Medieval Pottery from Romsey: an Overview

Home , Ash glaze, Border ware, Surrey whiteware

Proc. Hampshire Field Club Archaeol. Soc. 67 (pt. II), 2012, 323–346 (Hampshire Studies 2012)

MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW

By Ben Jervis

ABSTRACT is evidence of both prehistoric and Roman occupation, but this paper will deal only with This paper summarises the medieval pottery recovered the medieval archaeology, from the mid-Saxon from excavations undertaken by Test Valley Archae- period to the 16th century. ological Trust in Romsey, Hampshire from the Several excavations took place between the 1970’s–1990’s. A brief synthesis of the archaeology of 1970’s–90’s in the precinct of Romsey Abbey Romsey is presented followed by a dated catalogue of (see Scott 1996). It has been suggested on the pottery types identified, including discussions of the basis of historical evidence and a series fabric, form and wider affinities. The paper concludes of excavated, early, graves that the late Saxon with discussions of the supply of pottery to Romsey in abbey was built on the site of an existing eccle- the medieval period and also considers ceramic use siastical establishment, possibly a minster in the town. church (Collier 1990, 45; Scott 1996, 7). The foundation of the nunnery itself can be dated to the 10th century (Scott 1996, 158), but it INTRODUCTION was evacuated in AD 1001, due to the threat of Danish attack, being re-founded later in The small town of Romsey has been the focus the 11th century. The abbey expanded during of much archaeological excavation over the last the Norman period, with the building of the 30–40 years, but very little has been published choir and nave (Scott 1996, 7). The cloisters (although see Green & Lockyear 1992; Scott and western domestic range were constructed 1996). These excavations have recovered a in the 13th century and additional aisles were large quantity of medieval pottery, the details added in the 15th century (Scott 1996, 9). From of which have not been widely disseminated. the fourteenth century onwards the Abbey This paper synthesises the pottery from several went into decline, (Collier 1990, 50). Much of excavations carried out by the Test Valley the Abbey was demolished in the 16th century, Archaeological Trust, and is a product of the during the dissolution (Scott 1996, 9). authors PhD thesis. The paper begins with an Away from the abbey, the earliest evidence overview of the sites considered, before moving of medieval activity is a series of mid-late Saxon on to discuss the pottery itself. This discussion iron working layers, which may be indicative of consists of a catalogue of the types present, a Romsey’s role as estate centre in the mid-Saxon discussion of its distribution in the town, and period (7th–9th centuries) (Scott unpubl, 8). what this can tell us about the trade in pottery Evidence of iron working was recovered from and its use in medieval Romsey. excavations at Narrow Lane and Creatures Pet Shop. Faunal evidence from the mid-Saxon layers at the Midland Bank site are suggestive of THE HISTORY AND ARCHAEOLOGY OF high status occupation, and may be associated ROMSEY with a minster church or estate centre (Scott 1996, 158). At Bell Street, several late Saxon Romsey lies on the River Test, approximately buildings were excavated. They were associ- 9.5 miles north west of Southampton. There ated with late Saxon pottery and are of typical

323

hants 2012b.indb 323 20/08/2012 10:12:43 324 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

Fig. 1 Plan of archaeological excavations in Romsey

post-built construction (Scott unpubl, 13). the Abbey. At Narrow Lane, Bell Street and Amongst the sites discussed in this paper, very the Creatures Pet Shop site, several layers were few features of secure Anglo-Saxon date will identified dating to the 14th–15th centuries be discussed and therefore only a very broad and probably relate to gardening activity. His- overview of these wares will be presented. torical documents attest to many properties in The bulk of the pottery discussed in this Romsey having gardens (Merrick 1989, 4). At paper comes from deposits dated from the Love Lane a series of features were excavated 12th–14th centuries. The pottery from seven which contained small quantities of pottery, sites is discussed. In addition to the sites generally dating to the 11th–13th centuries. mentioned above, these consist of medieval At 11 The Hundred a length of late Saxon deposits excavated at 11 The Hundred, 15 The ditch was excavated, along with 14 pits and 15 Hundred, Church Street and Newton Lane postholes dating to the medieval period. The Link Road (Fig. 1). The comments on these ditch may have been a boundary feature and sites are based on the excavators site archive contained a very mixed pottery assemblage, reports, and archive pottery reports written by mostly of 10th–12th century date. It was cut the author. These can all be found with the site by two pits which can be dated to the post- archives, deposited with Hampshire County conquest period, on the basis of large pottery Museums Service. The dates for the pottery sherds. At least two of the medieval pits were types mentioned in the overview of these sites cess pits and these form an alignment, which are discussed in detail below. may have marked a property boundary. The Very little later activity was identified at the pottery from these pits is very fragmented and sites discussed above, with the exception of is likely to have been redeposited, with the pits

hants 2012b.indb 324 20/08/2012 10:12:47 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 325

possibly having been emptied, based on the were dug, and a new channel was dug into the fact that some were recut. A line of postholes ‘shitlake’. This is one of a number of streams may relate to a building which fronted onto the running through Romsey and acted as a drain Hundred. The bulk of these pits probably date for the privies for houses in this area (Scott to the 12th–13th centuries, with two having a 1996, 5). The pits at the site were filled in later, 14th–15th century, date on stratigraphic different ways, some principally contain rede- grounds. Much of the pottery in these pits is posited material, whilst others contain more residual, but a number of types are present intact deposits, seemingly dumped quickly, which are indicative of this later date, including directly into the features. Others have mixed coarse border ware and transitional sandy depositional histories. In the 14th–15th wares (see below). The cess layers are likely centuries a boundary ditch was dug along to have been periodically sealed with layers of Newton Lane. Two clusters of pits date to this redeposited waste material, and the pits closed phase. These contain a large quantity of frag- with dumps of similar material. mented, residual pottery sherds and are likely The medieval sequence at 15 The Hundred to have been filled with redeposited material. is less clearly defined and many of the pits had Analysis of the fragmentation of certain wares been cut by post-medieval features. A stretch suggests that this residual material was mixed of Anglo-Saxon ditch was excavated, which with contemporary rubbish, with sherds of may be related to that from 11 The Hundred. 14th–15th century types generally being the This contained a sherd of fine, wheelthrown largest in these features. sandy ware, of possible late Saxon date. In The final large assemblage included in this the southern part of the site, two pits were study is that from Church Street. A series of excavated. The primary fill of one dates to the channel and ditch sections were excavated 12th–13th centuries, with the uppermost layers here, the pottery in which generally dates to having a 14th–15th century date. The other the 11th–12th centuries. A yard, consisting of was filled with redeposited material, perhaps in several layers of oyster shell and earth, as well the 14th century. In the eastern part of the site as a contemporary gravel surface, was identi- two phases of pit digging can be identified on fied. The earliest pits at this site were heavily stratigraphic grounds. The earliest phase would disturbed, but the pottery is suggestive of an seem to date from the 11th–12th centuries, 11th–13th century date. A further series of based on the small quantities of pottery layers, structural features and a well could recovered, which include Wessex coarsewares be dated to the 14th–15th centuries. The and Flint and Sand Tempered wares. The later ceramic evidence suggests that the well was pits seem to date from the 13th–14th centuries, closed with a series of dumps of redeposited based on the presence of Laverstock-type wares, material in the late 14th century. Two pits South Hampshire redware and Wessex redware were filled in the 14th century, with a mixture see below). A series of structural features were of contemporary dumps and earlier, redepos- also excavated, which can tentatively be dated ited material. A single 15th century pit was to the 12th–14th centuries, on the basis of very excavated, which contained large sherds of small quantities of pottery. transitional sandy ware, Wessex redware and The Newton Lane Link Road excavations other late medieval types, including sherds recovered evidence of Prehistoric and Roman of Tudor green and Siegberg stoneware. A occupation, as well as a series of Anglo-Saxon, further well was also filled at around the same medieval and post-medieval features. Only time, principally with dumps of redeposited Roman pottery was excavated from the features material. The presence of cross fits between deemed by the excavator to be of Anglo-Saxon layers suggests that some of this material date. It is likely that in this phase Romsey was came from a common source. of very different character, with the focus of Much of the pottery considered in this occupation perhaps in an area which has not overview was recovered from negative features, been excavated (Frank Green pers. comm.). however the dating must largely remain fairly From the 12th–14th centuries a series of pits vague, or be based on analogies with other

hants 2012b.indb 325 20/08/2012 10:12:47 326 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

assemblages, as many pits were filled with a taken. Several measures were used to overcome mixture of redeposited, residual, material the problems of fragmentation, inherent in and contemporary waste. Waste also found its the study of pottery from urban assemblages way into the many streams running through (see Vince 1987, 202), and to assist in the Romsey. We do however, have excavated assem- answering of particular questions. The data was blages from several different areas of the recorded into an MS access database. A series medieval town, including the Abbey, which will of archive reports on the pottery were written permit us to carry out a study of pottery distri- by the author and have been deposited with bution and use in some detail. Hampshire County Museums Service, along with the data in MS Excel format.

