Excavations at Margetts Pit, Burham: large-scale manufacture of artefacts in late Bronze Age/early Iron Age
excavations at margetts pit, burham:
LARGE-SCALE MANUFACTURE OF ARTEFACTS in
late bronze Age/Early Iron Age
erica gittins
Excavations by Wessex Archaeology at Margetts Pit, a 1.9 ha area to the north-west of the village of Burham on the east side of the Medway valley on the lower slopes of the North Downs (NGR 572050 162220; Fig. 1), revealed evidence for activity dating from the Neolithic until the Saxon period. Limited use of the site in the Neolithic and earlier Bronze Age was followed by a set of Middle Bronze and Middle to Late Bronze Age inhumation and cremation burials which in turn preceded the establishment of a Late Bronze Age field system.
Evidence of a substantial industry manufacturing armlets from mudstone was also found, dating from the Late Bronze Age and continuing into the Early Iron Age. The early date of this industry gives the site a national importance. Activity in the Middle Iron Age was very much reduced in scale. A sub-rectangular double-ditched enclosure during the Late Iron Age continued to be used into the early Romano-British period. Saxon evidence is scant but appears to include domestic and industrial activities.
Early Prehistory
Despite the proximity of the site to a Neolithic causewayed enclosure, 60m to the north-west (Fig. 1), only two features dated to the Late Neolithic period. Pit 1969, towards the north-west corner of the site, was 0.6m in diameter and 0.26m deep, with a single fill containing one sherd of Woodlands-type Grooved Ware pottery, 20 pieces of struck flint, a small number of unidentifiable animal bone fragments and burnt flint. Pit 2932, 40m to the south-west, was 0.8m in diameter and 0.2m deep. Its lower fill contained two joining fragments of a Neolithic Cornish stone axe, an undiagnostic body sherd in a similar fabric to the Grooved Ware from pit 1969, animal bone and burnt flint. The upper fill comprised a dump of burnt material containing struck and burnt flint, stone and animal bone along with charred hazelnut shells. Charcoal from this deposit was dominated by Taxus baccata (yew), with a range of other species including Ulmus (elm), Corylus (hazel), Maloideae (hawthorn group) and Fraxinus (ash). The dominance of yew is unusual, as this is not usually favoured for fuel, but often associated with artefact making (Gale and Cutler 2000, 396-397), especially long-bows (Coles, Heal and Orme 1978, 10). The only evidence of Beaker and Early Bronze Age activity was ten abraded pot sherds (two possibly from Collared Urns) found redeposited in later features.
Middle to Late Bronze Age
Cremated bone derived from the remains of up to 17 unurned burials. Three other features contained what appears to represent formal deposits of pyre debris, and a fourth held a largely empty Earliest Iron Age vessel. The unburnt bone included remains from four inhumation burials; redeposited bone was found in one of the graves and two other features.
Radiocarbon dating of bone from four inhumation graves (408, 411, 419 and 982) and six cremation-related deposits (280, 316, 331, 380, 1415 and 1691) returned Middle Bronze Age dates for grave 982 and cremation-related deposit 1691. A Late Bronze Age date was returned for cremation burial 331, while all the others lay across the Middle to Late Bronze Age transition. It is therefore likely that contemporary Middle and Middle to Late Bronze Age settlement, agricultural and/or industrial activity was occurring nearby but outside the limit of excavation.
Most of the cremation-related deposits clustered in the south-west corner of the site forming a small, dispersed cemetery. With only a single exception, the inhumations and redeposited unburnt bone lay in an irregular east-west line across the northern portion of the site. Two of the cremation burials (279 and 315) contained fragments of amber, probably from beads or their manufacture. Otherwise, none of the burials were accompanied by any artefacts which could be considered as formal grave goods.
ADNA analysis was undertaken of the petrous bone from grave 982 and the human tooth from grave 408. The person buried in grave 982 may have moved to Kent from somewhere in present-day France or be a descendant of recent migrants. Strontium, oxygen, and carbon values from tooth enamel were within expected ranges for the local region. However, since these values are not unique to the UK and are comparable to those from continental Europe, including France (Willmes et al. 2018) alternative points of origin during childhood cannot be excluded. For the burial in grave 408, the data is more ambiguous and does not necessarily indicate that the individual was a migrant or descendant of recent migrants.
