an urban centre was
established on a nodal point of the imperial road network. From an
imperial perspective these centres would provide the contact point for
tribute collection and for tendering additional contracts for the support of
the superstructure: the army and the civil service.
In Britain the earliest phase of such “urban” development largely comprised administrative buildings, perhaps a bath house, and strip buildings. These last are supposedly commercial installations, but were
possibly also used in the early phase for tribute collection. “Town
houses”, the residences of the local leadership, generally supposed to be
the curiales, the councillors of the republic’s assembly, only began to proliferate in most cities towards the middle of the second century. It should
also be noted that prior to this the few villas in the countryside seem to
be on a rather grand scale, and that these increase in number and decrease in grandeur alongside the proliferation of the town house.
The gift of a constitution does not make a republic overnight In many
cases the gift itself may have been quite slow in coming. The earliest
handful of grand villas should probably be ascribed to a group we can
call, rather clumsily, “contact period negotiators”. This set might include
tribal kings, war leaders, the sort of people Caesar refers to as “leading
magistrates” and Quislings. For most of the first century, and indeed beyond, the imperial administration in Britain was largely military in
nature, and campaigning of one sort or another was almost continuous.
Allied British troops played a large part in these campaigns (although
this is often ignored by modern commentators), and relationships be-36 A. D. WOOLF
tween native elites and the imperial structure were probably forged as
much under canvas or in the saddle as at the new civic centres.
From the mid-second to the mid-third century the republican ideal was
probably as close to realisation as it would ever be. Local leaders competed for election to the council and support in other parts of the
democratic processes (according to Lex Imicana the citizen body had
voting rights on certain issues such as the election of judges). In the civic
centres they vied for the attention of provincial and central government
officers as they passed through on their various rounds, hoping to win
government supply contracts or to gain preferment for a kinsman or
client or even some imperial honour for themselves.
In the countryside this period is characterised by the relatively
widespread dissemination of “Roman” pottery types, such as Samian
ware, which turn up on many of the so-called native farms, and is often
the only indicator that the site is no longer in the Iron Age. Reece (1980,
86) and many others are happy to see this as a sign of a commercial boom
with the “peasants” selling their wares at the civic centres, and later the
small towns, and buying up imported goods. As Guy Bois (1992, 76-77)
has recently pointed out, however, the civic centres of these antique pagi
were far too widely spaced to have served any but a tiny minority of their
citizens as a market place. In Britain, Lincoln and Leicester are fifty-one
miles apart and Lands End is one hundred and twenty-three miles from
Exeter. Even if we allow ourselves the lUXUry of adding the “small
towns” of Roman Britain to this list we still cannot fill the gap adequately. As Bois correctly asserts, usefully spaced market towns are a
product of the tenth century and later in most of western Europe. If, on
the other hand, we allow ourselves to be convinced by the republican
ideal, as I tend to be, then we might interpret this phenomenon as evidence of the establishment of a patron-client network in which pots (or
perhaps primarily their contents?) were exchanged for popularity, support, and, ultimately, power. The curiales, actual or aspiring, owned town
houses and villas, living primarily on the latter and visiting the town for
curial functions and, importantly, for the visits of the civil governor, his
officers, or military commanders.
The horizons and aspirations of most Britons at this period, even
amongst the elites, probably extended little beyond their own individual
republics. For them the civic centre was not just the symbol but also the
source of empire. A major function of the town house may have been to
convince visiting provincial and imperial functionaries of the status and
sympathy of its owner. Those curiales who put on the most convincing TIlE URBAN LANDSCAPE IN LATE ROMAN BRITAIN 37
performance in the civic centre, probably the only part of most republics
a governor would ever see, would win contracts and favours. These in
turn could be used both to enrich himself and his immediate kinsmen
and to reward his clients for their support. Success in the civic centre,
however, was dependent upon attracting clients in the first place and this
was the function of the villa. With its rectilinear plan, its plaster facing,
and tiled roof, and no doubt many other features, the villa fulfilled the
same function on a local scale as the town did in the republic as a whole.
