A Roman Cremation Cemetery at Walmer

A RO MAN CREMATION CEMETERY AT WALMER RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT In 1901 workmen laying out a new sunken garden in the grounds of Walmer Lodge, an impressive Victorian mansion situated on the east Kent coast, some 650m north of Walmer Castle, discovered a number of Roman cremation burials. Kentish antiquary Cumberland Woodruff visited the site and was able to publish an account of the find (Woodruff 1902). It would seem that the burials had been found in two discrete clusters, about 6m apart, each containing several grave groups. Exact details could not be recorded as the evidence had been destroyed before Woodruff's arrival but he considered that ten or twelve separate burials were represented. A total of thirty-three vessels was recovered, including a fine two-handled glass vase. The collection was subsequently purchased for display by public subscription organised through Deal Chamber of Commerce (Chapman 1921, 117 & 132); the glass vessel and eighteen of the pots can still be identified in the Deal Town Hall Collection of Dover Museum; see Appendix. Further burials were discovered on the site by the late Mr A.G. Southam during building work in 1962, after the old mansion had been demolished and its grounds sold for redevelopment (Phillips 1964). Unfortunately, no report was ever published and with Southam's death, some years ago, all information on this important new discovery was presumed lost. However, in 1997 members of the Dover Archaeological Group, whilst searching through long-ignored archaeological material housed in the archive rooms of Sandwich Guildhall, unexpectedly discovered Mr Southam's report, together with virtually all of the pottery vessels excavated from the site (Hoskins, Holman and Parfitt, 2000). In the spring of 2000, two groups of cremated human bone were discovered at Southam's old house in Worth. The cemetery lay about 175m to the west of the present-day seashore, across a triangular piece of land bounded by Liverpool Road on the west, Granville Road on the south and Kingsdown Road on the east (Fig. 1). This is the area formerly occupied by Walmer Lodge (subsequently renamed Walmer Place) and its associated gardens (Broady 1989, 18). The land here slopes down from about 7m above oo at the north to some 345 3 RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT I 0 0 @ @: ® I 200 0 ®®@ ,® ® :@ t􀀄-- Original Driveway (Walmer Lodge) 0 1 2 3 4 0 3 6 9 12 15 5 200 400 © t 6 18 21 l\ I\ 11 \I 600 metres 7 metres 24 feet Fig. I Walmer Place; general plan of cremation burials recovered in 1962, with inset location map 346 L...1...1-1..:..;' .._' ___,: __ .:,___..:, I I I I A RO MAN CREMATION CEMETERY AT WALMER 3.50m at the south. The 1962 burials lay at about Sm OD and were cut into a natural deposit of marine sand containing sea shells, which extended across most of the site. The following account of Mr Southam's work has been compiled from his unpublished text. A total of 45 out of the 4 7 pots originally described have been relocated and 31 of these are illustrated here. Small boxes of cremated human bone survive from the Burials C and I, together with an empty box marked 'F'and a larger box, unmarked but containing a kilogram of bone. It would thus appear that the excavator had carefully retained all the bone for future study but most of this material must now be presumed lost. All the surviving records, pottery and cremated bone have been transferred to Dover Museum, to join those finds remaining from 1901. Discoveries in 1962 Aware of the 1901 discoveries, the approval of plans to build luxury flats across the area formerly occupied by Walmer Place immediately alerted Southam to the renewed archaeological potential of the area. The siteengineer and workmen were quickly informed of the possibility of further Roman finds being revealed and shortly after the contractor's machine exposed another Roman burial. Southam was told and a small team, led by Southam and the late Dr J.D. Ogilvie, was able to investigate the area the following weekend. A further three cremation groups were recovered. A plan was prepared (Archive Plan 