800 BCE – 42 CE
For the transition to Romano-British, please also read the page on Roman South Yorkshire here.
There is very little evidence for the late Bronze Age/early Iron Age transition in South Yorkshire – no settlements of this date have been identified and only a few artefacts, none from a stratified context. The Bronze Age cremation burial and associated pyre debris in a small pit and segmented gully enclosure at Sutton Common (Van de Noort 2007a), within the area later occupied by the Iron Age enclosed site, could suggest that certain places in the landscape held longer-term significance for prehistoric communities. Alternatively, such superimpositions may simply reflect the continuing utility of slightly raised areas within largely low-lying wetlands.
The earliest known Iron Age remains in the county are likely to be the enclosed sites at Wincobank and Sutton Common. Yet the former, possibly occupied circa. 500 BC has no reliable scientific dates associated with it, and there are no geophysical survey results showing any possible internal features. The main phase of construction at Sutton Common does not seem to have been initiated until the early 4th century BC (Nayling 2007; Van de Noort, Chapman and Collis 2007). Neither of these sites appear on present evidence to have had sustained ‘domestic’ settlement. Aside from a few metal detectorist finds, there is a large gap of some three to four centuries at present between the generally accepted beginning of the Iron Age around 800 BC and the first tentative evidence for occupation on several sites, although the nature and extent of the latter is at present unknown. Other possible earlier enclosed sites are either poorly recorded and unpublished (Croft Road, Finningley) or have never been investigated (Potteric Carr and Moorhouse Farm).
The first securely dated evidence for domestic settlement is from Balby Carr near Doncaster, where recent Bayesian modelling of several radiocarbon dates indicates a probable earlier ‘open’ phase of inhabitation during the 3rd rather than the 4th century BC or 410–350 BC (47.9% probability) and 315–215 BC at 95% probability (Daniel and Barclay 2016: 8). There may then have been a hiatus in occupation until the construction and use of an enclosed settlement from the 2nd century BC (215–85 BC at 95% probability or 205–135 BC at 68% probability). Settlement at Topham Farm, Sykehouse may also have begun during the 2nd century BC (Roberts 2003: 78). Much of the limited scientific dating and artefactual evidence for Iron Age inhabitation in South Yorkshire is from this later period.
Later Iron Age communities in South Yorkshire lived in small, dispersed enclosed farmsteads, and seem to have practiced mixed agriculture, though with pastoral production perhaps predominant. Cereal cultivation and crop processing took place on higher, drier ground, and some field systems and trackways were established during this period. There was a possible expansion or extensification of fields and production during the very late Iron Age, when some of the ‘brickwork’ fields and associated enclosures were initially laid out. Low-lying areas seem to have had an emphasis on pastoralism, with cattle probably being more numerous than sheep. People do not seem to have made or used much pottery, and items of metalwork such as brooches, rings and torcs appear to have been uncommon compared to other regions of Britain, even adjacent areas such as East Yorkshire, Lincolnshire and the midlands. There were contacts with other groups in modern north Nottinghamshire, Lincolnshire and West Yorkshire, but the nature and extent of these links is also unknown. No obvious ‘central sites’ have been identified, and it is possible that society was comparatively unstratified without social elites and little strong sense of ‘tribal’ identity. Any outward signs of higher wealth and status might have relied upon the numbers and quality of livestock, or perishable items of material culture for which no evidence remains.
After the invasion of Britain by Roman forces in AD 43, a temporary northern frontier was established around AD 55 that probably extended through South Yorkshire, but apart from the sites of Roman forts there is little evidence for the many complex interactions with local communities that took place during this period, and those immediately following the final invasion of the north in AD 70–71. There are indications of many intriguing continuities in cultural practices during the Romano-British period, however. Ceramic consumption and deposition at first were not concerted outside of Doncaster and its immediate hinterland, despite the presence of production sites. Overall the region remained relatively aceramic for many centuries – this was a long-term trend that seems to have persisted from the Iron Age right through into the post-Roman and early medieval periods. Some hand-made ceramics continued to be produced at a localised ‘household’ level until the 2nd century AD, the use of beehive rotary quernstones persisted at many settlement sites, roundhouses continued to be built and inhabited until possibly the 3rd century AD; and similar depositional practices took place in the Romano-British period as had occurred in the Iron Age, albeit sometimes utilising new artefact forms.
Other changes during the Roman period were more substantive – pottery production, consumption and deposition did increase from around AD 130 onwards, and during the 2nd and early 3rd centuries there may have been further expansion and extensification of field systems and agricultural production, perhaps in part to meet the needs of the Roman military. This may also have been accompanied by some intensification of production too. The scale and social impact of these changes is still unclear, however.
In many respects, the region departs markedly from conventional narratives of the Iron Age and Romano-British periods. There is a lack of ‘classic’ type sites such hillforts and oppida, and of villas, towns and other Roman-style settlements. Only a few vici were established. Ceramic use and deposition were scarce in the Iron Age, and not widespread or intensive during the Romano-British period, especially the first few decades after the conquest. This has been viewed by many authors over the years as problematic and even reflecting a poverty of evidence. It is a similar situation, however, to other areas of Britain such as north-east and south-west England, and much of Wales. Rather than continually regarding this as problematic or as a weakness, the archaeology of South Yorkshire instead offers the exciting possibility of exploring and constructing very different ideas and narratives of everyday life and identity during the Iron Age and Romano-British periods.
I would like to thank Andrew Lines, Jim McNeil, Zac Nellist, and Dinah Saich of the South Yorkshire Archaeology Service; for approaching me to see if I would be interested in contributing to the South Yorkshire Research Framework. The comments made during and after the workshop on the Iron Age and Roman periods by the many colleagues who attended were extremely useful, and wherever possible have been incorporated within this document.
I was privileged to spend some years living and working as an archaeologist in South Yorkshire. I would like to thank the following for many past conversations, information, and insightful ideas concerning fieldwork techniques and aspects of the Iron Age and Roman-British archaeology of South Yorkshire – Tim Allen, Simon Atkinson, Dr Bill Bevan, Mark Brennand, Professor Paul Buckland, Andrea Burgess, Dr Tim Cockrell, Professor John Collis, John Cruse, Dr Chris Cumberpatch, Patrick Daniel, Alison Deegan, Professor Mark Edmonds, Melanie Francis, Dr Eleanor Ghey, Dr Catriona Gibson, Dr Melanie Giles, Dr Andy Hammon, Kate Howell, Dom Latham, Ruth Leary, Andrew Lines, Louisa Matthews, Jim McNeil, Colin Merrony, Dr Debora Moretti, Richard O’Neill, Professor Mike Parker Pearson, Dr Rachael Reader, Dr Jane Richardson, Graham Robbins, Ian Roberts, Peter Robinson, Bob Sydes, Roy Sykes, Dr Jeremy Taylor, Reuben Thorpe, Steve Webster, Sarah Whiteley, and David Williams. I am grateful to Colin Merrony, Chris Grimbley and Dr Robert Johnston at the University of Sheffield for allowing me access in the past to the University of Sheffield Library of Aerial Photographs (SLAP) that belonged to the late Derrick Riley. I would also like to thank the indefatigable members of the Hunter Archaeological Society, Arteamus and the Doncaster Archaeology Group.
Unless otherwise stated, all dates including calibrated radiocarbon dates are presented as years BC or AD, but for ease of reading details of these dates including laboratory names and numbers have not been included. Readers are referred to the original references. No uncalibrated dates BP (Before Present) have been used.
ADS Archaeology Data Service
AMS Accelerator Mass Spectrometry
ARCUS Archaeological Research and Consultancy University of Sheffield (now defunct)
AS WYAS Archaeological Services West Yorkshire Archaeology Service
BAR British Archaeological Reports
BUFAU Birmingham University Field Archaeology Unit (now defunct)
CIfA Chartered Institute for Archaeologists
DCMS Department of Culture, Media and Sport
MAP MAP Archaeological Consultancy Ltd
NAA Northern Archaeological Associates
OA North Oxford Archaeology North
OSL Optically Stimulated Luminescence
PAS Portable Antiquities Scheme
SYAS South Yorkshire Archaeology Service
SYAU South Yorkshire Archaeology Unit (now defunct)
TVAS Thames Valley Archaeological Services
I would like to dedicate this section of the Research Framework to the memory of Flight Lieutenant Derrick Newton Riley DFC PhD (1915–1993), who was one of the first archaeologists to identify the extensive cropmark field systems of South Yorkshire and propose that some originated during the Iron Age. He was right.
This assessment as part of an archaeological research framework for South Yorkshire draws on other national and regional assessments and frameworks to inform its scope and research questions. These include Understanding the British Iron Age: An Agenda for Action (Haselgrove et al. 2001), itself based on an earlier online document (Haselgrove et al. 2000); as well as Britons and Romans: Advancing an Archaeological Agenda (James and Millett 2001). Although The Archaeology of Yorkshire: An Assessment at the Beginning of the 21st Century (Manby, Moorhouse and Ottaway 2003) purported to cover all of Yorkshire, in practice its flawed approach only focused on York and its environs, East Yorkshire, and to a lesser extent North Yorkshire. West Yorkshire received scant attention in that assessment, and South Yorkshire was barely mentioned at all.
This all reflects a continued trend to overlook the county at a national and regional scale (Chadwick 2008: x, 26-7; Cockrell 2016: 321; Cumberpatch and Robbins n.d.; Robbins 1999: 43-5). For example, there is little or no discussion of the South Yorkshire evidence in many published overviews of Iron Age Britain (e.g. Cunliffe 2005; Harding 2004, 2017). The South Yorkshire evidence has often been regarded by some archaeologists as problematic or mysterious (e.g. “It is as if this were a kind of ghost village, scarcely ever inhabited”, R. Van de Noort, quoted in Pratty 2002).
English Heritage (now Historic England) produced a draft Research Strategy for Prehistory in 2010 (Last 2010), which highlighted some broad themes for the whole of Britain. There are also useful regional documents from areas adjacent to South Yorkshire that have bearing on the county. These include archaeological resource assessments, research agendas and a strategy for the East Midlands incorporating modern Derbyshire, Nottinghamshire, Lincolnshire, Leicestershire, Rutland and Northamptonshire (Bishop 2001; Cooper 2006; Knight, Vyner and Allen 2012; Willis 2001). The most recent iteration of these assessments comprises the basic format on which this framework is based. There is also the Iron Age and Romano-British Research Agenda for West Yorkshire (Chadwick 2009), an area where there were many similarities (but also some significant differences) in the character of the archaeology during the Iron Age and Romano-British periods.
Prehistoric artefact typologies remain a key feature of dating Iron Age sites and finds. Outside central-southern England and eastern England, reliable stratigraphic sequences and diagnostic artefacts and assemblages from the Iron Age are uncommon (Haselgrove et al. 2001: 2-3), in northern England they are scarce (Bevan 2000: 145; Willis and Millett 2016: 226), and in South Yorkshire they are largely absent altogether. Until recently there were few reliable radiocarbon (14C) dates, and the relative lack of diagnostic Iron Age metalwork and ceramic finds has not permitted the construction of detailed typological sequences.
In 2001, the British Iron Age Agenda for Action noted that, except for the ‘Wessex’ area of south-central England, and eastern and south-eastern England; there were few parts of Britain with anything other than outline chronological frameworks for the Iron Age (Haselgrove et al. 2001: 2-3). This situation has improved somewhat at a national level in the past 18 years since the Iron Age Research Agenda was first produced, but many problems remain in South Yorkshire.
The Iron Age in central southern England has traditionally been divided into early, middle and late sub-periods (Cunliffe 2005). There is a significant problem with radiocarbon dating in Britain, however, with so-called ‘plateaux’ in the calibration curve, one of which falls between roughly 800–400 BC. This makes dating of the later Bronze Age and early to mid-Iron Age periods, and thus the late Bronze to early Iron Age transition, especially problematic (Needham 1996). These difficulties can be addressed to some extent using multiple samples, more accurate Accelerated Mass Spectrometer (AMS) dates, and Bayesian statistical analyses (Bayliss 2009; Hamilton 2010; Haselgrove and Pope 2007). The early, middle and late Iron Age scheme was developed for central-southern England, where pottery was relatively plentiful and some more distinctive forms and decoration allow tighter typologies to be constructed. In South Yorkshire, as in some other parts of Britain, there is very little artefactual material from the later Bronze Age and early to mid-Iron Age periods (see below), and generally little scope to distinguish between the middle and later Iron Age periods. A more realistic framework that has been proposed for much of Britain might therefore be an earlier Iron Age from circa. 850–400 BC, and a later Iron Age from c. 400 BC–AD 70 (Haselgrove and Pope 2007: 5-6); although there also needs to be clear recognition that many Iron Age traditions and practices continued well into the Romano-British period.
The soils over the Sherwood Sandstone, Coal Measures and Gritstone areas of South Yorkshire (comprising much of the bedrock geology of the county) are generally acidic, and there is often poor preservation of bone and organic material. To date, there is only one confirmed Iron Age inhumation burial from the county. There is thus much less potential for the radiocarbon dating of such material than in many other regions. Until recently, there were very few Iron Age radiocarbon dates available from South Yorkshire, and those that were (such as for Wincobank hillfort, see below) were unreliable due to their age and/or lack of secure stratigraphic context. Much of the available archaeological and radiocarbon dating evidence for South Yorkshire derives from the period circa 400 BC onwards. It is unclear if this reflects a genuine lack of material culture and inhabitation from the earlier period, a notable difference in the character of the archaeology from that date, making it less identifiable; or if this apparent absence of evidence is purely due to sample bias and the methodological difficulties noted above.
Traditional culture-history narratives, based in part upon written Roman accounts that were not contemporary to the 1st century AD, propose that following the Claudian invasion of AD 43, Roman forces advanced west and north across England. They halted and established forts or vexillation fortresses at Chesterfield, Templeborough, Rossington Bridge and Lincoln during AD 55–65 (Birley 1973; Hanson and Campbell 1986; Hartley 1980; May 1922; see below). This created a frontier along a roughly south-west to north-east line formed by the Rivers Dee or Mersey, Trent, Don, and Humber. How ‘hard’ or porous this temporary frontier was, the scale and nature of Roman interactions and the physical and social impact of occupation upon indigenous people are all unknown; but it took place over 10–15 years. This significant period is almost invisible in archaeological terms. Changes in occupation within forts and settlements, and their eventual abandonment, have few independent dates to support or contest the conventional historical narratives. After c. AD 120 South Yorkshire may have become incorporated within the administration of a civitas based at Aldborough (Isurium Brigantium) in North Yorkshire, although it is possible the area of the modern county south-east of the River Don might have been part of the civitas of the Corieltauvi/Coritani which was centred at Leicester (Ratae Corieltauvorum). There is simply no detailed historical documentation for this, however, and the archaeological evidence has very little to add; other than hints from the distribution of coin hoards (see below).
Roman ceramics were not widely distributed until the early 2nd century AD, and even after this period were not plentiful on many rural sites – this hinders attempts at establishing a more definitive chronology. Key Roman-period sites such as Templeborough were excavated and published a long time ago (May 1922), and more recent investigations by ARCUS and ArcHeritage have been small-scale (Davies 2013, 2016). The fort at Doncaster and surrounding settlement remains were destroyed during the 1960s and 1970s with only limited rescue work possible (Buckland and Magilton 1986, n.d.), and as no funding was available at the time for post-excavation work many aspects of the work remain unpublished. Subsequent investigations by SYAU, AS WYAS and Arc Heritage have also been relatively small-scale, and most results remain as unpublished interim notes and archive or assessment reports (Atkinson and Cumberpatch 1995; Chadwick and Burgess 2008; Chadwick and Lightfoot 2007; Chadwick, Martin and Richardson 2008; Davies 2013).
A once-useful overview of the Roman period (Buckland 1986) is now out of date due to discoveries made since 1990 as a result of developer-funded fieldwork. Detailed PhD research examining Iron Age and Roman-British rural settlement across north Nottinghamshire, South Yorkshire and West Yorkshire (Chadwick 2008a) has only been published online (2010); and it too is already out of date thanks to the results from recent commercial archaeological investigations, some of which have been on an unprecedented scale.
In very broad terms, the earlier Iron Age of Britain is characterised by two main categories of settlement or occupation. In upland areas, and on hills in lowland areas, hilltop enclosures developed in many regions, many of which became the hillforts once thought to be characteristic of Iron Age inhabitation. Many excavated British Iron Age hillforts have been found to have late Bronze Age antecedents, but hillforts were predominantly an earlier Iron Age phenomenon. For those that have been excavated in any detail (mainly in southern England, the Welsh Marches, Wales and Scotland), use seems to have declined after c. 400 BC, or by the later Iron Age. Some may have been re-occupied immediately following the Roman invasion of AD 43, but only specific examples. An influential model of hillfort inhabitation saw them as elite residences and strongholds, and centres for specialist craft production and exchange, agricultural storage and redistribution (Cunliffe 1995, 2003, 2005). This idea has been heavily critiqued and is not supported by archaeological evidence (e.g. Buckland et al. 2001; Collis 1996; Hill 1992, 1995a, 1996a; Sharples 2007, 2010). These more critical accounts have highlighted communal labour and social relations in hillfort construction, rather than the centralised authority of individuals.
Broadly contemporary with hillforts were lowland late Bronze Age and earlier Iron Age settlements. These often consisted of ‘open’ settlements of roundhouses and ancillary structures, sometimes defined by palisades; but rarely large enclosure ditches (Knight, Vyner and Allen 2012: 62). This was regionally variable, however, and might also reflect inhabitation becoming restricted to specific landscape zones (Haselgrove and Pope 2007: 5). Such settlements are hard to identify on aerial photographs of cropmarks, and on geophysical survey plots, and have often only been identified in large-scale archaeological excavations where extensive areas have been stripped and recorded. Two regional examples, one of which may not even be a ‘domestic’ settlement at all, were located just over the county boundary in West Yorkshire at South Elmsall; and there was an unusual palisaded enclosure at Swillington Common near Leeds (Grassam 2010; Howell 2001). The early phases of occupation at Balby Carr near Doncaster probably consisted of an ‘open’ settlement, though this dated to the 3rd century BC (Daniel and Barclay 2016).
During the later Iron Age, hillforts appear to have been largely abandoned, and across much of lowland England and Wales small enclosed settlements defined by ditches and/or stone and earth banks became commonplace, perhaps due to settlement and/or population expansion (Haselgrove et al. 2001: 29), and an increased emphasis on tightly-bound kin groups (Knight 2007: 197; Robbins 1998; Thomas 1997: 215). These settlements can be termed household ‘compounds’ (Hingley 1989: 55), or ‘farmsteads’. They were probably inhabited by one or more extended families, or several different households from the same kinship group. In upland areas, such settlements are notoriously difficult to date, but were constructed and used from the middle Bronze Age well into the Romano-British period. The very long-distance exchange networks characteristic of the middle and late Bronze Age contracted, and though wider movements of materials and artefacts still occurred (q.v. Knight 2002: 137-140; Moore 2007: 80-83), this was more often at a regional rather than an inter-regional or pan-European level.
Following the Roman occupation, adjacent areas such as Nottinghamshire, Lincolnshire and East Yorkshire developed networks of forts and stations, roads, small towns and rural villas, and Roman-style material culture became almost ubiquitous. This conforms in most respects to traditional views of ‘Romanised’ provincial settlement. This was less the case in West Yorkshire, though even here there were villas and roadside or aggregated settlements that seem to have prospered during the 1st to 4th centuries AD. The vicus at Castleford was a significant local centre and York was not too far away. In South Yorkshire there was a fortress at Rossington and forts with vici at Templeborough and Doncaster. Pottery production at Rossington Bridge became a significant regional industry. Yet few villas have been identified, and the area west and north-west of the River Don seems to have been far less ‘Romanised’ in terms of settlement form and material culture use.
As outlined below, in many respects the Iron Age and Roman-British evidence from South Yorkshire departs from these conventional broad archaeological narratives. Rather than continuing to contrast the county negatively with other regions, however, it is more appropriate to develop research questions, methodologies and interpretations specific to it (Robbins 1999: 46-7), a more local archaeology attentive to the details of its evidence. What makes South Yorkshire’s Iron Age and Romano-British periods different in character is also what makes them interesting.
The unusual pair of enclosures either side of the former course of the Hampole Beck at Sutton Common near Askern are situated within low-lying, once marshy land on the edge of the Humberhead Levels. They survived as earthworks and were Scheduled in 1937, but in 1980 the larger enclosure (Enclosure A) was bulldozed and ploughed. It was once interpreted as a Roman camp (Surtees 1868), and several 19th and early 20th century small-scale investigations have left no surviving records or finds. Whiting directed the first published excavations during 1933-1935, with trenches in both enclosures. These revealed a plank-lined pit, a possible solid wooden wheel, and a human skull fragment next to a post alignment linking the enclosures (Whiting 1936-38: 57). Enclosure A had a drystone revetment wall on its western side, and a timber palisade underneath the rampart (ibid.: 79). Whiting also identified at least 34 so-called ‘huts’ (Surtees’ ‘tent mounds’), on top of the ramparts and within the interiors of both enclosures. Many were surrounded by shallow gullies, some had stone-flagged floors and internal post-settings and traces of decayed wood and thatch; but they yielded no dating evidence.
Trial trenching by SYAU in 1987-88 found waterlogged and worked wood in the ditches of Enclosures A and B, including a notched wooden ladder; with radiocarbon dates indicating activity between c. 550–200 BC (Parker Pearson and Sydes 1997: 229). Joint SYAU and University of Sheffield investigations in 1992–93 recorded a marked deterioration in the condition of the wood (ibid.: 225-6); and indicated Enclosure B’s rampart had consisted of a clay base with turves above. One small ‘hut’ in Enclosure B was excavated, and its shallow subrectangular gully contained one sherd of ‘later prehistoric’ pottery (ibid.: 230). Fragments of saddle querns were also recovered. Continued degradation of previously waterlogged deposits by drainage schemes led to further work by the Universities of Hull and Exeter in 1998–99 and 2002–3; including open-area excavation of Enclosure A, though only 10% of features were investigated. This work revealed details of box timber ramparts and two substantial wooden gateways of Enclosure A, and the wooden causeway linking Enclosures A and B. Dendrochronology established that rampart timbers were felled between 372 and 362 BC (Van de Noort, Chapman and Collis 2007: 175). Hundreds of postholes within Enclosure A represented at least 115 four-post structures, fences and other features, but no roundhouses or other buildings were identified.
Following a period of disuse when ramparts rotted and partially collapsed, sometime between c. 400–200 BC at least 12 small subrectangular and penannular gullies may have been used for the secondary deposition of cremated human and animal remains, though evidence is ambiguous (Chapman and Fletcher 2007: 153-5). Given the overlap in dates with the construction of the ramparts, and that some of these features overlay four-post structures, this implies the original occupation of the enclosure was relatively short in duration. Only five gullies were sampled, and very minimally, rather than being 100% excavated as is almost standard with suspected funerary features. Just a few concentrations of bone were found, with occasional fragments of cremated bone noted elsewhere. Late Iron Age glass beads were recovered from one of these features, and the gold bracelet or ingot fragment found (Hill 2007) may also have been associated with them. Sadly, of the c. 10% of internal features actually excavated, most were only half-sectioned. An extremely limited sample of Enclosure A’s eastern entrance ditch terminals recorded possible ‘placed’ deposits of two human skulls, a quern fragment, and an antler weaving comb. Enclosure B was targeted by magnetometry survey and interpreted as a largely empty enclosure with a cross-wall, banks and ditches (Payne 2007: 68-71; Van de Noort and Chapman 2007: 37), although most of the numerous features in Enclosure A were not visible prior to excavation, so this may be supposition. It is also unclear from these excavations what the small ‘huts’ on the banks were or how they fit into the chronological sequence, but presumably they were part of a later phase of occupation, after the ramparts ceased to be functioning structures.
Sutton Common might not be exceptional – there are other enclosures in South Yorkshire on slight rises in otherwise low-lying areas. At Croft Road, Finningley, excavation by MAP ahead of gravel quarrying revealed a large curvilinear ditch at least 180m in length, with a possible ‘entrance’ up to 25m wide. This relatively narrow and shallow feature, part of a much larger oval feature visible on cropmarks, was interpreted as a possible Neolithic henge ditch (MAP 2006a: 18, 46-47), but there is no stratigraphic, artefactual or scientific dating evidence to support this assertion. No external bank was indicated, and no finds were recovered from the ditch, though very few slots were excavated across it. An outer series of shallow, curvilinear gullies or palisade slots respected the line of the innermost ditch, but also created an additional space or annex to the north-east. These may represent several later phases of activity. There were five pits inside the inner ditch – one contained later Bronze Age or earlier Iron Age pottery, and further sherds of this date came from the fill of an adjacent ditch. The area within the oval enclosure at Finningley, and to the north-east of it, then became the setting for a series of subrectangular enclosures, pits and waterholes of Iron Age and Romano-British date – radiocarbon dates of the 1st century BC were obtained from another ditch, and a pit.
The rather poorly written and at times internally contradictory 2006 report not only used segments of text uncredited from other sources (e.g. MAP 2006a: 50, para. 7.14, is largely plagiarised from Chadwick 1997), but also claimed that pottery of 3rd and 2nd millennium BC date was recovered from the enclosure (ibid.: 48). This is contradicted by the ceramics report, however, which suggested a late Bronze Age or early to mid-Iron Age date instead (Manby 2010: 180). Indeed, a subsequent Updated Project Design (MAP 2006c) admitted that there was ‘no conclusive evidence’ to support a Neolithic date. On balance, a later Bronze Age or earlier Iron Age date seems much more likely. At least one later prehistoric vessel had carbonised residue on the inside (Manby 2010: 180), potentially suitable for future AMS dating. Oddly, the ceramics report is dated four years after the main report it is presented within (MAP 2006b). This important site has still not been published, though the UPD dates to 2006 (MAP 2006c). Dating and publication should be a matter of urgency. Much of the oval enclosure lies outside the development area, and there is clear potential for future research-led investigations. It is hoped that any future fieldwork and reporting will be monitored more closely to ensure that it is of a much higher standard than the MAP investigation.
At Moorhouse Farm, Tickhill, a double or triple-ditched enclosure visible as cropmarks now lies beneath a modern farm (Riley 1980: 49, 66, plate 15). At Potteric Carr, a large (c. 1.7ha) irregular enclosure now partly under trees had up to three lines of ditches (Deegan 2004: 8, fig. 4; Magilton 1977: plate 4; Riley 1980: 91), and extant earthworks in Beeston Plantation. At least two, possibly three circular features up to c. 25m in diameter are visible it within it on some aerial photographs. Two smaller enclosures appear to have abutted the triple-ditched example. To the south, a series of trapezoidal fields and enclosures developed from the later Iron Age onwards (see below). Near Brodsworth, there is a cropmark of an irregular univallate curvilinear enclosure (Roberts, Deegan and Berg 2010: 27, fig. 33).
Several potentially similar bivallate and univallate irregular enclosures have been identified just outside South Yorkshire near Babworth, Loundfield Farm, Forest House Farm, Flint Hill, and Willow Holt in Nottinghamshire (Riley 1980: 48-49, maps 16, 20, 23), all located on floodplains or on slight rises surrounded by low-lying peaty and alluviated valleys. There is also a multivallate enclosure less than a hundred metres over the county boundary between Norton in South Yorkshire and Little Smeaton in North Yorkshire, on the River Went floodplain (Manby 1988: 26; Riley 1973: 212; Roberts, Deegan and Berg 2010: 37, 46, figs. 49, 46). Several possible ring ditches or roundhouses are visible on cropmarks, and further enclosures and trackways are situated to the south of it, within South Yorkshire.
Riley described such sites as ‘marsh forts’ (Riley 1980: 66), but this term is problematic, and it is unclear if they were ever defensive or even contemporary with one another. They were likely communal and social foci, as the numerous four-post possible granaries at Sutton Common suggest; but it is a significant stretch to propose that their distribution reflected a socio-political boundary between the limestone hills north and west of the Rivers Don and Idle, and the gravel lowlands to the south and east (Parker Pearson and Sydes 1997: 254). Earlier researchers, however, also suggested that this north-east to southwest line across South Yorkshire and north Nottinghamshire represented a Brigantian and Corieltauvi/Coritani tribal boundary (Haselgrove 1984: 16; Preston 1950a: 91). There is no real evidence for this.
Encompassing an area of c. 1.1 hectares, Wincobank is classified as a univallate hillfort (Pastscape 314855), with just one large surrounding ditch; though there are surviving remnants of a counterscarp bank too. There were small-scale investigations of Wincobank in 1899, but most of the interior was not examined and the results never published, although some records are held in Sheffield Museum. A summary of these unpublished results describes burnt masonry ramparts with the remains of timber lacing (Preston 1950a, 1954). Four possible entrances were noted in the inner ramparts; though the only dating evidence was some 2nd century Roman greyware found in upper ditch fills. An old radiocarbon date of c. 500 BC obtained from an unpublished evaluation by Beswick in 1979 may be unreliable (Beswick 1985; Coutts 1999: 78); although it has been suggested that Wincobank was probably disused by the later Iron Age (Buckland 1986: 6). In the 18th and 19th centuries mineral extraction pits were dug nearby, and during the Second World War a searchlight and an anti-aircraft gun were sited on the ramparts (Whiteley 1993). An analytical earthwork survey identified gaps in the earthworks to the north-east and south-west as possible original entrances (Pouncett 2001), but any interior features are still completely unknown. The site would therefore benefit considerably from detailed geophysical survey using the latest techniques and equipment. Located within an urban environment, Wincobank continues to suffer from erosion and vandalism, burning, and other negative impacts.
This unusual enclosure was broadly subrectangular in plan and c. 1.8 hectares in area, on a sloping Millstone Grit outcrop or promontory in moorland overlooking the Burbage valley. Along the northern and eastern sides of the enclosure the naturally steep-sided rock face provides an impressive facade, reinforced with some drystone walling. The southern slope of the outcrop is less steep, and the site is most accessible from the west. Along these two sides, unbonded drystone blocks were used to create imposing revetment walls up to three metres high, the western wall being up to eight metres wide at the base and reinforced with an earthen rampart. There is a single inturned entrance at the south-west, and the stones here are noticeably more regular and partly dressed. There are large earthfast boulders over much of the interior, though smaller stones inside and outside the walls appear to have been cleared and used in construction. No internal structures are visible. There was disturbance of the north-eastern side of Carl Wark during the 19th century by millstone quarrying. Erosion from walkers is an ongoing issue (Bevan 2006a).
Several antiquarian writers proposed that Carl Wark was a prehistoric stronghold (Gould 1903; Trustram 1911). During a survey in 1948, Stuart Piggott noted that the retaining wall and earthen bank resembled early medieval structures in Scotland. A small-scale trench by Simpson in 1950 found no dating evidence but revealed that the embankment was constructed of turves, thought to be another possible indicator of a ‘Dark Age’ date (Piggott 1951: 210-2). Despite its unusual form, however, it was accepted by several researchers as an Iron Age hillfort (Challis and Harding 1974: 47; Forde-Johnston 1976: 280-1); but it is very different to scarp-edge promontory forts or hillforts in the Peak District such as Comb’s Moss (Castle Naze), Fin Cop, and Mam Tor. Based on initial work at Gardom’s Edge, where a stone rubble bank enclosed a boulder-strewn area on a scarp edge, a Neolithic date for Carl Wark was proposed (Edmonds and Seabourne 2001: 73-4), but Gardom’s Edge has now been dated to 1400–800 BC or the later Bronze Age (Barnatt, Bevan and Edmonds 2017: 54). Work elsewhere in upland Britain suggests several other supposed Iron Age hillforts may in fact be Bronze Age in date (e.g. Jecock and Evans 2018). Though defensible in extremis, Carl Wark is a poor choice as a truly defensive structure, as it lacks a natural water source (Bevan 2006a: 8). Such sites may have instead fulfilled roles as community meeting places and ceremonial centres. Geophysical magnetometry survey at Carl Wark has been undertaken (Sykes and Brunning 2016), but with inconclusive or negative results. Further geophysical survey at the site may not be productive; but detailed 3-D laser scanning might prove useful.
The shape or plan and the size of enclosures has long exercised archaeologists who have attempted to classify them according to morphological criteria and chronological trends (e.g. Cox 1984; Riley 1980; Whimster 1989; Wilson 1987). For example, the recent Cropmark Landscapes of the Magnesian Limestone project identified curvilinear enclosures, D-shaped enclosures, rectilinear enclosures, enclosures with trackway entrances, enclosures with broad ditches, enclosures with outer compounds, field corner enclosures, enclosures with multiple circuits, and extensive enclosure groups (Roberts, Deegan and Berg 2010: 27-35). Excavation has shown that in most instances, the ditches were likely to have been external to earthen banks. Archaeology aims to identify patterning, but it must always be borne in mind that there is a danger of apophenia, and there will always be exceptions to broad generalisations or static typologies. In diachronic terms, it has been suggested that elsewhere in northern England, multi-vallate, irregular middle Iron Age enclosures became more regular and univallate during the later Iron Age, with single-ditched, subrectangular or rectangular forms common during the Roman period (Collens 1998). The many differences in enclosure size and form and variable dates from excavated examples suggest that this is too simplistic, and more local concerns of place, practice and identity might have been important. Many earthwork and cropmark enclosures remain unexcavated, and even those that have produced Romano-British material might have had earlier origins.
Medieval or post-medieval woodland plantings in South Yorkshire have preserved earthworks likely to be of Iron Age or Romano-British date (Buckland et al. forthcoming; Coutts 1999; Pouncett 2001, 2007; Whiteley 1992), including Scholes Coppice, Ecclesall Woods, Scabba Wood, Canklow Woods, Greno Wood, Wombwell Wood, Edlington Wood, and Peter Wood. An earthwork enclosure in Marr Thick Wood was levelled and ploughed in the early 1960s (Buckland 1986: 57; Cox 1984). Excavation showed that only the bases of the ditches survive (C. Merrony pers. comm.), though this remains unpublished. An enclosure in Roe Wood was destroyed in 1922 (Coutts 1999: 75). In non-wooded areas, earthwork enclosures shown on old maps or visible on historic aerial photographs have been ploughed-out or built on, as at Worsborough Common and Roughbirchworth Common. It is only comparatively recently that some enclosures have been surveyed in detail (e.g. Gowans and Pouncett 2010; Latham 1992; Lee 2005; Pouncett 2001, 2007).
Only a few enclosures in South Yorkshire’s woodland have been excavated. At Edlington Wood near Doncaster, a series of stone and earthen banks were noted by 19th century antiquaries (Armitage 1897: 36, 39; Hunter 1828: 90), but Roman brooch and coin hoard finds led to several surveys of a series of D-shaped and subrectangular enclosures built of limestone blocks facing stone and earth banks, and a series of stone buildings (Corder 1951: 66-69; Dolby 1973: 5-6; Ramm 1973: 28-31). Excavation of two superimposed enclosures during 1971-1973 found evidence for several occupation phases and pottery dating from the late 1st century AD (Sumpter 1973: 37-38). A complex sequence of modification, demolition and different uses followed, lasting until the early 4th century. The pottery, coin and metal finds indicate at least one relatively high-status settlement in the vicinity, either a successful ‘native’ farmstead that thrived during the Roman occupation, or perhaps a ‘Roman’ settler or someone retired from military service. Recent lidar analyses of Edlington Wood indicate that the enclosures formed part of a co-axial field system of low stone banks (Buckland et al. forthcoming).
Caesar’s Camp is situated within woodland in Scholes Coppice, on a slight rise on the edge of Rotherham. It was an ovate earthwork consisting of a single bank and ditch approximately 110m long and 95m wide, with a ditch up to 2.4m deep outside a bank originally around c. 3.5m high. The earthworks were surveyed in detail in 1992, when a narrow trench through the earthworks was also excavated. Only a few sherds of abraded 3rd to 4th-century Romano-British pottery were recovered from upper ditch fills (Atkinson, Latham and Sydes 1992: 40). A dip in the south-east side of the earthwork may reflect an original slumped entranceway, but due to later disturbance this is unclear. The enclosure had previously been interpreted as part of an Iron Age defensive network that included the Roman Ridge (Ashbee 1957). It was overlooked by an adjacent knoll, however, and so was unlikely to have been defensive.
At Scabba Wood, a sub-rectangular stone-walled enclosure investigated by the Doncaster Archaeological Group and the University of Sheffield had walls of double-orthostat limestone construction up to 0.5m wide with a stone rubble core (Merrony et al. 2017: 26). No clear entrance into the enclosure was identified, but a depression in the north-east corner led out into a possible holloway extending east-west past the site. The centre of the enclosure was hollowed as if worn or partly scooped out of the underlying limestone bedrock. There was probably an external ditch in places along with an internal ditch, which produced an Aucissa-type brooch of c. AD 40–65, stamped with the name ATGIVIOS. Though the internal ditch is an unusual feature, there are other regional parallels such as Royd Edge in West Yorkshire (Toomey 1982; Yarwood and Marriott 1988: 12). Worked flint, later prehistoric and Romano-British pottery, slag and a saddle quern fragment were recovered from within the interior, above an uneven surface of cobbling and natural limestone. The Roman pottery was mostly of 3rd to 4th century date. The enclosure was linked to one of a series of east-west banks faintly visible in a walkover survey and in lidar data (Chadwick and Robbins 1998; Merrony et al. 2017: 26, fig. 3), but also appears to have been within a block of prehistoric rectilinear fields and terraces, possibly overlain by the broad east-west banks. It is unclear if this was ever a permanently occupied ‘domestic’ enclosure, and it may have been principally used for livestock.
Castle Dike, Langsett used to be an earthwork but is now all but ploughed out. It is oval in plan (Bramwell 1973) and located on a south-west facing slope overlooking what is now Langsett Reservoir in the valley of The Porter or the Little Don river, on the Millstone Grit of the Pennine fringe at the west of the county. Geophysical survey demonstrated that it had two circuits of banks and ditches (Merrony et al. 1995: 90). No entrance is known. Features faintly visible on aerial photographs and on the geophysical survey plot might be possible roundhouses or building platforms.
Additional, smaller prehistoric enclosures may survive as earthworks on the Millstone Grit along the western edges of South Yorkshire and on into Derbyshire. Within South Yorkshire, more fragmentary fields and enclosures like those at Whitley, Wharncliffe (Makepeace 1985) may survive near Totley, Fox House, Higger Tor, Redmires, Broomhead Park, High Bradfield and Stocksbridge; but cultivation and woodland plantations may have damaged or destroyed many. To this author’s knowledge, none of the areas outside the Peak District National Park have been systematically mapped from historic aerial photographs or surveyed in detail and investigated further on the ground. Summaries of the Peak District evidence have been published elsewhere (e.g. Barnatt 1987, 2000, 2008; Bevan 2000).
These enclosures were subcircular or curvilinear in plan and defined by upstanding stone walls and banks rather than ditches. Remains of house platforms and stone roundhouses also survive. Such settlements often had irregular pens, garden plots or fields appended to them, and were rarely directly associated with larger, regular field systems or trackways. This might suggest a greater emphasis on pastoral agriculture, but cairnfields associated with some enclosures indicate clearance and cultivation. There may be a temporal dimension to this variation though – cairnfields might represent Bronze Age rather than Iron Age occupation (Barnatt 2000: 18-19). The settlements and fields on the Millstone Grit in the Peak District were somewhat different in character from those on the Magnesian Limestone and Coal Measures areas, and lowland locales (Barnatt 2000: 26-8; Bevan 2004: 56-65).
The date of upland enclosures and fields is difficult to establish, as many settlements could have been constructed and used from the middle Bronze Age well into the Romano-British period (Bevan 2000: 148). Few of these possible settlements have been excavated. At Gardom’s Edge near Baslow in Derbyshire, ceramics of late Bronze Age to earlier Iron Age date were associated with several roundhouses; and features with pottery along with residues on several sherds produced a broad range of radiocarbon dates ranging from 1300–700 BC, with some outliers in the lower range down to c. 400 BC (Beswick 2017a: 165-6). Swine Sty on Big Moor in Derbyshire produced ceramics and a radiocarbon date suggesting mid-Bronze Age inhabitation (Machin 1971; Machin and Beswick 1975), but 1st millennium pottery was also identified, though sadly this remains unpublished (Beswick 2017b: 186).
Returning to the cropmark landscape, so-called ‘clothes-line’ enclosures (Historic England thesaurus) were appended to or ‘hanging off’ pre-existing linear boundariesor trackways, and usually post-dating them, though later linear boundaries might have linked isolated enclosures. Excavated examples include Roebuck Hill, Jump (Robinson and Johnson 2007), Enclosures E4 and E5 at Redhouse Farm, Adwick-le-Street (Upson-Smith 2002), and High Street Shafton (Burgess 2001a). Few of these have been published. D-shaped enclosures, either isolated or integrated with field systems, have been excavated at Enclosure E7 at Redhouse Farm, Adwick-le-Street (Meadows and Chapman 2004; Upson-Smith 2002), and Engine Lane, Shafton (Burgess 2001b, 2003), both likely to be Iron Age in origin; as well as at Warning Tongue Lane, Bessacarr (Atkinson and Merrony 1994).
Subcircular or irregular enclosures that were isolated or in small groups may have been slightly earlier in date and linked to livestock herding. Some enclosures featured funnel-shaped or ‘antennae’ entrances and were often linked to trackways, and one example was excavated near Brodsworth (C. Merrony pers. comm.). This example is like some ‘banjo’ enclosures of central and southern England, where the few excavated examples originated in the middle or later Iron Age and were initially thought to be associated with livestock herding (Cunliffe 2005: 247; Fasham 1987: 8-9). Recent overviews have stressed that many were also occupation sites and thus a more complex and diverse group of features (Lang 2016: 356-7; McOmish 2011: 4). Subcircular enclosure E3 at Woodhead Opencast Site was linked to a trackway (Mudd and Webster 2001: 9, fig. 2), and two smaller subcircular enclosures were also excavated in Areas A and C at the same overall site (C. Jones 2003: 4, 5-6, figs 2, 3, 7). The sites at Brodsworth and Woodhead Opencast Site have never been published – publication of these should again be a priority.
Most enclosures on the Magnesian Limestone areas of South Yorkshire, and the co-axial ‘brickwork’ fields to the south-east, were rectangular or subrectangular in plan (Deegan 2007; Riley 1980: 49-50; Roberts, Deegan and Berg 2010: 28). Some enclosures had internal subdivisions. This is apparent on some cropmark examples, but also been shown in excavated examples such as Enclosure E7 at Redhouse Farm, Adwick-le-Street (Upson-Smith 2002) and Engine Lane, Shafton (Burgess 2001b, 2003). These subdivisions consisted of gullies, probably to support fences, or lines of postholes or stakeholes from fences and palisades. The subdivisions created different functional and social spaces, restricted entrances, and gradations of access and privacy. Such physical devices not only structured the movements of people and animals around enclosures but might have also controlled access according to kinship or clan, age, gender, or status (q.v. Giles 2007b: 242).
There is little evidence for what was originally present along the tops of banks. Some stone-walled enclosures may have wall footings topped with turves or fences (Buckland et al. forthcoming). At Balby Carr the remains of species such as alder, willow, hazel, hawthorn, and buckthorn were recovered from the base of some waterlogged field ditches, where tool marks, pollen and other indicators of cutting and coppicing suggest managed hedgerows (Gale 2007: 32; Greig 2007: 34-5). The banks around enclosures may have supported hurdle fences or timber palisades, with significant implications for the management of local woodland resources and tenure over these. Two planks found in the inner possible enclosure ditch at Balby Carr may have been from a palisade of split timbers (Daniel 2016: 19).
In the same way that some field systems in South Yorkshire have been shown to have an Iron Age origin (see below), the same is true of lowland enclosures. At Armthorpe, late Iron Age ceramics and hearths might have pre-dated enclosures and fields; but at Rossington Grange Farm, a subrectangular enclosure produced Iron Age and Iron Age tradition pottery from lower ditch fills and some pits, whilst a radiocarbon date from one pit indicated occupation during the 2nd to 1st centuries BC (Roberts and Weston 2016: 8). At Engine Lane, Shafton, the primary fill of the earlier phase of enclosure ditch yielded a radiocarbon date of 400–200 BC, a date of 380–50 BC from an internal partition ditch, and carbonised wood from the primary fill of the recut produced dates of 60 BC–AD 140 (Burgess 2003). A middle fill of an enclosure ditch at Wombwell Opencast site returned a 1st century BC to 1st century AD date (Mudd and Webster 2001). At Topham Farm, Sykehouse, the first occupation might have taken place in the 2nd century BC (Roberts 2003: 7-8, 27); and the layout might suggest an early ‘open’ settlement, with Iron Age and Romano-British pottery and radiocarbon dates indicating continuity until the early 3rd century AD. The evidence from a series of excavations by different units at Balby Carr indicates an initial ‘open’ settlement of roundhouses in the 3rd century BC, followed by a double-ditched enclosure during the 2nd century BC (Daniel and Barclay 2016; Jones 2007). At Rossington Grange, Sykehouse and Balby, many additional enclosures and fields were added during the late Iron Age and/or early Roman periods, indicating some continuities of landscape use (though not necessarily always continuous inhabitation), along with expansion or extensification (see below).
