The Medieval Site at Well Wood Aylesford

27 THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD BRIAN PHILP This site lies in the extreme south-west corner of the parish of Aylesford on what was Hermitage Farm (TQ7217 5645). It lies just south of the main railway in the former Well Wood, now swallowed up by ongoing ragstone quarrying and landfill operations (Fig. 1). The site appears on early editions of the Ordnance Survey map as a rectangular earthwork with two small keepers’ cottages across its south-east corner. The underlying soil forms part of the Hythe Beds of the Lower Greensand and the elevation is about 61m OD. From Hermitage Lane to the east, the site appears to be slightly elevated. The name Well Wood reflects the presence of a substantial stone-lined well, shown on early maps and found during the excavation. In 1988 planning consent was given for the creation of a large quarry across the west side of Hermitage Farm. This included about two-thirds of the rectangular earthwork, but excluded the site of the ruined Keepers’ Cottages. A planning condition requiring an archaeological investigation had been agreed with the Kent County Council. The Kent Archaeological Fig. 1 Site Location maps. BRIAN PHILP 28 Rescue Unit was then appointed by Mr Pat Gallagher, of Gallagher Ltd, to carry out a prior excavation. This was done as an urgent exercise in April and May, 1989. The excavation continued non-stop for 36 days and involved 20 members of the Unit team. In that time a total of 36 trenches was dug across the site covering an area of about 3,800m2 or about 31 per cent of the total area of the visible enclosure. This work substantially completed the site other than the remaining quarter in the vicinity of the ruined cottages which was not part of the quarry programme; this area should remain intact for future examination and has since been buried by dumped soil. THE EXCAVATION (FIG. 2) The excavation was largely confined to the area of the still visible earthwork (F23) which, in plan, was largely oblong with rounded corners. It lay on gently sloping ground on its main north-south axis and it was delimited by a ditch some 16-18m wide on its west and east sides and also on part of its north side. No trace of this large feature survived along the southern half of the site where it must have been destroyed by earlier quarrying. The internal east-west width of the area enclosed was about 85m and the minimum overall length about 126m. This wide, shallow ditch appears to be a late feature in the history of the site, only partly reflecting the medieval enclosure ditch (F1) which it clearly superseded. The area had been cleared of trees, bushes and soil in the months prior to the excavation and a large test-trench had been dug across its centre and later filled. In view of the limited time and the unknown extent of any surviving features, a pattern of long evaluation trenches was dug at close intervals across most of the enclosure. These revealed that most of the northern area was largely devoid of features, but some 26 individual features were found, mostly in the south-west quarter. Of these, a shallow stone-walled cellar, part of a later masonry building, a cluster of adjacent pits and gullies and the well were the more significant. These are described below and listed in Table 1. The Medieval Enclosure Ditch (F1) (Figs 2 and 4) The linear trenching discovered a continuous ditch enclosing an area slightly smaller than that of the large enclosure ditch, largely visible at the time of the excavation. This ditch was mostly only 2-3m in width and 1-2m in depth. It outlined a sub-rectangular area, with rounded corners, about 64m in internal width and with a minimum internal northsouth length of 126m. A slight curve at its south-west limit suggests that it was beginning to curve inwards to create the southern boundary. If so, then an original north-south internal length of about 132m would be THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 29 PLATE I The Well at Well Wood indicated. The ditch produced about 60 potsherds of twelfth- to fifteenthcentury date (Nos 1-7) and about 90 fragments of tile, some 40 of Roman date. Five sherds of Victorian pottery and fragments of house-brick here were clearly intrusive. The Well (F11 and F17) (Fig. 2) Plate I This structure had been filled and buried during the site-clearance operations ahead of the proposed quarrying and it was relocated