See important guidance on the use of this record.

If you have any comments or new information about this record, please email us.

Summary

Cadbury Castle is an Iron Age univallate hillfort with additional earthen ramparts situated on the summit of a prominent hill.

Location

Protected Status

• Scheduled Monument 1019540: Cadbury Castle
• SHINE: Cadbury Castle. Iron Age hillfort north east of Cadbury

Other References/Statuses

• Old DCC SMR Ref: SS90NW/9
• Old SAM County Ref: 98
• Old SAM Ref: 34254

Monument Type(s) and Dates

• HILLFORT (Early Iron Age to Roman - 700 BC to 409 AD)

Full description

Department of Environment, Cadbury Castle (Schedule Document). SDV340571.

The fort consists of a main enclosure defended by a single rampart and ditch with a narrow annexe added on the south west side. The approach was from the south through inturned entrances to both enclosures. A 30 foot well was found in the centre in 1843 containing Roman objects of the 3rd century AD which are at Fursdon House. The interior is occasionally ploughed and the rampart and entrance consequently reduced. Other details: Monument 98.

Ordnance Survey Archaeology Division, SS90NW1 (Ordnance Survey Archaeology Division Card). SDV340579.

English Heritage, 09/03/2015, Cadbury Castle, Devon (Correspondence). SDV357939.

Application for Scheduled Monument Consent granted to install interpretation display panels.

Woollcombe, H., 1839-1850, Woollcombe Manuscript (Un-published). SDV16214.

Visited on 3rd September 1836. Cadbury Castle encloses all of the hill summit with approximately outer slope of outer rampart circa 30 foot, inner slope of same circa 27 foot. Semi-circular entrenchment in west part of enclosure. Plan in manuscript.

Tucker, C., 1848, Untitled Source, 193 (Article in Serial). SDV340570.

A shaft within the fort was excavated by G Fursdon in 1843. Various Roman finds and a 17th century sword were recovered. Tucker suggests some pottery may have come from the tumulus used to fill the shaft. No trace of shaft now. Castle occupied by Fairfax in 1645.

Wall, J. C., 1906, Ancient Earthworks, 581-2 (Article in Monograph). SDV341465.

Cadbury Castle an Iron Age oval shaped fort of outer rampart with escarpment and inner scarp and outer scarp to the west, south and east. Entrances are in northeast and southeast. Other details: Plan.

Allcroft, A. H., 1908, Earthwork of England, 111-2,283-4 (Monograph). SDV11975.

Allcroft describes and illustrated the site. According to him the shaft was 58 feet deep, contained nothing of interest; and could be compared with a 'dene hole'. Other details: Plan.

Tyler, F. C., 1926 - 1927, Cadbury Castle, 296-7 (Article in Serial). SDV340575.

Tyler considers that the castle resembles a moot. There was a central pit.

Montague, L., 1929 - 1932, Untitled Source, 144 (Article in Serial). SDV340580.

Fox, A., 1952, Roman objects from Cadbury Castle, 111-2 (Article in Serial). SDV340586.

The Roman finds from the well or shaft are discussed in detail by Fox. They included coins, beads, bracelets, pot, finger rings etc. The possibility of ritual use of the hilltop in the later Roman period is raised.

Donn, B., 1965, A Map of the County of Devon, 1765 (Reprint) (Monograph). SDV336413.

Shown on Donn's map of 1765.

Ross, A., 1968, Untitled Source, 255-86 (Monograph). SDV340581.

Cambridge University, 1969, Untitled Source (Aerial Photograph). SDV340582.

Other details: Reference number not known.

Fox, A., 1973, South West England 3,500BC - AD600 (Revised Edition), 182,195 (Monograph). SDV16216.

Hogg, A. H. A., 1979, British Hillforts: An Index, 196 (Monograph). SDV7953.

National Monuments Record, 1979, SS9105 (Aerial Photograph). SDV340576.

Aerial photograph shows the site itself and all the surrounding fields under cultivation. The inner bank shows very well on this view. Other details: SS9105:SF 1510/330 copy in HER.

Silvester, R. J., 1980, Forts and Farms: The Iron Age in Devon, 47 (Article in Monograph). SDV343544.

The earliest construction phase is represented by a raised platform outlined by a scarp bank on the south side. Inturned entrance on east. Subsequently larger and more substantial rampart followed line of old rampart on north but enclosed a larger area on south east and west.

Timms, S. C., 1981, Untitled Source (Site Visit). SDV340572.

The field Ordnance Survey Number 2120 on the south west side of the hillfort has been ploughed right up to the base of the main rampart, the counterscarp bank being thus reduced on this side. Tractor tracks cutting into entrance. Interior under grass.

Greeves, T. A. P., 1982, Untitled Source (Personal Comment). SDV340574.

A collection of Prehistoric and Roman finds from Cadbury are in the possession of Mr Fursdon of Fursdon House in Cadbury.

Weston, S., 1983, List of Field Monument Warden Visits 1983 (Un-published). SDV343247.

Griffith, F. M., 1986, DAP/GA, 11,12 (Aerial Photograph). SDV340583.

