Figure 1: Herefordshire Beacon, a Hillfort in the Malvern Hills, Sept 2012. Source: Author.
The interpretation of hillforts, has, for the most part, been hampered by the lack of sustained and wide-ranging excavation (Hamilton & Manley 2001 :34). The number of sites that have been subjected to modern, systematic excavation is very small; Niall Sharples’ recent re-excavation of Maiden Castle, Dorset (Sharples 1991); and Barry Cunliffe’s long-running excavations at Danebury in Hampshire being among the most well-known and discussed examples. Cunliffe’s work at Danebury is the baseline for most other modern accounts (Hill 1995 :52), and informs his ‘Iron Age Communities in Britain’ (Cunliffe 2005). This disparity between the wealth of data from well-studied hillforts and the paucity of that from other sites leads to problems making comparisons to other sites where little information is available (Payne 2006 :9).
Another problem in the study and interpretation of hillforts is the lack of consensus about where the taxonomic distinction is to be made as regards the distinguishing of a hillfort from an enclosed hilltop settlement. Avery attempts to exclude from the definition any site where there is demonstrably inadequate defensive capabilities: adhering to the ‘fort’ aspect of the word (Avery 1976 :4). However, it shall be shown that the supposed defensive aspect of hillforts is being questioned (Bowden & McOmish 1997 cited in Sharples 1991:259), making a distinction on this basis problematic. The question of ‘what is a hillfort’ is one that will be returned to, as it will be shown that accounts vary, both on a regional basis, and with different interpretations of the same site.
Hillforts are not an evenly-distributed phenomenon, as Figure 2 shows for southern Britain, but sufficiently numerous that only a small sample of those studied can be discussed here. Preference has been given to more recent material, as earlier texts are based on a culture-historical theoretical framework that has been largely discredited (Hamilton & Manley 2001 :8). However, the theoretical framework upon which the various interpretations have been founded, is of importance in understanding how the various interpretations of hillforts came to be made (Lang 2009 :15), so although few sites will be considered here, the range of interpretations offers opportunity for evaluating the metanarratives that inform these interpretations.
Figure 2: Distribution of hillforts in southern Britain (source: Ordnance Survey, Map of Southern Britain in the Iron Age from Cunliffe 2005 Fig 15.1)
Danebury was excavated by Sir Barry Cunliffe and his team, with the intention of investigating not only the hillfort itself, but the surrounding ‘territory’ (Cunliffe 1983 :21). The findings of this project led to one of the first syntheses of Iron Age society (Clay 2001:1).
Based on evidence for large amounts of grain storage, and evidence for metalworking and imported pots, Cunliffe interpreted Danebury as being evidence for a hierarchical society, with hillforts embodying the idea of a societal élite controlling and demanding labour of those of subordinate settlements (Cunliffe 2005 :347). It was argued that Danebury represented a ‘redistribution’ centre and that this was managed by a ‘chieftain’, an interpretation that is no longer considered consistent with the evidence (Bradley 2005 :15). Little difference has been found between assemblages at Danebury and those found on contemporary non-hillfort settlement, and therefore more work is needed to establish any differences (Hill 1996 :97-101).
Sharples formed a different view of hillforts in Wessex, based on his recent excavations at Maiden Castle in Dorset (Sharples 1991); one that focussed more on the community, rather than the individual power Cunliffe concentrated in the idea of a chieftain. He argued that hillforts represented individual communities competing with their neighbours for resources, based on the idea that the early hillforts tended to be on ecotones with access to a mix of different resources. Over time, Maiden Castle out-competed the neighbouring hillforts, and the occupants of these shifted to unenclosed settlement (Sharples 1991 :264).
The construction of the banks and ditches is explained by display of the status of the group, and the elaboration and maintenance of them as a focus for the community, perhaps binding people into a cycle of dependency on the grain that the community collected (Sharples 1991:260).
Figure 3: The elaborate western entrance of Maiden Castle. April 2010. Source:Author.
