Chapter XXIX: Part 29

Between the door and the tower a sort of oriel has been corbelled out at an early period, possibly to defend the approach. At present it has a loop in its basement, and two Tudor windows above, and is surmounted by a small gable of the same date.

Round the base of the wall is a terrace, about 10 feet wide and 8 feet high, above the slope of the mound. The retaining wall is in part old, and is supported by short stout buttresses, apparently of Decorated date. This wall has been patched, and in places rebuilt, in Tudor and later times, and its low circumscribing parapet is mostly modern. What it was, or when constructed, is uncertain. It may have carried a low parapet, a sort of chemisette, defending the base of the keep wall, and intended to supplement the ditch at the foot of the mound. In the last century it was crossed on the south side by a wall, with a gate in it, but this probably was not original.

The buildings within the shell are next to be described. The entrance lies beneath a sort of gatehouse, of the date of the other buildings, having on the right the tower court, and on the left a small court having the outer wall for one of its sides, and in that wall a small doorway, whence a mural staircase ascends, winding with the wall, to the battlements. The inner entrance, opening to the private apartments and hall, is a rather elaborate doorway of the style of James I. This opens into a passage or lobby, having on the right the great hall, on the left a buttery, or modern housekeeper’s room, and in front the way to the kitchen.

The hall lies north and south, and occupies nearly the centre of the enclosure. At each end of it are distinct suites of apartments, having no direct communication save through the hall. On its east side is the tower and tower-court; on its west side the kitchen and kitchen-court.

The hall is 40 feet by 20 feet, the end of honour being the north. The entrance-door is on the middle of the south end. The north end is blank. Of the east side, about the south half is occupied by a large oak window-frame, with square apertures, glazed, reaching from about 5 feet high to the eaves of the roof, and looking into the tower-court. In the same side at the north end a door leads by a stair to the northern apartments and the tower. On the west side, in the centre, is a large fireplace; to its north a window similar to the other, but rather smaller; and to its south a door, opening on a stair, leading to the southern apartments.

The roof of the hall is of open-work, supported by two detached and two engaged principals, one against each gable. The stone floor has lately been replaced by boarding. The three doorways are round-headed, of the age of James I. The aspect of the hall is gloomy, the roof heavy and unskilful, the windows unpleasing, and the walls thin and of brick. There are here four good wrought-iron candelabra about 6 feet high.

The southern apartments upon the basement are, with the exception mentioned, private. On the first floor are the library and drawing-room, and one or two private rooms. The library is panelled with oak to the cornice, and along the upper tier of panels are painted Ferrers and his matches. The fireplace is very handsome, and above it is a large atchievement, carved in black oak, of Ferrers and his quarterings, crest, supporters, and motto. The drawing-room, also panelled and larger, has a good fireplace. Each room has a large Tudor window to the south, and is exceedingly cheerful. The second floor is not shown.

The northern apartments lie between the hall and the north wall, in which the windows are pierced. The basement is composed of cellars, on the ground level, opening from the tower and kitchen courts. The first floor contains a large drawing-room and two smaller lateral rooms. All are dismantled, stripped of the panelling, and in a state of decay. The second floor contains bedrooms, also disused. From this floor a door opens upon the rampart of the _enceinte_ wall, where it is seen to be 7 feet thick, and to have a parapet about 10 feet high. Below is the kitchen-court, and against the wall may be seen a sloping water table marking a roof, probably of an early kitchen. At the other end these rooms communicate with the tower, the floors of which are ruinous. The tower basement is entered from the court. The stairs throughout are in rectangular staircases, and each step is a heavy log of oak.

Between the hall and the west wall is the kitchen, fitted up with a modern roof and appliances, but, no doubt, on an old site. At one end of it is the well, about 5 feet diameter, lined with ashlar, and descending to the level of the river. At the other end is the kitchen-court, in which is seen a closed doorway leading into the basement of the northern apartments. It may be of Decorated or early Perpendicular date.

Looking to the rectangular and oblong outline of the defences, the cross-roads, and the position with one open side upon a river, it is difficult not to regard Tamworth as of Roman origin, or as modelled by Roman occupation. The Icknield-street, in its course from Birmingham towards Lichfield, passes, it is true, no nearer to Tamworth than Wall, the ancient Etocetum, six miles distant, where it is crossed by the Watling-street; but this latter, in its course to Atherstone or Mancetter, passes through Fazeley, only a mile south of Tamworth; and had it not been for its considerable angle at Wall, it would have passed directly through the town. Nevertheless, Roman towns are generally indicated by history or tradition, or the remains of Roman masonry, or articles of domestic use, and these evidences appear here to be entirely wanting.

But whatever may be the origin of the rectangular bank and ditch, there can be very little doubt but that the mound and platform of the castle were the works of Æthelflaed or her Saxon predecessors, the one to support the usual timber stronghold of the Saxon thanes, the other for the huts and sheds of their retainers and their cattle. Probably a ditch included both mound and platform on the three landward sides, and both these and the river front were strengthened by a palisade. As no mention is made of the town walls, no doubt a similar defence crested the bank all round. This is the arrangement well known to have been usual, both in Normandy and England, in the centuries preceding the Conquest, and a good and much earlier local example of it was given by the Romans at Wall, where a few years since the palisades were discovered preserved in a morass which formed their defence in front.

