Kent International Business Park, Manston: Excavations and Evaluations 1994-1997, Report 1

KENT INTERNATIONAL BUSINESS PARK, MANSTON: EXCAVATIONS AND EVALUATIONS 1994-1997, REPORT 1 D.R.J. PERKINS, E. BOAST, T. WILSON, AND N. MACPHERSON-GRANT INTRODUCTION It is about two hundred years since Thanet's village communities of 'fishermen with ploughs' found themselves engulfed by burgeoning towns, resulting from a new custom, the sea-side holiday. Sequential products of this was further urban developments as 'commuter dormitory', and 'retirement dreamland'. For the indigenous community the consequence was unemployment, at first slight and seasonal, then grossly endemic. During the twentieth century the percentage of Thanet's labour pool unemployed has seldom fallen below those for Britain's worst black-spots. Combating this iniquitous tradition required dramatic measures, so that, albeit sadly, the local authority decided to sacrifice large tracts of Thanet's highly productive farmland to the construction of industrial estates. The first to be proposed in the early 1990s was 'Kent International Business Park', at Manston, a development covering 70.82 ha. (175 acres). Originally the concept of Wiggins Group pic, the development was soon espoused by Thanet District Council and Kent County Council, and received support from the European Community. Any development on this scale would be likely to have archaeological implications, especially in Thanet where the average density of known sites is 8.9 per square kilometre (23.0 per square mile). For this reason, a programme of archaeological work was specified by the County Archaeologist Dr John Williams, and was undertaken by the Trust for Thanet Archaeology. The work was carried out in three phases; Phase 1, 1994, evaluation by trenching; Phase 2, 1996, evaluation and excavations; Phase 3, 1997, planned excavations. Of the total designated business park area of 70.82 ha. (175 acres), 217 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT 49.455 ha. (122 acres) has been evaluated, and three archaeological sites examined by excavation. The archaeological evaluations and excavations described in this paper constitute a landscape study on a scale rarely seen in Kent, and one augmented by archaeological work carried out nearby at the same time (1994) on the A253 (Bennett and Williams 1997). The evaluation has particular value in that the first 12 h. (30 acres) to be examined were obscured either by deep soil or periglacial features, so that crop-marks would not form, or could not be seen. Since large areas of east Kent share these geological conditions, the evaluation provides control data for future assessment. Similarly, the morphology of the three sites investigated provides new criteria for use in aerial landscape study. Site 18 proved to be a pre-Conquest Belgic settlement exhibiting little or no trace of Roman influence. The study of its ceramic assemblage will be of considerable value to Kentish archaeology. While currently under way, its anticipated size precludes incorporation herein. ACKNOWLEDGEMENTS Thanks are due to Wiggins Group pic who funded the archaeological work, also to their advisers to the Conspectus Partnership, Thanet District Council, Kent County Council and the landowner Mr Robertson, all of whom afforded the utmost co-operation. Not least though, the writer would like to thank the members of the Thanet Archaeological Trust evaluation team led by George Slade for their perseverance and dedication during many weeks of trenching and excavation work often in appalling weather conditions. THE SITE: TOPOGRAPHY AND GEOLOGY Kent International Business Park is situated on Thanet's high downland central plateau at elevations varying between 40 and 50 m. OD, see Fig. 1 and inset. From the high ridgeline occupied by the main runway of Manston Aerodrome, the land descends northwards towards Acol. A shallow valley curves at first north-west and then north-east through the centre of the site. According to the British Geological Survey map (Ramsgate, Sheet 274, 1:50,000, Drift Section) the site geology is that of a valley fill of Younger Brickearth between rises of the Upper Chalk, in other words typical Quaternary geology of the Thanet area. The landowner Mr Robertson said that local oral tradition had the overburden in the field as 'brickearth' 218 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 THANET Site 25 EAST KENT 10 km Site 26 Site 24 I. : Site 27 Evaluation Cleve Court Ar e a ^ Alland Grange 0 0 O Site 29 Site 28 Manston Aerodrome 500 m. Fig. 1. The Kent International Business Park development area and peripheral archaeology Inset: The Fig. 1 plan in its Thanet and east Kent setting. 219 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT between 8 feet (2.4 m.) and 22 feet (6.7 m.) deep'. In the event trenching revealed the geology as more varied, with some Tertiary strata exposed. Beneath, a modern agricultural topsoil (0.40 m.) and soil disturbed by deep ploughing (0.20 m.) three typical profiles were: (A) A surface of the Upper Chalk much disturbed by periglacial action (ice wedge clefts), with here and there hollows retaining an horizon of light brown loam; (B) Clay with flints (or this degraded to a thin layer of flints) over a red-brown ferruginous layer of sandy silt, the Older Brickearth. This was sectioned during evaluation (see Fig. 2, 4), and found to be 2.6 m. keep over the chalk. A phenomenon observed in this section and in other trenches was that the surface of the brickearth layer had been cut by channels of bowl-shaped section some 3.0 m. wide and 0.75 m. deep. These contained a 'porridge' of 'Coomb Rock'. The writer interprets these as gullies cut by run-off during glacial freeze-thaw conditions carrying down material from the chalk ridge to the south; (C) A hard fine ochre coloured sandy silt to an unknown depth, the lower measures of the Thanet Beds. In general, in the northern half of the site, a west to east profile expressed in terms of setting out squares would be B-C-A-A-A, while in the southern half it would be A-B-B-C-C-B-A. Like most of Thanet's farmland the business park fields have suffered from deep ploughing, and the evaluation team were told by the ploughman that the 'subsoiler' used to be set to 24 in. (0.60 m.) on the field. THE SITE: KNOWN ARCHAEOLOGY AND HISTORY Peripheral Archaeology Numbered archaeological remains are as shown in Fig. 1. Beyond the northern boundary of the business park crop-marks revealed scattered ring ditches, presumably ploughed-off round barrows and a complex (Site 25) that could be either a Romano- British building, or more likely, a medieval manor or farmhouse within a square ditched enclosure. No name or tradition survives to support this. To the east at Cheesmans Farm are several ring ditch cropmarks. Until recently obliteration by landscaping to create a paddock, an earthwork known as 'Cheesman's Camp' was visible (Site 26). It was a sub-rectangular enclosure formed by double parallel ditches with a causeway entrance. Illicit excavation of the ditch fills by a 'treasure seeker' in 1987 yielded a quantity of Romano-British pot sherds, 220 Site 22 •S \ ^ Inset: The trench layout in each grid box. g J 1 Measurements in metres. All trenches 20 x 1.5 m Site 21 Site 20 Site 13 (not located) Site 23 t l-f Site 19 Site 18 -:tSite 14 ;!cS+^i^i-: 1 Features or find spots: Q vo Fig. 2. Kent International Business Park, trenching system, sites, features and find spots D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT mostly BB2. Site 27 is a circular crop/soil mark associated with a bronze hoard (Perkins 1995a). Adjacent to the business park southern boundary on what is now Manston Aerodrome an extensive Romano-British iron working site (Site 28) was recorded by Professor W. F. Grimes in 1944 during work to extend the Manston runway. West beyond Cleve Court a crop-mark array suggests a barrow cemetery and prehistoric ditched field systems (Site/s 29). On-site archaeological remains and cropmarks Numbered archaeological remains and cropmarks are as shown in Fig. 2. The only remains exposed in the business park area prior to evaluation were revealed by the passage of a water main along the southern boundary in 1987. They consisted of patches of a subsoil horizon