In the process of assessing the applicability of existing models to the hillforts in the south-east of Britain, Hamilton and Manley visited each site to gain a more experiential knowledge of the site in its landscape, arguing that landscape and topography are immediately assessable, whereas detailed excavation is not achievable (Hamilton & Manley 2001 :34). They found that the early hillforts were about looking at, and physical access to, varied landscapes. Non-hillfort settlement seemed to be plentiful but generally where hillforts are not found: the hillforts are ‘locally peripheral’. Later, non-hillfort settlement declines and hillforts are constructed in different landscape positions and become more of a symbolic centre, uniting communities by the creation of a monumental representation of society (Hamilton & Manley 2001 :31). In agreement with Sharples (1991 :260), they then see Late Iron Age hillfort use as being about the rise of hierarchical structures: communities fragmenting into smaller, individualist groups, perhaps as a result of exploiting of local resources and specialised industries dependent on them (Hamilton & Manley 2001 :31).
In Wales, hillforts are most commonly found near good farmland near river valleys of the Marches, and near the coasts (Savory 1976 :237).
Savory had attempted to relate the named Iron Age tribes of Silures, Demetae, Ordovices and Deceangli/Cornovii to differences in the Late Bronze Age material culture but came to the conclusion that the variations could best be explained by networks of contacts, which varied as to whether the hillfort faced the coast, or was inland (Savory 1976 :241). This clustering of hillforts along the Welsh border has led to hillforts being considered typical of the region, despite the recent discovery that more low-lying settlement evidence has been reduced to cropmarks by later land use, demonstrating that the hillforts were surrounded by smaller settlements (Barker 2009 : 4).
Hillforts in the Marches often pre-date the Iron Age, with the Breiddin showing evidence for an open or possibly palisaded settlement before the first Iron Age ramparts were constructed at 800BC, which was a timber-framed box rampart similar to those found at Ivinghoe Beacon (Musson et al. 1991 :175-6). Lack of extensive excavation limits what can be stated with any certainty, but Musson et al. are reasonable certain that, like Danebury, the Breiddin supported a large population and shared many of the same features (Musson et al. 1991: 184).
Based on work in the Welsh Marches, Wigley (2002 :198) considered that hillforts represented the bringing together of Late Bronze Age social practices such as the consumption and deposition associated with burnt mounds, and those to do with building linears. He argued that just as the building of linears in various parts of the country in the Late Bronze Age, changed how people moved across the landscape, hillforts appeared not as a response to a crisis, but as an internal development to resolve conflict within society by modifying how people lived their lives.
Towards the Welsh coast, in North Ceredigion, Toby Driver demonstrated that the hillforts there can be shown to be about visual impact: the more visible parts of the structures are better built and sometimes include features specifically to enhance their appearance in the landscape, such as quartz (Driver 2007 :87). He gave the example of Caer Lletty Llwyd where the showy façade faces towards a pass connecting it to the hillfort at Pen Dinas. The other sides of Caer Lletty Llwyd are not elaborated and their visual impact seems to be of less importance (Driver 2007 :89). This deliberate attempt at monumentalisation demonstrates that hillforts are not just ‘hill-shaped’: the effort required to create the desired effect was often more than was simply required to enclose a space (Brown 2009 :35).
Harding (2004 :289) suggests that hillforts in northern Britain are not so numerous as in the south and may have been centres of occasional communal activity rather than the dense occupation that is found at sites in the south. In the Cheviot hills, a project to survey the hillforts there has resulted in the view that the idea of hillforts as defensive sites required closer examination and that the models developed in Wessex had little relevance to the evidence from these northern sites. What was found was that hillforts occurred in groups that appeared to have a relationship to each other, defined by the need to present a façade to the other sites (Frodsham et al. 2007). This relationship, however was suggested to be one of independent, equal communities: self-sufficient but interested in the surrounding communities (Frodsham et al. 2007 :261).
