Vadas Ferenc (szerk.): A Wosinszky Mór Múzeum Évkönyve 15. (Szekszárd, 1990)
Die awarischer Kultur - Márta Balla: Provenance studies of avar ceramics by neutron activation analysis
We draw 50-100 mg of powder samples by the help of a conical diamond drill. The samples are filled in small polyethylene capsules followed by accurate mass weighing. Then they are irradiated in the nuclear reactor of TU Budapest for 8-12 hours time of irradiation. Then follows the measurement and the evaluation. For gamma-ray spectrometry measurements we use semiconductor detectors and a multichannel analyser. The evaluation of the gamma-spectra is carried out by a computer. A great branch of analytical information can be obtained from the concentrations of elements but their succesful review, systematization and grouping of the samples similar to each other can be accomplished only by applying computerized mathematical statistical methods. The primary point of view when choosing the methods was that data processing should modell the questions arised by archaeological investigation and results should answer the questions directly. Two initial situations were modelled. First, we do not have „a priori" information on the origin of the samples. The ceramics found are grouped according to the distribution of their trace elements without knowing the workshops where they were produced. An adequate mathematical method of this approach is cluster analysis. The other approach named similarity analysis was applied when sherds of unknown origin were to be placed into one of the archaeologically and analytically well defined workshops. In this case on the base of the elemental concentrations we calculate probabilities by which we could give the chanche whether the sample belongs to a workshop or not. The cardinal question of chemical analysis of finds is whether it is possible to extend the pieces of information obtained from a small piece of sherd over potteries of bigger size i.e. what is the confidence or error of stating that the composition of the sherd is similar to that of the whole original pottery. That leads us to the calculation what deviation can be considered significant between two samples. So in the first step the investigation of homogenity must be carried out on each type of ceramics investigated by taking samples from several places of a sherd from inner and outer surfaces from fractures etc. and analyse them. After this investigation it can be decided undoubtedly whether the deviation of the sherds calculated from the analytical data is significant or not. Further on in a few words leaving the archaelogical interpretation to Gyula Rosner I would like to report about the examinations we carried out on ceramics of the avar age (2). In the first step examination of the two potting-workshops excavated in the avar village Szekszárd-Bogyiszlói road was carried out. On the basis of their different trace element distributions the two workshops can be separated which is an accomplishment of great importance considering that they were working at the same time of a small distance from each other and producing potteries of the same type and technique. After the analytical definition of the workshops we analysed about 20 pots found in the cemetery by the Szekszárd-Bogyiszlói road and attempted to answer the question by similarity analysis whether the sherds found in the graves were produced in the workshops of the village or not. The studies showed that the great part of the ceramics found in the cemetery is from the workshops of the settling. Only about pots from 2 graves can be stated with complete certainity 132
