The Use and Reuse of Stone Circles

by Richard Bradley

  The structural sequence at Clava Cairns and recumbent stone circles

  It is wrong to consider such structures in isolation – the stone circles were only one component of a more complex monument and generally date from a developed stage in their periods of use (Fig. 7.3). That is clear at both Clava Cairns and recumbent stone circles. In the case of the passage graves at Balnuaran the monoliths were erected after the cairns had been built and possibly when they were already going out of use. On that site the standing stones were supported by a ramp of rubble which had been piled against the outer edge of the tomb and obscured the decoration on its kerb. That deposit continued across the entrance to the passage and impeded access to the interior. At the same time, the history of these monuments followed an orderly sequence. At the central ring cairn at Balnuaran radial divisions which would have been visible in the surface of the cairn were extended outwards from the perimeter wall. They seem to have provided the foundations for some of the monoliths. At both passage graves on the same site individual stones in the outer circle were chosen to match the appearance of the nearest component of the kerb (Bradley 2000, 126–29).

  A similar sequence was followed at recumbent stone circles, and at Tomnaverie, as at Clava, radial divisions in the surface of the cairn established the positions where some of the standing stones would be erected in a subsequent phase. On the south-west side of that monument the cairn was built on a massive scale and its outer edge was supported by steps cut into the natural surface of the hill. It seems as though it was accepted that this part of the structure would support the weight of a recumbent stone, even though that was not to happen until the use of the monument came to an end (Bradley 2005, 48–51). In contrast to the improvisation that may have characterised the construction of older monuments, here the structural sequence seems to have been preordained. It is another reason for distinguishing these monuments from those of an earlier period.

  Figure 7.3. The structural sequences at Balnuaran of Clava northeast cairn and Tomnaverie. Information from Bradley 2000 and 2005.

  In one sense such developments owe their distinctive character to local architectural traditions. In another, they epitomise more general trends among the stone settings of the British Isles where some of the same components occur widely, although they are seldom discussed. One feature is the grading of the monoliths by height. It can be found at other monuments in the north but has also been recognised in Cornwall, Wessex, south Wales and Ireland (Burl 2000). Another is the emphasis on southern or south-western alignments. This is often overlooked but it is a feature that unites structures of very different dates.

  The most obvious examples are the recumbent stone circles of north-east Scotland, and a series of smaller structures in south-west Ireland. Both groups were considered together by Burl (2000) but their histories were separated by a thousand years (O’Brien 2004). What is significant is that similar concerns were expressed at monuments that were not contemporary with one another. Certain ideas lasted a particularly long time.

  Subsequent developments in northern Britain

  What light do new excavations shed on the chronology of Scottish stone circles? Three main elements can be identified.

  The first point to make is that Clava Cairns date from the Copper Age and Early Bronze Age. They are not Neolithic, as originally thought. The dates from Balnuaran, Newton of Petty and Raigmore suggest that they were constructed and used between about 2300 BC and 1800 BC (Bradley 2000, 160–61). Similarly, the dates from Hillhead suggest a terminus post quem of 2200–2000 BC, but those from Tomnaverie can be interpreted in more than one way. In one version the primary cairn has a terminus post quem in the 25th century BC. In another, it is in the 23rd or 22nd century (Curtis and Wilkin 2012, 241–24).

  Recumbent stone circles were not freestanding structures, but represented the end point in a protracted sequence. The distinctive form of the recumbent and the flankers could have reinforced that message, for these features resemble nothing so much as a closed door. That may be why the smoother, less-weathered surfaces of the monoliths are on the outside of the monument (Welfare 2011) – they were meant to be seen from the exterior because the interior was no longer accessible. But none of these features is universal, even within the heart of their distribution in north-east Scotland. Now it was quite common to erect a ring of monoliths with an emphasis on the south or southwest without including other elements. Conversely, the same arrangement could be observed at monuments which were not associated with stone circles. For example, the well-preserved ring cairn at Sketewan in Strath Tay enclosed a cremation pyre and was provided with a boulder kerb which was most conspicuous towards the southwest (Mercer and Midgley 1997).

