‘This book is about rethinking the relationship between urban settlements and water in the Roman period, focusing principally on the towns of Roman Britain, and what this theme can tell us about the way in which these urban spaces were experienced’ (1). The aim is laudable, but in order to rethink a subject it is useful to present an analysis of the current thinking and to explain why it needs to be rethought. Instead, the Introduction makes a series of claims about the need to consider cultural attitudes to water when examining Roman urbanism, while announcing that the book is not going to examine aqueducts, water supply, or bath-houses. We are therefore denied any opportunity to examine the interplay between the usage and experience of local, naturally occurring water sources, and those brought in from outside by means of artificial systems. The focus is to be on how water was experienced in Roman towns in Britain, but only in so far as that water consists of rivers, streams or marshes (in the event, springs are given very little discussion).

Ch. 2, ‘Five Urban Waterscapes’, examines London, Canterbury, Cirencester, Lincoln and Winchester, looking at their relationship to rivers and the creation of built waterfronts, derivation of rivers and land reclamation. Ch. 3, ‘Rivers, Lakes and Islands: Towns, Changing Waterscapes and Geography’, covers similar ground with a wider range of examples. Ch. 4, ‘Waterfronts: The Land/Water Interface and the Construction of Waterfront Installations’, focuses on waterfronts, returning in more detail to many of the examples in chs 2 and 3. It tells us repeatedly that waterfronts are located at the land/water interface. Ch. 5, ‘Wetlands: The Materiality of Land Drainage and Reclamation in Towns’, is actually largely about rural wetlands, only pp. 194–203 and 211–13 dealing with towns (mainly the examples already covered in ch. 2). Ch. 6, ‘Conclusions: Towns, Waters and Places’, reminds us again that it is important to look beyond infrastructural aspects of water supply and offers some ways in which we might focus on natural elements of water in Roman towns.

Despite his emphasis on the need for theoretical frameworks, the author does not in fact develop one beyond the reiterated claims that elements of the ‘waterscape’ (a landscape with lots of water in it) were ‘integral parts’ of urban settlement, were ‘important’ or ‘significant’, and that places ‘were imbued with meaning’ (phrases which occur repeatedly throughout the book). Why they were important, what constitutes their significance, what kinds of meaning were attached to place, and whether watery places had more meaning than other places, he never really investigates. Practical or economic explanations are set aside in favour of vague theoretical assertions that ritual or cultural significance will have been at least as important. The discussion proceeds by means of ‘may have been’ and ‘will have been’ (four times in a single paragraph on p. 18). The treatment of London's ‘waterscape’ in ch. 2 is emblematic, never rising above such banal statements as ‘Islands clearly formed a significant component of the urban topography in the Roman period’ (38) and ‘Water formed an integral element of the urban plan and the cultural attitudes to these water courses, such as the religious activity associated with the Walbrook, will have been a significant element of urban life and the urban biography’ (38–9). Yet although figs 2.1–2.5 show parts of Roman London, there is no one map that shows both banks of the Thames — thus rendering it utterly impossible for the reader to gauge the effect of the waterscape and the city's relation to it that Rogers is attempting to describe.

But one could in fact attempt a proper study of the ritual and cultural factors, rather than merely postulating their significance. Much more could be done with Celtic place-names than the very brief discussion on pp. 21–2: what proportion of toponyms contain elements relating to water, and what kinds of elements are they? A systematic presentation of the evidence for ritual deposition in streams, rivers and bogs in late Iron Age Britain would give us a yardstick to compare with the often speculative claims that Roman items found in ‘watery contexts’ should also be interpreted as ritual deposits. Bath is mentioned only once, in passing (137), as an example of a site with a ‘bathhouse’ in a temple context; one gets no sense that Aquae Sulis was a spa with hot and healing springs where religious activity included the deposition of curse tablets. The evidence of relevant dedications to the nymphs, or Neptune, worshipped in Britain as in other provinces as a general water deity at inland sites, remains unexploited.

The bibliography is almost entirely anglophone (10 non-English items in 37 pages of bibliography), which may explain some of the narrowness of perspective and the lack of exploration of whether Romano-British urbanism differed in its relation to watery features from Roman urbanism elsewhere. Presumably Brian Campbell's Rivers and the Power of Ancient Rome (2012) appeared too late to be taken into consideration. Overall, this book represents a missed opportunity to explore in any convincing or rigorous way the intersection of the practical, cultural and religious interaction with water in towns in Roman Britain; it is unclear that we have any real rethinking here.