An Archaeological Interpretive Survey of the Old Castle Scotney Part I

321 AN ARCHAEOLOGICAL INTERPRETATIVE SURVEY OF THE OLD CASTLE , SCOTNEY, LAMBERHURST: PART I – THE MEDIEVAL PERIOD david martin, barbara martin and jane clubb With Contributions by Rupert Goulding Historically, the moated site at Scotney was partly located within the parish of Goudhurst and partly within the Sussex moiety of Lamberhurst parish. It is thought that the county boundary followed the original course of the river Bewl prior to it being diverted southwards in order that the moat could be formed: today the river is contained within an artificial cut which skirts the southern boundary of the moat, but the pre-moat parish boundary was retained until the nineteenth century. Is it significant that none of the known early owners of Scotney are listed in the 1296, 1327 and 1332 subsidy returns for Sussex.1 At a later date, the Sussex return of 1411/12 includes Joan Ashburnham, assessed for a property in Lamberhurst (presumably Scotney),2 whilst the early sixteenth-century owner of Scotney appears in the 1524 Lay Subsidy for Sussex. So, if these later owners were assessed in Sussex, why was this not the case in the late thirteenth- and early fourteenth-century subsidies? Early documents refer to the property either as Courthopes or Scotney, perhaps suggesting two merged adjacent holdings. Could it be that the early owners of the property are missing from the Sussex returns because their early un-moated house was at Courthope in Kent, whereas the later moat house was built on the Scotney part of the property? Whatever the case, the moat at Scotney most likely dates from the fourteenth century – a period during which most of the local moats seem to have been dug – though it has to be admitted that a late thirteenth-century date is not impossible.3 Apart from the comments above regarding site relocation, it must be accepted that at present it is not clear which of the two possible scenarios is correct in relation to the moat at Scotney. Thus, it has to be accepted that the moat could have been excavated around already extant buildings upon the site. The Old Castle at Scotney is an extremely complex site, the interpretation DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 322 of which is not helped by the fact that only a fragment now survives, and even this is largely in ruins. Anything more than a cursory understanding of the building’s form would have been impossible had it not been for plans of the ground and first floor prepared by Edward Hussey immediately prior to demolition in 1837 (both re-drawn by him subsequently). A third plan showing the second-floor layout was subsequently prepared by him based upon memory, but the details are less certain. Furnished with this data, augmented by a series of late eighteenth-century and early nineteenth-century illustrations, and an inspection of parts of the structure not normally accessible, the sequence of the building’s development can be reconstructed with reasonable certainty. The fourteenth-century castle sits on the northern of three islands contained within a single moated enclosure. At this period the smallest, southernmost of the trio may not have had buildings upon it, but the central island did and almost certainly served as an outer court. It was via this that the main house was approached, across a bridge and through a gatehouse. The main enclosure itself was surrounded by a low curtain wall with drum towers at the corners, of which the Ashburnham Tower survives. Across the centre of the island extended the medieval house and its associated service building(s), dividing the site into a forecourt and a back service yard, probably with lesser outbuildings against the wall of the service yard, and perhaps with other buildings at the front. The sequence of medieval buildings may not have been developed at a single date, and no doubt continued to be modified through the fifteenth century, though the next discernible alterations date from the middle years of the sixteenth century (to be covered in Part II in the following volume). Historical Background By the early fourteenth century the property had come into the possession of John de Grofhurst who, in 1310, was granted the right of ‘free warren’ over Scotney Courthope. This may be the date of the Park laid out over land stretching west and north of the moated site, though it is unclear whether the moat itself existed at this time. John de Grofhurst was a member of Archbishop Winchelsey’s domestic household, hailing from Grovehurst in Horsmonden. In 1321 Grofhurst gained permission to build a private chapel at Scotney, suggesting he was resident at Scotney rather than Grovehurst by that date. (A chapel is known to have existed within the moat in 1558.) Thus there is a possibility that the house, and perhaps the moat and island complex, dates from this Grofhurst phase. John de Grofhurst’s widow, Isabel, married John Ashburnham of Ashburnham in Sussex and in 1358 their second son, Roger, inherited Scotney. By this point Sir Roger was already established in Sussex and A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 323 Kent society, having been a commissioner for walls and dykes on