The Normans seem to have begun by building the enclosing wall, the remaining part of which is certainly Norman, probably early, and older than the keep. The domestic buildings were probably at first below the court at the foot of the mound; afterwards the keep seems to have been filled up by buildings. The main entrance was evidently by the upper gatehouse from the town.

The curtain wall cannot be much later than the Conquest. It is clear that it never was prolonged across the top of the mound, as the slope of its rampart walk only points to the level of the top; probably, therefore, when it was built there was a structure of some sort upon the mound. It is also uncertain whether the curtain recommenced on the opposite slope of the mound, and was continued down by the present lodge towards the mill, in which case the area of the castle would have been about 100 yards across.

The present shell, upon the mound, seems later than the curtain. The entrance door and the middle band of the wall seem additions of one age, perhaps of the reign of John or Henry III. The terrace and the oldest parts of the interior are probably later, perhaps of the time of Edward I. or II.

Leland, writing in the reign of Henry VIII., says, “The castle of Tamworth standeth on a meetly high ground, at the south part of the towne, hard upon the ripe of Anker, at the mouth of it. The base court and great ward of the castle is clean decayed, and the wall fallen down, and therein be now but houses of office of no notable building. The Dungeon hill yet standeth, and a great round tower of stone, wherein Mr. Ferrers dwelleth and now repaireth it.... The town of Tamworth is all builded of timber.”—[Itin. iv. 122.]

The base court evidently was the platform, and the great ward no doubt included all the ground south of the present curtain, and between the mound and the mill.

In the east window of the church was a painting, of which a copy is preserved by Dugdale. It represents the Conqueror enfeoffing Robert Marmion with the castle. The king stands in front of a considerable building, fronted by two drum towers of two stories, with conical roofs, and connected by a curtain. In one tower is a gateway, and behind the two are seen, in perspective, the stepped gable of a hall, and the chisel-pointed roof of a rectangular tower.

On the proper right of the king and of the building, in the distance, is the mound, crowned with a wall. This is, no doubt, a representation, rather exaggerated, of the castle, as it stood in the later Plantagenet times.

Dugdale, writing after the civil wars, says, “The Norman castle stood below, towards the mercate-place, where the stables now are.” The mercate-house, rebuilt in Queen Anne’s days, remains; the stables are removed to the other side, towards the bridge. The Norman castle means the domestic buildings.

13 Edward I., Philip Marmion had made a certain “pour presture,” or encroachment, to the injury of the king’s market, on either side of Tamworth Castle, containing a width of 8 feet and a length of 40 feet.

The Mr. Ferrers whom Leland mentions was probably Sir John Ferrers (died 1576), who married Barbara Cockaigne; and the domestic buildings now standing were his work, and, perhaps, the work of his son and grandson.

What originally stood within the shell is unknown, probably some lean-to houses of early English and Decorated date, which were removed, or nearly so, for the present structures. These latest works are mainly of brick, with freestone dressings and door casings.

TAUNTON CASTLE.

Taunton Castle possesses an interest in the eyes of archæologists which its present appearance and its Norman history may not seem to justify, but which depends upon the fact that it is of English and not Norman foundation, that it dates from a period nearly two centuries earlier than any other fortress mentioned in the Saxon Chronicle, and not only is the date of its construction approximatively known, but its existing earthworks, though mutilated, are beyond question original.

The castle stands upon one of the many low hummocks of gravel, often with a base of red marl, which rise out of the extensive fen lands of this very singular district, and which, before agriculture had drained the marshes, were even more inaccessible (in military phrase were stronger ground), than even the hill fortresses of the upper country. The Thone, the river whence the town derives its name, rises by many and copious tributaries over a wide sweep of country, north, west, and south, and traversing the low land, which though neither so wet nor so extensive as many of the adjacent levels, was yet broad enough and marshy enough to serve every purpose of defence.

Here, upon the right bank of the stream, Ine, the celebrated leader and lawgiver of the West Saxons, is reputed to have established himself in the year 702, while engaged in securing his frontier against the western Britons, who, under the leading of Geraint, still maintained a footing in the broken ground east of the Tamar, upon Exmoor and among the Brendon and Quantock Hills, holding probably the camps which still remain, but little altered by the lapse of a thousand or eleven hundred years.

This seems to be the origin of the town of Taunton, and here, upon the edge of one of the inosculating branches of the sluggish stream, Ine founded his castle by throwing up banks of earth girdled with deep and formidable ditches, and no doubt further strengthened by stockades of timber, or at best by walls the workmanship of which scarcely deserved the name of masonry. Such as it was it was destroyed, that is, burned, by Queen Æthelburh in 722, who probably however left the earthworks, the better part of the defence, much as she found them.

The spot selected, resting upon the river, is covered by a loop, which has been converted into a mill stream, working a mill placed a little below the castle. This river or north front is tolerably straight and about 180 yards long. The west front, about 168 yards, is formed by what seems to have been a tributary stream called the Potwater, which here joined the river nearly at a right angle. The south and east fronts, of 340 yards, were formed by a curved watercourse, probably artificial, which connected the tributary, by a second junction, with the river, and thus completed the circuit of the defence. The enclosure was thus a sort of quadrant, the river and the brook being the radii, and the curved ditch the arc. The area thus enclosed measures about seven acres, and lies between the river and the town, which covers its east and south sides.

