2005:1175 - DUNBOYNE: Dunboyne Castle, Castlefarm, Meath

This was the second phase of excavation at ‘Dunboyne Castle’, an 18th-century house and demesne, the first having taken place in 2004 (Excavations 2004, No. 1230). A description of the site, some historical information and the background to the archaeological excavations can be found in the 2004 entry.

The 2004 testing/excavations took place mainly in advance of development and focused on the wider demesne and the gardens at the rear south side of the house. A substantial medieval ditched enclosure came to light on a ridge of higher ground in the latter area. The 18th-century house stands in the south half of the enclosure, with the truncated medieval ditch running under its south-west corner. Altogether a 75m-long segment of the south half of the ditch was excavated in 2004 and the finds indicated a 13th- or 13th/14th-century date for the backfill. A broad 13th-century date was subsequently confirmed by dendrochronological analysis of timbers from the waterlogged base of the ditch. The excavations suggested that there had been very little activity in that part of the site during the period from c. 1300 to 1750, when the present house was constructed. This was surprising, as a substantial building is shown at, or close to, the house on the mid-17th-century Down Survey map.

By autumn 2005, many of the housing units had been constructed and the conversion of the 18th-century house to a hotel and leisure complex was also well advanced. The 2005 excavations, in advance of the construction of a mains sewer pipeline and other more minor service trenches, were concentrated in the northern half of the medieval enclosure. The line of the north sector of the ditch was first traced by opening a series of shallow trenches. On foot of mitigation discussions, those service trenches that could be rerouted were moved further north, outside the enclosure. It did not prove possible to completely reroute the main sewer pipeline, however, and a line of least impact was agreed. This involved full excavation of a cross-section of the ditch at the east (Trench 1, Area C) and an L-shaped linear strip in the interior of the enclosure (Trench 1, Area B/Area B ext.). The discovery of the remains of a Bronze Age ringditch in the latter area resulted in the excavations being extended north of the pipeline trench, but still within the medieval enclosure (Trench 1B, Northern ext.). The remains of a stone building were found on the exterior of the ditch at the east and a limited amount of excavation was also carried out in that area (Trench 1, Area A). Unfortunately, part of the building had already been destroyed by the time excavations commenced. Finally, the initial testing had exposed the remains of revetment walling on the inner lip of the ditch at the west (Trench 2). The pipeline was rerouted to avoid the feature, but a limited amount of further excavation was carried out in an attempt to establish its context.

Excavations took place over a 14-week period from September to December 2005. A subsequent period of monitoring in other areas of the demesne (along the main avenue, in the old stables) did not reveal any further features of archaeological interest. Post-excavation analysis is ongoing at the time of writing.

Prehistoric
The Bronze Age ringditch was located on the ridge summit within the northern sector of the medieval enclosure; the interval between the prehistoric ringditch and the northern ditch of the medieval enclosure was only 4m. The monument was first identified after the removal of a post-medieval (probably 17th-century) metalled roadway (Trench 1B, F9) which sealed it. The forecourt area of the house had been scarped to create a level surface for the roadway and it is quite possible that the ringditch was truncated to some degree during that process. This could explain the flat, featureless and slightly sloping interior. The northern sector of the ringditch had earlier been truncated by medieval activity.

The monument was circular in plan, with an overall diameter of 7.8m and an internal diameter of less than 6m. The ditch had a surface width of 1.05m and was less than 0.94m in depth. It was cut into boulder clay; some patchy traces of the old ground level were recorded on the exterior west side.

Along most of its length, the ditch fill consisted of orange, yellow, grey and brown clays and silt, much of it sterile but with dispersed or, occasionally, concentrated charcoal flecking present intermittently. A small fragment of flint was found in the west sector of the ditch. The remains of a shattered and compacted cordoned urn were found in the eastern sector. The urn had been placed less than 0.25m below the top of the ditch and 0.35m above the base. No definite cut was identified, but the sediments underlying the urn were distinctive and confined to the immediate vicinity. As found, the vessel appeared to have compacted while lying on its side, but whether this was its original position or resulted from falling over was impossible to say. Post-excavation analysis is ongoing; at the time of excavation there were no visible fragments of burnt bone in the soils surrounding the urn.

Additional fragments of pottery were retrieved from the fill in the southern and northern sectors of the ditch. As the latter area was truncated by medieval activity, it is possible that there may originally have been other urn burials along that sector of the ditch.

