A Mesolithic Site at Tong's Meadow, West Street, Harrietsham

291 A Mesolithic site at Tong’s Meadow, West Street, Ha rrietsham paul riccoboni and dan swift with contributions by Lucy Allott, Chris Butler, Susan Pringle, Louise Rayner and Lucy Sibun Following on from archaeological evaluation Archaeology South-East carried out an excavation and watching brief in two targeted areas at Tong’s Meadow, Harrietsham. The excavations revealed evidence of Mesolithic/ Neolithic and Late Iron Age/Romano-British archaeology. Three-hundred and seventy-six worked flints largely of Late Mesolithic and Early Neolithic date were collected. The range of tools and presence of debitage within the assemblage suggest that a longer-stay base camp was situated on the site. Additionally, Romano-British agriculture and a masonry building in the area are suggested. Tong’s Meadow is located on the northern outskirts of Harrietsham village on a gradual slope that drops southwards, from c.89m to c 84m od (Fig. 1; NGR TQ 861 527). The underlying geology comprises Folkstone Beds overlain in places by River Gravels (British Geological Survey 1976). This was recorded at the site as sand, gravel and clay and was in certain areas degraded and variable and included unworked naturally deposited flint nodules. Planning consent for the building of a new school was granted by Maidstone Borough Council under the condition that a programme of archaeological work be conducted prior to any construction. Accordingly, Archaeology South-East (Centre for Applied Archaeology University College London), were commissioned by Kent County Council to undertake firstly a 14 trench evaluation (ASE 2002), and subsequently targeted excavation in Areas A and B, with a watching brief in Area E (ASE 2006; Fig. 2). The archaeological methodology was a standard machine-strip and map, hand-clean and excavation procedure as specified by KCC Heritage Conservation Group. Features were extensively sieved and sampled for 292 Fig. 1 Site location. 293 Fig. 2 Site Plan. PAUL RICCOBONI AND DAN SWIFT 294 microliths, debitage and environmental remains. Contexts are referred to in the text within [ ] brackets, prefixed by area A, B, and E or with no prefix denoting evaluation contexts. Late Mesolithic/Early Neolithic remains are known from the area with a quantity of scrapers, microliths and waste flakes found in and around Harrietsham, above the River Len, and from the surrounding area (Wymer 1977; Scott 2004). Late Iron Age and Romano-British activity is also evidenced in the vicinity. An archaeological evaluation at Glebeland, Marley Road, Harrietsham, recorded occupation dating from the Late Iron Age to the fourth century ad and, after a period of abandonment, fifth- to seventh-century buildings. All recorded archaeological features were sealed beneath layers of topsoil and subsoil with some evidence of reworking through ploughing in both areas; however, no disturbance to the substrate was detected. In the topographically slightly lower parts of Area B, a colluvial deposit was recorded beneath the subsoil; however, this also sealed all archaeological features where it occurred. Late Mesolithic / Early Neolithic (Figs 3-4) Numerous rounded, sub-rounded, oblong, ovoid and lobate features were recorded across the site although interpretation of these is problematic. Whilst some may be naturally formed, others may be representative of tree clearance episodes and tree-throws, or of small quarry pits for flint. Furthermore, these sub-surface hollows may subsequently have been used for flint-working and shelter and filled up slowly with local debris from the encampment, possibly after its desertion. Eight of the hollows contained evidence of flint-working in the form of debitage, finished pieces and cores, but most contained no dating evidence at all and only varying proportions of burnt flint and charcoal. From the features with worked flint, 376 mostly Late Mesolithic and Early Neolithic pieces were collected. The range of tools and the presence of debitage within this assemblage suggest that the site was used as a longerstay base camp. The eight features [A/52], [A/58], [A/114], [A/116], [A/118], [A/120], [A/122], [A/124] containing flintwork were all situated in the topographically higher southern-middle part of Area A suggesting that this was an area of focused activity (Fig. 3). Other features of similar form were recorded across areas A, B and E (Figs 3 and 4). Late Iron Age / Romano British (Figs 3-4) A ditch [A/04] orientated south-west/north-east across the northern part of Area A was recorded in