The self-evident romanitas of the villa was a symbol of the civic centre,
which was, for most people, the empire. If we consider the empire as a
potential source of power for the elites we can see the road network as a
mains system, the civic centre as a wall socket, and the villa as the
appliance. The villa signifies a man who has access to the ear of the governor, whose court of appeal annually toured the province, who might be
able to secure tax concessions or to find a place in imperial service for
one’s son. Before the middle of the second century none of this had been
necessary as the relationships of clientship, both between citizens and
curiales and between curiales and the imperial officers, had been carried
on in a military context The demilitarisation of civic Britain, perhaps
after the suppression of the Brigantian rebellion of c. 154 (Salway 1984,
199), had demanded the establishment of a new order in which
citizenship was not wholly dependent on military capacity, nor public
office on the qualities demanded by military leadership.
Reece, however, is correct to assert that in the third century the scene
was changing. The paradox of Rome’s republican gift to the civitates
under its imperium lay in the fact that it had failed to rectify the social
problems which had led to the disfunction of its own republic, indeed, it
had taken them on board the establishment Successful competition for
access to external sources of wealth, in the case of Rome the spoils of war
and “private diplomacy”, and in the case of the British republics the patronage (including commercial opportunities) of imperial functionaries,
led to a small number of families being able to gain a massive economic
advantage over their fellows. Wealth accumulated in this way was invested in land and larger agricultural estates began to emerge. The
maintenance of town houses in the civic centres was now dependent
upon slave labour and the need of the most successful curiales, the principales, for clients declined. As a result of these changes imported goods
ceased to reach the less-well-off households and they are henceforth
harder to spot in the archaeological record
In this way civic office and public service ceased to be the keys to con-38 A. D. WOOLF
tinuing patronage from the provincial circuit, but the circuit itself had not
declined in importance, indeed it was about to flower as it had never
done before. The rise of large estates should not be overestimated. Some
of the modem literature gives the impression, intentionally or not I am
not sure, that the vast majority of the land in late Roman Britain was
divided up between estates based on villas, and that all other rural
settlements were occupied by tenants and coloni. I find this very difficult
to believe. Esmonde Cleary, citing the work of Jean-Marie Carrie (1982, in
Esmonde Cleary 1989, 29), finnly refutes the dominance of the colonate in
the late antique west The problem is, as Carrie says, historiographical,
too many scholars wish to equate the fall of the Roman empire with the
rise of feudalism (e.g. Wickham 1986). When all is said and done lithe
Roman empire” is just a name and its death is of no more consequence
than that of Alexander. Those who see the empire as a system are misguided; it was an entity which utilised various systems in different places
and times. Feudalism, as has been said many times, has too many meanings, but the one most often used in this context is that utilised by Marx
which identifies the seigniorial ban, which reduced the peasantry to tied
tenants owing both rent and services to a territorial lord, as the primary
diagnostic. As many scholars, such as Guy Bois (1992) and Georges Duby
(e.g. 1981), have recently asserted this “mode of production”, if I may use
that phrase, did not become the dominant mode, though it may well have
existed before then, until the tenth and eleventh centuries, indeed Bois
would like to see Finley’s ancient economy (1973) persisting up until
c. 1000. This may be going a bit far but I suspect that a distinct postimperial phase may be required. Coloni existed, and their number almost
certainly increased under the late empire, but they were probably never a
major class presence in any landscape. The estates that grew up in Roman
Britain, equally, put their owners far ahead of his fellow citizens but not
over their heads in any legal sense, the bulk of land even in late Roman
Britain was probably the allodial property of small holders and the result
of this shift in balance was primarily to exclude them from the civic
centres.
