This book should be read by anyone interested in what it calls the ‘long fifth century’, a period encompassing not only ‘The Ruin of Roman Britain’ (whatever that is) but also the later Roman period in the diocese as well as what one might term the ‘successor peoples’ through the fifth and into the sixth century. It shows a wide-ranging knowledge of the archaeology (and history) of the period, is theoretically informed, and its bibliography shows the range of sources used, from excavation reports on individual sites to large-scale syntheses, and the approaches and questions that frame the work. Essentially it seeks to argue that the transition from Roman to early medieval Britain has been over-written and that rather than a sudden, major disruption in the first half of the fifth century, the agricultural economy and society continued much as before. What changed was much more to do with means of ėlite self-representation, with a shift away from an essentially civilian Roman mode based in the precepts of paideia to one where military prowess was the dominant male ėlite marker. It was this rather than socio-economic collapse consequent upon the withdrawal of Roman control that Gerrard argues to have driven the major changes in the forms and visibility of the archaeological record over the middle of the first millennium a.d., which as he acknowledges were profound.

As one who has himself recently written about modes of aristocratic display, particularly invoking the culture of paideia and the importance of militarised aristocratic display, I am hardly going to quarrel with this central thesis. But putting Britain in a wider, Western context a couple of comments seem worth offering. The first is that southern Britain was one of a series of regions, along with the Trier area, Aquitaine and parts of the Iberian peninsula, which invested in this ‘civilian’ tradition of display, showing that the British ėlite was part of a wider imperial network. The second is that the distinction between ‘civilian’ and ‘military’ display is perhaps overdrawn, since military-style display was also a form of ėlite representation in the West, and for fourth-century Britain it is interesting and unusual that the markers of military or official identity (brooches, belts etc.) occur mainly in the same southern regions as the markers of paideia rather than in the ‘military north’, unlike the situation on the Continent.

What the insistence on changing modes of ėlite display allows G. to do, of course, is to argue that the traditional distinctions between ‘Roman’, ‘Anglo-Saxon’ and ‘Western British’ are not as clear-cut as that, and that in particular structuring the fifth century by ‘ethnicity’ has badly skewed our perceptions and resulted in a deleterious compartmentalisation of research approaches and agendas. The ‘Western British’ with their reoccupied hillforts and the ‘Anglo-Saxons’ with their weapon burials play into this argument. So also do the debates about the relationship (if any) between ‘ethnicity’ and material culture, of which G. is well in command, and the consequent and current unwillingness to ascribe such material, particularly from funerary contexts, to the wearer's ethnicity rather than to their age, gender or status. Of course, G. is far too smart an operator to try to write out ethnicity entirely; as he points out the dating of and material from cemeteries such as Spong Hill are very difficult to explain other than in terms of migration. It is though perhaps worth noting that these militarised ‘successor peoples’ in the fifth century are largely to be found at the peripheries of ‘paideialand’, leaving one to wonder what was happening over those large and important areas where the old ėlite markers had failed but new ones cannot yet be observed.

These arguments play into G.'s concern to try to mitigate the concept of a ‘collapse’ early in the fifth century and thus to mitigate the perception of such a ‘collapse’ on our interpretations of what went on in the first half of the fifth century and why. This is the focus of ch. 3 where alternative models for what went on are put forward. The one G. favours, his ‘soft landing’, argues both that the problems of dating have allowed too precipitous a curve to be constructed, and that the ‘systemic shock’ model has overemphasised archaeologically visible material culture at the expense of the fundamental continuities of the agrarian economies and populations in their ‘small worlds’. I do not think anyone in the last quarter-century has argued other than that the fifth-century landscape remained open, peopled and worked. Equally it was recognised that the disappearance of visible material culture was that in which the ėlites were most implicated. But the ‘soft landing’ approach perhaps risks distancing the agrarian population from the processes and consequences of the collapse (yes, that word) of the ėlite economy and culture, with which that agrarian population had been socially, economically and culturally intertwined for a long time. It is still the case that, so far as we can see, relatively simple and non-ėlite technologies and products, for archaeologists above all pottery, did almost totally cease to be available to all sections of society, as G. so ably demonstrates.

Though the book is primarily concerned with archaeology, it does keep the textual evidence in sight, or more accurately in its sights, since it usually deconstructs the historical value of the texts to good effect. That said, I was a bit disappointed that so much of the discussion of the fifth century in ch. 7 responded to textual ‘narratives’; I could have done with a lot less Gildas and a lot more G. here. Mention of Gildas raises the question of religion; not the stale debates about Christian vs. pagan, but the whole area of evidence for religious and ritual practice. This is a very ‘secular’ book (save the obligatory references to ‘structured deposition’), yet given the structuring place of religion in daily life for the people of Roman-period Britain and the abundant evidence for religious sites, ritual practice and associated material cultures, this is an area which perhaps will need more thought. The acute shortage of such evidence for the fifth century, including for the ‘successor peoples’ suggests less a ‘soft landing’ than a ‘reformation’ of religion and its manifestations.

There are many other points it would be interesting to raise, not as criticism but because there is so much to debate in this book. I am sure G. does not see this as any sort of definitive statement but rather as a contribution to moving the debate about this period on. This it most certainly does and I am sure it will provoke the debates there is not the space for here, and many others, for a long time to come.