Settlement and Landscape Reorganisation from the Middle Iron Age to the Early Roman Period, Excavations South East of Park Fram, Ashford

A programme of archaeological fieldwork undertaken on land east of Park Farm, on the south side of Ashford between 2002 and 2007 revealed evidence for the extensive occupation and repeated reorganisation of the landscape during the late prehistoric and early Roman periods. A small number of medieval and later features were also recorded but are not described in this report.

The site, covering approximately 50ha (centred on NGR 6022 1386), comprised two blocks of land separated by a floodplain drained by the Ruckinge and Bilham dykes; these connect with a larger network of drainage ditches flowing north to the East Stour River. To the west of the valley are the Park Farm East and Park Farm South East residential and mixed-use developments, and to the east is land at Bilham Farm that is proposed for flood remediation measures. The main excavation, the results from which provide the evidence for this report, was undertaken in three adjacent areas in Park Farm East (Areas 1-3), totalling c.2.6ha; a further small area (Area 4) was excavated in Park Farm South East (Fig. 1).

The underlying geology comprises Cretaceous Wealden Clay (Geo-logical Survey of Great Britain 1990), with drift geology consisting of superficial caps of Pleistocene Third Terrace Gravel Deposits on the higher ground to the west, and recent Alluvium along the floodplain.

Middle and Late Bronze Age

An assemblage (378 sherds, 5666g) of Middle Bronze Age pottery, dominated by a single bucket-shaped jar, came from a shallow pit [31321] in Area 3 and the immediately adjacent section of a later ditch which may have disturbed the pit. Four pits in Area 1 produced slightly later pottery, broadly of 12th- to 10th-century bc date. The fabrics were flint- and grog-tempered, and forms included ovoid jars, a hooked rim jar and a cup.

The remains of a Late Bronze Age shouldered jar (137 sherds, 2147g) were recovered from an entrance terminal of a Middle Iron Age roundhouse ring gully (Roundhouse 2, below), and were clearly redeposited (Fig. 2). The vessel was in very poor condition with evidence of burning/re-firing and abrasion.

Middle to Late Iron Age

Part of a Middle to Late Iron Age settlement, comprising at least eight roundhouses represented by shallow curving gullies and a small number of other features, was recorded on the valley floor in Area 3 (Fig. 2). The settlement was associated with a number of trackway and field/enclosure ditches continuing north-west into Areas 1 and 2 (Fig. 2). These ditches represent the earliest elements of a complex array of overlapping field systems and boundary ditches that extended across the whole site (Fig. 1) and which indicate near-continuous organisation and exploitation of the landscape into the early Roman period (Fig. 3).

The phasing of these ditches, however, was hampered by the dearth of clear stratigraphical relationship at many of their intersections due in part to the soil conditions, and by the lack of chronological distinctiveness in their pottery assemblages. As a result, some aspects of the phasing for this and later periods are uncertain, and while the outline presented below is judged to be the most consistent with the available evidence, it is recognised that alternatives are possible.

Area 3 settlement (Phases 1-2)

The majority of the Area 3 roundhouses had entrances facing between east and south-east (Fig. 2), and in many cases the finds from them were concentrated within or near the gully terminals (including the possibly deliberately redeposited Late Bronze Age vessel in Roundhouse 2, see above).

All but one of the gullies had internal diameters of between 9.5 and 13m, the exception being Roundhouse 7 which was only 6m in diameter and may have been some form of ancillary structure associated with the adjacent Roundhouse 6. Three of the roundhouses had two concentric gullies. At Roundhouse 1 the outer gully, up to 1m wide and 0.4m deep, lay 3-5m outside the inner gully and appeared to form a small enclosure rather than part of the building, perhaps reflecting this building’s status. At Roundhouse 2, in contrast, the outer gully abutted the inner, and the pair could represent a repair to, or a slightly larger rebuild of, this structure. At Roundhouse 3, the two gullies, c.2-2.5m apart, were only present in the south-eastern quadrant, in which there was no entrance gap, and it is possible that they represent part of some other type of non-domestic structure.

