Archaeological investigations at the Cluniac House of Horton Priory, Monks Horton
With Luke Barber, Isa Benedetti-Whitton, Emily Johnson, Hayley Forsyth, Elke Raemen and Mariangela Vitolo
Since the 1990s Archaeology South-East has conducted several archaeological investigations using a variety of techniques, including excavation, at the small medieval Cluniac house of Horton Priory, Monks Horton, north of Hythe. While the individual investigations were all relatively limited in scope, cumulatively they succeed in greatly advancing our knowledge of the monastic precinct and its later fate. For the first time, previously unknown monastic buildings, notably the cloisters were identified. The fishponds were partially defined through survey and excavation and a significant finds assemblage, especially pottery and fish bones was recovered.
Horton Priory lies 14km south-east of Ashford near the village of Sellindge (Fig. 1) and is one of the smallest and least understood of the Cluniac houses established in England. Fortunately, however, the conversion soon after the Dissolution of some of the priory buildings, including part of the west front of the aisled twelfth- century church and the shell of the west range into a domestic residence, ensured their survival.
The site, now a Scheduled Monument (List entry number: 1018878) is still a private residence and is surrounded by mature gardens and lies in the shallow valley of the Horton stream, a tributary of the East Stour. The underlying geology is Head deposits of clay and silt above the Folkestone Sandstone formation.
For the last twenty years, as part of the periodic renovation of the house, Archaeology South-East has been conducting a series of archaeological investigations, including watching briefs, historic building survey, geophysical ground probing radar (GPR), topographic survey and evaluation trial trenching (ASE 1997, 2000, 2014, 2015; Acta 2014; Fig. 1 and Fig. 2). While much of this work was limited in scope, the combined results allow us to present the first consideration of the site based on archaeological evidence.
Historical background
The priory was founded by Robert de Vere, Constable to Henry I, at some time between 1121 and 1135, probably in the 1120s (Mayhew 2013, 46) and was among[pg1]the last Cluniac houses to be established in England. In 1144, the papacy confirmed the foundation although this has occasionally been confused as the date of its actual establishment. Although by no means the smallest, compared to the major Cluniac house in England, Horton was limited in both size and in the number of monks. Unlike many Cluniac houses that were sited close to towns, such as Lewes (Sussex), or castles (for example Castle Acre, Norfolk), Horton was a remote location and even within its own parish, the priory was some distance from the manorial centre.
[fg]png|Fig. 1 Location of the site and the archaeological investigations.|Image[/fg][pg2]
[fg]png|Fig. 2 Plan of geophysical survey and archaeological features.|Image[/fg]
[pg3][pg4][fg]png|Fig. 3 A plan of the suggested layout of Horton Priory.|Image[/fg]
Horton was a middling-sized dependency of Lewes with a resident community of between nine and nineteen monks. It maintained a full Cluniac liturgy, recruiting novices, offering hospitality to visitors and administering its estates (Mayhew 2013, 337). The buildings of Cluniac monasteries were laid out in a standard pattern within the constraints of the site, based on the original plan of Cluny itself. The usual arrangement was for the cloister to lie on the south side of a cruciform church abutting the nave and the corner formed with the south transept. The other three ranges were the dormitory above an undercroft to the east; the refectory on the south side; parlour or locutory with cellars to the west. In addition, there was usually a latrine or reredorter in the south-east corner and a kitchen in the south- west. The infirmary usually lay to the east of the church and the prior’s lodgings and guest accommodation to the north, although at Horton they were on the west side of the cloisters (Fig. 3). A brewery and bakery would normally have lain beyond this. Water management was an essential feature of any monastery and houses were expected to have an adequate water supply, a mill and often fish ponds.
Documentary sources and the results of the historic building survey of the priory allow some insight into its latter fortunes (ASE 1997; Acta 2014). Alterations to the west range were undertaken around 1400, and a timber-framed range was added to the north during the middle/late fifteenth century. The priory was dissolved in 1536 and the west range was converted for secular use, undergoing further modifications during the sixteenth and seventeenth centuries. Around the turn of the twentieth century, the southern end of the house was remodelled and, in 1913/14, extensive additions and some alterations to the main house were undertaken, designed by George Hornblower.
