The Archaeological Potential of Secondary Contexts

Module 5 - Introduction

The goals of the module were the assessment of the relative homogeneity and/or heterogeneity in space and time of Palaeolithic stone tool assemblages occurring in secondary context aggregate deposits, with particular reference to:

1. Clast entrainment, transportation and deposition
Extant research in fluvial engineering and physical geography has indicated the highly variable nature of clast transport at the micro-scale, reflecting localised stream bed conditions. However, the research also indicates some robust trends of considerable importance to the interpretation of transported, secondary context assemblages – principally, the absence of any clear relationship between clast size and transport distances, and the tendency for short step lengths and long burial phases. The clast size/transport distance relationship data stresses the stochastic nature of clast transport and highlights the importance of interpreting derived assemblages on an artefact by artefact basis. The short step length/long burial phase data promotes caution in the interpretation of abrasion data as an index of transport distance and an indicator of catchment source areas. Consequently, as with the interpretation of the Broom and Dunbridge data (Module 4), focus is placed on highly robust patterns, rather than on high-resolution trends. Finally, the identification of these trends has also highlighted the importance of further experimental research with respect to the duration of burial phases and the potential for abrasion development during periods of burial and partial burial.

2. Assessing stone tool assemblage data
With respect to the spatial homogeneity/heterogeneity of derived stone tool assemblages, artefacts’ physical condition data is obviously of prime importance. However, the work of Chambers (in prep.; Module 4) has indicated the importance of the état physique approach, emphasising zonal arête abrasion, edge damage micro-flaking, incipient cones of percussion, and the role of artefact morphology (e.g. cross-section profiles) in transportation. In an ideal world, field data from the associated sedimentary units (e.g. grain size, bed-forms, clast and artefact fabric data) would be informative with respect to reconstructing transport histories, fluvial regimes, and therefore the spatial origins of the material. Unfortunately, given the fragmentary nature of fluvial sedimentary sequences and the highly variable nature of entrainment and transportation (see above), such an approach would involve an unacceptable level of generalisation. Moreover, for nearly all extant assemblages, sedimentary field data of the type referred to above was not recorded. Therefore, modelling the spatial component of secondary context assemblages must focus upon the état physique of individual artefacts.

In cases where stratigraphic provenancing data are available, this information is important for an initial, crude assessment of the temporal homogeneity/heterogeneity of derived stone tool assemblages (e.g. whether the artefacts were deposited in a single ‘horizon’ or throughout a sedimentary sequence). However, to make a detailed assessment requires a high-resolution analysis of the preserved sedimentary sequence. This includes the geochronological framework (e.g. the duration of depositional events and sedimentary hiatuses – Module 2), and micro and macro-changes in the sedimentary sequence (represented by grain size distributions, sediment types, and bedforms). On a wider scale, understanding of ‘site’ formation processes are vital, with respect to the coarse-resolution chronologies of the initial entrainment of the artefacts (e.g. through floodplain ‘site’ erosion over historical rather than geological time-spans), and the depositional environment(s). Unfortunately, these data are often unavailable for extant assemblages, although the methodology is applicable to well-documented assemblages such as Swanscombe and Broom.

With respect to morphological data, the evidence from fluvial engineering research indicates that artefact size and dimensions cannot be used as an indicator of transport distances. However, this section has indicated the value of other elements of artefact morphology in the detailed assessment of transport history (e.g. the uses of cross-section profiles in Chambers’ (in prep.) transport modelling methodology).

3. Integrating laboratory, field and desktop research
Correlation of Chamber’s (in prep; Module 4) experimental flume research with the experimental fieldwork of Hosfield & Chambers (2002; this Module) has highlighted some robust patterns, most notably the rates and patterning associated with the development of arête abrasion on bifacial artefacts. The experimental research has also drawn clear links with the extant fluvial engineering research literature, through the stressing of local variations in field conditions (e.g. bed morphology) which have impacted upon individual clast transport behaviour.

However, it has been apparent from the field experiments that non-laboratory results will inevitably be more variable, reflecting the wider range of variable conditions, and in some cases unpredictable (e.g. the apparent role of algae in retarding abrasion and edge damage (micro-flaking) development). In contrast, the laboratory has provided tighter controls on experimental conditions and yielded data that is currently unknowable for field experiments (e.g. the mechanism of transport). The integration of the laboratory and field experimental research has therefore highlighted the need for further research, improved tracer recovery (magnetic tracers), improved data logging techniques (radio tracers) and the greater integration of archaeological and fluvial geomorphological research.

4. Taphonomy and secondary context assemblages
The demonstration of artefact transport inevitably raises the problem of explaining the formation of large artefact assemblages within secondary contexts. It is clear that artefacts recovered from a single site (e.g. Broom) have been transported a wide range of distances (based on the état physique of extant artefacts), while experimental fieldwork has demonstrated that materials from a single source will be dispersed over an increasingly wide area over time. The problem is therefore clear: derived artefacts originate from a wide range of sources, yet are ultimately deposited in a single sedimentary location. Why does this happen?

We suggest that fluvial geomorphological processes are the critical factor. Although hominid involvement is (inevitably) a required starting condition (e.g. the bifaces must be discarded in a valley for large concentrations to be formed downstream), it is fluvial processes and landforms that produce localised sedimentary ‘traps’ within which fluvial material and entrained artefacts are deposited, after a wide range of transport histories. The significant geomorphological processes are argued to include river confluences (resulting in localised loss of stream competence and therefore extensive depositional activity), the impact of bedrock types and valley forms upon fluvial behaviour (sediment re-working and river incision), and local depositional environments (e.g. braidplains and large-scale barforms).

In conclusion, it has been demonstrated that the majority of secondary context assemblages are the product of artefact transportation and deposition. With respect to the homogeneity and/or heterogeneity of these assemblages, investigations must consider the nature of clast transport within fluvial systems, the état physique of individual artefacts, and the taphonomic processes responsible for the deposition of sediments and artefacts in specific locations within fluvial landscapes. Through these approaches it is possible to begin to understand the nature of artefact discard and hominid behaviour within the contemporary fluvial landscapes of the Pleistocene.

Module 5 Outline

Module 5 Results

Module 5 Interm Report

References

Return to top