1) and the Ministry of Public Buildings and Works notified. The Ministry agreed to finance Southam to maintain a watching brief as the contractors continued their ground-works, in order to recover any more remains that might turn up. Working closely behind the machine, several more cremations were recovered in the same area over the next three days, the final total making eleven. The graves were generally located at a depth of about 0.15m below existing ground level but the cemetery area appeared to have been previously truncated by landscaping of the former mansion grounds. Southam was able to produce a site plan showing the distribution of all the burials (Archive Plans 2 and 3; Fig. 1), although individual grave plans do not seem to have been prepared (probably through lack of time). There are no surviving plans to indicate the precise location of the burials in relation to the new estate, although Southam's report implies that they were discovered during the construction of an access road (now Lord Warden Avenue) following the line of an earlier terraced driveway, running along the western side of the sunken garden that produced the 1901 finds (NOR TR 3770 5078). This siting is confirmed by the records of the O.S. Archaeology Division (Phillips 1964). The sunken garden was preserved in the new development and still remains, providing a useful fixed point for locating the cemetery site on the ground today. 347 RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT Al 􀀁 I I I I I I I I I I I I I I I I I I I I , B1 r' I " " ' B3 􀀂 I I Fig. 2 Walmer Place; pottery vessels from Burial Groups A and B ( 1 :4) 348 B2 A ROMA N CREMATION CEMETERY AT WALMER It appeared to Southam that the graves had been laid out in two rough rows aligned approximately north-south and about 1.50m apart (Fig. 1). In addition to the clearly defined grave-groups several scattered sherds of Roman pottery were also found, mainly in an area of modern disturbance. Much of this material seemed to be derived from the landscaping activities of 1901. Following the completion of the ground-works on site, a detailed account of the archaeological discoveries was produced by Southam and most of the pottery vessels drawn. Although largely completed by the end of 1966, this report was never published. INVENTORY OF BURIALS Burial Group A (Fig. 2) 1 Ovoid flagon of Canterbury sandy ware with buff core and surfaces. Neck and handle largely missing. c. l 00-150. 2 Plain globular flask with short neck and out-turned rim ( damaged). Smooth, hard Upchurch-type ware; sandwiched grey and brown core with light grey-brown surfaces. Monaghan (1987) Class lB. Secondcentury. 3 Lower half of a flagon in Canterbury orange-buff sandy ware. Traces of cream slip or colour-coat on exterior. c.75-150. (Not illustrated.) 4 Lower part of a flagon in Canterbury orange-buff sandy ware. Traces of cream slip or colour-coat on exterior. c.75-150. Nos 3 and 4 appear to be a matching pair. Date: first half of second century AD. Burial Group B (Fig. 2) 1 Large, undecorated jar with everted rim of hard Native Coarse Ware. Light grey core with light brown interior surface and patchy greybrown and orange-brown exterior. Lower half of exterior covered with near vertical striations caused by heavy wiping. c.150-225. 2 Small, carinated bowl of grog-tempered ware with slightly out-turned bead rim. Upright neck with cordon at the base. Grey core with mottled orange and grey-brown surfaces. Thompson (1982) Type E2- 1. First century. This vessel was probably at least half a century old before it was buried. 3 Plain jar with everted rim of Native Coarse Ware. Orange-brown core with light grey-brown interior surface and patchy light grey-brown and orange-brown exterior. Most of upper half missing. c.150-200. 4 Base of a large jar in Native Coarse Ware. Orange-brown core with grey-brown interior surface and orange-brown exterior. Traces of a lightly incised cross on the base. c. 150-200. (Not illustrated.) 349 , RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT I) I I I I I I I I I I I I I -- --􀀊----􀀋􀀌=--=--::.,.