raqefet cave (2004 – 2008 seasons of excavation)

edited by

Ofer Bar-Yosef&

François R. Valla

InternatIonal Monographs In prehIstory

Archaeological Series 19

Natufian Foragers in the Levant

Terminal Pleistocene Social Changes in Western Asia

Library of Congress Cataloging-in-Publication Data

Natufian foragers in the Levant : terminal Pleistocene social changes in Western Asia / edited by Ofer Bar-Yosef & François Valla. pages cm. -- (Archaeological series / International Monographs in Prehistory ; 19) Papers from a symposium held in 2009. Includes bibliographical references. ISBN 978-1-879621-45-9 (paperback : acid-free paper) -- ISBN 978-1-879621-46-6 (hard cover : acid-free paper) 1. Natufian culture--Middle East--Congresses. 2. Hunting and gathering societies--Middle East--Congresses. 3. Pleistocene-Holocene boundary--Congresses. 4. Social archaeology--Middle East--Congresses. 5. Social change--Middle East--History--To 1500--Congresses. 6. Excavations (Archaeology)--Middle East--Congresses. 7. Middle East--Antiquities--Congresses. I. Bar-Yosef, Ofer. II. Valla, François Raymond. GN774.3.N38N28 2013 306.3›640956--dc23 2013035516

© 2013 by International Monographs in PrehistoryAll rights reserved

Printed in the United States of AmericaAll rights reserved

Paperback:ISBN 978-1-879621-45-9Hard Cover:ISBN 978-1-879621-46-6

This book is printed on acid-free paper. ∞

International Monographs in PrehistoryAnn Arbor, MichiganU.S.A.

Printed with the support of the American School of Prehistoric Research (Peabody Museum, Harvard University)

Table of Contents

List of Contributors ................................................................................................................... vii

Preface – The Natufian Culture in the Levant: Twenty Years LaterOfer Bar-Yosef and François R. Valla ...............................................................................xv

Acknowledgements ....................................................................................................................xix

Northern Levant

Natufian Lifeways in the Eastern Foothills of the Anti-Lebanon MountainsNicholas J. Conard, Knut Bretzke, Katleen Deckers, Andrew W. Kandel, Mohamed Masri, Hannes Napierala, Simone Riehl and Mareike Stahlschmidt ..............................1

The Natufian of Moghr el-Ahwal in the Qadisha Valley, Northern LebanonAndrew Garrard and Corine Yazbeck ..............................................................................17

The Natufian of Southwestern Syria Sites in the Damascus ProvinceKurt Felix Hillgruber ........................................................................................................28

The Natufian Occupations of Qarassa 3 (Sweida, Southern Syria)Xavier Terradas, Juan José Ibáñez, Franck Braemer, Lionel Gourichon and Luis C. Teira ...................................................................................................................................45

The Early Natufian Site of Jeftelik (Homs Gap, Syria)Amelia del Carmen Rodríguez Rodríguez, Maya Haïdar-Boustani, Jesús E.González Urquijo, Juan José Ibáñez, Michel Al-Maqdissi, Xavier Terradasand Lydia Zapata ..............................................................................................................61

Fish in the Desert? The Younger Dryas and its Influence on the Paleoenvironment at Baaz Rockshelter, Syria

Hannes Napierala .............................................................................................................73

Preliminary Results from Analyses of Charred Plant Remains from a Burnt Natufian Building at Dederiyeh Cave in Northwest Syria

Ken-ichi Tanno, George Willcox, Sultan Muhesen, Yoshihiro Nishiaki, YousefKanjo and Takeru Akazawa..............................................................................................83

Southern Levant

El-Wad

Spatial Organization of Natufian el-Wad through Time: Combining the Results of Past and Present Excavations

Mina Weinstein-Evron, Daniel Kaufman and Reuven Yeshurun ...................................88

iv

The Last Natufian Inhabitants of el-Wad TerraceNoga Bachrach, Israel Hershkovitz, Daniel Kaufman and MinaWeinstein-Evron..............................................................................................................107

Domestic Refuse Maintenance in the Natufian: Faunal Evidence from el-Wad Terrace, Mount Carmel

Reuven Yeshurun, Guy Bar-Oz, Daniel Kaufman and Mina Weinstein-Evron ...........118

Natufian Green Stone Pendants from el-Wad: Characteristics and Cultural ImplicationsDaniella E. Bar-Yosef Mayer, Naomi Porat and Mina Weinstein-Evron ......................139

Eynan

The Final Natufian Structure 215-228 at Mallaha (Eynan), Israel: an Attempt at Spatial Analysis

François R. Valla, Hamoudi Khalaily, Nicolas Samuelian, Anne Bridault, Rivka Rabinovich, Tal Simmons, Gaëlle Le Dosseur and Shoshana Ashkenazi ....................146

A Study of two Natufian Residential Complexes: Structures 200 and 203 at Eynan (Ain Mallaha), Israel

Nicolas Samuelian ..........................................................................................................172

Graves in Context: Field Anthropology and the Investigation of Interstratified Floors and Burials

Fanny Bocquentin, Teresa Cabellos and Nicolas Samuelian ........................................185

Obsidian in Natufian Context: the Case of Eynan (Ain Mallaha), IsraelHamoudi Khalaily and François R. Valla ......................................................................193

Flint Knapping and its Objectives in the Early Natufian. The Example of Eynan- Ain Mallaha (Israel)

Boris Valentin, François R. Valla and Hugues Plisson with the collaboration of Fanny Bocquentin ...........................................................................................................203

Searching for the Functions of Fire Structures in Eynan (Mallaha) and their Formation Processes: a Geochemical Approach

Ramiro J. March ..............................................................................................................227

Avifauna of the Final Natufian of EynanTal Simmons ....................................................................................................................284

Bone Ornamental Elements and Decorated Objects of the Natufian from MallahaGaëlle Le Dosseur and Claudine Maréchal ...................................................................293

Reconstruction of the Habitats in the Ecosystem of the Final Natufian Site of Ain Mallaha (Eynan)

Shoshana Ashkenazi .......................................................................................................312

v

Southern Levant - other sites

Wadi Hammeh 27: an open-air ‘base-camp’ on the Fringe of the Natufian ‘homeland’Phillip C. Edwards, Fanny Bocquentin, Sue Colledge, Yvonne Edwards, Gaëlle Le Dosseur, Louise Martin, Zvonkica Stanin and John Webb ...........................................319

Art Items from Wadi Hammeh 27Janine Major ...................................................................................................................349

The Final Epipaleolithic / PPNA site of Huzuq Musa (Jordan Valley)Dani Nadel and Danny Rosenberg .................................................................................382

Natufian Settlement in the Wadi al-Qusayr, West-Central JordanMichael Neeley ................................................................................................................397

The Steppic Early Natufian: Investigations in the Wadi al-Hasa, JordanDeborah I. Olszewski ......................................................................................................412

The Natufian of the Azraq Basin: An AppraisalTobias Richter and Lisa A. Maher ..................................................................................429

Chert Procurement Patterns And Exploitation Territory: Case Study From Late Natufian Hayonim Terrace (Western Galilee, Israel)

Christophe Delage ...........................................................................................................449

A Faunal Perspective on the Relationship between the Natufian Occupations of Hayonim Cave and Hayonim Terrace

Natalie D. Munro ............................................................................................................463

The Natufian at Raqefet CaveGyörgy Lengyel, Dani Nadel and Fanny Bocquentin ....................................................478

Hof Shahaf: A New Natufian Site on the Shore of Lake KinneretOfer Marder, Reuven Yeshurun, Howard Smithline, Oren Ackermann, Daniella E. Bar-Yosef Mayer, Anna Belfer-Cohen, Leore Grosman, Israel Hershkovitz, Noa Klein and Lior Weissbrod ...............................................................................................505

The Life History of Macrolithic Tools at Hilazon Tachtit CaveLaure Dubreuil and Leore Grosman ..............................................................................527

General Reviews, Climate and Interpretations

Breaking the Mould: Phases and Facies in the Natufian of the Mediterranean ZoneAnna Belfer-Cohen and A. Nigel Goring-Morris ...........................................................544

Ruminations on the Role of Periphery and Center in the NatufianA. Nigel Goring-Morris and Anna Belfer-Cohen ...........................................................562

vi

The Natufian and the Younger DryasDonald O. Henry .............................................................................................................584

Scaphopod Shells in the Natufian CultureAldona Kurzawska, Daniella E. Bar-Yosef Mayer and Henk K. Mienis ......................611

The Natufian Chronological Scheme – New Insights and their ImplicationsLeore Grosman ................................................................................................................622

Natufian Foragers and the ‘Monocot Revolution’: A Phytolith PerspectiveArlene M. Rosen ..............................................................................................................638

Lithic Technology in the Late Natufian – Technological Differences between ‘Core-area’ and ‘Periphery’

Hila Ashkenazy ...............................................................................................................649

Variability of Lunates and Changes in Projectile Weapons Technology during the NatufianAlla Yaroshevich, Daniel Kaufman, Dmitri Nuzhnyy, Ofer Bar-Yosef and Mina Weinstein-Evron..............................................................................................................671

Specialized Hunting of Gazelle in the Natufian: Cultural Cause or Climatic Effect?Guy Bar-Oz, Reuven Yeshurun and Mina Weinstein-Evron .........................................685

Commensalism: was it Truly a Natufian Phenomenon? Recent Cntributions from Ethnoarchaeology and Ecology

Lior Weissbrod, Daniel Kaufman, Dani Nadel, Reuven Yeshurun and Mina Weinstein-Evron..............................................................................................................699

478

1. Introduction

Raqefet cave has a relatively short research history. It is a remarkable scar in the south-eastern extension of Mt. Carmel, which was discovered for scientific research only in 1956. More than a decade later, three seasons of excavations were carried out between 1970 and 1972. Although the cave yielded rich archaeological assemblages ranging from the Middle Paleolithic to the Roman period, the excavators hardly published their results. Several kinds of remains were later incorporated into various studies (see below), but the cave as an archaeological site remained neglected. Due to this situation the cave was hardly incorporated in Levantine Prehistory publications, though scanty data did appear in some (Garrard 1980; Ziffer 1978a, 1978b, 1981). In the 2000s several details of various Prehistoric occupations of the site were published, based on the 1970-1972 archives and data from our renewed fieldwork (Lengyel 2007, 2009a, 2009b, 2009c; Lengyel and Bocquentin 2005; Lengyel et al. 2005; Nadel et al. 2008, 2009a; Sarel 2004). Herein we summarize three aspects of the Natufian remains at the site, namely the flint industry, the burials and the bedrock features.

