Figure 1: Great Worge, Brightling. Ground-floor plan as extended in c.1700. Doorway locations are shown only where known
The preceding article by J. T. Smith raises a number of questions regarding a group of buildings published by the present authors in VA 28 as probable examples of detached kitchens. The points he raises are of such significance to our understanding of this important class of building that, rather than demanding a riposte, they deserve a more detailed response. In so doing, we not only address the points raised by Smith, but attempt to answer the key question: why should some households require a two-storey, multi-room detached kitchen whilst others managed without? Inevitably, in so doing we raise other questions which require further research.
A group of relatively large detached ancillary domestic buildings dating from the fifteenth and early/mid sixteenth centuries were published in summary form in VA 28. In that article it was suggested that, despite their inclusion of more than one ground-floor room and upper chambers, these buildings were synonymous with the larger examples of detached kitchens mentioned in documentary references of the period. The purpose of the article was to make researchers aware of this group of buildings and to stimulate debate and further documentary research into their precise use. The tenor of the argument was that, like the buildings which documents call aula (hall) and horreum (barn), the building called a coquina or kitchen in documentary references need not always be a single-cell structure as is often supposed, but could in some instances contain a number of rooms and chambers which fulfilled a range of functions in addition to that of cooking. It also suggested that, should this view be accepted, it would be worth reviewing previously recorded buildings to ascertain whether any examples have been wrongly interpreted either as houses (in instances where the main dwelling may have been demolished subsequently) or as examples of the 'unit system' a modern term used for two independent houses set in close proximity upon a single holding.
Inevitably, these views have been met with some scepticism (see J. T. Smith's preceding paper in this volume of VA). Smith does not deny that the buildings presented in our paper were kitchens serving 'a second more important household', what he justifiably argues is that our conclusions need more published detail if they are to be accepted. Apart from asking why these buildings should not be regarded as examples of the 'unit system', the main purpose of his article, it would seem, is to seek clarification to questions not answered by the summarised published details. To this end, he specifically requested that the editor allow us to respond. In this way, he felt, the debate could be usefully advanced.
Smith is not the only person to have been critical of the summarised, and in places over- simplistic content of our article. John Walker has rightly pointed out that it failed to put forward the criteria against which these buildings were tested in order to discount them as likely examples of normal houses. In his recent article regarding an Essex detached kitchen he corrected this oversight by publishing the criteria he uses in relation to Essex (Footnote 1). The criteria used by us in Sussex are substantially the same, but since there are regional variations, they are therefore worth giving here:
Not all the above features will be recognisable within every building, but more than two must be present before a building can be considered a possible detached kitchen. The more factors present, the more confident the interpretation. Examples which do not comply should be regarded as likely examples of the unit system. It is interesting to note that when this set of criteria was applied to the houses surveyed in eastern Sussex - in order to ascertain whether any 'detached kitchens' had been wrongly diagnosed as houses - only one clear example was found, at Grovelye in Warbleton (Footnote 3). Is it surely something of an unlikely coincidence that all other buildings which contain these non-house-like characteristics occur in conjunction with a building of standard house-like form. If these buildings were a hitherto unrecognised form of house, why are they not encountered standing in isolation?
In retrospect, some of the phrases used in our article could have been more carefully chosen and should have been qualified. In particular, the references in VA 28 to these buildings as 'house like' is unfortunate, and was not helped by Smith's decision to quote a shortened version of the more fully qualified sentence containing this phrase. The full quote is 'Although all incorporate non-standard features, in general appearance the surviving examples closely resemble small houses'. The statement that the structures have the general appearance of small houses means only that they are of similar size to small houses and have the same general external appearance in that they are two storeyed, have windows on each storey and incorporate some form of hearth or chimney. Our statement 'It is often only their location, close to the rear of a main house of more standard layout, which indicates their true function' is not intended to imply that this is the only way to differentiate them from houses - a pair of unit-system houses could easily have the same relationship. The intended meaning is that these buildings were not examples of non-domestic workshops etc, but served as an adjunct to the activities of the main house. These points need to be clarified in order that the main issues are not clouded by a debate regarding unit-system houses. We share Smith's general views regarding the unit system and Smith accepts that some houses have associated detached kitchens. However, we differ as to where the boundary should be placed between structures defined as two adjacent independent houses (true examples of the unit system) and those which should be classified as a house and kitchen. It is this difference in perception which highlights the importance of the point Smith is making. Are these multi-roomed buildings, which in some ways resemble houses but which we consider not to be houses, large examples of the kitchens mentioned in contemporary documents, or are they something else? The point may seem trivial - an exercise in splitting hairs - but defining whether these buildings, as well as those of more simple form, should be accepted as synonymous with the kitchens referred to in documents is important. Because many researchers have tended to regard all detached kitchens as simple single-cell structures, there is a danger of grouping too many buildings as examples of the unit system. This tends to debase the value of the term whilst at the same time masking important links between surviving buildings and the upper strata of a class of structure referred to in contemporary documentary sources as kitchens.