-;;;-=--) " C4 ' J D1 ,, '-- D2 (, ' Fig. 3 Walmer Place; pottery vessels from Burial Groups C, D and E (I :4) 350 ' ' ' ' \ \ A ROMAN CREMATION CEMETERY AT WALMER 5 Plain, globular beaker of samian ware, with small out-turned rim (damaged). Dechelette Form 72. c.150-200. Date: second half of second century AD. Burial Group C (Fig. 3) [About 850 grams of cremated human bone, probably from an adult; not analysed.] 1 Part of the base and wall ofa large, plainjar in Native Coarse Ware. Light grey core and orange-brown surfaces. c.150-200. (Not illustrated.) 2 Small, plain poppy-head beaker in smooth grey Upchurch-type ware. Everted rim with a shallow cordon at the base of the neck. c .100- 175. 3 Small, plain jar with everted, thickened rim in smooth Upchurch-type ware. Dark grey and orange-brown sandwiched core with patchy light grey and light orange-brown surfaces. About half missing. c.75-150. 4 Dish of samian ware, Ludovici Type Tg. Interior surface heavily eroded; half missing. Late 1st - early 3rd century. 5 Cup of samian ware, Dr. 33. Concave wall profile with external horizontal groove half-way down. Central Gaulish. Unstamped. Antonine. (Not illustrated.) Date: second half of second century AD. Burial Group D (Fig. 3) 1 Large plain jar with out-turned rim of hard Native Coarse Ware (most ofrim and upper body missing). Grey core with orange-brown interior and mottled orange-brown and dark grey exterior. c.125-175. 2 Dish with straight sides and simple upright rim in hard sandy Black Burnished Ware 2. Grey core with orange-brown interior and mottled grey-brown exterior. Horizontal groove below rim. c.150-250. 3 Lower portion of an elongated beaker in smooth grey Upchurch-type ware. Broad band of very shallow horizontal burnished rilling on the exterior. c. 75-150. 4 Lower half of a small, plain beaker in smooth Upchurch-type ware. Brown-pink core and interior with light grey-brown exterior. c.75- 150. (Not illustrated.) 5 Lower half of a probable flask in smooth Upchurch-type ware. Dark grey core with light grey-brown surfaces. Traces of a horizontal cordon around the girth. Monaghan (1987) Class lB l . l. c.100-150. (Not illustrated.) Date: second half of second century AD. Burial Group E (Fig. 3) 1 Large, narrow-mouthed jar with everted, slightly thickened rim of 351 RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT Gl I I / / / I / I / ' I \ ' II / "1 \I\ I 1 \/ \I 'I -1/. ' 'I / #. \ II \I // o I I I II /' \ H4 / Fig. 4 Walmer Place; pottery vessels from Burial Groups F, G and H (1:4) 352 Fl ) I----~\ A ROMAN CREMATION CEMETERY AT WALMER hard Native Coarse Ware. Light grey core with mottled orange, brown and grey surfaces. A pair of horizontal grooves occur on the shoulder, with a faintly tooled chevron pattern below. A very shallow cordon is present on the neck. Vertical finger smoothing appears on the lower halfofthe body. c.150-225. [Contains a sample of sand.] 2 Ovoid flask or bottle in smooth Upchurch-type ware. Everted rim with cordon below the neck and band of diagonal rouletting below the shoulder. Traces of external dark grey slip. Second-century. 3 Cup of samian ware, Dr. 33. Slightly concave wall profile with external horizontal groove half-way down. Possible eroded stamp, illegible. Antonine. (Not illustrated.) Date: second half of second century AD. Burial Group F (Fig. 4) 1 Large jar with out-curved rim of hard Native Coarse Ware. Light grey core with mottled light grey-brown surfaces. Traces of horizontal burnish on the neck and shoulder. Lower half of exterior covered with diagonal striations caused by heavy wiping. c.150-200. 2 Cup-mouthed flagon with an ovoid body and single handle. Smooth orange-red oxidised Upchurch-type fabric with traces of an external cream slip. Monaghan (1987) Class lE l .4. c.140-190. 3 Bowl of samian ware, Dr. 31. Central Gaulish. Potter's mark illegible, ----M. Antonine. (Not illustrated.) 4 Small, undecorated beaker with everted rim in smooth grey Upchurchtype ware. Band of shallow incised grooves on the shoulder. c.100- 150. Date: second half of second century. Burial Group G (Fig. 4) 1 Plain jar with everted, thickened rim of Native Coarse Ware. Greybrown core with orange-brown interior and patchy orange-brown and dark grey exterior. Lower third of exterior covered with shallow, near vertical striations caused by wiping. c.100-200. 