by the trial-trenching. It lay 58m west of the Keepers’ Cottages and inside the medieval enclosure ditch (F1). The well had been constructed inside a circular shaft 2.20m in diameter and cut vertically into the natural soils and rock to a depth of only 1.50m. Below that it tapered to about 2m, but for an unknown depth. The well proper was just 0.70m wide, but with its stone lining had an overall diameter of about 1.70m. The upper 2.00m of the lining consisted of large, roughly shaped ragstone blocks set in a cream-ginger mortar containing Greensand grits. Some of the blocks showed traces of a ginger-coloured mortar identical to that from one of the medieval walls nearby. It is likely that these had been re-used from demolished walls. From 2.00-3.00m (the maximum depth examined), the well lining consisted of four courses of yellow stock bricks, some 0.34m deep in all. BRIAN PHILP 30 Fig. 2 Overall Site Plan showing major features. Beneath were more courses of ragstone some 0.40m deep and these were sitting on at least three more courses of yellow stock bricks, the latter laid with alternate voids to allow water percolation. The upper 1.50m of the filling of the well consisted of spoil pushed in during the clearance operations (L.89-41). Beneath was a filling of ragstone rubble, clearly thrown in from the top. How deep the well was and what it might have contained at its base is not known. THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 31 As regards dating, the yellow stock bricks suggest that the well (as seen) was constructed in the nineteenth century to serve the cottages. It is strange that it was sunk inconveniently 58m from the cottages though it is possible that the Victorians re-lined a medieval well. It is anyway likely that one existed on this site in medieval times, but this did not exist in the areas excavated. BRIAN PHILP 32 TABLE 1. DETAILS OF FEATURES 1-26 (DIMENSIONS IN METRES) No. Type L W D Fill Pot Tile Brick Bone Shell Iron Mortar Total 1 Ditch linear 300 3.00 1.60 Sandy clays 58 85 3 18 28 192 2 Pit Circ 1.20 0.35 Sandy clays 0 3 Pit Circ? 1.10 0.70 0.60 Sandy clayloam 1 1 4 Pit Circ? 2.15 1.00 0.76 Various 0 5 Pit Circ 1.11 1.00 0.40 Sandy clayloam 1 3 4 6 Pit Circ 2.00 1.00 0.48 Very sandy clay 26 1 6 1 34 7 Gully linear 32.00 1.00 0.70 Various 75 7 3 31 1 117 8 Gully linear 2.20 1.40 Sandy clay 1 1 9 Pit oval 3.60 2.10 1.00 Stiff clay 2 2 10 Pit circ 2.60 0.40 Sandy clay 15 15 11 Well circ 0.70 1.60+ Dark brownblack clay loam 0 2 Pit oval 2.20 0.95 1.40 Various 50 2 52 13 Pit oval 2.50 1.70 2.00 Various 0 14 Pit rect 1.15 1.10 0.38 Sandy clay 18 2 20 15 Pit rect 0.74 0.70 0.16 Sandy clay 28 1 3 32 16 Gully linear 4.00 1.30 0.33 Sandy clay 0 17 Well cut circ 2.20 3.00+ Clay + ragstone 1 2 5 1 9 18 Pit oval 3.40 3.10 1.05 Loose clayloam 41 2 11 4 58 THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 33 No. Type L W D Fill Pot Tile Brick Bone Shell Iron Mortar Total 19 Ditch linear 5.00 1.60 0.75 Various 69 5 1 22 51 148 20 Ditch linear 8.60 0.80 0.20 Stiff clay 0 21 Pit oval 0.84 0.82 0.15 Sandy clayloam 3 3 22 Pit irreg 2.20 1.70 0.16 Stiff sandy-clay 10 10 23 Earth -work See text 24 Cellar See text 25 Wells A+B See text 26 T wall See text Total 383 121 6 67 116 2 3 698 TABLE 1 (cont). DETAILS OF FEATURES 1-26 (DIMENSIONS IN METRES) BRIAN PHILP 34 The Cellar (F24) (Fig. 3) Plate II A neat, mortared stone-lined room was found midway between the east and west medieval ditches (F1), but actually in the southern half of the enclosed area. This was set some 0.60-0.80m into the natural bedrock and thus must have formed a sub-basement, or shallow cellar. Allowing for the removal of soil during the site clearance it seems likely that it was originally about 1.00-1.20m in depth. About half its total area had been destroyed by a deep and wide machine-cut trench, running diagonally across it, dug earlier by the contractors testing the depth of the ragstone. In detail the cellar had been constructed in a large rectangular pit, about 6.70 x 4.50m, with its long axis roughly north-south. Both the east wall and the west wall had survived substantially intact, the former with an original opening later blocked. Only short sections of the south and north walls survived, some 0.95m and 1.60m respectively. The internal dimensions were 5.80m and 3.70m with the walls surviving to heights of 0.36-0.60m. The wall thickness varied from 0.24-0.50m and two forms of