2007:1498 - Kilteasheen, Roscommon

The 2007 field season at Kilteasheen, Co. Roscommon, run jointly by Christopher Read of IT Sligo and Thomas Finan of St Louis University, ran from 28 May to 7 July. The results of this year’s excavation, in addition to the analysis and dating of material excavated in 2005/2006 (Excavations 2005, No. 1339; Excavations 2006, No. 1736), has resulted in a dramatic rethinking of our interpretations of both the function and date of many features on the site, in addition to the realisation that the use of the site spans a much greater time period than originally thought.
We managed to significantly increase the numbers of excavated burials in 2007. In 2005, twenty burials were revealed and twelve were fully excavated. In 2006, a further 40 burials were uncovered, of which 27 were fully excavated (in addition to most of the eight left in situ in 2005). In 2007 a further 52 skeletons were revealed. Of these, 47 have been lifted, in addition to all thirteen left in situ in 2006. While there are only five burials now left in situ, there is clear evidence of further early phases of burial in both Cuttings F and G. A minimum of five distinct burial phases have been identified in B, F and G. Recently received 14C dates and other archaeological indicators such as differences in burial orientation and some indirect dating evidence point to at least two distinct levels of burial activity or, more likely, continuous use of the site as a burial-ground for over six centuries.
The excavation of Cutting D (located at the western end of the building formerly known as the church) was completed in 2007 and included the exposure of further industrial activity, post-dating the building’s construction, in addition to the discovery and excavation of some potential pre-building features. Three later child burials were also excavated in Cutting D. The results of both the excavation and the exposure of a definite base batter on the external wall faces support the current interpretation of the building as the likely remains of the cuirt built by Bishop O’Conor in 1253 and not the parish church as originally thought.
A single new cutting, H, was opened and fully excavated during the second half of the 2007 field season. Cutting H was centred on the bank/ditch of a large subcircular enclosure identified during the 2006 geophysical survey and visible on satellite photographs of the site. The excavation of Cutting H revealed a modest bank and a substantial external ditch, so wide its full extent could not be revealed, even with a 2m extension to the original cutting. While the enclosure could be from any period, with no dating evidence retrieved from either the ditch or bank, the 14C dates received for burials and features on the platform, which is contained within the enclosure, indicate a probable early medieval origin.
All of the specialist work undertaken so far has contributed greatly to our evolving interpretations of the site, and will be discussed at the end of the report, but none more so than the analysis of the 2501 lithics from the 2005/2006 field season and the recently received radiocarbon dates. The majority of the lithics come from secure early to later medieval contexts, and have been identified by Maria O’Hare as representing two distinct periods of prehistoric activity, Late Neolithic and Beaker/Early Bronze Age. The distribution of these items across the site and from widely varying stratigraphic levels clearly indicates they do not represent in situ prehistoric activity. It is more likely that this material originated in one or more locations in close proximity to the platform and were moved to this location, mixed, trampled and left exposed either in prehistory or later and subsequently disturbed by medieval construction and burial activity at the site, or the lithic-laden soil was imported specifically to cover burials (the majority of the lithics come from Cuttings A and F and, in particular, the many dump layers in Cutting F directly associated with the burials). In addition, the ratio of secondary tools/utilised pieces to primary reduction-related material is much higher than would be expected (1:2), indicating their original location(s) were unlikely to be associated with stone tool production and rather their use/storage or intentional interment (i.e. associated with burial activity/
settlement). Also, the ratio of flint to chert is quite high, indicating a likely local source for the flint. There is also evidence of worked quartz and one piece of imported exotic flint (the barbed and tanged arrowhead). More than 100 further pieces of worked flint and chert were retrieved during the course of the excavation in 2007.
The three dates (calibrated and 95.4% two sigma) received are as follows: Burial 5, Level 1 Cutting A: ad 661 to ad 784; Burial 27, Level 3 Cutting G: ad 1162 to ad 1275; Feature 31, Level 2 Cutting A: ad 679 to ad 828. The dates for both Burial 5 and Feature 31 are very close and strongly support a very early medieval date for the earliest burial in the Cuttings F/A sequence and for the actual fabric of the platform. While we clearly have much earlier burial activity than anticipated, the date for the charcoal sample is not especially helpful. The charcoal sample was taken from directly beneath the perimeter wall on top of the platform and was intended to give an earliest possible date for the wall’s construction. However, from stratigraphic evidence in Cutting B, burials found with 13th/14th-century finds were earlier than, or contemporary with, the wall in that cutting. While it is possible that the wall was rebuilt on one side of the platform at a much later date, it is unlikely, given the strong similarities in the form of the wall as revealed in Cuttings A and B. F31 can be considered to be part of the fabric of the platform itself. Thus there remains the question of whether the platform was an early or later medieval construction, although the latter is the current interpretation. This will be discussed in greater depth at the end of this report. The early date for F31 is very close to the date for Burial 5 and, regardless of its implications for the construction of the platform, it does confirm the early medieval use of the site.
The date for Burial 27 was equally surprising. Burial 27 was from the latest phase of burial activity in Cutting G, and was considered to be a likely candidate for mid-14th-century mass burial associated with the Black Death. The dates cannot be stretched that far and indicate a likely 13th-century date for the final episode of burial in this part of the site. This is generally supported by the artefacts associated with the latest burial in Cutting B that date anywhere from ad 1250 to ad 1350. As we could not date the chosen sample from the latest phase of burial in Cutting F and as of yet do not have any dates from the earliest burials in Cuttings B and G, it is unclear if we have a situation where there is extensive and continuous burial over the whole site, from the 8th century to the 13th, or if the burials in Cutting F indicate a separate area of much earlier burials than those revealed in Cuttings B and G. Only the completion of these cuttings in 2008 and the dating of further burials will answer this question. A further six weeks on site during 2008 is anticipated.