1

Exploring late Iron Age settlement in Britain and the near Continent:

Reading Edward Gibbon’s The Decline and Fall of the Roman Empire

and examining the significance of landscape, place, and water in

settlement studies

Adam Rogers

INTRODUCTION

This paper examines ways of considering landscapes, settlements, and significant places in

the late Iron Age and the development of Roman urbanism in North-West Europe. It is

divided into two related parts: firstly an historiographical section analysing an important

example of how early studies, and the social contexts in which they were written, have been

hugely influential in setting research agenda which remain strong in archaeological research;

and secondly a study of archaeological settlement evidence demonstrating how we can

interpret the data in different ways. The first section will look at the thought and work of the

ancient historian Edward Gibbon (1737–94) whilst the second will examine new perspectives

that are now emerging from detailed landscape studies focused on the later Iron Age.

Edward Gibbon’s six-volume work The History of the Decline and Fall of the Roman Empire

was published between 1776 and 1788 and was quickly established as a major work. It

became influential on studies not only of the later Roman period, playing a significant part in

the formulation of attitudes towards the later Empire, but the Roman Empire as a whole. The

text has, however, now been criticized for its emphasis on decline (e.g., Hingley and Rogers

forthcoming; McKitterick and Quinault eds. 1997; Porter 1988) and there is now more

caution in using terms such as decline in studies of the later Roman period (e.g. Lavan ed.

2001; Rogers 2005; Wickham 2005).

What has not been acknowledged so much, however, is the way in which Gibbon’s work

covered the rise and growth of the Roman Empire, including pre-existing settlement patterns

in Britain and the near Continent. His descriptions of this period were mainly used as a tool

for establishing what he saw as the greatness of conquest and the ‘Golden Age’, setting the

scene for its decline. This paper will explore the ways in which Gibbon conceptualized and

described pre-Roman settlement in Western Europe and the contribution that this has played

in the research tradition of late Iron Age settlement in this area. Also important is the social

context in which Gibbon was writing and how this will have contributed a major part to the

2

formulation of his attitudes. As a British historian, Gibbon’s work has had a great impact

amongst British scholars, but Gibbon has also been influential in mainland Europe.

Of course, it is not only Gibbon that has had an impact on archaeological work and the

development of research traditions, both of the decline of the Roman Empire and its growth

(work by Theodor Mommsen (1817–1903) and Camille Jullian (1859–1933) is also discussed

below), but his work has been highly prominent and deserves attention. Focusing on Gibbon

will demonstrate the potential of analysing past work in moving subjects forward. Recent

years have seen a rise in important work on the historiography of Roman and prehistoric

archaeology and the formation of tradition especially in Britain (e.g., Dyson 2006; Hingley

2000; Todd 2004) but also in France (e.g. Goudineau 1998) and other countries. Here,

acknowledging the complexities in the research traditions of late Iron Age settlement and

urbanization after the conquest will highlight the necessity to reconsider the way in which

these sites have been studied and conceptualized.

BACKGROUND TO GIBBON’S LIFE

There are numerous accounts of Edward Gibbon’s life including his own Memoirs (1966)

(e.g., Burrow 1985; Low 1937; Porter 1988). Gibbon was born into a relatively privileged

background in south London, his father being able to assume the life of a gentleman because

of the wealth made by his own father in business. This wealth allowed Gibbon to attend a

school that taught Latin and Greek and then, at 15 years, Magdalen College, Oxford. From an

early age Gibbon expressed considerable interest in Roman history and this was strengthened

by his trip to Italy on the Grand Tour but, as Pocock (1999, 276–7) has remarked, this

consisted of a voyage “through the history of taste” seeing and constructing ancient Rome and

Italy through eighteenth century eyes. Through his upbringing, Gibbon appreciated the British

aristocratic system, writing of the “superior prerogative of birth” (Gibbon 1994, vol. I, 188)1

and the validity of Empire and the civilization and order that it brought. He became a member

of parliament himself and belonged to a number of exclusive London clubs (Gibbon 1966,

155–6; Brownley 1976, 21). His inherited wealth allowed him to amass a large library, held in

his house in Bentinck Street, London, and devote his time to writing. Like the classical texts

Gibbon drew upon to write his work, The Decline and Fall itself soon became an essential

part of the education that was given to people of a similar privileged background to his own.

1 The references from The Decline and Fall are taken from the 1994 edited version of the work by David Womersley, published by Penguin in three volumes each containing two of Gibbon’s original volumes. The volume and page numbers in the text refer to the way in which Gibbon’s six volumes appear in this 1994 edition.

3

READING GIBBON

Gibbon was writing in a period in which the interest in history and historical works,

especially of the Roman Empire, was growing; the number of published historical works rose

considerably (McKitterick 1997, 164). History was hugely fashionable (Ghosh 1997, 277)

and as a statement in Horace Walpole’s (1717–97) letter to the poet William Mason (1725–

97) suggests – “there is just appeared a truly classic work” (W. Lewis ed. vol. 28, 1955, 243)

– Gibbon’s first volume was an immediate success and hailed a masterpiece. From that point

Gibbon became known as ‘the Historian of the Roman Empire’ (Pocock 1999, 292). The text

remained hugely popular and influential, and perhaps arguably even more so, into the late

nineteenth and early twentieth centuries. The image of Rome played an important part in the

nineteenth and early twentieth century social consciousness (Vance 1997); people were also

aware of the fate of the Roman Empire through Gibbon and wished to avoid a similar course

of events for the British Empire. The popularity and success of Gibbon’s work was partly

through his writing style and the language of his descriptions of Roman grandeur and

magnificence, but this was also combined with the lessons that could be learnt from his

graphic depictions of decline. Another lesson of Empire, however, was in the conquering and

transformation of land which Gibbon also described and provided images for his readers.

The antiquarians of Gibbon’s day, and earlier, were working in a largely similar social

context as Gibbon which would also have influenced their interpretations of the archaeology.

This intellectual atmosphere of the eighteenth century, to be examined below, would have had

a considerable impact upon Gibbon. It seems that there was a number of studies, attitudes,

and transformations at this time which came together and paved the way for nineteenth and

twentieth century developments, including archaeology.

