The Early Medieval Pottery assemblage from the Excavations at Lyminge, 2008-2015
This report presents the typological analysis of the early medieval pottery assemblage recovered from the University of Reading excavations at Lyminge between 2008-2015. The assemblage embraces the entirety of the settlement’s early medieval occupation sequence, dating between the c.mid-5th - 9th centuries, and is situated within a contextual, spatial, and chronological framework supported by radiocarbon dating (Thomas 2013, 2017; Yorke 2017). The report provides a typological characterisation of fabrics identified in the assemblage, followed by discussion of chronology, vessel types, and decoration.
The early medieval assemblage consisted of 15,805 sherds, weighing 130,512g, with a total rim Estimated Vessel Equivalent (EVE) of 84.3, which is exceptionally large for a settlement assemblage of this date. Given the average sherd weight of 8.3g, it is likely that the vast majority of the material reflects tertiary deposition (Backhouse, 2021, pp. 167-206). A total of fourteen local ware groups are present and divide into two distinct chronological phases: c.5th to 7th century wares and c.7th- to 9th-century wares. Eleven Continental imported wares (with four additional unsourced fabrics) and four non-local regional wares are also present. In addition to this material, there was a small assemblage of Roman pottery deposited primarily in archaeological features of c.5th- to 7th-century date (Backhouse, forthcoming; Timby 2021).
Located roughly 6 miles (10km) inland from the early medieval trading site of Sandtun, West Hythe (Fig. 1) with which it was closely associated, and 15 mile (24km) south of Canterbury, Lyminge sits in the North Downs, at the head of the Nailbourne valley.
[fg]jpg|Fig. 1 Lyminge location map.|Image[/fg][pg95]
[fg]jpg|Fig. 2 Site excavation plan of all excavation areas from 2008-2015 (© Lyminge Archaeological Project).|Image[/fg]
The Lyminge Archaeological Project was initiated in 2007 to explore and characterise the spatial and temporal development of Lyminge as an early medieval monastic landscape. The project consisted of two key phases of open area excavation in different locations within the village (Fig. 2). The first phase took place between 2008-10 to the south of the modern-day church and revealed a c.late 7th- to mid-9th-century monastic settlement The second excavation phase was conducted between 2012-15, primarily on the modern-day village green known as Tayne Field, and uncovered a c.5th- to mid-7th-century settlement (Thomas 2013, 2018). This second area of excavation has two stages of occupation:
c.5th- to late 6th-century occupation denoted by sunken-feature buildings and an infilled solution hollow replaced by ...
c.late 6th to mid-7th century a great hall complex interpreted as a royal estate centre (Thomas 2017b, 2018).
Early medieval Lyminge thus has two key chronologically and spatially distinct periods of occupation, and these underpin the chronological framework of the pottery assemblage.
Methodology
The entire assemblage was sorted and recorded by ceramic fabric type, vessel type, [pg96]sherd type, and decoration. It was quantified by sherd count, weight, maximum vessel count (hereafter MVC), and rim percentage calculated as estimated vessel equivalents (hereafter EVEs) (PCRG, SGRP and MPRG, 2016). It was not possible to identify vessel forms in detail, hence vessel types were identified and described as either ‘jar’, ‘bowl’, or ‘pitcher’.
An internal, site-specific type series was created, independent of all previously identified fabrics at Lyminge. Fabrics were classified by dominant inclusion (Orton and Hughes 2012, pp. 278-84). Due to the nature of hand-built pottery, there is a lack of standardisation in fabrics, thus individual fabrics are aggregated into coherent ware groups according to presence of the same dominant inclusion and, where relevant, level of inclusion processing (Orton and Hughes 2012, pp. 71-80). The term ‘local’ is given to ware groups that were most probably produced within the borders of Kent, on the basis of their material constituents and similarity to other pottery types representative of ‘Kentish’ ceramic traditions, identified primarily in the Canterbury Type Series (Macpherson-Grant 1995). The term ‘non- local’ describes regionally imported wares sourced from outside of Kent, primarily comprising known types such as Ipswich ware, recognisable from the raw materials employed and/or technology not locally available in early medieval Kent. Finally, the term ‘imports’ describes pottery produced on the near-Continent, embracing recognised ware types emanating from northern Francia and the Rhineland regions.
Where individual fabrics are discussed herein, their type series reference number is cited. (Individual fabric descriptions are available by contacting the author.)
Fabrics – c.fifth- to seventh-century ware groups
Wares of this date total 8,958 sherds, weighing 76,652g, accounting for approximately 59% by weight of the assemblage as a whole. Local ware groups principally comprise two local sandy ware types, iron-rich ware, and organic- tempered ware (Table 1). Sandy wares with chalk are less well-represented, and it is possible the low quantities of micaceous, gritty, and sandstone-tempered products indicate these are non-local types. Approximately 6% by weight are imported wares, principally sourced from northern Francia. Non-local regional wares account for less than 1% by weight.
Local ‘Kentish’ wares
Ill-Sorted Sandy ware: this is the largest ware group in the c.5th- to 7th-century assemblage (Table 1) and comprises fifteen individual fabrics. The group is characterised by dominant quartz-sand inclusions that vary in size from fine to coarse, are sub-rounded to angular in shape, and occur in different levels of abundance between fabrics and sometimes within sherds. Fabrics are differentiated by the relative quantity and size of inclusions, and some are further distinguished by additional occasional, fine, black ironstone inclusions. Surface texture ranges between fabrics from hard and rough to soft and smooth. Vessels are hand-built, reduce-fired at a low temperature, and colours vary from mid-brown to dark grey, even across a single sherd or vessel. This ware compares to EMS 1, Sandy ware in the Canterbury Ceramic Type Series.[pg97]
TABLE 1. QUANTITIES OF FIFTH- TO SEVENTH-CENTURY WARES
(ordered by local, non-local and imported types, and by relative abundance)
Where identifiable, the most commonly represented vessel type are jars. Bowls are only identifiable in the two most common fabrics of the ware. Three ‘miniature’ bowls are also present. Vessels are predominately globular in shape with simple, rounded everted rims. Ill-sorted Sandy ware accounts for almost 60% of all decorated sherds in this chronological range, of which incised decoration is most common. Bossed decoration in the form of large, finger-tip impressions is present on seven sherds and one sherd is decorated with multiple flower stamps, Type A7, in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983; see section below on decoration).
Iron-Rich ware: there are twelve fabrics in this ware, characterised by ferruginous inclusions ranging from black iron ore to red iron ore and naturally iron-rich argillaceous rock. Black iron ore is generally fine and sparse, whereas red iron ore and other ferruginous inclusions are medium in size and typically range from moderate to abundant in quantity. Inclusions are generally angular to sub-rounded in shape. Fabrics are hard and the surface texture is quite smooth. Vessels are hand-built and low-fired, varying from brown to grey in colour. Fabric 247 differs from the main group; it has finer inclusions and orange-buff surfaces. It may be wheelthrown and therefore later in date, although sherds are too small to confidently determine this. A similar fabric in the Canterbury Ceramic Type Series (MLS 15) is attributed to an east Kent source, thus it is possible that the Lyminge fabric may have a similar origin (Macpherson-Grant 1995, p. 823; Barber 2015, p. 201).
Identified vessels are almost exclusively jars. Jars have simple or slightly thickened, everted rims and a sagging, globular profile. One handle fragment cross-fitting with a rim sherd may be from a pitcher-like vessel. This rim differs from jar rims, being upright and rounded in profile. Twenty sherds are decorated, of which three-quarters are incised.[pg98]
Organic-tempered ware: comprises seven fabrics characterised by dense, chopped organic material that gives sherds a highly-laminated structure within a fine clay matrix. Fabrics are differentiated by the abundance of the organic material and the additional presence of sparse quartz inclusions. All fabrics are soft and smooth in texture. Fabrics are hand-built, reduce-fired, and range in colour from brown to black. This ware compares to EMS 4, Organic-tempered ware in the Canterbury Ceramic Type Series.
Identified vessels are mostly jars with simple, everted rims. Two flat bases are interpreted as bowls, or perhaps dishes. Eight decorated sherds are present, almost exclusively with simple incised lines.
Well-sorted Sandy ware, Group I: this ware comprises seven fabrics with well- sorted quartz-sand inclusions. All fabrics have well-sorted inclusions that vary in size and abundance to differentiate between fabrics. Inclusions range in size from fine to medium-coarse and in quantity from moderate to abundant, dependent on the fabric. Inclusions are either rounded or sub-rounded in shape. All fabrics are hand-built, with a smooth to slightly rough surface texture, and vary from light brown to mid-grey in colour. The ware is related to Ill-sorted Sandy ware, but inclusions are more consistent within fabrics in this group. It is also related to Well-sorted Sandy ware Group II, which is a chronologically later development (see below).
The majority of identified vessels are jars with everted, rounded rims. There are fifteen rims with a slightly clubbed profile. Three pitchers are interpreted from two rounded rod handles and one upright, rounded rim sherd, while five bowls are present in rounded profiles with flat bases. Decoration is present on 31 sherds, the greater proportion of which are incised with linear motifs. Bossed decoration occurs exclusively in Fabric 130, perhaps signifying it is the product of a single potter or workshop.
Fine Sandy ware: this ware comprises two fabrics characterised by fine to very fine quartz-sand inclusions. Inclusions consist of very well-sorted, common to moderately abundant, rounded quartz-grains. Both fabrics have a smooth surface texture and evidence of burnishing. Vessels are hand-built and reduce-fired, ranging in colour between fabrics from dark-grey to mid-brown, with a grey core. This ware is related to other sandy wares, but sherds are harder-fired, and quartz inclusions are finer in size and abundance.
There are ten rounded jars and two bowls. Rims are almost exclusively rounded and everted in profile, and one slightly clubbed rim is present. Incised decoration is most common, with chevron motifs featuring on four sherds. One sherd has an incised linear motif and stamped small circles, Type A1, in the Archive of Anglo- Saxon Pottery Stamps (Briscoe 1983).
Sandy ware with Chalk: there are four fabrics characterised by chalk inclusions in addition to quartz-sand. Inclusions are fine to moderately sized, infrequent, sub- rounded quartz-grains and frequent, angular chalk inclusions that vary in size from fine to moderate, depending on the fabric. Some fabrics have vesicules on the surface where chalk inclusions have leached out. All fabrics are hand-built and [pg99]some have a slightly rough surface texture, while others have a smooth texture. Fabrics are reduce-fired and range from light to dark grey in colour. This ware compares to EMS 2, Chalk-filled sandy ware in the Canterbury Ceramic Type Series.
