This is an odd book, which might have been more appropriately titled Late Roman Public Buildings in Britain. The majority of the chapters are concerned with a straightforward analysis of the condition and use of public buildings in the larger towns of Roman Britain in the late Roman period, the later third to early fifth century. However, prefacing these is a chapter (2) on Edward Gibbon and his History of the Decline and Fall of the Roman Empire, which provides the context and framework for situating the concept of decline in relation to the towns of Roman Britain in ch. 3. In addition, between these introductory chapters and the core of the book is a chapter (4) on ‘Establishing the Urban Context: Pre-Roman Place and Roman Urbanism’, which considers the Iron Age background to Roman developments, with a digression on the possible symbolic importance of water in their origin. All very well, but what this has to do with late Roman urbanism is not clear. It does, however, immediately draw attention to a yawning gap, the absence of any consideration of what towns were about in Britain in the first two to three centuries and thus how far their condition might or might not be perceived to have declined or changed in the late Roman period.

Since the peg on which the book is hung is that of ‘decline’, the author is not so much concerned with trying to characterise what towns were about in early or late Roman Britain, but with the analysis of the parts of the town as represented by public buildings. Thus the town in this context embraces the usual suspects of the larger examples where public buildings are to be found: coloniae, municipia, and civitas capitals. While the analysis of the selected evidence is certainly thorough, does it represent any more than the sum of the parts? There is no serious consideration of the town beyond the public buildings, so there is no glue with which to hold these parts together, no characterisation of the town in the late Roman period.

Nevertheless, from his very thorough analysis of the available evidence from the public buildings Rogers draws out two themes. The first is the increasing amount of evidence associated with the late Roman period for the use of public buildings for industrial purposes, particularly metalworking, while the second draws attention to the incidence of possible new timber structures within these buildings at the same time. However, the evidence for the latter is generally very fragmentary and dating such structures is problematic and, as the author acknowledges, some of them may be of post-Roman date.

The real problem in a book which is conceived against Gibbon's idea of decline is the lack of treatment of the public buildings in the early Roman period. Decline presupposes a need to make comparisons, but here the evidence for characterising the use of public buildings in the early Roman period is even more difficult, not least because there is still very little of it and what there is is complicated by problems of truncation or obliteration by later Roman activity. Perhaps there was not very much change between the second and the fourth century. Altogether this adds up to a book which neither addresses the question of change over time — though carefully rehearsing the late Roman evidence — nor attempts a holistic view of towns in Britain in the late Roman period. The unsurprising conclusion of the last chapter is that we are encouraged to think of ‘decline’ as ‘change’, even if we do not know from what.