The Roman Religious Sanctuary at Blacklands: Excavations at School Farm, Graveney Road, Faversham, Kent
In 2007 and 2008 the Kent Archaeological Field School (KAFS) carried out a campaign of investigation by trial trenching and limited open area excavation of an archaeological site to the east of Faversham, on land traditionally called ‘Blacklands’. The site had been found during field-walking as part of the Swale Archaeological Survey, conducted by the writer in 1995-1999.[fn1]
The results of the KAFS archaeological investigation were spectacular, with two buildings and a Roman rural sanctuary theatre investigated out of a potential ten buildings mapped by a geophysical survey, carried out by English Heritage in March 1997.
The purpose of the geophysical survey was to attempt to trace unexcavated archaeological features extending beyond the theatre area and to look for postulated features related to the temple complex, situated to the east of the theatre such as a temenos wall and ditch. In addition, the surveys were designed to look further afield for any other indications of buried remains which might add to the understanding of the archaeological context of the site under evaluation. Geophysical surveys were undertaken by Andy Payne of English Heritage, following the establishment of a 10m grid across the field by KAFS and Prof. Bill Martin.
[fg]jpg||Image[/fg]
[fg]jpg|Fig 1 and Fig 2 - The geophysical survey by English Heritage. As published in the Kent Archaeological Field School, The Roman Religious Sanctuary at ‘Blacklands’, School Farm, Graveney Road, Faversham, Kent, 25/12/12.|Image[/fg]
The geophysical survey had stunning results. The theatre is in the area highlighted with a red circle. The stage backs on to large freshwater springs, whist the seating area is cut into the hillside. Surrounding the theatre is a large complex of Roman buildings which include Building 1 (indicated by a red dot), possible temples (indicated by a blue dot) and Building 2 (indicated by a yellow dot).
The initial archaeological attention was focused on the south field in an attempt to locate the Roman buildings which had obviously been dismantled to provide building material for later local constructions. The soil was well drained arable soil, possibly forest at some stage. There are 16th century accounts of the School Lands providing timber for the new grammar school in Faversham, but now the land is ploughed.
Two ‘bald spots’ full of Roman building materials were noted almost immediately where modern ploughing had removed the ‘crown’ of soil in a colluvial action which meant the soil was creeping down the west slope towards the still working springs of Ewell Fleet.
The KAFS team concentrated on these two bald spots and began to find small, marble, multi-coloured tesserae, larger, tile-cut tessellated floor cubes, painted plaster, and large amounts of Roman building ceramics and Roman pottery. A small excavation was conducted and mortared stone walls were exposed, as well as a hypocaust system constructed out of chalk blocks. The hypocaust channels were filled with fragments of multi-coloured pictorial mosaics of the best quality and full colour pictorial painted plaster, some showing exquisite paintings of birds darting through foliage.
In view of the continuous damage to the environs of the site by agricultural operations and the danger of destruction to archaeological deposits by treasure hunters, it was felt that an archaeological evaluation of the site should be undertaken.
Negotiations with the landowner were completed satisfactorily, various funding bodies and local institutions were contacted, and a major campaign of evaluation was undertaken in August 2007-8. The success of this campaign was due to the generosity and kindness of the landowner, the sponsors, History Today magazine, Friends of the British Museum and the students from the Kent Archaeological Field School.
[fg]jpg|Fig 3 - Excavation of Building 1 in August 2007. The thinness of the topsoil covering can be seen. One of the excavators is kneeling on the southern flint exterior wall whilst a post hole dated by pottery to the late 5th century is being excavated. To the right another post hole is being drawn up. Three of eight postholes of a post-Roman timber building, constructed on top of the earlier mosaic floored bathhouse, can be seen. Note the massive wall on the right of the picture which is part of the base of an apsidal room.|Image[/fg]
The main aims of the evaluation programme were to define the central zone of the religious complex, to investigate the nature and building history of the Roman buildings. To seek any traces of a pre-Roman precursor to the Roman shrine and, by careful site recording, to dissect the latest remains and observe any indications of post-Roman modifications or reuse. It was also intended to record the positions three-dimensionally of as many of the finds as possible, in order that the distributions could be analysed.
The site produced evidence of Neolithic use, Bronze Age and Iron Age occupation. The Roman site seems to have been established in the late 1st century AD, and the theatre seems to have gone out of use by the mid-4th century.
The two Roman buildings excavated were substantial stone-built structures, both with hypocaust heating, and one (Building 1) embellished with fine pictorial wall plaster and full colour pictorial mosaics. Building 1 overlooked a large depression excavated out of the chalk hillside which upon investigation, has been identified as a Roman cockpit theatre of a type usually found on rural sanctuary sites in north-west Europe rather than Britain. The theatre overlooked large fresh-water springs from which a number of votive offerings have been retrieved; others were retrieved from field-walking on site.