Later Prehistoric settlement and stone working
One of the most notable results of the Margetts Pit excavations was the discovery of a very large and unparalleled stone working industry. Approximately 8.5kg of organic-rich black stone, mostly comprising armlet roughouts at various stages of manufacture, but also including broken examples of finished armlets and working debitage in the form of numerous very small mudstone chips, were recovered. Although the raw material has previously been referred to as shale subsequent analysis has demonstrated it to be a dark brown, unfossiliferous, ferruginous, laminated silty mudstone. Although the precise geological provenance is unknown, the most likely sources are the local Gault exposures (Ixer, pers. comm.), the closest deposits of which are only 500m to the south-west, or the Lower Greensand deposits (Steele, pers. comm.), 1.5km to the south-west. In terms of quality the raw material is far inferior to Kimmeridge shale, and it may be that its friability explains the large quantity of waste and reject material.
The manufacturing process seems to have been given over entirely to the creation of armlets (Fig. 2). With only a single exception (a very small tubular bead) all finished objects were of this type, as was the entirety of the part-finished component which could be identified with any certainty. In every instance, the technology appears to involve hand-working: there are no lathe cores or any other evidence for lathe turning, supporting the dating suggested by the associated ceramics to the Late Bronze Age to Early Iron Age period.
The range of the material recovered includes all stages of working from crudely shaped block to finished or near-finished object. There is no evidence for finished objects having been polished, although this stage may have been completed at another site or sites, with objects traded in an unpolished state.
While small quantities of mudstone were recovered along with domestic waste from a range of features across the site, the bulk of the assemblage was concentrated in two main areas: in spreads of material associated with groups of intercutting pits within a Bronze Age field system on the west of the site; and within a possible negative lynchet approximately 80m to the north-east. In both areas, small features in the immediate vicinity also contained mudstone.
A widely spaced rectilinear field system was laid out, probably in the Late Bronze Age, aligned nnw-sse. Potentially associated but undated features were found including pits, postholes, animal burials, and several possible small structures. Two of the ditches on the west lay parallel approximately 3.5m apart, possibly defining a length of trackway.
Numerous Late Bronze Age pits were recorded, occurring as intercutting groups, in clusters, in broad spreads and as single isolated features. Three of the groups of intercutting pits (3054, 3055 and 3056) had direct stratigraphic relationships with the field system, with some of the pits in each group cutting the infilled ditches.
The pits in groups 3055 (seven pits) and 3056 (at least 12 pits) were relatively small and shallow, typically 1m wide and up to 0.5m deep, with relatively small assemblages of pottery, worked and burnt flint and animal bone, probably representing domestic waste. Among these, pit 2266 contained a lead alloy miniature wheel ornament.
The wheel ornament (Fig. 3) is rare in Britain. One complete and one fragmentary example are known from Flag Fen (Northover and Gillis 1999; Rohl and Northover 2001; Coombs 2001), together with a similar object with no central hole and ‘teeth’ around the circumference. Another has been found at High Throston, near Hartlepool, Teesside. The principal distribution of comparable objects is in Switzerland and north Italy (Primas 1984; 1985). While the Margetts Pit wheel ornament shares the four spokes and central annulus of these other examples, and the chevron ornament is also paralleled on some of them, it is unique in having a double outer ring. All examples which have been subject to analysis are of tin or a tin-lead alloy. The Margetts Pit example is made of lead, adding to its already distinctive character.
The intercutting pits in group 3055 were similar in size to the earliest pits in groups 3053 and 3054. In contrast, the latest pits in groups 3053 and 3054 were significantly larger, up to 3m wide and 0.7m deep in group 3053, and 4m wide and 0.6m deep in group 3054. It was the upper fills of these pits that contained substantial deposits of stone-working waste. This indicates that the stone-working deposits are relatively late in the sequence of deposition within the Late Bronze Age pits.
Large quantities of struck flint were recovered from these same layers, mixed with the stone-working waste and pottery. A total of 4078 pieces was retrieved from samples from pit group 3054. Except for a single scraper and six pieces with short areas of retouch, all were debitage or cores. Samples from pit group 3053 immediately to the north contained 4994 pieces, only one of which was a retouched tool – in this instance an awl.
In both pit groups the material falls into two categories: assemblages which seem to have been concerned only with the reduction of nodules into smaller pieces, and other more deliberate knapping concerned with the production of flakes. Both types occur together in the same contexts and are unequivocally connected with the mudstone-working debris.
