Pleistocene Archaeological Cave Deposits

Rob Dinnis

Executive Summary

This case study is not an account of a specific site investigation; instead it outlines the likely context of cave deposits and identifies appropriate methods of investigation.

Caves have increased potential for the accumulation and preservation of archaeological material relative to open-air sites. They are therefore a valuable archaeological resource. Most Palaeolithic cave deposits date to the Late Pleistocene, and therefore document Late Middle Palaeolithic and Upper Palaeolithic occupations.

Caves provide the primary evidence for this period. Although development is unlikely to impact substantially on Palaeolithic caves, it should be recognised that surface fissures, smaller voids, the outermost parts of larger caves etc. also have the potential to contain archaeological Pleistocene deposits. Several sources of information can inform desk-based assessment of the potential of deposits, including literature produced or compiled by cavers.

Given the rarity of extant Palaeolithic deposits the identification of any new site, or unexcavated area of a known site, is of considerable importance. Invasive investigation should include methods that can account for deficiencies in extant archaeological archives, notably documenting localised geological complexities and the systematic collection of Upper Palaeolithic microliths. Consideration should also be given to the common presence of late prehistoric archaeological material and human remains.


Where: n/a
Region: South-West; West Midlands & the North
Palaeolithic period(s): late Middle Palaeolithic and Upper Palaeolithic
Type of investigation: n/a
Methods: n/a
Type(s) of deposit: Cave deposits
Features of interest: Potential for surface fissures and outermost portions of larger caves to retain Pleistocene deposits and be threatened by development

Development Context

Most known Palaeolithic caves are located in areas unlikely to see major development in the foreseeable future. Even in built-up areas around/above important caves, such as at Kent’s Cavern (Torquay, Devon), any development would encroach on peripheral parts of the cave system rather than the main cave voids, where archaeological material tends to be most abundant. Surface fissures, smaller voids and the outermost parts of larger caves are therefore more likely to be affected. However, recent work has confirmed that these areas too can retain valuable Pleistocene deposits that are more generally absent from the surrounding landscape.

To give two examples: a deep paleontological sequence spanning most of the Late Pleistocene was recently discovered in a formerly sealed entrance to Kent’s Cavern (Dinnis and Proctor 2015; Figure 1); and recent excavation on top of Chudleigh Rock has shown the survival in surface fissures of Pleistocene deposits comparable to those in the important Upper Palaeolithic Pixies Hole beneath, despite these deposits being absent elsewhere on the rock surface (C. Proctor pers. comm.).

Archaeological Context

Caves and rockshelters are well understood to have increased potential for the accumulation and preservation of archaeological material. Larger caves provide an obvious source of shelter, a useable workspace and in many cases also a viewpoint from which to observe the surrounding landscape.

In many cultures past and present they are also known to have had symbolic importance. Once archaeological material has accumulated, caves can serve to protect deposits and particularly fine sediment from external erosional agents.

Furthermore, caves’ stable environmental parameters, including reduced biological activity and relatively constant humidity and temperature, are conducive to the long-term preservation of bone and other organic material. Such material is usually missing from later Middle Palaeolithic and Upper Palaeolithic open-air sites. While the total volume of Upper Pleistocene deposits in caves is relatively small, the likelihood that it contains well-preserved archaeological and palaeontological material is therefore disproportionately high.

In Britain the invasive investigation of Pleistocene deposits in caves, rockshelters and fissures has a long history affecting many localities. The known extant resource is therefore much depleted. Invasive work in recent decades has primarily been research-driven, mostly to contextualise historic collections (e.g. Campbell 1977; Pettitt et al. 2009; Dinnis et al. 2019), but also includes the enhancement of conservation measures (Walker et al. 2014; Chamberlain and Wilmut 2017; Proctor et al. 2017a).

In England, Palaeolithic caves are clustered in limestone outcrops in the south-west, north and north midlands, with further clusters in north and south Wales. Most Pleistocene bone-bearing cave deposits accumulated during the Late Pleistocene.

Material from several caves together forms the period’s biostratigraphic framework (Currant and Jacobi 2011). Bone-bearing Middle Pleistocene deposits are in contrast relatively rare, and only a handful of caves have yielded archaeological material of this age. The Palaeolithic record from caves therefore chiefly reflects several late Middle Palaeolithic (LMP) and Upper Palaeolithic (UP) occupations.

Overall the British record for this period is sparse, with the majority of evidence, and undoubtedly the best quality evidence, coming from caves.