The gully of Roundhouse 4, of which the entire circuit was present, also had no entrance gap, and other aspects of this structure are also anomalous. The 10m diameter gully had a number of stakeholes recorded either in its base or on its edge at the south and south-east, perhaps indicating the presence of a wattle screen in this area. As at other roundhouses, this section contained almost all the finds from the gully (pottery, burnt flint, fired clay, animal bone – some burnt – and a number of pebbles including a possible hammerstone). Just inside the gully at the east and south there were two pairs of small circular features, each c.0.6m in diameter and 0.3m deep, spaced 1.8m apart and separated by a 3.6m gap. These features could be post-holes, possibly comprising structural timbers flanking a south-east-facing entrance (despite the absence of a gap in the gully). However, their fill sequences were more consistent with their having been small pits. In addition to further animal bone and pottery, the most southerly pit also contained a Late Iron Age potin coin, a find that may add weight to some non-domestic, possibly religious or ritual function for this structure, perhaps as a shrine. A small undated feature [31371] just east of the gully (and perhaps associated with it) was notable for the large quantity of charred cereal grains (barley, spelt/emmer and free-threshing wheat) which was concentrated in its upper fill.

There were relatively few other features apparently associated with the Area 3 roundhouses, and some of these were undated. They included a number of hearths and pits, as well as lengths of gully, some irregular in their lines and possibly defining other types of structure. In addition, a group of five undated post-holes, midway between Roundhouses 2 and 4, was located within the Phase 1 trackway (below) but formed no obvious structure, despite being quite regularly spaced. These features contained a range of finds characteristic of domestic waste – mainly pottery, but also worked and burnt flint, fired clay and animal bone, as well as the charred remains of emmer and spelt wheat, and charcoal reflecting the dominance of oak woodland in the immediate area.

Phase 1a-b ditches

The Area 3 roundhouses flanked a sinuous, 6m wide ditched trackway which ran north-west towards Area 2; the four larger roundhouses (11-13m in diameter) lay to its north-east, and the other four (9.5-10m in diameter) to its south-west (Fig. 2). Two sub-phases of this trackway were evident (1a and 1b).

In Phase 1a, the trackway, but not the ditch on its eastern side, appeared to end, and open out, within the settlement, giving the most direct access to the western area; the eastern ditch appeared to continue in widely spaced segments through the settlement. It is not possible, given the c.150m wide unexcavated area between Areas 3 and 2 (very few features were recorded here during the subsequent topsoil stripping – Wessex Archaeology 2008a), to directly link the arrangements of ditches in the two areas. However, the evidence suggests that Phase 1a trackway in Area 3 formed part of a longer land boundary continuing directly upslope through Area 2, where there were two subrectangular enclosures/fields, one abutting either side of it. Short lengths of curved gully south of the boundary in Area 2 may represent a small (8m diameter) possible roundhouse (Roundhouse 9), although this interpretation is far from certain.

Another slightly sinuous east-west ditch [12006] c.100 m to the north in Area 1 (its eastern part truncated by later – Phase 1b and Phase 4 – ditches, see below) may represent a comparable boundary, perhaps signifying the division of the Middle Iron Age landscape into substantial blocks, rather than into a close arrangement of contiguous fields.

The layout of the Phase 1a ditches appears to have been modified during the life of the settlement. In Phase 1b, it was now the ditch on the southern side of the Phase 1a trackway (in Area 3) that continued (unbroken) south-east through the settlement (curving around Roundhouse 5), with the result that the trackway now gave direct access only to the larger round-houses to its north-east. At the same time, the northern trackway ditch turned sharply to the north-east, effectively separating Roundhouse 1 (within its small rounded enclosure) from the other buildings in this area.

In Area 2, the line of the extended land boundary was shifted slightly to the south in Phase 1b (possibly resulting in places in a double-ditched boundary), and the rectangular enclosure on its south side was replaced by one c.20m to the west. This enclosure appears to have had an internal division (which would have cut across the northern edge of possible Roundhouse 9). The eastern part of the Phase 1a boundary ditch in Area 1 continued in use in Phase 1b [11871], but then turned south in the centre of the excavation area.