GIBBON’S ATTITUDE TOWARDS PRE-ROMAN SETTLEMENT

Gibbon not only researched books but was interested in the surviving remains of Roman

structures especially in Rome itself. These remains were important in formulating his ideas, as

one of the famous passages from his Memoirs states: “I can never forget nor express the

strong emotions which agitated my mind as I first approached and entered the eternal City”

(Gibbon 1966, 134). Originally, he had intended to write a book solely on the “decay of the

city” rather than the Empire as a whole (ibid, 136); he describes how he began to “collect the

substance of my Roman decay” (ibid, 146). Gibbon also knew the works of English

antiquarians, including Whitaker, Gale, Stukeley, Camden, Dugdale, and Horsley

(Womersley 1994, xii), and he drew upon and commented on their writings. For

understanding Britain under the Empire he refered to “our own antiquarians, Camden and

4

Horsley” (vol. I, 33, note 8) and especially valued Camden stating that he was “the British

Strabo” and “the father of our antiquities” (vol. II, 997, note 109; vol. III, 22, note 11). In his

study of Roman roads he refers to the itineraries of “Gale and Stukeley for Britain, and M.

d’Anville for Gaul and Italy” (vol. I, 77, note 85). Gibbon was especially interested in the

antiquarian work carried out within towns and sections of his work are devoted to Rome

which he thought reflected the Empire as a whole. He envisaged and described the decline of

the Roman Empire through the physical destruction of its public buildings and monuments.

For Gibbon, the public buildings and cities – interpreted through the ‘enlightened’ mindset –

were what represented the Empire’s greatness and civilization. He not only uses this to

emphasise decline in the later Roman period but also the benefits of romanization in the

conquest period in areas such as Britain, Gaul, and Germany.

Antiquarian work indicates some limited early knowledge of pre-Roman places in Britain by

this time. William Camden’s (1551–1623) and John Speed’s (1542–1629) work, for instance,

linked some names mentioned within the classical texts with those found on pre-Roman

coinage (Hingley 2006, 333), and John Horsley’s Britannia Romana (1974 [1732])

demonstrates awareness of sites of pre-Roman Britain. At this early date, pre-Roman peoples

were being identified with known places and monuments in the landscape (Hingley 2006,

333) which Gibbon would have been able to draw upon for his understanding of the pre-

Roman West. These places were identified at a period in time when the ways of

understanding and interpreting landscape and urbanism in society were changing, in turn,

influencing attitudes towards place and space in the pre-Roman and Roman periods. This

influence has survived in some form to this day.

That Gibbon considered pre-Roman settlement in inferior terms to Roman towns is

represented by his statement that with conquest “(T)he spirit of improvement had passed the

Alps, and been felt even in the woods of Britain, which were gradually cleared away to open a

free space for convenient and elegant habitations. York was the seat of government; London

was already enriched by commerce; and Bath was celebrated for the salutary effects of the

medicinal waters” (vol. I, 75), clearly applying modern views of the towns onto the past. For

details on pre-Roman settlement Gibbon drew heavily on Caesar’s De Bello Gallico. He

wrote that “(W)e can only suppose them to have been rude fortifications, constructed in the

centre of woods, and designed to secure the women, children and cattle, whilst the warriors of

the tribe marched out to repel a sudden invasion”. Caesar’s writing refers in a number of

occasions to the importance of woodland for indigenous meetings, activities and settlement.

He recorded that the Suebi sent “their children and all their stuff to the woods” (B Gall.

IV.19) and the “Menapii had all hidden in their densest forests” (IV.38). He also wrote that

“the Britons call it a stronghold when they have fortified a thick-set woodland with rampart

5

and trench” (V.21) and that the stronghold of Cassivellaunus was “fenced by woods and

marshes” (ibid).

Gibbon also used woods as a method of emphasizing the savagery and danger of the

indigenous peoples. Attacks from barbarians came from woods as a contrast from the

civilization of walled towns and forts: “The crafty barbarians, who had lined the woods,

suddenly attacked the legions” (vol. I, 308–9). Other phrases reflecting this are: “the savage

warriors of Scythia issued from their forests” (vol. III, 121), “a crowd of naked savages

rushed from the woods” (vol. III, 281); “the secret paths of the woods” (vol. II, 1066), “dark

recesses of the woods” (vol. II, 1077) and the “thick and gloomy woods” (vol. II, 124). The

term “woods and morasses” occurs repeatedly throughout the work as a means of

emphasizing barbarity compared with Roman civilization. Woodland clearance and the

drainage of marshland by the Romans, as in his own time, were regarded by Gibbon as acts of

improvement, civilization, and rationalization.

Commenting on later times, for instance, Gibbon wrote that in Germany the “immense woods

have been gradually cleared” and the “morasses have been drained” (vol. I, 232). It “is the

happy consequence of the progress of arts and agriculture” that instead of “some rude

villages, thinly scattered among its woods and morasses, Germany produced a list of two

thousand three hundred walled towns” (vol. III, 512). A highly influential German historian

of the Roman Empire was Theodor Mommsen (1817–1903) whose most well known work is

the three-volume History of Rome (Römische Geschichte) covering the Republic, published

between 1854 and 1856, and then The Provinces of the Roman Empire from Caesar to

Diocletian (Die Provinzen von Caesar bis Diocletian) which appeared in 1885. Elements of

Mommsen’s writing were influenced by Gibbon, describing the benefits of Roman conquest

and the unifying nature of imperialism. With clear resonances of Gibbon, on pre-Roman

settlement in northern Europe he wrote that the “population during war sought protection in

the morasses and forests rather than behind their walls, and beyond the Thames the primitive

defence of the wooden abatis altogether took the place of towns” (1866, 218). Mommsen

himself was influential on the views of a number of historians including Francis Haverfield

(1860–1919) in Britain and Camille Jullian (1859–1933) in France who studied in Berlin

under Mommsen (Goudineau 1998, 9–10; Hingley 2005, 35).

Like Mommsen, Jullian took the Romanist viewpoint in his early work including the first

three parts of his eight-volume Histoire de la Gaule published between 1907 and 1921.

Jullian attended the École Normale Supérieure in Paris at the age of 18 and received his

doctorate at 24 in 1883 at the Sorbonne (Goudineau 1998, 8). This route of education

undoubtedly encouraged traditional perspectives and approaches towards the Roman conquest

of France and pre-Roman settlement. Indeed, his earlier book Vercingétorix (1900) was

6

written predominantly from a Romanist perspective and he originally considered writing a

history of Rome, rather than Gaul (Goudineau 1998, 9, 21; King 2001, 120). Jullian knew the

Greek and Roman texts well but unlike Gibbon he did not put so much emphasis on structural

remains and antiquarian studies. Jullian’s work has been very influential in France although it

is now receiving critical analysis and it is recognized that his writing privileged certain

themes and periods (Goudineau 1998, 31).

