Stratigraphy
Four archaeological layers (Fig. 2) were defined. The lowermost Layer 4 (thickness c. 2 m) was excavated in a limited area (1.0 × 2.5 m). The sediment was red brownish clay, which contained few angular stones and small patches of fine gravel horizons. No architecture was discerned, although few stones and mud-brick remains were identified in the northwestern section. Layer 3 was composed of highly dense angular stone fragments, embedded within light gray sediment. This cobble layer (thickness c. 0.8 m) was excavated in 2.5 × 2.5 m area. Remains of what appeared to have been a two-course wall, built of unworked fieldstones (0.3 × 0.3 m), were found in the upper part of the layer. Layer 2 (thickness 0.9 m) was excavated in a somewhat bigger area (2.5 × 5.0 m) and its sediment was loose brown terra rossa soil with a few angular stones. The uppermost Layer 1 (5 × 5 m) was a plowed field (thickness c. 0.7 m—the reach of plow blades).

 

 The Lithic Assemblage

The lithic finds comprised 3476 artifacts from all layers. Layers 3 (N=1258) and 2 (N=1045) indicate a relative abundance of lithics, which is scarce in Layer 4 (N=536). The upper 0.1 m of Layer 3 and lower 0.1 m of Layer 2 were analyzed together (N=391), as they contained a mixture of Yarmukian and Wadi Rabah finds. The surface sample (N=246) was not sieved and therefore, it is small and incomplete.
Layer 4: The assemblage displayed dominance of ad hoc flake technology, evidenced by high frequencies of flakes (54.2%) within the debitage. Very few tools were made on bidirectional blades, characteristic of the PPNB (Fig. 3:1).
The six cores in this layer included an exhausted bidirectional-blade core, two flake cores, a radial and a core on flake, two fragments and a tested nodule—all extremely utilized. It is most likely that the only complete bidirectional-blade core (Fig. 3:2) in the dirt piles originated from this layer.
The tool assemblage included 36 items; the majority was made on flakes (36.1%), blades (27.8%) and primary elements (32%), although some were made on chips (5.6%), chunks (5.6%), and bladelets (2.8%). The most frequent tool type is the retouched flake (N=11) that varied in the retouch type and location. The second most frequent are the burins (N=6), mostly of dihedral type. It is important to note that three were fashioned on bidirectional blades, one of which was made on an arrowhead (see Fig. 3:1). Notches (N=4) are the third frequent type. Sickle blades and scrapers numbered three items each. The sickles included one plain item made on bidirectional blade and two denticulated sickles that most likely infiltrated from Layer 3. The remainder of the tools includes two bifaces, perforator, retouched blade, denticulate, microlith, a truncation and varia. The bifaces include a small polished axe and a trihedral pick. The retouched blade and truncation were fashioned on bidirectional blades and the microlith—on a patinated backed bladelet.
Layer 3: The Layer 3 assemblage displayed dominance of two technologies: ad hoc flakes (47.7%) and unidirectional blades (35.5%). This layer contained 15 cores, including six flakelet cores, two unidirectional blade cores, two bladelet cores, two flake cores (radial) and three fragments. Unidirectional blade cores with a semi-pyramidal shape and single platform bladelet cores made of lustrous flint characterize the Yarmukian lithic assemblage from Sha‘ar Ha-Golan (Barzilai O. and Garfinkel Y. 2006. Bidirectional Blade Technology after the PPNB: New Evidence from Shaar Hagolan, Israel. Neo-Lithics 1/06:27–31).
