See important guidance on the use of this record.

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

Summary

A hillfort of Iron Age to Roman date is visible as earthwork bank and ditches on aerial photographs of the 1940s and images derived from lidar data, defining a possible multi-phase site enclosing the hilltop under Lifton Wood. It is suggested, given the strategic importance of Lifton in the late Anglo Saxon period, that it may have seen a period of reuse at this time and that the outworks may in fact be Saxon in date.

Location

Protected Status: none recorded

Other References/Statuses: none recorded

Monument Type(s) and Dates

• HILLFORT (Iron Age - 700 BC to 42 AD)

Full description

Royal Air Force, 1946 - 1949, Royal Air Force Aerial Photographs (Aerial Photograph). SDV342938.

Possible hillfort or enclosure visible on GIS RAF aerial photograph. Bank and ditch forming an elongated enclosure. The Lidar highlights the bank and ditch along the southern part of the earthworks.

Environment Agency, 2000-2019, LiDAR DTM data (1m resolution) EA: Tamar Aerial Survey project area, LIDAR Environment Agency LAST RETURN 24-NOV-2016 (Cartographic). SDV363954.

Curvilinear banks and ditches and irregularly shaped pits are visible as earthworks.

Environment Agency, 2011, LIDAR data JPEG images (2 metre resolution) (Cartographic). SDV348634.

Possible hillfort identified on 2011 Lidar. Bank and ditch is visible. The south-west and north-east side of the bank and ditch is ephemeral and seems to respect the line of the contour.
Revisualation of this data using Hill-Shading in 16 directions revealed further earthworks across the summit of the hilltop.

Horner, B., 2014, Possible Hillfort, Lifton Wood, Lifton (Aerial Photograph). SDV357458.

Possible hillfort identified on 2011 Lidar in Lifton Wood to the south of Lifton. If an enclosure, it is perhaps more likely to have been associated with the later Anglo Saxon settlement of Lifton than the small enclosure of Castle Farm which lies about 1.5 kilometres to the east.

Hegarty, C., 2015, Comments arising from the transcription of earthworks at Lifton Wood from Lidar data (Interpretation). SDV358492.

A probable hillfort of Iron Age to Roman date is visible as earthwork bank and ditches on images derived from lidar data, defining a possible multi-phase site enclosing the hilltop under Lifton Wood.
The hillfort might comprise two or more phases of enclosure. A ditch and bank defined oval enclosure circa 80 by 40 metres in size towards the south-eastern end of the hilltop. A possible rectilinear or trapezium shaped annexe abuts the north-western side of the oval enclosure, taking in much of the summit of the hill. Possible linear outworks are visible on the north, south and north-western sides of the hillfort, the latter possibly defining two or three cross-ditches on the gentler approach to the hillfort. The earthwork ditch identified as a possible outwork on the south-west facing slopes might comprise an enhanced natural feature, or could simply be use of natural geological banding.
A probably post-medieval to 19th century quarry is located on the north-western tip of the hilltop spur and this, and a possible trackway from it to the south-east, might have impacted on the hillfort earthworks.

Horner, B., 2015, Possible Hillfort, Lifton Wood, Lifton (Ground Photograph). SDV358216.

Images showing vantage point of possible hillfort from road and internal features.

Gent, T. + Manning, P., 2016, Lifton Wood and the Development of the Surrounding Area. (Report - Assessment). SDV359657.

