In November 2015 and March 2016, two excavation seasons were conducted at Horbat Hanot, east of Bet Shemesh (Permit Nos. A-7533, A-7659; map ref. 204531–45/624369–83; Fig. 1), prior to conservation work at the site. The excavation, on behalf of the Israel Antiquities Authority and funded by Keren Kayemeth LeIsrael, was directed by I. Radashkovsky (field photography), with the assistance of N. Nehama and R. Abu Halaf (administration), B. Touri (safety), A. Hajian, M. Konin and M. Kahan (surveying and drafting), A. Peretz (photography), S. Gendler (metal detection), I. Taxel and P. Gendelman (pottery), I. Lidsky-Reznikov (finds drawing), A. Karasik (digital pottery documentation), C. Amit (finds photography), A. Buchennino (pottery report), N. Zak (map and plans), Y. Nagar (physical anthropology), T. Winter (glass), V. Nosikovsky (metallurgical laboratory), D.T. Ariel and G. Bijovsky (numismatics). Assisting were D. Ben-Ami, P. Betzer, D. Varga, A. Re’em, A. Shadman, Y. Dray, A. Dayan, N. Ben-Ari and B. Arubas (consultation), and the implementation unit of Keren Kayemeth LeIsrael.
Horbat Hanot (Khirbet el-Khân) is located on the northern slope of Mount Hanot that descends moderately down to Nahal Zanoah. The site is situated c. 18 km northeast of Bet Guvrin, along the ancient road that linked Jerusalem and Bet Guvrin/ Eleutheropolis (Taxel 2008:63). The road was probably paved during the suppression of the Bar Kokhba Revolt (132–135 CE), or possibly two years earlier at the time of Hadrian’s visit to the region (Shenhav 2003:269). In a salvage excavation carried out at the site in 1985–1986, remains of a church dating to the Byzantine period were uncovered, overlain by the remains of structures from the Mamluk and Ottoman periods. A multi-colored mosaic floor was uncovered in the church, and a four-meter long Greek inscription recorded that “Under the most pious and God-loving Theodore, priest and hegumen, was done all the work of the addition of the apse and of the painting and of the facing with marble of the end-wall of the presbytery, together with the diaconicon, from the foundations, in the month of April of the 12th indiction” (Shenhav 2003:271). The inscription dates the renovation of the church to the late sixth or early seventh century CE (Taxel 2008:64). In the Survey of the Nes Harim Map conducted in the 1990s, remains of buildings and of the church were documented at the site, some walls preserved to a height of two meters. Cisterns, a winepress, installations and a rock-cut pool were also recorded (Weiss, Zissu and Solimany 2005: Site 129).
In the two current excavation seasons, carried out prior to conservation work at the site, the eastern part of the Byzantine church, overlain by two construction phases from the Mamluk and Ottoman periods, was exposed. About ten meters to the north, an impressive, partly hewn and partly built winepress complex was unearthed, in which three construction phases were observed (Figs. 2–4). Since the church and the winepress were mostly exposed prior to the present excavation, the focus was on the documentation of the architectural elements. Other parts of the church were only drawn schematically (Fig. 2: marked CH and M). The excavation at the site was stopped before the church and the winepress complex were entirely exposed.
The Church (Fig. 2; Fig. 3: Sections 3–3, 5–5)
The Byzantine period. The remains of the nave with an underground crypt (Figs. 5, 6), as well as a diaconicon (northern room) and a prothesis (southern room), were attributed to the Byzantine period. A few of the lower courses of the southern, western and eastern enclosing walls of the nave were exposed (W305, W301, W317); at a later stage, new walls were built on top of the earlier walls (Fig. 3: Section 5–5). The original northern wall of the nave, which apparently lay north of the northern wall of the main hall of the church, was not preserved; a later wall (W300) was subsequently built here, enclosing the nave on the northern side (see below; Fig. 7). The best-preserved courses from the Byzantine period were visible in W305, the eastern wall of the nave. In this wall, three original courses, built of dressed stones with bonding material, and set on a fill of small stones, were preserved. A doorway (L325; Fig. 5) in the center of the southern wall of the nave (W317), led into another, poorly preserved room, located south of the nave. The remains of the springs of two arches were found built against the inner faces of the eastern and western walls of the nave, one arch in the northern part of the nave and the other in the southern part; they apparently separated off the crypt in the center of the nave from the northern and southern rooms (Figs. 5, 8, 9). In the northern part of the nave, a coarse white mosaic set on the bedrock, (L316) was apparently part of the original floor. The northern boundary of the diaconicon is unclear, due to damage in later periods. It is possible that two walls (W36, W37), and a pilaster (L45),uncovered north of the nave, were originally part of the diaconicon. The crypt uncovered in the center of the nave, consists of a rock-cut shaft with eight stairs (L315; Figs. 6, 8) leading down to a burial chamber (L313; Fig. 3: Section 3–3). The stairs were divided into two sets of four stairs, an upper and lower set. A large stone slab partially covered over part of the entrance into the shaft. The burial chamber is almost square (2.4 × 2.6 m; 1.4 m high); marks in the center of its bedrock floor indicate that a coffin had apparently stood here. To the east of the shaft opening, and near W305, a decorated capital from a Byzantine-period column was found (B403; Figs. 10, 11). The capital may have originally adorned Pilaster 45, from which the arch sprang.
