Oudenburg is a later-Roman coastal fort now situated inland, 7 km south-east of Ostend. This excellently presented set of books integrates reports on major excavations in the south-west quadrant of the fort, directed by the author between 2001 and 2005, with her PhD thesis which expanded and contextualised the work.

Volume 1 begins by enumerating the research questions and the ways in which these have been addressed by the Oudenburg data. There follows an overview of knowledge of the Roman coastal defences on both sides of the Channel – the close relationship between research on the so-called ‘Saxon Shore’ and the north Gallic maritime coast is stressed throughout the work – and a description and analysis of earlier excavations at Oudenburg, predominantly those of Martens in the 1960s and ’70s, and including the fourth- and fifth-century inhumation cemeteries, one of which overlay an extramural bathhouse.

The site reported on included a large area within the south-west corner of the fort and a stretch of its defences. The second chapter comprises the stratigraphic excavation report within an interpretative framework. The understanding of the stratigraphic sequence was largely based on the study of trench sections, with reference to features recorded in plan. The stratigraphy divides into a pre-Roman occupation level, followed by five main ‘levels’, equating to three earth and timber forts, followed by two phases of stone fort, within each of which there were many building phases, all contained within a depth of 1–1.5 m, and surmounted by a ‘dark earth’ horizon 1–1.3 m thick. Description begins with an account of the defensive sequence, including re-assessment of the work of Martens. Evidence for the stratigraphy of internal buildings is entirely provided by analysis of the south-west corner site. The coloured phase plans of the site within Volume 3 are admirably clear.

This area of the first timber fort (c. a.d. 180–200) was occupied by barracks, the structures much cut away by later disturbances. The second fort level (c. a.d. 220–245/50) shows a complete renovation or rebuild of the defences, and the construction of a timber-framed valetudinarium, virtually the complete plan of which was recovered, together with evidence of building and roofing materials. Extraordinarily, much of the interior was decorated with mural painting, sufficiently to reconstruct a decorative scheme. Much of this was found in situ at the base of walls. The third fort level (c. a.d. 245/50–260) again began with a levelling of the previous defences and the hospital building, and a complete rebuilding comprising timber buildings (termed ‘Units’), many with multi-phase hearths and showing evidence for frequent rebuilding and replanning. These are interpreted, no doubt correctly, as barracks, similar to the third-century barracks at (e.g.) Wallsend and include possible officers’ quarters and possibly stable barracks.

Level 4 comprises the first stone-walled fort (c. a.d. 260–290/300), the first for which the complete outline is known. In the south-west corner there is a change of use: small buildings, many hearths and several furnaces bear witness to its use as an area of workshops, with evidence of bronze-working, and waste from the manufacture of brooches and bracelets. Cereal processing and lead working also occurred here. In the second and final stone phase, Level 5 (c. a.d. 325/30–?430) the site was cleared again, and an internal bathhouse was built, with hypocausts and marble cladding, its water supply coming from a number of large timber-lined wells, within which the timber structures survived. Finally, after the demolition of the bathhouse this area appears to have been used for animal housing.

After two chapters discussing the landscape context of the forts by phase, and the previously published late Roman inhumation cemeteries, comes a synthetic chapter which brings discussion of the forts phasing and the finds assemblages together to illuminate aspects of trade and external contacts (an interesting aspect of which is an increasing orientation towards Britain), military identities and garrison change, the presence of women within the fort and the development of complex ‘ethnically ‘mixed’ societies at the end of the Roman period.

Volume 2 reports the huge quantity of material culture recovered. As the author claims, the material stands as a regional finds reference collection from the late second to the fifth century, and as such the volume cannot fail to have long-lasting value. An unusually large number of objects are illustrated, either in the text or (mostly) in the Plates volume – as an example, all 83 samian stamps are illustrated both as small photos with drawings of the associated vessel and also as large photographs. The 233-page pottery report is the most thorough and best presented that this reviewer has encountered. Each ware type is fully discussed including consideration of distribution on both continental and British sites (the late Malcolm Lyne contributes on the British imports). Many different patterns emerge, from diagrammatic representations of sherd matches by ware and vessel across the whole site to the evolution of supply patterns (e.g. a decline of German fine wares contrasted with the rise in British: fig 1.27). Text tables showing the distribution of forms and types, per fabric, across phases are supplemented by very useful illustrated ‘visualisations’ of these patterns.

All finds reports are closely aligned with phase and context information. Other than the ceramics, the largest category of finds material were the metal objects, with 46,083 iron objects and 4,149 copper alloy, of which only a proportion, of course, were diagnostic. The assemblage is discussed by material and by functional classification of diagnostic objects within each material category. Unsurprisingly among the copper alloy objects the brooches are studied in closest detail, not least because of the evidence of on-site brooch manufacture, and the objects representing stages in simple bow-brooch manufacture. The 52 crossbow brooches from the site represent one of the largest collections in the Roman Empire. The brooches are also assessed on the basis of alloys characterised from XRF analysis.

Finally, an appendix discusses entire material assemblages from particular ‘key contexts’. These are stratigraphically well defined and closed contexts from the fort levels, frequently the contents of wells or pits, containing large assemblages, both of pottery and other finds, which are key to the interpretation of the chronology of the site and the activities practised in the immediate area. With fewer residual finds than elsewhere, these in situ ‘key contexts’ are particularly important.

Volume 3 comprises 495 plates. Most are illustrations of the Volume 2 pottery and finds catalogues. Most of the non-ceramic finds are illustrated by photographs of individual objects. Though in general this reviewer prefers objects to be drawn, the photographs are of good quality, and are augmented by drawn elements.

Vanhoutte is to be congratulated for producing a model of how the absolute maximum of information may be derived from a highly complex excavation record and presented with maximum clarity. This is an essential source for the history and archaeology of the Roman coastal defence on the North Sea on both sides of the Channel, and for the study of cross-channel relations between Britain and the near continent in the Roman period.