The Archeological Features
Medieval Cluniac Priory buildings and features
Building 1: Cloisters
To the east of the existing building, in the area of the former cloisters, were three fragmentary lengths of masonry walls found in Trench 2 and Trench 11 (Figs 1, 3 and Figs 4 and 5). There were two north-south aligned portions ([2/007] and [2/013]) and a single east to west wall G22.
All of the walls had suffered from robbing events, and in no instances were the in situ masonry remains more than two courses high. The walls were built of un- mortared ragstone blocks and cobbles laid in irregular courses and occasionally bonded in off-white lime mortar. There were no associated finds recovered from the masonry, and these walls are dated by being stratigraphically earlier than the later robbing events.[pg5]
[fg]png|Fig. 4 Plan of Building 1 and other features in Trenches 2 and Trench 11.|Image[/fg]
While in themselves, the keyhole nature of the investigations greatly limited the amount of interpretation possible, they did serve to ‘ground truth’ the results of the geophysical survey. The walls identified by excavation and geophysical survey could well represent more than one phase of construction and all are not necessarily of contemporary date. Nevertheless, a c.25m by 25m ‘hollow square’ plan form typical of cloisters is clearly apparent in the northern portion.
This was particularly well-defined in the north-west cloister area by north to south walls [2/014] and east to west wall G22 and the prominent geophysical anomaly to the west. Together these define a narrow linear space that can be interpreted as the cloister walk and of approximately the same width as the cloister walk at Battle Abbey (c.4m wide; Hare 1985). As befitting a small establishment, the cloisters were a relatively modest in scale, smaller than those at Battle and Romsey abbeys (ibid.; Scott 1996). To the east, wall [2/006] and the other geophysical anomalies are harder to interpret, but they may have been part of the extension to the cloisters that visitors to the priory remarked upon in 1279 (Page 1974, 151-153).
Two small ditches [2/004] and [2/010] were also identified dug into the natural clay. Both were filled with brown gravelly silt [2/003] and [2/009], with the former containing a large assemblage of pottery sherds dating c.1225-1325, mostly from four green-glazed jugs and from fill [2/009] were six sherds of pottery dating[pg6] c.1250-1350. Interestingly, the jugs from [2/003] had a rough finish and appear to be kiln ‘seconds’, indicating that either the priory was deliberately purchasing ‘seconds’ or was actually involved in pottery production (see Barber below).
[fg]jpg|Fig. 5 Photograph of Trench 11 under excavating facing west.|Image[/fg]
[pg7][fg]png|Fig. 6 Plan of wall [1/007] (Building 2) and possible grave [1/003] in Trench 1.|Image[/fg]
These ditches, located within the cloisters are difficult to interpret as so little was seen, but as both were parallel to, and located either side of, partially robbed wall [2/007], they could feasibly represent contemporary drainage ditches. Alternatively, they could represent an entirely separate phase of land-use to the wall foundations.
Building 2: ?South transept of the church
Only a single trench (Trench 1) was located in the area of the former church, immediately north of the cloisters. Two features were identified, masonry wall ([1/007]) and a possible grave ([1/003]) (Figs 1, 3 and Figs 6 and 7).
Cut into natural [1/004] was construction cut [1/008] for masonry wall footing [1/007]. The wall, which was not excavated, was aligned north-south and built of un-mortared ragstone blocks laid in irregular courses. The wall had been partially robbed by feature [1/006], and was filled by orange brown silt clay [1/005] which produced no finds. If the eastern cloister alley has indeed been identified further south, then this wall, given its proximity to the site of the church, could perhaps represent the eastern wall of a suggested south transept.
In the eastern end of the trench was pit [1/003]. Only part of this feature was exposed and its exact form and nature are not clear. Pit fill [1/002] was an orange brown silt clay with occasional charcoal flecking and fragments of ceramic building material (CBM). Significantly, at the base of the fill was the fragmentary partial remains of a human skull, which includes adult parietal and temporal fragments.[pg8]
[fg]jpg|Fig. 7 Photograph of wall [1/007] facing west.|Image[/fg][pg9]
[fg]png|Fig. 8 Plan of wall [4/010] in Trench 4.|Image[/fg]
Several of these fragments have exhibited signs of possible ‘thickening’, however due to the limited size of the assemblage the aetiology remains unknown. Too little of the feature was seen to be certain if this skull was part of an articulated burial, although as the head was located in the west, this is a distinct possibility.