Romney Marsh. Between 1376 and 1380 he was one of three Conservators of the Peace in Sussex.4 It is the accepted (though unproven) belief that Sir Roger de Ashburnham built much of the Old Castle at Scotney around 1378. No licence to crenellate has been identified, though the character of the south-western tower, otherwise known as the ‘Ashburnham Tower’ is consistent with such a date and shows a number of similarities to the corner towers at Bodiam.5 Despite these similarities, there is no guarantee that all the fortifications at Scotney dated from this period: the moat could have been dug earlier and there is some evidence which could be taken to suggest that the gatehouse and perhaps the corner towers were added to an existing low curtain wall (see below). Likewise, as already pointed out, it is not known whether the fortifications were placed around an existing house or whether a new site was chosen for the moated house. The general climate of fear of French invasion resulted in a spate of fortifications to both new and existing gentry houses in Kent and Sussex during the fourteenth century. On Roger’s death his widow Joan, and son William, inherited Scotney. William died before his mother and without issue; in 1418 the manor passed to the Grofhurst’s heir-at-law, John Hall, who sold Scotney for 200 marks to Robert Chichele, brother of Henry Chichele, Archbishop of Canterbury. We know the archbishop visited his brother on at least one occasion, when he signed a mandate from Scotney on 3rd April 1418: possibly Scotney was used as a staging post when the archbishop was travelling between his palace at nearly Mayfield and one of his Kentish palaces. Scotney passed into new hands later in 1418 when it was settled on Chichele’s niece, Florence, on her marriage to a widower, John Darell of Calehill in Little Chart. Calehill continued to serve as the principal seat of the Darells, but Thomas Darell, the only son of John by his second marriage to Florence Chichele, founded a cadet branch who henceforth resided at Scotney. Little is known about this first Darell of Scotney, though more is known about his son, Henry Darell. Born in 1465, Henry Darell married Elizabeth Cheney and held Scotney until his death in 1534, aged 69. Henry, described as an esquire, was assessed at £50 in the Sussex 1524 Lay Subsidy, under the Hundred of Lokkesfeld, borough of Wadhurst – in which Scotney was then located. This is the joint sixth highest assessment within the Rape of Pevensey: four other persons were assessed for an identical sum.6 The 1535 Valor Ecclesiasticus records Robertsbridge Abbey holding the manor of Lamberhurst (centred within the Kent part of the parish) but it was leased to Henry Darell, suggesting that Darell was augmenting his estate by farming neighbouring property. He was an associate of Henry VIII’s chief minister, Thomas Cromwell, writing him two letters in 1531 and DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 324 1532 asking for positions for two of his sons. In 1534 Thomas Darell, the eldest of Henry’s sons, wrote to Cromwell with the news of his elderly father’s death. In 1539 Elizabeth Darell thanked Cromwell for the favour he showed her late husband, and asks him to take Richard, her clergyman son, into his service. Henry Darell’s heir, Thomas, was also prominent amongst the Sussex gentry. He was a surveyor for the dissolution of the chantries and a Justice of the Peace between at least 1538 and 1545. In 1540 he was involved in the ceremonial greeting of Henry VIII’s fourth wife, Anne of Cleves, on Blackheath near Greenwich – his name is listed amongst the esquires present. He served as Sheriff of Sussex in 1541 and the same year, alongside Robert Oxenbridge of Brede Place, heard the initial case against Lord Dacre of Herstmonceux. ‘Thomas Darrell de Skotney’ is listed in the muster books of 1544 in preparation for the French campaign. Thomas died in 1558 and his will provides a good record of early Tudor Scotney. A wide variety of rooms are mentioned: amongst them were a Chapel, Study, Parlour, Great Chamber, closet over the Inner Porch, galleries, Maiden’s Chamber, Kitchen and Hall. It is implied that some recent expansion had been undertaken through references to ‘my little chamber called the New Study’, ‘my new study in the tower’ and the ‘new parlour’. One passage is particularly illuminating: ‘I will therefore give and bequeath to the same Alice [Thomas’ widow] all that chambers, places and rooms in Scotney where she and I do lie, the maidens chamber, the tower from the grounds up to the top, the wardrobe, the loft above the new study with easement unto the privy’. The will also informs us that there was a room above the gatehouse (see below), as well as a barn, garden, stairs, back-house rooms and a milkhouse. It also mentions ‘my Mass books, the one in the Chapel and the other is an old written Mass book that is in my Study which was my ancestors’, a ‘great wall hanging over the stairs’, ‘Latin books for a scholars learning’, and ‘cushions made in Flanders’. Interestingly, the inclusion of ‘all my harnesses for war the javelins and pike and also all my sheaves of arrows and bills’ connects Thomas to Henry VIII’s 1544 French campaign. The will