The hypocaust channels of a Roman bathhouse in building 1 can be clearly seen in Fig 9, along with the slots (indicated by blue arrows) for hot air to heat the vertical walls through a system of box flue tiles. Fragments of full colour figurative marble mosaics of the finest quality were retrieved from the hypocaust channels. Stephen Cosh noted: “Exceptionally fine tesserae (averaging approx 6mm) in red, pink, ochre, white and dark grey. One fragment is clearly part of a figure and is probably an eye”.
[fg]jpg|Fig 4 - A quadrant of the theatre’s seating (cavea) was excavated in the north-east corner and just to the south of Building 1. The photograph is looking to the west and in the direction of the prolific freshwater springs of which the theatre overlooks. The original seating seems to be of turf on a sculptured base of rows of seating. Beyond, excavation is taking place on the front edge of the stage and back wall (scaenae frons).|Image[/fg]
Roman theatres hosted events such as plays, pantomimes, choral events, and orations. Their design, with its semicircular form, enhances the natural acoustics, unlike Roman amphitheatres constructed in the round. These buildings were semi-circular and possessed certain inherent architectural structures, with minor differences depending on the region in which they were constructed. The scaenae frons was a high back wall of the stage floor, supported by columns. The proscaenium was a wall that supported the front edge of the stage with ornately decorated niches off to the sides. The Hellenistic influence is seen through the use of the proscaenium. The Roman theatre also had a podium, which sometimes supported the columns of the scaenae frons. The scaenae was originally not part of the building itself, constructed only to provide sufficient background for the actors. Eventually, it became a part of the edifice itself, made out of concrete. The theatre was divided into the stage (orchestra) and the seating section (cavea).
[fg]jpg|Fig 5 - The theatre stage exposed by excavation at Blacklands in Trench 3. About five metres across the stage the stage is adjacent to the freshwater springs, and we are looking west with the seating behind us which overlooks the stage and has a view of the springs beyond. At the back of the stage the remains of a flint and tile wall, the scaenae frons, has survived as a truncated foundation (indicated by a red arrow).|Image[/fg]
In front of the stage, itself built of chalk and mortar on a clay base (see the green arrow in Fig 5), can be seen a slot possibly for the curtain or a retaining timber wall. The floor of the orchestra, the area by the 2m scale is littered with Roman pottery and building ceramics. A coin retrieved from the surface of the orchestra is of Constantine and dated to AD330-335. Note the postholes with white tags which may be the remains of an earlier stage.
Both in Trench 2 and 3 the rear wall (scaenae frons) was exposed by excavation. The 1m scale visible in Fig 6 lies alongside the truncated remains of the rear wall. The surface of the stage is visible, as is the drop to the orchestra. At the rear of the wall, a thick layer (1.27m) of deposited ‘mud’ was investigated which exposed the still bubbling springs. Finds retrieved from this deposit included votive offerings and a large number of broken pots thought to be ritual deposits.
[fg]jpg|Fig 6 - The rear wall exposed by excavation. The photo is facing east towards the orchestra and seating area (cavea).|Image[/fg]
This Roman pagan religious sanctuary site is recorded in the Anglo-Saxon charters[fn2] of AD 699 (BCS 99) and AD 815 (BCS 353) and Gordon Ward notes that: “The [Anglo-Saxon] use of the word ealh meant a pagan temple or, in more general terms, a sanctuary”.[fn3]
The site is considered to be of National Importance as it is the only known example in Britain of a Roman Rural Religious Sanctuary with a theatre built into the hillside.
[fg]jpg||Image[/fg]
[fg]jpg|Fig 7 and Fig 8 - The lip of the Roman theatre with a diameter of 65m can be clearly seen in both photographs. Fig 8 shows how the cavea of the theatre curved round towards the stage. No seats survived but the slope was turfed in the Roman period. In the trees beyond are the ‘sacred’ springs. To the right (north) the apse of Building 1 can be seen, and in the foreground the remains of Roman cobbling and post holes of the post-Roman building.|Image[/fg]
The post-Roman use of the site is of a probable large timber hall built on top of the mosaic floor of Building 1 and dated by pottery to the late 5th century AD. Anglo-Saxon activity from the 7th century is attested by two burials inserted into the fabric of Building 1.
A series of eight parallel postholes had been cut through the southern flint wall of the bathhouse measuring on average 600mm in diameter with a depth of 640mm. The length of the run of postholes was 16.5m with the east end of the timber building presumably turning north along the edge of the two stone platforms for 6.5m. To the west the postholes continued in a straight line beyond the curve of the Roman apse and possibly into an area of ground as yet unexcavated. The timber building could be longer than the 16.5m exposed.
Two fresh pottery sherds found in one of the postholes were dated by Malcolm Lyne to Middle Saxon whilst Nigel Macpherson-Grant suggests late 5th century, and notes that as they were fresh, and are most likely associated with the post-Roman timber building.