The Reece-Esmonde Cleary debate on the continued existence of towns
in the fourth century centres on their function in society. Reece argued
that they lost their social and economic r6le but may have retained their
administrative r6le, Esmonde Cleary that things had changed only superficially. Perversely, I would claim that they retained their social and
economic r61e but lost their administrative one. The archaeological evidence, put simply, shows a continuity of large town houses fairly widely TIlE URBAN LANDSCAPE INLA1E ROMAN BRITAlN 39
spaced, with a decline in the number of strip buildings, though not necessarily their disappearance, and, in some centres, the abandonment or
unorthodox use of civic buildings (for examples see Reece and Esmonde
Cleary, op. cit, or any of the standard texts on Roman Britain). My interpretation is probably clear by now, but before I go any further let us turn
to another theme that has played a great part in assessing the decline of
the civic centres, the fate of the decurionate.
In his monumental work The Later Roman Empire (1964) A. H. M. Jones
dealt at great length with the decline of the decurionate, the curial order.
We are told by various sources that evasion of service was rife and we
know, even from as early as the Lex irnicana, that curiales were empowered to conscript additional colleagues to the order in the event of no
candidates standing for election. The property qualification was gradually lowered to make up numbers as those families which had originally
supplied the curiales either became impoverished or gained exemption.1
The possibilities for exemption increased as time progressed due to a
gradual but far-reaching change in the structure of the empire. Since the
late republic it had been possible for the Roman state to award individual
allies with the honour of citizenship. Initially this was probably seen as
essentially honorific and in real terms meant that these new citizens were
treated more courteously by Roman officials and could appeal to Roman
magistrates, essentially provincial governors, if they felt unsatisfied with
the justice meted out to them in allied cities. Citizenship was hereditary,
however, and as time went by, and as the territory of the empire expanded, a higher and higher proportion of Roman citizens were of allied
descent Citizenship also carried with it the right to hold office within the
Roman state and so, as the military and· civil services expanded to administer the growing empire, more and more men in more and more
important positions were of allied rather than Latin descent From the
time of Marcus Aurelius (161-180) the number of grants of citizenship,
particularly in the west, where Romanisation was well under way
amongst the curial classes, increased enormously. In 212 Caracalla, himself of allied descent, extended the rights of citizenship to all free men
resident in the empire. Caracalla’s edict in effect recognised officially
what had already taken place; the colonial empire of Rome had become a
territorial state of Romania
In order to encourage the best of the allied citizenry to serve in the imperial service the late empire offered exemption from their native civic
duties for three generations to those appointed. Even a short commission
held such exemption and so service was avidly sought The adventus of 40 A. D. WOOLF
the governor, high-ranking military officers, and, after the establishment
of the tetrarchy, a caesar or an augustus in person, would have become
an eagerly anticipated event The local elites in our British republics
would have invested even more in their individual urban facades in the
quest for patronage, for now it meant not just the chance to compete
more effectively against their colleagues and compatriots but to escape
from that level of life altogether. Paradoxically those people who invested
most in their urban image were the very same as those who were least
willing to meet their civic costs; one might say that the town grew at the
expense of the city. In addition to these wealthy types who rose above the
demands of patriotism, Christian clerics were also, from the time of
Constantine, exempt from the civic munus (though one can hardly
imagine that this particular exemption had much effect on Britain). The
fourth-century army was also restructured, service becoming hereditary,
with those who survived receiving a small estate and exemption from tax
and civic duties; if their patrilineage was able to survive for a generation
or two even common soldiers might become significant men on a local
scale. Since civic expenditure was largely paid for out of the pockets of
magistrates and curiales the effective removal of large tracts of productive
land from the civic sphere, once its owners gained exemption, must have
had some effect upon the efficacy of civic government and justice.
Nick Higham (1992, 45) has succinctly summed up the resulting
dilemma of the surviving curiales now under compulsion from the state
and made corporately responsible for any shortfallings in tax payments
or services. Those amongst the wealthy who could not immediately
secure formal exemption might attempt to transfer their property to
another republic where they were not under obligation, perhaps most
safely in another province where they could not be informed against to
the governor. Indeed, those seeking further imperial preferment may
well have attempted to cluster their property in regions more likely to
experience adventus. The clustering of villas around places like Trier on
the continent, or Cirencester, capital of Britannia Prima and residence of
the governor, may reflect such strategies, as may urban housing in such
cities. The chances of such urban investment paying off were further
increased by the practices of the new field army, the praesentales or
comitatenses, who were better paid than the traditional legionaries and the
contemporary limitanei and who were billeted in the towns through
which they passed whilst on service (ibid., 45 and Jones 1964). Undoubtedly the most attractive residences would be allocated to the most senior
officers and equally undoubtedly this would be an honour worth TIIE URBAN LANDSCAPE IN LATE ROMAN BRITAIN 41
competing for.
