The Black Loch of Myrton: an Iron Age loch village in south-west Scotland
Wetland archaeology in Scotland has focused almost exclusively on a site type known as the crannog: an island, usually artificially constructed, with structures on top; their occupation spans nearly two millennia, from the first millennium BC to the seventeenth century AD (Crone 2012: 147; Henderson & Sands 2013: 269). It has long been recognised, however, that use of the generic label ‘crannog’ to describe sites whose occupation spans millennia implies a commonality of form and function that cannot be demonstrated (Harding 2004: 103).
Over the past decade, excavations in south-west Scotland for the ‘Scottish Wetland Archaeology Programme’ at Whitefield Loch (Cavers et al. 2011), Loch Arthur (Henderson & Cavers 2011) and Cults Loch, have gone some way to demonstrating the variability in construction, morphology and location that is subsumed under the generic ‘crannog’ label (Figure 1). Recent excavation at the Black Loch of Myrton (BLM) shows that there is probably an even greater variety of site types to be uncovered. The site had been classified as a crannog since its discovery in the nineteenth century, but investigations in 2013 demonstrated that there is no artificial foundation to the settlement. Excavation, survey and a sediment coring programme have revealed a settlement of at least seven structures—each surviving as mounds—built directly on the peat around the margins of a now dried-up loch (Figure 2). BLM is therefore a lochside settlement, although the more catchy description, loch village, has been used to draw comparisons with the iconic and better known Iron Age ‘lake villages’ of Glastonbury and Meare (see below).
The excavated evidence
So far only one of the structures has been excavated. Structure 1 is a roundhouse, some 11.7m in diameter (Figure 3). At its centre is a massive stone-built hearth that had been refurbished twice, presumably because its weight was causing it to sink into the underlying peat (Figure 4), but perhaps also as a result of repeated, even seasonal, reoccupations of the building. Each refurbished hearth consisted of a mound of cobbles contained within a kerb of large boulders and supporting either a stone slab or clay surface (Figure 5). At its most extensive, the hearth feature was at least 2.5m2 and would always have stood proud of the surrounding floor.
A surface of large stone slabs formed the foundation of the hearth and immediately around this a foundation layer of large, tightly packed and tangentially aligned alder logs was laid down. The surface beyond this consisted of a radial structure of alder logs interspersed with bundles of branchwood laid directly on the peat and forming the foundation for a plant-litter floor (Figure 6). The evidence for the superstructure consists of an internal post-ring and two outer stake-lines. A spread of small angular stones around the south-west perimeter, bounded by a stake-line, may represent an external yard and hints at an entrance on the south side of the structure, facing into the rest of the settlement.
In plan, structure 1 is very similar to other excavated Iron Age roundhouses in south-west Scotland, but what sets it apart is the survival of the organic component, the posts and the floor surfaces. As well as providing insights into the materials used in the construction, one of the most significant results to emerge from the analyses of the wood, plant macrofossils, insects and soil thin-sections is the evidence for spatial organisation within the structure. This was already apparent during the excavation in that there were clear differences in types of floor surface, particularly around the hearth. For instance, tangential log surfaces are found only on the south-west and south-east of the hearth, while on the north-east there is only a surface of radial logs (Figure 3). The later stony surface to the south-east of the hearth does not extend around the south-west or north-east sides. The occupation deposits in the south-west and north-east halves of the house also contrast. Different types of plant litter appear to have been used for flooring—sedges and rushes on the north-east and grasses and leaves on the south-west—but the most significant difference is the absence of domestic rubbish on the north-east side of the house. Food and hearth debris has been trampled into the floors on the south-west side, and although much debris had spilled from the hearth, it has built up only on its south-west and south-east edges. The artefacts were also found mainly on the south-west side of the house and around the south-west and south-east sides of the hearth. The implication is that most domestic activity was taking place in the west half of the house and possibly to the north and south of the hearth. There are complementary hints from the insect and soil thin-section evidence that animals were stabled on the north-east (exterior) side of the house.
