Craigpark House And Souterrain Excavation

THE EXCAVATION OF A HOUSE AND SOUTERRAIN AT CRAIG HILL, CO. ANTRIM1 by D. M. WATERMAN INTRODUCTION. DURING the process of stripping top-soil at the quarry owned by Mr. George Peden at Craig Hill, near Bushmills, early in 1954 the workmen partially uncovered a stone-built souterrain together with the fragmentary remains of an associated surface structure. The structures were preserved with commendable care by the foreman at the quarry, Mr. James McArthur, and the discovery was reported to Mr. A. McL. May, who subsequently informed the officers of the Archaeological Survey and visited the site in their company, Immediate excavation was impracticable, but through the efforts of Mr. McArthur the site was left undisturbed until later in the year when the writer completed the investigation of the structures on behalf of the Ancient Monuments Branch of the Ministry of Finance. ’ Craig Hill rises to a height of 355 ft. O.D. some two miles S.E. of Bushmills (O.S. 6 in. Antrim sheet 7). The site is on the northern slope of the hill a few yards south of the road which branches towards Drumyaran Bridge from the BushmillsBilly road; it is situated opposite the junction of a side road leading north towards the group of buildings called Turfahun. No indications of an earthwork enclosing the excavated structures can be seen and the first (1833) edition of the O.S. 6 in. map records no feature of this nature. It appears probable, therefore, that the souterrain and associated house comprised an open settlement, lacking the protection of any form of defensive enclosure. THE EXCAVATION. The House (fig. 1) -The surface structure associated with the souterrain was a timber-framed building, trapezoidal in plan with a central hearth, which may be identified as a house. It measured 12 feet in width north to south, varying in length from 18 feet on the north to 14feet on the south, and was entered by a stone-paved porch on the east. Before erection, a level site was prepared by cutting back into the slope of the hill, the rock-cut face of this excavation forming the upper limit of the terrace on which the house was situated. Outside the south (up-hill) wall of the structure a gutter, edged and paved in stone, was provided to intercept the surface water which otherwise would have entered the building; this feature was 12-18 ins. in width and a single stone-course in depth, embracing the length of the house and continued westwards to the face of a stone-built revetment retaining the stone and soil covering of the souterrain. On the east a stone curb, continuous with, and set obliquely to the north side of the gutter, extended to the stone paving of the porch and served to deflect the water away from the main building. ` The sub-soil comprising the floor of the dwelling-house consisted of friable, decomposed rock on the east and of boulder clay on the west, the junction of rock and clay occurring on a line extending north to south across the building east of the central hearth. The angles of the structure were indicated by post-holes, 6-12 ins. in depth, either rock-cut or excavated into the clay. A pair of these occupied the south-west angle and a further post-hole (no. 4 on plan, fig. 1), presumably 1

Ulster Journal of Archaeology 3rd series volume 18 1955

CRAIG HILL connected with a door providing access to the souterrain, was situated in the centre of the west side. At a distance of 7 feet east of the main structure an additional pair of post-holes (nos. 7 and 8), 3 feet apart and 8 ins. in depth, had held the timber supports of a porch protecting a door in the centre of the east wall of the house. The porch (pl. IX) was stone-paved, the paving consisting of flag-stones neatly placed and jointed with smaller stones filling the interstices between the larger slabs. Approximately in the centre of the house was a basin-shaped pit, about 9 ins. in depth, which, although previously emptied of filling by the quarrymen, still contained a small amount of dark sooty earth; it presumably served as a hearth. At a distance of 2 ft. 6 ins. south of the hearth was a small post-hole (no. 9), about 6 ins. in depth, around the mouth of which were lying a number of small packingstones; the significance of this post-hole is not clear. All sign of occupation within the house had been removed prior to excavation, save for the fillings of dark soil and charcoal within the post-holes, but over the paving of porch and gutter was a considerable amount of blackened soil and fire-reddened clay. The nature of the walling of the house is unknown, but a light construction seems to be indicated and it is possible that the clay represents the burnt debris of a daub super-structure. The adoption of a trapezoidal plan for the house is curious, but since careful search failed to reveal a post-hole in the appropriate position for the south-west angle of a rectilinear arrangement the outline of the structure, indicated by a dotted line in fig. l, is presumably correct. The plan adopted, therefore, left a triangular area, presumably unroofed, between the house and the stone revetment fronting the souterrain, but the presence of a post-hole (no. 10), 6 in. in depth, together with a couple of adjacent stake-holes, 2 ins. in diameter and 7-9 ins. deep, suggests some form of covered way between house and souterrain. The Souterrain.-The souterrain was undoubtedly in position from the first; the trench excavated to receive it was found to encroach upon the area of the house, and the post-hole (no. 5) at the north-west angle was dug in the loose filling replaced in the trench after construction of the souterrain had been completed. The souterrain on completion was covered with the surplus soil and stone excavated from the trench, this material being retained in position towards the house by a revetment of large stones on edge (pl. IX), which were shallowly bedded and rose about l ft. 9 ins. above the old ground level. At the mid-point of the revetment was a gap, inside of which was a stone-paved step, 6 ins. in height, providing access to an open, elbow-shaped passage, revetted in dry-stone walling (pl. The floor of the passage, cut in the boulder clay, descended steeply to the floor level of the souterrain, 4 feet below the old surface; at the foot of the slope was a lintelled opening (pl. X), 2 ft. 3 ins. in height and 1 ft. 6 ins. in width, leading directly into the souterrain. The souterrain passage assumed a quarter-circle on plan, curving round towards the north for a total length of 40 feet. It was revetted in dry·stone walling and covered by lintels, the walls sloping slightly inwards towards the roof; the passage was 2 ft.-3 ft. 6 ins. in width and 3-5 ft. in height. At about a third of its length from the entrance the floor of the souterrain dropped abruptly by means of a rock-cut slope, this descent, accompanied by a corresponding lowering of the roof, presenting a considerable obstacle to the negotiation of the passage. Thereafter,

