In December 2016 and January 2017, a salvage excavation was conducted at Har Karmi, west of Karmi’el (Permit No. A-7818; map coordinates 225200–85/757185–300; Fig. 1), prior to construction. The excavation, on behalf of the Israel Antiquities Authority and funded by the Ministry of Construction and Housing, was directed by R. Abu Raya (photography), with the assistance of Y. Yaakobi and B. Zidan (administration), R. Liran (surveying and drafting), A. Shapiro (GPS) and workers from Sha‘ab and Kafr Manda.
Area A, at the top of the mountain’s western slope, yielded three simple and particularly small rock-cut winepresses (1–3), four quarries (1–4) cut into the nari rock, as well as several cupmarks and a wide depression.
Winepress 1 (Figs. 3, 4) consisted of a rectangular treading floor (1.00 × 1.51 m, depth c. 0.4 m) and a round collecting vat (diam. c. 0.6 m, depth 0.6 m) to its north. The collecting vat was rather distant from the treading floor; the space between them may have been an earlier collecting vat that had been obviated, and the current collecting vat was hewn in its stead.
Winepress 2 (Figs. 3, 5) was the most complex of all the winepresses discovered in the excavation. It had a rectangular treading floor (1.35 × 1.70 m, depth c. 0.4 m), and to its west was a round collecting vat (diam. c. 1 m, depth 0.9 m). Cupmarks were hewn symmetrically around the winepress, along with four square depressions, triangular in section (upper parts c. 0.20 × 0.25 m, average depth 0.25 m). Two of the depressions flanked the treading floor, and the other two flanked the collecting vat. Winepresses of this type—Ta’anakh presses—are known in the north of Israel but are most common in the Jezreel Valley, and are dated to the Middle Bronze Age. Some scholars have suggested that the depressions served to affix wooden beams of a sack-press installation (Covello-Paran, Getzov and Tepper 2020).
Winepress 3 (Fig. 6) was largely destroyed, as only part of a shallow treading floor sloping westward was preserved (1.2 × 1.3 m, max. depth 0.1 m). The collecting vat seems to have been hewn to the west of the treading floor.
Quarry 1 was shallow (0.1–0.4 m; Fig. 7) and utilized a thin layer of rock. A large cave with a fallen roof was found below it (not excavated). The cave was apparently used to quarry chalk.
Quarry 2 was rather small (c. 3.5 × 4.5 m, depth 1.2 m; Figs. 8, 9) and comprised a single quarrying pit. South of the quarry were a depression (L107) and small, rock-cut cupmarks (diam. 0.1 m, depth 0.1 m). A grapevine grew in the depression, suggesting that the cupmarks were used to trellis offshoots.
Quarry 3 (Fig. 10) consisted of two adjacent quarry pits. The northern pit (2.5 × 4.0 m, depth 0.5 m) had one quarrying step, while the southern pit (c. 2 × 3 m, max. depth 1.1 m) had three such steps.
Quarry 4 (c. 5.5 × 7.5 m, max. depth c. 1.1.m; Figs. 6, 11) was the largest of the quarries, with three quarrying steps. Several large undetached stones (the largest stone measured 0.5 × 0.7 m, max. thickness 0.3 m) and severance channels (max. width 10 cm) were found on its western side.
The winepresses were filled with colluvial accumulations containing a few body sherds and handles of jars typical of the Roman and Byzantine periods. Under a layer of colluvium in the quarries was a fill of terra-rossa soil mixed with chips and quarry debris. In Quarry 1 yielded the rim of an ‘Uza jar Type 7, which is typical of the Byzantine period (end of the fourth and early fifth centuries CE; Fig. 12:1; L108, B1009), as well as two glass shards of wineglasses: a thickened, rounded rim (Fig. 12:2; L108, B1010/2) and a hollow ring base (Fig. 12:3; L108, B1010/1). The other quarries yielded only meager potsherds, as did the winepresses. In Depression 107, near Quarry 2, were potsherds belonging to two jars typical of the Roman period (not illus.) and a fragment of a flint chisel typical of the Neolithic period (Fig. 12:4; L107, B1008).
Two adjacent stone quarries (5 and 6) were uncovered at the top of the mountain’s southern slope.
Quarry 5 (c. 3 × 12 m, max. depth 1.6 m; Figs. 13, 14) consisted of four quarrying steps. One stone (c. 0.35 × 0.70 m, thickness c. 0.4 m) was left undetached. Below the lowest quarrying step on the eastern side of the quarry was a hearth (L502) surrounded by medium-sized, hard dolomite fieldstones, a type of stone which is not found in the vicinity. The hearth was full of ash and small body fragments of jars and cooking pots typical of the Byzantine period.
Quarry 6 was square (c. 4.5 × 5.0 m, max. depth c. 0.8 m; Fig. 15) with two quarrying steps.
The rock-cut agricultural installations revealed in the excavation were part of an agricultural zone in the Roman–Byzantine periods. The quarries were used to extract ashlars. The installations and the quarries were apparently associated with a site at the top of a spur, c. 400 m southwest of the excavation (Fig. 1: Unnamed ruin).
Abu-Raya R. and Stern E. 2011. Har Karmi. HA-ESI 123.
Covello-Paran C., Getzov N. and Tepper Y. 2020. The “Ta‘anakh Winepress” Revisited: Further Evidence of the Middle Bronze Age Wine Industry in the Jezreel Valley. In J. Ebeling and P. Guillaume eds. The Woman in the Pith Helmet: A Tribute to Archaeologist Norma Franklin. Atlanta. Pp. 123–147.