Desk-based assessment carried out between August 2015 and February 2016 to assist in the interpretation of a recently rediscovered hilltop enclosure and other features in Lifton Wood.
Lifton was one of the earliest villages to be founded in West Devon and was of considerable strategic importance, being close to the Tamar and the Cornish frontier, and on the major route, and Roman road, from Exeter (and beyond) into Cornwall. Lifton was a West Saxon royal estate, the head or centre and meeting place of Lifton Hundred, and possibly the site of an early minster. The first reference to Lifton is in King Alfred's Will (circa AD880) when it is called 'Liwtun'. It is referred to as 'the well known town' in a charter of 931 when it hosted the Witan assembly attended by King Aethalstan and his court, together with 100 magnates.
The hilltop enclosure in Lifton Wood comprises a ditched and lightly banked enclosure with a single entrance on the northwest side and which shows signs of elaboration and extension to the southwest. Although the construction and development of the enclosure has the potential to range from the Late Bronze Age to the Romano-British period, it is most likely to be Iron Age in origin. However, it is suggested that the outworks may date to the Saxon period; a number of hilltop sites in Devon have produced evidence for post Roman occupation. Given the importance of Lifton in the Saxon period, the reuse of such a hilltop site seems a reasonable supposition.
It is suspected that geophysical survey has a good chance of adding to an understanding of the earthworks (see report for full details).

Newman, P., 2016, Lifton Wood Hilltop Enclosure, Devon (Report - Survey). SDV359658.

An earthwork survey was commissioned by Bill Horner, Devon County Archaeologist, following encouraging results from a LiDAR transcription.
The hilltop enclosure occupies the broad, eastern end of the plateau. It comprises an irregular, closed circuit of earthworks, narrower at the western end but broadening and more bulbous to the east, with a perimeter of just under 300 metres, enclosing an area of 0.62ha. The interior of the enclosure is 110 metres long by a maximum of 80 metres wide.
Although the circuit appears complete, the character of the rampart varies. The most clearly defined and strongest section forms the southeast sector of the circuit (a). This comprises a partly silted ditch of up to 8.3m wide and 1.6m deep. A spread bank (a1) of up to 3 metres wide runs parallel with the interior edge of the ditch but is only 0.13 metres high at the southern point of the rampart; it becomes progressively higher to the east, developing into a more substantial bank of 0.6 metres high at the far eastern end, where it has been cut through by recent disturbance and now terminates (b). A counterscarp bank survives along the exterior (a2); it has become spread to almost 5 metres wide in places, and is on average only 0.3-0.4 metres high.
The southwest sector (c) was less developed than that to the southeast; here, the perimeter is defined in places only by a scarp. At its shallowest, this scarp is 0.7 metres high but it increases to 1.1 metres as it progresses to the northwest. There is no trace of an interior bank and the ditch is visible only over a short distance as an extremely slight scarp of 0.15 metres high. A counterscarp is visible but running parallel only with the western 25 metres stretch of the ditch (d). On the northwest run of the rampart, the substantial ditch (e) is again visible approximately 8m wide and up to 1.5 metres deep. However, the inner rampart bank (e1) survives only in patches to barely 0.2 metres high, though to the exterior of the ditch there is a continuous counterscarp bank of 0.2 metres high (e2). It terminates at a circular mound with a hollowed centre (f), which clearly represents a later intervention. The top of this mound is at the highest point of the Lifton Wood summit and corresponds with the approximate location of what is assumed to be a monument, known to have existed from a 19th century illustration (Gent and Manning 2016, 7), and depicted on the tithe map of 1840 as a small circle. The diameter of the top of the mound is 6.5 metres. Nothing remains of the monument, but the mound is strewn with fragments of shillet.

The base of the mound, which has a spread of 15 metres, has obscured the original terminal of the counterscarp and part of the ditch, at the site of what was probably the enclosure’s entrance, located in the angle at the northern sector of the circuit (g). This is visible as a short causeway, raised very slightly across the base of the ditch. At this point the scarp of the rampart is interrupted by a breach of approximately 3 metres wide with rounded terminals either side. The interior of the entrance is defined by a slight, semi-circular hollow between the two terminals, which intrudes into the interior space of the enclosure.