provided for his widow Alice’s keeping, specifically mentioning her allocation of butter, cheese, bacon, cider and apples. the buildings As Fig. 1 indicates, little standing medieval work survives at Scotney: all of that which does is likely to belong to the phase of the Ashburnhams’ ownership, most noticeably during the tenure of Sir Roger Ashburnham (d.1390s). Despite the paucity of remains, there is sufficient evidence to give some indication as to the form of the castle during the medieval period. A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 325 Fig. 1 Outline plan of the site showing known medieval features. DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 326 Outer Court How many buildings stood on the outer island during the medieval period is unclear, though based upon its external form the continuously-jettied northern range shown in S.H. Grimm’s illustration of ?1783 could easily have been built towards the end of this phase (Fig. 2). Similarly, the west range, foundations of which were picked up by a geophysical survey conducted in 2008 and which still shows as a slight earthwork, could be of medieval origin. To judge from other sites, these buildings would have formed a base court serving the house on the main island. They are likely to have included at least one barn, other agricultural buildings, stables for the horses of the owner and his guests, and perhaps lodgings for lesser members of the household and/or the servants of guests. Some buildings of this type survive within the outer court at Westenhanger Castle where they formed an un-moated complex standing in front of the castle gatehouse.7 Although not rare, moated outer courts are less common than un-moated versions.8 The remains of good un-moated outer courts survive locally at Knole, Sevenoaks; Ightham Mote; and Sissinghurst Castle, all in Kent. Fig. 2 Detail from drawing, c.1783, by Grimm showing outer island viewed from the north-west (BL Add. MS 5670, f.1). The continuously jettied building in the centre could well be of medieval date, ditto the close-studded building immediately to the left. A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 327 The Main Island Crossing eastwards from the outer court, the main part of the site was accessed by a bridge and gatehouse located approximately half way along the western side of the main island: this side of the island is deliberately canted in order to align it with the outer island. Even ignoring this point, the enclosure is somewhat irregular in that each side incorporates what might best be termed an outward bulge, particularly prominent on the north and south sides. On average the enclosure is just under 50m (just under 165ft) wide north-south and in the opposite direction tapers from 64m (210ft) on the north to 44m (144ft) on the south. Manor House Although the remains are slight, sufficient medieval work survives to indicate that the main dwelling had masonry walls and stood north of central, with its principal axis aligned north to south (see Fig. 3). One moulded jamb of the front door survives to show that the main entrance was positioned immediately opposite the gatehouse – a very common arrangement. Perhaps it was once protected by a porch, for the will of Thomas Darell made in 1558 mentions a closet over ‘the inner porch’: if so, it has been removed leaving no visible traces. The entrance itself led Fig. 3 Outline plan showing suggested building developments on the inner island during the medieval period. DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 328 directly into the ‘low’ (i.e. socially inferior) end of the open hall, with a less ornate back door immediately opposite. This rear entrance (now blocked) is shown in Fig. 4. It represents the best preserved of the manor house’s few remaining medieval features: one hollow-chamfered jamb and half the two-centred arch survives, as too does part of the rear arch. Having a clear span of 7.55m (24ft 9ins), the hall was of respectable width and compares favourably with several other un-aisled halls built by the local gentry at this general period. A short section of roof line built into the walls of the rebuilt adjacent part indicates not only the height of the medieval walls, but also gives an impression of the roof slope (Fig. 5). The hall stretched northwards from its entrance area, but the only hint as to its length is a foundation discovered in 1873, projecting forwards from the alignment of the main range (see Fig. 3). This could indicate the buttressed south-western corner of a high-end crosswing, projecting forwards from the main house – again, not an unusual arrangement. If this is so, it suggests a length for the hall of very approximately 13m (about 42ft 6ins). However, although by no means small such a size pales somewhat against the halls of the leading prelates and aristocracy, all of which were designed for serious entertainment and ceremony. Fig. 4 Remains of main arch (rebated for door) and rear arch to doorway leading out of the low end of the medieval hall, giving access to the rear yard. A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 329 To the south of the hall, housed within an (assumed floored) in-line end bay, were service rooms. Of