North Uist, Baleshare, Sloc Sabhaidh

Kitchen Midden, Ceardach Ruadh, Baleshare.

[NF76SE 9] About ½ mile north-west of Teampull Chriosd, on a high sandhill near the sea-shore, is a kitchen midden with some loose stones on the summit, perhaps the remains of a hut circle. Unornamented pottery, pumice, bones of animals and shells of winkles and limpets, the former most numerous, as well as a few pieces of iron, were noted. A large block of whale bone 9 ½ inches by 9 ½ inches by 7 inches, with two holes on top near one side 2 ½ inches and 1 5/8 inches in diameter and 2 ½ inches and 1 ½ inches deep, was found and retained for the National Museum. Human burials and pins of bronze and bone have been found here, as well as flints and bits of deer horn.

[NF76SE 19] About ½ mile south of above are extensive kitchen middens at a place called Sloc Sabhaidh. Burials have been found here, also plain pottery and animal bones, winkles predominating among shells. At one place cockles were the only shells. Two good selections of deposits were seen: one to the north, 1 foot thick, showed three layers of winkles with sand between, and another at the south showed a thickness of about 2 feet.

RCAHMS 1928, visited 7 September 1914.

OS map: North Uist xxxix (unnoted).

NF 7823 6085 A site on the W coast of the island of Baile Sear was uncovered and partially damaged in the storm of January 2005. The site had been monitored and recorded by the Access Archaeology group of North Uist, with the support of the SCAPE Trust’s Shorewatch Project. An excavation was undertaken in August 2007. Two trenches were opened on the beach in front of the site, in order to further investigate structures and deposits revealed during an evaluation in 2006.

Trench 1 was at the northern end of mound, and contained the western half of wheelhouse. This had been constructed within a large pit, which was lined by a single skin of undressed stonework, followed by an inner skin of coursed blocks. The wall was corbelled inwards, and was preserved to a maximum height of 1m. A crushed pot was found beneath the portion of the wall that was excavated to its foundations. Floor deposits and

two freestanding masonry piers survived in the northern part of the wheelhouse. The southern part of the structure had been modified by removing the piers to their footings and dismantling the exterior wall. This was then rebuilt half a metre to the N, thus slightly reducing the interior dimensions of the dwelling. The trench that had contained the primary wall was left open for a time, filling with midden deposits. An arc of stonework was also built inside the wheelhouse, concentric to the outer wall. After the structure had been abandoned, it filled with windblown sand. Several deep pits were then dug through these deposits from above. These penetrated the Iron Age floor in the centre of the wheelhouse, and truncated the two surviving piers. Finds, which are typical of later prehistoric domestic sites in the Western Isles, included quantities of animal and cetacean bone, worked antler, coarse stone tools, and Middle Iron Age pottery. Fragments of saddle and rotary querns, and a copper alloy ring, were also recovered.

Trench 2 was positioned at the southern end of the exposed features and was c25 x 10m in size, the main axis parallel to the beach. It was positioned in order to gain a cross-section of an eroding mound with walls, midden, bone and pottery spilling out onto the beach, as well as to investigate underlying deposits below the shingle. Within it a series of adjoining sections of dry stone walling were found which did not readily conform

to any known typological form, other than one corbelled cell which looked typically Iron Age. These parts of structures were all physically related and built into each other in a succession, the earliest of which seems to have been the orthostat-based curving wall to the N of the trench, long noted at the site by local people.

Various occupation deposits were excavated from within these walls, although nothing which could be confidently described as a floor layer. Midden material surrounded the structures to a depth of c0.40m. All other deposits consisted mainly of sand. At the southern extent a cist-like bowl (c1m wide and deep, orthostatic sides with overlying horizontal courses, all covered by large flagstones) was below the level of the other structures but was cut in from a contemporary context. It had been converted by secondary walling into a structure of unknown use, resembling something like a corn drying kiln in plan, with an adjoining stone

built flu- like feature which confusingly sloped downward into the bowl (unlike that required for a kiln).

Underneath these structures there was at least 1.5m of archaeological deposits. Many features were investigated, such as postholes, stone sockets, dog and other, mixed animal burials (pig, dog, sheep/goat), and large flat-bottomed pits of unknown use. Some of these were cut from the level of layers of peat ash and charcoal which were as much as1.2m below the uppermost structures, while others may have been cut from the occupation level of the structures, now lost due to the eroding profile of the beach.

Hammer/course stone tools were found in abundance all over the trench, as was worked, butchered and unworked bone, IA pottery and a small quantity of craggan ware. Many (c12) saddle querns have been found over the years in the immediate vicinity, including one noted by Beveridge (1911, 229) and four broken examples were found this year, together with a fragment of rotary quern. A beautiful polished ring-shaped bone bead, three bone needles and a bone spout or possibly mouthpiece were amongst the most eye-catching finds.

Archive to be deposited with RCAHMS upon completion of fieldwork.

Funder: Historic Scotland, the SCAPE Trust.

