Historical background

A hall is first recorded at Oakham in the Domesday survey of 1086, when William i had two ploughs belonging to the hall.Footnote ⁷ It is assumed that this hall stood, with other buildings, somewhere inside the 2-acre bailey, surrounded by an earthwork rampart and a ditch or moat, with a motte at the south-east corner (Fig 3). During the early twelfth century the manor passed to the Ferrers family, with Walkelin de Ferrers known to be in possession by 1166. The present castle hall, dated stylistically to c 1180–90, is attributed to a major building campaign by Walkelin, who died in 1201. Walkelin's son, Henry, inherited the manor in 1201 but left for Normandy in 1205, leaving Oakham to his sister, Isabella. She married Roger de Mortimer and, in 1207, obtained a grant for life of the manor of Oakham. The manor reverted to the Crown at Isabella's death in 1252, bringing the association with the Ferrers family to an end.

Fig 3

Key features of the castle site, with earthworks in the bailey. Drawing: Hartley Reference Hartley1983, 32

Henry iii granted Oakham to his younger brother, Richard, earl of Cornwall, in 1252, who held it until his death in 1272. It is to this unsettled period of English history that the strengthening of the earth ramparts with stone walling, together with the addition of at least two defensive towers, has often been attributed, though the evidence is far from conclusive.Footnote ⁸ Richard's son Edmund, earl of Cornwall, held the property from 1272 until his death in 1300. The castle passed through a number of hands in the first half of the fourteenth century, including those of the earls of Kent and Gloucester, before being granted by the king to William de Bohun, who became earl of Northampton and took up possession in 1347. An inquisition of 1340, a very valuable document, gives the first detailed description of the castle:

William de Bohun's young son, Humphrey, succeeded his father in 1361 but died in 1372. Oakham was then held directly by the king until 1386. In 1390 it was granted to Edward, earl of Rutland, and subsequently passed to Humphrey, earl of Stafford, in 1438. Humphrey became duke of Buckingham and his family held it, except for various vicissitudes, until the execution of Edward, duke of Buckingham, in 1521. An inquisition of this date says that ‘there is an old castle, all ruinous … the hall is in the best state of repair, and old-fashioned’.Footnote ¹⁰ Residential use had clearly been abandoned some time before, and it was its use as a court – first recorded in 1229Footnote ¹¹ – that led to the retention of the hall, and which has continued to protect it up to the present day, all the other buildings having been lost. Possession of Oakham by the Crown came to an end in 1596 when it was sold to Sir John Harington, whose family had built up a wealthy estate nearby at Exton and Burley on the Hill. In 1621 the manor was bought by George Villiers, a favourite of James i, who was created duke of Buckingham in 1623. Tradition records that he repaired the hall for continuing use as a courthouse, and he certainly rebuilt the gateway, as it matches others on his estate at Burley. Buckingham was murdered in 1628 and the property later passed to the Finch family, earls of Winchelsea and Nottingham, and eventually, in 1944, to Rutland County Council.

The main hall

The earliest depiction of Oakham Castle is on John Speed's map of Rutland of 1611, which includes a town plan of Oakham (Fig 4). The ‘Shire Hall’ stands alone, approached via an arched gateway through an encircling stone wall with a square tower indicated at the south-western corner. Within the bailey, various ‘humps and bumps’ on Speed's drawing indicate the remains of the rest of the castle buildings, which had all been lost by this date. It is similarly portrayed in an engraving of 1684 in Wright's history of Rutland (Fig 5) and another by the Buck brothers of 1730, which also shows Buckingham's gateway of the 1620s (Fig 6).

Fig 4

Plan of Oakham in 1611. Map: Speed Reference Speed1611

Fig 5

Oakham Castle Hall in 1684. Engraving: Wright Reference Wright1684, 104

Fig 6

Oakham Castle in 1730. Engraving: published by Samuel and Nathaniel Buck in London in 1730 as an individual print, entitled The South-East View of Okeham-Castle in the County of Rutland

The hall (Figs 7 and 8) is of four bays and measures 19.9m by 13.2m (65ft 3in by 43ft 4in) internally, with an aisle width of 2.8m (9ft 2in). The aisle roofs butt against the walls of the main ‘nave’ with a short upstand, there being no full clerestory. On the entrance front to the southern side, the doorway was originally at the east end, but was later moved to a central position. The doorway is round-arched with heavy roll mouldings and attached shafts. The windows are of paired form, with two pointed arched lights, having attached shafts to the central mullion and, originally, to the jambs. The window heads have solid tympana with shallow carved decoration. Both the front and rear aisle walls formerly had four windows, but two have been lost to the rear. The gable walls each have two prominent buttresses at the corners of the main ‘nave’, though the tops of these have been cut back to the east gable. The gables are crowned by the original finely carved finials, which are very rare survivals. That to the east is generally thought to represent Samson and the lion, while the western figure is a centaur. Less obvious, but of equally fine carving, are the gable corbels which survive on the north side, with twin heads turned to face each other. The east gable has two blocked doorways that connected to the service end, with a blocked six-light mullioned window above, a seventeenth-century insertion. Higher up in the gable is a large round-arched window with two pointed lights. The lower parts of the west gable are now obscured by extensions of the nineteenth and twentieth centuries, as is much of the rear wall (Fig 9). The walls are built mainly of the local ironstone, with limestone dressings, probably from quarries at Clipsham, only five miles away. The roofs are of the local Collyweston slate. Set securely within the castle bailey, the hall was clearly built to impress and had no need for any defensive character.

Fig 7

Oakham Castle Hall: ground-floor plan, drawn in 2011. Drawing: author and John Owen

Fig 8

Oakham Castle Hall: longitudinal section, drawn in 2011. Drawing: author and John Owen

Fig 9

View of Oakham Castle Hall from the north west, with later extensions. Photograph: author

Inside, the magnificent stone arcades have four bays of round arches, supported on circular columns with exquisitely carved capitals (Figs 10 and 11). Above the capitals are figures playing a variety of musical instruments (Fig 12), and the corbels supporting each end of the arcade have further fine carving, with beasts supported on twin heads. In the east gable, two blocked service doorways are symmetrically arranged in the centre, the door heads here having pointed arches (though the external arches are segmental). Offset in the north aisle is a third blocked doorway, this one being round-arched. Above it at first-floor level, squeezed tightly under the aisle roof, is a further round-arched doorway. Due to external refacing, only part of the lower doorway jamb is visible on the outside. In the west gable, there is an original doorway in the north-west corner, whose round rere-arch is visible, though the front face has been covered up. There is a further doorway in the south-west corner, but any details which might indicate whether this is an early opening are obscured. The west end of the hall retains court fittings dating from the early nineteenth century. The whole interior is dominated by the display of outsize horseshoes, it being the tradition since at least the fifteenth century that every peer of the realm must pay this forfeit to the lord of the manor on their first visit.

Fig 10

View of Oakham Castle Hall interior looking east. Photograph: author

Fig 11

Oakham Castle Hall: arcade capital with foliage decoration to the crockets and dog-tooth ornament. Photograph: author

Fig 12

Oakham Castle Hall: musician playing a viol. Photograph: author

The details of the architecture at Oakham have always been considered a classic example of the Transitional style. As Turner and Parker said in 1851: ‘The style of the building is that of the latter part of the twelfth century, being a transition from Norman to Early English, and the beauty of design and superiority of execution of its ornamental parts render this building one of the most valuable examples of that style which we possess.’Footnote ¹² Round and pointed arches are both used, sometimes together on the same building element. The windows, for example, have pointed lights externally, but a single round arch inside; and several of the external tympana, beneath the pointed lights, are decorated with round-arched arcading. Roll mouldings are used widely, on windows, doorways and the main arcade arches. Dog-tooth ornament is also applied very extensively, on the windows, doorways, arcades and even the capitals. As Parsons noted, dog-tooth is very rare in the area until the full establishment of the Early English style in the early thirteenth century, so the work at Oakham was certainly ahead of its time.Footnote ¹³ It is the capitals that have been used to establish a more precise date for Oakham's construction.Footnote ¹⁴ Their elegant leaf forms, with curling crockets, are considered so close to those at Canterbury Cathedral that they must have been carved by the same masons. The work at Canterbury started in 1175 and was finished by 1184, so it is thought that the masons arrived at Oakham some time during the 1180s. Some aspects of the fine figure carving at Oakham, such as the musician figures, have also been linked to the masons of Canterbury.Footnote ¹⁵

Buildings accompanying the hall: previous theories

The principal features of the main hall, as summarised so far, have been well known for many years. However, as noted above, the seeming completeness and perfection of the hall has meant that the hall itself has dominated most studies of the site, with little consideration of the buildings that would have accompanied it. Where previous studies have commented on adjoining buildings, there has been a considerable variety of views, with many points remaining unresolved.

At the east end, the four blocked doorways have always provided clear evidence that there was an original attached service end. Most accounts have referred to these only in very general terms. Wood, for example, in 1935 briefly mentioned the ‘blocked doorways leading to the offices in outbuildings now destroyed’ and ‘a first-floor doorway apparently approached by an internal stairway’ which ‘apparently led to an apartment above the offices’.Footnote ¹⁶ Holland Walker noted in 1924 that: ‘Tradition points to the lower of these two doors as being the entrance to the chapel, while the upper one led to the priest's room’, though he dismisses this idea as unlikely.Footnote ¹⁷ In 1955, Radford suggested that the doorway in the north aisle gave access to an external stair leading to a gallery above a screens passage.Footnote ¹⁸ He also thought, on the basis of ‘patching and fragmentary remains at the west end’ that there was a two-storeyed solar block here, flanked by a pent-roofed passage on the north side.