The tools included 118 items. Technologically, flakes (31.8%) and blades (32.7%) were equally used for tool making. Flakes dominated ad-hoc tools, namely scrapers, retouched pieces, denticulates and notches, while blades (mostly unidirectional) were used for standardized denticulated sickles blades. Primary elements (11.2%) were also used for making ad-hoc tools. Bladelets were occasionally used for making sickles, micro-borers and microliths. The remainder tools were made on chunks (4.7%), chips (1.9%), and a ridge blade (0.9%).
The most frequent tool type is the sickle blade (N=28), composed of three subtypes. The majority are complete, non-truncated and fine-denticulated sickles. Their retouch was made on the dorsal face with no backing. The second subtype is classic backed and truncated denticulated sickles, typical to the Yarmukian culture (Fig. 4:1–5). Their notches are bifacially retouched and display greater spacing than the first type. The third are backed and truncated sickles with no denticulation, which most likely penetrated from Layer 2. The second most frequent tool type are the scrapers (N=18), followed by retouched blades (N=15) and retouched pieces (N=10). The bifaces (N=9) comprised four chisels, one of which was polished, two axes, one was polished and the other was transformed into a hammer stone, and four fragments (Fig. 5). Perforators (N=8), denticulates (N=8) and notches (N=7) were equally represented. The perforators comprise three borers, two awls, a drill, a micro-borer and a bec (Fig. 4:8). Burins (N=5) consist of dihedral, transversal and one angled subtypes. Microliths include three items, namely two finely retouched bladelets and a lunette.
Rare but of great importance are the arrowheads (N=2). The first was an Amuq point with a tang fashioned by collateral pressure flaking, typical to the Yarmukian culture (Fig. 4:7). Comparisons to this subtype were found at Sha‘ar Ha-Golan and Munhata, Layer 2 (Gopher A. 1989. The Flint Assemblages of Munhata, Israel [Les Cahiers du Centre de Recherche Francais de Jerusalem 4]. Paris, Fig. 38:6, 8; Barzilai and Garfinkel, ibid., Fig. 3:1, 2). The second, which came from the dirt piles, was a small Nizzanim point (Fig. 4:6). Another diagnostic tool of the Yarmukian culture is the pressure flaked knife, of which two items were recovered from the site. The first was made on a flake and fashioned by pressure retouch on the dorsal face (Fig. 4:10) and the second, which probably belonged to this layer, was also found in the dirt piles (Fig. 4:9).
In addition to the flint, two obsidian flakes were found in this layer, a small fragment and a unidirectional micro blade (Fig. 6).
Layer 2: The lithic assemblage of this layer consisted of 1045 items. The ad hoc flake technology dominates the debitage (52.9%), although unidirectional blade and bladelet technologies are also evident. The 26 cores in this layer include 14 flakelet cores, five flake cores, four bladelets (Fig. 7:1–3), a unidirectional pyramidal blade core and a mixed blade flake core.
The tools include 91 items, the dominant forms of which are the sickle blades (N=18) and scrapers (N=13). The sickles include Wadi Rabah and Yarmukian subtypes. The latter were likely introduced from Layer 3. The Wadi Rabah sickle blades are made on unidirectional blade blanks (Fig. 7:6–8). They bear a rectangular shape that was fashioned by abrupt retouch at one edge and truncations at both ends. The glossed working edge is plain or finely retouched. Other formal tools are the arrowheads (N=4) and bifaces (N=4; Fig. 7:9). The projectiles include a transversal point (Fig. 7:5) and fragments. The remainder of tools include retouched blades (N=9), microliths/bladelets (N=7) and perforators (N=9), which include micro-borers (Fig. 7:4). Ad hoc tools include retouched pieces (N=5), truncations (N=6), denticulates (N=4), notches (N=2), burins (N=3) and varia (N=7).

