2008:495 - Church Road, Swords, Dublin

An assessment was undertaken for a proposed development site in Swords. The site is in close proximity to the major foundation attributed to St Colmcille, however on the basis of a noticeable curve on the adjacent road which follows the townland boundary, the development site would appear to lie outside and to the south-west of this, the most obvious ecclesiastical enclosure. It lies nonetheless within the RMP constraint area for the town of Swords. The testing programme was designed to investigate the possibility that an outer enclosure ditch might have survived between the townland boundary and the river, despite the recent reduction of much of the area. In addition there was the possibility of encountering human burials on the upper surviving slope, where little truncation appears to have occurred.
Swords is said to owe its origins to the establishment of an ecclesiastical centre by St Colmcille, who appointed St Finan Lobhar as abbot. This suggests a 6th-century date for the site’s foundation, however there are no contemporary documentary sources to confirm this. Local folk tradition records that, when St Colmcille established his church at the site, he took possession of the nearby pre-Christian well, blessed it and dedicated it to the service of religion. The name originally applied to the well was Sord Colmcille, from an old Irish word meaning ‘pure water source’. The well site (DU011–034/013) is located c. 100m to the south-east of the development site, on the far side of the river. Its landscape context has been somewhat diluted by an apartment block which has been constructed immediately adjacent to it and the well is now dry.
Three medieval churches have been recorded at Swords, dedicated to Saints Fintan, Brigid and Catherine. It has been supposed that these churches were all located within or close to the present Church of Ireland site, which has a standing round tower. Alild, Bishop of Swords and Lusk, died here in ad 965, so clearly the monastery was located at this site by the 10th century at least. The round tower is the only surviving fragment of the early monastic establishment. The western side of the enclosure relating to the ecclesiastical site has been eroded by modern housing; however the enclosure survives in the townland boundary of Swords Glebe.
Three trenches were mechanically excavated over the site with a grading bucket, to a width of 2m. Trench 1 was excavated for 40m approximately parallel to Church Road; Trench 2 was excavated for 35m to the south of Trench 1, the trench covering the remains of the slope and terminating close to the terminal of Trench 1. Both trenches were excavated across the area being directly impacted upon by the basement of the development. Trench 3 was excavated over the southern portion of the site, over an area which will not be directly impacted upon by the proposed development.
In Trench 1 bedrock was encountered just under the surface at the north-western terminal, which gradually fell off under a deposition of silt sealing a shell midden in a small pit cutting the subsoil. The slope of the bedrock was quite gradual and is not suggestive of there being a rock-cut ditch at this location. A left humerus was recovered from the ground surface just to the south of the trench. This is probably from the adjacent graveyard, an area of which was removed several years ago when the corner of Church Road was widened. At 24.8m from the north-western end of the trench, a pit filled with small cockle shells was sealed by the silt deposit at 1.7m below the modern surface. The pit was ovoid, measuring 0.6m by 0.7m; it was barely 0.2m in depth. An attempt was made to investigate whether the silt represented the fill of a curvilinear outer ditch of the nearby ecclesiastical site. However there was no indication that the deposit was filling a cut feature at its south-eastern end where the slope of the bedrock at the opposite end of the trench appeared to be natural.
Trench 2 was excavated along an east–west orientation; the initial 10m are described below from the west, where the trench was opened towards the top of the slope. The trench cut across the slope where it had been sharply truncated; this is likely to have occurred in the recent past as the exposed section was free of vegetation. Just above the modern break of slope at a depth of 0.46m from the modern surface, a pit containing a midden of small cockleshells was recorded. On the upper surface of the fill a tine was recovered, 87mm in length with a diameter of between 17–23mm. The pit cut an old soil horizon between the topsoil and the subsoil. The trench extended across the break in slope, where a modern concrete sump was dug into the subsoil. This was filled with a sweet-smelling oily liquid and probably relates to the usage of the site by a chemicals company in the 1970s. The final 13m was characterised by a French drain and demolition rubble. The drain was recorded crossing the trench obliquely at 1.5m below the surface. It comprised a trench filled with water-rolled gravels which, in the north-facing section, was contained within a formal stone culvert held together with lime mortar. The drain was 0.6m in width and was not investigated further. The rubble was concentrated over the final 8m and appears to have been derived from the demolition of the stone buildings which occupied the area, a wall fragment of which survived in the south-facing section.
