School of Historical and Philosophical Studies - Theses
Permanent URI for this collection
Search Results
1 - 2 of 2
-
Item
-
ItemProvenance and ware types: applications of proton induced x-ray emission and proton induced gamma-ray emission to analysis of northeast Anatolian pottery( 2004)Proton Induced X-ray Emission combined with Proton Induced Gamma-ray Emission offer an alternative to the more expensive Neutron Activation Analysis as a method to ascertain the elemental composition of pottery. The method makes use of the x-ray and gamma-ray energies released when a sample is bombarded with protons. These energies can be identified with specific elements and measured allowing quantification of the elemental composition of the samples. This study seeks to determine whether the techniques can be successfully applied to pottery from northeast Anatolia, and if this information is capable of meaningful archaeological interpretation leading to a contribution to further appreciating the nature of past societies in northeastern Anatolia. This area is mountainous, yet well watered and fertile. While the difficult terrain must have limited movement, the presence of four major routes linking central Anatolia with the Caucasus and Iran traversing the area ensured that regular contact was made with the cultures of neighbouring regions. The influence of these various cultures may be seen in both the general archaeological evidence, and in the ware types found in the area. Much of the pottery analysed in the study originated from an archaeological survey of the Bayburt Plain, one of the many east-west oriented depressions characterizing the region. Other sherds were obtained from excavation of a site in the Bayburt Plain, Büyüktepe Höyük, and from excavation of Sos Höyük, which is located in the Pasinler Plain, slightly over one hundred kilometers to the southeast. After PIXEPIGE analysis, the resulting raw data were log-transformed and standardised. Outliers were identified using box-plots and both average and single linkage cluster analysis, while possible conflicts between the assumptions inherent in cluster analysis were abrogated by eliminating non-normally distributed variables from the main analysis. The complete data set was analysed using Ward's Method of cluster analysis, complemented by average linkage and complete linkage. A series of subgroups based on the location of find spots and the chronological relationships between sherds were also analysed. The results of these analyses suggest that PIXE-PIGE may be used with reasonable success to identify sherds from two different areas, but that it was less successful in discriminating between sherds manufactured at the various sites throughout the Bayburt Plain. The isolation of two separate ware types into their own groups indicated the importance of the role of production technologies in addition to the source of the clay in contributing to the composition of the fabric of archaeological specimens.