During April–May 2007, an excavation was conducted south of Qibbuz Palmahim and c. 700 m north of the fortress at Yavne-Yam (Permit No. A-5086; central map ref. NIG 17190/64860; OIG 12190/14860), prior to the construction of a holiday village and in the wake of discovering antiquities during a preliminary survey by D. Golan. The excavation, undertaken on behalf of the Israel Antiquities Authority and underwritten by the Evlon and Ma‘oz Daniel Companies, was directed by M. Ajami and U. ‘Ad, with the assistance of E. Jakoel and L. Talmi (area supervision), S. Ya‘aqov-Jam (administration), A. Hajian (surveying), L. Yihye and A. Dagot (GPS), T. Sagiv (field photography), the Sky View Company (aerial photography), S. Al-Amla (metal detection), N. Zak and E. Belashov (drafting), C. Amit (studio photography), R. Vinitsky (metallurgical laboratory), P. Gendelman (ceramics), R. Gat (pottery restoration), M. Shuiskaya (find drawing), Y. Gorin-Rosen (glass) and D.T. Ariel (numismatics). Additional assistance was rendered by the project managers on behalf of the contractor and the antiquities inspectors of the IAA Tel Aviv district.
Several excavations had previously been conducted at the site of Yavne-Yam and in its vicinity (Fig. 1). J. Kaplan excavated inside the Yavne-Yam compound and the ramparts that surrounded it at the end of the 1960s (HA 30:16–17; 38:24–25 [Hebrew]) and M. Fisher of Tel Aviv University had excavated the site since 1992 (M. Fisher [ed.] 2005, Yavne, Yavne-Yam and Their Surroundings, Tel Aviv, pp. 173–208). During 1996 and 1997, E. Ayalon and Y. Drey exposed an irrigation system northeast of the compound, which included a saqiye well, a pool and plastered water channels from the Byzantine period (Yavne, Yavne-Yam and Their Surroundings, pp. 229–252). Several dozen rock-hewn and built tombs have been surveyed and excavated, dating from the Chalcolithic until the Byzantine periods; some tombs were located on the same kurkar ridge as Qibbuz Palmahim.
The excavation was carried out in the area between the Middle Bronze II fortification ramparts and the kurkar hill on which Qibbuz Palmahim is located (Fig. 1). Twenty-five excavation squares were opened in four areas (A, B1, B2, C) and twenty-five tombs (T1–T25), some of which included more than one burial, were identified to the west, southwest and east of Area B1, (Fig. 2). Industrial areas, an agricultural area and part of the northern cemetery of Yavne-Yam, which were used in the Late Bronze Age and from the Persian until the Byzantine periods, were exposed in the excavation.
Area A, the southernmost area, was opened at the top of a low kurkar hill. A well-built winepress (Fig. 3), in which two phases of use were discerned, was exposed. The winepress included a square treading surface (3.8 × 4.3 m) whose floor and walls were coated with thick white hydraulic plaster. Two vats were installed adjacent to the western side of the treading floor. The northern was a shallow intermediate vat (1.7 × 2.4 m, depth 0.6 m) that connected to the floor via a gutter and the southern was a deep collecting vat (1.75 × 2.10 m, depth 1.6 m). The two vats were linked by a channel cut in the middle of the wall that separated them. A sump (depth 0.15 m) was hewn in the southeastern corner of the collecting vat’s floor. The floors and sides of the two vats were coated with a thick layer of hydraulic plaster, identical to that on the treading floor. Based on the finds discovered in and around the winepress, its construction should be dated to the Persian or Hellenistic periods; it ceased to be used in the Roman period (first–second centuries CE).
At the foot of the hill, c. 5 m north of the winepress, a square building (3.5 × 3.9 m; Fig. 3) whose entrance was set in the northern wall, was exposed. Its walls were built of roughly hewn, medium-sized stones, with small fieldstones in-between, and it had a tamped-earth floor. Based on the pottery vessels found above the floor of the building and in the foundation trenches of the walls, it seems that the structure was built and used during the Early Roman period (first–second centuries CE), contemporaneous with the later phase of the winepress to its southwest. The building was probably connected to the later activity in the winepress, possibly as a storeroom for materials and tools that were used in the winepress.
