William Van Andringa (University of Lille III – CNRS, UMR 8164 – HALMA-IPEL)
Thomas Creissen (Éveha International – François Rabelais University, Tours, CNRS UMR 7324 – LAT)
Henri Duday (University of Bordeaux – Anthropological Laboratory of Bordeaux)
How Eveha Participates
The excavations carried out in 2014 were the first to take place as part of this new research programme. Their aim was to assess the archaeological potential of the site in order to establish a suitable strategy for the coming years. The aim was also to evaluate the usefulness of certain techniques adopted for this new project: photogrammetry for the recording of remains, and micromorphological analysis of soils.
The results were somewhat surprising, as the study area turned out to be quite different from what had been seen during the excavation of adjacent zones a few years earlier. While the funerary levels were only a few centimetres below the current ground level in enclosures 23 and 25, the zones investigated this year were covered by a large depth of fill layers. The removal and analysis of these layers slowed down the excavations considerably (fig 1).
It is important to note that in places, lapilli contemporary with the eruption of 79 AD were found in situ.
The earliest settlements
The earliest evidence of settlement currently known is attested by a large mausoleum from the Late Republican period (third quarter of the 1st century AD) which was found in the western part of the excavated area (Tomb 27 OS; fig. 2). Layers corresponding to activity contemporary with the monument could not be clearly identified, but were thought to be present at the bottom of a deep sondage, 1.70 metres below the current ground level.
At the other end of the explored zone, a funerary ensemble comprising five tombs marked by steles was identified (fig. 3). Two of these had been known since 2007. This enclosure was not fully excavated this year and it has not yet been precisely dated. It could be from the first two decades AD.
Between these two groups, the remains of a rectangular mausoleum were unearthed (fig. 4). This measured approximately 2.6 m x 2 m (part of it was under the baulk and could therefore not be excavated). One burial was found in the structure, and the bones were contained in a glass urn. The floor levels associated with the mausoleum have not yet been reached. It is thought to date from the reign of Augustus, though this has not been confirmed.
All of these structures are aligned along the south side of a stretch of the road which is still visible further to the east. The exact width of this road is not known, but it appears that it was gradually widened by successive additions of material. During these early periods, the road was built on top of a ridge, with the funerary structures located at the bottom of the slope.
From abandon to redevelopment
Subsequently, the ensemble undergoes considerable transformations, for which a detailed chronology has not yet been determined. Monument 27 OS is neglected and slowly decays. Mausoleum 26a is completely abandoned and falls into ruin. The small funerary zone to the east is gradually covered over, before eventually disappearing into the landscape; More widely, the area located below the road is gradually filled by alluvial deposits. In some places, it appears to have been almost at the same level as the road, which had by this time been widened.
The funerary monument of Castricia Prisca, which is still visible in the eastern part of the excavation area, is built on top of these natural fill layers. Its construction dates to around 60 AD.
It may be during this period that an inscription announcing a gladiatorial combat is painted on monument 27 OS. The text is patchy, but mentions the participation of twenty-five pairs of gladiators.
The last phases of occupation
During the last phases of occupation, the accumulation of natural deposits continues. Layers containing various materials (mortar, stones, plaster, pottery) build up on both sides of the mausoleum. A layer rich in material also accumulates within the building itself, in which several fragments of a Samian ware bowl with a marbled slip from the workshops at La Graufenesque were found (fig. 5). In particular, the decay of the mausoleum continues and this eventually affects the burial, which is partly torn open, causing bones to be be scattered both inside and outside this small funerary monument (fig. 6).
During this phase, colluvial deposits also continue to accumulate. The ground level rises considerably, until it completely covers mausoleum 26a. The analysis of these successive deposits has demonstrated that they are the result of a natural phenomenon: they were brought to this location by heavy rains. In its final state, it appears that the road was now lower than the ground on either side. We can therefore assume the existence of some kind of structure to keep the sediment off the road, or that the road was regularly cleared. At the top of this layer of colluvium, just under the layer of lapilli, the imprint of several roots were visible. A mould was taken of a particularly large one of these, and it was determined that it belonged to a medium-sized bush. The area was therefore under vegetation at the time of the eruption in 79 AD.
The photogrammetric surveys were used at two different scales: first, to place the study zone in the context of the surrounding area (fig. 7). Second, to record the most important structures, as well as the various phases of excavation. A general photogrammetric survey was therefore conducted each time another phase of the site was uncovered (figs. 8 and 9). Such work generates a considerable archive (several thousand photographs) and the processing of these data remains quite slow. This does not take from the fact that it is a very useful tool which enables us to record the structures destroyed by the excavations more effectively. It is also possible to generate as many cross-section profiles as were considered necessary to reconstruct the evolution of the site as a whole, or in relation to a particular structure.