Only some forty extant barrows can be considered as being visible from the Vale.These are restricted to a relatively small number of round barrows, many situated on the flanks of the Vale. This is in contrast to the massive concentration of barrows noted by McOmish et al. found on the eastern portion of the Salisbury Plain Training Area (2002, 33) and a smaller, yet significant, population of barrows on the Marlborough Downs (Cleal 2005). Some of those barrows on the very fringes of the Salisbury Plain cluster clearly relate to the Vale, however, occupying somewhat ambiguous topographical positions.
Barrows, topography and soils in and around the Vale of Pewsey
1. The Greensand Vale
A brief perusal of a map reveals that the vast majority of the barrows in the Vale are located on the chalk. Only two extant barrows, near Market Lavington (ST 9999 5641 & ST9995 5649), opened by Maud Cunnington in 1924, are found on Greensand soils. Field (1998) in his analysis of round barrows in Sussex, by way of contrast, noted concentrations of barrow cemeteries at the western end of the Greensands forming the Weald. These cemeteries were restricted to the Folkestone Beds formation which tended to form narrow prominent ridges running parallel to the chalk downs. The cemeteries were equally spaced, set back from, but aligned along small rivers (Field 1998, 313-4). There are apparent similarities between the location of the Market Lavington barrows and those at the western end of the Weald. The Market Lavington pair are located on the eastern spur of a small but prominent ridge running south east from Worton and flanked by streams on either side. The barrows are equidistant between the watercourses and seem to be addressing the narrowing shallow valley running north east to Stert. Field (1998, 315-6) states that there are three common themes in the siting of round barrows in the landscape in the Weald:
- Barrows are often located on ridges and are, therefore, highly visible;
- Barrows exhibit a tendency towards false cresting, highlighting the direction in which to view the monument from, and;
- Barrows tend to be located near rivers; i.e. mounds are situated close to but set back from rivers. Field claims this is so the monuments do not “interfere with agriculture or other activities”.
The monuments at Market Lavington certainly seem to fall into all of these three criteria, although whether they occupied land considered marginal in the Early Bronze Age is impossible to verify within the current bounds of this work. McOmish et al. (2002, 46) have noted similar locational traits in their study of the Bronze Age mounds of the Salisbury Plain Training Area notably around the Nine Mile River and River Bourne and Woodward (2000, 63-6) has noted a correlation between round barrows and watercourses on Cranborne Chase.
The topography of barrows and barrow cemeteries has been studied using a variety of approaches in the recent past. Woodward and Woodward (1996) analysed the topographical setting of barrows on the South Dorset Ridgeway, the environs of Stonehenge and Avebury. The paper claims to adopt ‘a systematic and detailed analysis of the geographical contexts of barrow cemeteries’ and yet no clear methodological approach is apparent beyond simple powers of observation. So called ‘analysis lines’ (ibid: Fig 1) applied to the barrows of the South Dorset Ridgeway simply follow the main ridges and linear barrow groups and arrows pointing from outlying clusters of barrows towards the main linear groups are apparently meant to imply some degree of connectivity, presumably through the medium of intervisibility although no clear explanation is provided. A similar approach is taken to the analysis of barrows around Avebury and Stonehenge, with inconclusive results for the former and a rather simplistic analysis of barrow distribution at the latter. Woodward & Woodward (ibid 283-5) state that there are two concentric bands of barrows running around Stonehenge, however their analysis omits a significant number of barrows including Winterbourne Stoke 38 to 43; Durrington 1 to 3; Winterbourne Stoke 46 to 50; Shrewton 13 to 22; the entire Lake Down Group and a large number of barrows left stranded between the two concentric circles. The main reason that barrows appear to be laid out in a circular distribution around Stonehenge is that the monument lies in the middle of a shallow basin surrounded by low ridges, typical locations for the siting of barrows (Field 1999: 315-6). The distribution of barrows around Stonehenge is partly a product of the topography of the area, not some overriding, conscious plan of Bronze Age barrow builders.
Tilley’s phenomenological analysis of the distribution of barrows on the Ebble-Nadder Ridge (2004: 185-203) suffers similarly from the lack of a defined methodology. The author admits following peer review (ibid: 199-203) that the methodology consisted simply of observations taken whilst walking (ibid: 202). Little reference is made to past archaeological investigation and the interpretation is purely based on the observer’s reactions to the landscape and position of the round barrows within it, what Tilley refers to as an ‘insider’s knowledge of the significance of place’ (ibid: 185). The most interesting observation Tilley makes is with reference to the tendency for some barrows to be grouped on the lower slopes of coombes for example Hydon Hill/Little Down (ibid: Fig 4; Fig 8: 189-191), an anomalous position if we accept that the majority of barrows are located on ridges, but no meaningful suggestions are made to explain their position beyond the observation that they are located at all the main transition points within the coombe complex. Certainly Tilley is correct to look for some of the explanations for the siting of barrows within the topography of the landscape but without a meaningful methodology and consideration of other factors, the approach has limited application.
