During June 2006, a salvage excavation was conducted in the site of Horbat Be’er Shema‘ (Permit No. A-4812; map ref. 1565/5740), in the wake of damage to the site caused by the plowing of the site’s western side. The excavation, undertaken on behalf of the Israel Antiquities Authority and underwritten by the agricultural union, Moshvē Ha-Negev, was directed by T. Erickson-Gini, with the assistance of B.J. Dolinka (University of Liverpool) and L. Shilov (area supervision), H. Lavi (administration), V. Essman, V. Pirsky and T. Kornfeld (surveying), L. Kupershmidt (metallurgical laboratory), O. Shorr (pottery restoration), C. Amit (studio photography), I. Lidski-Reznikov (drawing), D.T. Ariel (numismatics) and workmen from Rahat. Sonia Itkis (Ben-Gurion University, Be’er Sheva‘) conducted a magnetic prospecting survey prior to the excavation and D. Gazit provided helpful insights and information to the excavation team.
The ancient site of Horbat Be’er Shema‘ is located on a sandy plain in the Western Negev, 4 km east of Nahal Besor and 20 km west of Be’er Sheba‘. The site was first mapped and correctly identified as the ancient village of Birsama (Be’er Shema‘) by Alois Musil (1908. Arabia Petraea II: Edom – Topografischer Reisebericht. Vienna. Pp. 61–63). Early researchers noted the presence of a church, a large reservoir, and a fort with four corner towers. The fort has never been excavated. The site was documented in the survey map of the region (D. Gazit. 1996. Map of Urim . Jerusalem, Site 160). A church with mosaic floors and Greek inscriptions, southeast of the present excavation, was excavated in 1989 and 1990 (ESI 10:43–45).
The town of Birsama (Be’er Shema‘) first appears in ancient sources in the mid-second century CE in Claudius Ptolemy's Geography (5.16.10). Birsama later appears in the Notitia Dignitatum (late fourth–early fifth centuries CE), where it is described as a provincial military base with a castellum garrisoned by the Equites Thamudeni Illyriciani (Or. 34.10, 22). It also appears in the Codex Theodosianus (438 CE), where it is referred to as Castellum Versaminum. According to Sozomen's Ecclesiastical History (440–443 CE), a monastic community was present at the site, which he refers to as 'Geraris' (a name assigned to the ducal region in which it was situated). In Georgios Kyprios' Descriptio Orbis Romani (c. 600 CE), Birsama is listed as a regional administrative center for the territory of Gerar. Inscriptions discovered during the excavation of the church refer to one Helladios (Helladius) of the see of Gerara.
Nine areas (A–H and N), primarily oriented north–south, were excavated on the eastern edge of a plowed field in the western half of the site.
The excavation uncovered industrial installations, dating to the Late Byzantine period (sixth–seventh centuries CE), including a large winepress and storage facilities in Areas A, C and H. The winepress in Area C is a ‘four-square’ type that consists of a treading floor, storage compartments, a settling pool and collection vats in a square compound (Figs. 1–3).
Large quantities of Late Byzantine bag-shaped jars were revealed in a building excavated in Area A. An underground room, possibly created for use as a wine cellar, was excavated in Area H (Fig. 4). A number of nearly complete ceramic vessels, including an imported plate, flasks, jugs and deep bowls were discovered in the cellar and apparently had been discarded inside (Fig. 5). Each structure displayed a combination of construction elements that included well-cut hard limestone building blocks and floor slabs, medium-sized wadi cobbles and also walls that consisted of mud bricks or slurry. The walls and floors of all of the structures were heavily robbed out in the first half of the twentieth century.
Early Islamic (eighth century CE) and Mamluk (thirteenth–fourteenth centuries CE) remains were discovered in Area F. A number of small finds from the Early Islamic period included a glass cylindrical flask (Fig. 6), two metal weights (Fig. 7) and an incised potsherd (Fig. 8). Another sherd was inscribed with the name of Allah, in light blue paint.
Structures dating to the British Mandate era were discovered in Areas D, E and N, whereas no architectural remains were found in Areas B and G. Modern Black Gaza Ware jars were uncovered on the floors of the structure in Area N and the remains of a wooden box with metal fixtures (‘Persian wheel’; Fig. 9), used for raising water from a nearby well during the British Mandate era, was discovered on the surface of the site.
The excavation revealed multiple occupations, spanning the Late Byzantine (fifth–early seventh centuries CE) through the Early Islamic (eighth century CE) and the Mamluk (thirteenth–fourteenth centuries CE) periods, as well as the first half of the twentieth century CE (Late Ottoman and British Mandate periods). However, the site does not appear to have been continuously occupied and, with the exception of the Early Islamic and Mamluk occupations in Area F, different parts of the site were occupied in different periods.
In the Late Byzantine period, the western perimeter of the site served as an industrial quarter containing a large winepress, a building for storing jars that were apparently made in a nearby kiln, wasters of which were found on the surface of the immediate area, and at least one structure with an underground room or cellar used for storage and possibly fermenting wine. The large quantity of bag-shaped jars discovered in the storerooms suggests that these jars were produced at the site. An equally large number of Gaza wine jars were also found in the excavations. These may have also been produced at the site or nearby. This is one of the southernmost kiln sites for this type of jar located so far in southern Israel.
The 2006 excavations show that the site of ancient Be’er Shema‘ was probably much smaller in size than previously thought. This suggests that in the Byzantine period the site was a moderately sized village, housing a military installation and associated bathhouse, whose remains may have been located in 2006 a few meters east of Area A, and a monastic community that used the local church. Thus, the Byzantine ‘village’ must be located in the unexcavated area (c. 30 dunams), east of the 2006 excavations, south of the fort, north of the Byzantine cemetery, and west of the church.
The reason for the discontinued use of the Late Byzantine installations in the seventh century CE is unclear. This may have been the result of a decline in population after the mid-sixth century CE with the occurrence of the Justinian Plague and the collapse of the local economy, heavily dependent on the production and export of wine, following the Islamic conquest in 637 CE.
A meager occupation took place in the Early Islamic period in new areas of the site and the inhabitants in that period did not make use of the nearby facilities and chose to occupy an area (F) a bit farther to the north. The same area experienced some minor occupation and building activities took place during the Mamluk period. However, there does not appear to have been any Crusader occupation at the site in the intermediate period.