During October 2004 a salvage excavation was conducted in Gush Halav (Permit No. A-4257; map ref. NIG 24205/77020; OIG 19205/27020; HA-ESI 118), prior to construction. The excavation, on behalf of the Antiquities Authority and financed by A. Ashak, was directed by M. Hartal, assisted by Y. Ya‘aqoby (administration), V. Essman and V. Pirsky (surveying) and laborers from Kafr Manda and Maghar.
The excavation took place in a lot slated for construction in the high part of the village, at the foot of the school. Three contiguous buildings on the lot were marked on a map dating to 1945. The two eastern buildings were destroyed many years ago and the western building was demolished to allow the construction of the new building. Two excavation squares (50 sq m) were opened. Three strata were discerned, including rock-hewn installations and wall sections dating to the Roman period (second–fourth centuries CE). An installation, dating to the thirteenth century CE and containing cooking vessels, was also discovered.
Partial walls of modern houses (W1, W3; Fig. 1) that were apparently built in the twentieth century, some overlaying walls from the Roman period (Stratum 3), were found.
A single installation (0.8 × 1.0 m) that was set in the corner of two walls from Stratum 3 was attributed to this stratum. It was delimited in the west and north by a wall, one stone wide, which was preserved a single course high. It contained a group of complete pottery vessels, including an intact juglet (Fig. 2). Most of the vessels were handmade and decorated with painted patterns. A glazed cooking pot, glazed bowls and a lamp were found as well, but no imported wares. The vessels dated the installation to the thirteenth century CE. Fragmentary pottery vessels from the thirteenth century CE were detected in the accumulation within a rock-hewn pit (L108) in the southwestern corner of the excavation and in the upper layers of the accumulation in a pit (L113), in the northwestern part of the excavation.
Sections of walls and remains of a rock-hewn installation were found. A pool (L119; length 2.75 m) in the eastern square was excavated to a width of 1.1 m, since its eastern part was beyond the limits of the excavation. The pool was delimited by three walls, each one stone wide. Its floor and the lower part of the walls were coated with a thick layer of gray hydraulic plaster. The foundations of W1 (Stratum 1) penetrated down into the southern part of pool and in one spot, damaged the plastered floor (L107). The ceramic finds from the pool were mixed and dated to the Iron Age and the third century CE.
To the west of the pool were three rock-hewn installations (Loci 115, 120, 122; Fig. 3). The southeastern part of Installation 115 (depth 0.37 m), whose vertical walls and straight floor were carefully hewn, was exposed below the installation from Stratum 2 and contained potsherds from the third and fourth centuries CE. Next to its southern wall was a hewn settling pit (0.46 × 0.46 m, depth 0.36 m) that had a hewn cupmark (diam. 0.2 m, depth 7 cm) in its southeastern corner.
The second installation (L122; 0.55 × 1.40 m) was 0.7 m to the south of Installation 115. A hewn step (0.34 m long, 5 cm deep) to its east was overlaid with fragments of a Galilee Coarse Ware pithos. West of it was a hewn pit (0.55 × 0.63 m, depth 0.3 m) in whose northwestern wall was the bottom part of the same pithos type (diam. 0.25 m). West of the pit was a bedrock surface (length 0.45 m), covered with pottery fragments that belonged to jars, cooking pots and juglets. The finds dated the installation to the second century CE.
Installation 120, next to and south of Installation 122, was rectangular (length 2.2 m, width 0.65 m) and only its eastern part was excavated. In its northern wall was an entrance to a hewn cave (not excavated), above which Installation 122 was built. The cave entrance was blocked by a large stone doorjamb. Two thick walls (width: W6––0.75 m; W7––1.1 m) were built on bedrock between the rock-hewn installations.
Three rock-hewn installations (L108, L123 and L113) were discerned in the western square. The first, Installation 108, was a rock-cut pit in the southwestern corner, whose usage and date of hewing are unclear in the absence of finds. It was filled and blocked with stones and soil that included potsherds from the thirteenth century CE.
The second, Installation L123 (width 1 m, min. length 1.6 m, depth 0.65 m), whose southern end was not excavated, was in the eastern part of the square. Its floor and walls were straight and the finds it contained dated to the Iron Age. Its use, however, is unclear.
The third installation, west of Installation 123, was Pit 113 (1.1 × 1.4 m) whose upper part was hewn in cracked chalk bedrock. A retaining wall (1.1 m high; Fig. 4) was built in its southern side and in the north was an entry corridor to the pit (min. height 2.6 m), built of ashlar stones and covered with a vault (length 0.9 m, width 0.7 m; Fig. 5). At a depth of 1 m, bedrock turned into a stable uniform chalk and the pit was formed into a semi-elliptical shape (length 4.1 m, width 1.1 m). The pit was excavated to a depth of 3.2 m, although its base was not exposed. From a depth of 1.6 m the corridor’s opening was blocked by the wall, built of fieldstones and roughly hewn stones (Figs. 5, 6). An ashlar pavement was revealed at a depth of 1.7 m in the corridor and its excavation was suspended. North of the corridor was a chamber full of soft soil that was not excavated out of fear that the road above it would collapse. The pit and the corridor were filled with soil, stones and large quantities of potsherds. Vessels from the thirteenth century CE were predominant in the upper stratum (0.4 m). Below them, throughout the entire depth of the pit, were mostly vessels from the third and fourth centuries CE, and at a depth of 2.9 m the fill in the pit changed to debris from a chalk quarry, which contained vessel fragments from the third and fourth centuries CE. The walls of the pit in this phase were coated with gray plaster, characteristic of water installations. The pit is similar in shape to a water cistern but the corridor, extending the entire height of the northern wall, casts doubt on this assumption. The pit was probably hewn along a bedrock fissure, into which the corridor was built and was closed for most of its height by the blocking wall. This wall, however, faced north and no traces of plaster were noted on its southern side. Close to the top of the pit’s southern retaining wall was a small section of a beaten-earth floor (L103), overlaid with vessels from the third century CE (see Fig. 4). The floor level was higher than the walls of the pit and it probably covered the pit. No evidence for the roof of the pit was found; it was probably hewn into the floor, hence it could not predate the third century CE.