Twelve jars are represented. A single shouldered-jar profile is present featuring a flat base with rounded sides. Most rims are everted in a rounded profile and one rim is slightly clubbed. There are three sherds decorated with incised horizontal lines.
Mixed Grit-tempered ware: this ware comprises four fabrics with mixed inclusions of moderate to abundant, crushed flint, quartz, ironstone and infrequent quartzite. Inclusions vary in amount depending on individual fabrics. Inclusions are typically angular, coarse to very coarse in size and poorly sorted. This ware has a rough texture and a sandy matrix, and breaks with an irregular to jagged fracture. Fabrics are hand-built, reduce-fired, and appear grey to dark-brown in colour. Ten plain jars were identified; rims have rounded, everted profiles and there are two slightly sagging bases with rounded sides. Decoration is absent.
Sandstone-tempered Fabric: represented by a single fabric characterised by coarse, moderate, sub-rounded sandstone fragments; medium-sized, sparse quartz; and fine, sparse, black ironstone. The fabric has a slightly rough surface texture. All sherds are hand-built, oxidised and orange throughout. Comparable wares are more common in the Thames Basin, particularly in northern Kent and southern Essex (Blackmore and Vince 2008, p. 155); the most likely source of this fabric is therefore northern Kent.
Non-local wares
Micaceous ware: comprises two fabrics characterised by dominant white mica visible on sherd surfaces. Both fabrics have inclusions of poorly sorted, medium to fine, abundant white mica; unsorted, medium-sized, sparse to moderate felspar; and medium-fine sized, sparse, rounded quartz and angular granite. Vessels are hand-built, reduce-fired, and mid-brown to black in colour. A potentially similar ware is known from Lundenwic (London), where it is attributed to a source in the Charnwood Forest area to the south of Leicester and dated to the 5th to 7th centuries (Williams and Vince 1997, pp. 219-20; Cowie et al. 2012, p. 246). This ware would benefit from further analysis to confirm its source. Four jars are represented and have rounded, everted rims and there is one sagging base sherd.
Continental Imports
Continental imports account for 6% by weight and 5% by sherd count of the 5th to 7th-century assemblage (Table 2). Nine imported fabrics were identified. Fabrics are grouped together on the basis that they represent established ceramic traditions derived from sources on the near-Continent, primarily from northern Francia. These comprise five broad groups: northern French Greyware, Blackware, Low Countries Greyware, Late Mayen Ware, and an oxidised ware possibly sourced from northern France.[pg100]
TABLE 2. FIFTH- TO SEVENTH-CENTURY CONTINENTAL IMPORTED WARES
Northern French Greyware: two fabrics are present, characterised by a distinctive grey colour. Inclusions consist of well-sorted, medium to fine, moderate, sub- rounded quartz. Additional inclusions of sparse, fine, black ironstone are also present in both fabrics. The ware is hard, wheelthrown, and has a smooth surface texture. The defining characteristic is its grey colour; one fabric has a mid- to dark- grey core with dark-grey surfaces and the other fabric has a mid-grey core and dark-grey surfaces. This ware is recognised as a product of northern France, with a range of possible sources including Normandy and the Pas-de-Calais area. The latter is more likely given Lyminge’s position near to the southern Kentish coastline (Evison 1979, p. 48; Macpherson-Grant 1995, p. 823). This ware compares to EMS 9, north French Greyware in the Canterbury Ceramic Type Series.
Thirty-eight jars are identifiable. Rim sherds are almost exclusively rounded and everted in profile and one squared rim is present. Incised decoration is most common, comprising linear motifs and three sherds have chevrons. One sherd is decorated with an incised line and has small squares rouletting.
Low Countries Greyware: is represented by a single fabric and is a distinctive wheelthrown ware due to its ‘sandwich firing’. Inclusions consist of very well- sorted, fine to very fine, moderately-abundant, rounded quartz. Sherds are hard [pg101]and have a smooth surface; some examples have had their surfaces wiped prior to firing. The ware is partially oxidised with orange to mid-brown surfaces and a mid-grey core. A possible source is eastern Belgium, or perhaps the Meuse Valley (Evison 1987, p. 93; Blackmore 2003, p. 239). Thirteen jars were identified in this ware. There are nine rounded, everted rims and two flat base sherds. Decoration comprises incised linear motifs.
Blackware: four wheelthrown fabrics are present, characterised by black surfaces. Inclusions consist of well-sorted, medium-sized, moderate to moderately- abundant, well-rounded quartz. Two fabrics have additional inclusions of fine, sparse to moderately abundant, black ironstone. Sherds have a smooth surface texture and are reduce-fired. Sherds have black burnished surfaces and a reddish- brown core. This ware was produced at a range of centres in northern France and eastern Belgium (Blackmore 2001, pp. 193-4). It compares to EMS 8, Northern French Blackware in the Canterbury Ceramic Type Series.
A total of six jars were identified from five rounded, everted rims and a flat base sherd. A flat lid sherd decorated with an incised line is also present. Decoration is rare, the most common being incised lines. There is a stamped motif of small, plain rectangles, Type C1, in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983). One sherd is decorated with rectangular notch rouletting.
Late Mayen Ware: this is an imported late Roman coarseware characterised by its very hard, dense fabric. Inclusions consist of poorly-sorted, abundant, angular glassy mixed quartz and quartzite with occasional black, angular volcanic inclusions. The ware has a slightly rough surface texture and orange-brown surfaces with a dark grey core. The ware is sourced from the Eifel region of the Rhineland and dates to c.350-450 (Tyers 1996). The ware is represented by a single vessel that is a plain, lid-seated, rounded jar with a flat base. The vessel type is classified as R1 (Redknap 1987).
Unidentified/unsourced Oxidised Fabric: Fabric 297 is characterised by its orange colour. It has well-sorted, medium-sized, sparse, sub-rounded quartz inclusions, with rare, coarse, black mica. The ware has a smooth surface texture and is coloured orange throughout resulting from oxidisation during firing. This ware is possibly sourced from north-east France or eastern Belgium (Evison 1979, p. 43). Sherds are too fragmented to suggest vessel type.
Fabric 171 is characterised by red quartz-inclusions and may be wheelthrown, although sherds are too small to confirm this. Inclusions comprise well-sorted, fine, moderately-abundant to abundant, rounded red quartz. The fabric is slightly rough and reduce-fired, with a grey core and dark beige to mid-brown surfaces. The source is unknown, although given that it may be wheelthrown, an imported source is suggested. Three jars are present, represented by two rounded, everted rim sherds and a rounded base sherd. One pitcher is interpreted from a strap handle.[pg102]
FABRICS – SEVENTH- TO NINTH-CENTURY WARE GROUPS
Wares of this date total at 6,847 sherds, weighing 53,860g, accounting for 41% by weight of the whole assemblage (Table 3). The principal ware groups are local Sandy wares and local Shell-tempered wares. A small quantity of rock-tempered wares are also present, some of which may be from non-local sources. Non-local regional wares, principally Ipswich ware, comprise 2% by weight of wares in this date range, while Continental imports, mainly from northern France or Flanders, account for 6.1% by weight.
TABLE 3. SEVENTH- TO NINTH-CENTURY WARES
Local ‘Kentish’ wares
Well-sorted Sandy Ware, Group II: comprises four fabrics characterised by inclusions of very well-sorted quartz. Inclusions are consistent in size and quantity within each fabric but vary between fabrics. Quartz-inclusions range in size from medium-fine to moderate and vary in quantity from moderate to abundant, generally being rounded in shape. Fabrics are hand-built, have a slightly rough surface texture, and are reduce-fired, ranging in colour from mid-brown to dark grey. The ware compares to MLS 2, Canterbury-type sandy ware in the Canterbury Ceramic Type Series.
The main vessel type is jars. One vessel, possibly a bowl, has a pierced flat base. This may have been used for dairy processing, based on comparative examples suggested to have this function (Hamerow 1993, p. 44). Jars are mainly rounded in profile and a small number of slightly shouldered profiles are also present. Jars have a small range of rim forms, the most common being everted rims in a rounded profile. Other rim forms comprise slightly squared, everted rims, a few slightly clubbed rims, and three slightly hooked rims. A single inturned rim is present; this is a common characteristic of bowls with pierced bases, so may be part of the vessel with the pierced base. Most of the base sherds are flat or sagging, some of [pg103]which may derive from bowls or shallow dishes rather than jars. The presence of nine jar profiles with flat bases indicates at least some jars have flat bases. There are 42 decorated sherds showing a variety of techniques, of which bossing and incised decoration are most common. One sherd has a ring and dot motif, Type A2 ‘multiple circles’, in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983).
Shell-Tempered Ware: there are five fabrics characterised by the presence of crushed shell inclusions. Shell inclusions are well-sorted, range in size between coarse to medium, and in amount from abundant to moderate between individual fabrics. All fabrics are hand-built, and one may have been finished on a turntable. The surface texture varies from smooth to slightly rough, and colours vary from mid-brown to grey throughout, to oxidised orange throughout, dependent on fabric. One of the fabrics, Fabric 183, is identified as ‘Romney Marsh Shelly ware’ (Allen 1999, p. 12). This ware compares to MLS 4, Shell-tempered ware and MLS 5, coarse sand- and shell-tempered ware in the Canterbury Ceramic Type Series.
Jars are the most commonly identified vessel type, and five bowls are also present. Jar rims are everted and most have a rounded profile, with a small number of squared, slightly hooked, or clubbed examples also present. Bowls are interpreted from flat base sherds. Bossing is the most common form of decoration, occurring exclusively in Fabric 168.
Fine Sandy Ware: comprises two fabrics grouped on the basis of very well-sorted, fine quartz-grain inclusions. Fabrics have a fine matrix, very well-sorted, very fine, rounded quartz or possibly river silt inclusions. The fabrics are very hard, have a smooth surface texture and are reduce-fired, being grey to mid-brown in colour. The ware compares to MLS 2/LS1, Canterbury-type sandy ware (Mid/Late transition) in the Canterbury Ceramic Type Series.
The largest quantity of identified vessels are jars in a rounded profile. The majority have rounded, everted rims. Four slightly hooked rims are also present, and three rims are slightly clubbed and squared in profile. Most vessels are plain, although seven sherds have been decorated with pushed outwards “repoussé” bosses, and one sherd has an incised line.
