2003:891 - CARBURY HILL, Carbury, Kildare

A proposed waterworks scheme (Wellfield Development; see No. 976, Excavations 2003, 03E0208) by Kildare County Council includes the construction of a reservoir on Carbury Hill, plus associated access routes/pipeline corridor. The work was carried out in April/May 2003. Carbury Hill (142m OD), well known in the archaeological record, is located near the reputed source of the River Boyne. The most important body of evidence relating to the hill itself comes from excavations carried out there in 1938 by G.F. Wilmot (Three burial sites at Carbury, Co. Kildare, Journal of the Royal Society of Antiquities of Ireland 68, 130–42). All three burial sites consisted of earthwork barrows (SMR 8:3, 8:4, 8:5). No definite dating evidence was forthcoming, but most commentators suggest a fairly broad Iron Age date bracket extending down to c. AD 400. Other monuments in the immediate vicinity include Carbury Castle; the surviving structure is largely Jacobean but the founding of the castle can be traced back to the 13th century. A 19th-century church and school are located adjacent to farm buildings near the foot of the hill.

The proposed reservoir will be located just south of the hill summit. There are a number of monuments in the immediate vicinity, including the excavated SMR 8:3. Two alternative locations for the reservoir were investigated. In both cases, the access route/pipeline will commence at the Carbury–Broadford road and run for some 480m west-south-west to a field boundary just below the hill summit. The proposal for Option 1 (West Reservoir) is that the access road and pipeline would then follow the 125m contour around the southern side of the hill and terminate on the south-western side of the hill summit. For Option 2 (East Reservoir), the access road and pipeline would continue further uphill to the south-eastern portion of the hill summit. A 4–6m-wide access way/pipeline corridor has also been proposed for the eastern flank of the hill.

The hill is in pasture. Testing was carried out in four fields and was restricted to areas that might be impacted on by the proposed development. Six test-trenches were excavated. These followed the line of the pipeline corridor, traversed the two possible locations for the reservoir, and followed the lines of the two alternative ‘link routes’—Trench 1b/1/1a and Trench 1b/1/2—between the pipeline and reservoir.

All the trenches averaged about 1.65m in width. Depth varied according to the overburden of topsoil, the range being from 0.15 to 0.5m. The sod layer averaged about 0.1m in depth and was shallowest on the upper slopes and summit of the hill. With one exception (Trench 1, F1), features of potential archaeological interest occurred only towards the foot of the hill (Trench 1b). Modern/pre-modern debris was relatively rare in most of the trenches, a phenomenon that may be explained by the absence of ploughing. Non-archaeological features were investigated where feasible but features of possible or probable archaeological significance were left in situ and covered with polythene sheeting before the trenches were backfilled. Two discrete areas of modern disturbance were noted. The first lay adjacent to an old trackway at the eastern end of Trench 1. The second was at the eastern end of Trench 1b, where a dwelling had formerly existed.

Trench 1 extended from the Option 2 reservoir site on the hill eastwards to the Carbury–Broadford road—a total distance of some 425m. The trench (subdivided into Trench 1 and Trench 1b) crossed three fields, running at a distance of 3m out from the south field boundaries. Trench 1 was 185m long. It extended from the hill summit at the west to the old trackway at the east. The track is a continuation of the paved road that skirts the church and graveyard and continues to the castle. A 30m-long stretch of the pipeline corridor located to the east of the track was not tested, as the area is presently used to store fodder. The test-trench traversed steeply sloping ground, and ‘horizontal bands’ evident on the geophysics map proved to be stepped bedrock. The exposed rock was soft and, over most of the trench, the overburden of soil was extremely thin—about 0.1–0.15m on average. Some deeper (>0.3m) areas of soil were pocketed between the rock steps, especially at the lower, eastern, end of the trench. There was no perceptible change in the soil profile adjacent to barrow SMR 8:3.

Only one definite feature (Trench 1, F1) was recorded and this lay 25.8m from the east end of the trench. It was a largely natural, pit-like, hollow located at the foot of a bedrock step and neatly sheltered by the topology. The finds and other inclusions in the fill suggested that the feature was a smithing hearth. Its date is unknown at this stage. Further examination of a bronze piece, possibly a haft or binding strap, may provide a clue. A sherd of green-glazed ware of possible medieval date was recovered nearby. Stray finds of medieval pottery and a few animal bones were found further uphill, but no features or zones of archaeology were evident. At the east end of the trench, dump material incorporating brick fragments, etc., is probably associated with the construction of the trackway.