Griffith, F. M., 1986, Untitled Source (Personal Comment). SDV340577.

Condition stable. Under hay. Some badger damage to south rampart.

Griffith, F. M., 1987, DAP/HW, 4,5 (Aerial Photograph). SDV337520.

Griffith, F. M., 1990, DAP/RM, 11-15 (Aerial Photograph). SDV340584.

Grant, N., 1995, The Occupation of Hillforts in Devon during the Late Roman and Post Roman Periods, 102 (Article in Serial). SDV7954.

Fox, A., 1996, Prehistoric Hillforts in Devon, 25 (Monograph). SDV7958.

Department for Culture, Media and Sport, 2001, Cadbury Castle (Schedule Document). SDV340585.

Cadbury Castle includes a slight univallate hillfort with additional later earthen ramparts situated on the summit of a prominent hill overlooking tributaries of the River Exe. The monument survives as an ovalenclosure defined by a rampart and outer ditch to which a second rampart and ditch have been added at the eastern, southern and western sides. Beyond these lies a further outer bank confined to the south. The inner enclosure measures approximately 120 metres by 95 metres. To the north this is defined by a rampart up to 3 metres high. To the west and south the rampart survives up to 7.7 metres wide and 2.2 metres high. A modern entrance has been cut in the northeastern side to facilitate access but the original entrance probably lies more to the southeast. The construction of the second rampart enclosed a larger area to the east, south and west and consequently the outer ditch was modified to produce a curving flat area measuring up to 12.9 metres wide. The outer rampart to the south, west and east is up to 3 metres high with a modern entrance cut partially through it on the southern side. Around the outer rampart lie the largely buried remains of a ditch up to 9 metres wide by 0.3 metres deep and is visible on all sides of the hillfort. The remains of an outer bank to the south is up to 12.5 metres wide by 0.6 metres high and peters out to the east and west.

Wilkes, E., 2010, Cadbury Castle, Devon: Report and Technical Summary of Geophysical Survey by Magnetic Gradiometry and Earth Resistance Techniques 2009-2010 (Report - Geophysical Survey). SDV349871.

A geophysical survey was carried out on Cadbury Castle over 15 days in 2009-2010 in order to determine if such survey would provide information relating to subsurface archaeological features and to add details to known features, particularly the location of a well or shaft of possible Roman date. The interior of the monument together with a band of grids around the outside was surveyed by magnetic gradiometry, and in addition an evaluation strip of eight 30 by 30 metre grids was surveyed using an earth resistance meter. The surveys recorded anomalies of geological character and of archaeological potential. The results demonstrate that the monuments has at least two phases of construction. A third outer ditch circuit, previously unknown, was also identified lying to the south and west of the monument. The exact location of the shaft cannot be stated with certainty but a likely position was suggested (N) which corresponds with a depression in the ground which may be the sunken top of the shaft.

Griffith, F. M. + Wilkes, E. M., 2011, In the Footsteps of Pioneering Women; Some Recent Work on Devon Hillforts (Article in Serial). SDV361500.

Wilkes, E. M. + Griffith, F.M., 2012, Cadbury Castle, Devon, Reconsidered, 237 – 280 (Article in Serial). SDV354394.