Flint Tempered Ware: this ware comprises four fabrics characterised by inclusions of crushed flint. Flint inclusions are mostly white, poorly-sorted, coarse in size, moderate to common in abundance, and angular in shape. Some sherds also contain coarse, sparse red flint and medium, sparse quartz-sand. It is hand-built and generally hard, with a slightly rough surface texture. Vessels are reduce-fired and range from mid-brown to grey in colour. This ware compares to MLS 3, Sandy ware with sparse-moderate flint in the Canterbury Ceramic Type Series.
Sixteen jars were identified in this ware, primarily with rounded, everted rims. One squared rim sherd is also present. Jars are rounded in profile and three slightly sagging base sherds are present. Decoration is rare, occurring as bosses or incised lines.
Sandy Ware with Shell: comprises two fabrics characterised by quartz-sand inclusions mixed with crushed shell. Inclusions consist of moderately sorted, fine [pg104]to medium sized, common rounded and sub-rounded quartz mixed with coarse shell laminations that range from sparse to moderate in abundance. These fabrics are differentiated from Shell-tempered ware due to the dominant presence of quartz- sand. Sherds have a slightly rough surface and range from partially oxidised to reduce-fired, varying between orange to mid-brown in colour.
Twenty jars were identified and have everted rims in a rounded profile. Four rims exclusive to one fabric (Fabric 184) are everted and squared in profile, suggesting they are products of the same potter. Decoration is rare, with bossed sherds being dominant. A single stamped sherd is present with a flower motif, identified as Type A7 ‘flowers’ in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983). One rim sherd has thumbed decoration.
Shell- and Flint-Tempered fabric: a single fabric related to Shell-tempered ware but distinguished by the presence of flint mixed with crushed shell. Inclusions are moderately-sorted, medium-coarse, moderately-abundant, angular, crushed, white flint mixed with moderately-sorted, coarse to very coarse, moderately- abundant, shell laminations. The fabric is rough in texture and reduce-fired, having light brownish-orange surfaces and margins with a mid-grey core. One rounded, undecorated jar is present with a rounded, everted rim.
Non-Local Wares
Four non-local imported fabrics are represented, accounting for 2% by weight and 0.5% by sherd count of pottery of 7th- to 9th-century date (Table 4). These consist of Ipswich ware in two fabrics (fine and coarse variants), possible Winchester-type ware, and Oolitic Limestone-tempered ware.
Ipswich Ware: was produced in and around the wic town of Ipswich and is generally given a date range of c.720-850 (Blinkhorn 2012). Ipswich ware represents the first post-Roman pottery produced on an industrial scale and manufacture utilising a turntable in early medieval England (Blinkhorn 2012, p. 1). Two fabric variations are present at Lyminge: Group 1: Fine Sandy Ipswich ware, and Group 2: Coarse Sandy Ipswich ware, according to Blinkhorn’s fabric definitions (2012, pp. 16-7). The finer fabric is dominant at Lyminge.
TABLE 4. SEVENTH- TO NINTH-CENTURY NON-LOCAL WARES
[pg105]
Fine Sandy Ipswich Ware (Group 1): inclusions consist of moderately well-sorted, sub-angular quartz-grains; fine grains are abundant and medium-sized grains are common to moderate in amount. The fabric has a fine, dense matrix and is hard and well-fired, with a smooth surface texture, probably resulting from smoothing prior to firing. The fabric is typically grey throughout, although oxidised and partially- oxidised examples occur. This ware compares to MLS 7A, Ipswich-type ware in the Canterbury Ceramic Type Series.
Two jars and a pitcher were identified. One jar has a simple, everted rim in a rounded profile. The pitcher is thick walled and has a thickened, straight-edged rim with a strap handle. This vessel is decorated with lattice stamps, Type D2 ‘Grid Ovals’ in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983) which are located in triangular zones, and also has thick applied clay strips (see section on decoration below). Similar vessels are known from the early medieval trading site of Sandtun, near Hythe (Blackmore, 2001a, p. 206), and Lundenwic (Blackmore 2003, pp. 234-5).
Coarse Sandy Ipswich Ware (Group 2): inclusions of quartz are medium to coarse in size, sparse to moderate in abundance and rounded in shape. The fabric is very hard, well-fired, and has a ‘pimply’ surface texture created by inclusions protruding from the surface. Surfaces are greyish-black in colour with a brownish-grey core. This ware compares to MLS 7B, Pimply Ipswich-type ware in the Canterbury Ceramic Type Series. One sherd with a slightly clubbed rim in a rounded profile is interpreted as a jar based on form comparisons after Blinkhorn (2012, pp. 26-7).
Winchester-type Ware: a single sherd may be Winchester-type ware, or Michelmersh ware, based on comparison with a sample in Southampton Museum assessed by Jervis (2011, p. 23). It has inclusions of medium, sparse, rounded, and sub-angular quartz and occasional, fine black ironstone. The fabric is slightly rough in texture, wheelthrown and oxidised-orange throughout.
Oolitic Limestone-tempered Ware: Represented by a single sherd, this fabric is characterised by inclusions of coarse, rounded limestone and abundant, rounded voids, suggesting the presence of leached oolitic limestone. The fabric is low- fired, dark grey in colour, and hand-built. A similar fabric at Springhead, near Gravesend, is attributed to an East Anglian source (Mepham and Vince, 2011, p. 17). The Lyminge fabric may be from the same, or similar, source.
Imported Wares
There is a total of 335 sherds, from a maximum of 290 vessels identified as imported wares dating between the 7th to 9th centuries (3,278g, 1.7 EVE; Table 5). Some fabrics, particularly reduce-fired wares, are not easy to distinguish from one another, hence some superficially different types may well be related. Six sources have been identified, with two additional wares requiring further analysis to determine their provenance. The main groups comprise Greywares and Blackwares (both from northern France and the Low Countries), Whitewares (northern France) and Shell-tempered ware (Quentovic region). [pg106]
TABLE 5. SEVENTH- TO NINTH-CENTURY CONTINENTAL IMPORTED WARES
The imports identified at Lyminge were compared to the Hamwic (Southampton) Continental imports ceramic Type Series. The assemblage from Lyminge differs from Hamwic, where some 30% to 50% are suggested to have been supplied by potters in La Londe, near Rouen (Hodges, 1991, p. 884). The absence of directly comparable wares at Lyminge indicates imports were not supplied by the same industries as Hamwic. It is, however, evident from similarities in Greywares and Blackwares that the examples at Lyminge fall within the bounds of the same ceramic traditions as those represented at Hamwic.[pg107]
Northern French Greyware: is represented by one fabric, distinguished by its light grey colour. Inclusions consist of moderately-sorted, medium-sized, moderately abundant, rounded quartz-grains. The fabric is hard, wheelthrown, and has a smooth surface texture. The ware is light grey throughout. The most likely source is the Pas-de-Calais region (Blackmore 1988, p. 90; 2001, p. 194). This ware compares to MLS 12, north French Greyware in the Canterbury Ceramic Type Series.
Shelly Ware: three Shelly ware fabrics are present, characterised by sparse, crushed shell inclusions. One fabric (Fabric 178) is wheelthrown and has coarse, moderately-abundant, shell laminations; coarse, sparse, red flint; and fine, rounded quartz. Fabric 174 is oxidised throughout and has moderately-sorted, coarse to very coarse, abundant shell laminations. Of the identified types, seven jars are present which have rounded, everted rims. Two bowls are interpreted from flat base sherds associated with obtuse sides and two pitchers are interpreted from a strap handle and a clubbed rim. There are 24 sherds decorated with roulette motifs, the most common being small diamonds or squares in a horizontal linear motif. A single sherd is stamped with a ring and dot motif, Type A2 in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983). The third fabric (Fabric 194) is very hard and has inclusions of moderately abundant, iron rich argillaceous inclusions, with occasional fragments of coarse shell, and medium sized, moderate quartz. All of the fabrics correspond with imported wares identified at Sandtun; Fabric 178 compares to Sandtun Group 2, Fabric 174 compares to Sandtun Group 1, MSS 1b, and Fabric 194 compares to Sandtun Group 1, MSS1a (Blackmore 2001, p. 204). The region of Quentovic has been suggested as a possible source for the ware group (Blackmore 2001, p. 204), and its presence at both Lyminge and Sandtun supports the interpretation of written sources that these sites were connected in some way (Thomas, 2008).
Of the identified vessels, fourteen jars are present. All rims are everted, but the exact profiles vary slightly depending on the fabric. Rims in the oxidised and iron- rich fabrics are less angular compared to rims in the wheelthrown fabric. Three flat base sherds are present, and there is one base with a slightly sagging profile.
Blackware: two Blackware fabrics are present, characterised by their distinctive black surfaces. The principal inclusion in this ware is quartz. In Fabric 305, quartz inclusions are poorly-sorted, coarse to medium sized, and common grains. In Fabric 121, well-sorted, medium-fine, common grain. Both fabrics have reddish- brown cores and black surfaces, some examples are burnished. This ware was produced at a range of centres in northern France and eastern Belgium (Blackmore 2001, p. 193). It compares to north French Blackware in the Canterbury Ceramic Type Series.
The assemblage consists principally of body sherds, most likely derived from jars. One pitcher was confirmed by the presence of a spout decorated with oval lattice stamps, Type D2 ‘Grid Ovals’ in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983). Two sherds have small square rouletted motifs.
Badorf-Type Ware: three buff-coloured fabrics are identified as possible Badorf-type ware based on parallels at Lundenwic (Blackmore 2003, p. 240) and Hamwic [pg108]
(Jervis, 2011, p. 20); however, further analysis is preferable to confirm this. Inclusion consists of fine-medium sized, sparse, sub-angular quartz. The ware is hard and smooth in surface texture; it has pinkish-grey surfaces, and some sherds have a light-grey core. Fabric 313 is coarser, having additional inclusions of medium-sized, iron-stained quartz and a slightly pimply surface texture. Badorf ware derives from a range of sources in the Cologne region (Blackmore 2003, p. 240). The only confidently identified vessel is a pitcher with an upright, rounded rim with a tubular spout and a straight-sided vessel profile.
Northern French Reduced Sandy Whiteware: is represented by a single fabric, characterised by white/cream surfaces. Sherds have inclusions of medium-sized, moderately abundant, rounded quartz-grains, some of which are iron-stained. The ware is hard and has a pimply surface texture. Sherds are white/cream in colour throughout. Comparison with a similar ware at Sandtun, ‘Sandy Whiteware with Reduced Surfaces (Lundenwic NFWR)’, suggests a possible source in the Meuse valley (Blackmore 2001, p. 195). One sherd is painted with a red lattice motif.
