SECTION 9 Le Néolithique au Proche Orient et en Europe / The Neolithic in the Near East and Europe, SECTION 10 L'âge du cuivre au Proche Orient et en Europe / The Copper Age in the Near East and Europe: Sessions générales et posters / General Sessions and Posters 9781841716534, 9781407327280

The General Sessions and Posters from Sections 9 and 10 of the XIVth UISPP Congress, University of Liège, Belgium, 2-8 S

220 110 52MB

French Pages [407] Year 2004

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

SECTION 9 Le Néolithique au Proche Orient et en Europe / The Neolithic in the Near East and Europe, SECTION 10 L'âge du cuivre au Proche Orient et en Europe / The Copper Age in the Near East and Europe: Sessions générales et posters / General Sessions and Posters
 9781841716534, 9781407327280

Table of contents :
Front Cover
Title Page
Copyright
TABLE DES MATIÈRES / TABLE OF CONTENTS
THE FIRST FARMERS IN SILESIA
LE NÉOLITHIQUE ET L´ÉNÉOLITHIQUE DE LA SILÉSIE TCHÈQUE
ECONOMIE DU COMBUSTIBLE AU NÉOLITHIQUE FINAL : EXEMPLE DE LA STATION 19 DU LAC DE CHALAIN (JURA, FRANCE)
LONG BLADES/SUPERBLADES IN ANATOLIAN AND EAST BALKAN NEOLITHIC CONTEXTS
NEOLITHIC EXPLOITATION AND TRADE OF OBSIDIAN IN THE CENTRAL MEDITERRANEAN: NEW RESULTS AND IMPLICATIONS FOR CULTURAL INTERACTION
OUTILS AGRICOLES DES ASSEMBLAGES NÉOLITHIQUES EN SILEX DE LA BULGARIE - UN CONTEXTE PLUS LARGE
CAKMAKLI TEPE: AN ANCIENT NEOLITHIC VILLAGE IN THE AEGEAN COASTLAND OF NORTHWESTERN TURKEY
LA FONCTION POLYVALENTE DES GROTTES DU SUD DE LA GRÈCE PENDANT LE NÈOLITHIQUE : LE CAS DES DEUX GROTTES À ALEPOHORI (LACONIE)
THE NEOLITHIC SETTLEMENT OF CARRARA SAN FRANCESCO IN THE ITALIAN SOUTH-EASTERN NEOLITHIC
THE OLDEST WOVEN TEXTILE OF THE FUNNEL BEAKER CULTURE (4000-2900 CAL BC) IN NORTH AND CENTRAL EUROPE
LA NÉCROPOLE NÉOLITHIQUE DE LA JARDELLE À DISSAY (VIENNE, FRANCE)
LA NÉOLITHISATION EN HESBAYE (BELGIQUE): APPORTS DES FOUILLES RÉCENTES (1996-1999) SUR LE TRACÉ ORIENTAL DU TGV ET LE SITE DE ANS, À ALLEUR
1995-2001 : SEPT ANNÉES DE RECHERCHES MÉGALITHIQUES EN CALESTIENNE (BELGIQUE)
CERAMIC STYLE IN NEOLITHIC SOCIETIES IN GALICIA (NW IBERIAN PENINSULA). SIMILARITIES AND DIFFERENCES IN PATTERNS OF FORMAL REGULARITY
NEOLITHIC REMAINS IN UMBRIA (ITALY)
LES MINES NÉOLITHIQUES DE SPIENNES : ÉTAT DES CONNAISSANCES ET PERSPECTIVES DE RECHERCHE
LE GISEMENT NÉOLITHIQUE DE “EL RETAMAR”. UN EXEMPLE DES COMMUNAUTÉS DU VIe MILLÉNAIRE AVANT NOTRE ÈRE DANS LA BAIE DE CADIX (ESPAGNE)
INTERACTING COSMOLOGY: RITUAL USE OF FIRE DURING TWO PHASES OF THE SOUTHERN SCANDINAVIAN NEOLITHIC
THE OLDEST STONE RAMPART: ENCLOSURES AND MEGALITHIC TOMBS OF THE FUNNEL BEAKER CULTURE (4100-2800 CAL BC) IN NORTH AND CENTRAL EUROPE
LA PLACE DE LA CRÉMATION DANS LE TRAITEMENT DES DÉFUNTS À LA FIN DU NÉOLITHIQUE EN FRANCE
ARCHEOLOGICAL AND ANTHROPOLOGICAL RESEARCH ON NEOLITHIC BURIALS FROM SOUTH-EASTERN SICILY: VULPIGLIA (PACHINO, SIRACUSA)
HUMAN-ELK IMAGES IN NORTHWESTERN RUSSIAN NEOLITHIC ART
THE ROLE OF OBSIDIAN IN THE NEOLITHIC: A SYMBOLIC EXPRESSION OF HUMAN DOMESTICATION?
PREHISTORIC ART OF THE NEOLITHIC BRONZE AGE
LE RITUEL FUNÉRAIRE AU COURS DU NÉOLITHIQUE ITALIEN
LITHIC PROCUREMENT AND PRODUCTION IN THE LATE NEOLITHIC OF SOUTHWEST GERMANY
DÉBITAGE ET FAÇONNAGE DE LA MÉTADOLÉRITE DE TYPE A (PLUSSULIEN) : RÉALISATION DE TESTS ET IMPLICATIONS ARCHÉOLOGIQUES
PRELIMINARY RESULTS OF THE INVESTIGATION OF THE ENEOLITHIC SITE OF BODAKI (WESTERN UKRAINE) WITH LITHIC WORKSHOPS
LATE NEOLITHIC FLINT INDUSTRIES IN THE GERMAN AND SWISS ALPINE FORELAND: ECONOMIC PATTERNS AT A VARIETY OF SPATIAL SCALES
AGRICULTURAL LANDSCAPES IN THE TOPOGRAPHIC NEOLITHIC ENGRAVINGS OF THE ALPINE ARC
LE MATÉRIEL LITHIQUE DE SPIENNES DANS LE MUSÉE ARCHÉOLOGIQUE DE ZAGREB
CONTRIBUTION A L’ÉTUDE DE LA CIRCULATION DES MATIÈRES PREMIÈRES DANS LA RÉGION DES ALPES OCCIDENTALES : LE QUARTZ HYALIN
NEOLITHIC REMAINS IN GROTTA BELLA (TERNI, UMBRIA, CENTRAL ITALY)
PRE-POTTERY NEOLITHIC AMAZONITE BEAD WORKSHOPS IN SOUTHERN JORDAN
A GIFT FOR ETERNITY. ARCHAEOLOGICAL, EXPERIMENTAL AND TECHNOLOGICAL STUDY OF TEN SHELL-BEADS OF A NEOLITHIC BURIAL FROM VULPIGLIA (PACHINO, SIRACUSA)
CHARACTERIZATION OF NEW NEOLITHIC ENGRAVED PEBBLES IN SARDINIA
LES RESTES HUMAINS DE LA GROTTE DE L’ADAOUSTE DU NEOLITHIQUE ANCIEN FINAL (BOUCHES DU RHÔNE, FRANCE) : CANNIBALISME OU RITUEL FUNERAIRE ?
SALT MAKING IN EUROPEAN NEOLITHIC SOCIETIES
THE VISUAL MANIFESTATION OF THE CHALCOLITHIC RELIGION
ELEMENTS FOR AN ARCHAEOLOGY OF RELIGION IN ITALIAN PREHISTORY
RITUAL AND SYMBOLIC EVIDENCE IN THE NEOLITHIC OF SOUTH-EASTERN ITALY
FOSSES CULTURELLES (BÓTHROI) ET A CARACTÈRE CULTUREL DU NÉOLITHIQUE ET L’ÉNÉOLITHIQUE DE LA ROUMANIE
LA VALEUR SACRÉE DES NOMBRES DANS L’ÉNÉOLITHIQUE DE ROUMANIE
THE CHALCOLITHIC AND EARLY BRONZE AGE IN THE BIRECIK REGION(SOUTHEASTERN TURKEY)
ALLUVIAL LANDSCAPES IN THE TEMPERATE BALKAN NEOLITHIC: INVESTIGATING CHANGES IN FIFTH MILLENNIUM BC LAND-USE
NOUVELLES DONNÉES CONCERNANT LES CROYANCES MAGIQUES DES COMMUNAUTÉS DE LA CIVILISATION PRÉCUCUTENI (ÉNÉOLOTHIQUE ANCIEN) DE L’EST DE LA ROUMANIE
PODURI – DEALUL GHINDARU : UN TELL CHALCOLITHIQUE DANS L’EST DE LA ROUMANIE
THE PREHISTORIC SETTLEMENT AT MONTE COVOLO (BRESCIA-ITALY). EXCAVATIONS 1998-1999: THE COPPER AGE LEVELS
EXCAVATIONS AT THE PRE-NURAGIC SANCTUARY OF MONTE D’ACCODDI (SARDINIA, ITALY)
L’HYPOGÉE DES BOILEAU (VAUCLUSE, FRANCE): ÉTUDE FONCTIONNELLE ET ANTHROPOLOGIQUE
LE DÉVELOPPEMENT DE LA COMPLEXITÉ SOCIALE COMPRISE À LA FIN DU IIIeme ET PENDANT LE IIeme MILLÉNAIRES AV. J.-C. EN ANDALOUSIE OCCIDENTALE (ESPAGNE): LE CAMPANIFORME
THE PREHISTORIC SETTLEMENT AT REALMESE NEAR CALASCIBETTA (ENNA, ITALY)

Citation preview

BAR S1303

Actes du XIVème Congrès UISPP, Université de Liège, Belgique, 2–8 septembre 2001 Acts of the XIVth UISPP Congress, University of Liège, Belgium, 2–8 September 2001

2004 SECTIONS 9 & 10

SECTION 9

Le Néolithique au Proche Orient et en Europe The Neolithic in the Near East and Europe SECTION 10

THE NEOLITHIC AND THE COPPER AGE IN THE NEAR EAST AND EUROPE

B A R

L’âge du cuivre au Proche Orient et en Europe The Copper Age in the Near East and Europe Sessions générales et posters General Sessions and Posters Édité par / Edited by Le Secrétariat du Congrès

Présidents de la Section 9 : Ivan Jadin & Anne Hauzeur Présidents de la Section 10 : Nicolas Cauwe, Marc Vander Linde, Önhan Tunca & Marc Lebeau

BAR International Series 1303 2004

Actes du XIVème Congrès UISPP, Université de Liège, Belgique, 2-8 septembre 2001 Acts of the XIVth UISPP Congress, University of Liège, Belgium, 2-8 September 2001

SECTION 9

Le Néolithique au Proche Orient et en Europe The Neolithic in the Near East and Europe SECTION 10

L’âge du cuivre au Proche Orient et en Europe The Copper Age in the Near East and Europe Sessions générales et posters General Sessions and Posters Édité par / Edited by Le Secrétariat du Congrès Présidents de la Section 9 : Ivan Jadin & Anne Hauzeur Présidents de la Section 10 : Nicolas Cauwe, Marc Vander Linde, Önhan Tunca, Marc Lebeau

BAR International Series 1303 2004

Published in 2016 by BAR Publishing, Oxford BAR International Series 1303 Acts of the XIVth UISPP Congress, University of Liège, Belgium, 2-8 September 2001 Section 9: Le Néolithique au Proche Orient et en Europe / The Neolithic in the Near East and Europe. Section 10: L'âge du cuivre au Proche Orient et en Europe / The Copper Age in the Near East and Europe. Sessions générales et posters / General Sessions and Posters © The editors and contributors severally and the Publisher 2004 Avec la collaboration du Ministère de la Région Wallonne. Direction générale de l'Aménagement du territoire, du Logement et du Patrimoine. Subvention n˚03/15718 Mise en page / Editing : Rebecca MILLER Typesetting and layout: Darko Jerko Marcel OTTE, Secrétaire général du XIVème Congrès de l'U.I.S.P.P. Université de Liège, Service de Préhistoire 7, place du XX août, bât. A1, 4000 Liège Belgique Fax 0032/4/366.55.51 Tél. 0032/4/366.53.41 Email : [email protected] Web : http://www.ulg.ac.be/prehist The authors' moral rights under the 1988 UK Copyright, Designs and Patents Act are hereby expressly asserted. All rights reserved. No part of this work may be copied, reproduced, stored, sold, distributed, scanned, saved in any form of digital format or transmitted in any form digitally, without the written permission of the Publisher.

ISBN 9781841716534 paperback ISBN 9781407327280 e-format DOI https://doi.org/10.30861/9781841716534 A catalogue record for this book is available from the British Library BAR Publishing is the trading name of British Archaeological Reports (Oxford) Ltd. British Archaeological Reports was first incorporated in 1974 to publish the BAR Series, International and British. In 1992 Hadrian Books Ltd became part of the BAR group. This volume was originally published by Archaeopress in conjunction with British Archaeological Reports (Oxford) Ltd / Hadrian Books Ltd, the Series principal publisher, in 2004. This present volume is published by BAR Publishing, 2016.

BAR

PUBLISHING BAR titles are available from:

E MAIL P HONE F AX

BAR Publishing 122 Banbury Rd, Oxford, OX2 7BP, UK [email protected] +44 (0)1865 310431 +44 (0)1865 316916 www.barpublishing.com

TABLE DES MATIÈRES / TABLE OF CONTENTS SECTION 9 SESSION GÉNÉRALE / GENERAL SESSION 9-I Société, économie et études régionales Direction: Jak Yakar The First Farmers in Silesia .................................................................................... 1 M. Furmanek Le Néolithique et l’Énéolithique de la Silésie tchèque ............................................. 5 V. Janak Économie du combustible au Néolithique final : Exemple de la station 19 du lac de Chalain (Jura, France) ........................................................................ 9 A. Dufraisse

Technologies lithiques (première partie) Direction: Charles-Tanguy Le Roux Long Blades/Superblades in Anatolian and East Balkan Neolithic Contexts ........ 17 T. Tsonev Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean: New Results and Implications for Cultural Interaction ...................................... 25 R.H. Tykot Outils agricoles des assemblages néolithiques en silex de la Bulgarie un contexte plus large ...................................................................................... 37 M. Gurova

SESSION GÉNÉRALE / GENERAL SESSION 9-II Acualités de la recherche Direction: Jean-Pierre Pautreau Cakmakli Tepe: An Ancient Neolithic Village in the Aegean Coastland of Northwestern Turkey .................................................................................... 53 S. Tiné, A. Traverso La fonction polyvalente des grottes du Sud de la Grèce pendant le Néolithique: Le cas des deux grottes à Alepohori (Laconie) ........................ 57 C. Contaxi, H. Stravopodi i

The Neolithic Settlement of Carrara San Francesco in the Italian South-Eastern Neolithic ............................................................... 61 R. Striccoli The Oldest Woven Textile of the Funnel Beaker Culture (4000-2900 cal. BC) in North and Central Europe ............................................................................ 67 C. Baldia La nécropole néolithique de la Jardelle à Dissay (Vienne, France) ...................... 71 J.-P. Pautreau, B. Farago-Szekeres, P. Mornais La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999) sur le tracé oriental du TGV et le site de Ans, à Alleur ................ 81 D. Bosquet, H. Fock, C. Goffieul, J.-P. Marchal, D. Preud’homme 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique) ......................................................................................................... 95 C. Frébutte, M. Toussaint, F. Hubert, S. Pirson, P. Masy Ceramic Style in Neolithic Societies in Galicia (NW Iberian Peninsula) Similarities and Differences in Patterns of Formal Regularity ........................ 109 P. Prieto Martínez Neolithic Remains in Umbria (Italy) ..................................................................... 119 M.C. De Angelis, A. Moroni Les mines néolithiques de Spiennes : État des connaissances et perspectives de recherche ......................................................................... 129 H. Collet Le gisement Néolithique de «El Retamar». Un exemple des communautés du VIe millénaire avant notre ère dans la Baie de Cadiz (Espagne) .............. 135 J. Ramos, M. Lazarich, N. Herrero, V. Castañeda, M. Pérez, S. Domínguez-Bella, M.E. García & I. Cáceres

SESSION GÉNÉRALE / GENERAL SESSION 9-III Art et spiritualité Direction: Lars Larsson Interacting Cosmology. Ritual Use Of Fire during Two Phases of the Southern Scandinavian Neolithic ......................................................... 147 L. Larsson The Oldest Stone Rampart: Enclosures and Megalithic Tombs of the Funnel Beaker Culture (4100-2800 cal. BC) in North and Central Europe ................. 153 M.O. Baldia La place de la crémation dans le traitement des défunts à la fin du Néolithique en France .................................................................... 163 E. Gatto Archaeological and Anthropological Research on Neolithic Burials from South-Eastern Sicily: Vulpiglia (Pachino, Siracusa) ............................... 171 L. Guzzardi, M.R. Iovino, A. Rivoli, J. Robb Human-Elk Images in Northwestern Russian Neolithic Art ................................. 177 E. Kashina ii

The Role of Obsidian in the Neolithic: A Symbolic Expression of Human Domestication? .............................................................................. 181 B. Tripković Prehistoric Art of the Neolithic Bronze Age ......................................................... 191 N. Syriamina Le rituel funéraire au cours du Néolithique italien ............................................... 199 R. Grifoni Cremonesi

Technologies lithiques (seconde partie) Direction: Charles-Tanguy Le Roux Lithic Procurement and Production in the Late Neolithic of Southwestern Germany ............................................................................. 207 P. Kieselbach Débitage et façonnage de la métadolérite de type A (Plussulien) : Réalisation de tests et implications archéologiques ....................................... 217 C. Sestier, Ch.-T. Leroux, B. Faucillon Preliminary Results of the Investigation of the Eneolithic Site of Bodaki (Western Ukraine) with Lithic Workshops ...................................................... 225 N. Skakun, A. Samzun Late Neolithic Flint Industries in the German and Swiss Alpine Foreland: Economic Patterns at a Variety of Spatial Scales ........................................... 231 D. Underwood

SECTION 9 – POSTERS Agricultural Landscapes in the Topographic Neolithic Engravings of the Alpine Arc ............................................................................................. 235 A. Arcà Le matériel lithique de Spiennes dans le Musée Archéologique de Zagreb ........ 243 I. Karavanić, J. Balen Contribution a l’étude de la circulation des matières premières dans la région des alpes occidentales: le quartz hyalin ................................. 251 I. Baroni, G. Onoratini Neolithic Remains in Grotta Bella (Terni, Umbria, Central Italy) .......................... 257 M.C. De Angelis, G. Guerreschi, A. Moroni Lanfredini Pre-Pottery Neolithic B Amazonite Bead Workshops in Southern Jordan ........... 265 M. Fabiano, F. Berna, E. Borzatti von Löwenstern A Gift for Eternity. Archaeological, Experimental and Technological Study of Ten Shell-Beads of a Neolithic Burial from Vulpiglia (Pachino, Siracusa) ........................................................................................ 275 L. Guzzardi, S. Chilardi, M.R. Iovino, A. Rivoli, J. Robb Characterization of New Neolithic Engraved Pebbles in Sardinia ....................... 279 C. Lugliè iii

Les restes humains de la grotte de l’Adaouste du Néolithique ancien final (Bouches du Rhône, France) : Cannibalisme ou rituel funeraire ? ................ 289 B. Mafart, I. Baroni, G. Onoratini Salt Making in European Neolithic Societies ....................................................... 295 O. Weller

SECTION 10 COLLOQUE / SYMPOSIUM 10.1 Formes de manifestation du sacré dans le Néolithique et le Chalcolithique Coordinateur / Coordinator: Nicolae Ursulescu The Visual Manifestation of the Chalcolithic Religion .......................................... 299 T. Shay Elements for an Archaeology of Religion in Italian Prehistory ............................. 303 S. Tiné, A. Salerno, V. Tiné Ritual and Symbolic Evidence in the Neolithic of South-eastern Italy ................. 309 R. Striccoli Fosses culturelles (bóthroi) et a caractère culturel du néolithique et l’énéolithique de la Roumanie .................................................................... 317 V. Cotiugã, S. Haimovici La valeur sacrée des nombres dans l’Énéolithique de Roumanie ...................... 325 N. Ursulescu

SESSION GÉNÉRALE / GENERAL SESSION 10-I The Chalcolithic and Early Bronze Age in the Birecik Region (Southeastern Turkey).................................................................................... 333 J. Fuensanta, P. Charvát, M.S. Rothman Alluvial Landscapes in the Temperate Balkan Neolithic: Investigating Changes in Fifth Millennium BC Land-Use ............................... 339 D.W. Bailey, R. Andreescu, A. Howard, M. Macklin & S. Mills Nouvelles données concernant les croyances magiques des communautés de la civilisation Précucuteni (Énéolithique ancien) de l’est de la Roumanie .............................................................................................. 343 N. Ursulescu Poduri – Dealul Ghindaru: Un tell chalcolithique dans l’est de la Roumanie ....... 349 D. Monah, D.N. Popovici, Gh. Dumitroaia The Prehistoric Settlement at Monte Covolo (Brescia, Italy). Excavations 1998-1999: The Copper Age Levels .......................................... 359 R. Poggiani Keller, M. Baioni, F. Martini, D. Lo Vetro Excavations at the Pre-Nuragic Sanctuary of Monte d’Accoddi (Sardinia, Italy) ............................................................................................... 369 S. Tine, A. Traverso iv

L’hypogée des Boileau (Vaucluse, France): étude fonctionnelle et anthropologique ......................................................................................... 375 E. Mahieu, W. Devriendt, H. Barge, I. Lalys, P. Adalian, Y. Ardagna, M. Signoli Le développement de la complexité sociale comprise à la fin du IIIeme et pendant le IIeme millénaires av. J.-C. en Andalousie occidentale (Espagne): Le Campaniforme ........................................................................ 385 M. Lazarich González The Prehistoric Settlement at Realmese near Calascibetta (Enna, Italy) ............ 395 R.M. Albanese, E. Procelli

v

vi

M. Furmanek: The First Farmers in Silesia

THE FIRST FARMERS IN SILESIA Mirosław FURMANEK

Investigations on early farmers’ communities in Silesia were undertaken by German archaeologists before 1945. These investigations developed most extensively, however, in the past forty years, carried out by Prof. Anna KulczyckaLeciejowicz and Prof. Wlodzimierz Wojciechowski, the authors of numerous monographs concerning the sites Strzelin, Skoroszowice, Niemcza, Stary Zamek and Gniechowice. The results of these investigations formed the basis for further taxonomic considerations dealing with economic problems, the exchange of commodities and others. Of much importance for investigations on the expansion of settlements is the project of Archaeological Survey of Poland.

from the Cracow region seem to have stopped. The postlinear development of these communities is connected with the OLG (Oberschlesischen Lengyel-Gruppe; cf. Janak) or MK (Malice Kulture). At present, we have 12 C14 dates concerning Silesia, obtained from LBK settlements, as well as three dates from the SBK. These data concern settlements existing in later stages of the development of the LBK and the beginnings of the SBK. The data concerning stage IIc of the settlement at Niemcza are comparatively rather late. Summing them up, we may date them from 5260 to 5050 BC, i.e., parallel to the older stage in the development of the LBK III in Silesia. Comparing all the data from the settlement of Strachów (LBK III), the time of its existence might be dated from 5210 to 5050 BC (2 sigma: 5240-4990 BC). Although we may fully agree as far as the beginning of the LBK III is concerned, the final stage of this settlement may probably set it back to about 4950-4900 BC. The other LBK III data fully comply with the settlement dates earlier.

The area of Silesia comprises the southwestern part of Poland and the region of Opava, which belongs to the Czech Republic. The environmental conditions of this area are diversified. It may be divided into three fundamental regions: mountains, piedmont uplands and lowland plains. In the northern part of the piedmont uplands up the Oder river, there are fertile soils consisting of humus mould, loess, clay and loam. In the wooded regions around Niemodlin and in Nether Silesia, the soil is less fertile, consisting mainly of sand. The valleys of the Odra and most other rivers are covered with alluvial soil, mainly silt. The part of Silesia situated along the right-hand side of the Odra consists of sandy soils too.

In the case of the SBK, we have three dates from Silesia, connected with the development of the SBK I. Data from Niemcza are connected with stage Ia, those from Stary Zamek with stage Ib. Calibrating these data, we can determine the development of the SBK I in Silesia from 4950/4900 to 4800/ 4750 BC. It is, however, still impossible to find out the internal division of time.

The periodisation of the LBK (Linearbandkeramik) is based on patterns derived from Central Europe. Here we can distinguish three stages of its development. Studies concerning the ceramic styles have made it possible to particularize the excavations encountered in the region of Wrocław. Within stage II, we may separate three stylistic substages, and within stage II, two stylistic substages. The Nethersilesianpostlinear material belongs taxonomically to the SBK (Stichbandkeramik). Fragmentary publications concerning the most important sites impede more detailed taxonomic studies. Thus, the development of this culture would consist of two stages, the older one (SBK I) comprising two stylistic substages, and the younger one displaying distinctly some share of the Lengyel component of ceramic decorative style. Also this stage comprises two substages. In the region of Racibórz, the postlinear communities developed rather differently. Most probably between the decline of LBK and the appearance of post-linear formations, a settlers’ hiatus came into being. There do not exist any SBK I materials nor any pf the earliest Lengyel materials. Neither can it be excluded that groups of LBK communities existed, in the region of Wrocław, for a longer time, though it is more reasonable to assume the probability of a gap in settlement. It ought to be stressed that with the turning up of SBK I in the western part of Silesia, the supply of flint implements

Early farmers’ sites concentrated on the left bank of the Odra. On the right-hand bank, several ceramic and stone products have been found, which, however, are not connected with any early farmers’ settlements. Probably they are to be ascribed to contacts with some hunters and foragers. During the LBK period, two settlement concentrations are to be distinguished – the eastern one between the Odra and the Osobłoga, and the western one between the Nysa Klodzka and the Ślęza. Both are characterised by their compactness and a considerable density of settlements. A minor settlement concentration is located near Głogów. Beside these, there are only poor concentrations of settlements of only single sites. In the SBK period, the number of sites distinctly decreased. The two main concentrations around Wrocław and Racibórz were restricted, whereas the role of the concentrations around Głogów and Jawor were somewhat increased. Noticeable is the lack of settlement expansions in the course of the formation of the SBK. In the SBK period, the number of sites obviously decreased, particularly in the Upper Silesian region of Racibórz. In the 1

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

early stage of the SBK, settlements comprised principally the same regions as in the case of the LBK, i.e., the region of Wrocław. Excavations of the older stage of the SBK have been made in the region situated to the north of the Nysa, reaching to the east as far as the Osobloga. Around Racibórz, no finds are to be noted, corresponding chronologically to the SBK I. Neither are there any SBK I finds in the region of Głogów, Legnica and Jawor. During this time we seem to have a hiatus in the settlement process. During the SBK II, settlements occur in the same regions as during the LBK, i.e., the regions of Wrocław, Legnica, Jawor and Głogów. In Upper Silesia probably at that time, new groups of farmers turned up, identified as OLG I.

activity took place inside the houses and, to a certain extent, within its surroundings. The presence of storage pits evidences the storage of particular products outside the house. These features functioning as cellars-granaries were frequently secondarily used as a burial place. Spatial diversification is particularly well distinguishable here, which is related to the division of social roles (the southern part is associated with a woman and her occupations, e.g., food preparation, whereas the northern part is associated with a man’s occupations). During the LBK II-III, the majority of farming activities took place outside the house, mainly in the limits of the complexes of pits adjacent to the house. The zone of more remote features is mainly associated with the removal of impurities. A smaller number of features considered as cellars-granaries indicate that products were probably stored within the limits of the separated parts of the house. During the SBK, there is a lack of strict data referring to the form of activities within household limits. Probably they were similar to those during the LBK II-III. Attention is drawn to the features in the form of cellars, implying that the storage of products was “transferred” from houses to ground features.

Relations between the settlements and the environmental conditions indicate many similarities. If settlements were founded in large valleys, mainly the higher parts of these valleys were chosen, i.e., the slopes, SBK settlements normally being situated higher up than those of LBK sites. In both cases, exposed locations were preferred, rather to the south and east, a preference more characteristic for SBK settlements. Of much importance was also the kind of soil. Black and grey soils were usually preferred, mainly composed of loess and soils similar to this, although clay soil and sandy soils or basalt and metamorphic soils were not shunned.

In the species structure of animal bones of livestock, cattle hold a predominant position. A tendency of a falling number in small ruminants and a growing numbers of pigs’ bones can be observed. The breeding of animals in LBK farming overall seems to have played a supplementary role in relation to the decisive role of time-consuming jobs associated with the development of intensive farming. Generally, it is thought that the breeding of animals during the LBK was certainly aimed at the production of meat. The age distribution of cattle points out the trace presence of older individuals, which is interpreted as the effect of directing the breeding at the expansion of a herd or may result from the adaptation to milk production. This is also evidenced by the ceramic sieves that were found. Breeding, mostly cattle breeding, was also of social importance with regard to the sustenance and creation of group identity. This consumption was regulated culturally. The use of a pig as a slaughter animal is quite dubious. It may be reflected in spatial behaviours. The occurrence of pigs’ bones in features of Stary Zamek is precluded by the occurrence of the bones of small ruminants and wild animals.

Pits were more regular in the case of the LBK I. They were mostly round or rectangular. Irregular structures were comparatively rare, and there were no large pits at all. Metrical analyses have shown a high level of standardisation. LBK II and LBK III structures are characterised by the highest irregularity of their design. Structures which were constructed during the SBK unit characteristic features of both stages. They are more regular in their design (round or rectangular) with more regular cross-sections. There are, however, also complexes of pits of irregular shapes, but they never occur together with large houses. Differences between LBK I, LBK II-III and SBK are noticeable both in the case of dwellings embedded in the ground and in the shape of above-ground buildings. In the LBK I case, as an example may serve a house in Stary Zamek, rectangular, consisting of only one nave. During the LBK II-III, we have classical long, rectangular houses with distinct internal partitions. During the SBK, trapezoidal houses are to be found with an obvious evolution in time.

Paleobotanic analyses regarded mostly plant remains from puggings and ceramics. Wheat dominates during the LBK I (emmer and einkorn). During the LBK II, remnants are diversified. At Stary Zamek and Strzelin 16, wheat dominates, while at Sobocisko and Pietrowice Wielkie it is barley. This diversification is continued during the LBK III. Remnants of millet discovered at Strzelin 16 may prove that it had started to be grown during the LBK III. Also at Strzelin 16, it was observed that there is divisibility in the occurrence of plant remains in some sample, which implies the divisibility in the growing of consecutive grains.

In the spatial organisation of settlements, a basic role is played by households. Apart from a house, the households also include a different number of pits. During the LBK I, this number equals several or more than ten features. During the LBK II-III, a household is made up of the large complexes of pits situated by the house and the group of smaller features located farther from the house. During the SBK, there is a high number of pits near a house. Yet, in other cases, a household is formed by one house and a smaller number (13) of pits.

The role of hunting is signified by bone remnants found in those settlements. During the LBK I, the percentage of wild animals is relatively high and amounts to 41.97%. Relatively more were found in the largest groups of bones from Gniechowice, and their percentage at Stary Zamek is lower. In the following phases of the LBK and SBK I, the percentage

The differentiation of activity zones within the area of households is also noticeable. During the LBK I, economic 2

M. Furmanek: The First Farmers in Silesia

declines and is very low. Its frequency rises again during the SBK II and is 29.79%.

the houses indicates that they could be inhabited by small families as well as compound families.

Preferred raw materials connected with flint craft are different. Local erratic flint dominates during the LBK I. Jurassic Cracovian flint was used as a basic raw material in the phases of the LBK II-III in the western part. The role of local Baltic raw materials is generally minor. Good-quality erratic flints dominate in the eastern part. In small numbers, they are accompanied by the findings of Jurassic flint and chocolate flint as well as obsidian. Local Baltic raw materials begin to dominate again in the LBK I. During the LBK II, imported raw materials reappear in larger quantities. Flint craft was preponderantly aimed at the production of semiraw material in the form of chips. Various types of cores were applied. Most frequently it was schist. A peeling technique was commonly employed due to the shortage of raw material.

Consecutive small lineages were composed of larger lineages inhabiting the group of settlements that were used at different times, consisting of settlement microregions. Rotation within microregions did not have to include whole groups of villages that were consecutively occupying stage points and most probably embraced small lineages. These microregions established the higher levels of social organisation. In the area of Silesia, one can distinguish at least two subregions (Racibórz subregion and Wrocław subregion), characterised by the internal dynamics of changes and, e.g., stylistics of ceramic manufacture indicating the existence of diversified information exchange systems. Early-agricultural Silesian communities were functioning in the extensive system of the exchange and distribution of goods. It included local, Silesian groups as well as the groups of farmers inhabiting the Upper Vistula river basin and the area of the Polish Lowland. It is clearly visible, e.g., by the presence of stone products from the Lower Silesian raw materials in the Cracow region or Cracovian flint in the area of Silesia. The participation of hunting-gathering societies inhabiting the right bank of the Odra in the exchange network should be explained here. The contacts of hunters and gatherers are proven by numerous potteries and stone products discovered in the areas they inhabited. At the beginning, those contacts were of a cooperative character. It was not sooner than in the period of the functioning of postlinear groups that those contacts assumed a competitive character.

LBK and SBK societies were extensively using rock as raw material available in the area of Sudety, where was the base of raw material, which was located within the area or in the proximity of their settlement zone. In the production of stone tools, it is easy to identify the selection of types of lithic raw material conditioned by the intended application of a future tool. There is the diversification of raw materials used for the production of tools with a separated blade and the production of other types. Hoe tools were made of silica shale, silica phyllite and amphibolite; beaters and grinders were made of meta-greywacke and granite as well as sandstones and erratic raw materials. A basic social unit in these societies was a group of people inhabiting a house, representing small lineages. The size of

3

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

4

V. Janak: Le Néolithique et l’Énéolithique de la Silésie tchèque

LE NÉOLITHIQUE ET L´ÉNÉOLITHIQUE DE LA SILÉSIE TCHÈQUE Vratislav JANÁK

Résumé : La Silésie tchèque, il l´est possible de diviser en trois zones d´après des conditions naturelles pour le peuplement préhistorique – terres fertiles sur les sols loessiques ( le noyau du peuplement); les territoires d´un moindre fertilité (le peuplement rare); les montagnes (ont été seulement pénétrées). Les facteurs décisifs pour la recherche de l´histoire du peuplement, ce sont l´intégrité principielle de la partie tchèque et da la partie polonaise de la Haute Silésie et l´importance des voies de communication, avant tout en partie orientale de cette région (la Porte de Moravie). La recherche actuelle connaît des centaines de localités, mais pour la plupart, il s´agit de trouvailles de la seule industrie lithique; on ne connaît que des dizaines de sites et les autres catégories de monuments (les tombes, les dépôts etc.) ne sont, pour le moment, représentées que sporadiquement. La séquence des cultures archéologiques est semblable comme ailleurs en Europe centrale, il y a cependant des spécificités locales (on manque des certaines cultures jusqu´à présent) Abstract: The Czech Silesia can be divided into three settlement zones according to natural conditions suitable for the primeval settlements – fertile soil of the Silesian lowland (core settlement); less fertile lowland and sub-mountainous regions (rare settlement; mountains only penetrated). The essential integrity of the Czech and Polish upper Silesia and the communication importance of the region (especially its eastern part – the Moravian Gate) represent an indispensable factor in the historical research of settlements. Recent research of the Neolithic and Eneolithic ages identified several hundreds of locations which are characteristic of stone tools in tens of settlements. Other types of memorials (graves, depots, etc.) are rare. In terms of the cultural system there is analogy with the neighbouring areas of central Europe, yet with local specific features (some cultures are missing up to the present day).

LE NÉOLITHIQUE ET L´ÉNÉOLITHIQUE DE LA SILÉSIE TCHÈQUE

peuplement traditionnel, les alentours les plus proches d´Opava et de Krnov et une bande des terres fertiles entre les deux villes, dans la Porte de Moravie encore les environs de Nový Jièín; le reste de la région est une terre inconnue pour ainsi dire.

Sur le plan de l’archéologie, la Silésie tchèque est représentée par sept districts actuels dans le Nord-Est de la République tchèque, d’une étendue totale de 6274 km2. S’agissant des conditions naturelles du peuplement, on peut distinguer trois sphères principales :

2. Concernant la connaissance des étapes chronologiques ou des différentes cultures – nous disposons d´un fonds de trouvailles relativement vaste pour certaines périodes ou cultures, pour certaines autres nous n´avons presque rien (surtout pour l´Énéolithique récent et tardif).

1. Une zone des terres fertiles sur les loess de la Haute Silésie, avec un peuplement continu sur la base agricole, dès le néolithique jusqu’aujourd’hui.

S´agissant du traitement des découvertes, on peut constater la même chose. Nous manquons des connaissances générales basées sur une analyse détaillée du matériel, même les derniers essais datant du tournant des années soixante et soixante-dix du siècle précédent s´appuient avant tout sur la littérature et s´en tiennent à l´étude superficielle des problèmes; ils sont d´ailleurs dépassés de nos jours. Parmi les ouvrages partiels, il existe un vieux inventaire des localités avec la culture de la céramique rubanée qui date de la fin des années soixante, puis un nouvel inventaire des découvertes des haches-marteaux appartenant à la culture de la céramique cordée (dans le cadre du catalogue général des découvertes de la culture de la céramique cordée en Moravie et Silésie); ce n´est que la problématique du groupe lengyelien de la Haute Silésie qui a été traitée d´une manière complète et moderne. Le traitement des phénomènes supra-culturels est limité aux lests de pierre des filets de pêche. Il existe naturellement toute une série des autres articles importants, éventuellement des comptes-rendus, mais c´est le manque du système et le caractère décousu qui est typique pour l´état actuel de la connaissance.

2. Une zone plus ou moins boisée, avec des terres d’une moindre qualité, pour la plupart de l’origine glaciaire, notamment la Porte de Moravie et le pied des Beskides; à partir du Néolithique jusqu’à la colonisation médiévale, le peuplement dans cette zone fut enclavé, discontinu et en principe non agricole – c’étaient surtout le cours des voies de communication à longue distance et les sources des matières premières (le silex, le minerai de fer) qui jouaient un rôle décisif dans le processus de peuplement. 3. Les montagnes qui n’étaient plutôt que pénétrées. La région entière est caractérisée par son unité intégrale avec la partie polonaise de la Haute Silésie et par la possibilité de passage ; le tracé principal de la communication à grande distance menait par la Porte de Moravie. L’état de recherche est peu satisfaisant en général, c´est donc aussi le cas de l’état de recherche du Néolithique et de l’Énéolithique. Les recherches précédentes manquent avant tout d´équilibre : 1. Géographiquement – on peut considérer comme relativement bien exploités, sur le territoire avec un 5

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Du point de vue des découvertes prédominent des centaines de localités avec des trouvailles de la seule industrie lithique, mais on ne connaît que des dizaines de sites et les autres catégories de monuments, tels que les tombes, les dépôts etc. ne sont, pour le moment, représentées que sporadiquement. Les habitats sont connus quasi exclusivement dans les domaines des terres fertiles, dans la deuxième sphère on n´a repéré avec certitude que six villages. Par contre, c´est là que prédominent les localités avec une concentration de la seule industrie lithique, en silex pour la plupart, dont on connaît aussi des dizaines. Ce phénomène n´a pas été traité jusqu´ici. Dans certains cas, il s´agit probablement des ateliers, dans d´autres cas des habitats ou camps qui n´ont pas été discernés; dans la deuxième sphère spécialement peuvent être pris en considération les habitats sans céramique du soi-disant Néolithique forestier. L´image sociale et spirituelle de la société ne différait probablement pas en principe de celle du milieu environnant. Pourrait être considérée comme une spécificité économique la part énorme de la pêche indiquée par une présence massive des lests de pierre des filets qui est d´une grande rareté en Europe centrale.

pratiquement pas de la céramique lengyelienne des régions environnantes. Nous connaissons de nouveau une série de sites du domaine des terres fertiles (dans la Porte de Moravie trois sites seulement et nous supposons que le peuplement y a commencé assez tard). On ne peut qu´imaginer la forme des maisons, ainsi que celle des tombes, sur la base des découvertes des longues maisons rectangulaires légèrement enterrées et des sépultures accroupies (à la fin de l´évolution également des crémations) dans la partie polonaise du territoire traditionnel. La fin de l´évolution lengyelienne n´est pas représentée dans la région par la culture de Jordanów, comme on prétend parfois, il n´y a que des manifestations des influences jordanowiennes tant de la Moravie que de la Basse Silésie, mais aussi les influences du groupe de Ludanice, de la culture de Bodrogkeresztúr et de la culture des gobelets en entonnoir. Toutes les cultures du Néolithique se caractérisent par l´industrie en silex semblable, produite pour la plupart des matières premières du pays; parmi les matières importées on découvre surtout le silex corné de la Petite Pologne, moins fréquemment l´obsidienne, exceptionnellement la radiolarite. L´avènement de la culture des gobelets en entonnoir signifie le changement principal de forme de la culture matérielle et, aussi, le changement d´orientation géographique de la région qui ne s´adresse plus maintenant dans la vallée de Danube, mais vers le Nord. Nous connaissons encore nombre de sites, mais seulement dans la première sphère – dans la deuxième sphère on ne connaît avec certitude aucun site plus récent que le lengyel tardif – parmi lesquels aussi des enceintes en hauteur fortifiées. On connaît également le plan d´une maison trapézoïdale à trois lignes, construite en poteaux de bois. L´industrie lithique présente comme nouveauté les soi-disant haches de combat, les haches en silex corné et les lames énormément longues, produites, pour la plupart, des matières importées du Nord ou Nord-Est. Est connu aussi un dépôt unique en métal – une double spirale en cuivre et un disque en argent de type Stollhof, les deux objets anormalement grands. Malgré tous les changements radicaux, il faut pourtant compter sur une continuité au moins partielle de la population. Sauf les phénomènes tels que, par exemple, la maison trapézoïdale citée, construite entièrement dans les traditions anciennes du Néolithique, ce fait est indiqué surtout par la présence énorme, mentionnée ci-dessus, des lests de pierre des filets de pêche, à partir de la culture de la céramique rubanée jusqu´à la fin de la culture des gobelets en entonnoir. Pendant le passage à la phase tardive de la culture des gobelets en entonnoir, on constate dans la région les découvertes sporadiques de la céramique du type Retz, décorée au pointillé sillonné profondément, et les influences importantes de la culture de Baden, mais celle-ci n´est pas présente ici dans tout le sens du mot.

La séquence des cultures archéologiques est semblable comme ailleurs en Europe centrale, il y a cependant des spécificités locales. L´époque est ouverte par la culture de la céramique rubanée. Nous connaissons nombre d´habitats sur le territoire fertile et les habitats sporadiques dans la Porte de Moravie, nous ne connaissons cependant pas les plans des maisons et même pas les tombes. Toutes les phases sont représentées sur le territoire traditionnel d´habitation, dans la Porte de Moravie on ne rencontre que les phases récentes. Dans la phase finale de l´évolution on ne trouve pas la présence du groupe de Želiezovce et peut-être même pas de celui de Šárka, seulement leur influence sur la céramique la plus récente qui est marquée par la note de musique. La situation suivante n´est pas claire. Sur la frontière occidentale de la partie polonaise de la Haute Silésie on trouve des localités sporadiques de la culture de la céramique pointillée qui ne sont pas connues de la Silésie tchèque. On ne peut pas exclure un hiatus ou au moins une diminution massive de l´habitation, correspondant au degré ancien et partiellement aussi au degré récent de la culture de la céramique pointillée. La céramique au décor pointillé n´apparaît chez nous qu´en tant que partie de la plus vieille phase du groupe lengyelien de Haute Silésie. Celui-ci correspond chronologiquement au passage entre les deux degrés lengyeliens, l´ancien et le récent, et il est très proche aux groupes se trouvant plus loin au Nord et à l´Est, notamment au groupe de Malice. Le peuplement des vallées d´Oder et Vistule peut être compris à l´époque dans le cadre du complexe des groupes à la céramique mixte. Nous connaissons nombre d´habitats de la plus vieille phase, mais seulement sur le territoire traditionnel d´habitation; la forme des maisons est inconnue. On a aussi découvert deux tombes (des squelettes accroupis avec un mobilier pauvre – trois vases dans chaque tombe) sur le site. Ce type d´inhumation se fonde évidemment sur la tradition de la céramique occidentale rubanée et pointillée, on peut en trouver des analogies de la même époque surtout dans la région de Cracovie. La céramique des phases suivantes ne diffère

La culture des gobelets en entonnoir fut relayée, dans la région, par la culture des amphores sphériques dont on ne connaît qu´une petite quantité de trouvailles de la céramique. Suivit la culture de la céramique cordée dont les preuves se limitent cependant aux trouvailles exclusives des haches de combat typiques, à savoir du plus ancien type « A » jusqu´aux plus récentes. La culture des gobelets campaniformes est représentée, sauf une seule découverte d´une fosse très pauvre, par des trouvailles de la céramique sans contexte, dans deux 6

V. Janak: Le Néolithique et l’Énéolithique de la Silésie tchèque

cas probablement des tombes détruites, pour la plupart il ne s´agit toutefois que des fragments dans le placement secondaire. La séquence aurait dû être terminée par le groupe de Chlopice-Veselé; ce dernier est connu du nombre de sites juste derrière la frontière polonaise, pourtant nous en manquons jusqu´à présent de notre côté.

JANÁK, V., 1998, Přelom i pokračování (archeologie v českém Slezsku v 80. a první polovině 90. let ). Archeologické rozhledy 50, p. 455 – 464. JANÁK, V., 1998, Lengyelská keramika z Kotouče u Štramberka a počátky eneolitu v oderské části Moravské brány. Pravěk NŘ 7/ 1997, p. 105 – 127. JANÁK, V., 1999, Předběžné poznámky k neolitu a eneolitu českého Slezska. In : Otázky neolitu a eneolitu našich krajín – 1998, edited by I. Kuzma, Nitra: Archeologický ústav SAV Nitra, p. 95 – 109.

A partir de la culture des amphores sphériques, les sources commencent à faire défaut quasi total même dans le domaine des terres fertiles, on ne connaît rien du tout de la deuxième sphère, l´exception faite à l´industrie lithique. On ne peut se faire une idée sur l´état des choses en Silésie tchèque que sur la base de la situation mieux documentée de la partie polonaise du territoire avec un peuplement traditionnel.

