‘new’ Roman villa? – Colchester’s got villas!

On Sunday (17th April), the news was broadcast of the remarkable discovery of the remains of a very large Roman villa during works at a farmhouse in Wiltshire. The villa was given great media and social media coverage: you can read about the discovery at www.bbc.co.uk/news/uk-36066089 .

From the Essex County Council SEAX web-site: ‘… Roman villas were extensive houses, often equipped with baths, mosaics, and other trappings of the wealthy landowner, and sat at the centre of large rural estates. In Roman Britain they are seen as a symbol of status and of alliance with Rome, part of the process of Romanisation which took place across the country. Villa sites in Essex include the large site at Rivenhall, and others at Wendens Ambo, Gestingthorpe, Little Hallingbury, Boxted Wood, Hadstock, Pleshey and Chignall St. James. Much of the rural development which took place in the Roman period seems to date from shortly after the Boudiccan rebellion, about 70-75 AD, but the construction of villas and the system they represent reached its height in the 2nd century (100-200 AD) …’.

Over 60 Roman villa sites are known in Essex, but only a few of these have been investigated. The sites of Roman villas are known from aerial photo.s and substantial masonry remains or reports of Roman remains in farmland or on old building sites. One problem is that some Roman villas in Essex would have been built of timber and not masonry, and so their sites might not be identified. Another problem with villa sites is: when is a Roman villa not a villa? – that is, when it is a farmstead… Roman villas can range in size from the very large to the very small, when it becomes difficult to identify the site of a complex as a villa or as a farmstead; there is also a range of evidence of the Romanisation of such complexes.

The nearest known villa sites or suggested villa sites to Colchester include those at Great Tey, Fordham, Stanway, Layer Breton, Abberton, Alresford, Brightlingsea, and St Osyth, with several on Mersea Island. However, the Trust has discovered and investigated the sites of four Roman farmsteads or farmsteads/villas at Colchester itself, within the oppidum of Camulodunum (later the Roman town of Colchester), during our archaeological investigations at the old Colchester garrison over the past 20 years or so. These sites represent modest, functional farmsteads/villas and not great villas like the ‘new’ discovery in Wiltshire… The sites of our farmsteads/villas were at the former Meeanee and Hyderabad Barracks, at the Kirkee McMunn Barracks, at the former Goojerat Barracks, and at our Area E by the Mersea Road (see CAT Reports 412, 588, 607 and 788).

At the former Meeanee and Hyderabad Barracks off Mersea Road, in 2004 and 2011, we excavated Roman ditches, post-holes, and a quarry-pit. These features suggest that there was a Roman-period ditched enclosure here which contained at least one, relatively small, Romanised domestic brick building and several possible associated post-built structures. These elements seem to represent a small villa-type establishment of moderate status set in its own fields. We retrieved large quantities of Roman building material, including roof-tiles (tegula and imbrex), and Roman pottery, from the site. Fragments of flue-tile show that the ?villa included a hypocaust system, either for heating the house or a bath-house, so it would have been an establishment with some pretensions to a Romanised lifestyle. We also uncovered evidence for some activity outside the ditched enclosure, in the form of post-holes, ditches, and a possible gravel pathway to the enclosure which was itself flanked by ditches. This site is similar to another two sites which we investigated at the old garrison, ie in the Kirkee McMunn Barracks, 1,540 metres to the south-west, and in the former Goojerat Barracks, 883 metres to the south-west.

In 1994, at the Kirkee McMunn Barracks (now Merville Barracks, the new garrison), we discovered the site of a simple Roman farmstead/villa of moderate status. In fact, much of the area of the new garrison area seems to have formed part of the farmland of this farmstead/villa. The site produced Late Iron Age ditches associated with evidence of early Roman occupation, including a hypocaust pit typical of a small villa. The farmstead/villa was also close to a major road/trackway to the town. The farmstead seems to have been abandoned in the later 3rd century AD, perhaps due to the unstable times: there is evidence at Colchester that people of the settlements outside the walled town centre moved into the walled town in that period.

