KIDWELLY was originally the name of the district which included part of the coastlands lying between the estuaries of the Towy and the Loughor. In 1106, after the death of Howell ap Gronw, Henry I granted these lands to his minister, Roger, Bishop of Salisbury, who erected a castle at the mouth of the Gwaendraeth Fach. This formed one of a series of Norman strongholds designed to secure their newly won conquests in South Wales and to command the passage of the rivers across which the road to the west passed. A mention of the hall of the Castle in a document of 1115 or earlier shows that the building of Kidwelly must have been practically completed by that year. During the rising which followed the death of Henry I, the Battle of Maes Gwenllian was fought a short distance away from the castle (1136). The account speaks of Maurice de Londres, Lord of Kidwelly, and Geoffrey, Constable of the Bishop, as leaders of the Norman army. Maurice, who is mentioned for the first time in connection with this district, already possessed Ogmore in Glamorgan, where his father William de Londres appears to have been one of the original conquerors. The coupling of the two names suggests that Roger of Salisbury, while retaining possession of the castle, had granted the lordship of the district to Maurice de Londres, who probably acquired the castle also when the bishop died in the following year.

The Welsh chronicles record that, in 1190, the Lord Rhys built the castle of Kidwelly. This entry probably reflects a native conquest of the settlement, but the Normans must have recovered it before 1201, when Meredith, son of Rhys, was slain by the garrison of the castle. In 1215 Rhys Grug, another son of the Lord Rhys, captured Kidwelly and burnt the castle. He remained in possession until 1220, when Llywelyn the Great forced him to restore these conquests. The male line of the de Londres had become extinct during these troubles, and Kidwelly had passed to an heiress, Hawise. In 1225 she married Walter de Braose, who died during the campaign of 1233-4. Kidwelly Castle was by that date again in the possession of the Welsh as a result of the rising of 1231, when Llywelyn the Great, previously a supporter of the royal authority, had turned his arms against Henry III. Hawise de Londres, left a widow, was unable to regain possession of Kidwelly which, in 1242, was still held by Rhys's son Meredith. Two years later Hawise married Patrick de Chaworth, who seems to have recovered these lands soon after this date. The Welsh rising of 1257 involved the destruction of the settlement at Kidwelly, but the invaders failed to capture the castle. Patrick was slain during the campaign of the following year, and the wardship of his lands was granted to Hawise during the minority of their son Payn.

Payn de Chaworth must have attained his majority about 1270, as in that year he and his younger brother Patrick took the Cross. Hawise survived until 1274, and her death was soon followed by that of her two sons, Payn in 1279 and Patrick four years later. The elder brother died childless, and the younger left only an infant daughter, Matilda, who inherited Kidwelly and Ogmore. In 1291 the marriage of the young heiress was granted to the king's brother, Edmund, Earl of Lancaster, for the use of his second son, Henry, the union being celebrated in 1298.

The prominent part taken by Thomas, Earl of Lancaster, in the civil wars of Edward II's reign led to his execution after the Battle of Boroughbridge in 1322, but the forefeited title and estates were later restored to his younger brother, Henry. The extinction of the male line in 1361 caused a temporary partition of the Lancastrian possessions, but on the death of the elder co-heiress, Matilda, in the following year, the whole inheritance fell to her sister Blanche, wife of John of Gaunt, who became Earl and later Duke of Lancaster. On the accession of Henry IV, Kidwelly, together with the other Lancastrian possessions, passed into the hands of the Crown, and was of little importance during the fifteenth century. Henry VII granted the castle to Sir Rhys ap Tudor, whose grandson Rhys ap Griffith forefeited it in 1531. Later it was again alienated and passed to the Earls of Cawdor. The castle had long ceased to be habitable, but certain repairs were carried out during the nineteenth century. In 1927 the owner placed the ruins under the guardianship of the Commissioners of Works (now the Department of the Environment).

Since that date extensive works of preservation have been undertaken. In 1930 and 1931 excavations were carried out by Sir Cyril Fox and the writer in order to recover the earlier history of the castle. The results are embodied in the present guide, and the interesting series of relics recovered may be seen in the National Museum of Wales.