The northern side of the enclosure is the most complex section. An inner scarp (h), which includes the bullnose of the entrance breach, runs east for 37 metres , before curving towards the interior and fading out. At the foot of this scarp to the north, a shallow ditch (h1) has been formed by a substantial counterscarp (h2) running parallel with (h), which extends for 30 metres to the east but then disappears where a more dominant slope (i) takes over. This slope has itself been intruded into by a linear hollow (j) of 6.5 metres wide by 0.5 metres deep, running east to west from the outer rampart, but terminates after only 22 metres. The exact nature of this part of the rampart has proved difficult to interpret but has been further confused by the imposition of a later linear boundary bank, with inner ditch (see ‘t’ below), which runs along the foot of the scarp and may be disguising an external ditch and counterscarp.

Further to the east, as the rampart rounds its northeast corner, a slight inner bank again becomes visible (k) as does the lip of the counterscarp (m), representing fragmented remains of a ditch of 8 metres wide (consistent with the width elsewhere on the circuit) albeit the ditch has the later bank (t) covering much of it.

To the south of the enclosure, a natural escarpment has been adapted to form a possible extension (n). On the southwest side, this escarpment runs approximately in parallel with the main rampart, 15 metres away and at a lower altitude. It is up to 10 metres wide in places, with a drop of over 2 metres; the top and bottom edges are well defined, particularly at the southernmost point. Traces of a very slight bank are visible along the upper lip of the slope, becoming clearer near the western corner, from
where a substantial artificial scarp takes over (o), rounding the corner and continuing along the northwest side.

From the southern point of the natural escarpment (p), an artificial scarp of 0.5 metres high (q) branches at near right angles and runs to the northeast, joining the rampart near the eastern end of the enclosure. This combination of natural (n) and artificial (q) slopes forms a possible outer extension, and although slight, their character suggests these earthworks may have supported a light timber palisade, enclosing an additional 0.17ha attached to the southern exterior.

The interior of the enclosure has few, clear earthwork features that might be contemporary with the rampart. However, near the proposed entrance a subtle, circular hollow (r), with a diameter of 8.5m, has to be considered as a candidate for a hut platform.

On the broadest section of the enclosure a low, 4m-wide, flat-topped, linear bank (s) with, in places, a slight ditch on both sides, runs across the interior from northeast to southwest, dividing the enclosure into two sectors. It is first visible at the lip of the natural escarpment (n) and as it extends northwest it overlies the southern rampart (a) and is clearly later. To the north, this feature blends into the scarped corner of the cutting (j). This bank exists only within the confines of the enclosure and its southern extension. No trace of it could be found on the exterior and it appears to have been created specifically to divide the interior space of the enclosure. It is possible also that it worked in conjunction with the boundary of the southern extension (n), to which it joined at right angles, though this is much degraded. The remains of this bank (s) are notably more spread than other linear boundaries traversing the hilltop, and probably disused for far longer.

This survey has confirmed the significance of the earthworks previously recorded by LiDAR transcription, whilst also revealing their complexity and offering a basis for interpretation. The single ditched enclosure was probably constructed in the Iron Age. Its hilltop location, single ditched rampart and layout is typical of settlements of that date in Devon and elsewhere. However, although there is no evidence as yet for earlier settlement at this place, it is quite feasible that earlier 1st or 2nd millennium BC occupation may have preceded that of the Iron Age. The defensive circuit has proved to be mostly intact, though the character of the rampart is inconsistent as a whole and, in places, the earthworks have been altered by later interventions. For the strongest sections, on the southeast and northwest sectors, a moderate ditch survives, though rampart banks and counterscarp banks, where evident, are mostly small in scale. Traces of a ditch are visible along the northern flank, though there is no trace of it as the circuit progresses southeast, where it may have become effaced when a later field boundary was constructed along a similar course. For the southwest sector, the ditch is mostly absent, or extremely slight, and rampart banks are not present. A single entrance has been identified on the northwest corner of the rampart, and a part-natural, part-artificial earthwork on the southwest exterior, must be considered as a possible later extension to the enclosure. The interior contains few earthworks likely to be contemporary with the ramparts, although a subtle, circular hollow near the former entrance, is perhaps a candidate hut platform. A linear bank which transects the enclosure, is later.