Smith's main problem in coming to a conclusion regarding this latter point is a lack of published detail. Perhaps, therefore, the best way of addressing his questions and the issues he raises is to offer case studies of the five buildings he deals with. These will address the reasons for dismissing them as houses and, as far as is possible, other specific points raised by him. As with the original article, it will be necessary to give these studies in summarised form ™ those wishing further details are directed to the reports deposited at East Sussex Record Office (Footnote 4). Having dealt with the five sites, we will finish with some concluding remarks and observations.
NGR TQ 6630 2105 (Footnote 5) (Fig. 1)
Figure 1: Great Worge, Brightling. Ground-floor plan as extended in
c.1700. Doorway locations are shown only where known
There can be no argument that this example is a kitchen - it is described as such in a remarkably detailed survey of 1567. The problem is, the building to which it relates does not survive, and thus we cannot be absolutely certain that it resembled the other multi-roomed buildings discussed. The full documentary description of the kitchen is given in VA 28, p.89. In summary, it is described as located at a distance of 14ft (4.27m) from the rear of the house, of timber construction with a tiled roof. It measured 30ft x 16½ft (9.15m x 5.03m) and contained three ground-floor rooms. One of these - the room we call the kitchen - was used to dress meat in and had an oven and a malting oast within it. It was 'loathed over'. Despite the oven being in the main room, one of the other two rooms was called a bakehouse. Presumably this was where the baking was prepared rather than cooked. The other room was a milkhouse and would have served as a dairy. As with the main room, these were described as 'loathed over', but the phrase does not appear after the name of each room but as a single entry encapsulating both, perhaps suggesting that the 'loath' extended over both rooms in like fashion to the chamber over a pair of service rooms in a standard-layout medieval house. At 30ft long, the structure was probably of three bays - perhaps two bays for the main room with the other two located within the third bay. The details suggest a plan not dissimilar to that of the slightly smaller Comphurst example (see below) or the somewhat larger example recognised at Hempstead, Framfield, East Sussex in 1998 (Footnote 6).
What Smith finds problematic about the Great Worge description is the meaning of the phrase 'loathed over' - he is clearly unconvinced that this indicates a full-height upper storey of the type present in the other buildings and rather favours a low single-storey structure with a ceiled-in roof void, and it must be admitted that whereas the functions of all ground-floor rooms are given, those of the two 'loaths' are not. Luckily the 1567 kitchen description is preceded by an even more detailed description of the house, and here there can be no doubt as to what is meant by the terms used because the house still survives. The description reads as follows: '. . . one messuage or dwelling house made of timber puncheon and space and covered with tile, containing in length 47 foot and in breadth 19 foot, wherein is one hall containing in length 19 foot and in breadth 16½ foot over and beside the entry with a fair chimney of brick. There is also in the same house on the north side of the hall one little room called the parlour containing in length 11 foot and in breadth 9½ foot. And there is also in the same house on the north side of the hall another little room called the buttery containing in length 11 foot and in breadth 9½ foot. There is also on the south side of the same hall in the said house two other rooms whereof the one contains in length 12 foot and in breadth 9 foot and the other contains in length 12 foot and in breadth 9½ foot. There is also in the north end of the same house two loathes or chambers, the one of them having a chimney in it. There is also in the south end of the same house one loath or chamber' (Footnote 7).
The house at Great Worge is a closely-studded four-bay 'wealden' hall house of c.1500 date, having a standard two-room services, an overshot cross passage, single-bay open hall, and a divided high-end parlour. Already by 1567 the hall had been floored over to give a new hall chamber and a two-flue chimney had been inserted. Some of the first-floor ceilings may also have been inserted (Footnote 8). Excepting these modifications, the house still functioned in its original form, including the divided northern parlour bay with, on the ground floor, a small parlour and buttery. The buttery had diamond-section mullions to its windows, whilst those of the parlour were moulded. At this date there was no inter-communication between the service chamber and those over hall and parlour, thereby explaining the groupings given in the description.
It is important to note that, as with the description of the kitchen, the functions of the upper rooms are not given in the 1567 description of the house - they are merely described as 'loaths or chambers' (Footnote 9). But from the surviving building there can be no doubt that the term 'loath' indicated an upper room and not a roof void, for the 'loaths or chambers' within the house have side walls measuring 2.10m (6ft 10in) from first floor to top of wallplate. None of this proves categorically that the loaths within the kitchen at Great Worge had reasonable height walls - it is possible that the term was used for any upper room, regardless of whether it had lofty walls or was wholly within a roof area. However, it shows that the 1567 kitchen was two storeyed and makes it likely that it closely resembled the surviving examples presented here.