An east–west oriented channel (length c. 50 m, width 2.5 m) was dug next to the southern wall of the building. Flanking the trench on both sides was a series of small dirt mounds (average diam. 1.5 m) that were probably created when the trench was dug. Small and medium undressed kurkar stones were scattered on the northern mound near the building. The purpose of the channel is unknown, yet it is clear that it was dug after the building had been constructed and probably after the latter was no longer in use. It seems that in a later phase, after the channel was filled in, the northern mound served as a boundary enclosing an agricultural plot.
Area B1. A built installation was exposed c. 100 m north of the winepress in Area A. Although the installation was poorly preserved, two operating phases could still be discerned (Figs. 4, 5). In the first phase, the installation included a square plastered surface (3.8 [?] × 4.1 m), aligned north–south and divided into two (L241, L286) by a wall (W35) whose only southern part survived. In the second phase, W35 was dismantled and its foundation trench was blocked by medium and large fieldstones (L214). A square vat (L230; 1.4 × 1.4 m, depth 0.6 m) was installed near the center of the installation. It was lined with ashlar stones and light-colored mortar and its floor consisted of wadi pebbles and shells. The western part of the installation was re-paved: atop the earlier plastered surface, a floor of small stones bound with light-colored mortar, which abutted the vat in the middle of the installation, was placed. A lead amulet-pendant, bearing a Greek inscription on both of its sides, was discovered on the occupation level that abutted the southern side of the floor. Based on the few potsherds discovered in the foundation trenches and in the floor beddings of both operation phases, the installation’s use should be dated to the second–fourth centuries CE; its function, however, was not ascertained.
Three meters west of the installation was a heap of fieldstones and broken ashlar stones that contained numerous fragments of pottery vessels, dating to the end of the Roman–beginning of the Byzantine periods (fourth–fifth centuries CE). Most of the vessels were saqiye jars, the likes of which in large numbers were exposed in the excavation of the saqiye well, c. 120 m northwest of the excavation area (Fig. 1; Yavne, Yavne-Yam and Their Surroundings, p. 241, Fig. 14:5). It seems that the heaps of stones were formed when the installation walls were dismantled.
North of the installation was a low kurkar ridge that extended in an east–west direction and had a kind of a step (W23; length c. 0.55 m, height 0.5–0.8 m; Fig. 6); it had been hewn by removing a strip in the northern side of the ridge. A robber’s trench was exposed in the fill above the step and medium and large fieldstone collapse was discovered north of the step. It therefore seems that this step was used as a foundation for a wall, whose stones were mostly robbed. It is possible that the wall served to retain the fill upon which the installation was constructed or to separate between agricultural plots, such as the wall exposed in Area A.
Area B2. Two adjacent winepresses, oriented north–south, were exposed c. 40 m north of the installation in Area B1 (Fig. 7). The southern winepress, which was the larger of the two, had a square treading floor (2.8 × 3.2 m); a shallow intermediate vat (0.8 × 1.0 m, depth 0.3 m) installed west of the treading floor’s southwestern corner and a collecting vat (1.5 × 2.9 m, depth 1.2 m) with a sump in its northeastern corner that was installed north of the intermediate vat and west of the treading floor. The treading floor and the intermediate vat were connected by a perforated hole in the wall between them and the two vats were connected by a channel that was cut in the wall separating between them. The sides and floors of the treading floor and the vats were coated with a thick layer of white hydraulic plaster.
At least two operation phases were discerned in the northern winepress. The plan of the winepress in the early phase was almost a mirror image of the southern winepress, except for its smaller size. It consisted of a square treading floor (2 × 2 m), a shallow intermediate vat (0.8 × 0.9 m, depth 0.2 m) and an adjacent collecting vat (1 × 1 m, depth 1 m). In the late phase, the intermediate vat and the collecting vat were filled with brown soil that contained a few stones and plaster. At the same time, medium-sized fieldstones and light-colored lime-based mortar were placed over the treading floor and the walls, serving as a foundation for a layer of wadi pebbles and small fieldstones that was covered with a layer of light plaster. Based on the ceramic finds in the foundation trenches of the winepresses, between the stones in the walls and in the fill that was deposited in the vats of the southern winepress after it was no longer in use, it seems that the winepresses were built in the Persian or Hellenistic periods and ceased to be used in the Roman period.