Within this area, occupying its north-east corner and about a quarter of its extent, is the inner court or citadel of the place, roughly rectangular, and measuring about 123 yards east and west, by 73 yards north and south. Its east and north faces rest upon the main ditch and the river, and its south and west faces are covered by a curved ditch, artificial, which gives the eastern outer ditch a second connexion with the river, and divides the outer, called “Castle Green,” from the inner court. The position was a very strong one, having the river, and beyond it a morass, towards the north, or threatened side, and to the south a ditch, in part double, and always filled with water.

The inner court is further subdivided into two parts, of which the eastern half seems to have been raised into a sort of platform upon which probably Ine’s actual residence was placed.

The late Mr. Warre, a very great authority in Somerset earthworks, speaks of a mound here, but as I cannot make out that there is any record or tradition of a mound in the technical sense, that is, a conical heap of earth, probably he calls by that name the very considerable bank and contiguous platform of earth, much of which is still seen. What occurred here, and by whom occupied, or what changes took place between the reign of Ine and the end of the eleventh century is not known, but the Normans, accustomed, as far as practicable, to occupy the Saxon seats, soon perceived the advantages held out by the position and earthworks at Taunton, and William Gifford, who held the lordship as Bishop of Winchester in the reign of Henry I., seems to have decided upon building a regular castle. His successors, bishops of Winchester, were much here, and the castle received much addition at their hands, especially in the early Decorated period, of all of which traces more or less considerable still remain. The outer ward is traversed east and west by a road upon which were two gatehouses, of which the western was till recently represented by a fragment of wall and a stone bridge across the moat. Traces of a barbican, in part of timber, were discovered a few years ago while digging on the counterscarp. Of the eastern gatehouse the remains are still considerable. It was of large size, the entrance passage being 60 feet deep, with portals at each end and at the outer end, and a large square portcullis groove. The upper floor contained a fine room, of which on the north side there remain two windows in the early Decorated style, which is that of the whole gatehouse. The gateway was placed just within the ditch, on the counterscarp or town side, on which some foundations, probably of a barbican, were laid open a few years since. The wall of the outer court is gone, save a small fragment on the south-west quarter, neither are there any of the ancient buildings remaining within the area, Bishop Fox’s school, the oldest of them, being later than the period when the defences were of much value.

The defences and contents of the inner ward are less imperfect. The masonry here did not extend actually to the river, the immediate bank of which, as at Leicester, is very low, so that the enclosed ward occupied only about two-thirds of the whole moated area. The walled part is roughly triangular, the base being the east side, and the truncated part open to the west. This area seems further to have been divided by a cross wall into two parts, the keep, hall, and gatehouse being in the western, and in the eastern, the earthworks, which favours the notion of this having been the old English citadel. These earthworks are two banks along the east and south fronts, expanding at their junction into a rectangular platform of about 80 by 120 feet. The banks have been used as terraces or ramps, the Norman wall having been built against them and along the river edge of the ditch. These banks are about 18 to 24 feet broad and about 10 feet high. Along the east face about 150 feet of the original wall remains tolerably perfect, and is about 25 feet high outside. This is returned along the river or north front, and near the angle is a buried arch at present invisible, and which may have been a postern or a sewer. From the south face the wall has recently been removed. At the south-west corner of this court is a dwelling-house, part of the wall of which is old, either Norman or Edwardian.

The smaller or west court contains the chief remains in masonry, and of these the most remarkable is the keep. This is a well-defined though mutilated tower, standing upon the _enceinte_ wall, of which it forms the north-west angle. It is rectangular, 50 feet north and south, by 40 feet east and west, with walls about 13 feet thick. There is no chamber below ground. The basement is vaulted with a heavy barrel vault, apparently original, and round-headed. Outside are flat narrow pilaster strips, dying into the wall at about 30 feet. There are traces of Norman loops in the wall, which may have been 50 feet high, and probably included three stories. At the north-east angle is a well-staircase leading to the battlements, probably in part an Edwardian addition. The entrance is most likely to have been in the south face, no doubt on the first floor, though there is nothing left to show this.

From the keep, along the north front, the original, though much mutilated, Norman wall, with its flat pilasters and the jamb of one original window, crests the rising ground, as at Leicester, about 50 feet from the river, and, also as at Leicester, evidently formed one side of the hall. At the end of the wall, about 140 feet from the keep, is a postern, with a segmental arch, possibly in substance Norman, though mutilated.

In the centre of the south front, but at the south-east corner of this section of it, is the gatehouse, a rectangular structure, with an Edwardian portal, and some Perpendicular additions, square portcullis grooves, gates, and lodge. In the front are seen the holes for the chains supporting the drawbridge, now replaced by a permanent structure. Above the entrance passage is a chamber.

Right and left of the gatehouse the curtain extends about 70 feet, terminating a short time ago in bold drum towers, of which one is gone, and the other caps the south-west angle of the ward, and connects this front with a short curtain leading to the keep. Against this wall stands a line of buildings ranging with the gatehouse. Opposite, against the north wall, is the hall, modern as to its inner wall, fittings, and roof, but very evidently occupying the site of the original Norman hall and domestic buildings.

