See important guidance on the use of this record.

If you have any comments or new information about this record, please email us.

Summary

Excavations on land off Tiverton Road revealed a large number of pits, some of which were probably tree throws and others part of complex intercutting pit groups. On the basis of the few artefacts recovered these pits are dated to the later Neolithic period.

Location

Protected Status: none recorded

Other References/Statuses: none recorded

Monument Type(s) and Dates

• PIT CLUSTER (Neolithic - 4000 BC to 2201 BC (Between))
• TREE THROW (Neolithic - 4000 BC to 2201 BC (Between))

Full description

Rohan, N. + Morris, B., 15/08/2014, Land off Tiverton Road, Cullompton (Report - Excavation). SDV359363.

Excavation undertaken in advance of a proposed housing development.

Scattered across the site were a large number of pits, some of which were probably tree-throws, and others formed part of complex intercutting pit groups. On the basis of the few artefacts recovered, and a small number of contexts with suitable charcoal, these pits date to the later Neolithic. Comparable intercutting pit groups are a noted feature of the Neolithic period.

The defining characteristic of this phase is the complex intercutting pit groups and associated isolated pits. Only a very small proportion of these features contained artefactual material, and even then it was often undiagnostic. In addition, very few features contained any appreciable amount of charcoal, making a wider programme of scientific dating difficult to realise. The one pit group that produced both diagnostic artefacts and charcoal was dated to the later Neolithic, and it is by analogy that the rest of these features are assigned to this phase.

The fills of these features were very clean, and were often partly or wholly composed of re-deposited natural subsoil. This made identification and excavation very difficult and it is highly likely only a proportion were actually sampled. The pale and leached nature of these fills may reflect their great age, but is more probably related to the highly permeable sandy subsoil found across most of the northern part of the site, which facilitated the free movement of ‘anthropogenic indicators’ (see Appendix 16) within the soil profile.

The character of the fills would suggest that each feature was excavated and backfilled within a relatively short period. That would, in turn, imply they were dug for a specific purpose. Structured deposition of artefacts and faunal remains within pits is an acknowledged characteristic of the Neolithic in Britain (Thomas 2012), yet the Tiverton Road examples were notable only for the lack of artefactual evidence. Here, over 120 individual pits were excavated and only three contained diagnostic artefactual material, and even then they all belonged to a single pit group. The general lack of diagnostic material naturally hinders dating and interpretation, but similar undated and irregular pits have been found on a number of sites in Devon with evidence of Neolithic activity – e.g. the Donkey Sanctuary in Sidmouth (Gillard & Quinnell 2012), or work in advance of the North Devon Link Road (unpublished) – so it may in fact be part of a wider regional tradition.

The function and contemporary significance of these intercutting pit groups is open to question. Recent discussion would suggest that, in general, pit group sites provide the main evidence for settlement or settlement-related activity in most areas of the country (Thomas 1999, 72-84; Anderson-Whymark & Thomas 2012). Typically, such pits would contain domestic material, charcoal, and/or artefacts; the fact that most of the Tiverton Road pits were almost wholly devoid of such material seems rather unusual. The number and density of the intercutting pits groups would indicate that this particular location was repeatedly visited and used over a significant period of time, pointing to a place of wider significance in the Prehistoric landscape. Yet the lack of surviving diagnostic artefacts would suggest that these pits were not dug with structured deposition in mind, or if it was, deposition was structured in a way we can readily recover.

At Kilverstone in Norfolk, Neolithic pit groups were shown in a number of instances to contain discrete artefact assemblages, interpreted in that instance to suggest intermittent occupation by one or more groups of people (Garrow et al. 2006). By analogy, the pit groups encountered at Tiverton Road may represent a similar phenomenon. Pit clusters are often associated with slight eminences overlooking low-lying ground (Thomas 2012), and this is certainly the case at Tiverton Road. More pertinent to this discussion is the local prominence of St Andrew’s Hill, which may well have served as a focus for Neolithic activity within a partly-cleared landscape.

