In this work Collins has written a new kind of book on Hadrian's Wall — one concentrating on a specific period of its occupation and the latest period at that. It is a testimony to the last 25–30 years of research, most of it published, on the northern frontier, the frontiers of the Roman Empire and studies in Late Antiquity. Until now, however, we were lacking a thorough synthesis. C. has now admirably provided this, examining the fourth-century background to the end of the Wall frontier and the nature of continuous transformation on the frontier for a century before the end, to establish a trajectory towards change in the fifth and sixth centuries.
In the introduction, C. rightly sees Britain as a patchwork of different responses to the situation of the late fourth century. The northern frontier response is particularly resonant in that it is the response of the military ‘cut off from other imperial authorities’. He summarises the nature of Roman frontiers as transitional zones, frontiers in narratives of imperial collapse, chronological models for the collapse of the Roman West, and the paucity and relative reliability of different classes of archaeological and literary evidence. In ch. 2, C. examines the broad sweep of the history of Hadrian's Wall ‘from construction to collapse’ as understood and often repeated.
The key question of exactly what changed after the end of the fourth century occupies the three longest chapters. The first examines who the garrison of limitanei were, their logistics and organisation, and the extent to which they might have been called upon to engage in warfare. The command list of the Dux Britanniarum in the Notitia Dignitatum is scrutinised alongside the archaeological evidence for the presence/absence of occupation at fort sites, accepting that occupation need not have been wholly military in character, given the (limited) evidence for the abandonment of vici at the beginning of the fourth century. Current ideas of unit size in the later Roman army are discussed, but still we get no closer than a range of 7–15,000 troops in Northern England with very low numbers not occurring until the fifth century. Discussion of logistical issues concludes that recruitment was largely local, while the complex interlinked issue of pay and supplies are related firmly to local and regional provisioning as part of the tax regime. Important here is the PAS-generated pattern of coin distribution relating paucity of coinage on rural sites in the frontier zone to direct taxation through supply to the military. Pottery supply became regional rather than province-wide. The limitanei are seen as increasingly locally/regionally based for most basic purposes, though within a wider command organisation designed for frontier policing in a context of low-level conflict, primarily involving small-scale raiding.
Ch. 4, on ‘Interpreting Military Transformations’, examines the evidence for change at forts in the fourth and fifth centuries. It is remarkable how consistent these were. Barrack alterations are well known, but other phenomena — subdivision of rooms into smaller units, accommodation in former specialised buildings, the demolition and conversion of horrea, and the spread of metalworking from single to multiple locations — make the internal workings of the fourth-century fort very different from those of the second century. Defensive circuits were maintained throughout, ultimately including the use of earth and timber for repairs. Archaeozoological and paleobotanical evidence shows economic change, with compelling evidence for local supply, in the case of meat, and an increased use of barley. Market-place evidence has been found in some forts. Fundamental changes in the economic organisation and structural use of forts in the fourth century accelerated in mid-century to underlie developments observed in some forts in the fifth century. C. sees transformation from a Roman frontier force to a regionalised army and evidence of continuity reflecting ‘continued occupation by official Roman soldiers who experienced significant social change’ (109–10), and that these units became de facto war-bands. The conclusion is not new, though reaching it through mapping fourth-century change is.
The broader frontier landscape contrasts the distribution of villas and nucleated settlements in the south and east of the area with the farmsteads of the north and west, and is linked to differing tax structures. North of the linear frontier is evidence for subtle shifts in Roman selective diplomacy. Post-Roman settlement and burial evidence in the frontier zone shows a martial tradition in the fourth to fifth/seventh centuries north and south of the Wall, further reinforcing the idea of a continuing military occupation morphing into a war-band society. Palynology shows no indication of widespread reforestation, reinforcing continued occupation at an even level of intensity.
Comparison of the British evidence with that from other frontiers indicates that smaller unit size, changes in construction methods, and a tendency towards regionalisation are general trends on the imperial frontiers. Symptoms of regionalisation are evidence for frontier stability, and a consequent lack of need for central authority to reorganise. Thus the limitanei of stable areas could develop identities leading to their integration into post-Roman polities.
C. concludes that the northern frontier survived the separation of Britannia from the Roman Empire, fragmenting in the fifth and sixth century and becoming wholly irrelevant only when new polities, culminating in the hegemony of Anglian Northumbria, emerged. This has been argued before; however, thanks to C.'s excellent and thorough work, this conclusion now has a proper evidential and theoretical basis.