The south-west drum tower has been rebuilt or faced, but evidently represents the Edwardian or early Decorated works that replaced the old Norman curtain. The ditch along the west, and part of the south fronts of this ward, has been recently filled up. The drum towers, curtain, and keep stood on its edge, and formed its scarp.

Here, then, we have a combination of earthworks dating from the commencement of the eighth century; walls and keep the work of the early part of the twelfth century; and towers and gatehouses towards the end of the thirteenth century or early in the fourteenth century. Bishop Langton executed some additions here in 1490, and placed his arms outside the inner gatehouse. In 1496, the castle was taken by the Cornish rebels who rose against the close taxation of Henry VII., and here massacred the Provost of Penrhyn. Bishop Horne made further repairs here in 1557.

In the Parliamentary wars, Taunton was first occupied for the Parliament, then taken by Lord Hertford for the King, and finally retaken for the Parliament by Blake, who held it against a far superior force. The infamous Jefferies held the “Bloody Assize” in the present hall.

It has been thought that Ine’s castle was confined to the inner ward. No doubt his strong house was there, but the whole enclosure is not larger than Framlingham or other Saxon holds.

The absence of a mound is rather peculiar, and it is remarkable that the Normans should have placed this keep on the lowest ground. Altogether, looking to its very curious though scanty remains, and its very ancient history, Taunton Castle is a work of unusual interest, and it is much to the credit of the Somersetshire archæologists that the keep has been converted into a museum, and the walls and earthworks converted into an embellishment to the ancient town to which the castle unquestionably gave rise.

THURNHAM CASTLE, KENT.

Thurnham, called also, from the hill on which it stands, “Godard’s,” Castle, near Maidstone, is a curious example of a Norman castle placed upon what is evidently a British camp.

The camp crowned the high point of a very steep spur, which juts out between a depression on the one side and a small deep combe on the other, in the great escarpment of the lower chalk, about four miles east-north-east of Maidstone. The earthworks were formed by scarping the central knoll, and perhaps raising it a very little, so as to form a slight mound, and thus especially strengthening its weak sides, those towards the root of the spur. On the lower or Thurnham front the defences, naturally strong, are reinforced by a ditch and bank, placed some little way down the hill, far below the body of the fortress, and intended to command the road which here winds up the ridge from the village and church of Thurnham, at the foot of the hill. The ground within this outer defence has been extensively opened for chalk, and is so disfigured by heaps of quarry rubbish that but little can be accurately ascertained of its ancient dispositions.

The Norman castle occupied a platform close west of the mound, and probably included within the British camp. Here stand the remains of the gateway and court, but as a trace of masonry is still seen upon the mound, it may be that it was included in the _enceinte_ wall, or that upon it stood one of the circular or polygonal shell keeps which sometimes, with the Normans, took the place of the ordinary square keep, especially where there was an earlier mound to be fortified.

The ruins are not considerable. They are composed of the two parallel walls of a gatehouse, having on either hand two large round-headed recesses, dividing the passage into two bays, and there are besides two small lodge doors, also round-headed, on the east side.

Westwards from the gatehouse runs a low curtain wall, about 13 feet high, and 4 feet thick, for about 80 feet, ending in a broad flat buttress, perhaps the base of a square tower. The wall thence may be traced southwards, along the edge of the steep, whence it seems to have been continued towards the mound. A hollow way, cut in the chalk, winds from below, beneath and close to the west of this wall, and, making a bend, enters the gatehouse from the north. There are no traces of ashlar. Much of the wall shows a face of coursed flints. The work may be late Norman.

On the summit of the knoll is a depression in the soil. This part of the work is so covered with thick bramble and underwood that it cannot be very accurately examined.

Thurnham or Turnham occurs in Domesday, and was one of the numerous manors given by the Conqueror to Bishop Odo, and held under him at the survey by Ralph Curbespine. It then contained a church, and had been held under the Confessor by Sbern Biga.

On Odo’s fall, 19 W. C., it was granted to Gilbert Maminot by the tenure of castle guard under Dover Castle. The holders under Maminot were a knightly family, who took their name from the place. Robert de Turnham held it _temp._ Henry II., and founded Combwell Priory. Possibly he built the castle. Robert had Robert, who died _s.p._ 13 John; and Stephen, who died also _s.p._ 16 John.

In the reign of Edward I., Thurnham was held by Sir Roger de Northwode, who died 13 Edward I., leaving John, who married Joan de Badlesmere, and died 14 Edward II. Their son John died before his father, leaving Roger, who had Thurnham, and married Juliana, daughter of Geoffrey de Say, chief lord of the manor. Their son, Sir John, third baron, died 2 Richard II.; and his son, Roger de Northwode, who was never summoned to Parliament, alienated Thurnham, and died _s.p._, leaving a brother.

The Northwodes are said to have resided here, but the castle is thought to have been dismantled at an early period. Thurnham Castle is not named in the Ordnance map.

TICKHILL CASTLE, YORKSHIRE.