But what of the rural poor? What of those small allod holders who
owned too little land to qualify for the decurionate even in the fourth
century? I have already suggested that their support was no longer
sought by the local elites so that they might be noticed by the imperial
administrators. In consequence the gifts of imported and industrially
manufactured goods ceased to flow into their households, which consequently become much harder for archaeologists to find on the ground.
Without this hold on the curiales and magistrates the only way these
humiliores could have ensured support in civic or judicial proceedings
would have been to seek out patronage through offering voluntary tnbute to curiales or local principales who might exert pressure upon the
magistrates.
Such a course, almost certainly, was the road to the colonate; already
over taxed the expense of such gifts might prove ruinous to the fragile
economy of the small allod holder. As it was, social and demographic
pressures may have been eroding his position in any case. Several generations of pax romana must have led to the increasing sub-division of
allods through partiple inheritance. This was perhaps compounded by
the institution of a recruitment tax, which consortia of local honestiores
could pay to keep army recruiters out of the district Gones 1964,
In the pre-Roman and early colonial period military service and endemic
warfare must have ensured that many sons did not return to the
patrimony and that those who did married later, thus ensuring that there
would be fewer mouths to feed (early marriage leads to more individuals
and a greater number of generations coexisting within one household). A
similar crisis developed in the West Highlands and Islands of Scotland
during dernilitarisation from the early seventeenth to the late eighteenth
century. One result in Scotland was increased nucleation of settlement,
which minimised the destruction of agricultural land and allowed for
easy labour cooperation between close kinsmen. A second result was the
expansion of spade tillage which allowed unploughable marginal land to
be put under cultivation (Dodgshon 1992). One wonders whether either
of these two processes might be visible in the archaeological record from
late Roman Britain; unfortunately the dating of rural sites with no imported pottery and no coinage is exceedingly difficult
And what of the judicial and civic needs of those small allod holders
who were not prepared or constrained through proximity to beggar
themselves to the honestiores? The most likely course for them to have
taken would have been to retreat from civic activity altogether and to fall 42 A. D. WOOLF
back on local courts. We know these existed and that the civitas territories
were sub-divided into local districts (Salway 1984,534-535) even though
we have very little precise infonnation about them. An idea of how they
might have appeared and worked can be gained from Wendy Davies’
Small Worlds (1988), an excellent study of the plebes surrounding the
abbey of Redon, in eastern Brittany, in the ninth century. Place-name
evidence suggests that such plebes existed in Dumnonia also and they
may be a Romano-British institution, but even without demanding a relationship of descent we might still see their context as analogous to that
which we seek in fourth-century Britain. These small locales seem to have
averaged 40-50 square kilometres in extent, though with great variation
in individual cases, with a distance of between six and seven kilometres
between their foci, in other words an hour’s walk across (ibid., 64-65).
Disputes concerning property in the plebs would be heard before
the plebenses and settled locally, either formally under the presidency of a machtiern or a ruler’s representative, or infonnally; local
people served as judgement finders in fonnal cases and their
evidence as impartial witnesses might be taken. Local matters
were almost invariably settled by the community, and it is perfectly clear that a strong sense of community might be evidenced
in such proceedings, and a conscious one.
The machtiern mentioned in Davies account (whose title means “surety
lord “), was a local big landholder whose relationship to the plebs was
nevertheless that of patrone rather than seignior and any “gift” presented
in gratitude for his participation would be donated on behalf of the plebs
as a whole rather than by one or other of the litigants. Such civil divisions
seem to have existed elsewhere in early medieval Gaul where they where
known variously as agri and vicariae. Although even the estates of nobiles
technically lay within these territories, or even across them, there seems
to have been a process of self-selection whereby the rich attended the
count’s court and the poor the local assembly. The presence of a patrone
seems in no sense to have been a requisite for such courts and the maiores
and vicari who generally presided seem to have been local goodmen.