Low Countries Greyware: one fabric is present, characterised by ‘sandwich firing’. Inclusions consist of well-sorted, fine, common, rounded quartz-grains. The fabric is wheelthrown, has a smooth surface texture and is fired to produce a distinctive ‘sandwich’ effect of a mid-grey core, with oxidised orange surfaces. It is thought that Low Countries greyware, as a regional tradition, was produced over a wide area including the Pas-de-Calais, Picardy, and western Belgium, as well as further east in the Meuse Valley (Blackmore 2003, p. 239).
Unsourced imports: Three imported fabrics have not been identified by source.
Fabric 233 is hard-fired, with well-sorted inclusions comprising fine, moderate quartz; rare coarse, angular quartzite; and coarse, sparse calcite, which is mostly represented by voids. The fabric has a smooth texture and is coloured dark grey throughout.
Fabric 242 is a hard, wheelthrown fabric where inclusions comprise very fine, sparse, rounded quartz. The fabric is smooth and has a mid-grey core with dark brownish-grey surfaces.
Fabric 240 is a hard, wheelthrown fabric with inclusions of well-sorted, fine, moderate to common black ironstone and well-sorted, medium sized, sparse quartz. Sherds have a smooth surface texture and are mid-greyish-cream throughout. Sherds appear to be externally burnished and/or have been wiped internally to smooth the surface prior to firing.
Unidentified Sources and Residual Wares
A small group of fabrics have affinities with Romano-British wares but also compare with contemporary imports of early medieval date, meaning definitive identification is uncertain. The majority of these fabrics derive from 5th- to late-6th- century [pg109]
contexts, some which contained small quantities of residual material visibly Roman in date (such as Samian ware), alongside early medieval wares, including imports. Because of this, the contexts do not distinctively assist with dating.
Unsourced Very Fine Sandy Fabric 197: is distinguished by partially oxidised surfaces and very fine to silty inclusions of quartz-sand grains. Quartz inclusions comprise well-sorted, very fine, moderately-abundant, rounded grains of quartz and very occasional finely chopped, burnt organic matter. The fabric has a slightly rough texture. Sherds are mid-grey in colour and some examples have a patchy, slightly oxidised surface. The source is unknown; it is potentially late Roman in date because larger sherds suggest it is wheel-thrown (pers. comm. Macpherson- Grant 2018).
Unsourced Micaceous Fabric 195: is characterised by clear, micaceous inclusions. Quartz inclusions comprise fine, very common, rounded glassy grains in addition to sparse, fine, clear mica present on the outer surface. It is wheel-thrown, has a smooth texture, and a mid-grey core with light brownish-orange external surfaces and light brownish-cream internal surfaces. It is possible this fabric is Roman in date.
Unsourced Very Fine Fabric 288: is characterised by very fine quartz inclusions that are very well-sorted, very fine, rare, rounded quartz-grains which may be river-silt. This wheelthrown fabric is smooth in texture, and sherds range in colour from mid-brown to dark grey throughout. It is possible the fabric is Roman in date. Decoration is present on two sherds in the form of stamped motifs. The stamps comprise: small squares categorised as Type C1 ‘plain rectangles’ in the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983) and small ovals Type D2 ‘grid ovals’ in the Archive of Anglo-Saxon Pottery Stamps (Briscoe, p. 198).
Grog-tempered Fabric: the ware is represented by one fabric characterised by the presence of grog inclusions. Inclusions are coarse to medium sized, abundant, orange grog, and medium-sized, moderately-abundant, sub-rounded quartz. The ware is hard and rough, with patchy orange-grey surfaces and a black core. It is possible the ware is of Roman date; however, comparative wares are present in Flanders date to the 6th to 8th centuries, meaning this could be the source of the fabric.
Chronology
To build a reasonably secure ceramic chronology, local ware groups were examined for their occurrence in radiocarbon dated features and assigned date ranges accordingly (Table 6), on the basis of calibrated 2 sigma (95% certainty). These date ranges are estimated and have rounded-up limits; dates are intentionally broad to encompass the longest feasible duration of ware production. It is acknowledged that some dates will be narrower than what has been assigned. Where radiocarbon dates are not available, artefactual dating is used to provide a terminus ante quem for archaeological contexts. Imported and non-local wares are dated in accordance [pg110]
TABLE 6. DATE RANGES ASSIGNED TO ‘KENTISH’ WARES FROM STRATIFIED CONTEXTS ONLY, BY SHERD COUNT
with previously established chronological schemes, where possible, or assigned dates in line with those given for comparative assemblages, primarily Canterbury (Table 7) (MacPherson-Grant 1993; Macpherson-Grant 1995; Barber 2015), Sandtun (Blackmore 2001b) and Lundenwic (Blackmore 2003).
c.Fifth- to late sixth-century pottery
Ill-sorted Sandy ware occurred in every archaeological context radiocarbon dated between the 5th and late 6th centuries and its near-absence in stratified post 7th- century features reaffirms this date range for production and use (Table 6). This broad date range corresponds with the dating given to quartz-sandy ware (EMS 1) in the Canterbury Ceramic Type Series (Macpherson-Grant 1995, figs 364, 385). Four wares (Fine Sandy ware, Grit-tempered ware, Sandy ware with Chalk and Well-sorted Sandy ware Group I), appear to have narrower date ranges, ending by c.550 (Table 6). This is based on their greater presence in archaeological contexts radiocarbon-dated to between c.400-537 (C14 dated cal. ad c.401/5-537, [pg111]
TABLE 7. DATE RANGES ASSIGNED TO CONTINENTAL IMPORTS AND NON-LOCAL IMPORTS FROM STRATIFIED CONTEXTS ONLY
[pg112]
95% certainty, OxA-31785/ OxA-31786; C14 dated cal. ad c.396/421-537, 95% certainty, OxA-31722/ OxA-31723; C14 dated cal. ad c.392-537, 95% certainty, OxA-31961). These date ranges are reinforced by the low quantities of these wares in features with later radiocarbon dates of c.550-650 (C14 dated cal. ad c.537-644, 95% certainty, OxA-31713).
The date assigned to Organic-tempered ware deserves attention. The ware occurs in higher proportions in features that fall within a 5th- to late 6th- century date range (67% by MVC) compared to features of late 6th- to 7th-century date (30% by MVC; Table 6). Moreover, it derives primarily from two features that date no later than c.580. This indicates that the main period of production and use of Organic-tempered ware must predate c.600. When correlated with radiocarbon dates, it appears that Organic-tempered pottery was in use by at least the early part of the 6th century. This is some 50 years earlier than the date proposed for Organic-tempered ware at Marlowe Car Park in Canterbury, where it has been stratigraphically dated to the late 6th to 7th century (Macpherson-Grant 1995, figs 364, 385). It is possible therefore that Organic-tempered ware was either present earlier at Lyminge than at Canterbury, or that the date of Organic-tempering in Kent, which is derived principally from evidence at Canterbury, is earlier than previously thought and may require revision.
c.Late sixth- to late seventh-century pottery
The continuation of both major and minor wares from the preceding phase of occupation in stratified late 6th- to late 7th- century contexts affirms that a substantial proportion of the deposited material is residual in nature. This is further demonstrated by a high-level of sherd fragmentation and abrasion, particularly noted for Ill-sorted Sandy ware and Organic-tempered ware.
Shell-tempered ware stands out, however, and is quite distinctive as the only ware present in significantly greater abundance in discrete contexts of this date range. Shell-tempered ware primarily derives from SFB 1, and a small number of postholes and the wall trenches associated with the 7th-century Great Hall Complex. The primary fill of SFB 1 produced a radiocarbon date of cal. ad c.570- 650 (95% certainty, SUERC-35927) and datable finds including disc-headed pins and beads, all correspond with a 7th-century date (Scull 2012). Hence, the presence of Shell-tempered ware in SFB 1 provides a terminus post quem of the early/mid- 7th century. This is somewhat earlier than the date proposed for the production of Shell-tempered ware elsewhere in Kent. At Canterbury, Shell-tempered wares do not appear stratigraphically until the late 8th century (MacPherson-Grant 1993; Macpherson-Grant 1995, p. 823; figs 364, 384) and at Sandtun, it is broadly dated to the 7th to 10th centuries (Blackmore 2001, p. 205). Stratigraphic evidence at Lundenwic and Hamwic suggest a late 8th- to 9th-century date for local shelly wares (Timby 1988a, pp. 84; 114-6; Blackmore 2003, pp. 237-8). Elsewhere in early medieval England, Maxey-type ware, a shell-tempered pottery found in Lincolnshire and in the south-east Midlands, has been dated to c.650-850 (Hurst 1976; Young 1996), which corresponds more closely with the suggested inception of Shell-tempered ware at Lyminge. On the near-Continent, the development of shell-tempered pottery in Frisia, Flanders and Racquingham in the Pas-de-Calais [pg113]region has been dated between the final quarter of the 8th to 10th century (Stilke 1995, p. 15; Thuillier 2015, p. 130), whereas at Quentovic, a production date of the c. 7th to 10th centuries has been suggested (Coutts 1991, pp. 95-6). The contextual and scientific dating available at Lyminge suggest the introduction of shell-tempering is similar in date to its emergence in Quentovic. This could be related to the close relationship between communities in south-east Kent and northern Francia at this time, which is evidenced by the presence of contemporary Continental imports sourced from the same region (Table 7).
c.Late seventh to ninth-century pottery
Pottery assigned to this date range is dominated by local wares, with Shell- tempered ware accounting for the largest proportion. Shell-tempered ware occurs in all features that were radiocarbon-dated to Lyminge’s late 7th- to 9th- centuries monastic phase of occupation, demonstrating the extended currency of this pottery over two centuries, from the c.early 7th century until the mid-9th century. Sandy- ware with Shell is assigned a date of c.675-775 based on radiocarbon dates derived from its presence in a pit (cal. ad c.653-766, 95% certainty, OxA-31750) (Table 6) but it may have a longer period of production. During the 8th century, Sandy-ware with Shell and Fine Sandy ware co-existed, although Fine Sandy ware constitutes a significantly greater quantity and seems to have had a longer production period, continuing into the 9th century (Table 6). Fine Sandy ware is similar to Canterbury Sandy Ware (MLS3) and here is consistent with the 8th- to 9th-century date assigned at Canterbury (Macpherson-Grant 1995).