A further point is worth emphasising. When discussing the Comphurst example Smith states that ' a very important aspect of medieval domestic buildings, is width; and at about 4.8m the 'kitchen' would qualify as a small house'. The point he is trying to make is somewhat obscure, but the implication appears to be that although the width of 4.8m would be acceptable for a small house - an issue which is not in dispute - it is considered too large for the kitchen of a vernacular holding. If this is so, then it is worth emphasizing that the width of the kitchen at Great Worge is given as 16½ft (5.0m). Further, it can be shown by comparing the dimensions given in the document with those of the surviving house, that the documentary dimensions are taken internally and should be incremented by two feet in each direction in order to achieve external dimensions. Therefore the likely external dimensions for the kitchen are 32ft x 18½ft (9.75m x 5.65m). Clearly detached kitchens could resemble small houses in their dimensions.
Another point which worries Smith regarding the buildings we presented is their comparatively large size in relation to that of the main house. The ground-floor area of the kitchen at Great Worge is approximately 55 per cent that of the house (a factor of 9:5), considerably less than calculated by Smith for our other examples (but see below).
When the house at Great Worge was built it was owned by Robertsbridge Abbey who, in 1530, granted an eighty year lease to Richard Glydd, yeoman. It was his successor, Thomas Glydd, who built the kitchen, probably c.1550, though whether it replaced an earlier kitchen is not known. Although Great Worge now stands in isolation, the house formerly abutted a public footway and, to judge from the 1567 description, the kitchen and house stood in similar juxtaposition to the buildings at Comphurst (see below). The Great Worge holding was very large by local standards, amounting in total to 378¾ acres. Although a little under half the acreage was woodland, the profits of the woodland appear to have been included in the lease (Footnote 10). The Glydd family purchased the property in 1617 (Footnote 11). The date at which the kitchen was demolished is unknown. Today its place is taken by a c.1700 stone-built, combined parlour and kitchen range (Fig. 1). The fact that the new range included a kitchen supports Smith's statement that subsequent changes to a building ‘…can provide clues to the earlier uses of rooms or parts of them…’ (Footnote 12).
NGR TQ 7478 1587
(Footnote 13)
(Fig. 2 - see also VA 28, 86, fig. 1) 
Figure 2: 12/13 High Street, Battle. Ground-floor plan showing ‘kitchen’ building linked to the main house by a narrow added jettied range. Doorway locations are shown only where known
As with Great Worge, this 'kitchen' stands to the rear of a 'wealden' hall house, though in this instance there is a space of c.8.7m (compared to c.4.3m) between the two buildings and, rather than being rural, it is sited within the high street of a town.
Apart from the possibility of a narrow pedestrian alley at the southern end, the main 'wealden' house upon the site originally occupied the entire frontage of the tenement. It was constructed in the fifteenth century as a four-bay structure built parallel to and hard against the street. It incorporated a two-bay open hall which, almost certainly, had an overshot cross passage at its northern end. Beyond this were the usual services. The southern parlour bay appears to have been largely, if not entirely rebuilt, though there is conclusive evidence of its former existence. There are slight remains of an approximately contemporary two-bay rear range built, very unusually, against the rear wall of the hall, so the hall could only have been lit from the street facade at this period.
In the early/mid sixteenth century the rear range was extended back to link the house to the formerly detached building. Its construction blocked a ground-floor window in the front wall of the detached building. The link was jettied along both side walls and on the ground floor measured only 3m (9ft 7in) wide. It did not provide covered access to the rear range since it had two single-bay rooms on each storey. An enclosed courtyard was formed by the house, the linking range, the once detached building and the northern tenement boundary. Access from the street was only via the cross passage within the main house.
As initially constructed, the detached building measured 9.25m x 6.15m (30ft 4in x 20ft 1in) and on the ground floor comprised a principal two-bay room with either one or two smaller rooms within the southern bay (Footnote 14). The central bay was open to the roof, but the southern bay - and almost certainly the northern bay too - had a first floor. The first-floor area within the southern bay was subdivided by a longitudinal partition to form two lofts or chambers. This general pattern is one which has now been found in a number of 'kitchens', both of detached and attached type (Footnote 15), though the Battle example is a little unusual in that it did not incorporate a gallery across the open bay. The original method of access to the chambers is unknown. At 1.45m (4ft 9in) measured from original first-floor level to top of the wall, the heights within the chambers was acceptable, but not excessive. All chambers were open to the crown-post roof. The roof partitions extend only up to collar level so that the entire roof is sooted from the hearth located within the open bay.
Regardless of the fact that Smith considers this building ‘…not to differ from an orthodox house in any significant way, if at all’, it satisfies four, and probably five of our seven criteria for identifying detached kitchens. The window pattern does not comply with the standard format found in open halls; the building's position behind a house within a built-up street is unorthodox; the building is smaller than the main house, its floor area being less than 79 per cent (and probably less than 73 per cent) that of the house; there is no sign of internal decoration. Almost certainly the building had a two-bay room with one open bay, whilst the (assumed) floored bay had nothing beyond and cannot have been an overshot cross passage against services. Although the basic plan form is clear, lack of evidence for doorways means that the entry position and internal access points are unknown. In contemporary Sussex houses it would be unusual not to be able to recognise evidence of the hierarchical design which differentiated the high from the low end of the building. There are no such indicators in this example. Finally, there are a number of detailed manorial surveys of Battle, including surveys of 1433 and 1569 which suggest this was not a separate house (Footnote 16). Neither survey lists kitchens; instead, house and kitchen are covered under the blanket term 'messuage'. Despite this, the 1569 survey is good at identifying manorial holdings with two messuages upon them, including examples of two messuages under a single roof. Nos 12-13 High Street, Battle, is identifiable in 1569 as the freehold property of George Porter, held at a quit rent of 16d. It is described simply as a messuage and garden.