The most notable aspect of the lithics is the almost total lack of formal tools (less than 0.1%). This is perhaps explicable if the purpose of the reduction was not the creation of suitable tool-making blanks, but rather simply of sharp edges which could be used either unaltered or with minimal retouching. The occurrence of over 2,700 chips and pieces of micro-debitage indicate that retouching was occurring, and the few tools are merely thick irregular pieces with short areas of abrupt retouch along one portion of a longer concave edge. These may have served a purpose akin to scrapers and were undoubtedly used in the production of mudstone objects.
There was a very dispersed spread of pits and post-holes to the east of Trackway 1 (see below). These all contained dark charcoal-rich fills with finds typical of domestic waste disposal. A particularly rich pit (422) was 1.9m in diameter and 0.4m deep (the largest pit in this area). It contained a primary fill followed by three separate episodes of dumping which, in addition to domestic refuse, contained mudstone-working waste, copper pins and a bone object. One complete annular amber bead was also found in the pit, along with further fragments. The bead has a diameter of 10mm with a large (4mm) slightly off-centre perforation and has a thickness of 5mm; it falls into Beck and Shennan’s group 2 annular beads (1991, 52-53, fig. 4.1). A fragment of a second very small bead came from Late Bronze Age ditch 2402; it is straight sided with a central circular perforation, but unfortunately not enough survives to estimate its diameter.
Eighteen other late prehistoric contexts (possible cremation graves, pits, a ditch, a post-hole and layers) also contained amber fragments, some of which may also have originally been beads, as well as apparently unworked pieces of amber. These contexts are widely spread over the site with no apparent clustering. These remains may represent raw materials collected for bead production although no worked surfaces were identified on any of the fragments.
An isolated pit slightly north of the centre of the site (1685) contained a group of 22 small fragments of fired clay which appeared to derive from some object with smooth, curved surfaces; the fragments have a slightly powdery feel, and may represent mould fragments, from the casting of copper alloy objects. The pit is otherwise undated but may belong to the Late Bronze Age or Early Iron Age.
There are other suggestions of metalworking at Margetts Pit in the Late Bronze Age or Early Iron Age. A piece of copper alloy casting spillage and a piece of dross, collected from the surface of pit 3050 in the western part of the site, close to the large spreads of mudstone working waste, and possibly also a small awl, may be further elements indicating this activity. A thick plate fragment with an unpronounced ridge along its middle and bevelled edges on the long sides may be a wall fragment from a socketed axe, perhaps to be used as metal for remelting. A 3mm-thick sub-rectangular plate fragment, possibly also an axe fragment, was recovered from the heavily disturbed secondary fill of pit 1971 which also contained a partial lamb skeleton and Late Bronze Age pottery. Another 4.6mm-thick plate fragment comes from a fill of mudstone-working spread 3045. It is too small to be morphologically diagnostic but may support the general character of the activity in this part of the site.
Five pieces of lead waste and one spiral were found within and above a Late Bronze Age vessel placed in pit 403. The lead spiral is particularly interesting as it may be best interpreted as a spiral bead imitating a faience one. The presence of lead waste in pits 403 and 422 also attest to the working of lead on site. Lead objects in the later British Bronze Age are as scarce as those of tin (Needham and Hook 1988). Their contexts and associations place them in the Ewart Park period, although some could possibly date to the preceding Wilburton period given the abundant use of the metal for alloying at that time. This is also broadly the time to which the tin wheel ornaments can be assigned.
Five copper alloy objects can be assigned a Late Bronze Age date. These include four pins and one cast plate fragment. Additionally, a further six copper alloy objects can be assigned a Late Bronze Age or more broadly prehistoric date based on context association.
Three pins came from refuse pit 422, located slightly north of the centre of the site, while a fourth was found approximately 22m to the north-east in posthole 901. A shank fragment with a slightly bent tip from the lower secondary fill of ditch 2395 may have been reused as an awl.
Six partially complete Late Bronze Age or Late Bronze Age/Early Iron Age pots appeared to have been deliberately placed in the ground. The purpose of this deposition, whether for some primarily ritual or practical (such as storage) function, is unclear, although the former is more likely.
Seven possible Associated Bone Groups (ABGs) were identified, all but one of which consist of the disarticulated partial remains of sheep/goat. Four are Late Bronze Age: three of these are from pits (1971, 2023 and 2183, the latter part of pit group 3055) and one is from layer 1366. Two date to the Early Iron Age and come from pits 1388 and 1854. One (pit 416) is dated to the Late Bronze Age/Early Iron Age.