Assessing the likely presence of Pleistocene deposits

Several sources are available to inform desk-based assessment of the potential of cave deposits, but these are somewhat piecemeal and regionally uneven:

  • Scientific surveys, sometimes national but usually regional in scope (e.g. Barton and Collcutt 1986; Holderness et al. 2006).
  • Online databases, including the Cave Archaeology and Palaeontology Research Archive (CAPRA) (Chamberlain 2015), radiocarbon databases (e.g. ADS, ORAU) and the British Caving Association’s Cave Registry. Personal websites compiled by cavers can also prove very useful, but sometimes information requires a level of validation.
  • The British Cave Research Association cave studies library (Matlock, Derbyshire), which contains scientific publications as well as Caving Club publications and other documents.
  • Museum archives. Historic archaeological archives for the earlier excavated Palaeolithic caves tend to be poor, but with some exceptions. For example, although incomplete, Torquay Museum houses a reasonable documentary and artefact archive for the late 19th and early 20th century excavations at Kent’s Cavern and other Devon caves. For assessment of any location adjacent to or within known Palaeolithic caves, the Jacobi archive at the British Museum is also a useful resource.

Any assessment of a cave’s archaeological potential should first consider whether the site is a relict (=fossil) or active cave (i.e. with ongoing water action). Although Pleistocene deposits can survive in some areas of active caves, they are overall much less likely to retain deposits of this age. Secondly, caves near the site of interest should be considered, as caves in a particular area will often have comparable sedimentary fills. The archaeological and palaeontological record of many sites is limited to finds made during exposure of surface deposits by cavers; therefore for any given site a record limited to Holocene-age finds may not necessarily indicate the absence of pre-Holocene deposits. Consideration of a wider area can overcome this problem.

Some attempts have been made to formulate criteria that can be used in assessment of a cave’s likely archaeological value (e.g. Holderness et al. 2006; Dinnis et al. 2010). Several of the determining factors relate to local topography and would therefore necessitate on-site inspection.

Invasive work: some considerations and best practice

Extant Palaeolithic cave deposits are an acutely limited resource, and therefore identification of any new site or area containing archaeological material would be of considerable importance. Any invasive work should be undertaken with this in mind and should employ suitably rigorous excavation and sampling methods. If Palaeolithic material is identified it is especially important for methods to account for deficiencies in extant archaeological archives.

Current research on the late Middle Palaeolithic and Upper Palaeolithic addresses much higher resolution questions than that on earlier periods. In large part this is due to the applicability of radiocarbon dating, which offers a much greater precision than radiometric methods used for earlier periods, i.e. on a scale of hundreds or thousands of years, rather than tens of thousands. This increased precision allows examination of fine-scale inter- and intra-regional cultural processes. However, in conflict with this, most evidence for the period comes from caves, whose geology can be particularly complex, and therefore the precise relationships between archaeological objects can be difficult to determine.

Geological complexities are often extremely localised, and thus the geological sequence in one area may not fairly represent the situation in an immediately adjacent area. Because historic cave excavations inadequately recorded these complexities, key questions simply cannot be addressed from old collections.

A recent example is disagreement over the age of Britain’s earliest modern human fossil, ‘KC4’, from Kent’s Cavern (Higham et al. 2011; Jacobi and Higham 2011; White and Pettitt 2012; Proctor et al. 2017b): several lines of evidence point to profound but extremely localised stratigraphic anomalies close to KC4’s finds position, but the excavation archives are insufficient to understand what this means for the fossil’s age.

Should Palaeolithic material be found, excavation methods should therefore ensure adequate capture of geological data. This should include enough observations to reconstruct clast alignment post-excavation, in order to identify common issues such as localised collapses and sediment piping.

Excavation in 2015-16 of the intact Late Pleistocene sequence from Kent’s Cavern included Total Station recording of the footprint of all clasts over 10cm. It is unrealistic to expect this level of data collection at other sites, but faster and yet still adequate recording of similar data can be undertaken using inexpensive angle measurers and/or photogrammetry.

Extant cave collections are also notably selective, largely the result of inadequate excavation methods. Several periods of the earlier and later Upper Palaeolithic are now known to be characterised by microlithic retouched ‘tools’, sometimes ≤10mm in their largest dimension when complete (Anderson et al. 2016; Demidenko et al. 2018) but usually found in fragmentary form. Excavation of Upper Palaeolithic material should therefore not only account for the potential presence of microfauna, but also flint artefacts of ≤5mm.

Finally, it is worth noting that investigation at or adjacent to cave sites should anticipate late prehistoric activity and particularly human remains. The abundance of Neolithic and Bronze Age remains is now well documented (e.g. Chamberlain and Williams 2001), and newly dated material continues to expand the corpus. Based on their positions, these remains were deposited not only within caves’ recesses but also in small fissures close to or outside cave entrances. Invasive work should therefore anticipate this.


Anderson, L., Chesnaux, L., Rué, M., Picavet, R., Fernandes, P., Morala, A., Caux, S., Tallet, P., Caverne, J.-B. and Kawalek, E. 2016. 'Regards croisés sur la station aurignacienne de Brignol (Villeneuve-sur-Lot, Lot-et-Garonne, France): approches taphonomique, pétroarchéologique, technoéconomique et technofonctionnelle de l’industrie lithique'. Paléo 27: 11–42.

