by Andy Heald, AOC Archaeology Group
Below is an overview of some of the most well-researched Iron Age sites in South East Scotland: enclosures. The wealth of information available at these multi-period sites has drawn the attention of archaeologists for many years, and will likely attract additional research in the future.
Traprain Law, East Lothian
The craggy outline of Traprain Law is one of the most iconic Iron Age sites in northern Britain. Since the turn of the century parts of the site have been excavated by numerous individuals such as Curle, Cruden, Bersu and, more recently, Armit, Dunwell and Hunter.
As reported in the recent Atlas of Hillforts the excavations by Curle and Cree encountered deep stratified deposits on the plateau area halfway up the slope, recovering numerous hearths and fragments of both circular and rectangular structures, though the record of their discoveries is too all intents and purposes irrecoverable. The area had certainly been intensively occupied during the Roman Iron Age, providing a wealth of Roman goods that far outstrips any other site in Scotland and, unusually, seems to have prospered from the 1st century AD through at least the 3rd and 4th centuries and into the 5th century AD. Mainly excavated in spits, from the lower levels they also recovered a large assemblage of Late Bronze Age metalwork. While these finds suggest two principal horizons of occupation on the hill, the dating of the ramparts is imperfect, though single entity radiocarbon dates from contexts below the rampart of the 4ha enclosure on the summit fall in the Late Bronze Age, and an indistinguishable date was returned from an ashy deposit that accrued against its inner face (Armit, Dunwell, Hunter and Nelis 2005). Otherwise the Roman finds from the terrace bank, clearly place this in the early centuries AD.
Nevertheless, 3rd and 4th century finds recovered by Bersu and Cruden clearly provide a terminus post quem for the Cruden Wall, while Close-Brooks also suggests that a pin excavated by Cruden from an earlier structure above the quarry might place its construction in the 5th century AD. However, in the light of an excavation carried out in 1986 at the foot of the hill on the NE, which found that two palisade trenches had been replaced by a bank and ditch and finally superseded by an earthen rampart with an external stone revetment, it would be naïve to suggest that the plan of the visible elements of the defences can be understood with a simple model of expansion or contraction, and that the narrow trenches excavated to date provide any real insight into their construction or chronology (Strong 1986). Whether the Late Bronze Age activity attested by the finds on the western end of the hill fell within a Late Bronze Age enclosure, has yet to be demonstrated.
Radiocarbon dating from the most recent interventions broadly confirm the chronology presented by the artefacts, with a Late Bronze Age phase, possibly associated with the earliest enclosures on the hill, and an intensive Late Iron Age and Roman Iron Age phase, unusually continuing probably into the 5th century AD. Dates from the material recovered by Peter Strong, however, fall between these two extremes. Horn suggests 4ha summit enclosure start: 915-850 BC, end: 865-800 BC, 4ha summit enclosure span: 0-55 years, outer enclosure palisade: 480-390 BC, outer enclosure rampart: 370-200 BC (forthcoming).
Hownam Rings
The fort known as Hownam Rings, situated on the local summit of the NW spur of Windy Law, was the scene of excavations by Mrs C M Piggott in 1948 that played a major role in the evolution of hillfort studies in Scotland in the 1950s and 60s, giving rise to the much quoted Hownam sequence in which it was suggested that palisaded enclosures were successively replaced by univallate and multivallate defences, before being superseded by Roman Iron Age settlements built across derelict ramparts.
Unsurprisingly, the sequence has been argued to be overly simplistic; more so the relationship of the two palisade trenches that are supposed to form the earliest phases of the sequence is far from clear, Supposedly they are overlain by the ramparts of the multivallate phase, but in practice they were found only in an area where the ramparts had been ploughed down by the Roman Iron Age and their relationships to these defences is by no means secure. Finds from the excavations include a range of coarse stone tools and coarse pottery and an iron knife, while fragments of Roman pottery show that the late Iron Age settlement was certainly occupied into the 1st and 2nd centuries AD.
Broxmouth, East Lothian
Broxmouth was fully excavated in 1978-1979 prior to its destruction by quarrying. Excavation revealed a multi-period site beginning with the construction of a palisaded enclosure at about 640-570 cal BC, which was followed by an unenclosed settlement of massive timber round-houses.
After this phase a substantial hillfort was constructed, the inhabitants stamping their mark upon the land. The succession of defences was long and complex. The first rampart was constructed about 490-430 cal BC and formed an univallate enclosure. A second rampart and ditch were added subsequently and a monumental timber-lined entrance way was created. About 395-375 cal BC this entrance was blocked and a new entrance built on the SW, and the defences were elaborated in a series of stages, part of the circuit displaying three lines of ramparts and ditches. These went out of use around 295-235 BC.
Later, as the ramparts fell into disrepair a settlement expanded over the redundant defences, contracting again 235-210 cal BC. The final phase of occupation began about 100-60 BC, comprising a series of stone and timber round-houses, with the site finally abandoned around AD 155-210. Unusually for Scottish forts and settlements, the calcareous soils had not only preserved an extensive faunal assemblage, but also a wide range of bone tools and artefacts, as well as pottery, copper alloy objects and evidence of iron working; objects from the latest phase included items of Roman manufacture.
White Castle, East Lothian
Oval in plan, its defences comprise two main elements, an inner oval enclosure with a single rampart and ditch, encircled by a slightly eccentric outer pair of ramparts with a medial ditch. The inner enclosure measures 0.28h. Evaluation excavations in 2010-13 showed a rampart covered a row of timber posts. The most recent topographical survey during the recent evaluations has identified no fewer than eighteen platforms within the interior. On the strength of ten radiocarbon dates from a range of contexts, the excavators argue for three broad phases in which the eccentric and short-lived inner rampart was inserted into an earlier enclosure about 400 BC. However, the platforms represent ongoing occupation possibly as late as the beginning of the 2nd century BC; a single radiocarbon date from the outermost rampart falls in this last phase (Cook and Connolly 2013, 25-7).
References
Armit, I, Dunwell, A, Hunter, F, and Nelis, E 2005 ‘Traprain Law’, Past 49, 1–4.
Cook, M, and Connolly, D 2013 Rampart Scotland Project 001: The Hillforts of East Lothian Season 3. White Castle, Garvald, East Lothian, Data Structure Report.
Horn, J forthcoming The dating of hillforts in Britain and Ireland, Unpublished PhD thesis, University of Edinburgh.
Strong, P 1986 ‘Traprain Law (Prestonkirk p), fort, rampart’, Discovery and Excavation in Scotland 1986, 20.