Post-excavation work is well underway here at AOC’s lab, but while we work on undertaking various assessments and analyses which will contribute to the final report, we can share the interim results of our work.
Trench One aimed to investigate both whether the intramural cell opposite the entrance (Cell 2)was a primary or secondary feature to the broch construction, and the extent of the victorian excavation within the broch interior. The trench was positioned so that its northeast edge was at the mid point of Cell 2’s entrance, extending for 2m counter-clockwise from this point and extending 1m away from the broch wall.
Removal of the topsoil exposed a deposit of backfill from the Victorian excavations; a clay pipe bowl and a button support this dating. In the area immediately in front of the entrance to Cell 2 there was a concentration of larger stones more deliberately laid, forming a ramp into the cell entrance.
Underlying the rubble backfill was a deposit that may represent a former ground surface. In order to test this, the deposit was investigated by the excavation of a 1m by 0.5m sondage in the eastern corner of the trench. This exposed a possible paving slab set in a brownish black sandy silt with occasional charcoal flecks and heat cracked stone, a possible occupation deposit. This directly overlay the broch wall , below which was an old ground surface set upon the natural – a sandy silt glacial till.
As well as allowing us to examine a possible occupation deposit and ground surface, the excavation of trench one exposed the lowest courses of the inner face of the broch wall. There were three distinct levels of preservation within this wall. The upper, above ground, courses of the wall had been conserved and consolidated during the 1990s works and were in good condition. The portion of the wall extending for 0.5-0.6m below the ground had been heavily affected, probably by water flow and pooling through the structure of the broch and soil which had left stonework in this area full of voids and loose stone, with the remaining stone in a fragmentary and poor condition. The lowest two to three courses of the broch wall were well preserved.
Trench Two focussed on the guard cell accessed from the right hand (north) side of the entrance passage. This was reportedly cleared of loose material during the consolidation works during the 1990s; the aim of these excavations was to remove further collapsed rubble and blown topsoil and vegetation to return the chamber to its former condition.
Removal of accumulated vegetation across the guard cell revealed the backfill from the 1990s excavations. Removal of this rubble deposit exposed structural elements of the guard cell as well as soft deposits likely relating to the occupation of the broch.
At the southern end of the guard cell were a series of structural elements. Immediately within the guard cell was a single large paving slab, and to the north of this the remains of a blocking wall extending across the width of the guard cell. This blocking wall was a single course wide and survived up to three courses high. It abutted against the side walls of the guard cell, suggesting that it was constructed later than the cell itself. Further paving slabs lay to the north of blocking wall.
At the northern end of the guard cell were a series of deposits considered likely to relate to the occupation of the broch. Across most of the guard cell were two similar deposits interpreted as parts of the same occupation surface. These deposits were separated by a band of darker sandy silt with frequent charcoal flecks, which appeared to be the fill of a cut feature extending across the width of the guard cell and continuing below the side wall of the cell (Plate 6). This might suggest that there was a period of occupation predating the broch, or that the guard cell was a secondary construction or alteration.
Trench Three was located in the area of a opening in the interior broch wall identified during the 1990’s consolidation works. Work was halted here in the 1990s on the discovery of possible archaeological deposits. The abrupt cessation of works in this area facilitated the further decomposition and displacement of deposits and masonry without clarifying the nature of the stonework or deposits. Our limited excavations here aimed to determine the nature of both the structural and archaeological features and to enable the consolidation of the southern arc of the broch wall.
In the area of the opening, a loose mixed rubble deposit formed the upper most deposit. This was interpreted as backfill from the from the Victorian excavations. From within this deposit came two fragments of bone dice, perhaps from the same decorated bone die. It was quite a surprise to find these stunning artefacts among Victorian backfill, which must have either gone unnoticed or been overlooked by the Victorian excavators. These will be carefully conserved by AOC’s conservators, and closely examined so that we can learn as much as possible about them. Bone dice are rare and exciting finds but they are not considered unusual within Iron Age settings, so we will be looking to other broch sites for comparanda.
Underlying the rubble backfill were the remains of a small, possibly peat, fire. Built on the in-situ rubble collapse deposits by the Victorian excavators, it is not thought likely to be derived from the occupation of the broch in antiquity.
Removal of the rubble backfill and ashy deposit exposed the east and west walls of the opening into the broch wall, as well as the underlying rubble collapse deposit. Both the side walls of the opening were built upon the rubble deposit, implying that these were constructed after the collapse of the broch walls.
Structural Consolidation Works
Central to the aims of the project was the consolidation of the broch walls, which were in need of urgent repair in some areas.
The inner broch wall in the area of the guard cell was cleared of turf and topsoil and loose stonework was removed. In order to stabilise the corbelling of the cell and the strengthen the inner wall, counterbalancing stones were built up over the corbelling with larger stones used to tie the corbelling into the wall core. The wall was ramped up from the entrance passageway in order to form a buttress for the in-situ stonework of the inner broch wall.
The area most at risk was the southern arc of the broch, where the outer broch wall has collapsed down the slope of the river valley. The outer wall face is not visible in this area and the enclosing wall has been constructed over the remains of the broch. Of particular concern was the area where the outer wall face has broken, with individual stones of outer face having little support and there being nothing to stop the gradual splaying outwards of the entire wall.
In order to support the broken edge of the outer wall and to buttress the mass of rubble core of the broch wall, a supporting/retaining wall was constructed from the opening through the inner wall extending clockwise to the break point of the outer wall face, where the structure of the broch starts to rise. Turf and topsoil were removed from the rubble collapse of the wall and loose stone was removed and set aside to be re-used. A dry-stone wall was built up with larger stones being used to tie the construction into the rubble mass of the in-situ broch wall. The wall was capped with geotextile overlain with clay to prevent water and soil penetrating the wall. This was topped with topsoil and turf.
The wallhead was generally in poor condition and its core, although in part rebuilt, continues to move out and overhang the outer wall face in places. Areas with overhanging stonework were stripped of turf and topsoil, loose stonework was reseated or removed to return the wall head to a stable arrangement. On completion of the consolidation work, the wallhead was returned to a turved condition. Stripped areas were covered in geotextile, then a layer of clay to prevent water and soil penetrating the rubble fill of the wall. The clay was then topped with topsoil and turf.
Excavation of trench one exposed the inner face of the broch wall. There were three distinct levels of preservation within this wall. The upper, above ground, courses of the wall had been conserved and consolidated during the 1990s works and were in good condition. The portion of the wall extending for 0.5-0.6m below the ground had been heavily affected, probably by water flow and pooling through the structure of the broch and soil. The lowest 2-3 courses of the broch wall exposed in trench one were well preserved and required no intervention. This had left the inner broch wall in an unstable state that was unclear prior to the excavation of trench one.
The poorly preserved areas of the broch wall were cleared of loose and fragmentary material and the voids packed and pinned with new stonework to stabilise the wall in this area.
The field programme undertaken in conjunction with the conservation works at Dunbeath has advanced our understanding of the history of occupation of the monument and has stabilised the worst areas of structural instability at the monument. A full analysis of the structure and how these elements articulate with the excavated evidence described here comprises the first part of the post-excavation programme. Additionally, the analysis of environmental and artefactual material deriving from the excavations will complement this research. These programme objectives contribute to our post-excavation research design. We’ll share more results in due course!