West Midlands Regional Research Framework for Archaeology, Seminar 2: Palmer1
An Archaeological Resource Assessment for the Middle Bronze Age to Iron Age in Warwickshire and Solihull
Stuart C Palmer
Warwickshire Museum Field Services, Warwickshire County Council, The Butts, Warwick, CV34 4SS
Introduction
The following account attempts in summary to assess the resource for the Middle Bronze Age to Iron Age (MBA - IA) in the county of Warwickshire and the metropolitan borough of Solihull. Data has been amassed from all the sites within the study area which have been subject to sufficiently detailed survey, the minimum standards being fieldwalking, trial trenching or excavation.
A vast amount of data pertinent to the period is represented by cropmarks on the SMR and as these have generally been uncritically recorded as ‘later prehistoric or Romano-British’, they have not been included in this assessment. This obvious limitation may well skew the dataset, but some of the sites that have been included illustrate how cropmark morphology is not necessarily a reliable indicator of site type or even chronology, although this may be refined after a more detailed study.
Although the rate of later prehistoric fieldwork has increased in the survey area since Richard Hingley’s review (1996), much of it still awaits full publication. A substantial part of this assessment has by necessity been gleaned from the ‘grey literature’ and much from conversations with or details supplied by others.
Chronological issues
To even begin to understand the archaeological remains and therefore the archaeological potential of the survey area, it is imperative that we have in place a chronological framework by which we can organise the available data. This seemingly obvious requisite actually presents a fundamental difficulty because of the inconsistency of recording both in print and on the SMR. This problem does not just relate to the older records and publications, as the vagaries and confusion are evident even in relatively recent reports and summaries, albeit with some notable exceptions.
To structure this assessment, the data has been divided into five periods that broadly reflect the chronologies referred to in the available records. They are relative to the traditional three phase Bronze and Iron Ages. We begin with the Middle Bronze Age and follow with the Late Bronze Age. Thirdly we have the Late Bronze Age/Early Iron Age transition and the Early Iron Age, grouped here because of the difficulties in distinguishing the ceramics of the LBA and EIA on some sites and also because it was a system successfully utilised in the neighbouring East Midlands region (Willis nd), to which much of the Warwickshire evidence can be related. We then continue with the Middle Iron Age, a general coverall that forms a significant part of the record. Lastly the Late Iron Age, which on some sites is not necessarily pre-Roman and has generally been distinguished by the occurrence of wheel made and grog tempered ceramics on sites that might not otherwise be distinct from the MIA.
The key site within the study area from which a chronology for much of the region may yet be extrapolated is Wasperton in the Avon Valley. The importance of the ceramic assemblage from this site cannot be underestimated and its publication has been long awaited, not least because it should provide a well dated (both by C14 and thermoluminesence) ceramic sequence to which other sites could be related. Provisional information from this site has kindly been supplied by Ann Woodward.
The Archaeological Landscape
The Middle Bronze Age
Evidence for MBA settlement in Warwickshire and Solihull is scant. Traces of three possible round-houses excavated adjacent to the Neolithic complex at Barford have been posited as BA constructions, based on a few pieces of flint (Hingley 1996 12), but this site still awaits proper analysis. Pottery and a small copper alloy knife were found in a pit group at Coton Park, Rugby (Northants Archaeology 1998) and a second pit group associated with a significant assemblage of Deverel-Rimbury pottery was found more recently on an adjacent site (Maull 2001). In addition a scatter of possible MBA pottery was found amongst later material fieldwalked at Whitchurch (Hingley 1988).
The thin scatter of MBA metalwork across the study area defies analysis but the small group of burnt mounds in the northwest of the study area, adds some flesh to an otherwise indifferent dataset. Identification of these sites is a corollary of the intensive and systematic work undertaken by Barfield and Hodder (1989) and Hodder (1992) and more recently work on the Birmingham Northern Relief Road (M6 Toll). Two outliers in Leamington at Sharmer Farm (Barfield & Hodder 1989) were chance encounters, the more significant as they were found near a relic stream in the course of drainage work thereby proving the potential of investigating such features which must surely be widespread in the landscape.
The Late Bronze Age
Settlement evidence
The identification of sites of the LBA amongst cropmark palimpsests is near impossible despite erroneous claims by some (MPP listing) that clothes line enclosures, which are very common in Warwickshire, date from this period. Excavated settlement sites are rare, the bulk of the evidence currently deriving from beneath occupation sites of later dates such as Park Farm, Barford (Cracknell & Hingley 1994) and Ling Hall Quarry, Church Lawford (Palmer in prep c). LBA pits have however been recorded at Wasperton and Salford Priors (Palmer 2000a, 36-43) but we are far from sure how well such features reflect the settlement record.
The earliest known evidence for land boundaries in the study area comes from Wasperton where a major ‘territorial’ boundary ditch dated between 1300 – 850 BC effectively sealed off a land unit in a bow of the river. The remains of a contemporary field system are posited outside this land unit, which if substantiated, would certainly represent our earliest evidence of fields. A linear boundary ditch of this date with an associated palisade was suggested at Park Farm, although a later date is also feasible.