 

The Ground Stone Tools

Ianir Milevski
 
The stone tools assemblage from the site consisted of 27 items. Most stone tools were found out of context, in fills or topsoil. The larger part of the assemblage was composed of grinding-stone tools (73%).
Lower grinding slabs (N=3). Three fragments of grinding slabs were relatively small movable palettes, made of basalt or hard limestone, with one flat used surface.
Upper grinding stones (N=6). These are relatively small grinding stones, made of limestone or basalt. They are rounded and have one flat working surface grounded or polished. One item is a discoidal/planoconvex hand stone.
Hammer stones (N=3). These are core tools with battering marks all over the surface. Hammer stones are generally rounded and made from hard limestone or quartzite pebbles. It is not clear if they were used as pounders for food processing or for flint knapping. Two of the hammer stones from the upper layers were used and broken, with flaking all over the edges; they may have been reused as scrapers. 
Pestles (N=2). Two broken elongated pestles were unearthed in a PN context, made from basalt and hard limestone.
Polished pebbles (N=5). These water-worn pebbles, with one or a few polished surfaces, came from topsoil. Their function is unclear, but in Predynastic Egyptian burials they were usually associated with palettes and probably served for fine grinding.
Palette. One small rectangular palette (or platter) was retrieved from the PN layer.
Pebble mortar. The small limestone mortar (Fig. 8:1) was probably used for fine or small-scale pounding.
Disks or Stoppers (N=2). These are flat limestone disks, shaped by flaking. Two limestone disks from the PN layer have a lens form. Similar items found at Munhata were called choppers (Gopher A. and Orelle E. 1995. The Ground Stone Assemblages of Munhata, A Neolithc Site in the Jordan Valley, Israel: A Report [Les cahiers de Missions Archéologiques Françaises en Israël 7]. Jerusalem).
Perforated disks (N=2). Two limestone disks, with flattened and smoothed tops and bottoms and a drilled hole in the center, were found (Fig. 8:2). It is generally accepted to regard these disks as loom weights. Similar disks were discovered at the nearby site of Horbat Uza (Getzov N., Lieberman-Wander R., Smithline H. and Syon D. Uza I [IAA Reports, forthcoming]).
Weight(?). One elongated pebble with pecking on both sides seems to be an unfinished weight
Basalt bowl. A fragment of a basalt bowl was found in Layer 2. This is an elaborate thin section of a bowl with straight walls and the beginning of a ridge at the lower part. The inner surface is very polished. Although the fragment is small, it seems to belong to a pedestal bowl, which is common to the Ghassulian Chalcolithic period.
 
All types of ground stones, other than the basalt items, represent a domestic assemblage that could be manufactured locally. Most of the items came from topsoil and from Layer 1 (N=14; 48%). The Wadi Rabah Layer 2 contained a stone bowl and a disk. The Pottery Neolithic Layer 3 yielded a lower grinding stone, two hammer stones, two pestles, a platter, three disks, a perforated disk and unfinished weight. The Pre-Pottery Neolithic Layer 4 included a single perforated disk and an upper grinding stone. It is worth noting the wealth and variability of the Pottery Neolithic assemblage compared to the other layers.
 
The Pottery Assemblage and Other Finds
Anna Eirich-Rose
 
The pottery assemblage from the site includes 1055 potsherds from two major periods: Pottery Neolithic (the Yarmukian culture) and Early Chalcolithic (Wadi Rabah culture). Both layers contained mixed potsherds.
Pottery Neolithic ceramics is poorly preserved and crudely made. The decoration is limited to red slip and burnishing and in one case, a thick painted line was observed. No herringbone pattern was noted in the assemblage.
The typology is limited to a few simple shapes; the most common is a simple medium-sized bowl, mostly red slipped (Fig. 9:1–5). Other fragments included a rim of a pot (Fig. 9:6), a simple holemouth rim (Fig. 9:7), one basin (Fig. 9:8) and a large lug handle of a pithos (Fig. 9:9).
The Wadi Rabah assemblage is small (754 potsherds). The clay is levigated and well fired, with small gray and white tempers.  
The predominant decoration is red slip that was applied to all types of vessels. Black burnishing, combined with red, is restricted to bowls. Other decorations include combing, incisions, impressions, as well as lunar impressions (Fig. 10:1, 2) and plastic decorations.
The type diversity is comparatively limited and consists of bowls, jars, holemouth and pithoi. The bowls include specimens with a straight profile that are red slipped and burnished (Fig. 10:3, 4); comparisons are found at Nahal Bezet I (Gopher A. 1989. Mitekufat Haeven 22:82–93)and at Layer 19 at Horbat Uza (Getzov N., Lieberman-Wander R., Smithline H. and Syon D. Uza I [IAA Reports, forthcoming]. Jerusalem); small globular bowls with inverted profile; a carinated bowl (Fig. 10:5) that is similar to bowls from Nahal Bezet I (Gopher, ibid., Fig. 7:3) andLayer 19 at Horbat Uza; and a large bowl or basin with thick walls and a cut rim (Fig. 10:6) that has comparisons in Nahal Bezet I (Gopher, ibid., Fig. 7:7) and Horbat Uza, Layer 19.
The only jar type has a straight neck (Fig. 10:7) and can be compared to jars from Beisamoun (Rosenberg D. et al. 2006. Mitekufat Haeven 36:129–139) and at Horbat Uza, Layer 17. Undecorated holemouth jars, which have thick walls and a cut rim (Fig. 10:8, 9), are similar to Horbat Uza, Layer 19. The walls of the holemouth pithoi with cut rim at our site are thicker than those at Horbat Uza.
Two hollow high bases, probably from chalices (Fig. 10:10, 11) are similar to bases found at Horbat Uza, Layer 17.
 