METHODOLOGY AND THE HISTORY OF Terminology POTTERY STUDY Ware names were selected which reflect The pottery from Romsey has been studied by general usage in other studies of pottery in several specialists. The material from the early Hampshire, Wiltshire and Dorset. Generally Abbey sites, Narrow Lane and Bell Street was the names defined by Brown (2002) in relation studied by Dr. Andrew Russel, who produced a to Southampton have been used where appro- very detailed fabric type series and quantified priate. The term Laverstock-type fineware was the pottery by sherd count and sherd weight. adopted following the terminology used by This data has been used in this study, although Wessex Archaeology, whilst the term Wessex many of the fabric divisions have been merged redware has been retained, as this refers to a to create more general ware types. Later, the specific type of pottery identified in Christch- material was the subject of study by Helen Rees urch (Thomson et al. 1983, 53). The term and Charlotte Matthews. Rees has produced transitional sandy ware was preferred to Late unpublished overviews of the Anglo-Saxon and Medieval well fired sandy ware, as this demon- medieval pottery from several sites in Romsey. strates the continuity of some of these wares As Rees acknowledges, the value of these is into the post medieval period. Where new hindered by the over generalisation of certain wares were encountered they have generally ware types, which came about for various been defined by their colour, texture and reasons, not least the absence of published inclusions. parallels from elsewhere in Hampshire. This situation has changed recently, with the publication of assemblages from Southampton THE POTTERY (Brown 2002; Jervis 2009) and the forthcoming publication of some assemblages from Three broad ceramic phases have been Winchester (Holmes & Matthews forthcoming). identified; the Anglo-Saxon period (c.7th– Therefore, it is now possible to place Romsey’s 11th century), the post-conquest period pottery into a better defined regional context, (c.1050–1350) and the later medieval phase and to identify wares in more detail. (c.1350–1500). The Anglo-Saxon period can This re-assessment formed part of a wider be divided into mid-Saxon and late-Saxon sub- study of pottery from across Hampshire, under- phases. The post-conquest period has been taken as part of the authors PhD research defined in relation to the long lived nature of (Jervis forthcoming a). The pottery was divided Wessex coarseware, the principle type in this into ware groups and recorded by form, using phase. On the basis of material recovered from the guidelines defined by the Medieval Pottery excavations elsewhere, more closely defined Research Group (MPRG 1998; MPRG 2001). dating can be defined for some individual Elements of decoration were also noted. The wares. On the basis of this, the period can be pottery was quantified by sherd count, sherd sub-divided into two sub-phases, approximately weight and maximum vessel count. Rim 1050–1250 and 1250–1350. Only pottery from diameter and percentage measurements were medieval features have been considered.

hants 2012b.indb 326 20/08/2012 10:12:47 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 327

Table 1 Quantification of the Anglo-Saxon Pottery

Vessel Form: Jar Bowl Jug/Pitcher Unid. Total Ware Name SC SW SC SW SC SW SC SW SC SW

Mid-Saxon Organic Tempered 1 13 4 43 5 56 Organic Tempered Ware with Flint 1 9 1 9 Flint Tempered Ware 20 320 41 335 61 655 Chalk Tempered Ware 27 249 1 25 27 296 55 570 Flint and Chalk Tempered Ware 3 12 12 80 15 92 Calcite Tempered Ware 1 2 1 2 Crystalline Tempered Ware 3 28 3 28 Michelmersh 10 70 6 35 16 105 Winchester-type ware 12 71 12 85 24 156 Grand Total 61 664 1 25 12 71 107 913 181 1673

Phase 1: the Anglo-Saxon period tempered ware has additional flint inclusions, and may be a transitional type. Very little Anglo-Saxon material present was recovered from the sites examined as part of Calcite tempered ware this study (Table 1). Larger groups have been A single small sherd is in a reduced ware with studied by other authors, for example material soft calcareous inclusions, which are probably from the Waitrose site (Lorraine Mepham calcite. A similar fabric has been identified at pers. comm.) and the Orchard Street Car Hamwic (Timby 1988, 89), so this sherd may be Park (De Rue unpubl.). Only 35 of the 178 of mid-Saxon date. sherds recovered actually came from deposits of Anglo-Saxon date. There are seven wares Phase 1b Mid-Late Saxon present which date to this phase: These wares are typical of late Saxon (9th–11th century) deposits in Winchester (Holmes Phase 1a: Mid Saxon & Matthews forthcoming) and Southampton The following types are paralleled in the mid- (Brown 1994). Saxon settlement of Hamwic (Timby 1988) and are likely to date to the mid-Saxon period Flint-tempered ware (7th–9th centuries). This is the most common Anglo-Saxon type in the assemblage and three fabrics have been Organic-tempered ware identified. Two (fabrics F1 and F2) are unevenly There are five fragments of organic-tempered fired and are characterised by the presence of ware, which, on the basis of parallels with large flint fragments, giving a pimply texture. nearby Anglo-Saxon settlements including These fabrics can be paralleled in Hamwic those around Winchester (Fasham & Whinney (Timby 1988, 88–9) and may date to the earlier 1991, 59) and Hamwic (Timby 1988), typically part of this sub-phase. date from the 6th–8th centuries. The sherds The third fabric (MQargf1) is sandier in are fairly soft and are characterised by the nature and may be related to the later flint and presence of voids, derived from burnt out sand-tempered wares. It is a closer match for organic temper. A further sherd of organic the flint tempered wares recovered from Late

hants 2012b.indb 327 20/08/2012 10:12:47 328 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

Saxon Southampton (Brown 1994), so could & Matthews forthcoming) and is known from be slightly later in date, perhaps the 10th, Southampton (Brown 1994). rather than 8th–9th centuries. Twenty sherds came from jars, three of which were rims, all of Sandy ware with rhomboid impressions simple, everted form (Fig. 2). Three sherds have been identified in a sandy ware with rhomboid impressions. The same Chalk-tempered ware ware has been identified in Winchester and Chalk-tempered ware is a common component Southampton (John Cotter, pers. comm.) of later Saxon assemblages from Romsey in contexts of late Saxon date. The fabric is (Lorraine Mepham pers. comm; de Rue a reduced sandy ware, characterised by the unpubl.). A total of 55 sherds are present in presence of rhomboid voids of regular size the assemblages discussed here. Two fabrics and shape, likely to be imprints from leached are present. Both are coarse sandy wares, dif- selanite crystals. The source of the ware is ferentiated by the quantity of chalk present. unknown, although thin section analysis of Although chalk-tempered wares are known sherds from Winchester demonstrated the from Hamwic (Timby 1988, 80–2), the coarse, presence of glauconite inclusions, suggesting a sandy nature of the fabrics present here is more Greensand (i.e. east Hampshire) source. similar to the late Saxon types known from Winchester (Holmes & Matthews forthcoming) Winchester-type ware and Southampton (Brown 1994). Most sherds There are twenty-four sherds of Winchester- in these fabrics were from jars/cooking pots, type ware, a wheelthrown glazed sandy ware, however a single bowl fragment sherd has been sometimes with slipped decoration under the identified. All are undecorated. Jar rims are not glaze (Biddle & Barclay 1974). It has been abundant, but two forms are present, a simple recovered from excavations in Winchester everted type (common in the flint tempered (Holmes & Matthews forthcoming) and is present ware) and an everted, flanged type, with a in small quantities in Southampton (Brown straight edged profile. A single flat jar base was 1994). It is most abundant at the Abbey and it present in the assemblage. The abundance of is tempting to relate its presence there to eccle- chalk tempered ware in Romsey would imply siastical consumption. Fourteen sherds have that these wares were locally produced. been assigned to jugs/pitchers., which are the most common Winchester-type ware form in Flint and chalk-tempered ware Winchester (Biddle & Barclay 1974, 164). Eleven sherds, in a single fabric, belong to this group, which is likely to be related to chalk- Summary tempered wares (above). The fabric is an The majority of the Anglo-Saxon pottery was unevenly fired, coarse sandy ware, containing locally produced and the fabrics and forms abundant chalk inclusions with occasional flint can be paralleled in assemblages from Win- fragments. Three of the sherds were identified chester (Holmes & Matthews forthcoming) and as coming from jars. Southampton (Brown 1994). Jars are the most common form in the locally produced coars- Michelmersh-type ware (see Mepham & ewares and are typically present with simple, Brown 2007) everted rims, typical of the forms present in late Sixteen sherds of wheelthrown Michelm- Saxon assemblages across Hampshire. Vessels ersh-type ware are present in the assemblage. are typically undecorated, but one sherd of flint Kilns associated with the production of this tempered ware has scratch marked decoration ware and dating to the 10th–11th centuries and another exhibits thumb impressed decora- have been excavated to the north of Romsey. tion. Fourteen sherds of Winchester-type ware Michelmersh-type ware is a common type in are likely to be from jugs/pitchers. In addition larger Anglo-Saxon assemblages in the town to the sherds discussed above, a single sherd (Mepham forthcoming; De Rue unpubl.). The of Portchester-type ware (see Cunliffe 1976) ware is also common in Winchester (Holmes was recovered from an unstratified deposit

