Turkdean Roman Villa


Heritage Category:
Scheduled Monument
List Entry Number:
Date first listed:
Location Description:
Situated to the north-west of Chalkhill Barn, Turkdean, SP 09857 18979


© Crown Copyright and database right 2021. All rights reserved. Ordnance Survey Licence number 100024900.
© British Crown and SeaZone Solutions Limited 2021. All rights reserved. Licence number 102006.006.
Use of this data is subject to Terms and Conditions.

The above map is for quick reference purposes only and may not be to scale. For a copy of the full scale map, please see the attached PDF - 1408772.pdf

The PDF will be generated from our live systems and may take a few minutes to download depending on how busy our servers are. We apologise for this delay.

This copy shows the entry on 25-Jul-2021 at 19:18:12.


The building or site itself may lie within the boundary of more than one authority.

Location Description:
Situated to the north-west of Chalkhill Barn, Turkdean, SP 09857 18979
Cotswold (District Authority)
National Grid Reference:


The buried remains of a Romano-British villa, with an associated area of building to the south-east, and a culvert of Roman or later date.

Reasons for Designation

The Turkdean Roman complex is scheduled for the following principal reasons: * Survival: as an important example of a Roman villa, which survives well in the form of buried archaeological features * Potential: partial excavation and other investigative work have indicated that the site retains valuable information about the development of the villa as a whole, and the nature and use of the individual structures * Association: the site lies within a significant area of Roman occupation, between two concentrations of villas


Romano-British villas were extensive rural settlements of domestic, agricultural and occasionally industrial buildings constructed in Britain throughout the Roman period, from the first to the fourth centuries AD. Villas were rural establishments, independent of larger settlements. They seem to have been a fundamental part of the model of Romanisation, with a villa-owning elite centred on agricultural estates. Villas are often thought of as high-status buildings, with hypocausts, architectural ornamentation and baths as common features. Interestingly though, most excavated sites in Britain appear to have developed from simpler, perhaps ‘lower status’, to ‘higher status’ or more substantial buildings. The term 'villa' is now commonly used to describe either the estate or the buildings themselves.

Villas are found throughout lowland Britain and occasionally beyond. The least elaborate were modest farmhouses whilst, for the most complex, the term 'palace' is not inappropriate. Most were partly or wholly stone-built, many with a timber-framed superstructure on masonry footings. Roofs were generally tiled and the house could feature tiled or mosaic floors, underfloor heating, wall plaster, glazed windows and cellars. Ancillary buildings, which may include workshops, storage for agricultural produce and accommodation for farm labourers, were typically arranged around or alongside a courtyard, and surrounded by paddocks, pens, yards and features such as granaries, threshing floors, wells and hearths.

Turkdean lies in an area rich in Roman history: the Roman town of Cirencester, second in importance to London, lies twelve miles to the south-west, whilst the Fosse Way runs approximately two miles to the south-east. The site is found between two of the Cotswolds' denser concentrations of Roman villas: to the south a group including Chedworth lies within in the Coln valley, and to the north is a concentration in the Windrush valley. The nearest known Roman settlement to Turkdean was at Farmington, lying beyond the Fosse Way to the south-east; a further two villas stood to the south-west, at Compton Grove and Withington.

The remains at Turkdean are situated on a limestone plateau with fine landscape views to the south and west; a spring rises to the north-east of the site. The earliest pottery found at the site suggests that occupation at Turkdean commenced towards the end of the C2 AD, and the villa continued in use until late in the C4, or the very beginning of the C5, when the Roman forces were withdrawn from Britain. The main villa complex was probably developed in a number of phases. The site of the north range it thought to have been the main residential area in the first arrangement, with some further construction of the northern courtyard. Considerable development appears to have taken place in the second part of the C4. A detached building to the south-east of the main villa was probably built in the late C2 or C3, in at least two phases, and continued in use until the end of the C4. The designation 'Cealcweallas' meaning chalk or limestone walls occurs as boundary mark in a Saxon charter of circa 743 AD; it is suggested that this lies just east of the present Chalkhill Barn, which may indicate some substantial survival of the villa at that date. The existence of a Roman site at Turkdean, 300 metres to the north-west of Chalkhill Barn, was identified prior to the C19; according to a letter of 1891, the remains were above ground level in circa 1800, at which time some of the quoins were removed. The 1891 account describes the site as a villa or station, and mentions the discovery of Roman artefacts: these included coins, part of a mortarium of pseudo Samian ware, and a few pieces of floor tiles, marked with circles in the middle and straight lines on the side, and some pieces of millstone. The correspondent also reports the discovery of a watercourse of faced stones running into a square tank.

It would appear that the remains became obscured, and knowledge of the existence of the site may have been lost. The next recorded identification of the site came in 1976, when the owner sketched a series of white parch marks which appeared in the grass. In 1995 an aerial photograph taken by an amateur archaeologist gave a clear indication of the presence of substantial remains, and in 1997 the site was the subject of a televised investigation by Time Team. Geophysical survey confirmed the outline of a large villa, and limited excavation provided evidence of occupation during the C3 and C4. The configuration of the buildings, which are built around three courtyards, appears to be consistent with the interpretation of these remains as a villa, though the presence of a large bath in central range has led to the suggestion that it may have been a religious sanctuary. The substantial number of finds included Roman coins, jewellery - the most notable piece, a brooch of enamelled bronze, having the inscription VTERE FELIX ('good luck to the wearer') - a hammer head, a decorative roof finial and many shards of Roman pottery, together with box tiles used in the construction of hypocausts, tesserae, glass fragments, and a quantity of animal bones. In 1998 a subsequent programme of Time Team investigated a number of subsidiary sites to the east of the main villa, with finds in the detached building including coins and tesserae.


