Although a sketchy narrative of the general political events related to the Roman conquest of northern Britain is offered by Roman literary sources, the social impact of the military takeover, and the later transformation of the region into a frontier area, with the building of Hadrian’s Wall, are poorly understood. There is likely to have been regional variation in both the pre-and post-conquest landscapes, and it may be that the permanent imposition of Hadrian’s Wall on the region had a much greater long-term impact than the initial conquest of half a century earlier (see section R4 Native and civilian life), when, as further south in Britain, a continuity of native settlement forms would have been the norm with little change (in the kinds of artefacts at the sites, for example) before the Hadrianic period.
One priority is increase the precision with which a number of changes in the settlement pattern, occurring at the beginning of, or early in, the Roman period, are dated. This is likely to be possible primarily through the widespread use of radiocarbon measurements from material in well-understood archaeological contexts. See below for the essential aspects of a successful radiocarbon sampling strategy. The essential need is to acquire statistically significant knowledge of which native sites were abandoned at the time of the conquest and which were retained, and then how long the surviving sites lasted into the Roman period before being abandoned or undergoing a significant transformation.
Previous discussions, and even the first version if this research framework, influenced by the apparent homogeneity and lack of visible hierarchy among late-Iron Age rectilinear settlements (‘Jobey sites’), are rather dismissive of site morphology as a key to social structure. More recent discoveries have shown that in the future an unexpected range of site types and sizes are likely to come to light; as the examples accumulate the materials will come to hand to reassess what the varying size and layout of settlements means for the structure of the society that built them, a process that will be aided by obtaining as many complete plans (including outworks) as possible. Although the oppidum of Stanwick is just outside the region, the recent publication of the excavations there hugely increase our understanding of a site whose influence may have been felt across much of south Durham and Cleveland. One major research area must be a more detailed exploration of the Roman military and administrative response to Stanwick and the recently discovered neighbouring site at Scotch Corner. How do the military dispositions relate to the presence of the oppidum and whatever social authority succeeded it? How far was the wider network of known and possible villas and rural production centres (for example, Quarry Farm, Ingelby Barwick; Faverdale) and small towns (for example, Piercebridge and Catterick) influenced by the presence of this important focus of pre-Roman Iron Age occupation, or to what extent were they an utterly artificial Roman imposition that bore little relationship to the pre-existing social structure?
Specifications for excavation on sites of late Iron Age or Romano-British date should have adequate provision for establishing precise site chronologies. This might include the use of high-resolution radiocarbon dating and full analysis of ceramic assemblages, including thermoluminescence dating. As already suggested, for sites with limited quantities of, or no, datable finds it is essential that independent chronologies are established using multiple radiocarbon or other scientific measurements. Individual or small numbers of dates are unreliable and for fully-excavated earthwork enclosure settlements and other extensive sites an absolute minimum of 10-15 single-entity short-lived material dates from well-understood context should be aimed for, and preferably at least 20, more if possible. Bayesian modelling should be undertaken to enhance the information obtained from the measurements. It should be recognised that a multiplicity of poorly preserved and artefact-poor sites whose periods of occupation have been defined by scientific dating will provide more information in combination than a single site with better preservation and more artefacts – this is why it is vital that developer funding is still obtained for the investigation of site types of which examples have already been excavated in the region. In order for this to happen, project designs for both research excavations and developer-funded work need to make clear that material suitable for radiocarbon dating should not only be collected when spotted by excavator: on plough-truncated sites featuring mainly natural cut features in acidic sands and clays, there is often little visibly apparent survival of bone and plant remains, and a sufficient number of organic samples suitable for radiocarbon dating is likely to be obtainable only by sieving bulk samples (which will have been taken also for the purpose of environmental assessment – see Rix below).
