Evidence for Roman Activity in Sholden Probably related to the Villa at Hull Place

87 Evidence for Roman domestic activity in Sholden, probably related to the villa at Hull PLace emma jeffery with contributions from James Newboult, Nuala Woodley, Caroline Norrman, Rob Perrin, David Henderson, Holly Duncan, Laura Bailey and Julie Lochrie In January 2013 an archaeological excavation by Headland Archaeology (UK) Ltd on land to the north-west of Sholden uncovered a series of Roman remains. This included two parallel ditches believed to delimit a trackway, several pits containing substantial quantities of domestic artefactual material, a cremation burial, and a shallow pit with a mysterious chalk slab in its base. These features are believed to be associated with the villa at Hull Place, and may represent part of the villa complex (possibly where rubbish was dumped), or the margins of a separate settlement positioned within the wider villa estate. Excavations carried out by Headland Archaeology (UK) Ltd in 2013 on land just to the north-west of Sholden uncovered a series of Roman remains, believed to be associated with the villa at Hull Place. This work was undertaken as a condition of planning consent in advance of the construction of residential dwellings on the site. This followed the compilation of a desk-based assessment (CgMs 2010) and trial-trench evaluation which revealed the remains of two Roman ditches in the south-eastern part of the development footprint (RPS Clouston 1993). The fieldwork took place between 2-24 January 2013. It consisted of archaeologically supervised stripping of the south-eastern area, followed by mapping, excavation, and recording of the remains uncovered. A watching brief was also carried out in the north-western corner of the site; however, this was discontinued as no remains were revealed. The site lies to the north-west of Sholden, west of Deal (centred at TR 35589 52695). It is broadly rectangular in shape, covering an area of c.7.3ha (Fig. 1). It is bounded by London Road to the west, Sholden New Road to the south, residential dwellings to the east, and agricultural land to the north. The focus for the archaeological work was the south-eastern corner of the site, an area of c.0.85ha. Prior to development the site was open arable farmland, sloping gently down towards the north-east (from c.15m to c.5m aod). It overlies Head Brickearth deposits (clay, silt, sand, and gravel), overlying the Upper Chalk Formation. There is some evidence for prehistoric activity in the vicinity of the site. This eMMA JEFFERY 88 Fig. 1 Site location plan. includes a Bronze Age barrow cemetery c.100m to the south-west (Kent HER: TR35 SW70; TR35 SE108; TR35 SE109; TR35 SE113; TR35 SE114), and a Bronze Age inhumation 800m to the south (Kent HER: TR35 SE9). This part of Kent was heavily occupied in the Roman period, with Richborough 8km to the north-west of Sholden; and Dover, c.12km to the south, becoming ROMAN DOMESTIC ACTIVITY IN SHOLDEN, PROBABLY RELATED TO VILLA 89 a thriving trading town. A road connected the two and ran c.5km to the west of Sholden. Roman activity has been uncovered in excavations at Deal, and it seems highly likely that a network of smaller villas and farmsteads would have developed in the surrounding areas. Of particular interest is the Roman villa at Hull Place, c.300m to the north of the site, which was excavated by Dover Archaeological Group between 2005 and 2007 (Parfitt 2009a; b). The villa consisted of two separate successive dwellings, plus evidence for earlier Iron Age activity. The earliest excavated structure (Building B) was dated to the early second century, with the second structure (Building A) being constructed in the late second /third century. Roman building material has also been identified to the north of the Hull Place excavated areas (Parfitt, pers. comm.), reflecting the probable existence of other structures (bath-houses, barns, outbuildings) which formed part of the villa complex. Other Roman remains have been uncovered in the vicinity of the site, including a number of cremation burials to the south-east (Kent HER: TR35 SE7; TR35 SE39), and Roman pottery c.100m to the north-east (KHER: TR35 SE5). Furthermore, the 1993 evaluation on the development site uncovered the remains of two large parallel Roman ditches, extending south-east from the area of the villa (RPS Clouston 1993). In contrast, there is limited evidence for Saxon, medieval, or post-medieval activity in this area. The site was positioned