A more fully developed theory was put forward by Faulkner in 1958.Footnote ¹⁹ He proposed that the eastern end formed a two-storey cross wing, with a solar above service rooms, approached by a stair from the main hall. Drawing on the analogy of the solar at Lincoln Bishop's Palace, he suggested that the solar cross-wing roof was supported on transverse stone arches (spanning west to east), which have since been replaced by the present buttresses. The upper part of the east gable to the hall ‘originally rose clear of the cross roof of the chamber block and contained a circular window, the upper half of which has been incorporated in the modern window now in its place’. At the west end, Faulkner thought the buttresses original, precluding any attached buildings. Oakham thus became, for Faulkner, a classic example of the ‘end hall house’, with the main chamber over the services at one end, and no chamber at the high end. The identification of this ‘end hall house’ type was fundamental to Faulkner's thesis on the development of English domestic plan forms from the twelfth to fourteenth centuries, which he believed preceded the fuller development of the classic medieval tripartite plan, the hall with two storeyed ends. Faulkner's theory for Oakham was repeated by Wood in 1965,Footnote ²⁰ but is not supported by the evidence presented below.

New evidence was discovered in 1956–7 by John Barber during excavations at the east end of the hall (see below). Barber found evidence for two service rooms at the east end, separated by a central passage which led to a detached kitchen. He believed that the service end was more likely to be of one storey than two, and followed Radford's explanation of the first-floor doorway, as leading on to a gallery over the hall screens. He also followed Radford's view that there was a two-storey solar block at the west end. However, he was the first to remark on a particularly important detail: a corbel block located high up on the outer face of the west gable. Barber noted that the same type of corbel is used to support the top wall plate for the aisle roofs, so there must have been a similar sloping roof which abutted the west gable.

Clough's guidebook of 1981 mentions the accompanying buildings only in general terms, referring to the services with first-floor rooms above at the east end, and a probable two-storey solar block to the west, without details of any supporting evidence.Footnote ²¹ Pevsner again makes only very brief reference to the missing service rooms and kitchen to the east, and a possible two-storey solar block at the west end.Footnote ²²

The most recent analysis was by Mark Gardiner in 2008.Footnote ²³ He concluded, like Faulkner, that the main chamber was at the east end, over the services. However, having spotted the corbel block to the west gable, Gardiner thought that the west end must have taken the form of a single-storey lean-to, a ‘return or end aisle’, with no chamber above. While Gardiner's analysis is very perceptive, it still does not take account of the whole of the evidence, and it is to this detail that we must now turn.

The excavations of 1956–7 at the east end of the hall

Excavations were undertaken to the east of the castle hall by John Barber, a teacher at Oakham School, over two seasons in 1956 and 1957. Although the excavation was undertaken by a party of schoolboys, Barber had considerable archaeological experience and was a Fellow of the Society of Antiquaries. Detailed record drawings were made of the three trenches which were opened up, but only a very brief note of the excavations was published.Footnote ²⁴ The details of the excavations were eventually drawn together, with Barber's other researches on Oakham Castle, in his memoirs of c 1980 and subsequently published digitally in 2013.Footnote ²⁵ As part of this publication, a report on the pottery finds was produced by Deborah Sawday in 1999 and updated in 2011.Footnote ²⁶

Barber dug three trenches (Figs 13–16), which enabled him to put forward a reconstructed plan for the lost buildings at the east end (Fig 17). The first, Trench 1, was 4ft (1.22m) wide, with its western edge about 15ft (4.58m) to the east of the hall's east gable. Evidence was found for two walls (nos 9 and 17 on Barber's drawings) running west–east, roughly aligned with the existing corners of the hall. Both walls had been robbed out, but they were clearly of considerable thickness, having been set into trenches 4ft (1.22m) wide. Two further walls (nos 14 and 15) were found to either side of the line of the central doorway to the hall gable, obviously defining a passage. One of these was robbed out, but the other retained stonework of a wall 2ft (0.61m) in thickness. Another wall (no. 10), with stonework 3ft (0.91m) thick, was found just to the north of the southern wall. Slighter traces were found of two other walls (nos 11 and 13) in the southern part of the trench, which were thought to relate to an earlier phase or building. A well (no. 16) was found in the northern part. In Trench 2, running eastwards from Trench 1, a robbed-out wall trench 3ft (0.91m) wide was found, indicating the east wall of the building. Barber concluded that there was a smaller service room to the north, a central passage to a kitchen, and a larger service room to the south, which may have subsequently been extended a little further southwards. He thought the southern room was the buttery and the north room the pantry, though it is perhaps more likely that the reverse was the case, with the buttery located on the shaded north side and supplied with a well. The floors of both service rooms and the passage were described as being of ‘cement’.

Fig 13

Plan of Barber's three trenches excavated in 1956–8. Drawing: J L Barber Archive, Rutland County Museum, redrawn by R Ovens

Fig 14

Barber's Trench 1. Drawing: J L Barber Archive, Rutland County Museum

Fig 15

Barber's Trench 2. Drawing: J L Barber Archive, Rutland County Museum

Fig 16

Barber's Trench 3. Drawing: J L Barber Archive, Rutland County Museum

Fig 17

Buildings at the east end of Oakham Castle Hall as reconstructed by Barber. Drawing: J L Barber Archive, Rutland County Museum, redrawn by R Ovens

In his second and third trenches, Barber revealed convincing evidence of a detached kitchen, set only 9ft (2.74m) to the east of the service building, the gap between being stone-paved. The kitchen measured approximately 26ft (7.92m) square internally, with walls around 3ft (0.91m) thick. Two footings, around 2ft (0.61m) square, were found for column bases, with two more presumably in the unexcavated area. A baking oven (or similar hearth structure) was found in the south-western corner. Trodden into the stone-paved flooring were eggshells and the skeleton of a carp, together with a bronze nutcracker. Most of the pottery excavated (see below) was found in Trench 2 and Trench 3 to the kitchen building; there was very little in Trench 1 to the service building, and nearly all of it was from the well. Overall, the pottery finds are consistent with occupation of the site from the twelfth century, with intensive use in the thirteenth and early fourteenth centuries, followed by lower levels of activity from the fifteenth century. However, the pottery did not provide any clear stratigraphic evidence for the date of the service building attached to the hall.

Building analysis of the structures attached to the hall

The first part of the recent programme of research at Oakham Castle was a detailed analysis of the building fabric of the hall, in particular of the two gable ends and any light they might throw on the differing views concerning the nature of any attached structures. From this analysis of the gable ends, a new theory has emerged for the form and development of the hall.

West end

The west gable presents the more easily intelligible evidence, so will be treated first (Fig 18). Most previous studies (except Barber and Gardiner, as noted above) had failed to take account of a fairly obvious feature here, the stone corbel block projecting from the centre of the gable end at high level (Fig 19). The corbel is carefully shaped, with a concave moulding to the underside and a rectangular notch to the top, clearly designed to carry a timber plate. Two rows of similar corbels can be seen inside the hall where the aisle roofs abut against the ‘nave’ of the hall, with a wall plate still in situ to the south side, carrying the rafters of the aisle roof. Each side had five corbels originally, though two have been lost as a result of later alterations. The corbels are set at regular intervals of around 3.3m, which relate to the span of each section of timber wall plate, and bear no relationship to the stone arcades. The mouldings, stone type and location in the building all point to these corbels being an original part of the twelfth-century structure. The corbel to the west gable would also therefore appear to be an original feature, though it is set at a rather higher level than those to the aisles.

Fig 18

Oakham Castle Hall: west elevation. Drawing: author and John Owen

Fig 19

Oakham Castle Hall: corbel block to the west gable, designed to carry a timber plate. Photograph: author

Confirmation that the west gable corbel is original and in situ is provided by a detail of the gable wall not noted by previous observers. Just below the level of the corbel block there is an irregular flat ledge, an offset around 75mm wide formed in the coursed rubble walling. This offset runs right out to the corners of the building, with the limestone quoins carefully constructed to incorporate the offset. This offset ledge, built of irregular rubble rather than dressed stone, appears to have been designed to be protected by an abutting roof rather than being exposed to the weather.

The principal features of the west gable are the two fine buttresses, which survive in very complete state (though partly covered up by modern extensions at lower level). They are built of good-sized ashlar blocks, with two plain offsets and a plain, sloping top. These buttresses are clearly intact, original features. An unusual feature is a slot carefully formed in the main wall to receive them, lined with quoins for the full height (Fig 20). This detail is seen also at the east end, and internally where the stone arcades meet the gable walls. At the north-west corner (and also to the north east), the fine detailing at the roof eaves survives (Fig 21). Large corbel blocks, carved with two heads, support a projecting kneeler stone, which in turn carries the stepped coping of the gable parapet. No doubt there was a similar detail to the south-west corner, but this has been truncated in later alterations.

Fig 20

Oakham Castle Hall: top of west buttress, with ashlar quoins in the main wall face. Photograph: author

Fig 21

Oakham Castle Hall: detail of the eaves at the north-east corner. Photograph: author

The main walling of the gable is of irregularly coursed ironstone rubble. Amongst the ironstone are occasional randomly placed pieces of irregularly shaped limestone, perhaps the waste from the production of the main limestone dressings, as some pieces have ashlared front faces. Originally, the rubble facing would have been covered with lime render, masking any differences in the underlying stone. An irregular band of this limestone material runs across just below the buttress tops, appearing at first sight as a later insertion. Further study, however, indicates that it is original, being fully incorporated in the main build of the wall. To either side of the central corbel, inset a little from the buttresses, there are blocked holes which might have housed the two other corbels necessary to carry the timber wall plate. However, these two blocked holes are set a little above the level of the remaining corbel, so it may be that they were not for corbels, but for scaffolding putlogs. No clear evidence can be seen for the location of other corbels at the same level as the central one, though several more blocked putlog holes can be traced at higher levels in the gable masonry, including one with the probable putlog timber still remaining in situ to the internal face. Close study of the gable thus indicates that the bulk of the walling is original, twelfth-century work.

