Subscribe now for full access and no adverts
The High Pasture Cave is the second-longest recorded cave system in Skye, with a length of more than 320m (1,000ft). Between AD 70 and AD 135 the main entrance was deliberately buried under a mound of stone and midden deposits in a series of acts designed to conceal the cave mouth and prevent access.
Abandoned and forgotten, the cave system saw little in the way of activity for the next 1,500 years until the mid-18th century, when the blocking stones were used to construct a group of cellular structures to create a small shieling settlement. This provided a seasonal base for people living in the nearby Kilbride township for tending their livestock during the summer months, as part of the island’s transhumance-farming economy. The Highland Clearances brought these activities to an abrupt end in the mid-19th century, but livestock grazing continued, and some of the other shafts leading into the cave system were backfilled with domestic waste and discarded agricultural machinery, partly as a means of preventing livestock from falling into the cavities.
The island’s curving 8km (5-mile) outcrop of Cambrian Durness limestone was recognised to be a likely cave site by geologists and caving clubs, who began looking for underground systems in the 1960s and 1970s, especially at the boundary between the water-soluble limestone and the harder granite and sandstone outcrops. Over time, more than 100 caves amounting to 3km (1.9 miles) worth of passages were mapped, but most of them are short – less than 100m (330ft) between the point that their small streams sink and re-emerge.
In 1972, members of the University College London Speleological Society, led by John Sanders, followed a stream bed to the point where it disappeared underground. By entering the sinkhole, they became the first people to enter the High Pasture Cave (Uamh An Ard Achadh) in nearly two millennia. The following year, a detailed survey of the cave was made and published by the Yorkshire-based Moldywarps Speleological Group, and the 320m-long (1,000ft) system rapidly became popular with cavers due to its relatively spacious passages and interesting calcite formations.
These early cave visitors recorded seeing ‘a large number of various bones, some of which had obviously been there a long time because they had been calcited over’ – that is to say, covered by the same calcium carbonate deposits that form cave stalactites and stalagmites. Trampling by cavers resulted in the calcite floor becoming eroded and broken, and their attempts to dig through the blocked southern entrance resulted in significant deposits of bone, charcoal, and fire-cracked stone being exposed.
By now it had become clear that damage was being caused to what could be a significant and complex set of archaeological deposits, and at this point various archaeological institutions became involved, including National Museums Scotland and what was then known as Historic Scotland. Archaeological excavations started in 2003, initially to survey and rescue disturbed material, which included worked bone and red-deer antler, ceramics of potential Iron Age date, stone tools, and a socketed iron-axe. Seasonal rescue excavations then took place between 2004 and 2010. The project co-directors, Steven Birch and Martin Wildgoose, were determined that this should be a community endeavour, involving volunteers who lived locally in Skye and from across Scotland, augmented by archaeology students from the UK, France, the Netherlands, Spain, Canada, and the USA.
Peculiar passages
Throughout the seven years of excavation, the major focus of investigation revolved around the part of the cave system that had been aptly named ‘Bone Passage’ by earlier cavers. Beyond its entrance, a lower passage, which fills to the roof with water during heavy rain, appears to have formed a natural barrier to activity during the prehistoric period, and there are no archaeological deposits past this feature.
Bone Passage is 9.2m (30.2ft) in length and is divided into two sections by ‘the narrows’, a constriction only 0.7m (2.3ft) wide, which separates the first section (a chamber 3.5m/11.5ft long, up to 2.4m/7.9ft wide, and 2.3m/7.5ft high) from the second chamber (2.7m/8.9ft long by 1.5m/4.9ft wide and 2.7m/8.9ft high). A small trial trench dug through the floor of the passage revealed a complex sequence of deposits more than 0.8m/2.6ft deep, with excellent preservation of organic and artefact remains.
Above ground, geophysical survey identified a horseshoe-shaped enclosure defined by a series of walls, centred around a natural amphitheatre and aligned with the southern end of Bone Passage. Removal of the turf revealed fire-cracked stones, hearths, and associated deposits to a depth of 3m (9.8ft) – comparable to the deep archaeological deposits more normally associated with long-lived urban settlements.
Clearing the chaotic jumble of stones and sediments, which had been used to backfill this enclosure, resulted in the discovery of a boulder-filled stairwell, including a series of narrow stone steps descending steeply towards the southern end of Bone Passage. This finding ended the 2005 season, and from this point speculation about the nature of this clearly very special site became ‘an intensely enjoyable topic of debate’ among everyone involved in its excavation.
It is true to say that this debate continues and has been further stimulated by the publication of a massive (more than 600 pages) comprehensive report (see ‘Further reading’ below). There is a big difference between describing what was found and understanding what it might tell us about prehistoric cave-use, the kinds of activity that the finds represent, and the role of natural places within prehistoric world views. As the authors of the report say in their final sentence: ‘this publication should be viewed as a first step towards a better understanding of activity in caves and underground spaces in later prehistory’.
