Have any of the questions and problems raised in the introduction been addressed, let alone answered, in the preceding pages? We believe the present volume does contribute to understanding the Upper Palaeolithic phenomenon in the Levant, at least in terms of summarizing recent and current field projects. Both larger and more limited issues have received attention, thus serving as a point of reference for future studies. As is self-evident from the commentators’ remarks, there is no consensus even at the level of the ‘big’ picture of Levantine Upper Palaeolithic developments. While Marks extols the virtues of the 1980’s parallel phyla ‘Ahmarian’/’Levantine Aurignacian’ model, Copeland cautions against ‘forcing’ new data within too rigid a framework (see also remarks by Kuhn et al. herein). Yet, perhaps this very diversity of opinion should be viewed in a positive light, reflecting the dynamic nature of the subject matter, though limited in general to the lithics. For, ultimately, the Upper Palaeolithic signifies the beginnings of systematic modern behaviour, as regards all domains of human existence. In the following pages we will briefly address some of the varied issues relevant to current research, and point out what we believe to be potentially rewarding lines for further investigations.
When presenting prehistoric evidence and its interpretation, besides the unavoidable opaqueness resulting from the chronological remoteness of the data, there is always the issue of the researcher’s scientific background. Prehistoric studies in the Near East and, especially, the Levant are blessed by the wealth of different research schools. These are probably more diverse than for any other region in the world. In addition to the largely British and French backgrounds of the pioneers of Levantine Upper Palaeolithic studies, there are German and American researchers, as well as native Levantines. Thus each of the researchers represents her/his own specific intellectual orientations and paradigms (see Clark 1991). The very wealth of variability can be two-edged, on occasion leading to dialogues in which people talk past one another. The assumption that we are indeed using a common analytical vocabulary and language needs to be critically examined. Yet this should be done without delving into the backgrounds of any researcher in particular, least we find ourselves in the absurd position of the judgemental and jingoistic ‘we know better where it all comes from’ (e.g., see Clark’s comment in Marks et al. 2001).
It is unfortunately still a truism that the different preconceptions exhibited by various schools of research active in the region (American, French, Israeli, etc.) are further compounded by modern political boundaries and language barriers. Clearly one can no longer simply employ the term ‘Aurignacian’ as a synonym for the ‘Upper Palaeolithic’, just as no one today uses ‘Neanderthal’ to automatically denote a Middle Palaeolithic association. It is fascinating to observe the so-called ‘Aurignacian’ of Üçağızlı becoming ‘Ahmarian-like’ within the space of no more than 5–7 years (compare Minzoni-Deroche 1992, 1995 vs. Kuhn et al. 1999, herein). A similar preconception can be detected in the interpretation of the later Upper Palaeolithic levels of Umm el-Tlel, which Ploux (1999) and Boëda (personal communication) continue to attribute to the ‘Aurignacian’, notwithstanding the distinctly Masraqan/Late Ahmarian appearance of the illustrated assemblages (see also remarks by Copeland, and Nadel herein). Much the same has happened, for example, concerning the preliminary attribution of the late Ahmarian WHS618 assemblage (see Coinman this volume) to the ‘Levantine Aurignacian’ by Clark and his colleagues (Clark et al. 1987, 1988). Undoubtedly, not all of us are using the same criteria and definitions when referring to an ‘Aurignacian’ entity. As an additional example we can bring up the question of the Upper Palaeolithic assemblages from Warwasi in the Zagros, which were defined as ‘Aurignacian’ by some (Olszewski 1994), as opposed to others (Bar-Yosef and Belfer-Cohen in preparation). Furthermore, it is quite clear that the situation in terms of the Ahmarian/Aurignacian complexes differs considerably between the various sub-regions of the Levant (and see remarks in the introduction referring to the geographical isolation of Ksar Akil).
Indeed the criteria used recently to differentiate between the Aurignacian and Ahmarian phenomena are problematic. Notwithstanding techno-typological standards (see Bergman herein), the differentiation involves presence (in the Aurignacian) vs. absence (in the Ahmarian) of ornaments and bone tools. The former do occur sporadically in Early Ahmarian and related assemblages (Kuhn et al. herein; Bar-Yosef and Phillips 1977; Gilead 1981b). But though Marks (herein) points out problems concerning differential preservation of such organic and related items, the sheer variety and quantity of the bone and antler finds (and a range of symbolic items including shell ornaments, bone pendants, bird of prey talons, and ochre – see Belfer-Cohen and Bar-Yosef 1999; Rabinovich herein) in the Aurignacian sensu stricto reflect more than taphonomy and do, we believe, justify the validity of both the criteria and the division between these Upper Palaeolithic techno-complexes.