What is surprising, though, is the apparent lack of barrows located on any of the prominent natural features present on the floor of the Vale. Etchilhampton Hill, for instance, a large chalk outlier located in the centre of the Vale just to the east of Devizes has no documented evidence for any prehistoric burial monuments. The presence of a ring-ditch situated on the western slope of the hill was detected on aerial photographs (Valentin & Moore 2002)has been interpreted as indicating the possible presence of a barrow but the evidence is inconclusive. Similarly Potterne Field, The Knoll at Bishop’s Cannings, Woodborough Hill and Swanborough Tump all apparently lack any surviving evidence of Early Bronze Age monument building. The explanations for this apparent absence are none too obvious.
Whether there are any other round barrows present on the Greensands is very difficult to say. As previously noted, the Greensands are inimical to aerial photography unless chalk is located just below the soil surface. There are a series of NMR entries for possible round barrows in the parishes of All Cannings and Alton, located in areas where the shallow layer of Greensand overlies Chalk (NMR Entries: SU 06 NE 91; SU 06 NE 94; SU 06 103; SU 16 SW 102). Field (1998, 314) has observed that it is very difficult to tell monuments from natural features on the ground on Greensand soils making identification of eroded or inconspicuous barrows on the ground very difficult. Field (2001, 59) noted an apparent barrow cemetery at Charlton based on data from an aerial photograph showing a cluster of ring-ditches centred on SU 0920 5748 (Field pers. comm.).
Barrows, Ring-ditches and Aerial Photography
There does seem to be a tendency on the part of some writers to automatically assume that ring-ditches, when found in association with water-courses, represent ploughed out round barrows (pace McOmish et al. 2002, 46). There are plenty of excavated sites to suggest that is not always the case, for example Shorncote, Gloucestershire (Barclay & Glass 1995), Standlake, Oxon, Cassington, Oxon and Barrow Hills, Radley where ring ditches and round barrow were found alongside each other (Case & Whittle 1982). The examples cited are all situated in close proximity to the Upper and Middle Thames and, therefore, may represent a localized anomaly but may also be suggestive of a tradition of constructing ring-ditch monuments near rivers. Many ring-ditches probably do represent ploughed-out or otherwise levelled round barrows but it is possible that a significant proportion of ring ditches identified from aerial photographs represent either funerary monuments built without recourse to the construction of a mound or other ritual structures. Indeed Catling (1982, 101-2) argues for a period of currency for the construction of funerary ring-ditches commencing in the latest Neolithic and ending in the Middle Bronze Age, contemporaneous with the main period of barrow construction in Southern England. Excavations at Shorncote Quarry, Somerford Keynes, Gloucestershire (Barclay & Glass 1995) lend some support to this suggestion. The excavation uncovered a number of ring-ditches including a multi-phased penannular ring-ditch from Area 5 that was probably constructed in the Late Neolithic and contained a series of Deverel-Rimbury period urned cremations. This ring-ditch was located in close proximity to another ring-ditch enclosing a central Beaker grave in Area 1. As no original ground surface survived, it was unclear whether the Beaker burial had been covered by a mound (ibid: 25-34)
The currency of ring-ditches as an Early Bronze Age funerary monument may also help to explain the origins of some types of “fancy” barrow. A ring-ditch with external or internal bank is not very far away from a pond barrow or disc barrow in terms of its morphology. Perhaps archaeologists have placed too much emphasis on the mound element of barrow building (pace Peters 2000) and not given enough consideration to the enclosing ditch and associated features. In the case of ring-ditch enclosures close to rivers, there is evidence to suggest some ditches contained water from the outset, for example the Neolithic ring-ditch at Newnham Murren, Oxon (Moorey 1982, 57). If this were so, the water and any associated bank would represent a definitive and most striking boundary between the enclosure and the surrounding landscape and perhaps go some way to explaining the presence of ring-ditches near watercourses in a manner similar to that already suggested for henges (Richards 1996)
The most extreme explanation for the apparent absence of round barrows from the greensands of the Vale is to suggest the development of a local burial tradition in the Early Bronze Age utilizing ring-ditches and “inconspicuous” barrows (Peters 2000). This burial tradition combined with the inability of aerial photography to detect subsurface features on greensand would render any Early Bronze Age burial activity almost invisible but, in the circumstances, this explanation seems most unlikely.
2. The Chalk Downs
Of the barrows located on chalk, very few have been investigated by modern archaeologists and, indeed, many appear not have attracted the attention of antiquarians either, or at least those that recorded their investigations. Consequently this account is very largely dependent on the sparse entries recorded in the Wiltshire SMR and National Monument Archive which are based on site visits carried out, on average, some thirty years and more ago. The discussion will follow a prescribed route around the fringes of the Vale starting in the north-east at Wootton Rivers and running anti-clockwise until concluding at Burbage.