  The first structure at Croftmoraig is relevant to the argument. With a diameter of about 12 m it was smaller than a recumbent stone circle and had fewer stones, but, as Chapter 10 will demonstrate, its monoliths were graded by height towards the southwest. Like Tomnaverie, its site emphasised the position of a prominent mountain. In this case there was no cairn, nor was there any specialised structure on its south-western perimeter. Its absolute date is not known, but a clue is provided by the putative graves beside the portal stones. (Unburnt bone would not survive in the acid soil.) Radiocarbon dates from other Scottish sites suggests that these features could date between 2400 and 1600 BC (Ashmore 2004). The absence of cremated bone in primary contexts may be equally significant as this practice was adopted about 2000 BC. The age of the first monument is difficult to assess, but the fact that it was so obviously graded towards the southwest rules out a Neolithic origin.

  Smaller stone settings of the Early and Middle Bronze Ages in the north

  The new excavation reassessed the structural history of Croftmoraig. Far from being one of the oldest structures on that site, the inner oval was among the latest. That is especially important in view of its resemblance to the damaged stone circle at Tuach. There are other monuments of much the same form, in particular a well-excavated site at Cairnwell, south of Aberdeen (Rees 1997). Taken together, these sites suggested that small circular or oval stone settings like these were the last to be built. Their dates cover the later part of the Early Bronze Age as well as the Middle Bronze Age. They fall between about 1900 and 1600 BC at Tuach, and between 1600 and 1200 BC at Croftmoraig. At Cairnwell, a circle of monoliths, 9 m in diameter, was erected around the perimeter of a small ring cairn. This structure is dated between approximately 1400 and 1000 BC. There was a similar sequence at Moncreiffe where a ring of standing stones occupied the site of an older henge associated with a timber circle, but there are no radiocarbon dates (Stewart 1985). A timber circle outside the south entrance of the henge at Broomend of Crichie could also be relevant here because it was only 9.5 m in diameter and the posts appear to have been graded by height; in this case they were tallest towards the northeast. It produced dates between about 1800 and 1500 BC (Bradley 2011, 40–42).

  In northern Scotland, earthwork enclosures of similar proportions date from the same period. One example at Lairg has a terminus post quem of about 1600–1300 BC and a comparable earthwork at Pullyhour had a terminus post quem of 1620–1450 BC for its first phase and 1370–1130 BC for its second (Bradley 2011, 133). Neither contained a setting of monoliths and both sites are considered as small henges. Their dimensions were like those of stone circles of similar age. The interior of the earthwork at Lairg measured 8 × 6 m, and at Pullyhour the equivalent figures are 6 × 7 m. These figures can be compared with the dimensions of the inner oval at Croftmoraig (9 × 6 m) and the diameter of the stone circle at Tuach, which is estimated as 8.5 m.

  Some of these structures resemble a group of stone circles on Tayside already discussed by Burl (2000, 243–51). They have few radiocarbon dates, but share a number of characteristics with the monuments considered so far. They are less than 10 m in diameter, but can be slightly elongated towards the southwest. There are comparatively few monoliths, generally no fewer than six and no more than ei
ght; this is consistent with the evidence from Croftmoraig and Tuach. In most cases the monoliths increased in height towards the southwest. A number of these structures are associated with cremation burials in Collared Urns and Cordoned Urns, like those at Tuach, or occasionally in undecorated vessels of the kind found at Croftmoraig.