NGR TQ 6025 1654 (Footnote 17)
(Fig. 3 - see also VA 28, 86, fig. 1) 
Figure 3: Beestons, Warbleton. Site plan as in 1783 showing the location of the historic buildings (based upon 1st edition 25” O.S. map)
The 'kitchen' at Beestons is the second which Smith considers does not 'differ from an orthodox house in any significant way, if at all'. In addition to the 'kitchen', a house and barn still survive upon this holding. The present seventeenth-century house, which faced east, away from the 'kitchen', is built at right-angles to the road and incorporates a mid sixteenth-century two-bay parlour cross wing.
The three-bay fifteenth- or early sixteenth-century building identified as a 'kitchen' stands at a distance of approximately 13m (42ft 8in) to the rear (west) of the house and, like the house itself, is built at right angles to the road. Of the known multi-room 'kitchens', this example stands furthest from its house. The distance tends to discount its use as an everyday cooking kitchen. At a guess, it is likely to have been used for the more dirty, smelly service functions such as slaughtering, brewing, washing, malting and the like, which - of course - may have been the primary function of the other examples too (see below). The building measures 11.40m x 6.20m (37ft 5in x 20ft 5in). Even allowing that the house may have had only a modest medieval main range, the 'kitchen' is still considerably smaller than the house itself.
Generally the layout is very similar to the example at Battle, but with a very short floored bay to the main room and a first-floor gallery crossing the open bay, linking the upper chambers at either end. These chambers were open to the roof and had walls which measured 1.90m (6ft 3in) from first floor to top of wall plate. In this instance the building's location is perhaps less diagnostic, but otherwise all the points made regarding the unusual features of the Battle example apply here. Here too the locations of all ground-floor doorways remain unknown. Despite Smith's comment to the contrary, no researcher working in Sussex would accept the building's layout, with an overshot isolated chamber and a gallery, as that of an orthodox medieval Sussex house.
The documentary material relating to Beestons is relatively good. When Robert Pettit, husbandman, enfranchised the property from copyhold to freehold in 1544 it was described as being a combined holding comprising four adjoining manorial tenements of 20 acres each (Footnote 18). The abutments given in the enfranchisement can be reconstructed and calculate as encompassing approximately 109 acres, indicating that the 20 acre sizes given were somewhat nominal - a not uncommon occurrence. Smith makes very little of the reference in a deed of the property, dated 1635, describing the holding as having upon it a messuage, barn and kitchen (Footnote 19). There is no reference to a second house, so the third surviving building on the site is surely identifiable as the kitchen. Deeds tend to slavishly repeat earlier descriptions, even after some of the circumstances have changed, and this probably explains why a will of just two years later refers to two houses but no kitchen on the holding, suggesting that by this date the kitchen had been converted into a house in its own right - in effect it had become an example of the unit system.
By the nineteenth century the building had been converted into an oasthouse. Again Smith quite rightly asks 'what is the evidence for its…[use]…as a house and for the date of its conversion to an oasthouse?'. In large part it is the documents which suggest this sequence of conversion from kitchen to house to oasthouse, and the architectural evidence supports the suggested sequence. Documents show that detached kitchens were going out of use during the second half of the sixteenth century, a decline which continued during the next century, and if these redundant buildings were to survive they had to be found an alternative use (Footnote 20). Becoming an independent house (whether occupied by a tenant, a farm worker or a family member) was one possibility. It is probably significant that around 1600 the building was modified and upgraded. A floor was inserted into the open bay and the northern bay of the eastern wall was reframed using small-panel framing. Significantly, the central girder of the new floor is neatly stop chamfered, whilst at its northern end it is supported by a purpose-cut octagonal post. As the roof has been rebuilt and the first floor of the northern bay replaced (the medieval joists now being evidenced by mortices only) any chimney which may have been added within this area has left no evidence. Around 1800 the building was stripped to its frame, extended and modified to take on its present form. From this date onwards the building was certainly an oasthouse and could already have been used for this function when it was modified.
NGR TQ 6465 1161 (Footnote 21)
(Figs. 4, 5 - see also VA 28, 88, fig. 2)
Figure 4: Comphurst, Wartling. Site plan as in 1873 showing the location of the historic buildings (based upon 1st edition 25” O.S. map).