A. The North of the Vale
The Wootton Rivers or Square copse barrow group (SU 2180 6375) is located on a south-east facing spur overlooking the course of the River Dun. Five barrows are clustered together and are now covered by woodland, whilst two outliers (one considered doubtful) lie in a field just to the east. This group is clearly intended to be seen from the river valley east of their position. A separate single barrow (SU 2096 6420) lies approximately 1 km to the north-west of the Square Copse group lying on the 175m contour overlooking a wide area due south of its position. Perhaps significantly, the next barrows on the northern edge of the Vale are located 7 km to the west. This “gap” is interesting including, as it does, remarkable natural features like Martinsell Hill and the Giant’s Grave spur but there is no clear explanation for the absence of round barrows except perhaps to observe that the “gap” coincides with a substantial deposit of clay-with-flints that extends to within metres of the scarp slope in many places. The Wootton Rivers barrows are all located on chalk, in the case of the single outlier barrow, placed within metres of the clay-with-flints. The exception to this is the Giant’s Grave spur near Oare which is composed entirely of Chalk but the extent and nature of later earthworks including the construction of a summer house for Rainscombe House (Colt Hoare 1821 Plate III) may have obscured or destroyed any barrow present. It is also worth noting that the scarp between Martinsell and Knap Hill is extremely steep, unlike the slope further west, a factor which may have precluded barrow construction.
The “gap” on the northern flank of the Vale is terminated by the siting of a barrow cemetery (SU1394 6397) to the south-west of Gopher Wood near Huish. The barrow group comprises three bowl barrows (Grinsell’s Wilcot 1, 2 and 5a), a cluster of three confluent bowl barrows (Wilcot 3a –c), two confluent bowl barrows (Wilcot 4) and a disc barrow (Wilcot 5). Colt Hoare (1821, 11) notes that Cunnington Snr. opened three barrows, although it is not possible to say which, and whilst pottery was recovered no mention is made of human remains, suggesting none were recovered. Thurnam excavated the triple barrow in the middle of the nineteenth century recovering cremations from central pits from Wilcot 3b and c (Thurnam 1869, 42) although no other grave goods were present. The barrow group is located on a steep-sided spur overlooking the centre of the Vale. High ground rises behind and to the west of the cemetery and most of the barrows are set back from the edge of the scarp, making it difficult to see the barrows from immediately below.
Colt-Hoare, in his perambulation along the ridge between Gopher Wood and Knap Hill, notes two barrows in his passing (Colt-Hoare 1821, 11) but there is no evidence of their existence today.
Three barrows are located on Knap Hill, a kilometre to the west of Gopher Wood. Two are situated within the Neolithic enclosure and one on the western slope of the hill (Connah 1965). Connah states that the small barrow on the western slope (Grinsell’s Alton 10 or 13) was excavated in 1939 but unpublished (Connah 1965, 22). This is probably the same barrow excavated by Thurnam without result (Thurnam 1866, 326-7). The barrow is inconspicuous, recorded by the Ordnance Survey in 1974 as being 0.4m high and 16m in diameter (NMR No: SU 16 SW 171). Darvill (2004, 61) states that the barrow may date from the early fourth millennium, based on the dimensions of the barrow, and the findings of the excavator, C W Phillips. The excavation of the barrow revealed “a central crouched inhumation surrounded by flint-working waste and broken plain bowl-style pottery” (Darvill 2004, 61). An alternative explanation is provided by the NMR Monument Record for the barrow (SU 16 SW 173) which states “Although there is a possibility that the barrow may be of Neolithic origin, the incorporation of Neolithic material in a mound constructed in an area of Neolithic activity is not entirely surprising”. The movement of Neolithic cultural material incorporated into topsoil downhill from the causewayed enclosure is entirely possible; Connah noted the lack of distinction between the downhill edge of the enclosure banks and the overlying topsoil (Connah 1965: 10). The subsequent inclusion of residual Neolithic pottery in the construction of this small burial mound is perfectly feasible. Alternatively, it may have been the intention of the barrow builders to incorporate Neolithic material into the mound (see below)
Colt Hoare notes the presence of two barrows within Knap Hill enclosure (1821, 11) but the barrow on the hill top was destroyed by flint digging at some point before Thurnam excavated Alton 8 in the 1850s (Thurnam 1866, 326-7). His excavation revealed some sheep bones in the upper stratigraphy of the mound and a circular cist, some 0.6m in diameter and 0.6m deep, in the chalk at the centre of the barrow “nearly full of ashes and burnt bones” (Thurnam 1866, 327)
The two extant barrows seem to be positioned to be seen from the col between Adam’s Grave and Knap Hill through which the Ridgeway runs today. They also overlook the relatively flat, fertile land of East Field, where a thin covering of Greensand overlies the chalk, some of the most easily worked and fertile soil in the Vale. The long since destroyed barrow, in contrast, seems to address both the dry valley to the north of Knap Hill where it would have been clearly framed against the southern horizon and, to a lesser extent, the Vale.
