MDE1020 - The Beacon Roman Fortlet, Martinhoe (Monument)

A Roman fortlet briefly occupied in the 1st century AD, comprising an inner square enclosure in a subcircular outer enclosure. Excavations in the 1960s revealed the footings of three building ranges and the site of a beacon fire signalling seawards.

The Beacon Roman Signal Station is marked at SS 6630 4933 on the 1963 Ordnance Survey map. [1] Martinhoe Beacon, the Roman fortlet, was excavated in 1960-1961 by Fox and Ravenhill. There is a small square centre enclosure, fortified with a rampart and double ditches, with a third rampart and ditch about 21.34 metres away on the landward side, forming a roughly circular outer enclosure. [2] The inner rampart was of turf, the outer of dump construction made from material from the ditch. The entrance to the inner enclosure was on the seaward side; a metalled causeway led to a double portalled gateway. The outer entrance was on the opposite (south) side, possible remains of signal fires within this enclosure. The inner enclosure contained three timber buildings; a pair of L-shaped barracks on the west and east, and a small rectangular building on the south side. Each barrack contained eight cubicles, and one had three extra rooms added some time later. The smaller building of two rooms had a small domed furnace or forge nearby, and might therefore have been used by a smith or armourer. Immediately within the intervallum were a number of clay and timber field ovens. Pottery from the excavations dated occupation to c.AD 55-75, and included Samian forms 15/17 and 18, one stamped Licinus (AD 46-65), stamped Mortarium, and native ware. Also found were two bronze coins of Nero. The fortlet is considered to have been occupied on a semi-permanent basis for about 10-20 years within the period AD 55-75, and to have been associated with the Roman Conquest of the Silures (a South Welsh tribe). There would have been sufficient room for a century of soldiers. [3] The Roman fortlet at Martinhoe has been much reduced by ploughing. Excavations in 1960/61 showed it to bear some resemblance to Hod Hill, consisting of a small central enclosure surrounded by a rampart and two ditches, and an outer enclosure formed by a rampart and ditch terminating on the cliff edge at each end. Within the outer enclosure, near the edge of the cliff and below the hill crest, was the site of a series of bonfires, indicating a beacon intended to be visible at sea rather than inland. The interior of the fortlet contained a rectangular building thought to be an armoury and two barrack blocks with accommodation for a century. Finds of Samian and coarse pottery and two coins of AD 64-8, indicated that occupation commenced in the reign of Nero and probably continued until the foundation of Caerleon circa AD 75. The excavation showed that Martinhoe succeeded the temporary fortlet at Old Burrow (MDE1223) and it therefore seems less likely that either formed part of a chain of signal posts. They were clearly placed to command an extensive view across the Bristol Channel and alert patrol ships in the event of activity by the Silures. [4] The fortlet at Martinhoe replaced, in more permanent form, the fortlet at Old Burrow with the same function of watching for landings across the Bristol Channel. [5] The earthwork is heavily ploughed; the central square enclosure is 27.43 metres in diameter, and barely visible on the surface. It is surrounded on three sides by a rampart and ditch, the fourth side being the eroded edge of the cliff. The site commands an extensive view over the Bristol Channel. The interior excavation has been infilled; the second rampart is traceable to the north and west as a bank and to the east and south as a scarp. The outer rampart is best preserved on the west side. Neither the inner nor outer entrance can be traced as a surveyable feature. The fortlet is in poor condition. Resurveyed at 1:2500. [6] The fortlet, at SS 6631 4934, is basically as described by previous authorities. The whole site is now in a rather poor condition. It has been cleared and fenced but, being ungrazed, the bracken and thistles are slowly returning. As stated the area has been much reduced by ploughing and the scarps are not quite as clear as the Ordnance Survey plan [6] suggests. The internal dimensions of the central sub-rectangular enclosure varies from about 14 metres east/west to 22 metres. [9] McDonnell [7] refers to an enclosure defined by a low bank, as being larger than that shown by the Ordnance Survey [6]. It is visible on 1946 aerial photographs [8], but doubtful on the aerial photographs of 1978 [9]. This low bank is the outer