Relatively little progress has been made in terms of enhancing an understanding of continued activity within Roman towns in the last ten years, although the publication of the post-Roman levels in the northern Lanes in Carlisle is now moving forward (Zant and Howard-Davis in prep). Work at the amphitheatre at Chester seems to support the contention that, as in London, the pre-burh (established 907; Lewis and Thacker 2003) activity was outside the Roman fortress, as radiocarbon dating has provided dates of c cal AD 720-950 for carbonised grain from features forming multiple phases of activity in the centre of the arena, in contrast to the walled area, where little activity from the ‘middle Saxon’ period has been recorded to date (Wilmott and Garner forthcoming). There continues to be an impression of continuing activity in Carlisle, although this is not often supported by scientific dating, being reliant on sequences of activity that clearly post-date clear Roman occupation, as at Scotch Street (Newman 2011). There, two phases of activity were clearly later than a structure dated to the mid- to late fourth century (NPA 2004), and these notably were of very different construction. The later Roman structures appeared to have been constructed using sleeper beams laid directly on the levelled ground, but these were overlain firstly by a building, presumably with a timber superstructure, laid on shallow cobble foundations, set into a slot, the alignment of which also differed markedly from the Roman buildings. This was subsequently replaced by a structure, presumably cruck-framed, set on large post pads, and measuring at least 10 x 14m (ibid). Scientific dating in the last ten years has, if anything, complicated the picture, as radiocarbon dating of material from the ‘dark earth’ on the northern Lanes site, in the Roman town, has suggested that there were areas of dereliction and abandonment, with the commensurate formation of such soils, long before the formal ending of Roman governance (Zant and Howard-Davis in press).
4.2 Late buildings Scotch Street, Carlisle (permission needed from Wardell Armstrong)
Similarly, there has been remarkably little work in association with Roman forts in the region. At Ribchester, ongoing excavations within the fort are producing evidence suggestive of very late Roman, or sub-Roman, activity (D Sayers pers comm), and a new site in Lancashire may contain spreads of stone overlying Roman deposits (R Philpott pers comm) reminiscent of late phases of activity at Stanwix, on Hadrian’s Wall (McCarthy 2002, 136). In Cumbria, excavations outside the forts at Maryport and Papcastle have also produced hints of early medieval activity. At Maryport, a large, seemingly aisled, timber building, apparently multiphase, was excavated on the highest land to the north-east of the Roman fort, immediately beyond the core of the civil settlement (Haynes and Wilmott 2016). This was clearly later than the construction of a circular ditch that contained Crambeck ware, and may have been associated with late Roman/early post-Roman graves in the vicinity. Nearby, on the River Derwent’s floodplain, below the fort at Papcastle, geophysical survey indicated the presence of large timber buildings. Initial evaluation has produced an early medieval radiocarbon date (cal AD 780-970) from a posthole, which suggests that at least one of the buildings may be of this period (Wardell Armstrong Archaeology 2015).
4.3 Late levels Ribchester – from Duncan Sayer
4.4 Maryport plan – though would need an ‘improved version from Newcastle University
Whilst the number of confirmed early medieval rural settlement sites is still very small, despite the English Heritage survey of upland landscapes in Cumbria and Lancashire (Oakey et al 2015), there have been significant advances in knowledge in the last ten years as a result of excavation, particularly in Cumbria. There, two sites have been dated scientifically to the early medieval period, in Nether Wasdale, and just to the north-west of Carlisle, near the present village of Stainton. In both cases, timber-built rectangular or rectilinear structures were revealed by excavation, constructed using a post in posthole technique, but associated with no obviously datable material culture. Radiocarbon dating, however, provided modelled dates of cal AD 650-850 to cal AD 730-910 for the early medieval occupation of Nether Wasdale (Brown and Evans in prep) and cal AD 710-880 to cal AD 780-950 for the site on the Carlisle Northern Development Route (Gregory in prep). The site at Nether Wasdale was associated with metalworking, but also provided a small number of medieval dates (fourteenth to sixteenth century), although it is unclear how the features these came from related to the postholes providing early medieval dates. A tantalising early medieval date (cal AD 676-876; 1248±30 BP; SUERC-69186) has also been obtained from material below a medieval stone-founded building in the Duddon Valley (OA North 2016), only a short distance to the south of Nether Wasdale.