The two bell barrows and one bowl barrow (SU 1157 6376) referred to by Grinsell as Alton 15, 16 and 6 respectively lie in the col next to the Workway Drove, part of the Ridgeway. The saddle in which the three barrows are situated is flat but the barrows are close to the scarp edge again overlooking East Field. Thurnam excavated the bowl barrow in the 1850s, reporting that he found a previously disturbed cremation and a probable Early Medieval extended female inhumation with infant (Thurnam 1860, 326).
Further west, Tan Hill was a focus for barrow construction in the Early Bronze Age with some 12 barrows still extant within a radius of 1200 metres of the summit. Furthest east, All Cannings 1b (SU 0931 6457) is located, unusually, on the edge of the scarp slope with a commanding view over Stanton St Bernard and Milk Hill. The barrow appears not have been investigated in the modern era. It has suffered plough damage with the ditch ploughed out on the north side and its true diameter of 15 metres is now only visible on aerial photographs. East of the summit of Tan Hill, four bowl barrows (All Cannings 1, 1a, 12 & 15) are heavily plough damaged (SU 0904 6463). All Cannings 1 was excavated by Thurnam (1860, 325) who found two circular pits containing ashes but no evidence of cremations. The mound also contained fragments of antler and sheep and cattle bone. It is debatable whether these barrows were sited with the intention of directly addressing the Vale. Rather they seem to be best seen from Milk Hill and area immediately behind the ridge.
Conversely, the siting of All Cannings 20 and 22, a bowl barrow and saucer barrow respectively, are clearly situated to be seen from the gently sloping part of the Vale situated to the west of Rybury. All Cannings 20 appears never to have been excavated in the modern era but the NMR entry notes “some slight mutilation at its centre” (NMR No: SU 06 SE 21). It is situated on a small spur of the main ridge overlooking a trackway running between All Cannings and Beckhampton and which, possibly, is the contemporary trace of a much older routeway.
All Cannings 22 is bifurcated by the parish boundary between Bishop’s Cannings and All Cannings. A well preserved saucer barrow with a diameter of 25 metres, it was excavated by Thurnam (1860, 325). The excavation failed to locate any primary inhumation but a secondary urned cremation was recovered. All Cannings 22 must be considered part of the group containing Bishop’s Cannings 46.47, 48 and 97 all situated on a spur to the east of Kitchen Barrow Hill. Respectively three bowl barrows and one disc barrow, the bowl barrows were, again, excavated by Thurnam in the 1850s (Thurnam 1860, 324-5). The excavator recovered a possible primary cremation and beads from Bishop Cannings 46 but discovered that 47 and 48 had been previously investigated. All the barrows have undergone quarrying and some subsequent plough damage. It would appear that the disc barrow, Bishops Cannings 97, has not been investigated at anytime in the past two hundred years at least. This group of barrows is arrayed along the south-western edge of the spur looking out over the western end of the Vale.
The bowl barrow Bishops Cannings 45 on Kitchen Barrow Hill is situated at the head of a narrow deep comb. The barrow is placed in such a way as to maximize its visibility towards the south-west and the centre of the western end of the Vale. It placing could not be more deliberate. The barrow, though, is very mutilated, indeed it is no longer visible on aerial photographs. As a consequence it escaped the attentions of antiquarians.
Further west, the presence of a single bowl barrow on each of the two spurs of Easton Hill would appear to reinforce the idea that the barrow-builders wished to ensure that the monument being built was obvious from the floor of the Vale. Bishops Cannings 43, however, may be a spoilheap from an adjacent quarry. It is markedly oval, higher at the south-western end than the north-eastern and, most tellingly, constructed on a very steep slope. Consequently doubts must be raised over the veracity of Grinsell’s identification.
Situated at the end of the western spur of Easton Hill, Bishops Canning 42 takes the form of a bowl barrow some 12 metres in diameter and 0.8 metres high. It appears not to have investigated by either archaeologists or antiquarians. It overlooks a broad area of the western end of the Vale.
B. Salisbury Plain
Turning south to the northern scarp of Salisbury Plain, there are far fewer barrows overlooking the Vale of Pewsey. The vast majority of barrows in the central and eastern portions of the Plain are located within the drainage basins of the Rivers Bourne and Avon (McOmish et al. 2002, 46), only a very small number are found elsewhere than the slopes of the river valleys that have cut through the chalk plateau.
Goddard’s Urchfont 1 and 2 are two such exceptions. Urchfont 1 is perched on the ridge between the furthest headwaters of the Water Dean Bottom system and the scarp slope in an ambiguous position, with a viewshed of some 280° to the west, south and east. Urchfont 2 lies on the north-west facing scarp slope overlooking the gentler slopes of the Vale. Both are bowl barrows but whereas Urchfont 1 still stands to a height of 1.7m, Urchfont 2 has been almost destroyed by ploughing. Neither appear to have been investigated.