Barton, N. and Collcutt, S. 1986. 'A Survey of English and Welsh Palaeolithic Cave Sites'. Unpublished report for English Heritage and Cadw.

Campbell, J B, 1977. 'The Upper Palaeolithic of Britain: A Study of Man and Nature in the Late Ice Age (2 volumes)'. Clarendon Press: Oxford.

Chamberlain, A. 2015. 'Cave Archaeology and Palaeontology Research Archive'. Archived at:

Chamberlain, A.T. and Williams, J.P. 2001. A Gazetteer of English Caves, Fissures and Rock Shelters Containing Human Remains. CAPRA 1.

Chamberlain, A. and Wilmut, J. 2017. 'Theft, Vandalism and Restoration at Joint Mitnor Cave, Devon'. Paper given at the 28th Cave Science Symposium of the British Cave Research Association, University of Leeds, 21 October 2017.

Currant, A.P. and Jacobi, R. 2011. 'The Mammal Faunas of the British Late Pleistocene'. In N.M. Ashton, S.G. Lewis and C.B. Stringer (eds.) The Ancient Human Occupation of Britain: 165–180. Elsevier: Amsterdam.

Demidenko, Y.E., Škrdla, P. and Rios-Garaizar, J. 2018. 'European perspectives of the East European LGM Epi-Aurignacian with Sagaidak-Muralovka-type microliths'. In P. Valde-Nowak, K. Sobczyk, M. Nowak and J. Źrałka (eds.) Multas per gentes et multa per saecula: 85–92. Alter Publishing: Kraków.

Dinnis, R., Davies, J. and Chamberlain, A. 2010. 'Non-invasive assessment of the archaeological potential of cave deposits: the example of Bishopston Valley Caves, Gower, South Wales'. Cave and Karst Science 37(2): 45–48.

Dinnis, R. and Proctor, C. 2015. '(Re)discovery of a major cave entrance at Kents Cavern, Devon, UK'. Cave and Karst Science 42(2): 58–59.

Dinnis, R., Bates, M.R., Bello, S.M., Buck, L.T., Davies, J.S., Preece, R.C., Walker, E.A., Boulton, J., Flas, D., Harris, S.-J., Mogg, J. and Schouten, R. 2019. 'Archaeological collections from Long Hole (Gower, Swansea, UK) and their place in the British Palaeolithic'. Cave and Karst Science 46(1): 37–46.

Higham, T., Compton, T., Stringer, C., Jacobi, R., Shapiro, B., Trinkaus, E., Chandler, B., Gröning, F., Collins, C., Hillson, S., O’Higgins, P., FitzGerald, C. and Fagan, M. 2011. 'The earliest evidence for anatomically modern humans in northwestern Europe'. Nature 479: 521–524.

Holderness, H., Davies, G., Chamberlain, A.T., and Donahue, R. 2006. Research report – A Conservation Audit of Archaeological Cave Resources in the Peak District and Yorkshire Dales. CAPRA 7.

Jacobi, R. and Higham, T. 2011. 'The British Earlier Upper Palaeolithic: settlement and chronology'. In N.M. Ashton, S.G. Lewis and C.B. Stringer (eds.) The Ancient Human Occupation of Britain: 191–222. Elsevier: Amsterdam.

Pettitt, P.B., Jacobi, R.M., Chamberlain, A.T., Pike, A.W.G., Schreve, D., Wall, I., Dinnis, R. and Wragg Sykes, R. 2009. 'Excavations outside Church Hole, Creswell Crags: the first three seasons (2006–8)'. Transactions of the Thoroton Society of Nottinghamshire 113: 35–53.

Proctor, C., Dinnis, R., Stewart, J., Knul, M., Gleed-Owen, C., Hunt, C.O., Challinor, D., Richards, D., Fairbank, V., Boulton, J., Riley, H., and Whitton, C. 2017a. 'Cow Cave Heritage at Risk S17 Project at Cow Cave, Chudleigh Rocks, Teignbridge, Devon (Scheduled Monument 1010726)'. Report prepared for English Heritage, March 2017.

Proctor, C., Douka, K., Proctor, J.W. and Higham, T. 2017b. 'The age and context of the KC4 Maxilla, Kent’s Cavern, UK'. European Journal of Archaeology 20: 74–97.

Walker, E.A., Case, D., Ingrem, C., Jones, J.R. and Mourne, R. 2014. 'Excavations at Cathole Cave, Gower, Swansea'. Proceedings of the University of Bristol Spelaeological Society 26: 131–169.

White, M. and Pettitt, P. 2012. 'Ancient digs and modern myths: the age and context of the Kent’s Cavern 4 maxilla and the earliest Homo sapiens specimens in Europe'. European Journal of Archaeology 15: 1–30.