Interpretations of Iron Age Society
From his work at Danebury, Cunliffe constructed a model of Iron Age society, seemingly based on documentary sources of ‘Celtic’ peoples, both from Classical and early Medieval Irish literature, an approach roundly criticised by Hill (1996:95) but supported by Karl owing to the potential for continuity of ideas (Karl 2008:70). Cunliffe envisioned a hierarchical society, with power and status concentrated in an elite, but demonstrated and persisted by control of agricultural production and also raiding and warfare (Cunliffe 1983 :85). Finney (2005: 242) argues, however, that raiding is not necessarily indicative of an elite as societies with no permanent leader engage in raiding behaviour to obtain marriage partners, esteem, and material goods. Whilst the contemporary reports of writers such as Julius Caesar may shed light on the final days of the Iron Age, the fact that the early hillforts were constructed nearly a millennium before this point necessarily calls into question the applicability of such ideas to the Early and Middle Iron Age, and therefore distances them from the Middle Iron Age, when hillforts like Danebury see most activity. This is especially true if it is considered that the factors that bring hillforts into being initially may have only affected a few generations (Frodsham et al. 2007 :258), and the idea that a ‘big man’ needs an impressive residence (Hill 1996 :95) is inappropriate if the ‘big man’ only appears towards the very end of the Iron Age.
It is worth noting that the Danebury project was initiated against a background of the ‘New Archaeology’ (Lock 2007:342), with its emphasis on models and an economic (J Collis 2010) understanding of past societies (Hamilton & Manley 2001 :8). This approach, based on the idea of ‘central place theory’ and the analysis of settlement positioning by means of Thiesson polygons, has been attacked vociferously by other Iron Age scholars (e.g. Hill 1996) on the grounds that the evidence put forward does not support this argument (Hill 1995 :58). Later fieldwork in the Danebury area has shown that the satellite sites that Danebury was ostensibly the ‘central place’ to, were abandoned at the time interpreted as Danebury’s ‘peak’ of use (Frodsham et al. 2007 :257).
The idea of the exchange network at the heart of the ‘central place’ model is based on the idea of pottery and other goods being thought of as commodities: to be exchanged with only consideration of the material value of the goods and ignoring any social or religious connotations (Sharples 2010 :106). Moore (2007 :93) preferred to see this exchange as the more culturally-laden exchange of gifts and obligations that created and maintained relationships. Sharples (2010 :94) argued that gift-giving rather than commodity exchange is a marker of a society based on ideas of clan rather than class; a society where one cannot own land or resources as an individual, thereby ensuring that material wealth and power do not concentrate in an individual.
Another criticism of Cunliffe’s (2005) model of Iron Age society is to do with the boundedness of his views of the Iron Age. Bradley commented that the Iron Age was treated as a self-contained unit, with scant regard for the Bronze Age that preceded it, and ending abruptly with the arrival of the legions (Bradley 2007 :226). The origins of hillforts and the changes in society that they are assumed to represent must be searched for in the Bronze Age. The apparent collapse of the long-distance exchange networks that figure prominently in discussions of the Bronze Age and the deposition of large amounts of increasingly-poor bronzes are held to represent the end of a particular form of society. This depositional practice has been suggested to represent either an elite trying to hold onto power and enhancing the scarcity of a prestige material, or, could represent the rejection of the old ways by a more egalitarian society and be a form of ritual closure (Jackson 1999 :214).
The decline of use of hillforts at around the 100BC mark, is interpreted as a change in society, perhaps an increase in trade links with the Mediterranean, which caused instability to what have been interpreted towards the end of the Iron Age as being specialist centres of production (Creighton 2009:7).
The different interpretations, of what constitutes a ‘hillfort’ and the design and purpose of such sites, means at the very least that an interpretation of society based on the excavations of Wessex ‘developed’ hillforts is not appropriate to apply to such sites in other parts of the country (Harding 2004 :31). There may not be such a thing as a ‘typical hillfort’ and therefore generalisation from a regional to a national view is problematic; even at the site level, each hillfort appears to have its own history (Cunliffe 2006 :154).
A model of a single monolithic ‘Celtic’ society, as per Cunliffe’s interpretation, is unlikely to be appropriate (Collis 2012 :229). Also increasingly clear, is that an interpretation of Iron Age societies in Britain must consider settlement beyond hillforts (Driver 2007:84). Not all of Britain constructed hillforts (Hill 1995 :68), and settlement, even in areas that do have hillforts, was predominantly in the form of farmsteads. Hillforts served a purpose for the societies that constructed them but was not everyday domestic life for most of the population. Lock (2013) suggested that hillforts served as an idealised world: a defence against threats to belief systems, where the beliefs of a community could be reinforced and made material in the construction and maintenance of boundaries. They were not primarily for domestic life, but represented the public life of the community (Lock 2013).