In a situation where a large number of poorly preserved sites contribute to a bigger emerging picture, and the percentage of sampling will inevitably vary according to individual circumstances and resources available, it is essential that the research frameworks call for standardisation so that the quantities of pottery and other finds occurring at different sites can be meaningfully compared. This means that it must become standard to report the square meterage of features sampled and to report the volume of the deposits from which pottery and other finds derive (as is already the practice in the case of the recovery of seeds and pant remains from standardised units of soil). Thus it may be possible to identify those sites which stand out as genuinely exceptional in terms of their production and use of pottery or other artefacts during the Iron Age-Roman transition. Any re-dating of previously known sites should be fed back into the region’s HERs.
The Roman communication network in the region is only superficially understood and a greater understanding of its development is a priority. This research is intimately linked to the development of the earliest military infrastructure of the region, where our understanding of 1st century communication routes is as hazy and incomplete as our knowledge of the complete distribution of Flavian forts.
The work of Margary (1973), still the main source for roads in the region, should be revisited. Particular focus needs to be placed on exploring the putative course of Cade’s Road and the installations along it, and the possibility of a road from Newcastle/Benwell to the Devil’s Causeway. There has been very limited excavation of roads in general and a better understanding of the network of minor roads and trackways in the region would be welcome, especially exploiting the potential of LIDAR, aerial photography and geophysical survey. There should be a particular focus on the Stanegate. Although some of the associated forts, such as Corbridge and Vindolanda have been investigated, we still know little about others, or about pre-Hadrianic dispositions east of Corbridge, and the relationship between sites, such as South Shields and Washingwells to the Stanegate; in the case of Washingwells further work should establish the chronology of the fort.
Undated marching camps might be targeted for small-scale excavation to try and establish their chronology and major re-evaluation of the aerial photographic evidence could identify further possible sites, like the camps recently identified by the National Mapping Programme at the confluence of the North and South Tynes, potentially marking the North Tyne crossing of the Stanegate. The role of riverine communication and coastal installations related to seaborne trade should be researched, by means of a systematic study of the occurrence of Roman period finds along the coast. Direct research on ports, lighthouses etc. remains impractical because of the elusiveness of the evidence, but as a known Roman fort with evidence for a possible surviving Roman shipwreck, continued research at South Shields has the potential to address these topics.
There should be consistency in approaches to recording probable and possible Roman roads on the region’s HERs/SMRs, exploiting the potential of LiDAR and GIS techniques. Coastal installations should be assessed for potential threat from erosion and where necessary appropriate management regimes put in place. In many parts of the region marching camps and small forts may not appear as cropmarks; large-scale geophysical survey must be incorporated into the site evaluation process.
A particular urgency here is the obtaining of a greater understanding of the early period of military occupation, when sites were largely of turf and timber construction, often occupied for only a short period of time, and are often undiscovered or archaeologically invisible. The map of 1st century (Flavian) dispositions is almost certainly far from complete, as exemplified by the failure to finds early military sites where Dere Street crosses the Tees to enter the region from the south, or where the Devil’s Causeway reaches the sea at Tweedmouth in the far north. Was Corbridge really the easternmost Flavian post on the Tyne, or is it simply a matter of archaeological visibility that we know of no Flavian forts at places like South Shields or Gateshead, to control the strategically important river mouth and lowest crossing of the Tyne at an early date? A number of possible early forts, at Wrekenton, Elstob and Picktree, have been identified using aerial photography by Raymond Selkirk. The nature of these sites should be evaluated through a programme of fieldwork.
Despite its focal role in archaeological research in the Roman period in the region, reflected in its inscription as a World Heritage Site, further research into Hadrian’s Wall is required as fundamental uncertainties remain over its form and appearance in various periods, and indeed its actual purpose. Though the Wall is visible along much of its course, further investigation is required into the course of the Wall in urban Tyneside. There is also potential for previously unrecognised contiguous elements of the Wall system to be discovered.