construction could be identified. The walls were all constructed of roughly hewn ragstone blocks, some large, set in a yellow sandy mortar with green grits. The walls were broadly similar, except at the north-east corner where both joining walls had a reduced thickness of only 0.30m. This variation in the structure may simply indicate a minor change of The Cellar at Well Wood PLATE II THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 35 technique, or equally some localised rebuild. Where most complete, the walls survived to a maximum of eight courses (Fig. 5). The gap in the east wall, at a point about 1.00m from the integral southeast corner, was about 1.70m in width. This ran the full depth of the wall and almost certainly represents an original access point, perhaps by means of steps, later removed. The opening had itself been completely filled Fig. 3 Plan of Cellar F24 and Later Building F25. BRIAN PHILP 36 by layers of unmortared ragstone blocks, capped by a very large slab 1.30 x 0.62m and 0.27m thick. Slight traces of an internal rendering were found, suggesting that the Cellar was originally plastered, but no trace of fallen rendering was found. Fig. 4 Sections of Ditch F1 and Wall F25B with adjacent pits. THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 37 In addition, the complete west wall had a mortared internal chamfer at its south end, perhaps indicating a window or light-well. About a third of the way down its length was a rectangular slot, about 0.28 x 0.16m, which may have held an upright post serving some form of structural function. The Cellar had been deliberately filled with a succession of sandy loams containing some light rubble. These sealed a primary deposit of loam containing charcoal specks and domestic pottery, which probably represents the final use of the Cellar. This included 25 sherds of medieval pottery dating from the twelfth/thirteenth centuries. As regards dating evidence, the west wall cut a large pit (F15) which contained a small amount of pottery and clearly the cellar must post-date that. It was itself cut by a pit (F14) also containing pottery. Together these two pits contained 46 potsherds, mostly shell-loaded wares but including sandy ware jugs. These are probably of twelfth/thirteenth century date. More significantly, both the cellar and its complete filling had been cut through by the massive foundations of a large wall (see below) relating to a later structure. On balance it seems likely that the Cellar was constructed in the thirteenth century and soon abandoned and superseded by a large medieval building. The Later Building (Figs 3, 4 and 5) Two isolated sections of masonry, roughly at right-angles to each other, were found in or close to the cellar. These are described below: Ginger-mortared wall (F25A) This very large wall, or rather foundation, was found cutting through all the layers filling the cellar. It had been built diagonally across the cellar and its ends butted to the north and south walls of the cellar. Its original length had been 6.35m, but over half of it had been destroyed by the contractors trial-trench. Enough survived to show that it had a width of 0.94-1.00m and still stood six courses high to a maximum height of 0.6m. It was built of shaped ragstone blocks set in a ginger-coloured mortar. Its function must have been to carry a wall, running roughly north-east/ south-west, across the soft filling of the Cellar. L-Shaped Wall (F25B) About 2-5m east of the Cellar were two short lengths of wall joining at right angles. The north-west arm was 2.50m and the south-east arm at least 2.00m. Both consisted of ragstone blocks set in a hard orange-yellow sandy mortar and were clearly of integral construction. The external corner sat over two pits (F13 and F22) containing only ten small potsherds and BRIAN PHILP 38 Fig. 5 Sections through Wall F25A and Cellar F24. THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 39 incorporated a large rectangular block, 0.48 x 0.41 x 0.11m., almost certainly a corner quoin. The walls were 0.70-0.75m in width and 0.11-0.26m in depth. Both were sealed by a layer of broken roof tile, probably from the roof of the actual building. The absence of fallen masonry debris under the tile, could indicate that the superstructure had been timberframed. This same tile layer contained domestic rubbish including some 85 potsherds, again of twelfth- to fourteenth-century date (Nos 19-20). It