With the obvious substitution of the civic magistrates or governor for the
count, I would strongly suggest that this kind of institution should be
traced back to late Roman times if not to a much earlier age.
Why small allod holders should resist the appeal of the colona/e, which
may well have offered greater security than their own precarious economy might provide, is probably to be answered by the social value of TIIE URBAN LANDSCAPE IN LATE ROMAN BRITAIN 43
land ownership in pre-capitalist societies and the honour and status that
an allod holder retains, however humble his material condition. One
symptom that may develop from this poverty is the need for the poor
man to distinguish himself visibly from the colonus and even the casatus,
who may well be his material superiors or at least equals. A visible and
portable sign of his freedom that he could have to hand at assemblies and
on other public occasions that would ensure his claim to being free is
what he required. In short, a weapon.
Lex Julia forbade the carriage of arms on the public highway, but not
their possession. Brent Shaw (1984) in his discussion of banditry in the
empire makes it quite clear that weapons were available to those who felt
the need for them, and Wolf Liebeschuetz (1991, 231-232) has pointed to
Synesius of Cyrene’s purchase of weapons on the market, and domestic
manufacture of others, in order to arm his private militia Going back to
the second century it is hard to imagine a complete disarming of the
British soldiery, and by the end of the third century, if not earlier, barbarian raids were a distinct possibility in many areas. This is not to
suggest that Roman Britain was bristling with armed men, at least not before the fourth century, but the parallel could be drawn perhaps with
Ireland over the last century, and other areas, such as the Balkans, in
which paramilitary activity has been recurrent, where weapons will often
go “under the floor-boards” for years, and sometimes decades, when the
owner feels that they are no longer necessary but seeks the reassurance of
their presence against the eventuality that circumstances may change.
With the increasing constraints on their access to civic justice the
plebenses may at times, particularly if in dispute with an inhabitant of a
neighbouring plebs (rather than of their own), have had to resort to the
exercise of distraint backed up with force. While it is not the place here to
discuss the continental phenomenon of bacaudae at any length, the gradual transition of the society of the rural poor to one dependent upon the
exercise of distraint, with its constant shadow the bloodfeud, may well
provide a background for the structures utilised by men like Tibatto in
their “uprisings”. It is a well-known feature of feuding societies that conditions occasionally lead to private vendettas escalating to a scale difficult
to distinguish from political activity (see Byock 1982; Wormald 1981; and
Hopwood 1990). In Britain there is little if any evidence for bacaudic activity (but then, to be frank, there is little evidence for anything at all),
and it may never have become necessary. Dynamic leadership for those
who wished to reach beyond the boundaries of the plebs may have been
provided by the conunanders of the North British raiding parties which 44 A. D. WOOLF
seem to have frequented the diocese throughout the period; linguistically
and culturally very close to the humiliores, they may well have accumulated adherents as they rode heroically across a landscape devoid of the
richer romani, who were doubtless either cowering behind the walls of
their civic centres or eagerly handing over their silver and horses to the
barbarians in order to save their beautiful villas from the torch. 2
CONCLUSION
The reader may have been surprised when I, earlier, suggested an expansion of the rural population at this period when it is commonplace to
claim a drastic reduction. My argument is that with the severing of the
patron-client relationships between the local elites and the humiliores, the
plebenses, the paradox emerged that whilst the population of western
Europe grew so the population of the empire, as a social construct, declined. The regional elites supporting themselves on estates worked by
slaves and some coloni had become, in effect, the bottom tier of the imperial structure, a structure based upon patrocinium, and the rural poor, the
plebenses, lay outside it all together. In Britain recent environmental evidence, mostly palynological, has put paid to the idea of a late antique return to wilderness and may even suggest an increase in the land under
cultivation (although as I have suggested this may not represent economic growth) during our period (Higham 1992, 77-80; also Whittaker 1976).