Trench 3 was excavated across the southern end of the site down the slope (to the north) and across a level area towards the site boundary. The only feature of archaeological significance was noted towards the southern end of the trench where the footings of a wall belonging to a structure long demolished were recorded. The soil profile down the slope was similar to that recorded in Trench 2, with the subsoil being slightly higher relative to the surface. At the base of the slope was a thin (0.25m) stratum of sandy clay over the subsoil. This became increasingly more silty towards the river and there was no definite point where one superseded the other. The footings of a mortared masonry wall cut the subsoil at a point c. 8m from the southern end of the trench. The wall was constructed from small (0.3–0.45m) blocks of limestone and a quantity of brick dust within the mortar matrix may suggest an attempt to waterproof the foundation. The structure is evident on the first edition of the OS and is depicted as a ‘ruin’ on the 1864 revision. The wall would appear to be the very bottom of the northern side of a millrace associated with a cornmill located to the west of the site. All evidence for the millrace had disappeared by the 1906 edition of the OS apart from the wall in question, which now acted as a property boundary. South of the wall was evidence for extensive dumping, which was hidden by the dense vegetation across this part of the site. The material was modern in origin and may have derived from clearance works along the river.
The most significant results from the assessment relate to the recovery of the fragment of worked antler found in association with the shell middens. The fact that there was no associated ceramic finds would suggest that the middens may be early medieval in date and the association with the adjacent ecclesiastical site is therefore difficult to ignore. There is extensive evidence for early medieval occupation in and around Swords. The hilltop at Mount Gamble, less than 1km to the east, produced evidence for over 280 individual burials dating to between 550–1150 ad on a site with no obvious ecclesiastical remains (E. O’Donovan, pers. comm.). Two kilometres to the north-west in Mooretown/
Oldtown was found evidence for an early medieval enclosure, again with an extensive cemetery and other occupation indicators connected with animal husbandry (B. Frazer, pers. comm.).
The main focus of this assessment was the investigation as to whether or not the graveyard associated with the ecclesiastical site extended as far as the river and, on the basis of the results outlined above, it would appear that this is not the case. In addition, the trenches were specifically located to test for the evidence of there being an outer enclosure at the base of the hill, associated with the ecclesiastical site. This does not appear to have been so and it is likely that the Ward River itself acted as a boundary for the site along its southern and eastern sides. While it is difficult to quantify the extent of survival of archaeological features across the site on the basis of three test-trenches, it is quite likely that the midden pits do not exist in isolation and that more evidence for this period of the site’s occupation survives, specifically in the area impacted upon by the proposed development. Where it is possible that the middens belong to different periods of the site’s occupation, it is more likely that they are relatively contemporary, which adds to the likelihood of there being further evidence of occupation throughout the site. The midden in Trench 1 was sealed by a deposition of silt which probably derived from flooding and should other occupational evidence survive over this area of the site, it is likely to be below this material. The midden in Trench 2 was located on the slope, the lower half of which was truncated in antiquity, which suggests that the occupation of the site was not confined to the level ground along the river.
The assessment site is located between an important early ecclesiastical site and the bank of a river which mirrors its enclosure; it is therefore likely that the occupation of the site has an association with the Early Christian activity on the adjacent hilltop. The middens were recovered from small pits within 30m of one another and it is quite likely that they relate to riverbank activity associated with the Early Christian settlement immediately to the north-west. The extent of the activity is at this stage unknown; however, it would appear likely that it extends upslope from the level area previously occupied.