D. M. WATERMAN the floor sloped gently to the end of the structure although the roof, save for a sudden drop to within 2 feet of the floor at one point, continued at a constant level, thus providing a continual increase in the height of the passage. Within the souterrain the only structural feature requiring mention was a low lintelled recess, about 1 ft. l0 ins. in height, situated at floor level at a distance of 5 feet from the north end of the passage; the back of the recess was formed by the rock face of the trench in which the souterrain was built. Overlying the surface of the sloping approach to the souterrain, and covering the floor of the passage as far as the rock-cut step was a considerable deposit of dark soil with some charcoal, containing numerous sherds of "souterrain-type" pottery, the bulk of which contributed to a restoration of the two vessels illustrated in fig. 2, 1, 2. The only other find associated with the souterrain was a bronze ring-headed pin found in the soil and stone covering of the structure.

CRAIG HILL THE FINDS. Pottery (fig. 2, 1-10).-The pottery recovered from the souterrain is all handmade; it is of reddish- or buff-coloured clay, normally with grey core and sometimes underfired on the inside surfaces which are then similarly grey in colour, The paste is usually rather soft, the surfaces easily marked with the finger-nail; gritting is fairly fine but includes small pebbles up to ⅜ in. diameter. Numerous mica particles are present in the clay. The outside, infrequently the inside, surface of the vessels shows impressions of vegetable matter, probably straw or grass, and this is most noticeable on the underside of the base which is heavily marked all over with such impressions. Finger-smoothing of the surfaces is sometimes apparent and fingermarking is frequently to be seen at the internal junction of wall and base. The form of the pottery calls for little comment. The normal shape is demonstrated by the reconstructed vessels, nos. 1 and 2, which are saucepan—shaped pots having a rim ornamented by close-set finger-impressions; when this decoration is absent, the rims are flattened or plainly rounded (nos. 4-6). A pot with S-curved profile is present in one instance (no. 3). Bases (nos. 7-10) are flat, with gently. rounded or slightly angular junction with the side of the vessel. Bronze (fig. 2, 11).-Apart from the pottery, the only object found was a bronze ring-headed pin from the stone-and-soil covering over the souterrain. The stem of the pin is circular in cross-section but flattened at top and bottom. The top is simply turned over to accommodate the tenon-like attachments of the plain ringhead, and is here ornamented by a number of parallel incisions; lower down, at the junction of the rectangular with the circular stem, are a number of transverse mouldings. At two-thirds of its length from the head the stem is shouldered, expanding slightly before narrowing towards the point, which now is missing. The ring-head is oval in shape and of circular cross-section. SUMMARY. The excavation is of interest in demonstrating the intimate association of a souterrain with a surface structure of domestic character. It also serves to emphasise the existence of such structures independent of a fort, or rath, a phenomenon already remarked in the case of another Co. Antrim souterrain, at Harryville, Ballymena (Ulster J. Archœol., 13 (1950), 53). In view of the discovery from time to time of apparently isolated souterrains, it may be urged that in future any excavation undertaken should be extended to determine the presence of an associated surface dwelling at ground level. In attempting to define the period of occupation of the Craig Hill site, the absence of closely datable relics leaves the matter in doubt. The ring-headed pin belongs to a group that broadly can be dated to the last four centuries of the first millennium A.D.; simple pins of this type occur at Ballinderry Crannogs No. 1 (10th century) and No. 2 (c. 8th century) and also, not closely dated, at Lagore Crannog (late 7th to 10th centuries). The greater incidence of these pins in the 9th to 10th centuries is emphasised by their not infrequent occurrence in graves of the Viking Age and it is to this period that the Craig Hill settlement may most probably be attributed.