these, all that survives today are the jamb of what appears to be a doorway leading southwards out of the hall, together with fragments of the front wall and perhaps the stub end of the original south wall. Regardless of whether the stub end is medieval in date, it follows the line of the original south wall of the house, for the massively wide foundations of the wall were recorded by the Central Excavation Unit of English Heritage in 1986. Despite its great width, the foundation was quite shallow, supported on timber piles. There are no indications to suggest the internal layout of the service area. On the sole evidence of the foundation revealed in 1873, it is suggested that the high-end accommodation followed the normal arrangement for a house of this status, being contained within a crosswing which projected Fig. 5 Reconstructed cross section through the low end of the hall. DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 330 forward. This arrangement shows parallels to houses of similar date built by members of the Kentish gentry.9 If the above interpretation is correct, the space between the end wall of the hall and the curtain wall is too great for the curtain wall to have doubled as the side wall of the crosswing. Further, the curtain wall is running at an angle to the house, which in any case would have made such an arrangement improbable. Whether this means that the manor house was freestanding upon the site, or whether there was a further wing (or wings) extending out to the curtain wall, as at Ightham Mote, is impossible to tell from what survives. Perhaps the likelihood is that it was freestanding, for at this date the house did not extend the full width of the site: instead, its southern wall was positioned only slightly more than halfway across, giving the house a very offcentred location. The position of the house, heavily biased toward one side of the site, may seem strange, but is almost certainly explained by the presence to the south of a detached kitchen/service block, an arrangement known from other sites. Though no above-ground material evidence of this building survives today, fossilised architectural elements contained within the later developments require the pre-existence of a structure in this location. For example, it can be shown that by the close of the medieval period there was a jettied structure abutting the curtain wall in this location, with the jettied upper storey (most likely an end wall) overhanging the moat. Furthermore, this is where the later kitchen was located (see below). Such a configuration of buildings would have meant that the two principal structures on the site stretched north-south across the centre of the enclosure, perhaps linked to each other by a covered walkway. Effectively they would have divided the enclosure into two parts: a public forecourt area and a far more private back service yard (Fig. 3). The arrangement has similarities to the original layout at Ightham Mote (1330s) where the enclosure was smaller and the rear yard tiny.10 In addition to the main house and its kitchen building, the possibility of outhouses built against the curtain wall should be borne in mind, particularly at the rear where the results of the geophysical survey carried out in 2008 suggest structures (date unknown, but probably medieval) edging the eastern side of the enclosure (see Fig. 6). The Defences: gatehouse, corner towers and curtain wall In addition to the moat, the known defences at Scotney amounted to a low curtain wall, four corner towers, and a gatehouse. It seems likely that these represented the total defences which protected the main island. Further, as will be indicated, they were minimal in that they provided nothing more than light defence to the enclosure. Though designed to withstand a lightly-armed attack, there seems little doubt that their A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 331 primary purpose was show. This is a point often made regarding nearby Bodiam Castle, though, even accepting this similarity, it needs to be stressed that Scotney was never anything like as strong or impressive as Dallingridge’s castle at Bodiam: not only were there no interval towers, but in addition the curtain walls were thin and exceptionally low. In comparison to Bodiam, the surviving tower at Scotney is exceptionally squat, though despite this, beside Scotney’s curtain walls it would have appeared lofty. Whereas at Bodiam the curtain walls were integral to the mansion’s structure, at Scotney they formed nothing more than a low surrounding protective skirt to the enclosure. As at some other local sites – for instance, Glottenham Castle, Mountfield, Sussex; Westenhanger Castle, Kent; and Ightham Mote, Kent – in its late-medieval form the site was in truth nothing more than a lightly defended moated manor house. In some ways the late fourteenth-century arrangement at Penshurst Place (a house of the 1340s with crenellated walls and towers added under licence dated 1392) gives a close parallel to Scotney. Although probably never moated and standing within a very regular enclosure protected by rectangular, rather than circular towers, Penshurst too stood within an enclosure protected by a thin, low