bronze age

5.4.2 Cosmology and Bronze Age monuments

Kilmartin Valley, showing the remains of standing stones and cairns. ©RCAHMS

The belief in an existence of an afterlife is amply attested by Chalcolithic and Early Bronze Age funerary practices (see section 5.5), and is also arguably implicit in the practice of cremation. Other information about the nature of belief systems is provided by ceremonial monuments, a wide variety of which were constructed in Bronze Age Scotland. These range from cairns built for the disposal of the dead to stone rows, stone circles and various different types of ‘hengiform’ monument. In many cases, these occurred in monument complexes that are likely to have spanned the third and second millennia BC. The dense concentration of monuments in Kilmartin Glen, for example, includes a linear cairn cemetery, henge, stone circles, stone rows and standing stones, creating an extensive and impressive sacred landscape (e.g. Scott 1991; Cook et al. 2010). At Broomend of Crichie, there were originally two stone circles, one of which incorporated a cairn while the other was enclosed by a henge; these were linked by a double stone row that ended at an early Bronze Age cist cemetery (Bradley 2011).

Despite the uncertainty with dating some monuments, the evolution of certain landscapes in Scotland is now better understood, although the history and significance of others remain a topic of considerable debate. The chronology of many of these classes of monument is beginning to be clarified by new fieldwork and by targeted radiocarbon dating. The dating of Clava cairns, for example, has long been a matter of conjecture. Found only in north-east Scotland, these monuments comprise a central corbelled chamber entered via a long passage oriented southwest and covered with a cairn (e.g. Henshall 1963). At many sites, the cairn is revetted by kerbstones and surrounded by a stone circle. The kerbstones and stone circles are often carefully graded in height, with the tallest stones to the south-west. The similarity of Clava cairns to passage tombs has meant that they have often been thought to date to the Neolithic, but work by Richard Bradley has now shown that they were constructed in the late third millennium BC (Bradley 2000b).

Plan of the excavated features at the henge and timber circle at Broomend of Crichie (Bradley 2011, 35; Illus 1.35)

Stone circles take different forms in different areas and a number of distinctive regional traditions can be identified. Perhaps the best known of these are the recumbent stone circles of Aberdeenshire (e.g. Burl 1973; Welfare in press). These are distinguished by the presence of a large slab laid on its side on the southwest side of the circle and usually flanked by two particularly tall stones. Similar to Clava cairns, the stones were often graded in height, with the smallest stones situated directly opposite the recumbent, and these monuments too were constructed in the late third millennium (Bradley 2005b). Recent research suggests that they may have played a role in funerary practices. The primary feature at Tomnaverie, for example, was a cremation pyre which was subsequently enclosed by a ring cairn and finally surrounded by the stone circle (ibid.); however, others have argued that they were not solely designed as mortuary monuments and that the treatment of the dead was only one element in the range of ritual practices carried out at these sites. In other regions, stone circles can be ascribed rather different date ranges. Some are likely to have been constructed during the early second millennium (although others had already been built around 3000 BC), while the small, slightly oval stone ‘circles’ of Tayside – such as that at Croft Moraig (Bradley and Sheridan 2005) – were probably erected as late as the early first millennium. Likewise, some timber circles, such as that at Broomend of Crichie in Aberdeenshire (Bradley 2011), were built during the first half of the second millennium, but they have precursors constructed around 3000 BC.


‘Henges’ and ‘hengiforms’ – terms that are, perhaps, in need of replacing – seem to have a very wide date span, with some (e.g. Stones of Stenness on Orkney: Ritchie 1976; Ashmore 2000; Ashmore 2001) being built during the Late Neolithic, others (e.g. Broomend of Crichie: Bradley 2011) around 2000 BC and some (Pullyhour: ibid.) during the first half of the second millennium. Recent research on these and similar sites by Richard Bradley (ibid.) is improving our understanding of their dating. In several cases, as at Broomend of Crichie, they seem to have been built to define a sacred precinct around a pre-existing monument (in this case a stone circle). At this site, a number of Early Bronze Age cremation burials were deposited at the foot of the stones in the earlier stone circle and a deep grave was dug in the centre of the henge; these acts appear to have taken place around the same time that the henge itself was built (ibid.) and serve to remind that monuments such as henges were usually only one component of more extensive sacred landscapes. This work highlights the point that perhaps the most interesting result of recent research and excavation on these various classes of monument is that some traditions of monument construction (henges and stone circles, for example) appear to have been very long-lived. This calls into question any simplistic division we might be tempted to make between ‘group-oriented’ Neolithic societies and ‘individualistic’ Bronze Age societies.