People tend to use these terms in a variety of ways. Some employ them quite flippantly, considering them as general-purpose repositories (indeed is every ‘blade-oriented’ assemblage in the Levant necessarily ‘Ahmarian’? see Bergman herein vs. Gilead 1989), while others are far more particularistic. It is not always clear within the literature on what plane these terms are used (see, for example, the evolutionary stages in the use of the respective terms by Coinman, Kerry and Henry herein and references therein). Though the recognition of the Ahmarian techno-complex within the Upper Palaeolithic sequence of the Near East was of crucial importance, we believe (for details see Belfer-Cohen and Goring-Morris 2002; Belfer-Cohen and Bar-Yosef 1999; and also see Kuhn et al. herein) that it is now the time to define new taxa, instead of trying to fit every Upper Palaeolithic archaeological occurrence either within the ‘Aurignacian’ or ‘Ahmarian’ sensu stricto. Examples for such an awkward fit can be drawn from the later part of the Upper Palaeolithic sequence. To illustrate, we do not negate the possibility that the Late Ahmarian/Masraqan ultimately derives from the Early Ahmarian, yet the chronological, technological and typological proclivities are sufficiently distinctive, we believe, to warrant a separate appellation (see also Ferring 1988). It seems to us that there is a wide range of unaccounted for variability that is lost if we employ just the broad-based ‘Aurignacian’ vs. ‘Ahmarian’ terminology. One should bear in mind that, in the final analysis, we are attempting to investigate actual prehistoric populations that exploited real territories by specific strategies. As research progresses and the data base increases, we are thus able to propose more detailed models. This is not to say that everything can be neatly apportioned into specific boxes. Various assemblages still remain enigmatic, and, for example, we continue to ponder the nature of some of the Ksar Akil levels (Bergman herein), Kebara IV–III (Bar-Yosef et al. 1996) and the Qafzeh sequence (Ronen and Vandermeersch 1972).
Undoubtedly the material culture remains reflect the nature, intensity and dynamics of activities on-site. But to what extent do the recovered assemblages represent the full suite of activities? This is most especially relevant with regards short-term occupations. The compositions of the recovered toolkits by no means represent always the full range of activities, as refitting studies indicate considerable mobility of tools into and out of sites. The cautionary tale of Early Ahmarian Nahal Nizzana XIII is revealing: almost no blade tools remained on-site, and although they were the primary focus of knapping activities, the occupants removed these tool types on abandonment of the site (Davidzon 2002; see also the papers by Becker and Monigal herein concerning Boker A and the Abu Noshra sites).
This raises the thorny problem concerning the criteria and preconceptions by which we attribute assemblages within concrete archaeological entities. A notable case study relates to the assemblage of Sde Divshon (D27B), which was assigned to the Early Ahmarian by Ferring (1976) and Marks (1977b), but to the Levantine Aurignacian by Gilead (1981a, 1989), as each of the researchers used different analytical criteria, although all three agree to the bi-partite division and contemporaneity of these Levantine Upper Palaeolithic strands. Following Gilead’s (1981a, b) definitions, the Nahal Nizzana XIII assemblage, where flakes outnumber blade/lets among both debitage and tools, should be attributed to the Aurignacian. Yet, the refitting programme at Nahal Nizzana XIII clearly demonstrates that typical Early Ahmarian blade/lets were the primary focus of blank and tool production.
Nomenclature is clearly significant also with regards to the definition and understanding of technological processes, which are of paramount importance when dealing with the definitions of archaeological cultural entities. Technology has gained much prestige and importance through the advent and adoption of the chaîne opératoire method, which is currently the main topic in technological studies. Yet even here, we can see differences in interpreting the chaîne opératoire phenomena if we compare the respective attitudes expressed by Bergman, Chazan and Sarel and Ronen (herein).
