Area A. Two half squares (1, 1A; each 3 × 4 m) were opened 50 m apart on an east–west axis, along a flat area north of the mound’s slope; Square 1A was inspected with the aid of a backhoe. The bedrock, which descends to the north as terraces, was exposed in both squares, directly below the surface (depth 0.17 m; Fig. 2). A minute amount of worn potsherds from the Late Roman period and a Hasmonean coin were found in the soil that had accumulated on the bedrock. It seems that this area was situated outside the limits of the settlement throughout its existence; the few potsherds were probably swept over from the nearby tell.
Area B was opened along a road that was breached in the northern fringes of the tell and created a c. 1.5 m high section, which was straightened and cleaned. A small probe (1 × 2 m, depth 0.3–0.4 m) was excavated at the base of the section down to bedrock, where a concentration of potsherds was noted. A half square (3 x 4 m) was excavated c. 1 m east of the probe. Two strata were discerned; the upper stratum (thickness c. 0.5 m) consisted of soil with a few stones and a small amount of potsherds. This was probably a surface layer that had been cultivated over the course of centuries. The lower stratum (thickness c. 1 m) was composed of earth with numerous rock fragments of medium size (0.2–0.3 m) and many potsherds, mainly dating to the third and fourth centuries CE. Sections of two terrace walls were founded on top of this stratum. The walls were built of one row of small fieldstones (width 0.3 m) that survived to c. 0.5 m high (Fig. 3). The lower stratum rested on bedrock, which rises in broad terraces to the east. A large bronze coin that was minted in Banias during the reign of Caracalla (211–217 CE) was discovered on the fractured and crumbling bedrock surface.
Area C. Two half squares (totaling 3 × 9 m) were opened along the bottom third of the slope, c. 10 m south of Area B. Two walls, probably terrace walls, were exposed close to the surface (Fig. 4). One of the walls consisted of two rows of stones and the other was built of a single row of stones. An accumulation of agricultural soil was found behind the walls. The walls were founded on top of inclined strata that sloped steeply to the north (Fig. 4) and contained small stones and earth, as well as a large amount of potsherds, mostly dating to the third and fourth centuries CE and some to earlier periods, Iron Age I and the Hellenistic and Early Roman periods. Many plaster pieces were found in several spots; these were mostly white but some belonged to colorful frescos (Fig. 5). Due to the safety precautions, the excavation only reached a 4 m depth, without reaching bedrock.
Area D (3 × 4 m, depth 2.3 m). This area was opened at the top of the slope, along the presumed boundary of the settlement that existed at the top of the tell. Just below surface level was a layer of soil (thickness 1 m) that contained fragments of coarse pottery vessels; no Kefar Hananya vessels were found in this layer, although they are very common in the rest of the areas. Below it was a layer of hard soil with pieces of plaster and a burnt level. The layer terminated above a building stone, the only one recovered from the excavation.The meager ceramic finds in this stratum dated to the Early Roman period; therefore, this stratum may be ascribed to the occupation of the city by Titus. Due to time constraints, this area was not excavated down to bedrock.
Area E (3 × 4 m, depth 4.5 m). This area, located west of Area C, was dug with the aid of a backhoe. Two sets of inclined strata were exposed. The top of the upper set was located at the surface, although it is clear that these strata had originally rose to a much higher elevation (Fig. 6). The strata were composed of earth and small stones, sorted according to size, from pieces of rock on the bottom to small stones and soil at the top, and contained large fragments of pottery vessels from the third and fourth centuries CE, as well as pieces of fresco. The bottom strata descended rather gently from the southwest to the north and east,whereasthe upper strata descended quite precipitously from the southwestern corner toward the northeast and severed the bottom strata.
Area F (3 x 4 m) was dug with a backhoe on the eastern part of the slope, east of Area C. Layers of rocks and earth that sloped from the northeast to the southwest were found; the top of the layers was on the surface and it is clear that here too, they had originally reached a much higher elevation. At a depth of 4.5 m, a probe (depth 2.7 m) was opened in the southeastern corner of the square and the continuation of the layers was identified, without exposing bedrock. Most of the potsherds in the lower layers were ascribed to the third and fourth centuries CE. These were also recovered from the upper layers, along with earlier pottery vessels whose quantities were larger than in other areas, which dated to the Early and Middle Bronze Ages, the Iron Age and the Hellenistic and Early Roman periods.
The small size of the excavation areas yielded finds that are insufficient to determine the nature of the site in this region. Apart from the later farming terrace retaining walls, no building remains were discovered. A large amount of potsherds was exposed in the inclined layers on the slope; most dated to the third and fourth centuries CE and a smaller amount dated to earlier periods; practically no potsherds later than the fourth century CE were found.
The layers in all the areas on the slope (Areas C–F) are inclined. Nevertheless, it seems that the top of the slope (Area D)—where layers are more compressed than those lower down the slope and a burnt layer from the Early Roman period was exposed—is located on the edge of the tell, perhaps near the fortifications from the time of the revolt. The continuation of the slope, between Areas D and B, consists of steep layers of stones and soil.Since bedrock, which was exposed in Area B on the northern fringes of the slope, was quite horizontal, it seems that the inclined layers accumulated to a height of five meters or more on top of flat ground.It can be concluded that if the tell itself was created on a hill that rose above the surrounding area its original slope was quite steep.However, only further examinations can determine the original topography of the site before the formation of the mound.
The excavation did not supply an archaeological explanation for the formation of the inclined layers. The angle of inclination indicates that the layers were not intended as fill or for leveling the area. Moreover, the large amount of finds in these layers and their stratification, which is sorted according to size, shows that they were a product of natural activity and could not have been created by man.It also seems that the layers were stratified within a relatively short period of time, perhaps due to a landslide, because they are compressed one atop the other and soil did not accumulate between them, which would have happened had the layers remained exposed on the surface over time. Therefore, it is concluded that all the material slid downhill in one or two events and not as a result of a gradual shifting of the ground over the course of years.
An additional excavation in 2008 expanded the size of Area D (HA-ESI 122 A-5471) and aided in understanding the finds in this area and the origins of the inclined layers. It turns out that the Early Roman period stratum in Area D belonged to the upper part of a protective rampart that dated to the time of the revolt against the Romans, as suggested by M. Aviam, following his excavation on the western slope of the tell (ESI 3:37; HA-ESI 109:7*–8*). It therefore seems that the origin of the inclined layers is in the earthworks of the rampart at the top of the tell and in the accumulations of debris that was discarded there over the course of generations after the revolt, in the third and fourth centuries CE. The latest finds in these layers dated to the fourth century CE; therefore, it seems that at least some of the layers’ landslide on the bottom part of the slope—probably the more massive event—occurred during the earthquake of 363 CE. The landslide revealed the layers from the Early Roman period at the top of the slope where they were discovered right below the surface.