4.5 Nether Wasdale plan
The buildings at Stainton seem to have belonged to a small nucleated settlement, consisting of five structures, perhaps representing two properties, as there was a suggestion of compounds divided by fencelines. Three of the buildings were on a north-west/south-east alignment, and were very similar in size and construction (6-7m long by c 3m wide), whilst the other two were larger (8-9m long by c 4m wide), and set at right-angles, to the north of the smaller structures. It has been suggested (Gregory in prep) that the larger buildings may have been domestic, with the smaller buildings being ancillary, and it is possible that one of these had been replaced by another.
4.6 Stainton plan
Although technically in Yorkshire, the western part of Craven is on the western side of the watershed, and there, a series of small-scale excavations of stone-footed structures has been undertaken by the Ingleborough Archaeology Group. Material from these has been routinely subjected to radiocarbon dating, and with almost no exception, these structures have proved to be of early medieval date, with a remarkably close date range of c AD 700-1000 (F Brown pers comm; Johnson 2012; 2013; 2015a; 2015b). This work would suggest that early medieval structures tended to use the materials to hand, with timber preferred in the lowlands and stone-footed structures perhaps more favoured in the uplands.
4.7 David Johnson plan – would need permission
In the south of the region, the multi-period site at Irby, in the Wirral, has been published (Philpott and Adams 2010), where a sequence of activity post-dating Roman occupation followed a period of abandonment. This centred on three elliptical or bow-sided buildings, which were superseded by another building, with a rock-cut foundation slot with clay packing at its base, which contained a Saxo-Norman spike lamp (ibid). More recently, work in advance of improvements to the A556 between Knutsford and Bowdon in Cheshire has produced possible evidence of early medieval settlement in the form of pits containing charred cereal remains. These have produced radiocarbon dates of cal AD 550-670 (1430±41 BP; UBA-30659; Wessex Archaeology 2016) and cal AD 540-660 (1467±35 BP; UBA-30658). Whilst just outside the region, a pipeline near Rhuddlan, immediately over the border in Wales, has produced hints of early medieval land divisions in the form of ditches, the fills of which have produced early medieval dates (Gregory et al in prep). This type of land division is just as likely to exist in Mercia and Northumbria in the period, and there are possible hints of one such system on the A556 in Cheshire (Wessex Archaeology 2016).
4.8 Irby pic – from Rob Philpott
As on the A556 improvement, cereal remains, particularly oats, are being recognised on increasing numbers of sites, as palaeoenvironmental studies become more routine, and these are increasingly being subject to scientific dating. While it would be a truism to suggest that the presence of oats necessarily suggests an early medieval site, the routine dating of oat grains has contributed quite a number of early medieval dates, and has led to the recognition of early medieval activity on sites where it could otherwise not be proven. This occurred, for instance, at Irton, Cumbria, where small-scale work in advance of a churchyard extension produced features, at first thought to be prehistoric in date, but radiocarbon dating of two oat grains, a flax seed and a hazelnut shell all produced early medieval dates (cal AD 621-770 (1349±33 BP); cal AD 661-773 (1301±29 BP); cal AD 688-879 (1234±29 BP); cal AD 777-968 (1162±29 BP); L Watson pers comm). These dates are perhaps hardly surprising, given the presence of a very fine ninth-century cross in the adjacent churchyard (Bailey and Cramp 1988), but are welcome additions to the corpus nevertheless. The general acidity of the soils of the North West makes the survival of animal bone from the period very rare, and there have been few advances in an understanding of animal husbandry in the last ten years.