JANÁK, V., 2000, Prehistorické kamenné artefakty v bruntálském muzeu a počátky osídlení severozápadní části Nízkého Jeseníku. In : Sborník bruntálského muzea 2000, edited by T.Niesner, P. Vojtal. Bruntál: Muzeum v Bruntále, p. 5-14. JANÁK, V., 2001, Hroby hornoslezské lengyelské skupiny z Velkých Hoštic a komplex skupin se smíšenou keramickou náplní (Příspěvek k poznání postlineárního osídlení severní části střední Evropy). Pravěk – suplementum 8, p. 325 – 351.

ADRESSE DE L’AUTEUR

JANÁK, V., KOUŘIL, P., 1991, Problémy a úkoly archeologie v českém Slezsku a na severovýchodní Moravě. Časopis Slezského zemského muzea, series B 40, p. 193 – 219.

Vratislav JANÁK Slezská univerzita v Opavě, Archeologický seminář ÚHM Masarykova 37 746 01 Opava RÉPUBLIQUE TCHÈQUE

JANÁK, V., KOUŘIL, P., 2001, Archeologie Pobeskydí (k nové polské práci o Těšínsku). Archeologické rozhledy 53, 2001, p. 372 – 386. JANÁK, V., PAVELČÍK, J., 1994, Die Lengyel -Kultur in Nordund Ostmähren. In : Internationales Symposium über die Lengyel - Kultur 1888 - 1988 (Znojmo - Kravsko - Těšetice),edited by P. Koštuřík. Brno – Lodź: Masarykova univerzita v Brně, p. 78 – 84.

BIBLIOGRAPHIE BÖHM, J., 1926, Slezsko v době předhistorické. In : Vlastivědný sborník slezský 2, edited by A. Lhotský. Opava : Ústřední spolek českého učitelstva ve Slezsku, p. 15 – 31.

JISL, L., 1952, Nálezy vypíchané keramiky ve Slezsku. Časopis Slezského muzea, series B 2, p. 1 – 5.

HETFLEISCH, E., 1931, Fundübersicht des nordwestlichen Schlesiens. In : Altvater, Festschrift zum 50-Jahrfeier des Sudetengebirgesvereines. Freiwaldau : Der Hauptvorstand des Sudetengebirgesvereines. Freiwaldau, p. 268 – 272.

JISL, L., 1955, Záchranný výzkum eneolitických jam z Holasovic a Opavy roku 1953. Časopis Slezského muzea, series B 4, p. 34 – 43.

HOERNES, M., 1901, Neolithische Wohnstätten bei Troppau, Mitteilungen der Prähistorischen Commission der kaiserlichen Akademie der Wissenschaften 1, p. 401 – 411.

JISL, L.1964 : Počátky archeologického bádání ve Slezsku a opavská muzea. In : 150 let Slezského Muzea. Ostrava: Krajské nakladatelství v Ostravě, p. 121 – 129.

JANÁK, V., 1990, Východoevropské elementy v lengyelské antropomorfní plastice na Moravě a ve Slezsku. In : Pravěké a slovanské osídlení Moravy (Sborník k 80. narozeninám J. Poulíka), edited by V. Nekuda, J. Unger, & M. Čižmář. Brno: Muzejní a vlastivědná společnost v Brně & Archeologický ústav ČSAV v Brně, p. 26 – 37.

JISL, L., 1968, Slezsko a Ostravsko v pravěku a rané době dějinné. In : Ostravsko do roku 1848, edited by A. Grobelný. Ostrava : Profil, p. 10 – 29. JISL, L., 1969, Sídliště volutové kultury na Ostravsku. Časopis Slezského muzea, series B 18, p. 3 – 8.

JANÁK, V., 1991, Severovýchodní vlivy v Horním Slezsku v časném a starém eneolitu. Časopis Slezského zemského muzea, series B 40, p. 97 – 109.

JISL, L., 1969, Kamenné nástroje a zbraně z území bývalého okresu Bílovec. Časopis Slezského muzea, series B 18, p. 97 – 106. KACZANOWSKA, M., KAMIEŃSKA, J., & KOZŁOWSKI, J.K., 1986, Kontakte zwischen der Lengyel-Kultur und Kultur mit der Stichbandkeramik in Südpolen. In : Internationales Symposium über die Lengyel-Kultur, edited by B. Chropovský, H. Friesinger. Nitra-Wien: Archeologisches Institut der Slowakischen Akademie der Wissenschaften in Nitra & Institut für Ur- und Frühgeschichte der Universität Wien, p. 95 – 120.

JANÁK, V., 1993, Die Lengyel - Kultur in Oberschlesien. In : Actes du XIIe Congrès International des Sciences Préhistoriques et Protohistoriques 2, edited by J. Pavúk. Bratislava: Institut archéologique de l¨Acadmie Slovaque des Sciences, Nitra, p. 553 – 556. JANÁK, V., 1994, Sídliště lidu s moravskou malovanou keramikou ve Vyškově - dědické cihelně a počátky mladšího lengyelského stupně na Moravě a ve Slezsku. Pravěk NŘ 2/1992, p. 105 – 130.

KARGER, V., 1922, Die Vorgeschichtsforschung in Schlesien und die Ausstellung vor- und frühgeschichtlicher Altertümer im Schlesischen Landesmuseum 1921 - 22, Anzeiger des Schlesischen Landesmuseums in Troppau 1, p. 2 – 32.

JANÁK, V., 1995, Sídliště lidu s moravskou malovanou keramikou v Příboře - Hájově, okr. Nový Jičín a závěr lengyelského vývoje na Moravě. Acta historica et museologica Universitatis Silesianae Opaviensis l/1995, p. 5 – 19. JANÁK, V., 1996, Kdy „Pravěk Slezska“ ? (Ad marginem „Pravěkých dějin Moravy“). Časopis Slezského zemského muzea, series B 43, p. 193 – 210.

KOŠTUŘÍK, P., 1996, Vlivy malické skupiny v českém Slezsku a na Moravě. In : Kultura malicka. Drugi etap adaptacji naddunajskich wzorców kulturowych w neolicie pólnocnej części środkowej Europy, edited by J. K. Kozłowski. Kraków : Polska Akademia Umiejętności, p. 111 - 117.

JANÁK, V., 1997, Keramická hrnčířská hladítka v lengyelu Horního Slezska,. Acta historica et museologica Universitatis Silesianae Opaviensis 3/1997, p. 11 – 19.

KOUŘIL, P., PAVELČÍK, J., 1989, Pravěké a časně historické osady z Opavy - Palhance I. Časopis Slezského muzea, series B 38, p. 133 – 147. 7

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe KOZŁOWSKI, J.K., 1996, Kultura malicka – The Malice Culture. In : Kultura malicka. Drugi etap adaptacji naddunajskich wzorców kulturowych w neolicie pólnocnej części środkowej Europy, edited by J. K. Kozłowski. Kraków : Polska Akademia Umiejętności, p. 153 – 164.

SCHIRMEISEN, K., 1934, Jungsteinzeitliche Siedlungen auf dem Burgberg bei Jägerndorf. Sudeta 10, p. 68 – 77.

MAŠKA, K. J., 1898, Z pravěku Kravařska. In : Moravské Kravařsko, edited by J. Mňuk, J. Kranich, & A. Ugwitz. Příbor : České učitelstvo okresu Novojického, p. 39 – 49.

SCHIRMEISEN, K., 1943, Die Vorzeit des Mähr. - Ostrauer Raumes. Mähr. Ostrau : Der Oberbürgermeister der Stadt Mährisch-Ostrau.

OPRAVIL, E., 1974, Moravskoslezský pomezní les do začátku kolonizace. In : Archeologický sborník (Ostravské muzeum), edited by L. Wiegandová. Ostrava: Profil, p. 113 – 133.

STUMPF, G., 1927, Die Vor- und Frühgeschichte Kuhländchens. In : Festschrift zum l. Kuhländler Heimatfest in Neu - Titschein, August 1927, edited by J. Ullrich. Neu – Titschein : L. D. Enders´sche, p. 7 – 24.

SCHIRMEISEN, K., 1935, Jungsteinzeitliche Siedlungen auf dem Burgberg bei Jägerndorf. Jägerndorfer Ländchen 9, p. 23 - 31, 41 - 43, 49 – 55.

PAVELČÍK, J., 1958, Několik poznámek k poznání kanelované keramiky na Moravě a ve Slezsku. Časopis Slezského muzea, series B 7, p. 89 – 109.

STABRAVA, P., 2000, Archeologický výzkum ve Vávrovicích, okres Opava, Badania archeologiczne na Górnym Ślasku i ziemiach pogranicznych w 1996 roku - Archeologické výzkumy v Horním Slezsku a na jeho pomezí v roce 1996, edited by E. Tomczak, V. Janák. Katowice: Centrum Dziedzictwa Kulturowego Górnego Śląska, p. 20 - 23.

PAVELČÍK, J., 1963, Eneolitické a starobronzové sídliště u Brumovic (okres Opava, do r. 1960 Krnov). Časopis Slezského muzea, series B 12, p. 1 – 17. PAVELČÍK, J., 1977, J. Bukowska - Gedigowa, Kultura pucharów lejkowatych w dorzeczu górnej Odry, Przeglad Archeologiczny 23, 1975, 83 – 186. Památky archeologické 68, p.222 – 224.

ŠEBELA L. 1999: The Corded Ware Culture in Moravia and in the Adjacent Part of Silesia (Catalogue), Brno: Archeologický ústav AV ČR Brno.

PAVELČÍK, J., 1974, K problematice neolitu a eneolitu českého Slezska. In : Archeologický sborník (Ostravské muzeum), edited by L. Wiegandová. Ostrava: Profil, p. 22 – 42.

ŠIKULOVÁ, V., 1961, K otázce rybolovu v mladší době kamenné. Časopis Slezského muzea, series B 10, p. 1 – 18.

PAVELČÍK, J., 1979, Problem zasiedlania Moraw północnych przez ludność ceramiki wstęgowej. In : Początki neolityzacji Polski południowo – zachodniej, edited by W. Wojciechowski, B. Gediga, & L. Leciejewicz. Wrocław : Polska Akademia Nauk – Oddział we Wroc ławiu, p. 93 – 103.

ŠIKULOVÁ, V.,1971, Záchranné výzkumy v pískovnách ve Vávrovicích - Palhanci ( okr. Opava ). Přehled výzkumů AÚ ČSAV Brno za r. 1970, edited by A. Medunová, J. Meduna, & J. Říhovský. Brno : Archeologický ústav ČSAV, p. 75 - 76

PAVELČÍK, J., 1994, Osady na Kostelním kopcu v Opavě - Jaktaři i ich miejsce w badaniach nad problematyka neolitu i eneolitu Górnego Ślaska. Przegląd Archeologiczny 42, p. 5 – 34.

ŠIKULOVÁ, V. 1971, Záchranné výzkumy na přestavbě státní silnice Opava - Krnov ( okr. Opava ). In : Přehled výzkumů AÚ ČSAV Brno za r. 1970, edited by A. Medunová, J. Meduna, & J. Říhovský . Brno : Archeologický ústav ČSAV, p. 76 – 77.

PAVELČÍK, J., JANÁK, V., 1989, Neueste Ergebnisse der Erforschung des Äneolithikums in Nordmähren. Praehistorica 15, p. 67 – 70.

TICHÝ, R. 1969, Území Ostravy a okolí v mladší době kamenné. Ostrava ( sborník příspěvků k dějinám a výstavbě města ) 5, p. 29 – 34.

PAVÚK, J., 1996, Poznámky ke vzťahu malickej a lengyelskej kultúry. In : Kultura malicka. Drugi etap adaptacji naddunajskich wzorców kulturowych w neolicie pólnocnej części środkowej Europy, edited by J. K. Kozłowski, Kraków: Polska Akademia Umiejętności , p. 119 - 135.

VENCL, S., 1964, K otázce datování temene vrchu Kotouče u Štramberku. Časopis Moravského musea - vědy společenské 49, p. 49 – 60. WEIGEL, S., 1919, Die prähistorischen Wohnstätten bei Blattendorf. Das Deutsche Kuhländchen l, p. 18 - 20.

PODBORSKÝ, V. et al., 1993, Pravěké dějiny Moravy. Brno : Muzejní a vlastivědná společnost v Brně.

8

A. Dufraisse: Économie du combustible au Néolithique final : Exemple de la station 19 du lac de Chalain (Jura, France)

ECONOMIE DU COMBUSTIBLE AU NÉOLITHIQUE FINAL : EXEMPLE DE LA STATION 19 DU LAC DE CHALAIN (JURA, FRANCE) Alexa DUFRAISSE

Résume : L’étude anthracologique du site de Chalain 19 (Jura, France) a permis de mettre en évidence une certaine compatibilité entre l’approche paléoécologique et celle des stratégies d’approvisionnement en combustibles. Le spectre anthracologique donne en effet une image très proche de la végétation locale exploitée. Il révèle une transition frênaie-hêtraie qui pourrait intégrer une cyclicité naturelle et/ou un changement d’aire d’exploitation, motivé par un épuisement progressif des peuplements de frênes et probablement une volonté de partage de l’espace forestier afin de gérer l’ensemble des activités. Abstract: Charcoal analysis of Chalain 19 (Jura, France) emphasizes the compatibility between palaeological approaches and woody combustible acquisition strategies. Results, palaecologically representative, show the local exploited vegetation and reveal a transition ash/ beech. This dynamic could integrate a natural cyclicity and/or a change of exploitation area motivated by a progressive exhaustion of ash and probably by a division of the forest space to manage all activities.

INTRODUCTION Basée sur l’analyse de charbons de bois, l’objectif de cette étude est d’apporter quelques éléments à la compréhension des modes de gestion de l’espace forestier pendant le Néolithique. Elle s’inscrit dans le cadre d’une étude qui consiste en une approche archéologique et écologique de la végétation autour des lacs de Chalain et Clairvaux du Néolithique moyen au Néolithique final. En général, les charbons de bois archéologiques issus d’un contexte domestique, sont, dans certaines conditions de dépôts et d’échantillonnage, considérés comme le fruit d’une collecte aléatoire en bois de feu et sont dans ce cas interprétés en terme de paléoécologie (Chabal, 1992, 1997). Mais dans l’hypothèse d’une collecte modulée par des motivations ou des contraintes particulières, le spectre anthracologique ne pourrait-il pas aussi refléter des comportements d’acquisition en bois de feu tant du point de vue du comportement de collecte au sein d’une aire (sélection ou protection de certaines essences) que des aires d’approvisionnement (changement ou extension des aires d’exploitation)? C’est en suivant cette hypothèse de travail que nous aborderons l’étude de la station 19 du lac de Chalain dont l’occupation est datée du Néolithique final entre 3100 et 2900 av. J.-C. Figure 1. Localisation du lac de Chalain

LE CADRE D’ÉTUDE Contexte environnemental : relief, climat et végétation

avec des températures moyennes annuelles de 8 à 9°C et une pluviométrie de 1200 à 1500 mm/ an.

Le lac de Chalain est situé à environ 500 m d’altitude dans la Combe d’Ain, sur le premier plateau du Jura (Fig. 1). Encadré à l’ouest par les plaines et à l’est par le deuxième plateau du Jura et les Hautes-Chaînes, le climat y est semi-continental

Le relief du massif jurassien est à l’origine d’un étagement de végétation (Beaufils, 1983, 1993). Le lac de Chalain se place ainsi au niveau de l’étage collinéen supérieur (Fig. 2), la série submontagnarde du hêtre (Fagus sylvatica), des 9

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 2. Etagement de la végétation dans le massif jurassien

chênes (Quercus sessiliflora et Quercus pedunculata) et du charme (Carpinus betulus) domine. Sont également représentées les « hêtraies chaudes » sur les versants exposés au sud. Ces formations sont accompagnées du frêne (Fraxinus excelsior), de l’érable champêtre et sycomore (Acer campestre et Acer pseudoplatanus), du tilleul à grandes feuilles (Tilia platyphyllos), de l’alisier blanc (Sorbus aria), de l’alisier torminal (Sorbus torminalis) et du sapin (Abies alba). A proximité des rives du lac, on notera la présence de forêts ripicoles caractéristiques des zones très humides, telles que les saulaies et les aulnaies-frênaies.

séquence stratigraphique comporte trois niveaux occupations successifs séparés par des craies stériles. Le niveau le plus ancien (couche O) est attribué à la culture du Horgen (3200-3100 av. J.-C.) et se caractérise par une couche d’incendie présent sur toute la surface du village. La couche K est attribuée au groupe du Clairvaux ancien (30002910 av. J.-C.), suivie de la couche H légèrement plus tardive. En raison des processus de sédimentation, ces deux derniers niveaux sont parfois « compactés ». L’espace villageois est desservi par un chemin de planches d’une centaine de mètres de longueur qui semble avoir fait l’objet de multiples réfections. Une palissade demi-circulaire de 75 m de longueur protège le site du côté de la terre ferme. A l’entrée du village une zone densément bâtie est constituée de plusieurs maisons. A l’opposé se situe un espace libre avec une seule construction, plus grande et d’architecture différente, et dont la fonction semble sortir du cadre domestique strict. Quant à la végétation environnante, elle est caractérisée par la prédominance du hêtre alors que la chênaie mixte est en constante régression ; seul le frêne réussit à se maintenir (Richard, 1997).

En aval du lac, dans les plaines, les séries de la chênaiehêtraie-charmaie dominent. Elles sont accompagnées des érables champêtre et sycomore, du frêne, du prunellier (Prunus avium), du noisetier (Corylus avellana). La série septentrionale du chêne pubescent (Quercus pubescens) apparaît sur les pentes les plus ensoleillées, parfois accompagnées du buis (Buxus sempervirens) et du prunier de Ste-Lucie (Prunus mahaleb). En amont, sur le deuxième plateau du Jura et les HautesChaînes, les chênes et les charmes sont absents. Le hêtre et le sapin dominent, accompagnés de l’érable sycomore, de l’orme des montagnes (Ulmus montana) et à plus haute altitude de l’épicéa (Picea abies) et du pin à crochet (Pinus uncinata). La strate arbustive est pauvre ; on note la présence de l’alisier blanc, du nerprun des Alpes (Rhamnus alpina), de l’amélanchier (Amelanchier ovalis).

MATÉRIEL ET MÉTHODE En contexte terrestre, les charbons sont représentés par des dépôts concentrés (foyers) ou des dépôts dispersés dans les niveaux d’occupation. De nombreuses études ont montré que ces différents dépôts n’avait pas la même signification au regard de la durée d’activité (les charbons contenus dans les foyers ne représentant que les dernières combustions ne sont pas représentatifs de la durée d’occupation), et donc de la

Contexte archéologique La station 19 est située à l’extrémité d’une presque-île peu marquée sur la rive occidentale du lac de Chalain. La 10

A. Dufraisse: Économie du combustible au Néolithique final : Exemple de la station 19 du lac de Chalain (Jura, France)

Figure 3. Plan de la station 19 du lac de Chalain

signification paléoécologique (Chabal, 1992). La compréhension du mode de dépôts des charbons dont dépend le protocole d’échantillonnage est donc une étape primordiale. C’est pourquoi une analyse spatiale des charbons de la station 19 a tout d’abord été réalisée. Elle a permis de montrer que les modes de dispersion des charbons sont très différents en milieu lacustre de celle en milieu terrestre (Dufraisse, à paraître). Ils sont en effet concentrés sous forme de cordons de flottage, parallèles au rivage et correspondent à l’accumulation de débris organiques liée aux battements de l’eau du lac. Malgré cette répartition sous forme « concentrée », ces dépôts sont le résultat d’un brassage de débris et sont représentatifs de la durée d’occupation. Les prélèvements ont donc été effectués selon ces horizons d’accumulations dans la zone du bâtiment isolé (Fig. 3). En revanche, la fouille de la zone densément bâtie étant terminée lors des prélèvements, ceux-ci ont été effectués par une série de forages à la tarière manuelle. Un premier transect, perpendiculaire à la rive du lac, a été réalisé de manière à repérer les cordons de flottage correspondant à chaque couche archéologique. Puis, les prélèvements ont été effectués le long de ces cordons suivant les courbes de niveaux. Les sédiments ont ensuite été tamisés à la maille de 2 mm et la détermination des charbons a été réalisée au microscope optique à réflexion. Le nombre de charbons nécessaire et suffisant a été estimé par l’établissement de courbes de saturation et jusqu’ à stabilisation des fréquences.

Horgen et Clairvaux ancien, révèle une dynamique de végétation reposant sur quatre taxons principaux : le hêtre (Fagus sylvatica), le frêne (Fraxinus excelsior), le noisetier (Corylus avellana) et le chêne à feuillage caduc (Quercus f.c.).

RÉSULTATS

Les fréquences des essences dans la couche H/K sont différentes : le frêne et le hêtre sont des proportions équivalentes (respectivement 27.1% et 23.4%) alors que le chêne atteint 17% ce qui n’est le cas dans aucun autre échantillon. Il ne s’agit pas là d’un défaut de fiabilité

Ainsi, durant le Horgen (couche O), le frêne domine le cortège floristique avec 68.8%. Il est accompagné du hêtre (6.9%), du noisetier (5.2%) et du chêne à feuillage caduc (3.5%). L’érable, le tilleul, l’orme et quelques rosacées sont aussi représentées. Notons enfin des essences de ripisylve telles que le saule et le peuplier. Ces résultats suggèrent l’exploitation d’un milieu plutôt humide (essences de ripisylve) et ouvert (présence de taxons héliophiles) dont l’équivalence actuelle pourrait se retrouver dans les végétations de lisières à proximité des cours d’eau et des lacs. Dans les couches H et K, le frêne est bien moins représenté (20 et 25%) alors que les proportions de hêtre augmente (35 et 45%). Les essences secondaires (noisetier, chênes, érables, tilleuls, ormes et rosacée) sont dans des proportions équivalentes à celles du Horgen. Les essences de ripisylve telles que le saule, le peuplier et probablement le frêne se maintiennent également (le frêne appartient à plusieurs formations végétales telles que la ripisylve, la chênaie et la hêtraie). Ces résultats indiquent l’exploitation de formations végétales telles qu’on les trouve actuellement dans les étages collinéen supérieur et montagnard inférieur (Fig. 2).

L’étude de la zone densément bâtie repose sur la détermination d’environ 2000 charbons (Tabl. 1, Fig. 4). L’analyse anthracologique des deux phases d’occupation, 11

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe Tableau 1. Dénombrements et pourcentages des charbons de bois de Chalain 19 Sites Aires architecturales Cultures Couches Essences /fréquences Abies alba Acer campestre Acer platanoides/pseudoplatanus Acer sp. Betula sp. Carpinus betulus Cornus sanguinea Corylus avellana Fabaceae Fagus sylvatica Angiospermae Frangula sp. Fraxinus excelsior Hippophae rhamnoides Ilex aquifolium Larix /Picea Monocotyledonae Rosaceae Pomoideae Sorbus sp. Sorbus/Crataegus Pomoideae type Pirus Prunoideae Prunus type avium/padus Quercus f.c. Salix sp. Populus sp. Salix/Populus Sambucus sp. Taxus baccata Tilia cordata Tilia platyphyllos Tilia sp. Ulmus minor Ulmus montana Ulmus sp. Viburnum sp. Total Non identifié Indéterminable

Chalain, station 19 Zone densément peuplée Horgen

Clairvaux ancien

O Nb – 4 6 2 – – – 15 – 20 9 – 198 – – – – – 2 – 1 – – – 10 1 – 4 – – 3 1 5 – 2 5 – 288 – 2

K % – 1,4% 2,1% 0,7% – – – 5,2% – 6,9% 3,1% – 68,8% – – – – – 0,7% – 0,4% – – – 3,5% 0,4% – 1,4% – – 1,0% 0,4% 1,7% – 0,7% 1,7% – 100%

Nb – 8 9 11 – – – 22 – 129 7 – 67 – – – – – – – – – – 1 12 5 1 1 – – – 2 9 1 1 2 2 290 – 4

H % – 2,8% 3,1% 3,8% – – – 7,6% – 44,5% 2,4% – 23,1% – – – – – – – – – – 0,3% 4,1% 1,7% 0,3% 0,3% – – – 0,7% 3,1% 0,3% 0,3% 0,7% 0,7% 100%

statistique puisque le nombre de charbons déterminés, environ 900, a permis une stabilisation des fréquences relatives. Malgré ces différences difficiles à expliquer, ces chiffres ont toutefois été pris en compte de manière à pouvoir établir des comparaisons avec les résultats du bâtiment isolé où les couches H et K se présentent également sous forme « compactées». Notons enfin la présence ponctuelle de quelques essences, jusqu’à maintenant absentes, telles que

Nb – 13 16 10 2 – – 45 – 161 10 – 101 – – – – – 3 – 1 – – – 20 31 5 9 – – 5 3 6 1 2 6 1 451 – 3

H/K % – 2,9% 3,6% 2,2% 0,4% – – 10,0% – 35,7% 2,2% – 22,4% – – – – – 0,7% – 0,2% – – – 4,4% 6,9% 1,1% 2,0% – – 1,1% 0,7% 1,3% 0,2% 0,4% 1,3% 0,2% 100%

Nb 11 23 24 19 9 1 1 77 12 209 27 – 242 – – – – – 8 – 1 2 – 1 153 38 7 12 – 2 3 – 5 – – 5 – 892 – 2

% 1,2% 2,6% 2,7% 2,1% 1,0% 0,1% 0,1% 8,6% 1,4% 23,4% 3,0% – 27,1% – – – – – 0,9% – 0,1% 0,2% – 0,1% 17,2% 4,3% 0,8% 1,4% – 0,2% 0,3% – 0,6% – – 0,6% – 100%

Bâtiment isolé Clairvaux ancien H/K Nb % 24 0,5% 147 3,1% 222 4,7% 29 0,6% 5 0,1% – – 15 0,3% 387 8,2% 1 0,02% 2680 56,5% 6 0,1% 2 0,04% 637 13,4% 2 0,04% 1 0,02% 1 0,02% 1 0,02% 1 0,02% 10 0,2% 4 0,08% 17 0,4% – – 1 0,02% 1 0,02% 232 4,9% 145 3,1% 32 0,7% 13 0,3% 1 0,02% – – 34 0,7% 11 0,2% 36 0,8% 16 0,3% 15 0,3% 12 0,3% 1 0,02% 4742 100% 18 168

le charme (Carpinus betulus), le cornouiller sanguin (Cornus sanguinea), ou encore l’if (Taxus baccata). L’étude du bâtiment isolé, réalisée uniquement sur le niveau du Clairvaux ancien (H/K), repose sur la détermination d’environ 5000 charbons (Tabl. 1). Le hêtre, le frêne, le noisetier et les chênes sont toujours les taxons principaux et leur ordre hiérarchique est identique. Toutefois, le hêtre est 12

A. Dufraisse: Économie du combustible au Néolithique final : Exemple de la station 19 du lac de Chalain (Jura, France)

mieux représenté (56.5%) alors que le frêne l’est moins (13.4%). Les essences secondaires sont également présentes dans des proportions équivalentes à celles observées dans la zone densément bâtie, mais la richesse taxonomique est plus importante avec des essences de ripisylve supplémentaires telles que la bourdaine (Frangula sp.) et le sureau (Sambucus sp.), des essences à affinités plus montagnardes telles que le houx (Ilex aquifolium). L’argousier (Hippophae rhamnoides), espèce appréciant les forêts ripicoles et les berges de ruisseau de montagne ou du littoral où il peut constituer des groupements pionniers, peut aussi être rattaché à l’étage montagnard. Notons aussi la présence de conifères tels que le sapin (Abies alba) et l’épicéa (Picea abies). Cette richesse taxonomique, plus grande que dans la zone densément bâtie, est certainement lié au nombre deux fois plus important de fragments étudiés. Dans l’ensemble, pour la maison isolée, le cortège floristique, riche et varié, semble traduire l’exploitation de formations forestières différentes qui trouvent leur équivalence dans les étages collinéen supérieur et montagnard inférieur/moyen actuel. Par conséquent, les spectres anthracologiques aussi bien dans la zone densément bâtie qu’au niveau du bâtiment isolé montrent une richesse taxonomique importante (plus d’une vingtaine de taxons) et une forte cohérence entre taxons dominants et taxons subordonnés par rapport à une végétation actuelle, ce qui plaide en faveur d’une bonne représentativité paléoenvironnementale. Les analyses polliniques (Richard, 1997) et les relevés phytosociologiques actuels effectués dans la Combe d’Ain (Beaufils, 1983) assurent également une bonne représentativité, excepté pour le chêne qui semble sousreprésenté dans le spectre anthracologique. Le spectre révèle l’exploitation d’une mosaïque de formations : celle de la ripisylve (composée du frêne, du saule, du peuplier), celle d’une formation « de lisière » à affinités collinéennes (composée de rosacées tel le sorbier ou l’aubépine, de frêne, de noisetier, des érables et de quelques chênes) et celle d’une végétation à affinités plus montagnarde (composée de hêtre, tilleul, orme, et frêne). INTERPRÉTATIONS L’ensemble du diagramme indique deux phases d’exploitation qui correspondent aux deux groupes culturels que sont le Horgen et le Clairvaux ancien. La première phase montre l’exploitation dominante du frêne, qui est une essence de lumière ou de demi-ombre, appréciant l’humidité des sols, à affinités collinéennes et montagnardes ; la deuxième est caractérisée par l’exploitation du hêtre, qui est une essence d’ombre, exigeant des précipitations annuelles et une humidité atmosphérique élevées, à affinités montagnardes mais pouvant également pousser à l’étage collinéen lorsque les conditions sont favorables. Dans l’hypothèse où le spectre anthracologique est représentatif d’un point de vue paléoécologique, différentes hypothèses peuvent être posées quant à la transition frêne/ hêtre.

Figure 4. Diagramme anthracologique de Chalain 19 13

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Elle peut tout d’abord refléter une dynamique de végétation spontanée. En effet, les forêts de hêtre d’Europe centrale présentent des cyclicités naturelles, caractérisées par une disparition d’une forêt dense au profit d’une végétation arbustive, suivie d’un stade à arbres pionniers qui est finalement remplacés par une nouvelle forêt dense (Remmert, 1987). L’écologie du frêne et du hêtre, détaillée ci-dessus, étaye cette hypothèse ; le frêne jouerait ainsi le rôle de pionnier secondaire. Quant aux analyses polliniques (Richard, 1995), elles révèlent des successions de baisses simultanées du hêtre et du chêne, les taux de pollen du hêtre étant les plus affectés. En revanche, les taux de frêne sont constants et interprétés comme un maintien de cette essence dans le paysage (Richard, 1997). Les analyses dendrologiques menées sur les bois d’architecture à Chalain 3 (Billard et al., 1997), station contemporaine de Chalain 19, montrent l’exploitation de frênes de petits calibres à vitesse de croissance rapide, probablement présents sous forme de taillis. La production pollinique du frêne, peut-être plus réduite si ces taillis sont très jeunes, pourrait alors atténuer notre perception de ces éventuelles variations. Dans cette hypothèse, on peut se demander si les fortes différences de proportions frêne/hêtre entre la zone densément bâtie et la maison isolée ne reflètent pas une occupation du bâtiment isolé très légèrement postérieure à la zone densément bâtie, ce qui conférerait une signification diachronique aux rapports frêne/hêtre observés. Toutefois, notons que ces alternances entre forêt claire et forêt dense restent à confirmer.

L’image perçue par le diagramme anthracologique peut aussi refléter un changement d’aire d’approvisionnement. Les fortes proportions de frêne au Horgen peuvent en effet résulter d’une exploitation intense de la végétation à proximité des rives du lac. Les forts taux de hêtre au Clairvaux ancien résulteraient alors d’une extension des aires de collecte vers les hauteurs. Face à l’épuisement progressif du frêne, une préservation et un partage de l’espace forestier de manière à gérer l’ensemble des activités (agricoles, collectes de matières premières, chasse, etc.) peut être à l’origine de ce changement. DISCUSSION En faveur de comportements de collecte particuliers, on remarquera la quasi absence des arbres et arbustes dont les graines et/ou les fruits ont été consommés (Schall, 2000). Ainsi, le prunellier (Prunus spinosa), l’églantier (Rosa canina), le framboisier (Rubus idaeus), le mûrier sauvage (Rubus fruticosus) ou encore le pommier (Malus silvestris) sont absents du spectre anthracologique. Certaines essences déterminées dans les brindilles (Duplaix-Rata, 1998) et utilisées pour les litières du sol et peut-être pour l’alimentation du bétail sont également absentes du spectre anthracologique telles que le buis (Buxus sempervirens), la clématite (Clematis), le fusain (Evonymus europaeus), le lierre (Hedera helix) ou encore le gui (Viscum). Toutefois, notons que la combustion de tiges de petits calibres (églantier, framboisier,

Figure 5. Stratégies d’exploitation des bois d’architecture à Chalain 14

A. Dufraisse: Économie du combustible au Néolithique final : Exemple de la station 19 du lac de Chalain (Jura, France)

mûrier, clématite, lierre, etc.) est tellement rapide que ces essences sont très vites réduites en cendres.

anthracologique donne en effet une image très proche de la végétation locale exploitée, les taxons « chefs de file » des associations représentées étant toujours présents. L’analyse révèle ainsi une transition frênaie-hêtraie, qui pourrait intégrer une cyclicité naturelle et/ou un changement d’aire d’exploitation, motivé par un épuisement progressif des peuplements de frênes et/ou une volonté de partage de l’espace forestier afin de gérer l’ensemble des activités. L’analyse de nouvelles stations néolithiques sur les lacs de Chalain et Clairvaux ainsi que l’étude des modes de croissance du bois de feu devrait nous permettre d’appuyer l’une et/ou l’autre de ces hypothèses. Quant à l’interférence possible entre l’aire de collecte du bois d’architecture (chêne) et celle du bois de feu (hêtre), l’étude de nouvelles stations est nécessaire.

Que l’on retienne ou non comme réelle cette distorsion floristique, une question concernant l’interférence possible entre l’aire d’acquisition du bois de construction et celle du combustible pourrait permettre de préciser ces stratégies d’approvisionnement. L’étude des bois de construction par l’analyse des modes de croissance des pieux (Fig. 5) a en effet montré un passage de l’utilisation de frêne (issu de peuplements jeunes) à celle de chêne (issu de peuplements âgés) lors de la transition Horgen/Clairvaux ancien. L’anthracologie par contre la caractérise par le passage d’une exploitation de frêne à celle de hêtre. Le hêtre est, de plus, utilisé de manière anecdotique en architecture quel que soit le niveau d’occupation alors qu’il domine la pluie pollinique (Richard, 1997). Pour l’architecture, il semble donc qu’un choix des essences puisse être justifié du fait de l’utilité d’une bonne durabilité des pieux (frêne et chêne, pas de hêtre) alors que le combustible représente apparemment mieux l’environnement (frêne et hêtre). Les pourcentages du frêne dans les spectres anthracologiques et son utilisation pour l’architecture et les outils (manche de hache) laissent supposer son abondance particulière près du site au Horgen. L’abandon du frêne au Clairvaux ancien que ce soit pour l’architecture ou le combustible laisse non seulement supposer sa raréfaction dans l’environnement (Cf. supra) mais aussi que les aires d’approvisionnement pour les bois d’architecture et le combustible étaient les mêmes au Horgen. Le frêne est alors remplacé par le chêne pour les pieux et par le hêtre pour le combustible. Selon les analyses polliniques, ces deux essences (chêne et hêtre), présentant la même dynamique de végétation selon les analyses polliniques, pourraient subir les mêmes pressions et donc appartenir aux mêmes aires d’exploitation. Dans ce cas, les aires de collectes au Clairvaux ancien entre bois d’architecture et bois de feu interféreraient tout comme au Horgen. Toutefois, notons qu’un peuplement de chêne bien moins important que dans la végétation actuelle est possible et expliquerait qu’il n’ait pas été privilégié dans le cadre d’un besoin quotidien. Son développement sur le premier plateau du Jura a en effet pu être favorisé lors de périodes plus tardives où les activités très consommatrices en combustible telles que la métallurgie et l’exploitation des sources salées ont nécessité la mise en place de gestions forestières favorisant le chêne et le charme et générant de nouveaux équilibres (Dufraisse, 1999). En revanche, la construction aurait largement motivé un transport à distance du chêne. Dans ce cas, la collecte du bois de construction et celle du bois de feu n’interfèrent pas puisque les stratégies d’approvisionnement reposent sur des aires de collecte différentes plutôt que sur un comportement de collecte sélectif au sein d’une même aire.

Ces différentes perspectives devraient permettre, associées aux résultats d’autres disciplines, une meilleure conception de la gestion des ressources naturelles à Chalain au Néolithique, où contraintes économiques et environnementales sont en interaction étroite, sans être nécessairement en opposition.

Adresse de l’auteur Alexa DUFRAISSE, doctorante Laboratoire de Chrono-Ecologie UMR 6565/ CNRS UFR Sciences et Techniques 16, route de Gray 25030 Besançon cedex france [email protected]

Bibliographie BEAUFILS, T. 1983, Végétation et sols des forêts de la Combe d’Ain et des reliefs environnants In L’Homme et la forêt en Franche-Comté. Structures et dynamisme de l’écosystème forêt en rapport avec les activités humaines passées et actuelles., Besançon. BEAUFILS, T. 1993, La végétation potentielle des proches environs de Clairvaux-les-lacs. In Les sites littoraux néolithiques de Clairvaux-les-Lacs et de Chalain (Jura). III, Chalain station 3, 3200-2900 av. J.-C., Paris. BILLARD, M., F. JOLY, K. LUNDSTRÖM-BAUDAIS, A. MAITRE, J.-L. MONNIER & P. PÉTREQUIN. 1997, Environnement végétal et sélection des bois d’oeuvre. In Les sites littoraux néolithiques de Clairvaux-les-Lacs et de Chalain (Jura). III, Chalain station 3, 3200-2900 av. J.-C. (ed.) P. Pétrequin. Paris: Maison des Sciences de l’Homme. CHABAL, L. 1992, La représentativité paléoécologique des charbons de bois archéologiques issus du bois de feu. In Les charbons de bois, les anciens écosystèmes et le rôle de l’homme, Montpellier.

CONCLUSION

CHABAL, L. 1997, Forêts et sociétés en Languedoc (Néolithique final, antiquité tardive). L’anthracologie, méthode et paléoécologie . Documents d’Archéologie Française (DAF) ; 63. Paris: Maison des Sciences de l’Homme.

L’étude anthracologique de cette station a permis de mettre en évidence une certaine compatibilité entre l’approche paléoécologique et celle des stratégies d’acquisition (aires d’approvisionnement, comportements de collecte et aménagement de l’espace forestier). Le spectre

DUFRAISSE, A. 1999, Analyses anthracologiques des zones d’exploitation de sources salées dans le Jura. Etude diagnostique 15

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe des sondages de Lons-le-Saunier, Salins-les-Bains et Tourmont. Paléoécologie et impact anthropique sur le milieu. DEA «Environnement et Archéologie «, Nanterre-Université.

tri-annuelle 1999. Rapport intermédiaire. Laboratoire de Chronoécologie. UMR 6565. CNRS-Université de Franche-Comté. REMMERT, H. 1987, Sukzession im klimax-system. Verhandlungen der Gesellschaft für Oekologie XVI, 27-34.

DUFRAISSE, A. à paraître, Charcoal analysis in a lake-dwelling site : Chalain 19 (Jura) : a sampling model in lacustrine context. In Second Colloque International d’Anthracologie. Paris 13-16 septembre 2000, Paris.

RICHARD, H. 1995, Analyse de l’anthropisation du milieu à partir de quelques exemples de variations de pollens d’arbres et d’arbustes. In L’homme et la dégradation de l’environnement. XVème Rencontres Internationales d’Archéologie et d’Histoire d’Antibes, Sophia Antipolis.

DUPLAIX-RATA, A. 1998, Les litières de sol. In Fontenu (Jura). Lac de Chalain station19. Rapport de synthèse. Pétrequin, P. & A.-M. Pétrequin. 1988. Le Néolithique des Lacs. Préhistoire des lacs de Chalain et de Clairvaux (4000-2000 av. J.-C.) Collection des Hespérides. Paris: Errance.

RICHARD, H. 1997, Analyse pollinique d’un sondage de 7,5 m. In Les sites littoraux néolithiques de Clairvaux-les-Lacs et de Chalain (Jura). III, Chalain station 3, 3200-2900 av. J.-C., (ed.) P. Pétrequin. Paris: Maison des Sciences de l’Homme, Paris.

PÉTREQUIN, P. 1996, Management of Architectural Woods and Variations in Population Density in the Fourth and Third Millennia B.C. (Lakes Chalain and Clairvaux, Jura, France). Journal of Anthropological Archaeology 15, 1-19.

SCHALL, C. 2000, Etude carpologique du groupe de Clairvaux ancien de la station 19 de Chalain (Jura). Maîtrise Université de Franche-Comté.

PÉTREQUIN, A.-M. & P. PÉTREQUIN. 1999, Lac de Chalain. Fontenu (Jura). Chalain 19. 1ère année de fouille programmée

16

T. Tsonev: Long Blades/Superblades in Anatolian and East Balkan Neolithic Contexts

LONG BLADES/SUPERBLADES IN ANATOLIAN AND EAST BALKAN NEOLITHIC CONTEXTS Tsoni TSONEV

Résumé : L’article proposé porte sur quatre sujets principaux: la reconstruction technique de l’exploitation des nucleus; les modèles de ‘life -cycles’ des longues lames/superlames en Anatolie et sur les Balkans de l’est; les choix préférentiels des matières premières et les messages sociales. On peut distinguer deux types principaux d’exploitation des nucleus dans la région des Balkans de l’est. ‘Life-cycles’ de cette catégorie d’artefacts montre un ‘comportement catastrophique’. Les données disponibles révèlent une préférence culturelle bien marquée menant à l’approvisionnement des matières premières de très bonne qualité: c’est à dire - l’obsidienne pour Anatolie et le silex provenant de la Bulgarie de NE - pour les Balkans de l’est. L’importance des modèles d’échanges à longue distance des objets prestigieux et exotiques comme le catalysateur des activités publiques a pu mener certainement à une perturbation d’équilibre des sources de pouvoir parmi les communautés organisées surtout en sites de type ‘tell’. Ergo - les impacts sur l’évolution sociale des ces sociétés sont deux. Le premier pouvais résulter en augmentation des tendances égalitaires qui dissimulent l’inégalité existant et les aspiration pour une pouvoir symbolique plus forte. Le deuxième impact provoque un effet contraire: l’augmentation de l’hiérarchie sociale et rituelle. Abstract: The article focuses on four main themes: technical reconstruction of core reduction sequences, patterns of ‘life-cycles’ of long blades/superblades in Anatolia and eastern Balkans, raw material preferences and social messages. It shows that there are two main types of core reduction sequences in eastern Balkans. ‘Life-cycles’ of these artefacts have catastrophic behaviour. The evidence available so far reveals a clear cultural preference for high quality raw material supply strategies: obsidian was preferred in Anatolia, while high quality flint varieties from northeastern Bulgaria - in eastern Balkans. The importance of long-distance exchange patterns of exotic and prestigious goods as a driving mechanism for increased public activities would have disturbed the balance of the sources of power within communities organised mostly in tell settlements. This has two-way impact on social evolution of these societies. The first one may have increased the egalitarian tendencies masking the existing inequalities and aspirations for greater symbolic power. The second one has the opposite effect: an increase of social and ritual hierarchy.

RECONSTRUCTION OF CORE REDUCTION SEQUENCES IN EASTERN BALKAN NEOLITHIC AND ENEOLITHIC (only for long blades and superblades)

considering the previous constraint and the high expenses for creating it I use only published materials and the facts that I personally know. – In my reconstruction I use different materials coming from different sites, workshops and collections that combined together outline, to some extent, the technology of blade production in the eastern Balkans.

The Anatolian obsidian workshops of long blades are better studied and well known, so I shall concentrate on eastern Balkan production and distribution of long blades. The presented here reconstruction of core reduction sequences is hypothetical because of the following reasons:

RECONSTRUCTION OF THE REDUCTION SEQUENCES OF THE EAST BALKAN LARGE CONICAL CORES

– A completely preserved workshop of long blades/ superblades has not yet been found in the eastern Balkans. – The typical large conical (pencil-like) cores are encountered only in northeastern Bulgaria, while long blades and superblades occur mostly on tell-sites (multi-layer sites) in Bulgaria, Macedonia and northern Greece.

I shall compare them with the Kaletepe bipolar core reduction sequence because it is published and is well-known. Both techniques differ significantly at an operational level. Yet, as a concept of highly efficient exploitation of particular raw materials and design for end-products they are quite similar.

– Most of the materials are not published and this poses problems with reconstruction of long blades/superblades spatial distribution.

At the operational level they differ in:

– My hypothesis would have been much more convincing if I could have talked in terms of a probability map of the spatial distribution of cores and the corresponding long blades/superblades in eastern Balkans and Anatolia. This could have shown in a better way the areas of intensive occurrence or, zones of absence of such artefacts. But

Kaletepe

East Balkans

mostly bipolar cores

single platform cores

obsidian sources

high quality flint sources

earlier appearance (9-8 mill. B.C.) later (6-4 mill. B.C.)

17

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Similarities:

rotation of the striking surface effectively replaces the alternation of the striking platforms in the case of the bipolar cores, the Kaletepe technique, for example. The result is a highly efficient relationship between the volume of the core and the number of detached blades (Fig. 1). The systematic error that stops this highly productive technological process is the occurrence of faults in the silica mass inside the cores. The artisans never bothered to amend or turn the damaged cores into smaller ones, in order to continue their exploitation. Apparently, in the workshops, (not settlements) they wanted only long blades. This is backed up by the fact that even if a small part of the core was damaged the artisans threw it away (we find these cores now) and started exploitation of other large cores profiting from the abundance of suitable in size and quality raw materials.