2. Research History of the Natufian at Raqefet

The cave is situated in southeastern Mount Carmel, in wadi Raqefet, about 0.5 km upstream from the confluence with wadi Yoqne’am that runs east into the Jezreel Valley (Fig. 1). It is located ca. 50 m above the wadi bed (230 m a.s.l.), at the bottom of a west facing cliff. In front of the cave, directly below the entrance, there is a narrow terrace. Five chambers form the cave, covering an area of ca. 500 m2 (Fig. 2). At the front of the second chamber, there is a large angular block of rock (ca. 5 m long) in a north-south orientation, probably detached from the north wall.

The Natufian at Raqefet Cave

György Lengyel, Dani Nadel and Fanny Bocquentin

The cave was discovered by Ya’aqov Olami in 1956 (Olami 1984). Eric Higgs of Cambridge University and Tamar Noy of the Israel Museum conducted joint excavations at the cave between 1970 and 1972 (Higgs et al. 1975; Noy and Higgs 1971). The excavations were carried out in the first and the second chambers, covering an area of 75 m2. The first report of the site mentions Natufian remains including some human burials. Later, the unpublished report by Higgs et al. (1975) describes abundant Natufian remains found in each of the excavation trenches. Unfortunately, most of the Natufian finds were recovered together with ar-tifacts of other archaeological cultures (Table 1). Among the stratigraphic units yielding Natufian finds, a few thin layers and the human graves have turned out to be post-depositionally undisturbed. In the 1970s two radiocarbon dates were ob-tained, associated with the age of the Natufian occupation. The first date, 10,980 ± 260 (I-7032), was measured on a piece of bone collected from within Pit L in square C 20 (Fig. 3). This pit was filled with archaeologically heterogeneous material, containing finds of Bronze Age, Late Natufian, Geo-metric Kebaran, and Levantine Aurignacian. The date however fits the Late Natufian chronology. The second date, 10,580 ± 140 (I-7030), was obtained from a bulk sample of charcoals from Layers IV, VI, and VII in square K28 (Fig. 4). Besides the Natufian items, these layers also yielded Neolithic and Geometric Kebaran artifacts. After thirty-two years, the excavations were resumed in a short reconnaissance season in 2004 (Lengyel et al. 2005). A strip of formerly unexcavated sediment and additional small patches of sediment covering bedrock mortars and cupmarks near the south wall were excavated to bedrock. The sediment strip was archaeologically mixed, but three large bedrock mortars contained in situ Natufian sedi-ments. Flints from these together with a few from the 1970s yielded a sample for the lithic technology

Fig. 1. Location of Raqefet cave and other major Natufian sites in Mt. Carmel.

479

480


Fig. 2. Raqefet Cave plans. Upper – excavated areas in 1970-1972 and 2004 (modified after Higgs et al 1975). Lower – the updated plan with excavation loci in 2006-2008.

481


study on the Upper Paleolithic and Epipaleolithic of Raqefet (Lengyel 2007). This short season also

revealed in situ human remains, which led us to expand the project.

Table 1. The stratigraphic sequence of archaeological finds at Raqefet cave after the 1970-1972 excavations (after Lengyel 2007). B-G/18-23, A-H/7-17, J-M/24-28 are excavation trenches, indicated following the square division of the grid on Fig 2. upper. Upper case abbreviations mark archaeological periods, lower case letters mark general finds such as pottery and flints of unidentified periods. In columns B-G/18-23 and J-M/24-28 upper case letters are pits and Roman numerals are layers. In the three columns under A-H/7-17 numbers are pits and Roman numerals are layers. Clean Late Natufian assemblages are shaded grey. BA – Bronze Age, C – Chalcolithic, EP – Early Epipalaeolithic, f – uncharacteristic flint products, GK - Geometric Kebaran, IA – Iron Age, IEUP – Indeterminate Early Upper Palaeolithic, ILUP – Indeterminate Late Upper Palaeolithic, LA – Levantine Aurignacian, LK - Late Kebaran, LN – Late Natufian, M – Mousterian, p – potsherds, PN – pottery Neolithic, PPNA – Pre-pottery Neolithic A, PPNB – Pre-Pottery Neolithic B, R – Roman, UP – unclassified Upper Palaeolithic.

B-G/18-23 A-H/7-17 J-M/24-28

C7-17, D-E/16-17 D-E/7-13 G-H/10-12

A p, f 1 p, LK, LN 1 p, GK, LN 1 R, GK, LN, p A p, F B PN 2 p, LK, LN 2 p, GK, LN I R, GK, LN, p B p, F

C p, LN, LK, PPNB 3 p, GK, LN 3 p, f II R, GK, LN, p C ?

D IA, LN, GK 4 p, GK, LN 4 p, f III p, LK I p, fE p, f 5 p, GK, LN 5 p, GK, LN IV p, GK II PN, PPNB

F p, LN, LK, GK I p, LA, GK,

LN 6 p, GK, LN V LN D PN, PPNB

G BA, LN II p, LN, PPNB I R, GK, LN, p E p, f

H PPNA III p, f II p, LN, GK F p, f, BA, R

J PPNA, LN IV LK, GK, LN, PN III GK, LN G p, f, BA, C

K f V EP IV LN III p, f

L BA, LN, GK, LA V LN IV p, f

I LK V PPNA, LK, GK, LN

II ILP VI GK, LN

III LA VII GK

IV IEUP H GK, LN

V Sterile VIII ?

VI M, UP IX LK?

VII M, UP J LA?

VIII M, UP X LA, M

K M, UP

XI ?

XII M

482


In 2006 we continued to expose unexcavated areas near the north cave wall of chamber 1 and at the entrance (Nadel et al. 2008). As was the case in the 1970s, most of the Natufian sediments were disturbed, and only the burials yielded unmixed assemblages. The north wall area, designated Loci 1 and 3, contained several primary Natufian burials. Locus 2, located at the entrance of the cave, preserved a mass of reworked sediment. In this season we carried out ground penetrating radar scanning in the cave and on the terrace. The GPR

reflection data collected both inside and outside the cave were of high quality (Nadel et al. 2008). Studies of geoarchaeology, the flint assemblage, the groundstone industry, the mammal and micro-mammals remains, the bone tools and the mollusc beads were also carried out. In 2008 fieldwork and the analyses started in 2006 were continued (Nadel et al. 2009a). For better characterization of features hewn into the cave bedrock a three dimensional documentation method based on photogrammetry was used.

3. The Natufian in the Stratigraphic Sequence of Raqefet

Excavations at Raqefet cave in the 1970s were carried out in three areas indicated by square grid coordinates: A-H/7-17, B-G/18-23 and J-M/24-28 (Fig. 2). In the field each area had its own strati-graphic sequence numbered separately and the correlation between layers was made later. The majority of the stratigraphic units exposed in these areas contained mixtures of artifacts from several periods. The mixed archaeological units, including pits and layers, amount to ca. 70% of the total number of units. Natufian material often occurs mixed with Neolithic, Kebaran and Geo-metric Kebaran artifacts. Clean Natufian material can be found in graves or in sediments filling the large bedrock features. The Natufian graves are found under historical sediments and the filling of the graves directly covers the bedrock. Grave

Fig. 3. Pit L, location of C14 Natufian date sample.

Fig. 4. Section of trench J-M/24-28 of 1972.

483


sediments are archaeologically uniform. According to geo-archaeological studies (see relevant sections in Nadel et al. 2008 and 2009a) these are strongly reworked, mainly anthropogenic, with a conspicu-ous ash component represented by calcite ash mixed with chips of bones, pulverized soil material and charcoal. It should be stressed that so far no in situ Natufian anthropogenic features such as hearths, floors or dwelling structures were encountered at the site.

4. Late Natufian Lithic Technology

The lithic technology study of knapped stones in this chapter involves finds recovered in 2004 from within C-I and C-II bedrock mortars and layers IV-V of squares D-E/7-13 of the 1970-1972 excavations. Lithics from 2006 and 2008 are still under study, but some preliminary results are presented as well.

