Case Study 2: Assessing the Potential of Ploughsoil Scatters: Fieldwork at Oily Hall, Lode, Cambridgeshire
Lawrence Billington (Oxford Archaeology East)
Introduction
This case study sets out the results of fieldwork undertaken at a ploughzone lithic scatter site at Oily Hall, Lode, in the Cambridgeshire fens. Ploughzone scatters such as this represent the most abundant evidence of prehistoric activity across most of lowland Britain, but they are generally seen, with some justification, as having low interpretative potential (see Bond 2011).
The following suggests that significant information can be recovered from ploughzone sites, providing that appropriate methodologies and approaches are employed in their investigation and interpretation.
Background
The fieldwork was undertaken as part of doctoral research concerned with assessing the interpretative potential of ploughzone lithic scatters of Late Upper Palaeolithic (LUP) and Mesolithic date (Billington 2016).
A major part of this research entailed the collation of information on LUP and Mesolithic finds across a large part of Eastern England (Fig 2.1), which identified almost 1000 accurately located Mesolithic findspots, and a large number of poorly-provenanced assemblages. The vast majority, probably over 80 per cent, of these findspots derived from ploughzone contexts –– and in many parts of the study area these disturbed scatters and stray finds provide the only evidence of Mesolithic activity.
Whilst this record of ploughzone sites is valuable in allowing the regional and landscape-scale distribution of findspots to be examined, in most cases the interpretative value of their lithic assemblages was limited. This was partly attributable to the inherent limitation of ploughzone archaeology, including a lack of stratified or dateable contexts, high degrees of spatial disturbance and the frequent superimposition of numerous episodes of activity, creating complex palimpsests.
However, other issues surrounding the interpretation of these sites related to the circumstances under which they were sampled and recorded, particularly in terms of the small size of most assemblages and the suspicion that many are biased towards certain classes of artefact. It was in light of such problems that a programme of fieldwork and analysis was carried out to explicitly examine the effects of collection methods on the interpretation of ploughzone sites, and to inform future investigations of such sites.
Fieldwork at Oily Hall
Oily Hall is located in the south-eastern Cambridgeshire fens, in the lower Cam valley. The area was subject to extensive fieldwalking during the Fenland Project which identified a series of lithic scatters strung out along a spread of terrace gravels adjacent to the River Cam (Fig 2.2; Hall 1996). The flintwork from these scatters included a high proportion of Mesolithic material, alongside some later flintwork, but they produced few chronologically sensitive pieces such as microliths, and remain difficult to characterise in any detail.
In 2008 an opportunity arose to reinvestigate the area, when the National Trust acquired a parcel of land at Oily Hall, Lode. The Trust intended to take the land out of cultivation but, conscious of the archaeological significance of the area, first invited a local archaeological society, the Cambridge Archaeological Field Group, to fieldwalk the site.
Work began with extensive fieldwalking carried out along transects spaced at 10m intervals, which revealed a distinct scatter of flintwork, running along the terrace edge (Fig 2.2). This was followed up by gridded collection over a selected part of the scatter, with total collection of finds from 253 10m² grid squares.
Finally, in 2015, small-scale excavation was undertaken in one of the densest areas of the lithic scatter, with the excavation of sixteen 1m² test pits and dry-sieving (with a 5mm mesh) of all excavated deposits.
Results
The fieldwork provides a valuable opportunity to assess the results of each phase of work. The initial, transect, fieldwalking of the site recovered a small assemblage of 151 worked flints but was important in establishing the location and extent of the scatter in the north-western part of the area and, in the wider landscape context, demonstrating the continuation of what appears to be a semi-continuous swathe of lithic scatters along the fen-edge in this area (Fig 2.2).
This initial work also allowed the second phase of surface collection to be carefully targeted on an area with high densities of flintwork. A total of 1,285 worked flints were recovered from the 253 10m² collection units, ranging from 0 to 34 pieces per unit with a mean of 5 pieces per 10m² (Fig 2.3).
The assemblage is chronologically mixed, with diagnostic forms dating from the Mesolithic to the Early Bronze Age, but typological and technological analyses and differential recortication (‘patination’) indicate that about 70 per cent of the assemblage is of Mesolithic date.
The test pitting recovered an assemblage of 573 worked flints, 493 of which came from the ploughsoil, with the remainder coming from remnant buried soil horizons. From the ploughsoil alone, the 16 test pits produced a mean of around 30 worked flints each, individual counts ranging from 10 to 68.
Calculations comparing the average test-pit density to the highest density recovered from the 10m² surface collection units (34 pieces) suggest that around 1 per cent of the total population of ploughsoil artefacts was collected from the surface, a figure far lower than those cited from experimental ploughsoil sampling, which typically suggest that between 3 and 7 per cent of the total artefact population is collected (eg Clark and Schofield 1991; Boismier 1997; Ammerman 1985).
Variation in the proportion of surface collected material can be attributed to many factors, including ground conditions and fieldworker experience/aptitude but, in the case of Oily Hall, the compositions of the surface and excavated assemblages suggest that a major factor was the difficulty in recognising and collecting smaller artefacts during surface collection. Thus, whilst almost 40 per cent of the test-pit assemblage was made up of small pieces less than 20mm in size, only 11 per cent of the surface collected material fell into this category.