Flint-tempered ware represents a ceramic tradition that straddles the two main periods of occupation at Lyminge (Table 6). The individual fabrics within the tradition divide into two sub-groups based on their inclusions: two fabrics have coarse flint and are present in earlier dating features, and two different fabrics have flint and sparse quartz inclusions and are present in later dating features. These differences imply the utilisation of different raw material sources. Alternatively, because flint is an uncommon temper in Kent but more commonly used further west in East Sussex (Jervis et al. 2016), it is possible that the ware is a non-local import and thus the differences between the fabrics represent a change in source and/or how the pottery supply to Lyminge was organised.
Vessel types
This section offers the first comprehensive study of vessel types for an assemblage of this size from early medieval Kent. Such a study is long overdue; hence the region can now be brought into wider discussions of the range of vessel types identified at early medieval settlements. The Lyminge assemblage does not have any complete vessels, although there are a handful of partial vessel profiles derived from cross fits and larger sherds. However, in general, few vessels are represented by more than a single sherd.
Jars ‘Cooking Pots’
The largest group of identified vessels in the assemblage are jars, represented [pg114]by 1,253 by maximum vessel count (MVC) from stratified contexts (Table 8, Fig. 3). Overall, the dominance of jars for the duration of Lyminge’s 400-year occupation indicates that the functional demands of pottery saw little change over time. Locally-produced sandy wares account for the largest proportion. Jars are generally globular in form, although a small number of c. late 7th- to 9th-century vessels have flat or slightly sagging bases. Earlier dating jars typically have rim diameters ranging between 140-180mm (Fig. 3, C, D). Jars in later dating wares [pg115]have more varied rim diameters; while the majority of post 7th-century wares have rim diameters ranging between 160-240mm (Fig. 3, A), some jars have diameters measuring between 220-280mm (Fig. 3, B). Jars occur in a range of sizes, from small to larger vessels, the former being more common. It is feasible that some larger jars were used for storage, but the commonality of sherds with sooting or evidence of being heated indicates that most jars were probably utilised for food preparation. Two jars have rounded, upright pierced lugs (Fig. 3, H), which were likely used to suspend the vessel; however, lugs are extremely rare in the assemblage showing this was not a common attribute.
[fg]jpg|Fig. 3. Examples of identified vessel types: Jars, E: Bowl, F: Handled bowl, G: Lug, H: Pierced base sherds, I: Spouted pitcher, J: Pitcher with strap handle (Drawings by Lesley Colett.)|Image[/fg]
TABLE 8. IDENTIFIED VESSEL TYPES IN STRATIFIED CONTEXTS
All local wares have everted rim forms. The majority are simple, everted rims in a rounded profile and there are a few examples in wares that post-date the c.mid 7th century where the rim has been slightly thickened. Shell-tempered ware has some examples of thickened, everted rims in a squared profile, suggesting the use of knife-trimming. Squared rim forms appear primarily in archaeological contexts dating to the second half of the 8th century, corresponding with the peak of Shell- tempered ware, and this is mirrored in the presence of squared rims on some c.8th- to 9th-century local wares at Canterbury (Macpherson-Grant 1995, p. 890; Barber 2015, p. 201). The use of tools on mid-8th-century rims suggests an increased use of the turntable to finish vessels because tools are more easily applied to constantly revolving vessels to produce a consistent, neatly trimmed edge (Macpherson-Grant 1995, p. 890). This could reflect broader technological developments in Kent at this time, as well as influences resulting from contact with pottery industries elsewhere, such as Ipswich (Blinkhorn, 2012), and the near-Continent.
Imported wares account for an MVC of 84 jars. It may be that at least some imported jars were initially employed as containers to transport goods/foodstuffs to Lyminge from the near Continent and were subsequently reused. The majority of imported jars are rounded in profile with simple, everted rims, in accordance with Evison’s (1979, pp. 43-5; Backhouse 2024) work on wheelthrown pottery in graves. Typologically, the earliest is an imported Late Mayen-ware jar with a lid- seated rim, rounded body, and flat base, classified as form R1 after Redknap (1987). This is the commonest and most widespread form in this ware and suggested to primarily functioned as containers (Tyers 1996). Three northern French Greyware jars have everted rims in a squared profile, corresponding with north French Greyware jars at Lundenwic (Cowie et al. 2012, p. 252).[pg116]
Bowls
Bowls are the second most common vessel type in the assemblage, of which 48 MVC are from stratified contexts (Table 8). As with jars, local sandy wares account for the majority. Bowls are represented throughout the occupation sequence, although fewer are present in c. 5th- to 7th-century wares (Table 8). Bowl rims measure between 220-320mm in diameter. Vessels have obtuse sides and flat bases (Fig. 3, E). The examples at Lyminge have two main profiles, having either rounded- sides with upright, rounded rims, or straight-sides with slightly flared, everted rims. There are also a few examples of rounded bowls with clubbed rims. Seven identified bowls have simple, inturned rims, of which one inturned rim belongs to a bowl with flared sides. One imported ware bowl has a slightly hooked rim. Three bowls are associated with handles. Examples comprise a small, rounded handle (Fig. 3, F), a thin strap handle, and a rod handle. The Lyminge examples appear to fit with a group of handled bowls found in southern England, such as at Hamwic, dating to the 9th to 10th centuries (Jervis, 2011).
Although the function of bowls is unclear, their size and the absence of sooting on sherds suggests they were not typically used as cooking vessels. Bowls may have been used for preparing and/or presenting foodstuff or eating. The handled bowls may have served different functions, perhaps being used as a ladle/scooping implement, or for measuring. The low quantity of ceramic bowls suggests bowl- like vessels were produced in other materials, probably wood (Wood 2005, 2008).
‘Miniature’ Bowls
‘Miniature’ bowls are very rare at Lyminge, represented by only three vessels. They have rounded profiles and are all in Ill-sorted Sandy ware, found in c. 5th- to 7th-century contexts. It is possible that two of the vessels were crucibles based on their size and shape; however, there is no evidence of residue to demonstrate this. The third ‘miniature’ bowl may be handled, inferred from a small rod handle that compares to examples found at Marlowe Car Park, Canterbury (MacPherson- Grant 1993, pp. 167, plate 1). Parallels for the miniature bowls are known from Northfleet (Mepham and Vince 2011, p. 4), Canterbury (Macpherson-Grant 1995, no. 65 and 202), and Bekesbourne (Myres, 1977, p. no.1079). All comparative examples date to the 6th or 7th centuries and this date seems appropriate for the ‘miniature’ bowls from Lyminge.
Pitchers
A maximum of eleven pitchers are identified in the assemblage, of which eight are from stratified contexts (Table 8). Imported wares account for the largest quantity. Pitchers were probably used for storing and pouring liquids (Coutts 1991, p. 125; Medieval Pottery Research Group, 1998). Complete profiles could not be determined due to a lack of sherds; it is probable that pitchers had an overall similar profile to jars, having a rounded, shouldered body, a flat base, and upright, rounded rims. Some vessels have applied spouts (Fig. 3. I). A crude, tubular spout in Shell- tempered ware may be a local attempt to copy imported pitchers with this feature. There is a single Ipswich ware pitcher that has thick walls in a rounded profile, [pg117]a thickened, straight-edged rim, and a simple strap handle (Fig. 3, J). Blinkhorn (2012 pp. 26, 87-9) found that Ipswich ware pitchers occur in higher quantities on sites at the edge of the known distribution area of the ware and typically at ecclesiastical sites; the pitcher at Lyminge accords with this.
Pitchers are associated with liquids and principally used for storage, carriage and serving (Barber 2015, p. 201). There are very few locally made pitchers and, given the rarity of non-local and Continental wares identified at Lyminge, there seems to have been little demand for pottery vessels intended primarily for serving liquid, perhaps because jars could be multipurposed and used for such tasks. The general lack of pitchers corresponds with the low representation of pitchers at contemporary early medieval settlements, including Canterbury (Barber 2015, p. 201) and Lundenwic (Cowie et al. 2012, pp. 251-2).
Pierced Sherd
One fragmented base has a circular pierced hole that has been added prior to firing, suggesting the vessel was produced with a specific function in mind, although what this function was is unclear (Fig. 3, G). Numerous examples of vessels with pierced bases are known from contemporary sites. For instance, 50 examples were identified at Mucking, Essex (Hamerow 1993), and a complete vessel with basal holes from Sutton Courtenay, Berks., was interpreted as a ‘brazier’ (Leeds 1926, p. 72, pl. VI, fig. 2, 10). Comparative examples have been identified at settlements in Gotland, Sweden, and Bornholm, Denmark; these are suggested to have been employed in dairy processing, where pierced holes around the edge of the base allowed for the draining of liquid (Stenberger 1955, p. 1118, fig. 47; Klindt-Jensen 1957, fig. 111.4). Alternatively, Perry (2013 p. 351) argued for the use of a pierced vessel in the beer production process, where liquid and spent grain were strained using a filtering medium.
Lid and Miscellaneous Base Sherd
One lid was identified in the assemblage in imported Blackware. It was not possible to identify the full lid profile but the flatness of the sherd suggests it came from a dish lid (Medieval Pottery Research Group 1998, p. 99). There is one example of a hollow pedestal base, which may have functioned as a bowl support or the base of a lamp.
Decoration
At 2.9% by sherd count of the assemblage, decorated pottery at Lyminge falls comfortably within the range identified at contemporary settlements (Blinkhorn 1997, p. 117). By comparison, the figure for contemporary settlements at West Stow, West Suffolk, and Carlton Colville, Suffolk, is 2% (Tipper 2009), rising to 5% at Mucking, Essex (Hamerow 1993, pp. 21-2). There is no quantification of the Canterbury assemblage; however, it is noted that decorated pottery is rare (Macpherson-Grant 1995, p. 864). In total, there are 493 decorated sherds from Lyminge, of which the majority (82% by sherd count) date to the 5th to 7th centuries.[pg118]
In her analysis of decorated pottery at Mucking, Essex, Hamerow (1993, p. 43) acknowledged that fragmentation causes great difficulty when identifying decorative schemes in full. The following section therefore relies on the presence of decoration on individual sherds. Alongside the decorative techniques discussed here, 232 sherds are burnished.