4.1. Raw materials

The lithic raw materials of the Natufian assem-blage are generally good quality flints. This means the siliceous rock used for knapping is homogenous, lacks inclusions and cracks, and its grain texture is fine. Sources of such lithic materials are located in the Deir Hannah formation (DH) and some 10 km away in Nahal Mearot (NM) (Fig. 5). Flakes of Deir Hannah formation raw materials are sometimes fully cortical which mean that nodules were brought to the site unworked and the whole knapping process was performed at the cave. It is of interest that the closest lithic source to the site, the Ramot Menashe (RAM) range, which provides the most abundant supply of knappable rocks but in lower quality than the other two sources, was infrequently exploited. The farthest exploited source is the Shamir formation (SH), located in the northwestern part of Mt. Carmel.

4.2. Blank production

Deir Hannah good quality flints were the most common raw material used to produce blades, followed by the Ramot Menashe flints. In the first step of blade production, a single flake was removed from the nodule to create the striking platform. Then one or two flakes were further detached from one side of the nodule and the ridges created by flake scars guided the first blade removals. The blade cores have a single striking platform, and the blades were detached from the direction of that striking platform. Abrasion on the edge between

the striking platform and the debitage surface often took place before blade detaching. A total of 64.8% of the blades with preserved butts show this abrasion on the dorsal face. Usually, the abrasion is a good marker of soft hammer use. The preference for using soft hammer in blade production is also detectable throughout the low frequency (19.6%) of impact point on the butts (a phenomenon typical of hard hammer use). Bladelets, similarly to the blades, were produced from good quality flints. Contrary to the low number in the blade assemblage the good quality flints of Nahal Mearot become more frequent. On one hand, the bladelet production was simply achieved with the reduction of the size of the blade cores. This can be stated due to the low number of blade cores and the relative large number of blades, bladelet cores and core platform rejuvenation flakes and tablets (CTE) from the same raw materials (Table 2). On the other hand, small flint nodules were also set up as bladelet cores. The latter case involved the same procedure for core initialization which was used in the blade production. However, the studied bladelet core assemblage contains six items that have a reddish color and their surfaces are matt, except for some scars which are shiny. These features are indicative signs of heat treatment of flint (Inizan and Tixier 2001; Inizan et al. 1976). In addition to the cores, several products also bear these signs. In spite of the visible features, the heat treatment details must be further verified. While other local Epipaleolithic cultures used the waste flakes of blade and bladelet productions for making flake tools (Lengyel 2007), the Natufian evolved a separate process for flake production. In the case of Raqefet the flake production incorpo-rated Ramot Menashe flints of mediocre quality, as well as some Deir Hannah formation and Nahal Mearot flints. Eight of the eleven flake cores signify unidirectional, two show opposite and one multiple orientation of debitage. The flakes have equal signs of both soft and hard hammers percussion. Data from the 2006 and 2008 excavation seasons show a similar pattern of the lithic blank and core types, with a higher ratio of flake cores in these assemblages. This result supports the conclusion that there was a clear flake debitage reduction sequence in the Late Natufian.

4.3. Tools

Of the total studied blade assemblage 15.9% are formal tools. Six out of ten blade tools are simply

484


Fig. 5. Raw material sources found at Raqefet cave. (Modified after the Map of Israeli Geological Survey: MIGS.)

485


retouched on one edge. The rest of the blade tools are a burin, a truncated item, a denticulate, and a backed specimen with sickle sheen on the sharp edge. A total of 13.1% of the bladelets are tools. While blades make up 2% of the formal tool assemblage, bladelets constitute the majority, 43.1 %. Bladelet tools are often geometric microliths. A common type is the abruptly backed lunate (Fig. 6). The length of the lunates ranges between 12.3 and 18.2 mm. The remaining bladelet tool types are truncations, notches, denticulates and retouched items. A small portion of the total flake assemblage, 7.1 %, was shaped into formal tools. In spite of this, 37.3 % of the formal tools are flakes. The common flake tool types are the denticulate, end-scraper and burin. The lithic tool type list of the 2006 and 2008 excavations is similar to the above (Table 3). A few additional tool types are the Helwan lunates, Hel-wan backed bladelets, and the trapeze-rectangle, all of which are common in Late Natufian context, and the microgravette that usually appears in earlier phases of the Epipaleolithic . From the lithic technology point of view, the Late Natufian Raqefet industry is similar to the Eynan Final Natufian assemblage (Valla et al. 1998, 2001). The similarities are apparent in characteristics such as the use of fine-grained raw material, heat treatment, soft stone hammer technique in laminar production, unidirectional blade debitage, and a pronounced flake debitage.

5. The human burials

Human burials identified as Natufian were discovered first in the 1970s (Higgs et al. 1975; Noy and Higgs 1971). Their features remained unpublished for a long time and the archaeological

context of the graves was unclear until recently when the original documents and human remains were reanalyzed (Lengyel and Bocquentin 2005). These first burials were recovered near the north cave wall in chamber 1 (our current Locus 1). Our renewed excavations here and in adjacent Locus 3 revealed a dense concentration of burials (Table 4). This part of the cave was certainly dedicated to funeral activities (Fig. 7). Below we focus on this group of burials as only one other isolated grave was found, at the entrance the cave (H12: cf. Nadel et al. 2008). Concerning the graves recently unearthed (2006 and 2008 seasons), only burial aspects will be considered as the bones are still under process of restoration (acid treatment and refitting), a necessity before bioanthropological analyses.

5.1 Adult burials

Seven adults have been found in Loci 1 and 3. All belong to primary single deposits. Homos 1 and 2 were found in 1970-1971 and Homos 8, 9, 14, 15 and 17 were discovered by us. The exact location of the skeletons found in 1970 and 1971 is only approximately known.

Homo 1

The first skeleton (H1) was discovered in square D13, south of the rock basin (Locus 1), in 1970. The skeleton did not lie directly on the sloping bedrock surface; a thin yellow layer was found between the burial and the bedrock. The area around the skeleton was intensively disturbed by pits which contained potsherds of later periods. The H1 remains belong to an adult female, more than 30 years old at death. In the archives of Raqefet, no notes mention the state of articulation of the skeleton. The only available data on the grave structure concerns a stone placed

Table 2. Frequency of lithic products by raw material, including complete and incomplete items and tools from 1970-1972 (area A-H/7-17 layers IV and V) and 2004 excavations (bedrock mortars C-I and C-II)

Source Blade Bladelet Flake Waste CTE Flakecore

Bladecore

Bladelet core Total

RAM 17 22 46 34 6 2 2 9 138DH 33 82 123 320 12 6 2 27 605

DC 1 2 10 2 15

NM 8 32 39 97 3 3 1 13 196

SH 1 12 7 9 3 1 33

unidentified 4 14 12 13 1 2 46

subtotal 64 164 237 473 27 11 5 52 1033

486


Fig. 6. Natufian lithic tools from Raqefet cave.

487


against the head. Currently, only parts of the skull (frontal bone; right zygomatic bone; right temporal bone; small fragment of the right ramus of the mandible) can be attributed with some certainty to this individual. Other bones (part of the right clavicle, scapula, humerus, coxae, patella and a few bones of hands and feet) found without labels in the stored collection, may also belong to this individual according to their taphonomical and anatomical features.

Homo 2

Homo 2 was found in square B11 near the north wall of the cave, embedded in a yellow deposit in the western half of the bedrock basin. The filling of the grave was undisturbed. The skeleton was

found under a heap of large stones. On top, in the middle of the stone pile and above the chest and skull, a capstone was placed. The corpse was lying on vertical stone slabs aligned in an east-west di-rection and filling up the basin. The stone structure was found separated from the underlying bedrock by a hard red layer. This is the grave of a mature woman (>30 years old). The bones remained mostly in anatomical articulation. The skeleton lay in an SE-NW direction, head to the SE, on its left side in a flexed position with the right hand placed close to the face. The head and the knees rested against the south wall of the natural basin. The position of the cervical and thoracic vertebrae suggests that the upper part of the back was initially in an upright position, maintained by either sediment or perishable rope or structure. The state of pres-

Table 3. Type list of flint tools from un-mixed Natufian sediments of all excavation seasons between 1970 and 2008

1970-2004 2006-2008

Tool types blanks

blade bladelet flake blade bladelet flake total

End-scraper 3 4 7 14

Burin 1 4 6 5 11 27

Retouched 6 1 4 11 21 15 58

Backed straight 1 3 27 31curved 1 1 3 5arched 1 5 6helwan 1 1microgravette 2 2

Point 6 6

Truncation 1 3 2 12 1 19

Geometric straight backed and straight truncated 2 2 4

blunt backed lunate 8 67 75Helwan lunate 4 4proto-trapeze 6 6trapeze 1 2 3trapeze rectangle 1 1triangle 1 2 3

Borer 1 2 1 4

Sickle 1 3 4

Notch 3 2 5 5 11 26

Denticulate 1 1 5 13 6 17 43

Composite end scraper–burin 2 2

Total 10 22 19 49 181 63 344

488


Table 4. List of human burials of Raqefet

Individual Excavation Locus Square Age; Sex Position; orientationHomo 1 1970 1 D13 > 30 years old; F unknownHomo 2 1970-1972 1 B11 > 30 years old; F Lateral left; SE-NWHomo 3 1970-1972 1 B12 Child 8-12 years old Abdomen; NE-SWHomo 3a 1970-1972 1 B12 Child 4-5 years old Unknown, with Homo 3Homo 4 1970-1972 1 C12 Young adolescent UnknownHomo 6 2006 1 C13a-b Child 6-9 years old Lateral R; E-WHomo 7 2006 1 B12d Child 5-8 years old Abdomen; NE-SWHomo 8 1970-2006 1 B11d/B12c Adult On its back; E-WHomo 9 2006 1 C12c-d Adult On its back; W-EHomo 10 2006 1 C12d Young adolescent On its back; W-EHomo 11 2006 3 C14c/D14a Child 0-6 months Seated?Homo 12 2006 2 G8b-c Adult On its back; SW-NEHomo 13 2006-2008 1 C12d Young adolescent On its back; NW-SE Homo 14 2008 3 C14b/C15a-b Adult Lateral right; E-WHomo 15 2008 1 B12c-d/C12b Adult On its back; SE-NWHomo 16 2008 3 C16c-d Child 18-24 months Lateral right; W-EHomo 17 2008 1 C12a-b Adult Lateral right; NW-SE

Fig. 7. Location of human burials found in loci 1 and 3.

