Abstract
p. 349-350
Texte intégral
1The work done at the Conjux site (Lake Bourget, Savoie) shines a light on the remains of a place of worship dating back to the Roman period that is currently more than 3 metres underwater. The site was primarily made up of hundreds of bases of wooden spikes and posts, with artefacts and ecofacts in the surrounding area.
2It was discovered in 2000 during underwater survey work led by André Marguet to draw up an archaeological map of large French perialpine lake sites. Ten years after it was discovered, a university conducted work on the site (2010) and suggested that these remains be interpreted as a site of worship from the Roman period. After this study, five years of fieldwork were done at the site (2011-2015), followed by three years of preparing for publication (2016-2018).
3The site is on the northern side of Lake Bourget, near where it empties into the Savière, its natural outlet. This small waterway plays a significant role in the water system formed by the Rhône river, the Chautagne and Lavours marshes, and the lake. Periodically, when the Rhône floods, the Savière changes course and the river flows back into the lake. When the water level is very high, the river and lake flood the entire marshland, and only limestone knolls known locally as “molards” emerge from the vast stretch of water.
4Taking the Rhône as a geographic boundary, the site was located near the border of the civitas of Vienna (present-day Vienne), and the former province Narbonensis. The lake was not named in sources during Antiquity, but some accounts may have indirectly referred to the lake when discussing the Rhône and the bodies of water in its catchment area. For travellers in the hilly region where the Alps meet the Jura, Lake Bourget offered an attractive alternative to riskier land routes. Populations were drawn to this body of water, as we can see from numerous occupations in the area. The northern part of the lake had the advantage of being located along the imperial road from present-day Vienne to Geneva, as well as being near the Savière, which emptied into the Rhône. During the Roman period, the Conjux site was partially submerged near the bank, at a shallow depth of less than a meter at the end of the 3rd century.
5Looking at the site overall, we see homogeneous artefact and ecofact remains in their original position.The topographic survey of the site documented over 1400 pieces of wood, and nearly 400 were collected for dendroarchaeological studies. These were mainly remains of stakes and posts loosely buried in the sediment. They were, for the most part, small in diameter and showed very few traces of human activity, apart from the sharpened base to help them penetrate layers of sediment. A total of eleven types of wood were represented, with oak by far the most common. Measurements showed that this wood was cut from very young trees and planted in the lake waters nearly every year from 123 to 185 A.D., most often in autumn or winter. More generally, this site appears to have had a deliberate layout but no buildings, being made up of stakes and posts that were added each year and arranged according to a system all their own.
6A total of 122 coins were found on site. All were minted in Rome. There were several Julio-Claudian coins, a number of Flavian coins, mainly minted during the reign of Domitian, but most of the coins dated from the Nerva-Antonine dynasty, primarily from the reigns of Antoninus Pius and Marcus Aurelius. The occasional coins from Commodus to Constantine rounded out the collection.
7The instrumentum was primarily made up of manufactured metal objects. A number of remarkable objects stood out due to their rare nature and excellent state of preservation. One such example was two knives that still had their sharp edge and a part of their wooden handles. One of them was stamped. There were also handbells and a stylus at the site. A large number of carpentry nails were discovered in the area around the stakes and posts, along with fishing and ballistic materials (hooks, stone and lead fishing net weights).
8There was a large quantity of ceramic ware, with a unique assemblage. It included a range of fine wares, including a number of intact cups in what is known as “terra sigillata chiara B” dating back to the second half of the 2nd century A.D. Common, or coarse, wares were represented primarily by a series of black-coloured dark common ceramic pots of different sizes. Various types of light common ceramic flagons were also present, a number of which showed signs of deliberate alterations.
9On closer analysis, we can determine that this site was arranged in different zones with specific characteristics. This information alone is not enough to determine the function of the site, but by examining the assemblage of materials, we can identify it as a place of worship. A number of details support this interpretation: the presence of a large number of small coins, numerous intact and intentionally mangled objects, and objects frequently seen in religious contexts (knives, handbells and a turibulum).