curtain wall which incorporated towers which rose quite high above the wall. Fig. 6 Outline plan showing the inner island’s known defences. DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 332 The Gatehouse Only the stub walls of the gatehouse survived the dismantling of the defences, probably in the seventeenth century. These rise directly out of the moat with a battered (i.e. sloping) external face to the lowest section of the wall, which part serves to retain the earth. Following the building’s demolition, those elements of the upstanding superstructure which were retained were tidied up so as to form an ornamental feature standing at the entrance to the site: unfortunately this ‘tidying’ robbed the structure of all medieval architectural features. Faced in neatly-dressed ashlar blocks, the gatehouse took the form of a square tower, of which the lowest storey was entirely occupied by the entrance passage. Significantly, and rather curiously, the remains of the gatehouse stand in front of the curtain wall (Fig. 3), projecting into the moat. Angle buttresses are incorporated at all four corners, including those which face back towards the site (Fig. 7). In this the structure more closely resembled a barbican than a gatehouse and has the appearance of being an afterthought, constructed subsequent to the curtain wall. This is Fig. 7 Outline plan showing footprint of the gatehouse in relation to curtain wall. A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 333 not the only piece of evidence which suggests that the defensive elements of the site may be of more than one phase (see below). There are no clues within the surviving remains to suggest how high the gatehouse rose, and no indications to suggest the location of a staircase, though the thickness of the walls are consistent with a structure incorporating at least one upper storey, and perhaps two. That the gatehouse had at least one storey above the entrance is indicated by the will of Thomas Darell I (d.1558), which mentions a room above the gatehouse, and by an account of a search for a catholic priest made in 1598, during which search Thomas Darell II’s widow and children were confined to ‘one room over the gate’. Corner Towers Although all four corner drum towers show as stone-edged projections (with battered faces) projecting out into the moat, and the foundations associated with the south-eastern tower were identified during the 2008 geophysical survey, upstanding remains are limited to the two western towers. Of these, all that now stands of the north-western tower is a short section of wall incorporating a ground-floor doorway with a two-centred arched head, now very much tidied up (Fig. 8). Steps lead down through the doorway, indicating that the floor of the tower was below the level of the outer court, from which it was reached. Being attached to the later house, the south-western tower (known as the Ashburnham Tower) survives very much as originally built: it has lost one of its internal floors and the present roof is modern, but otherwise the only missing elements are part of the machicolations and the [assumed crenellated] parapet. It is this tower, and the evidence as to how it met the curtain wall, which gives the best picture of the likely form of the medieval defences. With regards to construction, its character resembles the work at Bodiam Castle in that it rises directly out of the moat, supported by a batterfaced retaining wall, and is faced throughout in ashlar sandstone blocks quarried from the local Hastings beds. Around the top of the tower are machicolations with rounded corbels which carry relatively depressed twocentred arches, which in turn supported the former [assumed crenellated] parapet (Fig. 9). Today the circuit is interrupted by the present house which abuts the tower, but when built the upper parts of the tower stood clear of other structures and thus the circuit of machicolations was complete – one of the machicolations still survives trapped within the thickness of a wall of the building, now only visible from the roof void. The (assumed crenellated) parapet has been destroyed down to the level of the wall top, which at this point is capped by massive stones spanning the full width of the wall. The design of the machicolations indicates that the parapet itself was c.300 mm (1ft) thick. A conjectural reconstruction DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 334 of the parapet is shown in Fig. 10, where the height is based upon the normal height of a parapet above wall-top level (Bodiam Castle has been used as the model, but in that instance machicolations were used on the gatehouses only, not on the corner drum towers). Using these average Fig. 8 Doorway leading into former north-western tower. A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 335 Fig. 9 Details of machicolations, Ashburnham Tower. measurements produces a very dominant upper order to the tower – it looks out of proportion to the tower’s height, but the parapet could not have been lower if it was intended to be functional. This, of course, raises an important issue: were the machicolations and (assumed crenellated) parapet ever intended to be anything other than decorative? If their