PERIODS OF CONSTRUCTION

The oldest surviving masonry is that of the towers and the curtain enclosing the inner ward. This occupies a rectangular area with a circular tower at each angle. There are two gates, on the south and north sides, each protected by a portcullis. The erection of these defences within the circuit of the original bank and stockade marks the beginning of the refortification of the castle in accordance with the ideas of the late thirteenth century. The awkward way in which the two western towers are brought close to the foot of the earlier bank can only be understood when the pre-existence of this defence is realised, while the simple character of the gates, so different from the elaborate gatehouses of the normal concentric castle, must be similarly explained. The south-east tower was designed for occupation, but the hall and other structures of the earlier castle probably remained in use. Most of the dressed stonework of this period has disappeared, but the few remaining details are of thirteenth-century character, and this taken in conjunction with the plan suggests that the construction of the inner ward was carried out by Payn de Chaworth, c1275.

The replacement of the older buildings by a new hall, solar and kitchen and the erection of a chapel, form the next stage in the development of the castle. The different character of the masonry and the unusual position of the best-appointed private apartments in the southeast tower behind the screens of the hall prove that these additions were planned and built after the completion of the inner ward. The mouldings of the doors and windows belonging to this stage are all of late thirteenth- century character, and none need be earlier than 1300. The absence of glass grooves in the trefoiled lancet windows of the hall and chapel is also an early feature. It is perhaps unlikely that an extensive building programme would have been carried out during the minority of Matilda de Chaworth (1283-98), and a date before the death of her father Patrick is earlier than the architectural evidence would justify. The work should therefore in all probability be attributed to c1300, directly after the marriage of Matilda to Henry of Lancaster.

The design of the inner ward at Kidwelly presupposes an outer defensive zone, and the replacement of the original stockade with a stone curtain followed the completion of the living-quarters. The main gatehouse, the lesser northern gate, the outer curtain with its four flanking towers, and the mantlet between the north-eastern tower and the chapel, are part of a single design intended to bring Kidwelly into line with other concentric castles of this period. The position of the elaborate gatehouse, which could in case of need act as a separate defensive unit, on the line of the outer instead of the inner curtain is an abnormal feature imposed by the pre-existing layout of the site. The new masonry curtain of the outer ward was higher than the original stockade and necessitated the addition of a further storey to the towers of the inner ward in order that they might still command the outer defences. The details of this reconstruction are all of early fourteenth-century character, and the fortification of the outer ward must have been carried out during the first quarter of that century. On the completion of this work the surplus material from the bank and other rubbish was used to level up the interior of the castle, which it now covers to a depth varying from 2ft to 7ft. The closing of the open gorges of the towers of the outer curtain marks the last stage in the military development of Kidwelly and may also be ascribed to the fourteenth century. The existing gate in the south wall of the town also appears to be of fourteenth-century date, and it might be expected that the erection of these walls would have followed rather than preceded the reconstruction of the castle. But grants of murage in 1282 suggest an earlier date, and the solution of this problem must await further research.

There is reason to think that the great gatehouse may have stood unfinished through most of the fourteenth century and that the opening of a quarry `for the work of the new tower', recorded in 1388-9, may mark the beginning of the works needed to complete it. Ten years later, between 1399 and 1401, when the lord of the castle had become king in the person of Henry IV, there is another record of nearly £100 being spent ‘on the new work of the tower over the gates of the castle’, the near completion of which at this time can be inferred from an order of 1402 for it to be roofed in lead. But in the autumn of the following year the castle was besieged by the Welsh rebels, aided by a naval force from France and Brittany, and subsequent documents leave very little doubt that they succeeded in setting fire to the gatehouse and inflicting serious damage to it; so much so that between 1408 and 1422 a further £500-£600 had to be spent on it, and only in the latter year was it finally roofed with lead shipped from Bristol. A superficial indication of the new work may be seen in the patches of thin flat slabs which contrast with the irregularly coursed boulders of the original masonry. To it belong the triple machicolis high up on the outward face, the upper part of, the wall towards the courtyard, all the square-headed windows, the rectangular stair turret at the north-west corner, and the stone vaults inserted at different levels in the flanking towers as a protection against fire.

The last significant addition to the castle, probably made towards the end of the fifteenth century, was a large hall placed on the west side of the outer ward. This was connected with a kitchen in the south-west corner of the inner ward by the enlargement of a thirteenth century embrasure so as to form a passage-way. Buildings placed against the outer curtain reflect the increased complexity of life in the Tudor period, but the provision of an entirely new hall and kitchen suggests that the earlier hall was already ruinous.