The later volumes of Histoire de la Gaule were not so strongly biased towards Roman

civilization as the earlier ones (King 2001, 120), although the ways in which pre-Roman

settlements are described does in many ways remain similar to Mommsen and Gibbon. In

volume V, for example, he wrote that the “Empire était l’apothéose de le ville”2 emphasizing

the civilization that Roman towns brought and this was accompanied in volumes II and V

with descriptions of the transformations of “marécages et forêts” (marshes and forests) around

towns (vol. II, 260–3; vol. V, 177–80) comparable to Gibbon’s writing on woods and

morasses. He also states that the marshes and forests would have been good defences (vol. V,

261)3 and that they would have been more prominent in northern France; a result of this being

that the north of France was considered less civilised than the south: “les grandes villes

étaient plus rares au nord”.4 Jullian does, however, acknowledge the sacred nature of some

wooded areas (e.g., vol. II, 155), as is mentioned in some classical texts such as Lucan (also

used by Gibbon),5 but not of wetlands.

Gibbon continued to use the imagery of woodland and marsh in his work to illustrate the

“decline” of the West after Rome: “Gaul was again overspread with woods” (vol. III, 481)

and in Britain “an ample space of wood and morass was resigned to the vague dominion of

nature” and areas returned to their primitive state of a “savage and solitary forest” (vol. III,

502–3). Jullian, who refers to Gibbon’s work (vol. VII, 3), also writes of “La Gaule en ruine”,

the “ruine des campagnes”, and that “peu à peu la forêt se rapprochait” (vol. VII, 16–7).

Describing decline in such terms as encroaching woods and the loss of order in the

countryside, as Gibbon, Jullian, and others did, will have influenced attitudes not only of the

later Roman period settlement but also the pre-conquest period and led to assumptions about

the way in which these places were experienced, valued, and understood.

2 The Empire was the pinnacle of the town. 3 As in the case of marshes, “les forêts présentaient les mêmes avantages militaires” (vol. V, 261). 4 The large towns were rarer in the north. 5 In the Pharsalia (III.399–432), Lucan provides a well known description of a sacred grove in the vicinity of Marseilles.

7

GIBBON AND ‘LANDSCAPE’

Although Gibbon referred mainly to classical texts in his work, then, he was writing at a time

of increasing rationalization and commoditization of land which would have influenced his

interpretations. These changes contributed towards the modern concept of ‘landscape’ which

has been powerful in the way in which attempts to understand oppida, other contemporary

sites, and their settings have been approached. Gibbon was working amidst a growing

empiricist understanding of the environment. It is not the purpose of this paper to go into

detail on the origins of the term and understanding of ‘landscape’, and there is considerable

writing on this subject (e.g. Brück 2007; Cosgrove 1984; Johnson 2007), but it can be seen in

part as being a product of the eighteenth century ideology of improvement and other social

changes. The Dutch painting tradition introduced the term landschap into English; painting

the land, contributing towards such terms as ‘picturesque’, objectified and distanced it,

separating people from it (Bevan 1997, 181). Landscape also formed a significant parts in

changing attitudes to economic aspects of life. Like painting, antiquarianism was one of a

range of pursuits dominated by the aristocracy during the eighteenth century (along with

travelling, gardening and drawing) which influenced and created perceptions of landscape

(Brück 2007, 244). Antiquarianism, and then archaeology with its interest in the context of

finds through excavation, were part of the process of the modernization and rationalization of

land (Lemaire 1997, 16; Schnapp 1996, 179–219).

Also emerging were concepts of environmental and economic marginality for areas that were

deemed too unsuitable for exploitation and were considered to be backward and uncivilized.

Civilization was brought to areas through exploitation and drainage of the land, as in the case

of the Fenland in eastern England, where large-scale drainage began in the post-medieval

period (Darby 1973; cf. Rogers 2007). Comparisons with the Fenland have also been made

with contemporary overseas land reclamation in the colonies (Evans 1997, 117). Through

land clearance, drainage and exploitation, Western concepts of ‘landscape’ were implanted

onto other parts of the world. Gibbon’s awareness of this is reflected in works held in his

library including William Douglass’ A Summary of the first planting, progressive

improvements and present state of the British settlements in America published in 1760;

America “must preserve the manners of Europe” (Gibbon vol. III, 514).

Mommsen also suggests that most of the provincial land possessed by the Romans was

governed in the same way as “English possessions in the earliest time in America” (1866,

214). Similar attitudes are clearly reflected in Gibbon’s writings on the Roman colonization in

the West and influenced his descriptions of pre-Roman settlement and his negative view

towards wetlands and woods. Gibbon’s writing, amongst other influences, came to play a

significant part in the formulation of ideas and direction of approaches in the study of late

8

Iron Age oppida and contemporary settlement, via the perspective emphasizing economic

priorities and hierarchy/dominance in considerations of their location, organization and

function. Moving away from the narrative of colonial conquest and economic prioritization, it

is necessary to be reminded that in the Iron Age people would have had an entirely different

concept of time and being which impacts on studies of their inhabited world.

RETHINKING LATE IRON AGE LANDSCAPE AND SETTLEMENT

This second part of the paper provides an example of how we might interpret settlement and

landscape in the late Iron Age differently, looking especially at oppida. The term ‘oppidum’

used in the context of describing settlement types in the late Iron Age of Britain and the near

Continent originates from Caesar’s De Bello Gallico. He uses the term, Latin for ‘town’, to

describe the sites that he saw in Gaul which might imply that he needed a word that would be

understood by his readership in Rome. It is unlikely, however, that we have anything more

than only a very basic understanding of these sites and, in turn, the process of urbanization in

the Roman period. This section will reconsider late pre-Roman sites in France, and some in

Britain, looking especially at their relationship with Roman urbanism. It might appear here

that Roman towns are being used to interpret Iron Age settlement patterns (as was perhaps the

case in Gibbon’s work) but the opposite is intended. Analysis shows that some Roman towns

are preceded by oppida whilst others have evidence for pre-Roman activity without the

monumental earthworks of oppida. Looking at the meanings attached to their landscape

setting, however, this analysis will suggest that the significance of many of these places need

not have related to their monumentality or apparent proto-urban appearance.

Where oppida have been identified in France and elsewhere in Continental Europe, through

monumental earthworks, they have predominantly been interpreted in defensive terms, acting

as refuges and storage places, and in economic terms being centres of production and

exchange. Roymans (1990, 202), for example, wrote that “the size of the entire settlement

corresponds primarily to its defensive function, while the size of its permanently inhabited

area corresponds to its economic one”. Fichtl devotes a large part of La Ville Celtique (2005)

to the defences of oppida, and the evidence for craft production within them, whilst Collis’

(1975) Defended Sites of the Late La Tène, as the title suggests, puts an emphasis on the

defensive nature and function of sites. Woolf (1993, 223), moreover, draws attention to the

fact that the title of many of the works on oppida explicitly state the interpretation that they

are urban,6 making the assumption that we understand both the nature of sites in the pre-

6 These works include Audouze and Büchsenschütz (1991), Collis (1984), Cunliffe and Rowley (1976), Wells (1984), and also Fichtl (2005). Woolf (1998) did look more at the social aspects of oppida.