seven sondages. The ditch had a shallow A MESOLITHIC SITE AT TONG ’S MEADOW, WEST STREET, HARRIETSHAM 295 Fig. 3 Area A Plan. PAUL RICCOBONI AND DAN SWIFT 296 Fig. 4 Area B Plan. A MESOLITHIC SITE AT TONG ’S MEADOW, WEST STREET, HARRIETSHAM 297 rounded profile and its fill contained Roman pottery dated ad 45-70 and a Roman tegula and brick fragment. Traces of another smaller ditch [A/13] ran away from this at right-angles, but when investigated this was found to be very short and shallow. Two further ditches, recorded in Area B [119] and [B/93] in the southeastern part of the site did not produce any dateable finds but share a similar alignment to [A/04] and are probably contemporary. Both of these ditches had a similar shallow profile to [A/04]. Ditch [B/93] terminated in a butt-end near to the edge of Area B. However, ditch [119] was only detected in evaluation trench 4 and was not visible in the excavation despite hand-cleaning and exploratory sondage-excavation in the area it was projected to appear (i.e. west from the location of trench 4). Medieval and post-medieval Six wheels ruts were recorded in the north-western corner of Area B (Fig. 4). Several sondages [B/104], [B/106] and [B/108] were excavated through the ruts, which were loosely filled and appeared very recent. These are probably cart tracks relating to nineteenth-century sand quarrying known to have taken place in three separate areas of the new school. Prehistoric Flintwork by Chris Butler An assemblage of 376 pieces (1.86kg) of worked flint was recovered which due to the careful excavation and recovery methods used, included a substantial number of very small pieces (Table 1). The raw material comprised a number of different types of flint: 1. Most predominant, a mottled grey flint with buff coloured cortex; varying from a light grey through to a very dark grey, with variation sometimes visible on the same piece. 2. A well-patinated light blue-grey flint with a light buff cortex 3. A dark blue-black coloured flint with a smooth off-white coloured cortex. 4. Bullhead flint (two examples); one white patinated piece from a chalk source; and a few small pieces of orange-stained flint. Debitage: the majority of the assemblage comprised debitage, with some 40% being small chips removed as a bi-product of flaking, preparation or retouching, normally less than 10mm in size, and frequently with all the attributes of a flake (Butler 2005). In addition to the chips, there were numerous small flakes, fragments and shattered pieces and other debitage comprised both hard and soft hammer-struck flakes, together with smaller numbers of blades and bladelets, and associated fragments. Soft hammerPAUL RICCOBONI AND DAN SWIFT 298 struck pieces made up some 37% of the flakes, blades and bladelets. Many of the pieces exhibited evidence of platform preparation with previous regular removals on the dorsal side, suggestive of a systematic knapping strategy. A number of pieces appear to have been removed from the core with the aid of a punch. All four complete cores were two-platform cores, with the platforms at right angles to one another (for example Fig. 5, 1) and sometimes exhibiting evidence for the removal of both flakes and blades from the same core, although from different platforms. Evidence for core rejuvenation comprised a core tablet (Fig. 5, 2), a crested blade (Fig. 5, 5) and a core rejuvenation flake; in addition, one of the fragments may also be from a crested blade. The flint-knapping technology is typical of the Mesolithic period. The Implements: only four implements were recovered from the fieldwork. These comprised two scrapers; an end scraper on a punch-produced flake (Fig. 5, 4), and a side scraper on the slightly concave lateral edge of a long flake (Fig. 5, 3). A single obliquely truncated microlith (Fig. 5, 6) with additional retouch on the leading edge (Jacobi Type 1b) was Table 1: The Fli ntw ork Hard hammer-struck flakes 73 Soft hammer-struck flakes 28 Soft hammer-struck blades 4 Soft hammer-struck bladelets 8 Blade fragments 7 Bladelet fragments 9 Fragments 74 Chips 149 Shattered pieces 8 Spall 1 Core tablet 1 Crested blade 1 Core -rejuvenation flake 1 Cores 4 Core fragments 4 End scraper 1 Side scraper 1 Microlith 1 Pick/adze 1 Total 376 A MESOLITHIC SITE AT TONG ’S MEADOW, WEST STREET, HARRIETSHAM 299 2 3 4 7 5 6 0 5cm 1 Figure 5: Selected flintwork Fig. 5 Selected Flintwork. PAUL RICCOBONI AND DAN SWIFT 300 recovered from [53], and a small pick/adze from Area B. The pick/adze is manufactured on a cherty piece of flint, and has a tranchet flake removed from