seems highly likely, on circumstantial grounds, that these two large sections of masonry formed part of the same structure and clearly postdate the filled-in Cellar (F24). Either way, the two walls of T-shaped masonry can be projected forwards with the north-west arm joining the ginger mortared wall at 7m from its own external corner. This helps create a possible room of unknown width, but 7m in length. The ginger wall ideally needs a corresponding wall, or walls, but all traces of these had been removed. A substantial building, of two or more rooms and Lshaped in plan, could be indicated. The superstructure could have been timber-framed in whole or in part. T-Shaped Wall (F26) Figs 2 and 6 A small, isolated section of strong mortared masonry was found at the extreme south edge of the site. In plan it formed a T-shape with a short stub joining a rather larger wall. Only 2.10m of the larger north-south wall survived and this was 0.75m wide and 0.30m deep. The stub was only 0.70m long, 0.70m wide and 0.30m deep. Both walls formed an integral part of a structure, well built of coursed ragstone set in an orange sandy mortar containing Greensand grits. It is likely that these walls represent part of a much larger structure, later almost totally removed, situated about 20m south of the main Cellar (F24). The wall stub is likely to have been an internal wall and, if so, this would indicate at least two rooms. The survival of the mortared masonry here is almost certainly due to the presence of a small underlying gully (F19). Here the walls had been constructed into the fill of the ditch to help stabilise them, otherwise it is likely that no trace of this structure would have survived. The gully contained 69 potsherds and five fragments of tile. The pottery is again of twelfth/thirteenth-century date (Nos. 15-17). Gully F7 (Fig. 6) A long straight gully, at least 32m in length, was found running roughly north-east/south-west across the south-west sector of the site. This lay only 4m outside the masonry cellar and its function may have been to serve as a close boundary relating to it. BRIAN PHILP 40 In detail it was 1.05-1.32m in width, 0.45-0.67m in depth and mostly U-shaped in profile. It contained a fill of dark brown sandy and clay loams which contained 75 potsherds, seven fragments of tile and 31 shells. All the pottery (Nos 9-11) seems to be of twelfth/thirteenth-century date. Its south-west end was found to be curved, but its north-east end must have been just beyond the edge of the excavation for it did not exist in adjacent trenches. Short sections of two more gullies (F16 and F20) were found Fig. 6 Sections of Wall F25B, T-shaped Wall F26 and Gully F7. THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 41 nearby which, if contemporary, could have related to Gully F7. This could suggest that together they represent small enclosures, but none continued and this remains inconclusive. Gullies and Ditches (Fig. 2) A total of five gullies, or small ditches, were found on the site. These include F7, F16 and F20 that could, if contemporary, form part of a small enclosure, since badly damaged. Another small gully (F8) was found just outside the west ditch, but did not continue northwards. Another ditch (F19) was found beneath the T-shaped walls at the extreme south end of the site where it survived for only 5m before being destroyed by earlier quarrying. Pits (Figs 2-6) A total of 14 pits was found during the excavation, the majority in the south-west sector. All are listed in Table 1. Two of these, F5 and F9, contained Victorian pottery and probably relate to the Keepers’ Cottages. Seven more pits occur in or adjacent to the masonry Cellar (F24). Of these three (F21, 13 and 22) are cut by masonry and clearly pre-date its construction. Of the others, two (F14 and F15) cut the Cellar wall and must post-date its demolition and two more (F6 and F12) are adjacent, the second of which cuts one of the early pits (F13). The remaining five pits (F2, 3, 4, 10 and 18) are well-scattered and are not of special significance. The largest (F18) was at the south-west corner, close to the well and this contained 41 sherds of pottery, mostly of twelfth/thirteenth-century date (Nos 13-14). The other pits contained more pottery also of the same period (Nos 8, 12 and 18). THE FINDS A total of 874 objects was recovered from the site and these are enumerated below: TABLE 2. ARTEFACTS RECOVERED FROM THE