between the villages of Cottington and Sholden and presumably comprised agricultural/pasture land. No development is known to have taken place in the post-medieval period. Results of the 2013 Investigation The archaeological investigation revealed a series of Roman features, comprising three ditches, five pits containing substantial quantities of domestic material (particularly pottery and metalwork), a cremation burial, and a pit with a chalk slab in its base. This reflects general domestic activity, potentially part of the villa complex, or the margins of a separate settlement positioned within the wider villa estate. Residual lithics were also uncovered, comprising two cores, three retouched pieces, 87 flakes, 28 chunks, and 171 chips. These are dated to the late Neolithic and Bronze Age and indicate that there was some background prehistoric activity in this area, although its form and extent is unclear. Very little evidence for later (post-Roman) activity was observed, aside from one probably modern tree-throw. The impact of post-medieval ploughing could be seen in the disturbance of some of the upper fills of the Roman features and through visible plough marks. Ditches (G1, G2 and G3) Two parallel ditches, orientated north-east to south-west and positioned 8m apart, were excavated (Figs 2 and 3). These were the same as those identified in the 1993 evaluation. They measured c.33m in length, continued beyond the limit of excavation to the north-east, and appeared to end to the south-west (although they eMMA JEFFERY 90 may have petered out or been disturbed by ploughing as no defined cut edge was observed). Although they had different profiles (Fig. 4), their identical alignment and the regular spacing between them suggests that they worked in conjunction with each other. Similarly G3, although narrower, shallower, and shorter than G1 and G2, was also on the same alignment. Relatively few finds were recovered from these features – some Roman pottery, a few fragments of daub, and a few pieces of magnetic residue. The grain assemblage is indicative of low-level processing, storage, and food preparation in the general vicinity, rather than being specifically associated with this feature. This may have formed part of a trackway, with ditches G1 and G2 positioned either side of it to act as drainage gullies, and G3 possibly functioning as a secondary drainage gully. It may have led directly to the villa to the north-east, potentially connecting the villa to the Roman road to the west; or was merely part of a local network of other (unknown) roads and trackways. It may also have functioned as the northern boundary to the settlement or villa estate, as no evidence for Roman activity was discovered north of this. Fig. 2 Plan of Roman features. ROMAN DOMESTIC ACTIVITY IN SHOLDEN, PROBABLY RELATED TO VILLA 91 Fig. 3 Photograph of ditches G1 and G2, looking south-west. Fig. 4 Sections of ditches G1 and G2. eMMA JEFFERY 92 Fig. 5 Sections of pit G4. ROMAN DOMESTIC ACTIVITY IN SHOLDEN, PROBABLY RELATED TO VILLA 93 Pits (G4, G6) Five broadly circular pits, of varying sizes and depths, were excavated in this area. The function of these is unclear although, given their artefactual assemblage, some are likely to have been rubbish pits. Significant quantities of Roman pottery were recovered from these pits, particularly G4 where fragments from 55 vessels were recovered (Fig. 5). These were dated to between ad 150 and 175 and included two amphora fragments (one a handle probably from a Dressel 20 and one a body sherd from a Pelichet 47), several complete or near-complete samian vessels, and grog-tempered vessels (some with wiped surfaces). Many of the vessels can be paralleled in the products of the Thameside kilns (Monagham 1987), and others are likely to have been produced in the Canterbury kilns. Much of the pottery was from table-ware types such as cups, dishes, and bowls, and this fact, combined with the proportions of the different fabrics, suggests that it represents a mixture of utilitarian and higher status domestic activities. The pottery recovered from the other pits was similar in type and form, and included samian ware vesssels, a grog-tempered jar, a small flagon with a grooved disc-type rim, and two lids. Other finds recovered from the pits point towards a domestic origin. These included a first-/second-century copper alloy coin, a small black stone bead which would have been strung around the neck, a bucket side mount, fragments from a ladle, general