The accuracy of road distances in the Antonine itinerary

Sammanfattning: The Antonine itinerary was compiled in the early fourth century CE and is a key source to our know¬¬ledge about ancient geography and the Roman road system. However, the major research work dates more than 100 years back in time and few studies have been focussed on its overall reliability. The aim of the present work is to assess the accuracy of the distance figures in the itinerary by internal validation, by comparisons with other ancient sources and with the modern map. Geographic and numerical patterns in the data are also analysed. The evaluation of internal consistency and numerical patterns indicates that the distance figures are to a certain degree imprecise and likely to have been derived by estimation rather than by actual measurement. Although the overall accuracy is good, there is a geographic variation and gross deviations occur as well. Using the itinerary should therefore take these aspects into account, and the imprecision makes it less meaningful to use the distance figures to scrutinise individual road stretches. However, if this imprecision is recognized then the itinerary gives a good description of the Roman road system. The comparable relative proportions of deviations for repeated road stretches and for the mile sums cannot be explained by errors in transmission. The authenticity of the itinerary is further confirmed by the specific numerical patterns occurring, which can also be seen from other ancient sources. It therefore seems likely that errors in transmission play much less role than previously assumed. This also means that the reconstructed archetype is likely to be close to the original compilation, and thus gives us a glimpse of the nature and quality of geographic information available in antiquity.