containing Belgic and Roman sherds and midden material (Perkins 1989). Nothing of the sort was seen in the southernmost trenches of the 1994 evaluation. No crop-marks appear in the southern 30 acres of the business park. Factors of the geology over most of that area, deep subsoil and periglacial action, ensure that crop-marks either cannot develop, or, masked by ice-wedge crop-markings, could not be seen. Their absence does not therefore mean that no significant remains are present, but the paucity of finds and features encountered during Phase 1 of evaluation in 1994, together with the results of magnetometer survey, strongly indicates that this is so. The northern and eastern areas of the business park (58 ha. = 145 acres) have a geology conducive to crop-mark formation, and nine marks had been recorded by air photography prior to evaluation. The sources were: Thanet Archaeological Society/Trust annual surveys, Royal Commission for Historic Buildings and Monuments, Kent County Council and Thanet District Council, the Potato Marketing Board, O.G.S. Crawford, and S.G. Robertson (landowner). The crop-mark sites were: (i) Ring ditches representing ploughed off round barrows: Sites 13, 20, 21, and 22, see Fig. 2. (ii) Rectangular/sub-rectangular enclosures: Sites 14, 15 (only identified in photos after its discovery by trenching), and 18 (iii) Circular enclosure: Site 23, this has a diameter of c. 140 m. (iv) Site 24: A long dark crop-mark was photographed in the 1930s by O.G.S. Crawford. The 'father of crop-mark photography' thought that it might be 'Thunor's Pit' mentioned in the Dark Age legend of 222 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 the run of Queen Domneva's hind (Crawford, 1933, Antiquity, vol. 7, 25, 92 and plate 8). It is more likely perhaps to represent the site of a long barrow. The site history Not much is known of this tract of land, although there is some evidence from place names (Acol O.E. 'The Place of the oaks') and from John Lewis' History of Thanet (Lewis 1723) that the area was heavily wooded until after Tudor times. On the other hand evidence of medieval agriculture in the form of strip lynchets can still be discerned north of Alland Grange (O.E. 'Old Lands'). When William Cobbett came here in 1823 on his 'Rural Rides' the whole area was corn fields without hedgerows (Cobbet 1853). Manston Aerodrome began its life as a Royal Naval Air Service flying field in World War I. Later in 1918 it became a Royal Air Force bomber base. In the inter-war years it was a training camp, and there is some evidence that part of the business park site on its periphery may have been used as a bombing range, as a small bomb of the target marker type was found by the excavation team. In the later part of World War II, R.A.F. Manston was extended onto the site which became a dispersal area. Here in 1944 were kept Britain's 'most secret' aircraft, the Gloster Meteor jet fighters being used against the VI flying bombs. Afterwards part of the land became orchard, but this was cleared in the 1950s and has been in arable use since. EVALUATION: METHODS AND RESULTS Methods Evaluation was by machine-cut trench. In all 231 trenches each 20 x 1.75 m. were cut to a pattern recommended by the County Archaeologist, see Fig. 2. The pattern ensured that no two trenches were more than 50 m. apart (most less than 25 m apart) and their area represented sampling at a 2.5 per cent level. The efficacy of this evaluation system in locating archaeological sites or indicating their proximity is dealt with in discussion. Within each trench the subsoil was removed with a toothless bucket a few centimetres at a time under archaeological scrutiny, excavation being terminated when an archaeological feature was exposed or a natural horizon was reached. Occasionally deeper sondages were cut down to chalk to confirm the interpretation of the deposits as Tertiary 223 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT or Quaternary strata. As trench surfaces were reduced, they were scanned with a metal detector, as were the sides of the trenches. Simultaneously with the trenching operation the adjacent ground surface was field-walked and prospected by metal detector using the setting out grid of 75 m. squares as a plotting reference. Finally, on commencement of construction in 1997, the 'infrastructure' routes for roads and services were field-walked and detected after they had been stripped of topsoil. It should also be mentioned that selected areas of the business park were surveyed by magnetometer by Bartlett Clark Consultancy acting for Wiggins Group pic. This provided useful information to the evaluation team, particularly as to the location of Site 18. Results, surface survey Over the field surface, field-walking and detecting revealed a light scatter of finds of all periods from the Neolithic to the modern era. In general, finds consisted of: Flints: White patinated debitage with a few scrapers and calcined flint 'pot-boiler' nodules. Ceramics: Mostly small worn sherds of local sandy wares in medieval fabrics, followed by Tudor to modern era materials. Metallic objects: A miscellany of finds, eighteenth-century to modern era, the majority components of World Ware II ammunition for small arms and aircraft machine guns and cannon. Sites, isolated features and finds located by trenching The following sites and features are numbered as shown in Fig. 2: Features 1, 2, 3. These were patches of dark soil in the subsoil surface, typically round and about 1.0 m. across, and a few centimetres deep. No datable finds, but calcined flints. Feature 5. This was a cut in the subsoil surface down to chalk, oval in plan, 0.80 x 1.00 m. and 0.15 m deep. The sides of the cut had been exposed to heat such as to bake them into a sub-ceramic. Within this the loam fill contained charcoal traces and tiny pot fragments, a possible kiln base? Feature 6. Here the brickearth surface was cut by two shallow patches discoloured by ash or organics. On exploration one was found to contain many hundreds of sea shells (mussels) and sherds of prehistoric pottery. This was in a flint tempered Late Bronze Age fabric. Further trenching revealed the feature to be in isolation, but yielded Belgic and medieval sherds. 224 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 Feature 7. This was the base of a small pit surviving plough damage by being cut into chalk. It was of bowl-shaped profile, 0.60 m. across and 0.20 m. deep, with a fill of loam. It held several sherds of Late Bronze Age (Deverel Rimbury) pottery, with others plough-scattered along the length of the trench. A plough-damaged cremation burial? Feature/Find 8. A bronze axe fragment was found by metal detector. At this location the Upper Chalk has an overburden of topsoil only 0.40 m. in depth. The bronze was found in a bowl-shaped hollow, possibly natural, about 0.40 m. across and 0.20 m. deep. Site 13. This from crop mark photography had been predicted to be the site of a ploughed off round barrow. In the event extensive trenching at the co-ordinates revealed only some minor features producing a few flint flakes, and what could be vestigial traces of an oval plan ditch. A possible explanation is that these are the remains of a barrow ploughed off since it was recorded by air photography, but the excavation team have reservations as to this. Site 14. This large medieval enclosure was easily located and evaluated by trenching. Its subsequent excavation is described below in 'Excavations: Site 4/5 Medieval'. Sites 15 and 16. These were found by trenching in 1994 and recorded as a length of ditch and a small pit about 30 m. away. Further evaluation in 1996 revealed them to be features of a large Belgic enclosure which was part excavated. This will be the subject of a future report. Site 17. Part of this was within the business park with the bulk within the eventual boundaries and subject to future phases of evaluation. It appears to be a ditched enclosure, see above. Sections in two places 10 m. apart yielded no finds. See above in 'Known archaeology and history.' Site 18. This enclosure could not be located by trenching in 1994, as it had been misplotted from air photos. A search pattern of trenching in 1996 located it on the western boundary of the development area, see 'Excavations: Site 18 medieval.' Site 19. Here trenches exposed two pits and the remains of an urn in situ. One of the pits was of bowl shaped section, 0.30 m. deep and about 1.0 m. in diameter, and contained waste flakes. The other was very shallow and about 3.0 m. in diameter. Its fill of burnt and ash-darkened soil held sherds and a small shell midden. The urn fragments and sherds obtained from the midden pit and ploughdamaged interface were in Late Bronze Age, Deverel Rimbury, fabric. Sites 20-24 are within the eventual boundaries of the business park and subject to future phases of evaluation. For descriptions of them and Site 25 see above in 'Known archaeology and history'. 