This important addition to the prestigious ‘Documents d'archéologie française’ series presents the results of excavations during 1989–1994 at the site of Crévéchamps in the Moselle Valley south of Nancy in the Lorraine region of eastern France. During the last 50 years, large-scale archaeological interventions in advance of mineral extraction have become commonplace in the major river valleys of Europe, but in France, a disproportionate amount of this work has focused on the Paris Basin and the corpus of published sites from other regions is more limited. In the case of eastern France, at Crévéchamps, after trial trenching an area of 40ha to establish the extent of the archaeology (much of which lay buried beneath alluvial deposits over 1m deep) six areas covering around 15ha were excavated. These investigations revealed a history of settlement and land-use spanning three millennia from the Neolithic to the Gallo-Roman period. This work, however, took place before the legal framework that nowadays governs such interventions was introduced. It was thus not until 1999 that post-excavation study under the direction of Marie-Pierre Koenig began, funded by the state and the then newly formed Institut National de Recherches Archéologiques Preventives. This work has led to the current report, which is of far-reaching significance not only for its archaeological content, but also for the innovative manner in which the material is analysed and presented.

The volume structure is both thematic and diachronic, with six main chapters, supported by metrical data for all the buildings, radiocarbon dates, a corpus of drawn pottery (annexes 1–3), as well as detailed plans of each excavated area, a context catalogue and an index of archaeological sites mentioned in the text. After a short contextualisation of the project in Chapter 1, Chapter 2 analyses the natural and anthropogenic history of the valley and the farming strategies practised by its successive inhabitants. It draws on sedimentary evidence from late Glacial palaeochannels left behind in the valley floor from before the Moselle stabilised into a single, if meandering, course in the Boreal, along with on-site pollen and floral data and the little animal bone that survived the acidic soils. Particularly useful are a plan pinpointing the contexts yielding relevant data, a practice featuring throughout the volume, and tables comparing the archaeobotanical data from Crévéchamps with contemporary sites in France, Germany and Luxembourg. Chapter 3 examines the structural components of the different areas, predominantly posthole structures of earlier Iron Age date, but also in situ storage pots, wells, ovens, palisades that were used to divide up the settled area from the second millennium BC onward, and undated features that are probably remnants of cultivation. Elements of the Roman landscape include a dwelling area, a small cemetery, 300 pits dug to provide the raw material for a nearby tile factory, and a well-made road heading for a villa located—as in many other northern French river valleys—on the adjacent slopes, all within a system of Roman land-allotment on the same alignment as the road, completing the process of physical landscape division begun in the Bronze Age.

The material culture (Chapter 4) is dominated by pottery. This includes a regionally rare Early Bronze Age group from a single pit, probably from a habitation of which the only other trace was a post-built storage structure and a remarkable assemblage of 200 Middle Bronze Age vessels from one of the palaeochannels that criss-cross the site. Several pits, palisades and ovens belonging to this settlement survived along with a mass of heating stones also dumped in palaeochannels, but no recognisable dwellings. There is a sizeable assemblage of earlier Iron Age pottery, and some Gallo-Roman, but there is less ceramic evidence of either the Late Bronze Age or the Late Iron Age activity that is common in many river valleys. Other types of artefact are rarer, but include some residual lithic implements, indicating frequentation of the terraces by mobile groups prior to the first appreciable evidence of cereal cultivation in the Middle Neolithic—rather later than in the loess-covered areas of northern France—while a range of other utilitarian and decorative objects attest to the extent to which the local population were locked into wider procurement networks from the third millennium BC onward.

All this is brought together in the final two chapters, where the sequence is discussed and interpreted. Aside from the long span of activity in one locale, which alone makes Crévéchamps a key reference site beyond the region, important contributions to the bigger picture are the evidence of a Middle Bronze Age dwelling site and the model of shifting settlement derived from detailed analysis of the Iron Age structures. Notwithstanding the sheer density and extent of Iron Age features, this palimpsest is argued to be a product of cyclical shifts within a restricted zone by a handful of dwelling units over many generations. This kind of earlier Iron Age settlement nucleus is now well attested in the Paris Basin, but Crévéchamps is a still rare example in eastern France. Conversely, the lower visibility of late Iron Age activity allied to more in situ rebuilding is seen as reflecting a stabilisation of settlement linked to changes in the nature of land ownership.

Marie-Pierre Koenig and her collaborators are to be congratulated on a stimulating study, which is a model for the publication of development-led excavations across the continent. Of course, some criticisms could be levelled; most notably, for example, having obtained over 50 AMS dates, why not use Bayesian approaches to model the date and duration of the successive phases of occupation? But there is little else to quibble about. This excellent report deserves to be widely read by all scholars of later European prehistory, and it will be, thanks to the plentiful illustrations and comprehensive abstracts in French, German and English accompanying each chapter, ensuring that the results are accessible not only throughout the continent, but also to linguistically challenged Anglophone archaeologists.