Tickhill is a place of high antiquity, and both before and for some centuries after the Norman Conquest its importance as a strong place and the head of an extensive lordship was very considerable. Mr. Hunter, the accurate and accomplished historian of the district, suggests “The-Wickhill,” in allusion to the village mount, as a probable etymology for the name, and cites “Thunder-cliffe,” or “Th’ Under Cliffe,” as an analogous case. It seems that near Sheffield Castle was a small green called “The Wick-er,” and “Ticken-hall,” near Bewdley, was the seat of an early fortress.

Tickhill, however, though obviously an early name, is not recorded in Domesday, but is thought to be included in Dadsley, a name still extant in the immediate neighbourhood. In “Dadesleia, Stantone, and Helgæli,” Elsi and Siward held eight carucates,—Roger de Buisli held seven in demesne; there were also thirty-one burgesses, a class whose presence has been held to indicate a burh or castle. Roger de Buisli was tenant in chief of these and other manors, comprising the Honour of Tickhill, a division certainly based upon an earlier fee, of which Tickhill was the chief seat. The Norman Honour numbered sixty-five and three-quarters knights’ fees, and extended from Yorkshire into the shires of Derby, Lincoln, Notts, and Leicester, including one manor in Devon. Tickhill, which seems to have been sometimes called Blythe, which, however, was also the name of a place in the adjacent part of Lincolnshire, was the chief seat of the powerful house of de Buisli during their somewhat brief career.

Roger de Buisli received Tickhill from the Conqueror, who erected it into an Honour in his favour. Roger had a choice in his wide Yorkshire domains of three ancient English seats, Laughton-en-le-Morthen, Mexborough, and Tickhill. He selected Tickhill; and the two other sites retain their English earthworks, unaltered by Norman masonry, and changed only by time. Laughton seems to have been originally superior even to Tickhill, probably as the residence of Earl Edwin, being named in the Domesday Survey. Roger himself probably fortified Tickhill with masonry, the gatehouse and much of the wall being apparently his work. The foundations of the shell-keep look rather later, but may also be of this date. He was also the founder of Blythe Priory, in 1088. He died 1098, and was succeeded by his son Robert, who died childless in the reign of Henry I.

The descent of Tickhill now becomes obscure. Roger had a brother, Ernald, who held six fees under Tickhill; and a sister, Beatrix, from whom descended the Earls of Eu. On Robert’s death, however, Tickhill was claimed by Robert de Belesme, as the next heir; and, as he was powerful, and supported his claim by payment of a heavy fine, or bribe, he succeeded. On his death, however, King Henry stepped in and took possession. The castle remained for some time, with some brief intervals, in the Crown. William Fitz-Godric held it in 1142, and Stephen, Earl of Eu, for a time. Ralph, Earl of Chester, had it in 1151–3. Henry II. seems to have settled it upon Eleanor, his queen, who founded the “Chapel of St. Nicholas within the walls.”

It descended to Richard I., and in his absence was seized by Prince John, and besieged for Richard by Pudsey, bishop of Durham. When John inherited it as king, he annexed his mother’s chapel to the chapter of Rouen. John was frequently at Tickhill, which is remarkable, as there was no park or chase annexed to it. He was here six times between 1200 and 1216 for at least eleven days. Early in John’s reign, however, the Earl of Eu, being powerful, claimed Tickhill as the husband of Alice, heiress of Henry, Earl of Eu, and representative of Beatrix, sister of Roger de Buisli. Ralph de Issoudon, or De Lusignan, her husband, an earl in her right, in 1197, seems to have been son of Geoffrey de Lusignan, who was to marry a daughter of King John, in consequence of which the king agreed to restore Tickhill to Ralph and his wife, and John de Bassingbourn was ordered to give possession, which was supplemented by an order to the same effect, 1 Henry III.

Meantime, another claimant appeared, in the person of Idonea, representative of Ernald, Roger de Buisli’s brother, and who had married Robert de Vipont, a baron much employed by John, and not unfrequently in connexion with Tickhill. Thus, in 1204 he was concerned in certain repairs at the castle, as also in 1206, which included a barn and stables. In 1207 he was to be paid for these repairs, and he was also employed upon the king’s castles of Nottingham, Bolsover, the Peak, and Scarborough. In 1208, five dolia of red wine, such as would keep, were to be sent to him at Tickhill, for the king. Idonea, Vipont’s wife, held by descent six fees in Tickhill, and now claimed the rest against Countess Alice, 6 Henry III. Alice had the best of it; but in 9 Henry III. she went abroad, probably to her Norman estates, and in consequence Tickhill seems to have lapsed to the Crown.

Henry III., when king, granted it to Prince Edward, who, in 1254, settled it upon Eleanor of Castile, but in 1259–60, Edward de Lacy, Constable of Chester, used the phrase, “Baronia mea de Tikehull,” as though absolute lord. Prince Edward, however, granted it to his cousin Henry, son of Richard, King of the Romans, in 1263.

In 1296, John, Earl of Eu, revived the family claim. He was grandson of Alphonso (son of John, King of Jerusalem), by a daughter of Countess Alice. His claim was speedily set aside, he being an alien. This was the last assertion of the right, but a century later it was remembered, when Henry V. created William Bourchier Earl of Eu and Lord Bourchier of Tickhill, titles only, not connected with the property, which remained in the Crown.