If pit [1/003] does indeed represent a grave, then a cemetery was located to the immediately east of the church. However, disarticulated human bone has also been previously found elsewhere, suggesting that parts of the cemetery suffered disturbance and the bone is scattered (Griffin 2000).[pg10]
[fg]jpg|Fig. 9 Trench 4 wall [4/010] looking north (1m scale bars).|Image[/fg]
Building 3: South range
A single wall length ([4/010]) in Trench 4 was the only element of this building seen and its interpretation remains tentative (Fig. 1 and Figs 8 and 9). The wall foundation was constructed of uncoursed ragstone blocks which did not appear to be bonded by mortar. The wall itself was not excavated, and only a small amount of its backfill was dug. It is probable that the wall remains were in fact rubble infill, the facing masonry of which had been robbed.
The relatively large amount of pottery from the wall foundation backfills ([4/008], [4/009] & [4/013]) was generally of large, unabraded sherds, and dated to the latter twelfth to early thirteenth century. This suggests that the wall was likely to have been built some 50-100 years after the foundation of the priory in the 1120s and may represent an early extension or remodelling. Of particular interest is the results of environmental sample (<1>) from backfill [4/009], which produced a varied selection of remains, including hazelnut shells, wood charcoal, mammal and fish bones. Of the 265 fish bones present, 212 could be identified to species, including bones from thornback rays, cod, flat fish, whiting, smelt, and possibly plaice, seabass and herring. Interestingly, all are saltwater species, and no fish remains potentially associated with the adjacent fishponds were found. This assemblage is especially interesting as these fish are likely to have been provided to the priory by the tenants of its various possessions and a cartulary dating to 1438 details[pg11]the residents of Tinton giving eels and red herrings (Mayhew 2013, 355-356). Its closest possessions were scattered across south-eastern Kent and included several near the coast, particularly around Hythe (see Map 1 p.26). Also likely provided by the tenants was the c.1.5kg of oyster shell recovered from [4/009] and dominated by upper valves, indicative of waste from consumption.
The amount and variety of finds and environmental remains from fills [4/008], [4/009] and [4/013] is surprising for apparent wall backfill and is more characteristic of refuse disposal deposits. The limited nature of the intervention makes any interpretation difficult, but an alternative view is that these deposits were not wall backfill at all, but rather the fills of pits truncating earlier wall [4/010].
Other possible wall foundations can be identified as anomalies in the geophysical survey, although the amount of truncation and depth of demolition deposits in this area hamper the interpretation. Nevertheless, the results suggest the presence of a former L-shaped building at least 20m by 20m in size, and probably representing part of the south range. Normally in this location would be situated the refectory, kitchens and other ancillary structures and the varied finds assemblages recovered may derive from refuse disposal from the nearby kitchens.
Building 4: ?Extension of west range
The interpretation of this building is somewhat tenuous, based on two wall lengths and an adjacent rammed chalk floor in Trench 6 and a service trench recorded during the watching brief (Figs 1, 2 and Fig. 10). Walls [81] and [83] were seen in a narrow service trench recorded during the 1999 watching brief. Both uncoursed walls were up to 0.5m wide, aligned east to west and bonded in an off-white coarse lime mortar. Wall [81] was the better preserved, surviving to a height of 0.42m, but both had suffered later truncation. There was no dating evidence recovered in association with these walls.
To the immediate south, identified in an evaluation trench, was a rammed chalk floor [6/006]. The surface, likely an external yard, was not excavated but finds recovered from its uppermost horizon included thirteenth-century pottery sherds and a possible hearth furniture rod of uncertain date.