hants 2012b.indb 328 20/08/2012 10:12:47 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 329

at Romsey Abbey. The majority of sherds are with rounded (17 sherds) and straight edged small and undiagnostic, largely due to the fact (25 examples;) profiles, as well as handled that the few Anglo-Saxon deposits examined forms (Fig. 2i) Similarly, thickened and everted had been disturbed by later occupation. rims (Fig. 2a; 2b; 2h) are present with both rounded (17 examples) and straight edged Phase 2: the Post-Conquest period (c1050–1350) (one example) profiles. These are likely to be a developed version of the simple form. Two This phase covers a fairly large timespan as examples of a clubbed rim are present (Fig. the phasing some sites did not lend itself to 2d). Single examples of a lid seated form (also finer definition and because some elements known in Southampton, see Brown 1994) (Fig. of the ceramic traditions appear particularly 2j), a straight edged, flanged rim (Fig. 2g) and a long lived. Ceramic sub-phases have been slightly inturned, bevelled rim are also present. defined, on the basis of parallels with other There is no relationship between fabric and assemblages and the co-occurrence of wares rim form. Jars typically have a flat base, often within deposits. A single coarseware type can with a sharply obtuse basal angle. The only be assigned to the Saxo-Norman period (say decoration on these vessels is scratch marking 11th–12th centuries), on the basis that sherds (present on 16 sherds), a typical Saxo-Norman do not commonly occur with later glazed decorative form in this region (Spoerry 1990). sandy wares and similar wares have been iden- Four jars exhibit combed horizontal line deco- tified from features of this date elsewhere in ration. None of the jars are glazed. Hampshire. The remaining coarsewares likely A range of bowl forms are present, the only date from the 12th–14th centuries on the basis rim type to be represented by more than one of their co-occurrence with glazed sandy wares example is a thickened, inturned form (Fig. of known date and their occurrence elsewhere. 2e). Other types present include hammerhead A group of glazed sandy wares probably date to and simple, everted forms. Four sherds in fabric the 13th–14th centuries (see Table 2). FMQ5 are likely to be from tripod pitchers. These are all glazed, with horizontal combed Phase 2a: The Saxo-Norman Period decoration. Such decoration is common on Flint and sand-tempered ware similar Wessex coarseware vessels (see below). These wares are a transitional type, rarely A single jug/pitcher base is present, it is flat, occurring in quantity in deposits with 13th–14th with an obtuse angle and is thumbed. A further century glazed wares, but often being found six sherds of FMQ2 and FMQ5 are glazed, but with Wessex coarseware and later Saxon wares. could be assigned to a specific vessel form. Similar wares are present in the late Saxon and Anglo-Norman assemblage from Southampton Phase 2b: 12th–14th centuries (Brown 1994; 2002). The fabrics are generally The wares present in this phase can be divided reduced and have been differentiated by the into two categories, coarsewares and glazed size and shape of the quartz and flint inclu- sandy wares. The coarsewares consist of sions. Most are likely to be local fabrics, but one three ware groups; Wessex coarsewares, Early fabric (FMQ1) has affinities with Kennet Valley Medieval fine sandy ware and fine flint and wares of the 12th–13th centuries (Vince et al. sand-tempered ware. A range of local and non- 1997). It is characterised by abundant, angular local glazed sandy wares are present including flint fragments. One fabric (FMQ5) is oxidised Laverstock-type wares, South Hampshire and sherds are typically glazed. All are likely to redwares, Dorset and Surrey products and a have been locally produced. small quantity of French imports. Most diagnostic sherds are from jars/ cooking pots, however there a small number Wessex coarseware of fragments from bowls/dishes, and six sherds Wessex coarsewares are the most common ware are from jugs/pitchers. The jars exhibit a range in Romsey as a whole. This is a broad group of of rim forms. Simple everted rims (Figs 2c; 2f) coarse sandy wares with moderately abundant- are the most common. Examples are present abundant quartz temper and occasional rock

hants 2012b.indb 329 20/08/2012 10:12:47 330 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY 9 2 5 28 79 64 23 87 48 329 291 203 216 196 246 SW 9468 3625 1913 2400 1467 1025 36260 57984 3 8 7 3 6 4 1 1 Total 25 27 13 15 67 23 11 29 694 381 219 215 114 SC 3678 5544 97 50 99 95 68 68 79 55 24 48 10 564 375 346 SW 3909 1254 18464 25605 6 9 8 8 8 6 2 6 5 SC 82 62 44 10 12 463 207 Unid. 2777 3715 SW 151 151 1 1 SC Spouted Pitcher Spouted SW 264 264 3 3 SC Dripping Pan SW 343 306 649 Curfew 20 37 57 SC 45 45 SW 1 1 Pitcher Bunghole SC 6 59 14 96 SW 304 1967 2446 6 4 1 1 1 43 56 SC Bowl/ Dish 9 9 2 5 65 47 28 76 92 23 63 SW 874 421 775 241 798 236 1326 2005 7095 8 2 3 6 9 1 3 9 4 1 1 55 24 50 17 53 24 SC Jug/ Pitcher 134 151 555 9 14 36 185 250 104 159 SW 5190 1585 14197 21729 Jar 2 1 4 6 2 17 19 SC 217 779 109 1156

Quantification of the Post-Conquest Pottery Vessel Form: Vessel Name Ware Ware Flint and Sand Tempered Coarseware Wessex Fine Flint and Sand Tempered Ware Fine Sandy Ware Laverstock-type Ware South Hampshire Redware Southampton Sandy Ware Local Pink Sandy Ware Local Whiteware Medieval sandy ware with argillaceous inclusions Whiteware with argillaceous inclusions Iron Rich Sandy Ware Iron Rich Whiteware Medieval Fine Sandy Ware Silty Sandy Ware Surrey Whiteware Dorset whiteware Dorset Red Painted Ware Rouen-type Ware Saintonge Whiteware Saintonge Polychrome North French Whiteware Grand Total Table 2 Table

hants 2012b.indb 330 20/08/2012 10:12:47 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 331

Fig. 2 Flint and sand-tempered wares A: jar with thickened, everted rim; B: jar with thickened, everted rim; C: jar with simple, everted rim; D: jar with clubbed rim; E: bowl with clubbed rim; F: jar with simple, everted rim; G: simple flanged rim; H: jar with thickened, everted rim; I: two handled jar with simple, everted rim; J: jar with lid-seated rim

inclusions, likely to be derived from the sand vessels in this ware (Table 2). There is a great or clay. The group is paralleled in Southamp- deal of variability in the rim forms present, ton (Brown 2002, 10–11), eastern Dorset and which may relate to both the date and source Wiltshire (Brown 2002, 11; Mepham 2000a; of these wares. At Laverstock a basic chro- Mepham 2000b; Mepham 2003; Hurst & Hurst nology of rim forms was proposed, whereby 1967; Musty et al. 1969) and the Test Valley simple everted rims date to the 11th century, (Timby 2005; Matthews 1985). They are less whereas thickened and clubbed forms are common north of Andover, their place being more common in the 13th century (Musty et taken Kennet Valley wares. Several fabrics al. 1969, 100). Unfortunately it has not been were identified within this ware group, princi- possible to test this observation here. Simple, pally based on the size and abundance of the everted rims with a rounded profile (Fig. 3d) quartz inclusions. The coarser fabrics (CQ1 are the most common form (102 sherds), and CQ2) match well with material excavated being over twice as common as simple, everted at Laverstock (Musty et al. 1969), but these are straight edged rims (Fig. 3b) (43 sherds), occa- relatively uncommon in Romsey. sionally decorated with thumb impressions (11 Jars/cooking pots are the most common examples). Flanged rims are also common, with

hants 2012b.indb 331 20/08/2012 10:12:49 332 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