The west gable wall of the aisles provides further evidence for the nature of the attached building. At the south-west corner, the projecting chamfered plinth which runs across the south front of the hall turns on to the west gable, and the corner itself has a well-formed chamfer, with a shaped stop at high level. The same detail is found at all four corners of the building, though altered by later rebuilding to the north-east corner and somewhat concealed at the north-west corner. Clearly, any original building structure abutting the gable end must have had its walls inset from the corner, not in line with it. This detail survives best at the south-east corner, where the plinth and a higher-level moulding (which connected to the hood mould of the south door, before its relocation) extend 350mm across the gable end, indicating the extent to which the abutting walls were inset.

The north aisle has a doorway leading out of the hall into the attached structure. This doorway, which is quite wide and tall, has a rounded arch, with a plain chamfer to the outer (west) face. The inner face and much of the other masonry of the doorway is covered up, but enough is visible to prove that the doorway is original. The plain chamfer to the rear face of the arch is similar to those on the service doorways at the east end, and the inner face would no doubt have had a decorative moulding. There is a further doorway in the south aisle, but here all details are covered by modern fabric so no conclusions can be reached as to whether there was an original doorway in this location or not. The remainder of the west gable wall is also largely covered up at lower level, with no further evidence of doorways visible. Various early photographs show extensions that preceded the current buildings attached to the west gable, but no original fabric is evident.

Before considering the likely form of the attached structure at the west end, it will be best to turn to the rather more complex evidence at the east end of the hall.

East end

Concern over structural movement of the east gable led to a major programme of repair and restoration work to the east end of the hall, with a memorial tablet and bust of G H Finch placed here in 1911. The external face of this gable (Figs 22 and 23), although much refaced at lower level, preserves its original facing to the upper parts, with irregular ironstone rubble similar to that of the west end, though incorporating fewer pieces of limestone. A ledge offset around 75mm in width runs across the whole gable immediately below the level of the eaves corbel, just like the ledge to the west gable. The detail at the eaves is the same as the west gable, with finely carved heads to the north-east corbel, though the south-east corner has once again been truncated. The pointing to this gable is fuller than that to the west, so the features are less easy to read, but one possible putlog hole can be seen halfway up below the north parapet.

Fig 22

Oakham Castle Hall: the east gable. Photograph: author

Fig 23

Oakham Castle Hall: east gable elevation. Drawing: author and John Owen

The southern buttress is original and of the same form as the buttresses at the west end. The northern buttress has been rebuilt in recent times, reusing much of the masonry but with a depth of 800mm instead of the original 450mm. The buttress is shown as its original size on Clayton's plan of 1847 and on the elevation drawing in Hartshorne's article of 1848, so the rebuilding was probably part of the 1911 repair programme. Both buttresses have had their tops cut back, and are shown in this state on Wright's engraving of 1684 and the Bucks’ engraving of 1730. The quoins which run up the main wall to either side of the buttresses continue on up past the cut-back tops, showing that the original height of the buttresses was the same as at the west end.Footnote ²⁷

The north-east corner has clearly suffered from long-term structural movement. The corner has been rebuilt externally, except for the section of the wall below plinth level, which still leans outwards. When this work was done, the external faces of the ground- and first-floor doorways in the north aisle were lost. As the doorways are not recorded by Clayton in 1847, the rebuilding of the corner must have already occurred by that date. The doorways were rediscovered as part of the 1911 works (Fig 24).

Fig 24

Oakham Castle Hall: east gable internal elevation. Drawing: author and John Owen

In the upper part of the gable is a large, round-arched window. Internally this has splayed reveals with a full-height twelfth-century roll moulding, of the same type as the moulding to the interior of the service doorways. Externally, however, there are plain slightly splayed reveals, with no moulding or decoration whatsoever. The window now has two pointed lights divided by a square mullion, with small, plain chamfers. It is shown with pointed tracery on the Bucks’ engraving of 1730 but is shown as blocked externally on Hartshorne's elevation of 1848. Evidently, however, the window details were visible internally, as they appear on Clayton's drawing of 1847. It was noted on Gotch's drawing of 1909 as still blocked, but, by 1924, Holland Walker notes that the window had ‘modern tracery’, so the unblocking and new tracery presumably date from the 1911 works.Footnote ²⁸ Below the main gable window is a blocked six-light mullioned window. With its chamfered surround and neat cyma-moulded cornice of finer-grained limestone, this probably dates from the early seventeenth century. The section of stonework above the six-light window, up to the sill of the upper window, is a refacing, probably from when the lower window was inserted. The window is shown on the Bucks’ engraving of 1730 and appears on a photograph of c 1880 with external shutters.Footnote ²⁹ It must have been blocked up in 1911, when the upper gable window was reopened.

Interpretation

Having examined the detailed evidence presented by both gable ends, we can now turn to the interpretation. At the west end, the evidence for an attached building is clear. The corbel would have supported a wall plate in the same manner as the corbels of the aisle roof (though at a rather higher level), with the irregular offset ledge protected by the roof above. The roof itself would have been of lean-to form, like the side aisles.

At the east end the evidence is similar, except for the problem of the upper gable window. The presence of the gable window is in direct conflict with the offset ledge. Close examination shows that the external jambs of the window have to incorporate a step to accommodate the ledge, a very awkward detail which suggests the window is a later insertion. Further consideration of its strangely blank external surround also supports the suggestion that the window is inserted. All the other external openings to the hall are highly decorated, and a large and prominent gable window would surely have been given similar treatment. Although the roll moulding and splayed reveals to the interior are certainly genuine twelfth-century fabric, the conclusion has to be that they are reused from elsewhere on site. The only evidence which might counter this conclusion is that there is very little sign of a scar or disruption to the rubble stonework around the head of the window arch, though it may just be that the work has blended in well. With good grounds for concluding that the window is a later insertion, the identical arrangement of the offset ledge and buttresses at both gable ends suggests that the east gable, like the west, was originally built with a lean-to roof, with a wall plate supported on stone corbels. The central corbel, matching that at the west end, would have been removed when the gable window was inserted, if not before. The undisturbed masonry of the west end shows of course that there was never a gable window here.

A reconstruction of the proposed original arrangement is shown in Figures 25 and 26. The lean-to end roofs are assumed to have been at the same 45-degree angle as the current roofs on the side aisles, a minimum angle for local traditional roofing materials, which also avoids any awkward junction to the side aisle roof. Because the walls of the lean-to ends are inset 350mm from the corners of the hall block (from the evidence of the string course), the eaves height is 350mm higher than the eaves of the side aisles. The resultant lean-to compartments have a width of around 4.0m to the outside face of the walls (compared to the equivalent side aisle dimension of 3.7m). As the drawings show, the proposed roof reconstruction fits very well with the evidence of the upstanding building. An interesting point to note is that the top wall plate of the roof does not span the whole width of the gable end but stops well short of the corners. The north and south roof slopes sweep up to meet the wall plate at this point, leaving the eaves corbels, with their finely carved heads, outside the roof line and fully visible. The sloping roof also fits neatly over the top of the buttresses, which are taken up as high as possible to support the gable, without protruding awkwardly above the roof.

Fig 25

Oakham Castle Hall: a reconstruction of the original plan. Drawing: author and Stephen Oliver

Fig 26

Oakham Castle Hall: a reconstruction of the original building. Drawing: Allan Adams

This theory does, however, encounter a problem when compared to the archaeological evidence discovered by Barber. The east end wall of the service block, as excavated by Barber, had its external face about 7.0m from the gable of the hall block, rather than the 4.0m dimension of the proposed lean-to ends. A lean-to 7.0m wide is wholly implausible and would require a roof pitch for the eastern slope of 30–35 degrees, which would not be possible with the available roofing materials of stone slate, thatch or timber shingles (Fig 27). The conclusion has to be that the roof over Barber's building was formed as a cross wing, with a ridge running at 90 degrees to the hall's main roof. However, a cross-wing roof with its wall plate set at this height would have been a lofty structure, for which the extant low-set first-floor doorway, squeezed under the line of the aisle roof, makes no sense at all. A cross wing with a wall plate set at this height would also form an awkward junction with the hall block, requiring large additional masonry walls rising above the current gable ends of the side aisles. Also, while the use of corbels to support a wall plate is widely used for lean-to aisle roofs, one would expect a larger offset ledge to support a cross-wing roof structure here.Footnote ³⁰

Fig 27

Oakham Castle Hall: the projected outlines of the phase 1 lean-to end and the phase 2 cross wing. Drawing: author and Stephen Oliver

The solution to this conundrum is that Barber's walls must relate to a second phase of development, when a cross wing replaced the original lean-to end (see the phase 2 reconstruction drawing in Fig 39). The wall plate of this cross wing would have been placed at a lower level than the original wall plate for the lean-to. It is interesting to note that the tops of the buttresses have been cut back at the east end (though not at the west end), so that they did not rise above the wall-plate level of the phase 2 roof. Without this adaptation, the buttress tops would have created an awkward obstruction in the valley gutter (see Fig 27).

