CPAT logo
Cymraeg / English
Back Home
Projects index
Download CPAT report Hindwell double palisaded enclosure

Neolithic Palisaded Enclosures


Introduction

Cropmark evidence and and an on-going programme of geophiscal survey and trial excavation have now identified three Neolithic Palisaded Enclosures close to the village of Walton, in eastern Radnorshire, including the largest such site known in Britain. The presence of three palisaded enclosures in close proximity is perhaps a measure of the significance of this part of the Radnorshire during the Neolithic. The largest site, at Hindwell, has been known since the mid 1990s, along with a second, smaller site less than a kilometre to the south. More recently a third, double palisade has been discovered at Hindwell, which is the subject of continuing investigations.

Hindwell palisades

Right: The two palisaded enclosures at Hindwell © CPAT 2012

Hindwell palisaded enclosure

The Hindwell palisaded enclosure is a truly remarkable discovery, enclosing an area of around 34ha, which makes it by far the largest such site in Britain. It lies on the valley floor at 190m OD, centred at SO 25436070, encompassing Hindwell Farm, from which it takes its name.

Despite its size, the enclosure was only discovered by chance in 1994 when Alex Gibson photographed what turned out to be the western end, showing as a curving ditch and lying between Walton and the intended target of the flight, the excavation of a round barrow at Upper Ninepence. A review of existing photography showed that it had in fact been first recorded by J. K. St Joseph in 1969 in the field immediately to the east, although the main subject of that view was a Roman marching camp and the nature of the slightly curving ditch was not then apparent. In 1995 the north-eastern section of the enclosure was revealed in a photograph by Chris Musson, again as a cropmark, in a field to the north of Hindwell Roman fort. Further cropmarks were identified in 1996, and collectively aerial reconnaissance has accounted for around 1165m of the circuit (51%). Interestingly, the missing northern arc of the circuit appears to be followed by a lane between Hindwell and Four Stones, the curvature of which exactly matches that of the cropmarks at either end. This accounts for a further 550m of the circuit (21%), so that to date around 75% of the monument has been identified with some certainty. The eastern end remains problematic, however, since it lies beneath the Roman fort and although geophysics has provided a very detailed picture of the fort interior the arc of the enclosure is not clearly visible. A single, narrow entrance has been identified at the western end, where two larger posts flank a gap around 4m wide.

© CPAT 94-010-0022

Right: The western end of the Hindwell Palisaded Enclosure seen as a cropmark in 1994 photo number 94-010-0022 © CPAT

In plan the enclosure forms an oval at least 800m long and around 525m wide with the long axis aligned at 287 degrees west of grid north. The northern side, together with the western end, match the curvature of such an oval almost exactly, while the southern side has a much flatter curve.

The monument has been the subject several programmes of geophysical survey, commencing with a number of small areas in 1995 which were generally inconclusive. Of rather more significance, however, was an extensive survey conducted by Dr Helmut Becker in 1998 as part of the Walton Basin Project which investigated much of the circuit then known from cropmarks, together with significant areas of the interior, as well as the Hindwell Roman fort, which overlies the eastern end.

Four small excavations have now been conducted to investigate the enclosure, of which only those in 1995 completely excavated a number of post-pits, although the 1996 investigation did partly excavate four pits. The more recent excavations, in February and September 2011, focused on determining the relationship of the enclosure with other features. Respectively, the results confirmed that a large, presumably Roman ditch, cuts the enclosure at its south-eastern end, while the west end of enclosure has been cut into the fill of the northern ditch of the Hindwell Cursus.

© CPAT cs95-053-0036

Left: The 1995 excavations on the north-west side of the Hindwell Palisaded Enclosure revealed the outlines of individual charred posts photo number cs95-053-0036 © CPAT

The 1990s excavations revealed that the enclosure is formed by a perimeter of intersecting post-pits, each with an attendant post-ramp. The postholes averaged 2m in depth and would have contained posts 0.8m in diameter, which may have stood at least 6m above ground (assuming that at least one third of the post height would have been buried). The remains of carbonised oak posts were found within the post-pits, from which radiocarbon dates were obtained of 2900-2800 or 2700-2220 BC, and 2880-2800 or 2780-2460 BC. The spacing of the posts indicates that there were three posts every 5m, so that with a circumference of 2.35km, over 1400 posts would have been required to complete the perimeter. Evidence from the 1998 geophysics suggests that 18 posts in the north-east corner may have been burnt in situ, as may a further with eight or nine on the south-east.

The limited excavations have failed to produce any significant artefactual evidence and there is no clear indication of any internal structures or evidence for activity, although the 1998 geophysical survey did identify numerous large pits, at least some of which could be archaeological. There are two barrows within the interior, one of which appears to overlie a circle of pits revealed through geophysical survey, some showing signs of intense burning, suggesting the possibility of an earlier timber circle.

The original appearance of the enclosures also remains uncertain as it is impossible to determine whether the posts were free-standing, with gaps in between, or whether they may have been infilled with timbers or wattle to form a solid barrier.