one end with additional retouch on the opposite side, together forming its working end, and abrasion on its lateral edges, presumably from hafting (Fig. 5, 7). In addition to the formal implements, there was also a single retouched soft hammer-struck flake of Bullhead flint, which may have been used as a scraper, and a blade fragment, which had been blunted by abrupt retouch on one lateral edge, and had evidence for possible utilisation on the opposing edge. A second blade had possible utilisation damage on one lateral edge, with the opposing edge being cortical which would have enabled it to be comfortably held. Mesolithic flintwork groups: [A/52] 112 pieces of relatively fresh and un-abraded worked flint, including flakes, blades and bladelets, together with 32 chips and 32 fragments recovered from wet-sieved soil samples. A single core, a core fragment, together with a crested blade and a core tablet were also recovered from here. The only implement to be found in this context was a microlith, although a blade fragment may have been utilised. Four of the pieces are fire fractured. [A/58] 17 pieces of struck flint, all of which was debitage. Only three pieces (bladelet fragments) were diagnostically Mesolithic in character. However, the majority of the assemblage comprised chips and fragments, so all are likely to be Mesolithic. [A/114] 43 pieces of worked flint, of which 30 were chips. Four pieces, including a bladelet and some blade fragments, were diagnostically Mesolithic, but most are probably of this date. [A/116] 28 pieces of worked flint, including 20 chips. Three pieces, a soft hammer-struck blade (possibly utilised) and two flakes, were diagnostically Mesolithic but the entire assemblage is probably of this date. [A/118] 22 pieces of worked flint, including 15 chips. Four pieces, a bladelet and three soft hammer-struck flakes, were diagnostically Mesolithic. Prehistoric and Roman Pottery by Louise Rayner A small assemblage of prehistoric and early Roman pottery was recovered, totalling 109 sherds (879g). The majority of this was recovered from slots through ditch [A/04], dating to the Late Iron Age/early Roman period. A single, probably prehistoric, abraded flint-tempered sherd was recovered from slot [A/65]. The fabrics and forms present in the ditch [A/04] are mostly grog-tempered and sandy ware jars. A few diagnostic sherds A MESOLITHIC SITE AT TONG ’S MEADOW, WEST STREET, HARRIETSHAM 301 are present and these fabric types have their origin in the Iron Age but continue in use through to the late first century ad. The few rim sherds (necked and beaded-rim jars) suggest an early Roman date of c.ad 45- 70. Sherds in [A/11] appear to largely derive from one grog-tempered jar suggesting the source of this material is within the nearby vicinity and the material has not been extensively re-deposited before becoming incorporated in the ditch fills. Discussion by Chris Butler and Dan Swift Although only 64 pieces of flintwork in the recovered assemblage have the diagnostic characteristics of Mesolithic flintwork (Butler 2005), it is likely that the entire assemblage is of this date. The characteristics of a small number of pieces are more like those expected in the Early Neolithic period, and it is therefore possible that this is a Late Mesolithic – Early Neolithic transitional site. There are no obvious later examples in the assemblage, and although much did not exhibit typical characteristics of the Mesolithic, there is no reason to believe that any of these is of other date. The presence of so many chips in the assemblage suggests that flint knapping was taking place on the site. Other evidence for in-situ flint knapping comes from the cores and core fragments, together with the core-rejuvenation pieces. With so few implements however it is difficult to draw any conclusions about the function of the site, although the presence of both the microlith and the pick/adze and the range of debitage as well as the presence of scrapers, suggests that the site was not a temporary hunting camp, but part of a longer-stay base camp utilising the nearby streams and springs. A number of Mesolithic sites are known in the vicinity, which have produced quantities of flintwork (Wymer 1977). This is predominantly of debitage, but has also included microliths, including ‘Horsham Point’ types (Scott 2004), microburins, tranchet adzes and other implements, perhaps suggesting the presence of other longer-stay base camps in the area. An assemblage of Mesolithic flintwork similar to the Tong’s meadow assemblage came from Shorne Wood Country Park in