Many enclosure ditches around settlements were repeatedly re-cut, a phenomenon recognised across Iron Age and Roman Britain (Chadwick 1999: 160-4; Knight and Howard 2004: 93; Rees 2008: 73-7). Rather than routine clearing-out of ditches, re-cutting appears to have been more episodic, a set of inscriptive and re-inscriptive practices possibly linked to changes in tenure, calendrical rites or social events, or notions of identity (Chadwick 1999: 163, 2008c: 238; Giles 2000: 183, 2007b: 246; Sharples 1999: 106). There was emphasis on the ditch terminals by enclosure entrances and gateways, and these same locales were sometimes chosen for the more ritualised deposition of artefacts and animal or human remains (see below).
There is a significant bias in the data, with enclosures in lowland Sherwood Sandstone and Magnesian Limestone locales generally much more visible archaeologically through aerial photographs in arable fields than those on predominantly pastoral Coal Measures or Millstone Grit areas. Much development and construction in South Yorkshire also focuses on the same low-lying areas, further skewing the pattern. Without more detailed National Mapping Programme-style mapping of the western areas, accompanied by programmes of geophysical and topographic survey, it is difficult to see this situation being addressed.
There is a tendency to regard enclosures, especially those with evidence for ‘domestic’ occupation, as relatively permanent nodal points within the wider landscape. This is a static and historically contingent perspective (Aldred 2014: 24; Edgeworth 2014: 49-50; Fleming 2010: 15; Sheller and Urry 2006: 214). Many enclosures were inhabited or utilised for centuries, but several excavated examples do not appear to have had sustained dwelling within them, and it is possible that these were only constructed and used within a few generations; perhaps just one or two decades. Some may have been occupied intermittently as rights of tenure and the fortunes of families fluctuated over the years; others probably experienced periods of abandonment followed by later re-occupation, though not necessarily of the same character. Others may only ever have been occupied on a seasonal basis, and there may have been much greater movement around the landscape than often supposed. Paths and trackways might have been longer-lived features within landscapes than many enclosures (Chadwick 2016b: 106). Archaeologists need to explore these tempos and rhythms of inhabitation and movement more effectively, and rather than trying to pigeonhole enclosures into typologies, it would be more productive to investigate their different and dynamic biographies.
The extent of unenclosed, potentially earlier Iron Age settlement is largely unknown. Can archaeological prospection methodologies be refined in order to detect such slight features? Or will they continue to appear as chance discoveries in large developer-funded open area projects? Was the progression in some areas from open to enclosed settlements linked to increasing sedentism?
May the density and/or spatial extent of settlements of specific types and periods and within certain landscape zones be underestimated?
What factors, such as development pressure, skew the overall pattern? Can we use lidar and geophysical survey to prospect for such settlements on the Coal Measures and Gritstone areas in South Yorkshire, where cropmarks are rarer?
What internal features can be identified within these sites, and what subsistence and/or social practices were undertaken within them?
The nature and organisation of production, exchange and consumption at both Iron Age and Romano-British settlements requires further investigation.
Large ditches and associated earthen banks that extended for many kilometres are known from around Britain; and are notoriously difficult to date. Examples in East and North Yorkshire have been dated to the late Bronze Age and early Iron Age (e.g. Fenton-Thomas 2003, 2005; Giles 2007a; Spratt 1989; Stoertz 1997). In West Yorkshire, the Aberford Dykes complex was a series of linear earthworks located between the Rivers Aire and Wharfe, including features known as Grim’s Ditch, South Dyke, Becca Banks and The Rein. Grim’s Ditch produced radiocarbon dates of 777–396 BC, 790–400 BC and AD 86–335. Together with evidence for ditch re-cutting, this suggested construction during the earlier Iron Age, with redefinition in the late Iron Age, and/or during the late Roman period (Wheelhouse and Burgess 2001: 129). South Dyke yielded a date of 104 BC–AD 112 (ibid.: 132). The ditch was re-cut at least once, with secondary fills of the recut producing an amphora sherd and radiocarbon dates of AD 212–413 and AD 141–404 (ibid.: 135). Another intervention at South Dyke obtained a date of 740–390 BC from a pit underlying the bank (Gregory and Daniel 2013: 100-1), whilst the ditch produced a date of 370–160 BC. Some Iron Age or Romano-British field system boundaries in West Yorkshire clearly respected these earlier linear earthworks.
In South Yorkshire, the Roman Ridge or Roman Rig consisted of two lines of earthworks aligned south-west to north-east and totalling c. 27 kilometres in length, originating to the south-west of Wincobank hillfort and extending north-eastwards to Swinton Common and Mexborough, though the original limits of the banks and ditches are still unclear. The earthworks have been surveyed in the past (Preston 1950b) but are still undated despite several investigations (e.g. Atkinson 1994; Greene 1950; Greene and Preston 1957; Riley 1957). The Roman pottery recovered from secondary or tertiary ditch fills in excavated sections does not date the original construction; and it is not known if the two lines of earthworks were even contemporary. One investigation indicated that two stratigraphically earlier phases of ditch pre-dated at least one stretch of bank (Atkinson 1994: 47), suggesting that the earthworks have much more complex histories than is visible from the surface.
Ashbee (1957: 256-265) proposed that the Roman Rig was built hurriedly at the behest of the Brigantian leader Venutius in response to a deterioration in relation to Roman forces, as Alcock (1954) had posited for the Aberford Dykes; and that the earthworks thus formed part of an Iron Age defensive network along with Wincobank hillfort and the small earthwork enclosure of Caesar’s Camp at Scholes Coppice. This is highly problematic, however, not least because the linear earthworks never appear to have linked to either enclosed site. The hillfort and enclosure are also poorly dated, and a planned system of ‘defence in depth’ seems unlikely. The South Yorkshire linear earthworks have few if any clear relationships with field system boundaries visible on aerial photographs, so even their relative position in a landscape stratigraphy remains hard to establish.
Boldrini (1999: 103) also suggested an Iron Age date for the Rig, though admitted the earthworks could have been renewed in later periods. He believed that the banks and ditches were not defensive but social and territorial markers, marking the divide between Brigantian and Corieltauvian territories. Nonetheless, he considered it possible that they could also have dated to the 1st century AD, built as a reaction to the temporary Roman frontier between Chesterfield and Rossington. A post-Roman date has also been proposed (Cronk 2004), perhaps linked to the kingdoms of Elmet in the 5th to 6th century or Northumbria in the 7th to 9th century AD; but a Roman hoard of a brooch and 19 coins was discovered inserted into the Roman Rig in 1891 during railway construction (Addy 1893: 249), though it is possible that the earthworks were renewed in places in the early medieval period.
Such ideas may be trying to fit the evidence into too narrow a historical context, though overall an Iron Age origin (earlier or later) seems most likely. It is unlikely that the earthworks were defensive barriers – any overtly hostile forces could simply have outflanked them from the north-east and south-west. They may have been designed to be seen from the south-east though, helping to define territories and social identities, and perhaps channeling movements of livestock and people (Giles 2007a). The Roman Rig earthworks do have intriguing relationships with springs and streams, cutting across the line of many of these. Their construction would have required considerable time and labour from a significant proportion of the population. A more recent overview (Chadwick 2016a) stressed the dynamic and mutable nature of the Roman Rig earthworks and the need to establish the character of the relationships between banks and ditches, landscape, people, and materials as they were potentially understood in the past.
Shorter linear earthworks have also been identified in South Yorkshire. In Edlington Woods, the Double Dike earthwork is of at least post-Roman origin, and probably much later (Dolby 1973: 5-6; Ramm 1973: 28-31). It has a noticeable ‘kink’ in it where it respects an Iron Age or Romano-British earthwork enclosure. The full extent of the Double Dike is unclear, though Google Earth imagery suggests that it may have extended further to the east. There are also three possible curvilinear boundaries surviving as faint ploughed-out earthworks and cropmarks near Thrybergh Park (Deegan 2007: 25; Roberts, Deegan and Berg 2010: 47, fig. 60), following the contours and effectively cutting off an area between tributaries of the River Don – but unusually they curve inwards to the enclosed area, unlike promontory forts or cross-ridge dykes. One kilometre to the west at Ravenfield is what appears to be a curving multivallate defence work on the western side of the River Don valley (Deegan 2007: 25). It may be that they were a Corieltauvian counter to the Roman Rig earthworks. These undated features have never been surveyed in detail or excavated.
Aerial photography has been key to identifying Iron Age landscapes in South Yorkshire, with much pioneering work undertaken by Keith St Joseph and especially Derrick Riley (Riley 1975, 1977, 1978; St Joseph 1969, 1977). Riley outlined three basic categories of fields and field systems (1980: 13). His most famous were the so-called ‘brickwork’ fields, named because of their regular co-axial appearance and found on the Sherwood Sandstone areas of South Yorkshire and north Nottinghamshire. Riley also identified fields ‘nucleated’ around enclosures and those more ‘irregular’ in pattern – these two forms were more common on Magnesian Limestone and Coal Measures areas. He noted that Roman forts and the lines of Roman roads were superimposed across some blocks of fields (ibid.: 25-6); and suggested that the nucleated and irregular fields pre-dated the brickwork examples. He thought the latter at least might be pre-Roman in date.
Riley (1980: 26) and Hayes (1981: 117) believed that the comparatively large size of many brickwork fields was too great for arable agriculture, and that due to poor grazing and lack of water sources they were probably pasture, perhaps for sheep. Branigan suggested that the brickwork fields were primarily for pastoral agriculture, with sheep kept not for meat as Hayes had suggested, but to supply an expanding Roman wool industry (Branigan 1989: 164). He thought that the apparent uniformity of the brickwork fields in plan was due to them being established by Roman surveyors almost like centuriation, with the fields part of extensive, centrally managed Roman estates, and the enclosures housing the estate workers. The many fallacies and assumptions underpinning such arguments have been outlined elsewhere (Chadwick 2008a: 104-9; 2008c: 224-6; 2013: 18-20, 25).
An overview of Romano-British field systems and rural settlement in England used the terms ‘cohesive’ and ‘aggregate’ to describe the differing field patterns across the wider region (Taylor 2007: 59, 62-66). Based on more detailed analyses, Deegan noted several inconsistencies with Riley’s scheme, not least of which being that the ‘brickwork’ fields were not arranged in a truly brickwork pattern – the short ‘cross’ boundaries were rarely staggered in alternating strips (Deegan 2007: 5-6; Roberts, Deegan and Berg 2010: 20). Riley’s ‘nuclear’ fields were illustrated with cropmarks from Hesley Hall, near Rossington Bridge (Riley 1980: 13, fig. 3); but Deegan argued that the enclosure concerned might have been a different date to the surrounding boundaries. Finally, it has become apparent that Riley’s ‘irregular’ category is rather an unsatisfactory grab-all type.
Deegan in contrast proposed just two main types of field system. Her ‘strip’ fields consisted of long boundaries at least 400m long and up to 100m apart with short cross boundaries, arranged in ‘bundles’ of four or more strips (Deegan 2007: 5, fig. 6.5; Roberts, Deegan and Berg 2010: 20-2). These correspond broadly to Riley’s brickwork fields and extended between Adwick-le-Street and Bentley, and north of Adwick upon Dearne near Barnburgh. The implication is that the ‘strips’ were laid out as long boundaries and then subdivided by shorter cross boundaries. It is a broader category that accounts for how the fields were probably created (q.v. Widgren 1990: 22), but the term ‘strip’ fields may cause confusion with the later medieval fields of the same name. It might also imply that there was greater centralised planning and a shorter and simpler developmental chronology. In contrast, ‘mixed field systems’ were much more variable in size, although sometimes fields of similar sizes seem to have clustered together (Roberts, Deegan and Berg 2010: 22).
Most blocks of brickwork or strip fields were terrain sensitive, avoiding river valley bottoms and following subtle ridges or promontories of slightly higher ground on low-lying Sherwood Sandstone areas, the gently undulating Magnesian Limestone and the broad, shallow valley of the River Don north of Doncaster (Deegan 2007: fig. 6V.5; Roberts, Deegan and Berg 2010: 22). Many long boundaries were orientated towards rivers (Deegan 1996; Robbins 1998). On Magnesian Limestone areas such as near Scabba Wood, Sprotbrough for example, several sinuous long boundaries and trackways sometimes acted as ‘axial spines’ for later land allotment and division. It has been proposed that their sinuous course might have been dug along the lines of geological bedding planes (Roberts 2008: 197; Roberts, Deegan and Berg 2007: 7), though there is no archaeological evidence for this. Alternatively, and more likely, they followed the edges of cleared parcels of land and existing woodland. In other instances, meandering lines of boundaries and trackways may also have reflected the irregular routes taken by livestock moving through the landscape. GIS-based analyses undertaken on a few case studies have revealed that fields of similar area were sometimes clustered together (Roberts, Deegan and Berg 2007: 20, fig. 25).
Many major boundaries were aligned roughly north-south and east-west, as at Lundwood, Adwick-le-Street, Barnburgh, Scawthorpe and Scabba Wood (Chadwick and Robbins 1998; Deegan 2000, 2001a, 2001b, 2007; Meadows and Chapman 2004), or on NNE–SSW orientations. This is similar to broad alignments of later prehistoric co-axial field systems identified elsewhere in Britain (English 2012; Green 2013, 2016; McOmish, Field and Brown 2002), although some areas of these field systems, such as those south and south-east of Doncaster, are less strongly aligned on such perpendicular bearings, with a possible slightly greater emphasis on east-west boundaries (Green 2014).
Recent lidar analyses of Edlington Wood identified systems of low stone earthwork banks, even with possible evidence for ploughing and hand-dug ‘cord rig’ (Buckland et al. forthcoming), though the latter is less certain. These rectangular fields appear to have had a longer north-south axis, unlike the fields detected through lidar at Birklands and Belhaugh Hays in Nottinghamshire (Malone 2017), which are much more like the predominantly east-west longer boundaries of the cropmark brickwork or strip fields at Barnby Moor and south of Rossington. This may imply different dates of construction and/or use; or different agricultural practices. It may even reflect social factors, perhaps even identity (see below).
Higher locales on the Magnesian Limestone and Coal Measures may not have been enclosed, though this is hard to ascertain due to later agriculture and erosion. Fields might have occupied the land between hilltops and ridgelines, and valley bottoms. Areas of co-axial brickwork or strip fields that appear uniform on large-scale plans have been shown to be more complex and not created in a single phase. At Edenthorpe and Armthorpe, individual fields, blocks of fields and trackways were added accretively to one another over time – there were different periods of ditch maintenance and re-cutting, in addition to changes in orientation and emphasis (Chadwick 1995, 1999, 2008c, 2013a; Richardson 2008). Even apparently simple sequences of ditch infilling probably reflected the thoroughness of frequent cleanings rather than short periods of use (Chadwick 1999: 161; Magilton 1978: 62).
It is impossible at present (and may always be) to establish if the fields within specific blocks of field systems were in use for arable or pasture at the same time (cf. Hayes 1981: 116-7). The size of the bounded field areas also need not reflect the areas that were in actual pastoral or arable use. In Sweden for example, prehistoric and early medieval boundaries did not define cultivated areas themselves, which were smaller plots within them, evidenced by clearance, lynchets or traces of fencing (Petersson 1999, 2008; Widgren 1990). These were only detected through the stripping and excavation of internal areas of fields. Land allotment, land division and land use are not the same as land use (Chadwick 2008b: 3-5). In South Yorkshire, the internal areas of fields have invariably been truncated by medieval or later ploughing. At Balby Carr, a rare waterlogged fenceline of oak stakes broadly followed the alignment of the ditched field boundaries (O’Neill 2005, fig. 5), although it was not clear if this fence was within a ditched field. The stakes themselves were sampled but unfortunately have never been analysed in detail or dated.
The dating of areas of fields is still problematic given the generally low levels of even Romano-British material culture outside of enclosures, the fact that artefacts tended to be deposited in particular areas (see below), and the usually poor preservation of organic material such as bone or wood that might be suitable for radiocarbon dating. At some sites such as Holly Grove Farm and Barnsley Road, both at Goldthorpe (Merrony 1993; Scales and Weston 2015), and Nether Moor Drive, Wickersley (Marot 2017) no dateable artefacts were recovered, not even worn Roman greyware. Given the relative scarcity of Iron Age ceramics (see below), a lack of Iron Age pottery need not necessarily indicate an absence of Iron Age activity (Chadwick 1997, 1999; Magilton 1978), and even where Romano-British pottery is found in upper ditch fills, this may only date the silting up of ditches, not their original digging. Trying to rigidly separate ‘Iron Age’ field systems from ‘Roman’-period fields is probably futile.
Radiocarbon dating has therefore been key. Radiocarbon dating of wood and bone from field ditches at Balby Carr returned three dates between 370 BC–AD 260 (Daniel and Barclay 2016; Jones 2007; Rose and Roberts 2006); indicating that the fields were probably later than the initial unenclosed settlement, but perhaps broadly contemporary with the enclosed phase. There was also one statistical outlier of 800–540 BC that may result from residual charcoal. Later Iron Age pottery sherds were also recovered from a series of narrow, sinuous gullies excavated at the Zone D2 site at Balby Carr, probably spade-dug slots designed to drain water from within a field (Cumberpatch 2008a: 13-14; Muldowney 2008). During the Finningley and Rossington Regeneration Route Scheme (FARRRS) across Potteric Carr, a date of 210–50 BC was obtained from alder cones at the base of one ditch (Daniel 2017: 7-8). A waterhole truncated by a later ditch yielded a date of 160 BC–AD 80 from waterlogged plant remains at the base of the feature, along with sherds of an unusual Roman mica-rich dish dating to the mid to later 1st century AD (ibid.: 17). Alder charcoal higher in the waterhole sequence returned a date of AD 20–130 indicating infilling during the late Iron Age or early Romano-British period. Later Iron Age pottery was also recovered from a linear ditch at Stainforth Marina (Strafford 2014). Pottery of late Iron Age date was also recovered from the primary fills of two ditches at Goldthorpe Industrial Estate, whilst charred wheat from the primary fill of the outer ditch of a field corner enclosure was radiocarbon dated to 46 BC–AD 76 (Ross 2014: 9, 20). As the enclosure was constructed at the intersection of two field system ditches, the latter must have been earlier. Mid-1st to mid-2nd century AD pottery was also recovered from some field ditches.
The Iron Age dates from Balby Carr, Stainforth, Goldthorpe, and FARRRS at Potteric Carr indicate the progressive enclosure of some areas from the later Iron Age onwards – in these instances this was of low-lying alder carr, damp pasture and sedge fen. The fields at Potteric Carr and to some extent at Goldthorpe Industrial Estate were also noticeably less regular in form than many co-axial brickwork examples – several at Potteric Carr were trapezoidal in shape; though in this area and at Balby Carr the earlier fields and enclosures seem to have been gradually incorporated within a more regular landscape. Most dateable ceramics recovered from excavated field system ditches, however, are of 2nd and 3rd century AD date (Roberts 2008: 193). This is evidenced at sites such as Gunhills, Armthorpe (Richardson 2008), Stainforth Marina (Strafford 2014), Rossington Grange Farm (Roberts and Weston 2016), and FARRRS (Daniel 2017). Although as noted above this Roman pottery does not always date the likely inception of the fields, it nonetheless suggests possible agricultural expansion during the Romano-British period. In the absence of any real evidence for the intensification of production (either arable or pasture), then it may be that extensification was being practiced (see below). The former raises output per unit area of land by increasing labour, soil fertility or other resources; whilst in the latter output is increased by enlarging the area under cultivation or pasture without a significant associated increase in labour or other inputs (van der Veen and O’Connor 1998: 127-9).
Many of the landscape changes in South Yorkshire during the later Iron Age and Romano-British periods suggest extensification rather than intensification, with areas of open heath or communal grazing gradually enclosed over time. Some Roman period landscapes were possibly more ‘open’ with less trees than even some contemporary farmed fields (Smith 2002). Areas adjacent to forts and vici may have seen more changes, particularly increases in livestock numbers, but overall much of the archaeological evidence does not suggest markedly more intensive, centralised modes of production implemented during the Roman occupation. There have been no large Roman-British crop processing facilities excavated, for example.
The palaeo-environmental evidence for the Roman period is complex. Along the River Trent in Nottinghamshire, there were some episodes of flooding and deposition of alluvial silts (Knight and Howard 2004: 117-120; Rackham 2000: 115). There might also have been late Roman flooding for areas beside the Rivers Don and Idle in South Yorkshire (Buckland and Sadler 1985: ch. 5; Dinnin, Ellis and Weir 1997: 124, 147). Loss of woodland and deeper ploughing with increased cultivation might have caused higher levels of surface run-off and soil loss (Didsbury 1992; Riley, Buckland and Wade 1995). Stimuli for the extensification or intensification may have been population increase, and growing demand for agricultural produce driven by the garrisons along the northern frontier and burgeoning centres such as the vici at Doncaster and Castleford. There might have been added need for meat, hides and draught animals, particularly by the military (q.v. Allen et al. 2018); and an increase in the cropping of winter wheat for tax payments.
It is notable that to date, no co-axial field systems of Bronze Age date have been identified in South Yorkshire, though some of the small irregular or nucleated fields surviving on the gritstone uplands may originate in this period (Barnatt 2008; Cockrell 2019). This is unlike areas of south-central and south-eastern England, where many so-called ‘Celtic’ fields on the chalk downlands, or large-scale systems across areas such as the Thames Valley, had their origins in the Bronze Age, even if some of them were subsequently reworked and altered during the Iron Age and Roman periods. This may imply that large-scale land division (but not necessarily land allotment) using formal boundaries such as ditches, banks and hedges did not take place until the Iron Age. Perhaps Neolithic and Bronze Age communities in the region were still quite mobile (T. Cockrell pers. comm.); but this must remain a supposition at present due to the lack of evidence. It is possible that if Bronze Age ditches and gullies were shallower and not as regular as Iron Age and Romano-British ditches, they did not survive millennia of later cultivation. There is some limited evidence, however, of later field systems respecting earlier features – at Goldthorpe Industrial Estate, one of the principal ditches of a later Iron Age and Romano-British field system that produced Iron Age and early Romano-British pottery sherds curved around a Bronze Age stone cairn (Ross 2014: 9-10, fig. 6); and field ditches at Rossington Grange Farm seem to have used Bronze Age barrows as markers (Weston and Roberts 2016: figs 2-3).
At the other temporal end of their construction and use-lives, it is difficult to assess when many of the fields went out of use and/or were abandoned. It was proposed that post-medieval field systems were usually very different in plan to the later Iron Age and Romano-British co-axial brickwork fields in Nottinghamshire, with little evidence for continuity of boundaries into the post-Roman and earlier medieval periods (Unwin 1983). Whilst this conclusion may be generally true for many parts of South Yorkshire too, there are exceptions. At Armthorpe, many of the co-axial field system ditches were slighted by boundaries that have to be post-medieval or at the earliest medieval in origin, yet some field boundaries at West Moor Park East were on the same orientation as early modern fields (Gidman and Rose 2004), suggesting the latter followed the alignment of existing earthworks. Some field boundaries may have survived as hedges and/or banks and ditches for considerable periods, although this need not indicate direct continuity. Rather, the weathered traces of earlier occupation would have influenced later generations of ditch diggers and hedge layers. At Goldthorpe Industrial Estate once again, two corn-drying kilns produced radiocarbon dates of AD 331–533 and AD 422–561 from primary fills (Ross 2014: 15). A charred cereal grain from the fill of an adjacent field system ditch returned a date of AD 540–644. The positions of both corn driers respected the field system – one was situated in a field corner, the other was parallel to an existing ditch (ibid.: figs. 8a, 9). The ditches and any associated banks and hedges must therefore still have been extant features in the post-Roman period.
There are some intriguing potential wider narratives with all this evidence. There were many apparent continuities between the later Iron Age and Romano-British field systems and patterns, which might indicate an inherent resilience, resistance to change, or even cultural continuities (M. Giles pers. comm.). On the other hand, some areas of South Yorkshire appear to have been transformed, particularly during the 2nd and 3rd centuries AD, and after this there is also possible evidence for over-exploitation and abandonment in places. There is much scope for those writing reports and interpretative accounts alike to explore some of these bigger themes.
There is little direct evidence for the agricultural regimes undertaken within the fields, other than palaeo-environmental analyses (see below). Some might have been used for arable cultivation, others may have lain fallow for several years. Tenurial rights of access and inheritance may have meant some fields were effectively abandoned for years or even decades (q.v. Chadwick 2008a: 207; 2008c: 224; Giles 2007a). Fields may have rotated between arable, fallow and pasture, and manure would have been needed to maintain soil fertility. In Iron Age Scotland and the Northern Isles there was careful stockpiling of midden material, which was then introduced into the soil (e.g. Guttmann 2005; Guttmann, Simpson and Davidson 2005). In the study region, this could also have been achieved through folding animals onto the fields.
In general terms, it has been noted that the archaeological patterns of a pastoral area are different from those produced by mixed farming where arable production is also important. Mixed farming regimes require the separation of livestock from areas of cultivation, which can mean more complex patterns with stock enclosures around settlements from which droveways lead through areas of fields to pastures beyond (Ramm 1980: 31). Pastoral areas have sparser settlement and simpler patterns.
The South Yorkshire field systems ranged from more mixed or irregular, nuclear and ‘ribbon’ arrangements to co-axial and ‘brickwork’ groupings. The mixed or irregular, nuclear and ‘ribbon’ arrangements of fields in South Yorkshire were more likely to have been associated with mixed farming regimes – this may have been the case with areas around Adwick upon Dearne, Barnburgh, Goldthorpe, Sprotbrough and Adwick-le-Street. Mixed farming may thus have been taking place on the Magnesian Limestone and Coal Measures areas (Deegan 2007). In contrast, co-axial brickwork or strip fields more probably resulted from a greater emphasis on large-scale animal husbandry (Chadwick 2008a: 207-10), and areas around Rossington and Edenthorpe may have been predominantly for livestock, though it is likely that arable farming also took place within them.
At Redhouse Farm, Adwick-le-Street, the Roman road between Doncaster and Castleford passed only c. 60m away from an Iron Age and Romano-British enclosure (Area 7 E1). Sealed beneath the agger of the road were a series of furrows. Some were deep, stone-filled features that were probably part of Roman road construction practices (Meadows and Chapman 2004: 14). Another group of smaller furrows were likely late Iron Age or very early Roman ploughing pre-dating the road, which was probably constructed in the 70s or 80s AD. Soil micromorphology indicated that the deposits found underneath the road were buried soils (Upson-Smith 2002: 57; Usai 2004: 25-30), but unfortunately not whether these were cultivated.
Double-ditched cropmark trackways within South Yorkshire were sometimes sinuous, especially on the Coal measures and Magnesian Limestone; elsewhere such as on Sherwood Sandstone areas they could be straighter and more regular. They often made use of subtle changes in slope, as near Goldthorpe where a trackway followed a slight depression extending down into a river valley. Some trackways may have been created along more intangible traces of previous movement – different vegetation, trampled or rutted ground, and other ‘ancestral marks’ (q.v. Chadwick 2016b: 111; Giles 2007a: 109). Sometimes even double-ditched trackways were not created in a single phase but were a result of fields being added successively to one another, and trackways might have become single units only in later recuts. This too suggests that trackways were used as routes before they were ‘formalised’ with double ditches at a later date, when some might have become subject to greater social control and surveillance (q.v. Giles 2000: 179; Fenton-Thomas 2005: 58-59). Although trackways often appear to have been amongst the first features in a field system landscape, in some instances they can be seen cutting across pre-existing fields (Roberts, Deegan and Berg 2010: 24).
Trackways were not necessarily droveways, but many of them were orientated towards watercourses and floodplains and were up to 15-20m wide in places. Some were associated with livestock-related features such as funnel-shaped entrances, as at Gunhills, Armthorpe; Broom Hill, Harworth; Edenthorpe and Redhouse Farm, Adwick-le-Street (e.g. Richardson 2008: 13, 18, figs 10, 14; Riley 1980: 90, 103, maps 4, 12; Upson-Smith 2002: fig. 9). Some trackways led to what may have been large livestock corrals, as at Norton and Tickhill (Roberts, Deegan and Berg 26, fig. 32). Others were associated with enclosures with pronounced trackway entrances or pens attached to them (ibid.: 30, fig. 38), as at Brodsworth, Marr Thick, and Sprotbrough. This evidence all suggests the movements of large numbers of livestock through the landscape. Some trackways featured bulges that may be passing places, others have narrow constrictions, some of which might be ‘races’ used for the selection of livestock (Pryor 1998: 100-5, figs 52-53). At trackway junctions and gateways there is also evidence for sorting gates and other means of changing and restricting access.
On aerial photographs of some trackways the dark marks of such holloways are visible in between the double ditches, as with examples near Adwick-upon-Dearne, Barnsdale Bar and Bolton-upon-Dearne (e.g. Chadwick 2016b: 99; Roberts, Deegan and Berg 2010: 24). This is evidence of the passage of innumerable feet, hooves, and cartwheels over time, the settings for countless daily and seasonal movements. Some trackways were at least four to five kilometres long, so represented significant features in the landscape and considerable investment of labour. In traditional archaeological narratives of Iron Age and Romano-British landscapes, trackways are usually represented merely as means of getting from one settlement to another, or as purely functional features to assist with the herding of animals. The role of trackways as places in themselves is often overlooked. So too are the complex relationships between people and animals enacted within trackways, field systems, and the wider landscape. In addition to their functional role in channeling human and animal movements, it is necessary to consider the many social practices that would have taken place in and along trackways, and their consequent significant role in identity, memory, human-animal relations and everyday life in Iron Age communities (Chadwick 2007, 2008c, 2016b). People and animals would have met at passing places and junctions along them – meetings that could have been convivial; or alternatively sources of tension, even conflict.
An Iron Age inhumation burial at Bilham, Brodsworth was within a trackway but also near the funnel-shaped entrance into a ‘banjo’ enclosure possibly associated with livestock herding (Merrony pers. comm; see below). It may be that the trackway and burial were of two very different dates. It is more likely, however, that either the trackway was constructed across a known burial; or that the burial was deliberately placed within the trackway, in a location perhaps considered to be of great honour. The deceased young male might have been a cattle herder, or the community might have wished to have a known kinsman or ancestor looking out for their herds and flocks. The trackway might have been a material metaphor for a life’s journey, or a material mnemonic of a past life. Other Iron Age and Romano-British human and animal burials closely associated with trackways are known from the wider Yorkshire region and elsewhere (Chadwick 2016b: 102-5), and this suggests that trackways held great social significance and symbolic as well as practical importance.
Later Iron Age settlements, fields and communities tend to be regarded as relatively fixed and static, but the evidence from several enclosures in the region suggests that some were relatively short-lived (see above), in use for just a few decades or generations. Others, especially those on higher ground, might only have been occupied on a seasonal basis. Many people in later prehistoric and Romano-British rural communities might have spent more time moving along trackways and around fields than inside buildings and settlement enclosures. If such a more dynamic interpretative framework is followed, then it becomes clear that whilst enclosures and fields may have been in flux, major trackways might have been semi-permanent fixtures in the landscape, surviving in use over many centuries. At Adwick-le-Street, the grave of an adult Viking woman who had probably spent her childhood in Norway was dug into the fill of a Romano-British trackway ditch during the 9th century AD (Speed and Rogers 2004). This not only demonstrates that the trackway remained in use for many centuries, but also that it retained its social significance. Trackways reveal the importance of exploring the dynamic nature of landscapes, and the routine daily and seasonal movements of people and livestock.
Until the advent of developer-funded archaeology, few roundhouses had been excavated in South Yorkshire, those at Pickburn Leys being amongst the first (Sydes 1993; Sydes and Symonds 1985). Examples are still not numerous. There have been no finds of the Bronze Age and late Bronze Age/early Iron Age roundhouses seen in West Yorkshire at sites such as Swillington Common South and South Elmsall (Howell 2001; Grassam 2010); or further south in Nottinghamshire at Holme Dyke, Gonalston (Knight and Elliott 2008). Instead, most if not all roundhouses excavated in South Yorkshire are later Iron Age or Romano-British in date, and usually only survive as curvilinear gullies and/or concentrations of postholes. Truncation by ploughing means that most internal features do not survive, and frequently only partial shallow curvilinear features or some postholes might survive. Roundhouses are rarely visible on aerial photographs, and often not in geophysical surveys.
In other regions including West Yorkshire, some deep, narrow and steep-walled roundhouse gullies were probably the slots for upright wattle and daub walls. In South Yorkshire, two roundhouses excavated at Pickburn Leys had steep-sided rock-cut gullies up to 0.5m deep (Sydes 1993) that were possible wall slots. Most ring gullies, however, were probably eavesdrip gullies, with shallow-dug scoops eroded further by run-off from roofs. Examples include Pastures Road, Mexborough (Williams and Weston 2008: 6, fig. 8), and at least three circular structures at Carr Lodge Farm (Stanley and Langley 2013: 14, fig. 6). To date, neither site is published. The internal diameters of 23 roundhouses in South Yorkshire are shown below (Figs 01-02). They varied between 4.5–17.5m in diameter, with the greatest number (4) at 9m across. Two size clusters are visible. This may imply that on some settlements there were primary roundhouses and secondary structures.
The best-preserved roundhouses in South Yorkshire have been excavated at Balby Carr and Topham Farm, Sykehouse. Both these sites were sealed underneath alluvium deposits, although they were disturbed by early modern and modern drainage and agricultural activity; and appear to have been truncated in the past. At Balby Carr, remains of at least ten roundhouses were excavated (Daniel 2016: 20, figs 2, 15, 18; Richardson and Rose 2005: 4.7, 4.16; Rose 2003: 4.7-4.9; Rose and Roberts 2003: 4.4-4.8), including two possible partial arcs of ring gully.
Roundhouse diameters in South Yorkshire | Diameter (m) |
Balby Carr D(ii) Harley Davidson (AS WYAS) | 9 |
Balby Carr D1 Roundhouse A (AS WYAS) | Unknown |
Balby Carr D1 Roundhouse B (AS WYAS) | 4.5 |
Balby Carr D1 Roundhouse C (AS WYAS) | 8 |
Balby Carr D1 Roundhouse D (AS WYAS) | 7 |
Balby Carr D1 Roundhouse E (AS WYAS) | 7 |
Balby Carr First Point (AOC Archaeology) | 9 |
Balby Carr First Point Strip, Map & Sample Roundhouse 1213 (WA) | 7.6 |
Balby Carr First Point Strip, Map & Sample Roundhouse 1173 (WA) | 5.3 |
Balby Carr First Point Strip, Map & Sample Roundhouse 1061 (WA) | Unknown |
Carr Lodge Farm 219/224/229/234 (On Site Archaeology) | 9 |
Carr Lodge Farm 240/246/252/255/257/259 (On Site Archaeology) | 6.5 |
Carr Lodge Farm 311/315/319/331/335/339 (On Site Archaeology) | 13 |
High Street Shafton (AS WYAS) | 8 |
Pastures Road Mexborough (AS WYAS) | 6.4 |
Pickburn Leys Structure A (SYAU) | 12.5 |
Pickburn Leys Structure B (SYAU) | 9 |
Redhouse Farm Adwick-le-Street RH2 | 12 |
Topham Farm, Sykehouse Structure 1 (AS WYAS) | 17.5 |
Topham Farm, Sykehouse Structure 2 (AS WYAS) | 16 |
Topham Farm, Sykehouse Structure 3 (AS WYAS) | 13 |
Topham Farm, Sykehouse Structure 4 (AS WYAS) | 14 |
Topham Farm, Sykehouse Structure 7 (AS WYAS) | 15 |
Topham Farm, Sykehouse Structure 8 (AS WYAS) | 12.5 |
Topham Farm, Sykehouse Structure 9 (AS WYAS) | 7.5 |
Figure 2. Excavated South Yorkshire roundhouses and their internal diameters where known.
Roundhouse B on the Balby Zone D1 site was only 4.5 metres in diameter but was surrounded by a subcircular ring ditch c. 12m across with a south-western entrance. Within the Zone D(ii) (Harley Davidson) site at Balby, Roundhouse gully 13 was 9m in diameter with a central hearth pit and a possible internal partition. It was adjacent to a plank-lined pit (Richardson and Rose 2005: 4.7, 4.16, fig. 3). These were all enclosed by a more substantial ring ditch 15m in diameter, 1m wide and 0.80m deep. In addition to artefacts, this ditch produced daub with wattle impressions, and there may have been a wooden bridging structure on its eastern side.
The roundhouses investigated by AS WYAS at Balby Carr produced dates of 400–200 BC, with an open settlement pre-dating an enclosed phase (Daniel 2017: 7-8). Roundhouse 1213 at the Balby First Point strip and record site produced a radiocarbon date of 197–44 BC (Daniel 2016: 7, 20), and only one posthole and traces of a possible hearth were found. To date the various Balby Carr excavations have not been fully published. At Topham Farm, Sykehouse, seven roundhouses were identified, one within an oval rather than circular ring gully; along with two more ambiguous circular structures (Roberts 2003: 27-30, fig. 24). Only a few internal postholes and possible hearths survived inside them. Further to the west, excavation revealed a D-shaped or trapezoidal enclosure with an internal roundhouse that yielded a date of 50 BC–90 AD (Muldowney 2008: 2; Wilson 2006: 4). A partial ring gully of a roundhouse possibly c. 6.4m in diameter and associated with 1st century BC pottery was excavated at Parrots Corner, Rossington Bridge (Bishop 2010: 11).
The sites of several possible roundhouses are only suggested by clusters of postholes – examples include Barnburgh Hall (Richardson 2005a: 6.13), Billingley Drive, Thurnscoe (Neal and Fraser 2004: 24, fig. 14), Roundhouses 1 and 3 in Enclosure E1 at Redhouse Farm, Adwick-le-Street (Meadows and Chapman 2004: 6-7); and two possible roundhouse structures within Enclosure 8 (Upson-Smith 2006: 4). No detailed plans have ever been published of the Redhouse Farm excavations, however. There were two possible structures identified from masses of postholes and pits at Roebuck Hill, Jump – one a possible circular building and the other interpreted by excavators as being rectangular in plan (Robinson and Johnson 2007: 8, 12, 18, fig. 6), though a circular structure would also fit. This site too is unpublished. Very fragmentary ring gullies, which might have been roundhouses but could also have been wind-breaks, were recorded at Croft Road, Finningley (MAP 2006a: 32, fig. 31), Enclosure E2 Trench 10 at Woodhead Opencast Site (Mudd and Webster 2001: 10, fig. 3, 9), Nether Moor Drive, Wickersley (Marot 2017: 13, fig. ); and perhaps at St Wilfred’s Road, Cantley (Daley 2007: 10, fig. 3).
The footings of possible stone-walled roundhouses similar to those from the Pennines and Peak District have been identified at Gosling, Long Heath and Whitley near Wharncliffe (Makepeace 1985), and Canklow Woods (Lee 2005); but to date, none have been excavated. At High Street, Shafton, however, a curvilinear ring gully 8m in diameter had stone wall footings, a south-east facing entrance defined by postholes, and remains of a possible flagged surface (Burgess 2001d; Howell 1999). It may have been Romano-British in date. Some timber roundhouses built and used in the Romano-British period include Barnburgh Hall (Richardson 2005a), and perhaps Roebuck Hill, Jump (Robinson and Johnson 2007: 8, 12, 18).
In some instances, there seems to have been a concern with rebuilding roundhouses on almost the same place, as at Topham Farm, Sykehouse (Roberts 2003: figs 4, 8). This does not seem to have been the repair of existing structures but rather the repeated replacement of them, suggesting that sometimes there was a need to retain ties to very specific places. There are similar Iron Age examples from across the midlands and northern England (Chadwick 2013b: 299-300, fig. 15.7), and this may have reflected seasonal re-occupation after periods of abandonment, but also a concern with identity and social memory.
Where only one roundhouse existed on a settlement this was probably inhabited by one extended family, but with two or more contemporary roundhouses additional structures could have been used by other members of the lineage, or by certain age or gender sets. Ethnographic evidence suggests that there could have been men’s houses and women’s houses, houses for senior men or women, for ritual specialists, for young unmarried men, or for women during menstrual or post-natal seclusion. Others might have been ancillary buildings used for a variety of craft or storage activities – one roundhouse at Carr Lodge Farm was almost certainly used for metalworking (Stanley and Langley 2013: 14). The two roundhouses at Balby within their own enclosures might indicate a different social status for the inhabitants – one enclosure had a quite large 9m diameter roundhouse within it, but the other enclosed roundhouse was only 4.5m across, the smallest recorded in South Yorkshire.
There has been considerable debate about roundhouse doorway orientations. Oswald’s ethnography-influenced study of roundhouse orientations (1991, 1997) appeared to show that the majority faced east or south-east, perhaps due to cosmological cardinal directions such as the equinoxes and the mid-winter sunrise. Such symbolic associations were also proposed by others, based on patterns of daily use and deposits of artefacts and human and animal remains sometimes found within roundhouses (Fitzpatrick 1994, 1997; Parker Pearson 1996, 1999; Parker Pearson and Sharples 1999). Traditions of construction, cosmology and inhabitation were passed down through everyday embodied movements and practices, and pragmatic and more spiritual ideas may have coexisted (Giles and Parker Pearson 1999). Pope (2003, 2007) argued that Oswald’s patterns were less clear-cut, however, and that the landscape setting of roundhouses was more significant.
An analysis of South Yorkshire roundhouse doorway orientations (Chadwick 2008a: 590, Fig. 3 below) was only based on 20 identifiable entrances from excavated examples, but this found that 12 faced east or south-east, with smaller numbers to the north-west and south-west. The distribution is still more restricted than one would expect with random patterning and appears to be independent of topography. This suggests that social conventions and traditions did influence doorway orientation, even if practical considerations of light and prevailing winds were factors.
If so, did these correspond to different functional areas? Can patterns of internal human and/or animal movement be detected?
Do any differences apparent across South Yorkshire reflect functional adaptations to different landscapes, or might these have corresponded to social differences between groups?
See Pope (2003, 2007). At present, it is also not known whether most buildings were abandoned and left to decay and fall, or whether they were deliberately dismantled or demolished and the materials used elsewhere. Can any archaeological techniques be refined or developed to investigate this?
This requires more detailed consideration on future projects, especially the issue of whether there were deliberate foundation and/or closure deposits.
Four-post structures in later prehistory are often presumed to have been elevated granaries. Some consisted of 5–9 posts (Cunliffe 1991, fig. 15.2), and might also have been fodder ricks, general storage structures, chicken coops, menstrual huts, or platforms for the exposure of the human dead. Older (and less likely) interpretations include watchtowers, ‘fighting platforms’, and even elevated huts (Challis and Harding 1975: 150-1).
At Sutton Common, hundreds of postholes were found within Enclosure A, from which at least 115 four-poster structures could be identified with some certainty, with another 30–40 possible examples (Chapman, Fletcher and Van de Noort 2007: 115). Many appeared to be aligned in broadly north-south or east-west rows of 4–6 structures. Several six and eight-post structures were also identified. Flat-bottomed oak posts were mainly utilised for their construction, with a few Scots pines. Unfortunately, relatively few of the four-post structures were fully excavated or sampled for palaeo-environmental remains, and so the function of them as granaries must remain inferential. Most wood from the interior of Enclosure A was too degraded for dendrochronological dating. For some inexplicable reason, no wood or charred grain from the four-poster postholes was submitted for radiocarbon dating. This might have allowed chronological groupings of four-posters to be identified, but it unfortunately remains unclear how many structures were built and used at any given period – it must surely have numbered dozens at the very least. This might indicate crop storage beyond the household level and may potentially have represented the accumulated stores of an entire community. The possible groups of structures might have represented contributions by individual households and farmsteads, or lineages and clans. What is still unknown is where all the earlier Iron Age domestic settlements contemporary with Sutton Common were.