Figure 5: Comphurst, Wartling. Ground-floor plans showing
‘kitchen’ buildings in relation to main house, c.1500 and c.1600. Doorway
locations are shown only where known.
Of all the examples we put forward as 'detached kitchens' this was always going to be the most provocative on account of its very showy exterior, with close studding, first-floor jetties and moulded fascia beams. When first built it stood just 2.50m (8ft 2in) from the rear of the medieval house, and because of the jetty this was reduced to just 2.10m (6ft 10in) on the first floor.
In its present form Comphurst House stands parallel to the south-eastern side of an un- made-up track. When the house was built, this track was the highway which led south- westwards from the main road to Lord Dacre's great brick castle and Herstmonceux church. The road skirted the western boundary of Lord Dacre's park, within which the castle stood; indeed, the park extended northwards to within a field of Comphurst House. Until relatively recently two large barns stood close to the south-western end of the house - their ages are not known (Fig. 4). In 1370 the atte Compe family held the property and it is assumed that the Compers, owners in the early-sixteenth century, were relations. John Comper was assessed in the lay subsidy return of 1524 at £13 - the eighteenth highest sum out of 200 entries in Foxearle Hundred. Yet, surprisingly, the holding amounted to only 40 or 48 acres (the total varies in the documents) of which twenty had only been acquired in c.1500. However, this low acreage may be misleading for in 1532 John, described as farmer of Lord Dacre, paid a fine releasing Lord Dacre from suit of court to a neighbouring manor. He was therefore leasing an unknown quantity of land from Lord Dacre. He died in 1539 and his son sold the property in 1544 (Footnote 22).
Comphurst House incorporates the remains of a late fifteenth century 'wealden' hall house. The hall had a single open bay with moulded cross beams in both end walls. To the south-west there was probably an overshot cross passage with services beyond, but this end was rebuilt in 1600. Despite the fact that the open part of the hall was limited to a single bay, the medieval house was of good size. On the ground floor it was 6.20m (20ft 4in) wide. The side walls of the surviving upper chamber were generous, even by Sussex standards, measuring 2.20m (7ft 3in) from first floor to top of wall plate. Because the service end has been rebuilt, the overall length of the house cannot be known for certain. However, the average ground-floor length for a Sussex service end incorporating an overshot cross passage was 5.05m (16ft 7in). Using this data as a guide, the approximate ground-floor length of the medieval house at Comphurst would have been 13.10m (43ft).
The three-bay 'kitchen' stands directly to the rear of the open hall bay of the house and dates from c.1500. It is therefore roughly contemporary with the house. On the ground floor it measures 7.55m x 5.05m (24ft 9in x 16ft 7in) (Footnote 23) and is approximately 51per cent the size of the house, substantially less than the 78-80 per cent estimated for it by Smith. The building's form is shown in VA 28, 88, fig. 2, and in summary it had an unequal two-bay main room, the shorter being an open smoke bay. The bay closest to the house contained two small rooms separated from one another by a 1.30m (4ft 3in) passage which gave access from the exterior to the main room. The layout on the first floor is uncertain: the two adjacent floored bays may each have contained a one-bay chamber, or there may have been a single chamber of two bays. Whichever was the case, the upper area was open to the crown-post roof and had walls which measured 1.90m (6ft 3in) high from floor to top of wall plate - only slightly less than within the main house. The location of the stairs is not currently known and external access cannot be ruled out. Externally both the south and west walls were jettied at first-floor level, though the end jetty was narrower than the other, the dragon beam which carries the joists being set at a 30 degrees° angle rather than the usual 45 degrees. Both jetties incorporated moulded fascia beams whilst the south elevation - and probably the other walls too - were close studded for extra external visual effect. However, it should be emphasised that, despite the high-quality external features, all visible internal work is plain.
The detached building survives because it was incorporated into the house as part of extensive modifications carried out by James Colbrand in 1600 (Footnote 24). The southern end of the medieval house was reconstructed as a lofty brick-built parlour cross wing, the gap between house and 'kitchen' was infilled and the 'kitchen' building was widened by constructing a leanto outshut (Fig. 5). Given Smith's comment that later development often preserves earlier room functions despite radical alterations, it is interesting to note that the formerly detached building continued to serve as a kitchen, albeit enlarged in size. It is only fair to emphasise, however, that as part of these alterations the service and parlour ends of the house were reversed, though the main entrance was preserved in its original position. From the style of the alterations it can be surmised that Colbrand intended to rebuild the remaining part of the medieval house on the same scale as the cross wing, thereby transforming it into a small gentry mansion. However, from the nature of the alterations he made to the rear part of the house it would seem that the kitchen was to remain. John Colbrand, James's successor, was styled 'esquire' when he sold the property in 1616 (Footnote 25).