Fig 3 Wessex coarsewares and Medieval fine sandy wares A: Wessex coarseware bowl with hammerhead rim; B: Wessex coarseware jar with simple, everted rim and scratch marked decoration; C: Wessex coarseware bowl with rolled rim; D: Wessex coarseware bowl with simple, everted rim; E: Wessex coarseware spouted pitcher; F: Wessex coarseware clubbed rim; G: Wessex coarseware jar with hammerhead rim; H: Wessex coarseware jar with thickened, everted rim; I: Wessex coarseware tripod pitcher foot; J: Wessex coarseware flanged rim; K: Fine Sandy Ware jar with simple, everted rim; L: Fine Sandy Ware jug base

flanges having both straight edged (Fig. 3j) are rarely decorated. Examples are present (46 sherds) and rounded (18 sherds) profiles. with scratch marking (three sherds) (Fig. 3b), Other rims occur less frequently. There are rouletting (two sherds), thumb impressions several developed versions of the simple rim; a (one sherd) and applied strips (one sherd). beaded everted rim, with a rounded profile (11 Twenty-two jar sherds are glazed, typically with sherds), everted clubbed rims (ten sherds) (Fig. a patchy clear internal glaze. 3f) and thickened (29 sherds) types. Lid-seated Other forms present include bowls, jugs/ rims occur infrequently (six sherds). There are pitchers, dripping pans and curfews. Twenty eight examples of everted, rounded hammer- sherds come from curfews, generally with head rims (Fig. 3g). Jar bases are typically flat, thickened, straight edged rims. Bowls most with an obtuse basal angle, although a small commonly have a hammerhead rim (Fig. 3a), number of sagging bases are also present. Jars either of everted (12 sherds) or inturned (six

hants 2012b.indb 332 20/08/2012 10:12:51 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 333

sherds) type. Thickened rims are the next most interior and exterior (two examples) of the common (Fig. 3h), typically with an upright vessels. profile (eight sherds) although everted (three A range of vessel forms were produced sherds) and inturned (three sherds) forms in this ware. It is likely that the coarse jugs/ also occur. Examples with simple everted (one pitchers were the forerunner to the medieval sherd), inverted (two sherds) and upright (four glazed sandy ware jug traditions described sherds), as well as rolled rims (two sherds) are below (see Brown 1992). Wessex coarsewares also present (Fig. 3c). Five bowl fragments have are the main jar type throughout this ceramic a partial, interior, clear glaze. Other types of phase. Unfortunately the dating of individual decoration are rare; five sherds exhibit thumb deposits is not defined enough to consider how impressed applied strip decoration. Two of the long lived the other forms were, but it is likely, bowls have sockets, where a wooden handle based on parallels with Southampton (Brown may have been inserted (see Jervis 2007). In 2002, 10–11), Winchester (Holmes & Matthews addition to these bowls, three sherds are from forthcoming), and Laverstock (Musty et al. 1962, dripping pans, based on the presence of a 100), that these chiefly belong to the late 11th- pouring lip or spout. early 13th centuries. Fifty-five jug/pitcher sherds are present. Most are likely to be tripod pitchers, typical of Fine sandy ware 12th–13th contexts in Wessex (Brown 1992). There are 381 sherds of fine sandy ware; fabrics Only five rims are present, as simple, everted characterised by the presence of abundant- forms with a rounded (two sherds) or straight- moderately abundant fine sized quartz grains, edged profile (one sherd), and as clubbed with few other inclusions. They can perhaps forms (two sherds). Most of the sherds are be seen as the finer end of the Wessex coars- glazed, usually on the exterior, with some eware tradition. The most common vessel form examples having additional rouletted (five in this ware is the jar. A similar range of rim sherds), combed (two examples), stamped forms are present to the coarsewares. Amongst (two examples) or applied strip (one example) the 12 rims present in the assemblage, the decoration. Single examples of unglazed most common are simple, everted rims (Fig. sherds with combed and painted decoration 3k) (four sherds) and flanged forms (four are also present. These vessels typically have sherds), similar to those present amongst the strap handles, only one example of a rod Wessex coarsewares. Bowls are present with handle is present. The bases are typically flat, simple upright (three sherds) and everted with an obtuse angle. Tripod pitchers have (one sherd) rims. No rims are present from applied, rather than pulled, feet (Fig. 3i). jugs/pitchers, although 24 sherds were attrib- There is a single spout and simple, everted uted to this vessel form, including thumbed rim from a spouted pitcher (Fig. 3e). bases (Fig. 3l), generally based on the presence A further 49 decorated, but unglazed of an exterior clear glaze. Thirteen of these sherds are present. The most common deco- glazed sherds exhibit combed decoration and rative types are scratch marking and combing there are single examples with applied scales (both 19 sherds); followed by applied strips and slipped lines, painted under the glaze. (two sherds), thumb impressed applied strips Only one of the jars is decorated, with thumb (six sherds), thumb impressions (two sherds) impressions. Fine sandy wares of this date are and painting (one sherds). There are 26 not common in Southampton (Brown 2002, glazed sherds from unidentified forms, which 12) but are known from Winchester (Holmes exhibit further decoration. Combing is most & Matthews forthcoming). common (11 examples), followed by applied strip decoration (ten sherds) and rouletting Fine flint and sand-tempered ware (five sherds). A further 151 sherds are glazed, The final group of early medieval coarsewares are but have no further decoration. The glaze the fine flint and sand-tempered wares. These is usually clear and applied patchily to the are quite distinct from the Saxo-Norman flint interior (53 sherds), exterior (96 sherds) or and sand-tempered wares, being wheelthrown

hants 2012b.indb 333 20/08/2012 10:12:51 334 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

and better fired. They can be distinguished by sharply obtuse or right angle basal angle, with their sandy texture and the presence of fine thumbed decoration. Two strap handles are fragments of (generally patinated) flint. Only 19 present. Vessels typically have a partial clear sherds could be assigned to specific vessel forms. or bright green glaze. There are some more Although 17 were from jars, these only represent elaborately decorated examples present; three two vessels. One has a simple, everted rim form sherds are combed and three have slip painted and is scratch marked. The other vessel has a lines under the glaze. These are amongst the flanged rim and is undecorated. Two jug sherds most common forms of decoration on material were identified , both have a clear, exterior from the Laverstock kilns (Musty et al. 1962). glaze. Similar wares are present amongst the Noticeably the elaborately decorated and assemblages from Foxcotte (Matthews 1985, anthropomorphic forms were absent from the 168) and Kings Somborne (Timby 2004, 152), material analysed. so they may be a northern Test Valley or Kennet Valley type. South Hampshire redware The next most common glazed sandy wares Medieval sandy wares are South Hampshire redwares. These date to There are a wide variety of glazed sandy the 13th–14th centuries and are common in wares present in the assemblage. The most Southampton (Brown 2002) and Winchester common are Laverstock-type fineware and (Holmes & Matthews forthcoming). They have South Hampshire redware, but a wide range also been recovered from excavations at Car- of sources are represented in varying quanti- isbrooke Castle (Mepham 2000a), Fareham ties; including the Southampton area and the (Brown unpubl.) and Portsmouth (Fox & Surrey/Hampshire border. All probably date Barton 1986). They are only present in very to the 13th–14th centuries. small quantities west of the New Forest and their presence in Romsey would suggest supply Laverstock-type fine ware from the south and east, as well as from the west These wares, produced at Laverstock near (as represented by the Laverstock type wares). Salisbury (Musty et al. 1962), between the The fabric is iron rich, pinkish in colour with 12th–14th centuries are abundant in Romsey. common quartz inclusions. They are also the most common glazed sandy As with the Laverstock-type wares, the vast ware recovered from excavations in Salisbury majority of vessels present are jugs, with a small (Mepham 2000b) and are found further south quantity of sherds being from bowls/dishes in Fordingbridge (Mepham 2003) and Christch- and jars/cooking pots. Amongst the jugs, urch (Barton et al. 1983) and further north there are two rim forms present; an inverted, around Andover (Matthews unpubl; 1985). thickened rim with a rounded profile (two They are present in small quantities at South- examples) (Fig. 4b) and a simple everted ampton (Brown 2002), Winchester (Holmes form (two examples). The inturned form is & Matthews forthcoming) and Newbury (Vince likely to be a slender, baluster form. The jugs et al. 1997). The fabric is a fine whiteware. typically have flat or slightly sagging thumbed Five fabrics have been identified, based on bases, with a sharply obtuse or right angle basal the sandiness of the matrix and the size and angle. A single example is slightly sagging abundance of the quartz inclusions. with applied feet. As with the Laverstock-type The majority of vessels present in these wares, most of the sherds exhibit only exterior fabrics are jugs, a small number of sherds may glaze, which is typically clear or dark green be from jars/cooking pots or bowls/dishes. Jug in colour. Seventeen sherds (probably from rims are surprisingly scarce. Both thickened a single vessel) exhibit rouletted decoration (e.g. Musty et al. 1962, no. 64) and clubbed (Fig. 3a). Other types of decoration present (e.g. Musty et al. 1962, no. 67) forms are rep- consist of combing (four examples) applied resented. No collared rims, typical of the most pellets (three sherds) or scales (one sherd), highly decorated Laverstock types were identi- stamping (two examples) (Fig. 3b) and slipped fied. The jugs generally have a flat base with a lines under the glaze (11 sherds).

hants 2012b.indb 334 20/08/2012 10:12:51 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 335