Despite the recovery of near-complete plans at Wallsend, Housesteads and Vindolanda, the interior arrangements of most Wall-forts remain obscure and this is equally true of most forts to the south and north of the Wall, though there have been some recent exceptions. For Roman forts in general in the region, gaining a more complete picture of their interior arrangements and how these changed over time remains a priority, particularly for the light it can shed on the ‘sociology’ of the Roman army – the social structure of the army and its attendant civilians in different periods. Attempting to relate this kind of data to the information offered by the written sources is challenging and frustrating, and much remains to be done. There are indications from the Notitia Dignitatum that the units in forts south of the Wall changed in the 4th century. Is this reflected archaeologically? How does this affect the vici?
Basic questions about the size of the units and the numbers of their members who would have been resident in a fort at any given time remain obscure and controversial. Signs of the presence of women and children, or, for that matter, male civilians, in forts (as well as the vici) at particular periods and in particular areas within the walls must be actively sought but this has been little done in the region to date. Geophysical survey has the potential to make dramatic advances in revealing internal plans (see the recently published geophysical plan of the interior of the outpost fort at Risingham), but only excavation can trace the development of these plans from phase to phase, tracking change over time, and that goes for the vici too. It is also the only way to recover the artefacts and biological materials that might provide dating evidence for the sequence of development and cast light on the nature and origin of fort inhabitants. In practical terms the opportunities for research of this kind will be strictly limited: modern excavation and analysis on such deeply stratified (and largely unthreatened) sites is often prohibitively expensive to carry out and fully publish on any significant scale.
The physical fabric of Roman forts and their environs remains under-researched. There is a vast opportunity in the region and a vast amount to do to better understand military architecture and reconstruct the scale, design, construction methods and external decoration of Roman buildings designed by architects and erected by soldiers building in a Mediterranean tradition. The variety of structures where further evidence and elucidation is needed includes, in this region: bridges, aqueducts, baths, defensive walls, with towers, ditches and gates, installations unique to Hadrian’s Wall (e.g. milecastles), peristyle houses (praetoria, town houses), granaries (horrea), headquarters buildings (principia), barracks, market buildings (macella), urban shops and industrial premises (‘strip houses’), temples and smaller shrines, tropaea (Victory monuments), tower tombs and other monumental funerary architecture, possible churches, and agricultural buildings, storehouses etc.
Other under-researched areas include the source of building materials (lime for mortar, stone quarries etc.), while increased environmental work will also improve our understanding of the environmental impact of the Wall, such as the extent of forest clearance and the resources needed for both its construction and the surrounding infrastructure (stone, lead, iron, wood, etc).
Finally, research is required into the function of forts and their relation to the civilian population in the vicinity. The well-known armlets or bangles, manufactured from re-used Roman glass, of the early-Roman period are ubiquitous site finds both at forts and native sites, demonstrating some relationship or link between the two: This needs to be more intensively researched. How much did the function of forts change over the period of their use? Forts should not be studied separately from their vici and vice versa; the populations and economies of these two site types would have been closely integrated and their development closely linked. There is still much basic work to be done in mapping the distribution and extent of vici, both at a regional level and for individual sites (see Riii). There is a need to improve our knowledge of the chronology of the vici, particularly the date at which they fall out of use. Who were the vicani? Was there any participation by the local indigenous population in the evolution of these settlements? What was the relationship between the vici and their forts? Was it always one of economic dependency or did some of the vici become self-sufficient?
Much of the Wall is a Scheduled Ancient Monument, although in built-up areas, both in urban Tyneside and in villages, remains are not protected in this way. Any development-driven archaeology here must be carried out within a research context. Reporting of such work should not be limited to ‘grey literature’. A better understanding of the sites identified along the Wall corridor will entail a campaign of targeted excavation, which balances conservation and research. Fieldwork should evaluate the date and preservation of sites, which must feed into site management and protection. It is clear from recent geophysical work that the vici often extended beyond the current scheduled area; full surveys of all the vici should be carried out in advance of re-consideration of protected areas. Finally, any work on Hadrian’s Wall and the associated military infrastructure must be placed firmly in an international context. The world importance of the Wall is highlighted by its status as a World Heritage Site, and moves to increase modern links between Hadrian’s Wall and the Antonine Wall, as well as the wider integration with research on other important Roman limes structures further emphasise this dimension of the region’s Roman heritage.