Although sited on a prominent hilltop, and possessing ‘defences’, the term ‘hillfort’ has been avoided in this report, in favour of ‘hilltop enclosure’, because the former somewhat overstates the character of the remains and the likely purpose of this site. The ramparts, such as exist, comprise only a single ditch, which varies in strength around the perimeter. A rampart bank, to augment the height of the ditch, is absent on most of the circuit, and where traceable, along the southwest sector, appears to have been very slight. Where traces of a counterscarp exist, they too are slight, and although now spread, they would offer very little in defensive terms. It seems likely that protection of the enclosure’s interior relied on a mostly shallow ditch, surmounted by a timber palisade. The total enclosed area is relatively small at 0.62ha; on its own this would give no grounds for arguing against this being a hillfort, but combined with the weak defences, it becomes a factor. Fox (1952) provided us with the term ‘hillslope enclosure’ to describe a particular group of lightly defended, Westcountry,
Iron Age sites located in elevated positions. Today this term is often applied to sites of similar character to Lifton Wood, such as Myrtleberry and Staddon on Exmoor, which, despite being smaller than Lifton Wood, are comparable in strength and in their spur locations. However, hilltop enclosure is favoured here and accords with the Historic England Thesaurus term for the ‘hilltop palisaded enclosure’, which is defined as follows: A small, defended settlement dating to the Iron Age, located on spurs, promontories or hilltops. The defences are marked by single or double trenches which originally held substantial palisades. Post-abandonment activity on this hilltop comprises a series of linear land divisions of medieval or later date and stone quarrying, which is most likely to have occurred in the later 2nd millennium AD. Trackways cut across the northern slopes of the hill, below the enclosure earthworks, and a monument, known to have been standing in the late 19th century at the highest point of the summit, has also left its mark in the form of a circular mound.

Hegarty, C., Houghton, E., Knight, S. and Sims, R., 2020-2021, Tamar/Lidar; A Single Source Approach to Landscape Survey and Socially Distanced Community Archaeology Area 2 (Culture Recovery Fund project) (Interpretation). SDV364011.

A possible hillfort of Iron Age date is visible as a series of curvilinear banks and ditches and irregularly shaped pits on visualisations derived from lidar data captured in 2016.
Two possible phases comprising an inner curvilinear enclosure and outer trapezoidal enclosure are evident.
The inner enclosure occupies an area of circa 0.55 hectares and takes in the summit of the hilltop. It is defined by a ditch, between 4-7 metres wide, with inner and outer banks, between 2-4 metres wide. A possible entranceway is visible along its north-west side, where this part of the enclosure is defined by a straighter edge.
The outer enclosure occupies an area of circa 1.5 hectares and is similarly defined along its south-east and south-west sides by a ditch, between 5-8 metres wide, and sections of inner and outer banks, circa 4 metres wide. Its north-east side more closely follows the defences of the inner enclosure where it has been disturbed by earthworks of a later field boundary (MDV112193). The nature of the earthworks that define the north-west side of the enclosure are more confused here where they have possibly been impacted by post-medieval or 19th century quarry pits (MDV73250), but possibly define between 2-3 cross-ridge entranceway defences.

Newman, P., 2022, Lifton Wood Hilltop Enclosure: An Earthwork Survey, 125-13; Figs 1-4 (Article in Serial). SDV365484.

