Excavation
Date June 2011 - July 2011
Event ID 966099
Category Recording
Type Excavation
Permalink http://canmore.org.uk/event/966099
NF 97000 91340 The site at Rubh’ an Teampuill represents one of the key archaeological sites on Harris, with evidence for human occupation ranging from the Mesolithic to post-medieval and beyond. The site is set within a landscape showing evidence of human activity over several millennia demonstrated as earthworks, possible platforms and collapsed ancillary structures, cultivation, possible graves, and more ritualistic monuments such as standing stones.
The site of a broch on the promontory at Toe Head is well known locally and survives as a grass-covered ruin partly superimposed by an early 16th-century chapel. The chapel itself is recorded in the RCAHMS volume of 1928, although the broch is not mentioned. It was the subject of surface stripping but has otherwise not been investigated further (Simpson 1965). The monument is visible on the ground as a turf-covered mound with evidence for both external and internal facing edges. The external diameter is c16m and the internal diameter c8m. The aim of the current investigations, undertaken June–July 2011, was to examine a sample of the monument with respect to understanding its constructional method, the nature of its survival, and its vulnerability to coastal and wind erosion.
The outer face of the broch had been laid directly on to bedrock (Lewisian gneiss; and had been levelled by underpinning. These outer facing stones were the most substantial encountered: the largest was c1.2m wide and 0.8m deep and sat against the highest point of bedrock exposure. The bedrock slopes away down to the N and suggests that the sea-facing aspect of the structure was deliberately positioned on the highest available point on the headland.
The outer facing stones formed part of a ring wall c1.2m thick with less substantial stones lining the inner face and with a wall fill of dark soil and small angular stones. This ring wall may have been used in order to define the footprint of the overall building. The inner face of the broch proper was also of less substantial facings and confirmed that the overall thickness of the broch wall was c4m.
In the area excavated the core of this walling contained part of a small curved chamber accessed from the interior of the broch through a narrow c0.55m wide faced passage c1.4m in length (Figure 1). Lewisian gneiss does not lend itself well to building and throughout the broch the stone faces were only roughly coursed, uneven and levelled as appropriate with small rubble. That said, there was some evidence in both the chamber and the passage to suggest that part-corbelling of the walling may have occurred. A small recess c0.1 x 0.1m had been constructed in the W wall of the passageway near the entrance, presumably to provide some form of securing mechanism against a door by way of a cross beam. A less formal angular arrangement of coursing provided a similar feature on the E face. The finds from Northton consisted of a small assemblage with a restricted range of artefacts, dominated by animal bone and handmade, decorated ceramic vessels. The form of the vessels and their applied decoration is widely paralleled in Iron Age Western Isles contexts and is typical of roundhouse-associated pottery in the area. All of the artefacts appear to have been produced from locally-sourced raw materials.
To the N of the broch, fragments of human remains were recovered during the excavation of a stone cist. Dating of this feature is difficult, although Iron Age or Norse is most likely.
Archive: University of Birmingham (currently) and Stornoway Museum (intended)
Funder: HLF via Landscape Partnership and HDL Ltd
University of Birminham, 2011