Interpretation of hillforts and their place within Iron Age society has depended heavily on trends in theoretical frameworks. Invasionist interpretations have given way to economic theories, which in turn have given way to a much more multi-faceted post-processual approach, drawing on ideas of agency, understandings of landscape and of anthropological work on inter- and intra-societal relationship-forming. Hillforts have altered in interpretation from being primarily defensive structures to being also vehicles for display and prestige; to representing communal, not elite, activity; from being ‘towns’ to being mostly seasonally occupied for the most part. Most importantly, the interpretation of hillforts has changed from being as a single, predictable class of monument, to an understanding that each site must be engaged with both at an individual level, and with coeval sites in its vicinity before it can be properly understood.
From the regional accounts, it is possible to derive some general themes of interpretation: previous consideration of ramparts as serving a purely functional purpose have been nuanced with considerations of display, of the meaning of boundaries and of the effort involved in construction and maintenance, and what this may mean about the societies that performed this work.
Reconstructing a single ‘Iron Age Society’ has been suggested to be an inappropriate goal, and instead a more regional approach, at the scale that would be appropriate to its time, should be the interpretative goal. This reconstruction would require consideration of the other forms of settlement in the Iron Age, not just hillforts, which, although they have historically loomed large in interpretations of the Iron Age, give only a partial view of a population largely living in other kinds of settlement. Wordcount 2775
Avery, M., 1976. Hillforts of the British Isles: A Student’s Introduction. In D. W. Harding, ed. Hillforts: Later Prehistoric Earthworks in Britain and Ireland. Academic Press, pp. 1–58.
Barker, L., 2009. Gaer Fawr hillfort: an analysis of the earthworks, Available at: http://www.rcahmw.gov.uk/media/75.pdf [Accessed April 13, 2013].
Bowden, M. & McOmish, D., 1997. The required barrier. Scottish Archaeological Review, 4(2), pp.76–84.
Bradley, R., 2005. Ritual and Domestic Life in Prehistoric Europe New editio., Routledge.
Bradley, R., 2007. The Prehistory of Britain and Ireland 1st ed., Cambridge University Press.
Brown, I., 2009. Beacons in the Landscape: The Hillforts of England and Wales, Oxford: Windgather Press.
Clay, P., 2001. Leicestershire and Rutland in the First Millennium BC. Transactions of the Leicestershire Archaeological and History Society, 75, pp.1–19. Available at: https://www.le.ac.uk/lahs/downloads/2001/PClayTLAHS2001.pdf [Accessed April 18, 2013].
Collis, J, 2010. Why do we still dig Iron Age ramparts? In S. Fichtl, ed. Murus celticus : architecture et fonctions des remaprts à l’âge du fer : table ronde internationale organisée par l’Unité mixte de recherche 7044 de Strasbourg, l’Unité mixte de recherche 6173 CITERES de Tours et Bibracte, Centre archéologique europ. Collection Bibracte — 19, pp. 27–35. Available at: http://www.hillfortsstudygroup.org.uk/2010 Celticus Bibracte19.pdf [Accessed April 1, 2013].
Collis, John, 2012. “Reconstructing Iron Age Society” Revisited. In T. Moore & X.-L. Armada, eds. Atlantic Europe in the First Millennium BC: Crossing the Divide. Oxford: Oxford University Press, pp. 223–241.
Creighton, J., 2009. Coins and Power in Late Iron Age Britain, Cambridge University Press.
Cunliffe, B., 1983. Danebury. Anatomy of an Iron Age Hillfort, London: Batsford Ltd.
Cunliffe, B., 2005. Iron Age Communities in Britain: An account of England, Scotland and Wales from the Seventh Century BC until the Roman Conquest 4th Editio., Abingdon: Routledge.
Cunliffe, B., 2006. Understanding hillforts : have we progressed ? In A. Payne, M. Corney, & B. Cunliffe, eds. The Wessex Hillforts Project: extensive survey of hillfort interiors in central southern England. London: English Heritage, pp. 151–162.
Driver, T., 2007. Hillforts and human movement: unlocking the Iron Age landscapes of mid Wales. In A. Fleming & R. Hingley, eds. Prehistoric and Roman Landscapes. Windgather Press.
