2004:0612 - KILSHANE, Dublin

NMI Burial Excavation Records

County: Dublin Site name: KILSHANE

Sites and Monuments Record No.: N/A Licence number: 03E1359 ext.

Author: Dermot G. Moore, for Cultural Resource Development Services Ltd, Unit 4, Dundrum Business Park, Dundrum, Dublin 14.

Site type: Neolithic segmented enclosure, Early Bronze Age activity

ITM: E 710927m, N 742924m

Latitude, Longitude (decimal degrees): 53.424752, -6.330966

This excavation was undertaken as part of the archaeological mitigation in advance of the N2 Finglas-Ashbourne road scheme (Appendix II). A geophysical survey was carried out by GSB Prospection in 2002, which recorded a number of possible archaeological features in Testing Area 5 (GS 2 Area 25). Pre-development testing subsequently carried out by David Bonner in October 2003 (03E1359) exposed a small number of archaeological deposits, interpreted as a ditch and ditch-like feature containing burnt stone, both undated. The licence was transferred to Dermot G. Moore in March 2004 and, from 15 March to 22 July 2004, excavation of Site 5 was carried out by a team of 43 archaeologists.

Site 5, which comprised three distinct areas, Sites 5a, 5b and 5c, was situated on a gently undulating gravel ridge associated with tributaries of the Ward River and was located in a large irregular-shaped field bordered by the Kilshane road to the south and the N2 to the east. Open drains and hedgerows formed of mature hedge plants and trees bounded the area of the site to the north and west. A commercial glasshouse was located on the north-western side of the enclosure. Site 5a was situated atop this ridge at 80.53-81.3m OD, while Site 5b was situated north of Site 5a at 80.3m OD. Site 5c, situated north-west of Site 5a, lies at 81.04m OD.

Initially, two areas were topsoil-stripped in February 2004 (consisting of a total of 199m2) around the area of a possible ditch/gully feature (later designated Site 5a) and a linear feature (designated Site 5b) identified during the testing phase in October 2003. The topsoil-stripping was carried out using a tracked mechanical digger equipped with a 2m-wide toothless ditching bucket. Upon commencing investigation of these features, it became obvious that the area of Site 5a was considerably greater in extent. A further programme of topsoil-stripping was therefore initiated, which expanded the Site 5a stripped area to 1335m2. This included the excavation of two geological test-trenches to determine the nature of the natural boulder clay and gravel deposits into which most of the archaeological deposits were cut. While the excavation was ongoing, the commercial glasshouse was demolished and the underlying topsoil stripped, adding a further 3500m2 to the opened area. A number of extensions to Site 5a were also added, as well as expansion of the area opened at Site 5b and a series of test-trenches to the west of Site 5a.

The main enclosure ditch on Site 5a was excavated in a series of fourteen box sections, ranging in length from 2.4m to 11.25m, to determine the structure and sequence of the ditch deposits and method of construction. Longitudinal sections were also cut through a number of the baulks to determine the sequence of deposition of individual deposits within identified ditch segments and to define an entranceway. The smaller causewayed ditch uncovered in Site 5c was also excavated by box section.

A detailed excavation strategy was put in place to retrieve as much information as possible from the enclosure ditch and its fills, especially the animal bone layer, due to the quality of preservation and the uniqueness of such a deposit from an Irish prehistoric site. All archaeological features interior and exterior to the enclosure were excavated, in addition to a number of geological features that were investigated to determine their archaeological potential.

The natural deposits defined on the site consisted of three types. The first was glacial gravel clay of unknown depth consisting of firm mixed grey/brown gritty gravel with frequent small stone and pebble inclusions. Ice wedges were identified within this deposit where exposed in the excavated ditch sections. Overlying this was a deposit of frequent angular and sub-angular stones and cobbles (of limestone/greywacke) within compacted mixed brown/grey silty clay with a depth of 0.2-0.35m, which occurred predominantly in the interior of Site 5a and to the west-north-west of the excavated area. This deposit appears to have become exposed due to a combination of natural erosion of the slight ridge on which the enclosure is set and plough action over the years.