The Wadi Rabah pottery at the site shows great resemblance to the pottery from the nearby sites of Horbat Uza and Nahal Bezet I. The limited variety of pottery types is due to the small-scale excavation. 
 
Other Finds
These included a clay figurine and a stamp seal.
Figurine. An upper fragment of a figurine (Diam. 1.1 cm, height 3.9 cm; Fig. 11:1) was unearthed in Layer 3. The body is cylindrical, smooth and has a uniform diameter. The clay is well levitated and tempered with very fine gray-black grits and straw. The head at the top end is modeled by pinching the clay to the back of the figurine. This creates a triangular head with a pointed nose, resembling a bird’s beak. The lower part of the figurine is broken. As in Munhata (Garfinkel Y. 1995. Human and Animal Figurines of Munhata, Israel [CCRFJ 8]. Paris), this object may be interpreted as a simple cylindrical standing figurine with a head at the top.
It is difficult to decide whether the figurine is a zoomorphic or an anthropomorphic depiction. A similar figurine was defined as conical-shaped faced and anthropomorphic (Commenge-Pellerin C. 1997. A Pottery Figurine from Yiftah’el. In E. Braun. Iftah’el [IAA Reports 2]. Jerusalem. Pp. 180–181). Garfinkel (ibid., pp. 34–40) classified similar figurines as schematic anthropomorphic presentations and Rollefson (BASOR 27, 1991, Pp. 95–116) saw a bird in a comparable figurine from ‘Ain Ghazal.
 
The stamp seal is made of black serpentine (Fig. 11:2) and was recovered from Layer 3. The extant seal has an oval shape (length 2.2 cm, width 2.4 cm) with a drilled hole in the center (diam. 0.24–0.70 cm). It is decorated with chevron rows that resemble a herringbone pattern. The artifact is broken in half, probably due to the unsuccessful drilling, which took place after the ornamentation. It probably belongs to the ‘low back seals’ (Buchanan B. and Moorey P.R.S. 1984. Catalogue of Ancient Near Eastern Seals in the Ashmolean Museum: The Prehistoric Stamp Seals. Oxford) and the drilled hole replaced a broken handle. This stamp seal was initially designated as an impressing tool, yet in secondary use, it was probably a personal amulet (Skeates R. 2007. CAJ 17:183–198).
Similar seals are known from other sites, while the design is rare and appears in Ha-Goshrim IV (Getzov N. 2005. NEAEHL, Vol. 5, pp. 1759–1761).
 
The Fauna
Nimrod Marom
 
The analysis of the entire faunal assemblage from the site yielded 169 identified bones from eight mammalian taxa. These include hippopotamus (Hippopotamus amphibious; NISP=1; incisor fragment), cattle (Bos sp.; NISP=44), Mesopotamian fallow deer (Dama mesopotamica; NISP=2), boar/pig (Sus scrofa; NISP=50), goat and sheep (NISP=66; Capra sp., NISP=1; Ovis aries NISP=7), gazelle (Gazella gazelle; NISP=14), dog or wolf (Canis sp.; NISP=10) and cape hare (Lepus capensis; NISP=1).
A diachronic change seems to have occurred in the frequency of representation of the three common ungulate taxa in the assemblage (cattle, pigs and caprovines) during the transition from the PPNB to the PN (pooled Yarmukian and Wadi Rabah loci), as cattle became much less abundant and caprovines more common (see Table 1). This change is statistically significant (chi-squared=6.47, p=0.04).
 