north Kent where mostly debitage and a similar range of cores and rejuvenation pieces, together with a limited range of implements including single examples of a microlith and tranchet adze was found (Butler, forthcoming). There are also some similarities between the Tong’s Meadow site and the site at Streat Lane in Sussex where an assemblage of 3,000 Mesolithic flintwork was found in association with pits and quantities of burnt flint as well as possible evidence of a shelter (Butler 2007). The Harrietsham area has also produced a number of other Mesolithic find spots, but the flints found at Tong’s Meadow are unusual as many of them occur within features, which is a rare phenomenon. PAUL RICCOBONI AND DAN SWIFT 302 As discussed, it is not clear how the sub-rounded, oblong, ovoid and lobate features were formed and it is not beyond possibility that they may have formed naturally. Additionally, evidence from the environmental samples suggesting disturbance of the features must not be discounted. However, regardless of the formation processes at work, it is almost certain that the features containing worked flints were open during episodes of Mesolithic activity on the site, and could therefore represent remnants of small flint quarry pits or tree throws. Indeed, the very occurrence of these natural flints could be what determined the location of the camp in the first place. The site is well-appointed and in close proximity to the River Len which would have provided a route through the landscape for hunting expeditions which doubtless ranged along its course. Interestingly, the Mesolithic remains at Sandway and Streat Lane were situated just on or slightly below the brow of the terrace, which is very similar to the position of Tong’s Meadow. Such sites may have been preferred flint knapping locations, both for the availability of flint and for their position in the landscape. Enigmatic shallow features seem to be a common characteristic on the Folkestone Beds (Wessex 1999) so the Tong’s Meadow evidence fits in well with the developing model of Mesolithic sites in the area. The Romano-British ditches are clearly part of the agricultural utilisation of the local area and are undoubtedly linked to a network of ditches forming a field system for a local farmstead or villa. The finding of building material, in particular brick although in very small quantities, suggests that a building, possibly with a hypocaust existed in the vicinity. Additionally, it is quite possible that some of the other undated features across Areas A and B if not naturally occurring may also date from the Romano-British period. acknowledgements Archaeology South-East would like to thank Kent County Council, and Wendy Rogers and Adam Single (Heritage Conservation Group), for commissioning and facilitating the project. The excavation was supervised by Paul Riccoboni, assisted by Deon Whittaker and Sid Jeffrey and managed by Darryl Palmer. Specialist reports drawn upon here comprise: Worked Flint (Chris Butler), Prehistoric and Roman Pottery (Louise Rayner), Medieval and Post-medieval Pottery, Metalwork and Glass (Luke Barber) Ceramic Building Material (Susan Pringle), Environmental Samples and Charcoal (Lucy Allott). The illustrations were prepared by Hannah Faux and Justin Russell. The post-excavation work was managed and report edited by Louise Rayner. The site archive will be deposited with Maidstone Museum. A MESOLITHIC SITE AT TONG ’S MEADOW, WEST STREET, HARRIETSHAM 303 bibliography ASE, 2002, ‘An Archaeological Evaluation at Harrietsham School, Tongs Meadow, West Street, Harrietsham, Kent’, ASE Report no. 1523. ASE, 2006, ‘Archaeological Investigations at Tongs Meadow, West Street, Harrietsham, Kent, Post-Excavation Assessment and Project Design Report’, ASE Report no. 1938. British Geological Survey, 1976, Maidstone, Sheet 288, Solid and Drift Edition. Butler, C., 2005, Prehistoric Flintwork, Tempus Publishing Ltd, Stroud. Butler, C., 2007, ‘A Mesolithic Site at Streat Lane, Streat, East Sussex’, Sussex Archaeological Collections, 145, 7-31. Butler, C., forthcoming, ‘The Flintwork from Shorne’, unpubl. KCC report. Scott, R., 2004, ‘Kentish Evidence of the Palaeolithic and Mesolithic Periods’, in Lawson, T. and Killingray, D. (eds), An Historical Atlas of Kent, Chichester, Phillimore. Wessex Archaeology, 1999, ‘Archaeological Excavation at Sandway Road (ARC SWR 99) Nr Sandway, Kent; Final Interim Report’, WA Report No. 45997c, Salisbury (Union Railways (South) Ltd. Wymer, J.J., 1977, Gazetteer of Mesolithic sites in England and Wales, CBA Res. Report no. 22.