Geophysical survey carried out in 2009-10 has shed new light on the history of Cadbury Castle and has both confirmed and amplified Fox's phasing of the site, though since it has not been tested by excavation, and no removal of vegetation was possible for the survey, the actual complexity of the monument could well be considerably greater. At least two prehistoric phases of development have been identified, as well as a previously unknown additional external line of defence which is tentatively suggested to date from the Civil War. The outer bank was seen by Fox as a secondary work, and the results of the geophysical survey bear this out.
Plough activity has greatly truncated the height of the inner bank and filled in the inner ditch, which can however still be seen clearly on the magnetometry plot. There is an entrance to the south, believed by Fox to relate to the second phase, and a modern entrance in the north east. The internal form of the inner bank is shown to be dividedby highly positive partitions spaced fairly evenly around the outer slope of the inner bank. This suggests the bank may be of a segmented box rampart construction. There is a suggestion of a terminal to the south of the entrance of the inner enclosure.
Where the entrance to the phase 2 hillfort encounters the ditch, however, there appears to be a deliberate blocking of the ditch to the left of the entrance.
Outside the upstanding remains of the monument, anomalies F and G are interpreted as ditches. Neither of these features had been previously recorded. The line of ditch F, to the south and west of the monument, but before the main break of slope downwards to the south, is concentric with the upstanding banks and can be suggested as a further range of fortification. There is a break in the ditch with enlarged terminals either side which appears deliberate although it only gives access to the main outer ditch of the hillfort. There are no indications on the ground of a bank between ditches F and E. At point F2 there is a curious dogleg in ditch F, approximately around a visible slight mound in the ploughed field.
Ditch F was not detected as a continuous anomaly around the monument. It is likely that the ditch never existed to the north and east, where the ground slopes much more steeply away from the hillfort and where perhaps the outer defence was not required. However, these are also areas of thinner soils where it is possible that plough activity has removed traces of the ditch if it had run through this area. The end of ditch F does not appear to be a proper ditch terminal, instead suggesting erosion here.
We have never seen a prehistoric ditch surrounding a hillfort with such a pronounced dogleg, and we incline to read this as a later feature. The geophysical signature of the ditch also differs from that of the hilifort ditches, being stronger and more consistent along its length. One possible context for both this ditch and the one or possibly two mounds may be the known temporary use of 'Cadbury Hill' by part of Fairfax's army in December 1645 and January 1646. In this interpretation, the mound or mounds might also relate to the defence of the site at this time.
Around the entire monument there are slight traces of a field system and plough lines. The field outlines radiate from the hillfort and are not seen within it, which suggests they post-date the construction of the hilifort. This is most clearly seen in the north-west area. The pattern radiates through the outer ditch, suggesting that the two are not contemporary, but it does not run into the main monument bounded by ditch E, a significant feature on the ground today which would have provided an obstacle for any later field system or possibly destroyed or masked any earlier traces. The 1947 RAF aerial photograph shows the lines of plough scarring, particularly to the east of the monument, that correspond to the alignments detected in this survey.
In the interior of the monument a number of anomalies were also detected, despite the clear impacts of heavy cultivation. Within the central area of the hillfort, three anomalies could be potential candidates for the site of the well or shaft. A corresponding slight circular depression at anomaly N is the most likely location of the well/shaft as described by Tucker and Fox.
In the north of the monument interior is a series of anomalies of sub-circular form, which may be interpreted as the gullies around structures (possibly round houses): some with possible internal post settings. The fact that they intersect indicates the presence of more than one phase of occupation.
In addition, the finds (bracelets and ceramics) from a nineteenth-century 'excavation' of a shaft or well in the interior have been reviewed and found to represent a significant assemblage of late Roman date, possibly suggstive of the survival of preChristian practices. It seems highly likely that there were aspects of structured deposition associated with the artefacts and ecofacts found in the Cadbury shaft. This deposition can now almost certainly be placed in the last quarter of the fourth century.
Two new fabrics are represented in the ceramic assemblage; a 'local Permian breccias fabric' and a South Devon variant 'granite derived fabric'. The identification of two new fabrics in the fourth century in Devon is of interest. The suggestion of pottery manufacture possibly in the region of Cadbury, as indicated by the local Permian breccia fabric, is unexpected. However a sherd in Taw/Torridge estuary clay and another possibly from the periphery of Dartmoor has recently been identified in a fourth century assemblage on Lundy Island. It may therefore be that in rural areas in the fourth century there was a scatter of sites in Devon manufacturing pottery on a small scale with products exchanged only within the local area.

Griffith, F.M. + Quinnell, H. + Wilkes, E, 2013, Hillforts of Devon, 28-30 (Monograph). SDV352457.

Hillfort which still preserves its massive ramparts. The original embanked enlcosure, which had an inturned entrance on its eastern side, was enlarged by the addition of an outer enclosure around its western and southern sides with an entranceway through on the south side. The original ditch was blocked inside the new entrance directing those entering the fort to turn along the ditch to enter the inner enclosure by its original entrance.
No excavation has taken place within the fort but an assemblage of fragments of 4th century AD copper alloy bracelets, fragments of shale bracelets, beads and 4th century Roman pottery was found by the landowner in a shaft or well in the interior in 1848. The interior of the hillfort has been subsequently heavily ploughed so that it is difficult now to locate the depression marking the location of the well.
Recent geophysical survey has provided detail of a segmented structure of the inner rampart (tree cover prevented magnetometry over the outer rampart), traces of possible roundhouses in the interior, the blocking of the phase 1 ditch by the phase 2 entrance, apparently deliberate kinks at the junctions of the phase 1 and phase 2 ramparts and a previously unknown outer ditch around the south and west sides. It is suggested that this ditch, which has a pronounded dogleg on its south side may be related to the hillforts use by Fairfax's army in 1645-6 during the Civil War.
See guide booklet for more information.

Tilley, C., 2017, Landscape in the Longue Durée, 299-319, tables 9.1-9.5, figs 9.7-9.17 (Monograph). SDV361032.

Discussion of Woodbury Castle and the other hillforts in the locality, making reference to the East Devon pebblebeds.

Carter, S., Unknown, In Search of Camelot - Cadbury Castle, Devon (Article in Serial). SDV355508.

Sources / Further Reading

Associated Monuments

Associated Finds

• FDV5683 - EAR RING (Undated)
• FDV5678 - ROD (Undated)
• FDV5681 - BEAD (Roman - 43 AD to 409 AD)
• FDV5684 - POT (Roman - 43 AD to 409 AD)
• FDV5680 - SPEAR (Roman - 43 AD to 409 AD)
• FDV5682 - FINGER RING (III - 201 AD to 300 AD)
• FDV5677 - BRACELET (IV - 301 AD to 400 AD)
• FDV5679 - METAL OBJECT (Post Medieval - 1540 AD to 1750 AD)

Associated Events

• EDV5935 - Geophysical Survey at Cadbury Castle, Devon

Search results generated by the HBSMR Gateway from exeGesIS SDM Ltd.