The excavated postholes of several Iron Age four-post structures at Sutton Common produced large quantities of carbonised cereals, mostly wheat and oats (Hall and Kenward 2007: 126). The grain to chaff and wheat to barley ratios sometimes suggested wheat being stored for human consumption, but other examples seemed to indicate mixed crops or maslins for animal fodder. In many older accounts it was proposed that charred grain in postholes was derived from fires in the structures above (either accidental or deliberate firing), but at Sutton Common it was noted that some charred remains were in the initial fills of the postholes, in some instances below the waterlogged remains of the posts themselves. This cannot be explained by above-ground conflagrations. It has thus been suggested that these were primary deposits, perhaps deliberate apotropaic offerings designed to bring good luck and ensure the safety of the stored crops (Van de Noort 2007: 133). Similarly, just one four-post structure at Sutton Common produced 18 saddle quern fragments from two of its postholes (Watts 2007: 145-6), and in addition to a prosaic function as post packing this might perhaps suggest deliberate deposition linked to ideas of fertility.
The rest of South Yorkshire provides a marked contrast to Sutton Common. Four-post structures, but mostly single examples, have been identified within Enclosure E1 and near Enclosure E8 at Redhouse Farm, Adwick-le-Street (Meadows and Chapman 2004: 6, fig. 4; Upson-Smith 2006: 6, fig. 3), and possibly at Roebuck Hill, Jump (Robinson and Johnson 2007: fig. 6). Several four and six-post structures were identified at High Street, Shafton (Burgess 2001a; Howell 1999, 2005); and some four-post structures may have continued in use well into the Roman period, as at Billingley Drive, Thurnscoe (Neal and Fraser 2004: 24, fig. 14). Otherwise however, they seem to have been relatively rare within the region.
Did they have other uses? Is grain storage merely an assumption? If used for grain, what storage capacity might have been represented by four and six-post structures? Did these represent household production, or more intensive/large-scale cultivation and crop processing?
Was it a late Bronze Age development, as 14C dates from sites such as South Elmsall (McNaught 1991) in West Yorkshire suggest?
There have been several small and more unusual gully or ditched structures excavated in South Yorkshire that cannot be readily be interpreted as roundhouses. Some are circular in plan, others are oval or even subrectangular in form, and which lack any internal features. All are associated with low-lying landscapes on river floodplains and/or areas that were likely to have been alder carr, wet meadow, or sedge fen in the past.
At Balby Carr, c. 750m to the north-west of the main area of Iron Age settlement and 150m east of a D-shaped enclosure containing a single roundhouse, there was an isolated oval gully approximately 9m across, with a 3m wide entrance in the north-east. This produced no dateable artefacts, only fire-cracked stones (Wilson 2006: 4). At Carr Lodge Farm, roughly 320m to the south-west of a group of roundhouses and some 30m from a ditched field corner of possible late Iron Age or early Romano-British date, there was an oval or subrectangular ditched feature 5.5m long and 2.5m wide. The ditch of this feature had no visible break in it and was 1.4m wide and 0.6m deep (Stanley and Langley 2013: 19). The fills of the ditch only contained some animal bone fragments and a piece of Roman imbrex tile, although post-medieval brick or tile and pottery was also recovered – it is unclear if these more modern finds were intrusive or not.
More recently during archaeological investigations ahead of the Rossington i-Port development, two subrectangular features were recorded close to the edge of a palaeochannel on the River Torne floodplain. One feature consisted of multiple rectangular postholes or plank slots defining a subrectangular area 6.8m long and 3.8m wide. The slots yielded no finds apart from a small fragment of burnt bone, but cleaning of the area recovered a struck flint flake and two amber beads (P. Daniel pers. comm.). At right angles and adjacent to this example was a second subrectangular feature 5.3m long and 3.3m wide, defined by a continuous shallow gully. No finds were recovered from this. Two subcircular features were situated 60m to the south and 150m east of the subrectangular examples. One was a penannular or C-shaped gully 4.4m across, the other a continuous circular gully 4.1m in diameter (ibid.). Again, neither feature produced finds.
The subrectangular features share some similarities with some of 70 or more small subrectangular gully and posthole-defined features excavated at East Carr, Mattersey in Nottinghamshire. These varied between 2–14m in length and 2–4m in width and were dispersed across the floodplain of the River Idle, many in clusters (Garton and Leary 2008; Morris and Garton 1998: 139). Some of the gullies contained Romano-British pottery, and at least three were cut by later co-axial Romano-British field ditches. These may have been drainage gullies around hay ricks or peat stacks, or drying stands for fodder, rushes or withies. It is also conceivable that several of the larger, more regular gullies might have surrounded tents or temporary seasonally occupied shieling-like structures built from peat, earth or turf (Chadwick 2008a: 157).
The flint and amber beads found above one of the subrectangular i-Port features suggest a prehistoric date (P. Daniel pers. comm.), in which case the Bronze Age and Iron Age ‘mortuary enclosures’ recorded at Sutton Common (Chapman and Fletcher 2007: 153-155; Van de Noort 2007a: 56) might be a better analogy, and perhaps these two performed a ritual or funerary function. The oval and subcircular features at i-Port, Balby Carr and Carr Lodge Farm have similarities to several features identified in West Yorkshire. These include a penannular gully 7.3m across excavated near Methley (MAP 1996: 29) on the floodplain of the River Aire, and a C-shaped gully 6m across recorded at Site 16 north of Wetherby, on the floodplain of
the River Nidd (Brown, Howard-Davis and Brennand 2007: 116-7, fig. 81). Iron Age pottery was recovered from the Site 16 gully terminals, and carbonised material from the gully produced a radiocarbon date of 100 BC–AD 90. They are also similar to small annular enclosures recorded at Moor Pool Close, Rampton near the River Trent (Knight 2000a, 2000b), which might have surrounded hay or fodder ricks. It may be of course that such features are only ostensibly similar and in fact performed many different practical and social roles, but their shared landscape setting and general lack of evidence for finds or occupation might be significant. More research on them is clearly required.
Can we use existing or new archaeological techniques to advance our understandings of them?
Roman-style villa complexes seem to have been rare in South Yorkshire, and across northern England were probably a late development during the 3rd and 4th centuries (Branigan 1980, 1984; Hingley 1989; Smith 2016; Wilson 1997). This lack of villas has biased discussions regarding ‘Romanisation’ and the region’s perceived marginality, yet this is only problematic for researchers familiar with southern England whose thinking is dominated by simplistic culture-history and core: periphery approaches. It is notable, however, that those settlements that have provided archaeological evidence for stone buildings or artefactual evidence for more ‘Romanised’ settlement were all located close to Doncaster and/or Roman roads.
At the likely villa site at Stancil north of Tickhill during the late 1930s, there was serious damage from a steam shovel digging a pipe trench, and salvage excavations were not stratigraphic and were poorly recorded (Whiting 1943). Several different phases of construction and occupation seem to have been identified, but the brief report is confusing and does not contain enough detailed illustrations or photographs, as was sadly typical of excavations of the time. A rectangular stone-walled building (Building I) on a north-west to south-east axis had a part-hypocaust floor with ceramic brick pilae stacks. On the opus signinum floor was a debris layer that included ceramic roof and flue tiles, decorated glazed floor tiles, and painted wall plaster that included floral motifs (Whiting 1943: 264). A possible north-west extension had a stone-flagged floor, and to the south-east there was an apsidal-ended room (Building II) with a red op. sig. floor that was probably a bathhouse. Traces of a third masonry structure were found south of the main complex.
Relatively small quantities of ceramics including mortaria and Derby ware cooking or storage vessels were recorded at Stancil, along with some platter rims (Whiting 1943: 267); but no samian, glass vessels, or coins. Many human remains were also identified, however, and these seem to have been mainly adult males, with some females and infants. These probably relate to a later re-use of the building as a church or cemetery, either in the late Roman or post-Roman periods. Unfortunately, their whereabouts are unknown and thus they have never been scientifically dated. There may have been geophysical survey and limited test pitting or trial trenching undertaken there by the University of Sheffield in the late 1990s or early 2000s, but no other details of such investigations are known.
Two possible villas or at least ‘higher-status’ Romano-British sites might have been located at Oldcoates and Conisbrough Parks (Buckland 1986: 38). The former was putatively associated with a patterned tessellated floor. The latter (south of Conisbrough itself) has had a geophysical survey undertaken, and apparently consisted of a bathhouse and timber framed buildings on stone footings (ibid.). It was excavated by the private landowner (A. Lines pers. comm.), and sadly no details concerning the site are available. Artefacts also hint at the presence of high-status sites at Loversall and Brodsworth (Cumberpatch 2004a; P. Robinson pers. comm.). At Chapel Hole, Braithwell, a possible Roman stone building was examined in the 1950s (Buckland 1986: 38) but again seems to have been poorly recorded.
Remains of a rectangular stone-footed building were excavated at Whirlow Hall Farm, on the south-west outskirts of Sheffield. The footings consisted of roughly coursed sandstone blocks without any bonding, presumably to support a timber superstructure (Waddington 2017: 37). On a broadly east-west axis, it was associated with an external metalled surface and close to an inner revetment wall for the perimeter ditch, but some internal stone paving survived within the structure too – it may have had an ancillary function. A stone block that might have been a support for wooden or stone columns in this or another structure was also recovered from a fill of the nearby ditch. The building had been quite heavily truncated by ploughing and much of the interior of the settlement lay outside the area of excavation, but relatively large quantities of pottery were recovered including grey wares, Derbyshire ware, gritty oxidised ware, Black Burnished ware, worn and abraded samian sherds, and shell-tempered wares of late 1st to 3rd or 4th century date. Lead waste and fragments of a blown glass vessel and a glass bead also perhaps indicate a more ‘Romanised’ identity for the inhabitants (ibid.: 121-2). The settlement enclosure was located on a spur of land on the side of a relatively gentle south-east facing slope but overlooking Limb Brook at the base of a steep slope to the west.
At Hazel Lane Quarry, Hampole, magnetometry survey identified a series of subrectangular ditched enclosures, but monitoring of soil stripping ahead of quarrying by Thames Valley Archaeological Services revealed limestone footings of an L-shaped building with an apsidal end featuring a hypocaust and a floor supported by ceramic pilae (Bevan 2006b: 26; Pine 2002: 3; Pine and Taylor 2006: 72). This was probably the bathhouse of an unknown villa or higher-status settlement. It is likely that the building was missed through geophysics as magnetometry often cannot detect buried masonry walls, unlike soil resistance survey, though the latter is a much more time-consuming method. Painted plaster and stone and ceramic roof tiles were recovered, along with a large quantity of pottery, mostly of 3rd century date. The decision was made to preserve the building in situ, so it was not fully excavated. To date, only a series of unpublished interim reports have been produced by TVAS rather than journal article or even a unified final assessment report. One of these has the only (small-scale) plan of the building (Pine 2002: fig. 4), and few details of the artefacts. It is also unknown where the main building or buildings associated with the bathhouse could have been located. Additional geophysics results suggested possible structures to the west (Pine and Taylor 2006: 72), but only a ditch was found. Google Earth imagery from June 2018 reveals the cropmark of a small rectangular ditched enclosure approximately 120m east of the quarry, but it seems likely that any associated dwelling would be much closer to the bathhouse.
The bathhouse at Hazel Lane Quarry was not situated within a double-ditched square or rectangular enclosure characteristic of many villas, and along with Stancil was not associated with any notable cropmark boundaries and/or field systems (e.g. Riley 1980: 92-94, maps 7, 8), aside from a funnel-ended trackway near Stancil probably associated with floodplain grazing. This could indicate that in the study region villas were established within existing patterns of land tenure. Even large villa complexes in south-central England have not had outer estate boundaries identified, however (Dark and Dark 1997: 73-74; Scott 2004: 54). Although villas were associated with some form of tenurial control, land-ownership and centralised control of production, and perhaps landowner and tenant (and slave owning) relationships; it is possible that patterns of land tenure existed which mean that such clear-cut boundaries may never be identified (Dark and Dark 1997: 74; Millett 1990: 203).
Detailed consideration of the landscape setting of these sites is also revealing. The Stancil site was situated on the north-eastern end of a low gravel ‘island’ or ridge between 5-10m OD in the otherwise extremely low-lying floodplain of the River Torne. Perhaps these were native ‘cattle barons’ who had supplied meat and hides to the garrison at Rossington Bridge, only 3.5km to the north-east. Alternatively, the complex was established by a serving or retired Roman officer or legionary. The possible north-south aligned Roman road identified by Deegan (Roberts, Deegan and Berg 2007: 17-18, fig. 8.4, 2010: 68, fig. 90) was only 1.75km east of Stancil, and this may have been another important reason behind its location. At Hazel Lane Quarry, the bathhouse building was sited on a slight plateau with the ground sloping away gently to the east and west, and more steeply to the south where it could potentially have overlooked the course of the Hampole Dike stream and the site of several springs. The Roman fort at Burghwallis was only 2.2km to the east, adjacent to the Roman road from Doncaster to Castleford. Iron Age and Romano-British farmsteads at South Elmsall and Redhouse Farm, Adwick-le-Street were close by.
Villas have long been interpreted as economic units in a Romanised, almost proto-capitalist economy (Branigan and Miles 1987; Rivet 1969), and as expressions of status and wealth. Millett (1990) saw villas as the products of success by Romanised native elites. Roman-style houses should not always be directly equated with wealth, however (Taylor 2001: 49). Other households might have chosen to invest and display wealth and status in livestock and arable land, in portable material culture, or through feasts (Hingley 1989: 159). Elaborate reception rooms of ostensibly rich owners might have sometimes masked financial problems (Samson 1990: 175), whilst nouveau-riche people may have had more rich furnishings than established families. Some villas may have had multiple occupancies with different resident households and families within them (Creighton 1992).
Roman-period timber-framed buildings are normally identified through postholes, post-bases and/or the narrow slots for timber sill beams or wattle and daub walls (e.g. Goodburn 1991, 1995; Smith 2016). These features may be relatively shallow and are thus particularly vulnerable to the truncation evidenced on many sites in the region as a result of more intensive ploughing since the Second World War. In some instances, horizontal timber sill beams might have sat directly on the original ground surface or on raised lines of clay, the weight of the completed building keeping it in place. This has been documented elsewhere in Roman Britain; and was also a recognised medieval timber building technique. Apart from traces of internal hearths or larger postholes or postpads for roofs, such buildings would leave very few archaeological traces. At Holme Hall Quarry there were remains of an oven and other features within the enclosure, along with ample evidence in the form of a midden deposit for sustained ‘domestic’ occupation, yet no actual structures could be identified (O’Neill and Raybould 2007: 102).
Within Enclosure A at Billingley Drive, Thurnscoe during Phase 1, a series of postholes marked what was regarded as an internal fenceline (Neal and Fraser 2004: 12-15, figs. 6-7), but a concentration in the north-west corner may instead have been remains of a possible rectangular structure. At St Wilfred’s Road, Cantley in Trench 3, shallow gullies included an L-shaped feature at least 14m long on its east-west axis cutting across the north-south ditch. The fill of another gully contained a fragment of Mayen lava quern (Daley 2007: 14). These gullies were either beam slots of timber buildings, or drainage features associated with such structures; of later 2nd or early 3rd century date. In Area 1 at High Street, Shafton (Burgess 2001a), at the central part of the enclosure a group of postholes may have been another poorly defined subrectangular structure. The rectilinear Enclosure 7 at Rossington Grange Farm may have enclosed a rectangular building (Richardson and Weston 2016: 32). At Warning Tongue Lane, Bessacarr, within the D-shaped enclosure a series of shallow rectilinear slots were probably the beam slots of rectangular timber buildings (Atkinson and Merrony 1994: 26). To date at least, the evidence suggests Roman-style rectangular timber buildings were scarce in the region.
Though Hunter (1831: 2) had described the surviving earthworks of the fort at Templeborough, it was not really investigated further until 1877–8 when discoveries of Roman artefacts on the surface of the by now ploughed site prompted excavations led by J.D. Leader and J. Guest on behalf of the Rotherham Literary and Scientific Society (Leader 1878). Though little detail was provided in Leader’s published report by modern standards, it is clear that substantial stone buildings were encountered in the southern part of the fort with more fragmented remains in part of the postulated vicus to the south-east outside.
In the First Word War, the expansion of the nearby steelworks led to salvage excavations directed by Thomas May during 1916–17. Despite difficult circumstances, this work revealed traces of heavily robbed stone buildings from different phases, including the headquarters building, which May called the praetorium, though he may be confusing this with what is normally termed the principia; and a ‘commandant’s house’, which elsewhere is normally called the praetorium. Several different phases of bathhouse, granaries and sections of some of the walls and gateways were also investigated (May 1922, plates iii-xi, xlvi-lii). Some stone walls supported sill beams for timber frameworks, but there were rows of stone column bases for the portico of the horrea (granaries). As was common practice at the time, work tended to follow the lines of walls whilst interiors and exteriors of buildings were dug out with little or no stratigraphic recording. Dorothy Greene later identified a wall from a possible large stone building several hundred metres south-west of the fort alongside the Roman road (Greene 1957b), whilst a proposed grid of roads and remnants of buildings on Brinsworth Common was interpreted as a putative Roman town (Greene 1957a, 1957c). The fragmentary remains and the confused reporting of them do not really support this notion though.
During more recent re-development of the former Templeborough steelworks site, developer-funded investigations recorded the truncated remnants of the fort ditches and other cut features, but also some clay-bonded sandstone wall footings in the former eastern vicus reflecting remains of a building or buildings at right angles to the road leading from the south-eastern gateway to the fort and probably fronting onto it (Davies 2013: 27-8, figs 15-16, 2016: 52-3, figs 16-17). Some stonework recovered from the 1916–17 Thomas May excavations (and possibly the earlier Leader investigations) is in Rotherham Museum, and might repay further detailed study.
In Doncaster, the rescue excavations in the 1960s and 1970s rarely encountered Roman building remains, due in part to centuries of later robbing and truncation, but also more modern disturbance including the developmental groundwork. Metalled Roman road surfaces and possible wall construction trenches were recorded in Frenchgate (Buckland and Magilton 1986: 45, 47-8, fig. 8 sites 2c, 2d), but at High Street a clay surface was overlain by stone wall footings and several postholes (ibid.: 57, fig. 12). This building was likely to have been of predominantly timber construction. At 8–10 High Street, fragmentary remains of Romano-British features included postholes, beam slots, stone wall footings and associated gravel floor surfaces (Chadwick and Burgess 2008: 41-3, 53-80, figs 15, 17, 22-3), though unfortunately only small interventions could be investigated due to the depth of stratigraphy within the constricted site and the hostile contractors who also hampered excavation. Linear beam slots within the early Roman fort at Doncaster were revealed at Church Walk, perhaps for military timber buildings (Chadwick, Martin and Richardson 2008: 14-15, fig. 5). The recovery of copper-alloy and iron lever lock rotary and latch keys from Doncaster, Rossington Bridge and Templeborough indicates the presence of buildings with more secure Roman-style doors (Lloyd Morgan 1986: 92-3, 2001: 20, fig. 23; May 1922: 77), but no similar artefacts have been found elsewhere in South Yorkshire. This has implications for notions of social space and privacy during the Romano-British period.
Can further buildings be identified at Stancil and Hazel Lane Quarry through the detailed use of geophysical survey techniques such as Ground Penetrating Radar? Would geophysical survey for possible villas or Roman-style buildings at Oldcoates, Loversall, Braithwell, Brodsworth and Conisbrough Parks also be useful?
Roman villas were not simply the country estates of wealthier rural incomers, retired soldiers, or the slavish emulations of Roman culture by aspirant native elites. Stancil and Hazel Lane Quarry could be interpreted in many of these ways, but are such arguments too simplistic?
Branigan (1980: 18) noted that most northern Romano-British villas were located east of the Pennines, mostly within the area of East and North Yorkshire considered to be the territory of the Parisi by culture-history approaches; north of the River Humber and north-east of South Yorkshire. He attributed the lack of villas elsewhere to a ‘positive aversion to the Roman way of life’ (ibid: 20), perhaps similar to the unconscious or deliberate ‘native resistance’ to Roman material culture proposed by Hingley (1996, 1997b). Can these arguments be proved or disproved?
In economic terms, most villas in southern England were associated with lowland arable production. Was one reason more villas were not constructed in the South Yorkshire area because of a corresponding lack of land suitable for arable production? The low number of urban centres in South Yorkshire would also have meant fewer local markets for the products of villa estates;
The past emphasis on establishing overarching explanations for the presence or absence of villas was derived from culture-historical and economic approaches. Although some of these interpretations are valid, they often do not adequately explain why villas were sited in particular places.
There are comparatively few examples of wells known from Iron Age Britain, particularly from rural settlements, so waterholes and natural watercourses must have sufficed in many instances for people and livestock. Many would have continued in use during the Romano-British period. Deep well shafts with timber or stone linings were largely a post-conquest development, however.
Five waterholes have been excavated at the FARRRS and i-Port investigations on low-lying land around Potteric Carr south-west of Doncaster. At FARRRS, the waterhole was originally subcircular in plan and up to 3m across but only 0.96m deep, with a bowl-shaped but slightly irregular profile (Daniel 2017: 17, fig. 9). It was probably dug in the later Iron Age and silted up in the early Romano-British period (see above); and was re-cut by a later boundary ditch. The finds and palaeo-environmental information from it are described below. At Rossington i-Port, four waterholes were investigated, the largest subrectangular in plan and 6.6m long, 4.7m wide and 1.7m deep, with quite steep sides and a flat base. Two also acted as sumps and were cut into underlying ditches (Daniel, Harrison and Powell 2014a: 18, 2014b: 6, figs 2, 4, 8). Roman pottery and stone finds were recovered from these, along with palaeo-environmental evidence (see below).
Two ‘ponds’ recorded at Finningley might also have acted as waterholes. One was up to 5.5m across and 0.95m deep; and was steep on three sides but had a more gradual slope on its northern edge for access. It was gently concave at its base and produced a relatively large amount of 2nd century Romano-British pottery and large pieces of timber, mostly tree branches but one interpreted as a step into the feature (MAP 2006a: 34-5, figs 23, 32). It also yielded a leather one-piece Roman shoe (see below). A second feature was of similar dimensions but did not produce any finds.
A slightly more complex feature was also excavated at Finningley Quarry. This was subcircular in plan and up to 5.4m across, with gradually sloping sides forming a cone or funnel. At a depth of 0.8m a square timber lining began, continuing to a depth of 1.6m (MAP 2006a: 35-6, fig. 32, plates 43-52). The near-vertical shaft continued to a depth of 3.25m. It produced 2nd century Romano-British pottery and stone fragments. Some Romano-British wells were relatively simple features. In one area of West Moor Park, Armthorpe, a well proved to be subcircular in plan and 4m across. It was excavated to a depth of 3m; and cored for another 1.1m without establishing the base of the feature (Hughes 2006: 13, figs 3, 8). The upper sides of the well had quite steeply sloping earthen sides, before becoming much steeper in a circular shaft. There were tool marks on the sides of the well, but only a small quantity of pottery was recovered. The report makes no mention of any well lining, but it seems unlikely that a near-vertical deep shaft could have survived for long without some support, unless it was dug out on a regular basis.
At 8–10 High Street Doncaster, a probable Roman well 2.15m in diameter was identified but it could only be excavated to a depth of 1.20m but auguring suggested it was 2.10m deep (Chadwick and Burgess 2008: 44, fig. 15, plates 10, 19-20). It did not have a timber lining.
At Templeborough one well was investigated by J.D. Leader’s team, three were excavated by May, and one recorded during construction work for the steel mill (May 1922: 35-6, 57, 59, figs xli, liii). The one dug in 1877–78 by Leader was ‘set round with sandstone rubble (ibid.: 59) which may refer to a stone lining or superstructure, and it was apparently ‘29 feet’ (8.8m) deep. Another investigated by May in 1917–18 was around 5.6m deep and had stone lining to a depth of 1.5m where a square framework of timbers began. Another was 1.2m in diameter and lined with stone to a depth of 2.7m, but it could not be bottomed due to water ingression (ibid. 57). The well recorded in 1918 had a steep-sided conical upper section that was 3m across and stone-lined, some of larger stones probably acting as steps down into the feature for drawing water (ibid. 35). From around 2.6m down there was an oak box-lining of the square shaft to a depth of 6.7m below the original ground surface, after which the well shaft continued to a total depth of 9.9 metres.
Can we develop methodologies to investigate them more effectively?
There is little evidence of Llyn Fawr period artefacts in South Yorkshire circa. 800–700 BC (O’Connor 2007). A decorated coper-alloy Sompting-type socketed axe (PAS 52EFB233001BD4; SYAS HER 599169), and a copper-alloy ‘swan’s neck type’ ring-headed pin (PAS 4E14515C00134B; SYAS HER 452337) recorded by the Portable Antiquities Scheme (PAS) may represent the earliest known Iron Age metal items. Sompting-type axes are broadly dated to 800–600 BC (Boughton 2015; Burgess 1971), whilst swan’s neck ring-headed pins are usually attributed to the Hallstatt C phase of the Iron Age or 700–400 BC (Becker 2008; Dunning 1934). The pin was originally identified as a late Iron Age or Romano-British end-looped cosmetic pestle. A fragment of gold bracelet or ingot was found with a metal detector at Sutton Common on the excavation project in 2003, in trench backfill (DCMS 2003: fig. 26; Hill 2007: 160-1). It is not closely dateable but was probably older than 200–100 BC. It may have been originally deposited near the western side of the main enclosure, close to some putative mortuary enclosures, and possibly associated with them.
Most of the few Iron Age-style metal objects recorded in South Yorkshire date to the later Iron Age or early Romano-British period – the 2nd century BC to the 1st century AD. A bronze sword or dagger chape found near Sprotbrough (still not fully described and published) and a decorated copper-alloy torc from Dinnington were all chance or metal-detecting finds (Beswick et al. 1990; Buckland 1986: 6). The torc is probably of later 1st or even early 2nd century AD date. It was found c. 140m away from a small subrectangular earthwork enclosure, and the find spot on a south-east facing slope was relatively unremarkable; although there was a predilection for the deposition of coin hoards on eastern and south-eastern slopes during the Iron Age and Roman periods (Bland et al. forthcoming). The possible significance of the Magnesian Limestone area for metalwork deposition is examined below. Found in six separate sections, the torc is of unique construction, but its decoration has affinities with other beaded torcs found in northern England (Beswick et al. 1990: 24-5), whilst its fastening is like examples in south-western England and Wales.
A copper-alloy tankard handle, an enamelled linch-pin, a horse harness toggle and terret ring were metal detecting finds from Rossington Bridge (O’Connor 2001: 91), and there have been metal detector finds of some brooches, harness fittings, and fasteners. A small cylindrical copper-alloy bead was found in a ditch fill outside the northern entrance of the enclosure at High Street, Shafton (Burgess 2001a), and two lead spindle whorls were recovered from Redhouse Farm, Adwick-le-Street (Upson Smith 2002: 24). At the subcircular enclosure at Marr Moor, in an otherwise finds-free ditch on the south-eastern side of the enclosure there was a complete, fresh-looking pennanular brooch of late Iron Age or early Romano-British date (C. Merrony pers. comm.). A headstud brooch of later 1st century date was excavated at Enclosure 8 at Redhouse Farm, Adwick-le-Street (Upson-Smith 2006: 8-9, plate 5), and the southern ditch of Enclosure E1 produced a copper-alloy Romano-British brooch from the fill of a recut (Meadows and Chapman 2004: 4). A trumpet brooch of later 1st century date was found at Billingley Drive, Thurnscoe (Cowgill 2004a: 49, fig. 27).
Brooches may have played important roles in projecting aspects of people’s identities, perhaps reflecting changing social ideas of the body, personhood and community during the Iron Age and Romano-British periods (Eckardt 2005, Farley 2012; Giles 2008; Hill 1997; Hunter 2007; Jundi and Hill 1998; Taylor 2013). In north-eastern England they were comparatively rare during much of the Iron Age, but their use became more common during the 1st and 2nd centuries AD, especially at vici, villas and roadside settlements (Brindle 2018: 28). Group C brooches (e.g. Nauheim Derivatives, Colchester and Colchester derivatives, headstud) were the most numerous (Mackreth 2011), with penannular and plate brooches also being well represented. A disc brooch with repoussé triskele design found at Rossington Bridge Pumping Station probably dated to the mid-1st century AD (Lloyd Morgan 2001: 18).
Dragonesque brooches were a distinctive form with a relatively restricted distribution in Britain, focused on north-eastern England (Brindle 2018: 30, fig. 2.28). The majority have been recorded by the Portable Antiquities Scheme, but an enamelled example was found at Templeborough (May 1922: 71, plate xiv.1), one at Holme Hall Quarry near Stainton (Bevan 2006b: 31), and another at Church Walk in Doncaster from an early Roman fort context (Cool 2008b: 136, 138, fig. 40). It has been suggested that this brooch form might have been associated with a particular social or tribal group, perhaps the Brigantes (Mackreth 2012: 12); but such distributions may also reflect localised workshops and marketing as there were probably at least two further regional sub-types (Eckardt 2014: 128-132; Hunter 2010: 100). Aucissa-type bow brooches were a novel Roman form introduced from the continent, with possible military associations. The one found in the ditch of the enclosure at Scabba Wood with the stamp or maker’s mark ATGIVIOS may date to c. AD 40–65 (Mackreth 2011: 132; Merrony et al. 2017: 55, fig. 30). This may be one of the earliest Roman artefacts in South Yorkshire. An unstamped example was recovered by metal detector near to the vexillation fortress at Rossington Bridge (O’Connor 2001, fig. 55.2). All told, 14 Romano-British brooches were recovered from the 1916–17 excavations at Templeborough (May 1922: 71-3, plate xiv), including two early 2nd century examples found in the same pit by the baths; and also a bow brooch fragment from the more recent investigations (Rogers 2016: 71). Ten brooches were found during the 1960s–1970s excavations in Doncaster (Lloyd Morgan 1986: 84-93, figs 19-20), with another few examples known from older finds.
Isolated finds of Iron Age and Romano-British brooches on excavations or by metal detectorists are usually regarded as accidental losses, yet many discoveries from South Yorkshire have been made away from or on the edges of settlements known from cropmarks and other evidence (Dearne and Parsons 1997; Lloyd Morgan 2001; O’Connor 2001). This phenomenon has been noted in other regions of Britain (e.g. Farley 2008, 2012). Worn on clothes and to fasten cloaks, some brooches may have caught on other clothing or been snagged by vegetation. But were Romano-British people always so careless with their personal items? Brooches were highly personal items, worn close to the body, and rubbed by hands and clothing (Giles 2012: 140). Some could have been heirloom objects, acquiring biographies and evoking memories. Even old or broken examples may have been valued therefore, and it is also possible that some breakage was deliberate. During the Romano-British period, some brooch types such as plate brooches and ‘horse and rider’ examples were possibly associated with visits to temple and shrine sites, and/or with certain deities. It is possible that some brooch finds were not accidental losses at all but rather were personal apotropaic or propitiatory offerings made at places in the landscape.
A group of late Iron Age and early Romano-British brooches were found in river deposits at Rossington Bridge and between the vexillation fortress at Rossington and the River Torne (Lloyd Morgan 2001: 16-18; O’Connor 2001: 91, fig. 55). Part of an iron cauldron chain and a ‘poker’ were also recovered from palaeochannel peat deposits at Rossington Bridge (Buckland, Hartley and Rigby 2001: 10), close to a possible wooden trackway. Many cauldrons and suspensory chains have been found in Iron Age or Romano-British river, hoard or grave contexts (Manning 1983), whilst the ‘poker’ with its paddle-shaped end is an enigmatic artefact whose exact function is unknown, though they may have been part of hearth furniture, or used in smithing. The Rossington Bridge finds may represent material in an eroding midden, as an ‘occupation layer’ of black organic sand with artefacts and bone was noted, though the records are poor. Alternatively, they could have been local objects discarded by the Roman military during their occupation nearby; or might have been late Iron Age or Roman deliberate deposits. Some human bone that was also recovered exhibited signs of deliberate disarticulation and defleshing (ibid.: 82). These finds may thus reflect the importance of wet and watery places in prehistoric and Romano-British depositional practices and cosmological beliefs (Bland et al. forthcoming; Chadwick and Ghey 2015; Walton 2016). The presence of the river, together with a possible wooden trackway and evidence for unusual depositional practices, suggests that this site had special social significance before or even during military and industrial use.
Additional iron objects from Rossington Bridge included hipposandals and a possible pilum or javelin head (Lloyd Morgan 2001: 20-23). A Roman military dagger found during dredging of the River Torne in 1988 near Rossington fortress and now in Doncaster Museum had an iron blade, a bone handle and a wood and leather sheath. Of likely mid-1st century date (P. Buckland pers. comm.), it has never been published. Possibly an accidental loss, it might also have been a deliberate deposit in a wet locale. The Doncaster shield was either from an auxiliary cavalry unit or was a campaign trophy (Buckland 1978, 1986: 51). At Doncaster, iron spear and pilum heads, copper-alloy sword and dagger chapes, terret rings, a phalera, a patera handle, lead spindle whorls and various iron and copper-alloy harness and military fittings were identified, and a small bronze child or cupid figurine was found in the 19th century (Buckland 1986: 14-15; Cool 2008c: 279-81; Lloyd Morgan 1986: 83-93, figs 18-20; Lloyd Morgan and Buckland n.d.: 7-11, 37-8, figs 2-4, 7-8). Roman metal items from Templeborough included iron tools such as a mattock and a scythe anvil, weapons such as spearheads, lead weights and a plumb-bob, and copper-alloy phalerae or harness mounts, terret rings and belt fasteners, ferrules and a handbell (May 1922: 74-80, plates xiv-xviii; Rogers 2016). An iron spade shoe from Gunhills, Armthorpe (Cool 2008a: 48, fig. 33) is an interesting everyday agricultural tool. Lead spindle whorls were found at Redhouse Farm (Upson-Smith 2002: 24).
The distribution of metalwork across South Yorkshire should not be taken at face value, however, as there are recognised biases in the PAS data (q.v. Brindle 2011; K. Robbins 2012, 2013, 2014). It is notable how many finds are from the Doncaster district. This reflects a greater preponderance of arable agriculture and ploughed fields on the Magnesian Limestone and Sherwood Sandstone areas at the east of the county, contrasted with mainly pasture and moorland to the west on Coal Measures and Millstone Grit locales. The Doncaster district also contains several accessible woodlands with upstanding Iron Age and Romano-British earthworks; and seems to have a much higher number of detectorists than other parts of South Yorkshire. Although many metal detecting finds from the area have been made by responsible detectorists and recorded by the Portable Antiquities Scheme, it is likely that several coin hoards and other metal objects have been illegally dug up without any recording (P. Robinson pers. comm.). There has long been a problem with ‘night hawkers’ in *these areas, and many known enclosure sites have been plundered. Iron Age coins may be under reported, and the Yorkshire region is one of the parts of Britain most adversely affected by illegal detecting (Oxford Archaeology 2009).
Nonetheless, better relationships between archaeologists and detectorists facilitated by the PAS have led to the recording of some interesting Roman-period metalwork too, including a Roman gold marriage ring found near Bawtry (DCMS 1999: fig. 21), and a wine strainer or dipper handle from Marr with a maker’s mark (DCMS 2006: 49; PAS SWYOR-1CD5F6) likely to be of mid-2nd to 3rd century date. This find might reflect more Roman-style drinking practices, but these artefacts could also have been used for beer or for medicinal purposes – an example deposited in the so-called ‘Doctor’s grave’ in 1st century AD Colchester contained residues of Artemisia or mugwort, commonly used for healing remedies (Crummy 2007: 397-8).
Cosmetic grinders, tweezers, scoops, probes and nail cleaners were excavated at Doncaster and have occurred in limited numbers on rural sites, or as isolated metal detector finds (Buckland 1986: 27; Dearne and Parsons 1997: 73, fig. 9; Lloyd Morgan 1986: 88, fig. 19.14). A cosmetic mortar was found at Edlington Wood (Corder 1951: 90-1, fig. 17: 9), along with trumpet and penannular brooches, and a miniature decorated bronze vessel. Two stone palettes that were part of toiletry sets were recorded in Doncaster (Lloyd Morgan 1986: 95-6, fig. 21.6-7). Overall though, toilet and grooming instruments were apparently rare on rural settlements. This might indicate that outside urban centres and ‘Romanised’ settlements, there was less concern to establish or maintain ‘Roman’ identities through personal grooming.
There have been detectorist finds recorded through Treasure Trove and the Portable Antiquities Scheme (PAS) of usually single gold staters, silver issues and bronze potins, mostly of Corieltauvi manufacture from further south, including a gold stater from Bawtry – interesting given the possible Romano-British shrine site nearby (see below). Overall, however, Iron Age coin finds from South Yorkshire are relatively rare in comparison to other regions, and no large hoards have been recorded.
Roman coinage was a little more common according to Treasure Trove and PAS data, as scattered detectorist finds of individual coins but also as coin hoards (but note the caveat about distributions above). Roman coinage seems to have been comparatively scarce in northern England away from urban centres such as Doncaster and York, vici and military bases. The use of coins was probably uncommon on many small-scale rural settlements until at least the 4th century (Allen et al. 2016; Walton 2012). This pattern challenges the long-assumed notion of a gradual ‘evolutionist’ trend in coin supply and use whereby monetisation occurred as part of a wider process of acculturation or ‘Romanisation’ (Walton and Moorhead 2016: 836). One stimulus to coin supply in northern England was the military presence immediately prior to and following the Roman invasion of the north – this may well explain the finds of Republican and early Imperial coins at Rossington Bridge (O’Connor 2001: 91). Another might have come in AD 208 with the campaigns in northern Britain and the movement of the Severan Imperial household to York (Creighton 2014: 19, 26). The larger quantities of coinage on some rural sites in the early 4th century AD could also have reflected changes in how the Roman administration organised and controlled agricultural output (Walton 2012, 2014).
The extent to which South Yorkshire and other areas of northern England, Wales and south-west England were incorporated within a monetary economy is debateable. Outside of Templeborough and Doncaster, few excavated sites in South Yorkshire have produced Roman coins. At Enclosure E1, Redhouse Farm, Adwick-le-Street, three copper-alloy coins were found, including a 1st or 2nd century AD coin and an unstratified 4th century issue (Upson-Smith 2002: 23). There were three copper-alloy coins at Holme Hall Quarry (Jones et al. 2007: 66), one of 1st century date and two from the 3rd century. Six coins were found at Billingley Drive, Thurnscoe, all antoniniani radiates or radiate copies (Brickstock 2004: 51), and two at Armthorpe, of 1st and 3rd century date (Cool 2008a: 47). There was a single 1st century coin at Rossington Grange Farm (Sitch 20: 26). Coin loss or discard is not the same as coin use, but nevertheless this evidence does not appear to indicate a monetised Roman rural economy. Barter and exchange may have remained important.
Most Roman coin hoards in South Yorkshire are older finds that have been poorly recorded, and a fuller account of their distribution will be published elsewhere (Bland et al. forthcoming), but several hoards have interesting and noteworthy depositional contexts. These include some of the Roman coin hoards found within Edlington Wood. Two were associated with the stone-walled enclosures (Corder 1951: 66-69; Dolby 1973: 5-6; Ramm 1973: 28-31). One consisted of 80 silver denarii and one antoninianus of the 3rd century, and a small piece of lead-tin alloy, lying in and around sherds of a small Castor ware ceramic beaker. Within one or two metres were fragments of a calcite-gritted, handmade ‘native style’ pot associated with 356 denarii and 172 antoniniani. It is not known if the pots were buried in small pits, postholes or stone-lined cists; but they had clearly been disturbed. Robertson (1935) believed that the two groups of coins were parts of a single hoard which overflowed from the small beaker into the larger pot. This seems rather dubious from a taphonomic perspective. It may be that although the two groups of coins were originally separate, they were deposited as part of the same event, or at least within a very short while of one other. Their proximity indicates knowledge of the previous coins by the person who undertook the subsequent deposition. Corder’s 1951 plan shows the location of the two pottery vessels and the coins.
Recent analyses of other Roman coin hoard find spots around England and Wales suggest that natural features in the landscape influenced the place and nature of deposition (Bland et al. forthcoming; Chadwick 2014). In Edlington Wood, another hoard found in 1935 was only 85m north-east of the two hoards above; and it featured 59 3rd century antoniniani, possibly within a ‘native-style’ calcite-gritted jar, sherds of which were found in the same area (Corder and Percy Hedley 1945). The hoard was below the rocky limestone outcrop of Edlington Crags. Several other coin groups subsequently found above and below the crags may be separate deposits but are more likely addenda to the original disturbed hoard. Many of the rock formations have an unusual protean quality to them, and although is impossible to know how vegetated the area would have been in the past, the find spot may have commanded views north-east and eastwards across the valley of Warmsworth Beck and St Catherine’s Well, a spring site. There used to be another water source at Alverley Spring, c. 200m to the east, which is now a pond. The coins were close to a Romano-British enclosure, however, and thus part of a prosaic landscape; though such modern distinctions may be far too simplistic.
Creighton (2014: 25) notes that although four of the radiate hoards from Edlington Wood date to around the AD 260s, a fifth was earlier and had a closing date of AD 225, along with a structure like coin hoards a decade earlier. He therefore suggests that hoards were being deposited at this locale over at least 50 years, and this depositional activity resembles the multiple deposits more commonly associated with Iron Age coin hoards at shrine sites. There is growing acceptance that the large quantities of coins recovered from Romano-British temple sites and from ‘votive’ deposits represent a significant accepted ‘function’ of Roman coinage at the time, as social and economic transactions with deities (Walton and Moorhead 2016: 841).
At Low Hall Wood north of Sheffield, a hoard of 738 3rd century antoniniani radiates was discovered and although only a six-figure grid reference was recorded, this places the presumed find spot on a north-east facing slope, probably with a stream within 50-100m, and also within c. 50-100m of a spring (Bland et al. forthcoming). Nineteen Roman coins along with an inlaid brooch were found in the Roman Rig linear earthwork during railway construction in 1891, at the base of a slope near Wincobank hillfort. The artefacts were underneath a flat stone inserted into either the linear earthwork bank or the ditch – there are conflicting accounts (Addy 1893: 249; Sheffield and Rotherham Independent 24th August 1891). Several other coin hoards were located close to the line of the Roman Rig, which may indicate its continued social and mnemonic significance over time (Chadwick 2016a).
There is an interesting recorded cluster of eight Roman coin hoards along the south and south-east facing bank of the River Don Gorge, near the villages of Cadeby and Sprotbrough. This would be a notable concentration anywhere (Bland et al. forthcoming); but is especially so for South Yorkshire. There were many differences between these hoards, however, so it is unlikely that all were related. One hoard discovered by a detectorist in 1980 led to a small excavation. The combined finds included 313 struck 3rd century barbarous radiates and minim copies, many hammered blanks of bronze, numerous sections of bronze rod and tubing, and copper-alloy strips, wire and sheeting. Some of these items were corroded together (Mattingly and Dolby 1982). Roman pottery was recovered from the same soil layer, forming over half of a calcite-gritted, Dales ware jar of early 3rd to mid-4th century date that was probably the hoard container. The finds suggested counterfeiting, and this was supported by the large number of die-linked coins, two groups alone accounting for more than half the total hoard. This locale would have commanded quite extensive views across the River Don gorge and floodplain towards the high rocky point of Levitt Hagg. If there was forgery taking place, it may have been useful to be in a place where anyone approaching could be seen from a long distance.
The counterfeiter’s hoard was probably hidden for pragmatic reasons, but this may not have been the case for some of the other Don Gorge hoards. Another hoard was found in 1978, in woodland on the north-eastern side of the former Dearne Railway cutting. Some coins were seen lying on the surface beside a large flat limestone slab, probably disturbed through animal burrowing, and the hoard was then traced under the stone. At least 1638 3rd century antoniniani were associated with a wheel-made, coarseware Romano-British greyware jar, though some coins were situated beneath the vessel (Manby and Burnett 1981). The stone slab was partly resting on a rock outcrop and the hoard was under the centre of the slab, in the angle between it and the natural earthfast stone. The base and part of the main body of the vessel apparently remained in situ; with its side touching the face of the underlying outcrop, and the underside of the slab. Unfortunately, the site was thoroughly disturbed when earth-moving equipment was used to move the slab. The find spot was on a steep, south-facing slope of the Don Gorge, opposite Conisbrough on the southern side of the river, overlooking where Kearsley Brook meets the River Don.
The most notable coin hoard from the Don Gorge, and from South Yorkshire as a whole, is undoubtedly the so-called Cadeby Hoard, found in Pot Ridings Wood in 1981. The hoard consisted of 103 silver denarii and 9 antoniniani dating to AD 194–251, within a globular ‘poppy’ or pedestal ceramic beaker that had ‘spalled’ in several places, perhaps from the action of freeze-thaw within the limestone fissure, and/or also because the vessel was already quite old and a curated item by the time the coins were deposited within it. There were also four silver bracelets. One pair of the bracelets was set with roughly cut cornelians, whilst the other pair consisted of so-called ‘snake’ bracelets (Buckland 1986: 41; Cool 2000a: 30). The beaker and the bracelets were deposited within a natural fissure in the limestone rock that had been capped with a limestone slab to form a small cist. The find spot was located on quite a steep south-east facing slope, overlooking the floodplain and the River Don.