Phase 2 ditches and Area 1-2 settlement/industrial features

The Phase 2 ditches represent a significant reorganisation of boundaries, particularly in Area 3, although this appears to have taken place within the period of occupation of the Area 3 settlement, since the newly laid out ditches, which have a far more regular rectilinear arrangement, continue to largely respect the positions of the roundhouses (Fig. 2). Moreover, they represent a comparable configuration of trackway and boundary elements (in relation to the settlement) to that in Phase 1b. In the new arrangement, for example, a 6.5m wide trackway approaching from the south-west appears to terminate within the settlement, and a ditch running perpendicular (towards north-west) from its end divides the settlement into the same two parts as before. Moreover, like the old trackway, the new one gives direct access only to the eastern group of larger roundhouses. A date for this phase of reorganisation in the later part of the settlement’s occupation (i.e. Late Iron Age) seems likely therefore.

In Area 2, the Phase 1b enclosure was crossed by three lengths of associated ditch, the longest of which cut across, and hence effectively decommissioned the earlier boundary ditches, before turning (at its northern end) towards the east. Possibly associated with this ditch were two clusters of small features to its immediate south – a tight cluster at the east of up to 16 small post-holes in an area less than 4m across (but forming no obvious structure), and a wider group to the west of at least 12 post-holes (again with no structure apparent), along with a number of other small pits and a hearth. These and other features in Area 2, such as a 2m-square, four-post structure (of the ‘granary’ type) at the west, and a number of further pits to the south, could feasibly belong to either Phase 1 or Phase 2 (or indeed to later phases).

In Area 1, the former boundary ditch appears to have been replaced in Phase 2 by a straighter ditch c.10-20m to the south, although on a similar east-west alignment and running parallel to the contemporary ditch in Area 2. A number of features in Area 1, including two roundhouse gullies, pits and a metal-working furnace, all yielding Late Iron Age pottery, appear to be broadly contemporary with perhaps the latter part of the Area 3 settlement.

The larger of the two roundhouses, Roundhouse 10 (13m in diameter), lay between Phase 1 and 2 boundary ditches. It was cut on its west side by a short length of ditch provisionally assigned to this phase, which appeared to be related to a length of smaller diameter gully (under 8 m) that lay inside the roundhouse on its west side.

Approximately 18m to the west was Roundhouse 11 (11m in diameter), which had a north-west facing entrance flanked by post-holes. This was bounded by an L-shaped arrangement of ditch segments, some marking the north side of short length of 3.6m wide trackway running beside the main boundary ditch.

Situated between the roundhouses was part of a metalworking furnace [11924], comprising a steep-sided, flat-bottomed pit with a rounded eastern end and a hard, fired clay lining (Fig. 2 inset). Although its western end had been cut by an early Romano-British pit [11987], and its northern side by a Late Iron Age pit [11951], both containing much redeposited furnace material, the surviving part is interpreted as the pit of an early form of ‘slag pit’ furnace. This would have had a clay shaft in which the iron ore was smelted built over the pit, with the iron bloom (to be worked) and the slag (to be discarded) collecting in the pit from where they could be extracted. Given the fact that all the slag recovered from the furnace (and the adjacent features) was the product of both iron smelting and smithing, it is significant that also recovered were parts of three crucibles used for the melting of copper alloys, as indicated by the residues adhering to them. The furnace’s fuel was dominated by oak.

There was a range of other features in Area 1, including pits, post-holes and hearths, that may have been broadly contemporary with this phase of activity. A Late Iron Age pit [11172] within the projected line of Roundhouse 11, for example, was notable for the high proportion of chaff from hulled wheat (relative to the number of grains); the pit also contained a large quantity of fired clay (c.10kg), possibly originating from the furnace to the east.