– appearance from a single highly specialized craft centre (few centres) – mass production of long blades of particular types – long distance exchange patterns – high efficiency in the production chain of desired endproducts. EFFICIENCY Both core reduction sequences are highly efficient in terms of the number of blades detached compared to the actual (not initial) volume of the cores (Fig. 1). In this respect the Kaletepe bipolar cores seem to be the most efficient ones. My simple calculations made from papers describing this technique show a monotonic decrease of the slope of the relationship: volume - number of blades (Fig. 1). The reduction process is not biased by serious systematic errors. The waste materials are only the shorter blades (two by two) detached from either side that prepare the flaking surface for detachment of straight pointed blades. The short upsilon blades that come before detachment of each pointed blade constitute another waste material from the point of view of the technological design and the desired end-product: straight, pointed blades.

The process, probably, begins with preparation of a large flint nodule of suitable silica mass (Fig. 2). Its proper exploitation starts from single, opposed or from several directions detachment of 1-2 or more super-long crested blades. Such super-crested blades are found during archaeological excavations. The next stage of large conical core exploitation begins with detachment of large series of superblades (30-40 cm. long), probably, by application of pressure-lever device. These are thin superblades with parallel edges that converge into a tip. They are slightly concave (maximum concavity) in the middle of their profile. By gradual enlargement of the flaking surface that rounds about the entire flaking surface of the core and after several rounds of clockwise detachment of superblades, the concavity of the flaking surface slides down towards the tip of the core. In order to correct this, a series of detachments with application of greater force (percussion with mediators or greater application of force to the levers or change of the levers and their tips) results into production of ‘lames outrepassées’. The greater thickness of their distal parts reduces the curvature that appeared close to the tip of the conical cores. While this technique amends the disadvantage of that kind of core reduction the decrease of the diameter of the striking surface that brings about the increase of the angle between the striking and flaking surfaces cannot be overcome so easily. In most cases the exploitation stops at this advanced but not entirely exploited stage. At the workshops in the northeastern Bulgaria many cores are left discarded at this stage of exploitation. Yet, there is another line of further reduction of these cores. After the large cores ceased to be effective for production of superblades, in order to repair the angle between the striking and the flaking surfaces, alternatively, the tip of the core or the opposite striking surface are corrected by detachment of large tablets or of series of small flakes. The single platform core turns into a double platform one exploited alternatively by either side. Overused cores of this type and long double platform blades do occur only in domestic contexts; rarely in workshops.

The Balkan large conical core reduction sequences have a number of advantages and disadvantages (Fig. 2). The first constraint comes up from the requirement of high quality materials suitable for this technological chain that localizes it only to northeastern Bulgaria and the middle Rhodopes mountains. Close to the Kaletepe technique in its efficiency is the II-nd (the smaller) type of conical cores (Fig. 2). The end-product of this technical design are long, straight blades, some of which when retouched by high, semi-steep lateral retouch form the so-called ‘Karanovo’ blades. The latter may be considered, with some caution, as the Balkan counterparts of the ‘Cayonu tools’ in Anatolia. The efficiency of this type of core reduction comes from the conical shape and perfect circular section of the striking platform. Thus the clockwise

PATTERNS OF ‘LIFE-CYCLES’ OF LONG BLADES AND SUPERBLADES IN ANATOLIA AND EASTERN BALKANS These highly specialized techniques require abundance, availability and suitability of raw materials. Indeed, both

Figure 1. 18

T. Tsonev: Long Blades/Superblades in Anatolian and East Balkan Neolithic Contexts

Figure 2.

obsidian sources and outcrops of high quality flint varieties from central Anatolia and eastern Balkans provide suitable materials for efficient production of long blades and superblades. This requirement confines these techniques of long blade production only to areas near these outcrops. On the other hand we cannot explain the mass production of long blades/superblades only at few workshops, their long distance exchange patterns, their patchy spatial distribution, the morphological preferences and the choice of materials exclusively in terms of production.

other words they appear at certain places, spread fast on long distances and then disappear rapidly. Formally, this process may be described with the Pontraigin’s principle of maximum including three parameters: work efficiency, population density and public demand (Tsonev 1996). Beyond certain critical moment work efficiency cannot be changed monotonically at any time and the increased population density requires fast imposition of new products according with the newly emerged public demand (Fig. 4). As a result of this analytical model the random local migration is kept at its minimal level and cannot serve as explanation of the wide distribution of these artefacts. The conclusion is that of primary importance for the sudden appearance and spread of highly specialized products such as long blades, various projectiles, microlithic points, Cayonu tools, Karanovo blades, etc. is not the production but the public demand that includes more product benefits and social messages than simple performance of a given task. This fact is better seen

Life-cycles If we join the time spans of the patterns of ‘life-cycles’ in a single chart of long blades/superblades in Anatolia and eastern Balkans we shall have the following picture (Fig. 3). This is a typical catastrophic behaviour of the appearances, spread and disappearances of long blades/superblades. In 19

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 3.

Figure 4.

from the point of view of the spatial distribution of long blades/superblades in Anatolia and eastern Balkans.

Kaletepe produces long pointed blades (15 cm. of length and 15 mm. of width). These artefacts constitute the typical blanks needed for prjectiles of the PPNB tradition in the Levant and southeast Anatolia (Cayonu, Cafer Hoyuk, Nevali Cori). On the other hand they are rare or absent in the central Anatolian assemblages. This style of debitage continues with the first pottery assemblages in the Levant (Balkan-Atli, at al 1999: 142-143) but remains totally unknown in the early pottery Neolithic in northwestern Anatolia (Mentese, Ilipinar) (Gatsov, pers. comm.; my own observation of the material from Mentese). The same is the distribution of long blades/ superblades in eastern Balkans: shorter blades tend to occur in northwestern Bulgaria, while in northeast and Thracian tells occur longer blades (Fig. 5). There are also some morphological preferences made in these regions. In the northwest blades seem to come from the second type of the presented above core reduction scheme, while in the other regions both techniques are present. At some sites (without

Spatial distribution of long blades/superblades If a universally important task performance stood behind the public demand for these artefacts then we would have expected that they would have had a gradual spatial distribution. This means that starting from a single point or few points these artefacts would have spread with ever diminishing pace in all possible directions. The intensity of their occurrences would have decreased slowly but steadily directly depending on the distance from the source. The power of the source would have kept the same energy in production and spread of such artefacts that would have appeared in every single site within the power range of the source. The real situation is far from such pattern of occurrences of long blades/superblades. The central Anatolian workshop at 20

T. Tsonev: Long Blades/Superblades in Anatolian and East Balkan Neolithic Contexts

Figure 5.

RAW MATERIAL PREFERENCES: “OPPOSITION” OR COMPLEMENTARY TO ONE ANOTHER CHOICES OF OBSIDIAN IN ANATOLIA AND HIGH QUALITY FLINTS IN EASTERN BALKANS

visible spatial order) long blades/superblades do not occur at all. Or, in this case, we can speak about existence of ‘communication corridors’ that better convey social messages along the line of least resistance. Such a corridor may be considered the Jantra valley and the two spatially close to one another early pottery Neolithic sites Koprivets and Orlovets that differ significantly in their techniques of core reduction sequences. In this case the technological preference relied not that much on the abundance, availability and suitability of the raw materials but the proximity to the large Jantra valley, along which high quality flints suitable for production of long blades travelled and were exchanged (Tsonev 2000). Compared to the spatial distribution of high quality raw materials the spread of long blades/superblades seems to be quite selective. It concentrates mostly on tell sites or bigger multi-layered sites. High quality flint varieties, however, appear almost everywhere though not in the same quantities. This pattern is better visible in northwestern Turkey (early pottery sites with presence of obsidian) and in southwestern Bulgaria. Both regions lack sources of obsidian and flint varieties respectively. Thus the long blades/ superblades show their own, particular distribution that differs from the simple spread of raw materials. All these suggest that we cannot expect a direct causal relationship along the line: raw material supply strategies - technological process morphology of artefacts - social messages they express. For example, the assumption that long blades associate only with long distance exchange mechanisms and with affirmation of maleness in the public and mortuary domains is, to a certain extent, deceptive. I shall consider in more detail this topic in the last section.

Recent analyses that aimed to determine the provenance of obsidian artefacts found in the archaeological sites from European Turkey have not been able to identify them with any already known obsidian source (Ozdogan, pers. comm.). On the other hand obsidian (or obsidian-like) source exists in the eastern Rhodopes, Bulgaria - exhibition at the Archaeological Museum, town of Kardzjali. My hypothesis is that if there is a possibility to locate the provenance of obsidian artefacts in Turkish Thrace to the eastern Rhodopes obsidian sources, then we would face a clear cultural preference of raw materials during the Neolithic. Obsidian was the preferred material in Anatolia. In most of the eastern Balkans, going as far east as Hodja Cesme site situated at the inlet of the Maritsa river, the high quality flints, that come predominantly from northeastern Bulgaria, constituted the systematic choice of the Neolithic population. This boundary is a clear-cut one and very few obsidian pieces have been found so far in few Neolithic sites in Bulgaria. In addition, in eastern Balkans, obsidian has never been used as a grave good. The question then arises whether this constitutes cultural “opposition” between different Neolithic populations or obsidian and “wax” flints express same social messages. The first question is why obsidian became so popular even in regions far from obsidian sources. The rapid increase of obsidian is observed in the middle and upper half of the 21

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

SOCIAL MESSAGES OF LONG BLADES, SUPERBLADES AND LONG CRESTED BLADES

Cayonu tell and it spreads throughout Anatolia, Mesopotamia and the Levant during the late pre-pottery Neolithic and later pottery periods. From the point of view of a chipped-stone knapper and user obsidian is not the best material for making tools. It is brittle and many high quality flint varieties, quartz and some quartzite have similar qualities or are harder than obsidian. They can easily be worked (even quartz and quartzite) and show greater durability of the working edges. For example, the daggers from Catalhoyuk were made from high quality flint varieties. Probably the value of obsidian lies not so much in its qualities as material for making tools but in its glossy surface, interesting texture and various colours. This new aesthetic adds additional value to it and attracts people living far away from obsidian sources. It is these features that give obsidian tools a ‘complete’ form. There is an additional fact that supports this supposition. Early farmers made a unique choice by making the so-called Cayonu tools exclusively from obsidian. There is no doubt that this choice was rather cultural though some functional characteristics of obsidian also prompt the choice in favour of this material (Anderson and Frometti, 1994). Similar features exhibit the flint varieties from northeastern Bulgaria used for manufacturing long blades and superblades. They are of wax colour and their fracture possesses “fat” surface (Sirakov, in press). The overall appearance of the superblades attracts with its glossy surface and with rich in shades and colours texture. As a matter of fact almost all of the so-called Karanovo blades, which to a certain extent may be considered as the Balkan counterparts of the Cayonu tools in Anatolia are made of high quality imported flint varieties. Though the Cayonu tools seem to be more specific and task oriented (Anderson and Frometti 1994) the Karanovo blades possess clear domestic character: generally, blades in the Karanovo I and II building horizons were used for variety of activities, such as: cutting meat, cereals, working hide, wood and bone (Gurova 1998). These tasks could have been done by blades made of quartzite or of local flint varieties but the inhabitants preferred the imported materials. Also some pressure made long blades with straight profile were little or had never been used. During the late Neolithic these long blades disappear from the Neolithic assemblages and only the big tells continued their supply with imported flints. Thus the general demand for obsidian or high quality flint varieties imposed a new aesthetic that formed a new group of highly standardized products - long, pressure made blades. Oppositions between long, pressure made blades and long, hammer made blades:

Long blade/superblade production and exchange has its climax during Eneolithic in eastern Balkans (Sirakov and Tsonev 1995). This is time of dramatic changes in the prehistoric societies. Innovations in ritual visible in the archaeological record remained unevenly distributed throughout the region. New social orders changed the focus towards the increased public consumption of exotic and prestige goods by individuals. The ideological background of constitution of personal identity involved many exchanged items (including superblades and long blades made on high quality imported raw materials) which symbolized a new incorporation of male ideology into political prestige hierarchies (Robb 1994). The importance of trade (especially the long-distance exchange patterns of exotic and prestigious goods) as a driving mechanism for increased public activity would have disturbed the balance of the sources of power within communities organised mostly in tell settlements. This had two-way impact on social evolution of these societies. The first one may have increased the egalitarian tendencies masking the existing inequalities and aspirations for greater symbolic power. The second has the opposite effect: an increase of social and ritual hierarchy. These two contradictory mechanisms may have caused the uneven geographical spread of long blades/ superblades and of high quality raw materials. (Fig. 5) The examples I gave below show a number of traits involved in the complex interplay of symbolic markers in constitution of personal identity. If we look from a gender perspective upon the above presented examples of technical and morphological choices we may easily come to direct conclusions about the male domination of some activities of the PPN population. For example, we may ask the question how straight is the relationship: raw material choice (obsidian) - the stable morphology of the Cayonu tools and their link with some male dominated social activities. This question concerns issues such as the imposition of control over long distance exchange mechanisms, over the highly specialized craft production and increased alliance building. At first glance Cayonu tools correspond perfectly to these social roles. Yet, such a straightforward relationship cannot be inferred. If we accept the conclusion reached by Patricia Anderson and Francoise Fromenti that these tools were used for stone pot decoration, or, if we just agree with the fact that Cayonu tools are much better represented at archaeological sites than the various projectiles, some of which are nothing else but perforators than the link pointed towards an increased male domination is not that obvious. All we can say is that these are objects of high craft specialization that consistently occur in domestic contexts.

• intact or accidentally broken - broken intentionally • extremely rare - occur as sickle inserts • used accidentally - traces of regular usage • imported - some imported; some local • standardized form - greater variation of width and breadth

The social messages expressed by lithic artefacts and long blades/superblades are better visible in mortuary domain. I shall present a short time evolutionary scheme of the presence of lithic artefacts in the Durankulak necropolis, northeastern coast of Bulgaria. According to N. Sirakov (2000) all lithic materials were divided into three assemblages.

These oppositions of ‘complete’ pressure made blades versus ‘broken’ sickle inserts clearly tends to reveal that ‘a message of wholeness and integration could betoken an effective contrast with the fragmentary, incomplete items symbolizing complex, enchained ties’ (Chapman 1996:215). 22

T. Tsonev: Long Blades/Superblades in Anatolian and East Balkan Neolithic Contexts Period

Presence of lithic artefacts in graves

Late Neolithic/Early Eneolithic

in poor and modest graves

Middle/Late Eneolithic

in poor and modest graves

Late Eneolithic

in relatively rich graves (richest graves in Varna necropolis, including superblades)

I shall compare the presence of lithic artefacts in the mortuary domain with the characteristics of the long blades/superblades in their domestic context. The following social traits have been distinguished:

of long blades/superblades and the high quality raw materials assumes certain social hierarchy build upon high craft production of valuable objects.

• existence of specialized workshops outside settlements

Author’s address

• fragility of long blades/superblades Tsoni TSONEV Institute of Archaeology and Museum Saborna str., 2 1000 Sofia, BULGARIA Email: [email protected]

• interesting texture and colour • direct distribution from workshops • long crested blades present in domestic context (settlement) • uneven spatial distribution

Bibliography

The various combinations of these traits reveal in a better way the social messages expressed by long blades/ superblades. For example, the direct distribution from workshops and the interesting texture and colour implies a lifestyle of their owners that is different from that of the rest of the community. This kind of silent discourse between the owners and the rest of the population helps justify and maintain hidden forms of social inequality. Those who possessed such objects justified their social status by their connection to the remote workshops and the control they, probably, exerted on the exchange mechanisms. The redistribution of high status objects (the uneven spatial distribution of long blades/superblades) created a dependence on certain social authorities and contributed to the formation of social identities. The fragility of the long blades and especially superblades suggests that they were used mostly for ritual purposes. This assumption is backed by the presence in some settlements of long crested blades without any other traces of long blade production of the kind presented above. For example, in Catalhoyuk a crested blade and some other tokens and a figurine were found in a basket. From a technical point of view crested blades start the process of production of a series of blades. This evidence implies performance of a kind of initiation rite.

ANDERSON, P. & FROMENTI, F., 1994. Exploring the use of abraded obsidian “Cayonu tools” using experimentation, optical and SEM microscopy, and EDA analysis. Archaeometry, Ankara, p. 553-566. BALKAN-ATLI, N., D. BINDER, & M.C. CAUVIN, 1999. Obsidian: Sources, Workshops and Trade in Central Anatolia. In Neolithic in Turkey, edited by M. Özdogan, & N. Basgelen. Istanbul: Arkeoloji ve Sanat Yayinlari, p. 133-143. CAPMAN, J., 1996. Enchainment, commodification, and gender in the Balkan Copper Age. Journal of European Archaeology, 4, p. 203-243. GUROVA, M., 1998. Gebrauchsspurenanalyse des neolithischen Feuersteininventars. In Osterreichisch- Bulgarische Ausgrabungen und Forschungen in Karanovo, edited by S. Hiller & V. Nikolov Bd.I, Archaologisches Institut der Universitat Salzburg, p. ROBB, J., 1994. Gender contradictions: moral coalitions and inequality in prehistoric Italy. Journal of European Archaeology 2, p. 20-49. SIRAKOV, N., 2000. Flint artefacts in Prehistoric grave-good assemblage from Durankulak necropolis. In Durankulak die Prähistorischen Gräberfelder, Band II, Teil 1, edited by H. Todorova (in press). SIRAKOV, N. & TSONEV, TS., 1995. Chipped-stone assemblage of Hotnitsa-Vodopada (Eneolithic/Early Bronze Age transition in northern Bulgaria) and the problem of the earliest “steppe invasion” in Balkans. Préhistoire Européenne, 7, p. 241-264.

The high quality raw material went through different ways of distribution than the long blades/superblades did. Though in the settlements artefacts made of these materials are abundant there is no trace of specialized flint production. Rarely occur smaller crested blades or tablets that suggest the presence of specialized workshops not far from the settlements. Thus the existing differential between the spread

TSONEV, TS., 1996. Neolithisation - a monotonic or catastrophic model for the transition to early farming. Journal of European Archaeology, 4, p. 87-99.

23

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

24

R.H. Tykot: Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean...

NEOLITHIC EXPLOITATION AND TRADE OF OBSIDIAN IN THE CENTRAL MEDITERRANEAN: NEW RESULTS AND IMPLICATIONS FOR CULTURAL INTERACTION Robert H. TYKOT

Résumé : Les objets fabriqués de l’obsidienne ont été trouvés à beaucoup sites préhistoriques dans le central Méditerranéen, jusqu’à centaines de kilomètres de leurs sources géologiques sur les îles de Lipari, Palmarola, Pantelleria, et Sardaigne. Nouvelles études de ces sources ont eu de l’emplacement plus précis et documentation de chaque courant de l’obsidienne ou affleurement pour résultat, et autorise la considération des facteurs tel qu’accès (par exemple topographie, distance de côte), dimension et fréquence des nodules, et mécanique et propriétés visuelles, dans leur exploitation. Physique et le chimique analyse de nombres grands des spécimens géologiques démontre l’utilité des méthodes pour identifier des sources, par exemple mesure précise de densité, et analyse d’élémentaire et de la composition isotopique. Analyses de nombres grands des objets fabriqués démontrent l’usage différentiel de subsources de la île, et suggère des motifs particuliers et des mécanismes pour leur exploitation. Les modèles spatiaux et chronologiques de distribution de l’obsidienne peuvent être utilisés pour adresser tel publie comme la colonisation des îles; la présentation des économies du Néolithique; et la complexité sociale croissante des Néolithiques et les sociétés de l’âge du bronze dans le central Méditerranéen. Abstract: Obsidian artifacts have been found at many prehistoric sites throughout the central Mediterranean, many of them up to hundreds of kilometers from their geological sources on the islands of Lipari, Palmarola, Pantelleria, and Sardinia. New geoarchaeological surveys of these sources have resulted in the more precise location and documentation of each obsidian flow or outcrop, and allow the consideration of factors such as access (e.g. topography, distance from coast), size and frequency of nodules, and mechanical and visual properties, in their exploitation. Physical and chemical analyses of large numbers of geological specimens demonstrates the utility of methods such as high precision density measurement, as well as elemental and isotopic fingerprinting, in identifying sources. Analyses of large numbers of artifacts demonstrate the differential use of island subsources, and suggest particular motives and mechanisms for their exploitation. The spatial and chronological patterns of obsidian distribution may be used to address such issues as the colonization of the islands; the introduction of Neolithic economies; and the increasing social complexity of Neolithic and Bronze Age societies in the central Mediterranean.

INTRODUCTION

OBSIDIAN IN THE CENTRAL MEDITERRANEAN

Obsidian sourcing has been a major aspect of archaeological research for more than a quarter-century (Cann & Renfrew 1964; Williams-Thorpe 1995; Tykot 2002a), yet until recently the central Mediterranean sources were not all fully documented (Fig. 1). Furthermore, while many studies (e.g. Hallam et al. 1976; Williams-Thorpe et al. 1979; 1984; Crisci et al. 1994; Bigazzi & Radi 1996) have contributed to a general picture of obsidian distribution, the source analysis of artifacts mostly has been limited to small numbers from any one site, limiting the determination of regional and chronological patterns of obsidian use (Tykot & Ammerman 1997).

Obsidian was used in the central Mediterranean primarily during the Neolithic period (ca. 6000-3000 BC) for blade and flake tool technologies. The sources on Lipari, Pantelleria and Palmarola are widely thought to have been a major factor in the initial settlement of these islands. Although Sardinia was settled earlier, the exploitation of its obsidian sources only began in the Neolithic. The detailed analysis of obsidian sources and mechanisms of exploitation, along with typological and use wear studies promise to provide important insights into the socioeconomic structures of Neolithic societies in this region (Tykot 1999).

Our current NSF-funded geoarchaeological survey of obsidian sources on Lipari, Palmarola and Pantelleria has employed a systematic approach to the documentation of the multiple localities where obsidian may be found, and complements previous work done on the Monte Arci sources in Sardinia (Tykot 1997; 1998). The analysis of over 1200 artifacts from dated archaeological contexts will allow geographic and chronological patterns of specific source exploitation to be recognized (for results on Sardinian obsidian, see Tykot 1996; 2001; 2002b; 2002c). These data will be used to test models of maritime capabilities, identify interconnections between island and mainland populations, and to reconstruct the economic and sociopolitical role of obsidian and other raw materials in prehistoric Mediterranean societies.

Previous research on the multiple obsidian sources in the Monte Arci region of Sardinia has revealed differential patterns of exploitation which may be attributed to chronological and cultural change, as well as geographic variation and the availability of alternative lithic raw materials. These patterns were only revealed through the systematic characterization of each obsidian source locality and the analysis of large numbers of archaeological artifacts. Distinctly different patterns of utilization of Sardinian obsidian existed at archaeological sites in France, where type SA accounts for more than 90% of the Sardinia obsidian, compared to Sardinia, Corsica and northern Italy, where it accounts for less than 50% (Tykot 1996; 2002b; 2002c; Williams-Thorpe et al. 1979; 1984; Crisci et al. 1994; Ammerman & Polglase 1997; Poupeau et al. 2000; de 25

R.H. Tykot: Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean...

significant percentage at sites in Sicily, Malta, and Tunisia (Bigazzi & Radi 1996; Francaviglia & Piperno 1987; Nicoletti 1997). This project is proceding to characterize these other island sources using the same detailed and comprehensive approach, with the expectation that previously unknown patterns of exploitation and distribution will be revealed. Geoarchaeological fieldwork was conducted in 2000 and 2001 to fully document each source and obtain samples for physical and chemical characterization studies. The preliminary results of this research are reported here.

distinguishable types of workable obsidian have been identified, including some specimens from the northern part of Punta Vardella that are quite transparent and visually similar to the most commonly used type of obsidian from Lipari, although the size and quantity of workable nodules is much more limited on Palmarola. The obsidian recovered from the southern part of Punta Vardella and from Monte Tramontana, however, is grey to black and nearly opaque (and difficult to visually distinguish from opaque Sardinian obsidian). Only devitrified obsidian of unworkable quality was found in situ anywhere else on the island. Preliminary results from elemental analysis of the geological specimens from Palmarola indicate that these three source localities are chemically distinguishable so it will be possible to determine whether obsidian for artifacts was selected from one or more of them.

LIPARI On Lipari, eruptions during the historic period have significantly changed the island’s landscape and obscured most of the obsidian outcrops available during the Neolithic (Buchner 1949; Pichler 1980). Settlements on Lipari dating to the Neolithic have been identified, however, as have large workshops in nearby western Calabria (Ammerman 1985). Earlier geochemical studies by Francaviglia (1986) and others were unable to distinguish more than one source on Lipari, and fission-track dates on 66 artifacts of Lipari obsidian from archaeological sites in Italy include only one determination older than 12,500 BP (Bigazzi & Radi 1996), suggesting that most if not all of the obsidian used in antiquity was the result of a single geological episode.

PANTELLERIA Pantelleria is located ca. 120 km from Sicily and ca. 90 km from Tunisia, making it the most remote obsidian source in the Mediterranean. Nevertheless, obsidian from Pantelleria is regularly found at Neolithic sites in Sicily and Malta, and may even have been settled itself in the Neolithic. Five chemical groups of Pantellerian obsidian had been described by Francaviglia (1988), but were not all based on the analysis of geological samples. While it is generally thought that most artifacts of Pantellerian obsidian probably come from the Balata dei Turchi sources at the southern end of the island, chemical analysis of Bronze Age artifacts from sites on Pantelleria (Francaviglia 1988), in Sicily (Francaviglia & Piperno 1987), and on Ustica (Tykot 1995) shows that obsidian from near Lago di Venere and Gelkhamar was also used. Furthermore, fission-track dates on archaeological artifacts from other sites have a range of nearly 100,000 years, suggesting the use of multiple source flows (Bigazzi & Radi 1996).

In our survey, more than 1200 geological samples were collected from over 40 localities, each precisely marked using a hand-held global positioning system unit (Fig. 3). Workable obsidian was found along the eastern and northern coasts (perhaps a mixture of prehistoric and recent flows), as well as inland at Gabellotto and above Canneto, Papesca, and Acquacalda. The in situ obsidian located on Monte della Guardia appears to be mostly of unworkable quality. Detailed examination of both these geological samples and prehistoric artifact assemblages has revealed at least two visual types of Lipari obsidian, one black and highly transparent, the other gray-banded, often with many spherulites present. Dating of each deposit will confirm its age, and chemical analyses will reveal whether more than a single ‘island’ fingerprint exists.

In 2000-2001, over 900 samples were collected from some 35 localities on Pantelleria (Fig. 5). Three vertically distinct obsidian flow layers were confirmed at both Balata dei Turchi and at Salta La Vecchia, apparently representing different eruptive cycles. Much smaller quantities of obsidian were found in situ within pumice deposits near Lago di Venere, but only surface finds of workable quality were observed near Gelkhamar, perhaps the product of airborn distribution or of erosion from a mostly buried deposit. Two extensive workshop areas were also located near Balata dei Turchi. Pantellerian obsidian may be easily distinguished from the other Mediterranean sources by its dark green color in transmitted light, and at least two visual types have been identified; preliminary results from elemental analysis confirm the existence of at least five chemical groups to which artifacts can be specifically attributed.

PALMAROLA Palmarola is a tiny island with no evidence of prehistoric settlement, and until recently its geological history has not been widely studied (Barberi et al. 1967; Tethys srl 2000). In the Neolithic, obsidian was likely obtained by seasonal fishermen coming from the mainland or the nearby island of Ponza. Obsidian on Palmarola is primarily to be found along its southeastern and northwestern shores, near Punta Vardella and Monte Tramontana, respectively. A geochemical investigation by Herold (1986) failed to reveal any differences among the deposits on the island, although fission track dates on artifacts (Bigazzi & Radi 1996) suggests a range in their age of formation.

MONTE ARCI, SARDINIA On Sardinia, all of the obsidian exploited in antiquity comes from the Monte Arci region in the west central part of the island. The existence of two or three chemical groups was

In the current study, over 300 geological samples were collected from 10 localities (Fig. 4). At least two visually 27

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 3. Map of Lipari showing localities where geological samples of obsidian were collected in this study.

revealed from early analyses of central Mediterranean artifacts (Cann & Renfrew 1964; Hallam et al. 1976), although it was not until later that the geological sources themselves were identified and fully characterized (Mackey & Warren 1983; Francaviglia 1986; Herold 1986; Tykot 1997; 1998) (Fig. 6). Best known is type SA from the Conca Cannas area, very glassy and highly translucent; the SC1 and SC2

chemical groups, represented by large blocks found between Punta Pizzighinu and Perdas Urias on the northeast side of Monte Arci, are less glassy and mostly opaque. Type SB2, which ranges from highly transparent to nearly opaque and with phenocrysts, also occurs in large natural blocks but apparently less frequently, on the western slopes of Monte Arci near Cucru Is Abis, Seddai, and Conca S’Ollastu. Type 28

R.H. Tykot: Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean...

Figure 4. Map of Palmarola showing localities where geological samples of obsidian were collected in this study. 29

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 5. Map of Pantelleria showing localities where geological samples of obsidian were collected in this study.

either very transparent or streaked with microlites and phenocrysts. Palmarola obsidian is black but opaque, although some transparent obsidian was found near Punta Vardella. Pantellerian obsidian is nearly opaque but its green color can be seen in thin edges. The color and translucency of the Pantellerian sources appears to have some systematic variation. Were it not for the several Sardinian obsidian sources, which appear similar to Lipari (SA, SB2) and to Palmarola (SB1, SB2, SC), it would be possible to reliably attribute artifacts found at Neolithic central Mediterranean sites at least to its source island based on visual means. By starting with visual assessments on artifacts from many Neolithic sites, some may be attributed without further analysis, allowing physical and chemical methods to be used as necessary on those still questionable. It must be remembered, however, that significant Neolithic usage patterns and differences have only been revealed through specific intra-island source attribution and visual analysis alone will rarely be sufficient for this purpose.

SB1, not nearly as well represented among artifacts, actually represents three chemically distinct source localities at high elevations east of Santa Maria Zuarbara (Tykot 2001). Survey of the Monte Arci region by Puxeddu (1958) revealed over 200 archaeological sites and several workshops, including the major production center of Mitza Sa Tassa near Perdas Urias where additional fieldwork to document this site has recently begun. VISUAL ANALYSIS Most provenance studies of obsidian have depended on chemical analyses which take advantage of the homogeneous elemental composition of most sources and the significant differences between each of them., although visual and physical characteristics are also useful in many cases (Tykot 2002a). Since the latter are inexpensive and non-destructive techniques, the ability to use them on very large numbers of artifacts, even if not 100% accurate, may counter-balance the statistical limitations of small numbers of chemicallysources artifacts (even if 100% accurate), and therefore should be investigated first (Tykot & Ammerman 1997). In our studies, physical attributes including workability, color, transparency, luster, and the presence and orientation of microlite crystals, larger phenocrysts, and banding are recorded for all samples. Lipari obsidian is black-grey and

DENSITY High resolution measurement of density (specific gravity) has also revealed systematic variation among Sardinian obsidian sources, and is an important factor in the calibration of hydration dates (Stevenson & Ellis 1998). Density was 30

R.H. Tykot: Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean...

Figure 7. Box-and-whisker plot of obsidian density.

analysis relied on neutron activation analysis (e.g. Hallam et al. 1976; Williams-Thorpe et al. 1979; 1984; Crummett & Warren 1985; Bigazzi et al. 1986; Ammerman et al. 1990; Randle et al. 1993; Ammerman & Polglase 1997) or X-ray fluorescence (Francaviglia 1986; Francaviglia & Piperno 1987; Crisci et al. 1994; Giardino et al. 1998; De Francesco et al. 1998; De Francesco & Crisci 2000; Poupeau et al. 2000). While all were successful in attributing artifacts to an island source, some were unable to specify a particular island sub-source either because their laboratory database did not include samples from the different subsources or they were unable to calibrate their results against already published data for these subsources. Of course none attributed artifacts to the specific Palmarola subsources only recognized in the present study, but it is particularly notable for the Sardinian artifacts in some of these studies as well. Results from the present study are now available for 320 geological samples from Lipari, Palmarola, Pantelleria, and Sardinia, comprehensively analyzed using several analytical techniques (at least 200 more are being analyzed). At the Missouri University Research Reactor Laboratories, instrumental neutron activation analysis was used to measure 27 major and trace elements, while X-ray fluorescence was used for 23 and laser ablation ICP mass spectrometry for more than 30 major and trace elements. This multi-method approach was designed to fully characterize the geological sources using the ‘technique of the future’ while allowing cross-referencing with previously published data on artifacts (mostly XRF and NAA) which may be reexamined and found sufficient to attribute them to specific island subsources. Abbreviated protocols for future analyses of obsidian artifacts are also being developed to maximize information about specific source exploitation while minimizing analytical costs and damage to artifacts. Laser ablation ICP-MS appears to best satisfy these requirements as it produces quantitative elemental (and even isotopic) data for the smallest of artifacts while leaving only the tiniest surface scar (Tykot & Young 1996; Gratuze 1999) (Fig. 8).

Figure 6. Map of Sardinia showing major geological source localities (after Tykot 1997).

therefore also determined for geological samples collected from Lipari, Palmarola, Pantelleria, and Sardinia using a Mettler electronic balance with precision to five decimal places. Cleaned pieces of obsidian were weighed and their volume determined using the water displacement method. Results indicate that Lipari, Palmarola and Pantelleria are mostly non-overlapping, while some of the Pantellerian sources (Lago di Venere: 2.51 “ 0.03; Balata dei Turchi 1: 2.46 “ 0.02; Balata dei Turchi 2: 2.62 “ 0.03) may also be differentiated (Fig. 7). The combination then of visual analysis and density measurement can be quite reliable for attributing artifacts to a single island, and in the case of Pantelleria some more specific source localities. ELEMENTAL ANALYSIS

As already indicated, trace element analysis of the Sardinian samples show seven distinct groups, but the four most significant (SA, SB1, SB2, SC) are also differentiable using major element analysis (Figs. 9-10). So when it is already

For the last quarter-century, the vast majority of provenience studies of central Mediterranean obsidian using chemical 31

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

SOURCE TRACING OF ARTIFACTS Obsidian is very commonly found at Neolithic sites in the central Mediterranean, although in generally decreasing frequency at greater distance from the sources. Long-standing research efforts by a few scholarly groups have revealed a general pattern of obsidian distribution in the region although, as mentioned above, the number of sites for any one cultural period and with more than a few analyzed artifacts is still quite limited (Williams-Thorpe 1995; Tykot 1996; Bigazzi & Radi 1996) (Fig. 11). Nevertheless, it is becoming clear that obsidian from Palmarola, rarely documented previously in northern Italy, is actually well represented at a few sites (Ammerman et al. 1990; Ammerman & Polglase 1997), while Pantellerian obsidian is continuously present in the Neolithic levels of Grotta dell’Uzzo in Sicily (Francaviglia & Piperno 1987) and common elsewhere (Nicoletti 1997), when Lipari was previously thought to be a nearly exclusive source for Sicily. Lipari appears to have been the main source, however, of obsidian distributed in the Adriatic, both along the Italian coast and across to the islands of Palagruza, Susac, Korcula, and elsewhere along the Dalmatian coast (Tykot et al. 2001).

Figure 8. Barely visible laser ablation ICP-MS craters.

known that artifacts are made of Sardinian obsidian (e.g. by a combination of site location and date, visual and density analysis), other techniques such as the electron microprobe are also sufficient for source specific attribution (Tykot 1997). The preliminary results available for Lipari fall into a single chemical group, although analyses for several localities have not yet been completed, so at present there is no advantage of one elemental technique over another for artifacts made of Lipari obsidian. Obsidian from Palmarola falls into three distinct groups, while at least five groups are present among the samples analyzed from Pantelleria; in both cases trace element analysis is necessary to attribute artifacts to their specific source locality. It must be noted that certain methods which have been successfully used to attribute artifacts to a particular island, including fission track dating and elemental analysis using a scanning electron microscope (Acquafredda et al. 1999), cannot distinguish among the island subsources.

The similar relative frequencies of the Sardinian obsidian sources in Early Neolithic assemblages in northern Sardinia, Corsica, and Tuscany are supportive of a simple down-theline trade mechanism, while the dominance of type SA obsidian in southern France must reflect specific selection for cultural preferences and/or functional reasons (Tykot 1996; 1999; 2002b; Tykot et al. 2002). Chronological change has also been documented at individual sites such as Arene Candide in Liguria, where it has been proposed that a shift occurred in the Late Neolithic towards the presence of

Figure 9. Major element plot of Mediterranean obsidian sources. Data from Francaviglia (1986; 1988) and Tykot (1997). 32

R.H. Tykot: Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean...

Figure 10. Trace element plot showing 7 sources on Sardinia. Similar plots show 3 sources on Palmarola and 5 on Pantelleria.

Figure 11. General distribution pattern of obsidian from central Mediterranean sources for the Neolithic period. All obsidian tested from Corsica and Sardinia (individual sites not shown) has been attributed to Monte Arci. At most sites, only a few artifacts have been tested, and the broad time span represented obscures chronological variation as well. 33

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Author’s address

exclusively small blades of Lipari obsidian while Sardinian obsidian was obtained as both prepared cores and larger blades (Ammerman & Polglase 1997), and at Filiestru in Sardinia, where the decreasing frequency of types SA and SB obsidian may be explained by increasingly intense exploitation of obsidian with particular physical properties (Tykot 1996; 2002c). These patterns would not have been observed if the individual Monte Arci obsidian sources had not been documented and characterized, and perhaps provide a glimpse of the patterns which may become noticeable for the other central Mediterranean island sources and the interpretations that may be made about their exploitation.

Robert H. TYKOT Department of Anthropology University of South Florida 4202 East Fowler Ave., SOC107 Tampa, Florida 33620 USA Email: [email protected]

Bibliography ACQUAFREDDA, P., ANDRIANI, T., LORENZONI, S., & ZANETTIN, E., 1999, Chemical characterization of obsidians from different Mediterranean sources by non-destructive SEMEDS Analytical Method. Journal of Archaeological Science 26, p. 315-325.

CONCLUSION While the characterization of the geological sources on Lipari, Palmarola and Pantelleria are still in progress, a much better understanding of the accessibility, quantity and quality of obsidian available from these islands for prehistoric exploitation already has been obtained, and the central Mediterranean island sources may be reliably distinguished based on their visual characteristics and density, both inexpensive and non-destructive techniques. Chemical analysis allows precise source attributions, and analyses of obsidian collections from significant archaeological sites are in progress which will double the number of artifacts tested for the entire region. This research will ultimately allow quantitative rather than qualitative assessment of obsidian exploitation and distribution during the Neolithic, and will be integrated with studies of typology, reduction technology and use function to reconstruct the entire chaîne opératoire from acquisition to tool disposal. Only in so doing can we interpret dynamic spatial and temporal behavior patterns in their functional and cultural contexts.

AMMERMAN, A.J., 1985, The Acconia Survey: Neolithic Settlement and the Obsidian Trade. London: Institute of Archaeology. AMMERMAN, A. J., CESANA, A., POLGLASE, C. & TERRANI, M., 1990, Neutron activation analysis of obsidian from two Neolithic sites in Italy. Journal of Archaeological Science 17, p. 209-20. AMMERMAN, A. & POLGLASE, C., 1997, Analyses and descriptions of the obsidian collections from Arene Candide. In Arene Candide: A Functional and Environmental Assessment of the Holocene Sequence, edited by R. Maggi. Rome: Il Calamo, p. 573-592. BARBERI, F., BORSI, S., FERRARA, G., & INNOCENTI, F., 1967, Contributo alla conoscenza vulcanologica e magmatologica delle isole dell’arcipelago Pontino. Memorie della Societa Geologica Italiana 6, p. 581-606. BIGAZZI, G., MELONI, S., ODDONE, M. & RADI, G., 1986, Provenance studies of obsidian artifacts: trace elements analysis and data reduction. Journal of Radioanalytical and Nuclear Chemistry 98, p. 353-63. BIGAZZI, G. & RADI, G., 1996, Prehistoric exploitation of obsidian for tool making in the Italian peninsula: a picture from a rich fission-track data-set. In XIII International. Congress Prehistoric and Protohistoric Sciences, edited by C. Arias, A. Bietti, L. Castelletti & C. Peretto. Forlí: A.B.A.C.O., vol. 1, p. 149156.

Acknowledgements This project is currently funded by an NSF grant (SBR0075535) to R.H. Tykot. A. Zarattini (Soprintendenza Archeologica per il Lazio), Umberto Spigo (Museo Archeologico Regionale Eoliano), and S. Tusa (Soprintendenza ai Beni Culturali di Trapani) gave permission for access to the obsidian sources of Palmarola, Lipari, and Pantelleria, respectively; A. Molinaro provided assistance on Palmarola, M.C. Martinelli on Lipari, and M. Tosi, V. Colella & R. di Fresco on Pantelleria. Many colleagues have also have provided access to artifact collections, including E. Atzeni, B. Bass, G. Bulgarelli, J. Cesari, S. Forenbaher, T. Kaiser, F. Lo Schiavo, J. Robb, V. Santoni, and C. Tozzi. Students L. Beyer, J. Bliss, T. Setzer, and B. Vargo assisted with the field collection in 2000 and 2001, while J. Bernal, J. Dukes, R. Glazer, S. McReady, G. Walton, and J. Winkler also assisted with the labwork at USF. Chemical analyses were done at the Archaeometry Laboratory at the Missouri University Research Reactor under the direction of M.D. Glascock and for LA-ICP-MS with the assistance of R.J. Speakman. Earlier work in Sardinia was supported by the American Schools of Prehistoric Research, Sigma Xi, Harvard University, and the University of South Florida.

BUCHNER, G., 1949, Ricerche sui giacimenti e sulle industrie di ossidiana in Italia. I. Rivista di Scienze Preistoriche 4, p. 162186. CANN, J.R. & RENFREW, C., 1964, The characterization of obsidian and its application to the Mediterranean region. Proceedings of the Prehistoric Society 30, p. 111-133. CRISCI, G.M., RICQ-DE BOUARD, M., LANZAFRAME, U., & DE FRANCESCO, A.M., 1994, Nouvelle méthode d’analyse et provenance de l’ensemble des obsidiennes néolithiques du Midi de la France. Gallia Préhistoire 36, p. 299-309. CRUMMETT, J.G. & WARREN, S.E., 1985, Appendix I: Chemical analysis of Calabrian obsidian. In The Acconia Survey: Neolithic Settlement and the Obsidian Trade, edited by A.J. Ammerman. London: Institute of Archaeology, p. 103-114. DE FRANCESCO, A.M. & CRISCI, G.M., 2000, Provenienza delle ossidiane dei siti archeologici di Pianosa (Arcipelago Toscano) e Lumaca (Corsica) con il metodo analitico non distruttivo in fluorescenza X. In Il primo popolamento Olocenico dell’area corso-toscana, edited by C. Tozzi & M.C. Weiss. Pisa: Edizioni ETS, p. 253-258. 34

R.H. Tykot: Neolithic Exploitation and Trade of Obsidian in the Central Mediterranean... DE FRANCESCO, A.M., CRISCI, G.M., & LANZAFAME, U., 1998, Nuovi sviluppi del metodo analitico non distruttivo in fluorescenza X per risalire all provenienza di ossidiane archeologiche. In Le Scienze della Terra e l’Archeometria, edited by C. D’Amico & C.A. Livadie. Napoli: Cuen, p. 141-145.

Soprintendenza ai Beni Culturali ed Ambientali Provincia di Palermo in Collaboration with Brown University in 1990 and 1991, edited by R.R. Holloway & S.S. Lukesh. Providence and Louvain-La-Neuve: Archaeologia Transatlantica XIV, p. 87-90. TYKOT, R.H., 1996, Obsidian procurement and distribution in the central and western Mediterranean. Journal of Mediterranean Archaeology 9, p. 39-82.

FRANCAVIGLIA, V.M., 1986, Characterization of Mediterranean obsidian sources by classical petrochemical methods. Preistoria Alpina 20, p. 311-332.

TYKOT, R.H., 1997, Characterization of the Monte Arci (Sardinia) obsidian sources. Journal of Archaeological Science 24, p. :467479.

FRANCAVIGLIA, V.M., 1988, Ancient obsidian sources on Pantelleria (Italy). Journal of Archaeological Science 15, p. 109122.

TYKOT, R.H., 1998, Mediterranean islands and multiple flows: the sources and exploitation of Sardinian obsidian. In Method and Theory in Archaeological Obsidian Studies, edited by M.S. Shackley. New York (NY): Plenum, p. 67-82.

FRANCAVIGLIA, V.M. & PIPERNO, M., 1987, La repartition et la provenance de l’obsidienne archeologique de la Grotta dell’Uzzo et de Monte Cofano (Sicile). Revue d’Archéométrie 11, p. 31-39.

TYKOT, R.H., 1999, Islands in the streams: stone age cultural dynamics in Sardinia and Corsica. In Social Dynamics of the Prehistoric Central Mediterranean, edited by R.H. Tykot, J. Morter & J.E. Robb. London: Accordia Research Institute, p. 67-82.

GIARDINO, C., GUIDI, G.F., MOIOLI, P. & NEGRI, P., 1998, Correlazione mediante spettrometra XRF di reperti i ossidiana provenienti da Capri con ossidiane del bacino del Mediterraneo. In Le Scienze della Terra e l’Archeometria, edited by C. D’Amico & C.A. Livadie. Napoli: Cuen, p. 295-299.

TYKOT, R.H., 2001, Chemical fingerprinting and source-tracing of obsidian: the central Mediterranean trade in black gold. Accounts of Chemical Research 36 (2002). Published on the web (http://dx.doi.org/10.1021/ar000208p) November 2001.

GRATUZE, B., 1999, Obsidian characterisation by laser ablation ICP-MS and its application to prehistoric trade in the Mediterranean and the Near East: Sources and distribution of obsidian within the Aegean and Anatolia. Journal of Archaeological Science 26, p. 869-881.

TYKOT, R.H., 2002a, Determining the source of lithic artifacts and reconstructing trade in the ancient world. In Written in Stone: The Multiple Dimensions of Lithic Analysis, edited by P.N. Kardulias & R. Yerkes. Maryland: Lexington Books.

HALLAM, B.R., WARREN, S.E., & RENFREW, C., 1976, Obsidian in the western Mediterranean: characterisation by neutron activation analysis and optical emission spectroscopy. Proceedings of the Prehistoric Society 42, p. 85-110.

TYKOT, R.H., 2002b, Geochemical analysis of obsidian and the reconstruction of trade mechanisms in the Early Neolithic of the western Mediterranean. In Archaeological Chemistry VI, edited by K. Jakes. American Chemical Society Symposium Series.