The Cadeby Hoard was acquired by Doncaster Museum in 1982, but the hoard has never been fully described, and only one set of drawings of the carnelian bracelets has been published (Buckland 1986: 41). Despite minor differences, the silver bracelets with carnelians appear to be part of a pair, perhaps made by the same craftsperson in Lincoln or York between 250 and 280 AD. The carnelians might have come from modern Cornwall, or possibly further afield in India or Afghanistan. No detailed mineralogical sourcing has been undertaken. In an interesting postscript, a third silver and carnelian bracelet was identified on the antiquities market in 2010, and XRF analysis of the surface of the bracelet indicated that it was broadly similar in silver content to the original four bracelets (SWYOR-B51685). The ‘new’ bracelet has been ruled by a coroner to be part of the same hoard. Its provenance is unclear, but it could certainly have been made by the same artisan.
In contrast, the silver snake bracelets display differences in wear and decoration, and one may have possible graffito, or a symbol scratched on the inner surface (P. Robinson pers. comm.), though no detailed microscopic study has been carried out. It appears to have been snapped or broken in antiquity. The second snake bracelet is simpler in execution, and could be a copy of the other, made by a different individual. It is less worn and scratched on the inner surface than the first bracelet, but is extremely worn on its external surface, as if it was regularly rubbed or polished. The two bracelets thus had very different biographies. Cool (2000a) discussed Romano-British hoards containing snake jewellery, including mid-2nd century AD examples at Snettisham in Norfolk, Backworth in Northumberland and Castlethorpe in Buckinghamshire. The Lightwood hoard from Longton, Stoke-on-Trent probably has a similar mid to late 3rd century provenance to the Cadeby artefacts (Mattingley 1963; Mountford 1963). The Lightwood and Cadeby hoards were deposited at a time when such bracelets were uncommon, and Cool thought it unlikely that they were deposited for safekeeping, or as bullion or ‘scrap metal’ (Cool 2000a: 38).
The symbolism of snakes and snake bracelets include associations with Mercury, Asclepios a god of healing, Glycon the hunter god, and/or with mother goddesses (Cool 2000a: 34-5). It is possible the bracelets were used or worn by religious specialists in ceremonies; or had amuletic or apotropaic properties. This might account for the wear or rubbing on the one snake bracelet. If this was the social context for the Cadeby hoard, then its deposition on a steep slope overlooking a river gorge might well have had ritualised significance. Such evidence suggests that there was something out of the ordinary occurring at Edlington Wood and along the Don Gorge in terms of Romano-British social practice, and the natural landscape may well have been an important influencing factor. Although no doubt targeted by metal detectorists, only one Roman coin hoard has been recorded from the south side of the Don Gorge around Conisbrough and Warmsworth, and quarrying and agricultural activities have yielded such finds either. This distribution may thus genuinely reflect past depositional practices. The River Don may have been a social boundary during the Iron Age between the Corieltauvi to the south and east, and the Brigantes to the north. It is possible that some of these differences were carried through as late as the 3rd century AD within the civitas administrative system. More likely, however, there was something numinous about the south and south-east facing slopes of the Don Gorge on the northern side that attracted coin hoarding activity.
The prevalence of coin hoards from the 3rd century matches a well-known national pattern. In the past this has been taken as evidence for a supposed ‘troubled’ period in Roman Britain, with military movements, possible uprisings as well as political and economic instability caused by various usurpers such as Gallienus. More coin hoards might thus have been buried for safekeeping during the period. It is likely that this has been considerably over-stressed however (Bland et al. forthcoming; Walton 2012; Walton and Moorhead 2016). The debasement of the coinage and subsequent inflation, coupled with the problems of owning coinage with images of a discredited emperor or usurper, meant that some people would have been left with large numbers of coins with low value. These coins would have been difficult to dispose of through financial transactions, particularly in a more rural context; and using them as ritualised offerings might have been a pragmatic response.
Evidence for iron and copper working is scarce, especially for the Iron Age; but developer-funded fieldwork has provided some examples. At Gunhills, Armthorpe, a hearth with evidence for high temperatures produced a radiocarbon date of 166 BC–AD 132, and was associated with handmade, grog-tempered pottery; whilst an excavated oven produced an archaeomagnetic date of 95 BC–AD 80 (Richardson 2008: 9-10). Though the high temperatures hint at metallurgical processes, there was no slag or hammerscale directly associated with those features, although there was elsewhere. Iron working and copper-alloy working residues were recovered from the ring ditch or eavesdrip gully of a roundhouse at Carr Lodge Farm, near Balby Carr and Loversall Carr south-west of Doncaster (Stanley and Langley 2013: 14), along with much charcoal, and fragments of triangular ceramic crucibles believed to be of Iron Age form and date. The iron-working slag included a small, rounded, part-complete smithing hearth bottom formed in the base of a smithing hearth (Mortimer 2013: 43-4). ‘Iron concretion’ was also recovered – possible bog iron? Several crucible fragments had traces of copper-alloy on them and fired clay and burnt stone was also found. At Engine Lane, Shafton, tap slag was associated with 1st century AD ceramics, and thus either late Iron Age or early Romano-British (Burgess 2003). Carr Lodge Farm and Engine Lane Shafton are unpublished. At Parrots Corner, Rossington Bridge, a roundhouse ring gully containing 1st century BC pottery also yielded a ceramic crucible fragment with copper-alloy residues, remains of two tuyères and iron hammerscale (Bishop 2010: 11).
At Deepwell Mews, Halfway, Sheffield, several pits produced hammerscale, charcoal, and Iron Age-style pottery (Wells 2018: 6). A lump of bog iron ore was also found in one pit, and early Romano-British pottery of 1st to 2nd century date in a few of these features too. There appears to have been continuity of occupation and ironworking at the site, and its location on quite a steep east-facing slope above the River Rother suggests the use of the updraught effect to achieve higher temperatures. At Billingley Drive, Thurnscoe, vesicular ferrous fragments of so-called ‘Iron Age Grey’ slags were identified, these almost always associated with late Iron Age sites (Cowgill 2004b: 54). The ditch section they were from was dated to the 2nd century AD by a single pottery sherd, so either indicates earlier Iron Age activity, or the persistence of metal working techniques into the Roman period. Iron block and tap slags at Armthorpe were predominantly from smelting (Cowgill 2008: 48), along with vitrified clay from furnace structures, and smithing slag and hammerscale. The two different types of smelting slag are rarely found together, with block slags usually considered to be mid to late Iron Age in date; and this may indicate a transitional technology. A clay crucible was found in a ditch on another part of the West Moor Park Armthorpe site, along with smithing slag, cinder and fuel ash slag (Hughes 2006: 13).
Iron-working evidence such as slag or vitrified hearth lining has also been found at Smarson Hill Plantation, South Anston (Radley and Plant 1969); Doncaster Park and Ride, Scawthorpe (Bishop 2005); and Ashfields, Stainforth (Dransfield and Harvey 2012: 5, 10), though all these examples may be of earlier Romano-British date. At Rossington Grange Farm, a ditch contained a spill of leaded tin bronze with a high tin content and some silver, possibly an alloy of Iron Age rather than Roman date (McDonnell 2016: 27). The ditch also contained Roman pottery of mid to late 3rd century date, however, and although it is unclear if these ceramics were from a tertiary fill (Roberts and Weston 2016: 17), it seems more likely that the alloy was of Roman date too. Further evidence for more specialist metalworking was found on a clay crucible excavated in a 2nd century pit at St Sepulchre Gate in Doncaster, which had traces of silver (Bailey 1986: 196, fig. 43); and litharge from another Doncaster pit of similar date at Frenchgate, this being a waste product associated with the recovery of silver from lead, perhaps from coins (Tylecote 1986: 196). A probable smithy, clay crucibles and iron slag were found within a walled area of the vicus at Templeborough (May 1922: 56-9), along with cisterns perhaps used for quenching.
There is more evidence of metalworking following the Roman occupation, though outside of military sites the ways in which this was undertaken and organised are still unclear. Iron objects became more common on small-scale rural sites, especially nails and metal fittings of various sorts; but so too does evidence for metalworking itself. Smelting as well as smithing was suggested at Manor Farm, Bessacarr (MAP 2014). Iron smelting tap slag and hearth bottom smithing slag was both found at Rossington Grange Farm (McDonnell 2016: 26). Smithing slags, ferrous slag bottoms, furnace lining and hammerscale were recovered from Billingley Drive, Thurnscoe and Ashfields, Stainforth (Cowgill 2004b; Wessex Archaeology 2012). At Holme Hall Quarry, the iron slag found indicated smithing rather than smelting, predominantly during the 3rd century AD (MacKenzie 2007: 73-4); and parts of hearth bottom slags from small-scale smithing were also found at Pastures Road, Mexborough (Jones 2008). At Whirlow Hall Farm, Roman lead and pewter working was indicated by lead waste and portable XRF geochemical analyses (Doonan 2017; Doonan and Slater 2017). A fragment of ferrous smithing hearth bottom was also recovered from the enclosure ditch (Lortie and Doonan 2017: 89).
Several other aspects of metalworking are noteworthy. Near the Romano-British pottery kilns excavated at Cantley an iron smelting furnace or bloomery was also excavated, and this partly stone-built structure contained iron slag, ash, charcoal, and the lower part of a rotary ‘beehive’ quern with ferrous staining on its flat surface was built into the walling (Cregeen 1956: 35-9, figs. 3-4, plates i-ii). The feature was backfilled with dark soil that included a pig scapula and a bronze plate, and there were some lumps of clay and fired clay above it; but no dateable artefacts were recovered. It was proposed that the raw material was in part composed of bog iron (ibid.: 41). The furnace’s proximity to the pottery kilns may suggest shared ‘high temperature’ practices, possibly with some of the same people involved. The re-use of a quernstone for metalworking at Cantley is paralleled by examples elsewhere with heat reddening and/or ferruginous deposits indicating their possible re-use as anvils; as at Balby Carr and Roebuck Hill, Jump (Heslop 2005; Wright 2007: 54), Ferrybridge in West Yorkshire (Heslop and Gaunt 2005: 149), and Gamston and Ramsdale in Nottinghamshire (Garton, Southgate and Leary 2000: 40; Knight 1992: 72). This might indicate pragmatic re-use of hard, flat stone surfaces but could also suggest symbolic links between metalworking and agricultural production.
During the Iron Age, iron objects could have been produced from ironstone sources in the Cleveland Hills and local Coal Measures, or as bog iron, found as iron pan in river valleys and the Humberhead Levels (q.v. Crew 1991; Halkon 1997, 1999; Halkon and Millett 2000, 2003). The enclosure at Whirlow Hall Farm was close to a local iron source in the Limb Valley (Doonan 2017: 106). Bog iron was found at Deepwell Mews, and an iron concretion from Carr Lodge Farm might also be derived from bog iron. The copper ore source for copper-alloy objects found in South Yorkshire is unknown, but the nearest are Alderley Edge in Cheshire or Ecton Hill in north Staffordshire (P. Buckland pers. comm.). Although basic smithing probably took place at many settlements, just a few highly skilled individuals or households may have produced high-status iron and copper-alloy objects. Those most skilled at metalworking could have held considerable power, but perhaps ambiguous social status. One study of copper-alloy working found that from the 4th century BC through to the early 1st century AD alloys in northern Britain were mainly composed of tin bronze in consistent proportions suggesting smiths followed a ‘recipe’ (Dungworth 1996: 404). Arsenic was a frequent impurity, indicating smelting at relatively low temperatures. From the early 1st century AD, copper alloys had varying levels of zinc incorporated within them, this probably derived from imported Roman brass.
Knowledge of iron and copper working might have been restricted during the Iron Age, surrounded by rites and proscriptions. These practices and the resulting materials and objects themselves could have been entangled with symbolism and metaphors associated with fertility and reproduction (Aldhouse-Green 2002: 16; Budd and Taylor 1995: 139; Giles 2007c: 398-399; Hingley 1997a: 12; 2006: 217-8). Evidence of depositional practices from Roman Britain (see below) suggests that some such meanings survived beyond the Roman conquest. The Roman occupation and the presence of additional smiths associated with the military may have disrupted existing indigenous metalworking social and economic links though, whilst Roman supply networks facilitated the import of ores from a much wider geographic area, though the extent of this is still unclear. Tin from Cornwall, for example, would have been present at Whirlow Hall Farm for pewter production (Doonan 2017).
When was block slag technology replaced by tapped shaft furnaces (q.v. Cowgill 2008: 50), and was this later in South Yorkshire than other regions?
What was the extent and nature of the use of bog iron? Are there any traces of ore extraction surviving, or has it all been destroyed by medieval and post-medieval surface workings?
In particular those artefacts recorded under the Portable Antiquities Scheme represent have significant research potential. How well are pre-PAS metal-detecting finds recorded on the HER? Many in museum collections may not be documented;
Are there any spatial overlaps, and is there any evidence that these might have reflected pre-conquest contacts?
Are there any correspondences in landscape associations for deposition? What roles did watery places, rock outcrops and other ‘natural’ locations perform regarding structured deposition, did some of these reflect ritualised practice, and how might these have changed over time?
Is any further investigation at the site possible, involving geophysics and/or targeted excavation?
Were patterns of late Roman artefact and coin deposition different to earlier centuries?
Did metalwork play a role in this, and can any possible social groupings be identified? Could these indicate territoriality? Can we identify any status differences between Iron Age and Romano-British individuals and settlements? How were such variations manifested? Is there any evidence for such groupings persisting into the Romano-British period?
Later Bronze Age or earlier Iron Age pottery has been recovered from Finningley (Manby 2010), and a few mostly unstratified sherds from Sutton Common (Cumberpatch, Vince and Knight 2007). For many years, the only later Iron Age pottery known from South Yorkshire was from Pickburn Leys (Sydes 1993: 39-41; Sydes and Symonds 1985), but this situation has changed with the increased volume of developer-funded fieldwork and improved on-site sampling. Much of the Iron Age pottery from South Yorkshire was coarse, poorly fired and fragile, and where organic or shell tempers were used these have often leached out leaving voids. Many sherds thus do not last long in ploughsoil and might not survive even in stratified contexts or be readily recognised during excavation (Cumberpatch and Robbins n.d.; Cumberpatch and Webster 1998). Nonetheless, pottery still seems to have been relatively scarce compared to some other regions. Decoration was rare or absent.
East Midlands Scored Ware was found at Redhouse Farm, Adwick-le-Street (Cumberpatch 2004b). These vessels had surfaces brushed with twigs or scored with vertical or curving lines, with more regular or comb decoration in later vessels (Elsdon 1992: 84; Knight 2002: 133-134). This tradition originated in the late 5th or earlier 4th centuries BC around the Nene, Welland, lower Trent and Ouse valleys, extending north to Staffordshire, Derbyshire and South Yorkshire, east to Lincolnshire, and south to Leicestershire and Hertfordshire. Lug-handled vessels have possibly been found at Topham Farm, Sykehouse (Cumberpatch 2003: 19). Also significant were Iron Age Shell Tempered Wares, usually hand-made, and possibly derived from sources in Lincolnshire or around the Humber estuary, but also possible sources in the Trent Valley. It has been found at Topham Farm, Sykehouse; Enclosure E1 at Redhouse Farm, Adwick-le-Street; at Balby Carr; Pastures Road, Mexborough; Marr; Goldthorpe; possibly Pickburn Leys and Ashfields, Stainforth; and at Rossington Grange and Rossington Bridge (Cumberpatch 1985, 2003, 2004b, 2005, 2006, 2008a, 2008c; Cumberpatch and Leary 2014; Dransfield and Harvey 2012; Leary 2010a; Rowlandson, Monteil and Hartley 2016). This pottery is fragile and prone to fragmentation. Its dating is problematic, and as with some Scored Ware, Shell Tempered Ware in late Iron Age forms continued in use into the 1st and 2nd centuries AD (Evans, Wild and Willis 2005: 135).
Hand-made Iron Age pottery, including some with a distinctive ‘soapy’ texture, has been recovered from sites such as Topham Farm, Sykehouse; Pastures Road, Mexborough; Balby Carr; and Nutwell Lane and Gunhills Armthorpe. These were produced using quartz, sand or sandstone tempers, and the former may have been manufactured locally (Cumberpatch 2003, 2008a, 2008b, 2016; Cumberpatch and Webster 1998: 21; Leary, Evans et al. 2008: 28). Other locally made vessels that probably extended in date from the late Iron Age into the early Romano-British period include grog-tempered wares from Armthorpe and Rossington Bridge (Buckland, Hartley and Rigby 2001: 79; Cumberpatch 2001; Evans 2001a; Leary, Evans et al. 2008). Grog was derived from older, broken-up vessels, and if these were associated with specific individuals and/or events this may have reinforced familial and symbolic links between old and the new (q.v. Hill 2002: 152; Woodward 2002: 109).
Sherds from possible briquetage or coarse ceramic salt containers have been identified at Topham Farm Sykehouse (Cumberpatch and Roberts 2003: 24); perhaps from coastal salterns in Lincolnshire. Fired clay loomweights are often a feature of Iron Age sites in other regions, but few have been identified in South Yorkshire. Fragments of some were excavated at Jump, however, from a pit also containing burnt bone, coarse vesicular late Iron Age or early Romano-British pottery, a burnt quern fragment, and charcoal (Robinson and Johnson 2007: 8-9). Possible late Iron Age or early Romano-British clay loomweights have been identified at Rossington Bridge (Buckland, Hartley and Rigby 2001: 34, fig. 32), and at Hatfield Lane Edenthorpe (Rowlandson 2015a: 31-32). A triangular clay loomweight was found at Church Walk, Doncaster (Cool 2008b: 139, fig. 40.11).
Most hand-made ceramics were probably produced at a domestic scale. The small numbers of pots produced by individuals might have had associations with those who had made them (Hill 2002: 153; Willis 1999: 90), especially where pots were physically marked by the fingertips and nails of their makers (q.v. Giles 2007b: 242). Most Iron Age pottery vessels found in South Yorkshire have been ‘closed’ forms such as barrel jars and open jars, although a few bowls have been identified (e.g. Cumberpatch 2003: 23, 2016: 10-11). This implies that ceramics were utilised primarily for the preparation and storage of food rather than its serving and consumption, for which wood and leather vessels and basketry may have been employed instead. A few jars were large vessels that would have been difficult to transport even when empty (ibid. 2003: 19). Together with its scarcity and restricted patterns of deposition, this all suggests that pottery was not a primary medium of everyday food production and consumption during the later Iron Age. Many households may have had a maximum of one or two ceramic vessels at any time.
Some Iron Age vessel forms were produced well into the 1st and 2nd centuries AD (Cumberpatch and Robbins n.d.; Darling 1995; Leary, Evans et al. 2008: 28). At many sites coarse or vesicular pottery may be Iron Age, or early Romano-British ceramics made in an Iron Age tradition. Such wares have been excavated at Church Field, Rossington (Atkinson 1998), Far Field Road, Edenthorpe (Darling 2004); Holme Hall Quarry, Stainton (Leary, Ward and Vince 2007), Windhill Plantation, Norton (Burgess 2001c), Barnburgh Hall (Evans and Ward 2005), Roebuck Hill, Jump (Didsbury 2007), Stainforth Marina (Rowlandson 2014); and at Deepwell Mews, Halfway and Whirlow Hall Farm, Sheffield (Vyner 2018; Waddington 2012).
The few thin-section studies that have taken place (e.g. Vince 2007) have tended to be site or project specific, limiting their usefulness for wider comparative purposes. Some vessels or clays might have been locally produced, others coming from north Nottinghamshire, Lincolnshire and the Humber area, but there are also parallels with ceramics in eastern Yorkshire (Cumberpatch 2008, 2016). A detailed comparative study of prehistoric ceramic forms and fabrics from across the region is highly desirable, but this must also be a comprehensive contextual-based study.
The Romano-British pottery industries of South Yorkshire are relatively well-known and published, but there are still many questions remaining concerning ceramic production, consumption, and deposition.
Some of the earliest Roman-style ceramics in South Yorkshire were probably produced by military pottery workshops or figlinae. These might have included some early mid to late 1st century vessels and wasters excavated at Templeborough (May 1922 235-7; Swan 2002: 35). Roof tiles with the stamp of the auxiliary unit cohors IV Gallorum but also the Legio IX Hispania were excavated at Templebrough (May 1922: 122-3, plate xxxvii), along with box flue tiles (tubuli) and ceramic roof antefixes. These were probably produced in Templeborough during the late 1st to early 2nd century AD (Stephens 1986: 17), a practice widely adopted during the reign of Trajan (AD 98–117). This is similar to the early military figlina excavated at Grimescar Wood near Huddersfield that supplied ceramic tiles and pottery vessels including bowls, cooking jars, flagons and mortaria to the forts at Slack and Castleford during the later 1st century AD (Betts 1998; Purdy and Manby 1973). There is an apparent temporal gap between the conquest of the north and these early ceramics products, and the wider adoption of Roman-style pottery, which does not really seem to have occurred in South Yorkshire until AD 120–130. It is unclear if there were as yet unlocated kilns producing pottery during this period, but the products were not widely distributed locally; as yet unexcavated assemblages (Daniel 2017: 29), or if there was a lack of a local market, perhaps caused by initial resistance to Romanised material culture.
The area around Doncaster saw a large number of pottery kilns established from the 2nd century AD onwards, many of which have been excavated and published (e.g. Annable 1956; Buckland and Dolby 1980; Buckland, Hartley and Rigby 2001; Buckland and Magilton 2005; Buckland, Magilton and Dolby 1980; Cregeen 1957; Gilmour 1954, 1955, 1956). Grey wares were produced at Cantley and Blaxton, with a form of Black Burnished ware and so-called ‘Parisian ware’ also being made at Rossington Bridge. Some of the Cantley kilns and pottery from them excavated by J.R. Lidster and J. Pallister in the 1950s and early 1960s have still not been fully published, however (Buckland and Magilton unpublished). Some of the vessels seem to have been transported directly up north to military garrisons along Hadrian’s Wall at Vindolanda and South Shields (Swan 2002: 59). This military demand seems to have declined by the later 2nd century however (Buckland, Magilton and Dolby 1980: 163). It is unclear if this reflects contracts lost to other producers, or if military preferences simply changed. Thereafter until the 4th century, much of the South Yorkshire pottery production was probably focused principally on local consumers.
It is still unclear how this industry was organised. A notable and intriguing figure is the potter called Sarrius who was producing mortaria in the Mancetter-Hartshill potteries in Warwickshire, but who then seems to have established subsidiary workshops including one at Rossington Bridge (Buckland 1986: 45; Buckland, Hartley and Rigby 2001: 45-7, 86), probably during AD 135–165/170. These may have been associated with two men called Secundua and Setibocius. It is unclear whether Sarrius would have employed freedmen or slaves, or both. The South Yorkshire potteries may have been located to take advantage of the presence of the road to Castleford, York and the north; but also the Rivers Torne and Don. Administrative and tax reasons might also have played a part (Buckland, Hartley and Rigby 2001: 86). Frustratingly, very few associated or ancillary features associated with the Cantley, Blaxton and Rossington Bridge kilns survived and/or were recorded during the 1950s–1960s excavations, so the spatial organisation of the production centres is unclear. No specialist structures or centralised workshops appear to have been recorded. More recent investigations, however, have provided some clues.
At St Wilfrid’s Road, Cantley, a late 2nd to early 3rd century ditched enclosure was associated with significant quantities of pottery including a large dump of kiln wasters, but also ‘domestic’ features such as small ovens and possible traces of buildings (Daley 2007: 11-14, fig. 4). At Rossington Grange Farm, the south-eastern part of Enclosure 6 contained remains of three pottery kilns (Roberts and Weston 2016: 16-17, fig. 8, plate 3), whilst a ditch to the north-east contained another large dump of ceramics, including warped waster vessels but also pottery from domestic use (Rowlandson, Monteil and Hartley 2016: 22). There was also evidence of crop processing. Warped and cracked vessels deposited in pits and ditches of enclosures at Hatfield Lane, Edenthorpe suggested the presence of one or more pottery kilns nearby (Rowlandson 2015a: 25), but crop processing was also taking place there (Weston and Roberts 2015: 47).
These examples suggest that for some at least, pottery production was not rigidly centrally organised, and the potters or their families were also farmers. Pottery production might have been undertaken in ‘slack periods’ during the agricultural year, and thus perhaps on a seasonal basis. Potting was probably not a high-status or lucrative occupation, and so incomes would either have had to have been supplemented by agricultural produce; or these formed the main basis of subsistence with ceramics merely an additional, even occasional source of income. To the east of Cantley a group of three to four unusual small oval and sub-oval enclosures were photographed by Derrick Riley (SLAP Riley 51 SE627025), but to this author’s knowledge these have never been investigated further. The purpose of the enclosures is unclear and their date is still unknown, but it is possible that they might have been associated in some way with pottery production.
There are additional gaps in knowledge regarding the South Yorkshire pottery industries. There are similarities between the ‘native-style’ jars found at sites such as Topham Farm, Sykehouse and a type of pottery (GTA 17 ware, usually in bead-rim jars) found at Doncaster, which may be further indication of local manufacture continued by indigenous potters (Leary, Williams et al. 2008: 67). Typologically similar vessels are also found in Nottinghamshire, north Lincolnshire and Humberside where they are most commonly found in post-conquest groups, but there seem to have been differences in the fabrics used, which thin-section analyses might clarify (ibid.: 51). The products from Blaxton, Cantley, and Rossington Bridge were produced over a lengthy period from the 2nd to 4th century and were stylistically conservative with comparatively little variation in forms over time. This creates problems not only in understanding any changes in production and supply over time, but also in dating features using such pottery. Detailed characteristics of individual types are chronological significant (ibid.), and this indicates why it can be important that even grey ware assemblages should be fully published and illustrated.
Some artefacts associated with pottery production have been excavated – stone burnishers at Rossington Bridge, including a re-used Neolithic polished stone axe; and a flat rotary quern modified as a possible potter’s kick wheel (Lloyd Morgan 2001: 27, 28, figs 25, 30). Other stones and re-used querns might have been the pivots for potters’ wheels, and one quern fragment may have been used for crushing red ochre pigment. A ceramic brush handle with holes for bristles and a stylised swan motif was found at Cantley kiln 33 (Buckland, Magilton and Dolby 1980: 152, fig. 5).
In terms of ceramic consumption, outside of Doncaster and military sites cooking pots and storage jars generally predominated in most South Yorkshire Romano-British assemblages, with smaller numbers of bowls. Tablewares such as platters, dishes, flagons, flasks, cups, tankards, beakers and other ceramic vessels used for the presentation and consumption of food and drink were usually rarer. This might imply a resistance to ‘Romanised’ practices of eating and drinking, but it is also possible that wooden vessels served this purpose on many rural settlements. Some ceramic colanders or cheese presses are also known (e.g. Rowlandson 2015b: 9).
Imported continental wares found in Doncaster included samian from southern and central Gaul, amphorae from Spain, colour-coated wares from the Rhineland, and a few mortaria from northern Gaul (Buckland 1986; Dickinson 1986; Hartley 1986; Leary, Williams et al. 2009). These types were relatively rare outside Doncaster, however, and samian ware was scarce on most Roman-period sites. At Armthorpe part of a samian dish dated to AD 70–100 was a rare early form (Leary, Evans et al. 2008: 42), and at Topham Farm, Sykehouse a roundhouse was associated with two samian sherds dating to AD 100–130 (Cumberpatch, Leary and Willis 2003: 22, nos 10, 14a). Overall however, much of the samian ware at rural sites such as Armthorpe and Holme Hall Farm was produced after the middle of the 2nd century AD. This was also the case for the unusual deposit of two near complete samian bowls in a pit at Hatfield Lane, Edenthorpe (Rowlandson and Monteil 2015: 25). Even the relatively ‘Romanised’ settlement at Holme Hall Quarry, Stainton produced just 45 sherds of samian from 27 vessels (Leary, Ward and Vince 2007: 47). Outside Doncaster and within or close to military sites such as Templeborough and Rossington Bridge (Leary 2010b), samian generally only occurred on a few rural settlements as small numbers of worn sherds; as at Whirlow Hall Farm, Topham Farm, Sykehouse; Billingley Drive, Thurnscoe; Armthorpe, and Rossington Grange Farm (Beswick 2017c: 60-1; Cumberpatch, Leary and Willis 2004: 22, 24; Didsbury, Dickinson and Hartley 2004: 34; Leary, Evans and Hartley 2008: 28-31; Rowlandson, Monteil and Hartley 2016: 22). This is similar to Romano-British rural settlements in West Yorkshire. Either people on rural settlements did not desire samian and other imported vessel types, or they were outside the networks of its distribution and sale or exchange.
The context in which samian was supposedly used across the Roman Empire, mostly as tableware for the presentation and consumption of food and drink, was not always how it was utilised in provincial households. There is evidence at some sites in Britain such as Scrooby Top in Nottinghamshire (Robbins 2000) of samian being sooted or burnt from cooking fires, showing that it was sometimes used to prepare food; and several such examples were noted at Church Walk Doncaster (Leary, Williams et al. 2008: 65). This indicates how Roman material culture could be reinterpreted in local ways. Residue studies of mortaria in Britain have also demonstrated much variety in their uses and that in addition to processing animal fats, they were also utilised on occasion for spice or plant preparation, but were generally not simply new forms of vessels put to existing Iron Age-style food preparation uses (Cramp, Evershed and Eckardt 2011: 1347-9). At Stanwick in North Yorkshire, however, mortaria do seem to have been used for milk or cheese processing, an adaption to localised traditions.
There may have been early military production of mortaria at Templeborough, as May (1922: 112) records finding wasters that were probably from nearby kilns. South Yorkshire white slipped ware, Mancetter-Hartshill, Swanpool and Verulamium-type mortaria have all been found at Doncaster (Buckland and Magilton 1986; Hartley 1986; Leary, Williams et al. 2008); but have also been recovered as small numbers of sherds from some domestic sites including Whirlow Hall Farm (Beswick 2017c: 57), Topham Farm, Sykehouse (Cumberpatch, Leary and Willis 2004: 24), Billingley Drive, Thurnscoe (Didsbury, Dickinson and Hartley 2004: 35-6), FARRRS (Rowlandson 2015b: 9-10), St Wilfrid’s Road, Cantley (Leary 2007), Holme Hall Quarry (Leary, Ward and Vince 2007: 28), Rossington Grange Farm (Rowlandson, Monteil and Hartley: 22-3) and probably from the bathhouse at Hazel Lane Quarry, Hampole (Bevan 2006b: 27), although the latter pottery assemblage remains unpublished. In the 3rd and 4th century black iron slag was crushed and used as trituration grits in grey ware mortaria (Cumberpatch, Leary and Willis 2004: 22). Like the iron smelting kiln at Cantley, this may again imply interesting economic and social connections between metalworking and pottery production.
Roman ceramic oil lamps or lucernae were rare in South Yorkshire. A partial example was recorded at Templeborough (May 1922: 117, plate xxxvii). One allegedly found at High Street, Doncaster before the First World War (Bailey 1986: 114, fig. 26) may be a relatively modern ‘import’ from Italy or North Africa rather than from an archaeological deposit. Four complete examples with makers’ stamps of late 1st to early 2nd century AD date were deposited in the cremation cemetery at Doncaster Waterdale; and these originated in East Gaul and Kӧln (Griffiths 2013: 85-6). A more fragmented example from the same context may have been made in Italy. Although Iron Age-style roundhouses might have relied on central hearths for internal lighting, if oil lamps were rare then how Roman-period rectangular buildings were lit inside and experienced by the occupants must so far remain unknown.
Apart from the early kiln at Templeborough, the nature and extent of Roman tile and brick production and consumption in South Yorkshire is also almost unknown. At Doncaster, the Roman brick and tile assemblage from the 1960s and 1970s rescue excavations has not been published, though the excavation report notes its occurrence in some features (e.g. Buckland and Magilton 1986: 52). At Church Walk, Roman tile tegulae and imbrices, box-flue tiles or tubuli, and brick bessales, pedales and bipedales were all identified (Tibbles and Tibbles 2008: 109-113), some with traces of mortar and opus signinum adhering to them; though much of the assemblage was residual in medieval contexts. The presence of several wasters and low-quality ‘seconds’ suggested at least some localised manufacture.
In line with the lack of evidence for Roman-style rectangular buildings, however, outside of Doncaster and Templeborough few sites have produced tiles. Hypocaust pilae and roof tiles were recovered from Hazel Lane Quarry, both in association with the possible bathhouse but also in some outlying features (e.g. Pine 2002; Pine and Taylor 2003, 2006). Occasional finds on rural settlements hint at tiled roofs on ‘Romanised’-style buildings located nearby. That no definitive remains of such buildings have been identified to date might mean that these were comprehensively robbed after abandonment, and the materials re-used. Eight fragments of tegula and imbrex roof tiles were found in a ditch and in two waterholes at Rossington Inland Port (Daniel, Harrison and Powell 2014b: 14), which along with a fragment of Roman window glass might hint at such a building somewhere in the area; and five tegulae fragments came from Rossington Grange Farm (Mills 2015: 60). Alternatively, these examples could simply represent materials re-used for other purposes.
Part of a moulded terracotta clay figurine was found at Doncaster High Street (Jenkins 1986: 112-3) and was probably from Cologne and of early 2nd century date. No other such ‘pipe-clay’ figurines are known from South Yorkshire.
Pottery production at the Rossington Bridge kilns and perhaps from as yet unexcavated kilns at Blaxton continued into the 4th century, but their products were increasingly only for local consumption. Across northern Britain, Dales ware and ceramics from York, Crambeck, and Huntcliff became dominant (Buckland, Magilton and Dolby 1980: 146-7; Swan 2002: 72-3). There was a general decline in tablewares. Much more work is still needed on how these macro-level artefactual and economic changes occurred across northern Britain, but also their impact at a local level in South Yorkshire. The end point of Romano-British ceramic use in South Yorkshire is difficult to ascertain. There are known problems with dating late Roman artefacts, including the fact that some metal objects and ceramic vessels appear to have been made in deliberately archaic styles harking back to the past (Cool 2000b; Fitzpatrick-Matthews and Fleming 2016; Gerrard 2016), so these might easily be attributed to earlier periods. The late Roman or post-Roman radiocarbon dates from two corn-drying kilns and a field system ditch at Goldthorpe Industrial Estate (Ross 2014: 15) indicate that great caution must be taken in future with late Roman and ‘post-Roman’ archaeological evidence. Independent scientific dates must be sought in such circumstances.
The predominantly aceramic nature of South Yorkshire during the Iron Age should be considered in its own right, not simply as an anomaly. The use and production of pottery in South Yorkshire was limited in both the Iron Age but also in the post-Roman/early medieval period, despite the major pottery industries that existed during the Roman period. What can this ‘cross-period’ phenomenon and analysis tell us about settlement, population and identity across the first millennium BC and the 1st millennium AD?
Do we need to reassess definitions of vessel and fabric types?
Does pottery represent the occasional movement of goods within ceramic containers, or is it evidence of the movement of people (either permanently or temporarily) from areas with radically different traditions of material culture use (Cumberpatch 2016: 13)?
Although pottery styles may be indicators of identity, South Yorkshire had varied sources of ceramic production and exchange during the Iron Age, including sources outside of the region. How does this equate with classic (cliched?) culture-history notions of the Brigantes as an identifiable tribal entity? Might this indicate a more complex situation? Can any ‘boundary zones’ be identified between the ceramic areas in wider Yorkshire and the aceramic areas of South Yorkshire, the Peak District and the Cheshire Plain?
The limited evidence suggests possible seasonal production on otherwise ‘domestic’ settlement sites. Can additional pottery kilns be identified though aerial imagery and/or geophysical survey, along with any associated features? How were they associated with the landscape of enclosures and field systems?
Are they Romano-British in date and associated with pottery production, or something completely different?
The development area at St Wilfrid’s Road, Cantley (Daley 2007) was meant to incorporate the locations of at least two known Romano-British pottery kilns, but no traces of them were found (Daley 2007: 17). Were such features misidentified during earlier investigations, had they already been destroyed by modern disturbance, or was there a problem with accurately establishing their original location? Can GIS help with this in future?
Can scientific dating independent means of assessing this?
Can any differences in food preparation and consumption be identified prior to or following the Roman conquest of the north? How was the earliest Roman pottery being used away from military sites? Was it utilised for food preparation and consumption, or perhaps just for cultural value? Did the uses of forms such as samian and mortaria always match Roman-style practices, or was there local diversion, subversion, or innovation?
Many later Iron Age and Romano-British beehive and flat quernstones in South Yorkshire were manufactured from Millstone Grit outcropping at Wharncliffe Crags near Deepcar (Challis and Harding 1975: 23-25; Wright 1988: 74, 2007: 24); and probably also from other outcrops along the Rivelin Valley. Other sources of querns found in South Yorkshire include Coal Measures sandstones (Cruse and Gaunt 2016: 24), and further afield, Woolley Edge Rock to the south of Wakefield or in the Normanton-Hopetown area (Heslop and Gaunt 2008: 19). The earlier Iron Age saddle quern fragments recovered at Sutton Common were of sandstone, possibly from drift deposits (Watts 2007: 145).
Many querns were distributed widely across the study region, probably as roughouts to be finished elsewhere (Wright 1988: 74-75). The production site at Wharncliffe was surveyed and partially excavated in 1950-1960 although this work remains unpublished, but part of the quern manufacturing site was surveyed in more detail in 1999. Over 2300 quern roughouts were identified, of which 272 were beehive forms, and 1960 were flat disc querns (Pearson and Oswald 2005). This indicates a preponderance of Roman-period and later forms. These different types had varying distributions, with flat disc ‘blanks’ occurring across the site, but the beehive roughouts located mostly along the eastern margins, again perhaps reflecting chronological trends in quern working.
Across the wider Yorkshire region, and in South Yorkshire, older beehive forms persisted in use well into the 3rd century AD. Even saddle querns occur in some Romano-British sites, as at Rossington Grange Farm for example (Cruse and Gaunt 2015a: 72). Whilst many ‘native’ sites would have carried on using beehive querns, some beehive querns have been found in Roman military contexts in northern England, including Templeborough (e.g. May 1922: plate xliii). A major study of Yorkshire querns was underway by Donald Spratt but following his death in 1992 it is not clear if this will ever be fully published, though his work forms part of the Yorkshire Quern Survey archive. Heslop published a study of 562 beehive querns from North Yorkshire and County Durham (2008a), but such work needs to be expanded to include South and West Yorkshire.
The organisation and nature of quern production is unknown. Specific groups may have used larger quernstone ‘quarries’, producing querns seasonally when not engaged in agriculture or other subsistence tasks; or there may have been specialist communities or individuals concentrating on stone working. Manufactured querns could then be traded with other communities to obtain agricultural produce, commodities such as salt, or items of material culture. Alternatively, although certain social groups might have controlled access to quern working sites, others may have had rights to work stone in them (q.v. Ballard 1996; Sundstrom 1996). Gaining access to quern working sites could have been achieved through ‘payments’ to the controlling group. Production required technical skill but might also have required rites and propitiations to ensure the co-operation of the stone and the future efficacy of the querns. Any changes following the Roman conquest are also unknown.
Quernstones and quern fragments have been found in and around many of the South Yorkshire rural settlement enclosure sites, too many to list here, although not usually in large groups. Those beehive querns recovered from Roebuck Hill, Jump are unusual in being stratified and associated with Iron Age pottery (Wright 2007: 24). Querns were often treated in quite specific ways prior to deposition (Cruse 2015; Cruse and Gaunt 2015a; Heslop 2008a; see below), many being heavily fragmented and beehive querns having had their outer surfaces struck off; and they often formed part of more ritualised ‘placed’ deposits (see below). A beehive quern base stone found at Rossington Grange Farm had a 120mm diameter and 25mm deep circular depression deliberately cut into the otherwise flat grinding surface, but which showed no obvious signs of functional use (Cruse and Gaunt 2015a: 73, fig. 23). In North Yorkshire a small number of beehive querns possessed the same modification (Heslop 2008a: 66-7), which Heslop interpreted as a possible stoup for holding or receiving offerings. Alternatively, this also represented use or re-use as a mortar; or was even a form of quern decommissioning. A rotary flat disc quern found at FARRRS had a pecked circular groove as decoration (Cruse 2015: 13).
Few millstones have been recovered in South Yorkshire – a photograph shows several examples including one inscribed with a harp pattern from Templeborough (May 1922: 124, plate xliii), though these were not described in detail. This indicates that whatever the agricultural regimes practiced by communities the processing of cereals was still important to their diets, but for most non-military people this might have largely remained at the household level of production, even following the Roman conquest. This also suggests that there were few Roman-style animal or water-powered mills in South Yorkshire outside of army contexts.
In the early Romano-British period, flat basalt lava quernstones were imported from the Niedermendig quarries in the Mayen region of Germany and were initially associated with the Roman military (Buckland 1986; Crawford and Röder 1955). In south-eastern England they became part of civilian trade, especially in areas with no suitable local stone for quern production, but in the north their distribution was more restricted. They might have been imported with colour-coated wares from the Rhineland as ballast for cargoes (Buckland 1986: 22). They have been found in fort and civilian contexts at Doncaster and Templeborough (Buckland 1986: 22; Buckland and Magilton 1986: 49 100-1; Cool 2008b: 140; Heslop 2008b: 290; Wright 2016: 73). There were fragments at the pottery production sites at Rossington Bridge (Buckland, Hartley and Rigby 2001: 27, 32) and St Wilfred’s Road, Cantley (Vince 2007b), perhaps originally from the fortress at Rossington. One lava quern fragment was re-worked into a rubber stone used at Billingley Drive, Thurnscoe (Wright 2004: 57-8, fig. 28 101). The import of querns from outside the region may have disrupted and undermined traditional stone-working practices and exchange, and the social and symbolic ‘meanings’ of querns might have changed for some indigenous people following the conquest. Individuals or communities moving into the region might not have shared such ideas. Flat Romano-British hand querns do not seem to have been treated in quite the same way as beehive querns, which might also imply differences.
One notable Romano-British find was an unusual complete but fragmented doublehopper hand quern of Millstone Grit excavated from a pit at Hatfield Lane, Edenthorpe (Cruse and Gaunt 2015b: 33); which has parallels with less-complete examples known from Doncaster (Buckland and Magilton 1986: 100, fig. 23.4), Templeborough (May 1922: 124, plate xlii), and Castleford (Buckley and Major 1998: 245, SF806). This was a post-conquest form whose main period of use was during AD 75–250 (Cruse and Gaunt 2015b: 34). They were commonly associated with auxiliary forts and might have been produced specifically for the military.
Can geochemical and lithological studies be refined in order to achieve this?
Querns are regularly recovered from excavations in South Yorkshire. There is also a manufacturing centre at Wharncliffe on the Millstone Grit (Makepeace 1984). However, there is no survey of the county’s querns comparable to that by David Heslop (2009) for North Yorkshire and County Durham. This is now needed to address such questions as the extent to which Wharncliffe products were distributed in the county and to what extent sources on the Coal Measures (reviewed by Gaunt 2008) and elsewhere were also used.
With fragmented querns, it is rare that all pieces are found on site, and the rest must be removed elsewhere. Where did these fragments go, with whom, and what were they used for? Why were such practices carried out?
Do any quernstones and millstones from Templeborough survive in Rotherham Museum? Could these be fully catalogued, described, and published?
Stone spindle whorls and whetstones have been recovered at Doncaster and Templeborough (Buckland 1986: 27, fig. 15.12; Lloyd Morgan and Buckland n.d.: 645, fig. 24; May 1922: 124, plate xxxvii); and Rossington Grange Farm (Cool, Drinkall and Sitch 2016: 25). Stone cosmetic pallets were found at Doncaster.
Some monumental masonry excavated from Templeborough may survive in Rotherham Museum and should be photographed and described. Few other sites have produced architectural fragments. At Templeborough, the sculpted stonework included tombstone fragments re-used in later Roman structures (Buckland 1986: 289, fig. 16; May 1922: 127-132, fig. xliv-xlv), including monuments to “Crotus, son of Vindex, veteran of the IV cohort of Gauls, aged 40 years. Flavia Peregrina his devoted wife…erected this memorial to a most devoted husband”; “Cintusmus, a soldier of the fourth cohort of Gauls”; and “Verecunda Rufilia, a citizen of the Dobunni, 35 years old. Exingus her husband erected [this].” A plain stone altar and a fragment of another were also illustrated (May 1922: plate xxxix), though are not mentioned in the 1922 report. A notable find from South Yorkshire is the small stone altar 0.75m high discovered in 1781 at St Sepulchre Gate in Doncaster, with an inscription dedicated “To the Mother Goddesses” by Marcus Nantonius Orbiotalis (Buckland and Magilton 1986: 63-5, fig. 14; see cover image). Another altar dedicated to Mars was found in c. 1782 at Staincross Common (Collingwood and Wright 1965: RIB 622; Jackson 1858: 233) but has subsequently been lost.