It is quite clear that the different backgrounds and perceptions of the various researchers will influence each and every aspect of technological studies. This can even be observed at the most basic level of primary definitions: for example, what is a cortical (primary) element? While some consider that these items should retain 50% of cortical surface coverage, others settle on 25–30%, yet others ignore them altogether (e.g., Fox herein). Another example concerns the notion of laminarity (Tostevin contra Bergman herein). There are also significant differences in describing the presence and nature of platform faceting – the textual description and illustration by Fox of the ‘ chapeau de gendarme’ contrasts with widely accepted definitions (e.g., see Tixier et al. 1980). When comparing the data presented by Tostevin (herein, Table 6.2) with the original publication of the same data by Marks and Kaufman (1983, p. 74, table 5.3), a major difference in the degree of faceting at Boker Tachtit is observed.
A slightly different example concerns the term ‘platform blade’, which is quite commonly employed by some researchers (see papers by Coinman, Fox, and Olszewski). The illustration in Figure 13.5 apparently represents what others over the years have called a ‘ridge blade’ (as opposed to the more distinctive ‘ lame à crête’ or crested blade that is a specific sub-category of ridge blade – see Clark and Rankine 1939). These core trimming elements often display unifacial lateral scar patterns on their distal portions. They are quite common throughout most Upper Palaeolithic and Epipalaeolithic industries in the Levant. Refitting studies demonstrate that they do not always derive from the striking platform edge. This would be counterproductive as a rejuvenation technique, increasing rather than decreasing the angle between the platform and the removal surface. Rather, they generally derive from the main removal surface often immediately after a core tablet was struck off, shortening the length of the removal surface. These ‘platform blade/lets’ are slightly overshot, including part of the keel at the base of the core. They also derive from lateral carination, which may be seen as a means of blunting to ease holding the core during knapping (ANGM personal observation; and Marder in preparation).
Differences in analytic procedures pertaining to technology are observed even among various participants of the same overall project (i.e., Wadi Hasa, and see Coinman, Olszewski, Fox herein; Neeley et al. 2000). And we are still not referring to the more significant aspects of technological studies. A case in point is the study of cores and their flaking scars. To what extent are cores in their final, abandoned and exhausted states really representative of the flaking mode actually used in reducing them (à propos Sarel and Ronen herein)? We doubt that there is often a direct correlation between the two – ultimately it is the original setting-up of the core into a preform that reflects the basic intentions of the knapper. Thus the shape of the abandoned core portrays rather the degree to which the knapper succeeded in managing the block of raw material at hand during the course of its reduction, which can be highly significant from a technological perspective. This is neatly illustrated again by refitting studies at Early Ahmarian Nahal Nizzana XIII, where a conceptually preconceived and redundant approach met the realities of variable raw materials, combined with probable differences in individual knapping skills (Davidzon 2002).
Over the past two decades research has focused primarily upon technological criteria in the investigation of the Levantine Upper Palaeolithic. Indeed, for the Ahmarian, the basic chaîne opératoire has lately been thoroughly documented (though for the Aurignacian and other flake industries the current state of technological knowledge still leaves much to be desired). Typological issues have mostly taken a secondary place with regards lithic analysis and, where relevant, are largely relegated to statistical issues. Indeed, amongst many researchers, antagonism against the very notion of ‘type fossils’, let alone their use, is almost palpable. Yet, whether one approves of the notion or not, it cannot be denied that a variety of distinctive tool types, restricted to certain chronological horizons and geographically limited, do occur through the Upper Palaeolithic sequence in the Near East. Examples include the Emireh point, the chanfrein, the Ksar Akil scraper, burins on Clactonian truncations/notches, etc. Ultimately, these do serve as cultural markers, just as we employ the presence/absence of bone and antler artefacts, provided we remain alert and use it wisely and in moderation (see summary in Flannery 1986).
Questions concerning typological definitions of modified items also reflect much wider issues. A troublesome example relates to the obvious differences in the definitions of point types (see Copeland herein for Upper Palaeolithic point types in the Levant). With reference just to el-Wad points, as employed by various researchers, it is of interest to see the differences in the approaches of Bergman, Phillips and Saca, and Becker herein, as well as those of Azoury (1986) and Ploux (1999). The situation is further complicated by the fact that Garrod originally considered the el-Wad (sic Font Yves) point as a type-fossil of the ‘Levantine’ Aurignacian. Today the el-Wad point is considered by many to be a primary hallmark of the Early Ahmarian (see Belfer-Cohen and Goring-Morris 2002 and references therein). Until such a time as we are able to either substantiate or negate claims that there is indeed variability between el-Wad points in Early Ahmarian as opposed to those in Aurignacian assemblages, we cannot use the el-Wad point as a cultural marker. The case of the el-Wad point is currently a weak link in the arguments for the validity of such criteria for the definition of the various cultural entities.
