• Background
• Holywood North
• Trench 1
• The Ditch
• Internal Features
• Postholes
• Pits
• Other Features
• Trench 2
• Holywood South
• The Cursus Ditch
• The Postholes
• The Pits
• The Ring-Ditch
• Other Features
• The Twelve Apostles Stone Circle
• Discussion
• References

Background

The monuments are located on a sand and gravel subsoil, and ground cover varies between arable, pasture and patches of woodland. Aside from the stones of the circle, there was little upstanding evidence of the structures prior to excavation, the only evidence of their existence having come from aerial photography. In places, particularly in the wooded areas, there was evidence of severe rabbit damage. The excavation was intended to test the preservation of the monuments in the face of animal and agricultural damage over a very long period. As such, it was conceived as a sampling exercise. Large areas were opened to provide an extensive plan view of the monuments, but the total excavation of the features revealed was not contemplated. Rather, a number of features were selected, with the aim in mind of acquiring representative sections, and samples for radiocarbon and environmental analysis. A principal objective of the project was the investigation of the chronological sequence of the construction and use of the monuments. index

Holywood North

The North cursus from the air showing trench 1 on the right and trench 2 on the left

At the northern cursus, a large cutting (Trench 1) was opened, over the northern terminal, while a smaller area was excavated to the south, in order to investigate an entrance in the side of the monuments (Trench 2). index

Trench 1

The Ditch

Above this primary material, a massive wedge of sand and gravel could be recognised, sometimes containing tip-lines and lenses of clayey sand (070, 213, etc.). This was often surmounted by a deposit of dirtier gravel (026, 194, etc.). Both of these layers were heavily skewed toward the inner side of the ditch, and for this reason they are interpreted as having derived from the collapse of an inner bank. The material which lay above this deposit appeared to rest upon it unconformably, and there was some indication that a degree of truncation had taken place before silting had commenced. In places, and particularly at the extreme terminal of the ditch, this truncation gave the impression of a cut feature, although it is within the bounds of possibility that the fans of collapsing bank material had simply come to rest at a relatively oblique angle. In any case, the truncation was recorded as a deliberate recut (060, 193, etc.), which is present at all points in the ditch which have been investigated. This feature was somewhat irregular, varying in width and sometimes being v-shaped while in other cases it had more of a u-profile. In some sections there was a hint of some redeposited gravel and stones in the bottom of the feature, but in all cases the major primary fill was a bright orange sandy silt (055, 162, etc.). This was the only layer in the cursus ditch from which artefacts were recovered, including a retouched flint knife from the south-eastern cutting. Above this orange silt was another, darker silt (055, 158 etc.). These fills which succeed the putative recutting of the ditch contrast with the primary ditch fill, and indicate that the history of the monument which followed any reinstantiation was quite different from its first phase of use.

In the two cuttings on the eastern side of the cursus, the recut fills were sealed beneath a layer of iron pan, and this and the layer of degraded sandstone fragments above it (151) rested on an unconformity with the earlier layers. Seemingly a horizon of erosion preceded the renewed filling of the ditch, which may have been associated with a ploughing of the site. A thin, creamy-grey layer of silt (141) lay above this, and some quantity of humic plough-soil had slumped into the top of the ditch (002). index

Internal Features (by Matt Leivers)

The north cursus from the air

Of these features, five were demonstrated to be subsoil variations, whilst a sixth - a very large area of dark soil positioned in the centre of the interior of the cursus at its northern terminal - was an area of plough soil caught in a shallow depression in the natural. This contained glazed sherds and clay pipe stems.Of those internal features excavated which were prehistoric, several classes can be distinguished for the sake of description:index

Postholes
By far the most numerous of these features were postholes. From the aerial photographs it was apparent that the cursus ditch was ringed around its terminal section by a line of internal posts. Upon excavation however, it became obvious that this interpretation was insufficient to explain the variety of features visible.