In 2015, an earthwork survey of the earthworks was undertaken following observation of earthwork features on lidar imagery.
The survey confirmed the significance of the earthworks, and revealed their complexity. The univallate enclosure was probably constructed in the Iron Age. However, although there is no evidence as yet for earlier settlement here, it is quite feasible that earlier 1st- or 2nd millennium BC occupation may have preceded.
The defensive circuit has proved to be mostly intact, though the character of the rampart is inconsistent as a whole and, in places, the earthworks have been altered by later interventions. For the strongest sections, on the south-east and north-west sectors, a moderate ditch survives, though rampart banks and counterscarp banks, where evident, are mostly small in scale. Traces of a ditch are visible along the northern flank, though there is no trace of it as the circuit progresses south-east, where it may have become effaced when a later field boundary was constructed along a similar course. For the south-west sector, the ditch is mostly absent, or extremely slight, and rampart banks are not present.
A single entrance has been identified on the north-west corner of the rampart, its position probably determined by the topography, providing a long, level approach along the hilltop. A part-natural earthwork on the south-west exterior must be considered as a possible later extension to the primary enclosure.
The interior contains few earthworks likely to be contemporary with the ramparts, although a subtle, circular hollow near the former entrance, is perhaps a candidate hut platform. A linear bank which transect the enclosure is later.
Post-abandonment activity on this hilltop comprises a series of linear land divisions of medieval or later date and stone quarrying, which is most likely to have occurred in the later 2nd millennium AD. Trackways were cut across the northern slopes of the hill, below the enclosure earthworks, and a monument known to have been standing in the late 19th century at the highest point of the summit has left its mark in the form of a circular mound. Recorded dates for its construction or demolition are not known.
Lifton Wood is one of several small, ‘defended’ earthwork enclosures known in this area of west Devon, and spread along the Tamar valley on the Devon side. Of these sites only Northcott Wood Camp (MDV2760) is of comparable dimensions, with most being quite a bit smaller. Some of these sites are in poor condition with minimal upstanding earthwork evidence, so the very visible remains at Lifton make its discovery an important addition to the late-prehistoric earthworks in this part of Devon. It is also the only one of these enclosures to be sited on a notable prominence.

Gent, T. + Manning, P., 2022, Lifton Wood Reused?, 135-145; Figs 1-3 (Article in Serial). SDV365485.