purpose nails, and a strap fragment. The remains of a young (<18 month old) cat was deposited in pit G4 after death – another indicator of domesticity. However there was also some suggestion of potential workshop activity, including the possible remains of a blade from an agricultural implement such as a reaping hook, and some magnetic residues with pieces of hammerscale (indicating that there was iron smithing taking place in the area, although it may have been blown in from a distance). Nonetheless, the finds recovered from the pits point towards a domestic origin (dating to the mid-later second century), with a mix of both higher status and utilitarian use. Some of these, such as G4 and 1023 (Fig. 2), clearly functioned as rubbish pits, whereas the others may have had a more general use, and simply been backfilled at the end of their lives. Chalk-slab lined pit (G5) A shallow rectangular pit was excavated which contained a chalk slab in its base (2m in length, 1.5m in width, 0.2m in depth). This slab was broken into several pieces, presumably after it had been placed in the pit (Fig. 6). The finds recovered from this pit included grey wares (particularly jars), a rectangular file (similar to one found at Gorhambury Roman villa, St Albans (Wardle 1990, fig. 131, 368-69), six nails, and a few fragments of daub. A date in the second half of the second century for deposition is probable. It is unclear why this chalk slab was positioned at the base of such a shallow pit. It did not seal anything, and yet does not appear to have been a base for a post-built structure as it was positioned on its own (unless it simply supported a single post). eMMA JEFFERY 94 Urned cremation burial (G7) A single urn containing cremated human bone and accompanied by two pottery vessels was uncovered c.20m to the south-west of the core of Roman activity (Fig. 2). The cremated human remains represent the interment of a single adult male, above 35 years of age. The roof of the right orbit showed a slight porosity which may indicate iron deficiency anaemia (Mays 1998, 142-3). The right frontal also displayed an incomplete supraorbital foramen (notch), and several large ossicles were recorded on a fragment of skull – these are both associated with nutritional stress in childhood (Mays 1998, 114-6). There was also an indication that the individual suffered from osteoarthritis of the lower spine and hip. Around 90% of the bone was fully calcined (i.e. heated to remove almost all the organic component of the bone, so that the remains were white in colour). This indicates sustained temperatures of over 600°C (McKinley 2000, 406). The remainder was bluey-grey in colour, indicating a slightly lower temperature. The areas of the skeleton which showed less complete burning suggest that the body was placed near the base of the pyre, with the hands by the sides and the head turned slightly to the right. The total weight of bone recovered is quite low for the cremated remains of Fig. 6 Photograph of the chalk slab in the shallow pit. ROMAN DOMESTIC ACTIVITY IN SHOLDEN, PROBABLY RELATED TO VILLA 95 an adult male, so it seems likely that not all of the remains were interred in this context. The central portion of the skeleton and largest pieces of bone appear to have been gathered first, followed by parts from higher up the body, concentrating on the skull. More marginal areas, such as the hands and feet, were not gathered very assiduously, whilst the less structurally robust ribs and vertebral bodies from the middle of the pyre may have been so fragmented that they were not collected. Of particular interest was the presence of a single faunal bone, fully calcined and highly fragmented. This was the tibiotarsus (the ‘drumstick’ bone) of a bird. It has not been possible to identify the species; however, it was probably from a bird about the size of a crow. This bird was placed in the middle of the pyre, above the body. There are other examples of birds being placed on cremation pyres in the Roman period, such as the individual from Stretton Sugwas (Herefordshire) who was cremated alongside a probable chicken (Mercian Archaeology 2005, 12); and the second century urn excavated at Alington Avenue, near Dorchester, which included cremated bird bones (Cooke 1998, 52). More locally, an early secondcentury cremation at Canterbury castle included a samian dish with the remains of a small bird (Ward 1990). The reasons behind positioning birds alongside human cremations are unclear – they may have been intended