Thirteen kilometres due east, Pewsey Down has the next concentration of round barrows on the southern flank of the Vale. Manningford 1 (Grinsell) (SU 1711 5552), a ditched bowl barrow, is situated on the upper parts of a dry valley running north-west to the Avon. The interlocking nature of the dry valley is such that the barrow can only be seen at close quarters, framed by the shallow slopes above it. Some 1100 metres to the north-west and situated on a west facing spur, a second bowl barrow (SU 1636 5632) overlooks the same valley. Unlisted by Grinsell, Meyrick retrieved a “chisel-shaped” palstave from this bowl barrow (Swanton 1987, 16).
The barrows located in the portion of Pewsey Down within Pewsey Parish have fared less well. Pewsey 1 (Grinsell) (SU 1760 5743), Pewsey 2 (Grinsell) (SU 1733 5728), the pair of barrows referred to as Pewsey 10A (Grinsell) (SU 1829 5740), and the unlisted probable bowl barrow at SU 1683 5777 have all been heavily plough damaged. Of these barrows, only the suspected bowl barrow (SU 1683 5777) appears positioned to be seen from the floor of the Vale proper. Pewsey 1 and the entirely ploughed out Pewsey 2 are situated at the head of a small re-entrant valley, part of the same dry-valley complex overlooked by the Manningford barrows. Grinsell’s 10a clearly share the riverine distribution of so many barrows on Salisbury Plain and can be seen as extreme outliers of the Nine Mile River concentration and, therefore, have more in common with barrow groups such as Down Farm and Everleigh Barrows.
Colt-Hoare excavated several of the Down Farm group, noting the mound of one barrow: ‘No.8 is remarkable for having a more pointed apex than any other barrow I remember to have seen, on which count I named it the Cone Barrow’. Colt-Hoare’s workmen encountered a considerable quantity of cremated bones, a bronze spearhead and a central cremation pit during the excavation of this barrow. Several other barrows were opened but Colt-Hoare soon tired of the operation on account of the difficult nature of the clay-with-flints soil (ibid: 191 ).
Two ditchless bowl barrows and three saucer barrows of the Down Farm barrow group were excavated by Faith de Mallet Vatcher in the Autumn of 1958 ( de Mallet Vatcher 1960: 339-51)as a result of ongoing agricultural damage. No burials were recovered from the saucer barrows which had all been seriously damaged by cultivation but pits containing ash and charcoal were recovered from Sites C and D. In Site D a hole measuring some 45 cm in diameter and depth was filled with a deposit of pure charcoal, no soil was present in the fill. Two ash-filled pits were located in Site C (ibid: 346) and several possible hearths were noted on the old ground surface beneath several of the mounds.
A burial within a coffin or timber lined pit was uncovered beneath the mound of Site A (ibid, 341-3). The human remains within were very fragmentary and badly decayed leading the excavator to suggest that the body was either in a decomposed condition when placed in the pit or the pit, with the body in situ, had been open to the elements for a considerable period before burial. The second bowl barrow, Site B, had been robbed at some time in the past (ibid 344-5).
Very little pottery was recovered from the excavation, literally a few fragments from each site. Typically, fragments of Beaker and Collared Urn was recovered but several small and weathered sherds of Early Neolithic Plain Ware were also found leading to the excavator suggesting that a Neolithic occupation site was located nearby (ibid 347), a claim supported by the presence of ‘occupation debris’ deposited in the ditch of Site C. This occupation debris was described as being ‘of relatively fine grain, dark in colour, and contained a considerable clay-fraction, with much organic matter, some charcoal and numerous broken snail shells’ (Dorell & Cornwall in de Mallet Vatcher: 348-9). The excavator suggested the layer had formed in the ditch early in the life of the site and was the accumulation of leaf mould and human refuse (de Mallet Vatcher: 345)
The northernmost barrow cemetery on Salisbury Plain is found north-east of Milton Hill Farm (SU 1999 5786) located on the watershed between the Nine Mile and Bourne drainage systems, approximately a kilometre from the edge of the Vale. In effect, the barrow group is situated on the south-eastern flank of a small slightly domed “island” of land situated between three drainage systems.
In the same year as Faith Vatcher excavated part of the Down Farm Group, Paul Ashbee excavated five barrows at Milton Hill Farm including a discrete group of an oval double disc barrow (Milton Lilbourne 1) and two bell-barrows conjoined by a small bowl barrow (Milton Lilbourne 2.3 & 4) and an outlying bowl barrow (Ashbee 1986: 23-96). The oval disc barrow had been previously opened by Thurnam who had recovered a cremation (Thurman 1870: 295) and Ashbee excavated another cremation and awl from the monument. The barrow is of an unusual form with only six comparable sites known in central southern Britain including Gussage St Michael 17a, Wimborne St Giles 8 (part of the Oakley Down group) and Bishops Cannings 95.
The conjoined barrows (Milton Lilbourne 2, 3 & 4) had been damaged by past activity. A pit had been sunk into Barrow 2 and the linking barrow, Barrow 3, had been entirely levelled possibly as a result of earlier excavation. The structure of Barrow 2 comprised a loam core covered by an outer layer of chalk rubble (ibid 1986: 35-8). The loam core was “augmented” by what Ashbee described as occupation debris, a feature the mound shared with Barrow 4.
