«Item type text; Dissertation-Reproduction (electronic) Authors Munro, Natalie Dawn Publisher The University of Arizona. Rights Copyright © is held ...»
el-Wad has a long excavation history that began in the Fall of 1928, when the site was tested by Charles Lambert. Lambert sunk two trenches on the terrace and one in the cave. His discovery of rich Natufian deposits both in and outside of the cave saved the Mount Carmel caves from plans to open a limestone quarry there (Bar-Yosef and Callander 1999; Weinstein-Evron 1998). One year later, Dorothy Garrod began an intensive excavation of the site that lasted five full seasons (1929-1933). Over the course of these excavations Garrod completely removed the fill from Chambers I, H, and part of Chamber III in the cave. Her team opened 270 square meters on the terrace and talus slope, digging until bedrock was reached (Garrod and Bate 1937). She discovered that elWad was a multicomponent site with few areas of clearly superimposed stratigraphy.
Figure 3.IS; Map of area excavated during Weinstein-Evron's salvage project in Chamber ill of el-Wad Cave.
Adapted from Weinstein-Evron (1998).
Late Natufian deposits in various patches within the cave, but only Natufian deposits on the terrace (Figure 3.15).
Garrod's excavations exposed extremely rich Natufian deposits, particularly those
Stone curb, a stone wall, a built hearth, and a slab pavement at the base of Layer B1 on the terrace. Layer BI was also rich in bone tools, ornaments (stone and bone beads), and mobiliary art, including two sculpted objects: a cervid carved from a large animal long bone and a roughly shaped human head manufactured from calcite. Though Layer B2 included many of the same classes of material objects as Layer B1, their diversity and abundance was much reduced in comparison to the preceding Early Natufian phase (Garrod and Bate 1937).
Bar-Yosef and Valla returned to the terrace at el-Wad in the 1980s to check and refine the stratigraphy using modem excavation methods. Valla et al. (1986) confirmed the presence of Garrod's Layers B1 and B2, and the assignment of B2 to the Early Natufian, but they fiirther divided Layer B1 into Late and Final Natufian components.
The presence of Early, Late, and Final industries makes el-Wad the only Natufian site other than Ain Mallaha to contain the full temporal sequence of the Natufian. According to archaeological indicators, the Early Natufian occupation at el-Wad was the most intensive, extending over the entire excavated area on the terrace and into Chambers 1,11, and in in the cave (Garrod and Bate 1937; Valla et al. 1986). The thinner Late and Final deposits are limited to the terrace, and cover less area than the Early Natufian layer.
Although, the only Paleolithic inhabitants of the terrace at el-Wad were the Natufians, the stratigraphy in the cave is much more complex. Deposits in the cave have formed in pockets that extend unevenly across its fioor. This means that several cultural deposits may be present at any given depth inside the cave, although the areal extent of each layer
sequence of layers than one excavated only a few meters away. Because of close contact between the layers and confusion over exactly where each begins and ends, some mixing between layers in the cave has undoubtedly occurred.
Although Garrod removed most of the deposits from Chambers I and II in the cave, Chamber in was not completely excavated, and some Early Natufian deposits were left intact. It is from these deposits that the el-Wad sample analyzed in this study was obtained (Figure 3.16). In 1988 a team led by Mina Weinstein-Evron of Haifa University initiated a salvage effort in Chamber III. Soon after, a walkway was constructed in the area to facilitate access to the cave for visitors (Weinstein-Evron 1998). Though most of the deposits in Chamber III have been assigned an Early Natufian age, the assemblage is characterized by much lower frequencies of lunates, sickle blades, and awls than the Early Natufian assemblages from Chambers I and II in the cave. The Chamber III deposits do, however, contain typical Natufian groundstone, artwork, and faunal remains, as well as some lunates. The latter led Weinstein-Evron (1998) to interpret the deposits as a specialized activity area from the Early Natufian phase. A second archaeological deposit in Chamber III causes some uncertainty. This layer underlies and extends beyond the Early Natufian layer in several directions. The lithic assemblage is virtually indistinguishable from the one assigned to the Early Natufian, except that no lunates and only a small proportion of microliths are represented. Although Weinstein-Evron (1998) suggests that the layer may be of Late Upper Paleolithic origin, she does not discount the possibility that it may also be associated with the Early Natufian layer above. This as yet unresolved issue does not affect the sample studied here, which originates solely from the overlying Early Natufian layer.
