Just to the north of Cirencester on the finger of limestone between the River Churn and Bagendon Brook lies the oppidum excavated by Clifford in the 1950s. Contemporary with this and a couple of kilometres further up that spur is the 4 ha Ditches enclosure, originating in the late Iron Age and containing within it an exceptionally early Roman villa. This volume reports on excavations carried out there in 1984–5 by Steve Trow and Simon James, combined with more recent geophysical contextualisation by Tom Moore. Earlier excavations in 1982–3 were published in the Transactions of the Bristol and Gloucestershire Archaeological Society in 1988 and inclusion of a better summary of these in this report would have been worthwhile.
The site began as a banjo-like enclosure in the late Iron Age, morphologically perhaps having more in common with other sites in Wessex than other hillforts on the Cotswolds. Indeed, a theme of the discussion of Ditches throughout the report is that of significant influences from elsewhere inspiring the development of this site; from its start it was an almost alien site in the West of England. In its late Iron Age phase there were no buildings excavated, but the currency bars from the outer ditch, the Augustan and Tiberian finewares, the Augustan gemstone and the coin or pellet moulds all point to a significantly high-status enclosure.
Similar to the development of Gorhambury very close to the oppidum at Verulamium, Ditches had significant storage capacity within the site's earthworks which the excavators related more to stock management than defence. They believed the 100 or so storage pits could have contained around 130 tonnes of grain if all were ever in use at the same time. Analysis by Huntley suggests this cereal came from damper fields than the limestone slopes of the Cotswolds, demonstrating the site's draw and importance in the late Iron Age.
The villa itself started as a simple structure in the early post-conquest period; there was no earlier timber building, meaning this stone-built house would have immediately appeared alien within this landscape, and clearly visible to passers-by on the Cirencester–Gloucester road to the west. Rapidly a corridor, cellar and wings were added to the structure over the next century; it fell into abandonment in the late second century with the collapse of a cellar and what was left was destroyed by fire in the early third. Its history is, therefore, in marked contrast to the floruit of third- and fourth-century villas in the area. The very early use of masonry, almost a century before it was being used in town-houses in Cirencester itself, suggests imported skills and labour.
Structured deposition is both observed and discussed, from the remains of six articulated cattle in the boundary ditch dating to the early post-conquest period paralleling deposits at other sites in the area (Roman Fields, Abbeymead and The Bowsings), to the phenomenon of copper brooches being deposited in the boundary ditch.
The authors provide a comprehensive synthesis, situating the site within contemporary discussion of ‘romanisation’ and identity. Much is made of parallels in early masonry architecture with other buildings in the South-East and Gaul, and clear changes in diet and ways of eating and drinking are also noted. Together with the forthcoming publication of Trow and Reece's earlier excavations at Bagendon and the current programme of geophysics there by Moore, this collection will provide a significant overview of one of the classic oppida outside the South-East.