Our appreciation of the relationship between the Roman military and the native and civilian populations of the region should be improved, whether the local population consists of indigenous British communities or groups from elsewhere in the Empire. To what extent was the economy of native communities influenced by Roman invasion and control? Did indigenous communities continue to farm and carry out industry in a native manner, or did they change their ways under Roman influence? What impact did the environment and native society have upon the deposition of Roman military forces during the conquest? How did native peoples react to Roman soldiers (and vice versa)?
Previous research frameworks in places assume that the late-Iron Age rectilinear enclosures of the region (“Jobey sites”) are broadly contemporary with Hadrian’s Wall. The apparent impact of the Wall on the people of the coastal plain to the north now disclosed by radiocarbon dating, the disappearance there of a traditional way of life, and the implications of that for the way in which the Roman frontier Wall was intended to function, challenge the former view that they flourished in symbiosis with the Roman occupation: it is now clear that their mainstream development was in the pre-Roman Iron Age. However it is important to continue to test these preliminary results of developer-funded work since 2000 by obtaining as many scientific dates as possible for the early Roman transition that occurs on sites in the Iron Age tradition both north and south of the Wall. Sites detected from the air immediately north of the Wall in its central sector remain uninvestgated and undated. Equally it becomes an urgent priority to find evidence for whatever less archaeologically visible form of society replaced the formerly dense agrarian settlement pattern in the lowland areas north of the Wall. It is essential that all elements of the settlement hierarchy are further investigated; in particular further work should be carried out to assess the extent and significance of the abandonment and possible Roman-period re-use of hillforts in the Cheviots (and elsewhere).
Are the military and native populations quite as distinct as traditional models make them seem? Do we have settlements that acted as local administrative centres and were independent (to a degree) from the army? It is important to avoid the easy elision of ‘native’ and ‘civilian’: many civilians in the region would have come from outside the region (either from elsewhere in the Britain or from abroad), whilst it is possible that some ‘native’ groups could have operated closely alongside the military, such as the areani.
In addition to native communities there were other important civilian centres, most notably the vici attached to many forts. To what extent was there real urbanisation in the North-East, with some fort vici emerging as regional centres in their own right? What distinguishes Corbridge and Piercebridge from the vici? What is the function of other civilian sites in the region such as East Park, Sedgefield?
A final research topic is the spread of the villa form in North-East England. A number of sites, such as Faverdale, Old Durham, Quarry Farm in Ingleby Barwick, and Dalton have been either suggested or firmly established as villa sites. This brings the distribution of northern villas closer to Hadrian’s Wall than would have been imagined only 20 years ago. More work should be carried out at these, and similar sites are bound to be discovered in the future. An important related question is who built and lived in them? Were they the homes of the descendants of the local, pre-Roman elite or immigrants in the aftermath of the Roman conquest? The spatial organisation and architecture of these sites should be explored more thoroughly. To what extent is masonry construction used? Is there regional consistency in their spatial layout or is there heterogeneity amongst this class of sites? Do their plans show influence form other areas of the empire? How did these villas fit into their wider landscape? Do networks of fields and paddocks surround them? What was their relationship to larger rural/urban settlements? For populations living in villas and other rural settlements, what do artefactual and ceramic assemblages tell us? How do they relate to assemblages at military sites and their vici? This artefactual material will also help improve our chronological understanding of these sites. Do they grow out of native Iron Age settlements or are they entirely new establishments? How late do they continue in use? Is there any evidence for sub-Roman occupation?