SITE Pottery 526 Tile 140 Brick 6 Bone 78 Shell 117 Iron 4 Mortar 3 Total 874 BRIAN PHILP 42 The Pottery (Nos. 1-20). The pottery is very largely medieval and of twelfth/fourteenth-century date. There is nothing earlier apart from a single small sherd of Roman samian ware. Similarly, nothing dating from about 1450-1800 was found and the remaining objects are nineteenth-century. The medieval material include a majority of shell-loaded cooking pots, some with vertical strap decoration. Fragments of about twenty sandy ware jugs are also present and these include two with green glaze and others with minor decoration. Tile The 140 fragments of tile are also most interesting for they include 44 fragments, some large, of Roman date! These include nine tegulae, five imbreces and three hypocaust tiles and at least two flat bricks. There were no Roman features or stratification anywhere on this site so clearly these tile fragments were brought here in medieval times. The remaining tiles are fragments of peg-tile of medieval date and of standard size. The other finds are of minimal importance, but have been retained together with the pottery and tile. DISCUSSION The primary feature on this site appears to have been a substantial ditched enclosure (F1), sub-rectangular in plan. This had an internal width of 64m and an internal length estimated to be at least 126m. The ditch was Ushaped in profile, 2-3m in width and 1-2m in depth. It was probably dug in the twelfth century. Centrally within the southern half of the enclosure was a pit (F21, and probably also F13 and F22), soon superseded by a rectangular stone-lined Cellar (F24), or sub-basement, some 5.80 x 3.70m internally. This may have supported a timber-framed building, later removed. The Cellar was soon abandoned and filled with soil and rubble and replaced by a large masonry building (F25), constructed diagonally across it. This included a massive new foundation (F25A) dug through the filling of the Cellar and to the east, related masonry, surviving only as an L-shaped (F25B) section. Together these suggest a substantial L-shaped masonry building of at least two rooms, but of unknown extent. Nearby, later pits (F6, F12, F14 and F15) were soon dug. Elsewhere within the main enclosure were gullies and ditches (F7, F8, F16, F19 and F20), perhaps parts of minor internal divisions. In addition there were more pits (F2, F3, F4, F10 and F18) and a small section of T-shaped masonry (F26) at the south end, which could represent part of another building. It is likely that when the original enclosing ditch had silted, the enclosure was enlarged on the THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 43 west side and delimited by a wide, shallow ditch (F23), still partially visible at the time of the archaeological excavation. The main dating evidence suggests that all the medieval activity, pits, gullies and buildings, was confined to the twelfth-fourteenth centuries. The surviving structural evidence, whilst not substantial, is just enough to suggest a middle-status holding, including at least one masonry building within a ditched enclosure. Detailed historical research could yet produce related documentary evidence, though the standard reference works mostly seem to be silent. The site was in an elevated position in a presumably wooded area, but on only average soils. Its position at the corner of Aylesford parish could also be significant. The Roman tile and single Roman potsherd must have been derived from a nottoo distant site. Apart from these the limited dating evidence suggests that the site was abandoned no later than about 1400 and not occupied again until the mid-nineteenth century when the Keepers’ Cottages were built. The large well (F11 and F17) was then dug (or even enlarged) and smaller pits (F5 and F9) dug. It is likely that any medieval masonry surviving above ground was then removed for the well-lining and the cottages. The site seems to have been abandoned again after the Second World War and substantially destroyed by ragstone quarrying in 1990. ACKNOWLEDGEMENTS The work on the site, directed by the writer, was assisted by Peter Keller, Derek Garrod and Maurice Chenery. Of the rest of the team, Mike Bennett, Barry Cork, Gerry Geradine, Alan Morley and Tom Tasker deserve a special mention for their extended work. The Planning Department of the KCC, in particular David Sayers and David King, greatly facilitated the