489


ervation of the bones is poor and they were highly calcified. Unfortunately the legs, removed during the first campaign in 1970 (feet, tibias and fibulas) are missing today.

Homo 8

The skeleton of H8, found in square B12c, is fragmentary (upper legs, lumbar column and hip bones) but enough is preserved to determine that the corpse was lying on its back, the left hand on the pelvis and the right hand under the left elbow. Concerning the lower limbs, only the right femur was found in situ, in extension. However, consid-ering the inventory and the location of the isolated pairs of fibulas and tibias found in Sq B11d by Noy and Higgs (Lengyel and Bocquentin 2005:279), it is very likely that those bones (labeled preliminary as “Homo 5”) belong to Homo 8. In this case, the corpse was deposited with both legs extended and covered by several big stones on which H2 was later deposited.

Homo 9

The left half of the upper skeleton of H9 (square C12cd) is preserved (Fig. 8). The ribs, scapula, humerus, radius, and ulna are connected but the clavicle shows a displacement. The body is obviously lying on its back, forearm extended along the body. The burial was later partly eroded and cut by a pit on the south. Skull and mandible were found ar-ticulated together right under the ribs in a reverse position (Fig. 9). If this location is a primary one, we must admit that the corpse was buried with the neck already partly disarticulated (cut deliberate-ly?); however, this treatment is unknown for the moment during the Natufian. Alternatively, this is the result of a secondary displacement, during decay process, due, for instance, to the decomposition of a voluminous headrest. However, the fact that the occipital is resting under the ribs still connected reduces the likelihood of the latter hypothesis. The absence of the vertebral column makes the resolu-tion to this enigma particularly difficult, unless cut marks will be observed in the mandible or occipital during laboratory work.

Homo 14

Homo 14 is a badly preserved adult skeleton that was found in cemented sediment, against the cave wall (squares C14b/C15a-b). The corps was deposited directly on the bedrock, lying on its right side, head facing the wall of the cave (north). Two medium size stones were found directly on the skull and thorax, with perfectly articulated bones under-neath. Both arms were placed parallel to the body, with the right forearm extended, hand open and the left forearm slightly flexed, hand closed above the right wrist. The position of the right clavicle (almost parallel to the vertebral column) as well as the preservation of part of the thoracic volume, as shown by the left ribs, should be interpreted as the result of a lateral compression of the body. This specific feature of decay might be explained here by the weight of the stones placed directly on the body. The lower half of the body was very fragmentary. A space empty of bone was observed between the last ribs, and a pair of knees recovered in a flexed position was found slightly towards the west. Additionally, a small coxal fragment, the di-aphysis of both femurs and a lumbar vertebra were found isolated nearby and were also temporarily attributed to the same individual. This assumption needs a definitive confirmation in the laboratory.

Fig. 8. Homo 9 (on the left).

490


Homo 15

The lower part of H15’s skeleton is preserved from the hip bones to the feet. The upper part of the body was probably cut by a later burial pit (probably Homo 17). However, H15 is preserved well enough to state that the corpse was deposited with both legs extended to the NW, feet against the cave wall. All preserved bones were found in tight anatom-ical connection. Their respective position and the fact that the patella and the bones of the feet are maintained in equilibrium in lateral position are a testimony of a burial in a narrow pit immediately filled by fine sediment (progressive infilling). Tibias and feet were lying right under a boulder cupmark partially embedded in breccia. A noted feature on this grave is the number of faunal remains (such as maxilla and scapula of Dama mesopotamica and maxilla of Gazella gazella) that were found in direct association with the skeleton.

Homo 17

Homo 17 was an adult represented only by the upper part of the body (Fig. 10), the rest probably cut by the double burial pit of Homos 10 and 13. The corpse, oriented NW-SE, was deposited on its right side, with both upper legs flexed. The thorax was directly covered by a big stone. The cervical column, preserved partly articulated, indicates that the head was originally in an upright position, and thus a perishable headrest was possibly placed under it. The two first cervical vertebrae have followed the head in its fall forward. A medium size stone de-posited on the right of the skull has moderated this later displacement and the head was kept upwards. The left hand appeared in pronation near the face, with all carpals, metacarpals and phalanges in an-atomical connection. The right upper extremity was maintained in a vertical position, partly covering the left hand. Again, the phalanges were found in

Fig. 9. The head of Homo 9.

491


anatomical position indicating that decay took place in a filled space with an immediate earth infilling (although a later bio-disturbance of the proximal part of the hand is likely). Close to the palm of the hand (although not in direct contact with it), was a horn core of Gazella gazella. Finally, it is worth noting that on top of Homo 17 several isolated hu-man bones (such as fragments of skull, mandible and the proximal half of a femur) not belonging to Homo 17 were recovered. Some of these remains may belong to Homo 15, which was cut at mid-skeleton.

5.2 Child burials

Nine immature individuals have been found in Loci 1 and 3. One of them is a double burial (H10 and H13), the other ones are single cases, except for the doubtful set of H3-3a. Homos 3, 3a and 4 were found in 1970-1971; Homos 6, 7, 10, 11, 13 and 16 were found by us.

Homos 3 and 3a

Homo 3 was found in square B12 beneath the overhanging cave wall, close to H2. The skeleton

belongs to a child between 8 and 12 years old at death. The grave contained numerous Bronze Age potsherds indicating a later disturbance. As in the case with H2, the excavators mention that a cap-stone was placed above the skeleton. The skeleton was found directly on the big cupmark boulder which was part of the stone-lined structure filling the bedrock basin under H2 and on top of H8. Accord-ing to the excavators, the skull, ribs and vertebrae remained in anatomical articulation (Higgs et al. 1975). A picture shows the upper part of the thorax appearing by the dorsal face and a crushed skull in situ; these suggest that the child was buried on its abdomen (cf. Lengyel and Bocquentin 2005:278, fig. 11). In the lab, the bones assemblage fits well with the field documents: parts of the skull and mandible are preserved as well as some of the upper body (left clavicle and scapula; humerus; vertebrae and ribs; sternum; hand bones) and fragments of the pelvis and femur. A few additional bones belonging to a younger child (Homo 3a: 4-5 years old) were found also in this grave. Given the numerous disturbances amongst the graves of Locus 1, these few pieces of bones can hardly be considered as the testimony of an intentional double burial.

Fig. 10. Homo 17.

492


Homo 4

The remains of H4 were found “near” H3, with two canines of wolf or fox. The features of the em-bedding layer are not mentioned in the documents. According to the archives only the skull, which was crushed but nearly complete, was removed (Higgs et al. 1975). However, other bones, preserved in good condition, were also recovered in the lab: the upper part of the body is almost complete including the hand bones and all the epiphyses, indicating a primary burial. In addition, a few fragments of the lower limbs and the complete right foot are also present. The skull and the skeletal parts correspond anatomically well to each other, showing that they belong to the same individual, a 13-15 year-old adolescent.

Homo 6

H6 (square C13a-b) is a child (6-9 years old) that was lying directly on the bedrock, on its right side with the knees in a flexed position, slightly straightened against the cave wall (Fig. 11). The skeleton was found directly beneath two large flat stones lying on the surface that was cleaned in 2004. This area was disturbed by numerous tree roots and, consequently, the upper part of the grave was filled with loose sediment. Thus, it is uncertain whether the two stones were contemporaneous with the burial. However, a third stone, lying directly on the head of the child was definitely associated. The lower part of the grave was heavily brecciated and the skeleton was taken out as a block.

Homo 7

Homo 7 (square B12c/d) is a very fragmentary skeleton of a child (5-8 years old) with only parts of the skull, mandible, clavicle, scapula, head of the humerus and cervical vertebrae preserved. The bones were found in their anatomical order, the face almost facing the ground. This assemblage is most probably the remains of a disturbed burial (later cut from the shoulders to the feet); it was also eroded at the top, as the skull vault is incomplete. A flat stone was found directly on top of it but considering the level of erosion, it may have been posterior to the grave.