Cavernous chronology
In subsequent seasons, it emerged that the enclosure created from the natural amphitheatre was a precinct and the focus of ritual activity, perhaps due to its location at the transitional area between daylight and the darkness of the cave. The enclosure incorporated walkways, walls, and a sequence of hearths, as well as access to the sequence of stairwell structures that lead into the cave. The stairwells were modified on numerous occasions during the cave’s active use, but each one was retained in situ, curated rather than being dismantled with each modification. Artefacts, as well as both human and animal bones, were found placed within the stairwell fabric, and it may also have once housed a wooden door or screen.
The earliest activity at the site (Phase 1) is evidenced by a cup-marked standing stone of Bronze Age date, a complex series of ard marks, pits, and stake-holes, as well as Beaker sherds that were found in the base deposits in Bone Passage, including one large group of ceramics deposited at the northern end. This activity could not be dated precisely, but probably belongs to sometime between 2140 BC at the earliest and 1510 BC at the latest.
The early Iron Age period – Phase 2 (starting c.800-745 BC and finishing c.730-640 BC) and Phase 3 (starting c.730-640 BC and finishing c.650-570 BC) – saw the construction of the first formal access to the cave and the development of the natural hollow around the entrance into a more formal ‘precinct’ with retaining walls, paved walkways, slab-built hearths, and three successive stone-built stairwells with access to Bone Passage. Artefacts and human bone, including skull fragments, were incorporated into these structures, while burnt and unburnt animal and fish bone, hazelnut shells, grain, fuel residue, and fire-cracked stones were deposited on the passage floor. Worked antler, bone points and needles, saddle-quern fragments, whetstones, and spindle whorls, along with broken shale bangles, are among the artefacts recovered from the hearth and midden deposits.
Phase 4 (starting c.630-515 BC and finishing c.540-435 BC) saw more intensive use of the precinct, with a new and more elaborate stairwell, side walls, and landings, new hearths constructed adjacent to the cave entrance, and the construction of a massive enclosure wall. The precinct was restructured with two activity zones on different levels: one designed to contain the cave entrance and the other emphasising or controlling access to this potent part of the site. Huge amounts of material were deposited within the precinct, including large quantities of artefacts, and a significant amount of fuel-rich material was also introduced into the passage.
Phase 5 (starting c.520-415 BC and finishing c.315-225 BC) is marked by subsidence within the enclosure. Deposits of clay and ash were laid over the resulting hollows to create a level surface, partly paved, sealing earlier hearths and fuel residues. The build-up of material in parts of the precinct was now such that the enclosure wall in the western arc had to be heightened and three new hearths were built, including one that was large and centrally placed. Modifications to the stairwell continued, and large amounts of ceramics, metal-working residue, and fire-cracked stones were found in features outside the precinct boundary.
At the beginning of Phase 6 (starting c.295-215 BC and finishing c.AD 50-130) the precinct was reduced in size to focus on the stairwell and cave entrance, with the stairwell acquiring a corbelled roof. A new paved floor was inserted in Bone Passage, sealing earlier deposits. The last phase of activity at the cave took place at the end of this period when a significant midden containing a large quantity of artefacts was deposited at the southern end of the passage (furthest from the entrance) on top of which was placed the semi-articulated remains of at least 23 pigs. This has been interpreted as a feasting deposit associated with the cave’s closure. A further artefact deposit was placed at the foot of the stairwell, which was then blocked using boulders and midden material. Midden deposits, incorporating a wide range of artefacts, were spread across the precinct’s surface.
In Phase 7 (AD 70-135), human and animal remains were placed into the top of the backfilled stairwell, and it is possible that this represents continued small-scale use of the site for anything up to 100 years, at the end of which the area around the entrance to the cave was covered in extensive deposits of midden material, fire-cracked stone, and boulders probably gathered from the immediate vicinity.
Subterranean significance
The central chapters of the High Pasture Cave monograph are devoted to impressively detailed reports on the material culture, the environmental assemblages, the stratigraphy, phasing, dating, and Bayesian analysis. All told, 1,700 artefacts were found, including tools of stone, bone, and iron used for hide-processing, textile-production, and the working of wood, stone, bone, and antler (more than 450 objects). There were also five spearheads and two daggers; 44 objects used for personal adornment (bangles, beads, and pins); fittings of bone, copper, and iron, consisting of handles and mounts; craft debris (slag, crucible, bone, antler, and copper off-cuts); and a significant number of querns, vessels, and stone objects associated with food-preparation and storage.
One artefact among many stands out for particular discussion: as widely reported in the media (including CA 267) at the time of its discovery, the remains of a lyre bridge led to recognition that music played some part in the rituals that took place at the site. Similar lyre bridges are known from post-Roman sites (including one from 10th-century York) but this is by far the earliest example found anywhere in Europe. The find stands as a powerful proxy for whatever music, poetry, and song was performed there, and is testimony to the fact that, as the finds report puts it, ‘during Phase 4, at least one person at High Pasture Cave was accustomed to exploiting sound in a conscious and sophisticated way’.
The lack of domestic structures in the vicinity of the cave raises questions about where the people who used the site were living, but they were probably local, and almost all of the artefacts were made from materials that could be sourced locally, with ‘the overall assemblage sit[ting] comfortably within the Atlantic Iron Age in terms of the number and range of artefacts’.