The Late Bronze Age to Iron Age Transition and the Early Iron Age
Settlement evidence
Settlement evidence for the LBA-IA transition and the EIA is more prevalent. At Wasperton, four small, house size enclosures constructed either side of the existing ‘territorial’ boundary have been dated 850 – 650 BC. Contemporary features include two large linear pit groups. At Coton Park, several arcs of shallow gullies have been suggested as evidence for a transitory episode of settlement (Northants Archaeology 1998) and a few features of this date were evaluated at Polesworth (Palmer 1992). Trial trenching at Wolston identified several areas of activity over a 140 hectare site. One such area included a large bell-shaped, clay-capped pit, providing a tantalising glimpse of the possibility of grain storage and production during the period (Palmer 1990a). Trial trenching at Charlecote also produced pottery attributable to this period (Hughes & Jones 1996, 79).
The earliest known enclosed settlement within the study area was recorded at Wasperton, where three large rectilinear enclosures are dated 650 – 550 BC. One enclosure is associated with two ditched trackways or drove roads, which by implication suggests fields of crops and areas of pasture. EIA activity is conspicuously absent on many other excavated sites although unenclosed pit groups are recorded at Burton Dassett (Booth 1989), High Cross (Palmer in prep a), and beneath the ramparts of Nadbury hillfort (McArthur 1990). Further evidence that local hillfort sites were utilised at this time was recovered during fieldwalking at Foxhill, Alderminster (Hingley 1987c).
The large LBA/EIA pottery scatter fieldwalked at Whitchurch, which included animal bone and quern fragments may be indicative of a midden site, activity here evidently continuing well into the Iron Age. Other fieldwalked assemblages likely to represent sites of this date and later are known from Ettington (Hingley 1987d), Idlicote (Hingley 1987e) and Halford (Hingley 1987g), all in the Feldon area.
Land boundaries
There is an increase in the evidence for boundary construction in this period with the ‘territorial’ boundary at Wasperton being re-enforced by a pit alignment dated850 – 650 BC and also possibly a trackway at Rollright (Lambrick 1988, 80-1).
It is in the EIA that we see the first evidence of landscape division at Ling Hall as evinced by two unusual posthole alignments with radiocarbon determinations in the middle of the first millennium BC. A third as yet undated alignment and other partial alignments await full excavation (Palmer 2002).
The Middle Iron Age
Settlement evidence
The ubiquitous rectangular ditched enclosure cropmarks of Warwickshire have generally been thought to be MIA in date (Hingley 1996, 16), although recent work at Marsh Farm suggests that they were still being constructed in the LIA (Palmer 2000d; in prep b). Other discrete enclosures such as that at Long Itchington (Palmer 1999a; 2000b)must also be considered as LIA. At Wasperton the EIA enclosure was abandoned between 500 – 250 BC and a replacement constructed to the south. Further enclosure modifications are made on the east of the boundary and open settlement occurred to the north. Other excavated settlement enclosures include Park Farm and Rollright, whilst the Fulbrook enclosure was apparently empty when evaluated (Palmer 1996; Palmer forthcoming a). An unusual polygonal enclosure settlement with double opposed entrances recently excavated at Meriden is thought likely to have origins in this period (Northants Archaeology 2001), although the little pottery recovered from the excavation was largely LIA.
Few of the 17 or so hillforts in the region have been examined in much detail and only the defences of Nadbury can be reliably dated at 400 - 600BC. Pottery likely to be of a similar date has been recovered from a so-called pit dwelling at Meon Hill (Hodges 1906; Price & Watson 1982), yet there is no record of its considerable defences having been examined.
On Dunsmore the washing line enclosures are MIA, examples at Ling Hall (Palmer 2002) and Bourton Heath (Hodgson 1991) proving so, although it is uncertain if all of them were actual settlements. A major complex of inter-linked enclosures has recently been excavated at Ling Hall, revealing several buildings. Preliminary analysis suggests that only one building existed within each enclosure at a time, with two examples of buildings being replaced (Palmer in prep c). A single example of a ‘banjo’ enclosure is known at Heathcote, and when evaluated produced a few sherds of probable MIA pottery (Coutts & Jones 1998).
Unenclosed settlement of this period is not demonstrable in cropmark form but two major sites recently examined were discovered as a result of geophysical survey. An extensive settlement with at least 25 buildings at Coton Park, appears at this early stage to be all the more important as it is the first in the study area to be found on Boulder Clay (Northants Archaeology 1998). The implications of this are potentially far reaching, as it appears that the site was far more sophisticated in terms of material culture than any of the sites so far excavated on gravel. Quite why this settlement produced substantial material evidence whilst the enclosures beneath the DMV to the north did not (Maull 2001), is yet to be explained.
A further extensive site at Walton, though only partially excavated, was apparently occupied through to the LIA (Palmer 2000c; in prep d). Two areas of MIA activity known from Tiddington are thought to represent closely spaced settlements (N Palmer pers. comm.), although it is not known if either were enclosed and a discrete possible round-house described by a curvilinear ditch at Wishaw also yielded pottery of this date (P Booth pers. comm.).