It must be admitted that, with its external display features, the building at Comphurst does, at first sight, seem too fine to have served as a menial kitchen. But, looking at the evidence objectively, if it was a second independent house, why was it not built where it could be seen easily rather than crowded in at the back of the main house, half way along its length? More important still, why was its primary, if not sole, means of access via a passage running through the services and reached by a doorway in its end wall? Although the design of part of the ground-floor south wall is not currently known, the available evidence does not suggest a doorway in this area - there are no closing rebates, latch points, pintle holes or external chamfers running down the principal posts of the wall. However, negative evidence of this type is not proof and thus a doorway in this area should not be entirely ruled out. Even so, it is certain that there was no doorway against the service partition in the traditional position for a medieval or transitional house in Sussex. This alone makes the building un-house-like. More important, a wide passage through the services, using valuable space, would be wasteful if the passage entrance was not the primary means of access to the building. Therefore, access to the 'kitchen' was almost certainly from the rear door of the house. Smith considers that the building's location and means of access are acceptable features of an independent house. This we cannot accept. The readers must decide for themselves.
So much for the location and access, what of the building's showy exterior? Would an owner lavish so much money upon a menial kitchen, and, if so, why? Indeed, regardless of the building's use, why expend so much money on a structure hidden away behind the house? The building was visible across a field from a road to the south, but although this might explain the close studding and perhaps the jetties, it would not explain the moulded fascia beams. The answer must surely be that, for what ever reason, the ground to the south-west of the house was accessible to visiting members of the public or important customers. Such an hypothesis might seem to favour the building as an independent house, but it should be remembered that all the show features are external - there is no decoration internally (Footnote 26). But would an owner choose to display wealth on the exterior of a utility building? There are precedents for this even away from important market centres. For example, Assleton House, in the small East Sussex village of Sedlescombe, has to one side of it a building which can only be regarded as a warehouse ™ three plain open storage bays and a lofted-over end bay reached by a stair trap in the centre of its floor. It was a menial building, yet it was jettied towards the house (Footnote 27). Likewise, although of a higher social standing, in the late-fifteenth century either Sir Thomas Etchingham or his son- in-law built a lodging range near the entrance to his manor house. It was close studded, jettied along both sides (complete with moulded fascia beams) and incorporated a run of arch-headed doorways. Yet these were not lodgings for important guests, but for menials. Internally the building was entirely plain and the freestanding crown posts were extremely crude (Footnote 28). In these instances it is the exterior of the building which portrays wealth, regardless of internal use. The likely builder of Comphurst was not poor - he was included amongst the top 10 per cent of the taxable population in the 1524 Lay Subsidy return - so a similar situation is conceivable.
NGR TQ 7096 1939 (Footnote 29) (Fig. 6 - see also VA 28, 89, fig.3)
Figure 6: Darwell Beech, Mountfield. Ground-floor plans showing development of the building. Doorway locations are shown only where known
When Darwell Beech was first surveyed in 1979 no timber framing was visible externally, whilst internally all but the main frame and a few other details were masked by later coverings. However, it was clear that, although a single dwelling since the early eighteenth century, it had started life as two separate mid sixteenth-century structures, both served by smoke bays. They were considered to be two fully independent houses sharing one yard - an example of the unit system. The southern 'house' had lost its end bay and, although lower than its neighbour and less well finished, it appeared to have been the larger of the two.
In 1986 the house was dismantled and re-erected allowing additional recording to be carried out. This showed that the northern house had likewise lost a bay. It also revealed that the internal layout of the southern structure was unusual in that the main room incorporated a wide passage next to the 'service room', leading northwards towards an external doorway close to the adjacent house. Furthermore, the small northern 'service' room was accessed from the passage, close to the external door, rather than from the main heated room as would be expected. The arrangement suggested that access to this room was most commonly from the external door rather than from the main room. The gap between the external doorway and the northern house was a mere 2.25m (7ft 4in) (Fig. 6, phase 1). A further non-standard feature was the fact that the entire first-floor area to the north of the smokebay formed a single room, open to the roof and crossed by a truss of utilitarian design. It was poorly finished with small, plain straight braces. The design closely resembles that found in local farm buildings. In contrast, the features uncovered within the northern house emphasised that building's relative high status :- most of its external walls were close studded, as too was the end wall of the floored-over hall, and there was extensive use of comb-decorated daub. These features are entirely absent from the south building.
In our opinion the newly-discovered features are sufficient to indicate that, rather than representing two independent houses as previously thought, the buildings functioned as two parts of a single household. Further, the southern building serviced the superior quality northern building - i.e. kitchen servicing house. Having said this, it must be admitted that the 'kitchen' building is fitted with its own independent front door and, as already pointed out by Smith, it is wider than the house. Although this latter point worries Smith, it does not overly concern us. We can see no reason why kitchens should always be narrower than their associated house, particularly where the extra height and more showy external treatment of the main house makes it quite clear which of the two is the superior structure. It is not difficult to put forward a hypothetical scenario which explains this difference in size: it might have been built for a small family, perhaps childless, with paid live-in staff to assist on the farm and carry out the more onerous household chores. In circumstances where the owner did not wish his staff to live under the same roof as his family and wished to separate the more smelly and dirty service functions from his living quarters, what better solution than the arrangement recorded at Darwell Beech? Perhaps some researchers would see such circumstances as complying with the requirements of the unit system. But in the arrangement outlined above the two buildings do not accommodate two independent households. If the 'kitchen' element of the complex were to cease to operate, the 'house' element could not function at the same social level without modification and extension.