9

conquest period and the nature of Roman urbanism. This attitude has had the effect of

simplifying our understanding of these sites in terms of their location, activity within them

and the way in which they were experienced. Even the ‘defences’ of many oppida seem to

have been far too impracticable for defence to have been their only or most important purpose

and other functions are also likely (Haselgrove 2007, 511). They also affirmed identities and

were significant aspects of the experience of the ‘landscape’.

Useful examples of the reanalysis of our understanding of sites and their settings include

Aitchison’s (1993) examination of the Dorsey in County Armagh, Northern Ireland, an

earthwork enclosure lying near the Dorsey River.7 The earthworks enclose a large area of

marshland which contains very little evidence for occupation or activity. Traditionally seen as

a military installation it may instead have functioned as some kind of sacred enclosure with

the marshland being deliberately defined in this way (Aitchison 1993), this landscape feature

forming an essential component of the site. Haselgrove and Millett’s (1997) reanalysis of pre-

Roman Verulamium, next to the River Ver, also puts an emphasis on the setting of the

earthworks here. They suggest that the enclosure in the valley floor with remains of

metalworking was a deliberate decision to locate the industrial activity, and probably also

political and religious events, here because of the religious associations of both the marshy

area and the metalworking itself. Actions were taken and experiences encountered that did not

relate to what would be regarded as rational and practical today.

Willis’ (1997) study of the settlement pattern of the East Midlands in Britain reminds us that

the monumentality of settlements may have been less important that their negotiation with

‘natural’ features in the landscape. Here, watery features were often integral parts of the

settlements which did not have large earthworks. Roman towns were placed within meaning-

laden landscapes and became a part of them (cf. Creighton 2006; Rogers 2008). Towns were

also landscape events where phenomenological aspects were just as important and meaningful

as they were with pre-Roman settlement (Willis 2007a). Town walls for instance, as with

oppida earthworks, undoubtedly had values beyond their functional use (cf. Woolf 2006).

Some work on the Continental oppida is now beginning to explore them as meaningful space

including evidence of sanctuaries (e.g., Fichtl et al. 2000; Kaenel 2006). The Titelberg

(Luxembourg), for example, seems to have had zones of different activities with a sanctuary

site in the centre, with its prominent position suggesting that it was an important part of the

site (Metzler et al. 2006). Other examples with sanctuaries as apparent foci are Martberg,

Wallendorf, and Mont Beuvray. Evidence for sanctuaries with pre-Roman origins near to

Roman towns might suggest that there were a number of closely related foci in the pre-Roman

7 Also see Rogers 2008.

10

landscape, one of which was then perhaps taken by the town. At Bayeux, settlement at the site

of the Roman town seems to have begun in the Augustan period but excavations have

discovered traces of a pre-Roman religious sanctuary on the site of the Saint-Vigor hill to the

east of the town (Bedon 1999, 178) indicating that this landscape was already in use and

ritualized, within which the Roman town will have become part. Sanctuaries need not be the

only way in which the religious meaning of sites is reflected; religious symbolism will also

have been attached to natural features and wider landscape settings. Roymans’ (2004; 2007)

recent work on the site of Kassel/Lith and its watery setting, for instance, demonstrates a

move away from a predominant defensive and economic interpretation of sites more common

in his earlier work (1990).

In France, Roman towns were not only associated with ‘oppida’ but also placed in locations

with traces apparently of other forms of settlement. In some cases even towns that seem to

have been founded in unoccupied landscapes may actually have been placed on already

significantly defined land.8 It is necessary to consider the symbolism that the wider landscape

might have held. ‘Natural’ places could be as significant and meaningful within landscapes as

‘man-made’ features, while culture and nature may not always have been consciously

distinguished (e.g., Bradley 2000; Insoll 2007): “natural places have an archaeology because

they acquired a significance in the minds of people in the past” (Bradley 2000, 35).

One factor concerning the location of many of the sites, and the meanings attached to them, is

their watery contexts. Gibbon, had he made a study of this, is likely to have interpreted it

through the necessity for landscape improvement with drainage and land reclamation. It is

unlikely, however, that given the significance, especially in religious belief, that seems to

have been attached to water and watery contexts in prehistoric Western Europe (Derks 1998;

Green 1986, 166; Webster 1995, 449–51) that these sites would not have been considered

solely in rational and practical terms. The watery context seems to have been important not

only for oppida but also other types of site (Figure 1). The significance of water within the

landscape, and to settlement in pre-Roman France, is indicated by the large number of

references to rivers and water in Gaul in the classical sources. Diodorus Siculus describes the

rivers (Diod. Sic. V.25.3–4) whilst Caesar refers to their defensive benefits around

settlements.

Bourges (Avaricum), for example, had great “natural strength, for it was surrounded by river

and marsh” (B Gall. VII.15) whilst around Paris (Lutetia) there was a “continuous expanse of

marsh which flowed into the Seine and formed a significant obstruction over the whole area”

(VII.57) and at Besançon (Vesontio; Walter and Barçon 2004) the “river Doubs practically

8 Cf. Brooks 2006.

11

surrounds the entire town” (B Gall. I.38). Strabo emphasizes Lyon’s (Lugdunum) location

where “the Avar and the Rhodanus mingle with one another” (IV.1.11), suggesting it was

significant. Eleven towns in Gallia Belgica, thirteen in Gallia Lugdunensis, and ten in Gallia

Aquitania were located at the confluence of rivers whilst some were within the meanders of

rivers, such as Cahors, Périgueux, Soissons, and Besançon and some in marshy locations

including Dax, Sens, and Amiens. There were, of course, also elevated sites that became

Roman civitas-capitals such as Langres and Metz in Gallia Belgica but, even here, rivers

played an important role in the landscape and the ramparts at Metz enclosed areas of

marshland; hilltop sites are likely to have had significances beyond their defensive function.

At a number of sites that became Roman towns there is evidence for pre-Roman oppida but in

no cases are their nature, function and interaction with the wider landscape fully understood.

The Roman town at Rheims, for example, in Gallia Belgica was located on the site of one of

the largest oppida in France, Durocortorum, which had two ramparts with the inner one

enclosing 90 hectares and the outer enclosing 550 hectares (Berthelot et al. 1994; Bedon

1999; Neiss 2005). It was situated between the River Vesle and two streams that have now

disappeared and whilst some structures and streets have been identified inside the oppidum, it

seems doubtful that the whole of the enclosed area would have been occupied. This

‘landscape’ element, enclosing large open spaces, and its watery context, suggests that there

was more meaning attached to it than a settlement site in modern terms.

