Area A (Fig. 3). An agricultural terrace system and farm installations were exposed, which together formed cultivation plots. A particularly tall, rectangular stone clearance heap (L27; height c. 4 m; Fig. 4) stood out prominently in the area. Its eastern part had collapsed, and therefore, its outline on that side was somewhat unclear. A trial trench excavated in the center of the heap showed that it was built of small to medium-sized fieldstones and quarrying chips, piled on the bedrock (Fig. 5). Retaining walls (W6, W11) that extended to the west were built north and south of the heap. The area delimited by the walls to the west of the heap was leveled using small fieldstones and quarrying chips (L17, L44, L45; Fig. 3: Section 1–1) and was covered with a tamped earth floor (L15), thus creating a work surface. Wall 6 and W11 served as simple retaining walls for agricultural terraces in the area west of the stone clearance heap. The walls were revealed for c. 15 m, and a trial trench revealed that they were dry built utilizing small to medium-sized fieldstones, founded on bedrock. A wall (W30), built of a single row of medium-sized fieldstones founded on top of L45, continued north of and parallel to W6. The wall consisted of three courses and was erected to ease the pressure of the fill inside the stone clearance heap.
The stone clearance heap might have been built in several phases: 1) clearance stones were deposited in the area (L45) between W6 and W11; 2) the pile was made higher by constructing Retaining Wall 30 and fill (L44) was added, adjoining this new wall; 3) W6 was made higher again and the surface was extended to the south (L17); 4) lastly, a tamped earth floor (L15) was installed above the raised portion and the area was turned into a work surface. To the east of the stone heap, stones continued to be stacked to a height of two stories (L27).
A rock-hewn cistern (L5) was discovered in the western part of the work surface, and a settling pit (L4) was located to its southwest. A built channel (L3) incorporated in W6 extended from the southern part of the cistern and was probably used to irrigate the fields south of the cistern (Fig. 6).
Two field walls (W12, W29) that formed three agricultural terraces descending to the east (L13, L14, L26; Fig. 7) abutted W11 from the north. The cultivation plots covered the bedrock and were characterized by fine-grained reddish rendzina fill (depth 0.4–0.5 m).
A wide wall (W33, Figs. 8, 9) founded on bedrock was constructed north of the clearance heap (L27). It was built of two rows of medium-sized to large clearance stones arranged to a height of eight courses and a fill of small fieldstones and gravel between them.
A natural cave (L43; Figs. 10, 11) was exposed beneath another stone clearance heap (L46) located west of Heap 27. It consisted of two elliptical spaces situated in the north (A1) and south (A2), separated by a natural partition wall. The opening of the northern space faced east and the southern space was equipped with a niche that was hewn in its southern wall. It seems that the cave continued eastward, but evidence of rock cutting above an opening in A2 suggests that its ceiling had collapsed or was removed by quarrying. An accumulation of light colored rendzina was revealed inside and in front of the cave (L10, L16, respectively). The ceramic artifacts recovered from the cave’s interior date to the Iron Age IIB (eighth–sixth centuries BCE) and the Roman–Byzantine periods (first–seventh centuries CE).
Area B (Fig. 2). A natural cave (L22) was excavated but yielded no archaeological finds. Two agricultural terrace retaining walls (W23, W24) were also exposed. Wall 23 (Figs. 12, 13) was built of roughly hewn fieldstones of diverse sizes and was founded on bedrock. An excavation square was opened in a section of the wall that incorporated dressed stones. Thin-grained terra rossa fill (L18, L20) was on the bedrock on either side of the wall. Wall 24 (Figs. 14, 15), built of two rows of medium-sized to large fieldstones on bedrock, was exposed c. 40 m east of W23. Thin-grained terra rossa fill (L19) was observed south of the wall and small fieldstone fill (L25), north of the wall. The soil fill covering the wall abutted the top of W23, and therefore, it seems that W24 predated W23.
Two adjoining water cisterns (L31, L32; Fig. 2) in the eastern part of the excavation area were discovered during an antiquities inspection and were documented during the excavation. The cisterns were treated with gray plaster and were bell-shaped. Although the cisterns were not excavated due to the large amount of accumulated refuse and water that had collected in them, it seems that they were built during the twentieth century. The pottery sherds recovered from the soil fill that abutted the cisterns’ walls represent periods when human activity occurred in the region; however, this data is insufficient to determine the date of the construction of the agricultural terraces. A bronze stamp engraved with the Arabic inscription “Sasson Mordokh 1928” is an unusual find from the area (Fig. 16).
Area C (Fig. 17). A shallow quarry (L35) in the layer of nari bedrock was excavated (Fig. 18). It was possible to reconstruct the size of the stones produced based on the severance channels and the negatives of stones visible on the bedrock (0.6 × 0.8 m, height 0.2 m). The quarry was covered with a layer of alluvium, devoid of datable finds.
A simple rock-hewn winepress (Fig. 19) exposed west of the quarry consisted of a treading floor (L38) with a shallow central channel that drained the liquid into a bell-shaped collecting vat (L37) with a shallow rock-cut sump in its floor. Remains of thick gray plaster (2–3 cm) were preserved on the sides and floor of the vat. In the eastern wall of the treading floor were two hewn niches for securing beam presses (L40, L41), used to squeeze the baskets containing the grapes. An elliptical rock-cutting (L42) in the western wall of the treading floor was probably an unsuccessful quarrying attempt. The winepress was covered with a layer of accumulated alluvium that contained fragments of pottery vessels from the Iron Age IIB, as well as sherds dating to the Hellenistic and Roman periods (second century BCE–fourth century CE).
In a later period, a stone quarry was opened in the area to exploit the nari layer that was suitable for producing building stones. The quarrying damaged the walls of the winepress, thereby negating its use: quarrying marks indicating the production of a single building stone (1.0 × 1.7 m, height 0.4 m) were discovered in the northern wall of the treading floor. The quarry apparently continued north beyond the limits of the excavation.
Pottery sherds dating to the Iron Age IIB were found in the three excavation areas: bowls (Fig. 20:1, 2), kraters (Fig. 20:3–5), a holemouth jar (Fig. 20:6), a jar (Fig. 20:7) and a lamp with a pinched spout (Fig. 20:8).
The archaeological finds revealed during this excavation join those of previous excavations in the area, which exposed the southwestern part of Jerusalem’s agricultural hinterland. The two caves were either devoid of finds (L22) or yielded meager pottery sherds (L43) that originated from erosion; thus, it was impossible to ascribe the finds to the time of the caves’ use or determine the nature of their use. The proximity of the caves to an agricultural area may indicate they were related to farming activity, such as storing tools or agricultural crops and lodging.
It is difficult to date agricultural terraces using material finds because there is no way of knowing the origin of the soil fill that abuts the terrace wall: was it brought from elsewhere or was the fill in situ
and cut and altered by the builders of the agricultural terrace? The pottery assemblage in the excavation may represent a period that predates or postdates the actual construction of the terrace (Davidowich et al. 2011
Like the agricultural terraces, the winepress and the quarry are difficult to date. The finds in the accumulation layer that covered them are insufficient to determine an absolute date for their construction and the time they were used. However, the finds can indicate periods of human activity in the area and hence, we can indirectly deduce the periods of use—from the Iron Age IIB to the Byzantine period.
The archaeological finds and historical sources indicate that settlement in the village of ‘En Kerem was renewed from the Crusader period (twelfth century CE onward), and from the Ottoman period, when there were two Arab villages in the area: Walaja, followed by Moshav Aminadav, and Jura, on which Moshav Ora was established.