The summit of a lofty mountain spur rising above Risca is enclosed by ramparts with a great castle mound or motte, set at its eastern end. These are the remains of a medieval castle, possibly set within an earlier hillfort. The enclosure circuit describes an elongated oval roughly 330m north-east to south-west by 135m, with an area of 4.14ha. It consists of a high rampart fronted by a ditch and counterscarp bank, set out to take advantage of the natural slopes. Gaps in the west and southern sides have been taken to indicate that the enclosure is unfinished although it is perhaps more likely that these result from later damage. The ground falls steeply away from the ramparts on all sides but the east, where they face onto the level neck of the spur. However, there is a possible approach from the north-west leading to the eastern gap which could be an original entrance.
The motte (NPRN 268157) commands or blocks the level approach from the east along the spur. It is a great circular mound 45m in diameter and 8.0m high with a level summit some 20m across. It may have carried a grand timber-framed tower within a small palisaded court. At its base is a deep ditch, rock-cut in places. This appears to cut the enclosure ditch. On the ditch's western lip, within the enclosure, is a round cairn some 10m in diameter and 0.6m high (NPRN 307849). This has been identified as a prehistoric funerary monument, however, it is tempting to regard it as a bridge abutment associated with the castle mound.
This is an unusually remote site for a medieval castle, particularly for a lordly territorial centre, and it is possible that the castle was built as a hunting seat within Newport lordship. Against this, if the enclosure is regarded as part of the castle, this would suggest that this was a place of assembly where vassals gathered to attend on their lord.
The site was restored and landscaped in 1984.
Source: CADW Guide to Glamorgan & Gwent (1992), 92-3
John Wiles, RCAHMW, 4 February 2008
According to Paul Davis: 'Grass fires in the summer of 2018 have caused considerable damage to the monument. Apart from superficial areas of burning, there are extensive areas where the soil has been destroyed, particularly along the north, west and east ramparts of the hillfort. The burning has revealed no obvious evidence for any structural features within the banks, which appear to be merely dumps of earth and rubble. Similarly, the slight counterscarp bank appears to be irregular rubble upcast from the ditch. Only at the SW end, in one of the gaps through the rampart, has the fire revealed any hint of a structure. An area of flat stones, very scattered, appears to form a rough line, perhaps the foundation of destroyed walling (see photo 2). Evidently this gap has been formed by the rampart being pushed down to infill the ditch, presumably in modern times, and perhaps associated with ploughing that took place within the fort. The small mound beside the castle motte has also been uncovered by the fire and appears to be a stony ring cut on one side by the ditch, which proves it to be earlier (see photo 6). It appears far more likely that it is a cairn rather than a bridge abutment for the motte as has been suggested.
No convincing evidence for any dwelling huts can be seen within the fort. There are three small stony rings set against the rampart on the SE and N sides, but these could be later sheepfolds (photos 1 & 4). The absence of any huts may be explained by the following: (a) they were of turf construction and have left no obvious trace, (b) the ploughing has utterly obliterated them, or (c) there never were any huts. The latter possibility would support the long-held tradition that the fort was unfinished. While it is clear that the rampart was indeed completed (and the gaps are due to later disturbance), its rather basic construction might be interpreted as a `work in progress?, and that the fort was never occupied as intended.'
Notes written by Paul Davis in April 2019
RCAHMW, 15th October 2019