2 Plain pie dish with bead rim of hard sandy Black Burnished 2 ware. Brown-grey core and dark grey surfaces. Exterior lightly burnished. c.150-225. 3 Upper part of a poppy-head beaker in smooth grey Upchurch-type ware. Everted rim with a shallow cordon at the base of the neck. Rectangular panels of barbotine dots on the body. c.75-150. (Not illustrated.) 4 Base and rim fragments of a beaker. Vessel now lost: description by Southam. (Not illustrated.) Date: second half of second century AD. 353 RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT 11 - Jl ,, .. . . . . I '1 ' I I - ... Fig. 5 Walmer Place; pottery vessels from Burial Groups I and J (1 :4) 354 0 , •' A ROMAN CREMATION CEMETERY AT WALMER Burial Group H (Fig. 4) 1 Upper portion of a fairly large, narrow-necked jar or flask in smooth Upchurch-type ware; rim missing. Dark grey core with light greybrown surfaces. Traces of a horizontal groove on the body with faint horizontal rouletting above. c.100-150. (Not illustrated.) 2 Flask in smooth light grey Upchurch-type ware. Most of neck and rim missing; broad, shallow cordon at the base of neck. c.75-150. 3 About 40 sherds representing part of a jar of Native Coarse Ware. Orange-brown core and dark grey surfaces. Lower part of exterior covered with diagonal striations caused by heavy wiping. c .100-17 5. (Not Illustrated.) 4 Small dish with straight sides, of coarse gritty ware: simple upright rim with an external groove below. Vessel now lost: description and drawing by Southam. Date: first half of second century. Burial Group I (Fig. 5) [About 60 grams of cremated human bone; not analysed. Part of an iron nail, probably from the pyre.] 1 Lower half of large jar in Native Coarse Ware. Light grey core and light orange-brown surfaces. Shallow, tooled lattice decoration on exterior. c.125-200. 2 Large, plain flask with short neck and out-turned, slightly thickened rim, in smooth grey Upchurch-type ware. Single bold cordon at the base of neck. c.125-175. 3 Small, plain poppy-head beaker with almost upright rim, in a hard grey sandy Upchurch-type ware. Traces of external dark grey burnished slip on the rim and upper body. c.125-175. Date: mid-late second century AD. Burial Group J (Fig. 5) 1 Large jar of Native Coarse Ware (rim and most of upper body missing). Grey core and interior surface with patchy dark grey and brown exterior. Traces of three shallow horizontal grooves on the shoulder. c.100-150. 2 Dish of samian ware, Curle 15. Central Gaulish. Unstamped. Antonine. (Not illustrated.) 3 Flask in smooth grey Upchurch-type ware, rim and neck missing. Double cordon at base of neck and diagonal rouletted band on shoulder. Second-century 4 About 25 sherds representing part of a small beaker of smooth grey Upchurch-type ware. Everted rim with a cordon at the base of the neck. Bands of horizontal rouletting on the body. c.75-150. (Not illustrated.) 355 RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT Kl 􀀇 ii:􀀈 :::-•· Fig. 6 Walmer Place; pottery vessels from Burial Group K (1:4) 356 A RO MAN CREMATION CEMETERY AT WALMER 5 Small globular flagon of hard Canterbury pink-buff sandy ware. Rim and neck missing; base of small ribbed handle remains. c.75-150. Date: mid-second century. Burial Group K (Fig. 6) 1 Large, undecorated jar (rim missing) of hard Native Coarse Ware. Grey core with light grey-brown interior surface and patchy light grey-brown exterior. Traces of a shallow horizontal groove at the base of neck. Lower half of exterior covered with near vertical striations caused by heavy wiping. c.150-200. 2 Lower half of a poppy-head beaker in smooth grey Upchurch-type ware. Traces of one panel of barbotine dots survive on the body. c.100-175. 3 Large, narrow-mouthed jar, with damaged everted rim, in a fairly hard sandy ware. Grey core and light grey-brown surfaces. Shallow cordon at the base of the neck and a single horizontal groove on shoulder. c.150-200. 