Overlying this substantial deposit of stone was a mixed deposit of firm pale-grey/orange coarse sand gravel. This deposit occurred predominantly at the northern and southern limits of the excavated area and was cut by the enclosure ditch and a number of features, especially in the northern portion of the site. This deposit appears to represent the ‘B’ horizon material of the overlying topsoil. In addition to the main geological deposits, a smaller series of deposits was identified during the period of excavation consisting of the remains of decomposed siltstones and calcareous rocks, while others were simply irregular depressions in the natural filled with yellow/yellow-brown sands and fine gravels.

Site 5a
The earliest activity is defined by the construction and infilling of a large ditched enclosure dating to mid-Neolithic times. The enclosure was almost egg-shaped, coming to a notable point (the ‘apex’) in the north. Once the limits were defined prior to excavation, it was noted that the shape of the ditch was very irregular and the reason for this became apparent during the excavation. The ditch had been excavated in a series of interconnecting regular and irregular segments.

The overall plan of the ditch shows that its long axis was aligned north-west/south-east and that it had maximum external dimensions of 45m northwest/south-east by 34m. The width of the ditch varied around its circumference (1.9-3.8m) and enclosed an area 38.5m north-west/south-east by 27.5m, totalling c. 850m2. The western side of the ditch bowed inwards, off line with the rest of the ditch, which followed a gentle curve north to south. However, there were sharp turns noted, particularly at three locations: the apex, the south-east and the south-west. This is quite probably due to the method of construction of the ditch.

The average length of a segment was 8.9m, the shortest being 6m and the longest c. 13m. The segments were interconnecting and were probably dug by different work groups. The alteration in direction of the line of the ditch at the apex and in other segments may be due to inaccurate ditch digging between the different work groups. This would also add to the suggestion that the segments were dug at different times.

During the excavation, the segments were primarily defined by changes in direction of the ditch and slight changes in height where the segments connected. The profile of the ditch, especially the individual ditch segments, varied from gentle U-shaped to V-shaped (generally, wide U-shaped profile in the centre of the segments and V-shaped at the ends).

In plan, some of the segments appeared to have considerable breaks between them but in most this can be seen as the remains/evidence of the segment terminals, which were sloping rather than vertical-sided terminals. Tentative evidence for the method of ditch excavation is shown by the presence of portions of antler tines in some of the ditch fills, although as yet only one was recognised from primary fill deposits.

Soon after the ditch was cut, it began to silt up (sometimes irregularly) around its circumference. Probably at this stage a segmented ditch with at least four defined causeways was constructed in the area designated as Site 5c to the north-west.

Once the initial natural slumping and silting in the base of the ditch began, a large volume of animal bone was deposited around the full circumference of the ditch. The bone assemblage, consisting of 60-70 individual cattle, is the largest Neolithic bone assemblage from an excavated context (Finbar McCormick, pers comm.). The cattle bone was placed in both a disarticulated and articulated state with apparent selection of certain bones, such as vertebra or long bones, to be deposited together.

In some areas of the ditch the bone appeared to have been deposited from the exterior, while in others it was deposited from the interior. None of the animal bone recovered appeared to have any distinct butchery marks (this has yet to be confirmed) and the deposition of the bone varied from each area of the ditch, indicating a number of possible phases of deposition, with some distinctions noted within individual segments. Fully articulated cattle skeletons were noted, primarily in the easternportion of the ditch, where at least three were found in close proximity. Amongst the bone were large stones and boulders, which may have eroded from the sides, possibly due to water action.

As the bone was being deposited, the silting continued and at the same time there was a series of slumps into the ditch, probably from the upcast material interior and exterior to the ditch. A further series of infillings took place, culminating in the placement of pottery within the ditch. This consisted of a large mid-Neolithic broad-rimmed, round bottomed vessel, which appears to have been deliberately placed on top of this sealing deposit in the south-eastern portion of the ditch. Other fragmentary pieces of ceramic material were recovered from just above the bone layer in the northwestern section of the enclosure ditch and these also may prove, using thin section analysis, to be Neolithic. The enclosure then appears to have been abandoned for a considerable period of time.