Principal Elements

The complex includes a Romano-British courtyard villa, with an associated area of building to its south-east, lying on an almost flat limestone promontory overlooking a dry valley. A spring rises on the hillside to the north-east, its waters channelled through a culvert of Roman or later date, emerging as a stream to the south of the aisled building.


The archaeological features are now entirely buried, but their presence has been confirmed by aerial photographs, geophysical survey, and archaeological excavation. All surviving walls are of limestone, and there are some cobbled surfaces within the courtyards.

The main villa complex is built around three courtyards, the whole being approximately 120 x 75m; the courtyards are aligned on roughly north/south axis, with the inner courtyard to the north. The inner and middle courtyards are separated by a cross range, with access through it, whilst a gateway leads from the middle courtyard to the southern outer courtyard. Around the exterior of the villa on the north, east and west sides is a perimeter wall, defining the compound.

The northern or inner courtyard measures 28 x 51m, and is traversed by a stone-capped drain, and by a second drain or path. Along the northern side of the inner courtyard is a range of rooms, 58m long, with corridors to the front and rear; one of the rooms, only 2.5m wide, forms a cross-passage. At either end of the range, larger wing rooms project northwards beyond the rear corridor by 2m. This range shows evidence of demolition and re-building. The west side of the inner courtyard is bordered by a portico or ambulatory, with an open yard beyond, bounded by the perimeter wall; there is a rectangular room in the southern part of the yard. The range on the eastern side of the inner courtyard is defined by two walls 10m apart; there are no visible partitions in this area.

The central or cross range is defined by northern and southern external walls 15m apart, with a spread of rubble between them; there are at least three rooms in this range. Across the centre is a rectangular room, 6m wide internally, and spanning the range, with a slight projection beyond the southern wall; at the southern end of this room is an infilled pool. The pool was lined with red-painted plaster, and had a flagged stone floor; there is a step on the northern wall. A fragment of a stone roof finial was retrieved from demolition rubble in this area. In the eastern part of this range, the room or rooms are bordered by corridors or porticos to north and south. To the west of the central room, a corridor or portico can be identified on the northern side of the range, possibly originally connecting with that on the inner side of the north-west range, but a north/south cross wall appears to have impeded access.

The middle courtyard, lying to the south of the cross range, is slightly irregular. On the western side of the courtyard are few defined structures. On the eastern side is a range 15m wide, in which an eastern corridor has been identified; on the western face of the eastern wall were fragments of painted plaster with a striped pattern. In the centre of the range was found the bowl furnace and flue of a small hearth.

Between the middle courtyard and the southern or outer courtyard are two parallel walls, 5m apart, on a slightly different alignment to the other west/east walls of the villa complex; there is some evidence of a room to the west of this range. There is evidence for a central entrance or gatehouse within the range, 6m wide. On the western side of the outer courtyard is a building or room measuring 16m by 9m; a spread of rubble on the eastern side of the courtyard suggests the presence of further structures. There is no evidence of a southern boundary wall.

Situated 30m to the south-east of the main villa complex are the remains of what appears to be a substantial detached structure, on a west/east alignment. Limited excavation work, together with evidence of alteration during occupation, makes this area difficult to interpret with any certainty. The structure may have been an aisled building, consisting of a nave with aisles to north and south. It has also been suggested that the area may hold parts of two distinct buildings, rather than one. The eastern end of the investigated area is 2.25 above the western area; the western end is therefore supposed to have been of more than one storey. A small sondage at the western end was faced and plastered, and probably represents a cellar or plunge pool.

A series of culverts of Roman or later date has been identified in the eastern half of the site; these are still fed by the spring to the north-east of the villa. A culvert leads southwards to a point to the north of the detached building, and then runs in subsidiary channels to the east and west of the detached building, discharging into the stream to the south. The areas investigated have shown the culverts to be lined with stone, with larger capping slabs.

Evidence of possible industrial activity, or perhaps a heating system, exists in an area investigated to the north-east of the detached building, where limestone rubble, silty clays, charcoal and fragments of iron slag were found. In addition, a circular quarry pit was identified to the south-east of the detached building. These features cannot be dated.


Books and journals
Grundy, G B, Saxon Charters and Field Names of Gloucestershire, (1935), 180
Holbrook, N, 'Britannia, 35' in Turkdean Roman Villa: Archaeological Investigations, 1997-1998, , Vol. 35, (2004), 39-76
Channel 4, Time Team, programmes recording investigations at Turkdean, 1997, 1998,
Cotswold Archaeological Trust, Turkdean Roman Villa, Gloucestershire: Archaeological Investigations 1997-1998, for Time Team, May 2000,
Gloucestershire Sites and Monuments Record Summary Report for Area 19798,
National Monuments Record Monument Report, Unique Identifier: 1185169,
National Monuments Record, Air Photographs dated 1946, 1947, 1974, 1993, 1999,


This monument is scheduled under the Ancient Monuments and Archaeological Areas Act 1979 as amended as it appears to the Secretary of State to be of national importance. This entry is a copy, the original is held by the Department for Digital, Culture, Media and Sport.

End of official listing

Your Contributions

Do you know more about this entry?

The following information has been contributed by users volunteering for our Enriching The List project. For small corrections to the List Entry please see our Minor Amendments procedure.

The information and images below are the opinion of the contributor, are not part of the official entry and do not represent the official position of Historic England. We have not checked that the contributions below are factually accurate. Please see our terms and conditions. If you wish to report an issue with a contribution or have a question please email [email protected].