Southampton sandy ware Medieval sandy ware with argillaceous Southampton sandy ware is a coarser sandy inclusions ware, present in 13th–14th century contexts in There are 13 sherds in two sandy ware fabrics, Southampton (Brown 2002, 14). It is charac- which have common argillaceous inclusions. terised by orange surfaces and a grey core, and Only four sherds could be assigned to a specific abundant quartz inclusions, with occasional vessel form, being from undecorated jars/ pieces of red iron ore or iron rich clay pellets. cooking pots with simple, everted rims. A single It has also been identified in Nursling. sherd from an unidentified form is glazed. In Southampton, jugs in this fabric appear to have had a relatively short life, often demon- Medieval whiteware with argillaceous strating different attrition patterns to the more inclusions highly decorated South Hampshire redwares A single whiteware fabric is present, which and imported wares (Jervis forthcoming a). They contains iron rich clay pellets. There are 15 perhaps had a function in food/drink prepa- sherds present, mostly from jugs, with a single ration and transport, rather than serving. In sherd being from a bunghole pitcher. Four of Romsey the most common form is the jug (50 the jug sherds have an exterior clear glaze, one sherds), followed by jars (19 sherds, represent- of which has slipped lines under the glaze. ing three vessels), with a single bowl. Sherds are rarely decorated (one sherd exhibits thumb Silty sandy ware impressions) and are sparsely glazed. Seven sherds are present in an exceptionally fine, pinkish sandy ware, with abundant, fine Local pink sandy ware quartz inclusions. A single sherd has been This ware is related to South Hampshire identified as coming from a jug, the remaining redware and is known from Southampton sherds are undiagnostic. They have a clear (Brown 2002, 15) and Winchester (Holmes & exterior glaze, but no further decoration. Matthews forthcoming). Only three sherds could be positively assigned to this group in Romsey, Iron rich sandy ware two of which are from jugs. Three iron rich sandy ware fabrics are present (67 sherds). The fabrics are all oxidised, with Local whiteware common inclusions of black iron ore. Most This ware is known from Southampton (Brown of the sherds have been identified as coming 2002, 16) and Winchester (Holmes Holmes from jugs. The jug bases are typically flat, with & Matthews forthcoming). It is a wheelthrown a sharply obtuse angle and are not thumbed. whiteware with common black iron ore inclu- Iron rich sandy wares are a common feature sions. Jugs are the only vessel form present, of medieval assemblages in west Hampshire often with a partial, bright-dark green, exterior (e.g. Brown 2002; Matthews 1985; Timby glaze. All of the rims present are simple or 2004). These jugs generally exhibit a bright thickened, everted forms, with a rounded green, external glaze, but exhibit no other profile. Bases are flat, with an obtuse basal decoration. angle and are thumbed (see Brown 2002, Fig. 15). A single rod handle is present. Two sherds Iron rich whiteware exhibit combed decoration and two have There are 23 sherds of an unprovenanced applied pellets. iron rich whiteware. The fabric fits into a Hampshire tradition of iron rich whitewares, Medieval fine sandy ware including Local whiteware (see above) and There are eight undiagnostic sherds, typically Southampton whiteware (Brown 2002). Of with a clear, external glaze, in a wheelthrown, the 23 sherds, nine are from jugs and two are fine, oxidised sandy ware, with occasional fine from undecorated jars/cooking pots. Four quartz inclusions. Sherds were principally sherds (one definitely from a jug) exhibit a recovered from features of 13th–14th century clear (three sherds) or dark green (one sherd) date. exterior glaze.

hants 2012b.indb 335 20/08/2012 10:12:51 336 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

Surrey whitewares at sites in the Test Valley, and were likely Three sherds of Surrey whitewares, including produced close to Romsey. A small number of coarse border ware (see Pearce & Vince 1988, vessels came from further north, being typical 9) are present. Other sherds are present in Kennet Valley types. The presence of South finer whitewares of Kingston/Cheam type. Hampshire redware and local pink sandy ware There are no diagnostic sherds present. indicates some pottery was acquired from the south and east of Romsey, and this may also be Dorset wares reflected in the presence of French imports, Six sherds of Dorset red painted ware and 11 of likely sourced through Southampton. Laver- Dorset whiteware are present. Dorset whiteware stock-type ware is the most common glazed is a fine whiteware with common iron stained sandy ware in Romsey and was sourced from quartz inclusions. It has been identified in the west. Small quantities of Dorset wares also Southampton (Brown 2002, 16), Poole (Barton came from this area. The Hampshire/Surrey et al. 1992) and Christchurch (Jervis forthcoming border was not a major source of pottery in b), and is likely to be a product of the Poole this period, the small number of sherds may Harbour area. Vessels often have dark brown have reached Romsey through means other slip trailed decoration under a yellow glaze. than sustained trade. Other wares cannot be All of the Romsey sherds would appear to be accurately provenanced, but likely represent a from jugs. Dorset red painted ware belongs range of sources in Hampshire, east Wiltshire to a tradition of white quartz tempered sandy and, potentially, Berkshire. There do appear to wares, possibly produced in the Christchurch be two classes of sandy ware, those which only area (Brown 2002, 16). The sherds present supplied jugs in any quantity and those which here are unglazed and have red painted lines provided a range of vessels including jars, jugs on the exterior surface. In Christchurch these and bowls. The differences in decoration and are believed to date from 13th–14th century glazing between these two classes may be indic- and this is corroborated by the dating from ative of functional differences between these Southampton. wares.

Imported wares Phase 3 (c.1350–1500) Medieval imported wares are not common in Romsey. The most common type is Saintonge As in Southampton (Brown 2002, 18), the late whiteware, a green glazed whiteware produced medieval period sees a shift in the pottery used near Bordeaux, and common in 13th–14th in Romsey. There are a smaller range of wares, century assemblages in Southampton (Brown in which a wider range of forms were produced. 2002, 26–7). One sherd of highly decorated Four Late Medieval sandy wares have been Saintonge polychrome ware is also present, defined, along with a number of transitional and further sherds are present in unstrati- sandy ware fabrics, which equate to Brown’s fied deposits related to the Abbey. There are (2002, 19) Late Well Fired Sandy Ware group. four sherds of Rouen-type ware, a buff bodied A range of imported wares are also present, sandy ware, typically with a yellow glaze and principally Rhenish stonewares, but also wares dating to the 12th–13th centuries (Brown from Spain and the Netherlands (Table 3). 2002, 23), as well as a further sherd of a North French whiteware of uncertain date. All of Wessex redware these imported wares are present in the form This is a wheelthrown sandy ware with oxidised of jugs. A further sherd of 11th–13th century surfaces and a reduced core, which has been Normandy gritty ware was recovered from an tempered with sand. The term Wessex redware unstratified context. was defined in relation to material from Christchurch (Thomson et al. 1983) and the Summary terminology has previously been transferred to The coarsewares present in phase 2 are all locally Romsey (Rees unpubl.). For this reason it has sourced. Wessex coarsewares are common been retained here. This is the most common

hants 2012b.indb 336 20/08/2012 10:12:51 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 337 6 23 85 15 15 12 250 195 506 SW 8629 19817 10081 Total 3 4 2 1 2 2 66 22 34 SC 570 695 1401 2 6 85 15 12 92 SW 116 129 9453 4811 4185 Unid. 1 4 2 2 2 SC 47 17 12 976 398 491 7 7 SW 1 1 Plate SC 14 14 SW 1 1 Albarello SC SW 150 150 Pipkin 1 1 SC 2 2 SW Dish 1 1 Chafing SC 6 21 15 SW 4 1 3 SC Cup/ Mug 42 SW 1137 1095 Pitcher 1 Bunghole SC 25 24 8 SW 805 407 390 7 1 SC 25 17 Bowl/ Dish 72 SW 126 297 4523 1824 2204 Jug 3 SC 15 74 18 93 203 31 72 SW 3705 1794 1808 Jar 2 3 SC 66 93 164 Quantification of the Late Medieval Pottery Grand Total Maiolica Spanish Coarseware Low Countries Redware Westerwald Stoneware Westerwald Raeren-type Stoneware Siegberg Stoneware Tudor Green Tudor Transitional Sandy Ware Transitional Late Medieval Sandy Ware Late Medieval Organic Sandy Ware Tempered Vessel Form: Vessel Name Ware Redware Wessex Table 3 Table

hants 2012b.indb 337 20/08/2012 10:12:51 338 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

