The Use and Reuse of Stone Circles

by Richard Bradley

  Trench 1 was designed to investigate the perimeter of the monument and its relationship to a nearby monolith (1) in the outer circle. The two were so close together that it was important to establish whether they dated from different phases. Moreover Piggott and Simpson never sectioned that boundary, with the result that its structure has been difficult to interpret – was it some kind of bank, as they suggested, or might it have been a drystone wall? On the western edge of the site it seemed to have been built on a shallow terrace cut into the edge of a natural mound. Trench 1 was located at one of the few points where the large stone blocks defining the limits of the monument had been displaced. It provided a rare opportunity to examine the perimeter down to its foundations.

  Trench 2 was to extend between a monolith (16) belonging to the inner oval and the centre of the site where Piggott and Simpson recorded ‘a flat natural boulder embedded in the old surface [and] a shallow hollow ... with a sparse scatter of comminuted charcoal, suggestive of a hearth’ (1971, 5). The boulder had obviously posed problems in the 1965 excavation as there are several detailed photographs of it in the project archive. It was important to establish whether the ‘hearth’ had been fully excavated. Part of Trench 1 was intended to follow an unexcavated baulk between two of the trenches dug 47 years before. It also examined Monolith 16 as the 1971 site plan suggested that it might have been bedded in the terminal of the ring ditch.

  Figure 4.10. The published plan of the 1965 excavation, after Piggott and Simpson 1971, with the positions of the trenches excavated in 2012.

  Figure 4.11. Sections of the monument during the 1965 excavation. After Piggott and Simpson 1971.

  Trench 3 was to follow a similar principle. Its aim was to investigate parts of two unexcavated baulks in the filling of the ring ditch at a point where the last excavators recorded fragments of charcoal and sherds. Indeed, most of the artefacts recovered in 1965 came from this part of the site. The trench was to extend between two monoliths in the inner oval (20 and 21), but in this case their relationship to the material in the ditch was already documented.

  Finally, Trench 4 was to investigate one of the large blocks of stone forming the perimeter of the monument on its eastern side. In the 1971 site plan it is shown overlying a pit containing charcoal, and in Piggott’s unpublished drawing in the project archive this feature is labelled as a ‘hearth’ – an interpretation that was not carried through to the final report. If any stratified charcoal still remained, it might provide a terminus post quem for the construction of the wall.

  The methods used in the 1965 excavation

  Before turning to the detailed results of fieldwork in 2012 it is important to understand the character of the previous excavation. This can be worked out by combining the information contained in photographs and colour slides of the original project with observations made in re-excavating Piggott’s and Simpson’s trenches. The 1965 excavation was one of a series of projects funded by the Ministry of Public Buildings and Works with a view to presenting important monuments to the public. The project file held by Historic Scotland shows that the work took a little over three weeks and was carried out by Edinburgh University students and workmen from Stirling Castle; I am grateful to Oliver Lewis for this information. The prehistoric deposits were removed and all the trenches extended down to the natural subsoil. That could not have happened unless they were removed rapidly – the very small areas investigated with hand tools in 2012 took almost a fortnight to excavate.

  It is clear that the superficial level had been mixed by the action of worms, with the result that separate layers could not be distinguished except where features were cut into the clay. Only where the sockets for posts or monoliths contained packing stones could their positions be recognised at a higher level. The surface from which these features had been cut was well above the material of the glacial mound. Piggott and Simpson documented this evidence in their photographs and section drawings but confused the issue by describing the clay forming that mound as the prehistoric land surface (Fig. 4.11). That was incorrect – in fact this was the first level at which the outlines of many of the features could be recognised in plan. It is clear that the excavators dug though the prehistoric ground level and into the subsoil. It was because the tops of the stone sockets seemed to be above the prehistoric ground surface that Welfare supposed that an artificial platform had been constructed over an earlier building.

  When the overburden was removed in 1965, the amount of quartz in the separate grid squares was carefully recorded, but this material was not retained. Perhaps it was because the excavation proceeded directly to the natural boulder clay that small fragments of cremated bone were overlooked. None is recorded in the publication of the 1965 excavation, but four pieces were identified in 2012, one of them in the filling of the earlier excavation. Their discovery is important for two reasons. Although Burl refers to the discovery of burnt bone at the centre of the monument (1995, 157), his statement is misleading, and no human remains were recorded in the first excavation. At the same time, the fragment found in Trench 2 in 2012 has a radiocarbon date in the Late Bronze Age: a period which is significantly later than any of the structural evidence from Croftmoraig.

  These are the only criticisms of an excellent excavation. The site plan is extremely accurate – where the two excavations overlapped no further features could be identified (those examined in 1965 were not reopened). Moreover, Piggott’s and Simpson’s trenches were exactly where they showed them in their report and the unexcavated baulks conformed precisely to the positions indicated in their survey of the site. Their published section drawings are equally accurate. It is only unfortunate that Piggott’s plan of the central part of the monument, now in the NMRS, was not used in the excavation report as it is more detailed and contains more information than the plan that was published in 1971.

