The Earlirst Prehistory of Cyprus Chapter 2

KALAVAS O S T E N TA R E V I S IT E D Five seasons of excavation were undertaken at the Aceramic Neolithic site of Kalavasos Tenta between the years 1976 and 1984. The site lies adjacent to the southern coast, between Limassol and Larnaca. Substantial stone and mud-brick architecture was excavated on the top of the site and on the southern slopes, comprising domestic buildings clustered in a small village on a strategically placed natural hill. In comparison with nearby Khirokitia Vouni the site was expected to date to the sixth millennium B.C.E. However, a series of radiocarbon assays included a number of dates in the eighth and seventh millennia B.C.E. The occupation of the site was divided into five chronological periods, the earliest of which (Period 5) was marked only by lightly built structures. Analysis of the stratigraphy, radiocarbon dates and some of the artifacts strongly suggests that Period 5 at Tenta is contemporary with the earlier Aceramic phases at Parekklisha Shillourokambos. T he Aceramic Neolithic site of Kalavasos Tenta is located in the lower reaches of the Vasilikos river valley, approximately half way between Limassol and Larnaca, in the southern coastal region of the island. It consists of a settlement on a small natural hill, 3.2 km north of the coast. Excavations were initially undertaken by P. Dikaios in 1947, but his campaign was brief and no full report was ever published. More extensive excavations were undertaken by the Vasilikos Valley Project, directed by the writer, in the years 1976–1979 and 1984. The first volume of the final excavation report, detailing the site and its setting, the architecture and chronology, was published in 1987 (Todd 1987); the second volume covering the artifacts, human burials, flora, fauna and site territory analysis, is about to be sent to press at the time of writing. The site is now open to visitors, and a summary guide (Todd 1998) has been published. The recent excavations were centered on the top of the site and the lower southern slopes. Substantial architectural remains of stone and mud-brick were encountered in both areas (figs. 1–2), in some cases with up to three superimposed building levels. Outside these areas, small soundings generally revealed a lack of well preserved architecture, and in some cases remains of the Ceramic Neolithic phase. This latter phase is marked on the site by areas of pits containing ceramics and other artifacts, but no contemporary standing structures were encountered. Although one of the original aims of the excavation was to try to find evidence of continuity of occupation from the Aceramic to the Ceramic Neolithic phase, all of the information retrieved from the excavation pointed in the opposite direction. There seems to have been a lengthy gap between these two main phases of utilization of the site. The focus of this paper is on the earlier, Aceramic settlement, and no further consideration will be given here to the later phase. The site of Tenta is strategically located on a small natural hill which affords an excellent view up the valley toward the Troodos Mountains and down the valley toward the coast. It overlooks an easy crossing of the river, lying on the major east-west route along the southern coastal fringe of the island Fig. 1. Plan of major architectural and other features at Tenta. Mud-brick walls are shown in solid black. Fig. 2. General view of upper part of Tenta with Structures 11 and 42 in foreground, from the southsoutheast. which is still very important today. The circular, or at least curvilinear domestic structures of the village clustered on the upper reaches of the hill, surrounded, in an early phase of the settlement, by a substantial encircling wall and a ditch cut in the havara (secondary limestone ) immediately outside it. While the factors governing the choice of location of the settlement are unknown, the desire for security clearly played a significant part. Kalavasos Tenta is not the only Aceramic Neolithic settlement known in the southern reaches of the Vasilikos valley. An extensive field survey has been undertaken of the valley from the Kalavasos Dam down to the coast, the results of which, in conjunction with information derived from the construction of the Nicosia-Limassol highway, have provided evidence of Aceramic Neolithic occupation at four localities in addition to the settlement at Tenta. The most southerly site named Mari Mesovouni, located 1 km north of the coast, comprised a village settlement on a steep-sided, flat-topped hill in a strategic position overlooking the coast. The