In February, 1322, the castle was besieged for three weeks by Thomas of Lancaster, and this siege was one of the charges brought against him on his trial:—“Et misit homines suos ... ad obsidendum castrum domini Regis de Tikhull; et quædam ingenia, ad projiciendum petras grossas super castrum prædictum, et homines in eodem castro ex parte domini Regis existentes; qui quidem proditores castrum illud, per tres septimanus continuè insultando et debellando, obsederunt, et quosdem homines Regis ibidem interfecerunt.” The castle was defended gallantly by Sir William de Anne, and relieved by the king in person. Tickhill was again settled upon a queen in the person of Philippa, who died in 1369. In 1362, Edward exchanged Tickhill with John of Gaunt, against the honour of Richmond, and it descended with the other estates of the duchy of Lancaster. In the Parliamentary struggle it was held for the king, but surrendered after Marston Moor, and was dismantled. When thus held, its outer defences were the moat and the palisades on the counterscarp. The foundation of the chapel was dissolved, 1 Edward VI. The castle now belongs to the Earl of Scarborough.

Tickhill Castle is an excellent example of a pre-Norman or English earthwork, composed of mound, fosse, and lower ward, converted into a Norman castle. It exemplifies exactly the manner in which the Norman engineers treated earthworks of this description, and how such works gave rise to one of the two great types of a Norman castle, that with the shell-keep. Tickhill is Laughton on a larger scale, the only difference being that the ditch of the mound is not carried wholly round it, but is wanting towards the attached area. Either it was never formed, or, what is not improbable, was filled up when the Norman works were constructed, or, as at Cardiff, at a much later period. Something analogous to this seems to have taken place at Kenilworth.

In the original construction of this fortress advantage was taken of a knoll of soft sandstone rock to form the base of the mound. This was scarped, and the ditch dug, and the material employed in forming the upper two-thirds of the mound. A modern cave in the side shows this natural base. The castle is composed of the mound, and a court or ward appended to its western side, the whole included within a ditch. The mound is conical, about 60 feet diameter at its table-top, and about 60 feet high, above the ward. The ward is a rounded and more or less circular area, save where it touches the mound, and includes about one quarter of its circumference. The exterior ditch follows the figure of this ward, and of the uncovered three-quarters of the mound: hence in plan it resembles somewhat a figure of 8, and it is this notch in the outline that makes it probable that the mound ditch was once complete, and the two parts of the fortress were, as at Barwick, distinct. The domestic buildings stood in the lower ward, on its western edge, opposite to the mound upon which was the keep. The gatehouse stands on the southern edge of the ward, between the domestic buildings and the keep. The curtain is broken down to the east, but elsewhere tolerably perfect. The ditch is filled up on the same side, and its place occupied as a kitchen-garden and by stables.

Upon the summit of the mound are seen the foundations of the keep, a decagon, the sides of which average 16 feet 11 inches, of which each angle was covered by a flat pilaster of 4 feet broad, and very slight projection. The door seems to have been towards the south-west, of 4 feet 6 inches opening. It lay between two sides of the exceptional length of 20 feet. The wall seems to have been, at the top of the plinth, 10 feet thick. The shell was apparently faced with ashlar. The whole building has been taken down with some care to the top of the plinth, a mere plain chamfer, formerly about 6 inches above the ground, and now covered to its level. Thus the actual dimensions of the plan are preserved, and the position and breadth of the entrance. It is said there is a well within the area, a few feet inside the place of the door. If so, it is at present effectually concealed.

The keep at this time is ascended by seventy-five stone steps in a straight line on the western face. Possibly this was the original approach. If so, the path from the head of the stair must, as at Tamworth, have passed for 20 feet round the outside of the keep. The steps terminate below under the shelter of the curtain. The two ends of the curtain ascend the mound about two-thirds of its height. Probably they were continued to the summit, but no foundations are now seen at the keep level, and the plinth of the keep shows there was no bond. The curtains which thus ascended these mounds were rarely bonded into the keep, and do not seem ever to have risen to its full height. On the contrary, they seem to have only risen to the level of the top of the mound or to the base of those of the keep, the parapet probably being continued so as to stop the passage round the exterior. This seems to have been the case at Tunbridge, Berkhampstead, and Tamworth. At Hawarden the curtain abuts against the keep about 10 feet high, but with no original bond, and with a doorway in it opening outside the base of the keep.

The curtain which enclosed the lower ward is here from 10 feet to 13 feet thick, and 20 feet to 30 feet high, with a plinth at its interior base. It rises out of a bank which forms a ramp or terrace 15 feet broad, on both outer and inner sides. Inside, this ramp is about 8 feet above the court level. Outside, it forms a walk all round the fortress, being carried by a bridge over the gateway, and in a step or notch round the slope of the mound. The curtain is entire from the mound to the dwelling-house, about 240 feet, along the north front, but to the west it is concealed by the house which represents the domestic buildings of the castle. It also remains from the house to the gatehouse, and about 50 feet or 60 feet beyond it along the south front. Towards the south-east about 300 feet are gone, but the last 78 feet, where it again ascends the mound, are tolerably perfect.

Where the outer wall skirts the mound it forms a revetment 6 feet high, and which may have been higher, and crested with a parapet to defend this front.