  Details of the 2012 excavation

  Because such limited areas were investigated in 2012, this account begins by considering each of the trenches separately.

  Trench 1

  To the west, Trench 1 extended from the perimeter of the platform, across the line of the ‘rubble bank’ described by Piggott and Simpson (Figs 4.12–4.14). It continued down the steep slope outside it. To the east, it incorporated part of the socket for Monolith 1.

  The excavation identified an artificial ledge or terrace dug into the surface of the natural mound (Context 3). It was filled with large boulders which constitute Context 2. On the bottom of this feature was a thin layer of soil (Context 5) which probably resulted from a brief period of erosion before the stones were introduced. They should not be interpreted as a ‘kerb’. Instead Contexts 2, 3 and 5 represent the foundations of a substantial wall. The blocks which defined this feature had been displaced, which is why it was accessible for excavation. The radiocarbon dates associated with the boundary wall are considered on p. 69.

  To the west, the ground was scarped by Context 4 which had the effect of making the perimeter wall look higher than was actually the case. To the east it was notable that the socket for Monolith 1 was exceptionally shallow. The reason for this anomaly will be discussed in Chapter 10.

  Trench 2

  Trench 2 extended from Monolith 16 in the inner stone setting as far as the central stone and the ‘hearth’ that was identified in the 1965 excavation (Figs 4.12, 4.15). It followed the course of an unexcavated baulk left between two trenches dug during the earlier project.

  Three features were identified. The first was a shallow basin-profile cutting into the natural clay (Context 5). It could have been the western terminal of the ring ditch reported by Piggott and Simpson, but too little of this feature remained undisturbed for this idea to be taken further.

  It was cut by the socket for Monolith 16 (Context 3). It had been packed with large rubble including a block of quartz, and had been enlarged by an animal burrow containing a large piece of worked quartz.

  The excavation also investigated the area of the ‘hearth’ r
ecorded in 1965. According to Piggott and Simpson, it was associated with a natural stone at the centre of the monument. This was relocated in 2012, but was not horizontal, as the previous excavators had implied. In fact it dipped towards the west where the deposit of charcoal had been found.

  The central stone was clearly a glacial erratic bedded in the natural boulder clay (Context 5), but its upper surface would have been visible at the level from which the stone sockets were cut (Fig. 4.16). The previous excavators had dug round the edge of this stone to work out its relationship to the local geology and even lifted one end using an A-Frame before they were satisfied that it was of natural origin. That may be why it is no longer level in the ground (I am grateful to Helen Mulholland, who worked on the project, for this information). Their site plan suggests that they were not the first to investigate it (Fig. 4.17). It seems as if part of the erratic rock had been exposed by removing the glacial clay during the prehistoric period. It was the hollow that resulted from this process that contained the charcoal interpreted as a hearth. No trace of this material survived.

  Figure 4.12. Plans and sections of Trenches 1 and 2.

  The central stone is described in more detail in Chapter 10.

  Figure 4.13. The stone foundation for the enclosure wall viewed from the interior of the stone circle with Monolith 1 in the foreground (Aaron Watson).

  Figure 4.14. The rubble foundation of the enclosure wall, set in a shallow trench on the west side of the circuit. The large stone blocks making up the wall have been displaced. From a colour slide of the 1965 excavation now held by NMRS.

  Figure 4.15. Trench 2 at the 2012 excavation showing an unexcavated baulk from the 1965 project and the position of the central erratic in the foreground (Aaron Watson).

  Figure 4.16. The glacial erratic at the centre of the stone circle showing where large flakes had been detached in the past (Aaron Watson).

  Figure 4.17. The centre of the stone circle during the first project, showing the glacial erratic and evidence that people had dug around it in the past (scale in feet). From a colour slide of the 1965 excavation now held by NMRS.

  Figure 4.18. Plans and sections of Trenches 3 and 4.

  Trench 3

  Trench 3 extended between two monoliths in the oval setting (20 and 21) and sampled the surviving parts of the baulks between four trenches excavated in 1965 (Figs 4.18–4.19). They were located over the unexcavated eastern arc of the ring ditch.

  Contexts 3 and 4 consisted of thin lenses of silt left on the base of the ring ditch in two trenches excavated by Piggott and Simpson. This material became increasingly greasy and organic with depth. The deposit survived only locally and contained small fragments of charcoal. Context 8 was a shallow hollow on the bottom of that ditch, filled with similar material.

  Contexts 5, 6 and 7 shared the same composition, but in this case they survived in the narrow baulks in between the areas investigated in 1965. Again these deposits became more greasy and organic towards the bottom of the ‘ditch’. They were excavated in 2.5 cm spits and contained traces of charcoal.

  Taken together, these layers are consistent with the filling of the ring ditch described in the report on the 1965 excavation. As well as charcoal, they contained two sherds and two tiny fragments of cremated bone. They also included a number of pieces of flaked or broken quartz. Radiocarbon dates for samples associated with this feature are considered on p. 69.

  The excavation also confirmed that the sockets for Monoliths 20 and 21 cut through the filling of the ring ditch. They were recorded in 1965, as was a stone-packed posthole. In 2012 they were drawn but not re-excavated.

