The Glamorgan-Gwent Archaeological
Trust Historic Environment Record
following information has been provided under the terms and conditions
of access as detailed on GGAT’s website www.ggat.org.uk.
Copyright is reserved on all data supplied by the GGAT HER Charitable Trust.
All output resulting from the use of the data must acknowledge the source
from information held by the GGAT HER Charitable Trust copyright.
data below is intended to be used for information and research only and
is not for use as part of a commercial project. If you wish to use
information derived from material held by the GGAT HER Charitable Trust
for publication in printed or multimedia form or to compile resources for
commercial use, prior permission must be obtained in writing. For further
information or to arrange a visit to the Trust please send an enquiry form
CASTLE TUMP (WHITWELL BRAKE)
Primary Reference Number (PRN) : 01034g Trust : Glamorgan Gwent Community : Caerwent Unitary authority : Monmouthshire NGR : ST47539115 Site Type (preferred type first) : Roman Villa Status : Scheduled Monument
Summary : Roman villa site, excavated in the latter half of the 19th century by G Colston. Unpublished, but a plan was deposited at the National Museum of Wales. The site is the subject of ongoing excavations by the Defence Archaeology Group, in association with Operation Nightingale.
Description : Excavated in the latter half of the 19th century by G Colston. Unpublished, but a plan was deposited at the National Museum of Wales. This show the S and E sides and the E half on the N sides of what may be a courtyard with a range of rooms of varying sizes on the E side, a slightly narrower but largely undivided range to the N, and a single small room in the SW corner. To the E of the E range is an undivided space which extends from the N to the S side of the complex and is somewhat wider than the E range. A mosaic pavement 6ft square was found, tesserae 1 in square, with in the middle a chequer design in black and white 1ft 6in x 1ft. Flue tiles also known. Box tiles and roof tiles were visible in 1987. Field visit and WB on geophysical test pits 1994 (Source 8): The Scheduled area is co-terminus with the small wood (Whitewall Brake), but the plan of the 19th century excavation shows that the whole of the building complex was not uncovered, and there is a high possibility that the Roman buildings would originally have covered a larger area. In addition, a Roman villa would normally be expected to stand at the centre of a field-system. The difficulty lies in determining how much of the Roman buildings/landscape survives over a wider area. The 1841 Tithe Award show that the surrounding areas were arable, and damage is therefore likely to have been caused to shallowly-buried or insubstantial feature by ploughing. In addition, construction works and earth-moving associated with the construction of the Royal Naval Propellant Factory is likely to have caused further damage, particularly on the north and south sides. The villa occupies an outcrop of rock raised above the surrounding area on all sides except the north. On the north, modern building extends to within 10m of the northern side of the Scheduled area. On the south, the natural fall of the ground is steepest, masked to some extent now by the railway line, which is here embanked. It is not possible to determine whether the villa would originally have continued downslope under the embankment. On the west side of the villa, a considerable amount of small limestone rubble was noted among the roots of a belt of trees running roughly N-S some 50m west of the wood (probably representing the line of the boundary between the fields known as Cae Gore and Limekiln Piece on the 1841 Tithe Award), and a lesser amount among the trees between this belt and the wood. It was not possible to determine whether most of this stone was natural in origin or had been derived from the Roman building, but a wedge-shaped piece of tufa from the N-S trees (043) is likely to have come from a Roman vault or arch. Nothing was noted in the area between the trees and the wood, but the vegetation cover was virtually complete. On the east side of the wood is the field in which trial pit no 10 was dug. The trial pit was just below the crest of a slight 'platform' which continued into the wood. Before the trial pit was dug, it was thought that this 'platform' might represent a continuation of the Roman building, but the results of the pit suggest that it is geological in origin. A small piece of tufa was picked up from a wheel-rut near the SW corner of the field (044), but on the E side of the hollow which runs N-S across the field.
A geophysical survey on the site in 2012 revealed a north-south transect from the northern boundary to the southern boundary walls, as well as ranges to the north and south of the courtyard. An outline of an orthogonal-shaped structure (a type often associated with high-status shrines, dining/reception rooms or possibly baths) within the main body of the Roman structure was also noted, of which robbed-out walls are visible on the ground. An interim report of the associated excavation notes that a number of 19th century excavation trenches and spoil heaps were identified, some of which had been modified at a later date for military sangars (firing positions) and bashers. These 19th century trenches focused on the Roman walls, of which a tesserae was mentioned by antiquarians in the area. A small assemblage of artefacts including ceramics and imported Samian ware from Gaul were found indicating a date of 3rd-4th century occupation. Roof slate was also identified indicating a fairly large roof (Brown 2012).
Phillips notes that the Castle Tump is located at the disused RAF facility at Caerwent and was searched for any possible remains of earthworks during a Chepstow Archaeological Society field-trip. All that were found were identified as a possible Roman villa by Prof Martin Hennig (pers comm; Phillips 2004).
An excavation in 2013 (E005176) found evidence of a well constructed, flashy hypocaust room in the southwest corner of the site, where mosaic tesserae and flue fragments were recovered, along with stone roofing slabs that match those used on Caerwent's urban basilica in the later 4th century, suggesting a similar date for this building. Evidence that use of this site may have been terminated by fire was recorded in the form of heavily burnt pottery, dating to the late Roman period.
However, while evidence for the later occupation, and perhaps abandonment of this site have been recorded, the date of its establishment is less clear. There was a general dearth of material culture recovered during this season's excavations. However, those recovered are indicative of occupation during the 3rd and 4th centuries, with the only tentative evidence of earlier activity being in the form of some unusual flue tiles of a type normally assigned a 1st century date. It is hoped that further study and future excavation will enable a better understanding of the earlier phases of the site.
Evidence for the demolition of the Roman walls was recorded, and was the cause of the destruction of the former mosaic in the hypocaust room. this phase of destruction is undated, but other features recorded overlying the Roman ones, which comprised an oven and stone flagged floor, are thought to be early medieval. Two possible graves were also encountered associated with these features, with a fragment of probable human cranium recovered from the fill of one. Map regression has also been used to suggest a more agricultural post medieval use of the site (Abramson, James & Walshe 2013).
Excavations in 2014, as part of Operation Nightingale, recorded further pilae in the area of the hypocaust room, along with fragments of mosaic, rather than just individual tesserae. However, the close dating of the site, and its ultimate function still remains undiscovered (Abramson, James & Walshe 2014).