proximity of the site to Tenta is surprising, but the two sites may not have been occupied at exactly the same time. The site has unfortunately been destroyed in recent years and the hill quarried away. During the construction of the bridge over the Vasilikos river for the new Nicosia-Limassol highway, one of the foundation trenches for the bridge supports revealed a small ash-filled pit at a depth of 5.5 m below the present flood plain, in association with a small quantity of chipped stone tools. A radiocarbon date from the pit suggests use of this part of the river bank in a late phase of the Aceramic Neolithic. The extent of the site is unknown. To the north of Kalavasos village, remains of probably two more Aceramic Neolithic settlements have been located by the field survey, but the fieldwork has not yet been completed. The information presently available indicates that Aceramic Neolithic settlers had penetrated the northern part of the valley well to the north of Kalavasos, and occupation of this phase was by no means restricted to the coastal zone. This pattern of settlement must be borne in mind in any consideration of the history of the earliest settlement or utilization of the valley. AT TH E OTH E R E N D O F TH E SEQUENCE: THE CYPRIOT ACERAMIC NEOLITHIC AS SEEN FROM KHIROKITIA The 7th millennium B.C.E. site of Khirokitia Vouni illustrates the end of the cultural process which began with Parekklisha Shillourokambos. The purpose of this paper is to present the main discoveries of the excavations at Khirokitia and to concomitantly assess the differences between the beginning and the end of the sequence. The scale of communal projects recorded in all phases of the settlement suggests the existence of a well-structured society. This is the focus of the present study, which offers a tentative analysis based on the organization of domestic space and funerary customs. T he preceding contributions illustrate how greatly our knowledge of Cypriot prehistory has pro gressed in these last years. The excavations at Parekklisha Shillourokambos have cast light on the initial colonization of the island, hence on the beginning of the Cypriot Aceramic Neolithic period. In contrast, the site of Khirokitia Vouni (hereafter Khirokitia) belongs to the end of the Aceramic sequence in the 7th millennium cal. B.C.E., which corresponds to the apogee of this period’s cultural development just prior to its collapse. The Cypriot Aceramic civilization had evolved in the distinctive, closed environment of an island and once the founding wave of settlement ebbed, all relations with the outside world seem to have ceased. Khirokitia differs in many respects from Shillourokambos—by its location, its massive architecture, its chipped and ground stone industries, the scarcity of obsidian, and its faunal assemblage. Addressing the issues raised by these points, I shall attempt to present the view of Aceramic Neolithic Cyprus as seen from Khirokitia. The site was discovered in 1934 by Porphyrios Dikaios (Dikaios 1953) who, on behalf of the Department of Antiquities, conducted six field campaigns between 1936 and 1946. The exploration of the site was resumed in 1977 by a French mission (Le Brun 1984, 1989, 1994, in press b) sponsored by the Centre National de la Recherche Scientifique and the French Ministry of Foreign Affairs. The first objective was to continue the study of the process of neolithization in an insular context, which had begun with the excavation of the small contemporary settlement of Cape Andreas Kastros (LeBrun 1981). The second was to try to answer questions concerning the site’s internal organization and location in space. Khirokitia is situated about 6 km from the southern coast of the island, in the Maroni river valley, which stretches from the Troodos Mountain range to the sea. The Neolithic settlement covers an area of approximately one and a half hectares on the slope of a hill. Khirokitia thus fits the typical pattern of Aceramic sites Fig. 1. Khirokitia. General plan of the site. characterized, with the notable exception of Shillourokambos, as being located on hills, promontories, or some other naturally protected feature. The hill chosen for the Neolithic settlement lies within a sharp bend of the river, which protects it on the north, east and southeast (fig. 1). When the site was occupied, the river had a more substantial rate of flow than at present, as shown by recent research undertaken on the southeast flank of the settlement along the bank where several structures suffered from floods and shifts in the course of the river bed. This natural protection, however, does