1.3 Field survey methods

Field survey is the best-known component of the regional approach. While walking over the fields finds are collected and soil features visible at the surface are charted. The finds have either been present in the topsoil since the site was inhabited or have become incorporated in the topsoil through mechanical or natural causes. The visible and collected finds are always, to a greater or lesser degree, a distorted reflection of what is present in the soil. A great deal of methodological research has been done as part of the survey process into the assessment of factors such as sunlight, plant cover, depth of ploughing, speed of walking and soil texture on the visibility of different find categories (i.e. Shennan 1985). The fact remains that in a regional investigation we are saddled with a pseudo-random sample which may not represent the archaeological reality. It may result from a scientifically planned investigation surveying a field once or twice in 'controlled' conditions or a collection of finds by an amateur archaeologist who has been combing a field for a dozen years without any clear system. Moreover, experience has taught us that as regards the archaeological value the amateur archaeologist seems to beat the professional more often than expected.

In areas where there is no agriculture it is necessary to adapt the survey technique. Walking, for example, along ditches or molehills may also provide a relatively good picture of the surface finds. Meadows are sometimes ploughed up, revealing their secrets. In forested areas dirt tracks or drifting sand may offer clues for the presence or absence of sites. However, these finds should be considered less representative than in agricultural areas as they can only give a partial picture.

In areas where field survey is completely useless, for example in areas with recent geological (peat, clay) or anthropogenic (sod manure) surface deposition, other methods must be employed. In the West of the Netherlands investigation of drill cores is an important aid. The core diameter may vary from several cms, as in the investigation of the donk settlements in the Alblasserwaard (Verbruggen 1992), to some dozens of cms in the investigation with the 'mega auger' used in the mapping by the RAAP foundation (i.e. Odé & Verhagen 1992) or as part of the Meuse Valley Project (Verhart & Wansleeben 1990; 1992). The application of a tree auger (Assendelver polder Project, Brandt 1983) is essentially a transition towards a more systematic reconnaissance by way of shovelpits and small test digs. Although these methods do reflect what is present in the soil, they have often lost the all-encompassing character so necessary for a regional survey.