Some of these features took the form of near-circular postholes with pipes visible in either section (most often) or plan (rarely). Cuts 184, 059, 188, 223, 068, 129, 201, 029, 015, 167, 145, and 057 are of this type. The cuts of both the pipes and pits were sometimes very difficult to distinguish since all fills (and the natural subsoil) tended to be variations of more or less sandy gravels. Generally of the same order of size were cuts 211, 127, 182, 131, 204, 220, 031, 196, and 209. These are most probably postholes of the same sort, the pipes being indistinguishable as a result of the homogeneous character of the fills.

North cursus: General view of excavations

In plan, these features can be seen to form the arc of posts visible on the aerial photographs. Notable amongst these are 184 which contained a great deal of charcoal in the pipe, suggesting that some posts at least were burnt in situ prior to removal, and 015 which cuts an earlier pit at the apex of the cursus terminal, indicating that the site was one of significance prior to the construction of the upstanding monumental elements. Cut 167 is also worthy of note.

This feature was extremely shallow and had the appearance of an highly truncated posthole, the pipe of which contained much charcoal. Other postholes close to 167 were not truncated to this extent, and this may suggest that 167 is also an earlier feature.

Other postholes were of different form. 217, 087, and 104 appear to be ramped postholes with no apparent pipes. These three are all located at the southern end of the eastern side of the cutting, in an area where the majority of the cut features were unusual. 094, for instance, was a pit of extremely large dimensions (2 by 1.33 metres at the top, and 0.90 metres deep as opposed to 0.85 by 0.74 metres at the top, and 0.40 metres deep for 184 - an `average' example). It seemed to be ramped on the western side, but if it had ever held a post then it had been deliberately filled after this had been removed, since three distinct layers of fill were distinguished, the middle of which consisted of a large number of stones of boulder size in a concentration which does not occur naturally on the site (such a concentration was also found in pit 224 - see below). Other postholes in this area were unusual: 037 and 108 were kidney-shaped holes with deeper postholes at the centre; whilst 039 had a depth of 0.77 metres compared to a diameter at the top of only 0.48 metres (averaged); 106 appeared to have been re-used after it had been filled initially. This group of postholes seem to form a unit in the sense that they are all different from the other postholes on the site and cluster in one area. It is also notable that at the point where these postholes occur is also the point at which the cursus changes alignment slightly, as if around some pre-existing feature which it was deemed desirable to enclose between the ditches. One possibility is that this group of postholes - possibly forming a small structure - are that feature.

The remainder of the postholes suggest a unified plan. Given their location immediately inside the cursus ditch, it seems plausible that they represent part of a revetment structure supporting a bank. This, and the presence of burnt material inside some of the postholes, might serve to explain the dense deposit of bank material in the ditch. If a revetted structure had collapsed catastrophically, perhaps as a result of the firing of the timbers, the result would have been a single event of deposition. Seemingly, this destruction took place soon after the building of the monument, since only a small amount of silting preceded the entry of the bank material into the ditch.index

Pits

102 was a cut 0.78 by 0.83 metres at the top, and 0.36 metres deep which contained two fills. The upper fill (135) was a layer of silting which contained some occasional charcoal flecks, and it is most likely that this represents the natural filling of the pit once it had been partially filled by the original fill 101. This fill (101) contained large amounts of charcoal, as well as numerous fragments of carinated bowl, a single piece of bone (possibly cremated), and some fragments of burnt nut shell. Stones in the fill also appeared to have been exposed to fire, but there were no indications that burning had taken place in the pit itself, this suggesting that the material was burnt elsewhere and only subsequently deposited in this feature.