The elites whilst preferring, when possible, to enjoy the otium of their
country estates, as evidenced by the wealth of those few villas that survived into the later fourth century, were dependent upon the patronage
of emperors, counts, governors, and the like in order to enjoy it As van
Dam (1985) rightly states the adventus of the emperor or his senior representatives was the high point in provincial life. These adventus were
Ulban events and to make the most of them the British elites must have
maintained impressive urban establishments. The late Roman town
house, set in its spacious gardens, was that establishment We must not,
however, assume that one house represents one aristocratic family.
Comites and their kind, let alone emperors, were rare visitors to fourthcentury Britain and seldom stayed long. The full twenty-eight city tour
was probably not an option. To ensure access to the potentate it was
probably necessary to be prepared to anticipate his movements, and perhaps to maintain more than one town house. A rich Dobunnian, we
might imagine, while idling most of his days hunting across the hills
above his home at somewhere like Woodchester, may well have main-THE URBAN LANDSCAPE IN LATE ROMAN BRITAIN 45
tained houses at Gloucester, Cirencester, and Bath, and perhaps even one
in the diocesan capital at London. The account of King Gunthchrarnn’s
adventus into Orleans at the beginning of Book Four of Gregory of Tours’
Historia Francorum is instructive in its description of bishops and other
notables gathered from allover western Gaul to enjoy his presence.
For most of the year, perhaps for years at a time in some cases, these
houses may have housed only a handful of servants just as the strip
buildings provided acconunodation for plumbers, masons, brickmakers,
and other artisans necessary for the up keep of the fabric. Esmonde
Cleary (1989, 80-81) claims the evidence of the cemeteries at sites like
Cirencester, Dorchester, and Winchester indicates a large permanent
urban population, but I would urge caution on two counts. Firstly, we
know of few rural cemeteries from this period so these are perhaps civic
rather than urban burial grounds. Secondly, skeletal evidence suggests
that there is an unusual sex ratio, with the number of males represented
considerably outweighing the females in the sample; bearing in mind the
role of urban sites as accorrunodation for comitatenses units this might be
explained if a significant number of the burials are in fact soldiers or
heads of household socialising with imperial officers (for sex ratios see
Nelson 1985). The fourth-century town was, indeed, a different animal
from the second-century town and in its difference lay the seeds of its
destruction.
Esmonde Cleary clearly notes a rapid disappearance of “town life” in
the first decades of the fifth century, very different from Reece’s gradual
decline, and he is right to do so. The urban landscape of later Roman
Britain was the stage for the adventus, the set for the fawnings of the
provincial elite, reduced to the status of clients, desperately trying to impress their masters into casting them the crumbs of patronage. After 409
there were no more emperors, no more counts, no more governors, no
more patrons. The British elites still had their estates and in each of their
republics made up a handful of local aristocrats but if they were to compete with one another for dominance under the new order they had to
return to the only resource available to them, their own patrocinium, and
the potential clients amongst the plebenses. The cities were empty shells
and the emperor was gone, the villa that at one time had symbolised its
owner’s access to the source of power now only symbolised his emasculation. Pax romana was gone, barbarians filled the highways, protection
and physical force was the new Rome. A new structuring of leadership,
with a new assemblage of symbols was required if the oligarchy was to
maintain its hegemony. 46 A. D. WOOLF
ACKNOWLEDGEMENTS
I would like to thank the following for their comments on draughts and
discussions of this paper, in part or as a whole: Karen Meadows, Colin
Merrony, John Moreland, Mike Parker Pearson, Ron Ross, Ross Samson,
and Greg Woolf. All errors and eccentricities remain mine alone.