225 to as 1 m lit * 24 •~"--v•-; • • Complex • 25 o 27 KVJ /.••••::.-i ii f^i t.-»t. •V-.M 3H # illlililsillllK AreaC 4 i Fig. 3, Site 4/5. The evaluation area boundaries are contracted by 3 metres to save illustration space. KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 EXCAVATIONS: SITE 4 /5, MEDIEVAL Evaluation results in 1996 Trenching revealed that Site 4 was a ditched enclosure of the medieval period, and that Site 5 was merely a short right-angled length of ditch of the same period, perhaps an abandoned commencement cut, the construction of the enclosure then being for some reason re-located. The first evaluation phase revealed Site 4 to be contained within a four-sided polygonal ditched enclosure measuring roughly 53 m. by 38 m. at its maximum width, see Fig. 3. After the four corners of the enclosure had been found and the lengths of ditch between them had been sectioned in 15 places, the interior of the enclosure was sampled by cutting two trenches across it, see Fig. 3, A, B. These revealed a number of deep-cut features and surviving ancient horizons. The Enclosure Ditch This presumably defensive boundary ditch had originally been cut down through a much deeper overburden of brickearth than now exists. Ditch fills above natural chalk surface level with their associated ancient land surfaces have been truncated by ploughing over most of the enclosure area. Evidence for their existence is retained in the form of truncated flow-lines of chalk and flint nodules in the ditch section. The surviving chalk cut ditches are characteristically 1.5m wide and 0.70 m deep, of open-U or truncated-V section. Two causeway entrances made breaks in the south-eastern and north-western angles of the enclosure. The first of these was only 1.0 m. wide, and the rounded ends of the ditch were cut down to a depth of 1.5 m., the south-western termination having a post-hole (Fig. 3, Feature 19). The width of the north-western entrance could not be determined, as it coincided with a deep brickearth-filled natural hollow in the chalk. Because of worm action, the ends of the opposing and unaligned ditches could not be found with certainty, but the causeway must have been not less than 3.0 m. wide. Finds from the ditch section fills were sparse other than in Section 1, the south-eastern angle. At this point a stratified fill was encountered, with layers packed with shells, bones , and pot sherds. The Excavation phase 1997 As a preliminary, topsoil and subsoil disturbed by deep ploughing 227 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT were removed over the whole site and in a band 5 m. wide on all sides. Preservation of features and ancient horizons was found to be variable. Over much of the interior ploughing had created a plough-soil - Upper Chalk interface so that features such as pits and post-holes were truncated. In two areas however, marked C and D in Fig. 3 the chalk descended, and these hollows had allowed the survival of man-made and natural horizons respectively. That in Area C was a layer of light brown loam modified by human activity and about 0.30 m. in maximum depth. Sandwiched within this were at least two twelfth-thirteenth-century occupation layers with structural remains. To have recorded this complex stratification would have taken the excavation team rather longer than had been allowed to deal with the whole enclosure, so that with civil engineering work about to commence, investigations had to be limited to the cleaned features, sections and trenching shown in Fig. 3. The Area D horizon was natural, consisting of a fill of bright orange coarse sand. It had been cut by two features, one prehistoric and described below, the other, Fig 5, Feature 30, a fragment of machine-cut slit trench, presumably dating from the 1940s. Earlier presence on the site ?Neolithic - Early Bronze Age Residual evidence consist of waste flint flakes and hammer stones recovered from ditch fills. Three site features cannot be definitely dated, but seem likely to be prehistoric, they are: Feature 29, a ditch or palisade trench that passes through the south-east corner of the enclosure with its fill cut by two angles of the medieval ditch. It was of V-section, between 0.40 and 0.90 m. wide, with a depth of 0.40 to 0.60 m. Its fill yielded no finds, but was a sandy brown loam quite unlike any other horizons encountered on the site. Feature 28 was a bowlshaped pit cut into the sand of Area D, more or less circular and 4.0 m. in diameter, with a maximum depth of 0.80 m. Just above the bottom was a black ash layer extending across the central 2.0 m. of the pit. Finds consisted of shells, (marine) bone fragments, and a boulder of local sandstone, and a flint scraper. The upper fill of the pit was cut by Feature 26, a lens of bright red sub-ceramic about 1.0 m. in diameter, presumably a hearth. A flint end scraper was found on the burnt surface. The medieval construction and occupation c. A.D. 1200-1375 Structural remains interpreted as huts were: 228 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 Feature 1, part of a floor of rammed chalk cut by Feature 2, possibly a sill-beam slot, Feature 3, a post-hole, and Feature 4 a shallow cutting in the chalk. Five post-holes, Features 5-9, were in association, see Fig. 5, Trench A. These features were all chalk cut, and at the base of the Area C soil horizon. Feature 41, a hearth situated 1.5m. away from the Feature 1-9 complex was 0.30 m. above the chalk on the same level as: Feature 42, a rectangular floor of rammed chalk/soil mix measuring 4.5 x 7.0 m. This had been laid in a cutting through the Area C horizon about 0.25 m. deep that extended down into natural chalk for another few centimetres. It had sealed two features, a post-hole, Feature 45, and a rectangular pit Feature 47, and was itself cut by Feature 46, a circular hearth about 1.5 m. in diameter. Just west of the Feature 42 two pits, Features 44 and 48 had been cut into chalk. Feature 41, a circular hearth 2.0 m. in diameter could represent the plough abraded remnant of a third hut. Post-hole systems: Although the north-east corner of the enclosure had suffered heavy plough attrition, two distinct groups of post-holes had survived in severely truncated condition. The Feature 24 complex had 19 round post-holes, and ten vestigial 'ghosts' (shown in broken line in Fig. 3). While they can be seen to fall into linear arrangements, the whole does not make sense as a single structure. Features 20, 21, 22 and 23 were shallow square and rectangular cuts in the chalk, apparently forming a crude right-angle. If not post-hole bases of large dressed timbers, it is difficult to think of a function for these features. Pits: Fifteen pits were recorded. Of these 10 were of round or oval plan, and 5 were sub-rectangular. Of the 10, all were less than 1.0 m. in diameter and shallow, and only 4, Features 38, 39, and 40 (a group) and 33 yielded datable finds. Feature 33 lay on the edge of an area of ?organically discoloured subsoil surface into which were sunk Features 31 and 32, shallow flat-bottomed rectangular cuts filled with black ashes. Features 36, 43 and 47 were sub-rectangular cuts in the chalk with inwardly sloping sides and flat bottoms, all three about 2.0 m. long by 1.5m. wide, and about 1.0 m. deep. All three appeared to have been infilled rapidly and purposefully. Feature 47 had been sealed away under the Feature 42 hut floor. Features 36 and 43 had been joined by a cut, and both were connected by a shallow chalk cut shelf with Feature 35, an internal ditch. The latter was about 1.5 m. wide and 229 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT 