Agricultural Terrace Retaining Wall (Area A). A wall (W1; exposed length 6.5 m, width 0.3–0.4 m, max. height 0.4 m; Fig. 3), oriented northwest–southeast and built of various-sized fieldstones was exposed, founded on soil that had accumulated above the bedrock. The wall apparently stretched in either direction beyond the limits of the excavation. It is possible that W1 supported an agricultural terrace that extended west, on a slope descending northeast, toward Nahal Raqqafot. Dark brown soil fill (thickness c. 0.6 m) overlying the bedrock east (L106, L107) and west (L105) of the wall was unearthed. Two additional sections of walls (W2, W3), also founded on soil that had accumulated above bedrock, were discovered east of W1. Wall 2 (exposed length 1.5 m) was built of a single course of fieldstones and may have continued east, beyond the area. Wall 3 was built of worked stones, and was apparently used to support and reinforce W1. Collapsed stones that probably originated from the nearby walls were discovered east of W1 and above Fill 107.
Ceramic finds (see below, Dolinka) dating to the Iron Age II (Fig. 4:1–4) and the Hellenistic period (most of the artifacts; Fig. 4:5–12) were discovered in the soil fills (L105–L107) and stone collapse. A single fragment of a bowl from the Early Roman period (Fig. 4:13) was found in the soil that had accumulated above the top of W1 (L103). Apparently, the finds from the Iron Age were swept into the area prior to the construction of the wall. The single bowl fragment from the Early Roman period was not found in a stratigraphic context. The ceramic artifacts ascribed to the Hellenistic period were the latest to be found in the fills that abutted the wall, and they therefore seem to date the time of its construction.
 
Olive-Press Cave (Area C; Fig. 5). A rock-hewn oil press was found inside a karstic cave (max. dimensions 6 × 9 m, height 2.3 m), which was open prior to the excavation, and over the years, was penetrated by alluvium and with contemporary artifacts. The floor of the cave had been leveled in preparation for the olive press. Four adjacent stone paving slabs of various sizes were discovered on the floor of the western part of the cave and were apparently part of the cave’s floor. Two oil production installations were exposed in the cave: a crushing installation and a pressing installation (the oil press terminology is based on R. Frankel [1984] and see also Ben-David 1998; Porat, Frankel and Getzov 2012).
The crushing installation was in the northwestern part of the cave (Fig. 6) and included a crushing basin (yam) and a crushing stone (memmel). The crushing basin, discovered in situ, was conical (upper diam. 1.8 m, lower diam. 1.4 m, height 0.8 m) and hewn from a single boulder; hewn in the upper part of the basin were raised margins (width 0.2 m, height 0.10–0.15 m) and in the basin’s center was a hewn circular depression (upper diam. 0.2 m, lower diam. 0.15 m, depth 0.1 m) within a sunken rectangular frame. The round crushing stone (diam. 1.00–1.05 m, thickness 0.3 m) was also hewn from a single block of stone. It was discovered c. 1 m southwest of the crushing basin. A square central hole (c. 0.20 × 0.22 m) hewn in the crushing stone was surrounded by a square depression (0.35 × 0.35 m) that secured a shaft. A circular path (L303; width 1 m) around the crushing basin was intended for an animal or a person to walk, pushing the shaft that rotated the crushing stone inside the crushing basin.
The pressing installation was probably operated by two methods. The first method used a cylindrical weight, a beam and a screw, or a cylindrical weight, a press beam and a screw (Frankel 1999:107–118; Magen 2008:314). This installation most likely stood c. 2 m south of the crushing installation and two of its components were preserved: a niche for securing a wooden beam (L302; Fig. 7) and a hollow for the weight (L305). Niche 302 (c. 1.0 × 1.3 m, depth 0.5 m) was hewn in the southern side of the cave’s wall. It was meant to secure the end of a wooden beam, at the other end of which a stone weight was attached. Hollow 305 (diam. 1.5–1.6 m, depth 0.35 m) was intended to accommodate a stone weight that was tied to the wooden beam. Neither the pressing floor nor the collecting vat from this phase were preserved. The weight, probably cylindrical, and the wooden beam, were not preserved. Hollow 305, the pressing floor and Niche 302 were all hewn along the same axis. At a later stage, the pressing installation was converted to a direct-pressure screw system (Frankel 1999:130–131; Magen 2008:323). The pressing floor (diam. 1.2 m; Fig. 8) and Collecting Vat 304, which belong to this installation, were preserved. The pressing floor was concave and hewn in the cave’s floor; channels (width 5–7 cm, depth 5–7 cm) hewn in the floor conveyed the oil to Collecting Vat 304. Pits (L306, L307; 0.35 × 0.50 m, depth 0.5 m) hewn on either side of the floor were used to secure wooden beams or stone piers that supported and stabilized the ‘aqalim (the baskets used to hold the crushed olives). The hewn collecting vat (diam. 0.8 m, depth 1.15 m; Fig. 9) was treated with gray plaster that included fine-grain grog.
Artifacts discovered beneath alluvium on the cave floor included fragments of pottery vessels from different periods (see below, Dolinka): Early Roman (Fig. 10:1, 2), Byzantine (Fig. 10:3), Early Islamic (Fig. 10:4–6) and the Middle Ages (Fig. 10:7–9); several fragments of glassware, apparently part of a bottle characteristic of the Umayyad period that continued in use for a short time in the Abbasid period (Fig. 11; Gorin-Rosen, pers. comm.); a few unidentified metal items (not drawn); and a fals from the time of Noor a-Din Mahmud that was struck in the mint in Damascus (1146–1174 CE; IAA 143647). The finds in the cave were not discovered in a clear stratigraphic context.
 