4 Small cup of samian ware, Dr. 33. Slightly concave wall profile with external horizontal groove half􀁴way down. Central Gaulish. Illegible stamp. Antonine. (Not illustrated.) Date: second half of second century AD. Burial Group L 1 Dish of samian ware, Dr. 18/31. Central Gaulish. Illegible stamp. Antonine. This vessel was unearthed during bore-hole sampling east of the 1962 burials and was thought to derive from one of the gravegroups discovered in 1901. (Not illustrated.) DATING AND DISCUSSION There can be little doubt that the adjacent groups of cremations discovered in 1901 and 1962 are related to the same cemetery, the full extent of which has never been ascertained. At least twenty burials now seem to be represented and the great majority of these appear to date to the second century AD, with a few that could be of the late first century or early third century. The forms and fabrics of the vessels recovered are largely typical of those found in the east Kent region during this period (see for example Savage, in Hicks 1998), with Native Coarse Ware jars and a range of Upchurch wares being well represented. Samian ware is also common, although one or two unusual forms are present amongst the more standard types. There are a number of flagons in Canterbury Sandy Ware, together with a few vessels in Black-Burnished Ware. Significant assemblages of very similar material come from the Roman cemeteries 357 RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT at Ospringe, near Faversham (Whiting, Hawley and May 1931) and Each End, Ash, near Sandwich (Hicks 1998). The arrangement of the Walmer Place burials into rough rows implies organisation and suggests that graves were marked in some way. The cemetery must be connected with a nearby settlement, probably a native farmstead, although the precise location of this remains unknown. Evidence for Romano-British occupation has been previously recorded on the adjacent chalk-lands at Mill Hill (Parfitt 1990), New St. Mary's Church and nearby St. Mildred's (Chapman 1921) but a settlement closer to the present cemetery may be anticipated. Two pits recently located in Walmer Castle Meadow, some 400m to the south of the cemetery, provide some evidence for Roman activity in the more immediate area (Wessex Archaeology 2000). The coastal location of the Walmer Place cemetery (see Fig. 1) is of some interest. According to Southam, all the 1962 graves were 'imbedded in sand', although Woodruff's report states that the soil around the 1901 vessels was 'of a dark and peaty nature' with 'traces of decayed vegetable matter and sedge' adhering to some. This seems unlikely as bore-hole sampling by Southam across the entire area revealed only sand below the thin top-soil layer. A coastal strip of sand, partly blown into dunes and bounded for most of its length on the seaward ( eastern) side by a shingle bank, begins close to the present site and can be traced for some 12.5km northwards, through the town of Deal to Sandwich Bay. Evidence of Roman activity on these sands has been discovered from time to time since at least the early nineteenth century (Parfitt 1982). At Dickson's Corner, some 6.5km north of the present site, a first-second century AD occupation site was excavated in 1978-79 (Parfitt 1980; Parfitt 2000). Recent work at the former Royal Marines East Barracks site, on the southern side of Deal, has revealed early Iron Age occupation layers within the sands there (Parfitt 1999). It is also this stretch of the Kentish coast that is traditionally believed to have been the site of Julius Caesar's landings in 55 and 54 BC. Southam was much pre-occupied with the point and his report contains a good deal of speculation on the subject. However, it is now clear that the second century AD burials at Walmer Place can have had no direct connection with these much earlier events and they are unlikely to have been connected with anything other than a local native settlement. ACKNOWLEDGEMENTS In his report, Mr Southam wished to record his sincere thanks to the site developers, Messrs James Miller and Partners Limited, particularly their Site-Engineer Mr A. Boyson, for permitting the excavations in 1962 and providing every assistance on site. 358 A ROMAN CREMATION CEMETERY AT WALMER The present writers wish to record their thanks to Mr C.A. Wanostrocht, formerly