0.70 m. deep, and of open-U section. It was 10 m. long and ran parallel to the western length of the boundary ditch and about 2.5 m. in from it, cutting it just past the return of the south-west angle. Site 4/5, a brief summary This site seems to represent a large farming compound in occupation from the twelfth to fourteenth century, with, from ceramic evidence, the main emphasis on the later twelfth century. Although it has suffered plough damage, surviving horizons and features demonstrate two kinds of structure, huts with rammed chalk floors and sill-beam slots, and seemingly larger timber-framed buildings using upright posts as main supports. That rather less was found in the way of midden material than might have been expected, even at a low level of occupancy, may be due to domestic refuse being broadcast on the fields with manure. EXCAVATIONS: SITE 18, MEDIEVAL Evaluation results 1996 After a great deal of abortive trenching, this site was located from information supplied by magnetic survey. That and the successful trenches revealed the crop-mark to have been misplotted by about 50 m., and that the site was bisected by a public footpath and the business park boundary. This limited subsequent excavations to the eastern half of what is presumed to be an elongated sub-rectangular enclosure. Trench 1 (see Fig. 4) located parallel western boundary ditches, Features 18 and 19, and a pit Feature 20. These features were found to be under about 1.5 m. of colluvium, thus explaining why an incomplete crop-mark is exhibited by the site. The features consisted of loam-filled cuts through colluvium of unknown depth. Trench 2 exposed the chalk cut eastern angle of the enclosure ditch and part of Feature 10. Partial excavation in 1997 With the constraint imposed by the land boundary and public footpath, investigation of the site had to be made from the line shown as A-B in Fig. 4 - modern topsoil and colluvium was stripped back from this to the internal features and eastern boundary ditch, and for a further 5 m. The stripped area was found to have a marked east- west slope, the fall 230 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 (9661) Z H™HL iramui C/5 o 231 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT being one metre in twenty, other than in the south-west corner (Area C in Fig. 4) where it was much more steep and the colluvium sectioned by the bulk was 1.5 m. deep. Elsewhere the overburden was about 0.30-0.50 m. deep, and the chalk-cut features were exposed in a slightly plough-abraded surface of the Upper Chalk. An exception, shown as Feature 4 in Fig 4, consisted of a slight hollow in the chalk that had retained a medieval horizon of light brown field loam. The enclosure ditch: This was sectioned in four places, see Fig. 4. The sectional profiles numbered 1, 2 and 3 were very similar, each being cut into the chalk about 2.5-3.0 m. wide and 0.80 m. deep. The inner sides were cut nearly vertically with the outer sloping gently with a step-like change of angle at a depth of 0.40 m. Ditch fills were of light brown loam apparently slowly deposited over an initial chalk silting derived from within the enclosure. At the south-eastern corner (Section 4) the deep chalk-cut ditch came to a rounded end cut by a narrow shallow continuation that turned west and was lost where the chalk surface dipped below colluvium. Last observed as Feature 8 its dimensions and profile suggest a palisade trench. Internal ditches or elongated pits: These were present as Features 2, 9, 11, 12 and 16. All were rather shallow cuts with no apparent function other than to drain surface water west and down hill. Feature 12 could have been a hut drip trench. Pits: These fell into two distinct groups: Large shallow oval cuts of bowl-shaped profile exhibiting staged fills of loam and chalk rubble, Features 6, 10, and 17; and deeper sub-rectangular steep-sided cuts with flat bottoms, back-filled as middens, Features 1, 5, and 15. While the latter group resemble prehistoric storage pits, no functional evidence was obtained from any of these features. Structural remains: Two sets of structural remains were present. These were Feature 7, consisting of building foundations, and the Feature 13/14 complex, a sunken-floored building, the latter being dealt with in detail in Appendix 1 below. Feature 7 survived as a chalk-cut foundation trench, in plan an open-ended rectangle measuring roughly 8x10 m., see Fig. 6. It is possible that Feature 5, a pit back-filled as a midden, was originally 232 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 part of the structure, perhaps as a pier foundation. The trench was of truncated-V section and about 1.5 m. wide. On the eastern uphill side it was cut about 0.50 m. deep compared with a depth of 0.20 m. on western side, demonstrating a conscious attempt at levelling and terracing. The fill of the trench was of typical 'robber trench' nature except where it had been cut by Feature 1, a later pit back-filled as a midden. Building materials in the fill consisted of large water rolled flint nodules of the kind found eroded from the Thanet chalk cliffs, small boulders of a local sandstone from the Thanet Beds 'dogger' layer, and a sandy calcareous conglomeration, presumably degraded mortar. Site 18, a brief summary Extrapolating from the excavation evidence, Site 18 seems to have been a sub-rectangular ditched enclosure measuring about 25 x 55 m., and somewhat cluttered with internal features, among them a stone build edifice and at least one outbuilding. Ceramic evidence suggests that the period of occupation was close to that of Site 4/5, from the late twelfth to mid thirteenth century. DISCUSSION The evaluation in general In an era given to the industrial or urban development of green field sites, often on a massive scale, two questions loom large in the thoughts of concerned archaeologists: how effective is sample trenching in locating sites, and how representative of site distribution are crop-mark phenomena? The Manston evaluation gives rather sobering answers to both questions. Site 15 was contained within six survey squares holding a pattern of 24 trenches at a maximum distance apart of 50 m., with three quarters of them only 25 m. apart or less. However, as can be seen from Fig. 2, Site 15 measuring 90 x 150 m., fitted into the pattern so well that only three of the trenches touched it, and it was only detected in two! Moreover, although Site 15 was found in hindsight to exhibit a crop-mark; it had escaped the notice of those carefully searching aerial photographic prints and was hardly macroscopic. Site 18 was known from its crop-mark, but in a featureless landscape had been misplotted by at least 75 m. Another problem affecting the evaluation was the heavy plough and subsoiler damage to the site, a degree of attrition that one hopes, 233 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT perhaps vainly, is not universal in Kent. The writer and other have been able to demonstrate that, over much of upland Thanet, up to one metre of topsoil has been eroded in the last sixty years or so, and in some areas perhaps 20 cm. of chalk bedrock. A sure sign of such attrition, and one noticed at once during the evaluation, was that artefacts of all cultural periods, Neolithic to the modern era, were found mixed in the modern plough soil above an horizon that usually contained no trace of human activity. The medieval sites: Site 4/5 Though severe, plough attrition on this site has, overall, left a great deal more evidence of internal layout than in the case of Site 15, the Iron Age enclosure. A picture emerges of a compound within presumably defensive ditches, possibly with an inner bank and palisade. The north-west quarter seems to have been favoured for domestic accommodation, with at least two huts. Post-hole systems in the north-east quarter suggest one or more larger post-framed buildings, perhaps a series of barns. Most ceramic fabrics present were of local origin, Canterbury or Tyler Hill wares; this together with the dearth of small finds seems to indicate that the site was a working environment, its occupation lacking any quality of affluence. At some time about the middle of the thirteenth century, the south-east corner of the enclosure ditch becomes a midden, becoming filled to ground level with a mass of oyster shells, animal bones, and pot sherds. Does this event reflect the commencement of a down-grading of the enclosure, in terms of diminishing status and level of occupancy? The medieval sites: Site 18 Although only about half the size of the Site 4/5 enclosure, and with only halfofit able to be investigated, Site 18 yielded 25 per cent more sherds, suggesting a higher level of occupation and activity. The choice of venue for the enclosure, the fairly steep slope of a dry valley, is curious. Why go to the trouble of terracing, when level ground lies only 100 m. to the east? Could the modern footpath and field boundary represent an ancient trackway on which both the Site 4/5 and 18 enclosures once fronted? Interestingly, the occupation evidence on both medieval Site 4/5 and Iron Age Site 15 is concentrated on the downhill western extremity of each enclosure, although this would afford no shelter whatever from the prevailing south-westerlies. 