Analysis of the spatial distribution of Fourth Century Roman coins in
School of Geography, Queen's University, Belfast BT7 1NN
The reforms of the Emperor Diocletian in AD 294 brought to an end the last of the provincial coinages of the Roman Empire. Within four years of the reform there were 14 mints across the Empire striking coin with a broadly uniform design. The mintmarks given on these coins make it possible to analyse the spatial distribution of the coins on the basis of the mint from which they derive. In this paper, several methods were employed in the examination of the spatial distribution of base metal coins struck during the period from AD 294 to 402. The focus in this analysis was part of central Europe (comprising modern Slovenia and western Austria) that was formally within the Roman province of Pannonia. The consistency of designs on the reverses of Fourth Century Roman coins across the Empire facilitates their division into distinct chronological groups. In this research, the coins were split into 13 such groups. The analyses entailed the use of several distinct techniques to examine different characteristics of the data. Firstly, simple maps of site finds as proportions of coins of each issue period from each of the mints that struck coin in the region were generated. Following this, to assess the interaction between the products of different mints principal components analysis (PCA) was used and the first two principal components were examined. In addition, the first component was mapped and the maps derived were interpreted in conjunction with the plots of the first and second components. Finally, the relationship between mint proportions and distances from the respective mints were assessed.
Distances from mints to sites where coins have been found were measured in two ways. Firstly, Euclidean distance from the relevant mints was used and the form of the relationships between proportions of coins from the mints and the distance from those mints was assessed. Secondly, Diocletian's Edict of Maximum Prices of AD 301 was used to estimate the cost of travel from the relevant mints. The Edict included maximum costs of transporting goods by various means. Transport cost ratios derived from the figures in the Edict were used to weight travel by land and water. That is, travel by sea is likely to be cheaper over a fixed distance than travel over land. A Digital Elevation Model (DEM) was used as the basis of a cost surface from the mints. The slope of the terrain locally was derived from the DEM and transport cost ratios based on the Edict were used to modify the slope map. That is, areas with gradual slopes were considered less 'expensive' to cross than areas with steep slopes. Following this, friction surfaces were generated from the mints using the slope map modified by the travel cost ratios. The relationships between mint proportions and cost distances were then assessed. The analyses demonstrated that GIS, in conjunction with a range of statistical tools, could be a powerful means of enhancing our understanding of the supply and circulation of Roman coins in the Fourth Century AD. Following the analysis, potential improvements, problematic issues and future work were discussed.