Very few Iron Age and Romano-British artefacts made of other materials have been recorded from South Yorkshire. This may in part at least be due to the acidic soils across many parts of the county, but cultural and economic reasons are also likely.
An early La Tène-style purple glass bracelet fragment was excavated from the ring gully of a roundhouse at Balby Carr (Cool 2003a), perhaps a 1st century BC Continental import. Fragments of another glass bracelet were found at Engine Lane, Shafton (Cool 2003b) – though Roman and probably from the later 1st century AD, the primary fill of the feature it was recovered from was radiocarbon dated to 380–50 BC. A gully excavated within Enclosure E7 at Redhouse Farm, Adwick-le-Street contained two fragments of a blue glass bracelet with an applied blue and white glass strip and possibly of late 1st century AD date, in addition to a shale bracelet fragment (Upson-Smith 2002: 15). The enclosure ditch did not produce any dateable artefacts, but pottery from other features within the enclosure included late Iron Age ceramics and 2nd to 4th century Romano-British sherds.
Glass bangles or bracelets were more common in north-eastern England than in other regions (Brindle 2018: 27, fig. 2.25), although it is still unclear if these objects were actually worn by people on their arms. It has been suggested that they were used as hair rings or even as rings on horse manes or as other horse fittings (Stevenson 1976: 50, 53), and they might have held apotropaic or amuletic properties too. Few are ever found complete, and this fragmentation may have been deliberate. It is possible that after the death of an owner, they were broken-up and different people took away the fragments as mnemonic items (Pope 2005). Other kinship or social links could also have been symbolised by such fragmentation.
Five objects of Iron Age cobalt blue glass were recovered from one of the possible ‘mortuary rings’ at Sutton Common, including a bead decorated with an annular ‘eye’ of white glass (Henderson 2007: 158). The glass lumps are the first evidence for Iron Age glass production in northern England, let alone South Yorkshire. The blue and white glass bead is probably later Iron Age in date, but the chemical composition of one piece of glass suggested that it was made before the 2nd century BC (ibid.: 160). This may indicate objects of different times and traditions gathered together for a ritualised ‘placed’ deposit. Another unstratified cobalt bead and a lump of blue glass
were surface finds from previous fieldwork (Buckland et al. 1997: 237), and the intact bead may be another earlier Iron Age example. A globular amber bead and a globular shale bead were also recovered, the former a surface find; but the latter near the base of an outer gully forming part of Enclosure A’s ramparts on the western side (ibid.). A glass bead of indeterminate date was recovered from a hollow at Roebuck Hill, Jump (Wilkinson 2007: 23). Flint was recovered from the same context, and although this might be residual it could indicate a Bronze Age date for the bead.
There are difficulties in interpreting beads as items of Iron Age dress and personal identity when, outside of East Yorkshire Iron Age graves at least, they are more commonly found as individual objects not from burial contexts (Foulds 2017). Strings of beads can be separated, subdivided and exchanged with different individuals or passed on as heirloom objects (Giles 2012: 149), and may carry with them a variety of different social and symbolic meanings. A dark blue and yellow glass bead of Iron Age form was found at Rossington Bridge (Lloyd Morgan 2001: 24, fig. 20-33), along with probable Romano-British examples. There was a single find of a Roman glass bead at Armthorpe (Cool 2008a: 47), and part of a perforated Roman glass bead at Whirlow Hall Farm, Sheffield (Lortie and Doonan 2017: 86, fig. 63). Excavations at Templeborough produced glass melon beads and glass gaming counters (May 1922: 83, plate xxi-a); and glass annular beads, melon beads, armlets and gaming counters have been found in Doncaster in fort contexts and at Church Walk, High Street and Frenchgate (Cool 2008b: 137, fig. 3, 2008c: 282; Lloyd Morgan 1986: 93-5, fig. 21; Lloyd Morgan and Buckland n.d.: 43-6, fig. 10). Pieces of a blue and white glass bracelet were found at Redhouse Farm, Adwick-le-Street (Upson-Smith 2002: 23).
Roman glass vessels appear to have been quite rare in South Yorkshire, especially on rural settlements. Part of a blue-green Roman prismatic glass bottle of mid-1st to late 2nd century AD date was excavated at Holme Hall Quarry (Wilmott 2007: 75), and a fragment of a blown glass vessel was found at Whirlow Hall Farm (Lortie and Doonan 2017: 86, fig. 63). Roman window glass is virtually unknown on most rural sites, with just occasional fragments occurring as at Rossington Inland Port Phase 1b (Daniel, Harrison and Powell 2014b: 15). Glass vessels including jugs or ewers, beakers, bowls, bottles and unguentaria were recorded at Templeborough during the 1870s and 1916-17 investigations, along with window glass, glass gaming counters and part of a glass bracelet (May 1922: 82-4, plates xx-xxi). Fragments of bottles, beakers and bowls were found during the 1950s excavations at Rossington Bridge (Lloyd Morgan 2001: 34, fig. 31).
The most significant finds have occurred in Doncaster. Roman window and vessel glass were excavated during the 1970s rescue excavations – most notably from Frenchgate, including jugs, jars, beakers and bottles (Allen 1986: 103), many from a Flavian-period pit containing broken but substantially complete vessels. Roman vessel glass was also found at Church Walk and 8–10 High Street (Cool 2008b: 13940, 2008d: 288) and in the cremation cemetery at Doncaster Waterdale, much of this melted and fused and probably pyre debris derived from unguentaria that held perfumes used to anoint the bodies of the dead and scent the pyres along with other liquids used for libations during funeral rites (Cool 2013: 89-90). This was a relatively uncommon practice in Roman Britain and might reflect a military origin for those undertaking the ceremonies. The body of a conical blue-green glass jug with an ornate handle was also recovered at Waterdale during machining, a possible placed deposit. There was a neat break across the base of the neck and the handle. It was like a vessel from the Frenchgate pit (Allen 1986: 105, fig. 24.1).
As noted above, a globular amber bead and a globular shale bead of late Bronze Age or early Iron Age date were found at Sutton Common (Buckland et al. 1997: 237). An annular amber bead of late Roman date was recovered at Rossington Bridge (Lloyd Morgan 2001: 24); and another example from across the county boundary at Sandtoft in North Lincolnshire (Samuels and Buckland 1978, fig. 7).
A pierced jet disc pendant and a similar shale example were found at Templeborough (May 1922: 123-4, plate xxxvii), and fragments of a shale bracelet at Redhouse Farm (Upson-Smith 2002: 23). Jet and shale pins, spacer beads, bracelets and a jet counter were found at Doncaster (Buckland 1986: 27, fig. 15.10; Lloyd Morgan 1986: 96, fig. 94.8; Lloyd Morgan and Buckland n.d.: 60-3, fig. 23).
One onyx or agate intaglio was excavated at Templeborough (May 1922: 60, plate lv.a), a carnelian example (Leader 1878: 509); and a carnelian intaglio at Rossington Bridge (Henig 2001: 16, fig. 13). A small assemblage of carnelian and glass paste intaglios associated with silver, iron and bronze rings and a bezel was excavated at Doncaster, possibly part of a disturbed hoard containing several bronze brooches, a bronze scalpel handle, and nine coins (Henig 1986: 97-9, fig. 22).
Very few bone artefacts have been identified, and this likely to be a consequence of the acidic soils on all bar the Magnesian Limestone soils of South Yorkshire. Different depositional environments may help preserve organic materials though. An antler weaving comb was found in a ditch terminal by the eastern entrance of Sutton Common (Tuohy 2007), similar to examples more commonly found in southern England, though one is known from Harborough Cave near Brassington in Derbyshire (Armstrong and Jackson 1923). A bone awl was found at Area 7, Redhouse Farm, Adwick-le-Street (Upson-Smith 2002: 24).
At Church Field, Rossington a fragment of bone knife handle survived acidic soil conditions through being partly mineralised. Only one sherd of sand-tempered late Iron Age or early Romano-British pottery was also recovered (Atkinson 1998: 19), so the handle could be Iron Age or early Romano-British in date. A Roman bone pin was noted from the Market Place site in Doncaster (Buckland and Magilton 1986: 52), and several bone pins, gaming counters, dice and a late Roman or post-Roman bone comb have also been identified (Buckland 1986: 27, fig. 15.14; Cool 2008c: 282; Lloyd Morgan 1986: 96, fig. 21; Lloyd Morgan and Buckland n.d.: 46-52, figs 14-15). A bone toggle was found at Rossington Bridge (Lloyd Morgan 2001: 24, fig. 21).
Wooden Iron Age artefacts are incredibly rare within South Yorkshire as elsewhere in Britain but include the possible solid wooden wheel from Sutton Common, now unfortunately destroyed (Whiting 1936); and the notched log ladder of poplar or willow wood found in anaerobic waterlogged deposits at Sutton Common (Buckland et al. 1997: 233). The poor feature sampling strategies adopted during the open-area excavation of Enclosure A at Sutton Common undoubtedly restricted potential finds of further wooden objects. A wooden trug and a finely worked wooden comb of probable late Iron Age or Romano-British date were recovered from partly waterlogged deposits at Rossington Bridge (Lloyd Morgan 2001: 25, figs 22-3), amongst other worked wooden objects that unfortunately have not survived.
Few leather artefacts have been recorded as so few waterlogged or anaerobic contexts have been excavated. Leather shoe or sandal soles were reported by J.D. Leader from a well at Templeborough (Freemantle 1913: 101; May 1922: 59), though this is not described in his 1878 publication (Leader 1878). Leather shoe and carbatinae soles and other fragments were found at Rossington Bridge (Lloyd Morgan 2001: 25-7, fig. 24), though they were allowed to dry out so were not well conserved. Fragments of a leather one-piece shoe were found at the base of a ditch at FARRRS (Daniel 2017: 31-2), and a one-piece shoe came from a pond or waterhole at Finningley (Mildwaters 2008: 175). Elsewhere, a hob-nailed sole found at High Street, Doncaster (Lloyd Morgan 1986: 93) may have survived because iron minerals in the nails partly penetrated the leather.
Fragmentation was an important part of prehistoric and Romano-British depositional practices. Querns, glass and shale bracelets and necklaces, pottery, animal bone and even some human remains seem to have been broken down, some of this material later being reconstituted in novel combinations.
Many researchers have argued that from later prehistory, ritualised practices were focused upon houses, settlements and fields (Barrett 1989; Bradley 2003, 2005; Brück 1999; Williams 2003). Ditches of enclosures and fields, storage pits and wells, and the ring gullies, doorways and postholes of roundhouses were sometimes settings of everyday acts of magic and ritual. Concepts of ancestry and community, memory, tradition and oral history would have been important. For Iron Age and Romano-British people, cosmological, spiritual and religious ideas were a much greater part of daily life than today. Their relationships to animals and material culture may also have differed significantly from our own modern societies.
Several studies have examined the spatial and temporal distribution of artefacts, human and animal remains across Iron Age and Romano-British settlements. This work has shown that there were often patterns to this discard (e.g. Hill 1995b, 1996b; Hingley 2006; Willis 1997; Wilson 1992, 1999; Woodward 2002). Specific locations were the setting for a variety of depositional practices, some potentially associated with the symbolic importance and liminal nature of boundaries (e.g. Bowden and McComish 1987; Hingley 1990). Pits, roundhouse ring gullies, wells and waterholes, the corners of enclosure ditches, ditch terminals especially by entrances and gateways, and ditch junctions were often the focus for such deposition. The spatial location of activities within and around enclosures may have been structured by common cultural understandings of the use of domestic space and the socially ‘correct’ contexts in which to deposit materials including ‘refuse’ such as broken pottery, butchered animal bone, slag, charcoal and burnt stone (q.v. Cumberpatch and Robbins n.d.; Robbins 2000: 87). Refuse was thus disposed of in culturally prescribed locations, as in our own society, but on an unremarked everyday basis. This nevertheless reflected unconscious cosmological beliefs or habitual practices concerning boundaries, cleanliness and pollution.
One interesting and more unusual aspect of pottery deposition has been identified on some Romano-British rural sites in South Yorkshire. At Gunhills, Armthorpe, there was a general ‘background’ of ceramic deposition in ditches and pits, a dispersed, low-level of often fragmented and eroded sherds that is a common pattern on many Romano-British rural sites in the region and elsewhere. One otherwise unprepossessing stretch of boundary ditch between a trackway and a field, however, produced nearly 4000 sherds of Romano-British pottery extending in date from the late 2nd century to the later 3rd century (Richardson 2008: 15, 18, fig. 6; Leary, Evans et al. 2008: 31, 44). Some of the sherds were relatively large, ‘fresh’ and unabraded, whereas others were very worn and had clearly been exposed on the surface for protracted periods. There was a higher proportion of bowls, dishes and wide-mouthed jars in this group than in most other ceramic assemblages from the site, yet the site of deposition was relatively remote from other places of ceramic disposal and concentrations of human activities. The pottery dump was almost certainly a secondary deposit of material that had accumulated elsewhere, probably in a midden heap, before being gathered up together with fresher material and deposited c. 270 AD or soon after (Leary, Evans et al. 2008). Why though was this material transported away from the nearest settlement focus, and why was it dumped in that specific length of field system ditch?
Elsewhere at Armthorpe, substantial portions of pottery bases or rims were found as deposits in the bottom of ditches ‘nested’ within piles of burnt or fire-cracked stones, one being inverted but substantively complete (Chadwick, Powell and Richardson 2007: plates 1-2; Leary, Evans et al. 2008: 27, 44), whilst two disused ovens contained most of a lid-seated jar and a complete jar base. An iron plough share tip was also found in the backfill of a small oven or corn drier; and an iron spade shoe in a ditch could even have been deliberately deposited (Richardson 2008: 21). At High Street, Shafton, the eastern boundary ditch contained three almost complete Romano-British vessels (Burgess 2001a). At Enclosure E3 at Redhouse Farm, Adwick-le-Street, a substantial group of Black Burnished Ware sherds was deposited in a field ditch next to either a narrow enclosure entrance, or the point where the enclosure bank met an associated field boundary (Upson-Smith 2002: fig. 11).
Large dumps of ceramic sherds of mixed date and different degrees of wear and weathering have now been recorded at several sites. In Area 3 at FARRRS, several hand-dug slots across one ditch produced much of the overall Romano-British pottery assemblage from the site (68% by count, 56% by weight), with just one slot producing nearly 500 sherds of predominantly later 2nd to early 3rd century AD ceramics (Daniel 2017: 10). This consisted of grey wares, including the base from a large jar or bowl, most of a large ‘native-tradition’ jar or bowl, and fragments from a Central Gaulish samian bowl or dish. Nearby features contained little or no pottery. At Stainforth Marina, one excavated segment across a ditch in an evaluation trench produced fire-cracked pebbles and 120 pottery sherds, by far a larger group than the rest of the ceramic assemblage from the entire site (Strafford 2014: 15-16). Predominantly grey ware, some sherds were in a fresh condition, but others were notably worn and abraded, and may have been deposited in watery conditions. One fresh sherd from a colour-coated beaker from the Nene Valley or Lincoln, the only one identified from that type of vessel, featured a white dolphin as a raised barbotine decorative motif (Rowlandson 2014: 44, 47, plates 21-2).
At Hatfield Lane, Edenthorpe some large groups of Romano-British pottery were recovered from pits, several ditches and a ditch intersection (Weston 2015: 6-7). The pottery contained worn and sooted sherds from domestic occupation, but also warped and cracked ‘seconds’ and ‘wasters’ from pottery production. The quantities of pottery waste might reflect the use of these features as handy dumping areas following the rake-out of kiln structures. The kilns might have been shallow or above-ground features that did not survive later plough truncation (Rowlandson and Monteil 2015: 27) – the broadly contemporary kilns at Blaxton certainly had little surviving structural remains (Buckland and Dolby 1980). Alternatively, kilns were located close by but not actually within the excavated area. Two pits were especially interesting – one contained nearly two complete samian bowls on the base of the pit, along with a fragment from a samian cup (Rowlandson and Monteil 2015: 25). Most of the remaining fresh sherds were dominated by large bowls including substantial portions of three warped or heavily cracked large bowls. Another pit again contained fresh sherds including many that were warped and misfired, but also a large proportion of an unusual jar in an oxidised fabric with white painted horizontal bands (ibid.: 26-7).
At Rossington Grange Farm, one short length of ditch forming the western boundary of Enclosure 5 produced a dump of 3321 sherds from its tertiary fill, all apparently from one episode as evidenced by cross-joins (Roberts and Weston 2016: 15, plate 2; Rowlandson, Monteil and Hartley 2015: 42-3). These were mostly local grey ware storage jars, bowls and dishes of mid-2nd to late 2nd century date, including some substantial upper or lower portions of vessels and many of which had been heavily used for cooking as they had carbonised residues. There were also sherds from at least seven warped wasters and misfired vessels, and fragments of mortaria manufactured in Lincoln. This ceramic group again appeared to be a mixture of pottery production waste along with domestic discard that had been deposited together; and overall would have looked quite striking.
Such depositional patterns do not easily fit within functional, ‘common-sense’ explanations of refuse disposal and casual discard, but also not cliched ideas of very formally structured ‘ritual’ deposits either (Chadwick 2012: 288-9, 2015: 52-3). At Hatfield Lane, Edenthorpe, the selection of more prized and/or unusual vessels to be deposited alongside the pottery production waste might have represented a practice of tidying up or ‘closing down’ the process of pottery production (Rowlandson and Monteith 2015: 26-7). Such ‘decommissioning’ might also explain some of the deposits in the ovens or corn driers at Gunhills, Armthorpe (Chadwick, Powell and Richardson 2007: 65-6). Other groups of near complete vessels combined with burnt stone may have been the residues from feasts held to celebrate key stages in the agricultural cycle such as planting, harvests or autumn culling; or more irregular events such as births, marriages, and deaths (Chadwick 2012, 2015). The combination of fresh sherds with worn and abraded material that had probably been lying in surface middens prior to deposition might reflect episodes when farmsteads and enclosures were periodically cleared out, but it is unclear why this dumping seems to have been such an episodic process, or why the material was not incorporated in any manuring spreads – perhaps manuring in the medieval and post-medieval form did not take place, especially if animals were just folded onto arable fields after cultivation. It does seem odd that sometimes rather distant ditches were chosen for such deposition, rather than those closest to domestic inhabitation.
It is possible that the mixing of materials discarded from the household with the wider landscape reflected implicit or subconscious ideas about ties to the land and social identity (J.G. Evans 2003: 141-3, 2007: 182), whilst pottery’s associations with food preparation, storage and consumption might have been significant too. Pottery may have signified a human presence within the landscape, and such deposition may have been a deliberate ‘entexturing of the ground’ (J.G. Evans 2003: 126) or of ‘signing the land’. Occasional concentrated dumps of material found in field ditches may have been linked to notions of boundaries, tenure and identity. Such dumps could also have marked changes in household occupancy or to rights of access, tenure and ownership.
The depositional processes in Romano-British wells, deep pits or shafts have been explored by several detailed taphonomic studies (Ayton 2011; Berg 1990; Cool and Richardson 2013; Gerrard 2009; Millett and Taylor 2006; Poulton and Scott 1993; Roskams et al. 2013; van Driel-Murray 1999), and some of the regional evidence is presented elsewhere (Chadwick 2009, 2010, 2015). Not all Romano-British wells have produced ‘unusual’ deposits by any means, but small numbers have yielded extremely interesting sequences of artefacts and palaeo-environmental remains. Wells were used to dump domestic refuse, butchery waste and building rubble, and such deep features inevitably act as ‘traps’ for stray artefacts or for small animals. Other deposits, however, included human remains, animal remains (particularly dogs and the heads/skulls of cattle, sheep and horses), whole or near complete ceramic and metal vessels, leather shoes, beehive and flat rotary stone querns, brooches and iron tools. The presence of human and animal remains would have been incompatible with drawing drinking water, so this activity presumably followed disuse and might have reflected closure or termination rites prior to abandonment.
There was a strong tradition in many parts of Iron Age Britain for the deposition of metalwork, other artefacts and human and animal remains in river and lakes, but perhaps due to the lack of Iron Age wells this does not seem to have been a particularly significant social practice in such features, although some deposits from a late Iron Age well excavated at Silchester may be an exception (Fulford et al. 2018). As deep well shafts with timber or stone linings were largely a post-conquest development, placed deposits in wells might also have been a Romano-British phenomenon (Webster 1997a: 136-7), whereby indigenous traditions concerning water, pits and deposits were combined with ‘Roman’ chthonic rites and other beliefs concerning watery places. Hingley (2006: 238) proposed that during later prehistory iron objects were mainly placed within boundaries, but during the Romano-British period this changed to wells and deep pits.
A probably Iron Age waterhole excavated during the FARRRS road scheme did not produce any particularly ‘unusual’ animal remains, but its basal fill did have a stone block and a large part of a very abraded mica-rich dish probably from a Lincoln workshop, and dating to the mid- to later 1st century AD (Daniel 2017: 17). This artefact might have originated in the vexillation fortress at Rossington Bridge around 4 kilometres away and could have pre-dated the final invasion of the north, and one can speculate whether it was a diplomatic gift, war booty, or was already an many decades old before it was deposited (ibid.: 28). Four waterholes were excavated at Rossington Inland Port, some at least probably Romano-British in origin as they had been dug into a ditch that produced Romano-British finds. They produced some pottery and animal bone, two of them more than 30 sherds each which was slightly unusual for this rural site; but one contained a large limestone block, with a cup-shaped depression worn or drilled into one surface, perhaps a pivot-hole in a door jamb; and another contained a stone block (Daniel, Harrison and Powell 2014b: 6-7). This admittedly limited evidence may indicate that waterholes did not carry the same set of symbolic meanings as wells.
Few of the small number of Romano-British wells excavated in South Yorkshire have produced potentially deliberately placed deposits, though of this small number most have not been fully investigated. One well excavated by J.D. Leader at Templeborough contained quern stones and leather sandal soles (Freemantle 1913: 101). A timber-lined well within the principia or praetorium at Templeborough was investigated by a site foreman after the excavations by Thomas May had ceased, and this produced pig jaws, a stone column base, a ‘black vase’ (some form of small colour-coated or terra nigra beaker?), and an amphora (May 1922: plate liii). A well at 8–10 High Street in Doncaster produced a complete but fragmented samian cup, most of a mica-dusted beaker and the large inverted top half of a stamped Dressel 20 amphorae (Chadwick and Burgess 2008: 44; Leary 2008: 87).
Given the problems and constraints outlined above, it remains difficult to identify deposits of material culture and animal remains that seem to have been more deliberately placed or ‘structured’. The archaeological evidence is complicated by the fact that often it was the same locations in and around enclosures and fields that were also chosen as the settings for Iron Age and Romano-British depositional events that appear to have had a more conscious, deliberate aspect. These included complete or near-complete ceramic vessels, complete or deliberately broken quernstones, complete or partially-articulated animal remains (Associated Bone Groups or ABGs), and human remains (Chadwick 2004, 2008a, 2012, 2015; Chadwick, Martin and Richardson 2013; Morris 2008, 2011; Wilson 1992, 1999). Trying to separate ‘everyday’ refuse disposal from more ritualised practices is therefore extremely difficult, especially when it is unlikely that prehistoric people would have made such binary distinctions themselves (e.g. Brück 1999; Chadwick 2012; Cumberpatch and Van de Noort 2007). In some instances, these might have represented the deliberate burial or discard of objects belonging to the deceased; or parts of people and animals and/or food items that were offerings to gods, spirits or ancestors (Cunliffe 1992; Smith et al. 2018). Propitiatory or apotropaic concerns may have been behind some deposits, whilst other remains may have resulted from relatively informal, small-scale practices associated with events in human lives such as births and deaths; or the agricultural calendar, such as rites associated with planting and spring livestock births, harvests, and the autumn culling of animals. Other deposits might have been from divination, or residues from feasting and therefore even more difficult to discern from everyday discard.
Possible examples of placed deposits from the wider Yorkshire and north midlands region have been outlined previously (Chadwick 2004, 2008a: app. F, 2012, 2015). One factor that may hamper the identification of ABGs of animal and human remains is the generally poor preservation of bone on some South Yorkshire sites due to the more acidic soils on Coal Measures and Sherwood Sandstone geologies. But this might only be a partial explanation, as it is also possible that cultural practices in the past were also involved. During the Iron Age and Roman periods animal remains could have been butchered and disposed of in different ways to other regions (C. Merrony pers. comm.), and it is possible that bone was ground up as fertilizer or deposited on the ground away from settlements or in places that have left little or no archaeological signature. Detailed work on body part representation, fragmentation and degree of wear and erosion on bones is required to test this hypothesis.
At Sutton Common, two human skulls and some other human bone fragments were found in the northern outer ditch terminal of the eastern entrance, along with dog remains, a saddle quern fragment, and the only yew wood from the site (Knüsel 2007: 139; Outram 2007: 142; Thomas 2007: 147; Watts 2007: 145). The northern inner ditch terminal of the east entrance contained a human tibia fragment, the only stratified Iron Age pottery from the site, a small rim sherd (Cumberpatch, Vince and Knight 2007: 143); and an antler weaving comb (Tuohy 2007: 147). Given the general paucity of artefacts at Sutton Common (although the poor excavation and sampling strategy did not help), it seems likely that these locations held social or cosmological significance as thresholds or portals through enclosing boundaries. Boundaries and entrances are often sites of social tension and/or symbolic significance in many more recent ethnographically documented small-scale communities around the world.
On sites on the Magnesian Limestone, more detailed information is recoverable. At Pickburn Leys, two substantially complete late Iron Age vessels and animal bone were recovered from the fill of a recut enclosure ditch, on the south-eastern side of the enclosure near to an entrance (Sydes 1993: 37-8). A later recut of the ditch at this point produced a rotary quern stone fragment. A group of disarticulated cattle bones was found in the eastern terminal of the E1 enclosure internal sub-enclosure ditch at Adwick-le-Street, possibly a deliberate dump to mark the entrance (Meadows and Chapman 2004: 5). A dog was buried in the western ditch of Enclosure ditch E1 at Adwick-le-Street (Meadows and Chapman 2004: 5), another in a pit in Enclosure E3, around 20m away from two pits containing querns (Upson-Smith 2002: fig. 6; see below). Cremated animal bones found in a possible stone-lined cist in the enclosure at Roebuck Hill, Jump may have been dog remains (Robinson and Johnson 2007: 8-9). Another pit on this site contained burnt bone, loomweight fragments, late Iron Age or early Romano-British pottery, and a burnt quernstone fragment.
The treatment of quernstones and quern fragments is particularly interesting (Buckley 1991; Chadwick 2008a, 2015). They are often found in topsoil or the uppermost fills of features, suggesting that these could have been tertiary or closure deposits. Sometimes whole top stones or base stones of querns were deposited whole, as in the large ring ditch surrounding one of the roundhouses at Balby Carr D(ii), where three beehive quernstone bases were recorded (Heslop 2005; Richardson and Rose 2005: 4.7) along with cattle, sheep/goat, pig, and deer remains; in addition to Iron Age pottery and the La Tène-style glass bracelet noted above. Intriguingly, the ring ditch around the smaller roundhouse at Balby Carr D1 contained articulated remains from a sheep/goat near the entrance (Rose and Roberts 2006: 6.1.2). A beehive quern fragment was found with a deposit of charred cereals at the entrance terminal of the enclosure ditch at Roebuck Hill, Jump (Robinson and Johnson 2007: 7).
At other times, querns seem to have been very deliberately smashed, and/or have had their external grinding surfaces removed (Heslop 1988, 2008a), as at Rossington Bridge Farm (Cruse and Gaunt 2016). Eighteen saddle quern fragments were found in just two postholes from a single four-post structure at Sutton Common (Watts 2007: 145-6). Half an upper beehive quernstone was found in a pit opposite the entrance of Enclosure C at Woodhead Opencast Site, Wombwell (Jones 2003: 11), and at Enclosure E1 at Redhouse Farm, Adwick-le-Street, a pit contained late Iron Age or early Romano-British pottery, animal bone and three fragments (two conjoining) of a rotary quernstone (Meadows and Chapman 2004: 8). A near-complete but fragmented beehive upper stone was found in pit near Enclosure E3 at Redhouse Farm (Upson-Smith 2002: 14, 23), incorporated into the clay and stone lining. A second pit contained a dog burial, a third a broken upper beehive quern. Further south at Enclosure E8, another pit contained an inverted beehive quern top stone (Upson-Smith 2004: 9). At Enclosure 1 at Pastures Road, Mexborough, a partial beehive quernstone base was recovered from the south-east corner of the enclosure ditch, and another partial quern base stone from a pit adjacent to the northern terminus of the enclosure ditch (Williams and Weston 2008: 5-6).
Why some querns were selected for such fragmentation and why others were left complete or even in fresh condition is unclear; but might well reflect the individual biographies of the objects and those who used them. Sometimes they are ‘right side up’, other times inverted. Often the fragmented querns are incomplete, the pieces apparently removed from the feature or from the site altogether. The violence of such destruction is noteworthy and may reflect ritualised ‘killing’ of objects perceived to have great power. Alternatively, such querns may have belonged to the deceased and had thus become ‘polluted’ in some manner by their deaths, and perhaps had to be destroyed to safeguard the living.
There is overwhelming evidence across Britain for considerable continuity in such deposits from the Iron Age into the Romano-British period but also novel developments, perhaps reflecting a mixture of Iron Age and ‘Classical Roman’ traditions and also the beliefs of all those who came to northern England from across the Roman Empire (e.g. Aitchison 1987; Clarke 2000; Fulford 2001; Hingley 2006; Merrifield 1987; Millett 1994; Smith et al. 2018; Willis 1999; Wilson 1992; Woodward and Woodward 2004). A juvenile pig skeleton was found in a small pit within the Romano-British enclosure at Barnburgh Hall (Richardson 2005a: 6.13). At Billingley Drive, Thurnscoe, a complete iron snaffle bit of 1st century AD date was deposited in one of the ditch terminals of the Phase III D-shaped enclosure (Cowgill 2004a: 50). A near complete Roman two-link iron snaffle bit and quern fragments were also found in a pit at Rossington Grange Farm (Cool, Drinkall and Sitch 2016: 25; Roberts and Weston 2016: 17). At Hatfield Lane, Edenthorpe, one pit contained two almost complete samian bowls in addition to other pottery including kiln wasters (see above); and another pit had a rare double hopper quernstone complete but broken in situ, along with iron box fittings and lead waste (Weston 2015: 7). In a building that was probably part of the principia of the fort at Doncaster, a small pit cut into the floor was lined with sherds of amphorae, and a complete cooking pot, a samian platter and chicken remains were placed within it (Buckland 1986: 49). At Church Walk, Doncaster, the beam slot of an early Roman timber building produced a copper-alloy terret ring and part of an unusual ceramic platter (Chadwick, Martin and Richardson 2008: 15), but could also be accidental inclusions derived from elsewhere.
This identification of possible examples of ‘placed’ or ‘structured’ deposits, however, based on them being ‘unusual’ or subjectively different to other finds on site, does not constitute a robust programme of detailed objective analysis. Even if apparently intentional patterning on settlement sites can be identified, this need not imply overtly ritualised behaviour, especially if there were unspoken social conventions governing deposition during the Iron Age and Romano-British periods. Prosaic practices and more arcane rites may have left ostensibly identical material remains – burnt stone and broken pots might have resulted from a minor domestic accident, a need to clear out old rubbish, or a feast held to honour a deceased relative. These remains might also only represent the final stages of a wide range different practices, some of which perhaps involved organic items such as flowers or fruit that have left no traces (Smith et al. 2018: 124, 169). Such practices were unlikely to have been a set of rigidly proscribed rules, but rather a ‘suite or palette of conventions’ (Chadwick 2012: 301), a ‘shared lexicon’ reproduced but also transformed by each community (Giles 2000: 153). It is thus extremely difficult to define ‘structured’ deposits and to distinguish these from ‘everyday’ refuse (Chadwick 2012; Cumberpatch and Van de Noort 2007; Garrow 2012). This realisation should not be taken to mean, however, as an argument for the notion that archaeologists should not attempt to identify such distinctions and investigate them in detail (contra Roskams et al. 2013).
Detailed analyses of everyday acts of finds deposition and dispersal in addition to potentially more structured occurrences are therefore crucial; but have been lacking on most excavated sites in the region to date. This can be partly achieved through more rigorous and systematic excavation and sampling methodologies, accompanied by spatial and statistical analyses, and sophisticated combinations of taphonomic and interpretative theoretical ideas. Repeated patterns and associations can then be identified, as well as exceptions and discrepancies (Brudenell and Cooper 2008; Chadwick 2009; Garrow 2012). Wide-ranging synthetic research that examines taphonomic processes, spatial distributions and diachronic changes on Iron Age and Romano-British settlement sites at intra-regional and inter-regional levels is therefore required. There have been many site-specific investigations of social space and chronological variations of deposition in and around Iron Age and Romano-British rural settlements (e.g. Evans 2001b; Gwilt 1997; Martin 2007; Millett 2007; Mudd 1999; Robbins 2000; Taylor 1997, 2001; Willis 1997, 1999; Woodward 2002; Woodward and Hughes 2007). Such methods would be usefully employed in South Yorkshire but need to be expanded beyond such site- specific studies.
Were large ceramic deposits just prosaic dumps, or something more meaningful? Can we develop methodological criteria to assess depositional practices on Iron Age and Romano-British sites, and to try and distinguish casual discard and dumps from more deliberate ‘placed’ deposits?
Much finds and contextual data gathered on-site in commercial archaeology is of poor quality or even useless for detailed comparative inter-site investigations (Blinkhorn and Cumberpatch 1997; Cumberpatch and Blinkhorn 2001, Cumberpatch and Roberts 2012). This may be compounded by different organisations employing different excavation and sampling methodologies, and variations in data recording and presentation.
Shrines could be defined as relatively small-scale structures and features generally recorded on rural sites, and possibly the settings for relatively modest and informal ritualised practices; whereas temples refer to structures forming part of GrecoRoman and Romano-Celtic architectural traditions, and perhaps also more formalised ritual practices. There are considerable theoretical and methodological difficulties in recognising relatively ambiguous archaeological features as shrines, however, and/or attributing a specifically ritual function to them (Smith et al. 2018: 121). In some cases, this has been in terms of what features are not rather than more positive identifications. Within South Yorkshire there are several unusual or enigmatic features and sites that may suggest ritualised practices were taking place.
Near Enclosure 8 at Redhouse Farm, Adwick-le-Street, there was a square structure defined by beam slots c. 2.1m across, opposite an early enclosure entrance and beside a trackway (Upson-Smith 2006: 6). It is similar in plan to structures from the hillforts at Danebury and South Cadbury interpreted as shrines (Downes 1997: 148150; Wait 1985: 166-167); and other Iron Age and Romano-British features at Maiden Castle, Uley and Chelmsford (Drury 1980: 45-50, fig. 3.2). Although the Redhouse Farm structure could conceivably be a small rural shrine (Chadwick 2008a: 663-4, app. F), no finds were recovered and thus there is no archaeological evidence to indicate a ritual function. It was only four metres away from a possible four-poster granary, so may simply have been an ancillary storage feature.
At Topham Farm, Sykehouse, a continuous circular gully 12m in diameter contained burnt animal bone and fired clay or briquetage; and produced a radiocarbon date of 60 BC–AD 180. This was replaced by another unbroken ring gully only 5.5m across that contained two shallow rounded postholes within it and which produced a similar date. The ring gullies do not seem to have been roundhouses and might have been small shrines (Roberts 2003: 29-30). The earlier structure contained 40% of all the Romano-British pottery from the entire site, including several near complete vessels. These two features might alternatively have been haystack stands or fodder ricks, though this does not preclude their use for more ritualised deposition.
There appears to have been a notable lack of identifiable Romano-Celtic style temples in northern England and this was especially true of the north-east, even in areas such as East Yorkshire that had more villas than South Yorkshire (Smith et al. 2018: 135). This may indicate that temple distributions were not simply mirroring the extent of Roman-style architecture and more ‘Romanised’ lifestyles; but might suggest deeper differences in underlying traditions and understandings of sacred space and ritualised practices from those more prevalent in southern England. For instance, regional and local ideas concerning ritualised areas, and practices where any profane and sacred divides were regarded as more fluid and protean, might have been more important than ‘formal’ religious spaces. The Roman stone altars found at Doncaster, Templeborough and possibly at Staincross Common (see above) are the most obvious manifestations of ‘formal’ Roman religious practices.
On the floodplain of the River Idle near Bawtry, archaeological investigation ahead of work by the Environment Agency revealed the tops of several dressed stone columns, along with a large artefactual assemblage including 72 Roman coins, iron objects, a copper-alloy fibula, and rolled lead sheets that might be prayers, votives or ‘curse tablets’ – phylacteries and defixiones (Berg and Major 2006: 5; Roberts, Deegan and Berg 2010: 77). The very large ceramic assemblage contained a high proportion of bowls, dishes and beakers with large numbers of vessels sooted and burnt; whilst included small flanged vessels with internal scorching suggest their use as tazze or incense burners. This may well be the location of a Romano-Celtic temple site therefore, and it is potentially one of the most significant Romano-British sites ever discovered in South Yorkshire and the wider region.
A geophysical survey was undertaken across the site at Bawtry Carr, but much of the area did not appear to have significant buried remains (Harrison and Webb 2006). Only magnetometry was used, however, and not soil resistance survey or Ground Penetrating Radar (GPR) that might have better detected stone walls and other remains sealed beneath layers of alluvium. In the surrounding area, eight coins of Gallienus to Diocletian (AD 253-304) were found a little way from the site around 1840 (Magilton 1977: 13), and a gold Iron Age stater also found at Bawtry might indicate some earlier Iron Age significance to this locale. The association with rivers is a feature of other Iron Age and Romano-British temple sites such as Redhill in Nottinghamshire, and of some coin hoards and metalwork deposits (Blagg 1986; Bland et al. forthcoming; Palfreyman and Ebbins 2003). Further archaeological work is urgently required at this location, which in the short-term by the depredations of illegal metal-detectorists and in the long-term is threatened by flooding and river erosion as a result of climate change (see below).
Can we define ‘typical’ topographic locations for known ceremonial sites in Iron Age and Roman Britain, and use such data to prospect for more?
Until recently there was very little archaeological evidence for Iron Age burial practices in South Yorkshire, due in part to the lack of excavated sites and significant problems with bone preservation in the often-acidic soils.
In other areas of Britain, from the late Bronze Age onwards formal burials seem to have disappeared from the archaeological record (Brück 1995). Most of the human dead might have been excarnated, exposed on timber platforms or on the ground surface (Carr and Knüsel 1997), or disposed of in rivers, lakes and bogs. Some disarticulated human remains found on some late Bronze Age and Iron Age settlements might have been selected and circulated amongst the living prior to deposition (Brück 1995). During the mid to later Iron Age East Yorkshire had the square barrow and inhumation burial rite, but much of northern England was once regarded as having isolated pit or cist burials (Mackey 2003; Whimster 1981). The archaeologically visible Iron Age burial practice in West Yorkshire consisted of isolated crouched or flexed inhumations within pits in corners of enclosures, just outside enclosure ditches, or next to field ditches and trackways. The social meanings associated with boundaries may therefore have been important in some instances. In general, as only a relatively small number of Iron Age burials have been excavated in northern England, most members of the Iron Age population are still clearly missing from the archaeological record.
Just a few hundred metres north of the South Yorkshire county boundary at Enclosure B, Barnsdale Bar East, an enclosure ditch terminal was found to have cut through or across the position of a crouched inhumation burial. This was of an adult male 45+ years in age and radiocarbon dated to 380–170 BC (Grassam and Ford 2008: 5). This suggests either knowledge of the original burial, or recognition of it as a surface feature. The ditch was later recut and incorporated within a field system.
On the Magnesian Limestone of South Yorkshire at Bilham Farm near Brodsworth, excavations by the Universities of Sheffield and Hull in 2009 revealed the skeleton of a young male around 18 years old buried in a prone, flexed position, and with pierced pendants made from a boar’s tusk and a dog or wolf canine. Shortly after burial the body appears to have been partially exhumed and the torso moved (McIntyre 2009). This burial, the uncalibrated radiocarbon date of which suggests death occurred most likely sometime during 200–100 BC was within a double-ditched trackway, north-east of a funnel-shaped entrance into a ‘banjo’ enclosure with little evidence for domestic inhabitation, but probably associated with livestock herding (C. Merrony pers. comm.). The location of the grave and identity of the deceased were probably known to those who moved the body. The burial of a 5-6-year-old child in a crouched position was also found nearby in 2010 (McIntyre 2010). This important site remains unpublished, and no calibrated radiocarbon date has been disseminated.
The burials near Brodsworth might seem to be in unattractive positions beneath the passage of feet and hooves, yet these are paralleled by other graves in West and East Yorkshire (Chadwick 2016b: 102-5). It is possible that these were liminal places where those who had transgressed social norms or had died unlucky deaths could be safely interred without harming the living, but it is equally possible that these were honoured places, linking the deceased to the livestock on which life and status depended. The human dead could have kept watch over the herds and flocks of the living (Chadwick 2016b: 105; Giles 2012: 223).
Cremated human bone and pyre debris was recovered from a single posthole at Sutton Common, located 12m west of one of the ‘mortuary ring’ structures. The bone was from a subadult person more than 13 years old, but it was not possible to identify it any further (McKinley 2007: 156-7). A series of small pits and postholes associated with the mortuary rings also produced cremated or charred animal bone and charcoal. It is unclear if the animal remains were food offerings placed on cremation pyres; or if these animals had been accorded some form of relational equivalence to humans after death. All features had been heavily truncated by ploughing, but in addition only a few narrow sections appear to have been excavated across the small gully structures (cf. Chapman and Fletcher 2007, figs 8.2-8.4). It is unclear why these features were not 100% excavated when it was suspected that they might have a funerary purpose. It thus remains unproven whether these features had a connection to mortuary rituals, though with artefacts such as the glass beads and the gold bar being found in this same general part of Enclosure A at Sutton Common, it does seem likely. Cremated remains might have been placed within the gullies on the old ground surface, or at the base of marker posts, which is why little survives. No cremated human bone of possible Iron Age date has been found elsewhere.
Cremated or calcined animal bone is known from several other South Yorkshire Iron Age and Romano-British sites, including the ring gullies of roundhouses 2, 4 and 8 at Topham Farm, Sykehouse (Roberts 2003: 8, 10, 17), a pit at Pastures Road, Mexborough (Holst and Richardson 2008: 27), and from a stone-lined pit or cist at Roebuck Hill, Jump (Robinson and Johnson 2007: 9). A fragment of burnt bone found above one of the subrectangular gully features at Rossington i-Port has not yet been identified (P. Daniel pers. comm.). A deposit of cremated bone from the base of a ditch at Barnsley Road, Goldthorpe is not identifiable (Richardson 2015: 6). It is unclear if such bone resulted from the deliberate cremation of animals, or if it was an accidental by-product of intense cooking.
At Sutton Common, the outer ditch terminal on the northern side of the east-facing gateway contained the fragmented remains of the skulls of two adult males aged 2535, without mandibles and only traces of neck vertebrae (Knüsel 2007: 137-9). Some mandibular fragments, teeth and post-cranial bones were found in the same context, however. The inner ditch terminal on the northern side of the east entrance contained a human radius shaft fragment.
Human bones from Iron Age settlements might have been residual and accidentally incorporated into later features, but some may have been deliberately curated as mementos of individuals or more general ancestral relics. The two skulls at Sutton Common might have been deposited with the other objects and materials in the ditch terminal; or were originally set up above the entranceway. Finds of disarticulated skulls from Iron Age contexts across Britain suggest that significance may have been attributed to the head, either of relatives and revered ancestors or those of enemies (Aldhouse-Green 2004; Armit 2006, 2012; Armit et al. 2011).
During dredging of the Mother Drain and River Torne at Rossington Bridge Pumping Station in the 1950s, workers found human bones under a layer of peat, close to a series of waterlogged timber posts that were probably the footings of the Roman road bridge (Buckland, Hartley and Rigby 2001: 10). The human remains included cranial and pelvic fragments and were associated with much Roman pottery including kiln waste, animal bone, and leather remains of shoes and sandals. The bone might be derived from one or more disturbed inhumations (ibid.: 12), but some bones had been cut and modified. The presence of leather objects and iron artefacts such as a cauldron chain and poker, however, could also be indicative of a series of ritualised deposits at the same locale, some of which may even pre-date the Roman occupation. Animal and human remains might have formed a part of these. Radiocarbon dating of the surviving bones would be useful, which could be Iron Age or Romano-British. Such combinations of metalwork, ceramics, and disarticulated remains have been noted at other similar Romano-British ‘watery’ sites recorded elsewhere in Britain, including the River Thames near Kingston on Thames, and the River Tees at Piercebridge in County Durham (Bland et al. forthcoming; Heathcote 1990; Walton 2012, in press).