Homos 10 and 13

Homos 10 and 13 were buried together. Homo 13, lying directly under H10 was discovered at the

end of the 2006 season and was consequently un-earthed in 2008. Both are approximately the same age (young adolescents) and were buried in the same position on top of each other. The pit burial of H9 cut the double burials to the south. This explains why only the upper part of skeletons H10 and H13 are preserved. A long slab was placed vertically at the west edge of the grave, partly overlying the skulls of both children. H10 was lying on its back, the left forearm ex-tended along the body (Fig.8). The right humerus found connected to the scapula, was cut at midshaft by the burial pit of H9. A big stone was found on the neck and the bones were dislocated. This dis-turbance was made during the decay of the corpse as the skull and atlas (first vertebra) were still articulated (this is one of the latest articulations to decay) when it occurred. The mandible, broken to two parts, was found not far from its anatomical location. The axis, on the other hand, was found far away on top of the ribs. This short time span between the burial and the unsettling of some bones lead us to suggest that it might have been deliberate, maybe part of a secondary funerary treatment. H13 was lying on its back, slightly on the right side. Most of the bones recovered from this

Fig. 11. Homo 6.

493


individual were found in anatomical connection or almost so. The cervical vertebrae from C2 to C7 and the beginning of the thoracic column show that the neck and the head were originally in an upright position. However, the skull was displaced towards the back, dragging down the atlas and the left part of the mandible (broken) with it. These were discovered up-side-down, partly crushed but still articulated between them, as a testimony of a displacement as an anatomical block during the process of decay. In order to explain this collapse we must admit that a secondary empty space ef-fect was created somewhat after the burial, which could have been the disintegration of a perishable headrest. Additionally, the presence of the long stone slab in direct contact with the right side of the skull may have contributed to the described movement of the head.

Homo 11

H11, a young infant (0-6 months), was found in squares C14c/D14a. The skeleton was discovered with all bones compacted and calcified together with the skull up-side-down. The hands were preserved in articulation, indicating a primary burial. The skeleton was taken as a block and laboratory work will permit to define the position and decay process of the body. Preliminary observations in the field suggest that it might have been buried seated. The bottom of the burial pit, cutting the Mousterian layers (yellow compact sediments with numerous flint flakes) was clearly visible.

Homo 16

H16 is a young child (18 and 24 months old) buried lying on its right side, in a west-east ori-entation and facing SSE. The skull and mandible were in articulation, as well as the cervical and thoracic vertebrae, left ribs and clavicle, scapula and both humeri. It is worth mentioning that the left humerus was cut at midshaft. The right arm was elevated and flexed, with the hand found mostly articulated under the skull. The lower part of the body was very fragmentary, although the position of the right femur, tibia and fibula (that were re-covered close to their anatomical position), as well as other bones that were found isolated (such as the left tibia and several tarsals, metatarsals and foot phalanges), support the statement that this individual was probably buried with both lower extremities flexed, with the left ankle crossed un-der the right thigh at midshaft level. The rest of

the body was not found. It is likely that a pit cut the skeleton, as suggested by the thorax, partly missing. An adult skull discovered immediately at the south of the child might be part of a complete grave, which should explain the cutting of grave 16. However, these remains were not yet excavated and the skull was left in situ for the forthcoming 2010 season of excavation.

5. 3 Discussion of the graves

Loci 1 and 3 as a sepulchral unit

Almost all the skeletons found at Raqefet until now (with the exceptions of H2 and H11) are only partially preserved. Late post-depositional pro-cesses are sometimes the reason for that (in the case of H6, H9, and maybe H3 and H7 as well) but mostly, the disturbances are contemporaneous with the use of this place for funerary activities by the Natufians. This is due to the fact that they have buried their dead in a specific small area repeatedly. The skeletons are grouped together in the deep natural depression of Locus 1 where at least 9 dead were buried in a minimum of 4 successive events. Indeed, H15 was cut by the pit of H17, itself cut by the H10-13 burials which were partly destroyed by the inhumation of H9 (see the schematic repre-sentation on Fig. 7). The time span between these different events cannot be precisely evaluated, but the fact that the burial pits are progressing to the south testifies to a planning in space organization. Furthermore, there appears to have been a certain common memory, even approximate, of the preced-ing location of pits. The dig in progress in Locus 3 is also revealing a concentration of burials further east. Three skeletons have been unearthed and three other skulls were exposed in squares D14, D16 and C16. Altogether, this area, adjacent to the north wall of the cave, contrasts strongly with the rest of the first chamber where only one grave was dis-covered (H12 in Locus 2). Loci 1 and 3 are part of a mid- or long-term area dedicated for successive primary burials. In contrast to a real collective grave where the dead share the same space of resting, the apparent co-mingling of skeletons in this case is the result of a succession of different burial pits cutting each other. The will for grouping together some of the dead community is eloquent but the treatment of the corps remains individual (or double in exceptional cases). Therefore, the skeletons are partly cut and numerous supplementary bones are found in each burial pit.

494


Sepulchral units are known in other Natufian sites such as el-Wad, Hayonim and probably Kebara (Bocquentin 2003) but all are attributed to the Early Natufian. They present the common characteristic very likely to be family units. Indeed, the study of the epigenetic variations (discrete traits) of the skeletons buried together has shown a high concen-tration of some of them in comparison with the rest of the skeletons buried on those sites (Bocquentin 2003). This bio-archaeological study has not been done yet amongst the skeletons of Raqefet but it will certainly provide us in the future with some clues for interpreting this cluster of burials.

Treatment of the Dead

Altogether 17 graves have been found near the north wall of the cave (Table 5). These include eight adults, three adolescents, five children and one infant. Sex was determined only for two of the adults as laboratory work is still under process.

For the moment, no selective recruitment can be observed amongst the dead population.

Primary deposits

Although the burials are very disturbed, there is no doubt that all are the remains of primary depos-its: the skeletons are fragmentary but anatomical connections, even fragile, are preserved. All, except one, are single. The double burial of H10-13 includes two adolescents who are lying in the same position, and directly superimposed on each other. This dis-position of corpses resting on top of each other is the rule in case of double primary burials during the Natufian (Bocquentin 2003), in contrast to corpses lying next to each other which is exceptional (but see Homos 2 and 3 at el-Wad and Homos 9 and 11 in Hayonim Cave) (Belfer-Cohen 1988b; Garrod and Bate 1937). In the case of H10 and H13 like in the other Natufian occurrences, no bones of the upper corps (H10) enter the space of the lower one (H13), contrary to what gravity phenomena would

Table 5. The types of human-made bedrock holes at Raqefet cave and terrace.

Type Description Cave Terrace Total

A Small, round shallow holes, 2-5 cm across and 2-5 cm deep, width:depth=+1 12 12

B Small, round shallow holes, 5-10 cm across, 2-5 cm deep, width:depth>1 2 3 5

C Medium round holes (cupmarks), usually bowl-like in shape, 10-15 cm across, 5-10 cm deep, width:depth=+1 13 11 24

D Large round holes (cupmarks), usually bowl-like in shape, 15-30 cm across, 5-30 cm deep, width:depth=+1 22 14 36

EDeep narrow round cylinders (mortars), 10-20 cm wide along most of the shaft and very narrow at the bottom, 20-80 cm deep, width:depth<1

3 3

F

Deep narrow round cylinders, funnel-like in shape, 10-20 cm wide along most of the shaft and very narrow at the bottom, 20-80 cm deep, width:depth<1. A clear shoulder/break is present between the wider top and the narrow bottom

2 2

G Deep wide round/oval cylinders, wider than 20 cm, 20-80 cm deep, the top is much wider than the bottom 5 1 6

H Oval shallow features, width:depth>1 1 1

IElongated features, including short “channels”, width:depth>1(The specimen here is connected to a cupmark and not counted separately)

*1 *1

J Composite sets, of a pair (or more) combined together

K Varia (specimens that do not fit any of the above categories)

Total 60 29 89

495


have produced during decay process. This means that the two bodies were placed in independent spaces of decomposition (single funerary bags or shroud, for instance). This is confirmed by the fact that both skeletons show lateral constriction when they are not buried exactly according to the same axe (this rules out a simple narrow pit effect).

Orientation and position for a last resting

Bodies were mostly placed in an east-west ori-entation (N=6), with the head to the east (H6, H8, H14) or to the west (H9, H10, H16). This preference is in harmony with the major axe of the Locus 1 natural cavity. Some skeletons are shifted towards a NW-SE orientation (N=4), with the head to the SE (H2, H15) or to the NW (H13, H17); others (N=2), fragmentary, seem to follow a NE-SW orientation (H3, H7). In terms of position, lateral and back lying are dominant, however children show more diversity with an infant tightly flexed on itself, probably originally seated (H11) and two older ones resting on their abdomen (H3 and H7). Concerning the lateral position, the right side is more frequent (H6, H14, H16, H17) than the left (H2). The back position is common amongst the dead buried in the natural deep rock basin (H8, H9, H10, H13 and H15). Both individuals with lower legs preserved (H8 and H15) testify to an extended position. This is particularly interesting as this position was known only during the Early Natufian (Bocquentin 2003) before the Raqefet discoveries. We note that these graves must be part of the oldest ones in the cave, at the bottom of the rock basin and under the cupmark stone.

Pits and space of decay

As much as we can see, the dead were buried at Raqefet in narrow spaces. This is shown especially by the skeletons lying on the back: the shoulders are higher than the ribs, the clavicles are almost parallel to the humerus and part of the thoracic volume is preserved. When hip bones and lower legs are preserved they are maintained in equilibrium materializing a limit of shroud or narrow pit, today lost. In case of lateral position, a recurrent linear effect following the vertebrae, hip and feet bones is also a testimony of a limit outside of which bones could not fall during the decay process. The bodies were probably deposited against the pit wall in a narrow space. Moreover, all skeletons show signs of decomposition within a filled space in fine sediment

which has progressively replaced the body volume during the process of decomposition. Therefore, displacements due to gravity are minimal.