10By examining artefacts and ecofacts, we can speculate on activities and practices. Some objects were linked together, such as the coin discovered inside a spike, while other coins were linked to other small posts. If we consider the closest parallels, we can see that this was a symbolic practice. As for the other coins, some may have been tossed, but others could have been originally hung higher up on the spikes. Other coins were placed under overturned bottoms of pots, which is evidence of actions taken on site. Some of the ceramics were intact and likely served as containers for offerings and receptacles used for rituals. Others were damaged, broken or dispersed. The presence of a stylus and a fragment of pigment could be evidence that writing was practiced, or figurative representations were drawn. Lastly, the remnants of wood charcoal and fragments of an oil lamp and incense burner raise questions about whether fire was used at the site. Likewise, remains of consumed wildlife, including some young animals, bearing cut marks, could be potential evidence of sacrificial rites.
11The interplay between the artefacts and wooden stakes and posts raises questions about the role they played at the site of worship. The presence of wood lain horizontally shows that some of it was in the shape of large poles. Based on the remains identified and available archaeological, iconographic and ethnographic sources, these spikes and posts could be interpreted as stands used to hold ritual items (tituli, crowns, ribbons, handbells, offerings, torches, etc.). Others could have served to mark a boundary or a fence or as bases for altars, aediculae or lightweight structures. Any and all combinations are possible.
12Based on their position and their large number, we can consider the possibility that these stakes and posts may have played a role that was more than merely practical. They may have held a more important role in worship. The collection of materials found at the Conjux site has sparked discussion on wood in religious contexts and the symbolic value attached to it. Further research is called for on beliefs associated with different types of wood, the diverse range of shapes in wood statuary, including anthropomorphic stakes and herms, and more broadly all long wooden objects that may be encountered in religious contexts.
13In more concrete terms, thanks to the detailed data available to us, we can offer a reconstruction of how religion was practiced here. Worship seems to have been regular, on a yearly schedule, with a strong seasonal pattern. It likely took place in autumn or winter. Based on signs that fire was used and on what we find in documentary sources, it is possible that some rituals took place at night. Worshipers likely moved on foot, assisted by a layout specifically designed to provide access to collection points.
14There are several potential reasons why this location may have been selected for the worship site. In a broad sense, the remarkable characteristics of natural settings like lakes and the disruptions – routine or episodic – that they experience are often ascribed to the divine. For the Conjux site specifically, we can put forward a number of hypotheses. The worship could have been related to seasonal water level variations that submerged the entire Chautagne and Lavours marshland. It could also have been related to the change in course of the Savière, a remarkable phenomenon that is rarely found elsewhere in the world, or to the unusual and irregular changes in the lake caused by the climate variations in the 2nd and 3rd centuries. The patron god of the site could have been Savus, or other more common gods or goddesses of the provinces of Gaul and the Gallo-Roman city of Vienna.
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
D'Augusta Tricastinorum à Saint-Paul-Trois-Châteaux
Drôme
Thierry Odiot, Valérie Bel et Michèle Bois (dir.)
1992
Pots et potiers en Rhône-Alpes
Époque médiévale, époque moderne
Élise Faure-Boucharlat, Tommy Vicard, Bruna Maccari-Poisson et al.
1996
Autour du groupe épiscopal de Grenoble
Deux millénaires d'histoire
François Baucheron, Franck Gabayet et Alain De Montjoye
1998
Dynamique du paysage
Entretiens de géoarchéologie (Table ronde tenue à Lyon les 17 et 18 novembre 1995)
Jean-Paul Bravard et Michel Prestreau (dir.)
1997
Le pont de la Guillotière
Franchir le Rhône à Lyon
Joëlle Burnouf, Jean-Olivier Guilhot, Marie-Odile Mandy et al.
1991
Brandes-en-Oisans
La mine d'argent des Dauphins (XIIe-XIVe siècles), Isère
Marie-Christine Bailly-Maître et Joëlle Bruno-Dupraz
1994