The Roman conquest of the north brought changes in burial practice, though how far these applied to indigenous people or were just associated with ‘Roman’ immigrants is still unknown. The most visible Roman-style burials have been excavated in and around Doncaster. At a national level, it has been recognised that Roman-period cremation burials were most common during the earlier Roman period, with inhumation gradually becoming the dominant rite by the late Roman period (Philpott 1991: 217-227; Smith 2018: 216-220), though there was much regional variation.
Older finds of Roman-period burials within and around Doncaster are poorly recorded (Buckland and Magilton 1986: 27, 60), but one notable find was a probable burial mound removed in 1902 at Corporation Paddock, but which produced one or two complete ceramic flagons or jugs and a ceramic tetina – the latter objects have been variously interpreted as oil lamp fillers, specialist sauce dispensers or baby feeding bottles (Cool 2006). Close by in 1968, groundwork at the law courts at Waterdale, Glasgow Paddocks disturbed four complete ceramic vessels (a Black Burnished ware 1 jar, bowl and dish; and a white ware flagon) possibly associated with a small pit containing a few cremated bone fragments. Roman infant burials were found within the late Roman fort at Doncaster (Buckland 1986: 17) but remain unpublished. Near Templeborough, two urned cremation burials were discovered during construction work in the 1950s (Greene 1957b: 289-90, plate 4).
At Hallgate in 1995, five cremation burials and six inhumations were excavated by the South Yorkshire Archaeology Unit. The cremations were both urned and unurned, and one of the unurned examples was found in close association with a ceramic flagon (Atkinson and Cumberpatch 1995: 21-2). A cremation pyre pit was also excavated, with evidence for intense burning, heavily burnt human bone fragments and a complete and unburnt slip-decorated pot placed within it, perhaps as a closing part of the rite. Two inhumation burials were accompanied by grave goods, in one instance a small ceramic jar with an upturned dish covering it, near the left leg of the individual. The SYAU was disbanded by South Yorkshire Council before the Hallgate site could be fully analysed and published. At 8–10 High Street in 2003, AS WYAS excavated a double inhumation burial that contained the remains of two adult males – one individual was placed in a supine position on the base of the grave and the other on top, head to toe, in a prone position (Chadwick and Burgess 2008: 3940; Holst 2008: 327-31). Their pathologies indicated degeneration of the neck joints, sinusitis, and a healed ankle fracture. Several iron objects found above the right clavicle and left shoulder of one individual may have been nails or shroud pins, but only a few fragments of brick or tile, mid to late 2nd century pottery sherds and one piece of glass were retrieved from the grave fill, though a pot base might conceivably have been a token of some sort.
At Doncaster Waterdale, a cremation cemetery contained 18 cremation burial deposits, in two groups of eight and one of two. Most were in a series of intercutting pits containing cremated bone, pyre debris, artefacts and other offerings (Davies 2013: 7-11). One burial group (Group A) was partly enclosed by a rectilinear gully, and had plant remains including pine nut, walnut, fig, grapes, lentils, a possible date fragment and cereals. Artefacts included five ceramic oil lamps (Griffiths 2013). The second Group B cremations were associated with a more limited range of artefacts and burnt food offerings. Group C contained only two intercutting pits with small quantities of burnt bone but numerous large fragments of partially burnt and broken amphorae (mostly from Baetica in Spain, but also from Narbonnensis in Gaul, and from Verulamium), remains of melted glass unguentaria vessels, and further foodstuffs including hazelnuts, possible pine nuts, figs, dates, grapes, apple and lentils (Davies 2013: 8). Oil from the amphorae may have been used to anoint bodies, and/or as accelerant. Large numbers of iron nails may have been derived from biers, coffins or objects placed on the funeral pyre, and hobnails from shoes and sandals were also recovered. Status and other social distinctions might have been marked by these differences in placement and grave goods – Groups A and B might have been late 1st to early 2nd century AD in date, broadly contemporary with the first fort in Doncaster; whereas Group C might have extended into the 3rd century. Ten single urned and unurned cremation burials were also recorded, and some buried vessels without cremated remains that may have been cenotaph deposits. A probable funeral pyre was separated from the cemetery by an arc of postholes from a fence or screen (ibid.: 11). There were also two inhumation burials on the cemetery site, both fragmentary and poorly preserved, and one with a coin of AD 268–269.
Detailed analyses of the human remains from Waterdale revealed that most cremation deposits only featured a small proportion of the likely bone from the pyre, and many burials were disturbed and truncated (Caffell and Holst 2013: 43), though this might also imply only token quantities of bone were collected from pyres in some cases. Small quantities of cremated animal bone were also noted, presumably from offerings placed on pyres. Due to the generally very high temperatures it is unlikely that aDNA will have survived. There was a mix of adolescents and adults, with three out of 32 features (or 9.4%) containing definitively identifiable non-adults under 18 years of age, and the same proportion (9.4%) of identifiable adults (ibid.: 51). Of the two inhumations, one was an adult aged 35+ years, whilst the other was an older adolescent or young adult (Caffell and Holst 2013: 51). It was not possible to determine the biological sex of any of the cremated or inhumed remains.
There were several unusual features of the burials. The sherds from unburnt, non‐burial groups likely to come from associated commemorative rites included decorated samian (Leary 2013: 74), yet decorated samian is generally uncommon in cemeteries (Cool and Leary 2012: 313). Amphorae were generally scarce in other Doncaster contexts so the presence of several at Waterdale emphasises the ‘exotic’ and potentially higher-status nature of the offerings, along with the glass unguentaria. It was rare in Roman-period cemeteries associated with rural and small urban settlements for glass vessels to be placed on pyres (Cool 2013: 89-90) and might indicate that one of the groups using the cemetery was military in origin. The pyre goods included at least six small copper-alloy hooked fittings that have only previously been found in numbers at Caerleon and from the Walbrook stream in London (ibid.: 90). These results are all tremendously important at a national as well as a local and regional level and so although an excellently detailed archive-standard report on the excavations has been produced, it needs to be fully published.
Outside of Doncaster, a Roman gypsum inhumation burial might have been destroyed during construction of the railway line near Newton in 1908 (Buckland 1986: 49), and a cremation of a 12–13-year-old child was found within a locally produced Blaxton grey ware lid-seated jar at Barnsley Road, Scawsby (Buckland and Magilton 1986: 60, 183; Nellist 1986: 200). At Adwick-le-Street, Roman pottery and human skeletal remains were disturbed by workmen on a construction site. At least four inhumations were present, one possibly extended originally in a coffin or shroud; and another in a flexed position, possibly with its hands and feet cut off and removed peri- or post-mortem (Buckland 1986: 36, fig. 21; Buckland and Magilton 1986: 2147). A cremation burial might also have been present, along with a hearth or pyre site. Much of the site was destroyed without any archaeological recording possible. The five Roman tombstones recovered from Templeborough (May 1922; see above) were re-used as masonry in later contexts, so were no longer associated with any burials.
Occasional cremation burials have been encountered elsewhere. One at Cantley was found close to some of the pottery kilns (Annable 1954), although it may possibly have pre-dated these; and a small quantity of un-urned cremated bone was discovered in a ditch that pre-dated the Roman road at Redhouse Farm, Adwick-leStreet (Upson-Smith 2002: 14). Three truncated pits containing extremely small quantities of cremated remains were excavated in Area A at West Moor Park, Armthorpe, within one of the small enclosures – both human bone and animal bone might have been present, along with pyre debris (Clough 2006: 24-5). This phase of archaeological work at Armthorpe has never been fully published, however. A deposit of cremated human bone from an enclosure ditch was excavated as part of the i-Port development at Potteric Carr near Doncaster, and returned a radiocarbon date of AD 50–220 (Daniel, Harrison and Powell 2014b: 16, 20, fig. 6) – this is likely to be within the Romano-British part of that range. The remains were of an adult and were wellsealed by silts and apparently undisturbed – it might therefore have been deposited in a small bag of leather or cloth. The very small quantity of bone recovered suggests that only a token amount of bone was taken from the pyre, and animal bone from possible sheep or goat was also identified.
At Billingley Drive, Thurnscoe, five large, flat-bottomed rectangular pits excavated alongside the enclosure ditches might perhaps have been late Iron Age or Romano-British graves, where human bone did not survive the acidic soil conditions (Neal and Fraser 2004: 24). In the future, great care should be taken to identify such possible grave cuts and investigate them as thoroughly as possible – even a few fragments of teeth, bone or hobnails might indicate an inhumation used to be present.
Even allowing for inimical geologies and soils, and taphonomic factors and later disturbance, it seems that cremation burials and inhumations were either uncommon on or around rural settlements or have not survived for other unknown reasons. If only token amounts of bone were taken from cremation pyres and were deposited in unurned burials without any accompanying grave goods, then the bone might easily have become dispersed or small shallow features truncated by centuries of later cultivation. Where Iron Age and Romano-British inhumations have been identified in single graves or small groups of burials as in West Yorkshire, there are still many people missing from the burial record. This is even more apparent in South Yorkshire. Perhaps bodies were indeed exposed in some way or placed in rivers and lakes; or were taken away from settlements to still unknown cemeteries. It could also be that most people in South Yorkshire had a genuinely different burial rite or sets of practices that have left little archaeological trace. In Doncaster, there is still insufficient data to identify any real patterning in cemetery organisation according to cremation and inhumation, age, gender and identity, and social status, though the Waterdale evidence does suggest the presence of several slightly different ‘Romanised’ social groups, perhaps military and civilian individuals and families.
A human skull was found in a pit at Site DG in Frenchgate, Doncaster, along with large quantities of samian and coarsewares and seeds including apple and grape pips (Buckland and Magilton 1986: 48); but the skull might have been derived from an earlier feature. Another fragmented but substantially complete skull of a young male minus its lower jar was found in a pit at 8–10 High Street, and a human femoral shaft fragment in another pit (Chadwick and Burgess 2008: 34). The disarticulated human bones at Rossington Bridge have already been noted above.
Was it through excarnation on platforms or in trees, or were bodies placed on the ground surface? Were bodies placed in rivers and lakes? Why were only some people apparently buried during the Iron Age, and who were these individuals?
Were rural people still following Iron Age disposal practices?
The soils of the region are generally poor at preserving organic remains. The alkaline soils on Magnesian Limestone areas of South Yorkshire allow some preservation, but the better quality palaeo-environmental information has come from low-lying wetland sites. The Coal Measures geology is still largely a blank area, but there is some limited evidence for the Pennine uplands. There remains a problem with resolution – some site-specific palaeo-environmental evidence from South Yorkshire is good, as is some of the wider regional data; but intermediate sequences suitable for describing the entire county are lacking. The information below inevitably draws on additional evidence from further north in West Yorkshire, further west in Derbyshire, and further south in the Trent Valley of Nottinghamshire.
The first extensive woodland clearance in South Yorkshire appears to have taken place during the Bronze Age, with charcoal layers at Thorne Moor indicative of burning and an associated increase in grassland and perhaps pastoralism (Smith 2002: 35-6). There was alder and birch carr on much of north and central Thorne Moor (Chapman and Gearey 2013: 141). Evidence at Leash Fen and Stoke Flat in the Pennines of Derbyshire indicates clearance, grazing and cereal cultivation from the later Bronze Age (Heath 2003: 35; Long, Chambers and Barnatt 1998: 512). To the south-east, peat from River Trent palaeochannels reveals a decline in woodland and a rise in grasses and sedges from 1200–1000 BC, with possible cultivation and pastoralism (Brayshay and Dinnin 1999; Knight and Howard 2004; Smith and Howard 2004). The Rivers Trent, Don, Idle and Torne drained into the Humberhead Levels, which was a shifting mosaic of alder and birch carr, open water and reed swamp, peat bog and raised mires (Van de Noort and Ellis 1997, 1999). These areas may have been seasonally exploited for wildfowl, plant foods, construction materials and bog iron ore; and might also have held significant symbolic meanings.
At Hatfield Moors, there is evidence that small-scale Bronze Age woodland clearance accelerated during the earlier Iron Age (Buckland 1976, 1979; Chapman and Gearey 2013; Dinnin and Whitehouse 1997; Smith 2002). Peat formation in both the Pennine uplands and the raised mires of the Humberhead Levels was probably underway by the late Bronze Age. Pools of standing water and reed swamps in the levels were common, with some heathland on drier areas (Buckland 1976: 208-9; Chapman and Gearey 2013: 138; Smith 2002: 43). There was possibly increased run-off of calcareous water from Magnesian Limestone areas (Buckland 1976: 212; 1979: 63; Smith 2002: 91), perhaps linked to clearance and cultivation.
This increased clearance and peat formation in upland mires, lowland mires and river valleys was possibly exacerbated by a climatic downturn across north-western Europe between c. 1000–800 BC, perhaps indicating a change from a more continental to a more oceanic if warmer climatic regime (P. Buckland pers. comm.). There may have been marine transgression sea levels and thus also rising inland water tables in the East Anglian Fens and Humberhead Levels from around 500 BC onwards (Bell 1996; Dinnin, Ellis and Weir 1997). The exact nature and scale of these large-scale changes continues to be a matter of intense debate. The wetter, colder conditions were once linked to volcanic eruptions or even cometary showers (e.g. Baillie 1991; C. Burgess 1985, 1989), and were regarded as causing the abandonment of upland settlements, famine and population collapse, large-scale population movements to the lowlands, and a concomitant increase in conflict between social groups that lead to the creation of linear earthworks and hillforts.
Such overly environmentally deterministic arguments have been criticised, however (e.g. Buckland, Dugmore and Edwards 1997), and although the nature and scale of human occupation might have changed, there does not seem to have been a wholesale abandonment of upland areas (Cockrell 2016; Tipping 2002; Young and Simmonds 1995). There is evidence of some level of widespread change with colder, wetter conditions; but the mechanisms behind this and the impact upon human communities is still disputed, in some cases by researchers using data from the same regions (e.g. Amesbury et al. 2008; Armit et al. 2014; Turney et al. 2016).
At Thorne Moors there were increases in indicators of grassy environments recorded from c. 870–540 BC, and human activities do not appear to have significantly diminished there or across the Humberhead Levels, with evidence for pastoralism and arable agriculture (Smith 2002: 36, 48-9). Low-lying areas and river valleys in South Yorkshire at sites such as Sutton Common and Balby Carr were probably still dominated by alder carr, with some willow, birch, hazel and oak, and wet grassland (Boardman 1997: 245-247; Gearey 2007: 62-64; Greig 2007: 13; Roper and Whitehouse 1997: 244). Many valley bottoms and low-lying areas had large pools or lakes of shallow standing water called meres during winter and spring, but these formed lush pastures in summer and autumn. Areas such as Potteric Carr and Balby Carr would have had such meres until the later Iron Age and Romano-British periods when areas were reclaimed and drained by ditches to leave wet and dry meadows.
At Balby Carr there was evidence for wetter environments with carr woodland of alder and willow, water-filled ditches, open wet ground of swamp/fen and some standing water with sedges, rushes and bur-reed. There were drier areas of waste ground, hedgerow, heath and grassland with hazel and hawthorn, sloe, bramble and wild rose, and heather and bracken. There were wet meadows with sweet-grass and buttercup, and drier pasture, with a low-level of arable cultivation of barley (Hordeum), and emmer and spelt wheat (Triticum dicoccum/spelta) on higher, drier ground (Greig 2007: 34-6; Smith and Tetlow 2007: 44-5; Wyles 2016: 14-16).
During the Iron Age and Roman periods, much of the eastern lowland part of South Yorkshire would have consisted of open landscapes with pasture and arable fields interspersed with small copses of managed woodland (Buckland 1986: 4; Greig 2007: 36). Plants associated with hedges may have helped define some ditched boundaries. The introduction of hay cropping may have taken place after the Roman occupation, with no firm evidence of it beforehand (Greig 1984; M. Jones 1996; Lambrick 1992). Stands of wildwood might have remained on steeper hillsides and upland areas, but most tree cover had probably disappeared by the earlier Iron Age. Woodland management was probably undertaken through plot-felling, with managed stands coppiced in identifiable cycles (Buckland 1986: 4). Rod fragments of ash, and worked round wood or boards of oak, alder, beech and willow were found at Balby Carr (Allen 2005; Gale 2007; Hall et al. 2005). Mather (1991) recorded evidence of widespread woodland clearance during the Romano-British period near Rossington Bridge, with an expansion of grassland accompanied by the presence of barley, oats or wheat and possibly hemp. Much of this may have been driven by the demands of the military for timber, but also the pottery industries for fuel.
This evidence broadly corresponds with that derived from more recent investigations at Potteric Carr as part of the FARRRS and i-Port developments (e.g. Daniel 2017: 34, 37, 38-9; Daniel, Harrison and Powell 2014a: 18-20, 2014b: 18-19). In addition to cereal remains (see below), hazelnut, apple, and sloe fragments and an almost complete tuber of pignut were recovered from the waterholes of later Iron Age and Romano-British date. The weed assemblage was indicative of grassland, field margins and arable environments (Daniel 2017: 33-5). Wood charcoal included oak, alder, hazel, willow or poplar, wild cherry, hawthorn and ash. Much of this, from branches or coppiced stems, was probably remains of fuel waste from domestic hearths. Charred stems of grasses, sedges and occasionally heather found in some of the waterholes probably relate to the burning of turves (Daniel, Harrison and Powell 2014b: 18). Waterlogged remains of alder and willow were found in waterholes and boundary ditches. Damp and waterlogged conditions suggested by the alder, willow and poplar were further indicated by crowfoot, water pepper and marsh pennywort that grow in shallow margins of ponds, ditches and slow-flowing streams; along with evidence for blinks, spikerush, rushes and sedges, caddisfly larvae case fragments, and a single egg case of water flea.
Remains of thistle, creeping buttercup, and grasses indicated grazed grassland in the vicinity. Knotgrass, chickweed, nettles, nightshade and other indicators of disturbed or nitrogen-rich soils were also in evidence, along with bracken. Remains of beetles and mites were also found (Daniel 2017: 41) but have not yet been fully analysed. Pollen assemblages of the waterholes were dominated by oak, hazel, alder and birch, elm, lime and holly; either from residual primary woodland or more likely secondary woodland/scrub growth after a phase of clearance. Pollen from a ditch at FARRRS indicated a more open landscape with wet pasture and cultivation, but at a very local level a change from wet, grass-sedge vegetation to alder carr woodland. The ditches, well and waterholes at Finningley Quarry produced oak, alder and hazel charcoal, waterlogged remains of elder and bramble, and beetle remains that are also not fully analysed or published (Alldritt 2006a: 123).
This all indicates a mosaic of vegetation in an essentially open landscape with grazing on low-lying areas of wet grassland and cultivation on slightly higher, drier ground, as well as for the presence of heathland. Samples taken from the Scaftworth Roman fortlet also indicated that away from the alder and willow carr woodland on the River Idle floodplain, the landscape was essentially open and farmed during the Romano-British period (Gilbertson and Blackham 1985) with a mixed arable and pastoral regime. For Britain as a whole, the climate became warmer and drier from the later Iron Age onwards (Lamb 1981: 62-63; Simmons 2001: 53). Even on higher ground, however, water tables have probably been lowered in the past few hundred years due to post-medieval and early modern drainage, and there would have been more springs and minor becks (Berg 2001: 4), especially on hillsides.
On the eastern limestone areas of West Yorkshire there were probably extensive open grasslands, with evidence for ploughing and arable cropping (Long and Tipping 2001: 225; Richardson 2001: 248). The South Yorkshire evidence is less clear as there has been a general lack of published data from the smaller number of excavations on Magnesian Limestone areas, and several excavated sites unfortunately remain unpublished, such as the investigations around Brodsworth by the Universities of Sheffield and Hull. The plant remains recorded at Barnburgh Hall suggested cultivation and weeds of disturbed and open cultivated land, but only indications of cereal consumption rather than processing (Young and Alldritt 2005). Few plant and snail remains were recovered from Redhouse Farm, Adwick-le-Street, with little potential for palaeo-environmental information (Deighton 2004: 25).
There are limited results even from more intensive sampling of sites on Coal Measures geology and soils. At Pastures Road, Mexborough, analyses suggested some cultivation and access to woodland, peat, and heathland resources (Alldritt 2008: 27). At Roebuck Hill, Jump, there was also evidence for cultivation and weeds of an arable, relatively open landscape with some woodland, scrub or hedgerows featuring blackberries, hazelnuts and sloes (Schmidl, Jacques and Gardner 2007: 56-8). Some pine charcoal present might also be indicative of sandier heathland. Only a few charred grains of barley and spelt wheat were found at Whirlow Hall Farm, which might indicate only limited crop processing; whilst remains of weeds such as buttercup and burnet suggested open meadow or pasture environments (McLellan 2017: 114). Large charcoal concentrations consisted principally of mature oak wood and hazel branch wood, probably linked to industrial activities such as the metalworking evidenced at the site. The hazel branches may have been used for kindling and the mature oak for stoking fires to high and consistent temperatures.
Across Britain, the generally warmer and drier climatic conditions of the later Iron Age and earlier Roman period may have begun to deteriorate again during the 3rd to 4th centuries AD (Knight and Howard 2004: 116; Lamb 1981: 62-63; Simmons 2001: 53). Along the Rivers Trent, Don and Idle and within the Humberhead Levels there were episodes of flooding and alluviation, perhaps caused by loss of vegetation cover and/or more intensive agriculture with deeper ploughing and additional winter cropping, leading to higher surface run-off and soil loss (Buckland and Sadler 1985; Dinnin 1997; Knight and Howard 2004; Macklin 1999). There may also have been marine transgression in the Humberhead Levels (Dinnin 1997: 32; Van de Noort and Davies 1993: 18). By the late 3rd and 4th centuries AD, this was leading to increased flooding and alluviation. The anthropogenic processes causing or at least contributing to these regional trends are likely to have included further woodland clearance and increased cultivation, perhaps exacerbated by deep-ploughing techniques capable of severing root mats, and the sowing of winter as well as summer crops (Knight, Howard and Leary 2004: 120). Some areas of previous field system and settlement in low-lying areas such as Potteric Car may have been abandoned (Daniel 2017: 52). Field system ditches silted up and these areas abandoned once again, until drainage and reclamation schemes during the recent historic period.
More evidence from South Yorkshire is needed to investigate the cultivation of new crops, changes in the scale of agricultural production and changes in crop husbandry practices during the Iron Age and Romano-British periods and how these changes vary in different regions (Carruthers and Hunter Dowse 2019, 97-99; Hall and Huntley 2007, 93; Van der Veen 2007, 204). The continuing cultivation of emmer wheat during the Iron Age in some areas (Carruthers and Hunter Dowse 2019, 207) and the status of bread wheat in the Romano-British agricultural economy (Carruthers and Hunter Dowse 2019, 98; Hall and Huntley 2007, 93; Van der Veen 2016, 811), are current areas of archaeobotanical research. Finds of more than the occasional specimen of these crops are therefore a priority for analysis and dating. Large assemblages of rye and oats are also a priority for analysis and dating, in order to investigate the cultivation of these crops in areas of Britain with acidic sandy soils, which were less suitable for the cultivation of other cereals (Carruthers and Hunter Dowse 2019, 210; Robinson 2002, 108). The origin of the domestic apple is a current area of research and charred or waterlogged remains of apple from Romano-British sites should be considered for DNA analysis (Carruthers and Hunter Dowse 2019, 100; Van der Veen 2007, 205).
Rich assemblages of charred wild or weed plant seeds are a priority for analysis, to investigate changes in crop husbandry practices during the Iron Age and Romano-British periods such as the expansion of agriculture on to new soils, the cultivation of soils depleted of nutrients, the use of manuring to improve soil fertility and the use of fallow as part of a crop rotation regime (Carruthers and Hunter Dowse 2019, 97-98; Van der Veen 2007, 204). The appearance of stinking mayweed (Anthemis cotula) in the archaeobotanical record is significant as this is a common medieval crop weed which is rarely present in archaeobotanical samples dated earlier than the late Iron Age and indicates expansion of agriculture onto heavy clay soils (Jones 1981, 111; Jones 1988, 90).
The recovery of waterlogged plant remains is also a priority, particularly from rural sites, to provide evidence for fruit, vegetables, herbs, medicinal plants, oil rich seeds and exotic/imported foods during the Romano-British period (Carruthers and Hunter Dowse 2019, 100; Van der Veen 2007, 193). Wells, latrine pits, and deep pits/ditches are a priority for sampling, as are sites with deep stratigraphy where anoxic conditions may result in the preservation of uncharred plant material in the absence of full waterlogging (Van der Veen 2007, 193).
Rich assemblages of wood charcoal are needed to investigate the use of fuel in industrial processes such as pottery manufacture and metal working and compare these with fuel used in domestic contexts (Huntley 2010, 63). Smelting and kiln sites are therefore a priority for sampling, alongside hearths, ovens, and other domestic contexts. Cremation deposits are a priority for sampling to investigate the selection of wood for funerary pyres (Huntley 2010, 63). Charcoal analysis may also provide evidence for imported or newly introduced trees during the Romano-British period, such as sweet chestnut and walnut.
Larger numbers of charred plant macrofossil samples from Iron Age and Romano-British sites are also necessary to provide good quality data which can be used to make reliable inferences regarding agricultural practices or diet (Van der Veen 2007, 195). For example, larger numbers of samples are needed to investigate regional differences in the maximum density of charred remains per litre of soil and therefore investigate the role of agriculture in the Iron Age and Romano-British arable economy of different regions (Monckton 2006, 271). Low density assemblages of charred crop material may be related to the small scale of arable production, the use of crop processing waste for other purposes such as fodder or to an arable economy based more on pastoralism than arable agriculture. Investigation of the role of pastoralism in the arable economy can also be supported by evidence from other sources such as animal bones, pollen, molluscs and insects (Carruthers and Hunter Dowse 2019, 101).
A sufficient number of charred plant macrofossil samples from each phase, feature type or area of the site need to be taken in order to provide evidence for patterning within the assemblage (Carruthers and Hunter Dowse 2019, 47 Van der Veen 2007, 195). It is important that negative evidence is recorded, to understand differences in plant use in different areas of a settlement or between settlements (Van der Veen 2007, 204). Material should be collected from different areas within a context to obtain a representative sample (Campbell et al 2011, 9). Multiple samples should be taken from occupation deposits to explore variation in different areas of the deposit (ibid., 9). Samples need to be 40-60 litres in size (ibid., 12) to achieve recovery of at least 100 specimens of charred plant material, which is necessary for statistical analysis (Van der Veen 2007, 203). Processing of large sample volumes for the recovery of charred plant macrofossils is also more likely to produce sufficient charcoal fragments to provide a representative sample of woody taxa utilised for fuel. For urban Roman sites where the concentration of charred remains is more likely to be high, smaller samples of 20 litres may be sufficient (Carruthers and Hunter Dowse 2019, 10). Sorting of the heavy residues of flotation samples can also be useful for the recovery of small artefacts such as hammer scale or beads or organic material such as molluscs or fish bone (Campbell et al 2011, 12). Charred plant remains such as cereal grain, hazelnut shell and small diameter wood charcoal fragments, which are preserved in acidic soils (ibid., 5), also provide dating evidence which may not be available from other material such as bone, which is poorly preserved at sites on the acidic soils of the Coal Measures.
How did human communities in South Yorkshire respond to the possible amelioration of climate during the later Iron Age, and the potential deterioration in the later Roman period?
What was the impact upon alluviation and colluviation, and variations between different areas? Did the scale, character and types of crops/stocks change (not necessarily in association with the conquest)?
During the Iron Age, emmer and spelt wheat were important crops, the latter becoming more prevalent during the Romano-British period (Greig 1984; Jones 1996; van der Veen 1992). Bread wheat and club wheat also became more significant during the Roman period. Barley was important in northern Britain, and oats and rye were also harvested, though it is unclear if these were deliberately cultivated. Flax, peas, beans, brome, vetch and fat hen were either cultivated or tolerated, and the latter two may have provided leafy greens and animal fodder. Other plants and trees were used for food, fodder, medicines, and to provide fibres.
Some earlier authors claimed that after the Roman conquest of northern Britain the limited local indigenous cultivation was abandoned, and grain was imported from the south (e.g. Branigan 1984: 30; Seaward 1976: 22-23). Such assertions are now untenable in view of the evidence for arable production identified from across northern Britain (Haselgrove 1984; Huntley and Stallibrass 1995; Topping, Halliday and Welfare 1989; van der Veen 1992). Taphonomic factors are also important. In the past cereal remains were used to try and identify ‘producer’ or ‘consumer’ sites (e.g. Jones 1996), but this is a potentially flawed approach. Significant charred grain-rich deposits on excavated sites might represent large-scale production and/or consumption episodes such as feasting, rather than the relative contribution of cereals to everyday diets (van der Veen and Jones 2006, 2007). It is also feasible that some apparent dumps of plant remains were deliberate placed deposits.
Although evidence for arable cultivation in South Yorkshire during the Iron Age and Romano-British period is hampered by poor preservation, it was clearly taking place. Much of the evidence up to 2008 was outlined by Chadwick (2008: chapter 4, appendix A). Pollen and insect studies (see above) demonstrates cultivation by the later Bronze Age, expanding in significance during the Iron Age (Smith 2002; Smith and Howard 2004). The evidence of saddle and rotary beehive querns and four-post possible granary structures for the physical processing of cereals and other plants has been outlined above. Spelt, emmer and hulled barley were recovered from Sutton Common in the 1980s and 1990s investigations (Boardman and Charles 1997: 248-9). Fat hen and brome were also present. The 2001-2003 excavations of postpipes and postholes associated with the four-post structures contained charred grain, whole spikelets and detached spikelet forks of barley, spelt and emmer, originally from a cleaned crop (Hall and Kenward 2007: 126; Van de Noort 2007b: 133). Some pit and ditch fills also contained charred spelt, emmer and barley grains and spikelets (Hall and Kenward 2007: 129-30). There was some cereal pollen in soil samples (Geary 2007: 64), but although there was a transition from a partially wooded to an open landscape during the occupation of the site, there was little direct evidence for cultivation in the immediate vicinity of Sutton Common. No cereal remains at all were identified from Croft Road, Finningley Quarry (Alldritt 2006a).
At Balby Carr, grains and spikelets of hulled wheat, emmer or spelt, barley and a small fragment of possible oat have been recovered, though the latter could be a wild species (Alldritt 2006b; Hall et al. 2005; Wyles and Challinor 2016). Pollen analysis detected a small but consistent record of cereal cultivation, though in the wider area rather than the immediate vicinity of the settlement (Greig 2007: 35). Large quantities of barley and emmer or spelt on the Firstpoint site represented the dehusking of hulled grain stored as semi-cleaned grain or in spikelet form (Wyles and Challinor 2016: 15), and a late Iron Age radiocarbon date of 197–44 BC was obtained from hazelnut shells associated with a dump of charred grain. This is further clear evidence for arable agriculture nearby, with cereals being brought to the Balby Carr settlement and processed, stored and consumed (Daniel 2016: 21).
At Topham Farm, Sykehouse, probable six-row hulled barley grains were associated with the Structure 7 roundhouse, and no weeds or chaff were recovered. There was a single oat grain from the enigmatic Structure 5 (Richardson 2003: 26). At Redhouse Farm, Adwick-le-Street, an Iron Age and Romano-British enclosure (Enclosure E1) only produced two unidentified charred cereal grains, but six-row hulled barley, spelt, oat/rye and other indeterminate cereal grains were recovered from trackway ditches and Romano-British enclosures (Deighton 2002: 28). Fat hen was also noted, and a small pulse (Leguminosae). Weeds included species of arable or disturbed ground. Some of the deposits seemed to be cleaned grain, but others contained chaff and were the by-products of processing. At Billingley Drive, Thurnscoe, over 90% of the charred grain recovered was wheat, mostly unidentifiable but with some spelt and emmer, although a small number of rounded grains were from free-threshing bread wheat. There was also some six-row hulled barley (6%), and traces of oats and rye (Giorgi 2004: 64, 68). It was not clear if the oats were cultivated, or wild. Chaff was present in many samples, but especially those from a Romano-British phase oven or corn drier. There were also charred seeds of vetch or pea, and hazelnut shells. Wild species included weeds of disturbed ground with cleavers, an autumn germinating weed, indicating autumn sowing, probably of spelt. Other potential edible species were sorrel, brome and docks.
At High Street, Shafton, most of the cereal grains were derived from deposits within two excavated Romano-British ovens. Most of the cereal was probably emmer, with some spelt and six-row hulled barley. There were significant amounts of chaff, with glume bases from wheat and rachis internodes from barley (Young 2001). There were also many weeds of cultivated and/or disturbed land, in addition to brome, and legumes that might have been pea, Celtic bean or vetch. Fat hen was nearly as frequent as grain. The relatively large amounts of chaff and weed seeds suggested the enclosure could be a so-called ‘producer’ site, although no querns were excavated; and both emmer and spelt could have been imported and stored when only partly processed. Some pits and four-post structures were excavated on the site (Burgess 2001a). Excavations at Roebuck Hill, Jump, Barnsley recorded small quantities of emmer and spelt wheat, barley, oats and rye grains; and weeds of disturbed or cultivated ground (Schmidl, Jacques and Gardner 2007: 57-58). At Scawthorpe, excavation of a double-ditched trackway and an associated enclosure produced some barley grains and emmer chaff, brome, and a seed of pea or vetch (Akeret et al. 2004: 32-3). This lack of evidence from these sites might reflect a genuine absence of crop-processing practices at them in the past; or could be merely due to the acidic soil conditions.
More recent results include sites excavated as part of the FARRRS and i-Port developments near Doncaster. Five waterholes produced particularly useful palaeo-environmental evidence. Analyses of charred remains from early Romano-British features identified hulled wheat, emmer or spelt, and some barley; with some free-threshing wheat as grains but particularly as chaff (Daniel 2017: 33-4; Daniel, Harrison and Powell 2014b: 18-19). Weed seeds indicate grassland, field margins and arable environments, and include seeds of oat/brome, brassica, and vetch/wild pea, though the latter could also have been utilised for food. Waterlogged remains from the waterholes included some further cereal grains but also wild fruits such as apple and hazelnuts, and evidence for grazed grassland.
What was the impact upon alluviation and colluviation, and variations between different areas? Did the scale, character and types of crops/stocks change (not necessarily in association with the conquest)?
Can geoarchaeological or palaeo-environmental evidence be found that Iron Age and Romano-British communities spread midden material or manure onto land, or were animals grazed on stubble after harvests and over winters? Is there evidence for other management practices such as marling or liming? Were there any identifiable changes during the Romano-British period?
The geologies and soils over large parts of South Yorkshire are too acidic for bone preservation, and even on Magnesian Limestone bone condition may still be very poor, hindering identification, ageing and sexing of animal remains. Bones from neonates, juveniles and smaller species rarely survive; and yet such information is vital to considerations of past husbandry and butchery practices. Most excavations of Iron Age and Romano-British rural sites in the wider Yorkshire region produce less than 1000 bone fragments (Richardson 2001), but such samples are thought too small for meaningful statistical analyses (Hambleton 1999; Huntley and Stallibrass 1995). Animal bone was often not retained or analysed during older excavations. Variations in analyses and data presentation mean that even recently published assemblages may not be directly comparable. An overview of faunal assemblages from northern England ignored South Yorkshire (Huntley and Stallibrass 1995), whilst a study of British Iron Age animal remains did not examine any sites in South or West Yorkshire; as samples were considered too small (Hambleton 1999: 16).
Cattle, sheep and pigs were the main livestock in Iron Age Britain. Goats are rarely identified, as their remains are too similar to those of sheep. It was once thought that sheep were the most numerically important animal, kept for milk and wool (Grant 1984), but an analysis of Iron Age faunal assemblages from around Britain found most had roughly equal counts of sheep and cattle (Hambleton 1999). Cattle were probably exploited for milk, traction, breeding stock and for their manure (Albarella 2007; Maltby 1996). Northern bone assemblages are generally more varied than those in south-central England, with cattle and sheep ranging from 20–70% for both species (Hambleton 1999: 47). This reflected a broader range of husbandry practices. On most Iron Age sites there was a low incidence of pigs, but higher percentages of pigs on ‘high-status’ Iron Age settlements and in East Yorkshire Iron Age burials suggests that pork was a delicacy (Maltby 1996; Parker Pearson 1999).
Iconography suggests a regard for wild boar, deer, and more rarely, horses and domestic cattle, especially bulls (Green 1992). Only small quantities of deer, hare and other wild animal remains are found on Iron Age sites (Grant 1981; Hambleton 1999), suggesting that hunting was infrequent, and/or that there were social prohibitions on certain species. Horses might have had connotations of long-distance movement, speed and hunting, and required considerable time and resources to breed, train and equip them. This would have given them high status associations. Dogs often occupy liminal positions within human societies and may be reviled as much as they are appreciated. Such ambiguous and complex beliefs seem to have afforded the special treatment of their remains during the Iron Age and Romano-British periods (q.v. Black 1983; Serpell 1995; Smith 2006). The dog remains frequently found in Romano-British wells might have reflected the potential role of dogs as intercessors with the gods; and perhaps also referenced Cerberus, guardian of the underworld (Woodward and Woodward 2004: 78).
Romano-British faunal assemblages usually feature lower proportions of sheep and higher percentages of cattle and pigs than Iron Age remains (Albarella 2007: 396-7; Grant 1989: 136; Hambleton 1999: 44; King 1991: 17), the latter perhaps due to increased pork consumption. Livestock and domesticates gradually increased in size during the Roman period, particularly cattle and horses but also sheep, pigs and dogs – though some Roman ‘lapdogs’ were considerably smaller than Iron Age hounds. These changes were a consequence of importing new breeding stock and greater mobility of livestock, though might not have become pronounced in some areas until the 3rd century AD (Dobney 2001; Minniti et al. 2014; Rizzetto, Crabtree and Albarella 2017). Slight increases in height and changes in horn core shapes in Romano-British cattle might nonetheless have signified marked variations in appearance compared to native cattle, with differently coloured coats, smooth rather than longer hair, and different temperaments, milking qualities or productivity (Stallibrass 2000: 69-70). Some indigenous farmers might have regarded such introductions with resentment or disdain, others with enthusiasm.
A limited survey of faunal assemblages from northern England, mostly from military and urban sites, indicates the presence of small numbers of particularly large, non-native cattle throughout the Roman period (Dobney 2001: 39). These could have been draught animals used to pull heavy wagons, especially for the Roman military. Cattle dominated Roman military faunal assemblages in Britain, with animals killed at the prime meat age of 3-4 years (Cool 2006: 82-84; Dobney 2001: 37; King 1999: 189). Military sites and urban centres imported most of their cattle as adult beasts, either as carcasses or live animals. More sheep were killed when sub-adult or adult in the Romano-British period than during the Iron Age, which might imply that meat and wool production were emphasised. Particularly in urban centres and forts, slaughtering patterns and butchery techniques may have changed considerably following the Roman conquest. These may have included the introduction of cleavers, the hanging of large joints for curing or storage, and the production of smaller portions as ‘snack foods’ (Cool 2006: 89-91; Dobney 2001: 39-41; King 1984: 214, 1991: 17; Meadows 1997: 26-7). Wool also supposedly became finer, and the appearance of donkeys, mules and new breeds of horse, dog and domestic fowl again suggest an increasing interest in animal breeding (Grant 1989: 146).
Faunal analysts quantify remains as NISP (Number of Identifiable Specimens), and as MNI (Minimum Number of Individuals). The latter is usually more informative. The known evidence for animal husbandry and butchery in Iron Age and Roman-period South Yorkshire up to 2008 was again summarised by Chadwick (2008: chapters 5-6, appendix C). Most Iron Age rural sites have produced little or no animal remains at all. Only 385 animal bone fragments from stratified contexts were recovered at Sutton Common, and even despite the wholly inadequate excavation and sampling strategy, the bone was heavily fragmented and in poor condition. In terms of the NISP values cattle formed 34.9%, sheep/goats 61.7%, pig 1.8%, horse 1%, roe deer 0.6% and red deer 0.3% (Outram 2007: 139). The apparent high proportion of sheep/goat is unusual when compared to the later Iron Age sites, but there may be many taphonomic and contextual biases in the limited data. The age profiles of the sheep/goat remains suggested most were slaughtered for meat (ibid.: 141).
At Pickburn Leys approximately 1200 animal bone fragments were recovered, with no complete bones (Berg 1985: 11). Although these were not quantified, cattle formed the majority of animals, followed by sheep/goat, with some evidence of pig and horse. Skull and foot bones were probably disposed of and buried relatively quickly. At Hazel Lane Quarry, Hampole, the fragmented and weathered bones of sheep/goat and pigs were identified, but the majority of remains were from juvenile cattle. This might reflect the maintenance of particular herd structures, and even the culling of young male animals (O’Neill and Brown 1999: 108).
At Topham Farm, Sykehouse only 374 animal bone fragments were recovered, and preservation biased larger bones and robust teeth from bigger animals (Richardson 2003: 26-7). In the late Iron Age Phase 1, identifiable remains produced NISP values of 37% for cattle, 3% sheep/goat, 2% pig, 5% unidentified ‘cattle sized’ animals, and 53% of ‘sheep sized’ animals. In the late Iron Age/Romano-British Phase 2, cattle were 21% of NISP counts, horse 9%, sheep/goat 1%, pig 1%, cattle sized animals 29%, and sheep sized animals 39%. There was slightly better bone preservation at Balby Carr. Of 539 bone fragments from one AS WYAS project, cattle and ‘cattle-sized’ animals accounted for 73% of the assemblage NISP (Richardson 2005b), though this was biased because of a partial cattle skeleton or ABG buried in a ditch. A partially articulated group of sheep/goat bones from the First Point site at Balby Carr may also be a placed deposit (Richardson 2006). Sheep/goat were 21% and pig 5% of the bone fragments, with some deer and horse bones also recovered. Most cattle and sheep/goat remains were sub-adult suggesting the slaughter of younger animals for meat, although some older animals were probably retained for secondary products. A largely complete but fragmented cattle skull was found along with other animal bone in the base of a ditch at Balby Carr during the D2 phase of work (Muldowney 2008: 4). This work is not fully published. Further work at the Balby First Point site recovered poorly preserved remains of cattle and sheep/goat in almost equal proportions (Higbee 2016: 13). There was also one unstratified long bone from a horse. The inner enclosure ditch produced a small group of immature sheep/goats, and the skull and jaw of a dog which returned a radiocarbon date of 358–105 BC – these ABGs are like possible placed deposits excavated elsewhere in Yorkshire (q.v. Chadwick, Martin and Richardson 2013).
More indirect evidence includes dung beetle remains that suggest the large-scale grazing of animals. Remains of beetles and other insect species associated with the dung of large grazing animals were recovered from the waterlogged ditches at Balby Carr (Smith and Tetlow 2007: 44). More detailed data will hopefully come with full analysis and publication of insect assemblages recovered from samples taken from the waterholes and ditches at the various Rossington i-Port and FARRRS sites.
Despite the extensive nature of the FARRRS and i-Port archaeological investigations, comparatively little animal bone was recovered from these projects, with the poorly preserved and fragmented remains numbering in the hundreds rather than thousands. Full NISP and MNI details are not yet available, but the remains were mostly of cattle, with some sheep/goat, and occasional horse and red deer (Daniel 2017: 30-1; Daniel, Harrison and Powell 2014b: 16). Only a few teeth and some burnt and /or calcined bone fragments from cattle and sheep were excavated at West Moor Park II, Armthorpe (Richardson 2008: 51). Similarly, very little animal bone was found at Rossington Grange Farm, and that was from mostly later Roman phases (Richardson 2016: 27). At Billingley Drive, Thurnscoe, bone preservation was extremely poor. Only a few cattle and horse teeth fragments were identified (Gidney 2004: 58-59). No animal bone at all was identified at Pastures Road, Mexborough (Williams and Weston 2008), and a probable Anglo-Saxon inhumation burial on the site had no surviving bone either. Other sites on Coal Measures or Sherwood Sandstone geologies have produced similar negative results.
Assessing the impact of the Roman conquest upon pastoral practices (cf. Allen et al. 2017) is difficult given the problems with preservation. It is possible that the apparent expansion of field systems during the 2nd and 3rd centuries AD in South Yorkshire, especially on the Sherwood Sandstone areas, might have reflected an increase or extensification of pastoral production (Chadwick 2008a, 2008c); but this is hard to prove. One interesting possibility to investigate in future is whether the opening-up of larger areas around Rossington and Doncaster during the Romano-British period for grazing and the development of field systems might have been partly an unplanned consequence of the clearance of woodland initially for timber and fuel.