sole purpose was decorative, with no access to the wall top, then the height would have been irrelevant. The present roof which caps the tower is conical and is a nineteenthcentury reconstruction of an earlier roof of similar shape. The usual assumption is that the earlier roof replaced a medieval flat roof, and it is this assumption which has been used for the reconstruction in Fig. 10. If this were so, it would have allowed a ladder and trap door to give access to the top of the tower. However, it is worth questioning whether the roof was indeed flat, for it is almost certain that at Bodiam the drum towers DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 336 were capped by conical roofs, partly shielded by the encircling parapet – tapered West Country slates from a conical roof were recovered from the moat at Bodiam in 1970, whilst the presence of a dovecote in the upper stage of the castle’s south-western tower demands some form of pitched roof in order to support the cupola / lantern through which the doves would have entered. Despite the accepted opinion, there is therefore no certain evidence to indicate that the conical roof shown in late eighteenthcentury drawings was not the original medieval roof structure. If there was a medieval pitched roof, then the issue of access to the tower top becomes far more relevant, for although there is an intramural staircase, this rises from first floor to second floor only, and does not continue up to the wall top. Fig. 10 Reconstruction of the Ashburnham Tower as built in the fourteenth century, showing relationship to the adjacent curtain wall. (Roof design and height of parapet conjectural.) A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 337 Viewed against the present house which abuts it, the tower is exceptionally squat. However, as Fig. 10 depicts so clearly, so originally was the curtain wall which abutted it. This point can be certain, based upon the form of the external face of the tower. On the eastern side part of the curtain wall’s superstructure survives (see below) and now forms the ground-floor walls of the house. That they were not lowered when the house was built is indicated by the external face of the tower, which not only lacks a scar for a removed upper section of wall, but still retains its original neat dressing, now visible within a void which can be viewed from the roof space in the house. Indeed, rather than having been lowered in height, repairs and differences in masonry at the top of the wall suggest that it may have been raised a little when, later, it was incorporated into the newly-built west range of the house (see Part II). The extant part of the southern curtain wall appears bonded to the tower. Taken at face value this suggests the two are contemporary with one another (but see ‘Curtain Wall’ below). In the north face are two distinct scars (rough areas of wall facing) indicating where two further walls were bonded to the tower. One of these scars is visible within the toilet enclosure formed under the tower’s external stairs which at this point replaces the curtain wall: the scar indicates where the western curtain wall met the tower. The other is a short distance to the east, on the opposite side of the doorway which leads into the base of the tower. Because only the scar remains, it is impossible to judge the angle of the wall, but its presence indicates that some form of structure abutted the tower at this point. As with the evidence at the point where the southern curtain wall meets the tower, here too both wall scars stop abruptly at the level of the tower’s first floor, with no evidence whatsoever for a stone wall having ever abutted the tower above this level. Further, the tower’s first-floor and second-floor apartments were only ever accessible via a first-floor external doorway. This doorway is located immediately above the line of the western curtain wall, without any space next to it for a protecting parapet wall – the present parapet wall at this point projects beyond the line of the original curtain wall and largely masks the chamfer on the doorway’s western jamb. The doorway must have had some kind of wall between it and the moat, but, based upon the visible evidence, one is forced to the conclusion that any such structure was most likely of timber and was jettied out beyond the line of the curtain wall. The obvious answer would be that the doorway was contained within the jettied timber-framed upper storey of an abutting building, but, if so, the roof of any such building must have been kept very low and sloped away from the tower in order to fit beneath the machicolations. Further, the gutter formed where the roof met the tower would have needed to be weathered, but the tower’s wall face shows no weathering courses or other scars to indicate the former DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 338 existence of any such valley. This does not discount the presence of such a building, and certainly this section of the tower’s external wall face is straight rather than curved. But there is an alternative possibility which seems to fit the evidence better. The curtain wall may have been capped by a timber-framed wall walk which projected proud of the