The Potential for Heavy Metal Soil Analysis on Low Status Archaeological Sites at Shapwick, Somerset

Article excerpt


The problems of identifying low-status agricultural settlements without pottery in the period 400 to 900 AD are highlighted here in the background to the research at Shapwick in Somerset. A solution to these problems based on the use of habitative field names, field walking, geophysical survey and soil analysis is a practical possibility; although not all this evidence may be available.

An interest in the analysis of soils for evidence of human occupation or activity has arisen in the general search for a methodology to identify sites with otherwise few archaelogical remains and to elucidate a now all but vanished landscape around the village of Shapwick in Somerset. One of the prime aims of the Shapwick Project is to investigate the development of the settlement pattern in the parish. The proposed hypothesis is the replacement of a more dispersed settlement pattern of scattered farmsteads and hamlets by a planned village, between the 8th and 12th centuries AD, with the 10th century being (at the moment) the favoured period.

This hypothesis, based on the growing body of evidence from across England (and indeed Europe), seems to indicate villages were a relatively late development and that they replaced an older, widespread and more persistent pattern of habitation (Taylor 1983); this 'older' distribution may still survive in the dispersed patterns seen particularly in western Britain. The general model (Roberts 1987) for the development of post-Roman settlement in some parts of England may be one of a gradual but dramatic replacement of previous, i.e. late Roman, dispersed pattern of hamlets and farmsteads by fewer nucleated village settlements which in many cases were planted and planned. This process, we suggest, was associated with the development of common field systems and, indeed, the creation of these field systems may have been the impetus for nucleation (Fox 1981). This model was developed from archaeological research, particularly in the eastern England, with work on deserted villages elsewhere. Extensive fieldwork in the East Midlands and East Anglia by Wade-Martins (1980) and others located pottery scatters indicating pre-Conquest settlements in ploughed land outside villages. This evidence suggested that the settlement pattern had changed and that many pre-Conquest settlements had been replaced by medieval villages.


Two aspects make fieldwork in western England problematical. First, a great deal of the landscape is pasture and 'unavailable' for fieldwalking and, in general, the fields are smaller, with more hedges and woodlands (Rackham 1986). Secondly, in comparison to the dynamic pottery industry of eastern England from post-Roman to the Middle Ages (Hurst 1976; Cadman 1983), the west is rather different. Over much of the period 400 to 900 (or even 1100) AD there is little, if any, pottery in use -- the period is all but 'aceramic'. Indeed pottery does not reappear in many parts of the west until the 10th century (Rahtz 1979). Consequently, there is little chance of recognizing low-status settlement sites from archaeological evidence over wide areas of the country.

Nevertheless, fieldwalking available ploughed land around Shapwick, making a collection of all finds of all periods, is an important component of the evaluation because it provides a context for wider interpretation. Archaeological evidence suggests occupation under certain areas of Shapwick village by the 10th century based on the earliest likely date of the pottery in the region. Any earlier occupation would be aceremic and therefore largely invisible to us, using the usual, conventional archaeological techniques.

'Habitative' field names and the Shapwick Case Study

The hypothesis that the settlement pattern changed from dispersed hamlets anti farmsteads to a nucleated village at Shapwick in the 10th century has been pursued over the last eight years using documentary and cartographic evidence; from the 13th century for the former, and from the mid 18th century for the latter. …