Dans la cité des Santons, dès la conquête romaine, des constructions ont été édifiées sur un bassin sédimentaire offrant aux bâtisseurs un éventail de pierres calcaires de qualité. Une recherche élargie à l’ensemble du territoire a consisté à faire le lien entre extraction et mise en œuvre dans le bâti. Pour ce faire, les roches des principales carrières ont été identifiées chimiquement sous la forme de référentiels auxquels les blocs d’architecture mis au jour par les fouilles ont pu être comparés afin d’en connaître la provenance. La microscopie électronique à balayage (MEB) équipée d’un analyseur EDS (energy dispersive spectroscopy) pour l’analyse chimique multi-élémentaire des grains piégés par la sédimentation marine est l’outil principal de cette identification qui permet d’appréhender les circuits de distribution de la pierre. Le présent article fait le point de quinze années de recherche et d’analyses.
This article reviews archaeological and archaeometric research on stone and quarries of the Charente basin in southwestern France, a geographical area corresponding to the ancient civitas of the Santones. The study, which is ongoing and began over fifteen years ago, aims to identify the ancient quarries and to better understand the subsequent distribution of building stone. The briefly described geomorphological history, underscores the Pyrenean orogeny, a major event which transformed the Aquitaine basin of the secondary era into the current sedimentary plateau. Compressed into vast anticlinal and synclinal undulations, cut into by rivers, notably the Charente, it offered stonemasons a wide range of high-quality limestone, suitable for sculptural and architectural purposes, as well as all other construction needs, whether ashlar or rubble. Sedimentary formations ranging from the Jurassic to the Cretaceous are the ones most frequently found along lengthy stretches on either side of the Saintonge anticline axis. Turonian limestone was particularly appreciated in this context because of its fine grain and white color. Knowledge of ancient quarries and the individuals who exploited them requires a specific archaeological practice, focused on the reading of quarry faces and floors, in order to reveal the strategies implemented toward block extraction. This particular approach, however, implies the clearing of considerable quantities of ancient and modern waste, and it is therefore understandable that archaeology alone cannot identify all the quarries that punctuate the landscape and remain visible today. The Thénac (Charente-Maritime) quarry in Saintonge is the only one to have been the subject of programmed excavations and formally recognised as an ancient stone source, dated to the Julio-Claudian period. It was therefore necessary to use other methods to identify the stone from the large quarries in the Charente River basin, whatever the period of exploitation, and to create a sort of identification card for each of them. About fifteen quarries have been referenced throughout the area in question. How can we recognise the characteristics of the stone from a particular quarry that differentiate it from another, more or less distant one? And how can we establish that an architectural block encountered in an archaeological site comes from a particular quarry? These questions represent the main challenges of this research. The archaeologist cannot hope to identify the solution to these enigmas if they rely solely on limestone observation, using only the naked eye. Though undoubtedly better equipped in this endeavor, the geologist may still have difficulty distinguishing between limestone from two quarries located close to one another and attributable to the same geological stage. Thus, only the stonemason would be left with his intimate knowledge of the material, its lustre, or resistance to tools; yet this craftsman can only know the stone that they have worked themselves. It was therefore necessary to apply an original method, developed at the Université de La Rochelle by Professor Jean-Claude Mercier, based on the chemical nature of certain limestone components, and in particular by analysing the terrigenous grains trapped by marine sedimentation. These microscopic grains, transported from the surrounding land by the rivers, picked up by the currents, and deposited elsewhere depending on a variety of physical conditions, create a scenario in which the deposits at any given point are not the same as those at another location only slightly further away. This a basic principle of geographical discrimination, which is the basis for limestone recognition in each quarry and which the archaeometrist may be able to evaluate. In order to do so, they must methodically sample the quarry in question using vertically staggered samples, thus taking into consideration both the duration of the sedimentation and the time spent by the quarrymen exploiting the successive banks. Dissolution using hydrochloric acid allows the collection of these residual grains, which must then be prepared into thin sections. The multi-element analysis of these grains is carried out using a scanning electron microscope (SEM), the principles of which are described in this article. An emitter produces an electron probe that scans the six contiguous areas of the thin slide, which will cause secondary electrons to emerge from the surface of the sample, providing the topography of the grains, as well as backscattered electrons whose contrast is related to the atomic number of the chemical elements present. At the same time, the characteristic X-rays also emitted by the sample make it possible to determine the nature of the grains, the quantity and the identity of the chemical elements present on the surface by obtaining spectra and quantification tables. This technique is known as EDS (energy dispersive spectroscopy). Scanning electron microscope analysis reveals a mass of grains composed mainly of pure silica, with a minority of other more remarkable grains, wherein silica has combined with elements such as aluminium, potassium and iron. Other grains containing elements such as titanium or zirconium also occasionally appear independently. The list of these elements is not exhaustive. It remains for the archaeometrist to count, sort, and assign them false colours, as well as to arrange them on individualised layers with the help of image processing software in order to establish discriminating statistical groups of mineral species. This will ultimately provide a singular graphic synthesis: the quarry reference frame. Any archaeological sample, tested using this process in the hopes of confirming a specific quarry as its place of origin, must, statistically speaking, possess the same discriminating characteristics and fit into the ranges of the graphic synthesis of the quarry in question. Occasionally, the concordance is more or less precise, and in this case, we speak of assured, probable, possible, or unknown provenance. This method, which accompanies and reinforces the geological identification of a given limestone, also has the significant advantage of being reversible. If an archaeologically well-defined block is analysed using this method and its discriminatory characteristics relate to a quarry that is referenced, but whose original exploitation has not yet been documented, then it can be affirmed that this quarry was exploited in Antiquity. This article describes the conditions for use of local stone and rubble, and focuses on the main quarries exploited in Saintonge during Antiquity: Thénac, Saint-Vaize, Crazannes, Pons en Charente-Maritime, and Marcamps en Gironde. The major role played by the quarries of Thénac, Saint-Vaize and Pons in monumental construction is evident in the early days of Roman colonisation. The settlements of Saintes and Barzan (Charente-Maritime) present recurrent evidence of this. The conditions of exploitation and transport are described and the commercial areas established by maps and tables that correlate production and work sites. The general observation is that Saintonge is self-sufficient in building stone, apart from marble. It seems that the city’s context is a geographically ideal area for the trade of a material whose price increases considerably with the length of the journey, especially if it is by land. Saintes benefited greatly from the placid waters of the Charente, used to bring in stone from Saint-Vaize in abundance. The port town of Barzan, on the edge of the Gironde estuary, relied on a strategically placed road, specially created to link it to Saintes, its capital, with the Thénac quarry conveniently located along the route. Finally, this study reveals a rubble trade that is generally under studied, as is the case for the quarries of Marcamps, which supplied the second phase of the reconstruction of Barzan. To be viable, this trade benefited from the duration and abundance of a market facilitated by the downstream current of the estuary. Was Saintonge stone exported far beyond its territory, as the historian Camille Jullian speculated? To answer this question, archaeologists working on stone supply outside of this region and who suspect a possible origin in Charente should familiarise themselves with our diagnostic methods. Uncertainties and shortcomings remain in terms of an inventory and assessment which, by their very nature, can never truly be completed.