HEROLD, G., 1986, Mineralogische, chemische und physikalische untersuchungen an den obsidianen Sardiniens und Palmarolas: Grundlagen zur Rekonstruktion prähistorischer Handelswege im Mittelmeerraum. Unpublished Ph.D thesis, Universität (TH) Fridericiana Karlsruhe.

TYKOT, R.H., 2002c, New approaches to the characterization and interpretation of obsidian from the Mediterranean island sources. Materials Research Society Symposium Proceedings 712.

MACKEY, M. & WARREN, S.E., 1983, The identification of obsidian sources in the Monte Arci region of Sardinia. In Proceedings of the 22nd Symposium on Archaeometry, University of Bradford, Bradford, U.K. March 30th - April 3rd 1982, edited by A. Aspinall & S.E. Warren. Bradford, p. 420-431.

TYKOT, R.H. & AMMERMAN, A.J., 1997, New directions in central Mediterranean obsidian studies. Antiquity 71(274), p. 1000-1006. TYKOT, R.H., BASS, B. & DELLA CASA, P. 2001. Obsidian as an indicator of prehistoric trade and interaction in the Adriatic basin. Paper presented at the 66th Annual Meeting of the Society for American Archaeology, New Orleans.

NICOLETTI, F., 1997, Il commercio preistorico dell’ossidiana nel Mediterraneo ed il ruolo di Lipari e Pantelleria nel più antico sistema di scambio. In Prima Sicilia: alle origini della societä siciliana. Siracusa: Ediprint, p. 259-269.

TYKOT, R.H., VARGO, B.A., TOZZI, C. & AMMERMAN, A., 2002. Nuove analisi di reperti di ossidiana rinvenuti nella provincia di Livorno. In Atti del XXXV Riunione Scientifica, Le Comunità della Preistoria Italiana. Studi e Ricerche sul Neolitico e le Età dei Metalli in memoria di Luigi Bernabò Brea. Firenze: Istituto Italiano di Preistoria e Protostoria.

PICHLER, H., 1980, The island of Lipari. Rendiconti della Società Italiana della Mineralogia e Petrologia 36, p. 415-440. POUPEAU, G., BELLOT-GURLET, L., BRISOTTO, V., & DORIGHEL, O., 2000, Nouvelles données sur la provenance de l’obsidienne des sites néolithiques du sud-est de la France. C.R. Acad. Sci. Paris, Sciences de la Terre et des Planètes / Earth and Planetary Sciences 330, p. 297-303.

TYKOT, R.H. & YOUNG, S.M.M., 1996, Archaeological applications of inductively coupled plasma-mass spectrometry. In Archaeological Chemistry V, edited by M.V. Orna. American Chemical Society: Washington, DC., p. 116-130.

RANDLE, K., BARFIELD, L.H. & BAGOLINI, B., 1993, Recent Italian obsidian analysis. Journal of Archaeological Science 20, p. 503-9.

WILLIAMS-THORPE, O., 1995, Obsidian in the Mediterranean and the near East: a provenancing success story. Archaeometry 37, p. 217-248.

STEVENSON, C.M. & ELLIS, J.G., 1998, Dating Sardinian archaeological obsidian. In Sardinian and Aegean Chronology, edited by M.S. Balmuth & R.H. Tykot. Oxford: Oxbow, p. 1723.

WILLIAMS-THORPE, O., WARREN, S.E., & BARFIELD, L.H., 1979, The sources and distribution of archaeological obsidian in northern Italy. Preistoria Alpina 15, p. 73-92.

TETHYS SRL (a cura di), 2000. Evoluzione Quaternaria dell’Isola di Palmarola. Elementi geomorfologici emersi et sommersi. Report produced for the Soprintendenza Archeologica per il Lazio, direzione scientifica Dott.ssa Annalisa Zarattini, Rome.

WILLIAMS-THORPE, O., WARREN, S.E., & COURTIN, J., 1984, The distribution and sources of archaeological obsidian from southern France. Journal of Archaeological Science 11, p. 13546.

TYKOT, R.H., 1995, Appendix I: Obsidian Provenance. In Ustica I. The Results of the Excavations of the Regione Siciliana

35

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

36

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large

OUTILS AGRICOLES DES ASSEMBLAGES NÉOLITHIQUES EN SILEX DE LA BULGARIE - UN CONTEXTE PLUS LARGE Maria GUROVA

Résumé : Pendant les années 80, une démarche d’analyse tracéologique sur les collections néolithiques bulgares en silex a été effectuée par la spécialiste russe N. Skakun. Ses recherches portaient sur certains assemblages de la Bulgarie de l’Ouest et du Nord-Est. Ces dernières années je suis en train d’étudier les collections en silex provenant des tells Azmak, Karanovo et Kapitan Dimitrievo (Néolithique ancien et récent); du site Kovachevo (étude en cours - Néolithique ancien) et Drama-Gerena (Néolithique récent), situés en Bulgarie du Sud. J’ai pu, également, voir des assemblages du Néolithique nommé ‘monochrome’ - la phase la plus ancienne dans la séquence des cultures néolithiques sur les Balkans, provenant de certains sites en Bulgarie du Nord. Cela permet d’élargir les paramètres temporaires et spatiales de la problématique proposée. L’analyse tracéologique est effectuée à l’aide des microscopes MBS 10 (jusqu’à 100x) et METAM P1 (jusqu’à 500x). Les microphotographies sont prises à WILD MPS 51. Dès le début du Néolithique ancien, dont la culture est manifesté par le site Kovacevo et la période Karanovo I bien attestée et étudiée des tells Azmak, Karanovo, Kapitan Dimitrievo (Thrace bulgare), on constate une abondance et une variété typologique considérables des armatures de faucille. Presque toutes les pièces possèdent des polis angulaires indiquant une insertion oblique (en épi) des éléments dans une manche courbe, caractéristique pour la faucille classique et bien connue de ‘type Karanovo’. Les lames-faucilles avec du poli latéral et parallèle sont très rares. L’abondance des armatures de faucille et leur utilisation parfois exhaustive témoignent d’une pratique et d’une tradition agricole assez développées. Cette constatation s’accorde bien avec les résultats des études paléobotaniques qui révèlent un large registre des taxons des céréales domestiques. Au sein des collections lithiques du Néolithique récent on a, également, identifié certains éléments de tribulum. Leur apparition, même sporadique parmi des assemblages lithiques du Néolithique moyen/récent présume un niveau plus évolué des techniques agricoles. Jusqu’au présent les éléments de tribulum ont été découverts par N. Skakun au sein des collections chalcolithiques de la Bulgarie du Nord-Est et des collections néolithiques de la Bulgarie de l’Ouest, mais avec la position stratigraphique incertaine. En général l’étude des assemblages en silex de la séquence néolithique, et particulièrement de la région de la Thrace (Bulgarie du Sud), révèle un développement continu dans le cadre de l’agriculture comme une des activités principales de subsistance de la population sédentaire. Abstract: During the 80s a use-wear analysis of Bulgarian Neolithic flint collections was done by the Russian specialist N. Skakun. Her investigations include some assemblages from western and north-eastern Bulgaria. At the last years I have worked on the collections coming from the tells Azmak, Karanovo, Kapitan Dimitrievo (Early and Late Neolithic), from the Kovacevo site (Early Neolithic - the research is under progress) and from the Drama-Gerena site (Late Neolithic), situated in southern Bulgaria. I also studied some materials from the so-called ‘monochrome’ phase - the oldest one of the Neolithic cultures sequence on the Balkans-provided from northern Bulgaria. This fact allows an enlargement of the problematic on question. The use-wear analysis was made by the microscope MBS 10 (up to 100x), METAM P1 (up to 500x). The microphotos are taken by using WILD MPS 51. At very beginning of the Early Neolithic, manifested in the Kovacevo site and during the Karanovo I period, the sickle inserts are numerous and show a great typological variability. Almost all pieces have an angular cereal polish (gloss) that marks an oblique insertion of the flint inserts into the curved antler handle - the most characteristic trait of the well-known ‘Karanovo type’ sickle. The sickle-blades with a lateral and parallel polish are extremely rare. The abundance of sickle inserts and their exhaustive use suggests that farming activities were well developed. This fact coincides with the results of the palaeobotanical studies which reveal a large range of domesticated cereals taxa. Within the frame of the flint collections from the Late Neolithic, there were identified some tribulum inserts. Their occurrence, though rare in the Middle/Late Neolithic assemblages presumes a higher level of farming techniques. Up to the present day, tribulum inserts have been detected by N. Skakun within the frame of the Chalcolithic collection from northeastern, and in the Neolithic assemblages from western Bulgaria, but without clear stratigraphic position. Generally, the study of the flint assemblages of the Neolithic sequences, and particularly in the region of Thrace (south Bulgaria), suggests a continuous development of the agriculture that constitutes of the most important subsistence activities of the sedentary farming communities.

INTRODUCTION

• l’apparition au début du Néolithique bulgare et l’utilisation très longue et étendue de faucilles de ‘type Karanovo’ (cf. infra). Il faut mentionner ici que le tell éponyme de Karanovo est un site de repère pour la chronologie et la périodisation des cultures néolithiques, chalcolithiques et de l’âge du Bronze ancien dans l’Europe de Sud-Est;

La Bulgarie représente une zone intermédiaire et assez importante pour l’application et le traitement du problème des voies de la diffusion de l’agriculture comme un des attributs de la néolithisation. Cette position avantageuse se manifeste avec des faits suivants: 37

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

• mise en évidence l’usage du tribulum dont les éléments en silex inserrés sont actuellement attestées au sein des phases récentes du Néolithique (cf. infra);

spécialiste russe N. Skakun (Skakun 1993a,1993b, 1994a). Ses recherches portent sur certains assemblages de la Bulgarie de l’Ouest, dont une partie est du Néolithique ancien, une autre partie - du Néolithique moyen et récent, et également l’industrie d’Ussoe (Bulgarie du Nord-Est), associée à l’étape finale du Néolithique (Skakun 1994b). N. Skakun identifie un large registre de types fonctionnels d’outils en silex, y compris des pièces agricoles.

• les résultats des études paléobotaniques des sites du Néolithique ancien, qui révèlent les restes de céréales domestiquées (y compris une espèce prédecesseure) suivantes: Triticum monococcum L (engrain), Triticum boeoticum Boiss. (engrain sauvage), Triticum dicoccum Schrank. (amidonnier), Triticum durum/aestivum (blé dur), Triticum aestivum L (froment), Hordeum vulgare L (orge) (Doceva 1992; Lisitsina, Filipovich 1980; Popova 1989, 1991, 1995; Thranheiser 1997).

Cette problématique, pourtant, n’est pas épuisée, étant donné l’abondance des assemblages en silex apparus dernièrement en cours des fouilles de sites comme Koprivets et Pomoshtitsa - liés avec des phase initiale ( soi-disant ‘monochrome’) du Néolithique en Bulgarie du nord; des tells Karanovo, DramaGerena et Kapitan Dimitrievo (Thrace bulgare); du site de Kovacevo situé dans l’extrême sud-ouest du pays. Ces sites et leur cadre chrono-culturel donnent une mosaïque assez intéressante et variée de l’aspect temporaire et spatial (Fig. 1). Mes études tracéologiques sur les collections lithiques des sites néolithiques mentionnés ci-dessus datent des 5 dernières années et ont comme objectif d’élargir le cadre contextuel (y compris les aspects temporaires) de l’outillage agricole sur une base empirique assez importante et à l’aide des études comparatives avec des matériels des régions voisines. C’est une ambition pour l’avenir, vu que pour l’instant ce sont seulement les collections lithiques d’Ilipinar (Anatolie du nord-ouest) qui ont été examinées (Gurova 2001c). Je me suis initiée dans l’étude des assemblages de la région de

• le cadre chronologique des cultures néolithiques et leurs faciès locaux est basé sur les séries des dates C14 BC cal. dont les échantillons variés proviennent de sites assez nombreux (Boyadziev 1995; Görsdorf 1997; Görsdorf, Bojadziev 1996). Les données généralisées sont présentées ci-dessous: - Néolithique ancien - 6000/5900 - 5500/5450; - Néolithique moyen - 5500/5450 - 5200/5100; - Néolithique récent - 5200/5100 - 4900/4850 (Boyadziev 1995: 179). Le problème des outils agricoles lithiques provenant de la Bulgarie a été abordé et élucidé, autant que possible, pendant les années 80 grâce aux études tracéologiques menées par la

Figure 1. Carte de la Bulgarie avec la localisation des sites préhistoriques mentionnés dans le texte. 38

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large

Troade - Kumtepe et Besik Sivritepe et j’ai une forte espoir de commencer travailler sur certains site néolithiques en Crèce.

nom s’attribue à un aspect culturel et stylistique de cet instrument sans avoir une valeur chronologique/temporaire. Et c’est uniquement dans ce sens qu’on l’utilise dans l’exposé présenté ci-dessous (Fig. 2). Cet outil, qui possède une forme évoluée et une efficacité optimale, est attesté dès le début de la culture du Néolithique ancien Karanovo I qui s’étend sur un vaste territoire en Bulgarie du sud et dont la position chronologique est entre 6000 et 5750 BC ( C14 cal.) (Cörsdorf 1997: 370; Görsdorf, Bojadziev 1996). Son apparition mène à une étape même plus ancien, d’après les données (y compris les dates C14 pas encore publiées) du site Kovacevo (vide infra).

L’analyse tracéologique des matériels présentés ci-dessous est effectuée à l’aide des microscopes MBS 10 (x 100) et METAM P1 (x 500). Les microphotos sont prises avec WILD MPS 51 - leur magnification est x100. ÉLÉMENTS DE FAUCILLE Dès les années 30 du XX siècle, la faucille de ‘type Karanovo’, découverte sur le tell éponyme, devient un outil très connu (Georgiev 1958, 370-71, Fig. 2). Il s’agit de faucille à manche courbe et insertion oblique (en épi) des armatures, qui, depuis des décades, est citée dans la littérature sous le nom de faucille de ‘type Karanovo’ (Helmer 1983: 190; Korobkova 1978: 39; Moundrea-Agrafioti 1983: 204; Skakun 1993b: 364; 1994a: 295). Il faut préciser ici que ce

Les données sur les éléments de faucilles attestées parmi les assemblages lithiques étudiés sont présentées dans les tableaux 1 et 2. Les observations et les constatations sur les artefacts en silex, utilisés comme éléments de faucilles peuvent être généralisées de la manière suivante:

Figure 2. Faucille de ‘type Karanovo’ (d’après G. Georgiev, 1958, p. 371, fig.2) avec des éléments (lames brutes et retouchées) de la culture Karanovo I (de tells Karanovo et Azmak - Néolithique ancien) et de la microphotographie des traces d’usure sur la pièce N 4. 39

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe Tableau 1. Éléments de faucilles provenant des assemblages étudiés par périodes et sites. sites

période du Néolithique ancien

total

récent 80

Azmak

80

Karanovo

35

11

46

Kapitan Dimitrievo

4

12

16

26

26

Drama- Gerena Kovacevo

111

total

230

111 49

279

Tableau 2. Répartition des éléments de faucille par catégories d’artefacts. sites

outils typologiques ancien

récent

lames brutes ancien

total

récent

ancien

5

35

11

3

12

80

Azmak

47

Karanovo

19

6

33

Kapitan Dimitrievo

3

7

5

22

3

Drama- Gerena Kovacevo

54

total

123

total

16

86

158

13 99

• Malgré la quantité non comparable des pièces des différentes collections étudiées on peut noter que dans l’aspect diachronique le rôle des lames brutes (y compris les autres produits de débitage) comme éléments de faucille diminue, tandis que le nombre et la variété typologique des outils augmentent.

25 91

37 35

récent

209

48 257

mesure du fonctionnement des éléments de faucille sont également attestées. • La constatation qui s’impose, du point de vue morphométriques et typologique, c’est: les éléments de faucille sont très variés et ne se limitent pas aux lames standardisées comme il était d’ordinaire admis (Skakun 1993 a,b; 1994 a). La prédominance des éléments de faucille aux polis en diagonale, indiquant l’insertion en épi, est très bien marquée (Fig. 3 et 4). Les pièces-exemplaires avec des polis plutôt parallèles aux tranchants sont sporadiques et peuvent être interprétées comme couteaux à découper des tiges des céréales déjà moissonnées et stockées. Cette explication est admissible et s’accorde bien avec les observations de la spécialiste - paléobotaniste examinant les matériels du tell Kapitan Dimitrievo (Gurova 2001b: 46; Marinova 1999: 127).

• Il n’y a pas de paramètres morphologiques standardisés en ce qui concerne les lames utilisées en armatures de faucille. Il s’agit d’habitude de fragments proximaux ou mésiaux, qui varient en dimensions entre 2 et 5 cm longueur et 1 et 3 cm largeur. • Les catégories typologiques les plus fréquemment présentées parmi les pièces commentées sont les suivantes: lames retouchées (aux retouches marginales, partieles, semi-abruptes, scalariformes), éclats retouchés, lames tronquées, perçoirs, grattoirs, pièces à encoches. Il faut mentionner que la diversité des types augmente pendant les étapes récentes du Néolithique, ce qui est bien observable parmi les outillages du site Drama-Gerena (Gurova - étude à publier).

• En ce qui concerne les assemblages des sites, possédant les horizons ‘monochromes’, j’ai pu voir le matériel en silex de Koprivets, qui m’était gentillement mis en disposition par I. Vaisov. Parmi les artefacts (comptant au total une centaine de pièces), il n’y avait même pas un seul élément de faucille. Il y en a assez, par contre, au sein de la collection lithique de Pomoshtitsa (étude en cours), ou on présume un niveau ‘monochrome’, mais, étant donné le fait que la situation stratigraphique demande des précisions supplémentaires, je ne me permets pas d’interprétation précaire.

• Parmi les lames aux retouches abruptes et scalariformes, particulièrement répandus au sein des outillages de la culture Karanovo I, il y a des pièces dont les tranchants possèdent parfois des traces relictes d’éléments de faucilles. C’est à dire qu’une fois utilisés en cette fonction, ces outils ont été réaffûtés (réavivés) à l’aide de retouches abruptes modifiant fortement la morphologie des artefacts qui restent d’habitude sans usage postérieur. Les retouches partielles d’utilisation et de réavivage des tranchants au fur et à

Dans le tableau ne sont pas inclues 22 pièces de débitage ayant été utilisées comme éléments de faucille: 1 fragment 40

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large

Figure 3. Éléments de faucilles et microphotos des traces d’usure (x100) du site Kovacevo (Néolithique ancien).

Figure 4. Éléments de faucilles et microphotos des traces d’usure (x100) du tell Kapitan Dimitrievo (Néolithique récent).

du Néolithique ancien de Kapitan Dimitrievo; 10 éclats et 10 fragments de Kovacevo et 1 éclat de Drama.

tracéologiques entre le traitement des céréales sauvages et cultivées etc. (Anderson-Gerfaud 1988; Anderson 1994a, b, 1999; Gassin 1996; van Gijn 1990; Ibanez et all. 1998; Jensen 1993; Korobkova 1993; Skakun 1994a; Unger-Hamilton 1989, 1999). Il est déjà démontré, prouvé et admis que les faucilles à manche courbe et tranchant en épi sont très efficaces et convenables pour assurer et maintenir une bonne récolte des céréales (Ibanez et all. 1998: 143; Korobkova

Dans un aspect contextuel et plus général, il faut noter que le thème des faucilles est très exploité (au point d’être banal!) dans la littérature spécialisée et porte surtout sur des travaux expérimentaux solides, consacrées aux gestes de moisson et à l’efficacité des outils appliqués; aux différences 41

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

1978: 49; Skakun 1994a: 297). Le problème de l’origine et de la diffusion successive des faucilles de ce type reste, néanmoins, peu élucidé. Sans prétention d’éclaircir ce sujet compliqué, on peut résumer les données disponibles en soulignant certains points importants:

Vaughan 1983: 219). Du Néolithique ancien (~5700 bc), les artefacts lustrés sont très variés, mais ...”en fait sans rapport direct avec la consommation de céréales”. Autrement dit, un phénomène bizarre est mis en évidence: les céréales sauvages et domestiquées identifiés sans les faucilles attestées (Perlès, Vaughan, op. cit., 211). Elster mentionne la présence d’une lame- élément de faucille parmi le matériel du Néolithique ancien de Korinth, mais il n’y a pas de précision (Elster 1977: 55)

• Leur apparition la plus ancienne est attestée au sein du PPNB récent à Tell Halula (Euphrate moyen - Syrie) où, pendant le Neolithique, ces faucilles deviennent prédominantes en comparaison avec les lames-faucilles à poli parallèle au bord (Ivanez et all. 1998: 133-134). Synchroniquement, les lames avec usure en diagonale apparaissent à tells Assouad et Sabi Abyad II (Syrie) (Cauvin 1983, fig.6; Copeland, Verhoeven 1996). Pour les autres sites du Proche Orient, à part de nombreuses attestations de la présence de faucilles, il n’y a pas assez d’information détaillée pour permettre de reconnaître le type d’emmanchement et le mode d’emploi des outils.

• Les recherches sur les outillages lithiques de la Thessalie et des Sporades (Achilleion, Argissa, Sesklo, Magoula Karamourlar, Dimini, Agios-Petros) révèlent une longue séquence du Néolithique thessalien - de la période précéramique à la période récente (de la moitié du 7ème à la II moitié de 4ème mill. BC) (Moundrea-Agrafioti 1981, 1983; Elster 1977, 1989). Ce que est mis en évidence, c’est la coexistence de deux types d’armatures des faucilles: aux polis longitudinales et obliques, avec la prédominance du premier type, mais sans aucun rapport au rang chronologie/ typologie de faucilles (Moundrea-Agrafioti 1983: 205; Elster 1977: 61). Par contre, à Achilleion, sans précision de la localisation des polis sur les lames usées, ce n’est que la faucille de type Karanovo dont l’utilisation est considéré comme indubitable (Elster 1989: 295-6). On peut, néanmoins, constater sur la base des illustrations qu’il y a des pièces avec usure parallèle (Elster 1989, Fig. 10.3, 10.4). Pour Argissa préceramique, Prordomos et Otzaki anciens, ce sont sûrtout les éléments de faucilles à poli oblique qui sont attesté et commentées par Tellenbach (Tellenbach 1983: 82-83).

• Du Lavant vers les Balkans les voies ‘d’exode’ de cet outil sont théoriquement ou moins deux: à travers des îles méditerranéennes (Chypre et Crète) et du Péloponnèse au nord et/ou via Anatolie centrale et occidentale et la région de la mer de Marmara. Il est raisonnable de préciser ici que, sans aucune ambition de plonger dans le débat perpétuel de la néolithisation, j’essaie de poursuivre les racines et la distribution d’un type d’outil, dont la valeur utilitaire est très importante dans chaque milieu d’économie sédentaire néolithique. En plus, côté conservatif de la faucille, comme un outil de subsistance, est plutôt un avantage et une prépondérance qu’un défaut dans le contexte des paramètres plus changeants et incertains auxquels se réfèrent parfois les concepts, scénarios et paradigmes variés du puzzle de la néolithisation.

Les fouille récentes dans la grotte Cyclope sur l’ île Gioura (Sporade du nord) ne révèle qu’une lame-faucille provenant du Néolithique ancien (entre 6500-5600 cal BC) (Sampson et all., 1998: 131). Il s’agit, d’après le dessin d’une lame aux polis bilatéraux parallèles (Sampson et al., op.cit., Fig. 6, p. 137).

Le premier itinéraire est plus facile à reconstituer: on commence avec le Chypre où il y a des éléments de faucille recherchés à Shillourokambos pendant le Néolithique précéramique (7600-7500 BC), contemporain du PPNB du Levant. D’après les chercheurs, les outils permettent “...d’approcher les éventuelles parentés des plus anciens agropasteurs de l’île... avec leur souche supposée: Levant et/ ou Anatolie” (Guilaine et all. 1995: 29). A Khirokitia précéramique (8ème mill. BP) ce sont seulement les éléments de faucilles à lustre parallèle, qui ont été mis en évidence (Coqueugniot 1984). Les dernières recherches suggèrent que les premières populations exogènes se sont installées sur l’île vers 9300 BP et leur établissement se date entre 8000 et 7500 cal BC (Peltenburg et all, 2000). A part Shillourokambos, c’est à Kalavasos - Tenta que se retrouvent segments retouchés et microlithes à poli oblique, en même temps avec des lames et lamelles denticulées, dont les polis sont d’habitude parallèles aux tranchants. Les formes parentales du PPNB Chypriote mènent, d’après les chercheurs, vers la Syrie, mais sans une évidence directe sur la zone littorale levantine (Peltenburg et all., op. cit.: 851).

• Sur le lithique de la région plus septentrionale - de la Macédoine grèque - il n’y a pas beaucoup d’information disponible et détaillée. Pour le Néolithique ancien de Nea Nikomedea sont annoncés des artefacts polis, présumant l’utilisation en éléments de faucille de ‘type Karanovo” (Elster 1977; Tellenbach 1983: 83). En ce qui concerne les assemblages lithiques de Giannitsa, ils sont, malheureusement, présentés de la manière assez laconique et leurs photos ne sont pas significatifs relatifs à la présence éventuelle des éléments de faucille (Chrysostomou 1992, 1994). • De la Macédoine ex-yougoslave c’est le site d’Anzabegovo qui est intéressant dans le contexte recherché. Une collection de 128 lames - éléments de faucille- est étudiée et décrite par E. Elster. Il s’agit des lames assez standardisées en dimension, en jaspe rouge, qui sont interprétées toutes comme armatures de faucille de ‘type Karanovo’. Le fait d’utilisation bilatérale et de réavivage des tranchants usés est également mis en évidence. (Elster 1977: 205- 211). Malheureusement il n’y a pas la distinction chronologique des artefacts par périodes, mais il est sûr que certains proviennent de Anza I - Néolithique ancien. Divostin (en Serbie) est le dernier point au nord qui je

• Du Péloponnèse - c’est la séquence représentative de Franchthi qui est étudiée, mais les résultats sont plutôt surprenants: du Mésolithique, il n’existe qu’un éclat lustré sans indication réelle de fauchage des céréales sauvages; dès le Néolithique le plus ancien (~ 6000 bc), aucune des pièces étudiées ne porte le lustre de faucille (Perlès, 42

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large

considère intéressant pour le sujet étudié. Les recherches de Tringham sur l’industrie lithique du site mènent à déterminer 87 pièces usées, avec des poli de Si. Ce sont exclusivement les lames brutes, dont les parties actives obliques suggèrent leur fonction comme éléments de faucille, qui ont été réaffutées en cours de leur l’utilisation. Tringham aussi réfère directement à la faucille connue de Karanovo (Tringham et all., 1988: 219).

• La partie occidentale et nord-occidentale d’Anatolie reste pour l’instant une tache blanche à cause des manques d’information sur les outils agricoles en silex. Mon étude préliminaire sur une partie des matériels de Kumtepe (Troade) montre la présence plutôt des éléments de faucille à poli parallèle. Malheureusement la position chronologique des cette collection lithique n’est pas certaine, mais en tout cas elle n’est pas plus ancienne que de Néolithique récent correspondant à la période Karanovo III-IV (Gabriel 2000: 236; Gurova - étude en cours). Du point de vue de l’industrie lithique, les sites précéramiques de la Turquie nord-occidentale, ne donnent pas beaucoup d’issues favorables pour élucider les voies de la néolithisation des Balkans du sud-est (Gatsov 2000). Néanmoins, les assemblages en silex de Musluçesme et Calca (étudiés d’aspect techno-typologique) montrent, d’après leurs chercheurs, certaines indications pour les concevoir comme “..representing the initial wave of advance of the “Eastern” Neolithic communities” (Özdogan 2000: 168), ou au moins une nouvelle perspective pour le débat de la diffusion du Néolithique de l’est à l’ouest (Özdogan, Gatsov 1998: 223).

• En ce qui concerne la Bulgarie et à part les données déjà présentées (cf. supra), il est raisonnable de mentionner ici l’importance de la connaissance des matériels en silex de Kovacevo- un site qui se trouve dans la vallée de la Struma, une des rivières traçant les voies de la néolithisation des Balkans. Ce site représente une séquence du Néolithique ancien très importante qui s’inscrit à une communauté culturelle avec les sites de la Macédoine grèque (Nea Nicomedea et Jiannitsa) et les sites de l’Ovce Pole en Macédoine ex-yougoslave. Cette communauté précède la culture Karanovo I en Thrace bulgare, mais dans une étape suivante Kovacevo s’apparente au domaine de la culture de Karanovo I (Demoule, Lichardus-Itten 1994; GrebskaKulova 1998; Lichardus-Itten et all. 2000). A Kovacevo il y a une abondance des armatures de faucille aux polis en diagonale, mais aussi les manches concervé en bois de cervidé, ou ces armatures ont été inserrées en épi (Sidera 1998: 225; Gurova - étude en cours sur le lithique). On voit, donc, qu’en Bulgarie il y en a eu de faucille de ‘type Karanovo’ même à une époque un peu antérieure de la période Karanovo I elle-même. La vérification de cette position chronologique sera possible après l’annonce, impatiemment attendu, des dates C14 de Kovacevo. En ce qui concerne la culture Karanovo I et son contexte, l’industrie lithique, provenant des sites assez nombreux qui appartiennent à cette culture, et déjà étudiées et publiée (Gatsov 1993; Gatsov, Kurcatov 1997; Gürova 1997; Gürova, Gatsov 2000).

• Les concepts plus généraux sur la néolithisation de l’Anatolie du nord-oest sont toujours discutables, mais l’idée qui se lance déjà couramment (sans être suffisamment prouvée, faute des lacunes empiriques dans les recherches), décrit la zone littorale de cette région comme point de départ pour la néolithisation par la voie maritime des Balkans de sud-est (Thissen 2000a, b). Quelle que soit l’issue de ce débat, il est indubitable que la néolithisation de la Thrace turque, Tessalie et la Macédoine grecque et ex-yougoslave est un phénomène simultané (Erdogu 2000: 163), dont le Néolithique ancien bulgare est un élément inhérent et consécutif, mais pas encore suffisamment élucidé. La thèse de son origine et diffusion de l’ouest à l’est est de plus en plus admise et porte sur le rôle intermédiaire des vallées des rivières soit Struma et Mesta (Nikolov 1989a, b); soit Struma, Vardar, Morava (Todorova, Vaisov 1983); soit via la connexion Anza-la vallée de la Struma septentrionalela pleine de Sofia- Thrace) (Thissen 2000a: 195).

Quant au deuxième chemin envisagé de la diffusion des faucille en épi - à travers d’Anatolie méridionale, centrale et nord/nord-ouest- ce chemin est beaucoup plus difficile à reconstituer à cause des lacunes existantes dans les recherches et dans les données disponibles, et particulièrement relatif aux données sur les aspects fonctionnels de l’industrie lithique.

• Dans le cadre de la koiné culturelle Balkano-Anatolienne, il est assez intéressant de connaître l’aspect fonctionnel de l’industrie lithique des sites de la culture Fikirtepe (Fikirtepe, Pendik, Yarimgurgaz), qui s’étend sur le vaste territoire autour de la mer de Marmara et représentante une zone intermédiaire entre Anatolie nord-occidentale et la Thrace bulgare. Cette culture est considérée par certains chercheurs comme très significative afin de déterminer la diffusion initiale des éléments néolithiques vers l’Europe (Özdogan 1999a: 212). Dans ce sens un fort accent est mis sur le site Hoca Cesme comme une colonie anatolienne sur le littorale egéen du nord et ergo: comme pointmediateur des influences culturelles vers la Thrace bulgare (Ozdogan 1997: 26-27). Une autre point de vue tend à marginaliser le rôle des régions étroitement adjacentes de la mer de Marmara et Thrace turque dans la néolithisation de l’Europe de sud-est (Thissen 2000a: 193). Malgré ces perspectives contradictoires, il est important de connaître les aspects fonctionnels des assemblages en silex provenant de la région concernée. Dans ce sens il faut souligner

• D’Anatolie centrale ce sont les données sur les phases acéramiques d’Asikli, Catalhöyük, Hacilar et Suberde sur lesquelles on peut compter, mais, c’est seulement le dernier site d’ou provient un grand nombre de lames en silex lustrées, représentant “...une parenté avec le matériel sur le Moyen Euphrate et Levant du PPNB”. Il n’est pas pourtant précisé de quel type de poli (respectivement de faucille) s’agit- il au sein de l’assemblage lithique de ce bizarre “village de chasseurs”, ou la présence des lamesfaucille est abondante, mais l’absence de céréales identifiées est aussi indubitable (Nur Balkan 1994: 125). D’une période plus récente il faut mentionner la faucille de Hacilar (VI-II) en andouillers de cervidés avec l’insertion parallèle au manche courbe. Sous une forme fragmentée, ce type de faucille est également attesté à Catalhöyük (VI-V) et il s’approche plus au moins au type recherché (Cauvin 1983: 72, Fig. 5). 43

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

l’importance des recherches en cours de ma collègue turque Güner Coskunsu qui travaille sur la tracéologie des outillages lithiques de la région de la Thrace turque. D’après la communication personnelle de sa part, les éléments de faucille de Hoca Cesme consistent en pièces peu nombreuses, surtout éclats assez amorphes. Il est également intéressant de mettre en évidence les outils agricoles des sites comme Keçiçayiri précéramique (?) de la région de Eskisehir - la zone envisageable pour l’infiltration culturelle avec des habitants du plateau central (précisément de la pleine de Konya) et ergo - une issue éventuelle vers l’origine des populations néolithiques d’Anatolie du nordouest et des Balkans du sud-est (Efe 2000: 172).

de la culture ‘Varna’ (Bulgarie du nord-est) et même parmi certaines collections néolithiques de la Bulgarie occidentale (Skakun 1992, 1993a, 1993b, 1999). Les artefacts publiés n’ont pas toujours de position stratigraphique précise et ne peuvent être interprétés que contextuellement: le matériel chalcolithique date du Ve mill. BC, le matériel néolithique de la fin de VI mill. BC (dates C14 calibrées) (Boyadziev 1995). Les études que j’ai effectué récemment sur les assemblages néolithiques bulgares m’ont également permis d’identifier certains éléments de tribulum préhistoriques (très peu pour l’instant) qui s’inscrivent dans le même cadre chronologique, mais diffèrent sensiblement des exemplaires décrits par ma devancière.

Comme remarque finale et surtout, étant donné l’état de recherches, on peut noter que, à part les sites du Néolithique ancien de la Thrace bulgare, c’est seulement à Ilipinar ancien, ou la présence des éléments de faucilles à manche courbe et insertion oblique est totalement prédominante - sans les formes plus simples précédentes et même simultanées, envisageant un autre type de faucille (Gurova 2001 c). Sur la base de cette ressemblance des faucilles et de la contemporanéité d’Ilipinar ancien (phases X, IX) et de Karanovo I, attestée par les dates publiées - C14 calibrées (Boyadziev 1995: 179; Gördsdorf 1997: 379; Roodenberg et all. 1989/90: 76; Roodenberg 1995: 172; Roodenberg 1999: 200; Thissen 1999: 31), on peut supposer des connexions ou une adoption indépendante et simultané au niveau de leurs prédécesseurs, dont l’identification reste toujours énigmatique. Dans ce sens une étude éventuelle fonctionnelle sur le lithique de Mentese, dont la séquence précède celle d’Ilipinar, doit avoir, présumons-nous, une valeur importante.

En général les éléments de tribulum archéologiques et récents représentent une variété morphologiques bien marquée - des lames ‘cananéennes’ (Syrie-Palestine) à travers des éclats laminaires (Crète, Chypre, Grèce, Turquie, Bulgarie) aux éclats assez amorphes qu’on retrouve parfois en Turquie et Espagne. Certaines démarches et publications scientifiques portent sur l’ethnographie et comportent des descriptions technologiques afin de révéler les traditions et le savoir-faire très élaborés de cette production artisanale (Fox 1994; Whittaker 1996). Il y a aussi des travaux dont l’accent est mis sur le problème des critères diagnostiques et fiables pour identifier les éléments de tribulum et pour les distinguer des éléments de faucille (Ataman 1999; Fujii 1986; Kardulias, Yerkes 1996; Wallon 1978; Yerkes, Kardulias 1994). Souvent ce problème est présenté et discuté dans un contexte archéologique des pratiques agricoles où, inévitablement, les éléments de tribulum se comparent et s’opposent aux éléments de faucille (Anderson 1994, 1999; Anderson-Inizan 1994; Skakun 1994a; Skakun 1999).

Une conclusion s’impose de soi-même: les faucilles connues comme ‘de type Karanovo’ sont inévitablement présentes et prédominantes sur certains sites du Néolithique ancien (et surtout sur les Balkans), ou la cultivation et la domestication se représentent toutes faites et accomplies sans évidence des formes prédécesseurs locales. A l’état actuel des recherches et connaissances, il n’est pas possible de reconstituer le développement de cet outil composite au sein des assemblages lithiques des sites qui possèdent des longues séquences évolutives, basées sur un substrat indigène. La confusion dans les cas pareils consiste en difficulté de distinguer les paramètres des processus comme acculturation, adaptation ou diffusion éventuelle entre les communautés autochtones/immigrantes, un fait déjà constaté et discuté dans la littérature sur les Balkans (Kozlowski 1982; Ursulesku 1998). Et cette constatation est valable pour les régions tant éloignées que marginales (y compris voisines) de la zone nucléaire de la néolithisation primaire.

D’après les descriptions des éléments archéologiques provenant de la Bulgarie et identifiés par N. Skakun il s’agit d’habitude de grandes lames fracturées des dimensions standardisées (4-4,5 / 2-2,5 cm) dont les traces d’utilisation sont localisées dans les zones angulaires des tranchants, mais ne sont pas identiques aux traces des éléments de faucilles (Skakun 1992: 294; Skakun 1999: 202-205). Le problème assez nuancé (parfois même sophistiqué) d’identification et de la distinction des éléments de tribulum/faucille m’a stimulé d’acquérir un échantillon d’éléments de tribulum ethnographiques en qualité d’outils référentiels. Il s’agit d’artefacts que j’ai collectionné des villages Karanovo, Kapitan Dimitrievo et d’une collection privée de I. Shopov (professeur d’histoire, passionné par la préhistoire bulgare), provenant de 2 régions - Dobrudja et Rhodopes, qui m’était donnée pour mon étude de ma collègue S. Ivanova. J’ai pu enrichir mes observations et expérience personnel grace à mon travail et l’expérimentation agricole à Sophia Antipolis (France), dirigés par P. Anderson et mon voyage/prospection en Espagne. Les résultats de cette recherche, présentés au colloque tracéologique international à St.-Petersbourg, sont déjà publiés (Gurova 2001a). Ce qui mérite d’être noté (rappelé) dans le cas concret ce sont les points suivants:

ELÉMENTS DE TRIBULUM Un autre outil agricole qui, depuis plus de 20 ans, attire l’attention des spécialistes et auquel sont consacrées des recherches détaillées - c’est le tribulum (Fig. 5). N. Skakun, étudiant des collections lithiques préhistoriques de la Bulgarie, a identifié les éléments de tribulum parmi les assemblages chalcolithiques en silex provenant de la région

• Tous les éléments analysés de tribulums bulgares ethnographiques possèdent les paramètres dimensionnels assez similaires, qui sont d’autre coté très proches (avec des variations de 3,4 mm) des éléments de tribulum publiés 44

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large

Figure 5. Tribulum et mode d’emploi: 1/ tribulum ethnographique de la Bulgarie (d’après H.Vakarelski, 1974, p. 134); 2/ battage traditionnel en Bulgarie - XIX siècle (d’après N. Skakun, 1992, p. 295); 3/ tribulum privé, village Karanovo (photo M. Gurova); 4/ tribulum abandonné dans le village Tramacastillia (Espagne, photo - M. Gurova).

de Malloura Valley (Chypre) et de la Turquie (Ataman 1999; Kardulias,Yerkes 1996). Morphologiquement et typologiquement il s’agit des pièces soigneusement façonnées aux retouches bilatérales, bifaciales et transversales sur éclats laminaires. Dans certains cas isolés il est probable que les supports soient des lames (Gurova 2001a, Fig. 4-8). Ce façonnage résulte d’une démarche technologique intentionnel afin d’accommoder l’insertion des artefacts aux petites rainures individuelles sur la planche en bois et d’améliorer l’efficacité des tranchants de cet instrument composite. Morphologiquement et comme traces d’utilisation les éléments de tribulum bulgares sont très semblables aux collections ethnographiques et historiques provenant de la Grèce et de la Chypre et publiées par l’équipe Yerkes, Kardulias (Kardulias, Yerkes 1996; Yerkes, Kardulias 1994). A cette koiné on peut également inscrire certaines pièces de Crète et Anatolie (collection de P. Anderson).

• Visuellement les parties usées des éléments de tribulum sont bien marquées - la zone polie s’étend parallèlement au bord sur les 2 surfaces jusqu’à 1,0 cm du fil de tranchant. Certaines pièces possèdent des traces d’utilisation bilatérales, les zones polies s’approchent au point de donner un aspect de lustre bifacial total. Les arêtes de tranchants varient de légèrement ébréchés jusqu’au fortement aplatis et arrondis. Le poli des éléments de tribulum est visuellement frappant, mais l’observation microscopique des zones polies révèle un microrelief assez varié - avec beaucoup de microcavités sous forme de taches noires à dimensions et configurations différentes (y compris la forme des ‘comètes’) et contours assez vagues. Le contraste entre les zones lustrées et sombres n’est pas net. La texture du poli ne possède pas l’aspect lisse, brillant et ‘miroir’ attribuable aux éléments de faucille et aux certaines lames ‘cananéennes’ de Proche Orient (Anderson 1994b, Anderson-Inizan 1994, Chabot sous presse). Les stigmates 45

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

caractéristiques des artefacts décrits (aussi bien archéologiques qu’ethnographiques). Ce fait est assez significatif et convainquant en ce qui concerne la conclusion autant qu’il s’agit d’une catégorie de pièces agricoles dont les racines sont très anciennes et dont la tradition et le savoirfaire de production sont exclusivement solides et inchangeables. Évidemment le tribulum dans le contexte local peut être considéré comme un type d’outil agricole qui, par rapport à la faucille de ‘type Karanovo’, se manifeste dans une étape préhistorique plus tardive (datant du Néolithique récent), mais sous une forme et construction complètement évoluées comme, à propos, dans le cas de cette faucille.

linéaires sont très marqués, sous forme de dépressions longitudinales, larges et profondes et de ‘comètes’ dont les ‘queues’ sillonnent la texture rugueuse des zones polies (Gurova 2001a, Fig. 11, 12). • En ce qui concerne l’identification des éléments des tribulum parmi les outillages néolithiques étudiés par moi - ils sont très peu - 4 ex. Au sein des assemblages de la culture du Néolithique ancien (Karanovo I, II) les spécimens pareils ne sont pas attestés et en effet je n’attends pas de les trouver pendant une étape si ancienne du Néolithique (Gürova 1997; Gürova, Gatsov 2000). Parmi le matériel de la séquence ultérieure du tell Karanovo il y a un éclat retouché avec des traces d’utilisation peu développées, mais assez diagnostique pour une fonction comme élément de tribulum. Cette pièce provient d’un contexte bien stratifié de la période Karanovo III (Néolithique récent), datée entre 5500 -5280 BC (dates C14 calibrées) (Görsdorf 1997: 379; Gürova - sous presse). Un artefact provient du site Pomoshtitsa (Dobrudja) qui révèle une séquence néolithique assez longue (ici je tiens à remercier prof. V. Nikolov, qui m’a permis de mettre en évidence cette pièce non publié). Cet éclat retouché possède des microtraces d’usure très caractéristiques et tout à fait identique à ceux des pièces ethnographiques de tribulum (Fig. 7). Contextuellement l’artefact s’associée à la période récente du Néolithique (la fin du VI mill. BC) - matériel non publie). Deux spécimens avec des traces bien significatives, mais sans position stratigraphique précise proviennent du dernier horizon du site néolithique de Drama - Gerena in Thrace (Gurova - étude à publier).

Adresse de l’auteur Dr. Maria GUROVA Département de Préhistoire Institut d’Archéologie, auprès de l’Académie Bulgare des Sciences 2, rue Saborna 1000 Sofia, BULGARIE E-mail : [email protected] Bibliographie ANDERSON-GERFAUD, P., 1988, Using prehistoric stone tools to harvest cultivated wild cereals: preliminary observations of traces and impact. In Industries Lithiques. Traceologie et Technologie , V. 1 aspects archéologiques, edited by S. Beyries. Oxford: BAR International Series 411, p. 175-195.

Malgré cette quantité minime des pièces archéologiques, sur la base de tout ce qu’était mentionné ci-dessus, on peut se permettre de faire certaines remarques de conclusion qui me semblent raisonnables et vérifiables:

ANDERSON, P., 1994a, Insights into plant harvesting and other activities at Hatoula, as revealed by microscopic functional analysis of selected chipped stone tools. In Le gisement de Hatoula en Judée Occidentale, Israel, edited by. M. Lechevallier, A. Ronen. Paris, p. 277-193.

• Ces artefacts préhistoriques représentent des éclats retouchés assez semblables aux pièces ethnographiques et possèdent des microtraces d’utilisation identiques aux derniers (Fig. 6, 7). Leur très petit nombre est un peu surprenant, mais assez explicable étant donné le fait que les outillages lithiques du Néolithiques récents restent encore insuffisamment étudiés, une lacune que j’espère récompenser dans un futur envisageable.

Anderson, P., 1994b, Interpreting traces of Near Eastern Neolithic craft activities: ancestor of the threshing sladge for processing domestic crops?. Helinium XXXIV/2, p. 306-321. ANDERSON, P., 1999, Experimental Cultivation. Harvest and Threshing of Wild Cereals: Their Relevance for Interpreting the Use of Epipalaeolithic and Neolithic artifacts. In Prehistory of Agriculture. New Experimental and Ethnographic Approaches, edited by P. Anderson. Los Angeles: University of California, Monogpaph 40, p. 118-144.

