«Item type text; Dissertation-Reproduction (electronic) Authors Munro, Natalie Dawn Publisher The University of Arizona. Rights Copyright © is held ...»
circular stone wall and contains evidence of discrete building sequences of layers representing at least four consecutive occupation phases. Locus 6 also contains intact layers belonging to four phases, but the samples from the upper two phases (IV and V) are small due to the removal of fauna from potentially mixed contexts. The fauna quantified from potentially mixed contexts in Locus 6 are thus, included only for this analysis. Figures 7.8, 7.9, and 7.10 compare the relative proportions of slow versus fast small game for each consecutive phase in Locus 4, 6, and 8, respectively. Once again, the patterns described for the site as a while stand out with only minor variations.
With the partial exception of the sample from Locus 6 presented in Figure 7.9, these results are based only on material from undisturbed contexts at Hayonim Cave.
The removal of fauna from potentially mixed contexts significantly reduced the sample sizes of the Phase IV and V (Late Natufian) deposits at the top of the layer. The use of only the most reliable of all contexts slightly diminishes the trend in small game use, patterning is stronger when fauna from potentially mixed contexts are included in the sample, but the trend is still very clear and consistent with the results of site-wide comparisons. Figures 7.11 and 7.12 show the differences between the small game fraction represented in good contexts from Phases IV and V and those from some good and some potentially mixed contexts. Small game from potentially mixed contexts dating to the Late Natufian contain even higher proportions of tortoises than the "cleanest" assemblage. The removal of these potentially mixed faunas reduces only the intensity of
Figure 7.11: Relative abundance of small game taxa from "good contexts" (contexts from secure Natufian contexts) versus those from potentially mixed contexts (those originating in areas with high potential for mixing with non-Natufian layers) from Phase IV of the Natufian deposits at Hayonim Cave.
The trends described for the small game fraction in the Hayonim Cave fauna are real. Even in the face of problematic contexts, a very distinct trend emerges from the data: fast small game were most abundant in human diets during the Early phase of occupation at Hayonim Cave (Phases I-III),but much less abundant relative to slow small game during the Late Natufian phase (Phases IV and V).
Discussion of Results on Hayonim Cave Natufian High proportions of fast small game animals in Early Natufian deposits of Hayonim Cave indicate that encounter rates with high-ranked species were insufficient to meet human demands for animal resources. Natufian foragers compensated for this by bringing more low-ranked hares and partridges into their diets. Hunters continued to capture tortoises when encountered, but the increased demand for animal products may have been sufficient to deplete tortoise populations. After low-ranked species were added to human diets, the foraging system stabilized; human demands for small game did not grow beyond the reproductive capacity of hares and partridges.
By the Late Natufian a significant shift in human foraging strategies occurred.
This was essentially a return to greater dependence on tortoises, implying that previously depleted high-ranked resources were able to recover at least to levels observed for the Kebaran period. Though the continuing presence of hares and partridges attests to the Natufian's recognition of these animals as potential food sources, they did not invest as much energy in their capture. This suggests that human needs were greatly reduced in
abundant to meet these needs, and that Late Natufian population densities in the areas were lower than those of the Early Natufian.
Following the expectations of the small game index presented earlier, trends in the relative abundance of small game from Hayonim Cave appear to reflect a dramatic change in the intensity of site use from the Early to the Late Natufian. The integration of low-ranked animals into the Early Natufian hunting repertoire suggests more mouths to feed per unit foraging habitat, and thus greater hunting pressure on the surrounding ecosystem than any earlier Paleolithic inhabitants. Furthermore, increased demands for animal resources and associated increases in hunting pressure point to more intensive site use; that is, greater numbers of Cave occupants per unit time. In the Late Natufian people seemed to have exerted significantly less hunting pressure on the surrounding ecosystem than their Early Natufian predecessors, implying a substantial reduction in the number of people occupying Hayonim Cave per unit time and/or reduced number of visits, and perhaps in the Mediterranean hills in general.
