Despite being one of the most intensively studied sites of all Roman cities the detail of how and why London came into being remains intriguingly obscure. The historical sources are silent and the archaeological evidence inconclusive, leaving considerable scope for scholarly debate. In this updated version of her PhD thesis Wallace makes a valuable contribution to this debate by drawing together the results from over 100 archaeological excavations to have investigated London's earliest phases. The original settlement is particularly amenable to study, since London was essentially a virgin site at the time of the Roman conquest before its destruction in the revolt of a.d. 60/61, leaving a bright-red destruction horizon clearly visible on numerous sites. This allows archaeology to explore an unusually tightly defined chronological window, although W. sensibly reminds us that it is unsafe to treat the fire horizon as a window onto a particular moment in time (24).

A problem facing the study of Roman London is that too many important sites are incompletely published, while a mass of ongoing research can be difficult to access and assimilate. A major achievement of this book is that it includes so much new material, a feat made possible by the generosity of London's excavators in making data available ahead of publication. The meticulously prepared colour maps contain an impressive wealth of topographic and architectural detail, much of which is new. These show, in definitive illustration of an argument some years in the making, that London formed around an orthogonally planned core set over a hill on the north bank of the Thames (Cornhill), with distinct areas of settlement of different character flanking arterial roads to the west of the Walbrook stream (on Ludgate Hill) and south of the Thames (in Southwark).

After a well-judged review of previous research (ch. 1), W. gives most attention to the topographic and stratigraphic information. This is set out in three chapters that describe different stages in the urban development: features associated with setting out the site (ch. 2), the first early town (ch. 3), and the topography of a.d. 60/61 (ch. 4). We are offered valuable descriptions of the domestic architecture and an infrastructure of roads, ditches and quarries. Attention then turns to the artefactual evidence, with useful summaries of finds associated with a range of different social groups and practices (ch. 5). These include items likely to have been associated with the presence of soldiers, evidence for industry and commerce (including metalworking, milling, baking, carpentry, butchery, gem-cutting, pottery manufacture, etc.), differences in patterns of consumption (chiefly indicated by comparative measures of table wares and jars), alongside the evidence of coin loss patterns and burial practice.

W. sets her study within the current debate concerning the origins of London, and closely follows M. Millett (The Romanization of Britain (1990)) in seeing London as essentially a civilian site, dominated by merchants from Gaul who had formed a self-governing urban community here a few years after the conquest (perhaps in a conventus civium Romanorum). In firm rejection of the fort-into-town model, W. takes issue with this reviewer's recent revival of the suggestion that the town may have reused the abandoned site of a fort associated with the Claudian conquest. This reflects on a wider fault-line in our understanding of the extent to which the Roman administration directed or delegated the engineering of architectures of political and economic control in the newly conquered province. Much turns on our different readings of the significance of early ditches found on the boundaries of London's early orthogonal core. The most informative results are those obtained from excavations by Ian Blair at Walbrook House in 2006–7, but his work took place too late for inclusion in W.'s original PhD and even now is only accessible in ‘grey literature’. This indicates, contrary to W.'s assertions (70), that the ditches preceded the earliest street system, were not dug in direct anticipation of a programme of Flavian road construction, and are unlikely to have served as drainage features. The configuration of a wide inner ditch and narrow outer ditch suggests a defensive enclosure. Possible reasons for the presence of Claudian fortifications on the banks of the Walbrook are not considered here, in part because W. chooses to understate the evidence for a military character to the early London settlement (better illustrated in the finds assemblages than is acknowledged here) and in part because the evidence is difficult to access.

In reconstructing London's wider urban topography W. necessarily makes several difficult decisions about what weight to give to incompletely published data. The laudable attempt to provide a coherent description of the site inevitably results in some cherry-picking. One example is the weight given to evidence for an early Roman cemetery north of London's early forum (at Leadenhall Court). A burial-ground so close to the urban core suggests a tightly drawn boundary to the area of formal settlement. But the evidence for this cemetery consists of five incomplete vessels found within the backfills of a quarry pit, one of which was found to contain burnt bone (G. Milne (ed.), From Roman Basilica to Medieval Market (1992), 11). There is no certainty that any of the pots contained human cremations and the dating is imprecise. If this evidence is set to one side it becomes possible to argue that the planned city extended further to the north than W. allows. This would accommodate evidence recovered at Whittington Avenue, east of the forum, where a road that formed part of the orthogonal street-grid was buried by burnt debris that interim reports tentatively describe as Boudiccan (also in Milne 1992, 135). W. excludes this site from her survey, preferring a later Neronian date for the fire (77), without presenting evidence in support.

This book is, therefore, a reminder of both the fantastic quality of the archaeological evidence from London and the pitfalls faced by researchers trying to work with the partial results of incompletely published excavations. Because of this our different arguments concerning the origins of the Roman city remain to be tested in excavations and studies yet to come. Notwithstanding these differences, this book is a very welcome and important contribution to our understanding of London's Roman origins. It provides both the data and the arguments that will be a spur to further research.