In the past few years, another excavation has been opened on el-Wad Terrace, under the direction of Weinstein-Evron and Kaufrnan. The new excavations are located on the north end of the terrace just beyond the extent of Garrod's original excavations.
Because the Natufian deposits on the terrace have been disturbed, some mixing has occurred, making it difficult to isolate Early and Late Natufian assemblages (Guy Bar-Oz, Personal Communication, 2001).
The four Natufian assemblages analyzed for this study were collected by different teams of excavators who, fortunately, shared high standards of data recovery. Each assemblage was dry screened using 2 millimeter mesh, wet-sieved, and "picked" in some if not all stages of data recovery. The importance of high standards of recovery cannot be overemphasized because this project uses the relative abundance of small game faunas to address some major questions. In general, the preservation in each assemblage is very good, particularly in the three samples that originated from caves (Hayonim Cave, Hayonim Terrace, and Hilazon Tachtit). The fauna from Hayonim Terrace were coated in a calcareous outer sheath. Once removed, the underlying fauna is in remarkably good condition, yet there is no doubt that the strength of the concretions had the potential to crush small, fragile bones. This may explain the poor recovery of avian specimens from
Hayonim Cave Portions of the Hayonim Cave Natufian fauna have been studied by several researchers with a range of research goals. Tchemov (Bar-Yosef and Tchemov 1967) undertook the first analysis, studying the sample recovered fi-om the 1965 test excavations. The assemblage was small, but provided an accurate preview of some broad trends that would emerge later. Davis' (1978) dissertation research focused on the ungulates and carnivores excavated during the first field seasons at Hayonim Cave up to
1975. The excavations during these seasons were concentrated at the back of the cave where several graves were located (Rows 19, 20, 21, 22). Much of this area is excluded from the present analysis for fear of contamination, because the Natufians dug several of these graves into the underlying Aurignacian deposits. Although Davis' study was insightful, particularly his identification of a temporal change in the age composition of gazelle populations, much of the fauna he examined is excluded from this study due to potential mixing of the deposits with earlier layers. Davis mentions the presence of diverse small game species, he did not quantify them in his study.
In 1984, Pichon completed her dissertation on the avifauna of the Natufian layer from Hayonim Cave. Like Davis, she studied the birds recovered only in the first few excavation seasons, as the fauna excavated in the final seasons had not yet been sorted.
Pichon's study provides a thorough analysis of the wide spectrum of avian species deposited at Hayonim Cave, from the micro fauna to the largest raptors. Pichon's avian sample and her identifications are clearly labeled in the Hayonim Cave assemblage. A
In areas where the current sample overlapped with Pichon's, her identifications of rare species were checked against the comparative avian collection in the Department of Evolution, Systematics and Ecology at the Hebrew University and were then adopted here. The avian sample was bolstered significantly by fauna excavated in the late 1970s and 1990s.
Other, more specialized analyses of the Hayonim Cave fauna were undertaken by Cope (1991a, 1991b) who reanalyzed the ungulates from the sorted assemblage and argued for a specialized Natufian gazelle economy that preferentially targeted adult males. In the late 1980s Lieberman (1991, 1993) examined cementum from gazelle teeth to determine the seasonality of site occupation. Most recently, Bar-El and Tchemov (2001) sampled the hare assemblage as part of a region-wide examination of Lepus exploitation in the Natufian period.
Despite the plethora of research on the Hayonim fauna, the assemblage has never before been examined in its entirety or treated as a complete unit. This has precluded any comparative analyses between taxonomic groups, and has prevented researchers from detecting broad scale changes within the Natufian period itself All faunal samples that have been examined to date were treated as a single Natufian component, masking both temporal and spatial variation within the site. For this study, the entire Hayonim Cave assemblage was analyzed, with the exception of fauna from potentially mixed contexts, or fauna with incomplete provenience information, and microfauna. This required the study of collections that had been sorted and curated by taxonomic group in Eitan Tchemov's
contained a large sample of identifiable material that had not yet been removed (ca. 5000 pieces).