Excavation just to the west of Barber's Trench 1 was undertaken in June 2012 as part of an investigation of the site by the television programme Time Team. No evidence was found of a foundation or a robbed-out trench for the proposed lean-to end wall, but the foundation may have been quite shallow, and hence obliterated by the construction of the phase 2 cross wing.

It is immediately apparent that, with a width of only around 4.0m and a low sloping roof, the rooms in the lean-to ends were very much smaller spaces than the main hall and distinctly subservient to it. To the west, the known entry into the lean-to is at the north end. Measurement of the original size of this doorway, which is now somewhat concealed, reveals that it is of quite impressive size, with a width of 1.27m and a height of around 3.0m to the rere-arch. The two major service doorways at the east end are slightly smaller, with a width of 1.21m and an arch height of around 2.6m. The round-arched doorway in the north-east corner is of similar width but considerably lower, with an arch height of only 2.3m. The relative size suggests that the doorway at the west end was of considerable importance, and yet it led into a fairly constricted space beyond. The ground floor would have been long and narrow if it formed a single room, or two rather small rooms if divided into two. Allowing space for the lofty doorway, a first floor, if one existed, would have been even more awkward than that to the east, forming a narrow space with limited headroom under the double slope of the lean-to roof. The two original stone buttresses would also have formed a considerable intrusion into any rooms on the ground or first floor. Whatever the use of the west end, it is clear that it could never have accommodated a chamber of any significance, particularly anything to bear comparison with the magnificence of the main hall.

A more likely reason for the provision of a fine doorway at the north-west corner of the hall is that it led out through the lean-to end to another building – probably a free-standing chamber block. Some rather intriguing evidence on the west gable may support this theory. The external face of the wall above the north-west doorway has a short length of raking stone weathering, with a further section of horizontal weathering above it. These weatherings must relate to former roof slopes which abutted here, and represent two phases of work. The raking weathering is very poorly placed to have served as the abutment for the original lean-to roof, which would then have had to incorporate an awkward step. It seems more likely that this is a survival from the small, double-pitched roof of a pentice corridor, leading to a separate building such as a chamber block. When the rest of the lean-to end became redundant, its corridor function would have been replaced with the pentice. There are frequent documentary references from the thirteenth and fourteenth centuries for such pentice structures being used to connect various separate buildings.Footnote ³¹ At Oakham, documents record that a passageway of this sort was built in 1376 connecting a newly built chapel to the hall.Footnote ³² The surviving horizontal weathering would have served a much later lean-to extension, with most of the earlier raking weathering of the pentice taken out and reset in its current position.

To the east, there is clearer evidence for the arrangement of the ground floor to the lean-to end, though the first floor is more problematic. Barber's excavations established that the central service doorway of the phase 2 building led into a passage to the kitchen. This may well also have been the case in the original building, giving a larger service room to the south and a smaller one to the north, each accessed directly from the hall (not the kitchen passage) in the usual service-end arrangement. One might expect the cooler, northern room to be the buttery and the larger room to be the pantry. Alternatively (and perhaps less likely), the central and southern service doorways, which are symmetrically placed on the central axis of the hall, may have led into the buttery and pantry, with the northern doorway leading via a passage to a detached kitchen. Above, the first floor within the lean-to end would have formed an awkward space, long and narrow, with headroom restricted by the double slope of the roof, poorly lit and with buttresses protruding from its west wall. The access doorway was in the north aisle, squeezed in below the aisle roof slope and located tightly over the ground-floor doorway. The first-floor doorway has a plain-chamfered round arch and is quite low – only 1.35m to the arch springing and 1.80m to the arch top. Headroom on the first floor would have been further reduced as all three service doorways have an original stone step of around 230mm, suggesting that the original service-end floors were set above that of the hall (rather than below it, as is more common). All this indicates that the first floor was originally designed very much as a subordinate space, and could never have been intended to be a chamber of any significance. It seems likely that it was used for storage, or by servants.

The first-floor doorway seems to have been accessed by a short landing (over the buttery doorway) and a stair, in the north-east corner of the hall. On the north wall there is an offset ledge, set at the right height for a floor platform in front of the first-floor doorway. The offset ledge runs along the north wall for 2.8m, but it now overlaps the location for an original window reveal, so it seems the offset was extended when this window was (much later) converted into a doorway (now blocked). The stair was probably set against the north wall (rather than the east wall), but would have had to be very steep if it was to avoid running across the north-east window. The offset ledge may originally have been at a raking angle to support the stairs, and been tidied up to form a horizontal ledge after the stairs were removed.

It may be thought surprising that the stair was allowed to protrude into the hall, especially given the subservient nature of the first-floor space. This problem has exercised others, for example Radford, who suggested that the ground-floor doorway led through the east gable to a stair, which rose up outside the hall, emerging through the first-floor doorway on to a gallery across the east end of the hall. This seems extremely unlikely, as galleries over cross passages are now understood to be a rather later development, and a gallery in this location would clash with the finely carved corbels of the arcade. Support for the location of a stair inside the hall as part of the original design is also provided by the placing of the north-east window, which is set some distance away from the corner, unlike the windows at the west end. There are good parallels for the location of such a stair within the hall (see below) at King John's House, Warnford (Hampshire), Bishop Auckland (County Durham) and Bricquebec (Normandy), as well as the well-known late thirteenth-century example at Stokesay Castle (Shropshire).

A final point to make about the lean-to ends to the hall is the nature of their construction. It seems likely that the lean-to ends were built of timber, rather than stone. The locations where the external walls of the lean-to ends would have abutted the hall gables have been largely refaced or are covered up, so the evidence here is inconclusive, though one might expect to see some slight indication in the masonry if the attached walls were of stone. Much more significant is the evidence of the hall buttresses. The buttresses are fully finished pieces of masonry, with no allowance for connection to abutting walls. If the lean-to ends had been built of stone, it seems likely that the buttresses would have been integrated into the structure, not left as free-standing units. Rather than building carefully finished buttresses to receive the thrust from the hall arcades, the more straightforward solution in a masonry building would have been to include cross walls in the lean-to ends. There is also no evidence of scars from masonry to the internal walls of the presumed kitchen passage, with enough original stonework visible to prove this fairly convincingly. A comparable example is provided by Clarendon Palace (Wiltshire), a late twelfth-century hall which had a stone arcade. A butt joint where the service end met the hall gable indicated that the original service end was of timber, which was replaced by masonry in the thirteenth century.Footnote ³³ Such timber walls may have had only shallow foundations, which would help to explain why no foundation evidence was found in the Time Team trench.

The hall roof structure

The recent research initiative has also included a study of the hall roof structure, accompanied by a programme of tree-ring dating, which has shed new light on several aspects of the building's development.Footnote ³⁴ The hall roof has generally been dismissed as of relatively modern date. The roof of the main ‘nave’ is made up of two different elements. At upper level, the roof has five king-post trusses, dividing the building into six equal bays (Fig 28). At lower level, however, there are three large tie beams, with arch braces rising from timber corbels set over each of the stone arcade columns (Fig 29). The aisles have simpler lean-to roofs, with principal rafters and tenoned purlins. Hartshorne in 1848 said that ‘the oldest portions of the present [roof] are two red beams put up by Villiers duke of Buckingham, who also built the gateway’.Footnote ³⁵

Fig 28

Oakham Castle Hall: cross section with king-post truss. Drawing: author and John Owen

Fig 29

Oakham Castle Hall: cross section with arch-braced tie beam. Drawing: author and John Owen

Examination of the roof from ground level revealed that a number of timbers were clearly reused, with empty matrices for lap joints of early character. It was therefore decided to carry out a programme of tree-ring dating, with access to the high-level timbers via a scaffold tower. This examination found extensive evidence of reused timbers. Five of the ten struts to the king-post trusses have empty joint matrices from earlier use, and cut-down joint matrices or pegholes were visible in more than forty of the 250 or so lengths of ‘nave’ rafters. Six rafters of reused timber were also noted to the aisles. No doubt full inspection would identify many more reused timbers. A group of twenty-one reused timbers was found to have an estimated felling date range of 1160–85, and clearly must have come from the original twelfth-century roof. Although insufficient evidence remains to allow a definite reconstruction of its form, it seems that the original roof had common rafter couples, with lap-jointed collars and canted braces (soulaces) as tentatively indicated in Figure 30. The surviving masonry, with no evidence for corbels, indicates that there were no arch braces below wall-plate level. The considerable outward movement of the arcade walls (see below) might indicate that there were no original tie beams. Joints, as can be seen on the king-post strut timbers (Fig 31) were of notch-lapped form, a standard type on early roofs. The original rafters were around 160mm by 140mm in section (as seen on the struts), but had been sawn in half for reuse, resulting in the current rafter sections of around 80mm by 125mm.

Fig 30

Oakham Castle Hall: tentative reconstruction of the original roof (it is not certain that the roof had any tie beams). Drawing: author and Stephen Oliver

Fig 31

Oakham Castle Hall: a notch-lapped joint matrix to one of the reused 1180s timbers, with tree-ring sampling in progress. Photograph: author

A second group of thirteen reused timbers was found to date from the late sixteenth century, around 1575–1600. Four of these timbers, interestingly, had lap-joint matrices of similar type to the 1160–85 timbers. This indicates that the original roof still largely survived at this date, and that a series of repairs was carried out which followed the twelfth-century form rather than substituting more up-to-date construction.