While little was seen of these structural remains, they can be relatively confidently ascribed to a fifteenth-century extension of the west range. The existence of this former extension was proposed after a historic building survey of the surviving west range, identified inserted doorways in this location (ASE 1997, 9-10). As the survey concluded, this extension was a low timber-framed structure, walls [81] and [83] would have been the foundations of dwarf walls to accommodate the timber sill beams. However, few other structural details are apparent and this extension remains difficult to interpret.
External surface
A small portion of well-laid cobbled surface [8/003] was seen in Trench 8 (Fig. 1 and Fig. 11). The surface was constructed from water-rolled stones set without mortar and was recorded but not excavated. Although no datable finds were recovered, this feature was cut by a fifteenth/sixteenth-century pit ([8/004]),[pg12]suggesting it is likely to be medieval in date and contemporary with the life of the priory. Elsewhere, Trenches 5 and 7 contained only features relating to post- medieval and modern landscaping.[pg13]
[fg]png|Fig. 11 Plan, section and photograph of cobbled surface [8/003] of Trench 8.|Image[/fg]
Fishpond
To the south of the house, in the fields of open pasture, the earthwork remnants of a fishpond were visible (Fig. 1 and Figs 12 and 13). The earthworks were topographically surveyed and examined by two trenches (Trenches 9 and 10). Trench 9, located within the former southern fishpond, was excavated through a[pg14]series of water lain fills to the base of the pond. The pond was 2.1m deep and had been excavated through the cap of natural clay to the underlying natural gravels. The pond was not therefore lined or water-tight, but rather relied on the naturally high water-table as a source. The only finds recovered were fifteenth/sixteenth-century CBM from [9/003] and seventeenth/eighteenth-century CBM from [9/002]. These were the two uppermost deposits and the finds date the final filling of the pond. The environmental samples from the fills produced very little material.
[pg15]
[fg]png|Fig. 12 Plan of the Trenches 9 and 10 and the topographical survey of the fishpond and leat.|Image[/fg]
[pg16]
[fg]png|Fig. 13 Sections of Trenches 9 and 10.|Image[/fg]
The fishpond was connected to a leat to the west and this was investigated by Trench 10. Leat [10/006] was steep-sided and filled with a series of water-lain clays and silts with finds of fourteenth/fifteenth-century CBM recovered from the upper fills ([10/004] & [10/002]). An environmental sample from leat [10/006], fill [10/005] contained poorly-preserved nightshade (Solanum sp.) and carrot family (Apiaceae) seeds in addition to daphnia ephippia (water-flea resting eggs). Though no finds were recovered from the lowest fills of these man-made water management features, a medieval date for their construction is suggested by the presence of late medieval and post-medieval finds in the uppermost fills.
Dissolution and abandonment
The priory was formally dissolved under the Act of 1536 and in the following year the site was leased to Richard Tate (Page 1974, 152). Only the western range, which was converted into a dwelling, was retained with all the other structures demolished. The results of the geophysical survey and excavated trenches indicate that the areas to the immediate south of the western range, and at some distance to the south-east were utilised as dumps for the vast amount of demolition material. Presumably, much of the building material was salvaged for reuse or sale, with the remainder dumped.
Extensive deposits and features dating to the probable post-Dissolution abandonment were identified across the site. These most notably in the form of wall robber trenches and massive dumps of demolition material. These deposits and features contained abundant late medieval/early post-medieval CBM as well as large amounts of residual medieval pottery.
The robber trenches had affected all the wall foundations identified and were often the only indication for their former location. The robber trenches were seen in the cloisters area, in Trench 1, 2 and 11. Wall [G22] was robbed by [118], wall [1/007] by [1/006], wall [2/007] by [2/006] and wall [2/013] by [2/012] (Figs 4 and 6).
To the immediate south of the house in the area of Trench 4 were a series of demolition dumps, c.1.2m thick ([4/007], [4/006], [4/005] and [4/002]) containing fifteenth/sixteenth-century CBM and residual medieval pottery sherds (Fig. 8). Within the deposits was a single posthole [4/004] containing no finds and representing some form of short-lived occupation. The geophysical survey results also appeared to show another area of widespread demolition deposits to the immediate east of the cloisters, although this was not verified by excavation.