not exist to the west where the village is open to the neighboring hills. In place of such natural defenses, a long, linear stone structure was built, crossing the settlement from north to south, providing artificial protection (fig. 1, structure 100). It is not within the scope of this article to provide a detailed discussion of the work carried out on this structure or of our observations resulting from this research: suffice it to say that it is no longer viewed as the “main road” of the village, as posited by Dikaios, but as a perimeter wall. When the settlement spread to the west onto previously unoccupied land, the same pattern was repeated and the development was accompanied by the simultaneous building of a new boundary in the form of an impressive stone wall (fig. 1, structure 284). The concept of an intentionally enclosed, built space, involving the construction of an artificial boundary implies obvious architectural skill. The architectural endeavor, relaying on the use of stone and mud in the form of mudbricks and pisé, expresses itself in two ways: first in the creation of public works for the good of the community and second in the development of domestic architecture. Whichever category a structure at Khirokitia falls into, the architecture at the site is always substantial. Such solidity is concomitant with sophistication, a point emphasized by the way in which this enclosed world communicated with that outside, as represented by the access points incorporated into the village enclosure wall. In order to connect two surfaces located on the same level, but separated by a barrier, it is simply a matter of making an opening in the barrier. At Khirokitia, however, the process it not so simple as the hill slope complicates the issue. The two surfaces that need to intercommunicate are separated by a difference of more than 2 m and the thick wall only adds to the difficulty. The solution adopted to overcome this problem was to construct a stairway of twelve steps, comprising three flights at right angles to one another incorporated within a stone structure built against the exterior face of the perimeter wall (figs. 2, 3). As for domestic architecture, the basic architectural unit is a structure with a circular ground plan and not a rectangular one as seen in most contemporaneous villages in the Near East. This cultural choice remains constant throughout the occupation of the site and buildings are round from the earliest level to the latest. The exterior diameters of structures vary from between 2.3 m for the smallest and 9.8 m for the largest. Before construction began, the ground surface was roughly prepared to support the walls, which were then built directly onto the underlying deposits, usually without a foundation trench. Floors were covered with mud plaster also placed directly on the underlying deposits, or more rarely, on a layer of stones. The mud plaster was worked from the floor up against the interior face of the wall where it also served as a base for painted mural decoration. In the present article I do not plan to discuss the issue of roofing, but suffice it to say that all the data recovered from Khirokitia indicate the use of flat, terraced roofs. Each of these circular units defined a habitation area and was a component of a larger domestic space: the house. Evidence suggests that the space within individual circular units may sometimes have been increased by a kind of loft resting on pillars. The interior space could be without features, or was subdivided by low partition walls and platforms, usually trapezoidal in shape. Domestic installations such as fireplaces, pits, or basins are also found. The ideal house form, as reconstructed on the basis of information yielded by the excavation, may be defined as a compound with several units grouped around an open space acting as a sort of small, inner “courtyard” containing a grain grinding installation. AC E RAM I C N E O LI T H I C PLAN T REMAINS IN CYPRUS: CLUES TO THEIR ORIGINS? Plant remains recovered from Aceramic Neolithic sites in Cyprus are compared to the predominant plants found on PPNA and PPNB sites in the Near East (Israel, Jordan, Syria and Turkey) in an attempt to identify a plant or plants that may provide a clue to the origins of the early farming communities of Cyprus. The species identified in the Cypriot Aceramic Neolithic are all found in contemporary sites throughout the Near East and Turkey and do not point to a particular area of origin. If annual species of wild rye were to be found it may provide a more definite link with southern Turkey, but given the paucity