An exciting avenue for future research concerns isotope and ancient DNA (aDNA) analyses. Although isotope studies undertaken on human remains have been able to provide fascinating insights into the movement, diet and identities of people in the past, these techniques are now being applied to animal remains with often remarkable results (e.g. Gan et al. 2018; Madgwick, Lamb et al. 2019; Madgwick, Lewis et al 2019; Madgwick and Mulville 2015; Minniti et al. 2014; Towers et al 2017; Viner et al. 2010). South Yorkshire has great potential despite the often-poor bone preservation; as the different geological and topographic zones and soils should be conducive to examining whether animals were moved between these different areas. In the future, it might prove possible to determine whether livestock were moved between lowland floodplains in the east and the Magnesian Limestone or Coal Measures hills to the west. The nature of dairying practices could be researched, and the supply of livestock to urban areas such as Doncaster and to Roman military garrisons investigated (q.v. Gan et al. 2018; Madgwick, Lewis et al 2019; Stallibrass 2000, 2009). Some existing archive faunal collections could be utilised for this.
The use of aDNA analyses is providing new evidence for animal domestication, sexing, and origins of recent breeds (e.g. Davis et al. 2012; Flink et al. 2014; MacHugh, Larson and Orlando 2017). Such work is currently expensive, time consuming and often relies on uncontaminated samples being taken in the field; and to date has not been used for many faunal studies in Britain. In future however, it might provide insights not only into animal biogeographies and connections between animals at different settlements, breeding and herd management; but also by extension evidence for links between different human communities. Such work could also inform research into Roman ‘improvement’, the likely physical appearance of Iron Age and Romano-British livestock, and livestock diseases and pathologies.
Can geoarchaeological or palaeo-environmental evidence be found that Iron Age and Romano-British communities spread midden material or manure onto land, or were animals grazed on stubble after harvests and over winters? Is there evidence for other management practices such as marling or liming? Were there any identifiable changes during the Romano-British period?
Would isotope and aDNA analyses of livestock remains shed light on herding practices, trading and the exchange of livestock for breeding? Can we identify dairying and transhumance through such techniques?
Were the military farming their own livestock, or were animals supplied by local producers? What implications does this have for understanding interactions between military and civilian in Roman South Yorkshire?
Can we discern any social or ideological aspects of how particular animals were regarded in the past?
Based on Roman literary sources such as Tacitus in his Agricola of c. AD 98, and Ptolemy’s Geographia written in c. AD 120–160 (Jones and Mattingly 1990: 17); traditional culture-history approaches to the Yorkshire region hold that the native peoples belonging to the South Yorkshire area were part of the Brigantes, a large tribal federation thought to hold sway from the River Don up to Northumberland and County Durham (Hanson and Campbell 1986; Hartley 1980; Hartley and Fitts 1988).
There are many problems with this assumption, not the least being that the ‘Brigantes’ may never have existed as a single identifiable social or political entity and may be more a product of Roman simplifications or misunderstandings (q.v. Collis 2003; James 1999; Jones 1997; Moore 2011). Many small-scale societies do not draw such clear-cut ethnic distinctions, or only do so in times of social stress. The very presence of the Romans following their incursions in 54 and 52 BC and the invasion of AD 43 may have had a galvanising effect on previously loosely connected communities, causing them to assert or even invent a common identity. Indeed, the many variations in settlement form and artefact production and consumption between South Yorkshire and West and North Yorkshire suggest significant social differences between the communities living across these areas. This diversity is also apparent across many regions of Iron Age and Roman Britain (Moore 2011; Smith et al. 2018).
Enclosed earlier Iron Age sites such as Sutton Common and Wincobank do not exhibit characteristics of elite centres such as later Iron Age Stanwick, although they might have been communal foci; and Sutton Common may have been used for the communal storage of agricultural surplus. There was nothing similar in the late Iron Age. An apparent lack of ‘high status’ sites and material culture may not necessarily equate to a less hierarchical society, but it is possible that during much of the Iron Age most rural communities in South Yorkshire were relatively unstratified or markedly hierarchical societies, without a pyramid of power stretching up from a base of farmers to some small social elite (contra Cunliffe 1974: 112, 305; James 1993: 53). Indeed, earlier simplistic models of hierarchical organisation in Iron Age societies have been critiqued (e.g. Hill 1992, 1995a, 1996a). Differences in social status may have been minor, or else not expressed through material expressions of wealth such as larger and more imposing settlements, or finer and more varied metal objects. The numbers and quality of livestock might have instead been used as measures of identity, status and wealth (q.v. Giles 2007b: 244) – perhaps prized beasts were cherished and displayed. There was apparently no distinctive local pottery or burial tradition. Perhaps perishable items such as woven cloth or body tattoos distinguished people from different lineages or clans.
Ties of kinship and clan may have mattered more than modern notions of Brigantian ‘tribal’ identity (q.v. Moore 2011). No coins were apparently manufactured locally, though coinage from the Corieltauvi and other groups made its way into South Yorkshire, and coin moulds were found recently at Scotch Corner in North Yorkshire (Fell 2017). There is still much we do not know about everyday Iron Age societies – if communities were patrilineal or matrilineal, if adult men and women were polygynous or polyandrous, and if marriage partners were patrilocal or matrilocal.
Conventional histories propose that Roman forces moved northwards during the governorship of Didus Gallus in c. AD 48–50 establishing a series of military bases at Newton-on-Trent, Chesterfield, Templeborough, Rossington Bridge and Lincoln sometime during AD 55–65 (Hanson and Campbell 1986; Hartley 1980; cf. Tacitus Annals 12: 31). According to the Annals of Tacitus, Roman troops suppressed unrest in Brigantian territory in AD 48, in AD 54–57, in AD 60 with a campaign carried out by Caesius Nasica, and again in AD 68–69 (Hanson and Campbell 1986: 81-2). The AD 54–57 disturbances may have coincided with the construction of the bases at Templeborough and Rossington Bridge. There is little archaeological evidence and thus no real independent dating for this suggested narrative, nor for the actual line of the early northern frontier. It might have been a fluid border zone, relying on just a few large bases and smaller outposts rather than on fixed defences.
A supposed Brigantian leadership dispute from AD 69 prompted the final Roman invasion of the north under governor Quintus Petillius Cerialis in AD 70–71 (Richmond 1954), although as with 19th century colonial situations it is possible that an invented crisis or a minor incident might have merely been a convenient excuse for a planned long-term strategy. Roman forces might have advanced along one or both of the lines of the later Roman roads – from Lincoln to the River Humber at Old Winteringham and across to Brough-on-Humber (Ermine Street), and northwards to Malton and Newton Kyme; and/or perhaps from Rossington Bridge to Castleford and Roecliffe (Birley 1973; Hoffmann 2013). Forts or fortresses were built at these locales. The winter of AD 71–72 may have seen the consolidation of river crossings and roads with forts and stations established at Brough-on-Noe, Burghwallis, perhaps Long Sandall, York, Adel, Slack, Elslack, Castleshaw, Tadcaster and Ilkley (Buckland 1986; Dearne 1993). Many early sequences are poorly dated, however.
Governor Julius Agricola supposedly campaigned with the army to the north of the region in Scotland until c. AD 84–86, when Long Sandall, Castleford, Brough-on-Humber, York and smaller forts and military stations might have acted as supply bases and centres for the acquisition of crops and livestock, as well as to quell any lingering unrest amongst indigenous groups. Many Roman forts would have continued to maintain and project imperial power through a policy of ‘internal policing’ (James 2002: 37-8). It is also possible that some sites were part of an intensification in the collection of the annonae tax in the later Roman period and formed a network of collection points, as argued for forts in Wales (Arnold and Davies 2000: 28).
There may have been uprisings in the north during the last few years of the Emperor Trajan’s reign in AD 105 and 117 which were suppressed by Quintus Pompeius Falco, before further campaigning under Hadrian and the eventual establishment of a fixed fortified frontier from the River Solway to the River Tyne. The military advanced northwards to the Forth-Clyde line during the AD 140s which led to the building of the relatively short-lived Antonine Wall defences. It has been proposed that during AD 155–7 there was further unrest amongst the Brigantes in northern England that prompted the rebuilding of the forts at Brough-on-Noe, Templeborough, Doncaster and Burghwallis (Buckland 1986; Dearne 1993; Hartley 1980) and the movement of some Roman forces south, though the evidence is mostly circumstantial. Some Scottish lowland forts continued to be garrisoned, but much of the Antonine Wall may have been abandoned by the 160s. There might have been native uprisings and army revolts from AD 175 to 195, and some forts in northern England may once again have been re-occupied to act as supply bases during the campaigns in Scotland led by the Emperor Septimus Severus in AD 208–9. The revolt by the British usurper Carausius, his subsequent overthrow by Allectus and the latter’s defeat by Constantius Chlorus and Julius Asclepiodotus during AD 286–96 might all have caused further changes in the status of military garrisons and buildings.
Smaller ‘fortlets’ were established just outside South Yorkshire, alongside rivers at Thorpe Audlin and at Scaftworth near Bawtry. The latter site by the River Idle was known since the late 18th century, and limited excavation suggests it might have dated to the 4th century (Bartlett and Riley 1958; Dearne 1997). More recent analyses of aerial imagery indicated another probable Roman camp just a few hundred metres to the north-east, and a second fortlet approximately 1.5 kilometres further east (Deegan 2007: fig. 6.22; Roberts, Deegan and Berg 2010: 36-7, 67-8, fig. 48). A small riverside fortlet has also been identified by the River Aire at Roall near Kellington in North Yorkshire (Pastscape 900082). These sites may have originally controlled military river supply routes and later trade, but during the later 3rd and 4th centuries might also have served to deter any raiders coming upriver from the coast.
One problem with such historically led approaches is that archaeological evidence can be forced into dominant and constraining narratives, some based on biased or partial Roman sources. Older accounts used evidence for burnt layers within Roman forts as indicators of attacks and uprisings (e.g. Hartley 1980: 5-6), yet the burning of old and rotting wooden buildings in Roman military bases might have taken place at irregular intervals prior to rebuilding and/or in order to rid sites of insect pests and rats. Similarly, fort rebuilding or refurbishment may have corresponded to higher-level political and military edicts rather than any local unrest. No inscriptions related to rebuilding within forts in Britain mention enemy attacks; and most may have been ceremonial and symbolic in nature reflecting symbolic ideas of renewal rather than responses to specific events (Thomas and Witschel 1992; Welsby 1980). To date there is no independent archaeological and dating evidence for any physical attacks or disturbance in the study region during the Roman occupation.
At Templeborough, the three-phase model of fort construction and occupation originally proposed by Leader (1878) and May (1922) is probably incorrect. The postulated three successive and superimposed phases of occupation between the 1st and 4th centuries AD consisted of an early fort with earth and turf-built ramparts, and stone-faced earthen ramparts in the second and third phases, these being separated by episodes of widespread burning (May 1922: 14-18). The third and smallest phase fort was nebulous, however. Buckland (1986: 30-2) tried to put the somewhat vague ideas of Leader and May into a more secure chronological framework, with a first phase fort occupied from AD 54 until the mid-2nd century when it was abandoned for a time. Later in the 2nd century it was re-built as a slightly smaller base, but the third phase and final abandonment was again somewhat unclear and described by Buckland with understatement as ‘enigmatic’ (ibid.: 32).
The more recent investigations identified two key phases of activity indicated by the pottery recovered from the truncated ditches and the fragmentary vicus features. This scheme has a first phase Flavian–Trajanic fort during the late 1st to early 2nd centuries, and a second Hadrianic–Antonine phase rebuild in the mid- to late 2nd centuries (Davies 2016: 57-8). There is Neronian-period samian pottery from the older investigations which might suggest a construction date of c. AD 55–65 for the first fort, but this could have already been old at the time of deposition and no material of this date was found in more recent excavations. A third and later phase may never have existed. This later history of Templeborough fort, and the nature and full extent of the vicus, remain extremely unclear.
The basic layout of much of the fort was established through the rescue excavations of the 1960s–70s and developer-funded investigations since the 1990s, but much of the internal area and details of the phasing are unknown. The original fort may date to AD 70–1 and the Flavian period, with a possible rebuilding episode in the mid to late 80s AD and again in the early 2nd century (Buckland 1986: 12; Buckland and Magilton n.d.). The defences may have been a single ditch in front of a turf-faced earthen rampart, perhaps with stone added later. This accords with evidence from the Church Walk excavations (Chadwick, Martin and Richardson 2008). It might have been mostly auxiliaries garrisoning these earlier phases of Roman forts at Danum.
Buckland posited a period of abandonment of nearly 30 years until the mid-2nd century, when a smaller fort was constructed within the north-western corner of the earlier base. This may have been occupied until the late 3rd or early 4th century, when a new fort or defended settlement was built with stone ramparts up to 6 metres high and 2.5m thick, and some major stone buildings within its walls (Buckland 1986: 17). The rescue excavations found evidence for a more irregular layout and buildings during the later 4th century, the latter of stone footings with timber and turf or daub walls, and in one case built across an earlier metalled road surface. Considerable quantities of pottery and animal bone in ‘dark earth’ deposits were also recorded. The end years of the fort are not known, though a coin hoard of silver siliquae dated to AD 388 suggests occupation continued into the late 4th or early 5th century.
No detailed plans of the fort excavations during the 1960s–1970s are published, and nor any details of the ceramics and animal bone. The pottery will be particularly crucial for constructing a detailed sequence. Only a draft ‘small finds’ report exists (Lloyd Morgan and Buckland n.d.), and no funding was made available by Doncaster Council or Historic England and its forebears to write up the rescue excavations. A post-excavation report on the Roman fort that was in preparation (Buckland and Magilton n.d.) was unfortunately largely halted due to the death of one of the co-authors John Magilton. There was a council proposal to write up and publish all the different stages of investigations including some of the more recent developer-funded sites, as part of a wider scheme to promote and display the Roman and medieval remains and heritage, but apart from a synthetic study (Pollington 2007) this appears to have stalled with few tangible results. The Doncaster Roman fort investigations need to be published as soon as possible.
The military base at Rossington Bridge was first identified from the air by Keith St Joseph (St Joseph 1969: 104) and was classified as a new form of military site called a ‘vexillation fortress’ (q.v. Frere and St Joseph 1974: 6) which may have held both legionaries and auxiliary troops. It was subsequently photographed and recorded by Riley (1980: 57, 93, map 8). Limited geophysical survey and trenching took place over its north-eastern corner to assess preservation (Head et al. 1997: 275-8), but much of the interior has never been surveyed. Some aerial photographs hinted at possible external outworks to the east (Wilson 1984: 58), and/or at several different phases of fort or fortress – along the southern edge of the fortress an inner ditch appears to be slightly out of alignment with the outer defences, whilst some internal linear features might relate to field systems pre-dating the fortress, or an earlier phase of military construction.
In advance of the construction of a park and ride scheme there was further geophysical survey confined to the area between the B6463 and the line of the A638, including the north-eastern corner of the fortress area (Schofield 2003). Subsequent evaluation and excavation revealed a possible earlier enclosure with a roundhouse and evidence for Iron Age pottery metalworking (Bishop 2010: 10-11, fig. 12; see above), and field system ditches and a trackway on broadly north-south and east-west alignments, slightly different to that of the fortress and with 2nd to 3rd century pottery. Approximately 150m east of the fortress, one evaluation trench recorded a ditch corresponding to a geophysical feature with a rounded corner (possibly a fort ‘playing card’ corner), and this ditch yielded late Iron Age and Roman pottery (ibid.: 12, figs 9-10). This may be associated with an early military feature.
Although an evaluation report and a draft publication report were produced as a result of the work in the early 2000s (Bishop 2005, 2010), an actual publication has never materialised. Further detailed geophysical investigation of the entire Scheduled area of the fortress but especially the interior in the form of magnetometry, soil resistance survey, Ground Penetrating Radar and pXRF survey is long overdue, and could be followed up with targeted excavation. A full supervised, methodical archaeological metal detecting survey of the site to plot and remove any metal finds in the topsoil for safe conservation and analysis would also be useful, as the site has been subject to much illegal detecting over the decades.
Hunter (1831: 487) recorded finds of Roman coins and fibulae from the site at Robin Hood’s Well, Burghwallis, but the fact it was a military station was first recognised from aerial photographs (Buckland 1986: 8-9, fig. 7; Deegan 2007: fig. 6.22; Roberts, Deegan and Berg 2010: 37, 67, 72, figs 48, 91; Wilson 1972: 311). The site has been partly truncated by an old (now backfilled) quarry visible on the 1st Edition Ordnance Survey map; and is also partly overlain by a farm and probably by the modern line of the A1. The site consists of at least two, probably three phases of fort – two with double ditches and of slightly different sizes, and a third smaller fort with only one visible ditch and on a slightly different alignment to the other two. No vicus is visible, and on some aerial photographs the forts are clearly situated at an oblique angle across a series of earlier fields (contra Roberts, Deegan and Berg 2010: 72).
The forts may have been positioned to guard the crossing of a low-lying valley and The Skell watercourse by the Great North Road or Roman Ridge, now the modern A1 (Margary 1973: 415, RR28b). There has been no recorded geophysical survey of the forts, so their exact layout and relationship to one another is unknown; and no excavation so nothing is known of their date. Roman pottery including samian of Antonine date was found in one of two ditches visible in section in a disused quarry at the site (Wilson 1972: 311), and two 3rd century bronze radiate coins of Tetricus I and Carausius along with some pottery were picked up at the fort in 1973 (Pastscape 56125). The site is Scheduled but the field the forts are within still appears to be under cultivation and thus threatened with continued degradation by ploughing. The field must be taken out of cultivation using agri-environmental schemes.
The site is likely to have been subject to much illegal metal detecting. A full methodical archaeological metal detecting survey of the site to plot and remove any metal finds in the topsoil for safe conservation and analysis would also be useful. It is vital that detailed geophysical survey takes place over the site as soon as possible using all three major techniques (magnetometry, soil resistivity and Ground Penetrating Radar), along with pXRF. Targeted excavation should take place to establish the stratigraphic sequence of the different fort features and to recover dateable artefacts or materials and palaeo-environmental samples.
It had been previously proposed that there might have been a Roman fort at this location (e.g. Buckland and Magilton 1986: 208) but it was confirmed through aerial photographic analysis (Deegan 2007: fig. 6.22; Roberts, Deegan and Berg 2010: 36, 67-8, fig. 48). The double, possibly triple-ditched camp was on the southern bank of the River Don north-east of Doncaster, but might originally only have had three sides, the fourth formed by the river. It may have been built to protect a riverine supply route, either for the initial invasion of the north, or during the later Roman occupation – the former might be more likely. Part of the site now lies beneath industrial developments. Detailed survey by Ground Penetrating Radar along with targeted excavation to retrieve dating information should take place as soon as possible.
A square double-ditched enclosure with large internal postholes on Bole Hill near Sheffield investigated through geophysical survey and trial trenching was probably of late 1st to early 2nd century AD date; and might be a timber signal tower overlooking a proposed Roman road between the fort at Brough-on-Noe in Derbyshire (Navio) and Templeborough (Inglis 2016: 60; Waddington 2017: 46-9).
The landscape impact of the construction of Roman roads is clear in parts of West and South Yorkshire, as these cut across the line of many existing field systems at Hook Moor, South Elmsall, Robin Hood’s Well, Burghwallis, Adwick-le-Street, Bramham and Rossington (Riley 1980: 94-5; Roberts, Deegan and Berg 2010: 58, 71-2, fig. 90). Some Roman roads even truncated earlier enclosures as near Hesley Hall, Rossington (Roberts, Deegan and Berg 2010: fig. 75). These roads were projections of Roman imperial power, disrupting familiar routes and routines, ignoring local tenure and tradition, and creating new politics of movement (Chadwick 2016b; Forcey 1997; Witcher 1997). Such roads allowed Roman soldiers and tax officials to penetrate deeper into the countryside, whilst the forced labour of local inhabitants was sometimes used to maintain them (Given 2004: 54). People living alongside roads could have their livestock, wagons and food requisitioned by Roman soldiers or officials, or face demands for hospitality and accommodation. This may well have caused suspicion and resentment. Yet such roads would also have linked families and communities more effectively, allowed livestock and goods to be taken to markets more quickly and efficiently; and would have speeded up the transmission of news. There would thus have been benefits as well as drawbacks. Investigating how Roman roads articulated with pre-existing trackways, fields and settlements can be a focus for future archaeological research; as can examining in detail where routeways and field systems re-orientated themselves to the new regime of movement.
Although the main outlines of most of the principal Roman road routes in South Yorkshire are known (e.g. Roberts, Deegan and Berg 2010: 68, fig. 88), there is much that remains uncertain. Analyses of cropmarks, lidar and geophysical survey data, and the work of the Roman Roads Research Association and Time Travellers organisations has provided valuable information to compliment or correct details of the major surveys undertaken by Margary (1973). The dates and chronologies of each road are still largely unknown, and a biographical approach would be productive. Though some Roman roads might have been initially used by the military for the invasion and occupation of the north, it might have been decades before they had full metalled agger surfaces and any flanking ditches constructed. Conversely, some minor routes might have remained largely as tracks with little or no metalling.
Roman roads are especially beloved of more traditional approaches to Roman archaeology, and arguments about their routes seem to arouse strong passions. It might not be too heretical to propose however that Roman roads in themselves are of limited intrinsic interest. Instead, their archaeological value and the reasons for identifying, investigating or preserving them comes from the information they provide on the development and sequence of landscape features over time, the chronology and tempo of military and official administrative activity, the physical, economic and social connections between communities, and how settlements may have flourished and failed. Some of the major routes in South Yorkshire are outlined below.
The road from Lincoln (Lindum) to Doncaster (Danum) (Margary 1973: RR28a) crossed the River Trent at Littleborough-on-Trent (Segelocum)in Nottinghamshire and then extended north-westwards towards Bawtry where it is partly visible on lidar (http://roadsofromanbritain.org/gazetteer/yorkshire/rr28a), initially spanning the low-lying and marshy land on the River Idle floodplain on a corduroy causeway built of logs, brushwood and turf to the north of the later Roman fortlet (Kennedy 1984; Van de Noort et al. 1997). This was later replaced by a more solid metalled agger surface on a slightly different alignment, reinforced with rows of substantial timber piles to carry it relatively high across the peaty floodplain. One of these piles was dated to the 3rd century AD. After several probable repair episodes, the stone agger ceased to be maintained and eventually collapsed, spreading laterally across the surrounding peat surface. Samples taken from twigs in peat above this spread in order to try and date the abandonment returned an Iron Age date of 180 BC–AD 50 (1σ) or 360 BC–AD 120 (2σ), suggesting that much older material had been redeposited by River Idle flooding (Bayliss et al. 1995: 88-9).
Entering under modern Bawtry somewhere under Queen Street Crescent, much of the alignment of RR28a is today marked by the line of the Great North Road or A638 leading past Gally Hills, King’s Wood and Hurst Plantation (Margary 1973: 410-11), with a possible deviation of 60m close to the Northern Racing College, and passing c. 500m east of the vexillation fortress at Rossington Bridge. It crossed the River Torne to the east of Rossington Bridge and Rossington Bridge Farm. Around Rossington and Cantley there might have been several Roman road junctions, perhaps indicating several different phases of construction or changes in alignments (http://roadsofromanbritain.org/gazetteer/yorkshire/rr28a).
The line of the road to Doncaster was traced north-west by Margary past Bessacarr Grange, along Cantley/Sand Lane and the edge of Doncaster racecourse, and entering Doncaster under the modern A638 (Margary 1973: 411-12). This route proposed by Margary has been criticised for being rather indirect and unlikely (http://roadsofromanbritain.org/gazetteer/yorkshire/rr28a; White 1963). Instead, using evidence from 1st Edition Ordnance Survey maps and lidar, it is suggested that there was a crossroads just north of Rossington and/or at Cantley, with RR28a then a straighter route across Doncaster Carr and under the modern suburb of Belle Vue. This line may be preserved by Roman Road and South Parade in Doncaster, before linking up with Hallgate and High Street, then Frenchgate and a presumed bridge across the River Don (Buckland and Magilton 1986: 30, 49, 55, 208; Chadwick and Burgess 2008: 122).
Another road, the so-called Cantley Spur (Margary 1973: RR281), headed NNW along the line of Warning Tongue Lane, past Cantley Hall, Sandall Beat Wood and Armthorpe, towards the fort long suspected and now confirmed at Long Sandall on the River Don. It is still visible in places as an earthwork (Roberts, Deegan and Berg 2010: 68). This might have even been part of a larger, abandoned scheme to link RR28a from Bawtry northwards to the fort at Roall and across the Humberhead Levels (http://roadsofromanbritain.org/gazetteer/yorkshire/rr28a). Lidar imagery also suggests that there might have been a short spur along the northern edge of the vexillation fortress and linking it to RR28a.
Analysis of aerial photographs has identified a previously unknown probable Roman road nearly four kilometres in length that extended almost due south from the south-eastern corner of Rossington bridge fortress, but which was unknown to Margary or Riley. It too apparently cut across pre-existing field systems and trackways, and the enclosure near Hesley Hall, Rossington (Deegan 2007: 25; Roberts, Deegan and Berg 2010: 68, figs 75, 88, 90). If it continued southward for 5.5km on this same alignment it would have met the crossing of the River Ryton just north of Blyth. This has been numbered as RR282 and may represent an alternate route for RR28a from a different phase, a northwards spur from a road from Lincoln to Templeborough (RR289), or even a route that eventually joined the RR5f Fosse Way to Willoughby (Vernemetum) and Leicester (Ratae Corieltauvorum). This might have extended northwards from the Rossington bridge fortress to join RR28a near Cantley, where an ‘Old road’ marked on the 1854 1st Edition OS map and lidar provide hints of a route (http://roadsofromanbritain.org/gazetteer/yorkshire/rr28a). These various routes were probably constructed in different phases – it is notable that the vexillation fortress at Rossington Bridge was not aligned to the known route of RR28a but appears to be more in line with (and at a possible bend in) the recently discovered RR282. More detailed lidar analyses and targeted excavation coupled with scientific dating techniques are required to resolve some of these questions.
This proposed Roman route north-westwards from Doncaster – the Roman Ridge (Margary 1973: 415-6, RR28b) – also has problematic aspects. It survived as a prominent earthwork feature and was recorded by Daniel Defoe in the 1720s and Codrington in the early 20th century but has since been much denuded where it is not directly underneath the line of the modern A1. It appears to have been laid out as a series of straight but kinked sections skirting the low-lying and boggier ground in many places such as that east of Bentley and Adwick-le-Street.
After leaving Doncaster, the road’s exact crossing place over the River Don is unknown but may have lain under the present A638 and it then extended through Sunnyfields and Scawsby (Buckland and Magilton 1986: 208; Margary 1973: 415), continuing as a tarmacked bridleway and byway. After Green Lane it continues today as an elevated minor lane and causeway past Highfields and Adwick-le-Street, changing direction slightly several times. Several sections excavated across the presumed alignment at this point in advance of cycle route construction found remnants of a road surface at the southern end, but by Hanging Wood the supposed agger seemed to be a low bank some 25 metres east of the Roman Ridge (Tinsley and Pollington 2010). The ‘bank’ might therefore have been a medieval boundary or trackway parallel to the original Roman road. At Red House Farm, Adwick-le-Street, an unpublished excavation by Dorothy Greene in 1957 during dual carriageway construction recorded a well-preserved agger, and pottery from a roadside ditch or burial cist comprised early to mid-2nd century wares (Buckland and Magilton 1986: 193, 195, 208, fig. 42.351-2).
At Redhouse Farm, RR28b was only c. 60m away from an Iron Age and Romano-British enclosure (Area 7 Enclosure E1). Two sections excavated across it in the early 2000s found no surviving original road surface but the makeup layers of the agger were 8–10 metres wide and up to a metre thick. These consisted of layers of crushed stone and clay, with flanking sandy deposits. Beneath these were found different sets of furrows. Some were from pre-road cultivation, but others were deep, stone-filled features that were probably part of Roman road construction practices as outlined by the Roman writer Statius (Meadows and Chapman 2004: 11-15, figs 8-9). The Roman road cut across earlier field ditches, and one of these ditches produced an un-urned and undated cremation burial east of the Roman road (Upson-Smith 2002: 14). This important excavation remains unpublished.
At Robin Hood’s Well near Burghwallis, the modern A1 curves slightly to the east, but it is likely that the line of the Roman road was originally on a straighter alignment (http://roadsofromanbritain.org/gazetteer/yorkshire/rr28b). None of the three possible Roman forts immediately to the east were orientated to the road, which may indicate they pre-date it. How and exactly where the road crossed the low-lying valley and The Skell beck is unknown, though this crossing may lie beneath the modern upgraded A1. After climbing the slope on the other side of the valley, at Barnsdale Bar the Roman road changed orientation to the north-west, where today it mostly lies beneath the line of the modern A639 Roman Ridge – though sometimes it is adjacent to and parallel to the modern route. After crossing a sloping shoulder just north-east of Thorpe Audlin, the RR28a descended into the valley of the River Went.
After the River Went crossing the Roman Ridge changed alignment slightly and extended north-westwards until East Hardwick in West Yorkshire, with the road either under or immediately alongside the A639. It then diverged underneath Sandy Gate Lane and alongside Hundhill Farm and Potwells Farm. A Roman milestone was found at Hundhill Farm, and geophysical survey and excavation at the findspot by the Pontefract and District Archaeological Society in 2002 found the largely plough-removed agger and a roadside ditch (Ferguson 2003; Houlder 2003). The inscription on the milestone has been transcribed as IMPC MANNI OFL[..] RIANO PF AVG INV MP E XXVI, which can be interpreted as ‘Imp[eratori] C[aesari] M[arco] Anni|o Fl[orliano] p(io) f(elici) Aug(usto) Inv[ictus], or “For the Emperor Caesar Marcus Annius Florianus [Pius Felix Augustus], Augustus Undefeated, Miles to York 26” (Tomlin and Hassall 2003: 371). Florianus only ruled for a few months in AD 276.
Further to the north in West Yorkshire, between Micklefield and Aberford, near Hook Moor, sections along and across a steeply embanked section of the Roman Ridge road were excavated by AS WYAS as part of the M1–A1 Link Road project. An enclosure and boundary ditches pre-dating the road and likely to be of later Iron Age date were identified, which were later incorporated into a broadly co-axial field system featuring a prominent NWW–SEE aligned double-ditched trackway up to 10m wide (O’Neill 2001: 108-111, figs 85-6). Cropmarks and geophysical survey indicated that several subrectangular enclosures or annexes were appended to this trackway (Deegan 2001c: 33, fig. 17), although only one was excavated. Due to truncation by later quarry pits only a few internal features were identified within the enclosure, but animal bone, slag, cinder fragments, clay hearth lining and uncharred barley grains were recovered from some of these (O’Neill 2001: 114). The quarry pits were associated with the construction of the Roman road, but the fact that some of them respected the enclosure indicates that this was an extant landscape feature when the Roman engineers arrived. Prior to the laying of the agger of alternating layers of subsoil and crushed limestone obtained from the quarry pits, two linear stone banks were constructed from weathered surface stones which may have formed markers for the road’s route (O’Neill 2001: 114-5, figs 88-9). A similar line of marking out stones was excavated at 8–10 High Street in Doncaster (Chadwick and Burgess 2008: 25-6).
Perhaps the most perplexing putative route is the proposed Roman road between the fort at Brough-on-Noe (Navio) and Templeborough (Margary 1973: 361-2, RR 710b) which once again was suggested as taking a variety of alignments by antiquaries and early archaeologists. Even the initial route eastwards along the Hope Valley has proven problematic, as it was far from clear which side of the river this was on (Preston 1957; Welsh 1984; Wroe 1982). More recent analysis of lidar evidence now suggests that a line extending south of Bamford was most likely (http://roadsofromanbritain.org/gazetteer/yorkshire/rr710b).
One major flaw of the route proposed by Preston (1957) was that the road would have had to ascend through a steep and narrow cleft in Stanage Edge, which has always seemed unlikely – this is more probably a pack-horse route associated with post-medieval millstone quarrying. From there Preston suggested the road passed over Hallam Moors to cross Lodge Moor, and then along a route in front of Weston Park and the University of Sheffield to just below Park Hill where it then supposedly turned north-westwards towards Templeborough and met the Catcliffe to Oldcoates road RR189 headed east (Greene 1955, 1957a; Margary 1973). At Lodge Moor Preston’s investigations did reveal a metalled road surface nearly 9.5 metres wide with flanking ditches, but no real dating evidence (Preston 1957: 332-3). Welsh (1984) identified a possible stretch of Roman road extending from Scraperlow near Hathersage, passed near Carl Wark, ascended onto Burbage Moor and then descended towards Sheffield from Houndkirk Moor and past Ringinglow.
More recently, the broad outline of Welsh’s proposed route has been investigated by the Time Travellers archaeological society, who have analysed lidar imagery and conducted a series of geophysical surveys, earthwork surveys and small-scale excavations (http://roadsofromanbritain.org/gazetteer/yorkshire/rr710b). At Sheep Hill Farm, Ringinglow, excavations revealed a metalled road surface at least 12 metres wide with flanking ditches up to 5m wide and 1.3m deep (Inglis 2016: app. 4). This road seems to have extended as far as Limb Brook, where it might have been overlooked by the possible timber signal or watch tower on Bole Hill, which was approximately 250-300m to the north-east (Waddington 2017).
From the Limb Brook the road has not been traced further but might have crossed Ringinglow Road just west of Firs Farm, and then along the length of footpath that links Cottage Lane to Common Lane. It may then have changed alignment to the bottom of the hill on Whiteley Wood Road, crossing the Porter Brook before climbing to Ranmoor (http://roadsofromanbritain.org/gazetteer/yorkshire/rr710b). it could then have continued in a more easterly direction to descend into Sheffield broadly along the modern line of Fulwood Road, meeting Preston’s preferred route through Sheffield and eventually Greene’s Catcliffe to Oldcoates road. Alternatively, the Roman road may have turned north-east in Darnall and followed the southern bank of the River Don to Templeborough. Leader identified a road extending eastwards from Templeborough to Park Hill in Sheffield (Leader 1878).
Margary suggested a line from Lodge Moor to the former Lodge Moor Hospital, and from there to Hallam head, Sandygate Road, Lydgate Lane, Hallam Lane and then down past the University of Sheffield and Sheffield Museum and Art Gallery (Margary 1973: 362). This part of RR710b is noted on South Yorkshire HER records as dubious, however, and more likely a medieval or post-medieval pack horse route.
It has long been posited that there must have been a Roman road to enable direct movement from Templeborough further north to Castleford (Lagentium) by joining either RR28b or another road (e.g. Bishop 1999: 309), and antiquaries and older generations of archaeologists suggested (and contested) various proposed routes often based on supposition, place names and ambiguous earthwork or excavation evidence (e.g. Codrington 1903: 272; Greene 1950b: 168; Kitson Clark 1931: 186; Margary 1973: 413-4; Whiting 1931: 258-9). It was even provisionally numbered as 18f (x) by Margary and as RRX2 by the Ordnance Survey; and seen as essentially a continuation of RR18e Ryknild/Riknild/Icknield Street.
Aerial photographic analysis identified a c. 600m length of possible Roman road near South and North Elmsall in West Yorkshire (Deegan 2007: 25; Roberts, Deegan and Berg 2010: 70, fig. 90), orientated NNE–SSW and probably cutting obliquely across earlier field boundary ditches. Ahead of the A6201 Hemsworth to A1 Link Road project, a 2m wide section revealed two roadside ditches up to 4.7m wide and 1.22m deep with a central area of wheel rutting. No metalling was identified, and the ditch fills did not produce any dating evidence (Weston 2014: 24, 81, 177, figs 7, 19, plate 7). The projected line of the trackway or road extends to RR28b and the crossing over the River Went at Thorpe Audlin a few kilometres to the north.
This road was proposed as leading from east of the Roman fort at Templeborough and meeting RR18e Ryknild/Riknild/Icknield Street (Margary 1973: 412, RR710c) before extending to Rotherham broadly under the modern A6178. It might have then ascended to higher ground through Eastwood and to Dalton along the line of the modern A630 Doncaster Road, and then passed by or through Thrybergh, Hooton Roberts and Conisbrough, From Conisbrough the route might have broadly followed the southern side of the Don gorge before extending through Church Lane, Warmsworth and Littlemore Lane, Balby into the south-west side of Doncaster, just north of the modern A630. A slightly different route a little further south and mostly under the A640 might also be possible (Buckland and Magilton 1986: 209).
There is no archaeological evidence for this proposed road, however. A possible length of agger was investigated near Oswin Avenue School in Balby (Magilton 1977: 49; South Yorkshire HER 01859/01) but results were less than convincing. No roadworks or construction have revealed any traces of agger along the route. Near Hooton Roberts the A630 cuts across fields of probable medieval and post-medieval origin as parts of the Doncaster Road were constructed as the Tinsley and Doncaster Turnpike of 1764. The antiquity of the proposed route thus cannot be supported.
Another proposed route is Margary’s suggested road from Templeborough to Littleborough-on-Trent (Segelocum) via Catcliffe, Oldcoates and possibly Bawtry (Margary 1973: 414-5, RR189), an extension of the road from Brough-on-Noe RR710b. This was in turn based on excavations ahead of opencast mining that claimed to have discovered metalled Roman road surfaces at Spa House east of Catcliffe up to 4.8 metres wide with an associated stone-lined ditch (Greene and Smedley 1955, 1957). This road lay underneath a late 17th century access road, but otherwise no dating evidence was found. Aerial photographic analyses identified two broadly east-west parallel cropmarks up to 530m long to the east of Firbeck Hall that might be the flanking ditches of a Roman road (Roberts, Deegan and Berg 2010: 70, fig. 88), but these features have not been traced further to the east or west so this route must remain largely conjectural until more research is undertaken.
Mention must be made of RR18e Ryknild/Riknild/Icknield Street (Margary 1973: 12-13). This Roman road extended from the Fosse Way at Bourton-on-the-Water in Gloucestershire and traversed most of the English midlands; and was still a noted routeway in the medieval period. From Little Chester (Derventio) it led to the Roman fort at Chesterfield, but north of there its course is uncertain, and Margary proposed two possible routes, RR18e and RR18ee, the former taking a more direct route and the latter following highest ground a little further to the east (ibid.).
The western route RR18e passed close to Woodhouse, Orgreave and Catcliffe where it supposedly intersected with RR189. This is all conjectural. From Bonewood Moor it then purportedly headed across Brinsworth Common just east of the site of Brinsworth Grange where Dorothy Greene claimed to have located a Roman road leading to the south-east gate of the fort at Templeborough, along with remains of a small Roman town (Greene 1950b, 1957a-c). Although Greene excavated several narrow trenches across stone road surfaces (Greene 1957a: 81-6; Greene and Wakelin 1950: 164-6), much of the line of the roads was established through probing or small test pits, and no artefacts were found in association with them. Some of the drawn sections of packed stones on rammed clay subsoil (Greene 1957a: 82-3) do not have the appearance of layered agger, although other sections are more convincing (Greene and Wakelin 1950: 166-7). This may be an old post-medieval coach road and other minor trackways of medieval or later date, and as Greene’s evidence for her proposed Roman town is unclear and has been discredited, the evidence for RR18e at this locale may also be suspect.
Greene’s evidence for a road approaching the south-west gate of the fort at Templeborough on the north-west side of Brinsworth Common is more plausible. Published plan and sections show excavated surfaces on the same alignment as the south-west gate and the road recorded by Leader and May, and the presence of roadside ditches and slight cambers in the surface is suggestive (Greene 1957b: 286-8, figs. 2-3). Again however, no dateable artefacts were recovered.
The second proposed route RR18ee probably extended from Chesterfield fort towards Eckington and is identifiable on lidar just north of the River Rother (http://roadsofromanbritain.org/gazetteer/yorkshire/rr18e.html), where an agger and a cutting are both visible. The road is visible on aerial photographs near Hagge Farm, where excavation revealed successive phases of metalled surface (Hart 1981: 92-94, plate 2), and later geophysical survey and excavation revealed more of the alignment (Derbyshire HER 13022). North of Eckington the course is unclear, but perhaps led through Beighton and Aughton. From Aughton it may have climbed onto high ground and joined with a possible Roman road traced by Inglis and the Time Travellers near Ulley, thereafter continuing through Aughton and Guilthwaite to Templeborough (Margary 1973: 413). It is odd that a Roman road did not follow the River Rother valley northwards, unless the low-lying ground was too marshy. Clearly, there are problems with the evidence for both routes, and further detailed research is needed.
Much research to date has focused on the Roman roads, but the places where they crossed low-lying valleys and rivers via bridges or fords is also worthy of future study. Although the lines of the corduroy causeway and later timber and stone causeway for RR28a from Lincoln on the River Idle floodplain just north of the later fortlet at Scaftworth was examined, the exact place where the road crossed the river is still unknown and remains of the bridge have not been recorded. Ground Penetrating Radar could be used to try and identify any timber pilings or stone footings below the existing banks. Just to the south, there was a detectorist find below or close to the early modern stone road bridge of a small Roman ‘purse’ hoard of eight 3rd century radiates, recorded in the HER and on Pastscape (Pastscape 321031). This could have been chance loss by a traveller but could also have been an offering prior to or following a river crossing (Bland et al. forthcoming), similar perhaps to the much larger deposits at Piercebridge (Walton 2016).
Despite the subsequent canalisation and dredging of river channels in the post-medieval and early modern periods, the crossing place across the River Don north of the fort at Templeborough, the full alignment of the crossing of the River Torne at Rossington Bridge, the line of the Roman road across the low-lying valley and The Skell at Robin Hood’s Well are all potential sites where geophysical survey and targeted excavation might be undertaken. Much may now lie underneath modern roads such as the A1 and A639, but it would be worth further investigation. Timber piles of unknown date were observed in the River Don near Wheatley (White 1963), and possible fords or crossing places such as one at Sprotbrough would also be worthy of future work.
The Roman forts at Long Sandal and Doncaster (and later the vicus) would have received many of their supplies by boats coming up the River Don, and this might also have been true of Templeborough. Bawtry was an important inland medieval port (Cumberpatch and Dunkley 1996), and so the fortlets and camps at Scaftworth were also probably part of a riverine supply route. No Roman wharves have ever been identified in South Yorkshire, however.
At Doncaster one location of such wharves could have been beneath the area under or around Doncaster College. Due to time pressures and budgetary constraints, the North Bridge/Low Fishergate excavations only proceeded to a depth of c. 3.5m below modern ground level, or roughly 5.8–6.6m above Ordnance Datum (McComish et al. 2010: 87). The ground surface was continually raised during the medieval and post-medieval periods by successive phases of building and demolition. Riverside timber revetments and a stone-surfaced draw-dock with stone walls were recorded, but these were medieval. Residual Roman material included coins, glass fragments and brick or tile. A 1988 borehole recorded ‘concrete, ash and brick fragments’ at depths of between 3.57 and 3.87m AOD (ibid.: 86) – the ‘concrete’ may be opus signinum. Roman-period deposits might also have been sealed by a 0.30m thick layer of alluvial silt representing late Roman or post-Roman flooding episodes (Buckland, Magilton and Hayfield 1989: 15). Residual Roman pottery was also found during the Doncaster Waterfront investigation near Doncaster College roughly 250m to the north-east. This work again found evidence of medieval waterfront structures, with residual Roman pottery suggesting features of this period might be present below unexcavated layers of alluvium (Brown 2004, Trench I).
The Romans in South Yorkshire may have attempted some large-scale engineering schemes as they possibly did in East Anglia and on the Gwent and Somerset Levels – though the nature and extent of the latter are still unclear. The canalised course of the River Don north of Thorne and the eastwards change in course of the River Idle to the River Trent at Stockwith created the Turnbrigge Dike and Bickers or Bycarrsdike canals (Buckland 1986: 40-42; Gaunt 1975; P. Jones 1995). These channels are recorded as having been cleaned out during the medieval period even if subsequently re-used by Dutch drainage engineers in the post-medieval period, and so may be earlier in date. Due to the later drainage and scouring work, however, attempting to date the first such digging would be extremely difficult.
Sampling for such evidence would have to be supported by high-resolution AMS dates and Bayesian modelling;
It has been argued that some forts such as Burghwallis appear to have been placed in landscape areas with few pre-existing fields and enclosures (Roberts, Deegan and Berg 2010: 72), yet historic aerial photographs held in Doncaster Museum suggest that the fort and road were superimposed obliquely across earlier boundaries, these since quarried away (Buckland 1986: 8, P. Buckland pers. comm). The fortress at Rossington also appears to have been constructed at an angle across earlier field system boundaries
Were existing settlements providing the forts directly, or was land use changed to enable construction/provisioning?