It seems that this late sixteenth-century programme of repairs did not resolve the problems of the roof, as a further phase of work was carried out soon afterwards. The structure of the lower tie beams, with their arch braces, was found to date from 1619–44, which fits with the tradition that the hall was repaired by the duke of Buckingham, who owned Oakham from 1621 to 1628. Examination indicated that these were inserted as a remarkably innovative piece of structural repair. The tie beams are carried right through to the outer faces of the ‘nave’ walls, with vertical posts tenoned to the underside, sandwiching the wall top (see Fig 29). The three protruding beam ends can be clearly seen on Wright's drawing of the exterior in 1684 (Fig 4). At the foot, the posts are tenoned into horizontal timbers, inserted through the stone arcades. The whole assembly is carefully designed to provide restraint to the ‘nave’ walls, which (as can be seen) had developed a considerable outward lean by this date. The design provides an elegant solution to a structural problem, with the curving arch braces rising from well-shaped corbels. There was further ogee moulding to the drop finials which used to adorn the underside of the tie beams, as shown on Clayton's drawings of 1847. The complete absence of other timbers dating from this phase suggests that the original common rafter roof was still in situ at this time, with work restricted to the insertion of the tie beams.

The final major phase of work came around 1737, the felling date of twenty samples taken from the current king-post roof structure. The ‘nave’ and aisle roofs were completely replaced at this date, though reusing much of the earlier timber. The work involved the replacement of the central tie beam, with the two other tie beams and the arch braces being retained. The roof of 1737 probably had dormer windows at the west end of both the south and north aisles (like those on the drawings of 1684 and 1730), which ensured good light to the ‘high’ end of the hall, presumably the main focus of the courtroom. These two large dormers were subsequently replaced in the nineteenth century by the current six smaller dormer windows.

Other aspects of the original design

The overall design of the hall shows a high level of coherence and regularity. The four-bay stone arcades are well proportioned, and neatly buttressed at either end. To either side, matching aisles sit neatly against the ‘nave’, with the lean-to ends fitted against the gables. This is a building which has clearly been carefully planned as a single, well-integrated whole. Measured survey indicates that the building has been very accurately laid out in a rectilinear form. In recent decades the systematic use of proportional ratios, or ‘constructive geometry’, has been recognised in some major medieval ecclesiastical buildings (such as Norwich and Ely cathedralsFootnote ³⁶) and a few secular ones (such as Orford CastleFootnote ³⁷). The proportion most commonly used was the ratio between the side of a square and its diagonal, ie 1:√2 (1:1.414). Analysis of the plan at Oakham suggests that this system was used here also. If one lays out a square using the distance between the central axis of the columns for one side, then an arc struck from the diagonal of the square (using the 1:√2 system) almost exactly bisects the internal length of the building (Fig 32). The use of the centre line of the arcade walls and the internal faces of the end walls at Oakham follows the same system as that at Norwich Cathedral. The internal plan form is thus made up of two rectangles, with sides in the proportion of 1:1.414. Measured on the ground, the length of each rectangle is accurate according to this ratio to within around 130mm, or 0.6%. It seems quite likely that the master mason who laid out the plan of Oakham followed this system, using stakes and ropes as compasses rather than any complex calculation. At Norwich and elsewhere further detailed design of the building was made with the same 1:1.414 proportion, but its use at Oakham seems to be restricted to the setting out of the ‘nave’.

Fig 32

Oakham Castle Hall: the ground plan as set out with two 1:√2 rectangles. Drawing: author and Stephen Oliver

Another feature connected with contemporary masons’ practice on high-status buildings is the presence of masons' marks. Quite a number of these marks, of several different designs, can be seen on the internal jambs of the window and door openings. No doubt further close inspection at high level would reveal many more examples.

Turning to the exterior of the building, several points are worth noting. While the design of the hall presents a good face on all sides, the front elevation is accorded somewhat superior treatment to the rear. As well as the impressive main doorway to the front, the windows here have roll mouldings to their arches, while the rear windows are plain-chamfered. The plinth moulding to the rear is also given a plainer treatment. The external masonry has been subject to refacing in many areas, but the principal architectural dressings survive in remarkably intact state. One section of the original coursed rubble walling survives to the rear wall of the north aisle. Here the brown ironstone is interspersed with single courses of limestone ashlar, giving a striped, polychrome appearance (Fig 33). This is clearly intentional, and was no doubt a treatment given to all the aisle walling. This decorative effect, made possible by the ready availability of both ironstone and limestone, continued to appeal to local builders. It can be seen in the early fourteenth-century work at Ashwell church, three miles to the north of Oakham, as well as in a number of seventeenth-century houses in the area. A much grander application of polychrome banding, contemporary with that at Oakham, is seen in the late twelfth-century great tower at Dover Castle. The west and east gable walls, however, as noted above, are built of irregularly coursed rubble. Only the tops of the gables, above the level of the lean-to roofs, would have been exposed externally. No doubt, as with most medieval walling of irregular rubble, these wall faces were originally covered with render.

Fig 33

Oakham Castle Hall: polychrome stonework to the rear wall, with original windows. Photograph: author

Another question is whether the hall originally had a clerestory, with the walls rising above the stone arcades and windows to light the central nave. The halls at Leicester Castle (c 1150–60) and the bishop's palace, Hereford (1179), two of the best surviving comparables for Oakham, both have evidence for clerestories, though the arcade structures to both of these buildings were timber-framed.Footnote ³⁸ In his report of 1955, Radford noted that: ‘The side walls of the nave now rise barely 1ft above the top of the aisle roofs. But the substantial character of the arcades implies that they were originally designed to rise as a clerestory … This is a reconstruction of the early seventeenth century, when the outer surface of the nave walls was refaced above the aisle roofs with a single course of ashlar.’Footnote ³⁹

However, several factors suggest there may not have been a clerestory. The single course of ashlar noted by Radford is carefully set back by about 100mm from the line of the main walls, with a chamfered weathering below (see Fig 21). A recess of this type at the corner of a clerestory wall and gable end would be an unusual feature, the normal arrangement being a plain square corner (unless there are pilaster buttresses to both sides of the corner). It is very unlikely that this feature is the result of later alteration, for which there would be no rationale. The eaves detail to both gables, with its decorative corbels, kneelers and gable coping (Fig 21), is either the original undisturbed arrangement, or has been reset with unusual care. The chronology of work to the roof structure also mitigates against an original clerestory. The dismantling of a clerestory would have involved reconstruction of the original roof. As noted above, detailed study of the roof structure, together with extensive sampling for dendrochronology, has identified only two known phases of alterations to the original roof prior to the major reconstruction of 1737. The first phase of alteration was undertaken in 1575–1600, and the evidence suggests that this involved a series of repairs to the original roof, which was still in situ. If a clerestory had been removed, presumably because of decay or structural problems, one would expect a well-constructed new roof, of contemporary form, with considerable new timber. The second phase of alteration was in 1621–8, and involved the insertion of three arch-braced tie beams to restrain the spreading walls – clearly a problem not adequately addressed in 1575–1600. These tie beams were inserted as completely independent structures, again indicating that the original roof form was retained, not reconstructed. Speed's depiction of the hall in 1611 (see Fig 4), although only a thumbnail sketch, also shows the hall without a clerestory.

It is somewhat surprising that in such a fine building there seem to have been no original windows to a clerestory or to the gable ends. Lit only by the aisle windows, the interior of the hall would have been rather dark. It would have been straightforward to include small windows at high level in the gables, above the slope of the return aisle roofs. Twelfth-century masons were also perfectly able to build impressive clerestories, as seen in many churches of the period. However, there are other examples of early aisled halls with no clerestory, such as the timber-framed examples at Fyfield Hall (Essex), Temple Balsall (Warwickshire) and Burmington Manor (Warwickshire).Footnote ⁴⁰ The width of the hall at Oakham was less than these examples, and considerably less than the clerestoried halls at Leicester and Hereford, so the light from windows in the aisle walls would have reached further in. From around 1200, the grander stone-built aisled halls, such as the bishop's palace at Bishop Auckland, Winchester Castle or Lincoln Bishop's Palace, had tall aisle windows with dormered gables above, making a clerestory unnecessary.

The original covering to the roof was probably of local stone slates, like the existing Collyweston slates. In 1383, 5,000 ‘sklat’ were bought for Oakham Castle at Collyweston, at a cost of 6s 8d per 1,000.Footnote ⁴¹ Such stone slates were clearly already in use at Bulwick (Northamptonshire) by 1272, when the former hall was described as being ‘roofed with stone’.Footnote ⁴² Barber also noted Collyweston fragments in his excavations, and these were encountered widely in the various trenches dug in June 2012.

The placing of the openings in the front and rear walls is also of interest. In the front wall, the four windows were originally spaced at regular intervals, with the entrance doorway pushed up hard against the east end, as dictated by the low-end entry to the hall. The placing of windows in the rear wall was rather different. The eastern window, of which only one jamb remains, was located around 2.6m west of the corner, which allowed space, as noted above, for a stairway to the upper door. The next window to the west has been converted into a doorway, giving access into the northern extension. However, analysis indicates that this window surround has also been relocated, a point not previously noticed (compare Figs 7 and 25). Dryden's plan, published to accompany Hartshorne's article of 1848, shows the window around 1.5m further east, beside the south-east corner of the modern ‘Grand Jury Room’.Footnote ⁴³ Clayton's plan of 1847 shows the window further west, and forming a doorway on the central axis of the northern extension. Examination of the masonry of the internal arch to the window shows it is quite irregular in comparison with that of the other arches, confirming that it was reset in the mid-nineteenth century, no doubt to give a better, more central, location for the doorway. The original arrangement of the windows in the rear wall was thus quite uneven, with regular spacing to the west, a blank section of wall to the centre, and two close-set windows towards the east. Given the purposeful nature of all the other building details, this must have been a conscious decision. Perhaps there was some central feature or piece of fixed furniture here – or even a rear doorway, though this would be most unexpected.