  These monuments are commonest in the centre of Scotland but, as the evidence from Tuach shows, they can be found more widely. The same applies to the stone settings known as four posters (Burl 1988). This is really a misnomer, as excavation has sometimes located other components. These sites are usually characterised as square or rectangular monuments, at odds with the general preference for curvilinear structures, but they can be associated with circular mounds or kerbs. One example at Balneaves was inside a ditched enclosure interpreted as a henge (Russell-White et al. 1992, 289– 301). Like other small stone settings, four posters can be associated with cremation burials, some of which were in Collared Urns. That is consistent with the evidence from Balneaves where the cremations buried beside the stones are dated between 1850 and 1550 BC. Two characteristics of these sites are especially relevant here. Not only were they of similar dimensions to the circular stone settings – the enclosure at Balneaves had an internal diameter of 8 m – the monoliths at these sites were graded, with the tallest towards the southwest.

  The last ring cairns and kerb cairns in the north

  Ring cairns had a long history in the north. It ran in parallel with some of the monuments already considered in this chapter. For instance, the small stone circle at Moncreiffe was erected around one of these monuments (Stewart 1985), and the same seems to have happened at Cairnwell where the cairn replaced a wooden enclosure associated with burials deposited between about 1400 and 1050 BC (Rees 1997). At present it seems as if the setting of monoliths at that site was among the last of its kind, but the structure it surrounded possessed another distinctive feature. It had an entrance towards the south, but when the ring cairn was built this feature was not retained. Instead its position was occupied by a substantial slab set on end behind two uprights. The excavator compared this arrangement with the settings found at recumbent stone circles. The details of this sequence are important. It is sometimes claimed that the recumbent and its flankers represent a blocked entrance. At Cairnwell this was exactly what was found. But it happened almost a thousand years after recumbent stone circles were built.

  There is similar evidence from a few kerb cairns. Again they are a kind of monument with an extended history. At Temple Wood two of these structures were built inside an older stone circle whilst a third was located just outside it. They are associated with cremation burials dated between 1450 and 1200 BC (Cook et al. 2010). Although the cairns are solid structures, those inside the stone circle included a narrow entrance towards the south which had been carefully closed. That was not a unique instance. Like the site at Cairnwell, a kerb cairn on the Sands of Forvie included a large upright slab at the southernmost point on its perimeter (Ralston and Sabine 2000, chapter 2). In this case the monument was not built until the Late Bronze Age or conceivably the Early Iron Age. It is later than the structures at Cairnwell and Temple Wood and has a terminus post quem of 1000–380 BC; charcoal associated with a cremation burial inside the structure dates to 970–370 BC. This monument can be compared with the small ring cairns at Laikenbuie and Balnuaran of Clava South discussed in Chapter 6. They were directed towards the south or southwest respectively and both were probably built between about 770 and 400 BC. The example at Laikenbuie incorporated coloured stonework in the same manner as an older monument. Such evidence is very limited, but suggests that certain architectural devices had a history extending from the Chalcolithic period to the Early Iron Age.

  Why did it happen? Perhaps the important point is to pay less attention to the classification of these buildings and more to the ideas that they were meant to convey.

  Closing by enclosing

  The excavations undertaken at stone circles since 1994 have sprung a series of surprises – or, rather, they have sprung the same surprise in a number of different settings (Fig. 7.4). It was not entirely predictable that the stone circles around the Clava Cairns would turn out to be a secondary development – they may well have been constructed as those monuments went out of use. Exactly the same was discovered some years later during excavations at recumbent stone circles. Again the cairns were older than the monoliths that enclosed them, but in this case one of the last features to be built resembled a formal entrance blocked by an enormous stone. In both cases the main emphasis was towards the southwest.

  The excavation of Broomend of Crichie between 2005 and 2007 produced an equally unexpected result, for here a large henge monument turned out to be later in date than the stone setting inside it (Bradley 2011, chapters 1 and 2). A similar result came from a new excavation in Wales, and a review of comparable monuments suggested that this might have been a more general pattern (Gibson 2010b). Far from initiating activity at these locations, the people who constructed henges restricted access to sites with an already established significance and may even have closed them down. It is no accident that, like other monuments in Scotland, some of these earthworks are associated with Early Bronze Age burials.