KIDWELLY is one of the Norman foundations strung out along the coastal plain of South Wales. There is no evidence of any occupation before the grant to Roger, Bishop of Salisbury, and even if a small Celtic settlement existed, it has been without influence on the subsequent development of the site. Like many other Norman settlements which were then continually threatened by a hostile attack from the mountains, Kidwelly stands at the head of an estuary where the river was still navigable at high tide. This situation ensured a line of communications when the castle was surrounded and the roads cut by a Welsh rising.

THE SETTLEMENT

From the bridge the road to the castle leads through the defences of the old town. The walls have mostly disappeared, but the main gateway, apparently of early fourtheenth-century date, still spans the road. The line of the defences can still be traced by the earthen bank which preceded the walls. It encloses about eight acres including the castle which it surrounds on all sides except the east. A transverse ditch running west from the castle moat divides it into two nearly equal halves, of which it is probable that only the southern was walled. The defences consist of an earth bank and ditch except on the east, where the steep scarp above the river formed a natural protection. The date of these ramparts is not certainly known, but as the walling of the southern part is to be connected with grants of murage, c1280, and the erection of the gatehouse during the following century, there is good reason for suggesting that they form part of the original Norman settlement.

Although the medieval buildings within the walls have been replaced with modern houses, the line of the existing roads probably follows the original layout. Another feature of exceptional interest is the ruins of the medieval mill which with the contemporary weir and leat can be traced on the low ground between the old town and the river. At a comparatively modern date this was replaced by a more efficient type of mill, which in its turn is now disused.

THE CASTLE

The gatehall was closed at each end by a double gate preceded by a portcullis. On entering the outer ward the inner gate is seen in the centre of the south curtain of the inner ward. It is a simple structure, a mere arch through the curtain defended by a gate and a portcullis. The inner ward is rectangular with a circular tower at each angle. The earlier hall and solar lie on the east side, with the kitchen between the former and the gate. Behind the hall is the chapel, contained in a bastion projecting down the steep scarp above the river. A later kitchen occupies the south-west corner of the courtyard. The outer defences form a semi-circle based on the river. The masonry curtain and towers date from the fourteenth century, but they follow the lines of the original rampart which was disclosed by the 1931 excavations, and which can still be traced on the north and west sides. The Tudor hall standing free on the west side and several later buildings have encroached on the already restricted area of the outer ward. On the east, where the outer defences do not surround the inner ward, a small mantlet joins the north-east tower and the chapel.

The earlier domestic buildings

The south-east tower

The chapel

The north-east tower

The north-west tower

The south-west tower

The gatehouse

The principal chambers were on the first floor. The inner side formed the hall lighted by windows with cusped heads looking into the courtyard. Early in the fifteenth-century these were enlarged and traces of the hood-mould surmounting the new rectangular-headed windows can be seen. Originally this hall could only be reached by the inconvenient stairs already mentioned, but later, probably as part of the general remodelling of the gatehouse, a wide external stairway was added leading up from the outer ward and entering the hall by a doorway inserted behind the screens. To the east of the hall lay the vaulted kitchen with a large fireplace and oven. The towers were occupied by two smaller rooms, while a third filled the space to the west of the hall. From the kitchen a door led to the wall walk along the eastern ramparts, while that to the west was reached from a small lobby opening out of the narrow room beyond the hall. Above the hall and stretching over the vault of the kitchen was the solar, reached originally by a small staircase contrived in the inner wall of the hall. The rest of this storey contained three smaller rooms corresponding to those on the floor below. With the exception of the solar all the rooms on the highest storey are covered with flat stone vaults, but the former, like the hall and the smaller chambers on the first floor, had wooden ceilings. From the hall the circular stairs in the added northwest turret led up to the solar and the roof.

The outer curtain

The later domestic buildings

To the same period belong the buildings standing against the east, north and west curtains of the outer ward. The purpose of the first, a large chamber which is very similar in appearance to the late, hall, cannot be determined. The building to the west of the north gate has a large oven built in the thickness of the side wall and was the bakehouse. The remaining structure by the south-west towers has two long narrow rooms, one of which is provided with a fireplace.

A report by Sir Cyril Fox and the writer, giving a full account of the 1931 excavations and detailing the evidence on which this guide is based, will be found in Volume LXXXIII of Archaelogia.