To the west of the buildings were a series of large pits identified in Trenches 3 and 8 ([3/006], [3/008] and [8/004]; Figs 1, 11 and Fig. 14). These features contained moderate amounts of fifteenth to seventeenth-century CBM and are likely to associated with demolition events.[pg17][pg18]
[fg]png|Fig. 14 Plan of pits in Trench 3 and section.|Image[/fg]
Finds
The Pottery by Luke Barber
The archaeological work recovered a large assemblage of medieval pottery from the site: 1,178 sherds, weighing 14,864g. The vast majority of the assemblage comprises of medium to large sherds with no or minimal signs of abrasion. As such, the majority of the assemblage does not appear to have been subjected to repeated reworking though it is clear from the pottery and ceramic building material there are low levels of residuality and/or intrusiveness in some deposits. Although a fairly wide chronological spread is represented the vast majority of the assemblage can be dated to the mid twelfth to thirteenth century.
The pottery is significant as it is the first notable assemblage from the site. The earliest fabric consists of the sand and shell tempered coarsewares that probably mainly derive from the Ashford Potter’s Corner industry (EM.M5), though some of the coarser vessels could be from the coastal industry (EM3 type), part of which was based on Romney Marsh (Barber forthcoming). Although these wares may have been in use from the mid twelfth century the current types are considered more likely to belong to a c.1175/1200 to 1250 period, particularly considering the association with some of the sandy wares. Notable fresh groups were recovered from the fills ([4/008] and [4/009]) from wall construction cut [4/010] of Building 4, totalling 128 sherds (4,202g) and 62 sherds (1,386g) respectively. These consist of cooking pots with a range of rectangular, tapering and rounded club rims, sometimes with applied thumbed strips (Fig. 10, Cat. Nos 1-8) (Fig. 15). In the lower fill ([4/009]), the fresh EM.M5 sherds (from five different cooking pots) are associated with EM29 (10/134g from two cooking pots), M40A (9/128g from seven different cooking pots), M40B (7/80g from five different cooking pots), M40BR (1/2g from a clear/green glazed jug) and M7 (10/44g a jug with applied red clay strips under a clear glaze). The presence of significant quantities of more developed sandy wares in association with the EM.M5 vessels suggests a deposition date perhaps around the middle of the thirteenth century. The assemblage from fill [4/008] has a similar spread of fabrics but with a little intrusive late post-medieval material.
Sherds dominated by sand conform with the types typically ascribed an Ashford/ Wealden source in the CAT fabric series. M40A, with sparse/common calcareous inclusions, M40B, with essentially medium sand only and M40BR, a more highly- fired sandy type are all represented in small numbers, the latter two types often in the form of jugs. The final type M40C is present in much larger quantities, always as jugs. These often have a thin uneven white slip and patchy green glaze with incised line decoration being common (Fig. 16, Cat. Nos 10-12). The rod handles are typically decorated with ring and dot stamping. This type is not well understood, but probably spans c.1225 to 1325. Of particular interest in the current assemblage is the large group of 208 sherds (4,492g) of M40C fill [2/003] of ditch [2/004], that appear to derive from just four green glazed jugs. Some of these have a somewhat rough finish, and one vessel in particular shows some bubbling/ blowing of the internal surface during firing and has a large broken bodysherd from another jug firmly stuck to its glaze (Fig. 17, Cat. No. 11). As such, either the site was purchasing ‘seconds’ or these are wasters and the site was actually[pg19] [pg20]involved with production. The latter would be of particular interest, as the source of M40C is still not known. The only other sherd in this deposit is from an EM29 cooking pot.
[fg]png|Fig. 15 Medieval pottery illustrations 1-6.|Image[/fg]
The latest medieval sherd consists of a 30g fragment from a well-fired fine sandy transitional vessel recovered residually from modern pit [7/003], fill [7/004]). This is the only sherd that post-dates c.1350 suggesting a significant decrease in refuse disposal or at least within the investigated area after the mid fourteenth century.