Despite the large numbers of post-holes recorded on the site, few clearly defined structures typical of late prehistoric settlements, such as roundhouses or four-post ‘granaries’, could be identified. While it is possible to create several potential structures from apparent arcs of post-holes, or square or rectangular settings of four or more post-holes, particularly where the post-holes occurred in large concentrations, in very few cases were these arrangements unambiguous.
Late Iron Age and early Romano-British
Very little evidence was recovered of Middle Iron Age activity, limited to 10 sherds of pottery redeposited in features of Late Iron Age to Romano-British date. This absence of settlement is typical of most of the sites with Late Bronze Age and Early Iron Age settlement evidence in the area during the Middle Iron Age. Champion suggests that the emergence of settlements in the Middle Iron Age in different locations represents a ‘wider attempt to recolonize a landscape little used in previous centuries’ (in Booth et al. 2011, 213).
Evidence for the resettlement of the site dates to the Late Iron Age when a sub-rectangular ditched enclosure was constructed which remained the focus for activity into the early Romano-British period. The Romano-British pottery indicates that the site was probably abandoned in the second half of the 1st century ad. Only the southern end of the enclosure, including an entrance, lay within the limit of excavation in the access road part of the site, although the layout of the whole enclosure is known from geophysical survey. A number of other ditches closely associated with the enclosure suggest possible alterations to its layout and use during this period. A series of pits and post-holes also date to this general period, many of them clustered in the area south of the enclosure, including a Late Iron Age four-post structure. A single early Romano-British feature (435) was recorded in the western part of the site.
A linear spread of material, possibly derived from a midden, lay towards the centre of the site. It contained a substantial quantity of Late Bronze Age/Early Iron Age pottery (68 sherds, 904g) and a mudstone roughout, but also small quantities of Late Iron Age and Romano-British pottery (six sherds, 77g), along with a small fragment from a translucent blue globular Romano-British glass bead, fragments of ceramic building material, three fragments of an iron knife and a fragment from a bun-shaped fired clay loom weight of Middle/Late Saxon date. The mixed-date nature of this deposit suggests it is middened material, redeposited either by natural processes or perhaps by cultivation, and largely of Saxon date.
Terraces, lynchets and trackways
A series of shallow terraces ran from nnw-sse, tangentially to the main trend of the slope, possibly comprising negative lynchets resulting from cultivation within unditched fields. Their orientation matches that of the ditches of the Bronze Age field system and Late Iron Age/early Romano-British enclosure, although this is doubtless due to the topography of the site and contemporaneity cannot be assumed; several possible hedge-lines were also identified. Two shallow linear depressions on the same alignment may have been trackways. Trackway 1, in the centre of the main excavation area, was traced for 58m; Trackway 2 lay 50m to the north-east, adjacent to the enclosure. Shallow striations in the bases of the possible trackways appear to have been formed by the passage of wheeled vehicles, suggesting that the trackways at least are of a relatively late date.
Discussion
Margetts Pit is a site of some importance, with structural, artefactual and mortuary evidence suggesting continuous use from the Middle-Late Bronze Age to the Early Iron Age (1300-600/500 bc). A hiatus in occupation seems to have occurred during the Early and Middle Iron Ages, with occupation re-established towards the end of the Middle Iron Age continuing, at least intermittently, until the Middle Saxon period.
The most important result of the excavations at Margetts Pit was the discovery of waste material deriving from the manufacture of mudstone artefacts, mostly armlets. The quantity of waste material recovered leaves no doubt that this activity was being carried out at an industrial scale in the near-vicinity and that most of the waste was dumped in large quantities in a few events, rather than accumulating gradually.
Other industrial activities are attested on the site. From the finds of metal waste and tools it is apparent that metalworking formed part of the activity on site in the Late Bronze Age (Webley et al. 2020), although the limited quantity of related material suggests that it may have been of subordinate importance to the working of mudstone. It is tempting to see a link between this craft activity and the many undated lead weights found at Margetts Pit, but while the use of balances and weights in the Late Bronze Age is well documented (e.g., at Cliffs End Farm: McKinley et al. 2014, 179 fig. 6.15, 1 and 3; cf. Pare 1999; Rahmstorf and Pare 2007), the lack of closed context associations precludes any firm conclusions from being drawn.