• Les éléments de tribulum chalcolithiques bulgares décrits et illustrés par N. Skakun diffèrent morphologiquement des pièces ethnographiques et des spécimens exemplaires provenant du Néolithique et identifiés par moi. Ergo - la question qui apparaît est si cette différence est due à l’une certaine transformation chrono-culturelle qui mène à la prédominance des lames fracturées, utilisées en éléments de tribulum pendant le Chalcolithique dans la Bulgarie du nord? Et si une telle grave modification a réellement eu lieu - de chercher et trouver une explication admissible. L’autre possibilité (plus probable selon moi) est que dans les descriptions de Skakun il s’agit en effet des éléments de faucilles, pas de tribulum.

ATAMAN, K., 1999, Threshing Sladge and Archaeology. In : Prehistory of Agriculture. New Experimental and Ethnographic Approaches, edited by P. Anderson. Los Angeles: University of California, Monogpaph 40, p. 211-222.

Parmi les outillages chalcolithiques provenant du site Drama - Merdzumekia (Gurova - étude en cours) il y a une série d’éléments de tribulum, complètement correspondant aux

CAUVIN, M.-C., 1983, Les faucilles préhistoriques du ProcheOrient. Données morphologiques et fonctionnelles. Paléorient 9/1, p. 63-79.

ANDERSON, P. & INIZAN, M.-L., 1994, Utilisation du tribulum au debut du IIIe millénaire: des lames “cananéennes” lustrées à Kutan (Ninive V) dans la région de Mossoul, Iraq- Paléorient, 20/2, p. 85-103.

BALKAN-ATLI, N., 1994, La neolithisation de l’Anatolie Paris: De Boccard. BOYADZIEV, Y., 1995, Chronology of Prehistoric Cultures in Bulgaria. In Prehistoric Bulgaria, edited by D. Bailey & I. Panayotov. Madison Wiskonsin: Monographs in World Archaeology N 22, p. 149-191.

46

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large

Figure 6. Élément de tribulum ethnographique (de la collection de Shopov) et microphotos des traces d’usure (x100). 47

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 7. Élément de tribulum préhistorique (du site Pomoshtitsa) et microphotos des traces d’usure (x100).

48

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large typologische Charakteristik. In Karanovo, I. Die Ausgrabungen im Südsektor 1984-1992, edited by S. Hiller & V. Nikolov. Salzburg-Sofia, p. 213-227.

CHABOT, J. in print, Tell’Atij et Tell Gudeda: les artefacts en pierre taillée. Québec: Presse de l’Université Laval. CHRYSOSTOMOU, P., 1992, O neolithikos oikismos ton Giannitson B (The Neolithic settlement of Giannitsa B). To Arxaiologiko Ergo ste Makedonia kai Trake 3 (1989), p. 119-134.

GEORGIEV, G., 1958, Za niakoi oradija na proizvodstvo ot neolita i eneolita v Balgarija. V: Izsledvanija v chest na akad. D. Dechev po sluchai 80-godishninata mu. Sofija: Izdatelstvo na BAN, p. 369-387 (en bulgare).

CHRYSOSTOMOU, P., 1994, Oi neolithikes ereunes sten pole kaieparxia Giannitson kata to 1991 (Neoliyhic excavation in the city and the area of Giannitsa in 1991). To Arxaiologiko Ergo ste Makedonia kai Trake 5 (1991), p. 111-124.

GÖRSDORF, J., 1997, C 14 - Altersbestimmungen. - In Karanovo, I, Die Ausgrabungen im Sudsektor 1984-1992 , edited by S. Hiller & V. Nikolov. Salzburg-Sofia, p. 377-384.

COPELAND, L. & VERHOEVEN M., 1996, Bitumen-Coated Sickle-Blade Elements at Tell Sabi Abyad II, Northern Syria 1996. In Neolithic Chipped Stone Industruie of the Fertile Crescent, and Their Contemporaries in Adjacent Regions. Studies in Early Near Eastern production, and Environment 3, edited by S. Kozlovski & H. Gebel. Berlin, p. 327-330.

GÖRSDORF, J. & BOJADZIEV, J., 1996, Zur absoluten Chronologie der bulgarischen Urgeschichte. Eurasia Antiqua 2, p. 105-173. GREBSKA-KULOVA, M., 1998, Néolithiuque ancien dans la moyenne vallée de la Struma (Strymon), Bulgarie sud-occidentale. In Préhistoiore de l’Anatolie, Genèse de deux mondes, edited by M.Otte, ERAUL 85. Liège, p. 143-159.

COQUEUGNIOT, E., 1984, Premiers éléments concernant l’utilisation des outils de silex de Khirokitia (Chypre). Campagne de 1981. In Fouilles récentes à Khirokitia (Chypre) 1977-1981. Tome 1: texte, edited by A. Le Brun. Paris: Mémoire n 41, p. 1984, 89-95.

GUILAINE, J., BRIOIS, F., COULAROU, J., & CARRÈRE, I., 1995, L’établissement néolithique de Shillourokambos (Parekklisha, Chypre). Premiers résutats. Report of the Department of Antiquities. Nikosia, p. 11-32.

DEMOULE, J.-P., & LICHARDUS-ITTEN, M., 1994, Fouilles franco-bulgares du site Néolithique ancien Kovacevo (Bulgariedu sud-ouest). Bulletin de correspondance hellénique (BCH) 118, p. 561-618.

GÜROVA, M., 1997, Gebrauchsspurenanalyse des neolithischen Feuersteininventars. In Karanovo, I. Die Ausgrabungen im Südsektor 1984-1992, edited by S. Hiller & V. Nikolov. SalzburgSofia, p. 363-375.

DOCEVA, E., 1992, Rastitelni ostanki ot zilisteto. In V. Nikolov, Rannoneolitno ziliste ot Slatina (Sofia). Sofia: Razkopki i proucvanija 25, p. 144-152 (en bulgare).

GUROVA, M., 2001a, Eléments de tribulum de la Bulgarie références ethnographiques et contexte préhistorique. Archaeologia Bulgarica, V, p. 1 -19.

EFE, T., 2000, Recent investigation in Inland Northwestern Anatolia and Its Contribution to Early Balkan-Anatolian Connections. In Karanovo, III. Beiträge zum Neolithikum in Südosteuropa, edited by S. Hiller & V. Nikolov. Wien, p. 171-183.

GUROVA, M., 2001b, Funktsionalen analiz na kremachen ansambal ot selishtna mogila Kapitan Dimitrievo.- Arheologia, p. 38-47 (en bulgare).

ELSTER, E., 1977, Neolithic technology: a Case Study in Lirthic Analysis for Old Europe, 6500-4000 BC. Ph.D dissertation. Los Angeles: Univesity of California.

GUROVA, M. , 2001 c, Analyse fonctionnelle des assemblages en silex d’Ilipinar , phases X et IX. In The Ilipinar Excavation II, edited by J. Roodenberg & L. Thissen, Pihans 93. Leiden, p. 297-325.

ELSTER, E., 1989, The Chipped Stone industry. In Gimbutas M., Sh. Winn, D. Shimabuku. Achilleion, a Neolithic Settlement in Thessaly, Greece, 6400-5600 BC. Los Angeles: Monumenta archaeologica 14, p. 273-306.

GÜROVA, M., sous presse, Feursteininventar aus Sondage O 19 in Tell Karanovo - typologische und funktionale Analyse. In Karanovo, II. Ausgrabungen in Sondage O19, edited by S. Hiller & V. Nikolov. Salzburg - Sofia.

ERDOGU, B., 2000, The Problems of Dating Prehistoric Axe factories and Neolithisation in Turkish Thrace. In Documenta Praehistorica XXVII, edited by M. Budja, Ljubljana, p. 155166.

GÜROVA, M., & GATSOV, I., 2000, Research Problems of the Early Neolithic Flint Assemblages from Thrace (Bulgaria). In Karanovo, III. Beiträge zum Neolithikum in Südosteuropa, edited by S. Hiller & V. Nikolov. Wien, p. 155-163.

FOX, W., 1984, Dhoukani Flint Blade Production in Cyprus. Lithic Technology, 13, p. 62-67.

HELMER, D., 1983, Faucilles et gestes de la moisson . In Traces d’utilisation sur les outils néolithiques du Proche Orient, edited by M.-C. Cauvin, TMO N5. Lyon, p. 189-198.

FUJI, S., 1986, Criteria for the Identification of Threshing Sledge Blade. Bulletin of the Okayama Orient Museum 5, p. 1-34.

IBÁNEZ, J., GONZÁLEZ, J., PALOMO, A., & FERRER, A., 1998, Pre-Pottery Neolithic A and Pre-Pottery Neolithic B Lithic Agricultural Tools on the Middle Euphrates: The Sites of Tell Mureybit and Tell Halula. In The Origins of Agriculture and Crop Domestication, edited by A. Damania, J.Valkoun, G.Wilcox, C.Qualset. Aleppo, p. 132-144.

GABRIEL, U., 2000, Mitteilung zum Stand der Neolithikumsforschung in der Umgebung von Troia (Kumtepe 1993-1995; Besik-Sivritepe 1983-1984,1987,1998-1999). Studia Troica 10, p. 233-238. GASSIN B. 1996. Evolution socio-économique dans le Chasséen de la grotte de l’Eglise supérieure (Var). Apport de l’analyse fonctionnelle des industries lithiques. Paris: CNRS Editions.

JENSEN, H., 1993, Flint Tools and Plant Working. Hidden Traces of Stone Age Technology. A use-waer study of some Danish Mesolithic and TRB implements. Aarhus: Aarhus University Press.

GATSOV, I., 1993, Neolithic Chipped Stone Industries in Western Bulgaria. Krakow: Jagellonoan University Press. GATSOV, I., 2000, Chipped Stone Assemblages from South Bulgaria and North-West Turkey (Epipalaeolithic/Mesolithic and Neolithic). In Technology, Style and Society. Contributions to the Innovations between the Alps and the Black Sea in Prehistory, edited by L. Nikolova. Oxford: BAR International Series 854, p. 1-28.

KARDULIAS, N. & YERKES, R., 1996, Microwear and Metric Analysis of Threshing Sledge Flints from Greece and Cyprus. Journal of Archaeological Science, 23, p. 657 -666.

GATSOV, I. & KURCATOV, V., 1997, Neolithische Feuersteinartefakte. Mineralogische Untersuchung und technisch-

KOROBKOVA, G., 1993, La différenciation des outils de moisson d’après les données archéologiques, l’étude des traces et

KOROBKOVA, G., 1978, Drevneishie zhatvennie orudija i ih proizvoditelnost (v svete eksperimentalno-trasologicheskogo izuchenija). Sovetskaja Arheologija, 4, p. 36-52 (en russe).

49

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe POPOVA, C., 1995, Plant Remains from Bulgarian Prehistory (7000-2000 BC). In Prehistoric Bulgaria, edited by D. Bailey & I. Panayotov. Madison Wiskonsin: Monographs in World Archaeology N 22, p. 193-205.

l’expérimentation. In Traces et fonctions: les gestes retrouvés, edited by P. Anderson, S. Beyries, M. Otte, H. Plisson, ERAUL 50/ 2. Liege, p. 369-382. KOZLOWSKI., J., 1982, La néolithisqtion de ls zone balkanodanubiemme du point de vue des ondustries lithiques. In Origin of the Chipped Stone Industries of the Early Farming Cultures in Balkans, edited by J.K. Kozlowski. Warszawa-Krakow: Panstwowe Wydawnistwo Naukowe, p. 131-170.

ROODENBERG, J., 1995, Chronology and conclusions. In The Ilipinar Excavations I, Five Seasons of Fieldwork in NW Anatolia, 1987-91, edited by J. Roodenberg. Istanbul, p. 167-174. ROODENBERG, J., 1999, Ilipinar, An Early Farming village in the Iznik Lake Basin. In Neolithic in Turkey. The Cradle of Civilization, edited by M. Özdogan. Istanbul, p. 193-202.

LICHARDUS-ITTEN, M., DEMOULE, J.-P., PERNICEVA, L., GREBSKA-KULOVA, M., & KULOV, I., 2000, Zur bemalten Keramik aus der fruhneolithischen Siedlung von Kovacevo (SWBulgarien). In Karanovo, III. Beiträge zum Neolithikum in Südosteuropa, edited by S. Hiller & V. Nikolov. Wien, p. 27-50.

ROODENBERG , J., THISSEN, L., & BUITENHUIS, H., 1989/ 90, Preliminary report on the archaeological investigations at Ilipinar in NW Anatolia, Anatolica XVI, 1989/90, p. 61-144.

LISITSINA, G. & FILIPOVICH, L., 1980, Paleoetnobotanocheskie nafodki na Balkanskom poluostrove. Studia Praehistorica, 4, Sofia, p. 5-90.

SAMPSON, A., KOZLOWSKI, J., & KACZANOWSKA, M., 1998, Entre l’Anatolie et des Balkans: une séquence mésolithique de l’ile de Gioura (Sporades du Nord). In Préhistoiore de l’Anatolie, Genèse de deux mondes. ERAUL, 85, p. 125-141.

MARINOVA, E., 1999, Arheobotanichno izsledvane na ranno- i kasnoneolitni materiali. In Selishtna mogila Kapitan Dimitrievo. Razkopki 1998-1999, edited by V. Nilolov. Sofia -Peshtera, p. 123-131 (en bulgare).

SIDERA, I., 1998, Nouveaux éléments d’origine Proche-Orientale dans le Nélithique ancien Balkanique: Analyse de l’industrie osseuse. In Préhistoiore de l’Anatolie, Genèse de deux mondes, edited by M.Otte, ERAUL 85. Liège, p. 215-239.

MOUNDREA-AGRAFIOTI, A., 1981, La Thessalie du sud-est au Néolithique: outillage lithique et osseux. Thèse de doctorat, Université de Paris X.

SKAKUN, N., 1992, Evolution des techniques agricoles in Bulgarie chalcolithique (d’après les analyses tracéologiques). In Préhistoire de l’agriculture: nouvelles approches expérimentales et ethnographiques, edited by P. Anderson, Monographie de CRA, 6. Paris, p. 289-303.

MOUNDREA-AGRAFIOTI, A., 1983, Pièces lustrées du Néolithique thessalien: essai de classement. In Traces d’utilisation sur les outils néolithiques du Proche Orient, edited by M.-C. Cauvin. TMO N5, Lyon, p. 199-207.

SKAKUN, N., 1993a, Results of traseological Examination of flint implements from neolithic settlements in Western Bulgaria. In I. Gatsov. Neolithic Chipped Stone Industries in Western Bulgaria. Krakow, p. 52-54

NIKOLOV, V., 1989a, Das Flusstal der Struma als Teil der Strasse von Anantolien nach Mitteleuropa. In Neolithic of Southeastern Europe and its Near Eastern Connexions. Varia Archaeologica Hungarica II, p. 191-199. NIKOLOV, V., 1989b, Zu einigen Aspecten der Kultur Karanovo I.- In Tell Karanovo und das Balkan-Neolithikum. Salzburg, p. 29-30.

SKAKUN, N. 1993b, Agricultural implements in the Neolithic and Eneolithic cultures of Bulgaria. - In Traces et fonctions: les gestes retrouvés, edited by P. Anderson, S. Beyries, M. Otte, H. Plisson, ERAUL 50/ 2. Liege, p. 361-368.

ÖZDOGAN, M., 1997, The beginning of neolithic aconomies in Southeastern Europe: an Atatolian perspective.- Journal of European Archaeology, 5/2, p. 1-33.

SKAKUN, N., 1994a, Agricultural implements and the problem of spreading of agriculture in Southeastern Europe. Helinium, XXXIV/2, p. 294-305.

ÖZDOGAN, M., 1999a, Nortwestern Turkey:Neolithic cultures in Between the Balkans and Anatolia. In Neolithic in Turkey. The Cradle of Civilization, edited by M. Özdogan. Istanbul, p. 202224.

SKAKUN, N., 1994b, Rezultati issledovanija proizvodstvennogo inventarija neoliticeskogo poselenija Usoe I (Bolgarija). In Eksperimentalno-trasologiceskie issledovanija, edited by G. Korobkova. St.-Petersbourg, p. 85-118 (en russe).

ÖZDOGAN, M., 2000, The Appearance of Early neolithic cultures in Northwestern Turkey. Some Problems. In Karanovo, III. Beiträge zum Neolithikum in Südosteuropa, edited by S. Hiller & V. Nikolov. Wien, p. 165-170.

SKAKUN, N., 1999, Evolution of Agricultural Techniques in Eneolithic (Chalkolithic) Bulgaria. Data from Use-wear Analisis. In Prehistory of Agriculture. New Experimental and Ethnographic Approaches, edited by P. Anderson. Los Angeles: University of California , Monogpaph 40, p. 199-210).

ÖZDOGAN, M. & GATSOV, I., 1998, The Aceramic Neolithic Period in Western Turkey and in Aegean. Anatolica XXIV, p. 209-231.

TELLENBACH, M., 1983, Materialen zum Prakeramischen Neolithicum in Sud-Ost-Europa. Typologisch-stratigraphische Untersuchungen zu kithischen Geratschaften, Bericht der Römisch-Germanischen Kommission 64. Mainz/R: Ph. v. Zabern.

PELTENBURG E., COLLEDGE, S., CROFT, P., JACKSON, A., & MCCARTNEY, C., 2000, Agro-pastoralist colonization of Cyprus in the 10th millennium PB: initial assessments. Antiquity, 74/286, p. 844-853.

THANHEISER, U., 1997, Botanische Funde. In Karanovo, I. Die Ausgrabungen im Südsektor 1984-1992, edited by S. Hiller & V. Nikolov. Salzburg-Sofia, p. 429-454.

PERLÈS, C., & VAUGHAN, P., 1983, Pièces lustrées, travail des plantes et moisson à Franchthi (Xème-IVème mill. BC). In Traces d’utilisation sur les outils néolithiques du Proche Orient, edited by M.-C. Cauvin, TMO N5. Lyon, p. 209-229.

THISSEN, L., 1999, Trajectoire Towards the Neolithisation of NW Turkey. In Documenta Praehistorica XXVI, edited by M. Budja. Ljubljana, p. 29-39.

POPOVA, C., 1989, Palaeobotanical studies of the Neolithic and Eneolithic epoch in the territory of Bulgaria. In Palaeoethnobotany and Archaeology. Acta Interdisciplinaria archaeologica, VII. Nitra, p. 269-272.

THISSEN, L, 2000a, A Chronological Framework for the Neolithisation of the Southern Balkan. In Karanovo, III. Beiträge zum Neolithikum in Südosteuropa, edited by S. Hiller & V. Nikolov. Wien, p. 193-212.

POPOVA, C.,1991, Palaeoethnobotanical investigation in South Bulgaria. In Paleoecologia e Arqueologia, II, edited by F. Queiriga, A. Dinis. Vila Nova de Famalicão, p. 187-189.

THISSEN, L., 2000b, Thessaly, Franchti and Western Turkey: Clues to the Neolithisation of Greece? In Documenta Praehistorica XXVII, edited by M. Budja, Ljubljana, p. 141-154. 50

M. Gurova: Outils agricoles des assemblages néolithiques en silex de la Bulgarie - un contexte plus large TODOROVA, H. & VAISOV, I., 1993, Novokamennata epoha v Balgaria (kraja na sedmo-shesto hiljadoletie predi n.e.).Sofia: Nauka i izkustvo.

URSULESCU, N., 1998, Courants culturels d’origine Anatolienne dans le Néolithique Balkano-Carpatique.- In Préhistoiore de l’Anatolie, Genèse de deux mondes, edited by M. Otte, ERAUL 85. Liège, p. 193-213.

TRINGHAM, R., Mc PHERRON, A., GUNN, J. & ODDEL, G., 1988, The Flaked Stone Industy from Divostin and Banja. In Divostin and the Neolithic of Central Serbia, edited by A. Mc Pherron, D. Srejovic, Pittsbourgh.

VAN GIJN, A., 1990, The wear and tear of flint. Principles of functional analysis applied to Dutch Neolithic assemblages . In Analecta Praehistorica Leidensia 22. Leiden.

UNGER-HAMILTON, R., 1989, The Epi-Palaeolithic Southern Levant and the Origins of Cultivation. Current Anthropology, 30/ 1, p. 88-103.

WHALLON, R.-J., 1978. Threshing Sledge Flints. A Distinctive Pattern of Wear. Paléorient, 4, p. 319-324. WHITTAKER, J., 1996, Athkiajas: A Cypriot Flintknapper and the Threshing Sledge Industry. Lithic Technology, 21/2, p. 108-120.

UNGER-HAMILTON, R., 1999, Harvesting Wild Cereals and Other Plants: Experimental Abservations. In Prehistory of Agriculture. New Experimental and Ethnographic Approaches, edited by P. Anderson. Los Angeles: University of California, Monogpaph 40, p. 199-210.

YERKES, R., & KARDULIAS, N., 1994, Microwear Analysis of Threshing Sledge Flints from Cyprus and Greece: Implication for the Study of Ancient Agriculture. Helinium, XXXIV/2, p. 281- 293.

51

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

52

S. Tiné & A. Traverso: Cakmakli Tepe: An Ancient Neolithic Village in the Aegean Coastland of Northwestern Turkey

CAKMAKLI TEPE: AN ANCIENT NEOLITHIC VILLAGE IN THE AEGEAN COASTLAND OF NORTHWESTERN TURKEY Santo TINÉ & Antonella TRAVERSO

Abstract: This paper presents the first results of the Italian Expedition in Kyme close to Cakmakli Tepe, a village on the Aegean littoral. This research has been developed with the cooperation of the Izmir Museum. An ancient Neolithic village was discovered on the coast, the remains characterized by a particular kind of red pottery with white traces. Below 3 m of alluvial deposits, test pits yielded some sun–dried bricks and few stones from a wall. Résumé : On présente les premières données des recherches conduites par la Missione Archeologica Italiana à Kyme en collaboration avec le Musée d’Izmir, prés du village de Cakmakli sur la côte égéenne de la Turquie. On a découvert ici, pour la première fois prés de la mer, les restes d’un village du Néolithique ancien, caractérisées par des céramiques avec un engobe rouge et traces de peinture blanche. Les sondages ont permis de mettre au jour, au dessous trois mètres de dépôt alluvial, des morceaux de murs de huttes en brique crue avec fondation en pierre.

Until a few years ago (1999), for the publication of an important synthesis of the Turkish Neolithic made by scholars 7 researching in this region, the editor Prof. Mehmet Özdogan stated in his conclusion: “despite the insufficient evidence at present, it is not impossible to surmise the presence of a separate (now merely hypothetical) centre of Neolithisation along the Aegean littoral”.

sustaining enormous silos. The rescue was possible thanks to the Smyrna Museum and to Sebastiana Lagona who directs, for the Catania University, an archaeological mission since twenty years at the Aeolian Kyme. It has been Prof. Lagona who invited us to carry out a series of soundings last October, in collaboration with the Smyrna Museum.

Now, finally, we start to know something about this. It is not much, but we consider it enough in order to suggest that the diffusion of the agricultural culture from the Near East towards West followed, together with the well known route through the Anatolian plateau, another route along the Aegean coasts (Fig. 1).

The site is placed at 7 km west of Aliaga, near the mouth of the river that, during the Neolithic and at the time of Kyme, had a notable flow, but now reduced to a desolate and putrid stream. The deposit above the archaeological level shows clear signs of its formation due to alluvial deposition from the nearby river, here at its mouth. Over two meters of sands and silt have been deposited above a Hellenistic farm. This was one of the various farms of the Kyme kora. Here it is directly overlapping the Neolithic layer.

This is a predictable conclusion and it is strange that the archaeological documentation is lacking west of Mersin, although this region has been intensively frequented by archaeologists, since this is a region where the more important Greek colonies, like Miletus, Ephesus, Pergamum and Kyme, developed.

Even if less important, other floods should have taken place in the prehistory and during the life of the village, during the Early Neolithic. One of these should have destroyed the first village creating, between this and the following one, a silt layer of about 30 cm.

It is near the last city, exploiting the same ecological conditions that the first Neolithic agriculturists established a village. The discovery of the site could be considered some sort of “reparatory gift” from the main antagonist of archaeology: industrialisation which devastated the splendid bay of Kyme, now the location of various industrial complexes, from a cement works to an oil refinery, from a chemical plant to a coal factory.

We founded the remains of the first village, on the bottom of our small sounding (October 2000): a small piece of a plastered floor and few stones delimiting the foundation of a small wall, possibly with an elevation of sun-dried bricks. Of the latter, if ever used, it was not possible to find clear evidence, since the humid conditions of the soil.

Our village is placed actually between the latter structures and the excavation conditions are easily imaginable, working between the chemical miasma and the coal dust. The discovery is due to the digging of deep pits for pillar

The archaeological layer is about 1,10 m thick and it finish above the sterile blackish clay bedrock. Here we are about 3.60 m from the present day surface. 53

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. Map of the Aegean coast of Turkey.

Large flint flakes have been found in the lower layers, immediately above the sterile clay, with pronounced bulbs and no retouches. These have suggested the presence of a Lower Palaeolithic deposit. If this is true, our sounding should have intercepted its periphery. In fact, more numerous flakes, apparently belonging to the Clactonian typology, have been found in the nearby area where, in 1994, the pillar pits have been opened.

periphery of the nuclear area of the impressed pottery. This pottery, as far as we know, goes from Palestine to Cylicia, with its core along the Syrian coasts, at Ugarit. The sherds assemblage is characterised by vessels with plain walls, red slipped with a uniform colour on both surfaces, both internal and external, possibly with a waterproof function. We have mainly vessels with very thin walls, calyx-like shapes, with limited capacity. Actually, the sherds with a thickness suggesting large containers are rare (Fig. 2).

For the moment, we decided to leave open the issue of the presence of a Palaeolithic layer below the Neolithic one, at least since the new excavation in the near future. However, it should be stressed the absolute differences of these tools with those clearly associated with pottery in the whole Neolithic layer. Such a difference is also clear as for the raw material used.

Almost all vessels have a double thin cannon-like handle placed vertically on the shoulder. (Fig. 3) Small ledges placed near the rim are rare. Also these probably had the function to suspend the vessel with a little rope. The bases are rarely flat, more frequently ring or heel shaped.

While the latter are large flakes, small blades obtained from small flint cores of different colours form the typical Neolithic assemblage. These are associated with wares in the layer.

As already said, a large number of sherds have an uniform red slip, often extending also in the internal surface, possibly to secure a better retaining quality.

Only one sherd between the 1000 sherds found in our sounding and the 3000 collected in the 1994 soundings, shows a decoration with notches impressed before firing. On the basis of this meaningful rarity, we should take this sherd as imported or, at least, consider this region as the

A certain number of these vessels show a series of gaps in the slip, with the presence of white spots, often shapeless, but in some cases creating zigzag stripes which unlikely can be considered as fortuitous. 54

S. Tiné & A. Traverso: Cakmakli Tepe: An Ancient Neolithic Village in the Aegean Coastland of Northwestern Turkey

Figure 2. Vessels with plain walls, red slipped surfaces.

Figure 3. Vessels with cannon-like handle. 55

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

We would like to keep open also this issue, waiting for more clear evidence. But we want to stress how white decorations as the one suggested for the Cakmakli wares have been found in various area of Turkey, particularly at Hoca Cesme, a site like our along the Aegean coasts, north of the Dardanelles.

It is important, at the end, to report the presence at Cakmakli of clay disks, similar to those found in many Neolithic sites of the Italian peninsula. There are also two strange flat pebbles with chipped edges similar to those found in the Neolithic site of Passo di Corvo. In that case, these have been interpreted as pick framework used to dig the typical C-shaped ditches of the Tavoliere.

More generally, it is possible to underline the similarity, both in shape and decoration, between our wares and those of the Tessaly pre-Sesklo phase. It is important to do so because it is for this reason that, since 1994, when the colleague Sebastiana Lagona showed us the first photos of these pottery, who is speaking here, although not young anymore, he has been convinced to start this new adventure.

To conclude, the archaeological site is placed near a fishing village at 3 km east; its name Cakmakli means “flint”. Author’s address

Another more important reason is the possibility to verify if the impressed pottery culture extended behind the Cylician coasts, before the great leap towards the Adriatic coasts, Sicily and the western Mediterranean. Also for this second objective, our results are meaningful!

Santo TINÉ Antonella TRAVERSO Istituto Italiano per l’Archeologia Sperimentale via Polleri 3 Genova, ITALY

56

C. Contaxi & H. Stravopodi: La fonction polyvalente des grottes du Sud de la Grèce pendant le Néolithique...

LA FONCTION POLYVALENTE DES GROTTES DU SUD DE LA GRÈCE PENDANT LE NÈOLITHIQUE : LE CAS DES DEUX GROTTES À ALEPOHORI (LACONIE) Chryssanthie CONTAXI & Eleni STRAVOPODI

Résumé : Le projet de Kouveleiki a pour objet la recherche des deux grottes en Grèce du Sud, qui ont été utilisées pendant le Néolithique. Ces grottes font part d’un système des grottes, des abris et des formations karstiques, à une distance de 50 mètres l’une de l’autre. Parmi les trouvailles on reconnaît des vases d’usage domestique, ainsi que des exemples de céramique fine du Néolithique Récent et Final, un grand nombre d’outils en pierre polie et taillée, des figurines et des poids de tissage, des objets divers d’usage spécial, des constructions et des planchers pour des activités quotidiennes en relation avec des restes organiques provenant de la faune et la flore et des restes anthropologiques. Tout cela indique que des groupes de la même population ont occupé une vaste région, les deux grottes et le voisinage. Ces estimations sont confirmées par les analyses chronologiques de C14, les analyses sédimentaires et de paléo- nourriture, ainsi que de DNA des os, qui sont en pleine élaboration. Abstract: Kouveleiki project focuses on the research of a diptych of caves in Southern Greece, used during the Neolithic. The two caves developed as part of a system of caves, rockshelters and karstic formations, in a distance of 50 m. from one another. Rich material culture remains of fine and domestic pottery of Late and Final Neolithic, a large number of polished and chipped stone tools, items of figurines and whorls, various finds of special use, archaeological features and constructions, living floors and working areas in relation with the organic remains of rich faunal and botanic data along with the burials and anthropological material, all indicate groups of the same people occupying a boarder area, the two caves and the nearby vicinity. The above estimation is further confirmed by the C14, sedimentological, paleodietary and DNA analyses still in progress.

Les deux grottes ont été localisées après une recherche de surface réalisée par un groupe scientifique de l’Ephorie de Paleoanthropologie-Spéleologie du Ministère de la Culture de Grèce. Les dépôts archéologiques ainsi que les trouvailles de surface ont suggéré le commencement d’une fouille archéologique.

comprend des vases de forme ouverte ou fermée, mais aussi des jarres pour le stockage des produits.

Les grottes se trouvent aux pentes d’un bloc calcaire, 5 km. au sud du village d’Alepohori, à Laconie, la première étant située à une distance de 50 m. de la seconde.

D’ailleurs, l’industrie en pierre taillée présente un grand nombre de lames à retouche latérale, de grattoirs, de pointes de flèche, mais aussi beaucoup de produits de débitage, prouvant que les pièces finies ont été manufacturées sur place. L’obsidienne a été utilisée comme matière première, sur un tas 90%, tandis que le silex est représenté seulement avec quelques outils finis.

Les meules et les outils de moulage, les restes de graines et les poids de tissage (fuseaux) prouvent que la grotte a été utilisée comme un site des activités domestiques.

Aux fouilles, on a appliqué les méthodes connues d’une recherche archéologique : établissement de kanavos, stratigraphie, plans et photos, tamisage sec et humide, analyses chronologiques et sédimentologiques.

Parmi les trouvailles il faut mentionner trois figurines typiques de l’époque Néolithique qui suggèrent probablement des activités de culte.

KOUVELEIKI A´ La grotte se divise en deux parties, formées à cause d’une chute du plafond. Jusqu’à présent, 24 m2 ont été excavés à la partie extérieure et 6 m2 à l’intérieur de la grotte.

En ce qui concerne les trouvailles anthropologiques, le plus important est l’enterrement d’un enfant dans un vase qui a été posé à l’envers dans un autre, plus large. Tout l’ensemble a été couvert avec un galet. La sépulture contenait de la terre et des pierres.

Le procès des fouilles, qui ont atteint le 1,80 de profondeur, ont relevé quatre couches archéologiques qui présentent une homogénéité en ce qui concerne leur expansion et la distribution des trouvailles.

D’autres restes humains ont également été relevés : un fragment du crâne posé sur un galet de calcaire et quelques os dispersés. Tous ces spécimens témoignent de la variabilité des pratiques mortuaires pendant le Néolithique au Péloponnèse du sud.

Quant à la céramique, elle est d’une excellente qualité, elle représente tous les types caractéristiques de la décoration du Néolithique Récent (black, matte painted, fine gray) et elle 57

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Parmi les restes de la faune, on peut identifier des animaux domestiques et sauvages, tandis que les os des poissons présentent un grand intérêt.

Des analyses de l’obsidienne, réalisées au Centre de Recherches Physiques DIMOKRITOS, ont montré que la matière première des outils provenait de l’île de Mélos, fait qui témoigne que le site faisait part du réseau de la distribution de l’obsidienne à la Méditerranée pendant le Néolithique.

En outre, un foyer entouré de pierres, trouvé à l’intérieur de la grotte, donne des renseignements sur l’économie et la vie quotidienne des occupants de la grotte.

La typologie de la céramique place les deux grottes dans le même environnement culturel avec les grottes de Franchthi en Argolide, Kitsos en Attique, Alepotrypa en Laconie. Le temps d’usage de Kouveleiki A´ et B´ coïncide avec la période pendant laquelle la plupart des grottes ont été occupées, période d’augmentation de la population, d’amélioration et d’expansion de l’agriculture et de l’élevage. Les datations de 14 C varient entre 5210 B.C. et 3362 B.C. pour Kouveleiki A´ et 4490 B.C.-3792 B.C. pour Kouveleiki B´.

KOUVELEIKI B´ La grotte dispose une seule salle des dimensions 25 x 12 m. Les fouilles s’étendent à une surface de 80 m2 et ont relevé trois principales couches préhistoriques, localisées surtout à l’entrée et près de la paroi gauche de la grotte. Malgré quelques problèmes stratigraphiques et chronologiques à l’arrière parti de la salle, dus aux dépôts d’inondations, les couches archéologiques en général révèlent une homogénéité. La céramique présente des vases d’usage domestique surtout, des fragments des jarres à décoration en reliefs, presque uniquement des cordons, tandis que les pièces de céramique fine et décorée sont rares. En général, la céramique, par rapport à celle de la grotte A´, est de pire qualité.

Malgré l’homogénéité des trouvailles des deux grottes, il faut noter une certaine diversion en ce qui concerne la qualité de la céramique et de l’outillage, l’occupation de l’endroit, qui paraît être plus intense à Kouveleiki A´ et la période chronique d’usage, qui est plus longue pour Kouveleiki A´. La recherche se poursuit, en insistant aux analyses sédimentologiques, géochimiques et de paléo-nourriture, ainsi qu’aux analyses chimiques et pétrographiques de la céramique. De même, une recherche de surface à la plaine, qui se trouve en bas des deux grottes, pourrait peut-être localiser un établissement de plein air qui probablement faisait part de ce complexe d’occupation d’espace à l’époque Néolithique.

L’industrie en pierre polie consiste en haches, herminettes, meules et broyeurs, tandis que l’industrie en pierre taillée, presque uniquement en obsidienne, présente des lames, lamelles, pointes, etc. L’absence de grandes pièces corticales indique que les blocs de matière première venaient à un stade de préparation déjà avancé. Toutefois, la présence des produits de débitage montre que les outils ont été manufacturés sur place.

Adresse des auteurs

Quant aux trouvailles anthropologiques, il faut mentionner les deux sépultures ‘ in situ’, dont la deuxième et plus importante : il s’agit du squelette entier d’une femme, d’une trentaine d’années, qui était mis à position raccourcie, dans une fosse peu profonde, creusée sur le substrat de la grotte. Sous la tête et les genoux il y avait deux plaques calcaires. La sépulture était bornée de pierres, une pratique d’enterrement typique au Néolithique.

Chryssanthie CONTAXI Eleni STRAVOPODI Ministère de la Culture Grec Ephorie de Paléoanthropologie-Spéleologie Ardittou 34 b, 116 36 Athènes, GRÈCE Bibliographie KARKANAS, P., 1999. Micromorphological studies in prehistoric caves and rockshelters of Greece. Annual report, American School of Classical Studies at Athens.

DISCUSSION

KONTAXI, CH., KOTJAMBOPOULOU, E., STRAVOPODI, E., 1995, Prokatarktiki ekthesi anaskafon apo tin A’Kouveleiki Spilia Alepohoriou Laconias. Archeologika Annalekta ex Athinon, 22, p. 21-30

Les donnés archéologiques suggèrent que les deux grottes fassent part d’un double cas d’usage contemporain, chronologiquement et fonctionnellement. Elles ont probablement été utilisées pour résidence provisoire et pour des activités domestiques et économiques. Ainsi, les outils de coupage des gibiers et des végétaux, les pièces d’écorchement des peaux et les pointes de flèche témoignent des activités d’agriculture, d’élevage et de chasse. La présence, également, des fuseaux en terre cuite prouve que les activités de filage et de tissage étaient connues. En outre, les grottes ont servi pour des pratiques d’enterrement, suivant le rite très habituel des Néolithiques d’enterrer leurs morts dans la zone d’habitat.

KOUMOUZELI, M., 1989, I keramiki apo tin A’ Kouveleiki Spilia Alepohoriou Laconias. Archeologika Annalekta ex Athinon , 22, p. 143-160 PERLES, C., 1990, L’outillage de pierre taillée néolithique en Grèce : approvisionnement et exploitation des matières premières. Bulletin de Correspondance Héllenique 114, p. 1-42 PAPATHANASSOPOULOS, G., (ed.), 1996, Neolithikos Politismos stin Ellada RENARD, Josette, 1989, Le site Néolithique et Helladique ancien de Kouphovouno (Laconie)

58

C. Contaxi & H. Stravopodi: La fonction polyvalente des grottes du Sud de la Grèce pendant le Néolithique...

Figure 1. Fragments des jarres à décoration plastique, Kouveleiki A’.

Figure 2. Lames en obsidienne, Kouveleiki A’. 59

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 3. Le vase qui contenait les os d’un enfant, Kouveleiki A’.

Figure 4. La sépulture humaine, Kouveleiki B’.

60

R. Striccoli: The Neolithic Settlement of Carrara San Francesco in the Italian South-Eastern Neolithic

THE NEOLITHIC SETTLEMENT OF CARRARA SAN FRANCESCO IN THE ITALIAN SOUTH-EASTERN NEOLITHIC Rodolfo STRICCOLI

Résumé : Le texte qui suive concerne mes fouilles dans l’établissement néolithique de Carrara San Francesco. On a determinés trois différents moments de fréquentation du site. Les phases attestées vont du Néolithique Ancien aux phases finales de la même periode. On a retrouvées sporadiques découvertes attribuables ou Néolithique Moyen. On a retrouvé un fond de cabane et d’un petit muret à sec avec trois dépositions. La déposition retrouvée en 1997, à l’intérieur de la cabane, est attribuable à un individu adulte et a été datée avec le C14 (50904833 B. C.). Abstract: The following text concerns my excavation in the Neolithic settlement of Carrara San Francesco. Three different of settlements were found, from the middle phase of the Ancient Neolithic to the final phase of the same period. There are some findings of the following period that can be dated to the Middle Neolithic. We found the base of a hut with apse and the base of a dry-stone wall. Three burials were also found, the burial founds in 1997 was dated to 5090-4833 B .C., by C14.

The Neolithic settlement of Carrara S. Francesco is on the eastern outskirts of Bisceglie and belongs to an enclosed cultivated area. Since 1995 I have carried out four archaeological excavations, and I intend to speak about the most significant findings from these, with relevant comparisons with other south-eastern Italian settlements of the same period.

to judge from food remnants and pottery remains found with the bones. In number three sector, a stone apse-shaped structure was found; this was probably part of a hut. In number four sector, stones belonging to a fallen down dry-stone wall, were found. On these was a soil deposit that dated back to subsequent period with long interruptions. These findings were extremely encouraging.

So far 350 m² of surface have been investigated. 325 m² of them make up the main area of inquiry, which has yielded the most important archaeological findings; this area will be enlarged in the future (fig. 1). The other 25 m² were the basis for the first stratigraphic test, carried out during the first excavation on a dry area with steppe vegetation. This first test involved a shallow sedimentary deposit (cm 20-25), which was contaminated by previous agricultural works. It contained few findings, predominantly ceramics from various prehistoric and historic periods, and was thus of limited use to our research, the aim of which was to analyse evidence of the earliest human habitation of settlement.

In the 1997 excavations, the second excavation was enlarged along the western and southern side, over an area of 75 m², which was divided into three squares of 25 m². The purpose of this excavation test was to show the entire stone structure with the apse and the dimensions of the dry-stone wall that was on the western side of the previous excavation test. We excavated in the above mentioned squares down to the monumental evidences in situ. The stone structure with apse was brought to light it is the wide bottom of a hut with a rectangular north-south oriented layout. The hut is 6 m long and 5.5 m wide and formed by two contiguous rooms, both apses. The rooms have different entrances. The western room is bigger and it has the entrance along the east side at the beginning of the apse. The eastern room probably had its entrance along the southern side. These two rooms share the central area of flooring, with two large holes in the half that probably large pillars to support the roof. This area is paved with two lines of flat stones. The western area, belonging to the second room, has a lot of stones. The eastern zone, belonging to the second room, has no stones but a small stone structure near the apse that divides the northern area with the apse from the squared area. Structures like this were found in Balsignano (Modugno - Bari) (Radina 1994), Passo di Corvo (Tinè 1983), Lagnano da Piede (Mallory - Tinè 1984), Rendina (Cipolloni Sampò 1982). Probably there was a partition in the western room, where the areas were divided by the northern side of the stone flooring. There were no stones in the area with the apse, which was covered mainly in soil marking a second burial place where we found an adult, in a crouched position.

The second excavation was more encouraging. It was carried out in an area close to that of the previous dig. It showed a deep and was characterised by archaeological humus covered in blooming vegetation. It gave back, from the rocky bottom, (80 cm deep) up, it yielded interesting data and the first evidence of stone structures in situ. Working up from the bottom, we discovered evidence of three different periods of settlement, namely the ancient Neolithic (fig. 2) between the middle and the final periods (Guadone - Rendina II, Lagnano da Piede, Masseria La Quercia - Rendina III) and based on subsequent periods, both prehistoric and historic (facies Passo di Corvo, Serra d’Alto, Diana Bellavista). The best documented period is the third: evidence of this was found in three sectors, including, in sector number 2, the buried bones of an adult. These bones were arranged in a simple ditch. Probably the burial had a religious significance, 61

Figure 1. Carrara San Francesco (Bisceglie - Bari). Planimetry of the excavation with burials (1 – 3).

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

62

R. Striccoli: The Neolithic Settlement of Carrara San Francesco in the Italian South-Eastern Neolithic

Figure 2. Carrara San Francesco (Bisceglie - Bari). Base of the hut and the beginning of the dry-stone wall (excavation 1995-1997).

The burial was west-east oriented and occupied the area of the apse. This burial (fig. 3a), unlike the first, was in a simple ditch in the hut. It had neither funerary outfit nor ritual remains. This is a typical feature of the ancient Neolithic men of the Italian south-east. Both the findings and the C14 analysis on the human bones, carried out at the “Istituto di Fisica delle Particelle” (Politecnico Federale di Zurigo) (ETH 21724) dated the burial and the cultural background to the final ancient Neolithic, i.e. 5090-4833 BC.

the site to our previous discoveries. In 1999 two large areas of the site (110 m² and 45 m²) were excavated. The largest area linked the areas excavated the years before. So that the entire surface we have excavated so far is 325 m², excluding the 25 m² from the first test in 1995, which produced nothing of particular relevance. The excavation we made in 1998 yielded stones from a collapsed wall and parts of palaeo-soil with impressed wares and lithic instruments in the northern sector. These findings, which have been dated to the final period of the ancient Neolithic, were mixed with to animal bones. The flat stones are the remains of an oriented east-west dry-stone wall that joins the wall we found in the test excavation at a distance of 15 m. The presence of stone structures did not allow us to examine the deposit right down to the rocky bottom.

The other main element we found was the beginning of a dry-stone wall defined by two lines of flat stones. The wall is 1-1.5 m wide near the hut’s apse and is east-west oriented. In the north sector too, a short stretch of a dry-stone wall that was next to the hut’s apse, was recovered, but it was northsouth oriented. In 1998 and in 1999 excavations were made in the western area of the settlement. In 1998 the excavated surface was enlarged along the west side. We excavated a 30 m² area next to the previous one. The area was divided in two 15 m² squares. The main purpose was to highlight the dry-stone wall we found next to the hut and to examine the deposit down to the rocky bottom in order to document the earliest period of the settlement, hinted at the few elements we found in 1995. In 1998 the research was extended to another area (10x4 m) that was 15 m from the western side of the main excavation area. In this case the main purpose was to compare the archaeological deposit in another area of

Also in 1998 we found buried human bones which belonged to a 5-7 year old child. The age was deduced from the presence of permanent milk teeth in the jaw bone. The burial area was 60 cm long (east-west oriented); and 50 cm wide (north-south). It was lined, along the four sides, by flat stones. There were cranial and thoracic bones, but only in the eastern part of the burial zone. So we excavated deeper in the western part in order to find other bones, but found only burnished ware from the upper-final period of the ancient Neolithic. The excavation in this area, then, brought to light the above63

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 3. Carrara San Francesco (Bisceglie - Bari). Burial n.2 (a) and dry-stone wall (b). 64

R. Striccoli: The Neolithic Settlement of Carrara San Francesco in the Italian South-Eastern Neolithic

Author’s address

mentioned burial and the north-south oriented dry-stone wall that intersects with the other wall that runs from east to west.