The agricultural terrace retaining wall was probably built in the Hellenistic period. The Iron II finds discovered in the excavation of the wall were apparently swept into the area and indicate that a settlement from this period, yet undiscovered, was located nearby. This terrace was part of an extensive system of agricultural terraces on the adjacent slopes that probably dates from as early as the Hellenistic period. Olive trees and vineyards—the two basic economic branches of ancient agriculture—were grown on these terraces.
The press components that survived in the cave cannot assist in dating the oil press because olive-press installations of this type were used almost unchanged from the Hellenistic period until the present day (Frankel 1984:122; Ben David 1998:47). The two installation types discovered in the cave were common throughout the country in the Byzantine and Early Islamic periods (Frankel 1986:156) and are prevalent at sites where settlement continued from the Byzantine to the Early Islamic periods in Israel and abroad (Magen 2008:314–329). Although the ceramic finds on the floor of the oil press were not revealed in a stratigraphic context, they seem to date the periods of activity in the cave. The pottery dating to the Middle Ages apparently indicates a later phase of use in the oil press. The well-preserved installations from the early phase of the oil press were probably utilized when the pressing installation was converted to a direct-pressure screw-type facility. The oil press, which was part of the economy of mainly olive trees and grapevines, joins another oil press previously exposed in the nearby neighborhood of Bet Ha-Kerem, west of Nahal Raqqafot (Billig 2007).
 
The Pottery
Benjamin J. Dolinka
 
The excavations at Nahal Raqqafot revealed a small corpus of ceramic material dating from the Iron Age to the Middle Ages. The assemblage consists of bowls, cooking pots, jars, jugs and lamps. The materials originated from two separate contexts: Area A, which consisted of a terrace wall (Fig. 4), and Area C, an olive press cave (Fig. 10).
 
A small but noteworthy quantity of Iron Age II pottery was recovered from Area A, including well-known bowl forms (Fig. 4:1, 2), a holemouth jar (Fig. 4:3) and a jug with at least one handle (Fig. 4:4). Most of the ceramics from Area A dated to the late Hellenistic period. Among them was a large selection of storage jars (Fig. 4:5–12), including a handle decorated with stamped pentagram (Fig. 4:7). A thin-walled finely levigated bowl dating from the Early Roman period (Fig. 4:13) was also recovered from Area A.
Much of the ceramics found in Area C were of a later date than those in Area A. The earliest sherds were an Early Roman fine-ware bowl (Fig. 10:1) and a jug (Fig. 10:2). Pottery from the later periods in the Area C assemblage included a Byzantine storage jar (Fig. 10:3) dating from the fifth–sixth centuries CE; Early Islamic wares such as a late version of the so-called Fine Byzantine Ware bowl (Fig. 10:4), a flat base from a buff-ware Abbasid jug (Fig. 10:5) and a lamp (Fig. 10:6) dating from the mid-eighth to the tenth centuries CE; a slip-painted bowl with yellow glaze (Fig. 10:7), known from local Crusader/Ayyubid assemblages and dating from the mid-twelfth to the mid-thirteenth centuries CE; and a Mamluk handmade geometric-painted bowl (Fig. 10:8) and a cooking pot (Fig. 10:9), both dating to the fourteenth–fifteenth centuries CE.