Early medieval
Most commentators suggest that Dunboyne was probably the site of an early historic fort, but its exact location remains unknown. Dunboyne Castle occupies the only elevated ground in the immediate vicinity of the village. However, no definitive traces of a pre-Norman enclosure came to light during the present excavations. A portion of a lignite bracelet found in drain upcast just outside the Bronze Age ringditch may be of early medieval date. Two other features, as yet undated, could be either early medieval or medieval in date

The first was the remains of a circular double-ring stake-built structure (F324) and the second, which lay immediately beside it, was a kiln (F144). A third feature, a ‘four-poster’, is more likely to date to the medieval period. All three features lay in the west half of the medieval enclosure (Trench 1, Area B).

The stake structure measured 7m in external diameter (outer ring) and 3.2m internally (inner ring) and was respectively made up of 31 and 24 surviving stake-holes, set at average intervals of 0.38m. The individual stakes averaged 0.07m in diameter and had been driven to a depth of 0.15m into an old sod layer. Five other stake-holes were recorded adjacent to, or in the interior of, the structure. No definite entrance was identifiable, but there were a number of gaps, any of which could have been the site of a doorway. The interior was devoid of any contemporary features or surfaces, etc.

The kiln was roughly oval in plan (1.3m by 0.98m and less than 0.25m deep), with the remains of a flue at the south side. The fill layers contained frequent amounts of charcoal and seeds and the south side of the pit was also fire-baked. The only find was a small piece of flint from the flue. The most likely interpretation is that it was a corn-drying kiln.

The rectilinear four-post structure measured 4m by 2.3m in plan and was made up of substantial circular post-holes of comparable size (less than 0.55m in diameter and 0.4m in depth). The post-holes were cut down from the same horizon (the old sod layer) as the stake-built structure, but the ground plans of the two features overlapped. Even if the four-poster was a raised structure (e.g. a granary), it seems highly unlikely that it could have been in use at the same time as the circular building. Either structure could, however, have been contemporary with the kiln.

The medieval enclosure
The east sector of the medieval ditch was similar in most respects to the south sector excavated in 2004. It averaged 8.5m wide by 4.5m deep and was extremely steep on both the inner and outer sides. The lie of the land meant that the inner edge was higher than the outer. For the most part, the ditch was cut through hard stony boulder clays, but the upper outer edge was cut through much softer sandier clays.

The remains of a timber structure were recorded 1m inside the inner edge of the ditch. The outline was that of a four-post structure with linking slot timbers and a few subsidiary angled supports. At this stage in the post-excavation it is unclear whether a) this is the remains of a gate/guardhouse or b) it was part of a box-built palisade. The former seems more likely, although no trace of a bridge was noted within the ditch itself or on the exterior. The footprint of the feature is similar to the palisade uncovered during Alan Hayden’s excavation of the ringwork at Trim Castle. However, taking erosion into consideration, the Dunboyne structure appears to be placed too far back from the edge of the ditch to have acted as a defensive palisade. Further analysis is required before this, or indeed the bigger question—whether the site was a motte, ringwork, or some other class of defensive enclosure–can be addressed.

The remains of well-built revetment walling were exposed along the upper inner edge of the ditch at the west and north. The walling was not excavated and it remains unclear whether or not it was carried the full depth of the ditch. The construction technique was similar in both areas. The upper inner edge of the ditch was first cut back, leaving an open-sided trench, up to 1m deep. The battered face of the revetment wall and its supporting rough mortar bonding were then put in place. The interval between the rear of the wall and the back of the trench was then filled with a looser mixture of mortar and rubble. At the west, the surviving south end of the revetment wall was truncated by 19th-century outbuildings, but the north end terminated at a short return that ran back into the interior of the enclosure. The foundations of a faced opening midway along the surviving portion may be the remains of a chute, or a doorway that gave access to the ditch. A few individual sherds of 13th/14th-century pottery (Leinster cooking ware and Dublin-type fineware) were recovered from the rubble-filled void at the north, and from a construction pit (F26) associated with the return of the revetment wall at west. The ‘revetment’ does not appear to have been a continuous feature, as there was no evidence for any walling along the eastern and southern sectors of the ditch. One possibility is that the walling formed an integral part of buildings that were located immediately behind the defences.