rampart around the western side and the Ordnance Survey plan [6] was checked and found to be accurate. The reconstructed plan of the fortlet [4] is not a true survey and gives a false impression of uniformity and symmetry. [10] This is an elliptical earthwork. It is undoubtedly a signalling station. There is a central enclosure, damaged by modern vandals and frequent use as a bonfire site. It has a bank measuring 2 feet 6 inches high, and an outer bank measuring 3 feet high. [11] The site lies on rough ground just south of the coastal footpath. It is covered in gorse, but otherwise appears to be in stable condition. Finds from the Fox and Ravenhill excavation are on display in the Barnstaple Athenaeum. [12] The site is virtually invisible on the 1983 aerial photograph. At Martinhoe, traces of the foundations of timber buildings were identified in excavation, leading the excavators to suggest that this site was in use for a much longer period, possibly replacing the site at Old Burrow. [13-15] Wilson classes Martinhoe amongst his "2a" category of forts; where an outwork entirely surrounds a fort. He comments on the contrast between the square plan of the fort and the circular outer enclosure. He suggests that the lack of buildings in the outer enclosure means it was possibly used for beacon material and stores. [16] The condition of the site is stable apart from a small area of erosion on the outer bank, on the northwest side. The area is being grazed by sheep. [17] Martinhoe Roman Fortlet was surveyed at 1:500 scale by RCHME staff in July 1997. The earthworks of the inner square enclosure have been spread by ploughing, and have also been distorted into a parallelogram shape. The inner enclosure now survives as a substantial flat topped mound with very spread scarps. Its entrance could not be located on the ground. The outer enclosure survives best on the west and southwest, where it is visible as a well defined bank with traces of an outer ditch. Along the south and east, it becomes gradually fainter. The northern part of the outer enclosure was either never constructed or has slipped over the cliff edge. The site is now fenced off separately from the surrounding fields and is subject to low level management by the National Trust, which includes strimming and mowing the sward on the earthworks. [18-21] The shape and poor preservation of the earthworks can in part be explained from the aerial photograph, which shows an area of substantial medieval ridge and furrow to the east of the fort. It originally ran much closer to or even across the site.[39] The fortlet at Martinhoe is visible on aerial photographs as a series of low earthworks comprising the structures described above. There are no traces of any interior structures visible on the photographs. Due to the extent of plough damage the features are best viewed on oblique photography. The site is clearly also at risk from coastal erosion, and it appears from the aerial photographs that the northern bank and ditch of the outer enclosure have already been lost over the cliff edge. [40-42] The height called Martinhoe was prominent enough to give its name to the parish and is visible over much of northeast Devon and from the great Somerset beacon of Dunkery. It is an elliptical work with an outer vallum about 900 millimetres high and a small central ring with a damaged vallum about 760 millimetres high. [43] The Scheduled Monument Condition Assessment of 2009 gave the site a score of 5. [45] The site was surveyed in May 2015 as part of the 2015 Exmoor Scheduled Monument Condition Assessment. It was given a survival score of 9. [46] Monument record reviewed as part of NRHE to HER pilot project. [47,48] A paper suggests that the landscape position of the fortlet implies it was intended to strengthen preexisting measures to counter coastal raiding. [49] The site has extensive views of the Bristol Channel but only a limited view inland and is not intervisible with Old Burrow (MDE1223), with which it may or may not be contemporary. Jones disputes that the forts were positioned to deal with a naval threat and notes that visibility at both sites often does not allow a good view of Wales apart from in the very best weather conditions. He instead suggests the sites may have acted as observation posts or headland warning beacons, with nearby safe haven protection and possibly with a "lighthouse" capability. [50] A Roman road network system in South West Britain was mapped using Lidar and GIS spatial analysis producing a prediction of the likely layout of the road network. Routes are suggested from Martinhoe Beacon east to Old Burrow and south to Herner [51]