In the room south of the nave, part of a mosaic floor (L331), and remains of the mosaic floor’s foundation layer consisting of soil and small stones (L324), was found. The floor was damaged by a bell-shaped cistern (L326) that was cut into it. The cistern opening was covered with a round stone (Fig. 12), in whose surface a small hollow was carved (L327). The cistern was not excavated due to safety considerations.
East of the nave, a wall (W333) and the foundation layer of a floor consisting of soil and small stones (L337) were uncovered. These elements may also have been part of the church.
The Mamluk period. In this period, part of the church building was altered for other uses (Shenhav 2003). A new wall (also W305) was built on top of the eastern wall of the church nave. It seems that part of the exterior and the lower courses of the Byzantine western wall (W301) may also date to the Mamluk period.
The Ottoman period. In this period, new walls apparently reutilized the remains from earlier periods. The northern wall of the nave (W300; Fig. 7) was now built over the Byzantine period mosaic floor (L316). The interior face of the western wall (W301) was also constructed at this time (Shenhav 2003:270). Associated ceramic finds included body sherds of pottery vessels dating to the Ottoman period.
A large pile of debris (L306) was found in the church nave. Pottery fragments were discovered in the heap, dating to the Byzantine, Early Islamic, Crusader and Mamluk periods, as well as human bones. The heap of debris may have been the result of the looting the crypt. The ceramic finds from the Byzantine period include a thin-walled jug with a plain rim (Fig. 13:10) and an imported amphora with a cut rim (Fig. 13:12). The ceramic finds from the Early Islamic period included a glazed bowl with a simple rim, a thin curved wall and a ring base (Fig. 14:1), a glazed bowl with a plain rim and a slightly curving wall near the base (Fig. 14:2), and a glazed, globular cooking pot with a thin wall and a folded-out rim with no neck (Fig. 14:3). The ceramic finds from the Crusader period, included carinated glazed bowls of the slip-painted type (Fig. 14:6, 7) and a jug with a swollen neck (Fig. 14:8). The ceramic finds from the Mamluk period included a handmade bowl with a curved wall, a slightly folded-out rim and a flat base (Fig. 15:1), a carinated bowl with thickened wall and rim (Fig. 15:2), an imported bowl, probably Egyptian, featuring broad-ribbed walls (Fig. 15:3), a jar with a thickened rim and long neck with a ridge at the base of the neck (Fig. 15:4), of a type that perhaps continued in use until the beginning of the Ottoman period, a jug with a slightly folded-out rim and combed strips on the body (Fig. 15:5), and a flask with a thickened rim and a long neck with a mid-neck ridge (Fig. 15:6). The human bones in the refuse heap included a thoracic vertebra, fragments of a tibia and a few hand bones. The bones revealed crushed epiphyses and the ring epiphysis of the vertebra was also crushed (without osteoporosis), findings that are characteristic of an individual more than 20 years old (based on Johnston and Zimmer 1989). The sex of the individual is unclear.
The Winepress Complex (Fig. 2; Fig. 3: Sections 1–1, 2–2, 4–4)
Three construction phases were distinguished in the winepress complex. A small winepress was attributed to the early phase, dated to the Byzantine period. In the middle phase, also dated to the Byzantine period, the winepress was expanded into a complex winepress. It seems that in both these phases, the winepress complex functioned alongside the church. In the third phase, the winepress fell out of use, and was replaced by channels and walls, some apparently dating to the Ottoman period.
The early phase (Fig. 16). In this phase, the winepress was part of the Byzantine church complex, and it comprised a treading floor (L43), a square settling basin (L41; depth 0.45 m) and a square collecting vat (L2), all rock-cut elements. The settling basin was connected to the collecting vat via a circular hole (L55; diam. 0.15 m) cut in the joint wall. The walls of the treading floor (L44) were coated with a thin layer of plaster, and the floor was paved with a mosaic of red and white tesserae (L39). The western and northern walls of the treading floor were partially preserved, while the eastern wall was damaged by the expansion of the winepress. The size and shape of the collecting vat may have changed when the winepress was expanded in the second phase.
