2002:1072 - ADARE: Adare Castle, Limerick

This season’s excavations again relate to structural conservation works at Adare Castle. Dúchas has been undertaking long-term conservation works on this multi-phase castle complex for a number of years. The earliest phase of the castle structure dates to the first decades of the 13th century, although recent work may indicate that the castle was built on an earlier foundation. Much debate in the past has focused on whether Adare Castle was built on an earlier ringfort or ringwork. Excavations in the drawbridge pit in 2001 (Excavations 2001, No. 742) reopened that debate with the discovery of an early medieval baluster-headed ringed pin and half of a decorated rotary quern. The drawbridge pit was revisited to determine the nature of two large voids discovered there.

Two areas in the outer ward were the main focus of this season’s work. Area 1 is sandwiched between the western entrance gate-tower, the outer curtain wall and the early hall, and Area 2 is situated between the early hall, the outer curtain wall parallel to the river and the later hall. Area 1, encompassing a virtually hidden enclosed space, identified as a kitchen, was fully excavated. Work in Area 2 included limited monitoring of the removal of a temporary flood-relief bank and small-scale excavations of a stone-built well and bread oven.

Drawbridge pit
The previous season’s work related to the drawbridge pit, excavation of which revealed two large voids at the north-east and north-west corners of the northern wall of the pit, effectively undermining the drawbridge structure. A baluster-headed ringed pin was also recovered. Removal of the loose fill revealed that the voids had penetrated the western and eastern gables of the drawbridge structure. Examination of the exterior of the structure revealed a single, poorly executed, masonry-patched area, indicating clearly that both voids had fully penetrated the walls. It was concluded that these void features are consistent with deliberate undermining associated with an aggressive military attempt to penetrate the inner ward. Furthermore, when the loose fill was excavated from the voids, it was apparent that the rear inside northern wall of the drawbridge pit was built against a consolidated earthen bank. Limited and difficult cleaning revealed some stratigraphy, which was recorded and sampled. It is hoped to obtain a radiocarbon date from charcoal associated with it, which might help in determining whether the castle at Adare was built on an early medieval ringfort or ringwork as various scholars have suggested.

Area 1: kitchen
Area 1 is unusual in that it comprises a fully enclosed area between the high walls of the early hall, the curtain wall and the gate-tower, with a fourth inner wall enclosing and sealing off direct ground-level access. There are no doors or windows in the room. This enclosed space was filled with over 3m of material, and excavation could only be carried out from above, which required the construction of a scaffold onto the inner lower wall. The kitchen, as it turned out to be, miraculously survived the extensive Dunraven excavations of the mid-19th century unscathed; it measured 4m by 3.5m. The upper layers, c. 1m, primarily of collapsed masonry, stone slates and dark organic fill, effectively sealed all of the contexts below. A number of architectural fragments were recovered from the upper, tumbled masonry layers, including two sandstone window fragments that originally came from the early hall, carved with heavy roll-moulding and foliated decoration.

Stratified and sealed beneath the collapsed masonry was 2m of fill containing the discarded remains of the 13th-century domestic economy of Adare Castle. A pit-like depression was visible from the upper levels and continued down through all archaeological layers. Recovery of finds associated with the sealed context was very productive, and their preservation was good; consequently all fill was sieved on-site to maximise artefact and ecofact recovery. One of the most productive layers, F7, extended more or less over the entire area. This rich organic layer, initially assumed to be a pit, was a slump caused by natural phenomena and consistent with a drain-hole effect associated with flooding. Interleaved between and around dark organic deposits were heavy layers of peat ash and numerous hearths with associated draughting flues that meandered throughout the ash. A broken rotary quern had also been used as a hearth, but in general cooking was done on the ground in simple, scooped-out pits. The most artefactually productive context was C26. This dark organic deposit contained many sherds of 13th-century ceramics, the vast majority of which were fine-quality local wares.

Meat joints were also prepared in the kitchen, as virtually all of the bones have coherent butchery marks, both flensing and chopmarks. The sieving also recovered the tiny splintered pieces of bone from the chopping process, indicating clearly that meat preparation took place in situ.

A hemispherical wall slot was excavated running vertically down the inside of the lower enclosing wall of the kitchen. This accommodated a wooden post that possibly functioned as a crane. The waney edges of the post had been roughly hewn, and the negative impression of the wood was still visible on the mortar bonding in the slot, indicating that the post was inserted during construction work.

A paper-thin veneer of wooden planking was recorded in the lower levels. It extended across most of the kitchen and also covered the natural pit. As it was only a few millimetres thick, it was not possible to lift it and maintain its fragile integrity. However, it was possible to ascertain that the planking originally consisted of two layers of boards fixed at right angles to each other.