Taken together, these strands of evidence may suggest that Margetts Pit functioned as a kind of entrepôt and that the deposition of votive materials and everyday objects may perhaps have been connected with the propagation of trade; especially being located so close to the North Downs prehistoric trackway.
acknowledgements
The author would like to recognise the important contributions of the following people, who’s work contributed to this summary note: Ian Armit, Madeleine Bleasedale, Philippa Bradley, Dana Challinor, Nicholas Cooke, Claire-Elise Fischer, Jane Evans, L. Higbee, David Holman, Matt Leivers, Jacqueline I. McKinley, Barbara McNee, Lorraine Mepham, Peter Northover, Andrew B. Powell, Jörn Schuster, Rachael Seager Smith and Chris Stevens.
bibliography
Beck, C. and Shennan, S., 1991, Amber in Prehistoric Britain, Oxbow Monograph 8.
Champion, T., 2011, ‘Later Prehistory’, in Booth et al., 2011, On Track: the archaeology of High Speed 1, section 1, in Kent, Oxford Wessex Archaeology Monograph 4, 151-241.
Coles, J. M., Heal, S.V.E. and Orme, B., 1978, ‘The use and character of wood in Prehistoric Britain and Ireland’, Proceedings of the Prehistoric Society, 44, 1-45.
Coombs, D.G., 2001, ‘Metalwork’, in Pryor, F. (ed.), The Flag Fen Basin: archaeology and environment of a Fenland landscape, Swindon, English Heritage archaeological reports, 255-98.
Gale, R. and Cutler, D, 2000, Plants in Archaeology; Identification Manual of Vegetative Plant Materials used in Europe and the Southern Mediterranean to c.1500, Westbury & Royal Botanic Gardens, Kew.
McKinley, J.I., Leivers, M., Schuster, J., Marshall, P., Barclay, A.J. and Stoodley, N., 2014, Cliffs End Farm, Isle of Thanet, Kent. A mortuary and ritual site of the Bronze Age, Iron Age and Anglo-Saxon Period with evidence for long-distance maritime mobility, Salisbury, Wessex Archaeology Report 31.
Needham, S. and Hook, D.R., 1988, ‘Lead and lead alloys in the Bronze Age – recent finds from Runnymede Bridge, 259-74’, in Slater, E.A. and J.O. Tate (eds), Science and archaeology Glasgow 1987: proceedings of a conference on the application of scientific techniques to archaeology Glasgow, September 1987, Oxford, BAR, Brit. Ser. 196.
Northover, J.P. and Gillis, C., 1999, ‘Questions in the analysis of ancient tin, in S.M.M. Young, A.M. Pollard, P. Budd and R.A. Ixer’, Metals in Antiquity, 78-85. Oxford, BAR, Int. Ser. 792.
Pare, C.F.E., 1999, ‘Weights and weighing in Bronze Age Central Europe’, in Römisch-GermanischesZentralmuseumMainz (ed.), Eliten in der Bronzezeit: Ergebnisse zweier Kolloquien in Mainz und Athen, Teil 2, Mainz, Monographien des RGZM 43, 421-514.
Primas, M., 1984, ‘Bronzezeitliche Schmuck aus Zinn’, Helvetia Archaeologia, 15, no. 57/60, 33-42.
Primas, M., 1985, ‘Tin objects in Bronze Age Europe’, Studi di Paletnologia in onore di Salvatore M. Puglisi, Roma: Universita di Roma ‘La Sapienza’, 554-62.
Rahmstorf, L. and Pare, C., 2007, Zu Gewichtsteinen der Späthallstatt- und Latènezeit, Jahrbuch des Römisch-Germanischen Zentralmuseums 54, 265-95.
Rohl, B.M. and Northover, J.P., 2001, ‘The analysis of the metalwork’, in F. Pryor, The Flag Fen Basin: archaeology and environment of a Fenland landscape, 298-308. Swindon: English Heritage.
Webley, L., Adams, S. and Brück, J., 2020, The social context of technology: non-ferrous metalworking in later prehistoric Britain and Ireland, Oxford, Oxbow.
Willmes, M., Bataille, C.P., James, H.F., Moffat, I., McMorrow, L., Kinsley, L., Armstrong, R.A., Eggins, S., and Grün, R., 2018, ‘Mapping of bioavailable strontium isotope ratios in France for archaeological provenance studies’,
Fig. 1 Plan of archaeological features, Margetts Pit, Burham.
Fig. 2 Examples of the mudstone armlets discovered at Margetts Pit.
Fig. 3 Lead alloy miniature wheel ornament discovered at Margetts Pit.