as food offerings or had other symbolic connotations, or they may have had more personal connections with the individuals being cremated, as their pets or similar. The cremation was accompanied by a small grooved disc rim flagon and samian plate. The placing of grave-goods, often pottery, alongside Roman cremation burials was a common rite, as is seen in the pottery vessels accompanying the cremations to the south-east of the site (Kent HER: TR35 SE7; TR35 SE39). Similarly, the excavation of the Romano-British cremation cemetery at Ospringe uncovered 1.7 ancillary pottery vessels per grave, and 1.4 ancillary vessels per grave were uncovered at the cemetery at Otford (Ward 1990). The positioning of this cremation alongside a probable trackway and away from the core of settlement was relatively common in the Roman period, stemming from the Roman law that burials must be positioned outside towns (The Laws of the Twelve Tables: Law III, Table X: quoted on http://www.constitution.org/sps/sps01_1.htm). Individual/small groups of cremation burials were dotted around the landscape (as is seen in the other examples of individual or small groups of Roman cremations in this area), close to trackways and at the margins of settlements, rather than in large central cemeteries. Burial was therefore driven by individuals/small family groups burying their dead wherever they saw fit, rather than being part of a more centralised strategy. Finds and Environmental Samples The pottery recovered from the site comprised nearly 1,200 sherds, weighing over 25 kilos, and with a rim EVE of nearly 24. A limited range of fabrics occur, mainly grog-tempered wares, various reduced and oxidised wares, and Central Gaulish samian ware (CGS), together with some sherds of South Spanish amphorae, a Lower Rhineland (LRCC) beaker, and possible Colchester or Kent mortaria. A variety of types of pottery vessels were recovered, all indicating domestic activity (both utilitarian and higher-status). These included jars (of varying types eMMA JEFFERY 96 – storage, globular / bowl-shaped, etc), beakers, dishes, and flagons (Fig. 7). A variety of forms and types of decoration were noted; 18 or 19 samian vessels were recovered (mainly dishes), all of second-century date, and made in the kilns of Lezoux in Central Gaul. Much of the pottery would have been produced locally, although it would have been fired in clamps or bonfires which leave little archaeological trace (Pollard 1988, 183; Booth 2006, 10). The oxidised wares and some of the grey wares are likely to have been produced in Canterbury (Pollard 1988, 177-9), the grey, reddish brown, and brown coloured wares in finer fabrics are the products of the Upchurch and Thameside potteries (Monaghan 1987; Pollard 1988, 173-7), vessels in a grey fabric may have had a source somewhere in the Isle of Thanet (Rachael Seager Smith, pers. comm.), and the mortaria may be Colchester products (Hull 1963, fig. 60, 27-42, fig. 61, 43-51; Symonds and Wade 1999, fig. 4.10, 158). Some of the vessels were continental imports, including a tiny beaker from Cologne, eight near-complete vessels of Central Gaulish samian ware (three of which were decorated with hunting scenes), and some sherds from South Spanish amphorae. The other finds support the picture of domestic activity on or near the site. The nail assemblage conforms to Manning’s type 1b general purpose nails and may indicate the presence of buildings nearby (Manning 1985, 134-5), the ladle bowl can be paralleled on a number of Roman domestic sites such as Fishbourne (Cunliffe 1971, fig. 60, 55), as can the bucket side mount (Westhawk Farm, Ashford (Scott 2008, fig. 5.12, no. 16)). However there is also some evidence for workshop activity in the vicinity, with a file indicating potential woodworking or metalworking activity, and the possible remains of a blade from an agricultural implement such as a reaping hook (damage and corrosion makes this identification uncertain). The environmental samples recovered from the site comprise a mixture of charred cereal grain, plant, charcoal, oyster shell, and animal bone. These are typical of domestic activity, with oysters being a frequent consumable in Roman times (and unsurprising here given the site’s proximity to the sea), the charred cereal grain consisting of grains of bread wheat, emmer wheat, spelt wheat, and spelt (typical for Roman settlement sites), botanical remains consisting of seeds of the common wild species and grasses (believed to be from cereal fields