234 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 No archive record or tradition survives with reference to these enclosures, so that it can only be seen in association with one another and perhaps the enigmatic Site 25. The ceramic evidence suggests strongly that they were contemporaries, although not sharing equality in level of occupation or social status, a probability further emphasised by the stone-built building in Site 18. A tempting image is that of Site 18 as a small manor, with Site 4/5 as its farming compound with barns, byres and huts for labourers. Occupancy, at least in terms of ceramic dating, seems to have ended for both enclosures in the mid fourteenth century, ?coincidental with the arrival of the Black Death. Sites 4/5 and 18, comparanda Although Tim Tatton-Brown (1977) was able to gazetteer 120 medieval moated sites in Kent, for the most part in the west of the county, very few have been examined in any detail. Stuart Rigold (1983) remarked that by national standards medieval archaeology in Kent made a showing that 'has been a pretty poor one'. So few rural settlements have been investigated that comparisons with those at Manston can hardly be drawn. Exceptions are those of the moated site at Pivington (Rigold 1962), and Moat Farm, Leigh, (Rigold 1976) both of which, moats apart, are somewhat similar to the Site 18 ditched enclosure. DAVE PERKINS APPENDIX 1: THE SUNKEN -FEATURED BUILDING IN SITE 18 Excavation Method Feature 13/14 is a cut feature located on site 18, discovered during phase two of excavations at Kent International Business park. The feature was excavated by cutting sections across the deposits visible on the surface leaving two upstanding baulks where a stratigraphic section could be reconstructed. Excavation A large spread of red-brown burnt clay at the south eastern corner of the feature was excavated first as it was the most clearly defined surface deposit. By cutting a section through this deposit a raised 235 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT T i l l II B ^r 8% •.' • C * * A I rf &m&& &\ t/> a N) cs K G O 7 CsO Feature 17 D ^ 2m oa w vr-TT Section 1 Fig. 5. Plan and Sections of Feature 13/14, Site 18. hearth approximately 1.5 m. wide was exposed. This was constructed from red brown clay laid on a bed of tabular flint nodules on top of chalk rock which had been scraped flat, see Fig. 5, 'A'. Continuing the section back revealed a very compact chalk deposit at the northern side; this was initially interpreted as a natural deposit. Running parallel to the hearth on the north side the narrow raised 236 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 strip of rammed chalk was approximately 40 cm. wide and 130 cm. long, see Fig. 7, 'B' . The deposit was laid on chalk rock running east-west and was differentiated by being of a softer consistency and slightly more yellow in colour. This deposit has subsequently been interpreted as a rammed chalk foundation or wall similar to that found at Merston church (Allen 1957). A sharp break of slope from the surface beyond the rammed chalk defined the edge of a steep cut, see Fig. 5, 'C. The chalk deposit divided the raised hearth area from the cut. As no back edge had been established, a section was left running across the hearth and excavation continued to establish the extent of the feature, leaving a further section (section 2). The stratigraphy in sections 1 and 2 implies that the feature filled up after abandonment either through deliberate backfilling or natural decay. When fully excavated, feature 13/14 was shown to consist of a large cut in the natural chalk rock approximately 4 m. wide by 5 m. long with a maximum depth of 0.70 m. Area B appears as an evenly cut area with a fairly flat base. A raised platform was left for the hearth on which the rammed chalk foundation was built. Possibly associated with the hearth and the narrow raised area of rammed chalk were two post-holes on the south westerly side of the hearth, L and M. L was a shallow small hole approximately 10 cm. in diameter and 16 cm. deep, M was a larger slightly angled cut approximately 28 cm. in diameter and 30 cm. deep. On the southern edge of the feature 13/14 was a large hollow cut into the side of the feature (F). Opposite on the northern edge of 13/14 was a further flat-bottomed depression (E). Further possible post-holes were observed on the north western side of the feature G and G. G being approximately 20 cm. in diameter and H approximately 12 cm. in diameter. Both holes were observed either side of a shallow gully 5 cm. at its deepest which led away from the feature into Feature 17, a shallow depression. A large cut was found on the south-westerly side of the feature, approximately 90 cm. in diameter and 60 cm. deep cut into the chalk rock bottom, see Fig. 5, 'D' This feature undercut the sides for approximately 40 cm. from the edge, its lower fill being a layer some 20 cm. deep of dark ashy or organic fill; this is currently undergoing analysis. Next to this pit and running along the southern edge for 120 cm. was a shallow ledge which became a gentle slope at the end nearest the pit, see Fig. 5, 'K'. Possible interpretations of feature 13/14 Because of the excavation method there is insufficient evidence 237 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT with which to reconstruct complex stratigraphic relationships within the feature and the deposits which were contained within it. The pottery obtained from the excavation related to demolition and decay deposits. It is difficult to make firm conclusions about the phases of construction and decay of the feature. We only have a general idea of its time-span, beginning in the late twelfth century and eventually going out of use in the late thirteenth or early fourteenth century. The feature itself is located at the southern end of a ditched enclosure typical of the period; a similar farm complex was recently excavated at Monkton 3 km. south-west of the site (Piatt 1978, Bennett and Williams 1997). The feature is segregated from the robbed foundation for a stone building by a contemporary east-west ditch. This may imply a social or functional segregation of the feature. The location of the hearth and the presence of possible structural elements certainly lead to the conclusion that feature 13/14 was an element of more complex building. Burnt clay daub found scattered throughout the fill of feature 13/14 and visible on the surface after machining suggests that the upper parts of 13/14 were of a wattle and daub construction. Area F, the rounded cut feature is possibly a support for a large post to support the external walls or roof. Area E could also be a recessed post base but this is not certain because it is not as clearly defined a feature as F. Comparisons with the sunken featured buildings at Mucking and West Stow (Hamerow 1985, 1991), show some structural similarities. GH 39 at Mucking, shows post holes cut into the side of the excavated area in a similar way to areas E and F. The hearth at the south eastern corner of Feature 13/14 is a complex construction, the flints in the base are all of a similar size and laid to form a compact surface with a thin silt matrix. The chalk rock below the hearth had been cut and scraped flat to form a surface