Assemblage integrity is obviously a major issue to be addressed in evaluating archaeological developments. The problem is especially acute with regards multi-level occupations, especially in caves and rockshelters. The matter is further complicated when re-analyzing assemblages from early excavations, when field methodologies were relatively unsophisticated. One can compare the detailed stratigraphy of the later Upper Palaeolithic levels identified and described by Tixier and Inizan (1981) at Ksar Akil, with that provided by the earlier excavations at the site that were based on much larger stratigraphic units (see details in Bergman 1987).
Additionally there are the long-recognized possibilities of admixtures resulting from post-depositional taphonomic processes (Bar-Yosef and Vandermeersch 1972). For the Middle to Upper Palaeolithic transition these issues are especially crucial. Yet, while admitting the problem of potential mechanical admixtures in the Carmel and Galilee sites, Sarel and Ronen (herein) do not really address the problem in any detail. Much the same can be said of the described sequence at Tor Sadaf (Fox herein); do all three of the arbitrarily defined assemblages really reflect a single continuum? There is thus clearly a necessity for in-depth studies of site formation processes to evaluate the integrity of assemblages (see Goldberg herein).
Another factor to be considered relates to sample sizes and the areas exposed. The described assemblage from Tor Fawaz (Kerry and Henry herein) is quite limited, notwithstanding its derivation from 12m3, raising the question as to whether the described assemblage does indeed represent a single occupation horizon. Also relevant in this context is the matter of inter-assemblage comparisons. Both Nadel, and Kaufman (herein) compare late Upper Palaeolithic/early Epipalaeolithic assemblages deriving from quite different types and scales of excavation. The assemblages from Ohalo II, Ein Aqev, Ein Aqev East and Shunera XVI all derive from extensive surfaces, while for example the Fazael IX, X and Nahal Ein Gev I samples derive from small test pits. Inevitably this raises questions as to the validity of inter-assemblage comparisons.
Notwithstanding the rise in the number of radiometric dates currently available for the Upper Palaeolithic period in the Levant (see Appendix), it seems that instead of providing a solid chronological framework, they frequently obfuscate matters. In part, no doubt, this has to do with the fact that the earlier stages of the Upper Palaeolithic approach the limits of the 14C dating method, playing havoc with the size of standard deviations, etc. (even for later periods the issue is contentious in terms of the reliability of dates). Another matter to be integrated concerns the possibilities and necessities of calibration (see Schramm et al. 2000). Thus the site of Ohalo II, with a large series of radiometric dates at ca. 19,000 bp corresponds to ca. 22 ky when calibrated. The differences between radiocarbon years and their calibrated values can reach up to 5,000 years.
We frequently tend to ‘blame’ the absence of dates for our inability to comprehend the nature of the prehistoric record. Yet the above-mentioned case of Ohalo II is a cautionary tale. The splendid series of dates now available clearly indicates that the site was occupied for some 2000 radiocarbon years (Nadel et al. 1999). Still the stratigraphic observations at Ohalo II do not always accord with the dates obtained. One wonders whether the site as a whole was indeed continuously occupied over such a lengthy duration? Based on the intensity and nature of the finds we believe that the Ohalo II site most probably represents a sequence of short-duration palimpsests over a considerable period of time. A somewhat similar phenomenon can be cited with regards the wide range of dates (ca. 39–31,000 bp, excluding outliers) from the Abu Noshra sites. Yet lithic refits between some Abu Noshra assemblages indicate that they are absolutely contemporary (see Phillips 1991; Becker 1999, herein).
We are nevertheless fortunate in having the results of the innovative environmental research at Soreq cave, as well as detailed studies of the Lisan lake levels (Bar-Matthews and Ayalon herein; Bartov et al. 2002). Accelerator dates have been particularly beneficial for dating archaeological materials, yet until the recent extension of the calibration curve backwards to include the Upper Palaeolithic we were in somewhat of a quandary (Beck et al. 2001; Schramm et al. 2000). The combination of the calibration to calendric years, together with the high-resolution proxy climate studies, should enable more rigorous examination and integration of the archaeological and environmental data. This may facilitate direct correlation with specific geomorphological processes, such as erosional events, increased spring flow, ponding, localised lakes, dune incursions, terraces, palaeosols, etc.