Taxon
Body Part
Area C
Area F
All Areas and Periods
 
 
PPN
Yarm.
WRab
Total
PPN
Yarm.
Total
Total
%
Bos
Head
2
2
1
5
 
 
 
5
11
 
Axis
5
2
 
7
 
 
 
7
16
 
Upper
Forel
 
1
 
1
 
 
 
1
2
 
Lower
Forel
 
1
 
1
1
2
3
4
9
 
Upper
Hindl
 
 
 
 
 
 
 
 
 
 
Lower
Hindl
5
1
 
6
1
3
4
10
22
 
Feet
6
3
 
9
3
6
9
18
40
Total
 
18
10
1
29
5
11
16
45
34
Sus
Head
12
2
2
16
 
 
 
16
43
 
Axis
4
4
 
8
 
 
 
8
22
 
Upper
Forel
1
2
 
3
3
2
5
8
22
 
Lower
Forel
1
 
 
1
 
 
 
1
3
 
Upper
Hindl
 
 
 
 
 
 
 
 
 
 
Lower
Hindl
2
 
 
2
 
1
1
3
8
 
Feet
 
 
 
 
 
1
1
1
3
Total
 
20
8
2
30
3
4
7
37
28
Caprine
Head
3
11
2
16
 
 
 
16
42
 
Axis
3
3
2
8
 
 
 
8
21
 
Upper
Forel
1
3
 
4
 
 
 
4
11
 
Lower
Forel
2
1
 
3
 
 
 
3
8
 
Upper
Hindl
1
2
 
3
 
 
 
3
8
 
Lower
Hindl
2
2
 
4
 
 
 
4
11
 
Feet
 
 
 
 
 
 
 
 
 
Total
 
12
22
4
38
 
 
 
38
29
Gazella
Gazelle
Head
 
 
 
 
 
 
 
 
 
 
Axis
 
 
 
 
 
 
 
 
 
 
Upper
Forel
1
 
 
1
1
 
1
2
22
 
Lower
Forel
1
2
 
3
 
1
1
4
44
 
Upper
Hindl
 
1
 
1
 
 
 
1
11
 
Lower
Hindl
2
 
 
2
 
 
 
2
22
 
Feet
 
 
 
 
 
 
 
 
 
Total
 
4
3
 
7
1
1
2
9
7
Canis
Head
 
 
1
1
 
 
 
1
25
 
Axis
2
 
 
2
 
 
 
2
50
 
Upper
Forel
 
 
 
 
 
 
 
 
 
 
Lower
Forel
1
 
 
1
 
 
 
1
25
Total
 
3
 
1
4
 
 
 
4
3
NISP
 
57
43
8
108
9
16
25
133
100
%
NISP
 
53
40
7
100
36
64
100
 
 
 
Table Reading: Yarm.= Yarmukian; WRab=Wadi Rabah; Upper forl=upper forelimb; Lower forel=lower forelimb; Upper hindl=upper hind limb; Lower hindl=lower hind limb.
 
The status of domestication for the main livestock taxa is difficult to ascertain due to the insufficient sample size, which prevented the construction of mortality profiles and a detailed morphometric analysis. Nevertheless, the few measurable specimens are smaller than the reference standards, routinely used as a benchmark to determine the state of cattle and suiddomestication and it suggests the possible domesticated status of these taxa. However, this can not be determined with certainty, given the small number of measurements and the indeterminable sex ratio present in the sample.
Skeleton element abundance (SEA) profiles were compared for cattle, caprovines, and pigs. Since raw NISP counts were employed, the SEA profiles are not taken to reflect the actual representation of body parts at the site, since differential element representation, fragmentation and attrition bias the true distribution. However, it is plainly visible that a similarity exists in the SEA profiles of cattle, pigs and caprovines, which indicate similar patterns of human transportation and carcass processing, or, maybe more parsimoniously, the equifinal impact of various taphonomic processes.
Among the taphonomic processes that commonly shape the bone assemblages is carnivore gnawing. Such traces are apparent on 5% (N=9) of the identified bones and therefore, one may claim that carnivore modification had some deleterious impact on the assemblage. Post-depositional, subaerial weathering of bones was checked on a sample of 20 shaft fragments. Most (NISP=14; 70%) showed traces of light weathering and fewer (NISP=6; 30%) were hardly weathered at all. Only one specimen showed severe state of weathering. The low amount of weathering in the assemblage indicates that a short period of time elapsed between the deposition of the bones and their burial. Bone fracture morphology was examined on eight epiphyses, which retained a diaphysis fragment. Morphologies indicating breakage of fresh bones were dominant in all categories: fracture angle (N=4; 80%); fracture outline (N=5; 100%) and fracture edge (N=4; 66%). The differences from N=14 was due to a large number of morphologies that were not assignable to a specific category, hinting at bone breakage for marrow extraction at the site.
Butchery marks were observed on 28 of the NISP (18%). Most were caused by carcass disarticulation activities (N=20), few by skinning (N=5) and a single butchery mark was indicated by the filleting of meat off the bones. Cattle were butchered more intensely than other taxa, but a pattern of more intense butchery of the lower limbs was apparent in all three ungulate groups. Higher rates of burning on lower limbs and head bones may indicate frequent use of roasting, which is more likely to char these bony elements that are not covered with much flesh.
 