The provision of agricultural goods to markets by villa estates would have required good communication. Can the presence of roads be used as a predictive tool for identifying further sites? (see also Rii). Where were the markets? Whether through developer or research funded work, both military and civilian landscapes in the region call out for fuller exploration, through survey including field-walking, geophysical survey, LiDAR and the use of aerial photographs, but the significance and function of resulting new site types will only become clear if a sample of them can yield excavated datable materials.
When future sites are excavated, work should be preceded by large-scale geophysical survey and field-walking. Where possible, open-area, strip and plan excavation strategies should be employed. It is essential that to obtain as complete a picture as possible research designs explicitly call for search for outworks and environs features lying outside ‘visible’ site nuclei.
As already suggested, for sites with limited quantities of or no datable finds it is essential that independent chronologies are established using multiple radiocarbon or other scientific measurements. For extensively excavated earthwork enclosure settlements and other extensive sites an absolute minimum of 10-15 single-entity short-lived material dates from well-understood context should be aimed for, and preferably at least 20, more if possible, with Bayesian modelling applied. It should be recognised that a multiplicity of poorly preserved and artefact-poor sites whose periods of occupation have been defined by scientific dating will provide more information in combination than a single site with better preservation and more artefacts – again it is vital that developer funding is still provided for the investigation of site types of which examples have already been excavated in the region.
Again, excavators should be briefed that a sufficient number of organic samples suitable for radiocarbon dating is likely to be obtainable only by sieving bulk samples (which will have been taken also for the purpose of environmental assessment). As already stated, to achieve standardisation so that the quantities of pottery and other finds occurring at different sites can be meaningfully compared it is essential that research designs call for excavators to report the square meterage of features sampled and to report the volume of the deposits from which pottery and other finds derive (as is already the practice in the case of the recovery of seeds and pant remains from standardised units of soil).
More research into Roman artefacts from the region is necessary. Although they are mostly published, particularly those from military sites, there is a need to capitalise on this vast quantity of accessible material. What suite of finds might be expected from a rural settlement, a vicus or a fort? (Allason-Jones 1988; Reece 1995). How can these assemblages be used to explore topics such as age, gender, class and ethnic identity? In addition to studying patterns relating to the use of small finds, there is still potential to improve our appreciation of their production, for example where objects are being made and by whom – as has been done in a recent pioneering study of quern use in the region. It is important to be more alert to technical advances in the study of small finds research.
Despite the large quantity of excavated ceramics from the region, there is still much important work to carry out. The large, stratified collections from the three Tyneside forts offer an opportunity for a significant work of syntheses. Re-examination and synthesis of the pottery from turrets and milecastles has been undertaken but not published.
Mechanisms of pottery supply also require investigation. These include the trading links which bring in ceramics in from outside the region, as well as native pottery production. Further petrological analysis of coarsewares may help locate local production. There should be full publication of the kilns excavated at East Park, Sedgefield and Piercebridge.
There is still a need for a published synthetic study of the coin evidence from Hadrian’s Wall or the region as a whole.
Briefs for NPPF excavation on Roman period sites should ensure adequate provision for full analysis of all ceramics recovered. Where necessary, petrological analysis should be specified. The recording of Roman material through the Portable Antiquities Scheme must be consistent.
Although small finds from major research and development-driven excavations are likely to reach publication, it is important to ensure that finds, including pottery, recovered on National Planning Policy Framework (i.e. developer funded) excavations are properly published. In some recent printed reports, finds have been relegated to the archive in favour of structural description (or, worse, general discussion bound to be of an ephemeral nature). In the case of significant assemblages of pottery or other material this order of priority should be reversed. On-going training of small finds and pottery specialists is vital and students should be encouraged to carry out small finds and pottery-related dissertations at undergraduate and postgraduate level. New methodological approaches to the analysis and interpretation of material culture should be explored (for example, Cool 2004).
Further research is required into Roman industry, including primary mineral and coal extraction, the production of ceramics and metalwork, and timber supply.