The bell barrow (Milton Lilbourne 4) was offset within it’s ditched enclosure and contained a central cremation of a mature male accompanied by a miniature vessel and surrounded by the burnt remains of what Ashbee described as a coffin and mortuary house (ibid 45-46, 87). Again the structure of the barrow featured a loam core with added soil and occupation debris overlain with chalk.
The outlying bowl barrow (Milton Lilbourne 5) had exactly the same structure as Barrows 4 and 2 with an inverted urn containing a cremation located in a pit at the centre of the ditched enclosure accompanied by a charcoal spread. Ashbee suggested that Barrow 5 may be earlier in date than Barrow 4 on the basis of the presence of a large piece of carbonised wood, perhaps part of the wooded structure in Barrow 4, lying on top of the primary fill of the ditch of Barrow 5.
Ash, Occupation Debris, Round Barrows and Persistent Places
The inclusion of ash into burial mounds has already been discussed with reference to Hatfield Barrow, Marden. In that instance, the ash comprised part of the make-up of the barrow mound in contrast to the findings made at the Down Farm barrow group where ash and charcoal was deposited in pits cut into the original ground surface (Vatcher 1960). This is similar in character to the Grooved Ware pits full of charcoal and ash recovered from Woodlands, Countess Road, and Ratfyn, Amesbury (Stone 1935; Stone & Young 1948; Stone 1949) although not on the same scale nor in the same topographical setting. The same deliberate effort had been made, though, to burn and curate a significant amount of ash and charcoal with the intention of depositing it in specially dug pits. The contrast between the black of the charcoal and the white chalk would not have been lost on the participants in this process perhaps redolent of cremations and the colour and texture of cremated bone. Parker Pearson (2004) has drawn attention to the symbolic possibilities of the inclusion of ash into barrows.
Ashbee’s occupation debris at Milton Hill Farm was ‘often charcoal-laden, embodying potsherds, flint artefacts, knapping debris and bones’ (ibid; 72) and would have provided a strong contrast when forming the loam core of the barrows to the enveloping chalk outer layer as it was heaped up. Mary Ann Owoc has suggested that the deliberate inclusion of soils of different colours and types was a deliberate part of the process and ceremony associated with barrow construction in the Early Bronze Age (Owoc 2000; 2004) and Parker Pearson (2004:80) has drawn attention to the number of round barrows excavated that contain similar material. The occupation debris at Milton Hill Farm contained sherds of pottery ‘partially or wholly representative’ of the entire repertoire of Neolithic and Earlier Bronze Age pottery (ibid: 73). This is clearly different from the occupation debris Faith Vatcher recovered from the ditch of Site C at Down Farm in the autumn of 1958 which seems largely humic in nature and contained a large number of snail shells (de Mallet Vatcher 1960: 348-9). It seems most unlikely, bearing in mind that both sites were excavated within a few months of each other, that the excavators were unaware of each other’s activities and may well have visited each other’s sites and discussed their findings. Vatcher’s identification of occupation debris comes after the cessation of activity at Milton Hill Farm and was, probably, made in the light of Ashbee’s discoveries but the deposit is of an entirely different nature to that comprising the central cores of the Milton Lilbourne barrows, both pedologically and stratigraphically.
The range and relative proportions of pottery types contained within the occupation debris encountered by Ashbee is difficult to ascertain as the report tends to concentrate on Grooved Ware and Collared Urn sherds (ibid:73). The range of pottery types combined with the presence of flint tools and debitage and animal bones is highly suggestive of the material having been curated in some form of midden. A possible comparable site is the midden site excavated by Keiller adjacent to the West Kennett Avenue (Smith 1965b: 210-6) in 1934-5 where Peterborough Ware and Grooved Ware sherds had survived in considerable quantities suggestive of a thick protective layer of soil (Pollard 2005, 110-1). The deposit appears to have accumulated as a result of intervals of periodic occupation, though the specialised and unusual nature of the flint assemblage suggests that the activity taking place was not run-of-the-mill. The presence of postholes associated with the densest concentrations of material has led Pollard to suggest the midden was marked by posts and that pits adjacent to the postholes and containing structured deposits including Grooved Ware may have been dug in an act of final closure. The incorporation of the midden into the course of the West Kennett Avenue was a further stage in this process, Smith (1965b: 212) noting that a gap in the stone settings of the western side of the Avenue coincided with the densest accumulation of midden material.
Another comparable site Avebury G55, excavated by Smith (1965c: 24-46) is situated some 250m north-west of West Kennet Long Barrow, located on a north facing spur overlooking the bend of the River Kennett at Swallowhead Springs. The barrow had been dug into on three previous occasions and the central cremation pit was disturbed, although a number of satellite cremations were recovered. The ploughsoil surrounding the barrow suggested that the monument had been over or in the vicinity of a Neolithic midden. Sherds of Plain Ware, Peterborough Ware, Grooved Ware and beaker pottery were recovered from the pre-barrow layers along with animal bones and flint implements. In addition a number of pits were recorded containing similar material to the midden.