As the kitchen walls became visible, a number of regularly placed horizontal slots were evident, some of which penetrated the entire thickness of the walls and no doubt were used as putlogs for scaffolding within and without. Corner-seam abutment joints were also uncovered that illustrated the construction chronology of the four walls.

Artefacts recovered were all utilitarian and included sherds of medieval coloured window glass, a number of small knife blades, a 13th-century die, numerous nails, one of which has a rove on it and may be from a boat, two silver coins (as yet unidentified and in poor condition) and an amorphous copper object. Analysis of plant remains shows the complete range of medieval plants, including corn gromwell, which was most likely an accidental import. The faunal remains included macro- and micro-assemblages. Sieving was particularly helpful, and many fish scales and tiny bones were recovered. The fauna is currently undergoing specialist analysis. One interesting find from the ash was a semi-intact eggshell from a hen. A number of coprolites were also recovered.

The results of the excavation of this hidden room show quite clearly that it was used exclusively as a kitchen, and it can be dated by the ceramics to the 13th century. Access to the kitchen would have been by a wooden stairs fixed onto the north wall of the early hall. The doorway into the first-floor dining area of the hall is immediately above the kitchen, where it connects with the lowest wall of the kitchen.

Area 2
The second area monitored, and partially excavated, was situated between the two halls of the castle and immediately inside the outer curtain wall. The excavations were undertaken in an effort to alleviate flooding primarily associated with the well and to uncover the oven that is recorded on all of the earlier maps of the castle. The well is situated alongside and virtually abutting the curtain wall, outside of which runs the River Maigue. The river receives tidal push up to and just beyond the castle. At times of high tides and storms, the castle is still vulnerable to flooding. In the mid-1970s major drainage works were undertaken, and an earthen bank or levee was inserted along the inside of the castle complex, through all of the buildings and parallel to the river. However, flooding still occurs. One of the conduits for the inundation is the well. River water percolates through the curtain wall and rises up through the well. The purpose of the excavation was ultimately to plug and seal the well. Before excavation, the earthen bank was removed and monitored. The cleared site was studded with collapsed masonry from the curtain wall (since repaired by Dúchas), which was recorded and removed.

Stone well
The well was c. 2m deep, and access to the bowl was by five stone steps. The sediment at the bottom of the circular bowl was emptied, and a few broken rotary-quern fragments and some faunal remains were recovered. It is understood that Dunraven had emptied the well previously. The well must have been lidded in the past as intramural stairs exit from the curtain wall immediately over the bowl of the well.

Oven
Some 3m to the east of the well and abutting the exterior south-west corner of the later hall were the possible remains of an oven as denoted on numerous layout maps. Partial excavation revealed the remains of a brick oven that was set in an outer, hemispherical matrix of dense clay and limestone. Only a few courses of the brick interior survived. The inside of the oven was small, measuring 0.36m by 0.34m. A flat slab formed the inner base. A broken slate was recovered from the inside and may have been used as a baking tray. Immediately outside the front of the oven was a heavy deposit of indurate ash with a kneeling slate set win it. The oven was only revealed and not fully excavated. These dome-shaped ovens relied on radiant heat and were used for baking bread. There is a very fine, intact, medieval example at the Franciscan friary in nearby Askeaton.

Cooking pit
Cleaning beside the oven revealed a rectangular stone-lined and -floored pit that was filled with a black, burnt deposit. Removal of the fill revealed scorching on the basal stones. This pit was most likely used for slow cooking and for keeping food warm.

Stone drain
A very fine stone-built drain bisected Area 2 in a north-west direction and exited to the river alongside the top step of the well. The drain became evident on removal of the collapsed masonry from the curtain wall. After recording, the lintels were numbered and removed, and the excavated drain fill revealed a stone base. The drain did not require full excavation, and the lintels will be replaced and the drain left exposed. The drain may have been used as a mechanism to run off excess or stagnant water from the inner moat. However, work to establish fully the relationship of the drain with the moat is not completed.

Other features and artefacts
The basal remains of a number of walls were also revealed. Dunraven had recorded sections of these walls in the mid-19th century. Artefacts recovered from Area 2 were not numerous: two coloured glass beads were found, along with a discrete quantity of pottery sherds and miscellaneous metal objects.

A small number of stray finds were recovered from various locations throughout the castle complex. These include a very fine, perforated, circular gaming-piece, made from antler with a carving of a fabulous animal, possibly a unicorn. A two-pronged bone-handled fork, some imported pottery sherds and a few slates were also recovered.

Post-excavation work is continuing.

3 Canal Place, Tralee, Co. Kerry