The town at Soissons in Gallia Belgica was preceded by the oppidum of Pommiers, a large

promontory cut off by a massive bank and ditch enclosing around 40 hectares, which in turn

followed the oppidum at Villeneuve-St.-Germain, a low-lying site which was enclosed by a

meander of the River Aisne and associated with floodable areas (Brun et al. 2000; Haselgrove

1996, 149). It now seems, however, that none of the sites were wholly abandoned. Coins,

brooches and other material indicate that the chronologies of Pommiers and Villeneuve-St.-

Germain overlap; both sites, at least periodically, were being used at the same time

(Haselgrove 1996, 151). Pommiers also remained in some kind of use during the Roman

period (Knight 2001, 77–8). It may be possible to consider all three locations as places within

a ‘complex’ of sites that retained long-term significance and meaning. They may also have

been used as periodic religious and meeting places before the decision was made to make a

more permanent move (Haselgrove 2007, 509). Settlement and landscape use was far more

complex than linear studies of Roman urbanization often allow.

Bibracte (Mont Beuvray) is another important example. This oppidum was situated on the

promontory of the Autonois hills overlooking the flood plain of the River Arroux where there

was waterlogged ground and the possibility of flooding (Rebourg 1998). It is in this

floodplain that there was a sanctuary complex including the ‘Temple of Janus’ which

12

probably had pre-Roman origins. A theatre and another temple have also been identified here

by aerial photography and it seems to be surrounded by an enclosure ditch (Frezouls 1997,

158; Rebourg 1998, 158–60). The oppidum continued to attract activity in the Roman period

despite the Roman town being located at Autun 27 kilometres away and suggests that this site

remained part of a wider meaning-laden landscape.

Collis et al. (2000) have also emphasized the need to acknowledge the importance of the large

non-defended sites as part of the settlement pattern. Amiens is perhaps a good example of the

problematic use of the term oppidum for only large earthwork sites since Caesar refers to an

oppidum here (B Gall. V.24; V.47; V.53), and that the council of Gauls met here (V.24), but

so far no earthworks have been identified. If Caesar can be taken at his word it would suggest

that this place was special without much monumentality. The site, including the later Roman

town, was located in a marshy area and at a crossing point and confluence of the Rivers

Somme and Selle. It seems that this was a special site of meeting, interaction, ceremony and

ritual that was fully integrated into its landscape setting. Another example is Sens in Gallia

Lugdunensis which was located in a low-lying marshy area next to the River Yonne (Perrugot

1996) and although there does not seem to have been an ‘oppidum’ here there are traces of

pre-conquest activity. Sites require further study in terms of pre-Roman activity and the

meanings attached to them.

In Britain there is a complexity in the late pre-Roman settlement record that is often

overlooked because of the preoccupation with the practical functions and nature of oppida.

This is examined in more detail in Rogers (2008) which sought to explore the oppida and

other forms of settlement in terms of their significance as places, their landscape setting and

the way in which Roman urbanism interacted with them. In Britain, oppida tend to be divided

into ‘enclosed oppida’ like the Continental sites and ‘territorial oppida’ which appear to

embrace much larger areas without continuous earthworks (Haselgrove 1999, 121;

Haselgrove and Moore 2007, 6). In both cases, the earthworks have generally been interpreted

in defensive terms and their watery contexts were practical elements of their location.

Sites examined in this way include Verulamium (St Albans) (cf. Niblett 2005; Thompson

2005) and Camulodunum (Colchester) (Crummy 1997; Hawkes and Hull 1947; cf. Willis

2007b, 121–2)9 with their earthworks spanning large areas. These were associated with

watery settings and foci of activity including metalworking and coin production. The religious

significance and magical nature attached to metalworking and the minting of coinage in

prehistory (Budd and Taylor 1995; Haselgrove and Wigg-Wolf 2005, 12; Hingley 1997)

9 The names Verlamion and Camulodunon are often used to distinguish the Iron Age period settlements from the Roman towns, and are based largely on Iron Age coinage, although the exact names and spellings are still uncertain (Potter 2002, 21).

13

would also have contributed to the special nature of these sites, integrated with the

religiously-imbued watery setting as studied by Haselgrove and Millett (1997, 284).

Potentially, however, key late Iron Age sites lacked earthworks (cf. Rogers 2008): at Lincoln,

which lay near the River Witham, evidence for pre-fortress and colonia activity includes

traces of structures on what would have been an island within a large natural body of water

there known as the Brayford Pool (Figure 2; Darling and Jones 1988) and there are

suggestions that there may also have been activity beneath the colonia itself (Jones and

Stocker 2003, 28–30), much of which would have been lost through its construction. A triple

linear bank and ditch system to the north of the later town (ibid, 30–1) appears to lead to the

site and was possibly even designed to direct movement to this religiously-imbued location.

Other apparent concentrations of pre-Roman activity in watery and marshy areas include at

Cirencester (Moore 2006; Reece 2003), Exeter (Fox 1952), and Leicester (Cooper and

Buckley 2003) and suggest that these sites may not have been considered in inferior terms to

oppida and may have played an important part in the creation of the settlement pattern in the

Roman period. The location of the fortress and subsequent colonia at Lincoln, for instance

could even have been a deliberate attempt to control a sacred place here which was

considered a threat (cf. Brooks 2006; Millett 1990).

Roman London is another important example (see Rogers 2008), it being located next to the

Thames and divided by the Walbrook stream, both of which appear to have been used for the

religious deposition of objects in prehistory (Bradley and Gordon 1988; Merrifield 1983, 9).

Evidence for Iron Age occupation in the area is still slight although traces have been

identified on islands within the Thames at Southwark (Beard and Cowan 1988). Holder and

Jamieson’s study (2003) has demonstrated the extent of loss of early levels in London due to

truncation. Regardless of the extent of Iron Age occupation here, however, the site could well

have been an important meeting and religious place at this time.

It is important, then, for oppida and contemporary sites to be studied in more theoretically

rigorous ways, acknowledging the differences in which the landscape would have been

conceptualized and the meanings attached to certain places. In a number of cases this

significance may have survived through memories and myths making ‘place histories’. These

‘commemorative places’ (Van Dyke and Alcock 2003, 5) would have gathered people,

experiences, histories, and thoughts (Casey 1996, 14) and remained culturally charged

(Creighton 2006) despite new elements being brought to the sites through the foundation of

Roman towns.