  Figure 4.19. Trench 3 showing the positions of two of standing stones and the baulks left between the trenches excavated in 1965. The new trench extends between Monoliths 20 and 21 (Aaron Watson).

  Trench 4

  Trench 4 investigated a charcoal-filled pit (Context 6) which was shown on the site plan as underlying one of the stones of the outer wall (Figs 4.18, 4.20).

  On excavation it became clear that the stone was no longer in position. It had been moved slightly to the east. Context 5 marked its original location, as it was where the natural boulder clay had been compressed. The wall must have stood directly on the ground surface.

  The shallow pit recorded in 1965 extended beneath that stone. The upper edge of that feature had been overcut during the earlier project but its base was completely excavated and contained a small quantity of oak charcoal, a sample of which has been dated. If this feature did not underlie the wall, as Piggott and Simpson had implied, it may well have been dug against its outer face.

  It seems likely that the top of this pit had been disturbed by the plough before the stone was moved. There was an area of disturbed topsoil containing charcoal in the eastern half of the trench. The same deposit contained a tiny fragment of cremated bone.

  The excavated material

  Few finds were recovered in 2012, and some of them came from the filling of the previous excavation.

  Figure 4.20. Trench 4 showing the remains of the perimeter wall on the south-east side of the enclosure (Aaron Watson).

  Apart from charcoal samples, they consist of two body sherds from the filling of the ‘ring ditch’, four tiny fragments of cremated bone (two of them from the same context), and a small quantity of quartz flakes and smashed boulders.

  Bone

  Fiona Shapland

  The burnt bone is too small to be assigned to a species, although the single piece found in the filling of the first excavation comes from a long bone which had been fired to a particularly high temperature. It has been radiocarbon-dated to the Late Bronze Age. Two other fragments were in the filling of the ring ditch, and there was one in the disturbed ploughsoil close to the pit in Trench 4.

  Pottery

  Richard Bradley

  Both sherds are in the same fabrics as the Bronze Age pottery from the first excavation which was studied in detail by Alison Sheridan in 2005 (Bradley and Sheridan 2005, 275–81). They have no diagnostic features. They can be compared with the nine vessels found in 1965 which she dates to the later Bronze Age. Both are from the filling of the ‘ring ditch’ in Trench 3.

  Lithics

  Richard Bradley

  The stone artefacts found in 2012 are presumably related to the material mapped by Piggott and Simpson in their excavation report. It seems as if the pieces of quartz that they recovered were not retained. They stated that

  throughout the stony soil between the modern turf and the natural surface, and in the make-up material [the deposit overlying the ring ditch] there was a scatter of water-worn or angular lumps of white quartz. Such pebbles and fragments occur in the soil under natural conditions on the site, but the apparent increase of these within the area of the monument was tested by recording white quartz fragments by weight in each of the … cuttings … Figures [for a ten foot square] within the circle … frequently totalled between 30 and 40 lbs. (13.6 and 18.14 kg).

  (Piggot and Simpson 1971, 9)

  The small sample recovered during the 2012 excavation amounted to 25 unworked quartz pebbles and 42 other fragments, not all of them diagnostic. The number is extremely small but they did come from the same contexts as those recorded in 1965 and should have been deposited during a late phase in the history of the site, most probably when the inner stone setting and the perimeter wall were built. Since Piggott and Simpson recorded the weight of material in their excavation, it is useful to compare their figures with those from the new excavation. In the small collection from the 2012 excavation the average weight of a quartz pebble was 20.5 g, and for a worked or broken fragment it was just over ten times as high. If the same ratio applied to all the material recorded by Piggott and Simpson, the interior of the monument would have contained at least 1200 pieces of worked quartz and 700 pebbles. This assumption is impossible to verify, but is useful as a comparison with the material from Hillhead.

  Those recovered from stratified contexts in 2
012 take three forms. There are three small flakes of good quality quartz and 11 broken nodules of coarser quartz and quartzite, none over 4 cm in maximum dimensions. There is no evidence that any of these result from tool production. They contrast with a second group of much larger pieces with maximum dimensions between 8 and 19 cm, the majority being in the higher part of this range. They consist of large pieces of quartz and quartzite which had been smashed with some force. Few flakes were detached. These fragments were most frequent in the centre of the monument and could have formed part of the layer of ‘paving or cobbling’ reported by Piggott and Simpson. Two other pieces were used as packing for Monolith 16. This material is very similar to the fragments associated with the final phase at Hillhead reported in Chapter 2.

  Radiocarbon dates

  There are 12 radiocarbon dates from the project, eight of them funded by Historic Scotland and another by NMS. They come from four contexts: the base of the foundation trench for the perimeter wall in Trench 1; the lower filling of the ring ditch in Trench 3; the external pit in Trench 4; and the filling of the 1965 excavation in Trench 1. One sample was of burnt bone (probably human), and another was on a single piece of oak charcoal – the only species represented in the pit excavated in Trench 4. All the other samples were of single pieces of charcoal from short-lived species.