It has previously been assumed, on the basis of May’s excavation, that Templeborough was founded during Gallus’s campaign of c. 51-2 (Buckland 1986, 30). Recent work has cast doubt on this date (Leary 2016), although this was in a fairly restricted area and so we should still be asking when the fort at Templeborough was founded and what its context was in the history of the Roman conquest of the region. Interrogation of the archive may be of value, but further excavation is a priority.
Buckland (1986, 12) dates the foundation of Doncaster to Cerialis’s campaign (c. 70-71), but in light of limited excavation, largely on the defences, this appears to be far from certain and it may be later, perhaps replacing Long Sandall and belonging to the governorship of Agricola (c. 78-84). Further work on the excavation archives as well as excavation is needed to address the question of the date at which Doncaster Roman fort was founded.
Priorities and implementation
Approaches to the study of Roman Britain have varied enormously over the years. By and large, and with rare exceptions, the antiquaries and early archaeologists were drawn from the upper class landed gentry and upper middle-class professionals. Their views of archaeology and the Roman Empire were inevitably shaped by the wider political and social milieu of the British Empire within which they lived (Freeman 1996; Hingley 1996, 2000; Scott 1993).
“The old theory of an age of despotism and decay has been overthrown, and the believer in human nature can now feel confident that, whatever their limitations, the men [sic] of the Empire wrought for the betterment and the happiness of the world.
Their efforts took two forms. They defended the frontiers against the barbarians and secured internal peace; they developed the civilization of the provinces during that peace… […] The long peace made possible the second and more lasting achievement of the Empire. The lands which the legions sheltered were not merely blessed with quiet. They were also given a civilization, and that civilization had time to take strong root.” (Haverfield 1923: 11, my parentheses).
The Roman occupation and administration of Britain and north-west Europe was thus regarded as a form of largely benign authoritarianism, perhaps seen as similar in many respects to the perceived moral obligation of the ‘white man’s burden’ (Brantlinger 2007; Hingley 2000; Kipling 1899). The growth of urbanism and civic society, long-distance metalled roads and communications, trade and commerce, a monetised economy, aqueducts, underfloor heating, sanitation and increased literacy were all viewed rather uncritically as self-evident benefits of the imperial system. This had strong contemporary resonances with members of the Victorian and Edwardian educated classes who themselves sought to promulgate ‘orderly and coherent culture’ (Haverfield 1923: 11) across the globe.
Unlike the development of the study of prehistory in Britain and especially in Europe; Roman archaeology was traditionally often seen as an adjunct to Classics, where archaeology was used mostly to support or refute Roman written sources. The accounts of Roman writers were generally given credence, however; and the various social, ethnic, political and textual biases and distortions often ignored or under-played. Tacitus’ Agricola was partly a work of hagiography for his father-in-law for example. These learned men (and it was usually men, with the exception of notable women such as Tessa Verney Wheeler, Kathleen Kenyon, Mary Kitson Clark and Dorothy Greene) were often from ex-military or civil service backgrounds and tended to focus on Romano-British urban sites, large Roman-style buildings such as villas and bathhouses, and the activities and sites of the Roman military. To some extent echoes of these views persist to this day across Europe, with the sometimes overly obsessive study of military sites and equipment and of Roman roads. Much paper has been produced and ink printed concerning fortifications of Hadrian’s Wall and the Limes on the Rhine and Danube for example. The typologies of artefacts such as samian pottery often seemed more important than the social context of the production, consumption and discard of such material culture. Research such as the Roman Rural Economy Project at Oxford portrays the Roman Empire almost in proto-capitalist terms as a rational economic market economy (Taylor 2013: 3).
The legacy of these often-implicit perspectives continues to affect how Roman Britain and especially northern Britain is viewed by many archaeologists (Robbins 1999; Webster 1999). Such approaches sometimes forget that the Romans were not proto-Victorians but were an Iron Age people too. Some may have had writing, roads and rectangular buildings but they were different in many respects to us today, not the least because of the ways in which seemingly irrational aspects of ritual and religious behaviour were prevalent throughout everyday life. We should also not let celebration of the many achievements of the Roman Empire lead us to disregard institutionalised slavery and the use of warfare and conquest as a means of economic growth.
One highly pertinent series of academic studies in the late 1970s and 1980s focused on post-colonial approaches – the social, cultural and political and economic processes and legacies of colonialism and imperialism, both from the perspective of colonised people and their lands but also investigating the effects of imperialism and colonisation on the colonisers too. Edward Said (1978, 1993), Gayatri Chakravorty Spivak (1987) and Homi Bhabha (1994) are amongst the most influential of these writers and cultural critics. They have also been concerned with acculturation and hybridisation – the creation of ‘hybrid’, ‘creole’ or ‘syncretic’ cultures through what they see as active, two-way processes of social change. Colonising cultures are inevitably changed by the process of colonisation, and new cultural forms and identities emerge. The notion of subaltern voices is also important to many of these authors – the views of those who are systematically denied representation due to race and ethnicity, class and caste, gender, or economic and political hegemony by one dominant group or culture over another.
These approaches have been effectively utilised by archaeologists in North America, Australasia, Asia and Africa to examine the impact of colonial societies upon indigenous communities, how colonists were themselves changed by these encounters, and how slavery and racism were legitimated or resisted through material culture and architecture. Such ideas have also proved attractive to those studying the Roman Empire, and the complex social changes and creation of new hybrid identities that followed the Roman conquest of north-west Europe and Britain.
Some more traditional archaeologists had proposed that Rome encouraged the indigenous elites in Britain to adopt the material culture, practices and customs of Roman culture, along with Roman administrative institutions and a Latinised language (e.g. Frere 1987; Haverfield 1923). Through processes of emulation, the ‘trickle-down’ of material culture and acceptance of the self-evident benefits of Roman rule and culture, the whole of Roman Britain gradually adopted a new identity to greater or lesser degrees. Authors such as Vinogradoff (1911) and Reece (1988) disagreed, arguing that for many Roman culture remained a ‘veneer’ over existing native beliefs and practices, and only the social elites ever really became fully ‘Romanised’ as they had the most to gain politically and economically by doing so.
Millett (1990) suggested that the conquered population actively participated in this process of Romanisation, as the Roman conquest offered some opportunities for individual social advancement outside of previous social hierarchies. Provincial elites also deliberately adopted Roman cultural trappings to distinguish themselves from others in society and perpetuate their own power and authority, and this was emulated by those in lower social positions. Hingley (1997b, 2005) disagreed, and suggested that native Britons adopted ‘Roman’ items and practices as part of their own complex social relations, and to make their own informed statements about personal power and identity. There was also opposition and resistance to Roman rule and Roman cultural hegemony, both overt but also more subtle and expressed through material culture and social practice. Different British social groups and different regions would have all reacted in highly diverse and complex ways, and Roman culture and artefacts allowed them to create new identities.
Many authors have now stressed that the emergence of regional Romano-British and Gallo-Roman cultures and the creation of new languages, material culture forms and social identities was a result of complex and dynamic two-way processes (e.g. Creighton 2005; Grahame 1998; Mattingly 2006; Revell 2008; Taylor 2013; Webster 2001; Woolf 1995, 1998). These arguments move beyond simplistic notions of ‘native’ and ‘Roman’, ‘conqueror’ and ‘colonised’. Indigenous British elites did not adopt Roman material culture and practices simply to remain in position or gain social advantages, merely as a mask or ‘veneer’. Similarly, people within communities did not simply ape their conquerors or their own leaders but made a series of choices, some unconscious but others deliberate and knowledgeable. There was of course imperial might and control, taxation and a degree of cultural hegemony (Forcey 1997; Given 2004; Hanson 1997). But there was also cultural resistance to Roman culture and social practices, and these were also re-interpreted and negotiated to form novel identities and ways of living. ‘Roman’ soldiers, administrators, merchants and colonists settled and/or married in Britain, and some of them became influenced by the communities they were living amongst. Some indigenous religious beliefs and ritual practices remained the same, native and Roman deities and practices merged, and new forms of belief and practice were adopted or emerged (Webster 1997b).
In Britain, highly diverse Iron Age societies were conquered by a Roman military that was ethnically diverse, and a hybrid series of regional Romano-British communities gradually developed as a result, incorporating different elements from indigenous groups and colonising peoples. This affected language, identity, material culture, agricultural and industrial practices, ritual and religion.
During the period of c. 15 years that the northern frontier extended through Derbyshire, north Nottinghamshire and South Yorkshire, there would have been extremely complex two-way social, economic, political and military interactions taking place along the border. There would probably have been Roman diplomatic missions to meet and bribe local leaders, mapping and spying expeditions, and small-scale military raids for punitive reasons after raids, theft or livestock rustling, or to back allied factions. Simultaneously, members of indigenous communities would have been trying to manipulate Roman forces to further their own interests, drawing them into their own internal politics; and there may well have been pro- and anti-Roman factions. There may have been official gifts and trade in both directions, and sexual relationships between Roman troops and locals – officially tolerated, illicit and passionate, or violent and abusive. Early Roman artefacts in South Yorkshire might have included the Aucissa-type brooch of c. AD 40–60 found at Scabba Wood (Merrony et al. 2017: 52, 55, fig. 30), and a mica-rich dish of the later 1st century AD found at the base of a waterhole at Potteric Carr (Daniel 2017: 17), though the latter were also found in Flavian and later contexts in Doncaster (P. Buckland pers. comm). It is interesting to speculate if such objects were diplomatic or personal gifts, dowries, loot from raids, or items stolen from Roman stations.
The impact of the invasion of the north on indigenous peoples in Yorkshire is unknown. Even for people with first or second-hand knowledge of the Roman army, the march of legions through their land must have had profound social and psychological impacts. Any armed resistance would have been crushed, warriors killed or executed, and any captives sold into slavery; but to date there is no archaeological evidence for any widespread destruction of buildings and enclosures at this time, an AD 70–71 ‘scorched earth’ horizon. It may be that the acculturation that had occurred along the frontier in the past in the South Yorkshire area meant there was little appetite for overt resistance to overwhelming Roman military might, and in any case, there were few sites capable of withstanding serious assault by troops. Even where there was no resistance, however, it is likely that livestock would have been ‘appropriated’ or killed; and stored or standing crops stolen or burnt. This after all was a common feature of 18th and 19th century colonial situations in Africa, Asia, North America and Australasia. Turf would have been stripped from pastures to help build fort ramparts and causeways of roads, and many woods and copses would have been cut down for the prodigious quantities of timber required to construct forts and bridges and provide fuel for military ovens, kilns and smithies (Hanson 1978; Reece 1997). This probably violated local rights of tenure and deprived indigenous communities of these resources for many years, if not permanently.
Many Roman troops were drawn from far-flung regions of the Roman Empire (q.v. Swan 1992, 2009) – memorial stones from Templeborough in South Yorkshire record the cohors IV Gallorum (May 1922: 127), and the remains of the shield found at Doncaster may also be of Gallic origin (Buckland 1978: 260). Tiles found at Templeborough and Doncaster were stamped Legio IX Hispania but need not have reflected the presence of a garrison drawn from this legion (Buckland 1986: 12). The bronze Stannington Diploma (Buckland 1986: 38-40, fig. 23; Hunter 1819: 18-20) was presented in AD 124 to a discharged auxiliary soldier from the 1st cohort of Sunuci, a tribe thought to have inhabited the Roman province of Germania Inferior between the rivers Meuse and Ruhr. These non-Italian men would have had their own social, military and unit cultures, and potentially quite different dynamics with local people. Some Roman soldiers, particularly those of more senior rank, brought their own families, servants and slaves to live with them, sometimes within the walls of the forts (Allison 2006: 18-19; James 2001: 83, 2002: 42-3; van Driel-Murray 1995: 9-10). Others would have found mistresses and wives in Britain when they arrived – one of the inscriptions at Templeborough recorded a woman from the Dobunni (May 1922: 130), a tribe thought to have inhabited the Upper Severn area of south-west England and parts of Wales. Her widower husband Excingus was probably from southern Gaul. The administrators, colonists, merchants, freedmen and slaves were similarly have been drawn from many parts of the empire and would have had correspondingly varied identities, beliefs and social practices.
Initial native responses to the Roman occupation after AD 71 might have included confusion, anger and fear, even various forms of social and cultural resistance (Hingley 1997b); but the occupation also brought the potential to construct or renegotiate new identities for individuals and communities, perhaps circumventing previous social structures and any restrictive communal conventions. For some people, smaller social networks centred less on lineages or tribal identities and more on individual households might have become more important over time. For others, traditional kinship links and allegiances probably remained important. Stresses and opportunities created by the Roman occupation might have crosscut existing ties and social obligations. These would have been far more complex than the rather generic categories of ‘villa owners’ and ‘farmers’ that used to feature in discussions of Romano-British people, and more recent work has highlighted more complicated changes (q.v. Hill 2001; Reece 1988; Smith et al. 2018).
In South Yorkshire, many hand-made wares in traditional fabrics and essentially Iron Age styles probably continued to be locally produced into the 2nd century AD (C. Cumberpatch and R. Leary pers. comms.). This has interesting implications for discussions of acculturation and ‘Romanisation’. Did this reflect a relative lack of pre-Roman pottery use amongst most households, and this lack of interest in Roman products when these initially became available; or was it even some form of subconscious or explicit cultural resistance? It is interesting that this is one part of Britain that was almost aceramic during the early medieval period too.
The widespread adoption of Roman pottery did not take place until the early to mid-2nd century AD, although its use was often still limited on many rural settlements. There was a predominance of jars in most Romano-British ceramic assemblages, followed by bowls and dishes. Deep bowls may have been used for stews (Leary, Williams et al. 2008). Sooting is often found on the outside of the vessels, which is typical of many Romano-British rural sites (e.g. Cool 2006; Evans 1993; Robbins 2000). This suggests that pottery continued to be used mainly for cooking and storing food, although greyware bowl forms may have gradually replaced any wooden vessels used for eating. Sooting was often most pronounced on pot rims, suggesting that the bases of vessels were imbedded in ash accumulated within hearths (Cool 2006: 39). Some Iron Age traditions of food preparation and consumption therefore seem to have continued in rural areas. The use of platters, plates, cups, flagons, beakers and other items of Roman-style material culture for the presentation and consumption of food seem to have been confined mostly to Templeborough, Doncaster and a few more ‘Romanised’ settlements. Yet other quite profound changes in diet and drinking might have left little tangible evidence.
One notable social change was the appearance of numeracy and literacy, though the extent of this is unclear, and the latter at least was probably rare in the countryside. Some people might have been capable of little more than marking and reading their own names, but this nevertheless could have had important implications for notions of self-identity and how people were perceived by others. There is some interesting evidence for this, aside from the inscriptions on the Templeborough funerary monuments, the Doncaster altar and the Stannington Diploma. At Templeborough, lead weights from the ‘Commandant’s House’ carried numbers in the form of units of the libra commercial standard (May 1922: 76-9, plate xviii), and at Doncaster graffito on pottery included personal names of Latin and Celtic derivation (Buckland and Magilton 1986: 115, fig. 27; Leary, Williams et al. 2008: 65). Seal-boxes and a copper-alloy stylus have also been identified (Cool 2008b: 142; Lloyd Morgan 1986: 87, fig. 19.8). Two samian inkwells were excavated at 8–10 High Street in Doncaster, and still had traces of ink within them (Ward and Dickinson 2008: 186).
The Magnesian Limestone belt of West and South Yorkshire and parts of Nottinghamshire include some unusual settlement forms, such as the enclosure at Scratta Wood with stone-built roundhouses built onto or actually keyed into the enclosure wall (Challis and Harding 1975, 94; White 1966, n.d.), more reminiscent of northern enclosures in Cumbria and Northumberland. The area around Edlington, Cadeby and Sprotbrough also had several similar stone-walled enclosures which differed in many respects to most Romano-British settlements on the Sherwood Sandstone and Coal Measures areas. This area has also produced a concentration of coin Roman hoards and finds of Iron Age and Roman metalwork items such as brooches, cosmetic grinders, wine strainers and other relatively ‘Romanised’ or higher-status items; which might suggest some slightly different social practices or even identities. Whatever the reasons behind the deposition of individual objects or hoards, their presence in such large numbers must be explained. Peter Robinson of Doncaster Museum and Art Gallery has proposed (pers. comm.) that the Cadeby-Edlington-Sprotbrough area was used to settle discharged military veterans from the Roman army, and this is one avenue of research that could be pursued in future.
The organisation and form of Romano-British communities and settlements in the late 4th century and early 5th century AD is still largely unknown. They faced many difficulties from widespread political, economic and climate changes. The eventual arrival of Anglo-Saxon and then Scandinavian individuals, communities and material culture would have presented further challenges for culture and identity. The apparent resilience and accommodation by these communities in the face of so many transformations, however, may have some lessons for us today.
There is a major problem in South Yorkshire with the production and dissemination of developer-funded archaeological client reports – so-called ‘grey literature’, although this problem is not confined to South Yorkshire (e.g. Holbrook and Morton 2008), nor to commercial reports. Some reports may take years to be submitted to the HER, and these are often interim publications or assessment reports rather than full reports with final detailed specialist analyses. Assessment reports need to be produced much more rapidly after fieldwork has been completed, to ensure that funding remains available and that the further analyses identified can occur. These unpublished reports can be used very effectively in academic research (Chadwick 2008a; Hodgson 2012), and thus should be made as accessible as possible.
Although when finally submitted unpublished client reports may be accessed at the South Yorkshire Archaeology Service (SYAS) Historic Environment Record (HER), the consultation of many such reports by researchers can be a protracted and at times difficult process (Hodgson 2012: 39). Curatorial archaeology staff face increased time and budgetary constraints, and even with GIS-based mapping it is not always obvious how many reports might exist for certain areas. In the future, developer-funded client projects and reports should be available digitally to all researchers, either by supplying copies to the Archaeology Data Service (ADS) or by HERs having digital copies that can be downloaded from a web-based HER, as with the Know Your Place website at Bristol. University-led research projects and the work of independent or ‘amateur’ groups and individuals should also abide by these requirements. Time-lags may be necessary in a minority of cases to consider issues of client confidentiality and commercial sensitivity, but there should still be a clear, unambiguous onus on the wider dissemination of reports. Although on-line publication of full reports should be standard, it must not be a full substitute for academic presentation in monographs or archaeological journals.
Another key consideration concerns long-running projects such as quarries or industrial estates where there are often many different phases of work (aerial photograph analysis and desk-based assessments, geophysical survey, evaluation, excavation and watching briefs) undertaken by more than one archaeological organisation and commissioned by more than one developer. This was the case at Armthorpe, where work was undertaken by SYAU, Archaeological Services WYAS and OA North; and at Balby Carr, with work carried out by BUFAU, ARCUS, Archaeological Services WYAS, AOC Archaeology and Wessex Archaeology. Three of these field units no longer exist. It is obviously desirable to have these different phases integrated into one overall report and final publication on those landscapes, but how would such a synthesis be funded and who would undertake the work? One commercial unit could be selected to write up the results from all the different investigations, including those undertaken by other organisations, but they would need ring-fenced funding in order to undertake this.
Minimum standards of data presentation and publication need to be maintained by SYAS for developer-funded archaeology reports within South Yorkshire, building on but extending and improving Chartered Institute for Archaeologists (CIfA) and Historic England guidelines (e.g. CIfA 2014: 9), although some unit’s reports and publications do not match even these rather minimal standards. Those carrying out research-led projects must also be held accountable to these same standards.
As outlined above, one key area of recent research into the Iron Age and Romano-British periods concerns artefact, animal bone and burnt stone deposition in and around settlements (e.g. Chadwick 2004, 2008a, 2015; Chadwick, Martin and Richardson 2013; Willis 1997). Understanding the spatial patterning of such practices and any changes over time is important to understanding social life within these past communities. Within evaluation and excavation reports, the location of so-called ‘small finds’ such as quernstones, brooches, bracelets, whole or substantially complete pottery vessels, human and animal burials and possible identified placed deposits all need to be shown in relation to recorded features, as distribution plots and volumetric analyses (Chadwick 2009: 142-3; Haselgrove et al. 2001: 10, 15). The use of digital technology both on-site and during post-excavation such as Total Stations, differential GPS and AutoCAD and Illustrator would mean that such two-dimensional distribution plots would take relatively little time to generate. These could form part of more extensive archive reports published on ADS.
This form of post-excavation analysis and presentation was undertaken for the late Iron Age and Romano-British enclosure at Scrooby Top in north Nottinghamshire (Davies et al. 2000). Statistical and spatial analyses revealed that that the majority of Romano-British fine wares by weight and sherd count were deposited in the northern part of the enclosure ditch, whereas most coarse wares, sooted pottery sherds and the greatest quantities of burnt and heat-shattered stone were all found in the south and south-eastern parts of the enclosure ditch (Robbins 2000). This not only indicated a focus of cooking and heating activities in the southern part of the enclosure and a possible domestic emphasis on eating and discard to the north, but also that patterns of deposition varied according to material type and perhaps reflected underlying social attitudes to material culture or even cosmological beliefs of the inhabitants. Generating similar histograms of frequency and weight for excavated ditch segments, pits and other features would be relatively simple. Similarly, discussions of assemblages should be more contextually based and less rigorously divided into specialist categories (Haselgrove et al. 2001: 10).
There are some issues even when sites are fully published. For articles submitted to journals such as the Yorkshire Archaeological Journal, there is a tendency to reduce specialist reports to minimal summaries, and to not publish enough photographs or illustrations of excavated features and artefacts. Full artefact and palaeo-environmental data and analyses may also be missing. This is done to save space, when the cost of paying for page numbers is a consideration.
For example, whilst it was excellent that the enclosures and fields excavated at Rossington Grange Farm were published promptly in the YAJ (Roberts and Weston 2016), there are only illustrations of the Bronze Age cremation urns and two of the seven querns recovered, with no Roman pottery or ‘small finds’ illustrated at all. Although much Roman pottery was standardised and is relatively well-known, photographs and some drawings would still have been useful, especially of the waster vessels. The so-called ‘small finds’ included two whetstones, and a near complete two-link iron snaffle bit (Cool et al 2016), only the second such item recorded from South Yorkshire – the other has been published and illustrated (Cowgill 2004a: 49-50, fig. 27). The Rossington Grange example may have formed part of a possible placed deposit but was not even shown as a photograph, either as a conserved item or in situ on-site. The snaffle bit was from a pit that included quern fragments, but only a select list of some quern fragments was included, and no section or detailed plan of the pit. It is thus impossible from the YAJ publication to glean any idea of the depositional context of the snaffle bit and links with other objects and materials. Such quantitative and contextual data is essential, however, in allowing comparisons between finds assemblages and issues of spatial organisation, discard and structured deposition (Haselgrove et al. 2001: 10). More comprehensive reports with full illustrations and photographs could be published online with ADS, and then in more summary form in YAJ and other journals.
There is also an unfortunate problem with archaeology in South Yorkshire concerning university departments, local societies and commercial field units failing to publish their work. In the case of developer-funded projects, in many instances the original project supervisors have now left the companies or retired from archaeology altogether, so they cannot be blamed for a subsequent failure to publish. It is likely that in many instances, it is the original clients who have withheld or refused funding for full post-excavation and publication, often once planning permission has been granted and on-site excavation work has ceased. Some businesses might subsequently have become insolvent and been liquidated, but several extant firms number amongst the largest and wealthiest developers in Britain, however. Other developers are local authority bodies, which is especially galling.
Key South Yorkshire Iron Age and Romano-British sites that have not yet been fully published or still languish as interim and assessment reports as of 2019 include:
Full publication of all these sites should be an urgent priority. Pressure from the South Yorkshire Archaeology Service, Historic England, CIfA and legal advisers should be brought to bear upon field units or their clients, so that post-excavation analysis and publication can proceed as soon as possible. If the original developer/client has gone into liquidation, then funding should be sought from Historic England to allow these sites to be written up and published as fully as possible and placed on ADS. This dire situation cannot be allowed to continue, especially when the reports from more recent investigations have already been published. The many different phases of work undertaken by ARCUS, AOC Archaeology Ltd, AS WYAS and Wessex Archaeology at Balby Carr need to be fully written up and published as one synthetic volume. A recent useful summary (Daniel and Barclay 2016) nonetheless omitted the results of the ARCUS and AOC Archaeology work. One unit (either AS WYAS or Wessex) should be nominated (or tender) to do this, and funding sought from existing developers and Historic England.
Educational and outreach projects such as the Romans on the Don and Ancestors of the Don Gorge initiatives (Bevan 2006b; Bevan and Ho 2006), funded through the Aggregates Sustainability Levy Fund, should be encouraged and supported. In addition to full academic publication but never a substitute for it, ‘popular’ publication in booklets and on the Internet should also be undertaken to bring the results of work to the attention of a wider, non-specialist readership.
A series of potential research projects have been identified within this document that would be ideal opportunities for university departments, postgraduate researchers and/or independent or so-called ‘amateur’ archaeology groups and local societies to undertake fieldwork and other research, with support and collaborative assistance from other potential stakeholders including SYAS, Historic England and/or the Historic Lottery Fund, university departments, and commercial field units. To recapitulate, such potential projects in South Yorkshire include:
Before any new fieldwork is undertaken by archaeology departments at the Universities of Sheffield and Hull, however, it is imperative that they publish the results of fieldwork that have already been undertaken but never properly disseminated. These unpublished fieldwork projects stretch back over at least 12–15 years.
If funding can be identified and the agreement of a landowner obtained, one or two Iron Age and Romano-British enclosures and field blocks could be selected for longer-term research projects undertaken in conjunction with local commercial field units, material culture specialists, university departments and local archaeological societies. This would not only stimulate research into these landscapes but would facilitate interpretative dialogues between ‘academic’, ‘unit’, ‘specialist’ and ‘independent’ archaeologists. Such collaborative projects could be funded by Historic England, academic grant awarding bodies, the Historic Lottery Fund and other sources. Appropriate post-excavation and publication funding would need to be assured from the earliest stages, and materials specialists able to bring their own expertise and research questions should be involved from the beginning.
Developer-funded archaeology has undoubtedly revolutionised our understandings of the Iron Age and Roman periods of South Yorkshire (Chadwick 2008a; Hodgson 2012). It therefore remains key to improving our knowledge of these periods, but some changes to existing fieldwork and sampling practices would prove highly beneficial and would increase the amount and quality of information recovered from commercial investigations. A series of excavation and recording methodologies are proposed below that are based on extensive experience of investigating cropmark landscapes in the Yorkshire region and Iron Age settlements elsewhere by the author and former colleagues. The intention is not to make life more difficult for archaeological fieldworkers. Instead, it is to maximise the amount of information they can recover, and to make their hard efforts more worthwhile. These techniques are not impractical but informed by developer-funded fieldwork, and they have already been undertaken on specific commercial archaeological sites in the Yorkshire region or elsewhere. They are fully compatible with the needs for cost-effective targeted fieldwork, yet they address specific problems of sampling and artefact recovery highlighted in other research agendas (e.g. Haselgrove et al. 2001: 15).
Aerial photographic and lidar analyses and geophysical survey should be routinely used before any evaluation commences. Magnetometry, the standard geophysical technique for surveying extensive areas, is generally good at detecting ‘negative’ features such as ditches and pits on most geologies (Gaffney and Gater 2003: 123-8). Soil resistance survey is better at detecting buried stone walls (ibid.: 146-7). Magnetometry is thus not infallible, and mistaken interpretations can sometimes impact upon subsequent fieldwork. At Wattle Syke in West Yorkshire, magnetometry survey on Magnesian Limestone produced detailed plots of ditches, pits and circular ring gullies from roundhouses (Webb 2003, 2004). A series of large subrectangular features were interpreted as quarry pits and assigned little significance during evaluation – only one or two were investigated with narrow trenches. When the whole area was subjected to a strip, map and record exercise, however, these were revealed as sunken-floored late Roman-British buildings, some with stone walls and floors (Martin, Richardson and Roberts 2013: 9, 72-3, figs 58, 64, 69, 72). The unexpected buildings added to the costs of the full-scale excavation and interfered with the work schedule. It is possible that soil resistance survey may have identified some of these part-stone structures at an earlier stage. Over suspected complex Romano-British enclosures therefore, soil resistance survey could also be utilised to try and identify features such as stone structures in more detail.
Programmes of trial trenching must balance cost-effective approaches to investigation with the need to adequately sample often large areas, as well as targeting specific features identified through aerial photographic analysis, geophysical survey or analytical earthwork survey. A study focusing on rural assessments concluded however that although trenching evaluations usually recorded the presence and date of archaeological remains, identifying their nature, extent and ‘quality’ was often more problematic (Champion, Shennan and Cuming 1995: 40-50). Evaluation of 2–4% of the total area is often inadequate, and figures of 10–20% are more realistic. Linear trenches are ideal when attempting to intersect features such as ditches and trackways, but they may provide inadequate insights into the interiors of enclosures and the possible presence of pits, postholes and structures. It may be difficult to identify structural features within narrow trenches. Some flexibility in approach is therefore required – standard 1.8–2m wide linear evaluation trenches can be retained for identifying field system ditches and trackways, but for enclosures and suspected buildings it might be preferable to open up wider or squarer trenches instead so that more complex features can be identified, interpreted and their archaeological potential assessed more easily. Strip, map and record approaches ahead of major development schemes are a much better means of assessing and preparing costings for the numbers of features and likely quantities of finds that will be encountered.
When sites are stripped of topsoil prior to excavation, it is often the case (particularly on Sherwood Sandstone sand and gravels) that trenches and open areas might need to be left for a week or more before archaeologists record and excavate them. This allows time for archaeological features to ‘weather out’ and become more obvious through the effects of rain and sunshine. This would also avoid any problems of the initial under-representation of features as occurred during the Pre-Construct Archaeology (Lincoln) investigations at St Wilfried’s Road, Cantley, when less experienced archaeologists oversaw soil stripping (Daley 2007: 6). There are potential Health and Safety issues with this, however, in areas where there is public access, so this might not always be possible (J. Richardson pers. comm); unless individual trenches are surrounded with fencing and have appropriate warning signage in place. There is also a very real problem with commercial units being able to recruit experienced site staff at present, which anecdotal evidence suggests is inevitably impacting upon the ability of some field staff to identify and investigate archaeological deposits and features.
Prior to the excavation of enclosures, those areas should be rapidly fieldwalked. During topsoil stripping of the interiors of suspected ‘domestic’ enclosures, machining should proceed cautiously in spits in case there are concentrations or spreads of artefacts, charcoal and burnt stone surviving within the topsoil or subsoil, as may be the case even on previously cultivated sites. If identified during machining, then some topsoil or subsoil could be left in place and intensively sampled by hand and metal detector for artefacts that might otherwise be machined away. Possible middens and artefact spreads might be detected in this way, as was the case at Scrooby Top (Davies et al. 2000). If the initial results proved disappointing the remaining soil could always be machined down to undisturbed natural subsoil as usual. This might be especially productive on sites that might not have been deeply ploughed or where colluvium or alluvium has been deposited above archaeological deposits.
Similarly, where aerial photographic evidence or the results of geophysical surveys suggest the presence of roundhouses within enclosures, then machining should again proceed extremely gingerly, with 0.10–0.15m of topsoil and subsoil left in place above the expected structures. Hand excavation and 3-D plotting of artefacts should then proceed. If results are disappointing, then normal machining can be resumed. Roundhouse ring gullies and all internal features should be 100% excavated, with extensive palaeo-environmental bulk sampling. The interiors of roundhouses should be subjected to phosphate sampling and magnetic susceptibility testing as a matter of routine, and soil aDNA, soil lipid analyses and other geochemical analyses could be attempted where soil conditions are suitable.
Field system ditches need to be adequately sampled, and ideally at least 20% of their lengths needs to be excavated. Whilst most interventions typically produce few finds, occasional large dumps of burnt stone, ceramics and other materials may occur in unpredictable locations (see above). Such sampling should be pragmatic and realistic however – if field system ditches produce little material on initial investigation through hand-dug slots, then additional interventions across ditches can be excavated in controlled spits by machine under close archaeological supervision. The sections can always be cleaned up by hand and recorded.
On both developer-funded commercial excavations and research investigations, at least 20% of the total lengths of enclosure ditches need to be sampled through hand-dug sections. Excavation should focus on ditch corners and terminals by entrances, but instead of limited 1–2m wide hand-excavated sections it would be more productive to employ 2–5m wide sections near enclosure entrances and corners. Again, this maximises artefact and bone recovery, as it was these areas that appear to have formed a focus for deposition during the Iron Age and Romano-British periods, whether this was everyday discard or more ritualised behaviour.
The total 100% excavation of enclosure ditches in controlled spits by machine under archaeological observation should then take place, with particular care taken towards the bottom of ditches, which could always be quickly hand-excavated. This methodology should be adopted as standard practice. It produced very positive results when undertaken during the excavations at Wattle Syke in West Yorkshire in 2007 (Martin, Richardson and Roberts 2013: 11). Much greater quantities of animal bone, pottery and other artefacts were retrieved, large numbers of human neonate burials were found, and unsuspected placed deposits of pottery and animal remains were discovered. A previously undetected enclosure entrance invisible on the stripped surface due to a later re-cut was also identified and recorded.
Pits need to be 50% sampled as an absolute minimum, and if any placed deposits or Associated Bone Groups (ABGs) are encountered than the features need to be fully excavated. Possible placed deposits and ABGs need to be recorded in much more detail, in recorded spits if necessary, with detailed working shots and field drawings at scales of 1: 10 or 1: 5. Some of these can then be included in reports and publications. Suspected graves should be gradually excavated in rough spits as is usual fieldwork practice; but running sections should also be left in and drawn, before the fills are taken down once more. In this way, possible evidence for disturbance or re-visitation of the graves in the form of later re-cuts can be identified, as perhaps happened with the Iron Age inhumation burial at Bilham Farm, Brodsworth (McIntyre 2009). The evidence is much more likely to be visible in these temporary but contiguous sections, rather than in plan. This technique has been successfully used on developer-funded prehistoric excavation sites in Wiltshire and elsewhere, and in practice does not take excavators significantly more time (C. Gibson pers. comm).
When excavating funnel-shaped entrances opening out onto alluvial areas, and/or double-ditched trackways at the base of gentle slopes or those surviving as holloways, some topsoil or subsoil could be left in place during machining, perhaps in strips 10–20m wide and up to 0.15m thick. These topsoil or subsoil strips would then be hand-excavated with greater care in order to find any wheel ruts or animal hoof prints that might survive underneath.
Fieldwork staff should have much greater on-site training in taphonomic processes and the recognition of stratigraphic interfaces, particularly on the often-difficult soils and geologies encountered within South Yorkshire. The physical traces of ditch re-cutting episodes need to be identified and recorded in greater detail than at present (Chadwick 1999, 2008a). Examination of section drawings in both unpublished client reports and published monographs and journal articles suggests that in some instances excavators simply did not recognise recuts from differences in the profiles of ditches, or that they drew layer interfaces that were highly unlikely to have been present due to a lack of understanding of how silting and slumping takes place.
The notion of ‘small finds’ or ‘special finds’ is deeply embedded in field archaeology but is problematic and is rather a redundant hangover from antiquarian traditions. It privileges certain materials over others from the same context, based on contemporary perceptions of ‘value’. Why should a corroded and illegible Roman coin have greater status than a sherd of samian that could potentially be quite closely dated to 25–50 years? If a contextual bag-number system is adopted (q.v. Collis 2001: 84-6; Cumberpatch and Dunkley 1996: 7), then objects can still be bagged separately on-site for specialist conservation if required, but a separate special find numbering system is not required. The analysis and publication of finds could also be less determined by modern categories but instead by their contextual associations (Chadwick 2012: 303; Haselgrove et al. 2001: 15).
Artefact, animal bone, human bone and other specialists should be more closely involved with project planning, excavation and initial post-excavation analyses from the early stages and should visit or be present on larger projects to inform and manage appropriate sampling strategies. Field units should also ensure that small numbers of site staff are sent on training courses in the identification and excavation of artefacts and human and animal remains, although these personnel should not be used by units as substitutes for experienced specialists who will undertake the post-excavation work. Curators should ensure that those involved in archaeological work closely follow appropriate management guidelines for historic environment research.
Many excavation staff in field units are still not adequately trained to recognise the archaeological deposits most likely to produce productive results and may only sample features that ‘look interesting’. Consequently, although concentrations of carbonised grain or waterlogged plant material may be identified, smaller ‘background’ quantities of palaeo-environmental evidence can be missed. On larger projects in particular, on-site finds specialists should be present to inform and monitor the sampling process, and to take more specialist samples for pollen and soil micromorphology analyses. This is also relevant where waterlogged deposits may be encountered, where there may not be adequate expertise amongst site staff, or existing resources to deal with the material.
For larger excavations, curators, consultants, field units and clients must give more serious thought to the possibility of on-site processing of samples before such major projects commence, and the water sources, flotation tanks, silt traps and drying facilities that are required for this. Adequate storage space needs to be assigned beforehand, and field units may find it necessary to rent additional commercial storage space for this. All soil samples should be routinely tested for the presence of hammerscale using magnets (A. Burgess pers. comm.) – some but not all units currently do this. This is a relatively quick technique that has greatly expanded the evidence for Iron Age and Romano-British iron working within South Yorkshire.
More comprehensive and rigorous post-excavation analyses and data presentation have many practical and financial implications that must be taken into consideration. These include the provision of the necessary contextual data to specialists, along with sufficient resources and time to undertake such work; and closer co-operation between specialists and technical staff in the field units to produce plans, drawings and full stratigraphic matrices. Wherever possible, specialists should be involved with a project from start to finish rather than being allocated specific time slots to carry out analysis of whatever class of data is involved. For larger projects at least, this should involve participation in project planning and the development of sampling strategies before fieldwork commences, regular visits to the site during excavation, assessment of the material recovered and round table dialogues with the director/project manager, supervisor staff and other specialists concerning the research potential offered by the assemblage. The basic identification, cataloguing, spot dating and the selection of material for scientific or other analysis should follow this, and only then full analysis of the data garnered during assessment to examine whatever research questions have been identified as critical. This work might include detailed examinations of distributions across the site, cross-context links, links between different artefact classes and other similar analyses.
It is only after all these stages have occurred that final specialist reports should be prepared. What still happens in practice, however, is that external specialists are often just sent the material to be examined with little or no accompanying stratigraphic and contextual data, and there is usually little or no opportunity to consult with fieldwork staff and the other specialists involved. Such approaches need to change if we are to develop more effective holistic approaches to archaeology.
The overwhelming consensus of the international scientific community is that major climatic changes are now underway across the planet, and that to greater or lesser extents these are being driven by human practices since the late 19th century. Global organisations such as the Intergovernmental Panel on Climate Change (IPCC) and national bodies such as the Meteorological Office National Climate Information Centre have produced recent reports outlining the evidence for climate change and alterations to weather patterns recorded so far (IPCC 2018; Kendon et al. 2018). The Environment Agency and Historic England amongst other organisations have begun to produce guidelines for the likely impact of changes on British landscapes and archaeology (https://historicengland.org.uk/research/current/threats/heritage-climate-change). Climate change is increasingly discussed by many archaeologists, either in terms of new weather patterns and conditions affecting buried and upstanding archaeological remains in Britain and the world (Chapman 2002; Howard et al. 2008), but also how archaeological evidence and palaeo-environmental data might themselves contribute to wider understandings of climate change’s impact on human communities (Sandweiss and Kelley 2012). The West Yorkshire Iron Age and Romano-British Research Agenda (Chadwick 2009: section 11.4) was one of the first such archaeological frameworks to consider the likely impact upon heritage assets at a regional level, but it is now imperative that all such research agendas do so.
In terms of broad climate and weather changes, there will probably be milder but much wetter winters, and longer hotter, drier summers; trends which have already started to become apparent in the past few years (Kendon et al. 2018). Human activities have already caused a mean global temperature rise of c. 1.0°C above pre-industrial levels, with a likely range of 0.8° C to 1.2° C. Global warming is likely to reach 1.5° C between 2030 and 2052 if it continues at the present rate (IPPC 2018: 4), and 2° C by 2100. This might not seem much, but global average temperature rises of 2⁰ C have not occurred on earth for over 100, 000 years and the last such changes led to widespread and highly significant transformations in climate and weather patterns (https://historicengland.org.uk/research/current/threats/heritage-climate-change). July 2019 has been the hottest on record.
Such changes will cause rising sea-levels and greater coastal flooding and erosion. It is also very likely that there will be a greater frequency of extreme weather events. In Britain, upland areas may experience summer droughts, falling water tables and destructive wildfires. At the same time, winters will become milder, but overall there will be an increase in rainfall including more sudden rainstorm events, as warmer air can hold greater moisture (Howard et al. 2008: 407; IPPC 2018: 12). Such events in 2017 have been published by the Met Office (Kendon et al. 2018: 2, 24-7). Total rainfall from extremely wet days has increased by about 17% since the 1960s. Overall seven of the ten wettest years for the UK have occurred since 1998; and current predictions suggest increased storminess in future, with increased likelihood of heavier winds too, though this also occurred during the 1920s. There are significant regional variations to this overall national pattern (Kendon et al. 2018: 12-4). There will likely be more events in northern England such as the destructive floods in York in November 2000 (e.g. Lane 2004: 13-4) and those in Sheffield in June 2007, caused by short periods of extremely intense rainfall leading to dramatic sudden increases in river flows and hence water levels. Low-lying areas alongside the Rivers Don, Rother, Idle and Torne could thus be flooded and damaged, with archaeological deposits or remains either buried by silt and debris or eroded by increased storm surges of current (q.v. Challis, Brown and Kincy 2008; Chapman 2002: 243). This is a very real threat to the possible Roman temple at Bawtry Carr.
In upland areas, streams and runoff could cause great localised erosion, especially where moorlands and hillsides have previously been experiencing prolonged dry spells or have been largely stripped of vegetation by fires. This might affect upland Iron Age and Romano-British fields and enclosures, and those on slopes or with exposed earthworks where erosion has removed grass or other vegetation. This may in turn cause increased flooding within lowland river valleys, and some reclaimed low-lying former alluvial areas may need to be returned to water meadows, osier beds and wetlands (Challis, Brown and Kincy 2008). As Historic England note, both sudden heavy rainfall as well as the cumulative impact from less intense but repeated heavy rainfall events can all be damaging to buried archaeological deposits and remains (https://historicengland.org.uk/research/current/threats/heritage-climate-change). The groundwater chemistry could be altered, and there may be other unpredictable changes to buried waterlogged sites, including Sutton Common. Although the dry summer of 2018 was good for cropmark formation, increasingly saturated soils would make geophysical survey, soil stripping and excavation more problematic at times. Prolonged spells of intense heat and dry conditions would also cause problems for fieldwork.
The many future human responses to climate change in Britain range from increased flood defences and drainage, changes to buildings and in land management and environmental stewardship and changing farming practices (e.g. Howard et al. 2008: 405-6). These might impact directly on archaeological remains. Different crops and/or crop regimes may be required to cope with wider climatic changes, whilst in the future increased quantities of crops could be planted for biofuel production and to be used in biomass generators. These new crops or crop regimes may have unforeseen and highly varied impacts upon buried archaeology than those previously in use. Climate change will also have long-term changes upon energy production and consumption. The need for renewable energy sources may require the construction of wind and solar generation stations. Several wind turbine schemes have already been proposed on hilltop areas in South Yorkshire and have led to archaeological investigations to inform the planning process (e.g. Rose 2010). Careful planning and mitigation will be required to ensure that any negative impacts of future energy production on archaeological assets are minimised.
Following the withdrawal of the United Kingdom from the European Union, some small upland livestock farms may go out of business with implications for future land stewardship and management of heritage assets. In the future there may be a perceived need for greater food security and production, and perhaps more intensive arable agriculture. There may be increased political pressure for such intensification. Ploughing is already having a highly deleterious effect upon many South Yorkshire archaeological sites known from cropmarks. Future demand for housing and major infrastructure developments such as the High Speed 2 rail link would also have considerable archaeological implications and impacts upon archaeology in South Yorkshire.
A scoping document could be produced by the South Yorkshire Archaeology Service to establish the possible archaeological implications of climate change, agricultural and developmental impacts within South Yorkshire. The use of GIS as a predictive tool to examine and identify areas likely to be affected by flooding and run-off is also being developed (Cook 2018), which may help with such analyses.
Original text by Adrian Chadwick (2020), with a contribution by Ellen Simmons (archaeobotany)
The research questions from the entire document are gathered here for ease of reference. Questions relevant to each theme can also be found above at the end of each segment.