Statistički pristup u proučavanju keramike sa ranoneolitskih lokaliteta na teritoriji Srbije

Abstract

Najbrojniju trupu nalaza na gotovo svim ranoneolitskim lokalitetima predstavljaju keramički fragmenti. Iako veoma pogodne za proučavanje te vrste materijala, statističke analize se još uvek ne primenjuju u širem obimu. Prilog predstavlja pokušaj da se kroz statističke opise keramike sa ranoneolitskih naselja Srbije i njihovu interpretaciju utvrde osnovne karakteristike keramičke produkcije tih lokaliteta, a zatim njihova kulturna pripadnost i hronološki položaj.

The Early Neolithic in the territory of Serbia appears to be well studied, to judge by the number of excavated sites. Still the attempts at drawing some definitive conclusions about the Early Neolithic stumble upon a few major obstacles, most pf all the small scale of investigation into almost every site and, in the case of those systematically excavated, none of which though, is fully explored - a lack of monographic publications. Settlement sites in the territory of Serbia usually being unstratified, it is Early Neolithic, primarily painted, pottery that in the absence of other indicators is utilized as the exact and, as it seems, the only criterion for establishing the date and identity of cultures and for drawing comparisons. Pottery is the most copious category of movable finds on all Early Neolithic settlement sites. As typological change in pottery is hardly observable during the Early and Middle Neolithic, we have assumed that interrelationships between decorating techniques were determined by chronological and cultural change. The goal of this contribution is to consider the data about the finds of pottery from those sites where the results have been presented in quantitative terms (Table 1). From comparing the results of statistical analyses for six sites (Biserna obala-Nosa Ludaš-Budžak, Anište, Blagotin, Bukovačka česma and Drenovac) it has been inferred that all the sites are characterized by the domination of transitional and coarse pottery, most of it plain. The most numerous in the group of decorated pottery are fragments showing various kinds of impressions (45-90 %), while barbotine occurs in only 2.2-10 %. Of the total number of pottery finds, barbotine forms 0.4 (Blagotin) to 3.7 % (Bukovačka česma). Painted decoration occurs on four of the six sites analyzed. The painted component in the decorated pottery group forms 0.16 (Ludaš-Budžak) to 0.3 % (Biserna obala). The statistical indicators suggest that all the sites, considering that their pottery content fully corresponds to that of Proto-Starčevo II phase, not only belong to the same culture but also to the same phase. Although the varying frequency of individual features remains within the limits assumed for Proto-Starčevo II, it may be presumed that a slightly higher percentage of impressed pottery on the sites of Biserna obala and Ludaš-Budžak should be attributed to their geographical position and possible contacts with the Körös culture. The results of statistical analyses presented in this paper demonstrate the indisputable advantage of such analyses over descriptive methods, provided that the entire pottery material, or a carefully selected sample in the case of larger series should be included. They also show that the classification system is in need of a standardized terminology and precise definitions for all the features of pottery under scrutiny. Potentials of a statistical approach are not limited to the establishment and comparison of basic characteristics of pottery material. Depending on the research objectives and the material put through statistical analyses, its inclusion in the examination of broader archaeological problems is possible, and desirable.