The faunal assemblage from the site is dominated by caprovines, cattle and pigs. The relative frequency of these ungulate taxa changes significantly at the transition from the PPNA to the PN, when caprovine frequencies rise and cattle becomes drastically scarce. The taphonomic history of the assemblage seems to suggest that carcasses of all three ungulate taxa were butchered on-site, the meat was prepared by roasting and the bones fractured for marrow. The weathering state of the bones suggests quick post-depositional burial of the assemblage.
 
The Bone Tools
Gaëlle Le Dosseur
 
The bone-tool assemblage contains eight pieces. Most of them (N=5; width 10–19 mm, thickness 3 mm) are flat cutting or smoothing tools; three of these have very thin sides and an elliptical section and may have been used for cutting. The sides of one of them are completely smoothed by use. Another tool, thicker than the previous ones and having a rectilinear base, is broken at the distal end and the sides converge at the break point. The sides of the medial part are parallel, while the section is rectangular. The remaining distal part is smoothed by use at a distance of 11 mm from the broken extremity. Most of these flat tools were made of ribs of medium to large mammals. The bones were cut longitudinally and in one case, the bone was also cut transversally by sawing. The blanks were then shaped by shaving, while the bases were abraded.
The interpretations of the traces left on the three remaining pieces are not univocal and therefore, it can not be determined whether they were tools. One is a rib fragment, with a complete section and smoothed surfaces, on which very fine and transversal striations appear. These could result from post-depositional events, intentional polishing of the bone or from use. A second piece is a thin rib fragment, burnt to black, with visible traces of shaving, which were left during butchery or while producing a tool. The third piece comes from a tooth. One extremity near the root could have been sawn, although this is difficult to confirm as the tooth is so smoothed.

 

The Shell Finds

Inbar Ktalav
 
The assemblage from the site consisted of 14 shells; 12 are of Mediterranean origin and two are local freshwater shells. The finds are described by periods.
Surface: Four shells were found, including a fragment of a freshwater snail, Melanopsis buccioidea, a shell rim of Phalium undulatum and two bivalves of Cerastoderma glaucum, one of which has an artificial large hole at the top of the valve.
Early Chalcolithic: A single complete valve of Cerastoderma glaucum was found.
Pottery Neolithic A: Three shells were found, including a complete shell of Melanopsis buccioidea that has a natural hole in the undulation of the body and two valves of Glycymeris insubrica, a fragment and a complete valve with an artificial hole at the top of the valve.
Pre-Pottery Neolithic B: Six shells were found, including five valves of Glycymeris insubrica, two complete ones, one of which has an artificial hole at the top of the valve and three fragments, as well as a single valve of Cerastoderma glaucum, with a large artificial hole at the top.  
 
The shells with holes were probably used as pendants. A symbolic meaning was attributed to the shells, based on shape and texture. It was suggested that the round and smooth shape of the Glycymeris shells symbolize the moon and feminine fertility, whereas the ridged Cerastoderma glaucum shells could represent the rays of the sun and masculine powers. The ‘cassid lips’ may represent a crescent moon, symbolizing the new moon, birth or a new beginning. However, whether these shells were used as mere ornaments or expressed deep symbolic meaning is a mere conjecture.