The exterior ditch is broad and deep, and in part contains water. Formerly it was fed from an adjacent stream, which flowed all round it. Beyond the ditch was a bank of earth, of which traces and portions remain, especially towards the north. It is difficult to say whether there was a second ditch, owing to the encroachments of roads and buildings.

The gatehouse deserves special notice, as an original and early Norman structure. It is 36 feet square, with walls 7 feet 6 inches thick, and has a round-headed gateway at each end, of 12 feet opening, with a plain rebate for doors, but no portcullis or chamfer. The inner space was covered with timber, and there was an upper story. This may have been partially rebuilt; it contains in the wall over the inner door a large Tudor window, probably an insertion.

There is no staircase. The structure much resembles Porchester before the alterations. It is placed upon the curtain, with a bold exterior, and still bolder interior projection. The outer front of the first floor is ornamented with four stiff rude pediments, each a right-angled triangle, with a rude figure at the apex of each and in the hollow angle or gutter, joining each pair. The tympana are filled with small square blocks, each carved with an undeveloped dog-tooth ornament. A plain string marks the division of the two stages, and so far all is Norman.

But although the upper part is unaltered, the lower part has been masked by a Decorated gateway with portcullis groove and pointed arch, while in front of and flanking this arch two walls, 6 feet thick, project 15 feet, and contained between them the drawbridge. Above and upon these, concealing the upper part of the arch of entrance, is a low flat bridge, which carries the exterior walk, or _chemin de ronde_, over the entrance, and from which the grate was worked. Had it not been for this bridge, and its Decorated connexions, it might have been supposed that the _chemin de ronde_ was a mere modern pleasure walk, whereas it is clear that it was a part of the defence, a work covering the foot of the wall, and no doubt strongly palisaded. There is no trace of a parapet.

The gatehouse seems early Norman, probably with most of the curtain the work of Roger de Buisli before 1089. The keep looks much later, but it must have been part of the original design, and possibly the works begun by Roger were completed by his son. In the Decorated period there were probably considerable additions. Probably when the gatehouse was masked, and the bridge thrown over it, the curtain also was repaired, and a new parapet added, and the _chemin de ronde_ formed. Leland speaks of a hall, now gone. Where the chapel stood is not known. A door case which may have belonged to it has been removed and set up inside the gatehouse, and outside is an old oak door of the style of James I., on which are carved the words,—

Peace and grace
Be to this place.

The entrance-way now leads up to the gatehouse across a modern bridge, over the wet ditch. To the south of the place is a tributary of the river Torne or Thorne, which covered that front.

This is one of the most curious castles in Yorkshire, not only for its pure Norman gatehouse, and the undisturbed foundation of its shell keep, but because it shows how the Norman lords availed themselves of an English seat, and how their architects or engineers accommodated their defences to the already existing earthworks. It should be studied in conjunction with Pickering for the general plan and the Norman works, and with Barwick-in-Elmete and Laughton-en-le-Morthen for the general resemblance of the earthworks. Unfortunately there is no plan.

TRETOWER, BLAEN-LLYFNI, AND
CRICKHOWEL CASTLES, BRECKNOCKSHIRE.

Among the passes that communicate between the open country of the upper Usk and the Wye, in the neighbourhood of Talgarth, Glâsbury, and Hay, and the valley of the lower Usk, there are two especially dangerous, and which, therefore, it behoved the Norman settlers to guard with peculiar care. These are the pass of the Bwlch, between Buckland and Cefn Moel, and the valley of the Rhiangoll. The strip of land bordering on the Usk, which flows close to the west of Mynydd Buckland, is very narrow, and in wet seasons was formerly scarcely practicable; and was besides, to a considerable extent, guarded by the castle of Blaen-Llyfni, which was specially built for the defence of the Bwlch, close to the north of which it is placed.

The Rhiangoll rises in Cwm Catwg, a deep, dark hollow in the Black Mountains, and flows down to join the Usk about three miles above Crickhowel. Near its head, where it lies open towards the north-west, it was guarded by the hill-castle of Dinas, and lower down by Tretower. Still lower down, the Usk itself is protected by the castled mound of Crickhowel, and still lower by that of Abergavenny; so that between the seigniorial castles of Blaen-Llyfni, Dinas, and Abergavenny, and the private posts of Tretower and Crickhowel, the incursions of the Welsh of Brecknock and Radnor were liable to be checked in their advance, and altogether cut off in their retreat.

Tretower stands in the meads of the Rhiangoll, on the left bank of the stream, about a mile above its junction with the Usk, at a point where a tump of gravel rises on the margin of, and a little above, the general level of the meadow, once evidently an impassable morass, and even now by no means a favourable specimen of Breconshire farming. The meadow is traversed by several branches of the stream, here reinforced by divers springs and brooklets, which, added to the natural strength of the ground, made the place one of very difficult approach save from the slightly higher ground to the north-east.

Tretower is a chapelry in the parish of Llanvihangel Cwm Du, and seems to have been the private estate of the owners of the castle, who, when the conquests of Edward I. settled the Principality, deserted the small and inconvenient fortress for a larger house a little to its south, now occupied by a farmer, and showing some traces of the Decorated style of architecture.