Fig. 4 South Hampshire Redware and Wessex redware A: South Hampshire redware jug with rouletted decoration; B: South Hampshire redware jug with inturned rim and stamped decoration; C: Wessex redware jar with thickened, everted rim; D: Wessex redware jug; E: Wessex redware jar with hammerhead rim; F: Wessex redware jar with flanged rim; G: Wessex redware jar with simple, everted rim; H: Wessex redware jar with flanged rim with thumbed decoration

late medieval type in Romsey. The ware seems that these wares are a development of this to be introduced in the latter part of the 14th earlier type. century, and continues in use until the end of Wessex redware jars are rarely decorated. the 15th. There are single, unglazed examples with A wide range of forms are present in this combed, scratch marked and thumbed applied ware, jars and jugs (often with thumbed bases) strip decoration. There are six examples of are the most common, but bowls and bunghole glazed jars, which have no other form of deco- pitchers are also present. Single Examples of ration. The glaze is generally clear and applied clubbed and simple, everted jug rims are in a patchy fashion to the exterior of the vessel. present (Fig. 4d). Strap and rod handles both Jugs are more commonly decorated. There are occur. There is more variety in the jar forms. five examples of unglazed, slashed strap handles Simple, everted rims with rounded (17 sherds) and a single rod handle has thumb impressed or straight edged (eight sherds) profiles (Fig. decoration. There are nine glazed jug sherds, 4g) are most common, but thickened (five with no further decoration. A further 15 glazed sherds) (Fig. 4c), lid seated (one sherd), ham- examples are also decorated; with combed dec- merhead (six sherds) (Fig. 4e), flanged (five oration (11 sherds), applied strip (one sherd), sherds) (Fig. 4h; 4g), collared (one sherd) and underglaze slipped lines (one sherd), roulet- beaded (one sherd) examples are also present. ting (one sherd) and slashing (one sherd) Like the Wessex coarsewares, the bowls often occurring. Like the jars, the glaze is generally have hammerhead rims, perhaps suggesting clear and applied in a patchy fashion to the

hants 2012b.indb 338 20/08/2012 10:12:53 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 339

exterior of vessels. Bowls are rarely decorated. common type. Brown (2002, 19) has previously Three unglazed examples are combed. noted that there is a degree of standardisation amongst these wares in Southampton, both Late Medieval organic-tempered sandy ware within a particular vessel form and also between There are 34 sherds of this ware, paralleled in standard components used in the production Southampton (Brown 2002, 18) and Winches- of different forms. The evidence from Romsey ter (Holmes & Matthews forthcoming), where does not refute this suggestion. As with the they are relatively short lived types, dated to Wessex redwares, both rod and strap handles the late 14th century. Jugs/pitchers are the are present. Vessels typically have flat bases with most common form, but as in Southampton, an obtuse basal angle, although some jug bases jars and bowls are also present. Sherds are continue to be thumbed. Thirty-one examples generally undecorated; five sherds are glazed, of unglazed jars exhibit applied strip decora- and one jug exhibits slipped lines under a clear tion, whilst a further six sherds are glazed with glaze. no further decoration. The glaze is generally clear and applied to the interior of the vessels. Late Medieval sandy ware Jugs are not commonly decorated, beyond a This group includes 22 sherds in four fabrics, patchy exterior glaze. Decorated examples are which date to the late medieval period, but do present with combed (three sherds), thumbed not fit into the transitional sandy ware group (one sherd) and slipped lines (one sherd). as they are not well fired. Only five sherds could be assigned to specific vessel forms; jars Tudor green and jugs. One of the jug sherds is glazed. Of Sixty-six sherds of Tudor green ware, a thin the sherds which could not be assigned to a walled whiteware with a dark green glaze, specific vessel form, one unglazed sherd has are present. Sherds are generally small and combed, wavy line decoration and three have undiagnostic, although a small number were a partial, interior glaze with no other decora- identified as coming from mugs, jugs and a tion. Similar plain sandy wares of late medieval chafing dish. These wares were produced at date are known from Southampton (Brown the Surrey/Hampshire border and date from 2002, 20). the late 14th century (Pearce & Vince 1988, 17) and are a continued presence in early post- Transitional sandy ware medieval deposits in Romsey. There are 570 sherds of transitional sandy ware present; defined as being fine, well fired Imported wares sandy wares, dating to the transition between Rhenish stonewares are the most common the late medieval and post medieval traditions imported wares. There are two sherds of in Hampshire. Similar wares have been identi- Siegberg stoneware, a light grey, thin walled fied in Southampton (Brown 2002), Fareham fabric, commonly with an ash glaze (Brown (Brown unpubl), Winchester (Holmes & 2002, 35). It is a relatively early Rhenish Matthews forthcoming) and Alton (Blinkhorn stoneware type, dating to the 16th century. It 2007). Jars and jugs are most common forms, is not common in Southampton and this is also but bowls/dishes, bunghole pitchers and reflected at Romsey. Two sherds of Raeren- pipkins also occur. The presence of tripod type stoneware are present. This is the most pipkins, which were produced in post medieval common Rhenish Stoneware in Southamp- types such as Border ware and Verwood, may ton (Brown 2002, 35), and is common in post be indicative of a shift in pottery function as medieval deposits in Romsey. It has a dark grey well as pottery production. fabric and is salt glazed, giving a bright brown The jugs generally have thickened, everted or grey finish. The term Raeren-type has been rims (ten sherds), although an upright example adopted to account for the fact that similar is also present. There is little variation amongst wares were produced at other centres in the the jar rim forms, with simple, rounded, Rhineland. everted rims (eight sherds) being the most Two sherds of Low Countries redware were

hants 2012b.indb 339 20/08/2012 10:12:53 340 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

recovered from the Abbey excavations. It has can be said about its distribution through an iron rich orange-red sandy fabric with a clear the town. It is interesting to note that sandy (orange) internal glaze. This is a common type ware with rhomboid impressions was found at in Southampton (Brown 2002, 32–3), and its several sites however, perhaps suggesting that presence in Romsey may be indicative of trade this was marketed as a container. The distri- with Southampton. bution of Winchester-type ware is focussed Spanish types are represented by four sherds on the Abbey, perhaps indicating that this of Seville-type Spanish coarseware, probably all ware wasn’t widely marketed, but was sourced from olive jars. This is a coarse buff fabric which through institutional links. It was also present is micaceous with abundant quartz inclusions. at Bell Street, likely to have been a major street This type is common in Southampton and in the early medieval period (Frank Green, also occurs at other inland sites in Hampshire. pers. comm.). The presence of these wares is suggestive of The bulk of the pottery used in 11th–14th the re-distribution of products traded from century Romsey was locally sourced and Iberia, such as olive oil, through Southampton. largely consists of Wessex coarsewares (Table Iberian redwares were also recovered, but from 4). These and the flint and sand-tempered post medieval deposits. wares are present at all of the sites inves- Three very small sherds of Maiolica were tigated. South Hampshire redware is the recovered from the excavations at the Abbey, most commonly used glazed sandy ware in one is from an albarello (straight sided jar) domestic contexts in Romsey. At the Abbey, and one from a plate. It has not been possible Laverstock-type ware is most common. Both to source these. Maiolica from the Nether- types are present across the town, suggesting lands and Italy is common in some households that they were widely marketed, but it can be in Southampton, and is generally focussed in suggested that the Abbey was supplied through wealthier tenements (Jervis 2009a; Guttierez a different mechanism, perhaps directly from 2000). Guttierez also identified maiolica from the potters at Laverstock. This supply route excavations at Church Street, an assemblage may be related to the fact that the Abbey held not studied here as it is principally of post lands in Wiltshire, and therefore sourced their medieval date. pottery from a workshop which was centrally located in relation to its properties. Other Summary locally produced sandy wares are present The late medieval period is typified by a limited only in small quantities, so little can be read range of, generally locally produced, sparsely into their distribution, other than to suggest decorated sandy wares. There are a small that they may have been marketed intermit- quantity of imports present, which are typical tently in Romsey, or that people may have of those found at inland sites in Hampshire, sourced pottery from other markets, perhaps with the widely traded Rhenish stonewares in Southampton, Winchester or Salisbury. The being most common. The Spanish wares are distribution of Dorset wares is limited to three all containers for other products and probably sites, perhaps suggesting some link, commer- reached Romsey through Southampton. The cial or tenurial, between these tenements and Maiolica and Low Countries redware sherds Poole or Christchurch, a link supported by the from the Abbey are distinctive and may be rep- presence of south Hampshire types in small resentative of the movement of pottery through quantities in Christchurch (Jervis forthcoming trade with this institution. b). Imported wares are not common in Romsey and are most abundant at the Abbey, where they may have been supplied with wine, for THE SUPPLY OF POTTERY TO ROMSEY liturgical use. The range of wares present at 11 The Hundred perhaps suggests some link with The majority of the Anglo-Saxon pottery was Southampton merchants. It is noticeable that produced locally and is present only in small the types present; Saintonge polychrome ware quantities, meaning that little meaningful and North French whiteware, are not widely

hants 2012b.indb 340 20/08/2012 10:12:53 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 341

Table 4 Distribution of Post-Conquest Pottery in Romsey

Ware Name 11 The 15 The Church Newton Abbey Grand Total Hundred Hundred Street Lane Link

