Pakistan
Banbhore – 2015

In 2015 was led the last campaign of the four-year research programme on the ancient city of Banbhore. The aim of this French campaign was to find the first traces of occupation on the Banbhore site and to identify and characterize the pre-Islamic port zone.

Eight surveys were opened, their location based on the previous campaign research. They were located in the continuity of those of 2014, and following the northeast-southwest slope of the field. The excavation was made by digging more or less deep successive layers, depending on what was discovered in the upper levels. Ceramics then made it possible to propose a chronological range for each of these arbitrary strata.

The surveys delivered data on the evolution of the site occupation from the pre-Islamic period to the Islamic period, this latter being characterized by the construction of the great rampart still visible today.

The rampart clearly presents several states of construction, each of them showing as well different building phases. Survey IV-51.41 was carried out in front of the Islamic wall. It highlighted masonry constructions with ground levels, located ca. 5 m to the inner facing of the rampart. An unbuilt space was thus left free between the fortification and the buildings. Several states of construction were noticed. The first masonry state is a wall about 1 m wide, leaning against the rampart. A door seemed to open in this wall, though it remains to be confirmed. It is difficult to understand the function of this first state : was this wall a doubling of the rampart or a part of a building leaning against the rampart? This construction was later repeatedly reworked.

Under these layers of construction, a series of alternated alluvial and sandy deposits strata was seen. Each sandy layer testifies to a short occupation between each rise of the river water level. In one of these layers, a lay out in posts with remains of decayed wooden boards was found. Its orientation follows that of the arm of the river, as the Islamic rampart does. The structure is a light frame construction that has been used for a short period of time. The post holes and the wooden boards were covered with the alluvial deposit that followed. This structure must probably be linked with the idaily life of the pre-Islamic city inhabitants on the banks of the river upstream from the port (fishermen or people related to the life of the port).

The hypothesis of the harbour location on the western arm of the river was based on the discovery of these levels of alluvial deposits and of a 1.4 m wide wall in Survey V-95 in 2014. In actual fact, the alluvial levels can be observed throughout the site (still visible old stratigraphic sections make possible this observation, which is also supported by the excavation of the Italian team). The Survey V-95 was re-opened in 2015, making possible a new interpretation of the wall seen in 2014. This latter turned out to be a collage of two masonries, each about 70 cm wide. The first wall formed part of a rectangular north-south building about 2 m long. The second masonry corresponds to a mere posterior doubling, intended to reinforce this building implanted at the edge of a rupture of slope. This construction is therefore linked to the particular topography of the site, established on a succession of terraces. In the present state of knowledge, the original location of the port remains uncertain. In the 1960s, a port had been identified near the southern entrance: perhaps this hypothesis should be maintained?

The alternation between alluvial and sandy deposits was also seen in Surveys IV-6.15.16.25, IV-3.4.14 and V-96 (2014). Numerous areas of fire places and remains of fauna bones and shells have been observed on the sandy strata. They testify to the short occupation of this lower part of the city during the pre-Islamic period. A more sustainable habitat level, dated back to this pre-Islamic period, prior to the Islamic rampart construction, has also been observed in Survey IV-3.4.14. It was characterized by a mortar floor showing divisions that could correpond to the negative of prior internal partitions. This type of floor, characterized by its colour and composition, is found in different parts of the site, at least in its southern part.

The excavations in the surveys have all stopped on levels upon the natural rock, which was never reached. A last survey was opened during the last week, 120 m north of Survey IV-51.41. It was meant to reach the first levels of occupation of the site. In order to reach the bedrock, a well survey of 8 m deep was digged. This level being under that of the water table, a pump was installed. Despite the difficulty of observation (water table, very low ground-level of the survey), it remains possible that the bottom of this borehole corresponds to the geological level of the site. The ceramic shards should help dating this first occupation sequence.