Additional fieldwork in the North Pennines may locate traces of lead and silver working, including both mining and smelting sites (and silver-extraction works, if these were indeed separate from smelting sites). This should include geochemical approaches as well as conventional fieldwork.
Although briquetage is present at some late Iron Age and early-Roman sites throughout the region, very little is known about the location and technology of the industry for most of the Roman period: production sites and routes and methods of transport remain to be identified.
Evidence from the Vindolanda tablets suggests the army was sending some distance for supplies which would be expected to be available locally. How far does this hold out for material which is archaeologically visible, for example, small finds, pottery, etc?
The potential of isotope analysis to identify the places of origin of cattle moved through trade or requisition for consumption at Roman military sites and civilian centres has been demonstrated by work at South Shields. Livestock origins might be one way of marking trade routes; another might be the recognition of traders’ gathering places or campsites, an examples of which might now be recognised north of a major gate (the Portgate) through Hadrian’s Wall, at Great Whittington. More research into such places is needed.
Routine mortar sampling and analysis following the excavation of Roman mortared stone structures has enormous potential to inform on sources of lime and other building materials, and the distances and networks over which they were transported. Early mineral-working sites should be dated using absolute dating techniques, otherwise it may not be possible to distinguish Roman sites from earlier and later examples. Where located, appropriate conservation measures must be put in place, possibly through the DEFRA Environmental Stewardship Scheme or via Scheduling. Full analysis of ceramic assemblages and increased understanding of local metal production and processing are priorities.
Research into Roman quarrying and stone extraction in the Hadrian’s Wall zone has been conducted by a number of PhD students in recent years, and a published overview and synthesis of their results would be valuable.
The North-East possesses an extensive corpus of epigraphic material, and the excavation of a number of religious sites, such as Coventina’s Well, has taken place. However, many important questions remain to be asked about religious practices in the region.
Corbridge and several Wall-forts have yielded astonishingly rich epigraphic and structural evidence for Roman military religion and its built settings, but away from the military sites relatively few Roman or Romano-British temple sites are known in the region. Are these being missed by archaeologists or did religious observance take some other form? Much of the surviving evidence for religious practice relates to the public and official rites of the Roman army; can we identify changes in religious practice at military sites which might reflect changes in units? Do all forts have mithraea? What is the extent of Christianity in the late Roman period? Is it purely a military phenomenon?
Moving away from public practices, what kind of domestic religion was practiced? How can we recognise native religion when it is filtered through Roman material culture? Is this necessarily true, or are we just looking in the wrong way? Are the so-called ‘Celtic stone heads’ really native British, or are they, as seems more likely, of continental origin, an idea imported by immigrants in the Roman period?
A wider understanding of the ritual and symbolic landscapes of the region could be achieved, both north and south of the Wall. Further research and full publication of the material recovered from the river at Piercebridge is essential. How widespread was the deposition of material in natural features, such as bogs, hills and rivers?
A small number of excavations have taken place on Roman burial sites in the North-East (Lanchester, South Shields, High Rochester, and, most recently, Binchester), and these are supplemented by an extensive epigraphic record (see R7 above).
It is important to develop new research questions and ask for central questions to be addressed, such as: is the idea that cremation might have prevailed in the late-Roman period in the military north based on evidence, or merely the absence of evidence (successfully recovered late-Roman inhumations)?
Although some cemetery locations are known, there are likely to be many more. Parallels from elsewhere in Roman Britain are a useful indicator to their likely location. Targeted geophysical work may be useful. Where were the civilians buried? How is burial – or disposal of the dead – outside the immediate vicinity of Roman military sites characterised?
In addition to simply identifying sites, it is important to have large-scale cemetery excavations. Excavation of burial sites within the region has been piecemeal, and it is only by exploring rites and practices that many questions can be answered. What was the effect of change of military units on burial practices? What evidence is there for ethnic grouping in burial practice (Cool 2004)? How were graves marked before and after the use of gravestones?