Devon contains a number of large Neolithic sites (e.g. Hembury Hillfort, Liddell 1935), and a rather larger number of small isolated pits, sometimes found in small clusters (Sidmouth Donkey Sanctuary, Gillard & Quinnell 2012; Haldon Hill, Gent 1999). Until now, complex intercutting pit groups were not, however, a feature of the Neolithic in Devon, although they are sometimes encountered elsewhere (e.g. Kilverstone, Norfolk). The recently excavated site at Butts Road, Ottery St Mary, has produced a similar spread of pits – some conforming to an arc like Pit Group <2523> – with rather more in the way of material culture (Fiona Pink pers. comm.). Stratified pottery of this date in this area is itself rather rare, though excavations at Knowle Lane (AC Archaeology forthcoming), the Willand Road Medical Centre (Hood 2010) and Shortlands Lane (Morris 2014) have all produced Neolithic material, which would suggest it is the scale of intervention, rather than the intensity of Neolithic activity, that is the key in this instance, and this reflects a wider trend in Neolithic studies.

A large number of the isolated pits, and perhaps some of those within pit groups, conform to a highly distinctive morphology: curving or semi-circular in plan with an asymmetric profile. On this basis it is very likely these features were tree-throws. The mechanics of a tree-throw has been the subject of some discussion (Crombe 1993; Langohr 1999), but is essentially the sub-soil disturbance caused by the root plate of a tree as it falls. In most cases, the semi-circular ‘pit’ is the cavity created when the root plate is pulled out of the earth. The significance of these cavities is in the fact that pottery and lithic artefacts were occasionally deposited in these ‘natural’ pits, which might in turn be associated with more obviously anthropogenic pits or pit clusters. The symbolism of a structured deposition within a tree-throw may be related to place-marking within a largely wooded landscape, where other landmarks are concealed and line-of-sight obscured (see Evans et al. 1999 passim). However, in this context we should note that the concept of visual markers within a largely enclosed environment is slightly anachronistic and reflects the dominance of visual metaphors in our own society; woodland peoples may privilege the other senses when line-of-sight is limited (e.g. Gell 1995). Alternatively, as the material deposited in tree-throws is frequently domestic in character, it is argued it could be related to site clearance, when settlements were abandoned or re-occupied (Pollard 2001). This place-marking may simply be fortuitous, but equally could reflect the special character of certain trees, the evidence for which no longer exists. It is notable that the deposition of artefacts in tree-throws is usually seen as an Early Neolithic phenomenon, and pit clusters and intercutting groups are a later Neolithic practice (Thomas 2012, 7). If such a division holds true, then that would suggest the Tiverton Road site was a focus for activity throughout the Neolithic period. This is borne out by the fact that the only Early Neolithic pottery from the site was found in association with later Neolithic Grooved Ware, implying it was either curated or, perhaps more probably, found and redeposited.

However, we should exercise caution when invoking the role of the tree-throw in archaeological explanation. Recent articles have characterised several irregular shallow pits containing Neolithic material as tree-throws (Leverett & Quinnell 2010; Pearce et al. 2011, 31), when the evidence is itself equivocal. A caveat should, however, be borne in mind: reliably dated artefactual material was very scarce, and three of the tree-throws contained highly-abraded medieval pottery, and one cut a spread dated to the Romano-British period, so not all of the features identified need be Prehistoric in date.

SWArchaeology, 2014, SWARCH - Cullompton 2009, 2011, 5 (Article in Serial). SDV360659.

An unexpected find was the large number of pits, many of which date to the Neolithic period.

Ordnance Survey, 2016, MasterMap (Cartographic). SDV359352.

Sources / Further Reading

Associated Monuments: none recorded

Associated Finds

• FDV6084 - WASTE DEPOSIT (Prehistoric - 698000 BC to 42 AD)
• FDV6079 - FLINT (Neolithic - 4000 BC to 2201 BC)
• FDV6082 - SHERD (Neolithic - 4000 BC to 2201 BC)
• FDV6080 - SHERD (Medieval - 1066 AD to 1539 AD)

Associated Events

• EDV6902 - Evaluation, Land off Tiverton Road, Cullompton (Ref: 140815)

Search results generated by the HBSMR Gateway from exeGesIS SDM Ltd.