This clear bias towards the recovery of larger and more obvious pieces has implications for comparing assemblages derived from surface collection with those from excavations. This is of special importance for Mesolithic sites, where many of the chronologically sensitive and informative pieces, especially microliths, are diminutive, easily overlooked pieces.
The effects of this are well-illustrated by the data from Oily Hall (Figs 2.4 and 2.5). The assemblage derived from surface collection included a roughly equal (and small) number of ‘Early’ and ‘Later’ Mesolithic microliths, alongside a range of tools suggestive of a ‘balanced’, domestic-type assemblage (cf Mellars 1976; Myers 1987).
In contrast, the retouched tools from the test pits were overwhelmingly dominated by microliths, which were almost exclusively Later Mesolithic ‘narrow-blade forms’, more diminutive than their Early Mesolithic ‘broad-blade’ counterparts.
These differences have major implications for interpreting the Mesolithic activity at the site – not only does the excavated assemblage suggest that a far greater proportion of the material is likely to relate to activity during the later Mesolithic, but it also indicates that the tool assemblage is less balanced than the surface collection would suggest, with consequences for interpreting the character and duration of occupation at the site.
In the wider regional context, it is significant that later Mesolithic microliths are likely to be underrepresented in surface-collected material – a trend which may at least partly explain the smaller number of demonstrably Later, as opposed to Early, Mesolithic findspots across much of East Anglia (Jacobi 1996; Billington 2016).
Conclusions
Although the results of the fieldwork at Oily Hall indicate that caution must be exercised when interpreting assemblages derived from surface collection, it is important to emphasise that each of the stages of fieldwork summarised here played a complementary role in charactering the scatter at Oily Hall.
The initial fieldwalking effectively defined the extent of the main lithic scatter and allowed it to be very clearly related to its local topographic/landscape context, whilst gridded collection allowed intra-site distributions to be examined in detail and recovered a substantial assemblage which allowed the basic chronology and character of the scatter to be established.
Finally, the small-scale excavation of the scatter provided an unbiased assemblage which allowed the composition of surface collected material to be critically assessed and provided important, reliable, evidence for the chronology and character of activity at the site. More widely, the results provide some indication of what might be expected from more intensive investigation of ostensibly similar scatters in the wider landscape.
The work summarised here suggests that at least some of the interpretative problems presented by ploughsoil scatters can be addressed by deploying appropriate methodologies, and by recognising differences in the kinds of questions we can reasonably ask of assemblages recovered using different sampling methods.
The complementary nature of different methods of ploughsoil sampling highlights the potential for examining lithic scatters at varying spatial scales and intensities; integrating detailed, site-specific investigations into broader landscape and regional-scale studies. In this context, the inherent problems of ploughsoil scatters must be balanced against their virtues, especially in terms of their high archaeological visibility and the comparative ease with which they can be investigated and sampled, whilst in favourable circumstances the intensive excavation of ploughsoil scatters can be expected to yield important results no less significant than those from some ostensibly well-preserved sites.
Acknowledgements
The work described here was undertaken whilst in receipt of a doctoral studentship, funded by the Arts and Humanities Research Council and set up, supervised and resourced by Historic England and the University of Manchester.
The impetus for the work at Oily Hall came from Angus Wainwright of the National Trust and was carried out by dedicated members of the Cambridge Archaeological Field Group, to whom special thanks are extended, especially to Mike Coles and Terry Dymott, for allowing me to draw so extensively on the results of their work.
References
Ammerman, A J, 1985 ‘Plow-Zone experiments in Cantabria, Italy’. Journal of Field Archaeology 12, 33–40
Billington, L P, 2016 Lithic Scatters and Landscape Occupation in the Late Upper Palaeolithic and Mesolithic: a case study from eastern England. Unpublished PhD thesis, University of Manchester
Boismier, W, 1997 'Modelling the Effects of Tillage Processes on Artefact Distribution in the Ploughzone - a simulation study of tillage-induced pattern formation'. Oxford, BAR British Series 259
Bond, C J, 2011 ‘The value, meaning and protection of lithic scatters’. Lithics 32, 29–39
Clark, R H and Schofield, A J, 1991 ‘By experiment and calibration: an integrated approach to archaeology of the ploughsoil’, in Schofield, A J (ed) Interpreting Artefact Scatters: contributions to ploughzone archaeology. Oxford, Oxbow Books, 93–105
Hall, D, 1996 'The Fenland Project No. 10: Cambridgeshire Survey, Isle of Ely and Wisbech'. Cambridge, East Anglian Archaeology 79
Hall, D and Coles, J, 1994 'Fenland Survey: An Essay in Landscape and Persistence'. London, English Heritage
Jacobi, R M, 1996 ‘The Late Upper Palaeolithic and Mesolithic in Essex’ in Bedwin, O (ed) The Archaeology of Essex: Proceedings of the Writtle Conference. Chelmsford, Essex County Council, 10–14
Mellars, P A, 1976 ‘Settlement patterns and industrial variability in the British Mesolithic’ in Sieveking, G de G , Longworth, I H and Wilson, K E (eds) Problems in Social and Economic Archaeology. London, Duckworth, 375–99
Myers, A M, 1987 ‘All shot to pieces? Inter-assemblage variability, lithic analysis and Mesolithic assemblage “types”; some preliminary observations’, in Brown, A G and Edmonds, M R (eds) Lithic Analysis and Later British Prehistory: some problems and approaches. Oxford, BAR British Series 162, 137–53