Cup and ring marked standing stone at Nether Largie, Kilmartin valley ©RCAHMS

Stone rows also were also built in the Bronze Age. The pre-existing Neolithic stone circle at Calanais may well have had its avenues added during the late third or second millennium BC, and other avenues of stone and timber were erected during the Early Bronze Age, with the stone example at Broomend of Crichie linking a Beaker cist cemetery with the henge and a recumbent stone circle (Bradley 2011). Cremated bone sealed in the socket of one of the Ballymeanoch alignments in mid-Argyll has been dated to c 1200 BC (Sheridan 2005), while charcoal from a stone hole in the short stone row at Ardnacross, northern Mull, has produced a similar date (Martlew and Ruggles 1996). The more complex arrangement of standing stones at Nether Largie is also likely to date to this period. Both here and at Ballymeanoch, stretches of cup/cup and ring-marked outcrops were prised off and used in these monuments, arguably enhancing their ability to ‘catch’ divine power. The ‘four poster’ stone monuments which are mostly found in and around Tayside were in existence by 1600 BC, if not a few centuries before (Burl 1988). Single standing stones are far more difficult to date, but to judge from the interments of urned cremated remains found at the foot of at least one must have been in place by c. 1800 BC.

Astronomical alignments are important features of many of these classes of monuments (e.g. Burl 2000). From at least the late third to the late second millennium, there was an interest in marking key points in time by orientating monuments with respect to positions of the moon, sun and possibly other celestial bodies: this hints at the importance of solar and lunar cycles to Bronze Age cosmology. For example, recumbent stone circles are oriented so that the moon at its major standstill rises or sets over a significant point on the horizon framed by the recumbent stone and its flankers (Ruggles and Burl 1985). Cupmarks mark the recumbent and sometimes the flanking stones – and hence the position of the rising or setting moon – at sites such as Sunhoney (MacGregor 2002, 151). The presence of smashed quartz at many of these monuments suggests that this white, numinous material may have been associated with the moon (Bradley 2005b): quartz emits a greenish spark – triboluminescence – and a cordite-like smell when struck. The short stone rows of northern Mull were also oriented towards the southernmost rising and setting points of the moon at major standstill (Martlew and Ruggles 1996). At Glengorm, for example, the moon rises over Ben More, the highest mountain on Mull, at this point in the lunar cycle (ibid.); mountains such as this may have been seen as sacred places. Monuments also incorporated solar alignments. At Balnuaran of Clava, for instance, two clava cairns were aligned on the midwinter sunset (Bradley 2000b), a significant point in the calendar of any farming community, while at Croft Moraig in Perthshire, the entrance to the stone circle in its earliest phases frames the sunrise at the equinoxes (Bradley and Sheridan 2005).

Plan of the stone colours at the recumbent stone circle at Tomnaverie (from Bradley 2005b, 30)

Similar concerns are given architectural form in the choice and placement of particular types of stone within these monuments. At Balnuaran of Clava, red stones that absorb the light were most prevalent around the southwest of the cairns while grey and white stones that reflect the light were concentrated on the northeast (Bradley 2000b). In this way, a contrastive symbolism of life and death, light and darkness, became embedded into the monuments themselves, underlining the likely symbolism of their astronomical alignment. At recumbent stone circles, on the other hand, white/grey stones such as quartz may have symbolised the moon while red stones may have referenced fire (Bradley 2005b): there is archaeological evidence for major acts of burning at these sites, probably carried out as an element of funerary and/or purification rites. At Sunhoney, for example, all of the stones were of pink granite with the exception of the recumbent which is grey (MacGregor 2002, 149), while pieces of quartz were deposited on the old ground surface opposite the recumbent stone at Cothiemuir Wood (Bradley 2005b). At these sites – as at Clava cairns - a broad interest in concepts of fertility and regeneration (as symbolised by the cyclical movements of heavenly bodies) may be discerned that makes perfect sense in a mortuary context.