Honorary Curator of the Sandwich Guildhall Museum, for allowing access to the museum's archaeological archives. Miss Jo Bacon has re-drawn the pottery and Barry Corke produced the site plan from Mr Southam's field drawings. Nigel Macpherson Grant assisted with dating the pottery. Jon Iveson, Curator at Dover Museum, has given much practical help and encouragement throughout . Finally, the work o f M r Southam himself must b e fully recognised, for without his vigilance and efforts further evidence from this important site would have been lost without record. APPENDIX Inventory of pots from the 1901 excavation still held in the Dover Museum (Deal Town Hall Collection; Accession code, DEATH). These notes are taken from an unpublished 'Catalogue of Roman Pottery in the Collections of Deal Town Hall and Dover Museum', compiled for Dover Museum by Richard Pollard in 1984: 0.300 Samian cup, Dr. 27 (Central Gaul, Lezoux?). Stamped 'PRISCINUS F'. (late 1st - mid 2nd century) 0.301 Samian dish, Dr. 31 (Central Gaul?). Stamp illegible. (mid 2nd century) 0.302 Samian dish, Dr. 35/36 (Southern Gaul?). Stamp illegible. (late 1st - early 2nd century) 0.303 Samian dish, Dr. 36 (Southern Gaul?). Stamp illegible. (late 1st - early 2nd century) 0.307 Samian cup, Dr. 33 (Central Gaul, Lezoux?). Worn stamp, possibly Annius of Lezoux. (First half of 2nd century) 0.309 Samian dish, Dr. 18/31 (Lezoux). Stamp 'PRIMANI' . ( first half of 2nd century) 0.312 Samian cup, Dr. 27 (Southern Gaul?). Stamp illegible. (late 1 st - early 2nd century) 0.314 Samian cup, Dr. 33 (Southern Gaul?). Stamp illegible. (late 1st - early 2nd century) 0.318 Samian dish, Dr. 36/Ludovici TI. (Central Gaul or Eastern Gaul). Four-petal rosette. (late 1st century - early 3rd century) 0.320 Samian cup, Dr. 27 (Southern Gaul?). Stamp illegible. (late I st - early 2nd century) 0.323 Samian cup, Dr. 33 (Lezoux). Stamp 'CERIAL.M'. (c. 140-160 AD) 0.326 Samian cup, Dr. 27 (Central Gaul, les Martres-de-Veyre). Stamped ']CLN[' . (late 1st - early 2nd century) 0.380 Ring-necked flagon in pink-buff fabric with 'white' slip. (mid 2nd - early 3rd century) 359 RICHARD HOSKINS, DAVID HOLMAN AND KEITH PARFITT 0.389 Disc-rim flagon in pink-buff fabric. (late 1st - early 2nd century) 0.390 Ring-necked flagon in coarse pink-buff sandy fabric. (late 1st - early 2nd century) 0.391 Pulley-rim flagon in fine off-white fabric. (late 1st - early 2nd century) 0.299 Pulley-rim flagon in orange fabric with white slip. (late 1st - early 2nd century) 0.403 Biconical flask or jar in fine reduced ware. (late 1st - early 2nd century) Additionally (not described by Pollard): D 29 A fine, two-handled, globular vessel of light green glass with wide, tapered neck. BIBLIOGRAPHY Broady, J., 1989, Deal in Old Photographs (Alan Sutton). Chapman, H.S., 1921, The Story of Do/a. Julius Caesars Landing Place (London), 109-114. Hicks, A.J., 1998, 'Excavations at Each End, Ash, 1992', Archaeologia Cantiana, CXVIII, 91-172. Hoskins, R., Holman, D. and Parfitt, K., 2000, 'Roman Cremation Cemetery, Walmer Place, Deal', Kent Arch. Rev., 141, 7. Monaghan, J., 1987, Upchurch and Thameside Roman Pottery (BAR 173). Parfitt, K., 1980, 'A Roman Occupation Site at Worth', Kent Arch. Rev., 62, 46- 47. Parfitt, K., 1982, 'Roman Finds from the Sandhills, north of Deal', Kent Arch. Rev., 70, 225-227. Parfitt, K., 1990, 'Excavations at Mill Hill, Deal 1982-89: an Interim Report', Kent Arch. Rev., 101, 15-16. Parfitt, K., 1999, Archaeological Observations at East Barracks, Deal, 1999 (unpublished CAT report, Nov. 1999). Parfitt, K., 2000, 'A Roman Occupation Site at Dickson's Corner, Worth', Archaeologia Cantiana, cxx, 107-148. Phillips, A.S., 1964, O.S. Record Card, TR 35 SE 17. Thompson, I., 1982, Grog-tempered 'Belgic' Pottery of South-eastern England (B.A.R. 108). Wessex Archaeology, 2000, 'Archaeological Observations in Dover, Folkestone and Deal: 1993-1998', Archive Report. Whiting, W, Hawley, W. and May, T., 1931, Report on the Excavation of the Roman Cemetery at Ospringe, Kent (Soc. of Ants Report No. VIII). Woodruff, C.H., 1902, 'Romano-British Interments at Lower Walmer', Archaeologia Cantiana, xxv, 1-10. 360