ABSTRACT
As a starting point, we may revisit the classic sites of Hownam Rings and Bonchester Hill. At Hownam (Fig. 3.1) a stone-built house within the enclosure produced Roman material on its floor indicative of occupation in the later third and possibly fourth centuries. It was not clear whether this building was contemporary with the occupation of the homestead that straddled the former defences on the south-east side of the site, for which the dating evidence was more equivocal. Mrs Piggott nevertheless considered the possibility that occupation at Hownam extended into the post-Roman period. At Bonchester Hill (Fig. 8.1, 1) the intermediate circuit of enclosing ramparts, which unlike the earlier defences does not follow the contours of the hilltop, included a distinctive method of revetment of the front face of the rampart, using edge-set boulders with horizontal coursing above (Piggott, C. M., 1950: cutting X, Fig. 8). The excavator rightly observed that in Northern Britain this was a construction technique characteristic of post-Roman defences, and it closely resembles the method employed in cellular buildings in Atlantic Scotland in the mid-first millennium AD. Whilst material evidence for later occupation at Bonchester was sparse, and the blue glass bead claimed as Dark Age would no longer be regarded as diagnostically post-Roman, a later re-use of the hilltop should certainly not be discounted.