Incised Decoration
Incised decoration is the most common technique, frequently occurring as linear schemes of multiple lines arranged horizontally, diagonally and/or vertically (Fig. 4). It is present on all sandy ware variations. Ill-sorted Sandy ware accounts for the majority of sherds with incised decoration at 169 sherds, although this accounts for only 3% of the whole ware group (Table 9). Based on contextual dating, these sherds fall within 5th- to mid-6th-century stratigraphy, which corresponds with a recent study dating incised decoration on the near Continent to c.300-500 (Krol, Dee and Nieuwhof 2020, p. 16). This is reinforced by the minimal occurrence of incised decoration on post 7th-century wares, when it typically accounts for less than 1% of each group (Table 9). Where vessel types could be identified, 35 incised sherds are primarily from jars, while one bowl and one pitcher are recognisable, and the lid sherd is also incised.
Eleven sherds in locally-produced sandy fabrics and three imported sherds are incised with more elaborate motifs including chevrons or pendant-triangles. Two locally-produced sherds and one import are incised with chevrons and horizontal lines. One Fine Sandy ware jar stands out, having two raised ribs on the neck and girth of the body, which are decorated with short, incised, diagonal lines (Fig. 4, E). A comparably decorated vessel was deposited in grave 42 at Lyminge cemetery (Backhouse 2024). The second most common ware group with incised decoration are Continental imports, accounting for 43 sherds, which proportionally represent 9% of imported wares, collectively. The majority of these sherds are in northern French Greyware and there are a small number of Low Countries Greyware sherds.
Bossed Decoration
The second most common decorative technique is bossing, present on 139 sherds; five bossed sherds also have incised decoration. It is feasible at least some bossed sherds are from the same vessel, although only two sherds are cross-fits. Bosses are pushed out from the inside (repoussé), and result from two distinct, chronological production techniques. Bosses on c. 5th- to 6th-century wares are repoussé bosses made using a finger-tip; whereas repoussé bosses on late 7th- to 9th-century wares are produced using a tool, perhaps a nail or other implement (Fig. 4, J, K). It is unlikely that these techniques are a direct progression from one to the other due to a substantial interim period in the c.late 6th to mid/late 7th century when bossing is absent from all wares.
Among the identifiable types, there are a maximum of 37 bossed jars, pre- dominately in Shell-tempered ware, and two Ill-sorted Sandy ware jars. The majority of the bossed pottery survives as body sherds, demonstrating the body was [pg121]the main area of vessels with decoration. It is highly-probable that boss-decorated Shell-tempered body sherds are all from jars, though this cannot be confirmed.
[pg119][pg120]
[fg]jpg|Fig. 4. Examples of decorated sherds: incised decoration, F-G: 5th- to 7th-century bosses, H-K: 7th- 9th- century bosses, L-M: rouletting. (Drawings by Lesley Collett, photographs, author’s own.)|Image[/fg]
TABLE 9. INCISED DECORATED SHERDS BY WARE TYPE
Bosses on 5th- to 7th-century pottery are elongated, large bosses, measuring between 13-17mm in size. Bossed schemes are generally spaced in rows, most of which appear to be horizontal around a vessel. There are three sherds in Ill-sorted Sandy ware with elongated bosses beneath thin, horizonal incised lines (Fig. 4, F, G). These bossing schemes at Lyminge compare to bosses on locally-produced hand-built pottery found at contemporary early medieval sites across England as well as in the Rhineland and Holland (Evison 1979, p. 42).
Post 7th-century bosses are also repoussé and the vast majority measure between 13-17mm in diameter. Bosses on Well-sorted Sandy ware Group II sherds are typically smaller, measuring between 6-8mm in diameter. Post 7th- century bossing divides into two main examples:
Group 1 are close-set, continuous bosses;
Group 2 are bosses in spaced rows, some of which are more variably spread than others.
Both types of post 7th-century bossing compare to the bossed decoration identified at Sandtun (Blackmore 2001, p. 212). Some of the larger Lyminge sherds appear to show specific, planned patterns, rather than being applied at random. Two Shell- tempered sherds have schemes including incised line arrangements that give the appearance of framing the bosses, perhaps suggesting that bosses themselves conveyed a particular meaning. Comparable examples of more ‘complex bossing’ are known from Sandtun (Blackmore 2001, p. 212), and Minster-in-Sheppey (Blackmore 2001, fig. 30).
A total of 65 Group 1 boss-decorated sherds with external burnishing are contextually linked by their presence in an 8th-century boundary ditch associated with the monastic settlement. Although only two sherds are cross-fits, it is feasible that at least some sherds are from the same vessel. The sherds occur exclusively in one Shell-tempered fabric, suggesting the decoration and fabric are exclusive to each other and may be the products of a single workshop/potter. The addition of burnishing on sherds in the fabric makes it quite unusual compared to the other Shell-tempered fabrics at Lyminge and could indicate it has a different source. A possible source is Canterbury, where burnished, boss-decorated vessels of 8th- to 9th-century date occur in a coarse shell-tempered fabric (Mid/Late Saxon 4), that appears closely-related to the Lyminge fabric. Further analysis would be beneficial to confirm this.
Two locally-produced sandy wares have close-set repoussé bosses that are small and well-formed, fitting with Group 2. The inside of some bosses retain marks from the tools used in their application. The sherds are in an uncommon fabric in the Well-sorted Sandy ware Group II, which is finer and thinner-walled compared to contemporary fabrics and gives the appearance of superior potting ability, possibly indicating it is not a local product. This uncommon fabric and bossing style correspond with sherds recovered from the monastic sites of St Augustine’s Abbey and St Martin’s Hill in Canterbury, dated to the late 8th to first half of the 9th century (Macpherson-Grant 1995, p. 870; Barber 2015).
The presence of directly comparable wares with bossing at religious settlements [pg122]in Canterbury and Lyminge perhaps results from close contact between these communities, or may represent a distinct decorative tradition, which, for whatever reason, was associated with monastic communities. Further weight is added to this argument due to distinctive parallels of repoussé bossed pottery at Sandtun and Minster-in-Sheppey, and it appears that this decorative technique was confined to settlements with monastic associations in east Kent (Blackmore 2001b, p. 222). While there are a handful of sherds with bossing at Hamwic and at a kiln site in Ipswich (Blackmore 2001, p. 222), there are no obvious direct extra- regional parallels for its development, suggesting it is a local, south-east Kentish phenomenon which was copied elsewhere. Alternatively, a possible influence is the near-Continent, based on the presence of a boss decorated imported northern Blackware jar at Lundenwic (Blackmore, 1988, pp. 99-100). Although the primary stimulus of the decoration remains unknown, it is rare on post 7th-century pottery and its exclusively to monastic settlements possibly indicates it served a particular function.
Rouletting
Almost all of the 26 rouletted sherds are in northern French Greyware (24 sherds), with the exception of two Blackware sherds, all of which were deposited in post 7th-century features. The most common roulette motif is small triangles (Fig. 4, L); there are also some examples of small squares (Fig. 4, M). The former is paralleled at Sandtun (Blackmore 2001, p. 195) and the latter is paralleled at Hamwic (Timby 1988b, p. 97). The comparisons between vessels at Lyminge and Sandtun emphasise the close relationship between the trading port and Lyminge, and, by extension, between Lyminge and northern France or western Belgium (Blackmore 2001, pp. 193-4). Two jars were identified with rouletting. Although the majority of sherds were not definitive in type, the curved nature of body sherds suggests they derive from rounded jars. Some sherds may be from bottles, drawing on comparisons with similarly decorated vessels deposited in 7th-century Kentish graves (Backhouse 2023); however, no bottle sherds were confidently identified in the assemblage as a whole.
Stamped Decoration
Stamped decoration is rare, being present on fourteen sherds that are primarily Continental imports and non-local regional wares. Stamps are classified according to the Archive of Anglo-Saxon Pottery Stamps (Briscoe 1983). Stamped decorations on locally-produced wares are poorly represented, comprising a clover stamp (Type A7 ‘flower’; Fig. 5, B), a faint triangle stamp, a ring-dot stamp (Type A2, Fig. 5, E), and multiple small flower stamps (Type A7; Fig. 5, F). There is one imported Blackware partial spout and a Blackware body sherd that both have oval lattices (Type D2 ‘Grid ovals’; Fig. 5, C) and one northern French Greyware sherd has a ring-dot stamp (Type A2 ‘multiple circles’). Two unsourced imported wares have small squares stamps (Type C1 ‘plain rectangles’; Fig. 5, A) and small grid ovals (Type D2 ‘grid ovals’; Fig. 5, G). The Ipswich ware pitcher is the only non- local stamped product, here decorated with large oval lattices located in triangular zones, classified as Type D2 ‘Grid Ovals’ (Fig. 5, D). A similarly decorated vessel was identified at Sandtun (Blackmore, 2001b, p. 206).
[pg123][pg124]
[fg]jpg|Fig. 5. Examples of stamp-decorated sherds: A: Unsourced imported ware, B: Sandy ware with Shell, C: Ipswich ware, D: Blackware, E: Well-sorted Sandy ware Group II, F: Ill-sorted Sandy ware, G: Unsourced imported ware. (Drawings by Lesley Collett, photographs, author’s own.) |Image[/fg]
A possible explanation for the low number of stamped local wares is the greater investment of time and resources required to make the tools required for stamping. Because very few pots in general are decorated, it is reasonable to argue that local potters had no desire to invest their time into creating and utilising the tools needed for stamped decoration as they were able to produce popular decorative motifs by employing simpler decorative techniques. Yet, since the rarity of stamps at Lyminge is paralleled at Canterbury (Macpherson-Grant 1995, p. 869), this may suggest stamped decoration was not as culturally and/or socially accepted by communities in Kent. Stamped decoration on locally-produced pottery suggest a 6th- to 7th-century date, corresponding with the height of stamped decoration across early medieval England (Briscoe 1983; Blinkhorn 2012, p. 64). Blinkhorn (2012, p. 66) proposed stamped decoration on Ipswich ware resulted from Frisian influences, based on strong similarities between these industries.
Applied Clay
Of the six sherds with applied clay, four have a single narrow strip, one sherd has a thick, applied strip, and one sherd has an applied pellet. Applied clay is very rare in the assemblage and may have been added for practical rather than decorative purposes, for example, to add grip or to strengthen a vessel’s walls and hold them in shape while drying or during firing.
Thumbed Decoration
There is one Sandy-ware with Shell rim sherd decorated with thumbing, probably from a bowl or dish. The ware suggests an 8th- to 9th-century date.