Ditch fill
The ditch was bottomed in the stepped section opened at the east. Two broad horizons were identified (Phases 1 and 2) separated by a distinctive layer of soft yellow boulder clay that occurred about 1.5m below the top of the ditch. At the time of excavation, the boulder clay was considered to be the construction level for the late medieval building described below. Subsequent analysis of the pottery suggests that this might not, however, be the case. The basal fill was waterlogged and four complete ash bowls were recovered, as well as worked and unworked pieces of timber. As in 2004, there were no large structural timbers present. The Phase 1 and 2 fills included dumps of mortar and stone rubble; these horizons were absent from the south part of the ditch excavated in 2004. There were some notable differences also in terms of the finds recovered in both seasons (the pottery from the site has been analysed by Clare McCutcheon and other specialist work is ongoing). The south sector of the ditch yielded a relatively large quantity of metal objects, including a 13th/14th-century seal, iron keys and decorative strapwork. All of the 279 sherds of medieval pottery recovered in 2004 can be classified as ‘local’, with most belonging to the class known as Dublin-type wares. By comparison, the east sector of the ditch yielded very few metal objects, but 5% of the 233 medieval pottery sherds recovered were imports, all but two of those being French in origin. The most interesting piece was part of a jug with a zoomorphic spout, similar to three examples from Wood Quay and a possible fourth from Usher’s Quay. On the evidence of the pottery, the ditch could have been completely backfilled either by the end of the 13th century or at some stage during the 14th. While there is an ongoing debate regarding the possibly longer currency of some of the native wares, in the case of the medieval enclosure at Dunboyne, all the chronologically diagnostic artefacts would fit comfortably into the 13th/14th-century date bracket. However, as there is a 15th-century documentary reference to the granting of monies to build a castle at Dunboyne, the medieval ‘hiatus’ will have to be scrutinised more closely.

Apart from the truncated foundations of five roughly rectilinear features, the interior was devoid of any definite medieval stratigraphy. The rectilinear features presented as sunken trenches (average 0.6m deep) filled with mortared rubble, and may have been foundation platforms for timber buildings. If this was the case, the floors must have been well above the surviving ground level, as no trace of any post-holes or residues of occupation soils were evident. Indeed, less than half a dozen sherds of medieval pottery were recovered from the interior, suggesting that the whole area had been severely truncated. A sherd of late 12th/mid-13th-century Ham Green B ware came from the rubble fill of one of the rectilinear structures (F392). The structure had a peculiar dogleg outline and extended into the interior from the revetment walling on the north side of the ditch.

Late medieval
The foundations of a masonry building F503 were uncovered on the outer edge of the eastern sector of the medieval ditch. The western wall, north-west return, what may be the south-west return and a disconnected fragment of the south wall were recorded. The external dimensions of the building were c. 8m north–south by at least 8.65m and the walls were less than 1m thick. The north and south walls did not survive above foundation level. The west wall was reduced to contemporary ground level. Its battered outer face was carried down into the ditch for a distance of 0.6m. The wall ran at a slight angle relative to the ditch edge, with the result that the northern portion sat on ditch fill and the southern portion was set into an L-shaped cut in the outer side of the ditch. A slope-sided trench running parallel to the rear of the wall is probably the remains of the eroded outer edge of the ditch. The eroded edge was levelled up with large stones set an angle. Blue/white pottery found amongst the stones has been identified as part of a Malling jug, an import from the Low Countries, probably Antwerp (Clare McCutcheon). The origins of Malling ware go back to the mid-16th century and this dating is confirmed at Dunboyne by the finding of two closely associated mid-16th-century coins.

The interior of the building did not survive; this part of the site was also scarped when the post-medieval roadway (F9) was constructed.

Post-medieval
Apart from the roadway (F9), the 17th to mid-18th-century activity at the site was represented almost exclusively by pottery (Merida ware, Frechen ware, Westerwald, North Devon gravel-tempered and sgraffito ware, and tin-glazed earthenware). The majority of the 17th-century pottery came from Area A on the exterior of the ditch and is likely to be associated with the occupation of building F503. There is a strong possibility then that that structure is the remains of the building shown on the Down Survey map of c. 1660. Only a small quantity of 18th/19th-century pottery was found at the site. Most of this came from the ruins of demolished outbuildings located to the west of the present house.