of sites that have produced this plant, such a connection would be highly speculative. A lthough one site is now known to predate the Aceramic Neolithic of Cyprus (Simmons this volume), there is, as yet, no evidence for a continuous sequence of occupation on the island into the earliest Aceramic Neolithic. As a result of this lacuna it is impossible identify an indigenous development of agriculture and village life, and the origin of these early farming communities remains unclear. One avenue of inquiry concerns the crops that were introduced by the immigrants. If it were possible to identify among the Cypriot sites a similar assemblage of crops, perhaps a species with a narrowly focused distribution in the Near East, it might be feasible to locate the area of origin of the Cypriot Aceramic Neolithic culture. In this paper the data on the plant remains recovered from Aceramic Neolithic sites in Cyprus are presented and compared to the predominant plants found on PPNA and PPNB sites in the Near East (Israel, Jordan, Syria and Turkey) in an attempt to identify a plant or plants capable of providing a clue to the origins of the early farming communities of Cyprus. T H E NAT U R E O F T H E E V I D E N C E Before embarking on any discussion of plant remains it will be helpful to discuss some of the problems associated with comparative studies in palaeoethnobotany in general and with analysis of plant remains in Cyprus in particular. Attempts to compare plant remains from sites in the Near East with those from Cyprus require consideration of the differences in methods of recovery and sampling, taphonomic processes, and preservation on the various sites. Some have been systematically water sieved from the outset with the purpose of recovering botanical material. Others have been sampled post-excavation, from the balks of the trenches. Still others were not water sieved at all, having been excavated prior to the introduction of such procedures, and the recovered plant remains were only those that could be seen during excavation. Thus it is impossible to compare quantities of remains from one site to another as different recovery methods result in different quantities of material. Taphonomic processes must also be taken into account since sampling only deposits where carbon is evident, such as hearths, may not recover all types of plants utilized or preserved on the site. Floor deposits, dumps, or pits may have quite different assemblages representing different plant processing steps, disposal or storage areas. Samples taken only from balks generally have little contextual information attached to them. The problems of differential preservation are difficult to assess as the lack of material may be due to poor preservation conditions or true lack of deposition. Poor preservation is likely on fairly shallow sites. In the Mediterranean summer the ground dries sufficiently to dry out carbonized remains in the upper 50 cm to 1 m of sediment. When the rains come in October and November, water percolating into the cracks of the dried ground is quickly absorbed by the carbon that splits and fragments. Subsequent tectonic movement and bioturbation further break up the material and disperses it through the deposit. When excavated, the carbonized plant remains appear as small flecks of carbon that are not recovered well during flotation and are difficult to identify. On Cyprus, the problem of poor preservation (or lack of deposition) varies from site to site. At Khirokitia preservation is quite good and large quantities of plant remains have been recovered from many deposits, even though the site has not been intensively water sieved. At Kalavasos Tenta, just 6 km away, on the other hand, systematic water sieving was carried out from the beginning of the project and nearly every deposit excavated was sampled (Hansen forthcoming). However, relatively few remains were recovered compared to Khirokitia. More than 400 samples were processed at Kalavasos Tenta yet only 165 of the Aceramic Neolithic samples yielded plant remains totaling a mere 1076 items. At Khirokitia, 241 samples have been examined with approximately 9000 items identified1 (Hansen 1989, 1994; Miller 1984; Waines and Stanley Price 1977). The reason for this difference in the two sites is somewhat puzzling since the sediments, calcareous kafkalla, and the depth of deposits from the surface, 3–4 m, are similar at both. The site of Parekklisha Shillourokambos (Guilaine this volume) is very close to the surface and has so far produced no plant remains, despite water sieving attempts. Thus, the botanical evidence for early