At the former Goojerat Barracks, in 2010, we uncovered evidence of an L-shaped, post-built timber structure within a ditched enclosure. This seems to have consisted of only two chambers, one of which included a deep basin-like pit within a substructure, and which may have been used for the storage of foodstuffs or liquids or, even, for the extraction of salt from brine. The structure was associated with a well and a group of pits and it may have been a farmhouse or a barn, which pre-dated but may have continued in use into the Roman period. We retrieved fragments of Roman building materials – septaria, greensand, tufa and ceramic roof-tiles – from the wider area of the farm complex, as well as several tesserae. The roof tile included tegula and imbrex fragments. So the post-built structure may have included a tiled roof and, perhaps, some structural stone. We interpret this structure as a domestic building which was influenced by Roman style, even if it was not a fully Romanised building.

The best evidence which we have uncovered for a Roman villa at Colchester is from a site in the Trust’s Area E of the old garrison, adjacent to the Mersea Road. Here we discovered the site of a Roman villa/farmstead, again of moderate status, but also close to a major gravelled road/trackway to the town. A robbed-out Roman wall associated with post-holes suggests that an early timber-framed building here had been replaced with a building with masonry wall foundations, which seems to have included an internal clay floor and an external metalled surface. A layer of demolition material included a large quantity of Roman tile, including fragments of box-flue tile from a possible hypocaust, tessera cubes from a possible tessellated or mosaic floor, and fragments of tegula and imbrex roof-tiles. The building here must have been fairly substantial, with masonry walls, a hypocaust heating system, a tiled roof, and a paved courtyard. The pottery fragments from the site dates the occupation of the site to the late 1st/early 2nd to the mid/late 3rd century. The villa/farmstead was associated with one or two agricultural enclosures like farmyard compounds or paddocks.
From CAT Report 412: ‘… It is likely, given the location of the building and the associated finds, that it formed a Roman villa/farmstead that operated as an administrative centre for the agricultural land within which it was set. The pottery evidence also indicates that the people of this villa/farmstead were at least moderately wealthy as they had access to a range of Roman commodities (amphoras) and pottery types (including fine wares and utilitarian coarse wares), and it was a place where food was prepared, stored and consumed. Significantly, the building was located immediately to the west of Mersea Road that probably follows the approximate line of a Roman road leading to the south-east gate of the Roman town. Thus the farmstead had easy access to the town and its hinterland to the south …’. We think that both the Area E and the Kirkee McMunn Barracks farmstead/villas operated as administrative centres for the agricultural land to the south of Roman Colchester, which supplied the town with food or which may even have included stud farms associated with our Roman circus…

We got Roman villas! – sort of…

All the Trust’s fieldwork reports are published online at http://cat.essex.ac.uk/ . Visit the SEAX web-site at http://seax.essexcc.gov.uk/ .

There are several Roman villas which you can visit in Britain, including Fishbourne Roman palace, Lullingstone, Chedworth, and Bignor.
The site of the finest Roman villa in Essex is, perhaps, at Rivenhall, where the villa complex included the main building, which consisted of a main block with two wings around a courtyard, a second building which was connected to a bath complex by a paved corridor, an aisled barn, a corn-drying oven, a possible mill, a possible lantern tower, and the line of a Roman road (SMR no 19117 on the SEAX web-site). The only visible remains of the villa complex are the building platform on which the east end of the later parish church stands.

The images show a group of Roman building materials from the farmstead at the former Goojerat Barracks and the farmstead enclosure at the former Meeanee and Hyderabad Barracks, and a plan of the old Colchester garrison, showing the Trust’s archaeological site areas outlined in blue: the black shows excavated evidence in four of our areas. The four farmstead or villa/farmstead sites are each indicated by a red cross. (The precise locations at the former Goojerat Barracks and at Area E are speculative and that at the former Hyderabad and Meeanee Barracks is highly speculative!)

.