NOTES
1. Eventually the property qualification dropped to 25 iugera, about 7 hectares.
Laid out as a block 7 ha would form a square with sides of 265 metres and
would take approximately 10.5 minutes to circumperambulate. Three such
“estates” could fit within the defences of the Iron-Age hillfort at Hod Hill,
Dorset The Bordeaux aristocrat Ausonius inherited a patrimony of 270 ha,
which would have taken an hour and six minutes to walk around and could
theoretically have provided the property qualification for 38 curiales. The
property qualification was based upon land within the republic so it is just
possible that curiales were not getting poorer but that men with estates on
the scale of Ausonius were deliberately splitting their property between republics (we know for certain that Julius Ausonius, the poet’s father, was on
the curial list in two republics: Bordeaux and Bazas).
2. Sidonius Apollinaris’ letter to Riothamus (Ep. ill, ix) speaks of slaves being
lured oft their masters’ estates, in the Auvergne(?), by British soldiers c. 470,
and see also Wolfram Herwig’s (1988) discussion of Gothic “ethnogenesis”.
On the social background of bacaudae I was willing to be convinced by Ray
Van Dam’s (1985) argument until my brother, Greg Woolf, pointed out the
consistently and resoundingly Celtic names of their leaders, surely a sign of
humble status better fitting the model of a divided society I present here. On
another occasion I hope to be able to deal with Pictish “ethnogenesis”.
BIBLIOGRAPHY
Bois, G. 1992 The Transfonnation of the Year One Thousand: The Village of
Lournand from Antiquity to Feudalism. Manchester: University of
Manchester Press.
Byock, Jesse L. 1982. Feud in the Icelandic Saga. Berkeley: California
University Press.
Carrie, J.-M. 1982. Le “colonat du Bas-Empire”: un mythe historiographique? Opus 1 (2), 351-370.
Davies, Wendy 1988. Small Worlds: The Village Community in Early Medieval
Brittany. London: Duckworth.
Dodgshon, R. A. 1992. Farming practice in the Western Highlands and
Islands before crofting: a study in cultural inertia or opportunity costs? TIIE UREAL’I’ LANDSCAPE IN LA TE ROMAN BRITAIN
Rural History 3 (2), 173-189.
Duby, Georges 1981. The Three Orders: Feudal Society Imagined. Chicago and
London: Chicago University Press.
Esmonde Cleary, A. S. 1989, The Ending of Roman Britain. London: Batsford.
Finley, Moses 1973. The Ancient Economy. Harmondsworth: Penguin.
Gonzalez, J. 1986. Lex Irnicana.Journai of Roman Studies 76, 147-243.
Herwig, Wolfram 1988. History of the Goths, London: University of
California Press.
Higham, Nicholas 1992. Rome, Britain and the Anglo-Saxon. London: Seaby.
Hopwood, Keith 1990. Bandits, elites and rural order. InA. Wallace-Hadrill
(ed.), Patronage in Ancient Society, 171-188. London: Routledge.
Jones, A. H. M. 1964. The Later Roman Empire. Oxford: Blackwell.
Liebeschuetz, J. H. W. G. 1991. Barbarians and Bishops: Army, OlUrch and
State in the Age of Arcadius and Otrysostom. Oxford: Oxford University
Press.
Nelson, A. 1985. A study of stature, sex and age ratios and average age of
death from the Romano-British to the Late Anglo-Saxon period. Unpublished Sheffield MA dissertation.
Reece, Richard 1980. Town and country: the end of Roman Britain. World
Archaeology 12 (1), 77-92.
Reece, Richard 1988. My Roman Britain. Cirencester: The Old Apple Loft.
Salway, Peter 1984. Roman Britain. Oxford: Oxford University Press.
Shaw, B. D. 1984. Bandits in the Roman empire. Past and Present 105, 3-52.
Van Dam, Raymond 1985. Leadership and Community in Late Antique Gaul.
California: University of California Press.
Whittaker, Christopher R. 1976. Agri deserti. In M. I. Finley (ed.), Studies in
Roman Property, 137-165. Cambridge: Cambridge University Press.
Wickham, Christopher J. 1986. The other transition: from the ancient world
to feudalism. Past and Present 103, 3-36.
Wormald, Jennifer 1981. Bloodfeud, kindred and government in early
modern Scotland. Past and Present 87, 54-97.