224 revealed a much more complex sequence of events. The cut itself was almost three metres long and relatively shallow, but the various fills and cuts within in demonstrated an elaborate pattern of use. Whilst this pit was still open and empty it appears that a smaller, circular pit (average 0.33 metres in diameter, 0.27 metres deep) was cut (180) through its base, subsequently to be filled with a loamy sand (174) containing rounded stones distinct in size and shape from the natural gravels which also occurred in this layer, along with a quantity of charcoal. Above 174 was a deep layer consisting entirely of large stones (154) which had obviously been collected and deposited deliberately (this context is comparable to the middle fill of 094 - see above). Some of these stones (twenty percent) appeared to have been deliberately broken; the edges of some were still sharp. Through this stone layer had been cut a pit (152) to a depth of 0.38 metres with a length of 0.76m (WSW-NNE) and a breadth of 0.32 metres (SSE-NNW). This was filled with a compact loamy sand (168) which lay beneath a considerable quantity of charcoal (053) 2.60 metres long. The pattern of visible wood in this charcoal demonstrated that it was not a single object, but successive dumps of burnt material, possibly over a significant period (a lens of sand (054) within 053 suggests that these depositional actions were interrupted by some amount of time). 053 lay under a layer of sand mixed with gravel and large stones (051) which was in turn beneath a layer of granular pebbly sand (052). At some point the southern end of this complex was cut by 015 - a posthole with pipe forming part of the inter post arc.index

Other Features

Trench 2 (by Chris Fowler)

North cursus: Trench 2

Each of the postholes was excavated, with 006 producing a set of large packing stones rammed down the sides of the cut. There were no finds from these postholes, which were some 20-30 cm deep. While the innermost eight postholes were clear features with hard edges, 018, 113 and 111 were less distinct, and must be regarded as no more than putative features.

The ditch terminals were filled with a series of layers. Described in chronological order, the initial cuts 117 and 120 were respectively 105 cm and 110 cm deep below the current land-surface, which itself has been truncated since the Neolithic. These terminals both appear to be filled with gravelly and sandy material (075 and 122/132), probably from the bank which was constructed from the original ditch material. The north ditch terminal was noticeably square at its deepest points, while the south terminal is somewhat shallower and not so clearly defined at its sandy base. The shape of the cuts indicates a severe weathering cone, into which the bank material has slipped or been pushed quite uniformly. The weathering profile of the northern terminal (120) intersected with posthole 170. Despite careful excavation, it was not possible to ascertain the stratigraphic relationship between these two features. This means that three possible scenarios can be imagined to account for the relationship:

• the post line is earlier than the ditch, and thus the cursus, and cut by it; it represents one aspect of pre-cursus activity on the site.
• the post line is contemporary with the ditch, and is an integral element of the construction of the cursus; it was cut by the weathering back of the ditch edge.
• the post line is later than the cutting of the ditch, and cuts into the primary filling of 120.

The excavation of the ring-ditch did not produce any finds, and was bottomed some 20 cms below the subsoil surface level. Situated clearly within the entrance, the ring-ditch may have provided and obstacle to movement around or into the monument, or may have been a later addition to a partially buried monument, feeding upon its particular connotations of power and significance.index

Holywood South (by Maggie Ronayne and Matt Leivers)

The south cursus from the air

The crop mark shows two ditches which run for approx. 300m NW/SE and are between 30 and 40m apart at the northern and southern ends respectively. Unlike Holywood North, both ends of this cursus have square terminals. There are a number of gaps in the ditches, and several internal and associated features show on the aerial photographs.

The topsoil (001) was removed to a depth of 0.30 m. This material consisted of a sandy silt loam with a high gravel content in places. It did not produce any prehistoric stray finds. 001 lay directly on top of the natural subsoil (002) into which the prehistoric features were cut, apart from at the south-west corner and northern end (outside the terminal) of the trench where there appeared to be traces of the base of the mineral soil/layers. These were not investigated as a result of time constraints. In general, the features were heavily truncated, and as a result only the very bases of some features survived. The subsoil was very variable in texture and colour, ranging from fine and coarse sand, to gravel and sandy gravel and so on, although the consistency generally tended to be loose. The character of this material made those contexts with gravelly/sandy fills difficult to identify and excavate.index

The Cursus Ditch

South cursus

Within this cut was a complex series of fills. The lowest of these was 109. Occurring only on the northern (outer) side of the ditch, this was an area of very compacted gravel which probably represents the very first episode of weathering of the original ditch profile. Above this lay context 108, a loose sandy gravel, and above this was 105, a friable silt.

