Roman furnaces of the Montagne Noire (France), instructions for use. Results of twenty years of experiments between archaeology and archaeometry
Résumé
From the first century BC to the third century AD, the southern zone of the Montagne Noire located in the southern part of the French Massif Central, 15 km north of Carcassonne and 100 km east of Toulouse, was the place of an intensive production of iron. During the Roman Empire, this iron provided a long-distance trade like that shown by the chemical analyses (e.g. Baron et al., 2011) and was probably mostly for the Roman armies (Long et al., 2002).
Twenty-five years of interdisciplinary researches, recently published in a monograph (Fabre et al. dir., 2016), have led to define a “Martys type” furnace on the basis of the morphological similarities of metallurgical structures employed and put together in batteries. The number of furnaces and the considerable size of the slags amounts are part of the evidence for the intensity of the iron production (about 110000 tons of metal in three centuries according to the estimations, e.g. Fabre et al. dir., 2016). How were the metallurgical operations conducted in Roman times in the “Martys type” furnaces?
To find the way these metallurgical structures were used, more than twenty experiments have been carried on between the years 1991 and 2009, first in the archaeological furnaces reconstructed at Les Martys (Aude, France) and then on the experimental platform at Lastours (Aude, France) using furnaces as similar as possible to the archaeological ones. From case to case, natural draught or artificial blowing was employed. The products and wastes of the experimental reductions were compared with the archaeological ones on the basis of the mineralogical and chemical compositions of ores (carbonated derived vs sulfide derived) and slags showing the impact of the operating conditions. At the same time, two other experimental platforms were created: one was dedicated to experiments on a model (scale ½) equipped with measuring instruments (temperature, gas compositions…), the other was used to determine the role of different types of iron ores. The synthesis of the results from the three experimental platforms, including some archaeological comparisons with the same types of furnaces, “the thick-walled shaft furnaces” type established by Pleiner (Pleiner, 2000, 179), leads us today to confirm, in agreement with the archaeological data, that the Roman metallurgists selected one type of ore, the carbonated derived one, and to propose a procedure for the use of the Roman “Martys type” bloomery furnaces using at least partly natural draught.