Headrests

In three cases at least (H2, H13 and H17, but maybe also H9 and H10) the head of the dead was originally in an upright position. The skull of H2 was resting against the inner wall of the rock ba-sin but in the two other occurrences the presence of a headrest of perishable material is very likely. After some time, while the corps was not yet totally decomposed, the cushion disappeared dragging the skull and part of the cervical vertebrae down with it (Fig. 10). The choice of putting the head in an upright position was also made during the Final Natufian at Eynan where the necks are, in several cases, resting against the vertical wall of the burial pits (Bocquentin in Valla et al. 1998, 2001; Bocquen-tin and Cabellos in Valla et al. 2007). However, the use of perishable headrests by the Natufians may be specific to Raqefet.

Stone covers

Another interesting feature at Raqefet cave is the repetitive use of big stones covering the skel-etons. Big stones are covering preferentially the head and the thorax. In addition, Locus 1 shows a complex structure of commingled stones on top of all the skeletons (Fig. 12). We could not yet recog-nize any specific pattern in this assemblage and its function remains unknown. However, some stones were obviously added while funerary activities were still carried out, as is shown by the H10 case, disturbed not long after being buried. These stones may have had an important role in the treatment of the dead.

5.4 Conclusion

Loci 1 and 3 were dedicated, at least during part of the cave use, to burial activities. The nat-ural deep rock basin was preferentially used for this purpose. Towards the east, further inside the cave, the new excavations reveal an additional concentration of graves. These may also have been dug on top of another natural rock basin or placed here in order to be near the basin of Locus 1. The relative chronology of the graves from Loci 1 and 3 is unknown as the Natufian layer shows a great homogeneity. Moreover, two later deep pits destroyed the direct sedimentological link between

496


both Loci. The fact that the dead buried on their back are found only in Locus 1 may indicate some time lag between the use of the rock basin and its surrounding. The 17 graves discovered so far in this area reflect a tendency to group together the dead, near the north wall of the cave. However, each individual usually had its own space of rest-ing in a single narrow pit. The succession of the burial pits shows a certain planning of the space for, at least, a mid-term use. Some diversity is shown in terms of position and orientation; also, the treatment of the adults appears to have been more conventional. A tight association between the dead and big blocks of stone is recurrent as it is regularly the case during the Early and Late Natufian. Additional blocks were covering Locus 1 which might be part of an architectural funerary structure. The heads of at least three individuals

were resting on a perishable cushion, a custom unknown elsewhere.

6. Bedrock Features

During the Early Natufian the production and utilization of portable groundstone implements became prolific, and the level of manufacture, the quantities and variety reflect a turning point in comparison to all earlier stone technologies. By the Late Natufian, bedrock features also became abun-dant in some sites. This shift may reflect important economic changes, new settlement patterns and new ways of using natural resources (e.g. Rosenberg and Nadel 2011; Wright 1994). Bedrock features, also termed Human-made Bedrock Holes (HBHs), are general terms used here to encompass all basins, mortars, cupmarks, holes,

Fig. 12. Stones found above the burials in Locus 1.

497


cavities etc. – that were hewn into cave rocky floors and terraces, as well as those that were made on open-air rock exposures. ‘Mortars’ and ‘cupmarks’ are the most common terms for two kinds of these features, but they do not encompass the entire range of types (Nadel and Lengyel 2009; Nadel et al. 2008, 2009a, 2009b). Proliferation of bedrock mortars in Late Natu-fian sites is widely considered evidence for inten-sified food-processing habits, shifting to economies based on cereals and / or acorns (Rosenberg 2008 and ref. therein). Though commonly taken as a ‘truism’, relatively little work has been done in terms of directly correlating these remains to food processing (see, for example, Eitam 2008; Nadel and Lengyel 2009). Indeed, during ca. eighty years of Natufian research, one of the Natufian hallmarks in terms of numbers and work investment was poorly documented (until recently) and not really incorporated into syntheses regarding Natufian economic, mundane and spiritual life. In the fol-lowing section a summary of the bedrock features from Raqefet cave and terrace is presented. The numbers and variety here are among the largest in

any Natufian sites in the southern Levant. As such, and as some provide evidence for utilization in the spiritual rather than the food production spheres, their relevance to understanding past activities at the site is apparent. At Raqefet, the bedrock features were first noticed during the 1970-1972 excavations. In 2004, undisturbed Natufian sediments and artifacts were recovered from within three bedrock mortars. Since then we exposed and studied ca. 90 specimens (Table 6), and thus introduced a new classification system for describing them, as previous works did not focus on these (Nadel and Lengyel 2009; Nadel et al. 2008, 2009a, b; Wright 1991, 1992). These are found in the first chamber, on a large fallen rock nearby and on flat and sloping rock exposures on the terrace. There follows a description of selected cases.

Basins

Large natural depressions (here at least 1 m long) are common in the cave floor. Four basins are oval in shape and have a relatively smooth surface.

Fig. 13. Bedrock basin in Locus 1.

498


Though direct evidence for their manufacture is lacking, it is suggested that the natural cavities were modified by the Natufians to fit their needs (not counted in the table). In one irregular narrow basin near the north wall of the cave, a bedrock mortar was hewn, and bodies were repeatedly bur-ied there (see above, Locus 1) (Fig. 13). In another basin, adjacent to the opposite cave wall, a pair of bedrock mortars were hewn, containing in situ Natufian remains (C-I – C-II, see below) (Fig. 14).

Bedrock holes

There are nine general types of bedrock holes in terms of size and shapes, and there are also several cases, at Raqefet and other Late Natufian sites, of sets of two or more annexed features. The smallest examples are tiny holes, 2-5 cm across and 2-5 cm deep. There are at least 10 in the first chamber, mostly in the middle, within a radius of 1 m from the largest bedrock mortar (C-XXIII). Most of the HBHs at Raqefet and other Natufian sites were found empty, and therefore those with

in situ finds are of particular interest, and some are described hereby. C-XVI (type F) was found with a stone firmly placed 30 cm below the rim (Fig. 15). There were no additional stones above or below it in the shaft. Under it, there was fine reddish sediment containing small animal bones, 3 complete lunates, 3 blade/lets, and 2 minute flints. C-I – C-II is a pair of mortars hewn in basin 3 and full to their rims with Natufian material re-mains. An angular stone block was set on edge on the western rim of C-I (type D) (Fig. 16). The bottom of C-I is covered by tufa containing Natufian flints and small stones. A child’s parietal bone fragment (5 cm in diameter) was found lying horizontally (Lengyel et al. 2005). C-II is one of the largest bedrock mortars excavated so far (type G). Four stones were set on edge deep within it (parallel to each other), two of which are conjoinable pieces of one stone (Fig. 17). The four stones had to have been buried as found. The flint assemblage (561 specimens) includes more than 70 cores, 38 tools (lunates, 24%), and 106 bladelets.

Fig. 14. A vertical view of chamber 1 (center) taken from a pole using a wide-angle lens. Note the elon-gated rock cavity of Locus 1 (top right) and th`e variety of mortars, cupmarks and holes hewn into the bedrock floor (scale bar in center is 0.5 m).

499


C-XXIII (type G) is by far the largest at the cave and one of the largest in any Natufian site (65 cm deep and 80 cm wide at the top). It contained red sediment with Natufian flints and bones. The bottom is covered by tufa with adhering Natufian objects. Two flat stones set on edge were found within it. The flints (289 specimens) are dominated by blades/bladelets (53.6%), accompanied by 29 tools (lunates, 34.5%). The only flint assemblages retrieved from within bedrock mortars were compared (HBHs C-I – C-II and C-XXII, Nadel et al. 2008). Lunates were

found in both; they are typical Late Natufian, with abrupt or bipolar retouch, and an average length of 15.2 mm. The main differences are in the num-bers of cores, more than 70 in the pair of mortars, and only one in C-XXIII (though they derive from similar volumes and similar size of assemblages). It may be suggested that many cores were placed in C-I – C-II on purpose, though for an unknown reason. C-XLIV (type G) is located in basin 1 where the highest concentration of burials was encountered (see above). Several small and medium cupmarks and holes are located immediately above the basin. The bottom and the lower parts of the mortar’s walls are covered by tufa containing flints and bones. One individual (Homo 9) was buried horizontally at the top of the mortar, with the body (ribs) below the rim level.

Pairs

An interesting phenomenon is the presence of pairs of bedrock features. The pair of C-I – C-II has been described, and it is unusual within the sets of pairs because it is composed of two mortars

Fig. 15. Mortar C-XVI with a stone set inside.

500


(though not in the same size). In most cases re-corded so far at several sites, the pair is composed of one mortar accompanied by a much smaller feature – usually a tiny hole. Thus, at Raqefet, there is a small cupmark annexed to C-XXIII, and a tiny hole annexed to the deep narrow C-XXI mortar. Furthermore, a large ‘boulder’ mortar was found in Locus 2. At the top of the mortar there is a tiny hewn hole. Though not numerous, there are more examples at Raqefet, as well as other sites. An interesting set is located in the center of Rosh Zin, just in front of monolith (Nadel et al. 2009b). At Huzuk Musa there are several examples, too (see Nadel and Rosenberg herein, Rosenberg and Nadel 2011). Additional interesting cases of large ‘boulder mortars’ from Hayonim terrace and Eynan were recently published (Valla 2009).