Rodolfo STRICCOLI Cattedra di Preistoria e Protostoria Dipartimento di Scienze dell’Antichità-Facoltà di Lettere e Filosofia Università degli Studi di Bari-Palazzo Ateneo 70100 Bari ITALY

In 1999 we carried out a fourth excavation in the above mentioned areas. The larger area was 110 m² and covered the previous excavation areas. The smaller area was 50 m² and involved expanding the dig along the west and north side. The largest area in the third level (cm 40/42-cm 50/55) had a long wide space with stones, from a dry-stone wall (fig. 3:b). This wall crossed the area laterally and, on the west side, joined the wall excavated during the previous year. On the east side, after a break of 5 m, probably an entrance, this wall seemed to join the wall on the west side of the apse of the hut, found in 1997. In the two test excavation on the west side, the above-mentioned dry-stone wall seemed to join to another dry-stone wall running north-south across the whole surface we were excavating. This ended with what was probably an entrance to the area defined by the drystone wall. The entrance is 80 cm. wide and made up of flat stones on the ground and, on the north side, of a big squaredshaped east-west oriented stone (80x40 cm). Other small stretches of stone flooring were found in the southeast corner, and in the northeast corner of the largest area. This zone was covered in other stones, dating back to a more recent period, probably the final phase of the ancient Neolithic, or the Serra d’Alto phase, to judge from the pottery findings.

Bibliography CIPOLLONI SAMPÒ M., 1977-1982, Scavi nel villaggio neolitico di Rendina (1970-1976). Relazione preliminare. In Origini, XI, p. 183-323. COCCHI GENICK D., 1984, Manuale di Preistoria. II. Il Neolitico. Viareggio: Museo Preistorico e Archeologico “Alberto Carlo Blanc”. COPPOLA D., 1981, Nuove ricerche nell’insediamento di Torre Canne (Fasano - Brindisi). In Rivista di Scienze Preistoriche, XXXVI, p. 261-279. COPPOLA D., 1987 L’insediamento neolitico di Scamuso. In Atti XXV Riun.. Scient.. Ist. Ital. Preist. Prot., p. 223-234. CREMONESI G. - GUILAINE J., 1987, L’habitat neolithique de Torre S. Sabea, dans le cadre du Neolithique ancien de L’Italie du Sud-Est. In Actes Coll. Intern. C.N.R.S. Premieres communautes paysannes en Mediterranee occidentale, Paris, p. 377-385.

On the north-east side of the largest area, the excavation went down to the rocky bottom. In this area, too, working up from the bottom, we found a first level of yellow-reddish sand and clay, on a rock, bed with deep fissures. The surface of the first level (25 cm from the bottom), was inhabited during the middle-final period of the ancient Neolithic by human communities with a mixed economy (agriculture, breeding, fishing and hunting). They built a village probably surrounded by a large dry-stone wall with houses (hut with apse), and burials (simple ground-graves or defined by stones, both inside and outside the houses). Stone structures and burials lie on the third level, (cm 55/52-42/40), while on its surface were found pottery fragments that can be dated to the Serra d’Alto and Diana Bellavista phases, clear indications of subsequent human habitation. Confirmation of cultural heterogeneity is found at the second and the first levels too (20 cm each). Recent agricultural activity compromised our findings where, however, there is evidence of the ancient settlement in the second and third period of the ancient Neolithic of the Italian Southeast. They are similar to findings from coeval sites at Rendina II and III (Cipolloni Sampò 1982), at Guadone (Tinè Bernabò Brea 1980), of Monte Aquilone (Manfredini 1972), at Lagnano da Piede (Mallory 1984-1987), at Masseria la Quercia (Trump 1966), at Ripa Tetta (Tozzi 1980, Tozzi Verola 1991), at Balsignano (Radina 1994), at Scamuso (Coppola 1987), Trasano (Guillaine Cremonesi 1987) and from sites in the Balkan area and the Near East.

GUILAINE J. - CREMONESI G., 1987 L’habitat neolithique de Trasano, Premiers resultats. In Atti XXVI Riun. Scient. Ist. Ital. Preist. Prot., p.707-719 MALLORY J. P, 1984-1987, Lagnano da Piede. An Early Neolithic Village in the Tavoliere. In Origini, XIII, p.193-290. MANFREDINI A., 1972, Il villaggio trincerato di Monte Aquilone nel quadro del Neolitico dell’Italia meridionale. In Origini, VI, p. 29-274. RADINA F., 1994, Modugno (Bari), Balsignano. In Taras (Rivista di Archeologia), XIV, 1, Taranto, p. 39-40 STRICCOLI R, 1996, Bisceglie (Bari), Carrara San Francesco. In Taras (Rivista di Archeologia), XVI, 1, p. 18-20. STRICCOLI R., 1999, Bisceglie (Bari), Carrara San Francesco. In Taras (Rivista di Archeologia), XIX, 1, p. 21-22. STRICCOLI R., 2001, Primi scavi nel sito neolitico di Carrara San Francesco (Bisceglie - Bari). In Taras (Rivista di Archeologia), XXI, Taranto, (in corso di stampa). TINÉ S., 1983, Passo di Corvo e la civiltà’ neolitica del Tavoliere. Genova: Ed. Sages. TINÉ S. - BERNABÓ BREA M., 1980, Lo scavo nel villaggio del Guadone. In Riv. Scienze Preistoriche, XXXV, p. 45 ss. TOZZI C., 1988, Contributo alla conoscenza del villaggio neolitico di Ripa Tetta (Lucera). In Atti VI Convegno Preist.,Prot., St. Daunia, San Severo, p.11-19. TOZZI C - VEROLA M.L., 1991, La campagna di scavo 1990 a Ripa Tetta (Lucera). In Atti XII Convegno Preist., Prot., St. Daunia, San Severo, p.38-46. TRUMP D. H., 1966, Central and Southern Italy before Rome. Londra.

65

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 4. Carrara San Francesco (Bisceglie - Bari). Impressed ware (1,2,4,6), painted (7) impressed-painted (3,5). 66

C. Baldia: The Oldest Woven Textile of the Funnel Beaker Culture (4000-2900 cal. BC)...

THE OLDEST WOVEN TEXTILE OF THE FUNNEL BEAKER CULTURE (4000-2900 CAL BC) IN NORTH AND CENTRAL EUROPE Christel BALDIA

Résumé : Le fragment de textile datant du Néolithique /âge du Cuivre récemment découvert dans un « long-barrow » de terre en Moravie est l’un des plus anciens fragments bien préservés de vêtement conservés en Europe. Il s’agit de l’un des textiles attribuables à la culture des Vases en Entonnoir (TRB) d’Europe septentrionale et centrale. Il rejoint cinq autres textiles néolithiques, comme celui du tumulus de « Spitzes Hoch » en Allemagne centrale, décrit par Schlabow en 1959. Les descriptions initiales identifiaient ces textiles comme de la laine tissée. L’analyse du matériau du textile de Moravie montre qu’il s’agit de lin et de laine finement filés dans une armure toile. La qualité artisanale extraordinairement fine du textile, le contexte sépulcral et la proximité du rempart construit en pierre le plus ancien d’Europe septentrionale et centrale, indiquent une technologie étonnement complexe. Abstract: The recently discovered Neolithic/Copper Age textile fragment from an earthen long-barrow in Moravia, is among the earliest well preserved cloth fragments in Europe. It is one of the currently surviving textile attributable to the Funnel Beaker Culture (TRB) of North and Central Europe. It joins five other Neolithic textiles, such as the textile from the “Spitzes Hoch” long-mound in Central Germany, described by Schlabow in 1959. Initial reports identified these textiles as woven wool. Material analysis of the Moravian textile shows it to be of finely spun flax and woven in a tabby. The textile’s extraordinarily fine craftsmanship, the burial context and the proximity to the oldest stone-built rampart in North and Central Europe indicates a surprisingly complex technology.

INTRODUCTION

THE RESEARCH AREA

It is the purpose of this paper to present research results for the oldest preserved cloth of the funnel beaker culture (TRB), found at the Neolithic site of Rmíz in the Czech Republic. The archaeological research has been very sparse in especially on textiles. Therefore, special attention will be paid to the Funnel Beaker Culture itself, but also the periods before and after the TRB.

Located in Central Europe, the ancient hilltop site village Rmíz near Laskov, Czech Republic appears to be among the most ancient in a long chain of eleven similar “central places” dotting the Early Neolithic/Copper Age Moravian landscape. In Moravia, this period is characterized by the appearance of pottery from the Funnel Beaker Culture (TRB).

The Funnel Beaker Culture or TRB occurs in the European Neolithic, and dates approximately to 4000-3000 BC. The term in fact describes a specific form of ceramic vessel, shaped like a funnel that is widely spread all through Europe. Dates of certain cultural and technological developments or achievements, such as megalithic tombs and enclosures are often based on and compared to these ceramics and their spatial distribution. As a member of the Czech-American Research Project (CARPRO), I had the opportunity to participate in fieldwork in the Czech Republic in the summer of 1998. In the process, I was able to study the Rmíz textile at its curation facility in Olomouc. It became apparent then that there is very little known about TRB textiles in Central Europe. Neither textile technology, the material that was used nor even the exact temporal association have been thoroughly studied and are thus not well understood — this area of archaeological research is virtually unexplored. To understand the relevance of the Rmíz textile, it is necessary to (1) define the general research area, (2) summarize the cultural background and chronology, (3) understand the textile technology that was available during the Funnel Beaker Culture in the Neolithic Period, and finally (4) be aware of the ongoing research that has been conducted in Central Europe.

CULTURAL BACKGROUND AND CHRONOLOGY Centered far to the North, the TRB is only one of several overlapping and partly coexisting pottery styles in Moravia. The TRB is divided into several regional groups, which exhibit change through time and are in Moravia divided into two phases: TRB I = ca. 4000 to 3600/3400 BC TRB II= ca. 3600/ 3400 to 3200 BC Unfortunately, the precise dating of these phases is uncertain, but more work on the question is currently in process (see Baldia, M.O. in this volume). RMÍZ, THE TOMBS AND THE TEXTILE Rmíz is a multi-phase (TRB I and II) walled hilltop site with a TRB II satellite copper production site and two adjacent burial mound clusters. The pottery style found here indicates that it was erected during the TRB I B (Baalberge phase) which implies that this area 67

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Fabric Construction

was first occupied around 3800 BC. The tomb clusters adjacent to Rmíz are called Kremela I and II (SMID, 1981).

To produce a textile, really two components are needed. First there is the static element warp that is generally fastened on the loom somewhere. The number of warp threads per inch constitutes the EPI or ends per inch. The second element that is needed in the construction of a textile is the weft. The number of threads per inch of fabric is called PPI or picks per inch. The higher the number of threads in either warp or weft per inch, the finer the craftsmanship of the textile.

Kremela I consists of 58 mounds. They range from round tumuli to oval and rectilinear long-mounds, varying in length from 6 to 29 meters and in width from 5 to 12 meters. The excavation of Mound 2 revealed a rectangular drywall enclosure. Within the eastern end of the enclosure, along the central axis, two cremation burials were unearthed. Among these cremation remains were fragments of cloth (SMID, 1990).

Both weft and warp can be manipulated to create different patterns. Since the weft is the freely moving element, it seems easier to be creative with the weft, but even a plain weave, which is the simplest form of weave, can be arranged to create different effects, as can be seen in figure 1.

TEXTILE TECHNOLOGY To understand Neolithic textiles, the material, the fabric and the equipment used to produce them, must be examined.

Equipment Material The warp is put on a loom. Even primitive come in a variety of shapes and sizes and may be either horizontal or vertical. The construction always entails two beams, which allow the threads to be warped under tension. This tension can be created in warping the thread on the beam, but it can also be done by hanging loom weights on the ends of several warp threads. These weights are often just plain stones, sometimes shaped.

Various materials were used by Neolithic peoples to create threads, cordage and eventually textiles. Generally these are from either plant or animal origin. The Swiss lake dwellers seem to have mainly used tree bast and some flax (RASTEICHER 1995, 169-73), which was later at least partially replaced by wool. She also describes the labor intensive process of procuring bast fibers (RAST, 1990, 119). Bender Jørgensen, suggests that bast fibers from trees but also flax were used most frequently, and were a continuation of the Mesolithic. She also suggests that the use of wool is only adopted with the introduction of domesticated sheep (BENDER JØRGENSEN, 1991:116).

Depending on the spatial distribution of these artifacts (loom weights and whorls) within an archaeological site such as Rmiz, inferences about how and where they were used can be made. Questions such as whether ancient peoples wove in their houses or whether they had weaving houses can be answered this way.

Flax has been domesticated in the Levant area, and it is thought that it then slowly moved into central and northern Europe. So far, evidence seemed to point towards flax oil extraction preceding fiber utilization for about a millennium, which may however be due to a lack of artifacts that have survived the varying soil conditions.

PREHISTORIC TEXTILE RESEARCH Various archaeologists have done work on Neolithic textiles from different sites throughout Central Europe in the last 100 years, but overall research on this topic has been rather scarce for several different reasons.

The first step to textile production is spinning of the thread. This can be achieved by just twisting the fibers in counter direction between the fingers. However, the production is greatly enhanced and much more efficient when spinning whorls are used. The spinning whorl is really a weight on the end of a stick and works much like a top, twisting the fibers around each other faster than can be done by hand. The whorls can be made from stone, clay or wood. Their shapes differ—they can either be shaped concave or convex, and are often decorated. Thread is described as an S- or Zspin, depending on the direction of the twist: clockwise or counterclockwise Using spindle whorls allows for a finer thread to be produced (MESSIKOMMER 1913, VOGT, 1937). As the test excavation in Rmiz in 1998 showed, spindle whorls were found at Rmiz. However, the length of the fiber itself also greatly influences the fineness of the thread that can be created.

For one, not much evidence has been found, due to poor preservation and the time depth of 6000 years. Textiles do not preserve well in many environments, and Central Europe with plenty rain fall does not have the ideal climate to preserve textiles well as do dry caves or nitrogen rich bogs. So overall, there have not been many sites that yield Stone Age artifacts (HUNDT, 1967:39-67). Secondly, many artifacts that have been found were actually recovered relatively early in the development of scientific archaeological techniques, which now sometimes presents a problem. From the five Neolithic textiles in Germany for instance, one was described well in the 1938 by von Stokar, but the textile itself was lost in the war (BENDER JØRGENSEN, 1991:114). The analysis of the textiles can also be limited, due to their earlier treatments such as mounting textiles on tablets with aluminum containing lacquer which interferes with many tests today.

The results of my research indicate that the use of flax may have been more commonly exploited for its fibers than initially thought. 68

C. Baldia: The Oldest Woven Textile of the Funnel Beaker Culture (4000-2900 cal. BC)...

Figure 1. Plain Weave or tabby, (a) balanced tabby, (b) weft-faced, (c) warp-faced

Nevertheless, the textiles from the Spitzes Hoch site in Central Germany that were found and documented first in 1888 as extremely fine weavings with an EPI/PPI of 40/30. The Spitzes Hoch textile was reworked and described by Schlabow (1959, 101-20) in the 1950s as wool. A more recent analysis by the Danish Museum confirms that the doubts about Schlabow’s analysis that had been raised by Sherratt and Ryder (SHARRATT, 1983:103) were justified. The material was identified as flax fiber and not wool (BENDER JØRGENSEN, 1991:51). Also, all five described textile artifacts are woven in a tabby and are associated with cultures that date younger than the Rmiz cloth:

Kremela I Burial Mound yielded the cloth fragment. This burial is associated with TRB II A, which dates to 36003200 cal BC. The dimensions of the cloth are 3 x 3 cm. It is extremely charred, and it occurs in several layers—obviously it had been folded.

MACRO-ANALYSIS The color could be best describes as several gray shades to black. It was assumed at the time of the original write-up that the fibers that were used were woolen. Also, the impressions of wheat on the cloth were noticed, and two kernels of wheat Triticum dicoocum where found in its context (OPRAVIL, 1989). The thread is rather fine and quite even in thickness, somewhat unraveled on one of the edges. The numbers of threads were counted at 13 per cm, which is about equivalent to a EPI/PPI of 33/34. The piece verifies the existence of extraordinarily fine craftsmanship. Most of the piece allows for the weave to be identified as an even plain weave or tabby with no apparent selvedge: the fragment seems to be out of the middle of a bigger textile. The other section of the fragment can only be described as if it were melted together. Due to the folding, different degrees of charring can be observed, with ash deposits around each thread. Also, in one of the folds, there is a shiny, clear, almost resin-like deposit (BALDIA et al., 1998).

1. Wiepenkathen textile (bast/wool), associated with the Lomborg phase A, late Neolithic, ca. 2400 BC. 2. Gollwitz textile, no details given, associated with the TRB. 3. Kreienkopp textile, associated with corded ware culture, late Neolithic. 4. Spitzes Hoch textile, associated with the Bernburg II/III phase of the TRB. 5. Lützendorf textile, associated with the corded ware culture and was lost in the war (BENDER JØRGENSEN, 1991, 224). Poland, neighboring the Czech Republic to the northwest, only has evidence of Neolithic textiles are impressions on pottery that is associated with the Kultura Tripolska, and one other textile fragment without provenience.

MICRO-ANALYSIS

Otherwise, the Swiss lake-dwellings have yielded many textiles (MESSIKOMMER 1913; VOGT, 1937; RASTEICHER, 1995, 1998), using some interesting techniques to embellish their borders to create visual effects. The materials are generally identified as tree bast of various sources and flax. The oldest textile artifact seems to have been found at the Schenkton Trichtermoos. It was found in the same strata charcoal, which dated C14 to 4840 BP (no standard deviation given). According to the Othmar Wey, the artifacts belong to the Cortaillod Kultur at ca. 3550 cal BC (WEY, 1990, 281).

The sample was examined under low power magnification using a stereo-microscope, and was analysed for elemental composition. Dr. Jakes determined that “…a glassy type material” was surrounding the individual fibers that measured from about six microns to nine microns in diameter, and that the cross-sectional shape of the fibers as indicate that this is flax. Radiocarbon dating for this piece was originally planned. However, the fragment contains “carbon from three different sources” (JAKES, 1999), which must be cleaned before it is possible to run a C14 date of the sample. According to Christine Pryor from RAFTERS Labs Inc. in New Zealand, it is possible to clean textiles in the appropriate lye bath

RMÍZ TEXTILE The Rmiz cloth, discovered in 1988, may be the oldest existing TRB cloth. Flotation sampling the excavation of the 69

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Bibliography

without destroying the sample (personal communication). This will be undertaken as soon as funding is obtained.

BALDIA, M.O., BALDIA, C., FRINK, D., 1998, Activity Report of the 1998 Czech-American Rmiz Research Project, http:// www.comp-archaeology.org/RMIZ1998REPORT.HTM.

DISCUSSION

BENDER JØRGENSEN, L., 1991, North European Textiles until 1000 AD. Aarhus: University Press.

Flax is a native of the Middle East and is considered to have been domesticated rather early (HELBEAK, 1959) and requires specific climate conditions. The Czech Republic seems to have been well suited for growing flax since it even today still produces some. So do neighboring Poland and the former Soviet Union (DEMPSEY, 1975). The processing of the plant to get to the fiber is rather labor intensive, and requires time and energy to go through the process. It is therefore likely that only a somewhat specialized society can pursue these sorts of activities—fiber procurement, as well as textile production (PRICE, 1998).

DEMPSEY, J. M., 1975, Fiber Crops, Gainesville: University of Florida Press. ERNY-RODMAN, C., GROSS-KLEE, E., HAAS, J.N., JACOMET, S., ZOLLER, H., 1997, Früher “human impact” und Ackerbau im Übergangsbereich Spätneolitikum - Frühneolitikum in schweizerischen Mittelland. In Jahrbuch der Schweizerischen Gesellschaft für Ur- und Frühgeschichte 80, p. 27-56. HELBAEK, H., 1959, Notes on the Evolution and History of Linum, KUML. Aarhus, p. 103-129. HUNDT, H.-J., 1967, Über Vorgeschichtliche Seidenfunde, In Jahrbuch des Römisch-Germanischen Zentralmuseum Mainz 16, p. 59-71.

Though the textile was originally thought to be wool, this new research shows that flax fibers were used in the funnel beaker culture TRB II at about 3600/3400 cal BC. The occurrence of cloth in a cremation burial context with wheat present is not only interesting, but may also help us understand TRB life ways and burial customs beyond the ceramic details previously known.

JAKES, K., 1999, Materials Analyses of the Oldest Persevered Textile from the Neolithic/Eneolithic in Central / Northern Europe http://www.comp-archaeology.org/Jakes_rmizreport.htm MESSIKOMMER, H., 1913, Die Pfahlbauten von Robenhausen: L’Epoche Robenhausienne. Zürich. OPRAVIL, E., 1989, Rozbar Textilu ( pro dr. Miroslava Smida). PAVELČÍK, J., 2001, Erkenntnisse bezüglich auf die Bevölkerung der Höhensiedlung der Badener Kultur in Hlinsko bei Lipník nad Beèvou (Mährisches Tor). In Mensch und Umwelt während des Neolithikums und der Frühbronzezeit in Mitteleuropa, edited by A. Lippert, M. Schultz, S. Shennan, & M. Teschler-Nicola. Rahden/Westf.: Verlag Marie Leidorf., p. 73-82.

Flax pollen are reported in Switzerland in association with cereal production around ca. 6500-6000 cal BC. One unburned flax seed was found in Wallisellen-Langachermoos, (ERNY-RODMAN, C., et al., 1997), but textile finds are not reported in this context. In Moravia, Pavelčík reports emmer wheat and other cereals grain as well as a good representation of flax from Hlinsko, a site that is located only 45 km east from Rmiz (PAVELČÍK, 2001, 73). Hlinsko is also associated with TRB II, the same time frame as Rmiz. It seems then that flax and cereal farming was in place during this time, and that the flax was not just grown for its seeds or oil extraction, but for fiber procurement as well.

PRICE, D., Gruppe, G., Schröter, P., 1998, Migration in the Bell Beaker Period of Central Europe. Antiquity 276, p. 405-411. RAST, A., 1990a, Die Verarbeitung von Bast. In Die ersten Bauern, Vol 1, Pfahlbaubefunde Europas Forschungsberichte zur Ausstellung im schweizerischen Landesmuseum, Zürich, p.119121. RAST, A., 1990b, Jungsteinzeitliche Kleidung. In Die ersten Bauern, Vol 1, Pfahlbaubefunde Europas Forschungsberichte zur Ausstellung im schweizerischen Landesmuseum, Zürich, p.121126.

The Rmiz cloth bridges an important gap in our knowledge about textiles production in the archaeological record. Further research is warranted to seek additional evidence not only on early textile production in this region, but also to round out this picture of early craft specialization and agriculture in Central European prehistory.

RAST-EICHER, A., 1995, Gewebe und Geflechte. In Die Schweiz vom Paleolitikum bis zum Mittelalter II, Neolitikum, Basel, p. 169-173. RAST-EICHER, A., 1998, Gewebe und Geflechte. In Die Schweiz vom Paleolitikum bis zum Mittelalter III, Neolitikum, Basel, p. 285-290. SCHLABOW, K., 1959, Beiträge zur Erforschung der jungsteinzeitlichen und bronzezeitlichen Gewebetechnik. Jahresschrift für mitteldeutsche Vorgeschichte 34, p. 101-120.

Acknowledgements I would like to thank Dr. Kathryn Jakes from the Dept. of Textile Science for material analysis that she performed on the Rmiz cloth for the project.

SHARRATT, A., 1983, The secondary Exploitation of Animals in the Old World. World Archaeology 15, p. 261-305.

Author’s address

SMID, M., 1990, Pripevek k poznáni eneolitických mohilových porebišt’ na stredni Morave. Praveké a slovanské osidleni Moravi, p. 67-89.

SMID, M., 1981, Stratifické pozorování na výinném eneolitiékem sidlisti Rmíz u Laskova, okr. Prostéjov, Prehledy výzkum, Brno, 1979, p. 17.

Christel BALDIA The Ohio State University Consumer & Textile Science Dept. 1787 Neil Ave Columbus, Ohio 43210-1295, USA

VOGT, E., 1937, Geflechte und Gewebe der Steinzeit, Basel. WEY, O., 1990, Seeufersiedlungen am Sempachersee. In Die ersten Bauern 1: Pfahlbefunde Europas. Zürich: Schweizerisches Landesmuseum. 70

J.-P. Pautreau, B. Farago-Szekeres & P. Mornais: La nécropole néolithique de la Jardelle à Dissay (Vienne, France)

LA NÉCROPOLE NÉOLITHIQUE DE LA JARDELLE À DISSAY (VIENNE, FRANCE) Jean-Pierre PAUTREAU, Bernard FARAGO-SZEKERES & Patricia MORNAIS

Résumé : La nécropole de la Jardelle à Dissay, dans le seuil du Poitou (France), rassemble une dizaine de sépultures. Il s’agit de coffres en pierre plus ou moins enterrés, recouverts par un tertre en terre limité par une palissade dont subsistent les fondations, fossés circulaires, subcirculaires ou allongés, souvent ouverts en direction de l’Est-Sud-Est. Ce cimetière a fonctionné au début du Néolithique moyen mais aussi au Néolithique récent. Abstract: The cemetery of La Jardelle at Dissay, in the Poitou (France), consist of about ten graves. They are stone cists more or less set in the ground, covered by a mound of earth limited with a circular, sub-circular or oval lengthened trench. The ditch is often opened in the direction of the East-south-east. This barrow cemetery was used at the beginning of the middle phase of Neolithic but also during the recent Neolithic.

SITUATION

est constitué par une juxtaposition de tranchées longues d’un peu plus de 2 m et de profondeur inégale. Le comblement des portions de fossé diffère selon les secteurs ; mais nous trouvons partout : un éboulis primaire (sédiment brun à petites pierrailles jaunâtres) issu de l’intérieur de l’enclos, un remplissage axial original formé le plus souvent de matériaux caillouteux, des horizons calcifiés à la base du remplissage, un sédiment sombre, parfois calcifié, dans le comblement issu de l’extérieur de l’enceinte. La découverte de bois calciné à la base et au centre du fossé confirme la présence d’une palissade en matière périssable. Le matériel issu du comblement se limite à des éclats, à une lamelle de silex, à quelques percuteurs et à des tessons de poterie non tournée dont un bord assez épais. Ils apportent peu d’éléments pour l’attribution culturelle.

Le gisement de la Jardelle se trouve sur la terrasse, rive droite de la rivière le Clain, à environ 1 km au sud-ouest du bourg de Dissay. Dans la partie centrale du seuil du Poitou, le Clain coule ici du sud vers le nord. La vallée, large et plate, est formée par les alluvions du cours d’eau. Limons et argiles sont surmontés par les sables et graviers de la basse terrasse. L’altitude du site, sur la terrasse, (63 m) est à peine supérieure à celle du cours d’eau (environ 62 m), distant de 300 m au nord-ouest. La partie occidentale du gisement subit des inondations fréquentes. Les premières pentes du rebord abrupt du plateau calcaire se trouvent à moins de 100 m des enclos les plus orientaux. Le plateau culmine ici à 110 m d’altitude et domine donc la vallée d’une quarantaine de mètres. Les sols du plateau sont recouverts par une terre de groie, terre rouge argileuse renfermant des cailloux calcaires.

Un fossé circulaire ouvert est venu se greffer sur la partie orientale de la construction primitive. Ce fossé circulaire de 24 m de diamètre est largement interrompu vers l’est et vers l’ouest. L’ouverture orientale, la plus large (10 m), correspond sensiblement à la largeur de l’enclos allongé. Une fosse (trou de poteau ?) occupe le milieu de l’interruption. Vers l’ouest le fossé circulaire disparaît sur moins de 4 m. Ces deux interruptions peuvent s’expliquer par la présence du tertre primitif limité par le fossé allongé et perturbant (ou rendant inutile) le creusement de la seconde tranchée. La largeur de la tranchée varie entre 0,50 m et 1,20 m. La profondeur va de moins de 0,10 m à près de 0,50 m. Ces variations sont dues au creusement effectué au moyen d’une succession de petites fosses mesurant un peu plus de 2 m de long mais de profondeurs absolument inégales. Le mobilier issu du fossé se limite à trente tessons et quelques silex à forte patine blanchâtre parmi lesquels des éclats laminaires et un petit grattoir.

HISTORIQUE Le site de la Jardelle à Dissay a été repéré lors de prospections aériennes par A. Ollivier. Les recherches de terrain se sont déroulées entre 1995 et 1998, sous la direction de J.-P. Pautreau, avec l’aide du Service régional de l’Archéologie et l’appui financier du département de la Vienne. LA NECROPOLE Les monuments sont disposés de part et d’autre d’une zone humide, au ruissellement intermittent, qui relie la base du plateau au lit du Clain. Trois tertres allongés arasés limités par un fossé, possèdent une fosse axiale et une ouverture vers l’Est ; deux ont pu être étudiés.

La fosse sépulcrale mesure un peu moins de 2 m sur 1,80 m pour 0,50 m de profondeur. La tombe a été aménagée dans une fosse creusée dans le substrat calcaire. Cette fosse occupe la partie axiale des enclos quoique plus près du fossé nord.

Le monument A, mesure environ 35 m de long sur 10 m à 15 m de large (fig. 1). Le plan « en fer à cheval allongé » montre un élargissement dans la moitié orientale. Le fossé 71

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. La Jardelle, Dissay. Haut : Plan de la nécropole. Situation. (relevé et dessins J.-P. Pautreau). Bas : Tertre et sépulture A. (relevé et dessins J-P. Pautreau, P. Mornais). 72

J.-P. Pautreau, B. Farago-Szekeres & P. Mornais: La nécropole néolithique de la Jardelle à Dissay (Vienne, France)

La plus grande longueur est de direction E. – N.-E. / O. – S.O., légèrement décalée par rapport à l’axe de l’enclos allongé. La cavité mesure 2,10 m (O.-E.) et 1,90 m (S.-N.). La profondeur varie entre 0,30 m au centre et 0, 50 m à l’emplacement des sablières. Les flancs sont assez abrupts. Les dalles ont disparues, arrachées par l’agriculteur exploitant, mais les traces des sablières (larg. 0,15 m, prof. 0, 20 à 0,30 m) supportant les parois de la chambre, calcifiées, restent bien visibles. L’intérieur de la chambre mesure 1,40 m sur 0,70 m. Elle ne comporte pas (ou plus) de pavage. Elle contenait quelques ossements humains mal conservés et sans collagène. Ces restes humains, très fragmentés et complètement dispersés, pourraient appartenir à un même sujet adulte. De rares tessons minuscules se retrouvent disséminés au milieu des ossements ; les pâtes brunes ou beiges en surface montrent souvent un intérieur noir. Les dégraissants sableux sont assez fins. La cuisson est de qualité moyenne soulignée par une tendance prononcée à la desquamation et à l’effritement. Le mobilier lithique est extrait d’un silex de couleur blanche ou cire. Plusieurs éléments montrent une patine blanchâtre. Il s’agit d’un débitage sur éclats courts ou éclats laminaires conservant souvent une réserve corticale. Aucun outil n’a été mis au jour. La datation d’un bois calciné du fossé donne une date comprise entre 4496 et 4367 avant J.-C. (Ly-8616). L’utilisation du monument semble limitée au Néolithique moyen 1.

du lit actuel du Clain et est en zone exceptionnellement inondable. Malgré le fort arasement de l’ensemble, il est possible de l’interpréter comme une fondation de palissade limitant un tumulus de faible ampleur. Dans la partie axiale de l’enclos, nettement décalée vers l’interruption, la chambre funéraire a été aménagée dans une fosse de 2,60 m sur 1,70 m. La chambre rectangulaire n’est pas pavée. Mesurant environ 1,30 m sur 0,85 m, elle est limitée par des dalles de pierre disposées verticalement (la plus grande mesure 1,30 m de long, 0,65 m de hauteur et 0,30 m d’épaisseur). Les chevets sont formés par deux blocs trapus. La couverture de la tombe était constituée de deux dalles jointives assez soigneusement préparées. A l’exception d’une dalle calcaire venant probablement de la vallée, toutes les autres sont issues du plateau. Les ossements humains mal conservés semblent appartenir à un individu adulte. Leur dispersion interdit néanmoins toute restitution de la position du défunt. Ils ont été datés entre 4523 et 4339 av. J.-C. (LY9411). Le mobilier archéologique comprend : - De rares tessons minuscules (pas de forme, ni bord, ni décor) disséminés au milieu des ossements. Les pâtes dominantes sont brunes ou beiges en surface avec un intérieur noir aux dégraissants sableux assez fins. La cuisson est de qualité moyenne : tendance assez prononcée à la desquamation et à l’effritement.

Le tertre B est limité par un fossé de 25 m de long pour 13 m à 13,5 m de large (fig. 2). L’enclos est interrompu sur près de 4 m à son extrémité Est. L’enclos marque un léger resserrement dans la partie médiane de sa longueur. La tranchée, dont la largeur est comprise entre 1,20 m et 0, 80 m, atteint une profondeur variant selon les secteurs entre 0,60 m et plus de 1 m. Ces variations sont dues à l’état d’arasement mais aussi peut-être aux techniques de creusement par modules indépendants observées sur les autres enclos. Le fossé présente une relative unité dans ses profils et ses comblements. Le forme observée du fossé est majoritairement celle d’un « U » assez profond aux flancs légèrement évasés. Le comblement des portions de fossé apparaît différent selon les secteurs. Parmi les constantes observées nous retiendrons la présence d’un remplissage axial original, l’existence d’horizons calcifiés à la base du remplissage, sur les parois et le fond puis un comblement réalisé par une succession de nombreuses couches graveleuses variées.

- Quelques éclats courts ou laminaires en silex bleuté avec souvent une réserve corticale. Plusieurs montrent une patine blanchâtre. - Les deux valves d’une coquille, probablement de moule perlière d’eau douce, Margaratifera margaratifera, (détermination par Yves Gruet). Le monument conserve les traces d’une seule phase de construction. Son utilisation, probablement en sépulture individuelle, semble limitée au Néolithique moyen 1. Les autres monuments étudiés sont des tertres circulaires ou subcirculaires. Le monument C est limité par un fossé sub-circulaire mesurant entre 18 et 21 m de diamètre extérieur (fig. 2). La tranchée, dont la largeur varie entre 1,40 m et 0,60 m, atteint une profondeur de 0,70 m à 0,40 m. Le remplissage montre, au fond, un éboulis primaire puis un dépôt caillouteux en «Y», vestige d’une palissade et de son calage par remise des terres du creusement déposées de part et d’autre de la tranchée. Dans le secteur oriental un nombre important de pierres et dallettes calcaires de tailles diverses (pouvant atteindre 20 cm de long) disposées à plat, occupent la partie axiale du fossé. Les seuls vestiges issus du fossé sont quelques éclats de silex de qualité médiocre recueillis en surface.

Le matériel archéologique reste extrêmement pauvre : un seul éclat de silex trouvé à 55 cm de profondeur n’apporte rien quant à l’attribution culturelle de l’ensemble. Le cailloutis axial présent sur toute la surface peut correspondre au calage d’une ligne de trous de piquet plantés dans le fond du fossé. Le comblement issu de l’intérieur de l’enclos et celui venu de l’extérieur montrent de grandes similitudes. Un apport de sédiment sombre concrétionné venu de l’extérieur a été observé dans le secteur S.-O. Seuls certains relevés mettent un évidence un remplissage différent probablement issu de l’éboulis d’un tertre interne. Les horizons concrétionnés témoignent d’une circulation d’eau abondante. Cet enclos est, parmi ceux étudiés, le plus proche

Au centre de l’enclos, la chambre funéraire est aménagée dans une fosse mesurant 3,10 m dans le sens N.-O. – S.-E., 2,20 m selon l’axe S.-O. – N.-E. et profonde de 0, 60 m, creusée dans le substrat. Les flancs sont relativement abrupts. Sur le fond, une tranchée au plan en forme de « U » large de 73

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 2. La Jardelle, Dissay. Tertre et sépulture B. (relevé et dessins J.P. Pautreau, V. Mariuz). Tertre et sépulture C. (relevé et dessins J.-P. Pautreau, E. Rodriguez-Lopez, C. Barbotin).

0,20 m et 0,30 m et profonde de 0,10 m et 0,20 m est interrompue au S.-E. sur 0,70 m ; ceci correspond à la pente douce de la fosse (accès ?). Elle servait de fondation aux

dalles de la tombe, arrachées par l’exploitant agricole et dont subsiste un fragment brisé. A l’intérieur de la surface ainsi limitée, le sol de la chambre a été bien aménagé par un pavage 74

J.-P. Pautreau, B. Farago-Szekeres & P. Mornais: La nécropole néolithique de la Jardelle à Dissay (Vienne, France)

régulier constitué de petites pierres plates en calcaire mesurant de 10 à 30 cm de côté reposant à même le substrat.

ou beiges. Un récipient montre des surfaces rouges et un intérieur noir. Les dégraissants sableux sont assez fins. La cuisson est de qualité moyenne ; une tendance assez prononcée à la desquamation s’observe sur de nombreux tessons. Aucun décor, aucun élément de préhension n’ont été découverts. Un des vases possède un trou (de suspension ?) au niveau du col.

Le niveau sépulcral, épais de 10 à 20 cm au S.-E., en lambeaux ailleurs après l’arrachage des dalles, est un sédiment sombre avec de petites pierres, renfermant des fragments osseux, quelques tessons puis des parures en os et en coquillage. Sa base est au contact du dallage. Exception faite du quart N.O. de la chambre, particulièrement bien conservé, ce niveau sépulcral contient des restes humains fragmentés, pour la plupart considérablement dispersés et les corps auxquels ils appartiennent sont souvent incomplets. Si le décompte des principaux os longs ne fait apparaître guère plus d’une quinzaine de pièces par catégorie (humérus, fémur et tibia, droit et gauche confondus), l’examen des dents retrouvées dans la chambre permet en revanche d’affirmer l’utilisation du sépulcre pour une vingtaine d’individus, adultes et enfants de tous âges. Neuf adultes (de sexe indéterminé faute de coxaux suffisamment conservés) dont trois jeunes d’une vingtaine d’années ont ainsi été identifiés, ainsi qu’au moins neuf enfants (l’un âgé d’environ 3 ans, quatre autres entre 5 et 10 ans et trois entre 10 et 15 ans) parmi lesquels on remarquera la présence d’un nouveau-né.

Le mobilier lithique est extrait d’un matériau de couleur cire ou crème parfois à patine blanchâtre (fig. 4, n°6, 8-10). Il s’agit d’éclats courts ou lamellaires conservant souvent une réserve corticale (fig. 4, n°10). Quelques retouches marginales des bords ont été observées sur deux petits éclats lamellaires. Les parures se limitent à des éléments de collier, bracelet ou pendentifs : Trois incisives et une canine perforées (fig. 4, n°1-3) ; une lamelle de défense de sanglier montre un certain soin dans son façonnage, avec le dégagement en bélière de son extrémité perforée (fig. 4, n°4). Une série de dentales, plusieurs enfilées les unes dans les autres, proviennent de la même zone. Une datation à partir des ossements de la sépulture a donné l’âge de 5070 ± 50 BP soit entre 3959 et 3757 avant J.-C. (Ly-7641). Ce qui nous place chronologiquement au milieu du Néolithique moyen. Les parures et les céramiques, associées à quelques éclats de silex sont en accord avec cette datation. Vraisemblablement nous avons là un monument construit ou plutôt complètement remanié au Néolithique moyen II, servant de sépulture collective à une communauté, peut-être familiale.

Malgré les multiples remaniements qu’a subi le niveau funéraire (arrachage moderne des dalles et dégâts connexes, mais aussi piétinement et déplacement des ossements en cours d’utilisation, voire vidange partielle ou pillage de la tombe), la répartition générale des vestiges montre que les ossements d’enfants occupent majoritairement la moitié de la chambre située en avant de la porte et, d’autre part, que les quelques éléments de parure présents dans la tombe ont systématiquement été retrouvés à proximité d’un crâne et/ou d’une mandibule d’enfant. Seul le quart N.-O. de la chambre offre ici une image intacte du sépulcre et de son fonctionnement. Tous les ossements gisant sur cette portion de dallage semblent pouvoir être attribués aux six corps trois adultes, deux enfants et un nouveau-né - qui y ont été successivement inhumés et il subsiste suffisamment de connexions anatomiques pour lire ou restituer la position de chacun d’entre eux. Excepté le nouveau-né, dont seuls quelques ossements signalent la présence, les corps sont tous allongés sur le côté gauche, tête dirigée vers l’entrée de la chambre, membres inférieurs et supérieurs fléchis. Ils sont déposés les uns sur les autres, les deux premiers au fond de la chambre, les suivants progressivement vers la porte d’accès et chaque nouvelle inhumation s’accompagne d’un léger remaniement des squelettes antérieurs : c’est notamment le cas du crâne et de la mandibule, souvent repoussés ensembles. L’intrication générale des ossements, les dislocations et déplacements semblent contredire l’éventualité d’un dispositif durable isolant les corps les uns des autres, et, d’autre part, rien dans la posture de ces défunts ne permet d’imaginer un enveloppement contraignant du cadavre ni une inhumation en position assise.

Un autre monument, le monument D, sub-circulaire, ouvert au Sud-Est, montre deux phases de construction. Le tertre primitif est limité par une palissade de 14 m de diamètre, rentrante (angle de 120°) au niveau de l’interruption large de 1,60 m. Le centre de l’accès probable est occupé par une fosse (fig. 3). Les dimensions de la tranchée, plus importantes aux extrémités où l’on observe des trous de poteau, varient entre 0,90 m et 1,20 m (largeur) et 0,25 m à 0,40 m (profondeur). Le fossé, profilé en «U» aux flancs évasés, présente un comblement particulièrement homogène de sédiment brun, mélange d’éléments fins et de petites pierres (module avoisinant 5 cm à 6 cm) surtout concentrées au centre du fossé. Plus tard le monument est agrandi et limité par un fossé circulaire de 19 à 20 m de diamètre ; il conserve, sur 5 m de large, son accès à l’E. –S.-E.. La largeur en surface, varie entre 3 m et 1,10 m pour une profondeur moyenne de 20 cm mais qui atteint 50 cm près de l’interruption (trous de poteau). Une petite tranchée recoupant ce fossé doit correspondre à un aménagement tardif de l’entrée du monument. Une pierre dressée (larg. 0,85 m, ép. 0,15 m) s’élevait à l’extérieur, à 3 m en face de l’entrée, en direction du soleil levant (fig. 3).

Les céramiques apparaissent particulièrement fragmentées. Les bords découverts portent une lèvre arrondie, légèrement éversée ou dans le prolongement de la paroi. Les récipients sont tous des vases fermés à panse ventrue. L’un d’eux possède un fond arrondi. Les pâtes dominantes sont brunes

La fosse sépulcrale occupe la partie centrale-nord de l’enceinte délimitée par les deux enclos sub-circulaires concentriques. Elle est creusée dans le substrat calcaire de la terrasse. Sa plus grande longueur, de direction E. – N.-E. / O. – S.-O. (3,40 m) n’est pas exactement dans l’axe de 75

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 3. La Jardelle, Dissay. Haut : Tertre et sépulture D. (relevé et dessins J.-P. Pautreau, A. Matringhem, V. Cherrier). Bas : Sépulture double d’enfants (relevé et dessins P. Mornais). 1 à 4 : Céramiques de la sépulture double d’enfants. 5 à 9 : mobilier de la sépulture D (dessins J.-P. Pautreau).

l’interruption des fossés. Au fond de la fosse, une petite tranchée de fondation (largeur : 0,20 m et 0,30 m à l’ouverture, profondeur : 0,20 m.) a été creusée sur les deux grands côtés (fig. 3). A l’intérieur de la surface limitée par ces deux tranchées, le sol de la chambre a été bien aménagé par un pavage régulier constitué de petites pierres plates en calcaire

(10 à 30 cm de côté). Si l’on excepte les deux perturbations récentes qui amputent la fosse sépulcrale à ses extrémités est et ouest, le comblement de cette fosse (espace central et alentours) peut schématiquement se décomposer en trois grands ensembles. La moitié nord est principalement occupée, sur toute l’épaisseur du comblement, par des blocs de pierre, 76

J.-P. Pautreau, B. Farago-Szekeres & P. Mornais: La nécropole néolithique de la Jardelle à Dissay (Vienne, France)

petits et gros, mêlés de quelques rares ossements humains. A l’inverse, la moitié sud n’a livré que quelques pierres, et cet espace est comblé par une nappe d’ossements très fragmentés au sein de laquelle on ne distingue aucune connexion anatomique. Le recollage entre les principaux fragments d’os longs fait d’ailleurs apparaître deux lignes de fracture, signe que cette nappe s’est scindée en plusieurs parties probablement lors de l’arrachage des dalles de la chambre. Vers le milieu de la fosse, cette nappe s’interrompt nettement selon un tracé rectiligne, tracé qui correspond sensiblement à la limite du dallage calcaire sous-jacent. Enfin, le troisième ensemble est constitué par le dallage sur lequel reposent un certain nombre de restes humains en connexion anatomique.

Plusieurs fragments proviennent d’une autre céramique de type pot de fleur avec téton, similaire au précédent mais au fond légèrement arrondi (fig. 3, n°8). Un tesson de grande taille vient d’un pot plus élancé et légèrement plus pansu que les précédents ; il porte un court col éversé à lèvre arrondie (fig. 4, n°11). Un petit bord à lèvre amincie (fig. 3, n°6) et d’autres fragments semblent isolés dans le sédiment. Les pâtes dominantes, aux dégraissants sableux assez fins, sont brunes ou beiges en surface avec souvent un intérieur noir. La cuisson est de qualité moyenne (tendance prononcée à la desquamation et à l’effritement) s’observe sur de nombreux tessons. Le mobilier lithique est en silex de couleur blanche ou cire, parfois à patine blanchâtre. Il s’agit d’un débitage sur éclats courts ou éclats laminaires (fig. 3, n°9) conservant souvent une réserve corticale. Le seul outil est une armature de flèche tranchante proche du type Sublaines (fig. 4, n°7). L’outillage en os se cantonne à un ciseau trouvé dans un secteur perturbé (fig. 4, n°5).

Le décompte de la totalité des ossements contenus dans la fosse montre que cette tombe a reçu au moins une dizaine de corps, parmi lesquels neuf adultes dont un jeune (15-20 ans) et un unique très jeune enfant, représenté par quelques fragments de diaphyse. Cette fois, et contrairement au phénomène constaté pour la chambre du monument C, le nombre minimum d’individus reste sensiblement le même quel que soit le type d’os considéré (os longs droits et gauches, dents supérieures et inférieures).

Une datation à partir des ossements de la sépulture a donné l’âge de 4670 ± 60 BP soit entre 3634 et 3349 avant J.-C. (GRA-17363, Ly-1343). L’utilisation du monument dès le Néolithique moyen I avec dès lors plusieurs phases d’aménagement puis son fonctionnement au Néolithique récent, alors comme sépulture collective, semblent bien établis.

