2006:613 - Assay Office, Dublin Castle, Dublin, Dublin

Built between 1847 and 1864 and extended to the south with a basement in 1872, the Assay Office originally functioned as a canteen and bar for the NCOs garrisoned in Dublin Castle. Since 1925 it has been occupied by the Company of Goldsmiths of Dublin. The brief involved deepening the existing cellar, excavating through the sub-floor deposits of the primary structure and opening up an area outside and to the west of the standing building, located just inside the Ship Street gate of the castle. This work took place between September and Christmas.
The Assay Office is located on a perceptible slope, with ground levels at 7.6m OD to the south, dropping to 6.4m OD 16m to the north over the extent of the site. The southern culvert of the Poddle extends a further 16m to the north and was recorded by Linzi Simpson (Excavations 1994, No. 61, 94E0074), while Tim Coughlan had monitored services trenches immediately to the west of the Assay Office the previous year. In addition, that year Simpson undertook work under the southern extension to the Clock Tower Building, prior to its conversion to the Chester Beatty Library. She found that extensive late- or post-medieval quarrying had removed the medieval occupation and surface deposits, but there was no indication as to the extent of this activity.
The archaeological brief was somewhat complicated by the new design, which involved the underpinning of the exterior walls of the building along with the complete removal of the spine wall that divided the primary structure and its southern extension.
It was envisaged that subsoil would be located just over the formation level for the new structure (located at 4.495m OD), with the possibility of checking it for evidence of the early Hiberno-Norse occupation of the area. This, however, was not to be and subsoil was never located, even in a test-pit mechanically dug to 2.7m below formation, which reached 1.8m OD.
The strata recorded in the test-pit were sampled for environmental residues and they are being examined at the time of writing. They consisted in the main of well-stratified, organic deposits containing leather scraps in abundance along with marine shell and pottery, which on initial examination appears to consist of Dublin-type coarseware and the ubiquitous Leinster cooking ware. A deposit of pure silt, 0.12m in thickness at 3.05m OD, possibly represents the western edge of the Black Pool, which deepened further east over the area now occupied by the Castle Gardens.
Cutting this material was a previously unknown channel running east–west, parallel but uphill from the Poddle, with a width of 4.2m and a depth of at least 1.2m. The southern side of the channel was initially recorded in the basement and the yellow material was assumed to be subsoil. However, further excavation, confirmed by a section provided by an underpin trench, indicated that the material sealed a waterlogged organic deposit and further excavation was stymied by the volume of water pouring in from all sides. Excavation in the exterior trench indicated that the channel was associated with two different and quite substantial layers of redeposited subsoil (which may once have formed banks) and there was evidence along the northern side of the feature for a formal revetment, presumably of timber, owing to the edge being vertical.
Clarke illustrates a watercourse emerging from the centre of what is now the Castle Gardens, which join the Poddle at the King’s Mills under the Lower Yard, just to the north-east of the Chapel Royal. It is tempting to interpret the feature as a western continuation of this millrace that may have come off the Poddle just before the break of slope in front of the Pool Gate on Bride Street. The pottery recovered suggested a deliberate backfilling, possibly as early as the 14th century, and the line of the watercourse was certainly visible into the early 18th century, when a brick latrine was constructed through the softer fill, with the harder redeposited material at either side. The line of the watercourse may well have acted as a property boundary at a later stage in the medieval period, on the basis of a bank of redeposited material with timber uprights, recorded in the basement trench (but not in the exterior trench). The timber may represent a barrier to stop livestock straying down to the marshy banks of the Poddle.
The redeposited material to the north of the channel was sealed by the demolition material probably associated with the structures on Buckridge’s Court, a small street depicted on Rocque’s map (1756) that extended into what is now the castle precinct, prior to the construction of the Ship Street barracks in the first decades of the 19th century. A single coin was recovered from this material, which is now undergoing conservation at ArchCon Labs. The material to the south exhibited a more gradual slope and was sealed by late medieval garden soils, which merged into the fill of the channel. This material produced sherds of medieval pottery, scraps of leather, animal bone and metal objects, along with a fragment of a line-impressed square floor tile. This was an Eames and Fanning type L4 and depicted a lion rampant sinister in a pointed quatrefoil frame, apparently the most numerous type found in Ireland, with other examples found at Swords Castle, where it is the dominant motif on the pavement. It has also been found more locally in the Castle itself and in Ship Street and St Patrick’s Cathedral.
The 18th-century latrine was similar to those excavated by the writer in Smithfield (Excavations 2003, 581, 00E0272) and was constructed as two chambers, with access via stone steps in the north-eastern corner. The walls were supported on timbers and the brickwork was indicative of the period, the bricks being handmade and held together in a lime mortar with a high sand content. There were no seepage holes present and no characteristic lenses of ash or sand to mask the smell. Aficionados will know that the collection of waste from the city’s latrines was privatised in the 18th century, the material being regularly removed by scavengers and brought to lay-stalls, where it was stockpiled and then brought downriver to be used as landfill in the reclamation of the area between Trinity College and Ringsend and across the river at East Wall. It is thus difficult, if not impossible, to accurately date such structures from their contents and the earliest sherds recovered in this instance were of the North Devon variety, which would probably not have been produced before c. 1680. The latrine certainly pre-dates the military occupation of the area and was doubtless associated with one of the properties on Buckridge’s Court.
The relatively shallow formation level of the new structure, although truncating medieval deposits, did little to elucidate the topographical entity from which the city receives its name. The recovery of evidence for a formal channel separate to that of the Poddle is, however, of some significance, as is the material purposefully placed on either side of it. The F240 material on the northern side of the channel survived to a maximum depth of 1.43m and had definitely been cut by the feature. This contained as much stone as it did clay and may have been compacted by the stable block of the General Officer Commanding, which was directly above it. Hopefully, the coin recovered from the material will provide some indication as to the date of its deposition.
The F192 redeposited subsoil to the south of the channel had a more gradual slope and a distinctive yellow hue and may be the same material as that recorded by Coughlan in a services trench uphill to the south. If this is the case, it represents quite a substantial deposition of earth that cannot be explained away glibly. One explanation is that it represents the initial upcast from the quarrying activity recorded by Simpson just to the south, which was spread downhill towards the Poddle for convenience, or perhaps to consolidate what might have been marshy ground. What is evident, however, is the depth of material, at least 2.7m of stratified organic deposits, which survives under the Assay Office. This compares with Coughlan’s sections uphill, where the yellow material directly overlies the subsoil, except over an 8m stretch where an intervening ‘soft wet grey clay’ was recorded.
If the shallow nature of the new formation level precluded a satisfactory assessment of the lay of the land in the early medieval period, it is hoped the environmental evidence from the early deposits will inform the topographical analysis of the site and indicate some of the activities that were being undertaken on the southern banks of the Poddle opposite the castle in the early years of its occupation.
References
Clarke, H.B. 2002 Medieval Dublin c.840–c.1540. Loose map in H.B. Clarke, Dublin Part 1, to 1610, Irish Historic Towns Atlas No. 11. Dublin.
Eames, E.S. and Fanning, T. 1988 Irish medieval tiles. Dublin.