Bedrock features in the Late Natufian

Bedrock features were observed at various Late Natufian sites, el-Wad being exceptional, with five large cavities and several small holes hewn into a

leveled bedrock surface, dated to the Early Natufian period (Garrod and Bate 1937:10–11). At Hayonim, bedrock cupmarks were found in the vicinity of the cave (Belfer-Cohen 1988a:167). At Jericho, the Natufian layer, there were several deep and shallow ‘post-holes’ in the soft bedrock (Kenyon and Holland 1981:272, pl. 145a, b). In the first chamber and on the terrace of the Nahal Oren Cave, there are several cases. At Hatoula, there are hundreds of (mostly) small HBHs (Samzun 1994). However, their types and the presence of Pre-Pottery Neolithic A struc-tures at the site suggest that they are Neolithic. These were recently interpreted as the remains of flint quarrying (Grosman and Goren-Inbar 2007). At Huzuq Musa (Lower Jordan Valley), there are tens of bedrock features of various dimensions and types, dated to the Final Natufian (Eitam 2005, 2008; Rosenberg et al. 2011; and see Nadel and Rosenberg herein). At Rosh Zin such features were exposed in the past (Henry 1976). A renewed survey revealed a total of 25 specimens (Nadel et al. 2009b). At Saflulim, there are more than 150 (Goring-Morris 1999) with additional dozens at the nearby Romam and Rosh Horesha sites. At Upper Besor 6, Early Natufian HBHs were also used by Late Natufian occupants (Goring-Morris 1998; Horwitz and Goring-Morris 2001). At Wadi Mataha (Edom Mountains, south-west Jordan), “several bedrock mortars are present on a sandstone ledge… the deepest mortars (up to 72 cm) are grooved from heavy use,” probably of the Late Natufian period (Janetski and Chazan 2004:164). In situ remains from within HBHs were reported from several sites, and can be grouped into two main categories: stones set on edge and buried objects. At Raqefet, stones set on edge were found inside graves and in three excavated bedrock mortars. At el-Wad, “a rough lump of limestone was firmly wedged into Basin 2, and two blocks of the tabular variety into Basin 3” (Garrod and Bate 1937:11); note that Garrod’s “basins” are bedrock mortars in our terminology. Stones placed in the shafts of deep narrow bedrock mortars were found in C-XVI and C-XXI (Raqefet). At Rosh Zin “exhausted mortars, hewn through into non-lithic substratum, were rejuvenated by the positioning of a quartzite cobble in the shaft to seal again the bottom of the mortar” (Henry 1976:337). We have recently encountered five such cases at the site (Nadel et al. 2009b). Interestingly, at Raqefet and Rosh Zin, the stones were placed inside both complete and perforated specimens. Each shaft had only one stone inside, and it seems unreasonable to

Fig. 16. Plan and three sections of bedrock mortars C-I and C-II.

501


assume that one stone fell into each, fitting exactly the contour of the shaft. Flint cores were also placed inside deep narrow bedrock mortars. At Rosh Zin, a flint core was set at the bottom of mortar 17 (Henry 1976:337). This mortar was carved into the bedrock near a large pavement with a monolith incorporated in it. Several unique objects, including five large symmetrical py-ramidal cores, were found there. And as mentioned above a small shallow cupmark was carved adja-cent to the bedrock mortar. Henry suggested that these finds were part of “ritual activity…” (Henry 1976:319–320, figs.11–17.). Garrod reported from el-Wad “a complete limestone mortar… with a hole through its base into which was jammed a flint core” (Garrod and Bate 1937:10). At Nahal Oren, within the graveyard, a high-quality long pyramidal core was found at the bottom of a small conical ash pit (Nadel et al. 1997). At Raqefet, a concentration of flint cores was found in the C-I – C-II complex, and within the graves in Locus 1. Apparently, selected stones and flint cores were repeatedly set in deep bedrock features. The stones did not function as a bottom seal to rejuvenate the mortar, and the flint cores within the shafts are not

a random phenomenon. These deliberate actions do not reflect post-depositional processes, nor do they represent depositories or cashes (why store one stone or one flint core?). We have no doubt that some of the Natufian bedrock features were utilized in processing food (acorns? cereals?), though direct evidence is yet to be exposed. This state of things lags behind the analysis and relevant experiments of Natufian grinding implements (Dubreuil 2004; Dubreuil and Grosman 2009). However, the data from Raqefet and other sites present other aspects of use. These include perforation of deep objects, insertion of stones or cores and manufacturing ‘large-and-small’ pairs. Furthermore, the location and con-text of some specimens are somewhat ‘strange’ in terms of simple functional considerations. For example, on the Raqefet terrace, several features are located on un-even rock exposures, and even on inclined rocks. At Raqefet there are burials on top one bedrock mortar and near several others, and a similar setting was exposed on the terrace of el-Wad (Garrod and Bate 1937; Weinstein-Evron 2009). Furthermore, there are many examples of broken ‘boulder mortars’ set in graves in several

Fig. 17. Small human-made holes and cupmark adjacent to deep mortars (see center of Fig. 14) (scale bar 5 cm).

502


sites. It is thus suggested that the mortar, as an important economic device, was incorporated in the spiritual sphere of the Natufians. Accordingly, it was placed / hewn in association with graves, and in some cases specifically made for the spiritual, not for the mundane (see Nadel and Lengyel 2009 and Rosenberg and Nadel 2011 for further discussion).

7. Conclusions

Raqefet cave may be identified as a burial site, or even a cemetery used during the Late Natufian. The variety of burial customs is wide here, with an emphasis on natural and artificial bedrock basins and mortars. The archaeological remains and cir-cumstances, however, suggest that this was the final but probably not the exclusive function of the site. As we found the Natufian deposits covering the bedrock and filling up the mortars and cupmarks, the Late Natufians most probably cleaned up parts of the first chamber down to bedrock, removing all sedi-ments formed a priori. Thus it was possible to carve the mortars and cupmarks. The sediment removed was used to cover the graves, together with stones collected in the area of the cave and the wadi bed. As the geoarchaeology, archaeozoology and lithic analyses show, the grave fillings contain remains of fire, animal bones, and lithic remains of all stages of tool production. Although it is sometimes hard to differentiate between objects deliberately put in the graves and the general ‘back-ground’ noise, it seems that the sediment in the graves was of Late Natufian domestic origin, formed during the pre-burial stage of occupation at the site.

8. Post-script

The final version of the paper was submitted just after the end of the 2010 season, which focused on Locus 3. Accordingly, two points should be stressed. The first is that five additional Natufian burials were fully excavated, and the presence of more was documented. The second is that some of these burial pits were not found within a Natufi-an layer. Rather, they penetrated through a thin layer, densely packed with Middle Paleolithic flint artefacts.

9. Acknowledgements

We are grateful to Teresa Cabellos for the pro-fessional fieldwork in the 2008 excavation season at Raqefet. We would like to thank Israel Hershkovitz

for giving access to the human remains of Raqefet housed at the Sackler Faculty of Medicine in Tel Aviv. Also, we thank Andrew Garrard for the ar-chives of the excavations at Raqefet between 1970 and 1972. Last but not least, we wish rendering thanks to Ofer Bar-Yosef and François Valla for the invitation to the conference and this book.

References Cited

Belfer-Cohen, A. 1988a The Natufian settlement at Hayonim

Cave. Ph.D. dissertation, The Hebrew University of Jerusalem, Jerusalem.

1988b The Natufian graveyard in Hayonim Cave. Paléorient 14:297-308.

Bocquentin, F. 2003 Pratiques funéraires, paramètres bi-

ologiques et identités culturelles au Natou-fien: une analyse archéo-anthropologique. Ph.D. dissertation, Université Bordeaux 1, Bordeaux.

Dubreuil, L. 2004 Long-term trends in Natufian subsis-

tence: a use-wear analysis of ground stone tools. Journal of Archaeological Science 31:1613-1629.

Dubreuil, L. and L. Grosman 2009 Ochre and hide-working at a Natufian

burial place. Antiquity 83:935-954.Eitam, D.

2005 The food preparation installations and the stone tools. In The Manasse Hill Country Survey. Vol. IV – from Nahal Bezeq to the Sartaba, edited by A. Zertal, pp. 686-689. Appendix 2, site 47 Huzuk Musa. The University of Haifa Press and the Ministry of Defense Press, Haifa.

2008 Plant food in the Late Natufian: the oblong conical mortar as a case study. Journal of the Israel Prehistoric Society - Mitekufat Haeven 38:133-151.

Garrard, A. N. 1980 Man-Animal-Plant Relationships during

the Upper Pleistocene and Early Holocene. Ph.D. dissertation, University of Cam-bridge. Cambridge.

Garrod, D. A. E. and D. M. A. Bate 1937 The Stone Age of Mount Carmel. Claren-

don Press, Oxford. Goring-Morris, A. N.

1998 Mobiliary art from the Late Epipalaeo-

503


lithic of the Negev, Israel. Rock Art Re-search 15:81–88.

1999 Saflulim: a Late Natufian base camp in the central Negev highlands, Israel. Pal-estine Exploration Quarterly 131:36-64.