A desk-based assessment and walkover survey of the enclosure and surrounding area was undertaken by the authors between August 2015 and November 2016, and an earthwork survey was carried out by Newman in August 2015 and January 2016. Newman identifies a univallate enclosure of probable Iron Age date, with evidence of limited redevelopment and enhancement. Remains of a medieval field system meet, and in places overlie, the enclosure. Following updated lidar surveys, a potential defensive bank was identified, encircling much of the hill. A further field visit by the County Archaeologist and one of the authors, in August 2022, confirmed the survival of this bank.
The adoption of prehistoric hilltop enclosure in the Saxon period, as both burhs and the administrative centres of hundreds, is recognised, and it is known that Athelstan convened a meeting of the Witan at Lifton in 931. The possibility exists that a similar Saxon redevelopment also took place in Lifton Wood.
Lifton was a West Saxon royal estate, the head or centre and meeting place of the Lifton Hundred. Land bequeathed (circa 880) by King Alfred to his younger son, Aethelweard, includes that of ‘Liwtun’, and there is strong evidence to suggest that this can be identified with Lifton. A source dated 12th November 931 deals with a meeting of the Witan which was held by Aethelstan ‘in the town … called Leowtun’ (Lifton). The Lifton Witan was probably one of the largest assemblies held in the Anglo-Saxon period.
The presence of earthworks in Lifton Wood has been known since at least the early 19th century, when a lease of 1816 refers to what is now Lifton Park as Castle Park. It is possible the earthworks were referenced much earlier, if the ‘dun’ element of the local name Dunshill is derived from the British ‘duno’, meaning ‘fortified place’, rather than the more common Old English element ‘dun’ (meaning ‘hill’). Either way, it is probable that Dunshill was once the name of the hill of which Lifton Wood is part.
Although Newman identifies a southern extension to hilltop enclosure, this enhancement may have been more ambitious. While recent quarrying on the ridge is undoubtedly responsible for a number of features to the north-west of the enclosure, particularly those at some distance from the ramparts, it is suggested that at least a proportion may represent defensive additions, possibly the result, with the work to the south, of a unified phase of development. Considering the parallel and equidistant spacing of a number of the features on the ridge, further complexity and potentially more than one line of defensive banking may be represented. In terms of the date of these works, and in light of other suspected activity on the hill, a Saxon rather than a prehistoric initiative should be considered a possibility.
Bank W, first located by Newman’s survey, approximately 150m to the west of the enclosure, measures 1.5m high and 11m wide. The north end turns to the east and extends along the ridge for a short distance towards the enclosure before terminating at an inturned gateway. The south end of the bank also turns, and continues to the south-east. Due to the tight vegetation, the line of bank W to the north and north-east of the enclosure was not identified. However, it is strongly suspected that this bank lies along the northern edge of the ridge, extending north-west away from the enclosure before turning in to form the eastern side of the same gateway. This would result in a circuit that hugs close to the northern edge of the earlier enclosure.
A number of linear boundary features approach the ridge and enclosure from the south, in two cases overlying the ramparts. These banks are seen as the remains of a medieval field system, with one possibly representing a deer park boundary. Bank U differs from the others, following a distinctly sinuous course across the hillside. Although broken in places, most significantly by the buildings of Home Farm, it extends from the ridge as far as the River Lyd. At up to 4m wide, with evidence of a former ditch to either side, this winding boundary is much more substantial than others on the slope. On the basis of character and dimensions, it is suggested that Bank U represents a significant and early land division, and one that is likely to have influenced the character and location of all other land divisions on the hillside. Physical relationships show that it post-dates features of proposed Saxon date, and based on its general character it is thought likely to be medieval in date, also possibly early.
One of the two trackways which approach the enclosure (Track X) is within a cutting that appears to predate the medieval field system extending from the south. Dropping east away from the ridge, and following the northern boundary of Lifton Wood to meet the valley bottom at Lifton Bridge, it is suggested that this entire route from the river crossing to a potential early medieval entrance, may be contemporary. The track might even date to the initial prehistoric use of the ridge.
It is believed that the Iron Age enclosure of Stanborough Camp was utilised by Alfred as the location for one of possibly a dozen ‘emergency’ Devon burhs. Although the evidence is circumstantial, it is suggested that the enclosure in Lifton Wood, part of a royal estate, was given the same function, and that this already fortified military establishment, rather than Lydford, represents the most westerly burh listed in the Burghal Hidage. This certainly fits a demonstrated pattern of Alfredian reuse of earlier enclosures. As part of the reorganisation that saw the Stanborough and Pilton burhs moved respectively to Totnes and Barnstaple, it is suggested that this burh at Lifton may either have been relocated to Lydford or, more likely if the outer defensive bank is indeed Saxon, the two were coexistent. While the pressure to protect valuable in resources on the north-western edge of Dartmoor no doubt remained, or even grew, in the late 10th century, the motives behind the suggested establishment of an earlier burh at Lifton are unlikely to have become less significant. It is known that Aethelstan convened a great Witan at Lifton in 931, and it is argued a current or former burh would have been a fitting place for such a meeting. With commanding views over the River Tamar, the newly ratified boundary between Devon and Cornwall, the location would be convenient. If this was the case, the apparent enhancement of the enclosure and ridge including one, two, maybe even three possible outer defensive works, may date to this period.

Sources / Further Reading

Associated Monuments

Associated Finds: none recorded

Associated Events

• EDV6998 - EarthWork Survey, Lifton Wood Hilltop Enclosure
• EDV6997 - Desk-Based Assessment of a Hilltop Enclosure in Lifton Wood and the Development of the Surrounding Area. (Ref: 1155)
• EDV8356 - Tamar/Lidar; A Single Source Approach to Landscape Survey and Socially Distanced Community Archaeology Area 2 (Culture Recovery Fund) (Ref: ACD2424)

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