Phase 3 ditches: slighting the settlement

Whereas the Phase 1 and Phase 2 ditches appear to have largely respected the settlement structures, and so can be taken to have been broadly contemporary with their occupation, the arrangement of the Phase 3 ditches is notable for the frequency with which the roundhouses were cut by them (Fig. 3). In Area 3, for example, three roundhouses (2, 4 and 6) were centrally bisected by Phase 3 ditches. In Area 1, where the Phase 3 ditches comprise an array of small fields/paddocks flanking a boundary, in places either double-ditched or forming a narrow trackway running north-east down the slope to the valley floor, both roundhouses (10 and 11) were similarly bisected. None of the ditches in Area 2 were assigned to this phase.

While this may be a coincidental correlation, it is also possible that the laying out of the ditches, some of which appear to have been angled specifically to pass through the centres of the roundhouses (e.g. Roundhouses 2, 6 and 11), represented a deliberate and conscious slighting of the former settlement.

Area 4

Due to Area 4’s distance, c. 300-600m to the south, from the main excav-ation areas, it is hard to assign its features the same phases used in Areas 1-3 (Fig. 1). The excavation here revealed a small sub-rectangular enclosure set within an array of rectilinear field ditches which are comparable in their general appearance to the Phase 2 and later ditches to the north. There was a hearth inside the enclosure and number of pits both inside and outside, one of which outside contained Middle/Late Iron Age pottery; the other features, including the ditches, contained small numbers of Late Iron Age, or undiagnostic prehistoric sherds.

Late Iron Age to Early Romano-British

Activity in the early part of the post-Conquest period is indicated by a small assemblage of early Romano-British pottery (255 sherds, 1379g), as well as 11 Roman coins (all from the topsoil). The pottery was recovered from a small number of discrete features, mainly pits, but also from some ditch contexts suggesting that the final phase of landscape re-organisation took place within, or at least continued into, the start of the Roman period.

Phase 4: Late Iron Age/early Romano-British ditches and other features

The Phase 4 ditches, representing another major reorganisation of the landscape, comprise a widely variable arrangement of boundaries and trackways, and enclosures, fields and/or paddocks (Fig. 3). Part of the valley floor in Area 3, for example, appears to have divided up by long, straight, parallel ditches, 46m apart, with other ditches running perpendicular between them and off them; these ditches had no spatial relationship to earlier arrangements, although it is notable that three of them passed centrally through the locations of earlier roundhouses (Roundhouses 2, 3 and 8).

In Area 1 on the side of the valley, in contrast, the situation was more complex. The Phase 1 east-west land boundary appears to have been re-used for part of its length, although the new ditch which followed its line at the east diverged northwards from it at the west. To the north of this main spinal ditch the landscape was divided into an arrangement of largely rectilinear plots of varying size and shape, some apparently with internal divisions. Further ditches ran south from the spinal ditch, extending into but not across Area 2.

As in the previous phases there is limited dating evidence for this final reorganisation, but what evidence there is suggests it took place either at the very end of the Iron Age, or (partly at least) at the very beginning of the Roman period. The Area 3 ditches produced only Late Iron Age pottery, but in Area 1 a small quantity of early Romano-British pottery (42 sherds, 282g) was recovered from six of the ditches, including one sherd from the primary fill of one ditch section, and sherds from at least three vessels from the middle of three fills in another. (Romano-British pottery was also recovered from the upper fills of a number of earlier ditches in Area 2.)

A range of other features containing early Romano-British sherds among their predominantly Late Iron Age pottery assemblages were found rel-atively thinly distributed across Areas 1 and 2 (Fig. 3). Many were pits, including the large pit [11987] that cut the western end of the Late Iron Age furnace and which contained much material deriving from it, including part of its fired clay superstructure, slag and crucible fragments.

There were also a number of hearths, including one [21226] in Area 2 (Fig. 3) whose shallow concave base, 0.8 m wide, was lined with over 300 sherds (7.4 kg) from a Late Iron Age cordoned jar, above which was a heavily burnt layer of clay. From within the hearth, which had a narrow channel (possibly the flue) running to the north, a further 126 Late Iron Age sherds (958g) were recovered, along with 22 Romano-British sherds (91g).