external face of the curtain wall to form machicolations: such a wall walk would probably not have incorporate a roof. This walkway could have resembled the hoardings often argued for the tops of castle walls. Although the presence of such an arrangement would make a great deal of sense, it must be stressed that it is, at present, nothing more than hypothetical. Internally the tower is of three storeys, of which the floor of the lowest storey is positioned slightly below the present external ground level. The ground may, of course, have risen over time, but so has the internal floor, a point well illustrated by two gun ports, the openings of which are now level with the floor (Fig. 11). The presence of the gun ports indicates that the castle certainly incorporated at least some serious defensive features, a point which perhaps argues against the tower-top machicolations having been incorporated for solely decorative reasons. In fact, there are indications which suggest considerable thought was put into the placing of the gun ports, and indeed the design of the island itself. The sides of the island are cranked outwards and the ports positioned so as to give a clear line of fire along the moat without hitting either the gatehouse or the drum tower (Fig. 6) Apart from the two gun ports and the entrance doorway, the tower’s ground-floor room contains only one other feature – a single lancet window, providing minimal light. The ceiling is formed by the heavy medieval joists which support the floor above – the joists are mostly replaced and the survivors are now reset on their side. The configuration of the room does not suggest it was intended for accommodation. In contrast to the ground-floor arrangement, although the two upper storeys were accessed via an external first-floor doorway only, the arrangement at this higher level suggests they were designed to provide a two-roomed suite of accommodation with plastered walls. The principal room is that on the first floor. It is lit by two lancet windows and is heated by a contemporary fireplace as well as being served by a garderobe recess. Although the recess was closed by a door, it provided no privacy in that the recess did not continue down to ground level, and thus contained within it only the seat. Thus, the users remained with their knees and legs in the room. The door was used solely to hide the recess when not in use. A similar fourteenth-century garderobe exists in the northern interval tower at Westenhanger Castle. Rising through the eastern jamb of the entrance doorway, located within the thickness of the wall, is an intramural stone staircase giving access to the upper chamber, the floor of which is now missing, though its joists are evidenced by socket holes. The chamber is unheated and is not so well lit A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 339 Fig. 11 Eastern of the two gun ports in ground floor of Ashburnham Tower. is that below, having only one lancet window. Furthermore, because of the depth of the machicolations the window is closely flanked on the exterior by the machicolation’s corbels, still further reducing the level of light. The window is just visible on the extreme left in Fig. 12. Despite this, it is a comfortable enough chamber. Its internal appearance is not entirely certain: much would have depended upon whether the roof was flat or, as now, pitched: if the latter, the chamber would have been open to the roof above. Curtain Wall Although the gatehouse and corner drum towers would have contributed much to the appearance of the castle, perhaps even more important was the form of its curtain wall. As discussed above (see ‘Corner Towers’) the available evidence indicates that almost certainly the curtain wall was exceptionally low – only about 2.75m (about 9ft) up to the level of the DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 340 wall walk, capped by a projecting timber-framed arrangement, perhaps an upper storey but, more likely, some form of hoarding. Observations made during moat clearance in 1987 showed the base of the wall to be battered, following the design used at the base of the gatehouse and towers. The wall, where it survives incorporated into the later house on the southern side, adjacent to Ashburnham Tower, measures 800mm (2ft 8ins) thick and, rather than being faced in neat ashlar masonry as is the case with the gatehouse and tower, it is of roughly squared rubble. This in itself is surprising – why the variation in finish? An obvious explanation would be that the wall and tower are not of the same date and that the gatehouse and Ashburnham Tower represent later additions. Taken at face value such an interpretation is, of course, inconsistent with the fact that the tower is bonded to the curtain wall. However, such an interpretation could explain why the gatehouse stands entirely proud of the curtain wall, and why the drum towers project so boldly into the moat. If the curtain wall does predate the other defences, the fact that the two are bonded could be explained by the need to demolish the corners of the wall in order to fit the towers in: if this was the procedure adopted, they would be bonded. If the curtain wall and towers do prove to be of one