The Gargano Promontory Flint: Mining Practices and Archaeometric Characterisation

An archaeometric project on prehistoric mining has been carried out in the Gargano promontory, in south-eastern Italy, since 1986, leading to the discovery of a large network of at least twenty mining sites, which was active from the Early Neolithic to the Early Bronze Age. The large area in which the mines are located, the considerable size of some mines and the skills required, together with the long duration of mining activities (four millennia), enable us to regard the Gargano as one of the main areas of flint supply in the central-northern Mediterranean. Three geological Gargano formations were mined: the Maiolica, Scaglia and Peschici formations. They appear to differ significantly with regard to the morphology of the flint. In the mines opened in the Nummulite limestone formation, flint seems to occur only as generally rather large lenticular nodules (often more than one metre in diameter). The Maiolica and Scaglia formations are different: the exploited flints primarily occur in the form of spherical and irregular nodules, only in some cases associated with lenticular lists. During the sixth millennium BC on the Gargano promontory we exclusively find flint mines with entrances opened into the slope of a hill and developing sub-horizontally. This type of mining generated large amounts of mining debris that were accumulated outside the mine and that are sometimes still preserved in their entirety. Neolithic mining techniques in the Gargano underwent a radical transformation from the fifth millennium BC on and especially during the fourth millennium BC: rather small mines can be identified that are characterised by vertical access to the flint-bearing formations, with or without horizontal digging at the base of the shafts. Systematic characterisation studies of the flint from primary and secondary Gargano sources are still lacking and the paper presents the plan and the preliminary results of a new project, applying multi-parametric characterisation obtained by non-destructive approaches that are rapid and economic (during this initial stage of the project macroscopic/petrographic description, colour and reflectance/gloss measurements). A total of 151 flint samples, representative of the three flint-bearing limestone formations, the ancient exploitation of which is demonstrated by mining indicators, and of a fourth formation (Fucoid Marls), which has not provided evidence of flint mining yet, was collected. The flint samples were taken from Neolithic mines and from accessible flint outcrops throughout the Gargano promontory. The macroscopic description of the flint samples provided a matrix of categorical and numerical data which helps to identify the distinctive features of the four formations investigated. Spectro-colorimetry provided coordinates in CIE L*a*b* colour space and reflectance in the visible spectrum of the flint matrix. The two main mining districts hitherto identified by archaeologists, namely those of Vieste and Peschici, cut into specific formations (Peschici limestone and Maiolica respectively), with mines that exploited flint with various characteristics even within the same formation. However, we observed some chromatic homogeneity at site level. This classification of chromatic and reflectance data confirms the similarities visible to the naked eye between most of the samples from the Maiolica and the Peschici limestone formations, which also turned out to be more homogeneous than the Scaglia and the Fucoid Marls formations. No mines were found in the Fucoid Marls formation, though the quality of flint was similar to that of most of the mining sites in the other formations. Although our interpretations have to be tested with chemical and mineralogical data, the statistical approach using quantitative macroscopic data seems to be a promising way of classifying and correlating geological and archaeological flints from the Gargano. Following the steps of the multi-analytical protocol, flints will be analysed with Raman spectroscopy to determine their mineralogical composition, whereas the chemical composition will be obtained by LA-ICP-MS. A detailed micro-structural and palaeontological study will be carried out using a scanning electron microscopy in order to distinguish flint sources of different ages.