Reinterpreting a “Roman villa” site: First results of nondestructive archaeological research in Dračevica (Donji Radišići, Bosnia and Herzegovina)

Abstract

The article presents the results of archaeological prospection of site Dračevica (Donji Radišići, Ljubuški municipality, West Herzegovina Canton, Bosnia and Herzegovina) which was partly excavated in the late 19th century and so far has been interpreted as a Roman villa, possibly with a special function (road station) and a name attested in ancient sources (Bigeste). Through the use of various nondestructive prospecting methods (aerial photogrammetry, fieldwalking survey, magnetic susceptibility, magnetometry, earth resistance), precise location with RTK GNSS and integration of the data in GIS, the site was restudied and reinterpreted. It was established that the structures are not isolated, as was assumed in the earlier research but rather were a part of a much larger Roman settlement. The research proves that the internal spatial division of the site is more complicated than previously thought, consisting of several building complexes located along a road, as well as land division systems. Such settlement pattern suggests similar, average-sized land plots and farms. The importance of the research lies also in the methodological approach and the outcome of the geophysical research. Our research was the first application of nondestructive methods of archaeological prospection on Roman sites in the karst landscapes of West Herzegovina, proving the usefulness of the applied methodology in these specific and difficult conditions. We have found dependencies in between the geological background (thin layer of low resistive, colluvial soil on shallow limestone bedrock and the occurrence of highly magnetic terra rossa, formed on weathered limestone) and the geophysical responses registered by various instruments, that is, moderate to good magnetic and electric contrasts between the features and their surroundings, with relatively strong anomalies from geological features. The most important was finding relationships between particular magnetic and electrical anomalies and the presence of very shallow bedrock or faint archaeological remains (e.g. limestone walls). Our methodology, as well as some of our interpretations, will be helpful as a reference for further surveys in the region.