Diversity in time and space
It is now generally accepted that there is no definitive interpretation of the function of hillforts and enclosures, which probably conveyed a variety of ideological statements at different times and places to different people. There are a range of potential functions, attested or implied, for forts and other enclosed places (e.g. Ralston 2006, 19-24), but it is simplistic to use lists of possibilities to work out specific functions for enclosures from archaeological evidence (as discussed by Haselgrove and McCullagh 2000, 77). It is generally understood now that proposing a single purpose for a ‘fort‘ is illusory and probably not demonstrable through excavation. In addition, there are widely acknowledged difficulties of relating interior features to enclosure circuit sequences (e.g. Broxmouth).
It would be unwise to attempt to classify function from visible morphology alone: for example the Wessex Hillfort Geophysical Project (Cunliffe 2006) found no correlation between the morphology or setting of a hillfort and the nature of activity patterns within as evidenced by archaeological remains. But for many parts of Scotland there has been relatively little excavation or geophysical survey of enclosure interiors, and opportunities for both should be taken.
Patterns of interpretation of the context and role of forts of Scotland have largely reflected the paradigm shifts developed largely out of Wessex hillfort models. There have been some attempts to consider local and regional datasets on their own merits (e.g. TLEP; Frodsham et al. (2007) in the Cheviots), but in other areas without recent investigations new interpretations are placed upon an unchanging dataset (e.g. Nieke 1990 versus Harding 1997 on the character and permanence of occupation of Argyll forts and duns).
Enclosures as places of habitation
Archaeological evidence suggests that the majority of enclosures were occupied on some basis, but there is considerable debate as to the character of that occupation – whether year-round, seasonal, intermittent or episodic / or celebratory, and whether settlement was long-term and permanent or short-term and shifting. These issues transcend the interpretation of enclosed sites, and are fundamental to a modelling of the character of later prehistoric societies and landscapes. Topographic and locational considerations such as the high altitude and exposed location of many sites have been taken as evidence of at most seasonal occupation of forts and enclosed settlements (e.g. Bradley 2007, 254-6), and other exposed (e.g. stack) locations have similarly been interpreted as unlikely to have supported long-term occupation.
Recent excavations have begun to reveal evidence that supports the long-term but seasonal or episodic nature of activities at certain sites (e.g. Braehead , Port Seton West), and short-term use of others (e.g. TLEP ; Carghidown promontory fort, Toolis 2007). Excavation evidence is not always clear-cut – for example there seems little real difference in the character of the excavated remains and materials from Port Seton East and Port Seton West that justify interpreting the former as a permanent farmstead and the latter as a communal seasonal market or fair (Haselgrove and McCullagh 2000).
Understanding activities within fort interiors represents an important research theme, and consideration is required as to the best ways to do this, whether through excavation or remote-sensing techniques, or combinations thereof. General rules are most unlikely to apply, and more contextual local studies are needed. It is also important to consider what can be expected to survive within the archaeological record, particularly if plough- truncated.
Forts as nucleated settlements / oppida / tribal capitals
There is no good archaeological evidence for large pre-Roman Iron Age nucleated settlements in the same terms as European or Southern English oppida or minor oppida (cf Feachem 1966). Nor can tribes alluded to on Ptolemy’s map of Britain (probably illustrating the political situation in the 2nd century AD) be ‘matched’ to larger forts forts as a means of identifying ‘tribal capitals‘ (e.g. Rideout et al. 1992, 142-3 who regard this as a futile pursuit), even if on the Continent some ‘oppida‘ are ‘contact phase‘ phenomena. Significant Late Bronze Age and Roman Iron Age settlement is apparent on Traprain Law, but even there the character and permanence of occupation is uncertain. Some of the larger sites contain large numbers of house platforms within their outer circuits (e.g. Eildon Hill North, Tap o‘Noth), but little exists idea of the full date range, permanence and extent at any one time of the implied occupation (for example as discussed by RCAHMS (1997) in relation to the Boonies house sequence as demonstrated by excavation). Eildon Hill was interpreted by its excavators as a market or fair location primarily populated with flimsy, temporary structures (Rideout et al. 1992); conversely others still wish to regard it as an urban centre (e.g. Wilson 2010, 46). However, many still consider the larger forts as centres of power and authority. More evidence for the nature and chronology of activity at ‘big forts‘ is required.