operation and Tom Ladell provided the initial contact and drawings. Pat Gallagher helped finance the excavation and provided machinery for the work. Unfortunately, no provision had been made for the costs of any postexcavation analysis and it was not until March 2002 that Paul Oldham (a local resident) kindly offered to fund this work so that the site could be published. APPENDIX Catalogue of Illustrated Pottery Ditch F1 (Fig. 7) No. l (WWAF89:4) Jug with upright rim. Hard sandy fabric with buff-brown core and grey surface. Thirteenth-century. BRIAN PHILP 44 Fig. 7 Pottery from Ditch F1, Pit F6 and Gully F7 (1:4). THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 45 No. 2 (WWAF 89:9) Strap handle of jug in hard very sandy fabric with light grey core and mid grey surface. Five vertical slash marks on handle. Thirteenth-century. No. 3 (WWAF 89:17) Cooking pot with outcurved rim in hard shell-loaded fabric. Light brown core and orange-brown to grey surface. Twelfth/ thirteenth-century. No. 4 (WWAF 89:18) Bowl with plain rim in a hard smooth sandy fabric. Dark grey core and black burnished surface. Iron Age/Romano-British. No. 5 (WWAF 89:19) Jug with out-turned rim in a hard very sandy fabric. Light grey core and surface. Thirteenth-century. No. 6 (WWAF 89:20) Jug with upright pinched rim in hard sandy shell-loaded fabric. Dark grey core and grey-black surface. Thirteenth/fourteenthcentury. No. 7 (WWAF 89:20) Cooking pot with out-turned rim in hard shell-loaded fabric. Dark grey core and light brown to dark grey surface. Twelfth/ thirteenth-century. Pit F6 (Fig. 7) No. 8 (WWAF 89:25) Cooking pot with flattened rim. Hard shell-loaded fabric. Dark grey to buff-brown core and light grey to grey surface. Twelfth/ thirteenth-century. Gully F7. (Fig. 7) No. 9 (WWAF 89:26) Cooking pot with flattened rim in a hard shell-loaded fabric. Light grey core and orange-brown surface. Twelfth/thirteenthcentury. No. 10 (WWAF 89:36) Jar or bowl with thickened rim in very hard sandy fabric with slight shell-loading. Dark grey core and light grey surface. Twelfth/thirteenth-century. No. 11 (WWAF 89:39) Cooking pot with out-turned rim. Hard soapy fabric with much shell-loading. Dark grey core and orange-brown to dark brown surface. Twelfth/thirteenth-century. Pit F12 (Fig. 8) No. 12 (WWAF 89:38) Probably a curfew, or fire cover, fragment of the body and handle in a hard sandy shell loaded fabric. Light grey core and dark grey surface. Two perforations near handle. Decorated with incised foliate patterns and slashes. Thirteenth-century. Pit F18 (Fig. 8) No. 13 (WWAF 89:43) Cooking pot with out-turned rim in a hard heavily shellloaded fabric with dark grey core and buff-brown surface with extensive burning. Applied vertical strap. Twelfth-century. BRIAN PHILP 46 Fig. 8 Pottery from Pit F12, Pit F18 and Gully F19 (1:4). THE MEDIEVAL SITE AT WELL WOOD, AYLESFORD 47 No. 14 (WWAF 89:43) Cooking pot with out-turned rim in a sandy shell-loaded fabric with a black core and light brown to black surface. Two horizontal wavy incised lines on neck and shoulder. Twelfth/thirteenth-century. Ditch F19 (Fig 8) No. 15 (WWAF 89:46) Rim and handle of jug in hard sandy fabric with occasional shell-loading. The core is light grey and light brown with a light grey surface. The rim is out-turned and the handle is decorated with deep slashes and possible thumb nail impressions. Thirteenth-century. No. 16 (WWAF 89:45) Cooking pot with flattened rim in a hard sandy heavily shell-loaded fabric. The core is light grey and surface black. Twelfth/ thirteenth-century. (Fig. 9) No. 17 (WWAF 89:47) Cooking pot with flanged rim in a sandy fabric with some shell-loading. Core is dark brown and surface light brown. Twelfth/ thirteenth-century. Fig. 9 Pottery from Ditch F19, Pit F21 and Rubble over walls (1:4). BRIAN PHILP 48 Pit F21 No. 18 (WWAF 89:44) Cooking pot with out-turned rim in a smooth sandy fabric with dark grey core and light orange-brown surface. Twelfth/thirteenthcentury. Dump over Walls No. 19 (WWAF 89:29) Cooking pot with flanged rim in a hard sandy shell-loaded ware with a dark grey core and dark grey to black surface. Decorated with a vertical applied strap. Twelfth/thirteenth-century. No. 20 (WWAF 89:29) Base of large cooking pot in a hard sandy shell-loaded ware with dark grey core and dark grey to black surface. Incised lattice pattern around perimeter of base and two rows of stick stabbing. Twelfth/ thirteenth-century.