14

CONCLUSIONS

This paper took the form of two separate but interlinked parts. The first section used Edward

Gibbon’s The Decline and Fall of the Roman Empire to highlight the complex nature in

which the research tradition of pre-Roman settlement has been formed. Gibbon’s description

of watery and wooded areas relating to pre-conquest settlement in negative terms was written

in a period where ‘economic improvement’ was a dominant social concept, changing

understandings of landscape forever, and where the colonial exploitation of land and peoples

was at its height. His attitude towards urbanism and landscape in his own day influenced his

understanding of the Roman period as well as colouring his interpretation of the classical

sources. What makes Gibbon’s work in its social context especially important is that it has

had such a prominent position in academic study and has gone on to influence attitudes and

approaches towards the past into the present day. The same social context has led to the

situation whereby what archaeologists tend to see as ‘facts’ and ‘practical considerations’

concerning place and landscape are mostly modern, Western-derived assumptions (Johnson

2007, 129).

Authors such as Theodor Mommsen and Camille Jullian were also influential in the research

tradition but there are also influences from Gibbon in their work. The context in which these

influential works were written has had an impact on our study of Iron Age and Roman

settlement patterns including our understanding of late Iron Age oppida and their relationship

with less monumental contemporary sites. The second part of the paper discussed an example

of how with detailed archaeological data and theoretical analysis, we can explore ways in

which settlement, landscape, and place might be approached in different ways. This in turn

has led to more sophisticated understandings of the process of urbanization after the Roman

conquest. Gibbon’s work is a good reflection of the way in which people thought about place

and space in the eighteenth century and how this has influenced historical and archaeological

studies since. As the second part demonstrates, however, there are more rigorous ways in

which the available settlement evidence can be studied to raise new perspectives on place and

landscape in the late Iron Age and Roman periods.

REFERENCES

Ancient Sources

Caesar (Translated by H.J. Edwards 1917) De Bello Gallico, London: Heinemann.

Diodorus Siculus (Translated by C.H. Oldfather 1933) Bibliothēkē, London: Heinemann.

Lucan (Translated by J.D. Duff 1969) Pharsalia, London: Heinemann.

15

Strabo (Translated by H.L. Jones 1917-1932) Geographia, London: Heinemann.

Modern sources

Audouze, F. and Büchsenschütz, O. (1991) Towns, Villages and Countryside of Celtic

Europe, London: Batsford.

Beard, D. and Cowan, C. (1988) ‘Excavations at 15–23 Southwark Street’, The London

Archaeologist, 5 (14), 375–81.

Bedon, R. (1999) Les villes des trois Gaules: de César à Néron: dans leur contexte

historique, territorial et politique, Paris: Picard.

Berthelot, F., Balmelle, A. and Rollet, P. (1994) Reims fouilles archéologiques: site du

Conservatoire National de Région de Musique et de Danse, Rue Gambetta, Reims

(Marne), Reims: Société Archéologique Champenoise.

Bevan, B. (1997) ‘Bounding the Landscape: place and identity during the Yorkshire Wolds

Iron Age’, in A. Gwilt and C. Haselgrove (eds.) Reconstructing Iron Age Societies: new

approaches to the British Iron Age, Oxford: Oxbow Books, 181–91.

Bradley, R. (2000) An Archaeology of Natural Places, London and New York: Routledge.

Bradley, R. and Gordon, K. (1988) ‘Human skulls from the River Thames, their dating and

significance’, Antiquity, 62, 503–9.

Brooks, H. (2006) ‘Colchester before Colchester’, in P. Ottaway (ed.) A Victory Celebration:

Papers on the Archaeology of Colchester and Late Iron Age-Roman Britain presented

to Philip Crummy, Colchester: Colchester Archaeological Trust, 5–10.

Brownley, M.W. (1976) ‘Gibbon: The Formation of Mind and Character’, Daedalus, 105,

13–25.

Brück, J. (2007) ‘Landscape politics and colonial identities: Sir Richard Colt Hoare’s tour of

Ireland, 1806’, Journal of Social Archaeology, 7, 224–49.

Brun, P., Chartier, M. and Pion, P. (2000) ‘Le processus d’urbanisation dans la vallée de

l’Aisne’, in V. Guichard, S. Sievers and O.-H. Urban (eds.) Les processus

d'urbanisation à l'âge du Fer (Collection Bibracte 4), Glux-en-Glenne: Centre

archéologique européen, 83–96.

Budd, P. and Taylor, T. (1995) ‘The faerie smith meets the bronze industry: magic versus

science in the interpretation of prehistoric metal making’, World Archaeology, 27, 133–

43.

Burrow, J.W. (1985) Gibbon. Oxford: Oxford University Press.

16

Casey, E.S. (1996) ‘How to Get from Space to Place in a Fairly Short Stretch of Time:

Phenomenological Prolegomena’, in S. Feld and K.H. Basso (eds.) Senses of Place,

Santa Fe: School of American Research Press, 13–52.

Collis, J. (1975) Defended sites of the late La Tène in central and western Europe, British

Archaeological Reports International Series 2, Oxford.

Collis, J., Krausz, S. and Guichard, V. (2000) ‘Les villages ouverts en Gaule centrale aux IIe

et Ier s. av. J.-C.’, in V. Guichard, S. Sievers and O.-H. Urban (eds.) Les processus

d'urbanisation à l'âge du Fer (Collection Bibracte 4), Glux-en-Glenne: Centre

archéologique européen, 73–82.

Cooper, N.J. and Buckley, R. (2003) ‘New Light on Roman Leicester (Ratae

Corieltauvorum)’, in P. Wilson (ed.) The Archaeology of Roman Towns: Studies in

honour of John S. Wacher, Oxford: Oxbow Books, 31–43.

Cosgrove, D.E. (1984) Social Formation and Symbolic Landscape, London and Sydney:

Croom Helm.

Creighton, J. (2006) Britannia: The Creation of a Roman Province, London and New York:

Routledge.

Crummy, P. (1997) City of Victory: the story of Colchester – Britain’s first Roman town,

Colchester: Colchester Archaeological Trust.

Cunliffe, B. and Rowley, T. (eds.) (1976) Oppida: the beginnings of urbanisation in

barbarian Europe: papers presented to a conference at Oxford, October 1975, British

Archaeological Reports International Series 11, Oxford.

Darby, H.C. (1973) ‘The Age of the Improver: 1600–1800’, in H.C. Darby (ed.) A New

Historical Geography of England, Cambridge: Cambridge University Press, 302–88

Darling, M.J. and Jones, M.J. (1988) ‘Early Settlement at Lincoln’, Britannia, 19, 1–57.

Derks, T. (1998) Gods, temples and ritual practices. The transformation of religious ideas

and values in Roman Gaul, Amsterdam: Amsterdam University Press.

Douglass, W. (1760) A Summary of the first planting, progressive improvements and present

state of the British settlements in America, London.