NF 782 608 An Iron Age openwork mount was found casually on the beach at Baile Sear, near an eroding settlement site (NF76SE 19). The copper alloy mount has a loop for attaching a strap, and bears openwork decoration with a variant ying-yang motif. Parallels suggest a date of c200 BC–AD 200.

Claimed as treasure trove (TT 37/07) and allocated to Museum nan Eilean.

F Hunter 2007

NF 782 608 Salvage excavation of a wheelhouse partially revealed in 2007 at the northern end of an eroding mound continued from 13 September–5 October 2008. Excavation was undertaken by the Access Archaeology group of North Uist with the support of the Shorewatch Project.

A trench (c15 x 7m) was positioned over the trench from 2007. It was extended c1.85m further E in response to erosion of the coastal dunes that delineate the limit of excavation on the landward side. Extension of the trench E confirmed that the wheelhouse and associated archaeological deposits continue beneath the coastal dunes.

Some of the exterior wheelhouse walling on the W edge of the trench, identified in 2007, had suffered significant erosion. Deposits in this area had also been severely truncated. A number of small pits were excavated and contained semiarticulated animal remains and complete ceramic vessels. The stratigraphic context of these heavily truncated pits was difficult to ascertain due to the severe erosion. They were

probably cut beneath the wheelhouse wall and therefore represent some of the earliest deposits on the site.

The excavation confirmed that in the southern area of the site a section of the outer wheelhouse wall, including

one of the piers, had collapsed in antiquity. It is possible that this collapse was due to the inherent instability of this wall, which had been built above a large pit. The collapse of the wall appears to have marked a period of abandonment, long enough for the collapsed masonry to be covered with a deposit of windblown sand.

In the southern part of the trench, a series of deliberate deposits that were stratigraphically earlier than the collapsed masonry were revealed. These deposits were above what is presumed to be a floor surface relating to the occupation of the wheelhouse immediately before its disuse. The surfaces were not excavated. A rotary quern with a single stone placed so as to block the central hole was recorded. Burnt material had been dumped around and above the quern. Above this deposit a low kerb defined an area containing a further

sequence of burnt material including cremated animal bone and a mixture of disarticulated and semi-articulated animal remains. A lower human mandible had been placed in these deposits. The mandible showed significant tooth wear on the left side.

A later wall was built on the inside of the wheelhouse to replace the collapsed section of walling. The extension of the excavation trench to the E revealed that this later wall was similar in construction to the original wheelhouse wall. The wall was corbelled inwards and survived to a maximum height of nine courses, 1.49m. Like the earlier wheelhouse structure, this wall was built of large blocks of gneiss, maximum size 0.80 x 0.56 x 0.22m, with flatter blocks for the inner facing and rougher stones for the outer face. A former pier and collapsed masonry from the earlier wall were incorporated into the foundation. At the edge of the N/S section this wall included a small niche or alcove c0.2 x 0.18 x 0.3m. The later wheelhouse wall was built without

reconstructing the collapsed piers. The remains of a robber trench for the removal of pier 6 were revealed in the W of the site. During later occupation the structure therefore ceased to be a wheelhouse in the conventional sense of this term.

In the northern area of the site excavation focussed on exploring post-wheelhouse occupation deposits and surfaces. A succession of hearth deposits and structures were revealed immediately outside bay 1 in the central wheelhouse area. These included two rectangular stone-kerbed hearths with stone bases, and a rectangular hearth with a fired clay base. The clay base of this hearth had been incised with a cross,

apparently using three finger marks prior to firing. A similar fired clay hearth base with incised decoration was found in post-Atlantic roundhouse phases at Berigh, Lewis (Harding and Gilmour 2000). The hearth structures at Sloc Sabhaidh were positioned in the same area of the wheelhouse but the alignment altered slightly from E/W to NE/SW. Excavation of occupation surfaces in bays 1 and 2 revealed a sequence of stone kerbs which altered the way in which these spaces were delineated over time. Deposits were found above a layer of grey sand that appeared to distinguish between the primary wheelhouse and post-wheelhouse occupation.

Finds, were typical of later prehistoric domestic sites in the Western Isles, and included large quantities of animal bone, some cetacean and fish bone, worked antler, coarse stone tools and large amounts of Iron Age pottery. The most notable finds were two almost identical bone combs, several bone needles, a fishbone pin, and a worked scallop shell.

Rebecca Rennell (The SCAPE Trust), 2008

NF 782 608 Further excavation was undertaken 13 September–5 October 2008 in Trench 2, 50m N of trench 1

and in the same position as trench 2 in 2007 (DES 2007). The trench was c20 x 7m with the main axis parallel to the beach. As in 2007, it was positioned to cover the largest area of eroding features and to gain a full cross-section of an eroding mound containing structures and midden. This mound is being eroded by a shingle beach which is advancing inland, truncating deposits at a 30°–45° angle.

What had been ‘a series of adjoining sections of dry stone walling which did not readily conform to any known

typological form’ in 2007 resolved itself into the truncated remains of a wheelhouse the outer wall of which had

(probably) collapsed around a pier and had subsequently been rebuilt closer to the centre of the building. While it seems highly probable, conclusive proof of this interpretation awaits further erosion of the mound.