Land boundaries
Linear boundaries form a significant part of the database for the period and their significance ought not to be overlooked. On Dunsmore especially, they seem to represent a major phase of land division and allotment and they are also widely represented as cropmarks in the Avon Valley.
At Ling Hall a major ‘territorial’ pit alignment divides a landscape of discrete rectilinear enclosures to the southwest from an organised network of estates or land units defined by pit alignments that emanate from a single nodal point to the northeast. The alignments are used as a spine for a number of settlement and other enclosures. A further pit alignment of this date has recently been excavated at Wishaw (P Booth pers. comm.).
Linear boundaries also form integral parts of the settlements at Coton Park and Park Farm, although they seem to have become redundant at Wasperton, as none are attributable to this phase. At Walton a linear boundary ditch divided the area of settlement features and appears to have acted as a focus for activity and also provide the axis for the settlement features. It seems likely that this boundary represented the divide between two distinct land-units.
Other linear boundaries of suspected later prehistoric date are known throughout the study area yet apart from those previously mentioned only the major earthwork of Hobditch, Lapworth, has been examined in detail and this was suggested as Middle or Late Iron Age (Cracknell & Hingley 1995; Hingley 1996, 12).
The Late Iron Age
Settlement evidence
LIA settlement is marginally less well represented in the record than the MIA although this cannot be taken as a true reflection of period demographics. At Wasperton the main enclosure was enlarged and open settlement is associated with a large pit group in the northernmost part of the excavated area during the period 250 BC – 0. The main settlement enclosure was enlarged again during the conquest period but there was a focal shift to the south in the early Roman period.
Open and enclosed settlement has also been excavated side by side at Marsh Farm Quarry, Salford Priors (Palmer 2000d; in prep b) and a further enclosed settlement was examined at Brandon Grounds (Bateman 1978a). The fragmentary enclosures excavated beneath the Coton DMV were apparently occupied throughout this phase with a suggestion of continuity into the Roman period.
Further nucleated settlement is known from beneath the Romano-British village at Tiddington and also beneath the Roman temple at Coleshill (Magilton forthcoming). LIA pits were examined in an enclosure at King’s Newnham (Palmer 1990b; Palmer forthcoming b) and further limited evidence for activity was examined at Wixford (Palmer 2000a). A disparate group of features of LIA/early Romano-British date have been recorded under difficult salvage conditions at Stretton-on-Fosse (Gardner et al 1982) but it remains unclear as to their function and significance. LIA settlement features were also recorded at Napton with some evidence that occupation continued into the Romano-British period (Dalton & Booth 1997). A similar date range is suspected of the settlement evaluated west of Alcester (Jones & Palmer 1995; Warwickshire Museum 2000).
It is interesting to note that whereas unenclosed LIA settlement sites continue into the Roman period, only at Coton Park has Roman settlement evidence been recovered from an enclosed IA settlement site.
Land boundaries
No certain examples of boundary features of this period are known, although on two sites they have been extrapolated when found in association with MIA pit alignments. At Ling Hall many of the pit alignments were cut by shallow ditches which have been suggested as re-defining the earlier land units or estates, and a MIA pit alignment at Wishaw was followed by a segmented ditch of probable LIA date.
A series of fragmentary enclosures and possible field systems excavated beneath the DMV at Coton Park are suggested as being of this date, although distinguishing IA from Roman systems has evidently proven difficult (Maull 2001). Otherwise there has been no firm evidence for field systems of this date on any of the excavated settlement sites. However, a variety of linear features that have produced pottery ascribed to the MIA or LIA in evaluation reports have been described as field boundaries in the absence of evidence for anything else.
Material culture
The material culture of the Middle and Late BA is by and large represented by metalwork recovered mostly by accident or more recently by metal detectorists. Only two stratified pieces are known, a chisel from Barford and a knife from Coton Park. The disparate list on the SMR includes 11 palstaves, 8 axes, 4 spearheads, 3 dirks, 1 gouge, 1 adze and 2 examples of gold ring-money. Many of these pieces have been poorly recorded, are unprovenanced or long-lost, making it impossible to interpret their distribution, but none are known to have derived from riverine or watery places.
The only identifiable EIA metalwork recorded in the survey area is a cast, beaded torc, recently excavated from a Romano-British pit at Wishaw. An iron spearhead was recovered from a MIA pit at Ling Hall and an undated bronze head and an undated torc are also known from the wider region. There are two examples of LIA horse harness fittings, a hinged fitting (Wise 1997) and a harness mount (Bolton 1998), and there is also a single LIA terret ring. A total of only 39 coins are recorded on the SMR; although their recording is inconsistent they are predominantly Dobunnic although a few Corieltauvi occur, including a hoard of ten from the north of Warwickshire. Many appear to have been deposited in Romano-British contexts but their distribution may still relate to pre-Roman tribal territories (Booth 1996). Given the small number of other metalwork finds recovered from the study area it is perhaps a surprise that a significant quantity of currency bars have been recorded (Hingley 1996, 20; 1991). Their occurence in, or close by, boundary features at Nadbury, Park Farm Barford and Meon Hill strongly suggests structuration in their deposition.