Above 105 lay 104: a fairly compact silty clay, which in turn lay beneath 106 and 107 - friable sandy loams. This last two fills are likely to be the same context, as it is possible that they are cut by 113, which may be a recut. Alternatively, they may represent episodes of weathering of the upper edge of the ditch cut. At this point in the stratigraphy, many contexts merge and mix, and it is almost impossible to determine whether these represent a recut or not. The contexts in question are 113 (the possible cut), 103 and 102 (fairly compact silty clay loams on the northern and southern sides respectively), and 101 and 065 (friable silty clay loams on the northern and southern sides). These last four contexts appear to be cut in turn by 110 - a possible narrow steep-sided cut feature with a rounded base in the centre of the ditch - filled in turn by 111: a fairly compact clay loam which is very similar to the clay loam 077 above it). 077 seals all of the earlier contexts in the ditch and is in turn sealed by 058 and 037 which appear to be more recent deposits and base of topsoil.

The cutting opened across the `entrance' to the cursus on the south-east of the trench showed a somewhat different pattern. Here, the ditch was a maximum of 1.20m deep. A section was cut north-south along the ditch (rather than across it), and the profile of the terminal at this point showed a gradual slope to 0.80m deep and then a much steeper slope to a flat base. This steeper section of the profile (119) may be interpreted as a recutting of the ditch, but if this is so, then the recut at this point in considerably more substantial than at the northern terminal.

If this recut existed, then its primary fill was 117: a loose to friable sand and gravel matrix. It may be more likely however that this material represents early collapse of an internal bank and that 119 is in fact the same cut as 112 (that is, the original cut of the cursus ditch, rather than a recut). The sequence of siltings throughout this ditch is complex: at this low level 117 interleaves with 116 (a friable sandy matrix), 114 (a friable gravel), and 115 (a very loose and friable sandy silt with large pebbles). On balance, these contexts seem to be early bank collapse and weathering of the edge (in the case of 116).

Above these fills are a series of more humic silty clays (010, 011, 012), and silty sands (006, 007, 080). Over this latter series are a pair of contexts (047, 004) which are friable silt loams, cut by 118, which may be a small pit or posthole, filled by a firm silt loam (067). This was sealed by 038/028 - friable silt loams - the latter of which had a high proportion of stone in the matrix. Four silt loam contexts filled the ditch above this point (013 with charcoal flecking; 008 with a notable sand content and charcoal flecking; 006; and 003 - residual plough layer). No features were found in the entrance area outside of the ditch cutting.index

The Postholes

084 was a posthole, 0.36 m deep, cut into a large 'lump' of charcoal-rich material (092) inside the cursus ditch to the south. This feature contained three fills: a weathered charcoal-rich fill at the top (062) representing the collapse of the top of the burnt post; the post-pipe fill (063); and the post-pit fill (078).

081 was a postpipe located to the north of the cursus, outside of the terminal. There were two fills: a burnt post layer (082), and a humic fill (088). These were within a small post-pit (095, 120).

One posthole (074) was found immediately outside of the entrance to the cursus in the south-east corner of the trench. This had one sandy fill (073) and was 0.25m deep.index

The Pits

It was notable that a number of these cut features, both pits and postholes were located quite close to the edge of the ditch. If, as argued above, the bank was a simple dump of material quarried from the ditch, it must surely have spread over this area. This implies that either the features were cut through the bank, post-date the erosion of most of the bank material, or pre-date the construction of the cursus. The latter interpretation is preferred.index

The Ring-Ditch

A stakehole (018) in the south-west quadrant, immediately outside the ditch itself, was the only indication of any possible associated activity. This was close to the area of the recut (055) and deposit of compacted material and may be related to it.index

Other Features

Context 092 consisted of a compact loamy sand with a high proportion of rounded stones of all sizes. Two connected oval features made up this context, and a tenuous interpretation sees it as a post setting (context 084 which cut 092 appears to be a posthole).