This vast post-Roman timber building erected sometime from the late 5th century on top of the relict full colour mosaic floors of the Roman bathhouse, recalls the legend of Beowulf, some say set in the adjacent watery landscape of the Swale Estuary with its known island of ‘Heorot’ (Harty) located across the Swale estuary from the site at Blacklands. This is where Wealhtheow queen of Heorot crossed a fagne flor to offer Beowulf and his companion’s refreshment. Margaret Gelling points out that the road to Heorot, a straet is an Anglo-Saxon loan word meaning Roman road and fagne flor is a mosaic floor, whilst Wealhtheow is a Romano-British noble woman in an arranged marriage.[fn4]
[fg]jpg|Fig 9 - View of Building 1 looking to the east. Three post-hole slots of the post-Roman timber building can be seen (indicated by red arrows). Four had been chopped through the flint and mortar foundations of the Roman building whilst some of the other post-holes were only slight depressions or pads.|Image[/fg]
Another, more prosaic hypothesis is that this building is an example of a rare type of post-Roman timber building recognised by Tony Wilmott at Birdoswald on Hadrian’s Wall.[fn5] The sizes of the timber building at Blacklands and Birdoswald can be seen in the chart below.
It may be that the timber building at Blacklands belongs to a type also recognised at Yeavering by Hope-Taylor in 1977.[fn6] Few major timber buildings of late 5th-6th century date are known in Britain, and apart from Birdoswald, there are the timber buildings at Wroxeter and Cadbury Castle in Somerset.
Excavators of these sites have suggested that the halls “could have been the residence of a royal official, noble, or chieftain”[fn7] whilst Alcock suggests it was “the feasting hall of whatever noble warrior lived at Cadbury with his war-band”.[fn8]
[fg]jpg|Fig 10 - Reconstruction drawing of Area A in Period 6 at Birdoswald showing the hypothetical timber-framed post-Roman cruck-framed structure built on the base of a Roman granary at the Roman fort on Hadrian’s Wall (courtesy of English Heritage) As in Wilkinson, 2012.|Image[/fg]
Demolition of Building 1 continued during the medieval period with a skillet handle dating from c.1370-1500AD found in a robbed out Roman wall, and fresh sherds of various medieval cooking pots from AD1200-1300 were found in another area of robbing within the building.
In the 12th and 13th centuries most of the fabric of the Roman buildings was removed, and some of the material was used to construct kilns discovered in the field to the north of the Roman site.
[fg]jpg|Fig 11 - Site drawing of Building 1. As published in the Kent Archaeological Field School, The Roman Religious Sanctuary at ‘Blacklands’, School Farm, Graveney Road, Faversham, Kent, 25/12/12.|Image[/fg]
The archaeological excavations at Blacklands Field, School Farm have confirmed the presence of an important Roman Religious Sanctuary constructed originally in the 1st century AD, with occupation and alteration continuing beyond the 5th century AD. With the archaeological investigation of the adjacent Roman villa’s, and other Roman buildings known in the vicinity, it seems a substantial Roman hinterland was established soon after the conquest in AD 43 and continuously occupied until at least the 5th century in the area of Faversham and Sittingbourne. Fieldwork in the environs of the site suggest that the landscape was laid out with Roman field measurements, with Germanic and Anglo-Saxon layers added later. The surrounding features and buildings at Blacklands have had only limited excavation, and if protected from deep ploughing, further investigation may offer valuable new insight to archaeologists.
Dr Paul Wilkinson is the director of Swale and Thames Archaeology, the founder of The Kent Archaeological Field School and a published author on topics including Pompeii, Beowulf and Archaeological practice.
[All photographs courtesy of the author, and with permissions obtained, 2025.]
[fn]1|Wilkinson, P., 2000 Swale Archaeological Survey. KAFS website PDF.[/fn]
[fn]2|BCS = Cartularium Saxonicum: a Collection of Charters Relating to Anglo-Saxon History, ed. W. de G. Birch, 3 vols. (1885–93).[/fn]
[fn]3|Ward, G., Archaeologia Cantiana Vol XLVI (1934)[/fn]
[fn]4|Wilkinson, P., 1998 Beowulf in Kent. Faversham Paper No. 64[/fn]
[fn]5|Wilmott, T., 1997 Birdoswald English Heritage (Archaeological Report 14).[/fn]
[fn]6|Hope-Taylor, Brian. 1977. Yeavering: An Anglo-British centre of early Northumbria. English Heritage. Liverpool University Press, Historic England[/fn]
[fn]7|Ritchie & Breeze, 1990 Scotland BC HMSO[/fn]
[fn]8|Alcock, L, Alcock E A & Foster, S M 1986 Reconnaissance excavations on Early Historic fortifications and other royal sites in Scotland, 1974–84: 1, Excavations near St Abb's Head, Berwickshire, 1980, Proc Soc Antiq Scot, 116 (1986), 255-279[/fn]