Defensive properties of enclosures
A range of opinions have been expressed as to the extent to which defensive considerations influenced the choice of site location (e.g. ‘prominent landmarks‘ and ‘naturally defensible‘ locations) and the nature and strength of enclosing works, although most commentators agree that some influence is evident in many of the forts at least. But there are many examples where defensive considerations appear not to have been primary factors in the siting and/or form of enclosure – for example enclosures located on overlooked sites (Chesters, Drem is most frequently quoted), overprovided with enclosing works in relation to area enclosed, overprovided with entrances, or simply too diminutive to have had a defensive function (e.g. Woodend; Banks 2000). On the other hand, oblong forts have massive, apparently gateless walls and interiors that were invisible from the outside, but even these have been recently suggested as ceremonial enclosures (Harding 2004, 87).
The archaeological evidence for warfare, conflict and violence at Scottish forts and enclosures is rather limited (Toolis 2007, 308-10) and equivocal (c.f. Bowden and McComish 1987 on Wessex hillforts). Because we no longer trace a linear evolution of fence-wall-dump is no longer universally applicable, it is probable that changing enclosure forms do not reflect developments in methods of warfare, at least not in an easily legible or direct manner (c.f. Avery 1993).
Have assessments of defensive strength and capability been clouded by the tactics of modern military warfare as opposed to pre-state ritualised, embedded warfare and raiding (Sharples 1991; Armit 2007, James 2007)? If it sis accepted that vitrification might have occurred as a result of careful, possibly symbolic destruction of timber-laced walls, how is evidence of the specific circumstance of destruction found? Retribution or reprisal by ‘conquerors‘ and a sign of political instability is a generally preferred explanation, but it could be a peaceful ‘killing‘ by the occupants to put the site beyond use, as argued for the deliberate infilling of souterrains (Armit 1999). What evidence can be marshalled in support of either explanation?
Symbolic aspects of enclosures
Various examples of work exploring symbolic aspects of enclosures have been undertaken. This work includes the exploration of the nature of the aggrandisation of entrances (e.g. at Dunnideer, Aberdeen RCAHMS 2007, 103) as well as miniature earthworks, and token or diminutive ramparts such as exist at the White Meldon (Ralston 2006, 126-8). Other aspects that could be considered within a symbolic context include:
- the scale of enclosing works which are out of proportion to the internal space;
- the overprovision of entrances;
- phenomenological and landscape entrance alignment references;
- the reuse of earlier material (e.g. imported Roman stone fragments at Rubers Law, Roxburghs. and Clatchard Craig, Fife.); and
- associations with earlier monuments, whether iron Age or earlier, at site locations (currently in need of synthesis).
Special or structured deposits occur at boundaries, including people (e.g. Knowes cist burial, Haselgrove 2009) animals (Eildon Hill, Rideout et al. 1992), metalwork (Burnswark; Buchsenschutz and Ralston 2007) and domestic debris incorporated in ramparts (Kaimes fort, Simpson et al. 2004). However, there is a rarity of rich votive deposits compared to e.g. Wessex (Rideout et al. 1992, 142).
Enclosures as communal places
A number of examples of recently excavated forts or settlements have been proposed as ‘not farmsteads‘ – Brown Caterthun, Port Seton West, Over Rig, East Lothian Late Bronze Age enclosures (all based primarily on lack of evidence for occupation) and Eildon Hill North. Oblong forts have been proposed as having a ceremonial function by Harding (2004, 87). Promontory, stack and headland forts are all posited by Harding (2004, 94, 145, 298) as not for ‘regular occupation‘ and by Henderson (2007b, 308-9) as for a range of settlement and non-settlement functions.
Early Medieval forts
Early Medieval forts have been examined and interpreted within a different explanatory ‘documentarily-enhanced‘ and historically specific, framework, and generally by different researchers, sometimes with different preoccupations, from those interested in Iron Age constructions. The defensive properties of the ‘nuclear‘ forts are seldom questioned, since there are documentary records of burning/attempted burning, capture, destruction, and sieges (Alcock 2003, 180). They are generally regarded as ‘royal‘ strongholds between which kings progressed to secure territory and exact dues from subordinate aristocracy; some e.g. Dunadd, Argyll are seen as paramount capitals. Harding (2004, 235-6) has questioned that type of occupation, highlighting the apparent lack of grand buildings, and alternatively suggested that they were perhaps seasonal, special sites with few permanent buildings, as argued for Iron Age forts. On the other hand it may be that even grand buildings did not have earthfast foundations, through e.g. use of post-pads, and therefore would be difficult to detect archaeologically.
Understanding activities within forts represents an important research theme; geophysical survey has proven a useful tool in other areas.
The nature of the largest forts (e.g. Traprain Law, Eildon Hill North, Burnswark) in relation to the wider settlement sequence remains an area requiring further work.
What can researchers of Iron Age and Early Historic fortifications learn from each other, to allow the gap to be bridged? Does it need bridging?