9. Reflections and Prospect

The survey was successful in its principal aim of identifying Gallo-Roman and potentially Iron Age rural settlements in Arroux valley, even if this was less easily done than in other regions! In addition, a certain amount of progress has been made in addressing our supplementary research questions. The failure to recover more Late Iron Age material could support the hypothesis of a reduction in rural sites at the time when Mont Beuvray was intensively occupied. In the early Empire, site numbers rise noticeably, only to fall to back again in late antiquity. Different types of site can be recognised, and at least up until the 3rd century AD, the impression is of a dense, settled, and intensively exploited landscape, while the pottery assemblages suggest differential integration into local and regional supply networks.

Further fieldwalking would certainly be worthwhile in the Arroux valley, but would be best done by workers based in the region, who are able to seize such chances as present themselves, very much as Roland Niaux has done over the years. As a visiting team, the possibilities for fieldwalking were particularly restricted. More opportunities do occur, but only at irregular intervals when the land use swaps around, or when fields are reseeded. It is, however, important that future fieldwalking data are collected in a standardised manner that will allow valid quantitative comparison between individual field units and with the results of surveys elsewhere - although this will always be problematic given the poor surface conditions pertaining in southern Burgundy. Due to the bulk of our survey material coming from follow-up collections and/or repeat walking of promising sites to obtain dating evidence, we have not attempted to define ADABs (Areas of Density above Background) in the manner advocated by Carreté et al. (1995), although a relative approach of this kind is essential if we are to escape subjective definitions of sites.

Given the geology of the area, the results of geophysical survey were better than we could reasonably have hoped. On the lower alluvium, as at Chantal and Laizy, the layouts of major Roman establishments were successfully traced, while on the higher terraces, large-scale, low-resolution survey proved able to pick out 'noisy' areas and some enclosures, producing results that could then be followed up by more intensive surveys. Resistivity survey fared less well in characterising sites, but might achieve better results in the future using other configurations such as a linear Wenner array.

PREVIOUS NEXT CONTENTS HOME