Surface treatments: scratchmarking and rustication
Twenty sherds have scratch-marked surfaces consisting of multi-directional, thin lines. The depth and quantity of scratchmarking is variable, most noticeably between wares suggesting different tools were utilised. Rustication in the form of thumbnail impressions on the external surface is present on four sherds. Based on ware groups and contextual dating, these techniques occur on mid-6th- to mid- 7th-century sherds. These surface treatments may have been functional, possibly applied to increase a pot’s surface area, thereby increasing its resistance to thermal shock, although this is difficult to prove. It is plausible, however, that they also, at least in part, acted as embellishment.
Use-wear analysis
Skibo (1992) showed that, in addition to the study of vessel type, the study of use-wear on a vessel is the one most effective methods for determining how ceramic vessels were utilised. Two main use-wear categories were examined for the Lyminge assemblage: sooting and surface attrition.
[pg125]
Sooting
Sooting is the dominant type of use-wear, recorded on 431 sherds. Sooting is present on locally-produced pottery, imports, and non-local imports. More than half of the sooted sherds (221 sherds) were identified as jars, indicating their probable use in food preparation and culinary practices, alongside numerous functions that are not identifiable from use-wear evidence. In contrast, the general absence of sooting on bowl sherds suggests their primary function may have been for carrying, preparation or serving, rather than cooking (Levi-Strauss 1965; Hagen 1992).
Sooting is most common on the external surface of body sherds. This suggests vessels were placed at the edge of the hearth and this is likely to be the cause of external sooting on 95 rim sherds. Thirty-two base sherds have external sooting, suggesting they were heated by suspension directly over a fire. Internal sooting on 139 sherds may be caused by the carrying of fire or is perhaps carbonised residue from the burning of a vessel’s contents (i.e., foodstuff) (Skibo 1992, 2015). One jar has sooting on its external surface and white residue on the internal surface, the latter possibly being a mineralised deposit accumulated during heating (Hally 1983).
Surface Attrition
Surface attrition is present on six sherds that are all stratigraphically dated to the 5th to late 6th century. The low number is probably due to the high level of fragmentation and abrasion in the assemblage, making surface attrition difficult to confidently identify. Five of the six sherds with surface attrition are in locally-produced sandy wares and the sixth sherd is in imported northern French Greyware. Two Ill-sorted Sandy ware sherds with attrition have incised decoration, demonstrating that vessels with decoration were utilised in domestic processes (Perry 2013, pp. 64- 85). Two rim sherds have attrition on their internal surface, potentially caused by repeated pouring or stirring of the vessel’s contents. The remaining four sherds are all bases. Possible causes of basal attrition are repeated turning or tipping of vessels. Ethnographic studies, such as Skibo’s (1992, pp. 113-17) work on pottery from the Filipino Kalinga, find that abraders from the surface of the hearth or floor cause basal attrition when a pot is repeatedly placed or dragged across this surface, often during the serving of food. It is possible that the attrition in the Lyminge pottery assemblage may have a similar cause.
Discussion
Local wares at Lyminge correspond with broader characteristics observed across early medieval sites in south-east Kent (Jervis et al. 2016; Backhouse 2021, pp. 275-283). Sandy wares are dominant at Lyminge and, while organic-tempered ware and shell-tempered ware contribute a smaller proportion at Lyminge compared to elsewhere, their presence does follow the general trend observed at contemporary settlements in south-east Kent. Additionally, although found in small quantities, certain wares at Lyminge (e.g., Sandstone-tempered ware and Oolitic Limestone- tempered ware) fall under the category of rock-tempered wares and correspond [pg126]with ware distribution patterns observed in coastal north Kent. The presence of such wares could be related to Lyminge’s role as a royal centre, which attracted commodities from a range of sources across Kent and beyond.
The range of Continental imports at Lyminge reflects the communication networks operating between southern Kent and the near-Continent during the early medieval period. Beyond Kent, pre-7th-century Continental imports are rare, occurring at only a handful of sites in the Lower Thames Valley and coastal sites in Essex, indicating Continental imports of 5th- to 7th-century date form part of a distinct south-east Kent ceramic profile. It is important to recognise the range and quantity of 5th to 7th century Continental imports at Lyminge far exceeds those from elsewhere in Kent. This probably relates to a combination of factors, including Lyminge’s geographical proximity in southern Kent, which was closer to the coastal sites across the Channel on the near-Continent than settlements elsewhere in early medieval England, as well as its importance as a royal central place. By the 7th to 9th centuries, the range of imports represented at Lyminge increases, in line with ceramic trends identified across southern and eastern England during this period. This was triggered by an intensification of long-distance exchange and communication networks operating across the Southern North Sea region (Dunning 1974; Hodges 1982).
Although non-local regional wares are comparatively few at Lyminge, those present show contact between Lyminge and southern East Anglia. The most recognisable non-local import is Ipswich Ware and situates Lyminge within the distribution pattern observed for Ipswich ware at royal settlements and monastic centres along the Kentish coast (Blinkhorn 2012, pp. 91-8), potentially related to these sites being centres of consumption and redistribution.
The range of vessels at Lyminge is largely comparable to that at other excavated early medieval settlements in Kent (e.g., Canterbury (Macpherson-Grant, 1995), Springhead and Northfleet (Mepham and Vince 2011), and the East Kent Access Road excavation (Cotter 2015)), suggesting pottery was used for broadly similar practices across the region. The evidence for vessel types and identifiable formal aspects suggests that pottery had two main uses: for cooking and for storage. The variety of cooking techniques was probably quite limited, with vessels primarily being used for boiling or stewing (Hagen 1992; Thomas 2007; Frantzen 2014). Since cooking techniques and the vessel types used in cooking are inextricable from social and cultural identity, and arguably strongly influenced by habitus (Blinkhorn 1997), the comparatively similar proportions of each vessel type in different phases at Lyminge suggest there was little change in the overall use of pottery throughout Lyminge’s 400-year occupation.
The selection of decorative techniques represented at Lyminge conforms well with wider traditions in contemporary early medieval England, although regional preferences in the popularity of different techniques and motifs can be observed (Backhouse 2021, p. 278). At Lyminge, incised decoration is most common and was likely to have been heavily influenced by decorative systems employed on the near-Continent (Macpherson-Grant 1995, pp. 864, 869; Krol, Dee and Nieuwhof 2020). By contrast, stamped decoration is poorly represented at Lyminge compared to contemporary sites elsewhere in eastern England, for example at Mucking, Essex (Hamerow 1993, pp. 45–51). A possible explanation for the low number of stamped [pg127]local wares at Lyminge is the greater investment of time and resources required to make the tools required for stamping. Approximately 10% of Continental imports and non-local regional wares are decorated, accounting for proportionally double that of decorated local wares. This could suggest that imports embellished with decoration may have been more highly sought-after and perhaps carried more prestige than undecorated pottery.
Lyminge’s pottery assemblage was influenced by numerous interwoven factors, the most significant of which are: the settlement’s high-status, its roles as a royal and monastic central place, its proximity to and relationship with the near-Continent, and its geographically nodal position within the landscape at the intersection of multiple routeways (Maslin 2017). The quantification of the assemblage indicates that Lyminge has a distinctive ceramic profile, defined by the diversity of ware types. Although production and consumption trends of local wares, form, and decoration at Lyminge correlate with contemporary settlement assemblages in south-east Kent, Lyminge’s dimensions as a high-status central place are manifested explicitly in the range and quantity of Continental imported pottery. With the exception of the assemblage at Sandtun, which is known to have been a satellite settlement of Lyminge (Gardiner et al. 2001), there are no direct comparisons in Kent. This diversity of both local and imported wares in the assemblage thus indicates Lyminge’s far-reaching contacts and the gravitational pull of the settlement, related to its high-status identity.
Acknowledgements
This research was funded by the Arts and Humanities Research Council, South, West and Wales Doctoral Training Partnership (AHRC SWW DTP), grant number AH/L503939/1. Acknowledgements are due to Gabor Thomas, Ben Jervis and Duncan Brown who supervised the Ph.D. project that led to this work.
Bibliography
Allen, J.R.L., 1999, ‘The Rumenesea Wall and the early settled landscape of Romney Marsh (Kent)’, Landscape History, 21(1), pp. 5-18.
Backhouse, L.T.D., 2021, ‘Changing Social Relations and the Making of an Early Medieval Kingdom: People and Pottery in Anglo-Saxon Kent, c.5th-9th centuries A.D.’, unpubl. ph.d. thesis, University of Reading.
Backhouse, L.T.D., 2024, ‘Ceramic Grave Goods in 5th-7th Century Kent: A Social Interpretation’, Medieval Ceramics, 42.
Backhouse, L.T.D. (forthcoming) Roman curation, or re-use? A curious assemblage from 5th/6th century Kent.
Barber, L., 2015, ‘Post-Roman Pottery’, in Hicks, A. (ed.), Destined to Serve: Canterbury Christ Church University Excavations 1983-2007. Canterbury: Canterbury Archaeological Trust Ltd, pp. 197-228.
Blackmore, L., 1988, ‘The Pottery’, in Cowie, R., Whytehead, R. and Blackmore, L. (eds), ‘Two Middle Saxon Occupation Sites: Excavations at Jubilee Hall and 21-22 Maiden Lane’, 39th edn, Transactions of the London and Middlesex Archaeological Society, pp. 81-110 (47-164).
Blackmore, L., 2001a, ‘Pottery: Trade and Tradition’, in Hill, J.D. and Cowie, R. (eds), [pg128]Wics: The Early Medieval Trading Centres of Northern Europe, Sheffield: Sheffield Archaeology Monograph 14, pp. 22-42.
Blackmore, L., 2001b, ‘The Pottery in Continental Trade and Non-Urban Ports in Anglo- Saxon England: Excavations at Sandtun, West Hythe, Kent’, Archaeological journal, 158 (1), pp. 191-224.
Blackmore, L., 2003, ‘Specialist appendices: 7.8 The Pottery’, in Middle Saxon London,
Excavations at the Royal Opera House 1989-99, London, MoLAS, pp. 225-241. Blackmore, L. and Vince, A., 2008, ‘Pottery Supply’, in Early and Middle Saxon Rural
Settlement in the London Region. London: MoLAS Monograph 41, pp. 153-56. Blinkhorn, P., 1997, ‘Habitus, Social identity and Anglo-Saxon pottery’, in Blinkhorn, P. and Cumberpatch, C. (eds), Not so much a pot, more a way of life: current approaches to artefact analysis in archaeology, Oxford: Oxbow Books, pp. 113-124.
