North Uist, Beinn Na H-aire

NF 90725 58940 The site was identified during a fieldwalking trip undertaken as part of an archaeological survey of the Eaval peninsula, E of Loch Obisary on North Uist. Work was undertaken from 18 November 2014 – 13 May 2015.

The site is located on and around a D-shaped rocky knoll whose summit lies c15m above the surrounding moor. The summit of the knoll is covered in a thick layer of heather, which partly conceals a mass of loose stone, a number of stone features or structures, and several stone-lined pits. The fact that this is a previously unknown site, its size and its level of preservation, with limited damage from human or animal activity, set this site apart from other similar sites on Uist.

The main structural remains are located in c30 x 20m area on the summit of the knoll. A second set of structures, including a small sub-circular hut (possible shieling 1) and possible stock enclosures and a rock shelter lie at the foot of a cliff face to the N of the knoll. A third set of features to the S of the knoll, which are laid out along a c40m long dog-legged line, include a possible pathway, one or two stone-walled buildings, stone-lined pits, and stone-revetted terraces. For clarity of description, the site is split into three principal features hereafter referred to as PF1, PF2 and PF3.

PF1 appears to consist of a series of three sub-circular cell type structures and a longitudinal pit or passage built in close proximity to the structures. Existing tumble and morphology between the cells suggests some linkage, but if this exists, it is not clear. The main feature, and the most complete on the

whole site, is a corbelled cell with c30% of its roof intact. It measures c1.5 x 1m internally, with 1.2m high walls and a floor level which is c0.8m below the surrounding ground level. The upper walls and roof are constructed of single skin stone rubble, whilst those below ground may be more substantial. The entrance appears to face W. This is probably the most complete cell of this type on the whole of Uist.

Cell 2 is slightly larger and is similar in form and construction, but without any roof. There may be the remains of an entrance passage to the W. Unfortunately, only around two courses of stonework remain in the walls, so it is not possible to determine if these walls were corbelled. However, there is a good deal of stone tumble within, and this may represent the collapsed remains of a roof.

To the SW of Cell 2 lies the remains of what appears to be a boulder wall, which seems to run a few metres in a SE direction before merging into a low bank. Together these seem to form a rectangular enclosure which may link through into PF3 to the E.

Cell 3 is less distinct and includes two larger stone slabs, both lying on the ground and partially buried. It is possible this is not a cell but a continuation of the linear feature immediately to the NE, which could represent the end of a passage. This is a well-built structure and has walls on all sides with up to four courses of stone still extant. The N end of the feature has a clearly defined end; however, there are hints that at its S end it may continue through to Cell 3, via a route now blocked with fallen or blocking masonry. This linear feature has similarities to others on the site especially in PF2.

PF2 includes the two most enigmatic features on the site: a saddle quern with a perfectly fitting rubbing stone in place; and a well-preserved stone-lined passageway, which is at least 6m long. Alongside these are three or four stone-lined cells plus at least four pits. The diversity of these structures may mean they aren’t connected, but their proximity suggests they may be associated.

Cell 1 shares some characteristics with the corbelled cell in PF1 in terms of its size (c1.5m diameter internally), shape and construction. It is also corbelled to a degree with its walls standing to seven courses high at one point. Apart from a single slab, the floor is clear of tumble. The passage connected to it is an impressive and striking feature being well preserved along a good part of its length with walls c0.7m apart and up to 1.2m tall, with up to seven courses of intact stonework. Around 3m from the cell the northern side wall is balanced right on the edge of the hilltop and at some time in the past a section of wall has fallen away to the valley floor below. Along the passage floor lie up to seven stone slabs, which may have formed the original roof or opening lintels. Where the passage joins Cell 1 a discontinuity in the stonework suggests an opening has been deliberately blocked. At its E end, the passage may fork with one arm continuing as a possible sunken path for c8–10m to the E. Just after the fork a third c3m long arm turns to the S, and although this has less well defined sides, some stonework is still intact. A further large roofing or lintel slab lies near its end.

Around 3–4m S of the long passage in PF2 lie a second set of similar features, although these two joined cells, 2 and 3, which are again below ground level, may have originally been entered from the E by a much shorter passage. However, as the passage entrance, passage/Cell 3 opening, and Cell

2/Cell 3 opening are blocked this interpretation remains speculative. The opening between Cell 3 and the short passage has a possible opening jamb or pier on its N side, with the opening blocked with a substantial lintel or roofing slab, which appears to have been deliberately placed.