Sur la surface relativement restreinte du dallage calcaire reposent trois ensembles très incomplets d’ossements en connexion parfois lâche. Un premier sujet adulte n’est représenté que par son torse. Ses membres supérieurs sont bouleversés et les membres inférieurs absents : la position primitive du corps reste donc très incertaine, mais il est peu probable qu’il ait été couché sur le côté gauche. Un amas d’ossements occupe l’emplacement présumé des membres inférieurs du sujet précédent. Cet amas, qui ne comporte aucun doublon ou ossement surnuméraire, est constitué par un empilement de tronçons anatomiques : on serait donc tenté de l’attribuer au même sujet adulte et l’ordre de superposition des tronçons n’est pas sans évoquer une inhumation en position assise, à moins qu’il ne s’agisse de la réduction d’un cadavre en cours de décomposition. Le troisième ensemble, très fragmentaire, repose au-dessus de l’épaule droite du sujet allongé et présente de troublantes analogies avec l’amas précédent. Bien que les détails du fonctionnement de cette chambre funéraire soient loin d’être élucidés, il semble néanmoins y avoir de profondes différences dans l’utilisation des monuments C et D, tant dans le recrutement que dans le traitement des défunts.

Une sépulture double d’enfants a été aménagée en bordure du tertre D. La fosse sépulcrale recoupe partiellement le second fossé ; elle était comblée par un sédiment graveleux brun clair à jaunâtre. Les deux petits corps reposaient à l’intérieur d’une chambre funéraire probablement en bois, de 0,90 m sur 1,10 m de côté, matérialisée par ses contours calcifiés. Les cotes d’altitude prises sur la paroi d’une part, et à la base des ossements d’autre part, révèle une profondeur minimale de 20 cm environ. L’aspect des parois (verticales, effondrées, etc...) et du fond n’est pas connu, l’intérieur de la chambre étant comblé par un éboulis compact, concrétionné, en pendage vers le centre de la chambre ; il recouvre en partie les deux petits squelettes. Les deux enfants inhumés sont âgés de 4 ans (+/- 24mois) (Ubelaker). L’état de conservation des ossements est très moyen, mais la position générale des enfants reste lisible. Ils reposent l’un contre l’autre près de la paroi ouest de la chambre, couchés tous deux sur le côté gauche dans une position quasi identique : les membres supérieurs sont fléchis et croisés l’un sur l’autre, les mains ramenées sous le menton, les membres inférieurs sont fortement pliés ; leurs visages sont tournés vers l’est. L’un des enfants, nettement de profil, semble retenu par la masse du corps du second enfant placé devant lui, tandis que ce dernier est légèrement affaissé sur le ventre. Si l’inhumation simultanée ne peut être formellement démontrée ici, elle reste toutefois fort plausible.

De rares tessons minuscules se retrouvent disséminés au milieu des ossements. Un seul vase entier, de type pot de fleur, a été découvert au niveau de l’épaule droite du corps allongé sur le dallage De rares tessons minuscules se retrouvent disséminés au milieu des ossements. Un seul vase entier, de type pot de fleur, a été découvert, associé à l’inhumation reposant sur le dallage. Il est assez trapu, à fond plat non débordant, tronconique à parois rectilignes et possède une lèvre amincie légèrement rentrante. Il porte un seul téton à 2 cm sous le rebord (fig. 4, n° 12). L’ouverture est légèrement ovalaire (hauteur : 90 mm, diamètre au fond : 80 mm, largeur à l’ouverture : 100 mm à 120 mm, épaisseur des parois : 6 mm).

L’étonnante préservation de l’ensemble des connexions anatomiques peut paraître ici paradoxale dans une vaste chambre construite supposant une décomposition en espace vide. Peut-être ce phénomène est-il simplement dû à une 77

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 4. La Jardelle, Dissay. Mobilier de parure, outillage lithique et céramiques (dessins J.-P. Pautreau et M. Asencio Alcade). 78

J.-P. Pautreau, B. Farago-Szekeres & P. Mornais: La nécropole néolithique de la Jardelle à Dissay (Vienne, France)

infiltration progressive du sédiment, mais il pourrait aussi signaler la présence d’un quelconque moyen de contention autour des corps. Enfin, on ne peut manquer de s’interroger sur les proportions inattendues de la chambre sépulcrale par rapport à la place qu’occupent les deux petits corps. Peutêtre peut-on y voir un aménagement spacieux destiné par exemple à entreposer des biens d’accompagnement pour les défunts, ou bien encore un caveau prévu au départ pour accueillir à terme d’autres enfants ?

Le monument E se trouve à environ 4 m de l’enclos circulaire du monument A. Il est délimité par un fossé circulaire mesurant environ 25 m de diamètre. La largeur de la tranchée varie entre 1,50 m et 1 m pour une profondeur de 0,60 m à 0,90 m. Le fossé, profilé en «U» présente un comblement homogène de sédiments brun à marron contenant un cailloutis assez abondant. Ils sont issus tant de l’intérieur que de l’extérieur de la structure et séparés par un cailloutis axial.

Le mobilier retrouvé dans la partie haute du comblement comprend l’extrémité proximale d’une petite lame de silex brûlée et uniquement quatre tessons ; ceux-ci appartenant à des récipients différents sont tous des fragments de bord (ce qui semble difficilement être imputable au seul hasard). Trois d’entre eux correspondent à des vases à paroi sub-verticale avec un léger éversement au niveau du rebord à lèvre arrondie (fig. 3, n°1-3). Le quatrième récipient est un vase épais à paroi supérieure rentrant, court bord éversé et lèvre ornée d’impressions au doigt (fig. 3, n°4). La datation des ossements, entre 4448 et 4177 av. J.-C. (Ly594/Oxa7301), indique le début du Néolithique moyen, ce qui est en accord avec les rares vestiges mobiliers. Les deux enfants ont été déposés en dehors de la fosse centrale mais en limite du tertre. Nous ne pouvons pas savoir s’il s’agit là d’un traitement particulier consécutif à leur mort simultanée ou non. Quand la sépulture centrale du même tertre est réutilisée au Néolithique récent, toutes les classes d’âges s’y rencontrent, mais c’était peut-être aussi le cas auparavant. La relation génétique entre ces deux jeunes défunts mériterait d’être recherchée.

Le matériel archéologique est inégalement réparti ; seuls de très rares éclats de silex et tessons se rencontrent en surface. Quelques pierres chauffées, tessons, silex, un fragment de meule et des charbons de bois étaient concentrés dans la zone axiale. Ce fossé circulaire semble avoir fonctionné comme tranchée de fondation d’une palissade. L’absence de fosse sépulcrale creusée dans le substrat s’explique très vraisemblablement par sa disparition avec l’arasement du tumulus. Dans le secteur sud-ouest, la tranchée s’élargit en une vaste structure circulaire mesurant 1,70 m de diamètre et plus 90 cm de profondeur. A la base, des blocs de pierre disposés grossièrement en carré montrent qu’il s’agit de l’emplacement d’un énorme poteau qui a nécessité un calage particulier. Le remplissage, similaire à celui du reste du fossé avec un cailloutis axial bien net, montre bien la contemporanéité avec le reste du fossé. Deux sépultures individuelles d’adultes étaient implantées au sud et à quelques mètres du tertre C.

Figure 5. La Jardelle, Dissay. Sépultures du quart nord-ouest de la sépulture du monument C (étude et dessins B. Farago-Szekeres). 79

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

La première (95-4), complètement arasée, se trouve à deux mètres du monument. Il subsistait une zone sub-ovalaire d’environ 0,80 m à 0,60 m de large et 1 cm de profondeur, constituée par un sédiment brun clair comportant un abondant cailloutis. En surface se trouvaient les débris d’une poterie. Il s’agit d’un récipient à fond plat, en forme de tonnelet (fig. 4, n°13). Le bord, sub-vertical, très légèrement rentrant, prolonge la panse et se termine par une lèvre amincie à l’intérieur. Le vase mesure 110 mm de haut, 115 mm de diamètre à l’ouverture et 125 mm de diamètre maximum. Il est constitué par une pâte brune à dégraissant micacé et semble monté au colombin. Plaqués sur le substrat calcaire subsistaient une dent humaine, des fragments d’os long et au centre un os iliaque complètement écrasé. Il reste impossible de préciser la position du défunt. Une analyse C14 sur les ossements place la sépulture entre 3518 et 3137 av. J.-C. en datation calibrée (GRA-17365, Ly-1344). Le style céramique et la datation absolue sont concordants pour une attribution au Néolithique récent.

un faciès septentrional. Quelques ressemblances sont observées avec le mobilier du grand bâtiment de La Tricherie (Louboutin et alli, 1998). Le cimetière de la Jardelle n’est plus utilisé après la fin du 4ème millénaire. La fouille partielle de cette nécropole soulève bien des questions d’ordre culturel et chronologique notamment. Il serait aussi particulièrement intéressant de savoir si ses utilisateurs appartiennent à la même communauté ethnique, culturelle ou religieuse que les contemporains utilisateurs et constructeurs de mégalithes sur le même territoire géographique.

Adresses des auteurs Jean-Pierre PAUTREAU Directeur de recherches au CNRS UMR 6566 Université de Rennes Rennes FRANCE E-Mail : [email protected]

La seconde (95-5), à environ 2 m au sud de la précédente, a été entièrement détruite par les travaux agricoles. Il n’en subsiste qu’un mélange d’ossements humains, appartenant apparemment à un jeune adulte (15-20 ans) inhumé en position allongée, et de plaquettes calcaires. Ces dernières sont identiques à celles constituant le pavage des fosses sépulcrales des monuments C et D. Certaines étaient encore en place sur le substrat. Aucun vestige mobilier n’a été recueilli.

Bernard FARAGO-SZEKERES Archéologue AFAN 18, rue des Trois Rois F-86000 Poitiers FRANCE Patricia MORNAIS Archéologue AFAN Collaborateur à l’UMR 6566 du CNRS 18, rue des Trois Rois, F-86000 Poitiers FRANCE

Les tertres de Dissay La fouille partielle de la nécropole de la Jardelle permet de distinguer plusieurs phases de construction. L’ensemble doit être considéré comme le cimetière d’une petite communauté ayant été utilisé sans discontinuité durant près d’un millénaire entre le milieu du 5ème millénaire et le milieu du 4ème millénaire av. J.-C. Les monuments les plus anciens sont les tertres allongés (A et B) qui se se rattachent au monuments «de type Passy» bien connus en Bourgogne, sur le site de Richebourg (Yonne) mais aussi sur le site des Réaudins à Balloy (Seine-et-Marne) où ils appartiennent à la culture de Cerny (Mordant, in Guilaine 1998), tout comme en Normandie (Chancerel, Desloges puis, in Guilaine 1998, p. 91) . Localement ces monuments sont peut-être plus nombreux qu’il paraît, pouvant constituer la phase primitive de constructions plus imposantes (Burnez, Louboutin 1999) et leur assimilation à une influence des cultures du Bassin Parisien n’est pas prouvée. L’utilisation des coffres en pierre se rapporte probablement à une tradition de l’ouest de la France (Normandie, Bretagne, Centre-Ouest, Chancerel, Desloges puis Verjux, Simonin, Richard, in Guilaine 1998, p. 91, 61), même si elle est observée précocement bien ailleurs (Moinat, in Guilaine 1998, p. 129). Localement, elle est attestée dans la première moitié du 5ème millénaire à la Goumoizière (groupe de Chambon). Les monuments allongés de la nécropole n’ont pas été remaniés au contraire des monuments sub-circulaires, dont la phase initiale appartient à la même période. Probablement au Néolithique moyen II, on aménage leurs chambres avec un pavage. Ils vont dès lors recevoir un assez grand nombre de défunts. Ces mêmes tombeaux sont encore utilisés au Néolithique récent même si à ce moment on pratique sur le site des sépultures individuelles. Les quelques vestiges recueillis permettent de rattacher les gens du Néolithique récent à la mouvance «Vienne-Charente», probablement dans

Bibliographie BURNEZ, C., LOUBOUTIN, C., 1999, Le long tumulus du Cruchaud à Sainte-L’Heurine (Charente-Maritime), Bull? Soc. Préhist. Fr., 96, 3, p. 442-443, 2 fig. GUILAINE, J., dir., 1998, Sépultures d’Occident et genèses des mégalithismes (9000-3500 avant notre ère). Séminaire du Collège de France, Errance, Paris, 206 p. LOUBOUTIN, C., OLLIVIER, A., CONSTANTIN, C., SIDERA, I., TRESSET, A., FARUGGIA, J.-P., 1998, La Tricherie à Beaumont (Vienne), un site d’habitat du Néolithique récent, Le Néolithique du Centre-Ouest de la France, Actes du XXIe Colloque inter-régional sur le Néolithique, Poitiers, 14-16 octobre 1994, A.P.C., Mémoire XIV, 1998, p. 307-325, 12 fig. PAUTREAU, J.-P., 1997, Un nouveau tumulus allongé, Dissay (86), L’Archéologue, 33, 1997, p. 73-74, 1 fig. PAUTREAU, J.-P., 1997, Une structure de type Passy-Richebourg à la Jardelle de Dissay (Vienne), Journée préhistorique et protohistorique de Bretagne, UMR 6566 « Civilisations Atlantiques et Archéosciences », Rennes, 15 novembre 1997, résumés des communications, p. 41-42, 1 fig. PAUTREAU, J.-P., 1999, Fouille d’un enclos allongé à sépulture en caisson depierre à Dissay (Vienne), in Projet collectif de recherche, La néolithisation du Centre-Ouest de la France, rapport triennal 1996-1998, G.N.A.C.O., 1999, p. 57-60, 1 fig. PAUTREAU, J.-P., 2000, Dissay, La Jardelle, Néolithique, Funéraire, Bilan scientifique 1997, Service Régional de l’Archéologie, Direction Régionale des Affaires Culturelles Poitou-Charentes, 2000, p. 54 et 56. UBELAKER, D. H., 1978, Human Skeletal Remains. Chicago, Aldine Publishing Company. 80

D. Bosquet et al.: La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999)...

LA NÉOLITHISATION EN HESBAYE (BELGIQUE) : APPORTS DES FOUILLES RÉCENTES (1996-1999) SUR LE TRACÉ ORIENTAL DU TGV ET LE SITE DE ANS, À ALLEUR Dominique BOSQUET, Heike FOCK, Claire GOFFIOUL, Jean-Philippe MARCHAL & Dimitri PREUD’HOMME

Résumé : Le présent article passe en revue les données acquises lors des fouilles entreprises sur 7 sites rubanés entre 1996 et 2000, sur le tracé du TGV et le site de Ans, à Alleur. L’étude préliminaire de ces données permet de tracer quelques axes de réflexion sur le fonctionnement socio-économique des premiers groupes néolithiques dans la région. Abstract: The present article reviews the data acquired during the excavation campaigns undertaken on 7 LBK sites between 1996 and 2000, on the High Speed Train tracing and the site of Ans, in Alleur. The preliminary survey of these data permits the development of hypotheses concerning the socioeconomic structure of the first Neolithic groups in the region.

INTRODUCTION

LE SITE DE WAREMME-VINÂVE

Le Rubané est abondamment documenté chez nous et fait l’objet de recherches depuis plus d’un siècle. Cependant, ce n’est que récemment que l’on a procédé à des fouilles extensives embrassant des surfaces de plusieurs milliers de m² permettant ainsi une vision globale de l’organisation spatiale de l’habitat. Les premières fouilles de ce type ont été réalisées entre 1981 et 1989 par l’Institut royal des Sciences naturelles de Belgique, sur quatre sites principalement, situés sur le haut Geer autour de Waremme : Darion-Colia, Oleye-Al Zèpe, Waremme-Longchamps et Hollogne-Douze Bonniers (Cahen et al., 1990 ; fig. 1 a). Depuis, d’autres fouilles ont eu lieu en Hesbaye liégeoise (Caspard et al., 1992 ; Burnez-Lanotte & Allard, 1997 ; Frébutte & Marchal, 1998 ; fig. 1 a) mais aussi en Hainaut occidental (e. a. Constantin et al., 1979 et 1980), en Limbourg (Roosens, 1961) et en Brabant flamand (Lodewijckx, 1984 et 1990), portant à une trentaine le nombre de sites fouillés extensivement en Belgique.

Situation : coord. Lambert belges : 156,700 km nord / 214,600 km est. Superficie dégagée : 1.068 m². Nombre de structures néolithiques : 8. Nombre de maison : 1. Datation relative (étude préliminaire) : transition Rubané moyen-récent, sur base de la présence majoritaire de silex grenu au sein de l’assemblage lithique et des décors céramiques. Datation C14 : OxA-6688 : 6205 ± 45 BP, réalisée sur pisum sativum (Jadin, 1999 : 4-25 et 6.1-52). Matériel : un percuteur sur nucléus utilisé comme broyon à ocre. Bibliographie sélective : Bosquet & Fock, 1996 ; Jadin, 1999. Les quelques vestiges découverts à Waremme-Vinâve sont probablement liés à un village qui s’étendait au nord de l’emprise du TGV, c’est-à-dire à l’emplacement de l’autoroute E40 Bruxelles-Liège. Il est possible qu’une partie de l’occupation soit conservée au nord de l’E40, bien que les prospections de surface menées de ce côté soient négatives jusqu’à présent (J. Haeck, comm. pers.).

En 1995, en prélude à la construction du tronçon BruxellesLiège de la ligne du TGV, la Direction de l’Archéologie du Ministère de la Région wallonne, en collaboration avec l’Institut royal des Sciences naturelles de Belgique pour la Préhistoire, entreprend une vaste opération de sauvetage archéologique. A cette occasion, six sites rubanés, dont certains étaient connus mais n’avaient jamais été exploités, sont mis au jour (Bosquet & Fock, 1996 ; Bosquet & Preud’homme, 1998 ; Bosquet et al., 1997a, 1997b, 1998 et à paraître (a) ; Fock et al., 1998 ; Goffioul et al., 1999 ; Preud’homme et al., 1997 ; fig. 1 a). Conjointement, l’extension d’une zone industrielle sur le domaine militaire d’Ans, à Alleur occasionne la fouille, par l’ASBL “ Les Chercheurs de la Wallonie ”, d’un site repéré en prospection dans les années ’60 (Marchal, 1998 et 1999 ; fig. 1 a). Outre une présentation succincte de chaque village, le présent article se propose de mettre en exergue les apports fournis par les travaux récents, qu’ils soient originaux ou complémentaires aux acquis issus des fouilles anciennes.

LE SITE DE DE REMICOURT-TOMBE DE HODEIGE Situation : coord. Lambert belges : 154,700 km nord / 220 km est. Superficie dégagée : 838 m². Nombre de structures néolithiques : 4. Datation relative (étude préliminaire) : transition Rubané moyen-récent, sur base de la présence majoritaire de silex grenu au sein de l’assemblage lithique et des décors céramiques (phase II b à II c de Modderman, 1970). Matériel : vases d’enfant ou d’apprenti, un rebut issu de la fabrication de dégraissant, sinon assemblage classique pour la période. 81

D. Bosquet et al.: La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999)...

Bibliographie sélective : Preud’homme et al., 1997.

l’ouverture, tandis que l’absence de traces de poteau indiquerait que la palissade a été démontée. La raison d’une division de l’entrée du village en deux parties égales reste à expliquer, de même que les modalités du démantèlement du dispositif. Notons l’existence d’un aménagement identique sur le site de Fexhe-le-Haut-Clocher,Voroux-Goreux. En section, le fossé oriental comporte toutes les caractéristiques classiques des fossés rubanés, mais il faut cependant noter une originalité par rapport aux autres exemplaires connus dans la région. En effet, plusieurs profils montrent que le fond du fossé oriental a été vidé de sa boue peu de temps après son ouverture, étant donné la faible épaisseur des dépôts remaniés. Un décapage horizontal au niveau voulu a montré que ces traces n’étaient pas à mettre en relation avec la présence d’une palissade implantée au fond de l’ouvrage, comme dans le fossé de Langweiler 3 (Eckert et al., 1972 : 362-363). Ce geste, qui n’a pas été répété à un stade ultérieur du comblement, ne s’explique pas de façon univoque. S’il peut correspondre à la volonté d’entretenir le fossé au début de son utilisation, on pourrait également évoquer un prélèvement d’argile décantée pour d’autres usages, tels que la fabrication de poterie ou de torchis. Le fossé oriental était le seul à contenir du matériel archéologique. Les extrémités situées de part et d’autre de l’entrée ont toutes deux livré une meule entière posée à proximité du fond, ainsi qu’un vase complet au nord et un exemplaire plus fragmentaire au sud, situés un peu plus haut dans le remplissage.

A Remicourt-Tombe de Hodeige, le petit nombre de structures contraste avec la richesse et la variété du matériel. A ce titre, la présence d’éléments attribués au Rubané moyen retient plus particulièrement l’attention, dans la mesure où on en retrouve sur les sites de Remicourt-En Bia Flo II, de Remicourt-Fond de Momalle, secteurs I et III et de Fexhele-Haut-Clocher-Podrî l’Cortri, situés respectivement à 1, 2 et 3 km plus à l’est (voir ci-dessous). Signalons encore la présence d’autres fosses néolithiques à 80 m au sud de l’emprise du TGV, repérées en prospection. LE SITE DE REMICOURT-EN BIA FLO II Situation : coord. Lambert belges : 154,350 km nord / 221,200 km est. Superficie dégagée : 6.847 m². Nombre de structures néolithiques : 235. Nombre de maison : 5. Datation relative (sériation préliminaire) : maison V extramuros : Rubané moyen à récent (phases I b à II d de Modderman, 1970), maisons intra-muros : Rubané récent (phases I b à II d) et final sur base de la présence de décors au peigne pivotant. Matériel : nombre important de meules et molettes, y compris un “ moulin à ocre ” complet, sinon assemblage classique pour la période. Bibliographie sélective : Bosquet et al., 1997b ; Bosquet & Preud’homme, 1998 ; Bosquet et al., à paraître a et b.

Le fossé occidental est implanté à 8 m à peine du fond du vallon de Bia Flo et traverse l’emprise TGV d’un bord à l’autre sans interruption. Au-delà de l’enceinte, le versant du vallon de Bia Flo a été entièrement évalué et n’a livré aucune trace anthropique. Les coupes pratiquées dans le fossé révèlent également un profil en “ V ” évasé et des dépôts similaires à ceux du fossé oriental, les traces de curage en moins.

L’enceinte Le fossé délimite l’aire villageoise à l’ouest et à l’est (fig. 1 b). Les deux tronçons repérés dessinent une courbe qui se referme à une vingtaine de mètres au nord de la surface dégagée, c’est-à-dire à l’emplacement actuel de l’autoroute, tandis que la plus grande partie de l’enceinte s’étend au sud, en dehors de l’emprise du TGV.

Les vestiges découverts ne permettent pas de déterminer de façon fiable la fonction de cette enceinte. Tout comme à Darion, il est fort probable qu’il s’agisse d’un ouvrage à buts multiples, sans prépondérance décelable de l’un de ceux-ci sur les autres, du moins jusqu’à présent.

Le fossé oriental est creusé à la rupture de pente entre les deux parties du versant, comme le fossé est de Darion (Cahen, 1985 : 75). Il a été suivi sur 50 m et mesure 3,50 m de largeur moyenne au niveau du décapage. L’extrémité sud est marquée par une interruption de 4,50 m de large, interprétée comme une entrée du village. Cette ouverture est aménagée d’une Schlitzgrube, creusée au milieu de l’interruption et perpendiculairement au tracé du fossé. Ce type de fosse est connu sur plusieurs sites danubiens, entre autres sur le proche plateau d’Aldenhoven en Allemagne (Boelicke et al., 1988 : 64), mais aussi à Darion (Cahen et al., 1990), sur le secteur III du site de Remicourt-Fond de Momalle et à Fexhe-leHaut-Clocher,Voroux-Goreux (voir ci-dessous). Ici, la fosse peut être interprétée comme la tranchée de fondation d’une petite palissade qui partageait le portail d’entrée en deux passages de 2,25 m de large, bien que les coupes pratiquées dans la structure n’aient pas révélé de trace de poteau. Par contre, la présence dans le remblai de blocs des horizons de surface E et B dark dont la forme est encore perceptible, montre que la fosse a été rebouchée très peu de temps après

La palissade Composée de 11 trous de poteau, la palissade longe le bord intérieur du fossé oriental sur une distance de 13 m. L’absence de trous de poteaux de part et d’autre du tronçon découvert s’explique par l’érosion qui affecte le secteur. Le fossé occidental ne semble pas avoir été doublé d’une palissade. En effet, l’érosion nettement moindre ici qu’à hauteur du fossé oriental, ne peut expliquer à elle seule l’absence de poteaux. L’existence d’une simple levée de terre, sommée ou non d’une palissade, peut être discutée, mais l’étude des profils, qui ne montrent aucune asymétrie dans le remblai amont du fossé, ne plaide pas en faveur de cette hypothèse. Les maisons Elles sont au nombre de cinq, y compris une habitation isolée à 130 m à l’est de l’enceinte (M V). Tous les bâtiments 83

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

présentent des plans très fragmentaires et l’érosion, qui atteint 80 cm en moyenne, ne suffit pas à expliquer cet état de chose. Rappelons qu’à Darion, pour un taux d’érosion équivalent, les plans sont quasi complets (Langohr & Sanders, 1985 : 23). Dès lors, il faut admettre que les constructeurs de Remicourt ont enfoncé leurs poteaux moins profondément.

Datation relative (étude préliminaire) : Rubané moyen et récent, sur base des décors céramiques. Matériel : un tesson orné d’un décor fait d’un cordon pincé appliqué en spirale, sinon matériel particulièrement pauvre par rapport au nombre de structures. Bibliographie sélective : Bosquet et al., 1997a.

Les fosses-silo

L’ensemble ne représente qu’une partie d’un village qui s’étend considérablement au sud de l’emprise (voir Fond de Momalle, secteur III, ci-dessous et fig. 2 a). Une des maisons possède une tierce en “ Y ”, trait architectural typique des phases ancienne et moyenne du Rubané, peu attestées dans la région. Le décalage d’orientation entre les deux maisons, les recoupements entre fosses ou encore la proximité des structures entre elles, particulièrement au centre de la surface dégagée, semblent indiquer deux phases d’occupation ou de construction du village. L’hypothèse est appuyée par le matériel céramique attribuable au Rubané moyen et récent.

Les structures d’ensilage (Villes, 1981; Tapret & Villes, 1996 ; Cahen, 1985) sont au nombre de 43 sur un total de 182 fosses, soit 23 %. Elles se répartissent sur toute l’aire villageoise, y compris à l’intérieur de certaines habitations et leur remplissage varie. Pour certaines, la présence de cendres blanches au sommet de rejets massifs de charbon de bois indiquerait la présence de foyers qui, une fois recouverts de terre, ont continué à se consumer. D’autres silos étaient massivement remblayés de terre brûlée. Une des fosses contenait des fragments d’une épaisseur moyenne de 4 cm, dont la forme courbe évoque les restes d’une coupole de four. Enfin, une grande quantité de meules et de molettes entières ou fragmentaires, ainsi que quelques lames de faucille ont été recueillies dans plusieurs fosses et dans le fossé oriental. Ce faisceau d’éléments nous a amené à interpréter la partie du village mise au jour comme une aire de traitement des céréales : torréfaction ou séchage, mise en stock, mouture et fabrication du pain. Les résultats préliminaires de l’étude palynologique confirment cette hypothèse, dans la mesure où des pollens de céréales sont présents dans les fosses-silo, dans des proportions qui peuvent atteindre près de 50 % des espèces représentées dans l’échantillon (Munaut et Defgnée, rapport inédit).

Les activités artisanales Plusieurs éléments témoignent d’activités artisanales spécifiques. Ainsi, des fosses au profil en “ marches d’escalier ” comportent un premier remplissage résultant de la décomposition d’une quantité importante de matière organique. L’analyse palynologique d’un de ces dépôts indique un taux important de pollens de céréales, soit 50 % des espèces présentes, de même qu’une altération des pollens encore inexpliquée, révélée par leur aspect ridé (Defgnée, comm. pers.). Pour d’autres fosses, la nature des dépôts initiaux et des phénomènes annexes associés témoignent de la présence prolongée d’eau, évoquant des activités telles que le tannage des peaux, le rouissage des fibres végétales ou, plus classiquement, la fabrication du torchis.

Le matériel L’assemblage lithique comprend un nombre très important de meules et de molettes classiques, mais également un moulin à ocre, retrouvé appareillé dans le fossé oriental. Il s’agit un broyeur aménagé sur un nucléus-percuteur en silex comparable à celui du site de Waremme-Vinâve, associé à une meule dormante de type classique, mais densément chargée d’ocre.

LE SITE DE REMICOURT-FOND DE MOMALLE, SECTEUR III Situation : coord. Lambert belges : 153,900 km nord / 222,150 km est. Superficie dégagée : 6.050 m². Nombre de structures néolithiques : 515, dont 237 ont été recoupées. Nombre de maisons : 10. Datation relative (étude préliminaire) : Rubané ancien et moyen sur base des décors céramiques et de l’absence du silex gris à grain fin de Hesbaye au sein du matériel lithique, issu exclusivement d’un silex grenu ou moucheté. Matériel : un vase d’enfant ou d’apprenti, sinon on observe la même pauvreté en matériel archéologique que sur le secteur I de Fond de Momalle. Bibliographie sélective : Fock et al., 1998.

L’étude de la céramique révèle une différence stylistique entre les décors des vases, selon qu’ils proviennent de fosses situées dans l’enceinte ou hors de celle-ci. Les décors extra-muros, réalisés au poinçon, évoquent fortement le Rubané moyen à récent (fig. 1 c) alors que les décors intra-muros indiquent un Rubané récent et final, avec des décors parfois extrêmement couvrants, réalisés au poinçon ou au peigne à quatre dents pivotant (fig. 1 d). LE SITE DE REMICOURT-FOND DE MOMALLE, SECTEUR I

Le secteur fouillé (fig. 2 a) constitue probablement le prolongement de l’occupation de Fond de Momalle, secteur I. Il couvre l’extrémité orientale d’un village dont une partie importante, vers l’ouest et au sud, reste à explorer. Compte tenu de la périodisation relative des deux secteurs, on pourrait proposer un déplacement de l’occupation du sud vers le nord.

Situation : coord. Lambert belges : 153,900 km nord / 222,150 km est. Superficie dégagée : 2.728 m2. Nombre de structures néolithiques : 171. Nombre de maisons : 2. 84

D. Bosquet et al.: La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999)...

85

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Les maisons

1973). Les analyses des prélèvements palynologiques, archéobotaniques et pédologiques apporteront sans doute des informations complémentaires.

Dix habitations ont été relevées. L’ensemble se distingue par la répartition régulière des unités d’habitations et leur organisation spatiale “ en rue ”, agencement tout à fait original pour le Rubané hesbignon et comparable à celui du village de Cuiry-lès-Chaudardes (Aisne, France) attribué au Rubané récent du bassin parisien (Ilett et al., 1986). La disposition spatiale quasi parcellaire est accentuée par l’absence de structures archéologiques entre les différentes unités “ maison/fosses associées ” et par l’absence de recoupements entre celles-ci. Les orientations identiques de M I, M V, M VII et M X, et surtout de M IV et M VIII situées aux extrémités opposées du décapage, renforcent l’homogénéité de l’ensemble. La conservation imparfaite des plans des maisons s’explique en partie par un taux d’érosion croissant du sud-ouest au nord-est. Toutefois, la disparition fréquente des façades par opposition à la bonne conservation des poteaux de tierce et des poteaux de paroi, habituellement moins profondément implantés, témoigne clairement d’un choix technique délibéré des constructeurs. Le module “ espace central/couloirs adjacents ” se distingue par sa grande uniformité alors que la superficie des espaces avant et arrière, la disposition de la façade et l’aménagement du chevet sont plus difficilement saisissables, mais paraissent également obéir à une certaine normalisation. Il est intéressant de constater que les chapelets de fosses de construction s’arrêtent en général à hauteur de la seconde tierce, soit à hauteur du couloir avant : cette caractéristique commune fut d’ailleurs prise en considération comme critère supplémentaire lors du positionnement de certaines maisons dans le diagramme (fig. 2 b). Les chevets de M IV et M X conservent les vestiges d’une tranchée de fondation partielle. Cette technique de construction ne semble cependant pas uniformément appliquée : ainsi, l’extrémité occidentale de M VII, confinée par la disposition des fosses alentour, est formée de poteaux plus imposants, espacés suivant le même rythme que les tierces et qui alternaient probablement avec des poteaux d’un moindre calibre. La M VI se distingue du schéma observé par une disposition apparemment moins cohérente des tierces, par la présence de deux fosses au sein du bâtiment et par une multiplication exceptionnelle des poteaux, indices probables d’une réfection ou d’une double occupation du secteur.

Les trous de poteau Deux trous de poteau se distinguent par des dimensions particulièrement imposantes, tant en plan qu’en section. Leur implantation un peu à l’écart des habitations semble renvoyer à une fonction spécifique. LE SITE DE FEXHE-LE-HAUT-CLOCHER-PODRÎ L’CORTRI Situation : coord. Lambert belges : 153,450 km nord / 223,400 km est. Superficie dégagée : 8.900 m². Nombre de structures néolithiques : 410, dont 31 ont été recoupées. Nombre de maison : 7. Datation relative (étude préliminaire) : maison I isolée : Rubané moyen-récent ; maisons regroupées : Rubané récent et final sur base de la présence de décors au peigne à quatre dents traîné. Matériel : vases d’enfant ou d’apprenti, Céramique du Limbourg et un vase rubané à la forme et au décor inconnus à ce jour, brai de bouleau sur quatre vases. Bibliographie sélective : Bosquet et al., 1998 et 2001. La maison I La maison I est sans nul doute l’habitation rubanée la mieux conservée découverte sur le tracé du TGV oriental (fig. 3 a). Elle se distingue par la présence de deux tierces disposées en oblique prononcée par rapport à l’orientation générale du bâtiment, originalité que rien n’explique pour l’instant, mais remarquée sur le secteur III du site tout proche de Remicourt-Fond de Momalle. En outre, sa situation isolée, à une cinquantaine de mètres par rapport au groupe de maisons qui forme le village proprement dit, rappelle celle de la maison V du site rubané de Remicourt-En Bia Flo II. Seuls 10 trous de poteau de la maison I ont pu être recoupés. Le contraste entre le diamètre très important de certains trous de poteau et la faible profondeur d’enfoncement est frappant, d’autant plus que l’érosion est quasi nulle à cet endroit.

Les fosses-silo Elles se situent toutes à proximité des maisons, soit le long de la paroi nord, soit à l’angle nord-est du bâtiment, proche de l’entrée.

La céramique L’assemblage céramique issu d’une fosse longeant la paroi nord de la maison I se révèle complexe et particulièrement intéressant. Nous n’insisterons pas ici sur les vases rubanés classiques, tous compris entre les phases I b et II d de Modderman (1970 ; fig. 3 b). La Céramique du Limbourg, quant à elle, est illustrée par au moins deux individus (fig. 3 c). Le premier vase est représenté par un tesson dégraissé à l’os et portant un décor en arêtes de poisson qui ne laisse aucun doute quant à l’attribution culturelle. Pour le second tesson, c’est surtout l’épaississement du bord qui assure son appartenance à la Céramique du Limbourg. Le décor

La Schlitzgrube Une Schlitzgrube est implantée dans le nord-ouest, relativement isolée des maisons, comme sur le site alsacien d’Hensisheim, entre autres (Jeunesse & Sainty, 1992 : 96). Son remplissage formé de gros blocs de substrat atteste un remblai rapide. Aucun phénomène annexe (précipités, altération quelconque du substrat, ...) ne vient étayer une interprétation comme fosse de tannage des peaux ou “ tanpit ”, telle que proposée par P. van de Velde (van de Velde, 86

D. Bosquet et al.: La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999)...

87

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

L’enceinte

principal, un motif en angles déterminé par des rangs de points au poinçon traîné, est atypique mais un exemple comparable provenant d’Aubechies est réattribué par P.-L. van Berg au Groupe de Blicquy (van Berg, 1990 : 192). Enfin, deux tessons forment les vestiges d’un vase tout à fait original, à pâte noire et lustré sur la face externe (fig. 3 d). La finesse du dégraissant et l’homogénéité de la cuisson montrent une facture de bonne qualité. Le bord du vase est déversé à 90° par rapport à la panse, particularité absente du corpus néolithique ancien du Nord-Ouest connu, en ce compris les styles Limbourg et Hoguette. Le décor principal, composé de panneaux de quatre cannelures verticales à section en “ U ”, est à ce point profondément imprimé qu’il est visible sur la face interne du vase ; les panneaux sont liés par un motif similaire en miniature, disposé en métopes. Ce vase ne peut être rattaché directement aux styles non rubanés connus et clairement définis que sont la Céramique du Limbourg et la Céramique de La Hoguette. Seule une affinité avec la Céramique du Limbourg peut être reconnue dans la verticalité du décor et la disposition en panneaux (van Berg, 1990 : fig. 20 : nos 1, 2, 3 et 6). Par contre, l’association proposée par Chr. Jeunesse entre la Céramique d’accompagnement, de La Hoguette, du Limbourg et certains groupes néolithiques méridionaux nous a permis de faire les rapprochements les plus convaincants à ce jour : les styles céramiques du Néolithique ancien provençal comportent effectivement des décors en cannelures verticales, dont certains sont disposés en métopes (Jeunesse & Sainty, 1991 : 25 ; Jeunesse, 1994 : 15-16 ; Binder & Courtin, 1986 : 86, fig. 2 et 4). Une zone de contact possible, de plus en plus souvent évoquée dans la littérature, est l’axe rhodanojurassien. Une étude consacrée au sujet (Manen, 1997) replace dans un cadre large les tessons décorés de cannelures de la couche 5 de Bavans, qui viennent spontanément à l’esprit en examinant le tesson de Podrî l’Cortri (Manen, 1997 : 134, fig. 31, nos 7 et 8).

Le fossé qui ceinture le village délimite une aire ovale d’environ 120 m de long pour une largeur de 100 m (fig. 3 e). L’interruption, repérée dans la partie nord de l’enceinte et large de 3,50 m, a été interprétée comme une entrée. En son centre, une fosse étroite et allongée de type Schlitzgrube est creusée perpendiculairement à l’axe du fossé, selon un dispositif identique à celui mis en évidence sur le site de Remicourt-En Bia Flo II. Les deux campagnes de fouilles ont permis de suivre la portion nord de l’enceinte sur une longueur de 60 m. Ce tronçon, d’une largeur moyenne de 2,50 m pour une profondeur comprise entre 1,40 et 1,70 m, présente les caractéristiques habituelles des fossés rubanés, deux particularités mises à part. Une des coupes révèle un profil en cuvette profond d’à peine 1 m, trait particulier déjà observé à Vaux-et-Borset-Gibours (Caspar et al., 1992 : 77). De plus, un rejet de terre noire organique mélangée à des boulettes de substrat limoneux forme le dépôt initial de part et d’autre de l’entrée, sur une longueur d’au moins 60 m, l’épaisseur de cette couche diminuant au fur et à mesure que l’on s’éloigne de l’interruption. Un remplissage similaire est présent au fond de deux autres fosses, dont la Schlitzgrube. Pour l’heure, l’origine de ce dépôt reste incertaine. Deux hypothèses sont envisagées : soit un dépôt de fondation, interprétation renforcée par la présence de la bouteille complète retrouvée (fig. 3 f), soit un rejet détritique organique. En dehors de la fine couche noire, le remplissage des Schlitzgruben de Voroux-Goreux et de Remicourt-En Bia Flo II est très comparable et pose les mêmes problèmes, notamment l’absence de traces de poteaux. Comme pour Remicourt, la fonction de l’enceinte de Voroux-Goreux est pour l’instant malaisée à déterminer. La palissade Elle est matérialisée par un alignement de 5 trous de poteau, au gabarit standardisé. L’alignement est conservé sur une longueur de 11,20 m, parallèlement au fossé et distant de celui-ci d’environ 3 m. L’aspect incomplet de la palissade est très probablement dû à l’érosion importante qui affecte cette partie du site. Notons la présence d’une fosse-silo qui recoupe l’axe supposé de la palissade, vestige d’une phase d’occupation ou de construction ultérieure.

Le brai de bouleau L’identification de brai de bouleau sur quatre vases issus d’une fosse du village est une première pour le Rubané récent du Nord-Ouest. La préparation de l’adhésif, de même que son utilisation pour réparer les vases selon diverses modalités, semblent avoir fait l’objet d’une spécialisation, dans la mesure où tous les vestiges issus de ces activités étaient concentrés dans cette seule fosse.

La céramique Mentionnons la découverte des fragments d’une bouteille complète, dans le fossé à hauteur de l’entrée. Le vase était posé sur l’important rejet organique dont il a été question. L’analyse des résidus observés sur les parois du vase a été confiée à M. Regert (Laboratoires des Musées de France, Paris).

LE SITE DE FEXHE-LE-HAUT-CLOCHER, À VOROUX-GOREUX Situation : coord. Lambert belges : 150,800 km nord / 225,400 km est. Superficie dégagée : 2.200 m². Nombre de structures néolithiques : 48. Nombre de maison : 1. Datation relative (étude préliminaire) : Rubané récent. Matériel : une bouteille complète dans le fossé à hauteur de l’interruption, sinon assemblage classique pour la période. Bibliographie sélective : Goffioul et al., 1999.

LE SITE DE ANS, À ALLEUR Situation : coord. Lambert belge : 152,600 km nord / 232,500 km est. Superficie dégagée : env. 5.000 m². Nombre de structures néolithiques : 253. 88

D. Bosquet et al.: La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999)...

CONCLUSIONS

Nombre de maisons : 4. Datation relative (étude préliminaire) : transition Rubané moyen à récent et Rubané récent, sur bases des décors céramiques. Matériel : Céramique du Limbourg, sinon assemblage classique pour la période. Bibliographie sélective : Marchal, 1998 et 1999 ; Collin et Marchal, 2000.

Les nouvelles données rassemblées lors des fouilles réalisées depuis 1996 concernent tous les aspects de la culture matérielle rubanée et permettent de proposer quelques hypothèses de travail à suivre lors des études à venir. Au sujet des enceintes d’abord, l’alignement des quatre sites fossoyés connus jusqu’alors sur le haut Geer et la Mehaigne leur confère une valeur de limite de peuplement très clairement confirmée, mais également nuancée, par les résultats des fouilles TGV. En effet, les deux nouvelles enceintes sont situées nettement plus à l’est et une autre frontière régionale semble peu à peu se dessiner, adossée à la Meuse en arc de cercle (fig. 5 a). La prise en compte des aspects d’ordre socio-économique permet de documenter ce nouvel axe de réflexion. Ainsi, l’aire circonscrite par les enceintes peut être superposée aux groupes pétrographiques 3 et 4, mis en évidence par les frères Toussaint dans leur étude sur les herminettes danubiennes. Ces groupes sont caractérisés par la dominance des roches locales, contrairement aux groupes 1 et 2, dominés par les roches exogènes d’origine volcanique (Toussaint & Toussaint, 1982 : 554). En outre, le caractère spécialisé de la production dans certains villages fossoyés (Cahen et al., 1990) semble également s’affirmer. A Darion-Colia, la découverte de deux ateliers de taille a mis en évidence un débitage excédentaire par rapport aux besoins des habitants, tandis que l’approvisionnement en poteries se serait plutôt fait à l’extérieur (Jadin, 1990 : 151-152). Le site d’Oleye, très riche en documents céramiques et pauvre en production lithique, est le seul à avoir livré une structure clairement liée à la production de céramique (Cahen et al., 1990 : 134, fig. 8). A Remicourt-En Bia Flo II, c’est le nombre très important de vestiges en relation avec le traitement des céréales et leur mise en stock qui est frappant. Ainsi, les enceintes de Hesbaye, construites sur le même modèle architectural (van Berg, 1989), apparaissent peu à peu comme une entité synchrone et géographiquement cohérente. La spécialisation des productions au sein de ces villages pourrait alors refléter l’instauration d’une économie locale à tendance autarcique ; les échanges des denrées de base s’y feraient prioritairement au sein d’un terroir constitué de quelques villages proches (Bosquet et al., 1998 et à paraître (a) ; Heim & Jadin, 1998). Dans ce contexte, le caractère éminemment défensif des dispositifs d’entrée élaborés de Darion-Colia et de WaremmeLongchamps (Bosquet, 1992 et 1993), exemples jusqu’à présent uniques dans le monde rubané, traduit la position des Rubanés face à “ L’Autre ”. Celui-ci constituerait une menace potentielle, d’autant plus qu’à cette époque, probablement, le système rubané essoufflé et en passe de disparaître doit, plus que jamais, affirmer sa puissance.

Situé en zone d’activité industrielle, le site d’Alleur (fig. 4 a) a malheureusement été détruit dans ses parties ouest et nordouest, mais présente encore un important potentiel au nord et à l’est. Ces derniers secteurs font actuellement l’objet de fouilles programmées par le Service de Préhistoire de l’Université de Liège. Fait rarissime dans la région, le site se caractérise par une absence quasi totale d’érosion, ce qui a permis l’observation de structures faiblement creusées, faisant généralement défaut sur d’autres sites. Malgré la faible superficie dégagée, deux phases chronologiquement distinctes peuvent être distinguées sur base de recoupements entre structures, des plans et de l’orientation des maisons et du matériel céramique. Les maisons Trois maisons pratiquement complètes ont été mises au jour (fig. 4 a), dont la plus récente (M I) présente un plan particulièrement intéressant, caractérisé par un élargissement unilatéral de la cloison nord à sa jonction avec la façade avant et par une entrée sans doute axiale. La M III entre dans la catégorie des habitations monopartites, rares dans la région. Les aires de combustion La présence de trois aires de combustion fonctionnellement distinctes constitue un des intérêts majeurs du site (fig. 4 a : nos 1, 2 et 3). La première, implantée le long de la paroi nord de M I, présente un plan rectangulaire inhabituel pour la période. Dans l’état actuel des recherches, sa destination première ne peut être déterminée de façon précise. La seconde (fig. 4 a : n° 2), une fosse circulaire dans le remblai de laquelle ont été aménagées deux cuvettes successives, est entourée d’une dizaine de pieux, d’un diamètre moyen de 5 cm. Leur répartition sur le périmètre extérieur de la fosse ménage un accès probable du côté ouest, dépourvu de poteau. La présence de ces piquets indique soit une couverture de la structure, soit la présence d’une plate-forme à claire-voie qui aurait servi lors d’opérations de fumage. La dernière structure, enfin, a été aménagée volontairement au centre du comblement d’une fosse circulaire, implantée à l’angle sud-est de M I (fig. 4 a : n° 3). Après creusement, les parois ont été enduites de limon, tandis que la sole a été aménagée avec des fragments de terre cuite mélangés à du limon et disposés sur une épaisseur qui peut atteindre 10 cm (fig. 4 b). Une série d’expérimentations, réalisées en collaboration avec des ingénieurs spécialisés en thermodynamique, révèle que l’installation du foyer dans un remblai préexistant apportait une rentabilité calorifique supérieure d’environ 70% à celle d’un foyer en fosse classique. Ce phénomène serait dû aux propriétés isolantes de l’air compris dans la terre de remblai.