A firebox (?) was exposed some 20 m east of the winepresses. It was installed in the bottom of a circular kiln (diam. 2 m; preserved height 0.3 m) that was dug into the hamra soil. Based on the pottery vessels recovered from the fill in the kiln and its vicinity, it seems that the installation was used during the Persian period.
Area C. Several concentrations of fieldstones and small pebbles (length c. 0.1 m) that formed irregular-shaped surfaces (max. diam. 0.8 m; Fig. 8) were exposed close to the surface, northeast of Areas B1 and B2. The remains of a light gray material were discovered between the stones; however, no datable artifacts were found in or around them and the time and purpose of these surfaces are unclear. A square pit (0.5 × 0.5 m, depth 0.2 m), dug in the hamra soil and filled with shells and a sandy fill (L321; Fig. 9), was discovered east of the stone concentrations. The sandy fill contained a small amount of potsherds that dated to the Roman period. The shells in the pit included murex shells and were probably meant to be burnt and used as a component in the production of mortar and plaster for the winepresses.
Tombs. Five Late Bronze Age tombs were exposed west and southwest of Area B1, at the exterior foot of the eastern rampart (Tombs 21–25; see Fig. 2). These pit graves and cist tombs were covered with roughly dressed kurkar slabs, placed the length and width of the tomb (Fig. 10). The date of the tombs was based on their shape, location—outside the enclosure delimited by the rampart—and the potsherds recovered from the fill in their vicinity; numerous tombs from this period had previously been exposed in the region (Yavne, Yavne-Yam and Their Surroundings, pp. 15–32).
Twenty cist tombs (T1–T20; average dimensions 0.6 × 1.9 m; see Fig. 2) were identified east of Area B1, but it is clear that other graves exist in this area. The tombs, dug into the hamra soil, were lined with carefully dressed rectangular kurkar slabs; in most of the tombs the long sides were formed by two vertical slabs standing next to each other and the narrow sides consisted of a single slab. The covering, which was only exposed in some of the tombs, was composed of dressed kurkar slabs that were positioned in a gable shape. It was impossible to determine with certainty the date of tombs since they were not excavated. However, E. Ayalon exposed identical tombs c. 100 m south of here, at the bottom of the southern (inner) slope of the northern Middle Bronze II rampart, which he dated to the Early Roman period (see Fig. 1; Yavne, Yavne-Yam and Their Surroundings, pp. 217–221).
An agricultural area with installations was exposed in the excavation; it was used by the settlement at Yavne-Yam from the Persian (fifth century BCE) until the Byzantine periods. It seems that the agricultural activity in this region began to flourish during the Early Roman period (first–second centuries CE). By the end of the Byzantine period or the beginning of the Early Islamic period (sixth–seventh centuries CE), this area was almost completely abandoned. Furthermore, tombs that dated to the Late Bronze Age and the Early Roman period were discovered. These join the finds from previous excavations and surveys, indicating that this region served for burial over long periods of time.
Four periods of activity were identified in the installation remains:
(1) The Persian and Hellenistic periods (Stratum V).—three winepresses (Areas A, B2), two of which were adjacent to each other (Area B2) and a kiln for firing pottery or metal vessels (Area B2) were built.
(2) The Early Roman period (Stratum IV).—two of the winepresses were re-used after the installations were abandoned for a period of two to three hundred years. The winepress in Area A was expanded and a structure (a storeroom?) was built alongside it; the large winepress in Area B2 was probably renovated and an open surface was installed in place of the small winepress.
(3) The Late Roman period (Stratum III).—the area was adapted for farming after the winepresses were filled in and no longer used. A wall was erected in Area A to mark the boundary of a plot and the building from the Early Roman period was converted for use as a watchman’s hut. A wall was built in Area B1, probably to support an installation that had a plastered surface and in which a small vat was installed in the second phase; the wall may also have been used for a farming terrace that extended southward.
(4) The Byzantine period (Stratum II).—the region continued to be used for farming and at the same time, the building stones from the structure in Area A and the installation in Area B1 were robbed.