These results are corroborated by independent reconstructions of the occupation sequence at Hayonim Cave by Bar-Yosef and Belfer-Cohen (n.d.; Belfer-Cohen 1988).
These authors argue for three major stages of building and occupation during the Natufian period. The earliest stage (Phase I) was marked by the construction of the first and largest of the stone structures (Locus 3) at the entrance of the cave. The second stage (Phases I-III, with the exception of Locus 3) was characterized by the construction of all loci, with the exception of Locus 3, and several episodes of structural refurbishment.
the drip line of the cave. Each of these loci with the exception of 6 and possibly 9, 10, and 11, which have not been fully excavated, contain formal occupation floors and stonelined hearths, reflecting significant energy investment. These observations along with other artifactual evidence indicate that the loci served primarily as domestic structures during their initial occupation (Bar-Yosef and Belfer-Cohen n.d.; Belfer-Cohen 1988).
Some ritual paraphernalia are also present, suggesting that both ritual and domestic activities were integrated into daily life. After its initial construction, each locus followed an independent trajectory of rebuilding and reuse. During the same time span, burials were separated from living areas, and the dead were primarily interred in graves behind the loci, toward the rear of the cave. Bar-Yosef and Belfer-Cohen (n.d.) interpret this second period of building in Hayonim Cave as the most intensive Natufian occupation phase. Their interpretation is independently supported by the faunal data.
During the final stage of occupation (Phases IV and V), a large continuous area opened up along the eastern wall of the cave. Some of the archaeological deposits in this area have since been disturbed by the construction of a Byzantine glass furnace, but a general picture of activities in this area can still be constructed. There is a general lack of formal features in the open area, and there are three unusual caches against the cave wall, as described earlier. Evidence for the last Natufian occupation at Hayonim Cave (Phase
V) also comes primarily from this area, where several graves were dug into the fill of earlier loci. The presence of these graves clearly indicates that loci ceased to be used for domestic purposes, though some domestic activities still took place along the cave's
more sporadic occupation, in contrast to the Early Natufian. Their interpretation is based on the apparent lack of formal architectural features, the presence of caches with intentionally stockpiled materials for later use, and the addition of several graves to the fill of the loci. This assessment is independently corroborated by the faunal evidence, which indicates a much less intensive occupation of the cave during the Late Natufian Phase.
A sample of Natufian sites are compared to examine the relative intensity of occupation at each site, a proxy for evaluation of regional patterns of Natufian demography and subsistence change. The size of the comparative sample is limited by both time constraints on data collection and the lack of other published assemblages of Natufian small game fauna. Although researchers working with Levantine assemblages have occasionally reported the frequencies of one or two small game species such as hares (Bar-El and Tchemov 2000; Bouchud 1987; Davis et al. 1994; Pichon 1984;
Rabinovich 1998; Tchemov 1984), complete quantification of all major groups of small prey have been done only in a few cases and for small samples (Bar-Yosef and Tchemov 1967; Crabtree et al. 1991; Valla et al. 1986). Bouchud (1987) analyzed the full range of food species from two stmctures at Ain Mallaha but did not quantify materials from areas outside these stmctures noting only that the proportions were markedly different. Thus we are left with the data collected specifically for this study from Hayonim Cave,
that the data meet a nearly ideal standard of comparability because the analyst is the same for all cases to be compared. Presentation will follow the same analytical format used for Hayonim Cave.
The fauna from Hayonim Cave are collapsed into two groups for this comparison corresponding to the Early (Phase I-III) and Late (Phase IV and V) phases of the Natufian period. Otherwise, the sample is identical to that used in the preceding intra-site comparison of Hayonim Cave. The Hayonim Terrace, Hilazon Tachtit, and el-Wad assemblages are as described in Chapter 3.
