2010:598 - Knocknarea North, Sligo

A limited test excavation was carried out on a group of house structures located on a ledge on the steep northern slopes of Knocknarea Mountain, Co. Sligo. The excavation formed part of the Knocknarea Archaeological Project, which has included major surveys and trial excavations of mainly Neolithic remains on the mountain over the last ten years. The main aim of the test excavation of the house structures was to establish their cultural context, since their construction as well as location on a relatively inaccessible ledge on the steep northern slopes gave no clear indication of their date and function. The structures, which were discovered during field-walking, had no evident link to any of the other sites recorded on the mountain, and created thereby an isolated group of remains in a rather remarkable location on a narrow ledge overlooking the Atlantic.
This part of the northern slopes of Knocknarea is a conspicuous place, consisting of vertical cliffs interrupted by narrow ledges. The site is located at 210m above sea level on a ledge some 15m wide and c. 200m long. Vertical cliffs 5–10m high define the ledge both downhill and uphill. The view is breathtaking, overlooking Sligo Bay and the Atlantic. The ledge, even though isolated and extremely exposed, can be accessed from above by a path through a natural break in the cliff face, and also from the west where the ledge merges with more gently sloping ground.
Prior to the excavation the structures were identified as a rectangular setting of slabs, some placed lying down while others were set standing to a height of c. 0.8m. The outer measurement ranged from c. 4m x 5m to 8m x 6m. The clear rectangular symmetry of their layout indicated some form of foundation for a house structure, even though the building technique with rough slabs, some evidently set standing, was seen as rather unconventional, and made the interpretation of their cultural context unclear.
Houses
To keep interference with the sites at a minimum, the methodology employed was to de-sod one of the structures completely, to thereby get a clearer understanding of the actual layout and construction technique. The de-sodding was followed by limited test excavation of squares of a total of 3m2.
The structure chosen for de-sodding was the most eastern one (House 1). This site turned out to consist of two house structures. The first was placed parallel with the ledge some 4m out from the cliff behind it. This structure (House 1A) consisted of a clear wall of standing slabs set shoulder to shoulder to a height of up to 1m along its back wall (closest to the cliff behind it) and two short walls. The front of the house – i.e. the long side facing out from the ledge – had been built of much lower-standing slabs (c. 0.3m high). The external dimensions of House 1A were 5.9m x 3.1m. Besides the actual standing slabs, substantial stone packing was present along the wall lines. No clear evidence for an entrance could be established.
The second house structure (House 1B) was placed at right angles to House 1A, and it had no southern short side ‘of its own’ since this consisted of the long side of House 1A, which it abuts. The two long sides of House 1B consisted of a c. 1m-wide and 0.3m-high dwarf wall of limestone slabs, intermixed with packing. The northern short wall, which was located only c. 2m away from the c. 8m-high precipice defining the ledge, was considerably simpler and consisted in parts of a single line of small slabs.
In contrast to the previous house, House 1B had a clearly defined entrance halfway along the western long side. The entrance was 0.7m wide and defined by selected right-angle slabs. Along the inside of the wall opposite the entrance, six naturally shaped slabs (0.3–0.6m in height) were set standing shoulder to shoulder over the entire length of the wall, creating an imposing interior facing. In front of the middle two there seemed to be the remains of what originally might have been a small ledge created by slabs and smaller stones. The outer dimensions of House 1B were 5.6m x 5m, while the interior measured 4.8m x 3.6m. A limestone slab measuring 1.1m x 0.4m x 0.2m was lying flat on the ground in the middle of the house, together with two slightly smaller slabs.
All stones in both House 1A and 1B consisted of local limestone, either quarried or loose field stones.
Two square metres were excavated down to sterile ground, in the northern part of House 1B. No finds were made in the very thin soil (c. 0.15m) covering the underlying bedrock.
Some 15m north-west of House 1, a low foundation for a building (House 2) was noticed due to limestone slabs protruding in the rich grass and heather growth. House 2 had outer dimensions of 5.5m x 3m and was aligned east–west just at the very edge of the ledge.
Further to the west on the ledge, at a distance of some 50m from House 1, a third house foundation was found. This consists of two joined rectangular sections constructed partly by standing slabs up to 1m in height, and partly by lower slabs set on edge. At least one wall consists of double rows, c. 0.2m apart, of large limestone slabs, giving the wall a total width of c. 1m. This building is located at a point where the ledge is only c. 10m wide, and, since the section that is placed perpendicular to the alignment of the ledge is c. 9m long, it effectively cuts off the ledge at this point. This section of the building has outer dimensions of c. 9m x 5m. The second section of the building is set at a right angle to the first, and abuts its eastern long side midway. This probable foundation for a house is surprisingly narrow, with outer dimensions of c. 6m x 2.7m and with an original inner width of c. 2m. The walls of this structure consist of large limestone slabs (up to 1.4m in length and up to 0.8m in height), set on edge. No apparent doorways can be traced, but the structure is heavily overgrown by grass and heather. The western section is built straight out to the edge of the ledge.
A further 30m west of House 3 a small foundation of a building was found in heavy scrub. This is the smallest in the series, measuring 3m x 4m, and consists of large (0.5m) limestone slabs, with a possible entrance to the west. In the middle of the building there is a large limestone slab (c. 1.2 x 0.8m) occupying a considerable proportion of the presumed floor space.
There are possible remains of a slightly larger house foundation immediately to the east, measuring c. 5m x 3m. There are, however, only a few stones remaining to indicate this part.
Conclusion
The remains recorded on the extraordinary exposed ledge, on the northern slopes of Knocknarea, represents a small group of houses. In all, the four different sites (Houses 1–4) probably represent six different buildings. The character of some of the foundations indicate that they could have functioned as dwarf walls, on top of which sods could have been placed to support some sort of roof. Other sections indicate that the slabs might have supported posts, which then would have been the main structural element of the walls, supporting a roof.
Since no dating evidence has come to light, the cultural context of the houses is unknown. There are no known springs nearby, and the north-facing exposed mountainside would hardly have added to any kind of comfort on the ledge. A common interpretation of upland buildings of post-prehistoric date is that they might be associated with booleying. Neither the location, very close to arable lowland, nor the construction, which sometimes is very complex, supports this interpretation.
When at the site one is looking straight down on the ruins of Killaspughbrone medieval church, perched on the headland below, with the Atlantic creating the backdrop.
A possible link to the early church cannot be ruled out and possibly the houses should be interpreted in the context of the isolation that the narrow ledge up the hillside offered.
The date of the houses has to remain uncertain, but, taking the various constructional features into account, the location and also the total lack of any prehistoric artefacts, it may be suggested that they probably were built and used sometime between ad 1000 and 1500.