Dyson, S.L. (2006) In Pursuit of Ancient Pasts: A History of Classical Archaeology in the

Nineteenth and Twentieth Centuries, New Haven and London: Yale University Press.

Evans, C. (1997) ‘Sentimental Prehistories: The Construction of the Fenland Past’, Journal of

European Archaeology, 5, 105–36.

17

Fichtl, S. (2005) (2nd edition) La ville celtique: les oppida de 150 av. J.-C. à 15 ap. J.-C.,

Paris: éditions errance.

Fichtl, S., Metzler, J. and Sievers, S. (2000) ‘Le rôle des sanctuaires dans le processus

d’urbanisation’, in V. Guichard, S. Sievers and O.-H. Urban (eds.) Les processus

d'urbanisation à l'âge du Fer (Collection Bibracte 4), Glux-en-Glenne: Centre

archéologique européen, 179–86.

Fox, A. (1952) Roman Exeter (Isca Dumnoniorum) Excavations in the War Damaged Area

1945–1947, Manchester: Manchester University Press.

Frezouls, E. (1997) Les Villes antiques de la France. 3, Lyonnaise 1: Autun, Chartres,

Nevers, Paris: de Boccard.

Ghosh, P. (1997) ‘The Conception of Gibbon’s History’, in R. McKitterick and R. Quinault

(eds.) Edward Gibbon and Empire, Cambridge: Cambridge University Press, 271–316.

Gibbon, E. (1966) Memoirs of My Life (ed. G.A. Bonnard), London: Nelson.

Gibbon, E. (1994 [1776-1788]) (edited by D. Womersley), History of the Decline and Fall of

the Roman Empire, London: Penguin.

Goudineau, C. (1998) Regard sur la Gaule, Paris: éditions errance.

Green, M.J. (1986) The Gods of the Celts, Gloucester: Alan Sutton.

Haselgrove, C. (1996) ‘Roman Impact on rural settlement and society in southern Picardy’, in

N. Roymans (ed.) From the sword to the plough: three studies on the earliest

romanisation of Northern Gaul, Amsterdam: Amsterdam University Press, 127–87.

Haselgrove, C. (1999) ‘The Iron Age’, in J. Hunter and I. Ralston (eds.) The Archaeology of

Britain, London and New York: Routledge, 113–34.

Haselgrove, C. (2007) ‘The age of enclosure: Later Iron Age settlement and society in

northern France’, in C. Haselgrove and T. Moore (eds.) The Later Iron Age in Britain

and Beyond, Oxford: Oxbow Books, 492–522.

Haselgrove, C. and Millett, M. (1997) ‘Verlamion reconsidered’, in A. Gwilt and C.

Haselgrove (eds.) Reconstructing Iron Age Societies: new approaches to the British

Iron Age, Oxford: Oxbow Books, 282–96.

Haselgrove, C. and Moore, T. (2007) ‘New narratives of the Later Iron Age’, in C.

Haselgrove and T. Moore (eds.) The Later Iron Age in Britain and Beyond, Oxford:

Oxbow Books, 1–15.

18

Haselgrove, C. and Wigg-Wolf, D. (2005) ‘Introduction’, in C. Haselgrove and D. Wigg-

Wolf (eds.) Iron Age coinage and ritual practices, Mainz am Rhein: Van Zabern, 9–22.

Hawkes, C.F.C. and Hull, M.R. (1947) Camulodunum. First Report on the Excavation at

Colchester 1930–1939, Reports of the Research Committee of the Society of the

Antiquaries of London Number 14, London: The Society of the Antiquaries of London.

Hingley, R. (2000) Roman Officers and English Gentlemen, London and New York:

Routledge.

Hingley, R. (2005) Globalizing Roman Culture: unity, diversity and empire, London and New

York: Routledge.

Hingley, R. (2006) ‘Projecting empire: the mapping of Roman Britain’ Journal of Social

Archaeology, 6, 328–53.

Hingley, R. and Rogers, A. (forthcoming) ‘The use of decline and fall as an imperial motif

from the late eighteenth to the early twentieth centuries and the impact of this idea on

early archaeological endeavours’, in M. Bradley and E. Reisz (eds.) Hegemony and

Cornucopia: Classical Scholarship and the Ideology of Imperialism, Oxford: Oxford

University Press.

Holder, N. and Jamieson, D. (2003) ‘The Prehistory of the City of London: Myths and

Methodologies’, The Archaeological Journal, 160, 23–43.

Horsley, J. (1974 [1732]) Britannia Romana: or, the Roman antiquities of Britain, Newcastle

upon Tyne: Graham.

Insoll, T. (2007) ‘‘Natural’ or ‘Human’ Spaces? Tallensi Sacred Groves and Shrines and their

Potential Implications for Aspects of Northern European Prehistory and

Phenomenological Interpretation’, Norwegian Archaeological Review, 40, 138–58.

Johnson, M. (2007) Ideas of Landscape, Oxford: Blackwell Publishing.

Jones, M.J. and Stocker, D. (2003) ‘Settlement in the Lincoln area in the Prehistoric Era:

Archaeological Account’, in D. Stocker (ed.) The City by the Pool: Assessing the

Archaeology of the City of Lincoln, Oxford: Oxbow Books, 19–33.

Jullian, C. (1907–1921) Histoire de la Gaule, Paris: Librairie Hachette.

Kaenel, G. (2006) ‘Agglomérations et oppida de la fin de l’âge du fer: une vision

synthétique’, in C. Haselgrove (ed.) Les mutations de la fin de l’âge du Fer (Collection

Bibracte 12/4), Glux-en-Glenne: Centre archéologique européen, 17–39.

King, A. (2001) ‘Vercingetorix, Asterix and the Gauls: Gallic symbols in French politics and

culture’, in R. Hingley (ed.) Images of Rome: Perceptions of ancient Rome in Europe

19

and the United States in the modern age, Journal of Roman Archaeology

Supplementary Series 44, Portsmouth, Rhode Island, 113–25.

Lavan, L. ed. (2001) Recent Research in Late Antique Urbanism, Journal of Roman

Archaeology Supplementary Series 42, Portsmouth, Rhode Island.

Lemaire, T. (1997) ‘Archaeology between the Invention and Destruction of the Landscape’,

Archaeological Dialogues, 4, 5–21.

Lewis, W.S. (ed.) (1955) Horace Walpole’s Correspondence, London: Oxford University

Press.

Low, D.M. (1937) Edward Gibbon, 1737–1794, London: Chatto and Windus.

McKitterick, R. (1997) ‘Edward Gibbon and the early Middle ages in eighteenth-century

Europe’, in R. McKitterick and R. Quinault (eds.) Edward Gibbon and Empire,

Cambridge: Cambridge University Press, 162–89.