Finney, J.B., 2005. Middle Iron Age Warfare of the Hillfort Dominated Zone. Unpublised PhD Thesis. Bournemouth University.
Frodsham, P., Hedley, I. & Young, R., 2007. Putting the neighbours in their place? Displays of position and possession in northern Cheviot “hillfort” design. In C. Haselgrove, T. Moore, & E. Moore, eds. The Later Iron Age in Britain and Beyond. Oxford: Oxbow Books.
Hamilton, S. & Manley, J., 2001. Hillforts, Monumentality and Place: A Chronological and Topographic Review of First Millennium BC Hillforts of South-East England. European Journal of Archaeology, 4(1), pp.7–42. Available at: http://eja.sagepub.com/cgi/doi/10.1177/146195710100400101 [Accessed February 13, 2011].
Harding, D.W., 2004. The Iron Age in Northern Britain: Celts and Romans, Natives and Invaders, London: Routledge.
Hill, J.D., 1996. Hill-forts and the Iron Age of Wessex. In T. C. Champion & J. R. Collis, eds. The Iron Age in Britain and Ireland: Recent Trends. J. R. Collis Publications, pp. 95–116.
Hill, J.D., 1995. The Pre-Roman Iron Age in Britain and Ireland (ca. 800 B.C. to A.D. 100): An overview. Journal of World Prehistory, 9(1), pp.47–98. Available at: http://www.springerlink.com/content/b6351l872038hp50/ [Accessed February 23, 2012].
Jackson, D., 1999. Settlement and society in the Welsh Marches during the first millennium BC. Unpublished PhD Thesis. University of Durham.
Karl, R., 2008. Random Coincidences Or : the return of the Celtic to Iron Age Britain. Proceedings of the Prehistoric Society, 74, pp.69–78.
Lang, A.T.O., 2009. The Iron Age archaeology of the upper Thames and north Oxfordshire region, with especial reference to the eastern Cotswolds. University of Oxford. Available at: http://ora.ox.ac.uk/objects/uuid:6e97faa5-a3de-4ea0-a5e4-c59bc2d7a650.
Lock, G., 2013. Hillforts, Emotional Metaphors, and the Good Life: a Response to Armit. Proceedings of the Prehistoric Society, 77(March), pp.355–362. Available at: http://www.journals.cambridge.org/abstract_S0079497X00000724 [Accessed March 14, 2013].
Lock, G., 2007. Wessex hillforts after Danebury: exploring boundaries. In C. Gosden, ed. Communities and connections : essays in honour of Barry Cunliffe. Oxford: Oxford University Press.
Moore, T., 2007. Perceiving Communities: Exchange, Landscapes and Social Networks in the Later Iron Age of Western Britain. Oxford Journal of Archaeology, 26(1), pp.79–102. Available at: http://doi.wiley.com/10.1111/j.1468-0092.2007.00274.x.
Musson, C.R., Britnell, W.J. & Smith, A.G., 1991. The Breiddin Hillfort: A later prehistoric settlement in the Welsh Marches, CBA Research Report No 76. Available at: http://archaeologydataservice.ac.uk/archives/view/cba_rr/rr76.cfm [Accessed March 25, 2013].
Payne, A., 2006. Hillfort studies and the Wessex Project. In A. Payne, M. Corney, & B. Cunliffe, eds. The Wessex Hillforts Project: extensive survey of hillfort interiors in central southern England. London: English Heritage, pp. 1–38.
Savory, H.N., 1976. Welsh Hillforts: A Reappraisal of Recent Research. In D. W. Harding, ed. Hillforts: Later Prehistoric Earthworks in Britain and Ireland. London: Academic Press, pp. 237–293.
Sharples, N.M., 1991. Maiden Castle : excavations and field survey, 1985-6, Swindon: English Heritage.
Sharples, N.M., 2010. Social Relations in Later Prehistory: Wessex in the First Millennium BC, Oxford: Oxford University Press.
Wigley, A., 2002. Building monuments, constructing communities : landscapes of the first millenium BC in the central Welsh Marches. University of Sheffield. Unpublished PhD Thesis. Available at: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.269386 [Accessed March 19, 2013].