Hall and chambers

As noted above, most previous assessments of Oakham have concluded that the hall had an attached block which contained the principal chamber. Radford (Reference Radford1955) believed there was a two-storeyed solar block at the west end, with a lesser two-storeyed block to the east. Faulkner (Reference Faulkner1958) thought there was a two-storeyed cross wing at the east end, containing a solar over the service rooms. He believed there was no attached block at the west end, Oakham thus forming a classic example of his ‘end hall house’. Gardiner (Reference Gardiner2008) concluded that the main chamber was over the services at the east end, with only a single-storey block to the west.

In his seminal article of 1993, John Blair, FSA, challenged the previous consensus that the two-storeyed blocks of late twelfth- or early thirteenth-century date found on high-status sites were ‘first-floor halls’. He laid out convincing arguments, now widely accepted, that these ranges were in fact independent chamber blocks, originally accompanied by free-standing halls. Blair noted that: ‘The typical late medieval arrangement at the lower end of the hall, with a cross passage, two service rooms and secondary chambers above, developed during c 1180–95. From the early thirteenth century it was increasingly common for the main chamber block to be attached to the upper end of the hall, producing the standard late medieval English house.’Footnote ⁴⁴

The new interpretation of Oakham's original form that is put forward here enables this classic example of an early English hall to be placed firmly within the pattern of development that Blair proposed. The lean-to form of the original blocks attached to each gable end meant that the space for an upper floor, as noted above, was very constricted. Any chamber in this location would have been narrow, with low headroom, further obstructed by the protruding buttresses. We can conclude that the room accessed by the stairs in the north-east corner of the hall was very much a subsidiary space, as would also be the case for any upper room at the west end. The secondary character of these appendages, in comparison with the magnificence of the main hall, is further confirmed by the indications that they were built of timber, not masonry. Instead, the principal chamber that one would expect to find in association with such a fine hall must have been housed in an independent block, somewhere else on the site.

The earliest documentary reference for a chamber at Oakham is the Rutland eyre of 1253, when the judges of the civil court withdrew to a separate chamber to consider the levels of fines (see below). It seems likely that this would have been in the original late twelfth-century chamber block. By the time of the inquisition of 1340 (see above), it was noted that the castle contained four chambers (iiii camerae), as well as one chamber for the gatekeeper. By this date, the lean-to end to the east had been rebuilt as a cross wing (see below), which would have allowed space for one, or perhaps two, chambers over the rebuilt service rooms. There is a further useful reference to chambers in 1375, when ‘the king's two great chambers’ were pargeted and whitewashed.Footnote ⁴⁵

The location of Oakham's original principal chamber block, like that of most of the other buildings listed in the 1340 inquisition, has not yet been established. Extensive earthworks within the bailey present a complex picture. A geophysical survey, using both magnetometry and resistivity was carried out in 2005,Footnote ⁴⁶ and another in June 2012 was undertaken as part of the Time Team investigations, with the addition of ground-penetrating radar.Footnote ⁴⁷ Unfortunately none of these surveys produced clear results. One might expect the chamber block to be located to the west, beyond the high end of the hall, in the typical position that Blair describes. However, the space between the hall and the surrounding ramparts is quite constricted, especially in comparison with the extensive main area of the bailey. Excavation of several trenches by Time Team, including to the west, north west and north of the hall, found some evidence of various walls, including possible remains of a pentice/corridor to the west of the hall, but no clear evidence for a substantial chamber block of typical rectilinear plan form.Footnote ⁴⁸

Another possible location for chambers would be within a stone tower on the motte, as occurs at many castle sites of the twelfth and early thirteenth centuries. But this is very unlikely at Oakham, where any structure on the motte seems to have been lost at an early date. There is no mention of a tower at Oakham in any of the early documents, including the inquisition of 1340. By contrast, for example, the entries which follow Oakham in that inquisition note a ‘stone tower’ at Fotheringhay Castle and an ‘old tower’ at Stamford Castle.Footnote ⁴⁹

Besides this theoretical evidence for a lost chamber block, elements of surviving twelfth-century fabric can be identified that may have belonged to the missing building. The masonry to the inside of the large window in the east gable, with its round arch and roll-moulded reveals, is similar to the front doorway and clearly of late twelfth-century date, but (as set out above) the gable window has been inserted at a later date. It could not have formed part of the original hall, so must come from another building on the site of the same date. The window was probably moved to the end gable of the hall in the fifteenth or sixteenth century, when the other buildings except for the hall were cleared away (see below). Given the size of the arched opening (which is around 200mm wider than the hall doorway) this masonry must have formed either a large window or a doorway to a very substantial building. The most likely location for such a fine feature would of course be an original twelfth-century chamber block. It seems unlikely that a chamber would have a larger doorway than the hall. Combined with the more pronounced splay to the reveals of the gable opening, this suggests it was probably always a window, rather than a doorway. A window of such size, unusual at such an early date, shows that the building it came from was an impressive structure.

As well as the gable window, there is further twelfth-century fabric that does not belong to the hall, in the shape of various pieces of reset sculpture now dispersed about the building. The sculptural scheme to the hall itself is complete, with each available location, as defined by the architecture, already occupied. The reset pieces of sculpture must therefore have come from other buildings on site, and were no doubt incorporated into the fabric of the hall when the site was cleared, or were retrieved at some later stage. There are three upright figure sculptures, two set into the front south wall of the hall and another of a heraldic lion, now displayed on the floor inside (Fig 34). The Bucks' engraving of 1730 shows two extra upright sculptures to the south front, one probably being the lion now relocated inside. There are also two sculptures of crouching beasts, one on the south front and the other inside, beneath the east gable window. A smaller, less distinctive sculpture, probably of a head, is set into the exterior of the east gable.

Fig 34

Oakham Castle Hall: the lion sculpture, now on display inside the hall. Photograph: author

All this sculpture is of rather unusual form, particularly the upright figures. Medieval sculpture is often incorporated in the form of corbels, but the reset pieces at Oakham would not fit into such contexts. The upright figures are rather similar to the musician figures in the hall arcade and, like them, must have formed part of an unusual type of decorative scheme. One of the reset figures, at the east end of the south front, is of similar size to the musician figures, and has the same seated, open-kneed posture. The other human figure and the lion, however, are of differing sizes and character, so no coherent pattern can be deduced. Similarly, an unusual architectural scheme would be necessary to accommodate the two crouching beasts. Perhaps the chamber block incorporated decorative blind arcading, like that in the late twelfth-century chambers at Bricquebec, Beaumont-le-Richard and Creully in Normandy.Footnote ⁵⁰

The combined evidence of the reset gable window and the sculpture strongly suggests that an impressive chamber block was built by Walkelin de Ferrers around the same time as the hall. It is also of course feasible that the reused masonry came from a chapel (documented in existence from 1300Footnote ⁵¹), but a chamber block seems a more likely source. One would also expect sculpture such as the heraldic lion to be found in a secular, rather than an ecclesiastical, context. The lost chamber block was thus large enough to have at least one window of considerable size for the period, and was decorated with an unusual array of sculpture, echoing that in the hall.

Oakham and the development of the medieval hall

Having examined Oakham in some detail, it will be useful to consider its development in a wider context.

Plan form and function

Oakham stands at an important point in the development of the standard plan form of the English medieval hall. Comparable buildings of early date and similar high status are generally of aisled construction, though few survive with much of their fabric intact.

The hall at Oakham originally had the front entrance placed tightly against one of the end walls. In later halls, there was an opposing doorway in the rear wall, creating the cross passage, which became a defining feature of the standard medieval plan. At Oakham, however, despite the presence of three service-end doorways in the entry end, there was no full cross passage, as the stairway blocked the rear wall. As Blair has said, the earliest halls generally had the front doorway placed more centrally, with no proto-passage, as at the bishops’ palaces at Hereford (1179) and Old Sarum (c 1180), and still found at the archbishop's palace at Canterbury in the early thirteenth century.Footnote ⁵² Other examples of halls like Oakham with an end entry but no full cross passage can be seen at King John's House, Warnford, of c 1180–1210 (Fig 35),Footnote ⁵³ Lincoln Bishop's Palace (early thirteenth century) and probably at Clarendon Palace (Wiltshire) of c 1180. Site topography of course could prevent the possibility of a rear door. At Leicester Castle (c 1150–60), steeply sloping ground would have made any door in the rear west wall difficult. The rear wall here had a row of six original windows and no doorway, though the location of the original front door is not known. At the much later Stokesay Castle, Shropshire (1285–90), the hall is built against the moat, so no cross passage was possible. The earliest proven example of a full cross passage is in the bishop's palace at Bishop Auckland of c 1190, where the blocked rear doorway is still visible (Fig 36).Footnote ⁵⁴ Farnham Castle (Surrey) probably had a full cross passage, and was previously thought to date from c 1150–60, though tree-ring dating now indicates a date of c 1190–1220.Footnote ⁵⁵

Fig 35

King John's House, Warnford: a reconstruction of the original plan. Drawing: author and Robert Ovens

Fig 36

The hall at Bishop Auckland, County Durham: a reconstruction of the original plan. Drawing: author and Robert Ovens

The placing of service doors in one end wall of the hall is another key feature of the typical medieval plan. As can be seen at Oakham, this development slightly pre-dates the adoption of the full cross passage. Leicester Castle (c 1150–60) already had paired service doorways in the north gable wall. Bishop Auckland (c 1190) shows the fully developed arrangement with three doors, set out with careful symmetry in the central ‘nave’ of the hall. However, at Warnford (c 1180–1210) there are only two doorways, of different sizes and set asymmetrically to the ‘nave’. Farnham Castle (c 1190–1220) has three doors, the central larger one leading to the kitchen. During the early thirteenth century, the symmetrical, three-door arrangement became the standard form, as seen at Lincoln Bishop's Palace, where the three linked service doorways were given a fine architectural treatment, separated by jambs with attached shafts. It is interesting to note that at Oakham the two central doors were carefully set symmetrically, as if to lead to the standard paired rooms of buttery and pantry. But instead, the left door seems to have led to the kitchen passage, with a third door (probably for the buttery) pushed out to the north-east corner under the stairs, and given an older-style round arch.