Perhaps because of the opaqueness of the Upper Palaeolithic archaeological record there is a tendency (even unconsciously) to interpret the observed archaeological phenomena within a strictly ecological dictum. Granted, there is an almost innate connection between water resources and the location of sites especially in the marginal zones; still, it is quite obvious that once we get a little more detailed information, both environmental and archaeological, the picture generated is much more complicated. We should be wary of comprehending the dichotomy between the Levantine Aurignacian and the Early Ahmarian only from an ecological perspective (Kerry and Henry, and Williams herein). This approach may relate to the backgrounds of the researchers, their theoretical paradigms and the notion that archaeological variability, more than anything, reflects environmental adaptation. Whether right or wrong, most probably the truth (if at all), is somewhere in between, i.e., the environment and climatic conditions are part of the considerations which led Upper Palaeolithic peoples to create their own particular network of sites, routes of communication, etc. Given the complexity of human behaviour, sheer proximity to permanent water sources did not make individuals more ‘Ahmarian’ or less ‘Aurignacian’ and vice versa.
As noted long ago most Upper Palaeolithic sites tend to be quite ephemeral, presumably reflecting highly mobile, small bands exploiting extensive territories, not only in the more arid regions but also within the Mediterranean zone. One speculates about the scale of movements by specific bands within the south-central Levant. It seems unlikely that specific groups would have had seasonal rounds encompassing both the central Negev and the Lebanese coast (a span of some 400 km). Yet the particularistic techno-typological similarities of some of the central Negev (i.e., Boker BE) and Wadi Hasa (Thalab al-Buhayra) assemblages, less than 150 km apart, each featuring distinctive Ksar Akil scrapers and truncated blades, would appear to confirm the settlement model predicted by Larson (1979), prior to any research in southern Jordan. Interestingly this is notwithstanding various subsequent studies contrasting the so-called central Negev model with that for south-central Jordan (see references in Coinman, and Kerry and Henry herein).
Was the Middle to Upper Palaeolithic transition a gradual transformation developing locally, in situ, or was it a swift revolution, a fresh start with new ideas and/or people coming from elsewhere? Given the limits of space we will only briefly comment on the issues of Levantine Upper Palaeolithic origins and the Middle to Upper Palaeolithic and the Upper to Epipalaeolithic transitions as touched upon in the present volume.
Sarel and Ronen see a local continuity, while Tostevin strongly argues for a revolution, initially taking place in an as yet unknown locality, negating all possible ties between the local Middle and Upper Palaeolithic. Both approaches leave much to be desired. Sarel and Ronen avoid the thorny issue of the postulated mechanical admixture of the ‘transitional’ assemblage (and see above). Tostevin’s argument encapsulates what we consider to be an inherent flaw in terms of general techno-typological trends being translated into absolute values, which are then translated into ‘indices’ of change (see also remarks by Marks herein). There are also some major differences in the ecological settings of the assemblages he used for comparison – e.g., Kebara VI vs. Boker Tachtit. Tostevin does acknowledge that in terms of the assemblages he analysed, his were more or less ‘no other choice’ choices, still there is no justification for accepting the comparisons at face value.
If, for the sake or argument, we accept that there was no endemic, local Middle to Upper Palaeolithic transition in the Levant, the question then arises as to potential antecedents and directions of diffusion. Tostevin discusses the possibility for the origins of the initial Upper Palaeolithic package deriving from either the Nile Valley, or from somewhere between the Levant and central-eastern Europe (Anatolia?). Yet, as he himself admits, the evidence is sparse or non-existent. Those assemblages from central-eastern Europe that are similar to the initial Upper Palaeolithic ones in the Levant are unequivocally later in time. With respect to north Africa, the occurrence of chamfered pieces in the Dabban of Cyrenaica (which are at least superficially similar to those in Ksar Akil, Lebanon) is intriguing (see remarks by Marks herein), yet there is a notable absence of other relevant terminal Middle to initial Upper Palaeolithic occurrences. In light of the intensive research in the Nile Valley over the past four decades this absence seems to be significant and likely reflects the actual situation. Indeed the absence of contacts between the Nile Valley and the Levant throughout almost the entire span of the Upper Palaeolithic is conspicuous.