Blinkhorn, P., 2012, The Ipswich Ware Project. Ceramics, trade and society in Middle Saxon England, The Medieval Pottery Research Group.
Briscoe, T., 1983, ‘A classification of Anglo-Saxon pot stamp motifs and proposed terminology’, Studien zu Sachsenforschung, 4, pp. 57-71.
Cotter, J., 2015, ‘Post-Roman Pottery’, in Andrews, P. et al. (eds), Digging at the Gateway Archaeological landscapes of South Thanet, The Archaeology of East Kent Access (Phase II) Volume 2: The Finds, Environmental and Dating Reports, Oxford and Sailsbury: Oxford Wessex Archaeology, pp. 247-278.
Coutts, C., 1991, Imported Pottery in Middle Saxon England, with particular reference to Ipswich, University of Sheffield.
Cowie, R. et al., 2012, Lundenwic. Excavations in Middle Saxon London, 1987-2000, Dorset: MoLAS.
Dunning, G., 1974, The Trade in Medieval Pottery around the North Sea, Rotterdam Papers: A Contribution to Medieval Archaeology.
Evison, V.I., 1979, Wheel-thrown Pottery in Anglo-Saxon Graves. Leeds: The Royal Archaeological Institute.
Evison, V.I., 1987, Dover: Buckland Anglo-Saxon Cemetery, London: Historic Buildings and Monuments Commissoin for England.
Frantzen, A., 2014, Food, Eating and Identity in Early Medieval England, Woodbridge: Boydell and Brewer.
Gardiner, M. et al., 2001, ‘Continental Trade and Non-Urban Ports in Mid-Anglo-Saxon
England: Excavations at Sandtun, West Hythe, Kent’, Archaeological journal, 158(1), pp. 161-290. doi: 10.1080/00665983.2001.11079010.
Hagen, A., 1992, A Handbook of Anglo-Saxon Food: Processing and Consumption, 2 vols,Hockwld-cum-Wilton: Anglo-Saxon Books.
Hally, D., 1983, ‘Use alteration of pottery vessel surfaces: an important source of evidence for the identification of vessel function’, North American Archaeologist, 4(1), pp. 3-36. Hamerow, H., 1993, Excavations at Mucking. Volume 2: the Anglo-Saxon settlement, London: English Heritage and British Museum Press.
Hodges, R., 1982, Dark Age Economics: Origins of Towns and Trade, A.D. 600-1000, London: Gerald Duckworth & Co. Ltd.
Hodges, R., 1991, ‘The 8th-century pottery industry at La Lande, near Rouen, and its implications for cross-channel trade with Hamwic, Anglo-Saxon Southampton’, Antiquity, 65, pp. 882-7.
Hurst, J.G., 1976, ‘The Pottery’, in Wilson, D.M. (ed.), The Archaeology of Anglo-Saxon England, Cambridge, pp. 283-348.
Jervis, B., 2011, Assessment of Pottery Recovered from Excavations at Lyminge , Kent.
Jervis, B. et al., 2016, ‘Early Anglo-Saxon Pottery in South East England : Recent Work and a Research Framework for the Future’, Medieval Ceramics, 36, pp. 17-26.[pg129]
Klindt-Jensen, O., 1957, Bornholm i folkevandringstiden og forudsaetningerne i tidlig ternalder, Copenhagen: Nationalmuseet.
Krol, T.N., Dee, M. and Nieuwhof, A., 2020, ‘The Chronology of Anglo-Saxon Style Pottery in Radiocarbon Dates: Improving the Typo-Chronology’, Oxford Journal of Archaeology (9999). doi: 10.1111/ojoa.12202.
Leeds, E.T., 1926, ‘A Saxon village at Sutton Courtenary, Berkshire’, Archaeologia, 73, pp. 59-80.
Levi-Strauss, C., 1965, ‘Le triangle culinaire’, Food and History, 2(1), pp. 7-20. Macpherson-Grant, N., 1995, ‘Post-Roman’, in Canterbury: Marlowe Car Park and Surrounding Areas, Part II: The Finds, Canterbury: Canterbury Archaeological Trust, pp. 815-896.
MacPherson-Grant, N., 1993, ‘Early-Late Saxon continental imports in Kent’, in Piton, D. (ed.), La Ceramique du Veme au Xeme Siecle dans l’Europe du nord-Ouest, Arras, Mainman: Nord-Oust Archaelogie hors-serie, pp. 165-187.
Maslin, S.P., 2017, The Landscape and Ecology of the Anglo-Saxon Conversion: a multiproxy environmental and geoarchaeological contextualisation of the high-status settlement at Lyminge, Kent, University of Reading.
Medieval Pottery Research Group, 1998, ‘A Guide to the Classification of Medieval Ceramic Forms: Occasional Paper 1’, p. 246.
Mepham, L. and Vince, A., 2011, ‘Saxon Pottery’, in Settling the Ebbsfleet Valley: High Speed I Excavations at Springhead and Northfleet, Kent. The Late Iron Age, Roman, Saxon and Medieval Landscape. Volume 4: Saxon and Later Finds and Environmental Reports. Oxford and Sailsbury: Oxford Wessex Archaeology, pp. 1-18.
Myres, J.N.L., 1977, A Corpus of Anglo-Saxon Pottery of the Pagan Period, Oxford: Clarendon Press.
Orton, C. and Hughes, M., 2012, Pottery in Archaeology, 2nd edn, Cambridge: CUP. PCRG, SGRP and MPRG, 2016, A Standard for Pottery Studies in Archaeology, London: Historic England.
Perry, G., 2013, United in Death : The Pre-Burial Origins of Anglo-Saxon Cremation Urns, University of Sheffield.
Redknap, M., 1987, Mayenerware and Eifelkeramik: The Roman and Medieval pottery industries of the West German Eifel, University of London.
Richards, J.D., 1987, The significance of Form and Decoration of Anglo-Saxon Cremation Urns, British Archaeological Reports.
Scull, C., 2012, ‘Lyminge 2010: Assessment of Material Culture Assemblages (Copper-Alloy, Bone/Antler and Vessel Glass) from SFBS 1, 2, and 3’, University of Reading, unpubl. report. Skibo, J., 1992, Pottery Function: A Use-Alteration Perspective, New York: Plenum Publishing Corporation.
Skibo, J., 2015, ‘Pottery Use-Alteration Analysis’, in Marreiros, J.M., GibajaBao, J. and Bicho, N. (eds), Use-wear and residue analysis in Archaeology, Springer, pp. 189-198. Stenberger, M., 1955, Vallhager: A Migration Period settlement on Gotland, Sweden, Copenhagen: Munksgaard.
Stilke, H., 1995, ‘Shell-Tempered Ware from the Frisian Coastal Area’, Medieval Ceramics, 19, pp. 11-17.
Thomas, G., 2008, Uncovering an Anglo-Saxon Monastery in Kent: Interim Report on University of Reading Excavations at Lyminge, 2008. Available at: http://www.reading. ac.uk/archaeology/research/Projects/arch_Lyminge.aspx.
Thomas, G., 2013, ‘Life before the Minster: the Social Dynamics of Monastic Foundation at Anglo-Saxon Lyminge, Kent’, The Antiquaries Journal, 93, pp. 109-145. doi: 10.1017/ S0003581513000206. [pg130]
Thomas, G., 2017, ‘Monasteries and places of power in pre-Viking England: trajectories, relationships and interactions’, in Thomas, G. and Knox, A. (eds) Early Medieval Monasticism in the North Sea Zone: Proceedings of a Conference Held to celebrate the Conclusion of the Lyminge Excavations 2008 -2015. Anglo-Saxon Studies in Archaeology and History, Oxford: Oxford University School of Archaeology.
Thomas, G., 2018, ‘Mead-halls of the Oiscingas: A new Kentish perspective on the Anglo- Saxon great hall complex phenomenon’, Medieval Archaeology. Routledge, 62(2), pp. 262-303. doi: 10.1080/00766097.2018.1535386.
Thomas, R., 2007, ‘Maintaining social boundaries through the consumption of food in medieval England’, in Twiss, K. (ed.), The Archaeology of Food and Identity. Carbondale: Center for Archaeological Investigations Occas. Publ. No. 34, pp. 130-151.
Thuillier, F., 2015, ‘L’atelier de potiers carolingien de la Rue de Cassel à Racquinghem (Pas-de-Calais)’, in Thullier, F. and Louis, E. (eds), Tourner autour du pot... les ateliers de potiers médiévaux du Ve au XIIe siècle dans l’espace Européen, Caen: Presses Universitaires de Caen, pp. 123-137.
Timby, J., 1988, ‘The Middle Saxon Pottery’, in Andrews, P. (ed.), The Coins and Pottery from Hamwic, Southampton: Southampton City Museum, pp. 73-122.
Timby, J., 2021, ‘Roman Pottery’, unpubl. report.
Tipper, J., 2009, ‘Pottery’, in Lucy, S., Tipper, J. and Dickens, A. (eds), The Anglo-Saxon Settlement and Cemetery at Bloodmoor Hill, Carlton Colville, Suffolk. Cambridge: Cambridge Archaeological Unit, pp. 202-243.
Tyers, P., 1996, Potsherd: Atlas of Roman Pottery. Late Roman Mayen Ware, Available at: https://potsherd.net/atlas/Ware/MAYN (accessed: 21 January 2021).
Williams, D.F. and Vince, A., 1997, ‘The characterisation and interpretation of early to middle Saxon granitic tempered pottery in England’, Medieval Archaeology, 61, pp. 214- 20.
Wood, R., 2005, The Wooden Bowl, Ammanford: Stobart Davies Ltd.
Wood, R., 2008, ‘What did medieval people eat from?’, Medieval Ceramics, 29, pp. 19-20. Yorke, B., 2017, ‘Queen Balthild’s “monastic policy” and the origins of female religious houses in southern England’, in Thomas, G. and Knox, A. (eds), Early Medieval Monasticism in the North Sea Zone: Proceedings of a conference held to celebrate the conclusion of the Lyminge excavations 2008-15. Oxford: Oxford University School of Archaeology, pp. 7-16.
Young, J., 1996, ‘The Pottery in “Two Middle Saxon Grubenhauser at St Nicholas school, Church Road, Boston”’, Lincolnshire History and Archaeology, 31, pp. 11-12.[pg131]