Topographically the location of the enclosure is interesting in as much as the majority lies on a level terrace, with the exception being the southern part of the perimeter which drops off the terrace to what is presumed to be the bed of a post-glacial lake. At the eastern end of the southern side there is currently no evidence to extend the palisade beyond a large and prominent palaeochannel between the Summergil and Hindwell Brooks. Indeed, geophysical survey has now been conducted on either side of the channel and shows clearly the arc of the palisaded extending to the western edge of the channel, but there is no indication of its continuation beyond the channel to the east. Unfortunately, there is currently no evidence with which to date the channel, although it is possible that this represents an earlier course of the Summergil Brook, which later became diverted further to the east. The Hindwell Brook now issues from Hindwell Pool, which was enlarged, perhaps during the late 18th century, as a picturesque feature, but may well have incorporated the site of a natural pond. On current evidence it appears that the palisaded enclosure adopted the channel as its south-eastern boundary, whether it was a contemporary watercourse or not. The possible presence of a natural pond within the enclosure may also be significant.

There is no indication as to whether the enclosure was constructed in a single season, although it is clear that significant manpower would have been required to fell, shape and transport the timber, as well as to excavate the post pits and erect the posts. Consequently, it may be presumed that some form of contemporary settlement should be present within the immediate area. It is interesting that the geophysical survey has not identified potentially significant areas of burning which might be associated with domestic hearths or fires, or indeed for larger fires which might have been used to char the base of the posts. This could suggest that even during its construction the internal space was regarded as ‘special’, or at least separate from day to day life.

Walton palisaded enclosure and pit avenue

© CPAT Walton

Right: The known extents of the Walton Palisaded Enclosure © CPAT

The enclosure was discovered by J. K. S. St Joseph in 1975 and consists of a curvilinear alignment of pits visible as cropmarks on aerial photographs. The site lies to the west and north of Walton village, between the Summergil Brook and Riddings Brook (SO 25355986) at an altitude of approximately 190mOD. The cropmarks are only apparent in fields to the west of the B4357, curving southwards across the A44 and to the south-east as far as the Riddings Brook.

To date cropmarks have only identified the western side of the palisaded enclosure, with an arc of individual pits extending for 410m from west side of the B4357 (SO 2549 6001), south across the A44, and to within 25m of the present course of the Summergil Brook (SO 2545 5970). A total of 55 pits are currently known with an average spacing for the posts of around 6m, although there is no evidence to suggest whether the intervening spaces may have been infilled. The enclosure is possibly associated with a double alignment of pits forming an avenue to the south-west, composed of two parallel rows of pits fourteen pits around 12m apart and extending for 75m in length. The presence of the avenue has drawn comparisons with Meldon Bridge, Peeblesshire. The site is overlain by two Roman marching camps.

Trial excavations were conducted in 1998 and 2010, investigating two of the post-pits immediately to the west of the B4357. The earlier investigation identified an oval pit c. 4.3m long and 2m wide, for a post 0.4m or more in diameter, with a post ramp on the south-west side. Unlike the Hindwell palisaded enclosure there was no evidence to suggest that the posts had been charred prior to construction. However, a sample of oak charcoal was recovered from the post-pipe which provided a radiocarbon date of 2840-2470 cal. BC (SUERC-32384).

3045-0028

Left: The 2010 excavation of one of the massive post-pits with the post ramp in the foreground photo number 3045-0028 © CPAT

The 2010 excavations, immediately to the west of the earlier trench, uncovered a substantial post pit about 1.1m in diameter and up to 2.05m in depth, with near vertical sides. There was clear evidence for a post pipe between 0.65m and 0.7m in diameter, and voids were noted against the outer edge of the post-pit, presumably formed by the impact of the post as it was erected. The upper fill of the post pipe indicated the presence of a weathering cone, suggesting that the post had rotted in situ. The only artefactual evidence came from a single piece of good quality, but unworked flint from the fill of the post pipe.

A large post ramp extended south-west from the post-pit for around 3.6m, with a maximum depth of 1.5m, sloping at an angle of 23 degrees. The ramp contained obvious tip lines which suggested that material had been deposited initially against the post to provide support before the remainder of the ramp was infilled. A fragment of hazel charcoal from near the base of the ramp, against the post-pit, produced a radiocarbon date of 2570-2290 cal. BC. The post ramps are positioned in such a way that the posts are likely to have been erected in an anticlockwise sequence. The reason for this is uncertain, although it may be related to the direction from which the timber was being brought, construction starting at the furthest point so that the erected posts did not form an obstacle as work progressed.

The eastern side of the enclosure remains elusive being generally under pasture. Even when conditions have been favourable for the formation of cropmarks the presence of palaeochannels in this area effectively masked the identification of potential pits. The geological conditions have also thwarted three programmes of geophysical survey, in 1995, 2009 and 2010. The results of each survey were somewhat inconclusive, not least because of the large number of pit-type features which were present, most of which are likely to be natural variations in the gravels. Careful analysis of the data from 2010 has, however, led to the tentative suggestion of two potential arcs of pits extending for up to 120m which might indicate the perimeter of the palisaded enclosure, although this is far from certain and without the known alignment to the west of the road these anomalies would not have been recognised as potentially significant. Nevertheless this is the only indication of the eastern side of the palisaded enclosure and we can only hope that further aerial reconnaissance will eventually either confirm these results, or produce more compelling evidence for a different alignment.