farming or plant use on Cyprus is rather poor, with only five Aceramic Neolithic sites having produced plant remains so far: Kalavasos Tenta, Khirokitia, Cape Andreas Kastros, Kholetria Ortos and Dhali Agridhi (Table 1, fig. 1). Dhali Agridhi has not been included in the following discussion, however, because of the paucity of remains at that site. Utilizing the data from the other four sites, we can begin to look at the types of plants recovered and compare them to some of the contemporary Near Eastern sites. C Y P R I O T A C E R A M I C N E O L I T H I C P LA N T R E M A I N S Table 1 provides some of the data for the sites in Cyprus that have produced sufficient plant material to allow an examination of early farming. Only the primary domesticated cereals, legumes, and a few of the dominant wild plants recovered from these sites are included in Table 1 and fig. 1, but it should be noted that none produced large quantities of “weed” seeds with the exception of ryegrass (Lolium sp.). This may be due to the fact that the crops were largely processed outside of the areas excavated, perhaps beyond the confines of the village, or alternatively, that there were relatively few other weeds associated with the crops. Fig. 1 shows the ubiquity of some key species at the four sites that produced sufficient material. Ubiquity, or presence analysis, looks at the number of samples in which a particular taxon is represented at a site and expresses that as a percentage of the total number of samples (Popper 1988). In this 1 At this writing, not all of the Khirokitia material has been analyzed or counted. Table 1. Ubiquity of selected species from Cypriot Aceramic Neolithic sites.* Species Einkorn Emmer Barley Lentil Pea Bitter Vetch Vetch Fig Olive Pistachio Ryegrass Cape Andreas 23 samples # % 18 78 20 87 15 65 21 91 3 13 0 0 1 4 2 9 7 30 3 13 20 87 Khirokitia 241 samples # % 97 40 64 26 34 14 94 39 1 <1 3 1 15 6 34 14 8 3 43 18 82 34 Tenta 165 samples # % 31 19 26 16 35 21 60 36 5 3 5 3 12 7 11 7 0 0 17 10 3 2 Ortos 40 samples # % 9 22 3 7 8 20 32 80 1 2 1 2 5 12 2 5 0 0 19 47 29 72 way it is possible to look at the relative representation of the various species at different sites. This should not be interpreted as equivalent to relative importance of different species, however, since we are not comparing similar contexts from site to site and the types of remains may, in part, depend on the type of context. For example, a hearth might be expected to contain a different assemblage of taxa than a storage pit or midden. Nonetheless, ubiquity provides a useful indicator of which species are better represented on these sites. Clearly, einkorn (Triticum monococcum), emmer (Triticum dicoccum), lentil (Lens culinaris), and ryegrass (Lolium sp.) are well represented on all four sites. It should be noted that the relatively low representation of ryegrass at Kalavasos Tenta might be due to poor preservation. The high representation of unidentified grasses (Gramineae) at this site may include ryegrass caryopses that were not identifiable as such. As already noted, ryegrass is a weed of cereal crops, but it is so abundant at Cape Andreas Kastros that van Zeist (1981) was led to wonder whether it might have been utilized as a food resource. It is impossible, however, at this point to make that claim for any of the sites until a more thorough analysis of the contextual data has been undertaken. Among the cereal crops at all sites, it appears that emmer and einkorn are fairly equally represented, with barley appearing less often. Whether these were individual crops or grown as maslin is difficult to say, although at Khirokitia relatively pure deposits of both einkorn and emmer grain and chaff have been recovered, suggesting that at this site they were separate crops. It is interesting that no * References for Tables and Figures: Abu Hureyra – de Moulins 1997; Ain Ghazal – Rollefson and Kafafi 1994; Rollefson and Simmons 1988, 1985; Rollefson et al. 1985; Aswad – van Zeist and Bakker-Heeres 1979, 1982; Basta – Gebel et al. 1988; Beidha – Helbaek 1966; Bouqras – van Zeist and Waterbolk-van Rooijin 1985; Cafer Huyuk – de Moulins 1997; Can Hasan III – Hillman 1972, 1978; Cape Andreas Kastros – van Zeist 1981; Cayönü – van Zeist 1972; Dhali Agridi – Stewart 1974; El Kowm II – de Moulins 1997; Ghoraifé – van Zeist and Bakker-Heeres 1982; Hacilar – Helbaek 1970; Jericho – Hopf 1983; Khirokitia – Waines and Stanley-Price 1974; Miller 1984; Hansen 1989, 1994; Mureybit – van Zeist 1970; van Zeist and Bakker-Heeres 1984b; Nahal Hemar – Kislev 1988; Nahal Oren – Legge 1986; Noy et al. 1973; Netiv Hagdud – Kislev 1985b; Ortos – O’Brien and Hansen in press; Ramad I and II – van Zeist and Bakker-Heeres 1982; Ras Shamra Vc – van Zeist and Bakker-Heeres 1982; Tenta – Hansen forthcoming; Yifth’el – Kislev 1985a T H E W O R L D A RO U N D C Y P R U S : F R O M E PI - PALE O LI TH I C F O RAG E R S T O T H E C O LLAPS E O F TH E PPN B C IVI LI ZATI O N This paper is an overview of the Epi-Paleolithic and early Neolithic of the Levant and Anatolia, based on the calibrated B.C.E. chronology as calculated using INTCAL98 (or Calib 4.1). Despite certain chronological ambiguities, and with reference to proxy climatic data, the cultural sequence is described against a background of fluctuating environmental conditions. Among the most decisive paleo-climatic changes were the Younger Dryas (11/10.800–9.600/ 9.500 B.C.E.), at the end of which intentional cultivation was established in the Levant, and the climatic crisis of ca. 6400/6200 B.C.E., which marked the collapse of the PPNB civilization. The onset of the cultural changes occurred with the appearance of the Early Natufian nonegalitarian society. The cold and dry conditions in the Levant during the Younger Dryas led to increased mobility and efforts to negotiate equality amongst members of various groups. The systematic cultivation of cereals and legumes resulted within a short time in major social change. Most prominent was the rapid population increase during the PPNA, which led to the emergence of the PPNB civilization—an expanding agglomeration of several social entities (tribes?). A variable settlement pattern characterizes each of the archaeologically identified territories of the PPNB civilization, where large central villages differ from ceremonial centers, hamlets, and sacred locales. The interaction sphere incorporates the marginal semi-sedentary and mobile foragers, as well as special quarry sites (e.g., for obsidian). Long distance connections are marked by the exchange and trade of chlorite stone bowls, obsidian, stone bangles, and marine shells. The final collapse of this civilization is attributed to an abrupt climatic change that is well-marked in the date sequences of both speleothems and pollen cores, an event recognized globally. The population of the large settlements in part dispersed and founded small hamlets and farmsteads, and in part moved westward and northward into more humid regions. T his paper is not intended as a full survey of our knowledge concerning the late Epi-Paleolithic and Early Neolithic of the eastern Mediterranean, a vast region within which Cyprus is located. Various summaries are already available (e.g., Aurenche and KozŒowski 1999; Banning 1998; Bar-Yosef and Meadow 1995; Harris 1998; KozŒowski 1999; Özdo©an 1997). It is instead aimed at those interested in the origin of the early colonizers of Cyprus. As these people crossed the Mediterranean Sea from the mainland, it is appropriate to try and gain some insights in this home area concerning the transition from late forgers to early farmers-herders. This was a complex socioeconomic process. It began when foragers implemented intentional cultivation of wild cereals and legumes, and culminated in the establishment of fully-fledged crop agriculture. Penning and tending wild populations of goat, sheep, cattle, and pig occurred later. Domesticated forms appeared in due course after a period of several centuries. Following Childe, we incorporate the intricate, step-wise economic changes, which were intertwined with major upheavals of social structures, under the term the “Neolithic Revolution.” The major cultural outcome of this process is labeled here, following the traditional terminology, as the “PPNB civilization.” I believe that the use of such a highly charged term is justified in view of the striking archaeological discoveries of major PPNB sites during the last two decades (Özdo©an and Basgelen 1999). Moreover, given the demonstrated high level of technology achieved during this period (including building two floor houses, stone tool making, early use of metals, emergence of pottery, intensive use of fibers, extensive trade and exchange, and the like), it is not surprising that seafaring became part and parcel of societal ventures (e.g., Cherry 1990). The first visitors evidenced in the prehistory of Cyprus were late Epi-Paleolithic foragers (Simmons, this volume). They were followed by Early Neolithic villagers (Guilaine, this volume; Peltenburg et al., this volume), bringing the issues involved in this island’s colonizations to the forefront of prehistoric research. Although most of the Mediterranean islands are not as far away from the continent as are those of the Pacific Ocean, similar questions concerning the technology of seafaring have been raised (e.g., Kirch 1997). The questions are often ‘when,’ ‘who,’ and ‘why.’ Dating the first colonization provides the answer to ‘when.’ For the ‘why‘ we may consider curiosity as a basic instinct of Homo sapiens sapiens. Development of adequate technology enabled humans to reach previously unattainable islands. Suggesting ‘who’ were the first to land on an island necessitates the identification of the original homeland of the travelers. At that point in the discussion we enter the treacherous field of equating archaeological finds with people. However, tracking the cultural markers can be done if the information from both areas, the homeland and the new land, is well documented. Archaeologists working in the Near East have long been busy sourcing the movements of specific raw materials, finished objects, seeds for crops, and even live animals. In the case of the early colonization of Cyprus, the review of neighboring continental regions could greatly assist in determining from where the early Cypriots arrived. A survey of the archaeological evidence may provide insights concerning the advantages and the difficulties that faced those who lived in the ‘old homeland’ and decided to cross the sea to settle in a ‘new land.’ Recognizing the nature and kind of immigrations is not solely a problem of archaeologists. Bio-geographers and sociologists study the same issues, and to an equal extent the impacts caused by colonizing species, including people, on island ecosystems (e.g., Cherry 1990; Kirch 1997 and references therein). These systems are often ecologically unique as they have definite boundaries and a lower degree of biodiversity, which is therefore prone to even minor disturbing effects, such as the introduction of new crops or animals, overkill of local fauna, and the like. The focus of this paper, however, is not the island of Cyprus itself but the source areas for its prehistoric populations. Using the archaeological records of late Epi-Paleolithic and early Neolithic periods of the Near East—information concerning the material culture components such as house and tool types, vegetal species, and animals, in conjunction with tentatively reconstructed cosmologies—it may be possible to characterize potential areas from which the colonizing groups sailed. THE CHRONOLOGY This review proceeds from a short survey of the chronology of the Terminal Pleistocene through the first half of the Holocene, based on calibrated radiocarbon dates. It incorporates the main archaeological entities of the late Epi-Paleolithic and the early Neolithic. The latter is also known as the PrePottery (or Aceramic) Neolithic, subdivided following Kenyon’s terms as PPNA and PPNB. The current dating for these subdivisions is presented in fig. 1. This version provides the reader with the Fig. 1. The calibrated chronology of the Levant. timescales of uncalibrated and calibrated B.P. and B.C.E. dates. INTCAL98, or Calib. 4.2 (Stuiver et al. 1998), was employed in creating fig. 1. In order to keep the chronology in a simple form, I made the following choice. A radiocarbon date often has more than one optional calibrated date, especially if calculated within two S.D.(standard deviations) (95% of probability). However, I used the median figure produced by the software, as calculated with only one standard deviation. Obviously this procedure does not mean that the calibrated date is calendrically accurate. The cultural chronology of the late Epi-Paleolithic and Neolithic periods in the eastern Mediterranean—based on stratified and radiocarbon dated sites—is relatively well-known. The increasing number of published 14C readings, produced either by the traditional method or by Accelerator Mass Spectrometer, facilitates long distance correlations within this vast region. In addition, the realization that the AMS measurements can use very small samples has encouraged archaeologists to submit seeds for direct dating. However, in the enthusiasm accompanying recognition of the amount of real time that is allocated to cultural entities, we should not forget that the need to record and scrutinize the context APPENDIX: A NOTE ON DATING T here are many different ways of telling time. Even when we use presumed calendar years, there is some variation on what dates actually mean. This can be a particularly vexing problem when dealing with ancient literate societies that had their own calendar systems, but even in prehistoric archaeology, “absolute” dates in fact are not absolute. An absolute