Ri Cruin end slab ©RCAHMS

It is, of course, difficult to reconstruct the content of the belief systems that underpinned the ceremonies carried out at Bronze Age monuments in any more detail: iconography dating to the Bronze Age is virtually absent in Scotland (the axes carved onto cist slabs at Ri Cruin and Nether Largie in the Kilmartin Valley are exceptions). However, comparisons with Continental material may prove instructive. Studies of Bronze Age iconography, particularly in northern Germany and Scandinavia, have identified a number of possible religious symbols including ships, waterbirds, chariots, horses, the sun and the wheel (e.g. Gelling and Davidson 1969; Kaul 2004). These were produced in a variety of media, of which Scandinavian rock art is perhaps the best known (e.g. Kristiansen 2010). This body of images, along with bronze ceremonial items such as the Trundholm chariot (Glob 1973, 99-103) and the Nebra disc (see the ScARF Case Study:The Nebra sky-disc), have facilitated the reconstruction of elements of Bronze Age cosmology. This seems to have been concerned with the movement of the sun, moon and other celestial bodies, the passing of the seasons and the cyclical regeneration of life. Bronze Age societies are thought to have regarded the sun as a deity that travelled by day in a chariot through the heavens and by night in a ship across the watery underworld (Kaul 2004). As such, boats and chariots were linked with cycles of death and rebirth and appear to have been considered a suitable means of transporting the dead to the afterlife.

The astronomical orientations discussed briefly above hint that similar concerns may have informed Scottish belief systems during this period. Likewise, the large number of votive deposits in wet contexts in Scotland suggests that water was a significant component of Bronze Age cosmographies in this region. The cup-and-ring motifs that dominate Scottish rock art have sometimes been interpreted as sun symbols (e.g. Beckensall 2009, 62), while quartz pebbles (which some have linked with the sun and others the moon: e.g. Darvill 2002; Bradley 2005b) are frequently included with Early Bronze Age burials: the grave goods that accompanied a double cremation at Beech Hill House, Perthshire, included a bone pommel, bone toggle and a quartz pebble (Stevenson 1995). The gold discs that accompanied the cremation burial at the Knowes of Trotty, Orkney (Sheridan et al. 2003), may make reference to the sun, and it is not hard to imagine how conceptual links may have been drawn between the sun, fire and the magical activity of metalworking (e.g. Budd and Taylor 1995); as such, either making or owning items of this sort may have been a source of social power (Sheridan and Shortland 2003). One of the inhumation burials at Barns Farm, Fife, had been deposited in a possible curricle or small, skin-covered boat (Watkins 1983), while elsewhere, for example at Forteviot on the River Earn (e.g. St Joseph 1976; 1978), concentrations of barrows in river valleys created a link between the dead and travel by water. Together, these examples suggest that although the belief systems of Bronze Age Scotland were doubtless regionally specific, significant elements were shared with other parts of northwest Europe, underlining the strength and importance of inter-regional interaction during the period.

Whatever the case, the rituals carried out at Scottish Bronze Age monuments would have facilitated the creation, reproduction and transformation of social identities. Those who played a key role in the rituals carried out at these sites, or who had preferential access to ancestors buried there, are likely to have been people of particular social standing: it was they, after all, who ensured the cyclical regeneration of the natural world. Indeed, the architecture of monuments such as stone circles or henges may have ensured that some members of the community were admitted while others were excluded. The construction of special-purpose monuments that may have been accessible only to certain people suggests that, on occasion, architecture was used as a technology of social differentiation.

Temple Wood, Kilmartin Valley, Copyright RCAHMS

Finally, it is important to note that earlier monuments were frequently re-used many centuries after they were originally constructed. Cremation burials dating to the Middle and Late Bronze Age were inserted at both recumbent stone circles and Clava cairns, for example at Tomnaverie (Bradley 2000b). At some of these sites, metalworking – itself perhaps seen as a magical, transformative activity – seems to have taken place during the Late Bronze Age. It may be that locations long associated with the ancestors and with ritual practices involving fire were considered especially suitable for the production of bronze objects: part of a sword mould was found at Loanhead of Daviot, for example (Kilbride-Jones 1936, 290). Chambered tombs of Neolithic date were also re-used as places of burial and for other depositional practices during the Bronze Age: at Isbister on South Ronaldsay, for example, the remains of a number of white-tailed sea eagles were deposited in the tomb during the Early Bronze Age, around a millennium after the tomb itself was built (Sheridan 2005). Such practices suggest that communities drew on the potent symbolism of ancestral places in the construction of social identity; doubtless, however, the significance of these places and the beliefs that underpinned the activities carried out at them changed over time.