The castle was a three-cornered enclosure, having two round towers at its northern and southern angles, or the ends of its base, and the keep at its western angle or apex. The two sides are each about 60 yards in length, and the base about 80 yards. The entrance was probably near the centre of the base or eastern curtain, in the line of the present road to the farm-buildings which occupy this outer ward. Fragments remain of the two round towers, and about half of the northern curtain. At the gorge of the southern tower, nearly buried, is seen the crown of a large pointed arch which probably opened into its basement, though its breadth is unusual.

The inner ward, within, or rather capping which, is the keep, is a square of about 23 yards by 18 yards outside, and 17 yards inside measurement, placed at the apex of the triangle; so that three sides of it are exterior to, and one within, the outer ward. It has been destroyed on the north and east sides; but the south side is perfect, and the west very nearly so. The foundations can be traced all round. This ward occupies the whole of a tump of gravel rising about 12 feet above the meadow, which is its defence on the west, south, and east; and it is so built that its lower part serves as a scarp wall, the ground within having been about 6 feet above the narrow bank or terrace outside. There were no towers at the two northern angles, which seem to have been chamfered off with a sort of spreading or buttressed base. From the south wall, at its west end, projects a half-octagonal tower, and at its east end another, much larger. They are but half-octagons, and have no internal projection. The wall was 8 feet thick, and 25 feet high to the battlements. In the remains of the west wall are traces of a recess, perhaps for a loop, a fireplace with a clumsy drop-arch, and higher up a window in the same style. These, in their present form, are late insertions, but probably represent original openings, as on the exterior is a broad flat buttress intended to carry the chimney-shaft. This wall has been a good deal pulled about at an early period. The south wall is exceedingly curious, and though evidently much altered, bears large marks of its original style. Near its west end, in the hollow angle, is a small round-headed doorway opening into a well-stair which led to the battlements, but has been blocked up. East of this, an opening in the wall (probably the place of a second door) shows a flight of steps descending in the wall to another door, also round headed, and now nearly buried. The stairs are lighted by an exterior loop. Higher up, at a different level, is a window recess having bold, heavy, flanking, engaged piers fluted in zigzag, and supporting a fragment of an arch-moulding worked in the same chevron pattern. It may be that this was a double window, or two arches of an arcade, for the fragments of the arch-head have somewhat that appearance.

TRETOWER CASTLE—GROUND PLAN.]

Next, east of this, at the same level, is a doorway of 4 feet by 6 feet opening, in the same fashion and style as the window, with jambs fluted in zigzag and the chevron moulding above. The work is rude, and the ornamentation only partially executed. All these openings have been walled up with early masonry. The east half of this east wall is occupied by a rectangular projection, forming a chamber 10 feet deep by 18 feet broad, which may have been a kitchen, since it contains a fireplace with a round back and gathering up to a cylindrical shaft, and at the ground level is a water-drain; also there are three round-headed loops, of which one is in the west wall, and in the east wall is a small round-headed door, 2 feet 9 inches broad, and opening upon the outer ward; it has no portcullis.

The openings, of which traces remain in this wall, seem to show that the building was originally a rectangular Norman keep, having a basement and a first and a second floor, indicated by the lower door, the entrance to the wall stairs, and the door and window above. The southern projection is probably original, as is the small door in its side. The battlements, of which a part remain on the west wall, are, no doubt, restorations. They are composed of broad merlons with a cruciform loop in each, the cross-arm being very short.

In the exterior of this building, near its ground level, a breach in the masonry discloses a cavity about 14 inches square, which pierces the wall parallel to its faces. This was evidently occupied by a beam, inserted as a tie, as at Brunlas, where also the timber has rotted away, and the cavity been exposed, and has given rise to a number of ingenious conjectures. In the centre of this square enclosure, and leaving only a space of from 7 feet to 9 feet between its exterior and the centre of each side, is the interior keep. This keep is a cylindrical tower, at its base about 43 feet in diameter, and battering inwards to a height of about 10 feet, where it is girt by a very bold, rather more than half-round bead or cordon, which marks the summit of the slope and the first-floor level. Here the tower is about 38 feet in diameter. From thence it continues to batter, though slightly, perhaps 2 feet, to the base of the battlements, which are gone. It is now 70 feet high from the ground level, which originally may have been 5 feet lower, and the tower 5 feet higher, making a total of 80 feet. About half-way up is a plain set-off of about 6 inches, reducing the diameter of the tower by a foot. With this exception the exterior is quite plain. The present entrance to the basement or ground floor is on the north-east side, by a broken entrance of modern date, the making of which has been aided by the first-floor entrance, which is immediately above it, and of which it now makes a part. The tower is composed of a basement and three floors, and its interior diameter at the base and first floor is 20 feet. The floors were of timber, and are gone; and the roof seems to have been flat. The basement, now about half filled up, was about 10 feet high, and the wall at its floor about 11 feet 6 inches thick. It was lighted by two loops in the north-west side, the splayed recesses of which are square-headed, and, as in Canterbury keep, have their inner opening closed above, so that the loop was on the level of the first floor. These loops are stopped up outside, evidently by very early masonry. In the south-west side is a door, of which the top is seen unburied, whence a flight of steps leads in the wall upwards and southward into the floor above, and this was evidently the only entrance to the basement floor.