La tranchée supérieure de l’aqueduc du Gier à Saint-Joseph (Loire) : nouvelles données et synthèse

Abstract

The aim of this paper is to provide a short synthesis of old studies carried out on the upper trench of the Gier aqueduct in the light of new data provided by two preventive archaeology operations conducted in Saint-Joseph (Loire) at a place called Grange-Blanche: an evaluation in 2007 and the excavation that followed in 2008. In particular, these new data opened up new ideas on the nature and function of this trench, and cast light on questions about the route and construction of the aqueduct.In 2007, the archaeological evaluation conducted by Catherine Coquidé (Inrap) prior to the construction of a housing estate brought to light for the first time a buried section of the upper trench of the Gier aqueduct alongside the remains of its channel, which it doubles from Saint-Chamond (Loire) to Chabanière (Rhône) between ten and fifteen metres above its base. Known since the 19th century, it had until then only been identified above ground, and had been explored in just a few specific studies and syntheses. The few complete maps that we have date from the first studies, in the second half of the 19th century and the beginning of the 20th century. In these, a number of hypotheses had been proposed as to its nature and function, the most commonly accepted being that it represented an early route of the Gier aqueduct which was abandoned because the work involved in developing it in the large valleys that it bypasses would have been too extensive and costly.The 2008 excavation that followed this evaluation, led by Guillaume Maza (Evéha), focused only the upper trench, its immediate surroundings and an ancient ground level section observed to the northeast of it. The excavation followed its course over an 89 m stretch at the place where, like the aqueduct, it begins a turn from the southwest. In addition, a thorough examination of its profile and filling was carried out by means of four deep soundings and two verification surveys. Its profile, similar to that observed above ground at Chagnon during earlier surveys by Jean Burdy and and Jean-Claude Litaudon, appeared very similar to that of the aqueduct’s installation trench, though with smaller dimensions and significant irregularities. Although the date of the trench could not be clarified, examination of its filling produced a wealth of information about its abandonment. This probably happened quickly and took place in several stages. No trace of masonry was found in the infill or near the trench.These new unpublished data add to our knowledge of this structure, an almost unique and somewhat poorly documented example in the Roman world. While it is undoubtedly linked with the Gier aqueduct, it seems to be more than a levelling error, but rather part of a different aqueduct project from the one finally executed –an aqueduct that would have been around ten metres higher, narrower and likely to have developed more significantly in the Janon, Langonand and Durèze valleys. While the reasons for its abandonment remain unknown, it would appear that a construction project, even one that has started, does not stand still. The discovery of the upper trench during the evaluation of 2007 and the 2008 excavation was an opportunity to document the structure in a new, buried form, and to consider whether it is worth protecting. Although it is sometimes included in the the Gier aqueduct’s protected zone, it does not benefit from any specific protection.