Catalogue (Figs 15-17)
Cooking pot. Light grey core with pale orange surfaces. EM29. Cut [4011],
fill [4/008] (G26 of B2). (Fig. 15)
Cooking pot. Mid grey core with grey-brown surfaces. Some external soot-
ing. EM.M5. Cut [4011], fill [4/008] (G26 of B2). (Fig. 15)
Cooking pot. Mid grey core with grey-brown surfaces. Some external soot-
ing. EM.M5. Cut [4011], fill [4/008] (G26 of B2). (Fig. 15)
Cooking pot. Mid grey core with grey-brown surfaces. Some external soot-
ing. EM.M5. Cut [4011], fill [4/008] (G26 of B2). (Fig. 15)
Cooking pot. Mid/dark grey core with grey-brown surfaces. Some internal
sooting. EM.M5. Cut [4011], fill [4/008] (G26 of B2). (Fig. 15)
Cooking pot. Mid grey core with grey-brown outer surface and dull brown interior surface. Some external sooting. EM.M5. Cut [4011], fill [4/008] (G26 of B2). (Fig. 15)
Cooking pot. Mid grey core with dull orange surfaces. Some patchy external
sooting. EM.M5. Cut [4011], fill [4/008] (G26 of B2). (Fig. 16)
Cooking pot. Dark grey core with dark grey interior and patchy dull orange brown/grey exterior surfaces. Some external sooting. EM.M5. Cut [4011], fill [4/008] (G26 of B2). (Fig. 16)
Jug. Blue grey core where vessel walls thicker, otherwise dull orange throughout. Stub of a stabbed rod handle but no glaze apparent on the sur- viving part of the vessel. M40BR. Cut [4011], fill [4/008] (G26 of B2). (Fig. 16)
Highly-decorated jug with simple pulled spout, oval-sectioned rod handle and plain base. Pale grey core with pale brown surfaces. Incised horizontal lines around neck with incised lattice design over upper half of body and ring and dot stamping down rear of handle. Poorly applied dull green glaze over neck and upper parts of body, becoming much sparser toward girth. Some traces of thin white slip notable on lower body. M40C. Fill [2/003] of ditch [2/004] (G23). (Fig. 16)
Highly-decorated jug with supported pulled spout, oval-sectioned rod han- dle and plain base. Mid grey core with brown surfaces. Incised vertical wavy lines between vertical straight lines over top half of body and ring and dot stamping down rear of handle. Poorly applied dull green glaze over neck and upper parts of body, becoming much sparser toward girth. Some traces of thin white slip notable in places. This vessel certainly appears to be a waster[pg21]or a ‘second’. In places the vessel wall has blown from internal air pockets
(b) and there is at least one large bodysherd from a similar vessel to No. 10 (though not the same vessel) that has fused itself to the external glaze (c). In addition there are adhering pellets of red clay in places around the neck. M40C. Fill [2/003] of ditch [2/004] (G23). (Fig. 17)
Highly-decorated jug with well-formed rounded rim. Pale grey core with pale brown surfaces. Incised vertical lines forming zones in which there is alter- nating horizontal incised lines and vertical rows of impressed ‘dots’. Patchily applied dull green glaze over neck and upper parts of body, becoming much sparser toward girth. M40C. Fill [2/003] of ditch [2/004] (G23). (Fig. 16)
[fg]png|Fig. 16 Medieval pottery illustrations 7-10, 12.|Image[/fg]
[pg22][pg23]
[fg]png|Fig. 17 Medieval pottery illustrations 11a-11c.|Image[/fg]
The Ceramic Building Material by Isa Benedetti-Whitton
A moderate assemblage of 643 fragments of ceramic building material weighing a total of 34,827g was recovered. Roof tile is often difficult to date in isolation, as the general form of peg tile changes little from the fourteenth to the nineteenth centuries, and many fabrics were in use from the medieval and throughout the post-medieval period. However, certain fabrics were used for shorter periods of time, for example MOLA 2273, which was only used from the late twelfth-mid thirteenth century. A number of 2273-type tiles were yellow brown or dark brown glazed, and glazing is a technique generally becomes less common after c.1400 (Drury 2000, 58).