As yet, the only evidence for a palisade consists of a nineteenth-century antiquarian description and two oak stakes found in a drainage ditch to the west of the settlement, which may or may not relate to this enclosure. One of these posts was radiocarbon dated to 380–200 cal BC (2215 ± 30 BP; SUERC 32598). One of the alder posts within structure 1, however, was radiocarbon dated to 770–410 cal BC (2470 ± 35 BP; SUERC 32597), indicating that there were at least two phases of building activity on the site. The chronology has been further refined by a dendro date from oak timbers used in structure 1 that were felled between 461 BC and 429 BC, and by Bayesian analysis of radiocarbon-dated foundation timbers that indicates construction between 510 and 435 cal BC. A fifth century BC date is consistent with the evidence from other crannogs where detailed chronological work has been undertaken (Crone 2012: 161).
As a lochside settlement, BLM represents an entirely new site type for Scotland. The only contemporary parallels that come readily to mind are the Iron Age lake villages of Glastonbury and Meare in Somerset (Coles & Minnitt 2000), and the settlement at Ballycagen Lough, Isle of Man, which was built out on the floodplain of a small river (Bersu 1977). While many of the features of the Glastonbury and Meare sites can be recognised at BLM, such as the refurbishment of sinking hearths, BLM is perhaps structurally more akin to the Irish Bronze Age settlements of Clonfinlough (Moloney 1993) and Cullyhanna (Hodges 1958), both of which consist of a small group of house platforms built directly over fen peat on the shores of a small lough. These are all exceptional sites, with few, if any comparable examples elsewhere in the British Isles.
BLM is also significant in a Scottish context because the site lies on a spectrum between the classic wetland site, the crannog and the roundhouses of the terrestrial record. It has always been difficult to transpose the evidence of the better preserved, organic structural evidence from one site to the other, but BLM provides an intermediary stage, a well-defined roundhouse with its organic components relatively intact and with evidence of sophisticated structural responses to challenging environments. BLM demonstrates that in south-west Scotland during the Iron Age a wide range of wetland environments were being sought out for settlement, and this may reflect a different social and economic stratum to that represented by crannogs in terms of access to resource bases and social organisation.
Although the BLM settlement is unique in its choice of location, the form of the settlement is typical of the earlier Iron Age in south-west Scotland. It is with this potential to fill out our otherwise skeletal knowledge of how early Iron Age roundhouses were built, occupied, re-built and abandoned that BLM has most to offer future studies.
Fieldwork at the site continues in 2015.
We are grateful to the landowners, Dourrie Farming, and in particular to Rory Christie for his permission to excavate the site and for his continuing interest and support of the project. The fieldwork and post-excavation programme were grant-aided by Historic Scotland.
- BERSU, G. 1977. Three Iron Age round houses in the Isle of Man. Douglas: Manx Museum & National Trust.
- CAVERS, G., A. CRONE, R. ENGL, L. FOURACRE, F. HUNTER, J. ROBERTSON & J. THOMS. 2011. Refining chronological resolution in Iron Age Scotland: excavations at Dorman’s Island Crannog, Dumfries & Galloway. Journal of Wetland Archaeology 10: 71–108.
- COLES, J. & S. MINNITT. 2000. Industrious and fairly civilized: the Glastonbury Lake Village. Exeter: Somerset County Council Museums Service.
- CRONE, A. 2012. Forging a chronological framework for Scottish crannogs; the radiocarbon and dendrochronological evidence, in M. Midgley & J. Sanders (ed.) Lake-dwellings after Robert Munro: 139–68. Leiden: Sidestone.
- HARDING, D. 2004. The Iron Age in northern Britain: Celts and Romans, natives and invaders. London: Routledge. http://dx.doi.org/10.4324/9780203326107
- HENDERSON, J.C. & M.G. CAVERS. 2011. An Iron Age crannog in south-west Scotland: underwater survey and excavations at Loch Arthur. Proceedings of the Society of Antiquaries of Scotland 141: 103–24.
- HENDERSON, J. & R. SANDS. 2013. Irish and Scottish crannogs, in F. Menotti & A. O’Sullivan (ed.) The Oxford handbook of wetland archaeology: 269–82. Oxford: Oxford University Press.
- HODGES, H.W.M. 1958. A hunting camp at Cullyhanna Lough, near Newtown Hamilton, County Armagh. Ulster Journal of Archaeology 21: 7–13.
- MOLONEY, A. 1993. Excavations at Clonfinlough, County Offaly (IWAU Transactions 2). Dublin: Irish Wetland Archaeology Unit.
* Author for correspondence.