The treatment of the cross passage and service doors at the impressive aisled hall of Winchester Castle (1222–35) throws an interesting light on the priorities which influenced the design of early halls.Footnote ⁵⁶ This very large hall, rebuilt within the existing castle site, has clearly been squeezed into a space which imposed specific constraints, owing to pre-existing structures. Several compromises therefore had to be made (Fig 37). The normal desired plan would be to have the cross-entry doorway in the end bay, with the service doors to one side, leading into the service end. As the east gable of the hall is built right up against the outer castle wall, there was no space here for the service rooms, which had to be located separately, to the north and south. The cross-entry doors could just have been placed in the east-end bay but, interestingly, that was not the design choice. Instead, they are pushed westwards into the second bay, which enables doorways for the service rooms to enter the end east bay – albeit from the north and south sides, not the east end. It seems it was more important to preserve the concept of the service entry being separated from the main entry, and service entry being into the end bay, than it was to have a cross entry at the far end of the hall. This is probably linked to the need for a processional route from the service areas, especially the kitchen, to the high table. A further point to note is that the cross-entry doorways are deliberately positioned to the east side of the second bay, not central to it. This gives an awkward, asymmetrical appearance externally, on both the north and south sides. No doubt this decision reflects the normal practice of the cross-entry doors being placed close to the end gable, not centrally to the end bay.

Fig 37

Winchester Castle hall: a reconstruction of the original plan. Drawing: © Winchester Excavations Committee; redrawn by author and Robert Ovens

As noted above, a service block attached to one end of the hall, with a chamber above, was identified by Faulkner as the ‘end hall house’ type, and played an important part in his theory of the development of the medieval English house. The attached service end at Warnford provided Faulkner with ‘an almost perfect example’ of his ‘end hall house’ type.Footnote ⁵⁷ At Warnford, the surviving evidence shows that the stairs to the chamber rose from within the north aisle, as at Oakham, precluding the possibility of a cross passage. Unlike Oakham, however, the stair at Warnford led through a fine doorway into a spacious chamber, provided with a garderobe, an inner room and probably a fireplace. A slightly different arrangement can be seen at Bishop Auckland, where the stair was squeezed into the north-east corner of the hall, leaving space for a full cross passage. Here the stairs passed right through the north wall of the hall, where an external timber staircase under a pentice roof gave access to the chamber. However, the constricted access route and the small size of the chamber in comparison with the hall suggest that, as at Oakham, the chamber over the service end here was of secondary importance. The principal chamber block, located to the south west of the hall beyond the high end, is usually thought to have been built by Bishop Bek around 1300. However, the form of this range, with a large, tall chamber over an undercroft, is typical of the early chamber blocks described by Blair, and it seems likely to have had an earlier origin.

An important aspect of the medieval hall, which followed on from the adoption of the end entry, was the distinction between the ‘low’ end, with the service entries, and the ‘high’ end, with the principal table on a raised dais. At Oakham, no evidence is visible for a dais, though there may have been a raised platform set inside the western columns. At Warnford, two corbels and slots in the attached columns of the east gable which carried a beam provide rare evidence for an early dais. Until the fuller development of the dais and its accompanying oriel window, there was little distinctive differentiation in the architecture to signify the high end, though furniture and wall hangings would no doubt have made the difference very plain. At Oakham, Emmerson (Reference Emmerson1981), followed by Pevsner and Williamson (Reference Pevsner and Williamson1984), noted that the south-west arcade capital, at the high end, has leaves carved in a more naturalistic style. The difference from the other capitals, it has to be said, is quite subtle, and would probably have gone largely unnoticed. A more appreciable signifier of high and low ends may have been the subject of the musician figures over the column capitals. The two figures at the service end of the hall are of animals, while those to the rest of the hall are human. Another standard feature of the high end was the door which led out of the hall towards the principal chamber, a pattern which probably, as noted above, points to the location of Oakham's lost chamber block.

The musician figures themselves are no doubt a representation of the band of performers who would be employed to play at feasts and celebrations. In a study of the important role which musicians played in the courtly life of twelfth-century Britain, Wright illustrates the theme with examples of the sculpture from Oakham.Footnote ⁵⁸ He quotes the contemporary poet, Chrétien de Troyes, who describes a scene remarkably reminiscent of Oakham:

Great was the joy in the hall …

This one plays the harp, this the rote,

This the gigue, this the fiddle,

This the flute and this the shawm.

The original hall: conclusions

The hall at Oakham Castle has been widely recognised as an extraordinary survival. Although much evidence has been lost, the study of comparable buildings suggests that its defining feature, the stone-built aisle arcade, was always exceptional, used only in houses of very high status, and by no means in all of these. The fact, however, that many early timber aisled buildings imitate masonry construction indicates a clear aspiration towards the use of stone. Stone was of course the standard material used for the aisle arcades of every parish church, and was generally considered the only suitable material for the house of God. There are twenty-two churches with surviving late twelfth- or early thirteenth-century arcades in the surrounding county of Rutland alone.Footnote ⁹¹ Stone arcades also feature in the grand infirmary halls of early hospitals, such as the one at Canterbury Cathedral Priory, or that at Ramsey, Huntingdonshire, which survives in remarkably complete condition, as it was later converted into a church.Footnote ⁹²

So who was the man who built this exceptional structure and what were the circumstances that led to its construction? The information available is of course very limited, but it is possible to draw some clues together.Footnote ⁹³ Oakham's builder, Walkelin de Ferrers, inherited the property in 1156/7 and died in 1201. Walkelin's great-grandfather, Henry de Ferrers, had come to England around the time of the Conquest and was ‘among the most powerful Anglo-Norman magnates’.Footnote ⁹⁴ The family's seat in Normandy was at Ferrières-St-Hilaire (Eure) and before his death around 1100 Henry had built up extensive English estates, with a major holding centred on Tutbury in Staffordshire. After his death, it seems that the estate in Normandy stayed with his eldest son, William, but the bulk of his English estates passed to his youngest son, Robert. This younger branch of the family went on to become the earls of Derby. William de Ferrers, with his Norman lands, fought alongside Duke Robert of Normandy in his defeat by Henry i at the battle of Tinchebray in 1106. Despite this, Henry i seems to have forgiven William, and probably granted Oakham to him before his death in 1131. On William's death, the Norman estates passed to his son Henry de Ferrers, who is also documented as in possession of Oakham from this date.

Walkelin de Ferrers is known to have inherited Oakham as a minor on his father Henry's death in 1156/7. In 1166, Walkelin is recorded as holding Oakham for 1½ knights’ service – a rather modest level of commitment. Although he probably also held a smaller estate at Lechlade (Gloucestershire), the principal focus of his power base clearly remained in Normandy, with 47¾ knights in his service in 1172. To judge from the surviving references, Walkelin seems to have spent little time in England. In 1174 he was a witness in Henry ii's court at Caen and in 1184 a case is referred to in his own court in Normandy. From 1189 he seems to have become, as Loyd states, ‘one of the personal following of Richard i’. In 1190 he accompanied Richard on the Crusade and he makes several appearances in the chronicle that describes the expedition. In July of that year, arriving to reinforce the siege of Acre, he appears at the head of a list of several crusaders who are specifically noted as being from Normandy (de Normannia). Early in 1191 he gains specific mention as vir ille nobilis Walkelinus de Ferrariis, assisting in the relief of poorer crusaders suffering from famine. In September 1191 he took part in the battle of Arsâf. In 1193–4, Walkelin seems to have visited Richard i during his captivity in Germany, delivering to the king £140 from the treasury at Caen. Following Richard's release, Walkelin probably continued to spend most of his time abroad in the king's retinue. He appears as a witness to royal charters in 1194 at Spires, in 1195 at Vaudreuil and in 1197 at Château Gaillard. In 1201 he paid scutage on his lands in Rutland, but died later the same year. In 1204, after the conquest of Normandy by the French king, the Anglo-Norman landowners had to choose between retaining their estates in England or in France. For Walkelin's son Henry, the Norman power base was much more important than his English lands, so Oakham was forfeited to the English Crown.

The picture that emerges from this in regard to Oakham is very interesting. From the early twelfth century onwards the Norman estates always took precedence for the senior branch of the Ferrers family. Despite investing in the construction of fine architecture at Oakham, with an elegant hall, chamber block and probably other structures, the main focus of Walkelin de Ferrers’ activities lay elsewhere, beyond England. Perhaps we have here part of the reason for the exceptional character of Walkelin's buildings at Oakham – making use of the very latest developments in architectural style, but employing these in a domestic, residential context. In contrast, at the principal castle sites owned by English nobles, the twelfth century saw an emphasis on the construction of impressive, defensive stone towers – as happened at Tutbury under the Ferrers earls of Derby. For Walkelin, his major castle base remained in Normandy, so he may have felt no need to compete with this at Oakham. However, as a rising Anglo-Norman figure, he may have wanted to ensure that his principal English estate was at least provided with some high-class accommodation for his own occasional use and to act as a high-status statement, the focal point of his English lordship. Oakham's hall no doubt served from the beginning as an important courthouse. Dating from the 1180s, Oakham would have been built when Walkelin was in his forties, or perhaps early fifties, immediately before his successful later career under Richard i. Perhaps his Continental connections had encouraged an interest in the latest architectural developments, and even led to the contacts with the French and English craftsmen at Canterbury Cathedral which made Oakham possible.

