2005:360 - MILLMOUNT, CARROWREAGH, Down

Upon topsoil-stripping it was revealed that the southern area of a development site at Millmount, Carrowreagh, Dundonald, contained an Early Christian rath. The presence of this site was anticipated, as an earlier series of test-trenching by Norman Crothers in 2003 had indicated as much. That this feature represented a rath was assumed by the presence of a large ring-ditch, with an external diameter of c. 50m and a quantity of souterrain ware present in the upper ditch-fill. However, in addition to this there also appeared to be another large, similarly sized, ring-ditch to the north-east, which had not been picked up previously during the limited test-trenching. Prior to excavation it was thought that this might represent another rath, although whether it was earlier or later than the first could not be resolved until excavation took place. It appears at the current time that this may be the remains of a field system associated with the rath. It was also clear that any bank that may have been present in the initial rath had been ploughed out. The ditch was up to 2.5m in depth and 3m in width. There was no evidence for internal structures within the enclosure.
Three other areas of archaeological interest were also uncovered. These lay to the north in an area of gravelly soil. Area B represented Neolithic activity, Area C consisted of three ploughed-out ring-ditches and Area D was the extremely truncated remains of what may have been a circular structure containing an internal hearth.
Area B contained quite a number of features of archaeological significance. These consisted of various-sized pits and spreads. Most significant was the presence of what appeared to be a very truncated building. Two sides (north and east) survived, measuring c. 4m by 6m. This appeared to be slot constructed with sizeable corner-posts. There was no indication of an entrance, so it is assumed that this must have been located to the south or west.
A large amount of early Neolithic pottery was uncovered from the slots and post-holes associated with this structure. The quality of some sherds was exceptional. Most were extremely well-fired, highly burnished carinated bowls. No pottery displayed any sign of suspension-holes or lugs or decoration of any kind. Two flint projectile points were recovered from the north-west corner post of this structure, but these were exceptions, as most of the flint found was quite unremarkable.
It is also worth noting that some of the small post-hole-type features produced a tremendous amount of pottery in relation to their size. Clearly this pottery had been deposited for a reason. One post-hole or pit in particular was completely crammed with pottery, to such a level that it could be suggested that the reason it had been dug was to provide somewhere to deposit pottery. In this case the pottery represented different types of varying quality and the weatherworn appearance of most sherds implies that in many cases it was residual at the time of deposition.
Before full post-excavation analysis, it would not be prudent to suggest a function for any other excavated features in Area B. Furthermore, only a very approximate date range can be given in relation to the site. This is based on the early forms of western Neolithic pottery found across Area B and would suggest that the site was in use some time around the first half of the fourth millennium BC.
Area C comprised three ring-ditches in very close proximity. Two had an external diameter of around 7m, while the other was smaller, with a diameter of 4m. The smaller ring-ditch was also the shallowest, with a depth of only 0.12m, suggesting truncation. One of the other ring-ditches (RD-2) had cut an earlier shallow pit containing burnt bone. No other burnt bone was found in association with the other ring-ditches. In fact, apart from some weathered flint, no other finds were recorded.
The most northerly of the three ring-ditches (RD-3) did, however, reveal four burnt lengths of wood at the southern side of its base. This was a strange and unexpected discovery. The ring-ditch had been recut and this recut had disturbed one of the lengths of wood. The primary fill showed no sign of weathering or exposure and so appears to represent an immediate backfill. The four lengths of wood measured in excess of 1m and were all apparently placed in position, rather than simply being deposited here without forethought. They lay against the inner cut and all sat in the same east–west alignment.
Although these lengths of burnt wood were all located in the same place, there were other large chunks of charcoal distributed throughout the primary fill. The fairly uniform distribution pattern of this charcoal throughout the primary fill of this ring-ditch may allow it to be conjectured that these were purposely deposited. None of this wood was burnt in situ; it was burnt elsewhere and brought here for deposition. No sign of toolmarks was evident. However, it was noted that all the wood was very thin, no more than 20mm or 30mm thick, and this gave the distinct impression that we were looking at split wood planks rather than burnt branches.
It was quickly apparent that Area D had been greatly truncated, probably due to relatively recent ploughing. All that remained was a semicircular arc, with a shallow pit slightly offset from where the conjectured centre would have been. Similar to Area C, other than some unimpressive flint, there were no finds recovered.
It is probable that the semicircular feature represents the truncated base of a foundation slot for a roundhouse, with the shallow off-centre pit being the remains of a hearth. If the remaining slot were to be hypothetically extended, it would form a structure with a diameter of around 12m. The off-centre shallow pit contained burnt flint and burnt stones, but a full analysis of the soil samples taken from this feature will be required before a function can be suggested.
Work at this site is still ongoing.