Several T1 tiles and occasional examples of T2A and T2B also had either surviving patches of glaze or glaze splatter, including a piece of medieval ridge tile. Unglazed fragments of ridge tile, or potentially from which the glaze had become abraded, were also present, in fabrics T2B and T3, although there is always the possibility that these represent later renovation. Although the priory was built during the latter twelfth century, it was partially remodelled during the fourteenth century (Baily 1876). With the exception of the fragments of extremely sandy roof tile fabric 2273, the bulk of the tile assemblage could belong to either phase of work.
No pieces of roof tile were recovered in situ from medieval contexts and the vast majority was recovered from the post-medieval demolition deposits and features, mainly from the robbing of wall G22 by [118]. In addition, 12 fragments in fabrics T1, T2A, T2B and T3 were recovered from the fills of modern drain [131]. Round and diamond-shaped peg holes were noted in all fabric types apart from fabric T1 which were only round.
Only three pieces of floor tile were recovered, all in fabric FT1. Fabric FT1 was had very similar inclusions to roof tile fabric T2A and it is likely they were derived from a common clay source, potentially by the same makers. The characteristics of the tile pieces, although all very fragmentary, suggest a common type. Two fragments were glazed in monochrome: one was green-black, and the other a similar shade of yellow-brown to the glaze found on the roof tile. All the tiles were 21-22mm thick, and are most probably of a type similar to thirteenth- and fourteenth-century ‘Westminster’ or Penn tiles, which were fairly small square tiles, either glazed monochrome like the present examples or more decorative polychrome, and are often found on ecclesiastic sites.[pg24]
[fg]png|Fig. 18 Registered finds illustrations.|Image[/fg]
The Glass by Luke Barber and Elke Raemen
A total of six shards of glass weighing 642g was recovered from two different contexts. The earliest and most interesting fragment comprises a blue shard of medieval grisaille window with flower pattern from fill [109] of modern drain [110] (Fig. 18). It has been grozed on three sides and was snapped off on the fourth. It is likely that it derived from a diamond-shaped quarry (3.4mm thick), originally grozed on all four sides. The fragment dates to the thirteenth to fifteenth century; however, the window could have been damaged much later, for example during the Dissolution when they were collecting the lead cames. Decorative fragments of window glass were also often reused in later glazing regimes (John Shepherd pers. comm.).
The Animal Bone by Emily Johnson and Hayley Forsyth
Less than 2kg of animal bone was recovered and a limited range of taxa are represented including cattle, sheep/goat and pig. The assemblage is dominated by cattle and large-mammal which are represented by meat-bearing and non-meat bearing bone. Sheep/goat and pig are represented by mandibles and teeth only including two large canines from a boar. Butchery was noted on two specimens including a cattle pelvis and a pig mandible. It seems likely that the bones represent a sample of domestic waste though the assemblage is too small to provide any meaningful data regarding contemporary animal husbandry techniques.
In addition, 265 fish bones, mostly post-cranial elements were recovered from bulk sample <1> of wall backfill [4/009]. The fish bone was in moderate to poor condition with signs of surface erosion, and 212 bones could be identified to species. The fish bone consists of common species expected from the area, the majority of which include dorsal and ventral buckler spines from thornback rays. Also present were vertebrae from the family Gadidae including cod and whiting, flat fish (Pleuronectidae) including plaice, smelt, possible seabass and herring.[pg25]
The Registered Finds by Elke Raemen
Of interest is a copper-alloy thimble fragment from robber trench fill [121] of wall G22 (RF <1>; Fig. 18) dating between c.1300 and 1650. The thimble is probably hammered and stamped and shows circular indentations in concentric lines as well as a ‘tonsure’ on top. It measures 16mm high and c.16mm in diameter.
Environmental remains by Mariangela Vitolo
In the six bulk soil samples charred plant remains were scarce. These were principally hazel (Corylus avellana) nutshell fragments from wall backfill [4/009] and poorly-preserved nightshade (Solanum sp.) and carrot family (Apiaceae) seeds [pg26]from leat [10/006], fill [10/005], in addition to daphnia ephippia (water-flea resting eggs).