Trench 1b was 240m long. It extended from the field fence east of Trench 1 to the Carbury–Broadford road. From the low-lying eastern end, the ground rises gently at first and then more steeply within the western third of the trench. In the latter area, bedrock lies fairly close to the surface (average depth 0.15–0.2m) and shattered parent rock is common. At the lower eastern end of the trench, the natural was represented by either compact chalky white boulder clay or softer yellow clays, surfacing at a depth of 0.2–0.25m.

Over fifteen features of archaeological or possible archaeological potential were exposed in the trench, most being concentrated on the lower ground. The features were diverse in character and included spreads of archaeological soil, pits, gullies and more substantial but relatively shallow ditch-like features. Without excavation, it is impossible to assess what their function might have been. Definition was often unclear, but some features (e.g. F10, 11, 12 and possibly 9) may be related. The finding of a clay mould (F24) and a flint flake (F10) hints at prehistoric activity, while the presence of green-glazed pottery suggests medieval activity. The NMI register includes a relatively large number of casual medieval/post-medieval finds from the vicinity of the adjacent Carbury Castle. A number of additional features at the eastern end of the trench proved to be modern, or of no archaeological interest.

Trench 2, 110m long, zigzagged downhill from the proposed West Reservoir site on the hill shoulder and summit. On the downhill eastern side, the trench terminated at an east–west field boundary. No definite features were exposed, but a small number of finds were recovered. The soils were noticeably clean and devoid of casual inclusions or bric-à-brac. In the western uphill part of the trench, platey bedrock, or shattered bedrock, lay at an average depth of 0.1–0.15m below the surface.

The only finds recovered in this part of the trench were two fragments of pottery, both possibly prehistoric in date. The pottery came from loose gritty soils at the interface between the rock and subsoil. Further downslope, the trench angled south-westwards and ran down off the hill shoulder. The overburden of soil became increasingly thicker, with the stepped surface of the bedrock dropping from 0.25 to >0.6m below present ground level.

Trenches 3/4, in the East Reservoir site, were a cruciform set of trenches which traversed the most easterly of the two proposed locations for the reservoir. Trench 3 (46m long by 1.6m wide) ran north–south and Trench 4 (32.8 by 1.6m) ran east–west. Both were located at a slight shoulder just below the hill summit and within each the ground sloped away very gently.

No features of potential archaeological interest were uncovered. The stratigraphy was more or less the same throughout both trenches. Any variations noted in the relative depth of soils, etc., was due to gradations in the natural topology of the bedrock. Natural, represented either by bedrock, rotten stone, pockets/bands of yellow boulder clay or very compact pockets of white chalky boulder clay, occurred at a depth of 0.15–0.25m below present ground level.

At the West Reservoir site, two cruciform trenches, Trenches 5/6, were located in the field to the west of Trenches 3 and 4. Trench 5 ran north–south and was 34m long; the east–west-running Trench 6 was 45m long. The stratigraphy was the same as in Trenches 3 and 4, with bedrock/rotten rock being exposed at an average depth of 0.25m. It was deeper (<0.4m) in some places where the surface stepped down and pockets or bands of sticky yellow boulder clay occurred occasionally. The soils were sterile and there were no traces of any features, archaeological or otherwise. A small rim-sherd of possible medieval pottery was found in the topsoil 3m from the west end of Trench 6. Trench 5 was subsequently extended a few metres northwards towards the field boundary, but the stratigraphy, etc., remained unchanged and no finds were recovered.

The results of the testing uncovered over a dozen features of potential interest in Trench 1b but only one (the smithing hearth in Trench 1) outside that area. Some of the features in Trench 1b occurred in close proximity to each other and are likely to be related, if not physically connected. The most potentially interesting features were F9–12, which may be part of a ditch enclosing some internal features. If this were the case, a ring-ditch or barrow would be a likely interpretation, given the archaeological background of Carbury Hill. The association between the possible ditch and the relevant finds – a flint flake, what appears to be part of a bivalve clay mould and the few sherds of prehistoric pottery – is unclear at this stage. A cursory examination of the pottery suggests that it is more likely to be early, rather than late, prehistoric. The mould cannot be earlier than the Late Bronze Age, while the date of the flint flake could span a very broad period. The frequency and randomness of the features uncovered in Trench 1b, and the fact that many continue beyond the confines of the trench, strongly suggests that further examples will occur in the adjoining ground.

Given the proximity of the castle and its long history of occupation, the amount of medieval activity exposed in the trenches was quite low. Apart from the smithing hearth, there were no potential medieval structures and the quantity of medieval pottery would be commensurate with ‘plough soil’ or ‘garden soil’.

Recommended mitigation measures include further archaeological work at the site.

7 De Burgh Road, Dublin 7