Home | Main Menu | Castle Index | Historical Essays | Recommended Castles | What's New | Links

Rumney Castle
A Ringwork and Manorial Centre
in South Glamorgan

by K. W. B. Lightfoot
Text and photographs copyright © by K. W. B. Lightfoot
and the Glamorgan-Gwent Archaeological Trust

Map link for Rumney Castle

We are pleased to welcome Kenneth Lightfoot as a contributor and consultant to the Castles of Wales web site. Mr. Lightfoot was the principal director of excavations at Rumney Castle in southern Wales for two archaeological digs that took place in 1978 and 1980-81. Mr. Lightfoot is also the author of the subsequent excavation report produced for the Glamorgan-Gwent Archaeological Trust.

Rather than reprint the entire report, we've decided instead to present only some basic background on the castle along with some of the conclusions reached as a result of the excavations. The full report can be found in Medieval Archaeology, vol. XXXVI 1992.


Below (2): map showing the approximate location of Rumney Castle (left) and Period VI hall, Building H, after removal of roof debris and before excavation of the interior (right)



Rumney Castle, a small ringwork historically part of the marcher lordship of Gwynllwg, was situated above a steep natural scarp overlooking the river Rhymney. First mentioned in A.D. 1184-85, the castle guarded the western boundary of the lordship and the river crossing. The defenses consisted of a ditch and clay rampart constructed around three sides of the site. Initially incorporated into the defenses along the fourth side was a large timber building and possibly a palisade. The entrance was originally defended by a large timber gate tower, later superseded by a smaller timber structure. Following this, the defences were strengthened by the widening of the rampart and the construction of a small tower or keep alongside the entrance. Several phases of timber building, including two large halls, were arranged around a courtyard. During a later period the entrance was relocated and a stone gate tower constructed.

During the second half of the 13th century the site was converted for use as a manorial centre. The rampart was levelled, the interior of the site infilled, and a range of buildings constructed along the edges of the mound. A well-sealed coin hoard of C.A.D. 1288-89 discovered in a destruction deposit provides a terminus post quem for the abandonment of the site.

The castle was sited to utilize the protection of two steep natural scarps, one forming the NW edge of the site created by the river Rhymney which flows at its base and the other forming the NE edge of the site created by a now dry small valley leading away at right angles from the Rhymney. The site was isolated from the rest of Rumney Hill by the digging of a ditch from the stream valley in a south-easterly direction then curving to the NW and linking with the natural scarp above the Rhymney. This created a D-shaped mound c.40 m by 45 m, similar in size and shape to Grosmont Castle in Gwent.

Two separate excavations of the castle were conducted by the Glamorgan-Gwent Archaeological Trust in advance of development, sponsored on both occasions by the I.A.M. (Welsh Office, now Cadw) and a Manpower Services Commission Special Temporary Employment Programme. The entire summit of the mound was excavated, except for a 3 m wide baulk along the modern property line, and a limited portion of the ditch was also examined. The first excavation was behind the Oaklands Hotel, directed by P. Stanley and K.W.B. Lightfoot between 6 April and 4 July 1978. The second excavation, directed by K.W.B. Lightfoot, took place from 8 May 1980 to 1 December 1981 on the NE three-quarters of the mound and ditch behind no.637 Newport Road. Both time and labour were far more limited in 1978 compared to 1980-81 and a more complete and reliable history of the castle was recovered from the latter excavation, particularly for the earlier periods of occupation which were not uncovered during the 1978 season. Certain finds reports are incomplete due to fire in the Trust's Headquarters in 1983.


Below: Northeast wall of Period VI hall, Building H, pierced by a drain

The excavations at Rumney represent the most extensive examination of a Norman ringwork castle in the British Isles. The interpretations offered here are certainly not beyond question, however. Yet, when the recorded evidence is analysed it is possible to trace a broad pattern of development for the site. More difficult, however, is the dating of this sequence. The foundations of the castle could be related to the establishment in 1081 of a Norman enclave centred around Cardiff, or later, during Robert Fitzhamon's annexation of Gwynllwg in c.1093. The overall development of the site continued throughout the 12th and 13th centuries. The dating of the six different periods of building identified within this span of two hundred years, however, can be tentatively suggested:

Period I - c.1081-93
Period II - early to mid-12th century
Period III - c.1184
Period IV - late 12th century or early 13th century
Period V - early 13th century
Period VI - post 1270 to c.1295

The dates offered for Periods III and VI are, like that for Period I, based on historical references. The extensive works carried out during Period III represent a complete upgrading of the defences, which would have involved considerable expenditure. Allowing for the two earlier period of building, it would seem reasonable to ascribe these works to either William, Earl of Gloucester, who held the lordships of Gwynllwg and Glamorgan from 1147 until his death in 1183, or his successor, Henry II. The importance of the castle as an instrument of control would have been amply demonstrated at the outset of Henry's tenure when the Welsh rose in serious revolt lasting until the summer of 1184. Henry retained possession of the lordships until his death in 1189, and he is known to have repaired or strengthened several castles throughout Wales. Much of his work is recorded in the Pipe Rolls, but other projects which are not documented are thought to have been commissioned by the King, most notably the extensive rebuilding in stone of Newcastle in Mid Glamorgan. It is not impossible that the works at Newcastle and Rumney were carried out as part of a royal policy which was financed by a source other than that recorded in the Pipe Rolls; for example expenditure of castles under Edward I is not all represented in surviving accounts.

Below: Photo of ram's head jug discovered at the site

It is difficult to envisage the conversion of the castle into a lightly defended manor of Period VI before 1270. Until this date Rumney would have retained something of its strategic front-line importance, with the Welsh in continuous possession of the two commotes, Senghennydd and Machen, just to the north. Before these territories were annexed by Gilbert de Clare in 1267 and 1270 respectively, they posed a distinct threat to the low-lying lands of Gwynllwg and Glamorgan. The construction of Caerphilly Castle immediately following these annexations clearly demonstrates both Earl Gilbert's respect for the potential scale of the Welsh threat, particularly under the influence of Llywelyn ap Gruffydd, and his determination to secure these newly seized territories. The conversion of Rumney Castle may therefore be directly linked with the construction of Caerphilly Castle and is likely to have occurred when Gilbert's mother, the Countess Maud, was in possession of the manor as part of her dower from 1267 until her death in 1289.

The work at Rumney also represents a further development in the archaeological investigation of Norman ringworks in Wales. The results of the Rumney excavations add considerably to our knowledge of this type of castle and together with the evidence from other excavated sites in the region, both ringworks and mottes, provide the basis for a detailed comparative study of related castles within a well-defined geographic area - the former lordships of Gwynllwg, Glamorgan and Gower. Just such a study covering the majority of this area has been recently published by the Royal Commission in their authoritative survey of early castles in Glamorgan.

K. W. B. Lightfoot

The Norman Invasion of South Wales and Rumney Castle


Follow this link to learn more about ringwork and motte-and-bailey castles in Wales

Home | Main Menu | Castle Index | Historical Essays | Recommended Castles | What's New | Links

Copyright © 2009 by K. W. B. Lightfoot and the Castles of Wales Website