The compact distributions of such distinctive monuments as recumbent stone circles and Clava passage graves, leave little doubt that various aspects of Bronze Age life and practice were expressed differently from place to place. This is particularly so of funerary and ceremonial monuments. Thus some of the ideas that seem to be inherent in recumbent stone circles in the North‐east, which dramatically emphasise the south‐western arc of the monument and extensively draw on the symbolism of quartz (see Welfare 2011), are not only inherent within the Clava‐type cairns around the inner Moray Firth, but are found across much wider areas of Scotland. The grading of the kerbstones towards the south‐west, for example, is not only a notable feature of many large cairns with well‐preserved kerbs, but equally a feature of the classic small kerb cairns, with their disproportionately large kerbstones and extensive use of quartz (Ritchie and Maclaren 1972); these latter are mainly found in Argyll and Perthshire, but the distribution is by no means exclusive. Indeed, any regional patterns to be observed in the character of the kerbs of cairns is much more diffuse and difficult to discern, though the apparent absence of such features from the south‐east of Scotland might be construed as some form of regional expression in its own right. Even here, however, the faint echo of the same beliefs is perhaps detectable in the Harehope Cairn, where with the primary cairn was enlarged eccentrically with a much thicker band of new cairn material on the south‐west (Jobey 1981). The significance of these regional distributions, however, are not understood, and it would be naïve to define a region on the distribution of one particular type of monument or artefact. The brash architecture of a recumbent stone circles becomes self‐appointing as a regional expression, and ignores the a host of other circles in the North‐east that present shades of the same design, but without the addition of the massive recumbent block. The presence of an internal cairn, for example, is also attested in other circles, as is the grading of ringstones, though in some cases the evidence suggests no more than an open ring of ungraded orthostats. In this sense, the North‐east forms part of a much larger region stretching southwards into Perthshire, an area equally notable for its concentration of small megalithic rings and settings. Many of these are also graded towards the south‐west, including the complex circle at Croft Moraig, while others, such as Moncrieffe House, reveal other patterns. This area and neighbouring Angus are also notable for their four‐poster stone settings, in which the corner stones are also typically graded in height

Key research questions include:

  1. Perhaps the most interesting result of recent research and excavation on these various classes of monument is that some traditions of monument construction (stone circles, for example) appear to have been very long-lived. Were there changes over time in the architecture, use and significance of classes of monument such as henges and stone circles?
  2. Monument complexes in certain regions have been well studied but there are considerable gaps in our knowledge of the chronology and development of others. Work such as that currently being undertaken by the Strathearn Environs and Royal Forteviot project (SERF) has much to offer here.
  3. When were single standing stones erected? Are they, like stone and timber circles and henges, a kind of monument that was erected at different times from the Neolithic onwards?
  4. Although dates have been obtained for stone rows in some regions, these remain poorly dated elsewhere, for example in Caithness and Sutherland where these are found in spatial association with Bronze Age monuments. Likewise, there may be regional differences in the chronology of other monument types and these require further study.
  5. When exactly, and why, were ‘four posters’ built?
  6. Although it is widely accepted that Orkney’s souterrains date primarily to the Iron Age, new dates call for a review of these monuments.
  7. The chronology of some classes of chambered tomb remains poorly understood. For example, it is often assumed that Bargrennan tombs in southwest Scotland are Neolithic, but the only dateable evidence from recent excavations at two of these sites comprises a series of Early Bronze Age cremation burials deposited around the edge of the monuments (Cummings and Fowler 2007). Clava cairns and Irish wedge tombs are now known to date to the Early Bronze Age and it is possible that other types of megalithic tomb may also have been constructed in this period. Shetland’s chambered tombs also remain poorly dated.
  8. How many of the monuments of the Chalcolithic and Bronze Age were designed with specific astronomical orientations?
  9. What does the regionally-specific character of Scottish Bronze Age monuments say about the construction of social identity during this period?
  10. To what extent were belief systems regionalised? Conversely, can sharing of beliefs be detected far beyond Scotland during this time and, if so, in what respects?
  11. The nature of religious beliefs and practices in parts of Scotland (e.g. Shetland) is very poorly understood. How can these be characterised?
  12. Was the re-use of older monuments particularly prevalent during certain periods of the Bronze Age and, if so, what does this mean?
  13. Although the sources of stone used to build certain classes of monument (e.g. recumbent stone circles) have been well-studied, this is not the case for other monument types and further work could be carried out on this topic: the capstone from the cist at Forteviot, for example, appears to have been taken from the bed of the nearby river (K. Brophy pers. comm.).
  14. The spatial relationships between Bronze Age monuments and contemporary settlements requires further study in many areas.
  15. What was the relationship between monumental and non-monumental elements of the landscapes? For example, rock art can be seen as a cultural elaboration of a natural formation while it has been suggested that some of the stones at recumbent stone circles were quarried from mountains considered to be sacred places (MacGregor 2002). What does this say about the interplay between culture and nature during the Bronze Age?
  16. If certain features of the natural landscape (e.g. mountains, rivers, the moon) were imbued with sacred power, how did that effect land-use and agricultural practice?

See also the ScARF Case Studies: Cladh Hallan / Ballachullish / Sculptor's Cave