2003:0017 - KILROOT: The Bishop’s Palace, Kilroot, Antrim

The Bishop’s House and bawn at Kilroot, near Carrickfergus, Co. Antrim, are located within an earlier ecclesiastical site whose origins probably date to the Early Christian period. The evidence for early ecclesiastical activity at the site includes a bullaun stone (now removed and replaced by a cement replica), a holy well recorded as being ‘an unlined spring [located] a few rods south of the old graveyard’ (O’Laverty 1884, 82; Cordner 1940, 158–9) and two Anglo-Norman grave slabs (Bigger 1895–96; Cordner 1947). The ruins of the site’s medieval church were recorded on the second-edition OS 6-inch map (1857) but are no longer visible.

The construction of the house and bawn, built on the site for the Bishop of Down and Connor, dates to the early 17th century (Brett and O’Connell 1996, 19). The house is now a three-storey ruin, while only the northern and eastern elements of the associated bawn and part of a possibly later bawn flanker survive. The area enclosed by the bawn included part of the earlier ecclesiastical site’s graveyard and, by the early 18th century, use of the area immediately to the east of the house, which included the ruins of the medieval church, had reverted back to a graveyard. The western edge of this near-rectangular graveyard is partly defined by the eastern gable wall of the Bishop’s House, whilst its northern and eastern edges are defined by the bawn wall. The southern edge of the graveyard was defined by a retaining wall of uncertain date which abutted the south-eastern corner of the Bishop’s House. The earliest recorded stone in the graveyard dates to 1709. The graveyard is still used by a small number of families and contains an interesting collection of gravestones dating to the 18th, 19th and 20th centuries. The site was scheduled in November 2000.

The southern retaining wall of the graveyard collapsed in November 2002. Carrickfergus Borough Council intend to replace the collapsed wall and the Environment and Heritage Service have undertaken to carry out any necessary archaeological mitigation in advance of this building work. Following test-pitting undertaken by John Ó Néill (see No. 16, Excavations 2003), it was considered necessary to excavate, by hand, the foundation trench for the new wall to the surface of the natural weathered bedrock. The proposed trench was to be c. 29m long, 1.5m wide and located 1m to the south of the line of the collapsed wall.

Excavation took place between 10 November 2003 and 28 January 2004. Following the clearance of relict traces of the collapsed wall, the excavation of a small part of the modern road surface laid by the contractors during the clearance of the wall and the re-excavation of John Ó Néill’s three test-pits, the humic topsoil was excavated. At the western end of the trench the topsoil contained a distinctive clinker-rich lens, which probably represents rake-out from a relatively late, industrial phase of activity in the Bishop’s House.

In addition to the remains of a modern chicken coop, underlying the topsoil was a stratified sequence of deposits which represents at least three phases of a post-medieval garden. The earliest phase of the garden was represented by a series of five truncated, hourglass-shaped negative features which were cut into the underlying graveyard deposits and aligned along the line of the collapsed wall. They are provisionally interpreted as planting holes for trees which were moved once, either closer to or away from the southern wall of the graveyard (the edges between the two cuts for each plant were not recognised during the course of excavation). Apparently contemporary with this possible episode of planting were two small spreads of ‘chalk’ nodules and three other negative features of uncertain purpose. No artefactual dating evidence for this phase of the garden was recovered; however, the fact that the planting holes were aligned on the collapsed southern wall of the graveyard suggests that this earliest recognised phase of the garden does not pre-date the graveyard.

Overlying the first phase of the garden was a series of levelling deposits, which increased significantly in depth towards the eastern end of the trench. These deposits were humic in character, suggesting they were formed from organic-rich soil excavated elsewhere. Overlying them was a gravel path, aligned along the length of the collapsed wall, which ran through the entire length of the trench. Where the path was built over the levelling deposits in the eastern half of the trench, it was built on a foundation of large stones and coarse gravel that had been laid in a deep cut. The exposed, northern edge of the path was lined with a stone kerb and its surface was covered in relatively fine gravel. Originally, the path turned through a right angle at the eastern end of the trench, where it respected the junction of the collapsed southern wall with the north–south-running bawn wall. The gap between the path’s kerb and the two walls was c. 1.75m; this area would presumably have formed a garden bed. In the garden’s final phase, a c. 2.55m-wide breach was made through the bawn wall and the path was extended through this gap. The secondary character of this alteration is demonstrated by the path’s foundation cut and fills not extending through the gap as well as a marked change in the character of the path’s kerb and gravel surface at the eastern end of the trench. The layout of this third phase of the garden is featured on the first-edition OS 6-inch map (1834), suggesting that it survived until at least the 1830s.

Underlying the various garden phase deposits was an inhumation cemetery. The skeletal remains of fifty adults, seventeen sub-adults (children and adolescents) and three infants were excavated. In addition, a large amount of disarticulated human bone was recovered. All but two of the burials were orientated east–west; the other two were aligned north–south (with their heads to the north). The hunched posture of a large number of the skeletons suggests that at least a significant number represent shroud burials. Although the superimposition of burials made the recognition of individual grave-cuts and fills problematic, a relatively detailed stratigraphic sequence was recorded. Provisional analysis of this sequence suggests that the cemetery developed outwards, to both the east and west, from a point c. 9m to the east of the Bishop’s House. Evidence of unusual burial rites includes the two north–south-aligned burials noted above, two skeletons whose grave fills contained a large number of ‘chalk’ nodules (which were possibly intended to form grave-markers), and the careful curation and redeposition of three separate clusters of long bones disturbed in the cutting of graves through earlier burials. In addition, the skull of one skeleton shows evidence of two trauma wounds, apparently inflicted with a heavy blade, one of which had healed and one of which was fatal. Close dating of the burials is not, at this stage, possible. It is reasonable to assume that they pre-date the construction of the Bishop’s House. The evidence for the practice of shroud burial and the recovery of a small number of sherds of souterrain ware and everted-rim ware from the grave fills suggests that a medieval date for the cemetery is most likely.

References
Bigger, F.J. 1895–96 Cross at Kilroot, Ulster Journal of Archaeology (second series) 2, 209.
Brett, C.E.B. and O’Connell, M. 1996 Buildings of County Antrim, Ulster Architectural Heritage Society and Ulster Historical Foundation, Belfast.
Cordner, W.S. 1940 Some old wells in Antrim and Down, Ulster Journal of Archaeology (third series) 3, 156–61.
Cordner, W.S. 1947 Kilroot, Ulster Journal of Archaeology (third series) 10, 71.
O’Laverty, J. 1878 An historical account of the Diocese of Down and Connor, ancient and modern. Volume III. Dublin.

School of Archaeology & Palaeoecology, Queen’s University, Belfast, BT17 1NN