As Smith suspected, the subsequent developments at Darwell Beech are particularly instructive and help clarify the earlier phases. The principal early alteration made to both structures was the insertion of brick chimneys into the smoke bays. Some of the first-floor ceilings may also have been added. Significantly, the chimney inserted into the 'kitchen' had a wide cooking hearth, whilst that introduced into the 'house' was much narrower and had splayed jambs, intended for heating and not for working (Fig. 6, phase 2). Thus, at this period the original relationship between house and kitchen was maintained. However, relatively soon afterwards further alterations were made. Both buildings were reduced by one bay creating two cottage-like structures, each with an end chimney. More significant still, the heating hearth within the 'house' had its splayed jambs rebuilt narrower in order to widen the fireplace and convert it into a cooking hearth (Fig.6, phase 3) (Footnote 30). Dating these alterations is difficult, but to judge by the heavy weathering on the end of the house from where the bay had been demolished, the truss was external for some considerable time before once again becoming an internal wall early in the eighteenth century. The authors have no doubt that this evidence indicates that either at the close of the sixteenth, or during the first half of the seventeenth century, the two buildings ceased to function as 'house' and 'kitchen', and instead became two independent two-cell houses. At this period the farm was owned by absentee landlords and it is not known whether the occupiers of the houses were related or merely farm workers employed either by the absentee owner or by a tenant who lived elsewhere. Whichever was the case, early in the eighteenth century circumstances changed again and the two houses were integrated to become a single dwelling (Fig.6, phase 4).
No details are known regarding the documentary history of the property prior to 1618, though the farm appears always to have been large. In 1683 it was described as a messuage, garden, and 200 acres of land and wood (Footnote 31). The description of the buildings upon the holding is clearly deficient since it implies one messuage only at that date and does not mention the extant barn.
It is hoped that the details summarised above are sufficient to place the five most controversial buildings mentioned in our VA 28 article into their wider context. Similar accounts could be given for other buildings, but they are unlikely to add anything substantive. However, some of the points addressed by Smith remain unanswered. In addition, by grouping together all the 'kitchens' of more than one room in the VA 28 article, a somewhat false impression has been given as to how rare the larger three- and four-bay examples are likely to have been. A further point which needs to be explored is the existence of attached kitchen ranges of very similar layout to the detached examples presented. It is this latter point which will be addressed first.
When built, Combe Manor, Wadhurst, had a small single-room attached kitchen (VA 28, 90, fig.4). For whatever reason, the occupants soon found their small kitchen unacceptable, for within a generation a further three bays were added to give an attached multi-room kitchen similar to the detached versions discussed here (Fig. 7). In its enlarged form the 'kitchen' part of the building had a ground-floor area 87 per cent the size of the house part - a similarly high percentage to that noted in many of the larger detached examples (Footnote 32).
Figure 7: Longitudinal sections through Combe Manor, Wadhurst, and Beaumans, Ticehurst, showing both buildings as extended in the sixteenth century
At Beaumans, Ticehurst, there is a 'standard' late-fifteenth-century three-bay detached and galleried 'kitchen' of the type present at Beestons, Warbleton. In c.1525 it was attached to the house by rebuilding the latter against the end wall of the kitchen (Fig. 7) (Footnote 33). Another attached example is Wardsbrook, Ticehurst (c.1565). Wardsbrook was one of the first multi-roomed and galleried kitchens published by the authors, though at that time it was considered to have been a detached structure. However, it has recently been shown that when built it was attached to a now- demolished part of the main house (Fig. 8) (Footnote 34).
Figure 8: Ground-floor plans of two houses dating from the second
half of the sixteenth century showing attached kitchen ranges
A number of other attached examples are now known, mostly dating from the mid/late sixteenth century. Unlike Combe, Beaumans and Wardsbrook, most take the form of a range extending back from the service end of the house, as illustrated by Bugsell, Salehurst (c.1560) (Fig. 8). What these examples appear to illustrate is the increasing tendency during the sixteenth century to build attached rather than detached 'kitchen' ranges.
With one possible exception – Comphurst – all the large rural multi-room detached 'kitchens' studied by us are located upon what were, by local East Sussex standards, exceptionally large holdings. Even the owner of Comphurst is known to have leased an undefined acreage of demesne and thus, despite owning less than fifty acres, he too may have run a substantial farming enterprise. None of the other holdings was less than 100 acres in extent, and four exceeded 200 acres. Significantly, the same is true of those houses which incorporate attached multi-room kitchen ranges. This is surely no coincidence and perhaps begins to explain the existence of these exceptionally large detached 'kitchens'.