archaeological timescale must date from, or to, a fixed point in time. Generally, archaeologists refer to time within the Christian calendar, which by convention is taken as the birth of Christ, supposedly in the year A.D. 1. In this calendar B.C. represents “before Christ,” while A.D. represents “Anno Domini,” or “the year of our lord.” In some instances, to avoid cultural insensitivity, B.C. may be referred to as “B.C.E.,” meaning “Before the Common Era,” and A.D. may be represented as “C.E.,” or “Common Era.” In other cases, some archaeologists avoid the B.C./A.D. dichotomy altogether, preferring instead to refer to dates as “b.p.,” or “before present,” with “present” being established as 1950. In this volume the “B.C.E.” convention, as well as b.p have been used. There is, unfortunately, often a large degree of inconsistency in how dates are presented in terms of upper case (e.g., B.C.) or lower case (e.g., b.p.). Technically speaking, upper case is the “scientific” convention preferred by radiocarbon laboratories, while lower case is the “historical” convention often preferred by archaeologists. As noted, however, this often is inconsistently presented. In this volume, we have attempted consistency by using upper case for B.C./A.D. dates and lower case for B.P. dates. Further clarification is required on the presentation of dates in this volume when they are based on radiocarbon chronologies, as are most prehistoric time frames. Radiocarbon dating, developed in 1949 by the physicists J. Arnold and W. Libby, represented a true revolution in prehistoric dating. With recent developments, virtually anything that once was living can be radiocarbon dated, although certain materials, charcoal especially, are better than others, such as shell for example. Different materials present different problems in reliability, but the radiocarbon method has overall proven to be highly effective for dating samples up to approximately 50,000 years old. One must realize, however, that radiocarbon determinations are, in fact, statistical approximations. That is to say, when a laboratory calculates a radiocarbon determination, as with any radiometric dating method, the results are always presented with a standard deviation or sigma figure, which is a statistical plus or minus factor. One standard deviation, which is the convention with which dates are presented, represents only a 68% statistical certainty that the given date falls within the range represented by the plus/minus factor. If the standard deviation is doubled, however, the statistical certainty jumps to approximately 95%, and if tripled, it is around 99%. Given these statistical uncertainties, it is clear that the smaller the standard deviation, the better. For many years, archaeologists were content with their radiocarbon determinations. In the 1970s, however, additional research questioned the reliability of radiocarbon dating. Dendrocrhonology (treering dating), which is one of our most accurate dating methods, often reliable to the calendar year, demonstrated both the inaccuracy and provided the solution to the problem. When some ancient Bristlecone pines that had been tree-ring dated were also radiocarbon dated, the results were surprising in that they were consistently younger, especially for trees dating to 1200 B.C.E. or earlier. The reason for this variation, it was determined, was that a basic assumption for radiocarbon dating—that the concentration of 14C in the atmosphere was constant through time—was shown to be incorrect. We now know that it has varied, largely due to changes in the earth’s magnetic field so that the concentration of radiocarbon in the atmosphere has varied in the past, rather than remained steady, as previously assumed. This meant that in many cases, the radiocarbon determinations were in fact often too young. Fortunately, it is possible to correct this error back to around 12,000 years ago. The correction is done by calibrating the radiocarbon determinations with tree-ring chronologies. Thus in contemporary archaeology, many radiocarbon determinations are presented in both their uncalibrated form, usually as a B.C./A.D. or B.P. date, and in their calibrated form, which is more accurate in relation to an actual calendar date. The calibrated dates are often expressed as “cal. B.C.,” or, in this volume, as “cal. B.C.E.” Note that calibrated dates are not presented as a mean figure with the + error. Additional information on dating may be found in many publications, such as Fagan (1999: 82– 90) or Renfrew and Bahn (2000:117–70), as well as from laboratories, such as Beta-Analytic (www.radiocarbon.com).