Three charcoal rich `mounds' (076, 085, and 086) were investigated. These may possibly be the basal fills of pits which stand proud of the present subsoil surface due to the greater truncation of the softer material through which these features cut. index

The Twelve Apostles Stone Circle

The stone circle

However, the results of the survey were largely negative, perhaps showing no more than undulations in the subsoil. In order to test this outcome, a trench 20 x 10 metres in extent was excavated on the southern side of the circle, cutting across the circuit of the stones. Here, the topsoil came down onto an earlier, more mineralised ploughsoil. Beneath this, the natural was found to be featureless.

Discussion

Both the monuments seem to have stood for a comparatively short while before the bank material started to return to the ditch. Possibly, the southern cursus had a dump bank which began to erode back into the ditch quite quickly. In the case of the northern cursus, the collapse of the bank may have been deliberate, yet here there are strong indications that this did not bring the use of the site to an end. The post fa?ade across the eastern entrance may post-date the bank collapse, and may be contemporary with the recutting of the ditch, and the deposit of pottery and stones in the entrance terminal. Furthermore, there are ring-ditches on both sites, located in both cases in direct association with an entrance through the cursus ditch, positioned so as to dominate the point of access to the monument. So, although both monuments may have had a principal phase of use which was very short, this fits into a history of the use and modification of the immediate landscape which was much more extensive.

Another point which is of interest relates to the unusual use of stones within the features associated with both monuments. Small boulders or cobbles formed distinct layers in the unusually large posthole 094, and the pre-monument pit 224, both inside the northern cursus. In both of these cases, the act of deposition appears to have been quite deliberate, forming one element in an elaborate sequence of fillings. Also in the northern cursus, numerous pebbles were deposited together with pottery and flint waste in the northern terminal of the eastern entrance, subsequent to the collapse of the bank. Neither stones nor cultural material was found in the southern terminal. In the southern cursus, angular gritstone pebbles were found in pit 091, another large feature with a singularly complex series of fillings. Moreover, a group of cobbles was found at the bottom of the cursus ditch at the entrance terminal, echoing the terminal deposit at the northern cursus. In all of these examples, the stones concerned were quite distinct from those found in the surrounding subsoil, and they were certainly more numerous.

Recently, a number of authors have pointed out the importance of materials and substances drawn from the earth (chalk, soil, stone, wood) to Neolithic architecture and cosmology (e.g. Bender 1992). I have used the term `cosmological engineering' (Thomas forthcoming) to describe the way in which the use of these materials involves not simply the manipulation of their substance, but of their meanings and connotations as well. In building monuments and placing deposits in the earth, people were able to re-organise the landscape in microcosm. At Holywood, the massive feat of constructing the cursus monuments, re-creating places and establishing pathways across the local topography, was complemented by more intimate events, which reconfigured the land in subtler ways.index

References

Barrett, J.C., Bradley, R.J. and Green, M. Landscape, Monuments and Society: The Prehistory of Cranborne Chase. Cambridge: Cambridge University Press, 1991.

Bender, B. Theorising landscapes, and the prehistoric landscapes of Stonehenge. Man 27, 1992, 735-56.

Bradley, R.J. and Chambers, R.A. A new study of the cursus complex at Dorchester on Thames. Oxford Journal of Archaeology 7, 1988, 271-290.

Brophy, K. Undergraduate dissertation. University of Glasgow, 1995.

Thomas, J.S. An economy of substances in earlier Neolithic Britain. In: J. Robb (ed.)Material Symbols: Culture and Economy in Prehistory. Carbondale: Southern Illinois University Press, forthcoming.

 index