for laying the flints. Such a labour intensive construction implies longevity and durability. The flint base suggests the possibility of periodic maintenance, when the clay surface became worn it only needed to be renovated as far as the flint bed and a new clay surface laid over the top. This suggests a sustained use rather than a one-shot process requiring the relaying of the new hearth. Additionally, the flint nodules could have acted as a heat store for the hearth enabling it to sustain a steady temperature over a longer period. There is no evidence of a similar type of hearth in a domestic context in a sunken feature building. With the exception of the ashy layer in section 2, none of the excavated deposits suggest an obvious function for the hearth such as pottery or ironworking waste; domestic hearths being more commonly centrally located. It is certain, however, that a heat 238 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 source was a valuable resource for secondary agricultural processes such as baking drying or brewing. Examples of buildings of this type have been identified at Rochester, New Cemetery, and at Coppergate, York (Hall 1987), a semi-sunken malting kiln and a wattle built bake house respectively. The close proximity of the post-holes, L and M, suggest that they may also be associated with the rammed chalk foundations perhaps as props for a screen or as part of a construction to separate the hearth from the remaining space. The chalk deposit, 6 of section 1, is of a similar rammed chalk material. It is possible that this is a structural element built to narrow the entrance to ' C from the remainder of the sunken cut, perhaps as a way of channelling air. It is also possible that the deposit is part of an external chalk foundation which slumped into 'C when the building decayed. While it is possible that J and Feature 17 could have been formed naturally, it is also possible that they were part of a structure which acted as a form of external flue for the hearth. It maybe that post-holes H and G supported a light screen which could be opened and closed to regulate the flow of air to the hearth. The relationship of feature D to the building is unclear, no pottery of any kind was observed in the fill or in the section 2. The fill of the pit was characteristically different from the deposits backfilling the main body of the sunken feature being composed of a very fine soft brown loamy deposit with occasional chalk inclusion. At the bottom of the pit a very fine grey ashy layer was observed. Altogether this suggests that the fill of the pit was originally organic and was already backfilled or silted up when the building began to decay. Its purpose could be that of a raking pit, into which debris contained within the sunken of the feature was raked. The fill of post-hole K is stratigraphically above the backfilling deposits, showing that it is a later structural feature imposed after the feature was backfilled possibly part of a later boundary or fence line suggesting a continued occupation of the site. There are many examples in the archaeological record of buildings with sunken structural elements in the region of the site. Recently Roman storage huts with sunken floors have been excavated at Monkton (Bennett and Williams, 1997). Sunken floors have been used to serve a number of structural and spatial purposes. It seems likely that the hearth that dominated this feature required a sunken floor as a design solution to its construction or its function. EMMA BOAST 239 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT a rtMJXm+^^T ve-« Feature I I I E i >« I i Feature 7 I I Feature 5 g i * I s I J 3 Presumed limit of foundation ^ < I 5m Fig. 6. Site 18, Feature 7 building foundations. APPENDIX 2: THE LITHICS A total of 175 flints were collected during the archaeological programme at the site of Kent International Business Park. Of these 7 are natural, 3 are burnt and unworked and a further 4 are burnt and unidentifiable; these pieces will not be considered any further during the course of this report. Flint artefacts were recovered in varying quantities from sites 4/5, 6, 7,15/16, 18 and 19. The distribution and composition of the assemblage is summarised in Table 1. 240 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 TABLE 1: FLINT QUANTITIES BY SITE AND CLASS Site 4/5 6 7 15/16 18 19 Total Blade - - - 1 2 - 3 Chip - - - 2 - - 2 Chunk - - - 5 - - 5 Core - - - 1 - - 1 Flake 4 5 2 117 7 3 138 Retouched 4 - - 7 1 - 12 Total 8 5 2 133 10 3 161 There are no apparent distinctions between the artefacts retrieved from the different sites and therefore the material will be considered as one assemblage. Raw Materials The types of raw material used throughout this assemblage vary. A black semi-translucent flint, a grey-brown semi-translucent flint and glauconitic (Bullhead) flint are all well represented. Some variations of the black and the grey-brown flint contain grey, opaque almost cherty inclusions. A small number of other types are also present in lesser quantities. All of the material used has a hard weathered cortex. Some corticated naturally shattered pieces have also been used. Condition The majority of the assemblage is in a fresh condition, a number of pieces have slight cortication, few are heavily corticated. In addition a small quantity are burnt. Little under half of the artefacts are incomplete or damaged. The damage consists mainly of irregular chipping along the edges of the pieces and would not be inconsistent with such damage as that caused by farming practices. Technology Only one core is present within this assemblage; it is fragmentary but appears to have been struck from at least two directions. In general, there is little evidence of core preparation and no evidence of 241 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT attempts at core rejuvenation. Examination of the flakes indicates that hard hammer was used almost exclusively. The exception are a small number of pieces with dihedral or faceted striking platforms and others which display some platform abrasion. In addition there is a small number of flakes which are blade-like. A total of 12 retouched pieces were collected, these are summarised in table 2. TABLE 2: RETOUCHED FORMS Class Borer Denticulate Knife Notched Flake Scraper Retouched Flake Quantity 1 1 1 1 4 4 Most of the retouched pieces are of fairly simple form with rather limited retouch. Of particular note is a wedge-shaped piece, retouch is present on all of the edges. One edge is blunted by abrupt retouch, the opposite of which has denticulated retouch. All of the scrapers are retouched at the distal end only. Three have semi-abrupt retouch, one of which from site 15/16 appears to be worn and very smooth. One scraper is formed on a blade with abrupt retouch at the distal end. Discussion It is likely that raw material was collected from the surface within the vicinity. The weathered cortex and the use of shattered nodules suggests that little effort was made in the selection of material and the consideration of quality. The quantities of flakes and associated waste suggest a degree of knapping taking place within the vicinity. Only one core, however, was recovered. A search for con-joins was made with the material recovered from F14, none were achieved. This deposit alone may represent the deliberate disposal of knapping waste. There is no evidence of any knapping in situ, however, and it is likely that the assemblage is largely derived. 