The next major activity occurs in the Early to Middle Bronze Age, with the deposition within the ditch of a relatively uniform deposit of orange sandy clay. This appears to have been deliberately placed into the ditch around its full circumference, possibly to seal the earlier (Neolithic) activity. The deposition of the orange clay appears to have been immediately preceded by deposition of charcoal/wood lenses, especially in the western portion of the enclosure.

The deposition of the orange sandy clay within the ditch is another intriguing aspect of the site, with a number of questions relating to the origin of this material and why such a considerable deposit was placed in the ditch. Artefactual material and animal-bone fragments were also recovered from this orange clay deposit. Two suggestions are that the orange clay material was derived from the basal topsoil material or that it was derived from either the interior or exterior of the site, possibly from the creation of a bank. However, there was no evidence of an external or internal bank encountered during the excavation, but this may have been ploughed out.

The next defined phase of activity on Site 5a occurs during the Early Bronze Age. This activity consists primarily of a series of deposits and features associated with the later stages of the main enclosure ditch and a series of cut features, some of which, based on ceramic associations, may date to the Earlier Bronze Age.

Set on to and in many cases cut into the orange clay deposit sealing the fills of the enclosure ditch was a series of deposits, shallow scoops and pits. Most of these were located along the eastern portion of the ditch. Many of the scoops and pits intercut each other and almost all were filled with the same generally homogeneous fill, which seems to represent a midden deposit. The size and depth of these pits and scoops varied considerably, but none appear to have exceeded 0.2m in depth. Within these scoops were ash/cinder deposits and burnt and unburnt bone (some of which appears to be human). The animal-bone remains appear, on preliminary identification, to be pig and possibly ovicaprid. Some of the bone had been worked into pins or awls. Also within this deposit was a large range of lithic material in the form of flint manufacturing debris and finished tools. The predominant components of the flint manufacturing debris consisted of small pebble cores and fine micro-debitage. The secondary worked material consisted predominantly of small, high-quality thumbnail scrapers, a fine hollow-based flint arrowhead and a reworked small barb and tanged arrowhead. A well-made flint piercer and a large hollow scraper (of non-pebble flint) were also recovered. The hollow scraper represents a non-Early Bronze Age tool type and may be directly associated with the initial construction of the enclosure. Chert, quartz and other coarse stone material were also well represented.

However, it is the ceramic remains which dominate the artefactual assemblage. These consist of a substantial quantity of high-quality Early Bronze Age ceramic styles in the form of funerary and high status/ceremonial wares such as food vessels, cordoned urns, cinerary urns, Irish bowl food vessels and a small range of as yet unidentified ceramics. In many cases several ceramic styles were deposited together, with at least six different types (based on decoration and form) being recovered from one single square metre. On some occasions, burnt or partially burnt stone was also found within the deposit.

Generally overlying the artefact-rich deposit was a relatively compact metalled surface, which had its greatest extent in the extreme eastern area of the ditch. The function of this metalled deposit may have been to formally seal the midden deposit. Overlying the metalled deposit was a less artefact-rich horizon, which appears to have been partially disturbed. The extent of this deposit is greater than the underlying deposit and variants were found in the northern, southern and western portions of the enclosure ditch. A small number of inter-connecting pits containing material similar to the artefact-rich horizon were also uncovered in the northern area of the enclosure ditch (near the apex).

Associated with this Early Bronze Age activity in the ditch were a number of features located within and without the enclosure. In the northern area of the site, three rather mysterious features were also uncovered. These appear to be cremation pits, which contained unusual sloping red-orange burnt soil deposits upon which were set thin deposits of finely ‘pounded’ or crushed burnt bone. As two of these pits are directly associated with burials of single bones, their true function still awaits clarification. They do, however, seem to be connected with the artefact-rich horizon in the upper portion of the enclosure ditch.

During the course of the excavation of the interior of the main enclosure, a number of features were uncovered which gave the impression of having been cleaned out (sterilised) in antiquity. Several appear to have been pits for probable unprotected cremations, with much of the cremation deposits (and the putative pots into which they were placed) having been ‘cleaned out’ of the pits as the material was deposited into the ditch.