  Figure 7.4. Four Scottish monuments with blocked entrances or other features focusing on the south or southwest. Information from Ritchie 1971; Rees 1997; Ralston and Sabine 2000; and Cook et al. 2010.

  The same project reviewed the evidence from a series of small henges in northern and north-eastern Scotland. It showed that a number of them shared the same orientation as the later stone circles, but at the time the significance of another observation was overlooked. A recurrent feature of these enclosures is that their entrances had been completely or partly blocked during the prehistoric period (Bradley 2011, chapter 5). It could be done by erecting a monolith to restrict access to the interior; this happened at Migdale. The entrance could be put out of use by a deposit of stones, as evidenced at Pullyhour. Otherwise the causeway in the enclosure ditch could be partly removed. A good example is found at Wormy Hillock (RCAHMS 2007, 56–58). The point that should have been made is that all three categories of monument seem to have been closed.

  It happened over a wide area. There are ring cairns in other regions that were provided with an entrance, and examples where that feature was closed when the monument went out of use (Ward 1988). In Northern Scotland, the Clava chambered tombs were provided with an external platform that impeded access to the passage; at recumbent stone circles in the northeast of the country an enormous block was placed between the tallest stones; and at some of the henge monuments various obstacles were placed between the ditch terminals in the entrance. The same processes can be seen at the monuments excavated since 2011. At Hillhead they are illustrated by the remains of the recumbent and its flankers which block an earlier entrance, while a break in the earthwork perimeter at Tuach could have been removed during a subsequent phase. At Croftmoraig a gap in the outer wall was filled by a decorated stone and may have been yet another blocked entrance. The argument could go even further. For example, at the Sketewan ring cairn and the Middle Bronze Age kerb cairn at Stromtoller, neither of which had any entrance, the outer kerb was built on a larger scale along the south-western perimeter of the monument (Mercer and Midgley 1997; Ritchie 1971). It occurred at other sites in Northern Britain where the evidence is less clearly documented.

  Such observations call for a single explanation. Perhaps this was done to prevent something threatening from entering or leaving the interior. These monuments are commonly associated with human remains and might have been where the rites of passage concluded. Were these buildings closed to prevent the dead from returning to the living, or did it happen in order to protect the power of these places and to separate them from the world outside? Both suggestions amount to saying that certain directions might have been considered especially dangerous. They were associated with those who had died, and the boundary betw
een life and death was expressed by constructing a blocked entrance facing southwest. This can be thought of as the dark side of the sky, for it is where the sun goes down and sets (Bradley 2016; Henty 2014). In ancient Ireland this direction was associated with the dead (O’Brien 2002), and a similar idea is recorded in Viking Age Scandinavia (Eriksen 2013).

  Perhaps it was this notion, or something like it, that proved so tenacious that it found expression in a range of monuments built in Scotland over a thousand years. It extended from a large structure like that at Hillhead to the small ring cairn at Laikenbuie. The last structures were built on a lesser scale than those of the Neolithic period or even the earliest Bronze Age and no doubt acquired new connotations as older beliefs were allowed to lapse. But there never was a complete rupture with the past. For that reason the next step is to consider the ways in which stone circles were reused over the course of time. That question provides the subject of Chapter 8.

  CHAPTER EIGHT

  Histories of Reuse

  Richard Bradley

  Introduction: closing and reopening

  The previous chapter suggested that many monuments were closed as their histories came to an end. That applied to stone circles like those associated with ring cairns and passage graves, but the same argument extended to later structures including earthwork enclosures and kerb cairns. It is ironic that the closure of these sites was rarely lasting, for recent fieldwork has shown how often they were reused. Rings of stones remained a significant feature of the landscape for such a long time that they were associated with renewed activity in four different phases: the later Bronze Age, the Early Iron Age, the Roman Iron Age, and the Pictish period.