A clear priority is the location and investigation of an inhumation cemetery (by definition later-Roman in date) with enough well-preserved skeletal material for quantative scientific analysis. Many basic questions are still unanswered relating to stature, age and pathologies. Sexual identification can now be more easily achieved by means of peptide analysis. There is also scope for exploiting the potential of isotopic analysis on skeletal material, which may identify the geographical origin and biographies of buried communities.
Roman burials must be fully excavated and analysed, particularly if skeletal material is preserved. Where possible absolute dating should be carried out. Large-scale geophysical survey might be used to identify burial sites related to forts/vici, and, especially, late-Roman urban centres such as Corbridge. If burial sites related to known military or civilian sites are identified, the state and extent of preservation of the burials should be evaluated and, where necessary, protection extended to cover the burial ground.
Our knowledge of the landscape of the Roman North-East might well be expanded, including environmental evidence for farming practices.
The predominance of spelt wheat cultivation in the late Iron Age and early Roman period throughout most of the lowland part of the region – definitely running north of Tyne – is now established by more extensive sampling of later-Iron Age sites (contra Van der Veen 1992), but it remains to be seen where the northern border of the crop husbandry regime lies, before the predominantly emmer-using Lothian is reached. On the other hand, the overall picture of trends in crop husbandry in the mid- and later-Roman periods, both north and south of the Wall, remains obscure. Are former identifications of ‘Roman’ bread wheat secure, and can bread wheat imported from outside the region be recognised?
More pollen cores are needed in most areas, but especially away from the uplands and especially on the Northumberland coastal plain to see whether the apparently widespread abandonment of agrarian settlement in the 2nd century AD is reflected in the pollen record (the basic problem here though is an absence of suitably waterlogged and accessible deposits). This material should be synthesized to inform the wider picture of environmental change, which would take account of information already obtained from sites along the Wall, for example Fozy Moss, Crag Lough. What is the potential of wetland sites elsewhere? Further investigation of fossil soils under the Wall and its earthworks, and of ditch infills within its earthworks is vital.
Full analysis of plant macrofossils and faunal data should be part of the brief for development-driven archaeology. These samples will also be the main source of organic materials for radiocarbon dating. On both counts it is essential that briefs make it clear that the value of these materials is in building up a broad picture over many sites, so the fact that similar sites nearby have had samples taken and analysed should not be relevant in determining the necessity to sample and assess. Samples should be taken in standard sizes and the volumes of soil both excavated and sampled should be recorded and reported in order to allow standardised comparison between sites. There should be adequate provision for sampling and analysis of small peat deposits discovered during development-driven fieldwork.
This is a key area of research as sequences of activity extending into the 5th century and beyond have come to be recognised on Roman period sites, especially the military and urban sites. The Roman-early medieval transition requires clarification and it is essential that the structural and artefactual archaeology is closely studied in each potential case with an open mind and without preconceptions. In interpreting structural evidence, it must be asked whether it possible to distinguish between continuity or personnel and re-use/re-occupation on Roman period sites in the 5th century. The potential role of Christianity and of post-Roman civilian inhabitants in the late 4th and 5th centuries must be carefully considered. These investigations should extend to the common but largely undated types of ‘native’ settlement, to investigate whether forms like upland enclosures or roundhouses continue into the post-Roman period.
A re-assessment of post-Roman finds from South Shields is currently underway; similar work should be carried out to search for such finds at other major excavated Roman sites, such as Corbridge. Margaret Snape has identified a distinct sub-Roman variant of D7 penannular brooches. Further work could identify other potentially diagnostic examples of very late/sub-Roman material culture in the region.
Absolute dating techniques must be employed wherever possible to date very late sequences on Roman sites. There are difficulties in a period when the radiocarbon curve had a particularly unhelpful plateau after about 430 cal AD, but it should be clear to all that the only effective response to this is to obtain as many dates as possible from well-understood sequences of the later-4th century onwards.
Part of the Research Frameworks Network