So, what do the report’s authors think was going on at High Pasture Cave in the Iron Age? The most prosaic answer is to point to the hearths, the ash and fire-cracked stones, and the objects in the artefact assemblage that relate to craft-working. There is a wider pattern across Britain and Ireland of craft being associated with ritual and magic, and of craft-workers occupying ‘liminal and seemingly impractical parts of the landscape’. Examples include the workshop associated with a subterranean structure at Mine Howe, Orkney (CA 199), the metalworking and bone-/antler-work site on the exposed promontory of Knowe of Skea, Westray (CA 199), and, closer to home, albeit slightly later in date, at the Fiskavaig rock shelter on Skye, where craft-production was accompanied by feasts involving a large number of cattle. There is a case here for further research to examine on a larger scale the range of sites associated with craft-production more generally, and to ask whether the nature of the manufacturing site adds potency to the objects being made.
But craft-working alone cannot explain the longevity of the site – the process of transforming rock into metal might have seemed like magic in the early days of metallurgy, but it must surely have become commonplace and ubiquitous as time went on. And adding to the complexity of the site are the modifications to the entrance and stairwell, the carefully placed caches of ceramics, lithics, and bone tools, and the incorporation of human remains into the structures and deposits.
At least seven, possibly 11, individuals are represented, including two to four adults, one or two children, one or two infants, two perinates (babies who died at or around the time of birth), and one foetus. The majority are represented by teeth or bone fragments; only one person, from an inhumation at the top of the backfilled stairwell, dating from Phase 7, was found as an articulated and complete skeleton. This is a very rare find, as funerary remains are few and far between across Iron Age Britain, where exposure, rather than formal burial, seems to have been the traditional means of disposing of the dead.
These remains are aptly described as ‘amongst the most complex and intriguing British prehistoric assemblages of the period’ and there are very few parallels, leading numerous contributors to offer various suggestions for the meaning of these ‘non-normative’ deposits. Was the adult female an incomer? Was she the mother of one or more of the infants whose remains were also found in the stairwell backfill? Was she a midwife or wet nurse? Was she a ‘witch’ or wise woman, malefactor or hostage, social outcast or the victim of human sacrifice, playing a role in some sort of exchange with the spiritual world?
Contemporary comparisons
The report’s final chapter seeks to answer questions around the site’s use, as well as understand the experiences of the people who came to this enigmatic place, by comparing High Pasture Cave with other subterranean sites from prehistory. In particular, the authors note that the 1st millennium BC was a time of architectural innovation in Scotland’s Atlantic Iron Age, with the construction in Skye and surrounding islands of complex circular, sub-circular, or oval stone-walled roundhouses with double walls and intramural galleries and staircases. This is the age, too, of the wheelhouse – a structure unique to the Western Isles and largely hidden below ground – and subterranean passages known as souterrains.
Renowned examples include the brochs of Dun Telve and Dun Troddan at Glenelg, Inverness-shire, Dun Carloway in Lewis, and Mousa and Gurness in the Northern Isles (see CA 274). Such carefully constructed dry-stone buildings are among the best-preserved Iron Age monuments in Europe, and they have similarities to High Pasture Cave in being very long-lived (in some cases, from 850 BC to AD 800), with multiple hearths, curation of earlier features, and incorporating human and animal remains in the floors and wall cavities. Other sites that closely resemble High Pasture Cave are the two so-called ‘wells’ at Gurness and Mine Howe, both in Orkney, with their elaborate dry-stone stairs leading to stone-lined cisterns. In particular, the underground chamber at Mine Howe sits at the hub of a metalworking workshop where the floors also incorporate human remains.
Fieldwork on Skye during the last 20 years has substantially increased the number of known souterrains, with a distribution that relates to the most fertile land for growing grain. The excavation of one souterrain at South Skye Medical Practice in Broadford, recovered evidence for grain-processing, -drying, and -storage from 374 BC to AD 25. Storage of agricultural produce seems to be the main use, but souterrains have also been interpreted as places of refuge and as water-storage cisterns, all of which may well have had a ritual element. Indeed, it is a commonplace of archaeological interpretation (based on ethnographic research) that people in the past did not perceive ritual and everyday domestic activity as separate domains – prosaic tasks are often permeated with ritual significance and, even in recent times, it was the practice in the crofting towns of the Western Isles to drive cattle between two bonfires at the start of the transhumance journey to the summer shieling grounds.
There is much still to be learned about the use of natural and ‘man-made’ caves, but there is no doubting that the combination of firelight from the hearths, the music and song, and the passage into another world with its strange calcite formations would have had an impact on everyone who took part, combining theatre, mystery, and magic to powerful effect.
Further reading:
Steven Birch and Jo McKenzie (2025) High Pasture Cave: ritual, memory, and identity in the Iron Age of Skye (Oxbow Books, ISBN 978-1785709500, £44).
All images: The High Pasture Cave and Environs Project
You must be logged in to post a comment.