Results of Inter-Site Natufian Comparison Figure 7.13 shows the relative proportions of three broad taxonomic groups at each site. Although the results indicate some variation, there is no trend in the relative abundance of ungulates and small game among the sites. The high proportion of small game, which comprises between 40 and 60% of each assemblage, is an exclusively Natufian phenomenon, uniformly exceeding proportions of small game from preceding Paleolithic cultures in the Levant.
Analysis of small game proportions alone confirms that much of the diversity in Natufian faunas occurs within the small end of the dietary spectrum. Figure 7.14 reveals dramatic differences in small game type frequencies in the Natufian samples. Both of the Early Natufian assemblages, el-Wad and Hayonim Cave, are rich in low-ranked small game. The situation reverses in the Late Natufian assemblages from Hayonim Cave, Hayonim Terrace, and Hilazon Tachtit, when high-ranked species resurge on par with the
Figure 7.13: Relative abundance of three broad taxonomic groups from five Natufian occupations in northern Israel.
ELWD = el-Wad, HAYC = Hayonim Cave, HAVT = Hayonim Terrace, and HLZT = Hilazon Tachtit. Early and Late refer to temporal phase divisions within the Natufian period.
These data comment on both regional patterns of game use (broad taxonomic groups) and the site occupation intensity of each site (relative proportions of small game).
The high proportions of small game remains at Natufian sites indicate both increased human predation pressure and associated decreases in human foraging efficiency on a regional scale compared to earlier periods. This pattern clearly contrasts with that observed by Stiner et al (1999, 2000) for earlier cultural periods in the Levant, although the scale of contradiction is only noticeable if confined to variation within the Epipaleolithic, and is of great interest for dynamics within the Natufian in particular. As discussed in Chapter 1, a shift from high-ranked, large-bodied ungulate species to small, low-ranked ones suggests that ungulates were no longer sufficiently abundant to meet the demands of the human population. This condition typifies all Natufian phases. Increased reliance on small, locally available animals points to a decrease in encounters with large mammals or an increase in human demands for meat due to larger population sizes, or both. As a result of depletion, environmental deterioration, or increased demand, hunting efforts were intensified in the Natufian period. Natufian populations in the Mediterranean zone likely exerted sufficient pressure to depress, though never entirely deplete ungulate resources. When researcher biases are eliminated (Chapters I and 3), the proportions of small to large game are similar at the four Natufian sites, regardless of time period, suggesting that human pressure on large game populations was constant on a regional scale throughout the Natufian period.
At the local scale, the changing compositions of small game assemblages (Figure
occupation intensity. The Early Natufian occupations at Hayonim Cave and el-Wad provide the clearest indications of intensive site use (Garrod and Bate 1937; Valla et al.
1986). Foragers at both sites hunted large quantities of low-ranked prey, but continued to capture high-ranked species when encountered, particularly tortoises and gazelles. In fact, large sites generally date to the Early Natufian phase.
Like the story at Hayonim Cave, the Early Natufian layer at el-Wad (Garrod's Layer Bl) is both thicker and larger in area than the Late Natufian deposits of this site.
Moreover, el-Wad Terrace exhibits several architectural features attributed to the Early Natufian (Garrod and Bate 1937). The Early Natufian bone assemblages are also much richer and more diverse than those from earlier Paleolithic periods. These characteristics meet the criteria established by Bar-Yosef (1970) for Natufian base camps. Variation in the types of features present in Early versus Late Natufian phases signals an important difference in the intensity of site occupation, which was clearly greatest in the Early phase. The high proportion of low-ranked game in the faunal assemblage from inside elWad Cave supports this interpretation of the Early Natufian occupation; however, the Late Natufian sample from el-Wad was insufficient for comparison, due to the mixing of site deposits.
We have already seen that the Late Natufian occupation at Hayonim Cave was much more ephemeral and sporadic than the Early phase. The likelihood that Hayonim Cave and Terrace represent one very large site makes it essential to compare the Late Natufian deposits from both sites. It is important to clarify whether changes in the Cave
function of the Cave versus the Terrace during the Late Natufian as has been suggested by other researchers.