Nevertheless, recent genetic studies do lend credence to at least elements of the Out-of-Africa model (e.g., Semino et al. 2000). Thus it is quite possible that certain elements of the Upper Palaeolithic package do indeed derive from outside the Levant, perhaps combining with local developments as a catalyst for change. These may have resulted in more profound changes than the sum of the separate parts. In such a scenario some elements of the initial Upper Palaeolithic package may perhaps be profitably sought through another and equally, if not more, plausible alternative route for an Out-of-Africa model. This is by way of the Horn of Africa to Yemen/Saudi Arabia and thence up the Red Sea coast, an area that remains archaeologically terra incognita.
Still, the option of an in situ transition from Middle to Upper Palaeolithic in the Levant cannot be ruled out entirely. Lately, McBrearty and Brooks (2000), in a detailed compilation of early modern human behavioural correlates throughout Africa, argue for African origins of the Upper Palaeolithic phenomenon. They propose that these indicate continuity across the local Middle-Upper Palaeolithic boundary. Almost all of the evidence they cite appears equally early in the Near Eastern Middle Palaeolithic (ochre, symbolic/artistic endeavours, etc.). Other aspects of extra-somatic modern human behaviour (e.g., ‘out of the brain’ storage of symbolism, systematic spatial organization of activities) in Africa recently discussed by Wadley (2001) occur also in the Near Eastern Middle Palaeolithic.
The picture is, of course, even more complicated, as the transitional, Middle to Upper Palaeolithic assemblages display considerable variability among themselves – compare Umm el-Tlel, Ksar Akil, Emireh, Boqer Tachtit, Tor Sadaf, WHS 623X, etc. In this context of the transitional/initial Upper Palaeolithic the small Tor Fawaz assemblage (Kerry and Henry herein) is likely also relevant, as well as other as yet unpublished sites, e.g., Nahal Eilonim in the central Negev (personal observation).
What do the changes reflected by the transitional assemblages signify in behavioural terms? Indeed, to what extent are the changes even visible in the material record? What is the role of the posited changes in hafting and/or means of propulsion for projectile points? Certainly lighter and more symmetrical projectiles replace heavier, and sturdier ones. But is there also a shift to more discrete functional classes from Middle to Upper Palaeolithic lithic toolkits, together with a concomitant tendency to increased standardization in the morphological aspects of the toolkits (see Mellars 1989; Mellars et al. 1999 vs. Marks et al. 2001)?
Are the changes in knapping from the Middle Palaeolithic to Ahmarian really as dramatic as Chazan (herein) contends? In our view the conceptual changes of volumetric vs. surficial (see Géneste 1985; Boëda 1994, 1995a, b) are, perhaps, sometimes over-emphasized, reflecting the intrusive nature of the Upper Palaeolithic phenomenon in Europe that tends to contrast the differences in a starker light.
There appears to be a consensus that transition to modern human behaviour involved broader conceptual changes in the organization of activities, functional, symbolic and spatial. Evidence for definitely modern behavioural correlates already occur sporadically during the Middle and, even, Lower Palaeolithic – and various examples can be cited in the Near East, e.g., Berekhat Ram, Quneitra, and Qafzeh (Goren-Inbar 1986, 1990; Hovers et al. 1997; Marshack 1997). Ultimately, the issue revolves around the matter of systematic, patterned behaviours. The neural ‘wiring’ may have been in place for a considerable period of time, but the overall integration of the system as a whole took a while to become normative.
And what about the Upper Palaeolithic sequence itself? The relationship between the ‘Ahmarian’ and the ‘Aurignacian’ is still to be satisfactorily defined, let alone explained (see above and Williams herein). We do know by now that the Levantine Aurignacian sensu stricto derives from outside the region, though where from remains speculative. As data accumulate the picture becomes more complicated. Thus is it possible to hypothesise that there was more than one event that brought about the Upper Palaeolithic as we know it? Certainly the evidence Bergman presents in his paper on Ksar Akil, with two separate stratigraphic blocks of ‘Aurignacian’ type industries, appears to indicate two separate intrusive ‘pulses’.
Here we re-encounter the question of ‘Aurignacian’ definitions, not just in the immediate Levantine sphere but also further afield. While cursorily dealt with in a previous section of this chapter, it is important to raise yet again the issue of common denominations when dealing with local archaeological entities. While there may be increasing consensus with regards the issue of ‘modernity’, it should nevertheless be emphasized that whereas the Initial Upper Palaeolithic in the Levant concerns the ‘Ahmarian’ sensu lato, for most Europeans the equivalent relates to the ‘Aurignacian’ entity. Conversely, in the Levant the Aurignacian is a relative latecomer with a restricted geographical distribution.
