| Abstract: |
The groundworks consisted of a single foundation trench measuring 18.69m in length; between approximately 1.00m to 1.20m in width, and up to approximately 1.00m in depth; the ground was hand-excavated by groundworkers. Where archaeological horizons were encountered, they were cleaned by the archaeologist by hand. In all cases, investigation of the archaeological horizons and features was minimally invasive, with preference given to in situ preservation of the discovered remains where possible. Standard John Moore Heritage Services recording techniques were employed throughout, involving the completion of a written record for each deposit encountered, with scale plans and section drawings compiled where appropriate. A photographic record was also produced. The spoil from all of the works was visually scanned, especially for finds relating to the Roman period. Where a Roman wall was discovered, the ground reduction was halted to protect the masonry. Discussion between John Moore Heritage Services, the architect, and the Historic England Inspector of Ancient Monuments led to a revision of the modern wall construction plans, which allowed shallower foundations in parts. Furthermore, when ground reduction was complete, a soil ‘blinding’ layer was laid over the base of the whole trench, followed by a sheet of geotextile, followed by a second soil ‘blinding’ layer. These three layers protected the underlying archaeology ahead of the pouring of concrete, as agreed with the Historic England Inspector of Ancient Monuments. Overall, the watching brief findings provide some insight into the research questions regarding the construction of this part of the boundary wall. Whilst the late post-medieval boundary wall was constructed just above the Roman deposits, antiquarian protective sandy layer (3) probably shielded much of the Roman masonry from damage. The wall recorded as 4a by Hurst et al. (1987) survived at a higher altitude than any other Roman masonry in this area; layer (3) was very thin at this point, but the boundary wall does not appear to have caused considerable damage to the stone of wall 4a. The unmapped section of wall running perpendicularly (recorded during this watching brief as structure 14) also had a recognisable face and form, though showed signs of disturbance, especially at its east-northeast end. It is possible that this disturbance represents partial wall collapse or robbing following the villa’s abandonment; however, the possibility that the construction of the boundary wall damaged structure 14 cannot be discounted. To the west of wall 4a, no further Roman masonry was encountered at the depth of the limit of excavation. It is highly likely therefore that layer (3) continues to protect the walls recorded by Hurst et al. 1987 as 4, 3, and 19, with limited or no impact from the construction of the boundary wall. The exception may be construction cut 8, which was excavated to an unknown depth to house an additional foundation for the boundary wall. Due to its location, construction cut 8 may have physically impacted upon the Roman deposits located internally to apsidal Room 26, as well as the room’s western wall (wall 19 (Hurst et al. 1987)). The extent of the impact on the Roman remains is dependent on the depth of cut 8, which could not be investigated during this stage of works. Ditch 6 was the only feature discovered which may have related to garden use prior to the importation of soil. The quantity of ex situ Roman material recovered from the top of the ditch fill indicates that post-Roman garden use did impact upon archaeological deposits, at least in this location, just outside of villa rooms 26 and 25.1. Further excavation of The Wilderness garden would be required to further substantiate this conclusion. |