Thirteenth-century developments

Oakham Castle seems to have remained in active use throughout the thirteenth century. King John visited the castle three times in 1206–7 and Henry ii visited seven times between 1219 and 1258.Footnote ⁹⁵ The first recorded assize at Oakham was in 1229, though a gaol is referred to in 1208–9.Footnote ⁹⁶ An interesting reference for the Rutland eyre of the autumn of 1253 indicates that, on this occasion, the Crown court pleas were being heard in quadam grangia castri de Okham, presumably a large barn within the bailey. At the same time, the court of civil pleas was being held in the castle hall (in aula eiusdem castri), though the judges here withdrew to a separate chamber (in quadam camera) to agree the fines payable.Footnote ⁹⁷ During the Barons’ War of 1264–5 the castle was captured ‘by the king's enemies’ and there is a reference to the hall being damaged by fire; complaint was made in 1275 that money raised in the nearby hundred of Martinsley for rebuilding the hall, ‘which was burned in the time of the war’, had been wrongly retained by the collectors.Footnote ⁹⁸ However, either this was a different building, or any damage must have been fairly slight, as there is no evidence of scorching to the original masonry and many twelfth-century roof timbers still survive, with no timber of late thirteenth-century date identified in the dendrochronological analysis.

As argued above, the original lean-to at the east end of the hall, which housed the services, was replaced in phase 2 with a wider and taller service end with chamber above, forming a cross wing. It is this phase 2 building whose foundations were excavated by Barber in 1956–7 (see Fig 17). The only dating evidence that Barber's excavations found consisted of three pottery sherds dated to c 1230–1300 from the lowest deposits in the well located in the north service room.Footnote ⁹⁹ By the later thirteenth century, the lean-to end would probably have seemed old-fashioned, as fully formed cross wings attached to a hall were now becoming established as the standard form. It seems likely that the phase 2 development at Oakham is of this date, perhaps following the fire damage to the hall of 1264. If the lean-to ends were built of timber, as argued, they would have been the most vulnerable parts of the building.

A reconstructed plan form for phase 2 is shown in Figure 38, together with the reconstructed view in Figure 39. The phase 1 service doorway locations dictated the layout on the ground floor, with an offset passage to the kitchen, set between unequally sized service rooms. It is interesting to note that the passage walls (the line of which was also found in the 2012 Time Team excavations) are not set symmetrically around the doorway into the hall. The north wall is pulled quite tightly against the door jamb, though it still relates rather awkwardly to the phase 1 buttress to its north. To allow for a reasonable passage width, the south wall is pushed well to the south of the door jamb. On the upper floor, the gabled form and higher walls of the cross wing would have allowed for a reasonably spacious solar, still approached by the stairs and doorway in the north-east corner of the hall. One would expect this solar to have been heated, probably by a lateral fireplace on its east wall. The principal chamber at this date, however, would still have been in the independent late twelfth-century chamber block. Built with straight joints where its stonework abutted the original hall, the phase 2 cross wing has left no visible evidence of its presence on the surviving masonry, except for the reduction in height of the gable buttresses (as noted above). At the west end, the phase 1 timber-framed lean-to end was probably replaced at the same time by a pentice, connecting the hall to another building, perhaps the main chamber block.

Fig 38

Oakham Castle Hall: a reconstruction of the phase 2 plan of c 1300. Drawing: author and Stephen Oliver

Fig 39

Oakham Castle Hall: a reconstruction of the phase 2 building. Drawing: Allan Adams

A further building that may well date from a similar period is the free-standing kitchen discovered by Barber immediately to the east of the cross wing. This was a square stone-walled building, measuring 26ft (7.92m) internally, with its roof supported on four timber posts. A large hearth was located in the south-west corner, though the building was only partially excavated. A kitchen of this type, 36ft (10.97m) square internally and with four timber posts, was excavated at Clarendon Palace and is thought to be the second (north) kitchen, built in 1244.Footnote ¹⁰⁰ The roofs of such kitchens were presumably of pyramidal form, with a louvre to the upper part. In tracing the development of early kitchens, Brears notes a number of square stone-walled kitchens, with hearths against the external walls, that became a common type from the early fourteenth century.Footnote ¹⁰¹ The kitchen in the bishop's palace at Chichester, with a surviving pyramidal roof tree-ring dated to 1293–1320, is one of the earliest known examples with a hammer-beam roof, which enabled the elimination of the obstructive timber posts.Footnote ¹⁰² At Oakham, one pottery sherd was found under the north wall with a date of c 1150–1250 and nine further sherds of the late twelfth to early thirteenth century under the adjoining stone-paved floor.Footnote ¹⁰³ However, these sherds might relate to an earlier building, as the form of the excavated building appears more consistent with a date of the mid-thirteenth to early fourteenth century. The bulk of the pottery found in the kitchen area had a broader date range, of the late twelfth to early fourteenth century.

Later development

Oakham Castle continued in regular use through the fourteenth century, with six royal visits between 1301 and 1380.Footnote ¹⁰⁴ In 1300 there is the first reference to a chapel at the castle.Footnote ¹⁰⁵ The inquisition of 1340 (see above) provides an extensive list of the castle's buildings at that date, which now included four chambers, as well as a chamber in the gatehouse. An inquisition of 1347 confirms that the castle was provided with latrines, and the inquisition of 1361 mentions similar buildings to that of 1340.Footnote ¹⁰⁶ Between 1373 and 1385 Oakham was held directly by the king, so good documentary records survive for this period.Footnote ¹⁰⁷ In 1375–7 work included pargeting and whitewashing of ‘the king's two great chambers’ and the ‘great chapel’. In the following year a new chamber and chapel were built. The chapel (which was quite small, costing only £30) was built with a passageway connecting it to the hall – a further indication of the dispersed nature of the castle's buildings, even at this later date. Various other works were carried out during this period, including repairs to the kitchen and the king's chamber, and the construction of a new roasting house, in preparation for Richard ii's visit in 1380. In 1383, 5,000 roofing slates were bought at Collyweston.Footnote ¹⁰⁸ It is somewhat puzzling to find that, despite these ongoing repairs, the inquisition of 1388 states that: ‘in the castle are certain buildings in a poor and ruinous state and of no net value’.Footnote ¹⁰⁹ No further records are available to shed light on the state of the castle's buildings until an inquisition of 1521, which says that: ‘There is an old castle all ruinous, being a large ground within the mantell wall. The hall is in the best reparation, and of an old fashion, and most convenient to be upholden and kept with covering, because of the courts [that] be kept in the same.’Footnote ¹¹⁰

As noted above, the earliest depiction of Oakham Castle is the thumbnail sketch included in Speed's town plan of Oakham from 1611 (see Fig 4), which shows the hall surrounded only by ‘humps and bumps’ with no other upstanding remains of buildings. Taken with the evidence of the 1521 inquisition, this suggests that all the castle buildings except for the hall were abandoned in the course of the fifteenth century, and that there was subsequently an active programme of dismantling to clear the castle bailey of building remains. As part of this work, the cross wing at the east end of the hall must have been removed. With the east gable now exposed, the large late twelfth-century window must have been retrieved from the original chamber block and inserted into the hall gable, to throw additional lighting into the courtroom interior. Since the masonry of the window survives in fairly intact form, it seems that the chamber block must still have been standing at this point, rather than being a collapsed ruin. Most of the various pieces of late twelfth-century sculpture which were reset in the hall (as noted above) but originate from other buildings on site (perhaps from the chamber block) must also have been collected up at this period – an unusual early programme of artefact preservation. The location of one of the pieces of reset sculpture, set at the foot of the inserted east window, is a further indication that all this work may have been conducted as a single campaign. It is possible that it was carried out in the late sixteenth century, at the same time as the programme of repairs to the hall roof that has been identified by dendrochronology, though it seems more likely that it was undertaken at an earlier date.

It remains only to trace the final sequence of alterations to the hall. As noted above, three arch-braced tie beams were inserted in the hall in the 1620s, an elegant solution to combat the structural problem of the outward-leaning arcade walls. This work, together with the rebuilding of the entrance gateway, was undertaken by the duke of Buckingham, who owned Oakham in 1621–8. It seems likely that the six-light mullioned window to the lower section of the east gable, with its cyma-moulded cornice, was also inserted around this time. This window would have thrown additional light into the east end of the hall, with a large dormer window (as shown on Wright's engraving of 1684) to light the west end. It was perhaps also around this time that the north-east window was converted to a doorway (later blocked), providing the hall – at long last – with a full cross passage. At some date after the 1730 Bucks' engraving, the front doorway was moved to the centre of the building. It seems likely that this was undertaken during the eighteenth century, so that two courts for criminal and civil cases could be better accommodated, at either end of the hall.Footnote ¹¹¹ Perhaps this alteration was carried out at the same time as the construction of the new king-post roof structure (see above), with its tree-ring date of 1737. During the earlier nineteenth century, an extra room was added to the north side, noted as a ‘Modern Room’ on the earliest surviving plan of the hall, drawn by John Clayton in 1847 and in the possession of the Society of Antiquaries.