The longevity of the period of deposition, no matter how episodic, defines both the West Kennett Avenue site and the midden adjacent to Avebury G55 as ‘persistent places’; successive generations have returned to the same location over long periods of time and, apparently carried out broadly similar activity much of which can only be described as being of a ritual nature. This sort of activity takes the concept of persistence of place far beyond the narrow functionalist definition laid down by Schlanger (1992: 94-112). These sites clearly have a symbolic significance far beyond their original meaning to subsequent groups, they serve as mnemonics reminding people of past human activity and importance of a location (Pollard 2005: 105)(Gosden & Lock 1998). Potentially a midden would become every bit as much a monument in the landscape as a more formal sites such as a long barrow. The colour of the soil, the presence of a mound, posts and the scatter of pottery would all serve to identify the location as one of special significance.
The symbolic and mnemonic importance of Neolithic middens may help to explain the incorporation of so-called occupation debris into the structure of the round barrows at Milton Lilbourne and elsewhere. It is difficult to justify the argument that large amounts of midden material were carted in specifically for the construction of the round barrows at Milton Hill Farm. Rather it could suggest that the location for the barrows was at least partly chosen because of the presence of a Neolithic midden on or close by the site as seems to have been the case at Avebury G55. Here a riverside spur below the crest of the hill coincided with the presence of a Neolithic midden – perhaps the most auspicious position for the construction of a round barrow.
The significance of Neolithic middens for the location of some Bronze Age round barrows is further emphasised by the results of Edwina Proudfoot’s excavation of Bishop’ Cannings 61, 62 and 62a (Grinsell) on Roughridge Hill in 1964(Pollard 1993; Chap 3.2). The barrows were situated on the south-eastern slope of Bishop’s Cannings Down approximately 450m west-south-west of Easton Down Oval Barrow.
Beneath G61, a single large pit measuring 2.5m in diameter and 0.8m deep produced a considerable amount of Early Neolithic material. Six more early Neolithic pits were located beneath and around G62a together with three less tightly dated pits. The pottery and lithic assemblages from the pits on both sites suggests a very early Neolithic date, perhaps before the main periods of long barrow and enclosure construction. Pollard suggests the contents of the pits had previously formed part of a midden accumulation as animal bones were disarticulated but softer parts such as phalanges were present and the pottery was largely unweathered. Therefore it is not unreasonable to suggest that evidence of either the existence of an Early Neolithic midden or the presence of the pits acted as a stimulus to the construction of the round barrows at that location.
Care must be taken not to extrapolate these findings to explain the location of every Bronze Age round barrow. In the case of Avebury G55 and Bishop’s Cannings G61, 62 & 62a the round barrows have been constructed in close proximity to long barrows. The same is true of the Milton Hill farm group with Giant’s Grave lying some 1100m to the west-north-west. The sequence of events is complicated, with the pits at Roughridge Hill predating the mound at Easton Down and apparently episodic depositional events at Avebury G55 but there does seem to be correlation between at least some Bronze Age round barrows and earlier Neolithic activity.
This association is all the more striking when analysing the spatial relationship between Neolithic monuments and Bronze Age round barrows in the study area. With a few exceptions, Bronze Age barrows are not found in association with earlier monuments unlike the case on the South Dorset Ridgeway (Woodward & Woodward 1991: 140-6) and Cranborne Chase (Barrett et al 1991: 124-39). The most notable exception is the presence of two round barrows (one now destroyed) within the circuit at Knap Hill. The location of these barrows is analogous to those of Winterbourne Monkton 1 and 2 situated on the highest point of Windmill Hill (McOmish in Whittle et al 1999:16). By way of contrast, Rybury appears to have no barrows in association, although the interior of the enclosure is so badly mutilated by quarrying that it is difficult to say with any certainty that no barrows existed within its banks.
The incorporation of Neolithic cultural material into the monument would be an inevitable consequence of building a Bronze Age round barrow on such a site as Windmill Hill or Knap Hill as, perhaps, we see in the case of Alton 10/13 on the slopes of Knap Hill where sherds of Plain Ware were encountered by the excavator (Darvill 2004: 61). Field’s criteria for the location of round barrows (1999: 315-6) may need to be fine tuned in the case of some barrows to take account of the reuse and incorporation of Neolithic material from around the barrow’s location.
The side valleys of the Bourne system continue to be the focus for round barrows in the parish of Easton. Grinsell’s Easton 1 (SU 2099 5775) and Easton 1B (SU 2160 5822) are located at the head of a small re-entrant running south-south-west into a dry valley. Unusually, an unlisted bowl barrow (SU 2148 5802) is located in this dry valley, close to its base. None of these bowl barrows have been investigated.