or waste ground), and the charcoal (mainly oak) relating to background burning and settlement debris. These reflect incidental deposition of domestic and settlement waste, typical of a settlement site or the margins of a villa complex, rather than indicating that more substantial food-processing and/or storage was taking place here. Discussion The Roman features and artefacts discovered on this site represent domestic activity, dated to the second century, in the vicinity of the villa at Hull Place. However it is unclear if this area of activity was part of the villa complex itself, potentially being where rubbish from the villa was dumped, or formed part of a separate settlement within the wider villa estate. This site is an interesting example of the type of activity positioned around Roman villas, and can help illuminate how such activity operated in relation to ROMAN DOMESTIC ACTIVITY IN SHOLDEN, PROBABLY RELATED TO VILLA 97 them. This is a stimulating area of research at present, with the ongoing ‘Rural Settlement of Roman Britain’ project (a collaboration by the University of Reading and Cotswold Archaeology) collecting and analysing all data from developer- Fig. 7 Drawings of a selection of Roman pottery recovered on the site. a) Storage jar b) Bowl with a curved rim c) Carinated jar with a bead rim d) Narrow-mouthed jar with a square rim e) Mortarium, with a bead and flange rim f) Ja r with a curved rim g) Samian bowl (Dr37) eMMA JEFFERY 98 funded excavations in an attempt to understand how the rural settlement of Roman Britain was organized and operated. The results of this excavation at Sholden are clearly relevant to this research. It would appear that there was a connection between this area of domestic activity and the Hull Place villa. The two sites are contemporaneous – the pottery recovered from this site points to a general mid-late second century date, and Building B (at Hull Place) is believed to have been constructed in the early second century and replaced in the late second/third century. The discovery of high status pottery vessels in some of the pits on this site suggests a direct connection to the villa. Indeed the pottery is similar to that recovered during the Hull Place excavations, with the majority of 2nd-century pottery from both collections being grog-tempered wares – 39% of pottery recovered from this site, and just over a third from the Hull Place excavations (Lyne 2009). Furthermore, many of the oxidised wares from both Hull Place and this site were products of the Canterbury industry, with Upchurch grey ware also being recorded at both sites. The finer vessels recovered from both sites included samian ware; however that from this site is believed to have originated from Central Gaul whereas that from Hull Place originated in Southern Gaul, with Gallo-Belgic Whiteware lagenae only being recorded at Hull Place. The similarities in 2nd-century pottery recovered from both sites further supports suggestions that there was a connection between the two sites; however, the discovery of earlier (Iron Age) and later (4th-century) vessels at Hull Place suggests that activity at this site was relatively short-lived in comparison with that at the villa. It is not known whether the features uncovered on this site represent the margins of the villa complex itself, or a separate domestic focus within the wider villa estate. If the former, it is possible that this was where waste from the villa was dumped, particularly considering the high-status pottery assemblage recovered. If the latter, it is still very likely that there would have been a connection between the settlement and the villa, possibly being where people who worked on the villa estate lived. It is interesting that no evidence for Iron Age activity was uncovered on this site. This is unusual, as 53% of Roman settlement sites in the south-east region were continuations of earlier Iron Age settlements (Smith 2013, 14), and evidence for Iron Age activity was uncovered beneath the villa at Hull Place (Parfitt 2009a; b). This suggests that this site may simply have been used for the dumping of rubbish (and therefore did not evolve from an earlier settlement) or, if a separate domestic focus within the villa estate, that it developed alongside the construction of the villa (and potentially because of it), and not out of an earlier settlement. The area was also used for the interment of individual members of the community, with the cremation of the man alongside the bird (possibly