Although perhaps only a quarter of the circumference of the full house remained (diameter estimated at c12m), a full sequence from pre-wheelhouse levels all the way through to the re-build was preserved. The wheelhouse was built upon a level area of clean yellow sand, although the entrance passage, guard cell and structure 4 (see below) were laid onto or cut through pre-existing midden. No evidence for deliberate levelling was seen, although this seems likely. To the N of the trench, the sand contained a flat-bottomed c0.9m wide trench-like feature with a clean, slightly darker fill which did not respect any alignment of later walls, and was interpreted as pre-wheelhouse. A c0.4m diameter pit with undercut edges, c0.3m deep, was cut through the edge of this trench, almost directly under a wheelhouse pier. It seemed to have been subject to in situ burning. It is possible that this feature is related to the wheelhouse construction, perhaps during ground preparation. Partly over it a shallow, linear cut was made, in which a dark fill and then the first course of a pier were laid. The first course of slabs had smaller packing stones at the sides and/or corners which probably acted as levellers. The sand beneath these deposits was very compacted. Nearby another pier demonstrated a similar sequence (and levelling stones) but was cut directly into the yellow sand. After this a layer of fine-grained sand

appears to have been deliberately spread, as the outer wall of the wheelhouse sits on top of it, with no construction cut. Only two piers remained, one extant to 1m and the other to only two or three foundation courses, with an internal dividing wall between them in a similar spoke-like alignment. The wall was constructed of orthostatic slabs, making it a proper pier, and like the outer wall it was laid on top of the

fine-grained sand. Between this wall and the southern pier a great deal of burning had occurred. Between this dividing wall and the N pier, and to the N and S of each pier, were compacted sandy layers interpreted as floors.

The floors contained finds such as pottery and worked bone, including what appears to be a bone brooch or pendant. Overlying the sand layers was a slimy black organic-rich layer, containing pottery shell and bone (some worked). This layer was not compacted. Directly over it, to the N of the southern pier, was a large amount of tumbled stonework which had been re-built to create a new wall face, presumably concurrent with the original. This new wall abutted the southern pier, the northern pier having collapsed and been rebuilt into it.

To the S of the trench, only one side of the original entrance passageway remained. This contained the small corbelled cell (the ‘Guard Cell’) described in 2007, and led to the subterranean structure tentatively interpreted in that year as a corn-drying kiln. This ‘structure 4’ was further excavated this year and discovered to have been substantially modified at some point in its use to create the approximate plan of a corn-dryer. The original structure was almost egg-shaped, and built for most of its length from orthostats cut c0.8m into midden

material overlain with two or three courses of horizontal slabs. At its northern, thinner, extreme these were missing, suggesting either an opening or robbing. Its floor was a layer of peat ash in the midden, thinning towards the wider S end. The primary deposit was a large stone in the middle of this S end. The presence of a piece of pottery and darker sand (not midden) immediately beneath the stone suggest it may have been placed in a cut through the peat ash, although no distinct cut was seen. In the N end a rough pile of beach cobbles was mixed with a cow horn, cow and possibly other animal bone and a large concentration of pure white quartz

pebbles as well as two hammer stones. Two rough parallel walls were superimposed on these deposits to create a narrow trench running from the N to a smaller circular bowl at the S end, ie the plan shape previously thought of as the possible corn-dryer. It has subsequently been noticed by the author that the original structure 4 here is very similar to structure 5 at the Cnip wheelhouse, Lewis (Armit I 2006, 68), and is similarly situated in an area close to the entrance.

Archive: RCAHMS (intended)

Funder: The SCAPE Trust

Ian McHardy (SCAPE Trust), 2008

NF 7823 6085

The excavation of the eroding wheelhouse at Baile Sear continued from 27 August–3 October 2010. Excavation was undertaken by the North Uist local archaeology

society, Access Archaeology and SCAPE. In addition to the great support from local volunteers, the excavation also hosted visits by students from Taigh Chearsabhagh, who used the site as inspiration for their artwork.

This final season completed the excavation of the half of the wheelhouse that lies closest to the coast edge. The other half of the wheelhouse remains within the dune, but is extremely vulnerable to future erosion. The excavation confirmed that the wheelhouse had collapsed and been rebuilt in antiquity. The rebuilding changed the shape and form of the structure, and the southern part was deliberately backfilled with sand. The sand was held in place with upright stone slabs positioned in the area of the hearths described in 2008 (DES 2008, 181–182).

A series of floor deposits relating to the period prior to the collapse of the building were excavated. Within and below these were numerous large and small pits. The greatest concentration of pits was in the central area and many were cut against the ends of the piers. Most pits contained large amounts of burnt and unburnt animal bone, and many also contained pottery. One pit at the end of a pier contained an almost complete crucible. Analysis is ongoing and results will be conveyed in the next edition of DES.

Funder: Historic Scotland

Archive: RCAHMS (intended). Report: The Scape website (intended)

T Dawson 2010

Site covered in shingle and a freshly exposed stone wall on far edge.

Visited by Scotland's Coastal Heritage at Risk (SCHARP) 11 January 2014