These two volumes arise from Michael Fulford's career-long programme of fieldwork research at the Roman town of Silchester, which is 80km west of London at the intersection of two important roads. The Little London report is part of a wider project examining the developments that took place at Silchester in the first few decades of the Roman occupation of Britain. It is concerned with the excavation of pottery and tile kilns 3km from the city. Oppidum to Roman city is the final volume of five describing the excavations between 1997 and 2014 within insula IX in the heart of the city.
The motivation for the work at Little London was the prospect to investigate the locality where a tile stamped with the abbreviated name and titles of the emperor Nero (AD 54–68) was found in 1926. This has fostered the belief that Little London was the location of the kilns manufacturing these tiles which have previously been found only within Silchester itself. The imperially stamped tiles are unique finds in Roman Britain and have prompted debate as to what they signify. The excavation was therefore an opportunity to discover more about the context of their production. Two trenches that targeted anomalies detected in a geophysical survey revealed five kilns, not all in use contemporaneously—one was used to make tile, three for pottery and one is very poorly preserved so its purpose is uncertain. Only three fragments of Nero-stamped tile were found among the 4.5 tonnes of ceramic building material (CBM) recovered from the trenches, so it is unlikely that the excavated tile kiln was the one that made those actual tiles. CBM is a ubiquitous find at many Roman sites but—stamped examples aside—for a long time it was an under-researched artefact category. That situation changed in recent decades, with Gerald Brodribb's Reference Brodribb1987 book influential in demonstrating the kind of information that could be gleaned from careful study; since then, understanding has been advanced considerably through post-graduate research and specialists examining the assemblages produced by development-led excavations. The Little London project provided a chance to record the totality of the CBM from a kiln site to modern standards and evaluate the information that this approach could deliver.
While the pottery kilns were not anticipated, their presence was not overly surprising either because the production of pottery and tiles often occurred side-by-side. The pottery can be dated by typology to the Neronian to early Flavian period and was most likely the work of immigrant potters who drew upon continental rather than indigenous traditions for their repertoire of forms. The mortaria (a type of kitchen vessel) have affinities with types found on military sites in the South West Peninsula which were imported, most likely from western Gaul. Perhaps the potters came from there, or else were miliary artisans who had previously served in the South West. Dating the production of the tile kilns is hampered by a lack of independent dating evidence but they are certainly first century AD; the excavators suggest a range c. AD 50/60–80/90 or even slightly later. Notable finds were fragments of relief-patterned flue tile which, if they date to the early part of the period of production, would provide an earlier date for the start of production of these distinctive tiles than has hitherto been thought (the conventional dating is c. AD 75–155/75).
The Little London results have prompted Fulford to consider a series of broader questions. Refreshingly he is not afraid to fly a few kites in his effort to make sense of the findings. His proposition is that Silchester was adopted by the imperial government as a base from which to co-ordinate and administer the reconstruction of provincial infrastructure devastated by the Boudican revolt of AD 60/1. He considers the imperial tilery a short-lived venture, but the big question is where were these buildings in Silchester and what was their purpose that required the use of imperial tiles? The town's Roman baths, re-excavated by Fulford between 2018 and 2021, seem the most likely candidate, perhaps accompanied by other unidentified structures. The topic is taken further in an associated paper by the book's author, which examines the distribution of the relief-patterned tiles in more detail (Fulford & Machin Reference Fulford and Machin2021). It is argued that the essential provincial infrastructure of South East England—and in particular official roadside buildings—dates earlier than has hitherto been generally thought as the relief-patterned tiles destined for bath-houses at such establishments are essentially pre-Flavian (before AD 69). These would be precursors to the known second-century structures to which the tag mansiones or mutationes is commonly applied. The authors consider that a parallel enterprise to Little London is a tilery at Oaksey Common, Minety, near Cirencester, where a handful of fragments of relief-patterned tile has been found. We currently lack securely dated examples of these postulated early roadside structures, and especially so structures that can be securely associated with relief-patterned tiles. Until such time as that evidence is forthcoming, the arguments will remain unproven, although there is a reasonable likelihood that the recent excavations of the Silchester town baths will yield valuable new evidence.
Fabric analysis is an important strand in advancing the arguments made above. The recent excavation of a tile kiln at Brandier, Minety (2km distant from the Oaksey Common site), produced a CBM assemblage of 9.4 tonnes derived from three successive kilns that all exploited doubtless the same clay source. Variation in the fabric of the tiles is visually very marked due to the firing technology utilised in the kilns, and if similar specimens were found elsewhere, nobody would assume they were all the product of kilns at a single locality. Thin sectioning has been the favoured scientific characterisation method to date; pioneering attempts to utilise other techniques such as pXRF have demonstrated the interpretative challenges of the results obtained. What is needed is a comprehensive scientific study employing multiple techniques and blind testing to establish whether characterisation can take us beyond visual inspection and securely identify the products of specific industries. Such work has yet to be done on any scale with kiln assemblages.