McKitterick, R. and Quinault, R. eds. (1997) Edward Gibbon and Empire, Cambridge:

Cambridge University Press.

Merrifield, R. (1983) London: City of the Romans, London: Batsford.

Metzler, J., Méniel, P. and Gaeng, C. (2006) ‘Oppida et espaces publics’, in C. Haselgrove

(ed.) Les mutations de la fin de l’âge du Fer (Collection Bibracte 12/4), Glux-en-

Glenne: Centre archéologique européen, 201–24.

Millett, M. (1990) The Romanization of Britain: an essay in archaeological interpretation,

Cambridge: Cambridge University Press.

Mommsen, T. (1866 [first published in German in 1856]) The History of Rome Volume III

(translated by W.P. Dickson and published as Volume IV – Part 1), London: Richard

Bentley.

Moore, T. (2006) Iron Age Societies in the Severn-Cotswolds: Developing narratives of social

and landscape change, British Archaeological Reports British Series 421, Oxford:

Archaeopress.

Neiss, R. (1984) ‘La Structure Urbaine de Reims Antique et son Évolution du Ier au IIIe Siècle

ap. J.C.’, Revue Archéologique de Picardie, 3–4, 171–91.

Neiss, R. (2005) Durocortorum. Etat des recherches après vingt années d’archéologie

préventive à Reims, Revue archéologique.

20

Niblett, R. (2005) ‘Roman Verulamium’, in R. Niblett and I. Thompson Alban’s Buried

Towns: An Assessment of St Albans’ Archaeology up to AD 1600, Oxford: Oxbow

Books, 41–165.

Perrugot, D. (1996) ‘Sens: Origine, développement et repli du Ier siècle au début du Vème

siécle. Aux origines de la ville antique’, in R. Bedon (ed.) Les Villes de la Gaule

Lyonnaise, Caesarodunum 30, PULIM: Limoges, 263–78.

Pocock, J.G.A. (1999) Barbarism and Religion Vol. 1: The Enlightenment of Edward Gibbon,

1737–1764, Cambridge: Cambridge University Press.

Porter, R. (1988) Edward Gibbon: Making History, London: Weidenfeld and Nicholson.

Rebourg, A. (1998) ‘L'urbanisme d'Augustodunum (Autun, Saône-et-Loire)’, Gallia, 55,

141–236.

Reece, R. (2003) ‘The siting of Roman Corinium’, Britannia, 34, 276–80.

Rogers, A. (2005) ‘Metalworking and Late Roman Power: a study of towns in later Roman

Britain’, in J. Bruhn, B. Croxford and D. Grigoropoulos (eds.) TRAC 2004:

Proceedings of the Fourteenth Annual Theoretical Roman Archaeology Conference

Durham 2004, Oxford: Oxbow Books, 27–38.

Rogers, A. (2007) ‘Beyond the Economic in the Roman Fenland: Reconsidering land, water,

hoards and religion’, in A. Fleming and R. Hingley (eds.) The Making of the British

Landscape: Fifty Years After Hoskins: Prehistoric and Roman periods, Macclesfield:

Windgather Press, 113–30.

Rogers, A. (2008) ‘Religious place and its interaction with urbanization in the Roman era’,

Journal of Social Archaeology, 8(i), 37–62.

Roymans, N. (1990) Tribal Societies in Northern Gaul: An Anthropological Perspective,

Amsterdam: University of Amsterdam.

Roymans, N. (2004) Ethnic Identity and Imperial Power: The Batavians in the Early Roman

Empire, Amsterdam: Amsterdam University Press.

Roymans, N. (2007) ‘Understanding social change in the Late Iron Age Lower Rhine region’,

in C. Haselgrove and T. Moore (eds.) The Later Iron Age in Britain and Beyond,

Oxford: Oxbow Books, 478– 91.

Schnapp, A. (1996) The discovery of the past: the origins of archaeology, London: British

Museum Press.

21

Thompson, I. (2005) ‘Verlamion in the late pre-Roman Iron Age’, in R. Niblett and I.

Thompson Alban’s Buried Towns: An Assessment of St Albans’ Archaeology up to AD

1600, Oxford: Oxbow Books, 23–40.

Todd, M. (2004) ‘The Rediscovery of Roman Britain’, in M. Todd (ed.) A Companion to

Roman Britain, Oxford: Blackwell Publishing, 443–59.

Van Dyke, R.M. and Alcock, S.E. (2003) ‘Archaeologies of Memory: An Introduction’, in

R.M. Van Dyke and S.E. Alcock (eds.) Archaeologies of Memory, Oxford: Blackwell

Publishing, 1–13.

Vance, N. (1997) The Victorians and Ancient Rome, Oxford: Blackwell Publishing.

Webster, J. (1995) ‘Sanctuaries and Sacred Places’, in M. Green (ed.) The Celtic World,

London and New York, Routledge, 441– 64.

Wickham, C. (2005) Framing the Early Middle Ages: Europe and the Mediterranean 400–

800, Oxford: Oxford University Press.

Willis, S. (1997) ‘Settlement, materiality and landscape in the Iron Age of the East Midlands:

evidence, interpretation and wider resonance’, in A. Gwilt and C. Haselgrove (eds.)

Reconstructing Iron Age Societies: new approaches to the British Iron Age, Oxford:

Oxbow Books, 205–15.

Willis, S. (2007a) ‘Roman Towns, Roman Landscapes: the cultural terrain of town and

country in the Roman period’, in A. Fleming and R. Hingley (eds.) The Making of the

British Landscape: Fifty Years After Hoskins: Prehistoric and Roman periods,

Macclesfield: Windgather Press, 143–64.

Willis, S. (2007b) ‘Sea, coast, estuary, land and culture in Britain during the Iron Age’, in C.

Haselgrove and T. Moore (eds.) The Later Iron Age in Britain and Beyond, Oxford:

Oxbow Books, 107–29.

Womersley, D. (1994) ‘Introduction’, in E. Gibbon The History of the Decline and Fall of the

Roman Empire, London: Penguin Books, xi–cvi.

Woolf, G. (1993) ‘Rethinking the Oppida’, Oxford Journal of Archaeology, 12, 223–34.

Woolf, G. (1998) Becoming Roman: the Origins of Provisional Civilisation in Gaul,

Cambridge: Cambridge University Press.

Woolf, G. (2006) ‘The end of “the end of the Iron Age”?’, in C. Haselgrove (ed.) Les

mutations de la fin de l’âge du Fer (Collection Bibracte 12/4), Glux-en-Glenne: Centre

archéologique européen, 267–76.