4.2.1 Cave and Rock Shelter Sites

Reconstruction of the Mesolithic settlement at Sand, Applecross in Wester Ross, by Phil Austin.

As shelters, depots, landmarks, and tombs, caves have provided a focal point for human ritual and subsistence activities over a significant time period. Caves also preserve a record of past faunas and environments, as well as documenting , as natural geological structures, the geomorphological history of the wider landscapes and littoral zones. Due to their relatively stable microclimates and natural protection from the more usual external taphonomic threats which impact on archaeological deposits on the surface, cave sites generally provide enhanced preservation including organic materials. As such, archaeological evidence from caves allows a glimpse of past societies' cultural understanding of natural places in the landscape. The inaccessibility of some caves also suggests that specialised knowledge and equipment would have been essential prerequisites for their utilisation, rather than simple opportunism.

While Scotland has a record of cave archaeology, most of this work was undertaken during the late 19th and early 20th centuries, at a time when archaeological techniques and fieldwork publication were not comparable with those today. There has also been a profound coastal bias, e.g. the cave and rock shelters around Oban (Anderson 1895; 1898; Turner 1895; Bonsall and Robinson 1992; Bonsall et al.1993; Bonsall and Sutherland 1992; Bonsall 1996; Saville and Hallen 1994). The coastal bias is replicated in many wider landscape survey projects investigating the earliest settlement of Scotland. The geomorphology of natural caves and rock shelters has also received limited research, a major gap in knowledge that will have to be addressed if sites where potentially Mesolithic and more specifically Palaeolithic archaeological deposits exist, are to be targeted. It is widely accepted that the major glacial events of the Devensian and the Loch Lomond Readvance are responsible for the destruction of sites on the surface relating to the earliest settlement of Scotland. Therefore, the potential preservation of Palaeolithic and Mesolithic deposits in caves is significant, including sites submerged through the effects of deglaciation and changing sea-levels. 

One promising non-coastal cave complex that has witnessed much attention is at Creag nan Uamh, Assynt, Highland, originally interpreted as a likely site of Upper Palaeolithic activity on the basis of a rich faunal assemblage including reindeer (Peach and Horne 1917; Callander et al. 1927; Cree 1927), the deposits have since been reinterpreted (Murray et al. 1993; Birch and Young 2009; Saville 2005; Kitchener 1998; Kitchener and Bonsall 1997). The Pleistocene and early Holocene faunal assemblages from the Creag nan Uamh caves are of significant importance to Quaternary studies in Scotland, while subsequent analysis and documentary research (Lawson 1981), artefactual analysis and radiocarbon dating of human remains (Saville 2005) have proved that the recovered archaeological material relates to post-Mesolithic activity.

Alt nan Uamh Cave (Creag nan Uamh), Assynt © RCAHMS 

Recent archaeological fieldwork projects investigating cave sites have included Ulva Cave, Mull (Bonsall et al. 1989; Bonsall et al.1994), the caves and rock shelters of Argyll (Smith 2001), and the Scotland's First Settlers Project (Hardy and Wickham-Jones 2009) that investigated the seascapes of the Inner Sound between Skye and the adjacent west coast Mainland. Results of this work indicate the widespread human use of caves and rock shelters during the Mesolithic and through subsequent periods of time. The prolonged use of sites evidenced in these projects is generally typical of cave occupation in Scotland (Leitch 1987; Martin 1984: 122-7) and elsewhere in Britain (Branigan and Dearne 1992) and Ireland (Drew 2006). Little new fieldwork is currently being undertaken in cave sites in Scotland today and as a result there are significant gaps in knowledge of cave use from the Palaeolithic and earlier Mesolithic periods. It is here that re-evaluation of existing material, offers most potential. One example of this is the reassessment of the lithic assemblage from Kilmelfort Cave, Argyll (Coles 1983) where a late Upper Palaeolithic date is now proposed on typological grounds (Saville 2004a, 2010; Saville and Ballin 2009).

As repositories of archaeological deposits and information through time it is important that the contents of caves are investigated in an all-embracing way including the varied inputs of material into the site, the study and analysis of all materials and their relationships (whether deposited by humans or other agencies). Only in such a way can researchers expect to understand the complex taphonomic variables that exist in such sites. Caves also act as good natural sediment traps and are significant for their potential to preserve deposits pre-dating the last glacial maximum, as evidenced in the recent work in caverns at Uamh na Claonaite, Assynt, Highland (Birch and Young 2009).

Studies that quantify and evaluate the value and conservation status of archaeological cave deposits (e.g. Holderness et al.2007; Chamberlain and Williams 2000; Drew 2006) and promote more integrated research networks (see http://www.uplandcavesnetwork.org/). Working with caving groups to help identify archaeological deposits and avoid disturbance can yield worthwhile results, as at High Pasture Cave and the Grampian Speleological Group is now setting a good example of this with its recent work in the Assynt caves.

The extensive investigation of cave sites elsewhere in Britain and Ireland has revealed a wealth of information relating to their use. The discovery of 'Cheddar Man' in Gough's Cave (Jacobi 1985; Stringer 1985; Cook 1986) and the only early postglacial cemetery at Aveline's Hole in the Mendip Hills (Garrod 1926; Schulting 2005), have shown the potential of caves to inform knowledge of funerary practices. Rock art has also been more recently identified at Aveline's Hole and at Creswell (Pettitt et al. 2006). Using sound heuristic methodological approaches it may be possible to produce new and important information from cave sites pertaining to the earliest settlement of Scotland although the time-consuming and long-term commitment required for new field work campaigns (and the potential for negative returns) also needs to be acknowledged.