The Natufian layer from Hayonim Cave is in direct contact with other cultural layers (including Byzantine, Kebaran, Aurignacian, and Mousterian). The Natufians excavated many of their graves into older cultural deposits, and later occupants disturbed the surface of the Natufian layer, particularly the Byzantines, who dug a glass furnace into Late Natufian layer. By "potentially mixed context" I refer to deposits that are stratigraphically assigned to the Natufian, due to their proximity to deposits fi^om other layers, but have a greater chance of containing intrusive materials. Potentially mixed contexts are expected to contain predominantly Natufian-aged faunas, but they are excluded from analyses of relative species abundance to ensure that the patterns reported in upcoming chapters are clean and accurate. Secure cut-off depths for the Natufian deposits were determined in consultation with Ofer Bar-Yosef and Arma Belfer-Cohen and based on field maps, notes, and firsthand knowledge of the complex stratigraphy of the Natufian layer. The upper and lower cutoff depths for each square and subsquare from the Natufian excavations are presented in Appendix 2. At 15,000 NISP, the sample from secure contexts is ample, and each phase is represented by roughly equal numbers of bones, eliminating the potential for sample size bias (Table 3.4). Comparisons indicate that the fauna analyzed from potentially mixed contexts before the final cut-off points were determined (NISP = 3000), closely resembles those from secure contexts (see
Table 3.4: Size (NISP) of faunal samples by Natufian phase at Hayonira Cave.
Hayonim Terrace The fauna recovered from the original test pit dug by Bar-Yosef (Bar-Yosef and Tchemov 1967) and Henry's excavation have been analyzed in part (Henry et al. 1981).
In addition to the Hayonim Cave fauna, Davis (1978) reported on the ungulates and carnivores from Henry's excavation in his dissertation. Cope (1991a, 1991b) and Lieberman (1991, 1993) also sampled the ungulates from Henry's excavations in their dissertation research. More recently, Horvvitz (n.d.) studied the fauna from Valla's mixed Niveau I, and attempted to separate the PPNA and Natufian components using taphonomic indicators. Aside from an in-depth study of the Canis sp. burials (Tchemov and Valla 1997), no other analyses on the Natufian faunal assemblage from Niveau II have been undertaken.
The sample from Hayonim Terrace studied here was recovered entirely from
Preliminary work on the assemblage was completed under the direction of Rivka Rabinovich. This included cleaning the bones with weak acetic acid to remove concretions, and sorting identifiable fragments into drawers organized by element. For this study, the NISP of the Terrace collection was calculated by tallying the taxon and element of all identifiable fauna, excluding microfauna from the sorted drawers.
To ensure that the exclusion of identifiable fragments in the unidentifiable fraction would not bias the results presented here, a sample of "unidentifiable splinters" was examined and all identifiable material was counted and compared to the species represented in the sorted fraction (Figure 3.16). The resulting profiles differ significantly because of a large quantity of tortoise carapace and plastron fragments that remain mixed in with the unidentifiable splinters. Because Figure 3.16 compares proportions of major game groups, the large discrepancy in the number of tortoises make it appear as if the remaining groups are underrepresented in the unidentifiable fraction. In Figure 3.17 the tortoises have been removed from the analysis, and the resulting proportions from the sorted and splintered fractions are virtually identical. Clearly, it is only a significant difference in the abundance of tortoises that causes the discrepancy in Figure 3.16, and 1 am confident that the relative proportions of all taxa other than tortoises reported here are accurate, even without additional sorting of the splintered material. The tortoise fraction is, however, underrepresented, as are corresponding proportions of small game. Because both tortoises and small game are already common in the Terrace assemblage, this
Figure 3.17: Relative abundance of major game groups from tlie Late Natufian occupation at Hayonim Terrace.
"Sorted" bars represent identifiable bones that were removed from the "unidentifiable" splinters prior to this analysis. "Splinter" bars represent identifiable bones removed from a sample sorted from the "unidentifiable" fraction.