his pet!) providing a more personal insight into the lives of those who lived and worked here. The precise connections between this man and the area are unclear; however it can be presumed that there was some form of more personal connection, and that it was a personal decision to be buried alongside the bird. This investigation has highlighted the importance of excavating in the vicinity of villas, as a way of gaining a greater understanding of how the surrounding landscape was utilised. Perhaps similar excavations could provide further evidence ROMAN DOMESTIC ACTIVITY IN SHOLDEN, PROBABLY RELATED TO VILLA 99 on the type of activity taking place around villa complexes, potentially shedding further light on the lives of the villa estate inhabitants. acknowledgements The work was funded by Ward Homes, and commissioned by Matthew Smith of CgMS Consulting. The work was monitored by Ben Found, Kent County Council Heritage Conservation Group. Thanks to Keith Parfitt, of Dover Archaeological Group, for his help and provision of information in relation to the excavations at Hull Place. references Allen, M., 2013, ‘The Roman Rural Settlement Project. Preliminary Results from the South-East: Settlement and Land-Use’ (presentation given to the Society of Antiquaries). Booth, P., 2006, ‘Late Iron Age and Roman pottery’, in Booth, P. (ed.), Ceramics from Section 1 of the Channel Tunnel rail Link, Kent, CTRL Scheme-wide Specialist Report Series, 120-228. CgMs, 2010, ‘Archaeological Desk-Based Assessment: Land to the North-West of Sholden, near Deal, Kent’. Cooke, N., 1998, ‘The definition and interpretation of Late Roman burial rite in the Western Empire’, ph.d. thesis, UCL. Cunliffe, B., 1971, Excavations at Fishbourne 1961-1969: 2 The finds, Reports of the Research Committee Society of Antiquaries London, 27. English Heritage, 2011, Introduction to Heritage Assets: Roman Settlements. Headland Archaeology, 2013, ‘Land NW of Sholden, Deal, Kent. Updated Project Design and Assessment of Results’. Hull, M.R., 1963, The Roman Potters’ Kilns of Colchester, Reports of the Research Committee of the Society of Antiquaries of London, 21. Kent County Council, 2012, South East Research Framework Conference Agenda Documents: http://www.kent.gov.uk/leisure_and_culture/heritage/south_east_research_ framework/serf_agenda_conference_docs.aspx Kent Historic Environment Record. Accessible online at http://www.kent.gov.uk/ ExploringKentsPast/ Lyne, M., 2009, ‘The changing pattern of pottery supply to the Sholden Villa site’, unpubl. report. Manning, W.H., 1985, Catalogue of the Romano-British Iron Tools, Fittings and Weapons in the British Museum, British Museum. Mays, S., 1998, The Archaeology of Human Bones, Routledge. McKinley, J., 2000, ‘The analysis of cremated bone’, in M. Cox and S. Mays (eds), Human Osteology in Archaeology and Forensic Science, Cambridge. Mercian Archaeology, 2005, ‘Osteological Analysis of the Cremated Bone from Roman Road, near Stretton Sugwas, Herefordshire’, unpubl. Monaghan, J., 1987, Upchurch and Thameside Roman Pottery. A ceramic typology for northern Kent, first to third centuries AD, BAR British Series 173. Parfitt, K., 1992, ‘Observations on the Hull Place, Sholden, Foul Sewerage Scheme’, Kent Archaeological Review, 108, 174-176. Parfitt, K., 2009a, ‘Excavations at the Hull Place Roman Villa, Sholden, 2005-7’, Kent Archaeological Review, 176, 120-124. Parfitt, K., 2009b, ‘Preliminary Report on Excavations at Hull Place Roman Villa, Sholden’, Archaeologia Cantiana, cxxix, 103-112. eMMA JEFFERY 100 Pollard, J., 1988, The Roman Pottery of Kent, Kent Archaeological Society Monograph 5. RPS Clouston, 1993, Sholden, Deal: an Archaeological Evaluation. Scott, I.R., 2006, ‘Ironwork’, in Booth, P., 174-88. Smith, A., 2013, The Roman Rural Settlement Project. Romano-British Rural Burial Practices in the South-East (presentation given to the Society of Antiquaries). South-East Research Framework: Public Seminar on the Roman Period (notes) (27.10.07). Symonds, R.P. and Wade, S., 1990, Roman pottery from excavations in Colchester 1971- 86, Colchester Archaeological Report 10. Ward, C.P., 1990, The Romano-British Cremation Cemetery at Frog Farm, Otford, Kent, in the context of contemporary funerary practices in South-East England, Otford and District Archaeological Group. Wardle, A., 1990, ‘Objects of iron’, in Neal, D.S, Wardle, A. and Hunn, J., Excavation of the Iron Age, Roman and Medieval settlement at Gorhambury, St Albans, EH Archaeological Report 14, 138-55.