Flint and Sand Tempered Ware 11% 31% 7% 22% 8% 16% Wessex Coarseware 73% 56% 72% 55% 51% 63% Fine Flint and Sand Tempered Ware 0% 2% 2% 1% Fine Sandy 3% 6% 5% 6% 16% 6% Laverstock-type Ware 2% 3% 2% 5% 6% 3% South Hampshire Redware 4% 2% 7% 7% 2% 4% Southampton Sandy Ware 1% 0% 1% 1% 11% 2% Local Pink Sandy Ware 0% 0% Local Whiteware 1% 1% 0% 1% 1% Medieval sandy ware with argilla- 0% 1% 0% 0% 0% ceous inclusions Whiteware with argillaceous 0% 1% 0% 1% 0% inclusions Iron Rich Sandy Ware 4% 1% 0% 2% 2% Iron Rich Whiteware 0% 0% 1% 0% Fine Sandy Ware 0% 0% 0% 0% 0% Silty Sandy Ware 0% 0% Surrey Whiteware 0% 0% 0% 0% Dorset whiteware 0% 0% 1% 0% Dorset Red Painted Ware 0% 0% 0% Rouen-type Ware 0% 0% Saintonge Whiteware 0% 0% 0% 2% 0% Saintonge Polychrome 0% 0% North French Whiteware 0% 0% Grand Total 17141 12643 9068 7985 8675 55512

distributed in Southampton, so these wares were principally recovered from dissolution are not likely to have been sourced directly deposits, probably because this type of pottery through the market place in Southampton. was being used at the turn of the 16th century. Virtually all of the pottery used in late The occupants of Church Street appear to have medieval Romsey was sourced locally (Table been particularly well connected, with their 5). Wessex redwares are a common presence assemblage containing sherds of imported at all sites, and along with the small quanti- pottery, both from the continent, perhaps ties of Late Medieval organic-tempered sandy acquired through Southampton, and from ware, probably represent the earliest late the Surrey/Hampshire border, possibly also medieval types. Transitional sandy wares were acquired through the market in Southampton, also widely used. These post-date the Wessex or through trade with east Hampshire. At least redwares and this is perhaps illustrated by their one house on this street was the property of a general absence from the Abbey. Here they rural manor and others may have been asso-

hants 2012b.indb 341 20/08/2012 10:12:53 342 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

Table 5 Distribution of late medieval pottery in Romsey

Ware Name 11 The 15 The Church Street Newton Lane Abbey Grand Total Hundred Hundred Link Wessex Redware 81% 84% 20% 33% 84% 45% Late Medieval Organic 4% 1% 9% 1% Tempered Sandy Ware Late Medieval Sandy Ware 2% 0% 0% 6% 1% Transitional Sandy Ware 12% 15% 78% 50% 0% 51% Tudor Green 1% 0% 0% 9% 1% Siegberg Stoneware 0% 0% Raeren-type Stoneware 0% 0% Westerwald Stoneware 0% 0% Low Countries Redware 0% 0% Spanish Coarseware 5% 0% Maiolica 1% 0% Grand Total 3175 2259 10859 835 1798 18926

Table 6 Quantification of Post-Conquest pottery from Romsey by vessel form

Vessel Name 11 The 15 The Church Street Newton Lane Abbey Grand Total Hundred Hundred Link Jar 32% 43% 24% 40% 55% 37% Coarse Jug/Pitcher 1% 5% 1% 3% 2% 2% Glazed Sandy Ware 10% 5% 9% 12% 18% 10% Jug/Pitcher Total Jug/Pitcher 11% 9% 10% 15% 19% 12% Bowl/Dish 6% 3% 8% 3% 2% 5% Bunghole Pitcher 1% 0% Curfew 1% 7% 1% Dripping Pan 4% 0% Spouted Pitcher 2% 0% Unid. 51% 44% 58% 40% 14% 47% Grand Total 17141 12643 9068 7985 7174 54011

ciated with the Abbey. Excavated buildings retained, despite the general perception that it appear to have been major stone built struc- was in decline during this period. tures with garderobes (Frank Green, pers. comm.). The general absence of Tudor Green from domestic tenements in Romsey implies USING POTTERY IN MEDIEVAL ROMSEY that it was not widely marketed. It is noticeable that the bulk of the imported pottery As with the distribution, the low quantity of was recovered from the Abbey, illustrative of Anglo-Saxon pottery, coupled with its frag- the commercial links which this institution mented nature, means that it is difficult to

hants 2012b.indb 342 20/08/2012 10:12:53 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 343

Table 7 Quantification of Late Medieval pottery from Romsey by vessel form

Vessel Name 11 The Hundred 15 The Hundred Church Street Newton Lane Link Abbey Grand Total

Jar 20% 17% 16% 27% 17% 17% Jug/Pitcher 17% 27% 18% 20% 22% 22% Bowl/Dish 4% 1% 4% 1% 4% 4% Bhole Pitcher 10% 6% 6% Chafing Dish 0% 0% Cup/Mug 0% 0% 0% 0% Pipkin 1% 1% 1% Albarello 0% 0% Plate 0% 0% Unid. 59% 55% 50% 52% 50% 50% Grand Total 3175 2259 10859 835 18926 18926

pass meaningful comment on ceramic use in ampton, where jugs account for a third of the this period. Virtually all of the vessel forms assemblage by weight. It should be noted that identified were jars, which likely had a range the high level of fragmentation means that a of functions including as cooking, storage and large number of sherds could not be assigned processing vessels. The presence of Winches- to form. Even when this is taken into account ter-type ware jugs or pitchers at the Abbey may however, the quantity of jugs is lower than even be related to a function in communal dining, in the lowest status tenements in Southampton or in the liturgy. (see Jervis 2009 a). The types of jugs present are The composition of the phase 2 assem- also noticeably different to Southampton, the blages by vessel form is surprisingly uniform vast majority are sparsely decorated, suggest- throughout Romsey. Jars are generally the ing that the occupants were more interested in most common form, typically accounting for the functionality of these vessels, rather than between a third and half of the pottery by their aesthetic value. Clearly the occupants of weight. Coarseware jugs or tripod pitchers Romsey wanted something different from their account for only small proportions of assem- jugs than the wealthiest inhabitants of South- blages, whereas glazed sandy wares are much ampton, who had at least a small number of more common, typically accounting for highly decorated serving vessels. Such vessels between 5%–10% of assemblages. At Church are present in negligible quantities in Romsey Street and the sites in The Hundred, bowls and were perhaps seen more as curiosities than are present in similar quantities to jugs. Bowls as a well defined functional group. The sparsely do not appear to have commonly been used decorated vessels used in Romsey are best par- for serving in the medieval period, so their alleled in Southampton by the group from York presence is often indicative of some processing Buildings, a group of tenements occupied by role, perhaps as measures, or for use in dairying smiths (Jervis 2009 a). Even these consumers (Blinkhorn 1999, 44). Their presence at 11 and used a small quantity of more highly decorated 15 The Hundred, sites at the edge of Romsey, wares however. This perhaps suggests that the is perhaps relevant to this, as the occupants of occupants of Romsey did not live a fully ‘urban’ these sites may have engaged in some agricul- lifestyle. This is further supported by the preva- tural activity. The quantity of jugs present is lence of bowls, which are often largely absent low, even in comparison to small towns such as from urban assemblages. Parallels can perhaps Andover (Matthews unpubl.), let alone South- be drawn with the assemblage from the village

hants 2012b.indb 343 20/08/2012 10:12:53 344 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

of Popham in north Hampshire (Hawkes been living an almost rural lifestyle, based on 1987). Here, forms were quantified by vessel the pottery they used. This is an issue that count, and bowls account for 7% of the assem- needs to be further addressed through study of blage. It is also noticeable that the jugs used the other finds and faunal remains from exca- at this site are generally sparsely decorated vations in the town. and locally sourced, much like the vessels This overview should be regarded as a used in Romsey. We can perhaps suggest then starting point for future research on material that although a small town, the occupants of from Romsey. Now that wares have been Romsey were living something of a semi-rural defined and the range of vessels identified, we domestic life in the 11th–14th centuries, based can begin to compare elements of these assem- on the ceramic evidence. blages with one another, as well as attempt At all of the sites, around half of the late to place Romsey into a wider context. This medieval pottery (by weight) was too fragmented process has already begun in this paper, but as to assign to a particular form, largely due to new material is excavated, both in Romsey and the fact that vessels often shared standardised in other settlements in the region, and other component parts (see above). In functional finds from the town are analysed, further inter- terms, the assemblages are fairly similar, with jars pretation can be attempted. This paper has and jugs being the most common forms. Bowls focussed only on the medieval pottery and the are less common than in phase 2, although this large post medieval assemblages are equally could be due to fragmentation. Vessels fulfill- worthy of study. Fragmentation is clearly an ing new functions emerge, Tudor Green and issue to be addressed in any analysis of this Rhenish Stoneware drinking vessels, a Tudor material and further attempts need to be made Green chafing dish and highly decorated to integrate the pottery, excavation and other maiolica table vessels, recovered from the finds data to better understand issues of depo- Abbey. Small quantities of drinking vessels were sition and site formation processes. This has identified at several sites in Romsey, illustrat- already begun through the Romsey Rubbish ing the start of a general trend which continues Project (see Green and Lockyear 1992), and into the post-medieval period.. The presence of it is hoped that the clearer definition of the imported pottery, particularly Spanish pottery, pottery provided by this paper will aid such suggests that the Abbey community had access analysis in the future. to a distinctive range of foodstuffs, and it would seem that at least some of this community had adopted new methods of consuming these foods ACKNOWLEDGEMENTS too. We can tentatively see a distinction between the Abbey community, who were engaging in This research derives from my doctoral new table practices and the secular occupants research undertaken at the University of South- of Romsey, who continued to use pottery largely ampton under the supervision of Dr. Andrew as a functional tool, rather than a medium for Jones and funded by the AHRC. Thanks go to adopting new fashions in dining. the staff of Hampshire Museums; Dave Allen, Kay Ainsworth and particularly Alan Jacobs for supporting this research and providing access SUMMARY AND CONCLUSIONS to the relevant site archives. I would also like to express my thanks to Dr. Frank Green, Dr. This paper has demonstrated that much of the Andy Russel and Helen Rees who passed on pottery used in medieval Romsey was locally much valuable insight from their previous produced. The Abbey would appear to have work in Romsey, and commented on drafts of been supplied in a different manner to the rest this paper. Thanks also go to Duncan Brown, of the population, and differences can also be John Cotter, Lorraine Mepham and Yvonne de observed in the way that pottery was used here. Rue for conversations which have influenced When compared to other towns in the region, my thinking. The pottery was drawn by Frances the occupants of Romsey can be seen to have Saxton.