Notable among the features associated with the Phase 4 field system in Area 1 was a localised cluster of 56 post-holes in an area 11 m by 14 m, many of them arranged in lines. The southern and western edges of the group lay parallel to the adjacent ditches, although separated from them by gaps of c.3-4m; these ditches defined the south-western subdivision of a larger rectangular field. Among the group of post-holes were two square, four-post possible granaries (in one, two post-holes appear to have been replaced, or added), and two possible fence-lines, but identifying other possible structures in the group was not possible. Only a small amount of exclusively Late Iron Age pottery was recovered from these post-holes, but they are clearly associated with the Phase 4 ditches. Other groups of post-holes and pits within these same fields (shown on Fig. 2) could equally belong to this phase.

Discussion

The archaeological works east of Park Farm have revealed a settled landscape subject to continuous occupation and re-organisation from the Middle-Late Iron Age into the start of the Roman period. These findings complement the results from other excavations in the Ashford area, such as at Brisley Farm (Archaeology South-East 2006), Waterbrook Park (Wessex Archaeology 2008b) and Foster Road (Powell 2010; Powell and Birbeck 2010), as well as more widely in Kent (Champion 2007, 120) (Fig. 1).

There was a low level of later Bronze Age activity on the site, but the dispersed and limited nature of the evidence makes it hard to interpret the features beyond indicating the likelihood of settlement activity in the vicinity. Such activity is demonstrated by other findings in the area. A possible roundhouse was recorded at Waterbrook Farm, for example, and at Foster Road this period was represented by possible field and droveway ditches, a roundhouse and an area of craft/industrial activity. Evidence for the division of the landscape has also been recorded at Brisley Farm and Westhawk Farm (Archaeology South-East 2006; Booth et al. 2008), but there was no evidence to suggest that any of the field boundaries recorded at this site were of later Bronze Age date.

This activity was followed by an apparent break in occupation until the establishment of a small, nucleated, valley-floor settlement in the Middle-Late Iron Age. A level of social differentiation may be evident in the sizes and locations of the settlement’s roundhouses, one being located in a small enclosure; others may not have been domestic structures at all. This appears to have been a typical farming settlement, involving animal husbandry (principally cattle and sheep/goat) and the cultivation of barley and wheat, and with a range of domestic and craft activities undertaken. A furnace, at a distance from the main focus of settlement, provided evidence of both iron working (smelting and smithing) and the melting of copper alloy, suggesting that it was used by a person (or persons) possessing a range of metallurgical skills, a little at odds with the relatively modest status of the settlement as seen in the rest of its finds assemblage. However, it is possible that these operations were undertaken by one or more peripatetic metalworkers.

The settlement initially occupied a landscape divided into relatively large blocks by a series of boundary ditches, with lengths of trackway and subrectangular enclosures associated with them, although the organ-isation of these elements appears to have been modified on more than one occasion during the settlement’s life.

A further re-organisation, however, clearly marked the end of the settle-ment, with some of the ditches apparently deliberately aligned so as to bisect the former roundhouses. This not only effectively decommissioned the settlement, but may have been designed to make an explicit statement about a transformation of the social landscape at a time of general political uncertainty and economic and social change.

The final phase of reorganisation may have taken place at the very start of the Roman period. Although no settlement structures were identified, a series of pits and hearths containing early Romano-British pottery, along with larger quantities of Late Iron Age sherds, suggests possibly short-lived settlement close to the site. The site lies just 400m south of the Roman road between Lympne and Maidstone, on which, within a gen-eration of the Conquest, an extensive Romano-British settlement was established at Westhawk Farm, 2.5km to the north-east of the site (Fig. 1).

Acknowledgements

The fieldwork was commissioned by CgMs Consulting (Duncan Hawkins) on behalf of Taylor Woodrow Developments Ltd and Persimmon Homes, and monitored for Kent County Council by Casper Johnson and Wendy Rogers.

