The excavation (50 sq m) uncovered remains of walls (Figs. 2, 3) near the northern bank of Nahal Keziv, c. 300 m west of Tel ‘Avdon. Based on the pottery and flint finds, the walls were dated to the Early Bronze IA. The area where the excavation was conducted was surveyed in the past and no antiquities were discovered (Frankel and Getzov 1997); prior to the current excavation, no sites from the Early Bronze IA were known in the vicinity. To the east of the current excavation area, near Tel ‘Avdon, stone quarries, hewn burial caves, cupmarks and a field wall were discovered in the past; they were probably used by the settlement that existed at Tel ‘Avdon in the Roman and Byzantine periods (Lerer 2011).
 
Six sections of walls were uncovered (W108–W111, W116, W118), some built of two rows of stones and some of a single row of stones. They do not form a coherent plan. The ceramic finds from the excavation include bowls with a rope-decorated band (Fig. 4:1, 2), a carinated bowl (Fig. 4:3), gray burnished-ware bowls (Fig. 4:4, 5), a holemouth jar with an applied rope decoration (Fig. 4:6), a jar made of light-colored fabric with inclusions (Fig. 4:7) and straight-rimmed jars bearing rope and reed decorations (Fig. 4:8, 9). Similar vessels were found in the past in Stratum II at Yiftah’el (Braun 1997) and in the earliest phase of Period A at Tel Bet Yerah (Greenberg, Rotem and Paz 2013), where they were dated to the Early Bronze IA.
 
Flint Finds
Masha Krakovsky
 

The excavation yielded 280 flint items (Table 1) from a variety of industries and in different states of preservation, most of them dating from the Early Bronze I and II and a few from the Middle Paleolithic period. The items were produced from three types of raw material: two types of good-quality, dark brown or brownish beige flint, and limestone (six items). Thirty-two of the items (11%) are burnt. Sixteen of the items (6%) are poorly preserved and coated in patina, with signs of wear and blunt edges. These finds cannot be dated and their poorly preserved state indicates that the assemblage, or at least part of it, was not found in situ.

All the cores except two have a single striking platform (N=10). One of the cores has two adjacent striking platforms. The cores were made from small nodules of flint, some of which were almost exhausted (length c. 0.2 m) while some were discarded because of hinge fractures. About half of the cores were used to produce bladelets and small flakes (N=5), others were used to produce bladelets (N=3) or small flakes (N=3) in their final knapping stages before they were discarded. Similar cores have been discovered at sites from the Early Bronze I and II (Rosen 1997:106–108, Figs. 3.9, 3.24, 3.43). A Levallois core from the Middle Paleolithic period (250,000–50,000 years BP) that was also discovered was made of a large limestone pebble and was used to produce a series of preferred flakes using the recurrent centripetal Levallois mode of core exploitation.

The core-production debitage (c. 9% of the flint finds) includes naturally backed pieces (NBP; Bordes 1961), core trimming elements (CTE), a core tablet (CT) and ridge blades (RB). The type of debitage indicates a local ad hoc industry for the production of flakes (NBP and CTE) and blades (CT and RB), together with the almost complete absence of small knapping debitage (chips and chunks) and the presence of intrusive and worn items bearing patina, showing that some of the items were swept to the site. One of the items with a natural back was produced using the unidirectional Levallois technique and has a faceted striking platform (Fig. 5:1).

The tools include four blades produced by the Canaanite method known from the Early Bronze I and II (Rosen 1997:44–60), including a blade whose proximal end is broken and has not been retouched, a blade retouched on both edges and truncated at both ends (Fig. 5:2) and two sickle blades, one broken at both ends, and the other burnt, broken at one end and truncated at the other (Fig. 5:3). Other tools include a blade-shaped on a broken flake using a method that is not Canaanite, with a double truncation and a retouch on both edges; a steep scraper made on a core-trimming flake; an awl made on a transverse flake; a burin on a retouch made on a large, thick flake; and five retouched flakes, including two primary flakes with two or three scars on the ventral face where the percussion bulb was removed. A bladelet was also found that was broken at one end and truncated at the other and had a retouch on both edges, and a steep scraper that has a little cortex on the dorsal side and was made on a thick, short blade (2.6 × 7.0 cm), retouched around the outside and not on the ventral face. These two tools, especially the scraper, resemble Chalcolithic tools in the style of retouch and the flakes on which they were made (Rosen 1997:93–98), but they cannot be dated with any certainty to this period and may also appear in Early Bronze Age assemblages.

The flint assemblage is not chronologically homogeneous; it contains finds from two periods, namely the Middle Paleolithic and the Early Bronze I and II. The mixture of periods and the poorly preserved state of some of the items show that at least some of the assemblage was not found in situ. Nevertheless, the Canaanite blades and the tools made from them date from the Early Bronze Age, as does the pottery retrieved from the excavation. The Middle Paleolithic flints, especially the well-preserved one with a natural back (see Fig. 5:1), indicate the presence of a Mousterian site nearby. No Middle Paleolithic sites are known in the region, but a previous survey retrieved flint items from this period on the surface, some of which are in fresh condition (Frankel and Getzov 1997: Sites 82.2, 173.2, 190.2). Key sites from the Middle Paleolithic are known to the north (Hayonim Cave in Upper Galilee), the east (Nahal ‘Amud Cave and Qafzeh Cave in Lower Galilee) and the southwest (the Carmel Caves). In Eastern Galilee, several flint-quarrying and knapping sites discovered in the past were used in the Lower and Middle Paleolithic periods as well as in later periods (Neolithic and Chalcolithic; Finkel, Gopher and Barkai 2016).

 
Table 1. Flint Finds
Item
Quantity
%
Primary items (flakes and blades)
82
33.3
Flakes
127
51.6
Blades/bladelets
11
4.5
Canaanite blade
1
0.4
Core trimming element (CTE)
11
4.5
Natural back piece (NBP)
10
4.1
Core tablet (CT)
1
0.4
Ridge blade (RB)
1
0.4
Burin spall (BS)
2
0.8
Total knapping debitage
246
100
Chunks
3
50
Chips
3
50
Total fragments
6
100
Knapping debitage
246
88
Fragments
6
2
Tools
16
6
Cores
12
4
Total
280
100
 
Animal Bones
Zohar Turgeman-Yaffe
 
Eight identifiable bones were recovered (appendix). The most common species is sheep/goat (Ovis aries/Capra hircus). Other species found are cattle (Bos taurus) and Greek tortoise (Testudo graeca). The limited assemblage does not permit a full interpretation of the site’s economy, but it can be assumed that it was based mainly on domestic animals and not on hunting, since no game bones were found. The tortoiseshell may not belong to the anthropogenic assemblage and it may have reached the site as a result of a later influx of material or by some other means, not associated with any human activity. The Early Bronze I site of Meʻona, c. 10 km east of the current excavation, yielded a similar bone assemblage, including mostly sheep and cattle as well as one tortoise bone (Horwitz 1996).