hants 2012b.indb 344 20/08/2012 10:12:53 JERVIS: MEDIEVAL POTTERY FROM ROMSEY: AN OVERVIEW 345

REFERENCES

Barton, K, Cartwright, L, Jarvis, K & Thomson, Hawkes, J 1987 The Pottery, in Fasham, P The R 1992 Catalogue of the Pottery, in Medieval Settlement at Popham, Exca- Horsey, I & Jarvis, K (eds), Excavations vations 1975 and 1983, Proc Hampshire in Poole 1973–1983, (Dorset Natural Fld Club Archaeol Soc 43 83–124. History and Archaeological Society Holmes, K & Matthews, C forthcoming 1500 Years of Monograph Series 10), Dorchester, Pottery from Winchester: excavations 1972– 65–52. 1986, Winchester. Biddle, M & Barclay, K 1974 Winchester Ware, in Hurst, G & Hurst, J 1967 Excavation of Two Moated Evison, V, Hodges, H & Hurst, J (eds), Sites: Milton, Hampshire, and Ashwell, Medieval Pottery from Excavations. Studies Hertfordshire, J Brit Archaeol Ass 3 ser 30 presented to Gerald Clough Dunning, with 48–86. a Bibliography of his Works, London, Jervis, B 2009a For Richer, For Poorer: A synthesis 137–66. and discussion of Medieval Pottery from Blinkhorn, P 1999 The Trials of Being a Utensil: eastern Southampton in the context pottery function at the Medieval hamlet of the High and Late Medieval Towns, of West Cotton, Northamptonshire, Medieval Ceramics 30 73–94. Medieval Ceramics 22–23 37–46. Jervis, B 2009b Pottery from Late Saxon Chichester: Brown, D 1992 A Note on Tripod Pitchers, in Horsey, a reassessment of the evidence, Sussex I & Jarvis, K (eds), Excavations in Poole Archaeol Collect 147 61–76. 1973–1983, (Dorset Natural History Jervis, B forthcoming a Placing Pots: an investigation into and Archaeological Society Monograph the use and perceptions of pottery in Saxon Series 10), Dorchester,102–6. and Medieval Southampton, in its local Brown, D 1994 Pottery in Late Saxon Southampton, and European context (cAD700–1400), Proc Hampshire Fld Club Archaeol Soc 50 unpubl Ph.D. Thesis, University of 153–68. Southampton. Brown, D 2002 Pottery in Medieval Southampton Jervis, B forthcoming b Pottery from Two Medieval c1066–1510, (CBA Research Report Tenements in Christchurch, Dorset, 133), York. Proc Dorset Natur Hist Archaeol Soc. Brown, D unpubl Pottery, in Gifford & Partners, Matthews, C 1985 Pottery, in Russel, A Foxcotte: The Market Quay Fareham. Excavation Report. archaeology and history of a Hampshire Report no. 10677.ROI, unpubl typescript Hamlet, Proc Hampshire Fld Club Archaeol in Hampshire museum. Soc 41 149–224. Collier, C 1990 Romsey Minster in Saxon Times, Matthews, C unpubl 50/50A Chantry Street, Andover Proc Hampshire Fld Club Archaeol Soc 46 (A1982.6) and Winchester Street, Andover 41–52. (A1984.2) The Pottery Report. TVAT Report Cunliffe, B 1976 Excavations at Portchester Castle 23/97, Deposited with Hampshire (Volume II, Saxon), London. Museums Service. De Rue, Y unpubl Pottery from the Orchard Street Car Mepham, L 2000a Pottery, in Young, C (ed.), Exca- Park, Romsey, unpubl report deposited vations at Carisbrooke Castle, Isle of Wight, with Hampshire Museums Service. 1921–1996, (Wessex Archaeology Fasham, P & Whinney, R 1991 Archaeology and the M3, Report 18), Salisbury, 98–131. (Hampshire Field Club Monograph 7), Mepham, L 2000b The Pottery in Rawlings, M Exca- Winchester. vations at Ivy Street and Brown Street, Fox, R & Barton, K 1986 Excavations at Oyster Street, Salisbury, 1994 Wiltshire Archaeol Natur Portsmouth, Hampshire, 1968–71, Post- Hist Mag 93 29–37. Medieval Archaeology 20 31–255. Mepham, L 2003 Pottery, in Harding, P & Light, Green, F & Lockyear K 1992 Plant Remains from A Excavations in Fordingbridge, 1989 Buried Soils at Romsey, Hampshire, and 1997: The Former Albany and Review of Palaeobotany and Palynology 73 Greyhound Hotel Site, Proc Hampshire 57–70. Fld Club Archaeol Soc 58 150–59. Gutierrez, A 2000 Mediterranean Pottery in Wessex Mepham, L & Brown, L 2007 The Broughton to Households (13th–17th Centuries), (BAR Timsbury Pipeline, Part 1: a late Saxon British Series 306), Oxford. pottery kiln and the production centre

hants 2012b.indb 345 20/08/2012 10:12:53 346 HAMPSHIRE FIELD CLUB AND ARCHAEOLOGICAL SOCIETY

at Michelmersh, Proc Hampshire Fld Club Scott, I unpubl The Evidence from Excavation for Archaeol Soc 62 35–68. late Iron Age, Roman and Saxon Occu- Merrick, P 1989 Medieval Romsey as Revealed by pation in Romsey, unpubl typescript Winchester College Muniments, Pots deposited with Hampshire Museums and Papers 4 1–12. Service. MPRG 1998 Medieval Ceramic Forms, (Medieval Spoerry, P 1990 Ceramic Production in Medieval Pottery Research Group Occasional Dorset and the Surrounding Region, Paper 1), London. Medieval Ceramics 14 3–17. MPRG 2001 Minimum Standards For The Processing, Thomson, R, Barton, K & Jarvis, K 1983 The Pottery Recording, Analysis and Publication of Jarvis, K (ed.), Excavations in Christchurch Post-Roman Ceramics, (Medieval Pottery 1969–1980, (Dorset Natural History Research Group Occasional Paper 2), and Archaeological Society Monograph London. Series 5), Dorchester, 53–68. Musty, J, Algar, D & Ewence, P 1969 The Medieval Timby, J 1988 The Middle Saxon Pottery in Andrews, Pottery Kilns at Laverstock, near Salisbury, P (ed.), The Coins and Pottery from Wiltshire, Archaeologia 52 83–150. Hamwic, Southampton. Pearce, J & Vince, A 1988 Surrey Whitewares, Timby, J 2004 Pottery, in Pine, J & Preston, S Early London. Medieval Settlement on Land Adjoining Rees, H unpubl Late Medieval and Early Post- Froman’s, Cow Drove Hill, King’s Medieval Pottery, in Rees, H, Bourdillon, Somborne, Hampshire, Proc Hampshire J, Crosby, V & Scott, I Aspects of Medieval Fld Club Archaeol Soc 59 149–54. and Post-Medieval Occupation in Romsey, Vince, A 1987 The study of pottery from urban TVAT Report 29–92, unpubl typescript excavations, in Schofield, J & Leech, R deposited with Hampshire Museums (eds), Urban archaeology in Britain, (CBA Service. Research Report 61), 201–13. Scott, I 1996 Romsey Abbey. Report on the excavations Vince, A, Lobb, S, Richards, J & Mepham, L 1997 1973–1991, (Hampshire Field Club and Excavations in Newbury, Berkshire 1979– Archaeological Society Monograph 8), 1990, (Wessex Archaeology Report 13), Winchester. Salisbury.

Author: Dr Ben Jervis, Archaeology, School of Humanities, University of Southampton, Highfield, Southampton, SO17 1BJ

hants 2012b.indb 346 20/08/2012 10:12:53