The disc barrow listed as Easton 1A (Grinsell) (SU 2108 5927) is anomalous in respect of its location and size. Situated just below the summit of Easton Hill, a prominent chalk ridge that juts out into the eastern end of the Vale, it is clearly meant to be seen from the north, from the floor of the Vale. The size of the monument is most striking, an overall diameter of some 67 metres. This is one of the largest disc barrows in southern Britain, similar, if not larger than Grinsell’s Wimbourne St Giles 28, part of the Oakley Down cemetery. The morphology of the barrow is unusual, too, with the presence of an external ditch. The barrow appears not have been investigated.
One final anomalous barrow is the large oval barrow located near Burbage (SU 2381 6100). This innominate and uninvestigated barrow is located on the narrow watershed between the Rivers Bourne, Avon and Dun. It appears to address the Dun system but would not, given the flatness of the terrain, have been very visible but its location set between the furthest limits of these three river systems would have been highly significant, given the association between river valleys and round barrows.
The correlation between clay-with-flints and barrow location may well be significant. A brief perusal of soil types and the distribution of round barrows on the Marlborough Downs and in the Savernake Forest area appears to demonstrate a similar aversion to clay-with-flints. This is not the case on the northern fringes of Salisbury Plain where the Down Farm group, among others were constructed on top of Clay-with-flints. From a practical, modern perspective clay-with-flints presents a number of difficulties. Vehicles become easily mired, farm machinery must avoid overdue compression of the soil or cause crop failure and it is particularly awkward to walk or cycle across. Both Colt Hoare (1810: 191) and de Mallet Vatcher (1960: 340) remarked on the difficulties or ‘clamminess’ of the soil. The construction of mounds would not be aided by the nature of the soil and perhaps the lower concentrations of barrows on the northern edge of Salisbury plain demonstrates the reluctance of barrow builders to construct monuments on this type of soil. Furthermore, it is often suggested that Clay-with-flints supported dense woodland (Barber et al 1999: 57) in the Neolithic and, perhaps, this was still true in the Early Bronze Age, possibly limiting both the choices of location for barrow construction and potential viewsheds towards and away from the mound. Alternatively, Field (1998, 316) suggests round barrows were located on marginal land and Peters (2000, 353) notes that what she terms “Inconspicuous barrows” are located on the edge of MBA field systems. Despite its innate difficulties, clay-with-flints is a fertile soil and, perhaps, the siting of barrows reflects the development of an early agricultural landscape.
Round Barrows and Routeways
In summary, then, the distribution of round barrows in the Vale is unequal. There are proportionately more round barrows sited along the northern flank of the Vale than anywhere else. Perhaps significantly, these barrows are found in groups in association with routes running transversely across the Vale (i.e. north – south). The Wootton Rivers group are located adjacent to a series of tracks passing through Ram Alley and heading towards the byways running along the dry valleys flanking either side of Easton Hill and leading into the eastern portion of Salisbury Plain. The Gopher Wood group overlook two routes running from Hare Street in Wilcot which converge 600 metres north of the barrows before following a dry valley north across the clay-with-flints of West Woods and meeting the chalk at Fyfield Down. Similar locations for the barrows at Knap Hill and west of Tan Hill have already been noted but the siting of the barrows at these respective locations does seem to exhibit a desire to be seen from the routeway.
The possible explanations for this deliberate and clear patterning are not straightforward. The lack of much positive evidence for EBA activity on the greensands of the Vale has already been noted. Consequently, the answer may lie to the north. It has already been suggested that there was a drift of activity away from the Vale towards the end of the Neolithic towards the twin poles of Avebury and Stonehenge, exemplified by the lack of EBA finds from Marden henge. Perhaps the placing of barrows on the edge of the chalk scarp, close to major transverse routeways was intended as a statement for travellers along those routes. If we accept the ecotone between the chalk and greensand as both a natural and human boundary zone, the siting of funerary monuments, metaphors for a journey between this world and another, along this liminal zone makes sense. Travellers approaching the Pewsey Downs/Avebury complex passed through a series of loose “zones” as they journeyed towards the henge. The outermost of these zones was concerned with dialogues with other worlds, the world of the dead and the outside world or perhaps the reverse is true with the “dead zone” being at the heart of the landscape around Avebury. The siting of barrows along natural routeways served as “signposts” for travellers that they were entering a new territory/polity, reminding them of the ancestral claim the present residents had on the area, the antiquity of that claim and the wealth and status contained both within the mounds and the territory.
Barrows along the southern side of the Vale seem to fit into the pattern of distribution along river valleys noted by McOmish et al. (2002, 46). Even the two bowl barrows located at Urchfont are very close to the furthest arm of the Water Dean System. The only barrow that, perhaps, defies this analysis is Easton 1A, the large disc barrow situated at the northern end of Easton Hill. As noted above, the barrow is flanked on either side by byways of some antiquity, following dry valleys linked to the Bourne system. It is also worth noting that the Vale reaches its narrowest width north of Easton, only some 3 kilometres wide and that the barrow overlooks the headwaters of the Avon, Bourne and Dun. Perhaps the size of the disc barrow reflects the significance of the location and the aspirations of its builders