Four pits are located between the two sets of conjoined cells and passages. All four take the form of inverted truncated cones c1.5–2.5m in diameter at ground level and c1m deep with some stonework visible in their side walls. Three of the pits are closely grouped and may be connected by passageways.

A saddle quern complete with an in place rubbing stone was recorded lying on open ground in the centre of PF2. The quern which measured c550 x 400mm was clearly not in situ.

Of the three principle features, PF3 was perhaps the hardest to interpret. It includes a mass of loose stone much of it unlinked to any particular structure, and a large recumbent slab c1.7 x 1.0m which is buried beneath thick heather. There is a second smaller slab lying c1m to its S. The clearest feature of PF3 is Cell 1, which has sections of walling on its W and NE sides and an internal diameter of c4m, making it the largest cellular structure on the hilltop.

A straight line of stonework on its SE side has changed the original circular plan to a D-shape. A possible passage, c0.7m wide by 3–4m long appears to extend NW from the cell. A doorway between the cell and passage was formed on its E side with a jamb or buttress of three stones that have now fallen. At the N end of the passage a depression in the ground appears to connect with a part stone-lined pit nearby. Finally on the eastern side of PF3 lies a curved line of what appear to be deliberately placed stones c3m long, 1m wide and 1m high of indeterminate purpose, and a few metres further E another large pit, which was completely concealed by heather.

3D models and plans were created of these features and these can be accessed via our website. Website: http:// uistsummerwine.weebly.com/eaval---the-wild-side.html

Roger Auger, David Newman and Simon Davies

(Source: DES, Volume 16)

What may be a fort, two souterrains, and as many as eight other structures are situated upon a rocky knoll on the E flank of Beinn na h-Aire. The summit of the knoll, which is defined by cliffs to the N and S, is a roughly trapezoidal area measuring approximately 20m from N to S by at least 20m transversely, delimited on the E, S and W by a denuded wall (see plan, 1). The wall survives as a single course of boulders with a stony spread behind, measuring at least 1m thick and 0.3m high, which has been carefully set on to bedrock wherever possible. There is no formal entrance.

There are two concentrations of rubble within the uneven interior. One broad band on the NW includes at least three drystone cells (2-4), two of which appear to be interconnected. They are built of coursed rubble walls standing up to 0.6m high. In addition, a well-preserved corbelled hut (5) with a W-facing entrance is situated near the centre where it seems to overlie earlier features. It is oval on plan and measures 1.5m from E to W by 1.3m transversely, within a drystone wall at least 0.5m thick standing up to 1.3m high.

A second broad band of rubble on the E includes at least five structures, which may post-date the construction of the boulder wall. These include the two souterrains (6 and 7), a well preserved corbelled hut (8), a sub-rectangular building (9) and the footings of another hut (10). The northern souterrain, which measures at least 7.3m long and 0.7m wide, is formed from parallel drystone walls 0.3m thick standing up to 0.9m high. Its S wall is revetted into the slope, but its N wall is freestanding and at least three lintels lie in the passage between them. The E end of the souterrain terminates in an oval cell, which has partly collapsed down the steep slope, while the W end is overlain by the well preserved remains of the corbelled hut (8). This measures 1.8m from N to S by 1.3m transversely, within a wall up to 1m in thick and 1.3m high, with an entrance on the N. The southern souterrain, which measures at least 4.5m long and 0.7m wide is formed by a drystone wall on the SSW standing up to 1m high, but its NNE side is less well defined and its WNW end has been disturbed. A tenth structure is represented by a small arc of walling just outside the fort wall to the SE (11).

There are also at least four shallow pits within the N half of the fort and one just outside it on the NE. These may represent a primitive form of excavation rather than stone quarrying. Such activity may also explain the condition of the souterrains, which appear to have been cleared out and their lintels robbed. A saddle quern (NF95NW 27.01) with its rubber is situated on the ground surface between them and it is possible that this was discovered at the same time.

This site was discovered by Nick Ingeldew on 18 November 2014.

Visited by HES Survey and Recording (GFG, AW, AGCH) 20-21 April 2016

Surveyed with dGPS and at 1:250

HES surveyed Beinn na h-Aire fort between 20-21 April 2016 with plane-table and self-reducing alidade at a scale of 1:250. The final drawing includes amendments from dGPS data that was collected during the survey.