This volume reports on the extensive investigation of a roadside settlement on the edge of the Weald, significant for its well-preserved layout, the details of its long-term continuities and development, and the light it sheds on the organisation of iron-working, undertaken on a considerable scale and seemingly connected in some way with the imperial system. It constitutes a welcome addition to our knowledge of a type of site hardly known in Kent away from the northern corridor of the county; indeed, on current knowledge, smaller civil centres are virtually absent from central, eastern and southern Kent and East Sussex (cf. fig. 1.2). That being so, it is possible to deduce a particular set of reasons as to why the site developed where it did, in terms of topography, environment and the contemporary road system, known or suspected. The site, established from geophysics, covers at least 18 ha, a clear 10 ha of which, including the surmised focal area, was taken out of development for long-term preservation; this volume thus covers the excavation of c. 6 ha at the southern third of the site, spanning a period of some 200 years prior to its cessation in the mid-third century.

Kent contains many striking Roman sites with a long history of discoveries and investigation, yet despite this it is neither well understood nor readily characterised. Its proximity to the Channel and the Continent, together with its geological and physiological variability, make it distinctive, but with the focus of so much archaeological endeavour in specific parts of the county, configuring settlement and society, economy and the role of the imperial hand remains restricted. Until recently much was dependent on a series of excavation reports that were modest in number and aging in terms of methods and interpretative models. More recently, several major site reports have appeared to fill the gap, among them the current volume on Westhawk Farm.

All sites have their own identity and Westhawk Farm is no exception. Settlement morphology, spatial organisation and functional areas, including a shrine, are clearly delineated through time. Intriguingly the origin of the site's developed morphology may lie in the co-axial land divisions commencing perhaps in the Later Bronze Age (365–6). Around 20 timber structures were encountered, including contemporary circular and rectilinear forms. Particularly distinctive are the property plots and compounds and the indications that areas near and adjacent to the buildings were used for agriculture and in particular grazing (Robinson, 357–63).

Sarah Paynter's analysis of the metalworking remains (ch. 7) constitutes a major study, combining much data with helpful illustrative material, including annotated SEM images and spatial plots. She establishes the nature of the iron-smelting and primary smithing undertaken in the well-defined workshops and how its chronology conforms to the pattern postulated for the Wealden iron industry which ceased around the mid-third century. The ore was probably local and products included high-phosphorous blooms, a potentially harder product suitable for making tools and billets for export (298). The scale of the iron-processing is modelled but a firm conclusion about its magnitude remains elusive; the synthetic discussion favours the view that this was a site where iron was produced as opposed to one where specialised industrial activity was its main raison d'être. As with other iron-producing sites, pottery may have been manufactured, this being unsurprising given the similar resources and logistics involved (Lyne, 257). The settlement had an agricultural side and probably performed local marketing and servicing functions — weighing commodities was clearly important (Cool, ch. 5) — while the shrine potentially served the hinterland community (cf. 370).

The smoke and smell of industry are effectively evoked by both the cover illustration and several of the individual sections, but there are mixed indications as to whether the community gained much wealth from its labours. While Cool notes that the finds reveal an interest in the acquisition of Roman material culture and the means to afford it (156), it is also clear that an exceptional proportion of the decorated samian bowls had been repaired, which might indicate a lack of ability to afford replacements (Bird, 252); while amphorae include vessels seemingly used as plunge-tanks leading Lyne to postulate (209) that they arrived for re-use after their original contents had been consumed elsewhere. Evidence for recreation is all but absent (Cool, 153–4). There is also little evidence of building elaboration, while the burial groups dating to the main phase of activity yielded generally very modest accompaniments; samian features in some, in line with burials at other rural and smaller civil sites as compared with cemeteries at urban centres or military installations. Two exceptional burials were recorded, namely a Late Iron Age box burial with metal vessels and evidence for linen and a Late Roman cremation burial of a young adult male, with jet necklace and jet and lignite armlet (with a hint of gold), possibly of Germanic origin (chs 3 and 8).

While the authors provide a balanced discussion about an imperial dimension to the organisation of the iron-working, there are no firm grounds for any particular conclusion (384). It is possible that wealth, like the iron products, went elsewhere, reinforcing a suggestion that the community and its enterprises were subject to a controlling military and imperial force; the involvement of the Classis Britannica here as elsewhere in the region cannot be assumed, however, though the chronologies are suggestive (394–6). A strong military influence seems certain, nonetheless, from an assessment of the coins (Guest, 139). It may be no coincidence, therefore, that the two exceptionally furnished graves ‘bookend’ the Roman settlement, perhaps suggesting that the hand of Roman officialdom was indeed at work; and that when this yoke was absent (in the Late Iron Age and Late Roman periods), a different level of wealth existed to be expressed in funerary rites.

Malcolm Lyne deserves praise for working his way through 73,000 stratified pottery sherds, all of which are assigned to fabric categories and presented in quantified tables by phase (ch. 6); the assemblage is then scrutinised by a ‘dated groups’ approach. These are valuable data. The super abundance of the pottery is in marked contrast to the recovery of bone, fewer than 2,500 fragments of which were recovered, no doubt a function of the acidic soils.

Ultimately this report reveals much about how space was organised, material forms employed and productive activities undertaken, such that the milieu of community life becomes apparent. All aspects are discussed in well-informed and thoughtful contributions. This is no less than we have come to expect from a Unit that is a leader in bringing sites to full academic publication; the resulting volume will doubtless be constantly referred to as a model of its kind.