Grosman, L. and N. Goren-Inbar 2007 “Taming” rocks and changing landscapes:

a new interpretation of Neolithic cup-marks. Current Anthropology 48:732-740.

Henry, D. O. 1976 Rosh Zin: a Natufian settlement near

Ein Avdat. In Prehistory and Palaeoen-vironments in the Central Negev, Israel. Vol.1, edited by A. E. Marks, pp. 317-347. Southern Methodist University Press, Dallas.

Higgs, E. S., Garrard, A. N., Noy, T. and D. Ziffer 1975 Report on excavation at Rakefet, Mount

Carmel, Israel. Unpublished Raqefet Archives.

Horwitz, L. K. and A. N. Goring-Morris 2001 Fauna from the Early Natufian site of

Upper Besor 6 in the Central Negev, Israel. Paléorient 26:111-128.

Inizan, M.-L., Roche, H. and J. Tixier 1976 Avantages d’un traitement thermique

pour la taille des roches siliceuses. Quaternaria 19 (1975-1976):1-18.

Inizan, M.-L. and J. Tixier 2001 L’émergence des arts du feu: le traitement

thermique des roches siliceuses. Paléori-ent 26/2:23-36.

Janetski, J. C. and M. Chazan 2004 Shifts in Natufian strategies and the

Younger Dryas: evidence from Wadi Mataha, southern Jordan. In The Last Hunter-Gatherers in the Near East, edited by C. Delage, pp. 161-168. British Archaeological Reports, International Series 1320. Oxford.

Kenyon, K. M. and T. A. Holland 1981 Excavations at Jericho, Vol.III. The

British School of Archaeology, Jerusalem.Lengyel, G.

2007 Upper Palaeolithic and Epipalaeolithic Lithic Technologies at Raqefet Cave, Mount Carmel East, Israel. BAR Inter-national Series 1681. Oxford.

2009a Operational Schemas in the Upper Palaeolithic and Epipalaeolithic of Raqefet Cave. Analele Universităţii din Oradea, Seria Istorie – Arheologie 19:5-12.

2009b Lithic economy of the Upper Palaeolithic

and Epipalaeolithic of Raqefet cave. Human Evolution 24/1:27-41.

2009c Technical behavior of the Levantine Aurignacian at Raqefet Cave, Mount Carmel, Israel. In Humans: Evolution and Environment, edited by E. Crubezy, E. Cunha, and S. M. de Souza, pp. 59-68. Proceedings of the XV UISPP World Congress (Lisbon, 4-9 September 2006). BAR International Series 2026. Oxford.

Lengyel, G. and F. Bocquentin 2005 Burials of Raqefet Cave in the context of

the Late Natufian. Journal of the Israel Prehistoric Society - Mitekufat Haeven 35:271-284.

Lengyel, G., Nadel, D., Tsatskin, A., Bar-Oz, G., Bar-Yosef Mayer, D. E., Be’eri, R., and I. Hershkovitz

2005 Back to Raqefet Cave, Mount Carmel, Israel. Journal of the Israel Prehistoric Society - Mitekufat Haeven 35:245-270.

Nadel, D. and G. Lengyel 2009 Human-made Bedrock Holes (mortars

and cupmarks) as a Late Natufian social phenomenon. Archaeology, Anthropology and Ethnology in Euroasia 37:37-48.

Nadel, D., Noy, T., Kolska-Horwitz, L. and I. Zohar

1997 A note on new finds from the Natufian graveyard at Nahal Oren. Journal of the Israel Prehistoric Society - Mitekufat Haeven 27:63-74.

Nadel, D., Lengyel, G., Bocquentin, F., Tsatskin, A., Rosenberg, D., Yeshurun, R., Bar-Oz, G., Bar-Yosef Mayer, D. E., Be’eri, R., Conyers, L., Filin, S., Hershkovitz, I. and A. Kurzawska

2008 Raqefet Cave: the 2006 Excavation Season. Journal of the Israel Prehistoric Society - Mitekufat Haeven 38:59-131.

Nadel, D., Lengyel, G., Cabellos Panades, T., Bocquentin, F., Rosenberg, D., Yeshurun, R., Brown-Goodman, R., Tsatskin, A., Bar-Oz, G. and S. Filin

2009a The Raqefet Cave 2008 excavation season. Journal of the Israel Prehistoric Society - Mitekufat Haeven 39:21-61.

Nadel, D., Rosenberg, D. and R. Yeshurun 2009b The Deep and the shallow: The role of

Natufian bedrock features at Rosh Zin, Central Negev, Israel. Bulletin of the American Schools of Oriental Research 355:1-29.

Nadel, D. and D. Rosenberg herein The Final Epipaleolithic / PPNA site of

504


Huzuq Musa (Jordan Valley). Edited by O. Bar-Yosef and F. R. Valla. International Monographs in Prehistory, Ann Arbor.

Noy, T. and E. S. Higgs 1971 Raqefet Cave. Israel Exploration Journal

21:225-226.Olami, Y.

1984 Prehistoric Carmel. Israel Exploration Society and M. Stekelis Museum of Prehistory, Jerusalem and Haifa.

Rosenberg, D. 2008 The possible use of acorns in past

economies of the Southern Levant: A staple food or a negligible food source? Levant 40/2:167-175.

Rosenberg, D. and D. Nadel 2011 On floor level: PPNA indoor cupmarks and

their Natufian forerunners. In Studies in Technology, Environment, Production, and Society, edited by E. Healey, S. Campbell and O. Maeda, pp. 99-108. Studies in Early Near Eastern Production, Subsistence, and Environment 13. ex oriente, Berlin.

Rosenberg, D., Groman-Yaroslavski, I., Winter, H., Yeshurun, R., Zertal, A., Brown-Goodman, R. and D. Nadel

2010 Huzuq Musa – A Final Epipalaeolithic / PPNA site in the Jordan Valley. Paléorient 36(2):189-204.

Sarel, J. 2004 The Middle-Upper Palaeolithic Transition

in Israel. BAR International Series 1229. Oxford.

Samzun, A. 1994 Le mobilier en pierre. In Le Gisement

de Hatula en Judée Occidentale, Israël, edited by M. Lechevallier, A. Ronen, pp. 211–226. Mémoires et Travaux du Centre de Recherche Français de Jérusalem 8. Association Paléorient, Paris.

Valla, F. R. 2009 Une énigme natoufienne: les ‘mortiers’

enterrés. De Méditerrannée et d’ail-leurs… Mélanges offerts à Jean Guilaine, edited by D. Fabre, pp. 751-759. Archives d’Écologie Préhistorique, Toulouse.

Valla, F. R., Khalaily, H., Samuelian, N., Bocquentin, F., Delage, C., Valentin, B., Plisson, H., Rabinovich, R. and A. Belfer Cohen

1998 Le Natoufien Final et les Nouvelles Fouilles à Mallaha (Eynan), Israël 1996-1997. Mitekufat Haeve -. Journal of the

Israel Prehistoric Society - Mitekufat Haeven 28:105-176.

Valla, F. R., Khalaily, H., Samuelian, N., March, R., Bocquentin, F., Valentin, B., Marder, O., Rabinovich, R., Le Dosseur, G., Dubreuil, L. and A. Belfer-Cohen

2001 Le Natoufien Final de Mallaha (Eynan), Deuxième Rapport Préliminaire: Les Fouilles de 1998 et 1999. Journal of the Israel Prehistoric Society - Mitekufat Haeven 31:43-184.

Valla, F. R., Khalaily, H., Valladas, H., Kaltnecker, E., Bocquentin, F., Cabellos, T., Bar-Yosef Mayer, D., Le Dosseur, G., Regev, L., Chu, V., Weiner, S., Boaretto, E., Samuelian, N., Valentin, B., Delerue, S., Poupeau, G., Bridault, A., Rabinovitch, R., Simmons, T., Zohar, I., Ashkenazi, S., Delgado Huertas, A., Spiro, B., Mienis, H. K., Rosen, M., Porat, N. and A. Belfer-Cohen

2007 Les fouilles de Aïn Mallaha (Eynan) de 2003-2005: quatrième rapport prélimi-naire. Journal of the Israel Prehistoric Society - Mitekufat Haeven 37:135-383.

Weinstein-Evron, M. 2009 Archaeology in the Archives. Unveiling

the Natufian Culture of Mount Carmel. Brill, Boston.

Wright, K. 1991 The origins and development of ground

stone assemblages in late Pleistocene Southwest Asia. Paléorient 17/1:19-45.

1992 Ground Stone Assemblage Variation and Subsistence Strategies in the Levant, 22.000 to 5.000 b.p. Ph.D. dissertation, Yale University. Yale.

1994 Ground-stone tools and hunter-gatherer subsistence in Southwest Asia: impli-cations for the transition to farming. American Antiquity 59/2:238-263.

Ziffer, D. 1978a The use of technological and metric data

in the study of four Levantine Aurigna-cian sites in the Mount Carmel region. A preliminary study. Paléorient 4:71-94.

1978b A Re-evaluation of the Upper Palaeolithic industries of the Kebara Cave and their place in the Aurignacian Culture of the Levant. Paléorient 4:273-293.

1981 Yabrud Shelter II – A reconsideration of its cultural composition and its rel-evance to the Upper Palaeolithic cul-tural sequence in the Levant. Quartär 31–32:69–91.

raqefet cave (2004 – 2008 seasons of excavation)

Documents