The early evaluations of the site were undertaken by Pre-Construct Archaeology (Pre-Construct Archaeology 2003) and Archaeology South-East (Archaeology South-East 2003), with all subsequent phases of work undertaken by Wessex Archaeology and managed by Richard Greatorex (Wessex Archaeology 2004; 2007a-c; 2008c-d), apart from a watching brief by CgMs Consulting and Pre-Construct Archaeology (reported in Wessex Archaeology 2008a). The post-excavation stage was managed by Alistair Barclay and Andrew Powell with grateful assistance from Grace Jones (Finds) and Sarah Wyles (Environmental). The illustrations are by E.S. James.

A full report on the fieldwork east of Park Farm, including finds and environmental reports, will be published on the KAS website (www.kentarchaeology.ac). The archive will be held at Wessex Archaeology under the project codes 52891 and 65480-5 until accepted by a Kent museum.

andrew b. powell

[The full report can be found on the KAS website kentarchaeology.ac]

bibliography

Archaeology South-East, 2003, ‘Bilham Farm, Park Farm East: archaeological evaluation report’, Ditchling, unpubl.

Archaeology South-East, 2006, ‘Brisley Farm, Ashford’, Archaeology South-East website: www.archaeologyse.co.uk/04-Projects/Kent/Brisley-Farm/index.htm.

Booth, P., Bingham, A.M. and Lawrence, S., 2008, The Roman Roadside Settlement at Westhawk Farm, Ashford, Kent: Excavations 1998-9, Oxford, Oxford Arch. Monogr.

Champion, T., 2007, ‘Prehistoric Kent’, in J.H. Williams (ed.), The Archaeology of Kent to AD 800, The Boydell Press/Kent County Council.

Geological Survey of Great Britain (England and Wales),1990, Folkestone & Dover A, 1:50,000, Drift Sheet 305-6.

Powell, A.B., 2010, ‘Foster Road, Ashford’, Archaeologia Cantiana, cxxx, 376-9.

Powell, A.B. and Birbeck, V., 2010, ‘Excavation of a multi-period site at Foster Road, Ashford, 2006’, KAS website on-line report: www.kentarchaeology.ac/archrep/archrep.html.

Pre-Construct Archaeology, 2003, ‘An Archaeological Evaluation: Park Farm East, Ashford, Kent’, CgMs revision of unpubl. 2002 PCA rep.

Wessex Archaeology, 2004, ‘Park Farm East, Ashford, Kent: post-excavation assessment and updated project design’, Salisbury, unpubl. rep. 52891.01.

Wessex Archaeology, 2007a, ‘Park Farm East, Ashford, Kent: assessment report for additional trial trenching’, Salisbury, unpubl. rep. 65481.01.

Wessex Archaeology, 2007b, ‘Park Farm East, phase II, Land at Bilham Farm, Ashford, Kent: results of an archaeological evaluation’, Salisbury, unpubl. rep. 65480.01.

Wessex Archaeology, July 2007c, ‘Park Farm South-East, Phase II, Ashford Kent, Assessment report for additional Archaeological Excavation’, Ref. 65483.01 unpubl.

Wessex Archaeology, 2008a, ‘Park Farm East/South-East, phase II, Ashford, Kent: results of a programme of strip, map and sample monitoring (Areas A-D)’. Salisbury, unpubl. rep. 65484.02.

Wessex Archaeology, 2008b, ‘Area A, Waterbrook Park, Ashford, Kent: post-excavation assessment’, Salisbury, unpubl. rep. 65742.02.

Wessex Archaeology, 2008c, ‘Park Farm South-East, phase II, Ashford, Kent: post-excavation assessment report for additional archaeological excavation’, Salisbury, unpubl. rep. 65483.02.

Wessex Archaeology, 2008d, ‘Park Farm East, phase II, Land at Bilham Farm, Ashford, Kent: results of a strip map and record excavation’, Salisbury, unpubl. rep. 65482.02.