date, perhaps Fig. 12 South elevation showing low height of curtain wall in relation to Ashburnham Tower. The gothic doorway was most likely a postern gate accessed by water. The flanking windows are later. A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 341 Fig. 13 Blocked arrow slit in fragment of curtain wall. Photograph by Peter Leach 1987. the curtain wall was originally rendered, as was certainly the case with the town defences at Winchelsea, Sussex. If this were the case, the ashlar of the drum towers and gatehouse would have been left exposed as a feature. With so little of the curtain wall surviving, it is hardly surprising that it contains only two medieval architectural features. The most obvious is the two-centred doorway a little to the east of the Ashburnham Tower (Fig. 6). The existence of this doorway, visible in Fig. 12, could indicate that a building always stood on the site of the present west range, though (perhaps more likely) the doorway merely served as an opening through the curtain wall. Whichever the case, it must have been a postern, for until the present external platform was made much later, the moat lapped against this wall and thus the doorway could only have been reached by boat. The second curtain-wall feature, now almost hidden by foliage, is even more interesting. It is an arrow slit located within a small fragment of curtain wall incorporated into the south-eastern corner of the later main range (Fig. 13; see Fig. 6 for location). Here too the external facings are DAVID MARTIN, BARBARA MARTIN AND JANE CLUBB 342 of rubble, but the slit itself is of dressed sandstone. Originally the slit was very narrow, but it was widened somewhat when it was blocked. Internally it has splayed jambs [Part II, detailing the later development of Scotney Old Castle, will appear in the next volume.] acknowledgements In 2008 a programme of quinquennial repairs were put in hand at Scotney Castle, and, as part of this work the National Trust took the opportunity to carry out limited targeted recording of those parts not normally accessible. This work coincided with the preparation of a Conservation Management Plan for Scotney. With this in mind, the Trust extended the recording by commissioning Archaeology South-East, University College London, to carry out an interpretative survey of the house as a whole. Throughout the project there was a close working relationship between the project staff of Archaeology South-East and the regional office of the National Trust at Scotney, in particular with Rupert Goulding, who was responsible for preparing a Conservation Management Plan. Elements of Rupert’s work have been incorporated within this report, in particular with regards to his research into the documentary background of the site. We are extremely grateful to him for making available pictorial information regarding the castle, particularly 18th- and 19thcentury plans and drawings. In addition to Rupert, we would also like to extend our warmest thanks to the National Trust staff in general, and in particular to Caroline Thackray (former Territory Archaeologist) and Emma Slocombe (Curator) for their help and encouragement throughout the project. A close working relationship between Archaeology South- East and the contractors, A.T. Palmer Ltd was essential for the success of the project: their cooperation throughout cannot be faulted, and for this we wish to thank, in particular, the help of their site representative, Colin Maddocks. Finally, we would like to extend our thanks to the well-known expert on Polite architecture, Nicholas Cooper. endnotes 1 Sussex Record Society, vol. 10, 1909. 2 Sussex Archaeological Collections, 10, 1858, 129-146. 3 Some moats (such as, for instance, Glottenham in Mountfield, Sussex, built by Sir Robert de Etchingham early in the 14th century) were dug around earlier buildings, whilst in other instances the moat occupies a virgin site and replaces an earlier house elsewhere upon the property in question. A good example of the latter phenomenon is Sir Edward Dallingridge’s Bodiam Castle, Sussex, built under licence dated 1385. A SURVEY OF THE OLD CASTLE, SCOTNEY, LAMBERHURST: PART 1: MEDIEVAL 343 4 Sir Roger served alongside Sir William de Echyngham (who owned two moated sites locally, one at Etchingham, the other at Udimore) and Sir Edward Dallyngridge of Bodiam (the subsequent builder of Bodiam Castle). 5 Unlike at Scotney, the corner towers at Bodiam lack machicolations, but these features are found on its gatehouses. 6 Sussex Record Society, 1956, vol. 56, 124-137. 7 D. and B. Martin, 2001, ‘Westenhanger Castle – a Revised Interpretation’, Archaeologia Cantiana, 121, 203-236; a similar complex, again un-moated and standing before the main moated house, is depicted in an estate plan showing the principal seat of the Ashburnham family next to Ashburnham Church. 8 Good local examples are Sir Edmund de Pashley’s moated site of about 1318 at The Mote, Iden, Sussex and John de Cobham’s castle at Cooling, Kent, built 1381. 9 Copton Manor; Old Soar, Plaxtol; and Ightham Mote – see Barnwell and Adams, The House Within, HMSO, 1994. 10 See Pearson, Barnwell and Adams, A Gazetteer of Medieval Houses in Kent, HMSO, 1994, pp. 74-77.