[fg]png|Map 1. The possessions of Monks Horton Priory in Kent (from Mayhew 2013, fig. 27, reproduced by kind permission of the author).|Image[/fg]
The identified taxa of the ten charcoal fragments from wall backfill [4/009] included oak (Quercus sp.), beech (Fagus sylvatica) and possible hornbeam (cf. Carpinus betulus). The oak fragments were from both slow and quick grown wood and no small round wood was present. The wood of beech and oak makes an excellent fuel and both taxa might have been chosen specifically for this characteristic. They are indicative of deciduous woodland, where the possible hornbeam could also have been present.
Conclusions
Compared to many Cluniac foundations, Horton Priory was a modest and remote establishment which, until now, has remained archaeologically unexplored (Aston 2002). While limited in scope, these surveys and investigations have together succeeded in greatly advancing our knowledge of the monastic precinct and its later fate. For the first time, previously unknown monastic buildings, notably the cloisters were identified. The fishponds were partially defined through survey and excavation and a significant finds assemblage, especially pottery and fish bones was recovered.
In addition to achieving a fuller understanding of the priory’s ground plan, the extent of the cloister can be suggested and the location of the church and associated cemetery has been proposed. From the archaeological evidence the priory appears to have been established in the mid/later twelfth to thirteenth centuries, which corresponds reasonably well with the probable date of its founding in the 1120s. After the Dissolution in the 1530s and 1540s, most of the priory seems to have been formally demolished and much of the building material robbed for use elsewhere.
Acknowledgements
ASE would like to thank Curt and Carmen Englehorn for commissioning the work and for their assistance throughout the project, and Paul Roberts, Historic England Archaeologist, for his guidance and monitoring. The investigation was directed by Giles Dawkes. The author would like to thank all archaeologists who worked on the excavations. Fiona Griffin, Lauren Gibson (pottery illustrations) and Antonio Reis (finds photographs) produced the figures for the report, Neil Griffin project managed the excavations and Jim Stevenson and Andy Margetts who project managed the post-excavation process.
Bibliography
Acta, 2014, Horton Priory, Monks Horton, Kent: Conservation Management Plan, unpubl. report.
ASE, 1997, Horton Priory, Monks Horton, Kent, Historic Building Survey, unpubl. report. ASE, 2000, An Archaeological Watching Brief at Horton Priory, Monks Horton, Sellindge, Kent, unpubl. report.
ASE, 2014, Geophysical Survey Report, Horton Priory, Monks Horton, Kent, unpubl. report.[pg27]
ASE, 2015, Archaeological evaluation report: land at Horton Priory, Monks Horton, Kent, unpubl. report.
ASE, 2016, Horton Priory, Monks Horton, Kent Written Scheme of Investigation for Archaeological Mitigation. Archaeology South-East unpubl. material.
Aston, M., 2002, Monasteries in the Landscape, Tempus.
Baily, C., 1876, ‘Monks Horton Priory’, Archaoelogia Cantiana, 10, 81-89.
Barber, L., forthcoming. ‘The pottery’, in Archaeological investigations on the New Romney and Greatstone 1st time sewerage scheme, Canterbury Archaeological Trust monograph.
Drury, P., 2000, Aspects of the production, evolution and use of ceramic building materials in the Middle Ages, Medieval Ceramics, 24, 56-62.
Hare, J.N., 1985, Battle Abbey, The Eastern Range and the Excavations of 1978-80, HBMCE, Arch. Rep. No. 2.
Mayhew, G., 2013, The Monks of Saint Pancras: Lewes Priory, England’s Premier Cluniac Monastery and its Dependencies 1076-1537, Lewes History Press.
MoLA, 2014a, Medieval and post-medieval roof tile fabric codes, available online: https://www.mola.org.uk/medieval-and-post-medieval-roof-tile-fabric-codes.
MoLA, 2014b, Medieval and post-medieval brick and drain fabric codes, available online: https://www.mola.org.uk/medieval-and-post-medieval-brick-and-drain-fabric-codes
Page, W. (ed.), 1974, The Victoria History of the County of Kent, Vol 2.
Scott, I.R., 1996, Romsey Abbey report on the Excavations 1973-1991, Test Valley Arch Trust Mono 8.[pg28]