The Little London project has prompted interesting new lines of research and has demonstrated the potential of CBM to contribute to broader topics such as the nature of the Romano-British economy and the chronology of provincial infrastructure. As more CBM from stratified, independently dated deposits, is catalogued to modern standards, the conclusions of pioneering typological studies will doubtless be refined, although the longevity of these most durable and eminently reuseable artefacts warns us to expect a high degree of residuality.
While the discussion of Little London is predominately focused on the pre-Flavian period, Oppidum to Roman city is concerned with the development of insula IX between c. AD 85–125/150. The insula IX project has been one of the most important pieces of British field research of the early twenty-first century, and the publication of the fifth and final volume is a reason to celebrate the achievement of Fulford and his many collaborators in delivering work on this scale over 18 seasons. The volume begins with a description of a moment of extreme dislocation in the development of Silchester with the laying out of an orthogonal street-grid c. AD 85 on a different alignment from that used from the Late Iron Age onwards. Was the decision to establish the new grid on the cardinal points of the compass a pragmatic decision to better fit the evolving town with the alignments of the two major roads that crossed here? Or was it a more overt, almost sadistic, demonstration of power by the new town council, a physical manifestation of their will? But if the new grid was intended to dictate uniformity, it did not work. One of the great curiosities of Silchester is the persistence into the later Roman period of buildings built afresh but preserving the old Iron Age orientation, and thus askew to the street grid. Landowners made a conscious decision to build on the old alignment and the town council was either unwilling or unable to enforce a rigid adherence to the street orientation. This suggests that landowners could pretty much do what they wanted on the ground they owned, provided they did not impinge on to the streets.
The sequence of changing land use and timber-building construction is carefully teased out in the report. In many ways, this is the archaeology of the ordinary townsperson as the structures are unremarkable and there are only occasional traces of architectural elaboration and adornment. A case is made from the small finds for a circular building to be associated with ‘magico-medical’ practices.
It would be valuable to have a chronological discussion of the totality of the insula IX sequence exploring the changes in landholding and building location over four centuries. While this information is available in the separate monographs and is summarised in a more accessible account (Fulford Reference Fulford2021), a retrospective discussion that includes a reflection on the excavation strategy would be informative because the project provides such a fine-grained, chronologically precise, visualisation of how urban space was used. This evidence has been gained through a labour-intensive and long-lasting field school that has educated scores of students. While the investment has been high, there is no other way to obtain data of this quality. Indeed, insula IX stands in contrast to many other research programmes on ‘greenfield’ Romano-British towns where the approach is commonly extensive remote sensing (primarily geophysics) followed by sample excavation of relatively restricted areas to test interesting spaces. Given its ambitious scale, insula IX provided an opportunity to examine the spaces between buildings, where artefact and ecofact assemblages that might be diagnostic of function are most likely to be found. The structures themselves tended to be kept relatively clean, with insightful occupation deposits relatively uncommon.
The Silchester excavations will surely be regarded in the years to come as just as influential as Sheppard Frere's were at Verulamium in the third quarter of the twentieth century, especially the late Iron Age and earliest Roman deposits which yielded the most surprising discoveries (Frere Reference Frere1972; Reference Frere1983). Indeed, the completion of the insula IX excavations is a significant moment in the history of Romano-British urban studies. Fulford has displayed phenomenal energy, determination and fund-raising prowess to deliver the project, but who will follow in his footsteps? Many younger academics might sensibly seek quicker wins for their research outputs. Many towns will continue to be explored through fieldwork before development, which can sometimes attract funding on a considerable scale. London tends to dominate the provincial picture, especially for the early Roman period where preservation is typically better than for later Roman deposits. Further opportunities may come as the retail developments of the 1970s, 1980s and 1990s face uncertain futures in whatever form town centres will take in the coming decades. Redevelopment is likely but the emphasis on preservation in situ wherever possible will likely mean that the areas to be explored will be relatively restricted, and it will be a surprise if fresh excavations occur anytime soon on the scale of those of previous decades in, for example, Canterbury, Exeter and Leicester. The immediate future for the investigation of Roman towns in Britain therefore looks likely to be a return to relatively limited excavation areas, but what we may lack in physical extent should be made up by the application of the latest scientific techniques to gain knowledge that has not hitherto been available.