Fragments of pottery from the Byzantine and the Crusader periods were found on the treading floor. The finds from the Byzantine period, include a bowl with a thin wall and an folded-out rim (Fig. 13:1), an imported bowl of Type LRC3, with a carinated rim decorated with two deep, rouletted grooves (Fig. 13:2); a krater with an arched rim bearing a wavy, combed decoration (Fig. 13:3), a deep cooking bowl with a straight wall with close ribbing on the outside and a cut rim (Fig. 13:4), a cooking bowl lid (Fig. 13:5), closed cooking pots with upright necks (Fig. 13:6–8), including one with two loop handles pulled from rim to shoulder, and a Gaza Ware jar with a cylindrical body, no neck and lumps of clay on rim and shoulder (Fig. 13:11). A carinated slip-painted glazed bowl with a ring base, dating to the Crusader period, was also unearthed here (Fig. 14:5).
The middle phase. During this phase, also dated within the Byzantine period, the winepress was expanded into a complex winepress (c. 9.0 × 14.5 m; Fig. 17). It now comprised a large, central treading floor (L1), a collecting vat that had apparently been expanded from the early phase (L2) and five surrounding units (CI–CV) that were built around the central treading floor. The central rock-cut treading floor L1 was paved with white industrial mosaic tesserae (1 × 1 cm). The walls were partly preserved, apart from the northern wall which was not extant. In the center of the floor, a square pit was hewn (L3; 0.8 × 0.8 m) to anchor a wooden beam or screw press. At the bottom of Pit 3, a channel (L42; 3.05 m long) led under the floor into collecting Vat 2 (Fig. 18). The vat floor was paved with white mosaic (L54), a small settling basin (L57) was installed in its southwestern corner, and the vat walls were coated with plaster. Two depressions, apparently serving as steps, were hewn in the eastern and northern walls of the vat. In the western wall, below Channel 42, a shelf was installed (L56; 0.3 × 1.6 m), the upper part of which was paved with mosaic; vessels were apparently set on it when the vats filled with the must. In the southern wall of Vat 2, remains of a vault were identified, indicating that at least the southern part of the vat was roofed.
The five Units CI–CV, built around the treading floor, were bounded by walls. They were apparently used as small winepresses, each unit comprising a treading floor and a collecting vat; some of these units did not survive due to later constructions, and there may have been additional units that were not preserved. Similar units around a central treading floor are known from other winepresses, e.g. at Horbat Siv, north of Tulkarm (Dray 2011). The units were built on a leveled rock ledge that was overlain with a foundation of small stones. Mosaic floors of white tesserae (1 × 1 cm) laid on the foundation layer, served as treading floors. The three western units (CII–CIV) were similarly built, and they were better preserved than the others. Units CIII and CIV were built on a lower level than Unit CII. In the two former units, the treading floors, and the rock-hewn collecting vats (L5, L13) that were lined with stones and coated with plaster, were well preserved (Fig. 19). A channel containing a metal pipe (L53) led from the bottom of Vat 5 to the central treading floor (L1). The channel opening was found blocked with a compacted soil fill; it is unclear whether the blockage was intentional, or the result of soil accumulation. Unit CV was poorly preserved. A jug with a folded-out rim (Fig. 13:9) dating to the Byzantine period, was retrieved here on the mosaic floor foundation (L70), and a coin (IAA 156962), dated to the fourth–fifth centuries CE, was discovered on the mosaic floor remains (L71). In this period, a long wall (W11/W72/W75) was built in the southern part of the winepress. A zir jar with a particularly large body, thickened rim and walls and a long neck with a ridge at the base of the neck (Fig. 14:4), dating to the Early Islamic period, was found in the stone remains of W75. South of Unit CI, the remains of a mosaic floor (L38), defined by walls (W50, W51), may have been part of another unit in the winery. An Umayyad post-reform coin (IAA 156963), dating to the eighth century CE, was found in the foundations of this mosaic floor.
The late phase. In this phase, the winepress complex fell out of use, and three rock-cut drainage channels were cut into it; walls post-dating the channels were also found. The drainage channels (L18, L26, L27; Fig. 3: Section 4–4; Fig. 20) were cut into the rock slope, plastered and covered with stones. They may have been associated with a bathhouse that was uncovered nearby in the past (Shenhav 1986:25). A later wall (W30) was built of one row of stones, above and damaging the eastern Unit CV. A coin from the Mamluk period (IAA 156964) was found in W30, possibly dating the wall. Above the walls of the winepress’ western and southern units, four broad walls were built (W4, W10, W21, W80), delimiting rooms whose purpose is unclear. These walls were built of two rows of various-sized dressed stones with bonding material and small stones, and preserved to a height of four to seven courses. Some of the walls were built above the drainage channels, putting them out of use. Based on their construction technique, these walls were dated to the Ottoman period.