The only intact burial was that of a single crouched inhumation, located south of the centre point of the enclosure. The burial was orientated east-west in a shallow oval pit with no evident grave goods. It was in an extremely degraded condition due to the nature of preservation. The grave might have been tampered with, which may account for the lack of grave goods.

A further series of rather irregular features was also encountered within the interior of the enclosure and these consisted of irregularly shaped pits, which contained small amounts of charcoal and occasionally burnt bone and pottery. The pottery recovered appeared to date to the Early Bronze Age. Only one feature, a hearth, represents activity later in the Bronze Age.

Some medieval activity also occurred on Site 5a, which took the form of a large pit group with deposits of stone and medieval pottery, and another single large pit associated with a north-south-running field boundary. A second parallel field boundary was noted on the south-eastern part of the site. The last phase of activity consisted of a large east-west post-medieval culvert drain and a number of north-southrunning culvert drains, which were all part of agricultural improvements to the land, which had since been used for pasture and crops.

Site 5b
The archaeological activity located within the confines of this area consisted of a south-northoriented linear feature c. 9.5m in length. It varied in width from 0.44m to 0.96m and up to 1.05m where it became very shallow at its northern end. It had a variable depth of 0.1-0.2m and contained only two distinct fills. The basal fill consisted of partially burnt sandy clay with some charcoal flecking, while the upper fill consisted of grey/black sandy clay with much charcoal flecking and occasional burnt stone. Small quantities of burnt bone (and snail shell) were also recovered from the upper fill. Each of the deposits was sampled and when these are analysed a fuller determination can be made as to the function of this feature. For the moment, the linear feature in Site 5b is interpreted as a burnt-out field boundary, with the burnt bone possibly representing small rodents or birds trapped within the hedgerow. However, the fills of the feature also may suggest burnt-mound activity.

Site 5c
The area designated as Site 5c (located in the area of the now demolished commercial greenhouse) produced better and more definitive evidence of a causewayed ditch. This feature ran approximately north-south and had an excavated extent of c. 70m. The ditch itself had a variable depth of 0.13-0.48m, with the smaller depths occurring at the terminals of segments and the greater depths at the centre of segments. Width also varied between 0.99m and 1.77m and this corresponds to the centre and terminals of segments. At the northern and southern ends of this ditched feature, the width narrowed considerably, to 0.5m.

At present, at least four narrow causeways have been identified. The ditch itself was filled by a series of deposits, some of which contained charcoal, animal bone (predominantly cattle) and mollusc (snail) remains. Although, a number of lithic finds were recovered from the various deposits, the only artefact of note was a complete leaf-shaped arrowhead from the uppermost fill of one ditch segment. The recovery of this projectile point from such a location would tend to indicate that the causewayed ditch is Neolithic in construction.

The form of the ditch also varied, especially in the southern area of Site 5c, where the ditch not only narrowed but also divided into two. Although severely truncated by the insertion of the concrete reservoir associated with the commercial glasshouse, the ditch then appeared to deepen and widen before it was lost under modern activity. In this location, two distinct fills were evident, one of which produced a small irregular pebble core and a quantity of mollusc shell.

Also on Site 5c, the only other probable prehistoric feature was a small hearth pit containing burnt and unburnt animal bone. It is also significant to note that the upper homogeneous fill of the large west-east-running double culvert produced two retouched pieces of flint and one small thumbnail scraper. It would appear likely that the deposit within which these three secondary worked pieces were found was derived from somewhere close by, possibly from the two large pit features on the northern edge of Site 5a.

In summary, the main enclosure ditch seems to fall into the causewayed enclosure tradition or at the very least a variation of it, if not by the presence of causeways across the line of the ditch (which may have been removed) then by the segmented nature of the ditch construction; also the apparent deposition in individual segments of grouped cattle-bone deposits, of which there is an exceptionally large quantity, and the presence on top of the sealing deposits of a mid-late decorated Neolithic vessel of broad-rimmed type. In addition, the presence of the outer segmented ditch to the north-west (Site 5c) would lend further weight to the causewayed enclosure hypothesis.