242 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 It is likely that this assemblage is associated with the Middle to Late Bronze Age activity on the site. The lack of preparation and minimal retouch would be consistent with material of this period. There is, however, not enough evidence to indicate knapping of any scale occurring on site at this time. The artefacts which do demonstrate some preparation, those with prepared striking platforms, are probably indicative of an earlier phase of activity. In addition, the end-on blade scraper and a second well formed scraper, which has a greater degree of cortication than the bulk of the assemblage, may be of Neolithic date. TANIA WILSON APPENDIX 3: THE MEDIEVAL POTTERY Introduction An overall total of 2,357 sherds (weight 34 kg. 027 gms.) were recovered during the excavations. This total sub-divides into the following broad 5 ceramic phases: Ceramic Phase 1 Ceramic Phase 2 Ceramic Phase 3 Ceramic Phase 4 Ceramic Phase 5 ?Early-Mid Bronze Age Beaker or Urn 2 Late Bronze Age Deverel-Rimbury 175 Late Iron Age 1607 Early medieval-medieval 527 Post-medieval 1 Despite the presence of two early Saxon beads no ceramic evidence was recovered for the entire period c. A.D. 75-1050, though their presence ought to suggest a degree of activity at some time during the fifth-sixth century A.D. The implications of the recovered lithic assemblages and the first three ceramic phases will be dealt with in a forthcoming article; the present notes are a summary of the main pottery trends associated with two Phase 4 enclosures 4/5 and 18. Discussions of the parallels and dating can frequently be lengthy and tedious. The necessary justification for the dating given in the two phase-based vessel types and frequency tables and the text are held with the author and as part of the site archive. 243 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT The medieval Pottery In terms of individual site and period totals, the recovered excavated total of 648 post-Roman sherds sub-divides into: Site 4/5 Early medieval-late medieval wares 183 Site 18 Early medieval-late medieval wares 465 With one or two exceptions noted below all this material falls into the general broad approximately 200-year dating given to Ceramic Phase 4: c. A.D. 1075-1350/1375. Despite the potential opportunity neither funding nor time was available to excavate totally either of the two enclosures represented by these sites; recovered site totals are therefore partial. As a result assessment and graphic presentation of sherd discard patterns and quantities are meaningless; only brief summaries are given below. However, with the exception of 4 vessel rims from Enclosure 4/5's ditch (which are broadly paralleled by material from Enclosure 18), most drawable items are presented in Figs. 7-10 together with an aide-memoire enclosure plan reproductions labelled with significant source contexts. /. Enclosure 4/5 (Fig. 7): 1.1 Fabric types, period-relayed quantities, vessel numbers Early medieval: Unsourced shell-filled ware with 4 sherds 1 vessel coarse ?beach sand: Canterbury sandy ware: 11 sherds 6 vessels (incl. no. 1) Early medieval/medieval transition: N. Kent (E. of Medway) shell-filled 37 sherds c. 8 vessels sandy wares (3 vars): (incl. nos. 3, 5-6) Tyler Hill (Canterbury) shell-dusted 1 sherd lvessel(incl.no.2) sandy ware: Tyler Hill sandy ware: 1 sherd 1 vessel Medieval: Tyler Hill shell-dusted sandy ware: 96 sherds 7 vessels (incl. no. 4) Tyler Hill sandy ware: 33 sherds 9 vessels 244 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 •3 o a i "3 > o •3 u 6 "S w 60 51 u. CO « 245 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT Overall 5 main fabrics are represented, 4 if early Medieval Canterbury sandy ware is considered as sourced to the Tyler Hill area. Canterbury products dominate throughout all chronological sub-phases; in common with regional trends shelly wares are a significant component of the later twelfth-early thirteenth phase. No fineware imports were recovered. Overall no more than 35 vessels are represented. 1.2 Sherd and vessel distributions. In terms of recovered sherd quantities, only moderate amounts were recovered from ditch fills: 70 sherds compared with 113 from internal features; for the latter the bulk came from Pit F33 (81 sherds, mostly body sherds representing two vessels). For the remaining internal features sherd totals average up to 10 per feature, usually less; context within the rectangular structure F42 collectively produced only 15 sherds. However, in terms of estimated vessel quantities, c. 14 vessels are represented by material from internal features and 20 from ditch fills. The following table summarises frequencies per vessel class for each chronological sub-phase (based on identifiable individuals): Phase Pre-c. 1150A.D. c.l 150-1175 A.D. c.l 175-1250 A.D. Post--^v ^ 17 W a to S CD tZl H r t/3 o I o Fig. 8. Early medieval-medieval pottery from Enclosure 18 (1:4 scale). F10 I F13/14 F11 « 19 F 2 / ~ r^. 22 to 3 I CO c co co i Co Fig. 9. Medieval pottery from Enclosure 18 (1:4 scale). D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT F10 x: = \ 26 F11 ? 27 fe 28 \ Fig. 10 Early medieval-medieval pottery from Enclosure 18 (1:4 scale). Medieval: N. Kent (E. of Medway) shell-filled 14 sherds 1 vessel sandy ware: Tyler Hill shell-filled sandy ware: 2 sherds 1 vessel (no. 28) Tyler Hill shell-dusted ware: 9 sherds 7 vessel (inc. nos. 8, 20, 26) London-type ware: 1 sherd 1 vessel Tyler Hill sandy ware: 410 sherds 79-89 vessels (inc nos. 7, 9, 11-19,21-25) Again Canterbury products dominate throughout; north Kent shellfilled wares are a minor component of the overall assemblage. One fine-ware jug is represented by the imported London Ware sherd. Collectively, between 106-117 vessels are represented. 2.2 Sherd and vessel distributions The unusual nature of the internal structures dictated excavation strategy. Internal features were sampled more thoroughly than the enclosure ditch with the result that original discard patterns cannot be assessed accurately. However, in terms of sherd quantities, as recovered, the highest quantities come from the backfills of Structure F7 and the ?robber pits Fl and F5 (168, 139 and 99 sherds, 250 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 respectively). The remaining features averaged 10 sherds or less; the sunken-floored building F13/14 contained only 24 sherds. In terms of recovered vessel quantities the majority came from the Fl, F5 and F7, the relatively high quantity of vessels involved, frequently represented by single sherds, ensures that these features were backfilled with domestic rubbish. Structure F13/14 contained a total of 4 vessels most of which should represent abandonment discard. There are two instances of sherds from the same vessels occurring in different contexts. Sherds from the same jug as no. 16 occur in Pit F5 and Structure F7; sherds from a later fourteenth-century jug are scattered between features Fl, F5 and F7. Associated implications are discussed in 2.4 below. The following table summarises frequencies per vessel class for each chronological sub-phase (based on identifiable individuals): Phase Pre-c. 1150 A.D. c.l 150-1200 A.D. c. 1200-1250 A.D. c. 1250-1325 A.D. c. 1325-1375 A.D. Pitchers/ Jugs - - 5 15 3 Dishes - - . . - Bowls - - 7 9 - Cookpots 1 2 13 9 1 Storage jars - 72 ?1 . - Totals 1 3 ?25 33 4 2.3 Chronological and site-related implications Again confident commencement dates for the enclosure are qualified by unavoidable limited excavation. The presence of pre-1150 A.D. sherds (particularly the c. 1075-1100 A.D. rim from hollow F4) could imply an eleventh-century construction-date but the quantities are so low and heavily worn that they could equally well represent residual material arriving via pre-enclosure manuring practices. We are on firmer ground with later twelfth-century material; the quantities are still low but evenly distributed throughout the enclosure, and best represented by no. 27 from Fl 1. If we cannot be confident regarding a later eleventh-century commencement, the enclosure was certainly in existence between c. 1150- 1175. Gully Fl 1 is the earliest datable internal feature which, with no. 28 in mind, remained open, if not in regular use, at least until c. 1250 A.D. There is an initial dichotomy in dating the end of Structure SI 3/14. The London Ware sherd is more wisely dated fairly late, but it could be a c. 1225 product; conversely the possible shell-filled storage-jar 251 D.R.J. PERKINS, E. BOAST, T. WILSON &N. MACPHERSON-GRANT sherds could be dated later but, in Canterbury at least, are much more typical of mid-later twelfth-century products. Though the latter comes from post-use infill layers, its relatively fresh condition and quantity suggest that its deposition is unlikely to be significantly later than its manufacture date. If we accept the Canterbury evidence for the ware concerned (Canterbury fabric EM3A) discard into S13/14, and the latter's abandonment, should not be much later than c. 1225, with the London Ware sherd perhaps