Hindwell double palisaded enclosure

Hindwell double palisade

The enclosure lies to the south-east of Hindwell Farm, centred at SO 25976044. Its presence was first noted in 1998 when a geophysical survey was conducted by Dr Helmut Becker as part of the Walton Basin Project. This focused primarily on the Hindwell palisaded enclosure but also included the Hindwell Roman fort; and it revealed the partial circuit of a double-ditched enclosure, apparently lying in part beneath the fort and its eastern vicus. At the time it was thought to be associated with a set of triple ditches further to the east, but this was subsequently shown to be a separate feature following aerial reconnaissance by Toby Driver of RCAHMW in 2006 which located the eastern side of the enclosure, as well as identifying a short section between the Summergil and Hindwell Brooks. Further geophysical survey by CPAT in 2010 to the east and south of the fort provided additional evidence for the enclosure ditches.

© CPAT 2012, photo 3433-0047

Left: The partly excavated trench for the outer palisade showing the intense localised burning of the fill and the charred outer edges of perhaps four contiguous posts. © CPAT 2012, photo number 3433-0047

The combination of aerial photography and geophysical survey has so far identified at least 55% of the enclosure, which measures 300m by more than 250m across, with the two circuits 25m to 30m apart. A trial trench across the outer palisade in 2011 showed that this was formed by a series of intercutting pits up to 2.7m wide and over 0.8m deep, which had held closely spaced oak timbers, about 0.25m across. A re-examination of the evidence has recently concluded that the posts may have been burnt in situ as an act of deliberate destruction. Charcoal from one of the timbers has been dated to 2866–2574 cal. BC (SUERC-35386). The upper fill of the weathering cone above the post-pipes contained several sherds of Roman pottery and a coin, indicating that the line of the palisade remained visible, albeit as a slight earthwork, into the Roman period.

© CPAT 2012

Right: The 2012 excavation of the inner palisade of the Hindwell double-palisaded enclosure. The plan shows the excavation at an early stage with posts defined by the charring of the outer surfaces. © CPAT 2012

Recent investigations in March 2012 have confirmed the position of the inner palisade on the eastern side, demonstrating in the process that it had been cut by the canalised section of the Hindwell Brook. Like the outer palisade, this was formed by intercutting pits up to 2.8m across and at least 1.45m deep below the natural gravel, which held close-set posts around 0.35-0.4m in diameter. The upper part of the posts had clearly been burnt in situ, causing intense reddening of the gravelly fill pack against the outer face of the posts, but not on the inner side. This suggests the deliberate destruction of the monument by fire, a feature which has been recorded at a number of palisaded enclosures elsewhere.

© CPAT 2012, photo 3433-0047

Left: The partly excavated trench for the outer palisade showing the intense localised burning of the fill and the charred outer edges of perhaps four contiguous posts. © CPAT 2012, photo number 3433-0047

The enclosure at Hindwell may also have a relationship with the Summergil Brook, for its southern extent as currently known stops just 10m to the north of the brook, with no evidence to suggest a continuation beyond the watercourse. Given the likely movement of the brook over the last three millennia it is possible that the southern side of the enclosure has been lost.

The position of the monument with respect to both the Hindwell and Summergil Brooks also raises an interesting question regarding its relationship with the larger Hindwell palisaded enclosure to the west. As noted previously, the latter site appears to respect a large palaeochannel lying between the two brooks and does not continue to the east, into the area occupied by the double-circuit enclosure. Whether either of these channels was active during the Neolithic is perhaps questionable, but as landscape features they appear to have exerted some influence over the siting of both monuments.

Conclusions

The discovery of three Neolithic palisaded enclosures in the Walton Basin is clearly of considerable significance. Even more so when one considers that this area also contains what may be the second longest Neolithic cursus in Britain, as well as a second cursus, a Neolithic causewayed enclosure and numerous Bronze Age burial mounds.

The limited excavations have failed to produce any significant artefactual evidence and there is no clear indication of any internal structures or evidence for activity, although the 1998 caesium magnetometry survey did identify numerous large pits, at least some of which could be archaeological. The original appearance of the enclosures also remains uncertain as it is impossible to determine whether the posts were free-standing, with gaps in between, or whether horizontal timbers were used to form a solid barrier. Topographical and geophysical surveys were carried out over the Hindwell enclosure in 1998, but these revealed no evidence of surviving earthworks.

There is no indication as to whether either enclosure was constructed in a single season, although it is clear that both, but Hindwell in particular, would have required significant manpower to fell, shape and transport the timber, as well as to excavate the post pits and erect the posts. Consequently, it may be presumed that some form of contemporary settlement should be present within the immediate area. It is interesting that on neither site has geophysical survey identified potentially significant areas of burning which might be associated with domestic hearths or fires, or indeed for larger fires at Hindwell which might have been used to char the base of the posts. This could suggest that even during their construction the internal spaces were regarded as ‘special’, or at least separate from day to day life.


Privacy and cookies