We should bear in mind that connections between areas are and were dynamic, so that what moves in one direction at one point in time can subsequently return from the opposite direction in a totally different guise. The European Aurignacian may be explained as the sum of the ingredients accumulated on the long trek by the pioneering Upper Palaeolithic bands from Africa to Europe – thus the Aurignacian recognized in western Europe is actually autochthonic, as only there it exhibits all the classic Aurignacian attributes (Bar-Yosef 2000; Tostevin herein).
Progressing through time we encounter yet again the question of the nature and significance of transitions, this time the one to the Epipalaeolithic. Gilead (1984) has argued for continuity through to the Natufian entity, which proclaimed incipient sedentism and increasing social complexity. Indeed, few if any, would argue against the Early and Middle Epipalaeolithic representing anything but the continuity of a mobile hunter-gatherer foraging mode. Nevertheless the pace of change alters quite dramatically after ca. 22,000 bp (see Goring-Morris and Belfer-Cohen 1997). Overall, the behavioural changes from the Upper Palaeolithic through the Epipalaeolithic are generally accretional, as many characteristic elements occur sporadically earlier. As regards hunting practices there is a clear widening of the equipment repertoire, with the beginnings of systematic (?) trapping/snaring and sometimes fishing during the course of the Upper Palaeolithic (see Rabinovich herein). Bone and antler (and, perhaps speculatively, wood) were used more commonly as raw materials for hunting and processing implements. There are even hints for the beginnings of weaving of nets and baskets (Nadel et al. 1994), for storage and other purposes, while ostrich eggshell containers may also be documented (Gilead 1977; Goring-Morris 1987).
Yet, perhaps the most significant difference between the Upper Palaeolithic and the Epipalaeolithic relates to the clearly demarcated evidence for territoriality, reflected by the rapid changes in the emblemic style of microlith proclivities from ca. 22,000 bp (Fellner 1995; Belfer-Cohen and Goring-Morris 2002; and see Wadley 2000 for defining stylistic changes pertaining to social and environmental challenges). This seems to represent a conscious and intentional tightening of social networks, perhaps in response to the onset of the extreme conditions during the Pleniglacial in the Levant. As conditions deteriorated competition would have increased, leading to more restricted (refugia) territories (the coastal plain, the Rift Valley, and inland lacustrine basins) and a concomitant widening of the resource base (see Goring-Morris and Belfer-Cohen 1997). It is in this light that the nature of Ohalo II (and the Ein Gev sites) should be understood (Bar-Yosef 1970; and the papers of Nadel, Kaufman, and Olszewski herein).
It is the beginning and end of the Upper Palaeolithic that illustrate most clearly the significance of this period. Gone is the ambiguity of parallel human lineages sharing cultural affinities (Neanderthals and Anatomically Modern Homo Sapiens), and we are now facing Homo Sapiens sapiens as the sole representative of humanity all over the world.
The end of the Upper Palaeolithic brings us into the domain of rapid changes, the dominant tempo of eras to follow, and one can nearly touch the great transformations to come: sedentism, agriculture, pastoralism, incipient formalisation of belief systems, etc.
The fact that we are dealing with a single species (i.e., us), as well as the comparative chronological proximity to the present, justifies the use of ethnographic data and the wider human experience for comparative and inferential purposes – so that speculations are now based on sharing the basic characteristics (of individuals and groups) as fellow humans.
The above makes life more straightforward for researchers of the Upper Palaeolithic in comparison to earlier periods. Even so, only smatterings of Upper Palaeolithic behaviours can be deduced from the archaeological material at hand. One can despair and let go. Yet, there is also the possibility to try and tackle the archaeological vestiges, however fragmentary and sporadic they may be.
Whatever the origins of the Upper Palaeolithic and all that it encapsulates, there can be no denying that the Near East and the Levant in particular play a pivotal role concerning the origins and dissemination of modern human behaviours throughout the Old World. The contents of the present volume present summaries of the nuts and bolts perspective of the Upper Palaeolithic phenomenon. Yet it seems that whatever insights we have gained in the last few years of research, they have come about precisely through the investigation and integration of such mundane aspects. It is our firm belief that, though tedious and sometimes despairing, this is the only avenue open for archaeologists to appreciate the wider implications of modern human behaviour in all its diversity during the Upper Palaeolithic.