arriving somewhat later into an already abandoned structure. Alternatively if the whole sherd package recovered was dumped in one go, and the shelly sherds referred to somewhat later than city trends consistently indicate, the size and relatively unworn condition of the overall assemblage suggests fairly rapid burial; this scenario would suggest initial abandonment between c. 1250-1275. However, it is felt that this is unacceptably late, particularly since there is, unlike Fl, F5 and F7, a clear absence of obviously later material. Abandonment by c. 1225 A.D. is preferred. The above data are slightly inconclusive but the presence of the shelly EM3A does imply that S13/14 should have been in existence by c. 1175 at least. Structure F7 cannot be dated closely; it could date from the late eleventh-century or be contemporary with the site's increase in mid-later twelfth-century pottery. That it was in existence before c. 1200 is implied by the earlier-mid thirteenth-century sherds (including the greater part of the highly fragmented bowl 23) from its robbed foundation trenches. Much of this material consists of relatively fresh and large sherds, fired to temperatures that do not survive long-term exposure; most are likely to have been sealed quickly and their presence should date some of the foundation robbing to between c. 1225-1250 A.D. at latest. Both pits Fl and F5 have relatively small quantities of abraded or fresh earlier-mid-thirteenth-century pottery; in view of the following comments this material (at least for Fl) must be derived from adjacent sections of the already robbed F7 foundations, introduced at a subsequent time when, whatever Fl and F5 originally represented, were also removed. Features Fl, F5 and F7 all have two ceramic trends in common: considerable quantities of later thirteenth-earlier fourteenth-century pottery and a small quantity of either later fourteenth- or fifteenth-century pottery. The former is consistently smaller and more abraded than the pre-1250 A.D. pottery associated with the building's probable dismantling; in addition sherds from the same jug (no. 16) from both F5 and F7 indicate contemporary discard for at 252 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 least some of this material. The later sherds, though harder-fired, are unabraded and were almost certainly buried shortly after discard. In the first instance the discard pattern for the earlier material implies that, whatever the sequence of structural robbery, Features 1, 5 and 7 remained open to receive later rubbish. The numbers of vessels involved, sherd sizes, similar wear patterns and generally small numbers of sherds per vessel represented are very marked; this same trend seems to also apply to Features 2, 4 and 10. Whilst some of this material may represent accumulation and discard during continuing later thirteenth-century or later occupation of the enclosure (and sherd quantities and sizes for jug 7 could encourage this possibility) the overall similarities suggest that much of this later rubbish did not necessarily accumulate gradually but was all dumped at the same time. If this scenario is correct, when? The key aspect here is that the bulk of the material is abraded; this scenario would imply that it was allowed to weather in unsealed dump contexts for a reasonable period of time following a general discard date of no later than c. 1325/1350 A.D. But where do the harder-fired later sherds fit in? This dating cannot be resolved on the available evidence; whilst they might just be contemporary with the earlier fourteenth-century material, it is felt that they are more likely to represent later fourteenth-century or earlier fifteenth-century losses. Whichever, they may have arrived much later than the c. 1325/1350 A.D. date suggested above, either as fifteenth-century rubbish into still open contexts which were then sealed, or as part elements in a later fourteenth-century general site clearance assemblage which included weathered earlier material (swept up from the same or brought in from another site) following, since it has been suggested above, plague and settlement decimation. 3. Overall Summary Enclosure 4/5 certainly originated around c. 1075 A.D. Enclosure 18 may have a similar construction date but the evidence is unclear; it was certainly in existence by c. 1150 A.D. Irrespective both have low ceramic counts until c.l 175/1200. Both appear to share structural changes which took place around c. 1225 A.D.: F42 in Enclosure 4/5 may have been replaced by other buildings such as F1, Structure F7 in Enclosure 18 was being dismantled and the sunken-floored building S13/14 was probably being backfilled at this time. Enclosure 4/5's occupation appears to markedly decline after c. 1250 A.D. and ceramically appears to terminate by c. 1300 A.D. , if 253 D.R.J. PERKINS, E. BOAST, T. WILSON & N. MACPHERSON-GRANT not earlier. It is precisely this period that witnesses the marked surge in post c. 1250 pottery from Enclosure 18. The latter is a difficult entity to interpret: Structure F7 may not be domestic and neither may S13/14. Though half the enclosure remained unexcavated and domestic buildings may have been unrecorded, its internal layout appears markedly different from the probably domestic function of Enclosure 4/5. The limited quantities for this period from Enclosure 4/5 may be due to sampling bias but, in part, it is tempting to see the low post-1250 A.D. ceramic count not so much as a result of occupational decline, but as a direct by-product of the dismantling of structural elements within Enclosure 18 which provided convenient disposal points for at least some of 4/5's later thirteenth-century pottery. Irrespective both enclosures appear to have ceased functioning at much the same time somewhere around c. 1325 A.D. with an arguable final demise for Enclosure 18 at some point between c.l 375-1450 A.D. N. MACPHERSON-GRANT REFERENCES Allen, A.F., 1957, 'The lost village of Merston', Arch. Cant, lxxi (1957), 198-200. Bennett, P. and Williams, J., 1997, 'Monkton', Current Archaeology, 151 (1997), 258-264. Cobbett, W., 1854, The Rural Rides (various editions), 'From Dover to the Wen, Thursday Afternoon, September 14th' (1854). Cook, N., 1933, Arch. Cant., xlv, xliii (1933). Crawford, O.G.S., 1933, Antiquity, vii (1933), 25, 92, Plate 8. Hall, R., 1987, 'The Viking Dig' mMed. Arch., 31 (1987), 159, fig. 154,128. Hamerow, H. 'South west Stow, the Anglo-Saxon village', East Anglian Archaeology, 24 (1985); 'Excavations at Mucking II, the Anglo-Saxon Settlement', English Heritage Archaeological Report 21 (1991). Hillier, J., 1889, 'A British Village at Ramsgate', Arch. Cant., xviii (1889), 1-4. Hurd, H., 1909, 'On a Late Celtic Village near Dumpton Gap, Broadstairs', Archaeologia, lxi (1909), 427-38. Jessup, R.F. 1943/44, 'A Bronze Age Hoard from Sturry, Kent', Ant. Journ., xxiii (1943), xxiv (1944), 148. Lewis, J., 1723, History of Thanet (1723, 2nd edn. 1736). Perkins D.R.J., 1985, 'The Monkton Gas Pipeline', Arch. Cant., cii (1985), 53. 254 KENT INTERNATIONAL BUSINESS PARK, MANSTON 1994-1997 Perkins, D.R.J., 1989, 'The Selling-Thanet Water Main', Arch. Cant, cvii (1989), 276. Perkins, D.R.J., 1991, 'Sarre first time drainage schemer', in Thanet Archaeological Trust: Two Reports, 8-15. Perkins, D.R.J., 1992, Archaeological Sites in Thanet: their need for protection, Thanet Trust report to English Heritage. Perkins, D.R.J., 1995a, Arch. Cant, cxv (1995), 'Researches and Discoveries', 471. Perkins, D.R.J., 1995b, 'The Sandwich Bay Wastewater Treatment Scheme', Arch. Cant, cxv (1995), 263. Perkins, D.R.J. 1995c, Arch. Cant., cxv, 'Researches and Discoveries', 472 Piatt, C, 1978, Medieval England, 58. Rigold, S., 1962, 'Excavation of a Moated Site at Pivington', Arch. Cant., lxxxvii (1962), 27-47. Rigold, S., 1976, 'A Moated Site at Moat Farm, Leigh, Kent, The Significance of Building "A" ', Arch. Cant., xcii (1976), 177-82. Rigold, S., 1982, 'Medieval Archaeology in Kent', in Archaeology in Kent to A.D. 1500, CBA Research Report no. 48, 84. Roach-Smith, C. 1854, Collectanea Antiqua, iv (1854), 65. Tatton-Brown, T., 1977, 'Kent Manorial and Moated Sites Group report', Arch. Cant., xciii (1977), 219. 255

Previous
Previous

Worked Whale Vertebrae

Next
Next

Nineteenth Century Sandwich