Smith assumes that the holdings with detached kitchen blocks were probably occupied by a nuclear or at most three-generation family. This was almost certainly the norm for local households, but given the exceptionally large acreages of the farms which incorporated these detached and attached multi-room kitchens, is that assumption valid in these cases? Unfortunately there is very little data to help answer this question. Live-in farm servants appear to have been rare in the weald of East Sussex, but by the early eighteenth century they were usual on the large farms (Footnote 35). Whether the same household structure existed in the sixteenth century is unclear. However, given the large size of the holdings upon which these multi-room 'kitchens' are located, it is certainly possible that these buildings were being used in part to accommodate resident farm workers. Furthermore, sufficient is known regarding the farming practices of the region to be certain that, in general terms, the smaller the holding the more it tended towards subsistence farming, whilst the larger holdings were usually run as profit- making businesses. These large holdings are precisely the type which would have required additional storage space and processing areas associated with commercial farming.
The tendency to classify everything we study (in order to make sense of the data) risks forming artificial barriers which our ancestors would not have recognised. They lived and worked from home. Inevitably, some farming operations would have spilled over into domestic areas and vice versa. Further, the extent to which this occurred would have varied from household to household, and from generation to generation. As an example of this, granaries are rare in the weald of East Sussex. Documentary sources indicate that the threshed crops were usually stored in the house, either in an attic or an upper chamber. Occasionally they were stored on an upper floor at the end of the barn, or, if the holding had one, in an oasthouse (Footnote 36). In a large holding, what better place for the grain to be stored than in one of the upper chambers of the kitchen! The processing of grain - threshing, winnowing and sieving - was carried out in the barn, so usually only the storage of the end product may have involved the use of the kitchen building. But the Great Worge description mentions malting – an operation which could be classed as either domestic or agricultural/industrial, depending upon the scale of operation. Brewing – a common household task – is another. Some of the fruit and vegetables would have required preserving and, if grown on a commercial scale, both the processing and subsequent storage may have required considerable space.
The above addresses crop processing and storage only. But the weald was primarily a cattle fatstock region. Could the same logic apply to meat? The autumn cull is likely to have produced a not inconsiderable number of carcasses, some of which might have been smoked, others salted. The Great Worge description makes it clear that one of the uses of the principal room within the kitchen was dressing carcasses. This and the associated preservation process required space. If the carcasses were smoked, where did this take place? Could smoking explain the continued inclusion of an area open to the rafters within many kitchen buildings, long after the demise of the open hall? Could it in part explain the galleries which crossed most of these rooms? Regardless of whether cured or salted, the processed carcasses would require storage, and this is likely to have been considerable on large farms of the type being discussed here, especially if the meat was produced for the market as well as household requirements (Footnote 37).
On these large holdings, the choices as to where the owner processed and stored the products of his farm (whether for domestic consumption or the market) and where he lodged his agricultural servants probably answers one of Smith's main concerns. He asks why one house should have needed only a small kitchen whilst another required one more than three times as large. An equally important question is what form of agricultural regime was being followed on the holding. Was it principally fatstock, or did cereal crops or dairying play an important role? Was production primarily for domestic or local consumption or was it for market? The answers to these questions may explain the need for different levels of staffing and different sizes of processing and storage areas. A milkhouse required for the everyday needs of the household would, for example, be considerably smaller than one where dairying was carried out on a commercial scale. The one may have been located within the house, the other within part of a specialised building.
Whether these detached multi-purpose buildings should be called kitchens or
detached service blocks is a point which has been raised by Bob Meeson (Footnote 38). It
is far from clear how often (if ever) they were used for everyday cooking of the
type we today associate with a kitchen. As noted, they may have been used
primarily for such service functions as processing, storage, brewing, washing,
baking and malting. It is unlikely that the example at Beestons, located 13m
(42ft 8in) from the house was used to prepare and cook the family's everyday
meals. There can be no doubt that fifteenth- and sixteenth-century documents
call at least some of these multi- room, two-storey detached buildings kitchens,
but is it misleading for us to use this term today bearing in mind our
twenty-first-century concept of the word? There is certainly a precedent for
Meeson's suggested change of name, for although medieval documents often refer
to the main building as the aula (hall) or, more commonly in later documents,
the messuage, researchers use the modern term 'house' to prevent confusion with
the principal room of the house - the hall. This is despite the fact that in
early documents the word 'house' had a wider meaning than it does today and
rarely meant a self-contained dwelling. However, there is a worry in abandoning
the historical name for these buildings in favour of a more understandable
modern term. By so doing we risk divorcing a class of surviving building from
its documentary record. It is only by breaking the preconceived notion that
kitchens as mentioned in documents have to be single-cell menial structures used
solely for cooking (Footnote 39), and
by excepting that under certain circumstances kitchens can be multi-room,
two-storey detached buildings of some size, that we will be able to place them
within their true historical context. Kitchens, like houses, can take a number
of forms and have a considerable range in size.