Pour l’architecture, plusieurs éléments originaux ont été mis en évidence dans les domaines des techniques de construction et de l’organisation spatiale des villages. Ainsi, contrairement aux sites connus, la plupart des nouveaux habitats révèlent un enfoncement étonnamment faible des poteaux, y compris ceux des habitations peu érodées. Ce phénomène récurrent traduit un choix délibéré de la part des constructeurs et confirme que les techniques de construction sont soumises, elles aussi, à 89

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

90

D. Bosquet et al.: La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999)...

91

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

une évolution, qu’elle soit chronologique ou “ culturelle ” (Jadin, 1999 : 2-117). Il conviendra de confronter ces observations aux résultats des travaux expérimentaux réalisés par Fernand Collin et son équipe, visant à reconstituer la “ chaîne opératoire ” de construction des maisons (Collin et al., 1998). La mise en évidence d’un modèle de bâtiment très standardisé, illustrée par la superposition des plans des maisons du secteur III de Fond de Momalle, renforce effectivement l’hypothèse de l’existence d’une telle chaîne. Mieux encore, la corrélation entre ce schéma standard et la maison I de Podrî l’Cortri, révélée par la coïncidence remarquable dans le rythme des travées, le chevauchement exact du couloir avant et surtout une obliquité identique des seconde et troisième tierces, montre que ce modèle s’applique à deux villages au moins (fig. 5 b). Le fait que ces villages couvrent des périodes différentes du Rubané régional permettrait de conclure à la persistance d’un modèle architectural du Rubané ancien à moyen au Rubané récent, impliquant alors une unité de mesure commune, par ailleurs mise en évidence pour les enceintes de Hesbaye, entre autres (van Berg, 1989).

également une première pour la période. Les études permettront d’en préciser le mode de fonctionnement et de définir le type d’activité pour laquelle ils ont été utilisés. Malgré le caractère tout à fait préliminaire de l’étude du matériel archéologique récolté sur les nouveaux sites, l’enseignement principal à en tirer concerne la typochronologie de ces habitats. Depuis longtemps, différents chercheurs ont supposé que les sites rubanés les plus anciens de notre pays se situaient à l’est du territoire. Dans ce contexte, un gradient d’est en ouest, du nombre croissant de sites de plus en plus récents, révélerait la progression de la colonisation, au départ du Limbourg néerlandais (Cahen et al., 1986 : 5-9). Dès lors, il est piquant de constater que l’opération archéologique sur le tracé du TGV BruxellesLiège n’a pas occasionné la mise au jour de sites du Néolithique ancien hors de l’aire de peuplement du Rubané de Hesbaye ; par contre, dans sa progression vers l’est, cette opération a permis la détection de sites plus anciens. Un autre aspect concerne plus spécifiquement la méthode à utiliser pour l’étude des assemblages céramiques. Les différences typo-chronologiques et culturelles mises en évidence sur les maisons isolées de Remicourt-En Bia Flo II et Fexhe-le-Haut-Clocher-Podrî l’Cortri soulèvent effectivement des questions qui nous semblent difficiles à approcher par le biais exclusif de la sériation des styles décoratifs. En effet, si la céramique reste un des documents privilégiés des chercheurs intéressés par les premières civilisations sédentaires, c’est quasi uniquement via les formes et les décors que l’on envisage son étude, dans un but essentiellement chronologique. Loin de nier la légitimité des études typologiques, il nous semble toutefois opportun d’insister sur un aspect jusqu’ici trop peu abordé de l’artisanat néolithique : la chaîne opératoire de fabrication de la poterie (Bosquet et al., à paraître b ; Livingstone Smith, 1998 et 2001). Pour peu qu’il se systématise, ce type d’étude devrait fournir des données fraîches sur les modalités de production et d’échange des récipients en terre cuite, dans la continuité des travaux de P.-L. van Berg notamment (van Berg, 1987, 1988 et 1996).

Quant à l’organisation spatiale de l’habitat, on peut citer, en plus de l’implantation “ en rue ” des maisons de Fond de Momalle, secteur III, la présence d’habitations isolées pour les villages de Remicourt-En Bia Flo II et de Fexhe-le-HautClocher-Podrî l’Cortri, qui suscitent diverses interprétations. Les particularités typo-chronologiques de la céramique associée permettent d’évoquer une phase pionnière dans le peuplement rubané de la région, avec une première vague “ d’éclaireurs ”, suivis par le “ gros de la troupe ”. Il est aussi envisageable que deux groupes rubanés se soient installés au même endroit, à des époques suffisamment distantes pour que les derniers arrivants ignorent la présence des premiers. Enfin, on peut également évoquer des demeures à “ vocation particulière ”, expression du statut social des occupants. Certains vestiges originaux liés aux activités artisanales requièrent également une attention particulière. A Podrî l’Cortri, l’identification de brai de bouleau sur quatre vases a suscité quelques réflexions sur différents aspects de la production et de l’utilisation de cet adhésif (Bosquet et al., 2001). Ainsi, l’utilisation exclusive du bouleau pourrait traduire un choix délibéré, peut être dicté par le grand nombre de produits dérivés que fournit cette essence, et dont les rebuts de fabrication ont pu servir à la fabrication du brai. De plus, la similarité des profils chromatographiques observés permet de penser que l’adhésif est issu d’un mode de production relativement standardisé, comme cela a déjà été montré sur d’autres sites néolithiques, celui de Giribaldi en particulier (Regert et al., 2000 : 593-594). De plus, la concentration, au sein d’une même fosse, des tessons portant des traces de brai pourrait signaler une spécialisation dans la fabrication et l’utilisation de l’adhésif. Enfin, les collages effectués sur des vases inutilisables même après réparation, montrent que les rubanés attribuaient une valeur autre que simplement utilitaire à leur récipients. Celle-ci devait, en partie au moins, être esthétique, compte tenu de la facture exceptionnelle des décors qui ornent les deux vases restaurés.

Les quelques aspects abordés illustrent le potentiel scientifique incontestable des données rassemblées depuis 1996, mais montrent également l’ampleur du travail à accomplir pour renouveler la vision du peuplement de nos régions au Néolithique ancien. L’étude pluridisciplinaire entamée devrait nous permettre d’avancer dans la réponse à une question essentielle : comment fonctionnait la société rubanée et, en particulier, quels en étaient les facettes et les moteurs identitaires, sociaux et économiques ? Bibliography BINDER, D. & COURTIN, J., 1986, Les styles céramiques du Néolithique ancien provençal. In Le Néolithique de la France. Hommage à Gérard Bailloud, édité par Demoule, J.-P. & Guilaine, J. Paris, Picard, p. 83-94. BOELICKE, U., VON BRANDT, D., LÜNING, J., STEHLI, P. & ZIMMERMANN, A., 1988, Der bandkeramische Siedlungplatz Langweiler 8. Rheinische Ausgrabungen 28, p. 395-427.

A Ans/Alleur, la découverte de trois foyers, grâce à des conditions de conservation exceptionnelles, constitue 92

D. Bosquet et al.: La néolithisation en Hesbaye (Belgique) : Apports des fouilles récentes (1996-1999)... BOSQUET, D., 1992, Les dispositifs d’entrée des enceintes du Rubané de Belgique : interprétations et reconstitutions en trois dimensions des entrées de Darion et de Waremme-Longchamps. Mémoire de licence de l’Université Libre de Bruxelles, inédit.

CONSTANTIN, C., DEMAREZ, L., FARRUGIA, J.-P., COUDART, A., DEMOULE, J.-P. & ILETT, M., 1979, Le Néolithique omalien et post-omalien en Hainaut belge. Bulletin de la Société Préhistorique Française 76, p. 196.

BOSQUET, D., 1993, Essai de reconstitution des dispositifs d’entrée de deux villages fortifiés du Rubané de Hesbaye. Notae Praehistoricae 12, p. 123-129.

CONSTANTIN, C., FARRUGIA, J.-P. & DEMAREZ, L., 1980, Aubechies : site de la Céramique Linéaire en Hainaut occidental. Bulletin de la Société Préhistorique Française 76, p. 367-382.

BOSQUET, D. & FOCK, H., 1996, Vestiges rubanés à WaremmeVinâve. Notae Praehistoricae 16, p. 151-154. BOSQUET, D., FOCK, H. & PREUD’HOMME, D., 1997a, Découverte d’un village rubané au Fond de Momalle (comm. de Remicourt) (TGV oriental). Notae Praehistoricae 17, p. 111-115.

ECKERT, J., IHMIG, M., KUPER, R., LÖRH, H. & LÜNING, J., 1972, Untersuchungen zur neolitischen Besiedlung der Aldenhovener Platte II. Bonner Jahrbücher des Rheinischen Landesmuseums in Bonn und des Vereins von Altertumsfreunden im Rheinlande 172, p. 344-394.

BOSQUET, D., PREUD’HOMME, D., FOCK, H. & GOFFIOUL, Cl., 1997b, Découverte d’un village rubané fossoyé à Remicourt au lieu-dit En Bia Flo. Notae Praehistoricae 17, p. 103-110.

FOCK, H., GOFFIOUL, Cl. & CORNÉLUSSE, F., 1998, Un habitat rubané à Remicourt, au lieu-dit Fond de Momalle, secteur III. Notae Praehistoricae 18, p. 123-129.

BOSQUET, D., PREUD’HOMME, D., 1998, Dernière campagne de fouille sur le village rubané de Remicourt au lieu-dit En Bia Flo II. Notae Praehistoricae 18, p. 119-122.

FRÉBUTTE, Chr., & MARCHAL, J.-Ph., 1998, Implantation du Rubané récent au lieu-dit “ Ferme de l’Abbaye ” à Donceel (Province de Liège). Notae Praehistoricae 18, p. 141-148.

BOSQUET, D., FOCK, H., GOFFIOUL, Cl. & PREUD’HOMME, D., 1998, Le site rubané de Fexhe-le-Haut-Clocher-Podrî l’Cortri : résultats des fouilles. Notae Praehistoricae 18, p. 131140.

GOFFIOUL, Cl., FOCK, H., CORNÉLUSSE, F., BOSQUET, D. & PREUD’HOMME, D., 1999, Découverte d’un village rubané fossoyé à Voroux-Goreux (Comm. de Fexhe-le-Haut-Clocher) sur le tracé oriental du TGV. Notae Praehistoricae 19, p. 101106.

BOSQUET, D., REGERT, M., DUBOIS, N. & JADIN, I., 2001, Identification de brai de bouleau sur quatre vases du site rubané de Fexhe-le-Haut-Clocher “ Podrî l’Cortri ”. Premiers résultats. Notae Praehistoricae 21, p. 119-127.

HEIM, J. & JADIN, I., 1998, Sur les traces de l’orge et du pavot : nouvelles données carpologiques sur l’agriculture danubienne de Hesbaye, dans un cadre européen. In Organisation néolithique de l’espace en Europe du nord-ouest (Actes du XXIIIème Colloque Interrégional sur le Néolithique, Bruxelles 24-25 octobre 1997), édité par van Berg, P.-L. & Cauwe, N. Bruxelles, Bulletin de la Société royale belge d’Anthropologie et de Préhistoire 111, p. 164-187.

BOSQUET, D., GOFFIOUL, Cl., FOCK, H. & PREUD’HOMME, D., à paraître (a), Les enceintes rubanées de Belgique : apport des fouilles récentes. In Camps, enceintes et structures d’habitat en France septentrionale, (Actes du XXIVème Colloque interrégional sur le Néolithique, Orléans, 20 et 21 novembre 1999), dirigé par Verjux, Chr. & Leroy, D. Orléans.

ILETT, M., PLATEAUX, M. & COUDART, A., 1986, Douze années de sauvetage dans la vallée de l’Aisne. Analyse spatiale des habitats du Rubané récent. Problèmes actuels. In Le néolithique de la France. Hommage à Gérard Bailloud, dirigé par Demoule, J.-P. & Guilaine, J. Paris, Picard, p. 131-140.

BOSQUET, D., FOCK, H. & LIVINGSTONE SMITH, A., à paraître (b), La chaîne opératoire de la céramique rubanée : essai de reconstitution sur le matériel de la fosse 10 du site de Remicourt En Bia Flo II. In Pottery Manufacturing Processes : Reconstruction and Interpretation (Actes du XIVème Congrès UISPP, Liège, septembre 2002), dirigé par Livingstone Smith, A., Bosquet, D. & Martineau, R. Oxford, BAR International Series.

JADIN, I., 1988, Économie de production dans le Rubané récent de Belgique. In Rubané et Cardial (Actes du Colloque de Liège, novembre 1988), édité par Cahen, D. & Otte, M. Liège, ERAUL 39, p. 147-153.

BURNEZ-LANOTTE, L. & ALLARD, P., 1997, Le Site rubané du “ Petit-Paradis ” à Harduémont (Verlaine, Lg) : résultats de la campagne 1997. Notae Praehistoricae 17, p. 117-121.

JADIN, I., 2003, Trois petits tours et puis s’en vont. La fin de la présence rubanée en Moyenne Belgique. Thèse de Doctorat, Université de Liège. Liège, ERAUL 109.

CAHEN, D., 1985, Interprétations nouvelles du site de Darion. Bulletin de la Société royale Belge d’Anthropologie et de Préhistoire 96, p. 75-86.

JEUNESSE, Chr. & SAINTY, J., 1991, Bischoffsheim “ Le Village ” (Bas-Rhin). Un habitat rubané avec Céramique de La Hoguette. Cahier de l’Association pour la Promotion de la Recherche Archéologique en Alsace 7, p. 15-58.

CAHEN, D., OTTE, M. & CASPAR, J.-P., 1986, Industries lithiques danubiennes de Belgique. Liège, ERAUL 21, 88 p.

JEUNESSE, Chr. & SAINTY, J., 1992, L’habitat néolithique ancien du site d’Hensisheim “ Rattfeld ” (Haut Rhin). Cahiers de l’Association pour la Promotion de la Recherche Archéologique en Alsace 8 (1ère partie), p. 95-146.

CAHEN, D., KEELEY, L. H., JADIN, I. & VAN BERG, P.-L., 1990, Trois villages fortifiés du Rubané récent en Hesbaye liégeoise. In Rubané et Cardial (Actes du Colloque de Liège, novembre 1988), édité par Cahen, D. & Otte, M. Liège, ERAUL 39, p. 125-146.

JEUNESSE, Chr., 1994, Le Néolithique de la plaine du Rhin supérieur. Prähistorische Zeitschrift 9/1, p. 1-31.

CASPAR, J.-P., HAUZEUR, A., DOCQUIER, J., BIT, R., VAN ASSCHE, M. & TROMME, F., 1992, Le fossé rubané de Vauxet-Borset “ Gibours ”. Notae Praehistoricae 11, p. 77-83.

LANGOHR, R. & SANDERS, J., 1985, Etude pédologique du site de Darion : données préliminaires. Bulletin de la Société royale Belge d’Anthropologie et de Préhistoire 96, p. 17-30.

COLLIN, F., BILQUIN, L. & DUPAGNE, J., 1998, Construction d’une maison rubanée au Préhistosite de Ramioul. Expérimentations pour la mise en évidence d’une chaîne opératoire. Notae Praehistoricae 18, p. 149-156.

LIVINGSTONE SMITH, A., 1998, Poteries “ ethnographiques ” et archéologiques. Analyse de la chaîne opératoire. In Méthodes d’analyse de la terre cuite (Actes de la Journée d’Archéologie en province de Liège, Ocquier 28.11.1998), dirigé par Leotard, J.M. Jambes, p. 59-68.

COLLIN, F. & MARCHAL, J.-Ph., 2000, Approche expérimentale du Four 2.40 du site rubané d’Alleur. Notae Praehistoricae 20, P. 153-158.

LIVINGSTONE SMITH, A., 2001, Pottery Manufacturing Processes : Reconstruction and Interpretation. In Uan Tabu in 93

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe Settlement History of the Lybian Sahara, édité par Garcea, H. Rome, Arid Zone Archaeology (Monograph 2), p. 113-251.

TAPRET, A. & VILLES, A., 1996, Contribution de la Champagne à l’étude du Néolithique ancien. In La Bourgogne entre les bassins rhénan, rhodanien et parisien. Carrefour ou frontière ? (Actes du XVIIIème Colloque Interrégional sur le Néolithique, Dijon 25-27 octobre 1991), édité par Duhamel, P. Dijon, Revue Archéologique de l’Est (14ème supplément), p. 175-256.

LODEWIJCKX, M., 1984, Les deux sites rubanés de Landen-Wange et Linter-Overhespen après la campagne de fouilles. Notae Praehistoricae 4, p. 97-108. LODEWIJCKX, M., 1990, Les deux sites rubanés de Wange et d’Overhespen (Belgique, prov. Brabant). In Rubané et Cardial (Actes du Colloque de Liège, novembre 1988), édité par Cahen, D. & Otte, M. Liège, ERAUL 39, p. 105-116.

TOUSSAINT, M. & TOUSSAINT, G., 1982, Pétrographie et paléogéographie des herminettes omaliennes de Hesbaye. Bulletin de la Société Royale d’Études Géologiques et Archéologiques “ Les Chercheurs de la Wallonie ” XXV, p. 503-570.

MANEN, C., 1997, L’axe rhodano-jurassien dans le problème des relations sud-nord au Néolithique Ancien. Oxford, BAR International Series 665.

VAN BERG, P.-L., 1987, Rubané récent de Hesbaye : signatures récurrentes de maîtres potiers. Bulletin de la Société royale Belge d’Anthropologie et de Préhistoire 98, p. 197-222.

MARCHAL, J.- Ph., 1998, Sauvetage sur un site rubané à Alleur Domaine Militaire. Notae Praehistoricae 18, p. 107-117.

VAN BERG, P.-L., 1988, Le poinçon, le peigne et le code. Essai sur la structure du décor céramique dans le Rubané récent du nord-ouest. Thèse de Doctorat, Université de Liège, inédit.

MARCHAL, J.-Ph., 1999, Seconde campagne de fouilles sur le site rubané d’Alleur - Domaine Militaire. Note préliminaire. Notae Praehistoricae 19, p. 107-109.

VAN BERG, P.-L., 1989, Architecture et géométrie de quelques villages Rubané récent du Nord-Ouest. Helinium XXIX, p. 1341.

MODDERMAN, P. J. R., 1970, Linearbandkeramik aus Elsloo und Stein. Analecta Praehistorica Leidensia III (2 vol.), 218 p., 18 fig., 232 pl.

VAN BERG, P.-L., 1990, Céramique du Limbourg et néolithisation en Europe du Nord-Ouest. In Rubané et Cardial (Actes du Colloque de Liège, novembre 1988), édité par Cahen, D. & Otte, M. Liège, ERAUL 39, p. 161-208.

PREUD’HOMME, D., BOSQUET, D. & FOCK, H., 1997, Vestiges rubanés à Remicourt, au lieu-dit Tombe de Hodeige (TGV oriental). Notae Praehistoricae 17, p. 97-101.

VAN BERG, P.-L., 1996, Gauches, joueurs et apprentis : production des marges dans la céramique rubanée occidentale. In La Bourgogne entre les bassins rhénan, rhodanien et parisien. Carrefour ou frontière ? (Actes du XVIIIème Colloque Interrégional sur le Néolithique, Dijon 25-27 octobre 1991), édité par Duhamel, P. Dijon, Revue Archéologique de l’Est (14ème supplément), p. 29-53.

REGERT, M., GARNIER, N., BINDER, D. & PÉTREQUIN, P., 2000, Les adhésifs néolithiques : quels matériaux utilisés, quelles techniques de production dans quel contexte social ? L’exemple des adhésifs des sites de Giribaldi et de Chalain. In Arts du feu et productions artisanales (Actes des XXèmes Rencontres Internationales d’Archéologie et d’Histoire d’Antibes), dirigé par Pétrequin, P., Fluzin, P., Thiriot, J. & Benoit, P. Antibes, APCDA, p. 585-604.

VAN DE VELDE, P., 1973, Rituals, skins and Homer : the danubian “ tan-pits ”. Analecta Praehistorica Leidensia VI, p. 50-35.

ROOSENS, H., 1961, Rosmeer : Bandkeramische nederzetting. Archéologie 1, p. 519-520.

VILLES, A., 1981, Les silos de l’habitat protohistorique en Champagne crayeuse. In Les techniques de conservation des grains à long terme, édité par Sigaut, P. Paris, p. 194-225.

94

C. Frébutte et al.: 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique)

1995-2001 : SEPT ANNÉES DE RECHERCHES MÉGALITHIQUES EN CALESTIENNE (BELGIQUE) Christian FRÉBUTTE, Michel TOUSSAINT, François HUBERT, Stéphane PIRSON et Philippe MASY

1. INTRODUCTION

2. L’ALLÉE COUVERTE DE LAMSOUL

La pauvreté du patrimoine mégalithique belge est évidente. Elle résulte de la position du pays en marge des grandes aires touchées par ce phénomène néolithique et de nombreuses destructions effectuées au cours des derniers siècles. Ainsi, de la petite dizaine de sépultures mégalithiques recensées dans la littérature, quatre sont conservées de nos jours ; il s’agit des chambres funéraires de Lamsoul près de Rochefort, de “ Wéris I ” et de “ Wéris II ” près de Durbuy et de Laviô près de Bouillon.

Le monument mégalithique de Lamsoul a été édifié au pied du flanc méridional de la colline de “ Rubiémont ”, près d’un petit col qui domine la rive gauche de la Lomme. sur le territoire de l’ancienne commune de Jemelle, dans l’entité de Rochefort (province de Namur). Il a été repéré en 1971 par W. Lassance, puis exploré une première fois en 1976 et 1977 par un cercle d’amateurs local. Ces travaux ne répondant pas à toutes les questions que suscite une telle structure (Huysecom, 1983), la Direction de l’Archéologie du Ministère de la Région wallonne a procédé à une nouvelle fouille, au cours de l’hiver 1995-1996 (Jadin et al., 1998 ; Toussaint & Jadin, 1996 ; Toussaint et al., 1997b).

La situation des menhirs est également navrante. Quelques uns ont été débités en pavés, comme à Bray ou à Gozée. Les blocs subsistants sont dispersés en Wallonie, dans les provinces de Hainaut (à Hollain, Gozée, Baileux), de Namur (Velaine-sur-Sambre, Doyon) et de Luxembourg (à Wéris, Heyd et Izier).

Le monument est implanté sous le niveau du sol actuel, dans une large tranchée de fondation creusée suivant la ligne de déclivité du terrain environnant, dans le sol meuble puis, par endroits, dans le substrat schisteux. La partie arrière de l’édifice est recouverte par un tumulus.

Les connaissances à propos de ces vestiges ont longtemps reposé sur des fouilles peu rigoureuses, exécutées sur quelques monuments à la fin du XIXe siècle et au début du XXe siècle. Entre 1979 et 1986, de véritables campagnes archéologiques modernes sont consacrées à des éléments du “ champ mégalithique de Wéris ”. Ces interventions ont été conduites sous la direction de F. Hubert du Service national des Fouilles.

Le mégalithe est en partie détruit, surtout sa zone d’entrée. Son noyau central se compose de neuf grosses dalles de calcaire qui forment une chambre rectangulaire orientée approximativement du nord-ouest au sud-est. Seules cinq d’entre elles sont encore en place, les quatre autres ayant été basculées. Dans son état de conservation moderne, l’édifice a environ 5,10 m de longueur extérieure, près de 2,50 m de largeur externe et de 1,20 m à environ 1,60 m de largeur interne. Deux piliers sont encore dressés sur le côté gauche de la chambre. Les deux orthostates droits correspondants ont été couchés au cours des Temps modernes comme en atteste la datation 14C d’un petit foyer disposé juste sous l’un d’eux. Le chevet se compose de deux piliers convergents formant un “ V ” bien ouvert et de la seule dalle de couverture encore in situ. La disposition des dalles montre que l’ensemble du chevet a sans doute légèrement glissé vers l’arrière.

Afin de prolonger et de compléter ces investigations, la Direction de l’Archéologie du Ministère de la Région wallonne, à laquelle s’associa ultérieurement l’Association wallonne d’Etudes mégalithiques (AWEM), a entrepris, de 1995 à 2001, le premier volet d’un nouveau programme de recherches pluridisciplinaires. Durant cette phase, des fouilles ont été entreprises aux allées couvertes de Lamsoul, de “ Wéris I ” et de “ Wéris II ” et sur des sites du “ champ mégalithique de Wéris ” où gisaient des menhirs potentiels ; ces monuments se trouvent en Calestienne, bande de terrain calcaire du sud de la Famenne. La seconde phase du projet, qui se clôturera en 2003, consiste à préparer une monographie intégrant un bilan des connaissances relatives au champ mégalithique avant 1979, un rapport détaillé des fouilles exécutées entre 1979 et 2001, et des synthèses archéologiques et paléoenvironnementales élargies à d’autres problématiques. La troisième phase qui est prévue à partir de 2004 englobera des prospections systématiques de l’ensemble du champ mégalithique et de ses abords par diverses techniques géophysiques et mécaniques, ceci afin de détecter des monuments encore enfouis qui pourraient déboucher sur un nouveau programme archéologique.

La tranchée de fondation a un fond plus ou moins plat et des bords obliques pouvant atteindre 70°. Deux types de structures liées à l’implantation des orthostates y ont été aménagées. Il s’agit d’étroites rainures longitudinales creusées dans le substrat schisteux sous la base de certains orthostates et d’un dispositif particulier observé sous un des blocs effondrés dans la chambre, près du chevet. La base de ce bloc y est simplement adossée, côté extérieur, à la paroi latérale, très abrupte, d’un petit surcreusement de la tranchée de fondation. 95

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. Plan d’ensemble du “ champ mégalithique de Wéris ” et de ses alignements supposés. Les cercles blancs correspondent aux mégalithes douteux ou non encore étudiés, les cercles noirs aux monuments confirmés par les fouilles modernes : I. Allée couverte de “ Wéris I ” ; II. Allée couverte de “ Wéris II ” ; M1-3. Groupe des trois menhirs d’Oppagne ; P4-7, M8. Pierres dressées du vestibule et menhir indicateur de l’allée couverte de “ Wéris II ” ; M9-13. Groupe des cinq menhirs d’Oppagne, voisins de “ Wéris II ” ; M14. “ Menhir Danthine ” ; M15-16. Menhirs fouillés en 1985 au “ Champ de la Longue Pierre ” ; P17-18, M19-24. Pierres dressées du vestibule et menhirs proche de l’allée couverte de “ Wéris I ” ; M25. Menhir de Morville ; M26. Menhir du lieu-dit “ å Djèyî ” à Heyd ; M27. Menhir d’Ozo. 96

C. Frébutte et al.: 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique)

Une fois redressés, les orthostates ont été maintenus par des pierres de calage et surtout par de larges amas de blocs – de véritables contreforts ou murets de soutien - déversés dans la tranchée d’implantation, sur les deux flancs extérieurs et à l’arrière du monument. Les interstices entre les piliers ont été comblés par de la blocaille formant des petits murets de pierres sèches. Parfois, ces murets sont renforcés par une petite dalle de calcaire.

sera poursuivie de manière continue par les éventuelles découvertes à venir. L’ensemble mégalithique se compose d’une série d’axes parallèles, déterminés à partir de 2 à 3 sites chacun (Hubert, 2000) ; la réalité de certains axes, qui intègrent des blocs erratiques repérés au XIXe siècle par le géologue X. Stainier (Stainier, 1902), reste cependant encore hypothétique. Quoi qu’il en soit, l’axe de base s’étend de l’allée couverte nord, ou “ Wéris I ”, à l’allée sud ou “ Wéris II ”. Les trois menhirs d’Oppagne (M1 à M3) déterminent une seconde ligne avec un bloc erratique, à 50 m à l’ouest de l’axe matérialisé par les allées couvertes. A 25 m à l’est de cette ligne se trouve l’alignement le plus riche en menhirs avec, du sud au nord, les cinq menhirs annexes de l’allée couverte sud (M9 à M13), les trois menhirs du “ Champ de la Longue Pierre ” (M14 à M16) et le menhir “ å Djèyî ”, à Heyd (M26) ; le menhir de Morville (M25) est légèrement décalé par rapport à ce dernier axe. Plus à l’est, des erratiques de Stainier commanderaient encore deux autres lignes parallèles. Enfin, le menhir d’Ozo (M27) se trouve à quelque 300 m à l’est de la ligne définie par les allées couvertes et pourrait déterminer un dernier axe - pas strictement parallèle aux précédents - avec le “ Thierde-Sel ” situé à l’extrémité méridionale du champ mégalithique.

Les quelques plaquettes disposées horizontalement sur le sol du chevet signifieraient qu’un dallage recouvrait, au moins en partie, ce secteur. Devant le mégalithe s’étale un empierrement grossier, allongé perpendiculairement à l’axe de la chambre et constitué de blocs et de petites dalles irrégulières, essentiellement en grès, schiste et calcaire, de petites dimensions. Quelques indices liés au calibre et à la quantité des pierres pourraient permettre d’y voir une terrasse aménagée aux abords de l’entrée plutôt que le produit de la vidange de la chambre ou des contreforts latéraux. 3. LE “ CHAMP MÉGALITHIQUE DE WÉRIS ” 3.1. Généralités et contexte géologique

3.2. Les allées couvertes de “ Wéris I ” et “ Wéris II ”

La région de Wéris est située sur le flanc oriental du Synclinorium de Dinant, entité tectonique héritée de l’orogenèse varisque se rattachant à la Zone rhénohercynienne (Franke, 1989). Tous les mégalithes connus à ce jour dans les environs de Wéris se situent à la limite orientale du Synclinorium, dans la Calestienne, bande de terrains d’âge dévonien moyen dominée par les calcaires et séparant les terrains aréno-pélitiques des hauteurs de l’Ardenne (d’âge dévonien inférieur) des «schistes» de la dépression de la Famenne (d’âge dévonien supérieur). Ils sont en poudingue de la Formation de Hampteau, probablement d’âge emsien supérieur (Stainier, 1994). Cette roche affleure sur la crête située à quelques kilomètres à l’est des alignements (Toussaint et al., 1999 : 145).

3.2.1. “ Wéris I ” Signalée dès 1879 dans un texte d’A. Daufresne de la Chevalerie, l’allée couverte “ Wéris I ” ou dolmen nord est acquise par l’Etat belge en 1882 avant de faire l’objet d’une première restauration et d’être entourée d’une grille, en 18851886. Diverses fouilles sont alors menées à l’intérieur et à l’extérieur du monument jusqu’en 1906, notamment par A. de Loë du Service des Fouilles de l’Etat. Enfin, deux programmes de fouilles modernes y sont réalisés de 1979 à 1984 par l’ex-Service national des Fouilles (Hubert, 1980, 1982, 1985 ; Hubert & Huysecom, 1983, 1984), puis de 1999 à 2001 par la Direction de l’Archéologie du Ministère de la Région wallonne (Frébutte et al., 2000b, 2001a, 2001b ; Toussaint et al., 2000c).

Les éléments du champ mégalithique sont dispersés sur deux plateaux, bordés à l’ouest par la vallée de l’Ourthe et séparés par la vallée de l’Aisne, affluent de celle-ci. La plupart des menhirs attestés et deux sépultures collectives du type “ allée couverte ” sont implantés dans le plateau méridional, sur un substrat calcaire du Givétien. Un seul menhir (M27) a été découvert sur le plateau septentrional, à Ozo près d’Izier, sur un substrat de siltite eifélienne.

Dans l’état actuel de conservation, le sol environnant le monument est au même niveau que celui de la chambre funéraire. Il n’est pas possible d’affirmer si cette disposition est néolithique ou si elle résulte des nombreux remaniements modernes des abords ; ces derniers ont pu en effet effacer les indices d’éventuels aménagements tels qu’un tumulus et/ou une tranchée de fondation pareille à celle de “ Wéris II ”.

Dans l’état actuel des connaissances, le “ champ mégalithique de Wéris ” s’étend sur près de 8000 m de longueur, selon un azimut orienté grossièrement du sud-sud-ouest au nord-nordest, à environ 26° 50'. Depuis 2000 (Frébutte et al., 2000c), les pierres dressées des vestibules des deux allées couvertes de Wéris et les menhirs attestés sont désignés par un chiffre – précédé de la lettre M pour les menhirs et de la lettre P pour les piliers des vestibules - dont la numérotation suit un cheminement du sud-sud-ouest vers le nord-nord-est. Elle

Ce mégalithe se compose d’un vestibule, d’une chambre rectangulaire couverte et d’une dalle posée à plat derrière le chevet. Il mesure 10,80 m de longueur pour 4,60 m de largeur. Son grand axe est orienté du nord-nord-est au sud-sud-ouest. Le vestibule est délimité par deux piliers (P17-P18) dressés de part et d’autre de l’entrée ; son sol a livré un radier de blocs calcaires disposés soit à plat, soit de chant et associés à 97

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 2. Plan des fouilles de 1995-1996 à l’allée couverte de Lamsoul (dessin S. Lambermont).

quelques fragments de poudingue. Cet aménagement aurait servi d’assise à deux grands blocs de poudingue interprétés comme les portes du sépulcre et qui sont actuellement couchés dans ce secteur. Le vestibule est séparé de la chambre par la juxtaposition de deux dalles posées sur chant. Ces dernières ont été échancrées par bouchardage à leur point de contact pour composer un demi-hublot évasé vers le haut.

La chambre a environ 6 m de longueur, 1,70 m de largeur moyenne et environ 1,50 m de hauteur. Deux paires d’orthostates alignés forment les côtés longitudinaux. Les orthostates qui délimitent l’avant de la construction sont chacun quasiment aussi larges que la chambre elle-même ; ils supportent une énorme dalle de couverture de 5 m de long sur 3,60 m de largeur maximale. Les fouilles récentes ont 98

C. Frébutte et al.: 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique)

montré que les piliers arrière reposaient sur un lit de blocs de calcaires disposés de chant ou à plat et répartis en deux niveaux. Ces supports supportent la deuxième dalle de couverture du monument.

Un chemin fait de plaquettes de grès et de calcaire court tout au long du côté extérieur gauche de l’allée couverte, juste au-dessus des quelques blocs de pierre qui remplissaient le fond de la partie latérale de la tranchée d’implantation. La présence d’un tel aménagement, quasi à la base des piliers du vestibule et de la chambre, à peine un peu plus haut que le dallage interne, semble indiquer que le monument était destiné à être vu, au moins au début de son utilisation.

Le chevet ne présente plus de dalle de fermeture dressée. Un radier de blocs, qui joint les fondations repérées sous les piliers gauche et droit, montre cependant que la chambre était bien fermée par une dalle perpendiculaire à son axe. Peutêtre la petite dalle actuellement couchée à l’intérieur de la partie arrière de la chambre en est-elle un reliquat. La grande dalle, posée à plat derrière le monument, reste énigmatique.

Comme à “ Wéris I ”, la fonction de la grosse dalle de poudingue, qui repose derrière le monument, demeure inconnue car elle ne semble pas avoir contrebuter la dalle du chevet.

3.2.2. “ Wéris II ” 3.3. Les menhirs du “ champ mégalithique de Wéris ”

L’allée couverte sud ou “ Wéris II ” ou encore “ dolmen d’Oppagne ” a été découverte fortuitement au printemps 1888 par les fils de l’exploitant agricole, puis dégagée par A. Charneux. Vers 1889 ou 1890, la chambre sépulcrale est vidée par A. Orban tandis qu’en 1906, des fouilles sont réalisées par A. de Loë et E. Rahir. Enfin, en 1996 et 1997, la Direction de l’Archéologie (à l’époque Direction des Fouilles) du Ministère de la Région wallonne lui consacre une campagne de travaux dont les multiples résultats aboutissent à une restauration du monument (Toussaint et al., 1996, 1997a).

Les fouilles entreprises sur les menhirs cités ci-dessous ont été motivées par le caractère exogène de ces blocs de poudingue par rapport au substrat local (calcaire ou siltite) et par leur disposition au sein de certains alignements. 3.3.1. Les trois menhirs d’Oppagne (M1 à M3) Trois blocs de poudingue couchés côte à côte sont découverts en 1888, à Oppagne. Ils suscitent une fouille en 1906 par A. de Loë qui réalise des tranchées peu méthodiques. En 1933, le propriétaire du site, l’Institut archéologique du Luxembourg, décide de lever et d’aligner les trois pierres, interprétées comme des menhirs, sur des fondations maçonnées. En 1961, le Service technique de la province de Luxembourg relève un des poudingues qui s’est effondré.

L’allée couverte a été édifiée dans une tranchée creusée transversalement par rapport à la pente du terrain. L’axe principal du monument est orienté à peu près du nord-est au sud-ouest, avec l’ouverture au nord-est. Sa longueur maximale après restauration est de 9,90 m et sa largeur maximale de 4,55 m, sans le menhir-indicateur (P8) redressé à sa droite. Le vestibule, les fouilles récentes l’ont montré, était plus long qu’à “ Wéris I ” ; il était bordé par deux paires de piliers (P4 à P7) répartis de chaque côté de l’entrée de la chambre. Cet espace a livré deux dalles couchées interprétées comme des portes monumentales, comparables à celles de “ Wéris I ”.

La fouille organisée en 2001 (Frébutte et al., 2001a, 2002) a souligné deux faits : d’une part l’importance des modifications et des remaniements modernes qui ont clairement porté préjudice au contexte stratigraphique, d’autre part l’absence de structures néolithiques qui pourrait résulter de l’observation précédente. Le caractère mégalithique des blocs n’est toutefois pas remis en cause grâce aux arguments apportés par l’étude géologique. En outre, il semblerait que les pierres étaient primitivement alignées entre 4,50 et 5,50 m plus au sud ou au sud-ouest de leur implantation actuelle. Cette hypothèse résulte de certaines particularités des paliers d’interface entre les dépôts naturels et les dépôts anthropiques de cette zone, et de l’analyse d’une photographie de 1906 représentant les poudingues couchés.

Une dalle d’entrée posée de chant, avec ouverture aménagée par bouchardage en “ U ” ouvert vers le bas, sépare le vestibule de la chambre proprement dite. Le piédroit oriental de cette dalle d’entrée était posé sur une dalle de poudingue disposée horizontalement tandis que le piédroit occidental s’appuyait sur un radier de pierrailles. Devant l’entrée a été dégagée une petite fosse qui contenait quelques esquilles osseuses. De plan rectangulaire, la chambre mesure 5 m de longueur, 1 m de largeur près du chevet et 1,80 m de largeur maximale près de l’entrée. Ses flancs comportent chacun deux orthostates qui soutiennent trois dalles de couverture. Les interstices entre les piliers étaient bouchés par des murets de pierres sèches dont les niveaux inférieurs subsistaient à certains endroits. Le chevet est formé par une dalle placée de chant perpendiculairement aux deux piliers arrière. Le sol du sépulcre a été aménagé avec soin à l’aide d’un dallage de plaquettes de grès et de psammite. La hauteur intérieur entre ce dallage et les dalles de couverture s’élève entre 70 et 90 cm.

Conformément à la volonté du propriétaire, la présentation du site avec les trois menhirs dressés a été maintenue. Des panneaux didactiques expliqueront le caractère inexact de cette disposition, résultat de la reconstitution de 1933. 3.3.2. Les menhirs voisins de l’allée couverte de “ Wéris II ” (M9 à M13) A la fin du XIXe siècle, trois dalles de poudingue, considérés alors comme les éléments d’une chambre funéraire non achevée ou détruite, sont localisées à 25 m à l’est de “ Wéris II ”. Un siècle plus tard, en 1986, le Service national des 99

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 3. Plan des allées couvertes de “ Wéris I ” (A) et de “ Wéris II ” (B) avant leur restauration contemporaine (dessin S. Lambermont).

Fouilles (Hubert, 1987) leur consacre une campagne archéologique. Deux blocs supplémentaires sont repérés dans des fosses qui contenaient des tessons de La Tène finale et du XVIe-XVIIe siècle, des fragments de bas fourneau, des scories et des galets ayant servi de percuteurs.

poudingue qui gênait leurs activités. En 1947, Hélène Danthine, professeur de Préhistoire à l’Université de Liège, redégage la pierre auprès de laquelle est découverte une petite dalle annexe de poudingue (Danthine, 1961). La faible emprise de ces investigations explique peut-être qu’aucune structure néolithique n’ait été décelée. Le caractère exogène du poudingue par rapport au substrat calcaire tend toutefois à conférer au bloc une nature de menhir. La dalle annexe correspondrait à un élément de renfort posé sur le sol lors de l’érection et réemployé ensuite comme calage du monument dressé.

L’ensemble des données recueillies par la fouille permet de considérer les cinq dalles comme des menhirs abattus. Divers indices dont des pierres calcaires de calage encore en place ont autorisé, en avril 1997, le redressement de quatre de ces blocs sur un alignement sud/nord (Frébutte et al., 1998). 3.3.3. Le “ menhir Danthine ” (M14)

En 1948, suite à la demande de l’exploitant agricole, les deux pierres sont acheminées le long de la route Barvaux-Pont d’Erezée, à environ 130 m au sud-ouest du lieu de leur découverte. C’est à cet endroit qu’est élevé depuis lors le “ menhir Danthine ”.

Vers le milieu du XXe siècle à Wéris, au lieu-dit “ al Longue Pîre ” ou “ Champ de la Longue Pierre ”, des agriculteurs creusent une fosse pour enfouir plus profondément un gros 100

C. Frébutte et al.: 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique)

Figure 4. Pointes de flèche en silex recueillies sur les sites des allées couvertes de Lamsoul (1), de “ Wéris I ” (2-8) et de “ Wéris II ” (9), (dessin S. Lambermont). 101

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

La connaissance de ce mégalithe réclame clairement une nouvelle campagne de fouille dont les sondages seront plus étendus et plus systématiques. Les résultats de l’opération, projetée dans les années à venir, autoriseront éventuellement une mise en valeur plus adaptée.

Deux autres menhirs M20 et M21 se trouvaient respectivement à 21 m et à 28,50 m de l’entrée de l’allée couverte (Hubert, 1980). Ils constituent un axe nord-nordest. Le menhir M20 avait été enterré à proximité de sa fosse d’érection ; il a été redressé en 1981. La base du menhir M21 était encore fixée dans la structure de fondation néolithique ; sur ce morceau ont pu être recollés des fragments de la partie supérieure qui avaient été arrachés et jetés dans une excavation située à 1 m plus au nord (Hubert & Huysecom, 1983).

3.3.4. Les deux menhirs exhumés en 1985 au “ Champ de la Longue Pierre ” (M15 et M16) Deux menhirs supplémentaires ont été exhumés en 1985, à 50 m au nord-est de l’emplacement où reposait le menhir M14. Le Service national des Fouilles, responsable de l’intervention, a relevé sur une superficie d’à peine 24 m2 les structures liées à l’érection et à la condamnation de ces monuments (Hubert, 1985).

Les tranchées et les sondages complémentaires exécutés devant “ Wéris I ” entre 1999 et 2001 n’ont pas livré de vestiges supplémentaires. C’est donc avec prudence que trois grands poudingues, encore couchés dans ce secteur, pourraient être considérés comme des menhirs abattus. Ces pierres portent les numéros M22, M23 et M24.

Le menhir M15 a été fracturé en deux lors de sa destruction : sa partie supérieure a été rejetée dans une excavation tandis que la partie inférieure est demeurée en place, dressée dans la fosse d’érection primitive avec quelques éléments de calage en poudingue et en calcaire. Des tessons situent la destruction vers la fin du XVIe siècle ou durant le XVIIe siècle.

3.3.6. Le menhir de Morville (M25) Ce bloc de poudingue a été découvert dans la campagne de Morville ; il était dressé sur chant et émergeait à peine de la surface du sol. La fouille de 1995 (Toussaint & Hubert, 1998b) a montré que la pierre était installée dans une fosse d’érection et que sa base était maintenue par six petits poudingues. Du côte sud-ouest, les limites de la fosse paraissaient s’étendre jusqu’à un empierrement disposé à la même hauteur que les éléments de calage et à une quarantaine de centimètres du bloc. De forme subovale d’environ 1,80 m sur 0,75 m, cet empierrement se composait essentiellement de petits poudingues mais aussi de quartzites, de calcaires et de grès. Au terme de l’intervention et sur contrainte de l’exploitant, le menhir a été déplacé et redressé à la limite de la parcelle et d’un chemin, à 4 m vers le sud-sud-ouest de son implantation néolithique.

Avant d’être basculé dans une autre structure de condamnation, le menhir M16 était érigé à une distance d’1 m par rapport au premier bloc ; il formait avec celui-ci un axe sud/nord. Les vestiges de son dispositif de fondation correspondent à plusieurs lits de petites pierres qui bordaient trois côtés d’une fosse. Les dimensions intérieures de cette dernière s’élèvent à 1,10 m sur 60 cm. Dans la même structure de condamnation a été dégagée une dalle de poudingue triangulaire dont l’utilisation serait similaire à celle associée au menhir M14. Les conditions émises par l’exploitant agricole n’ont pas permis une restauration du site ; par conséquent, les poudingues ont dû être “ stockés ” au fond de la parcelle de “ Wéris I ”.

3.3.7. Le menhir “ å Djèyî ” (M26) En 1947, Hélène Danthine révélait la présence d’un poudingue couché à Heyd. La fouille qui s’est déroulée en 1998 a mis en évidence les traces d’une fosse d’érection et d’une fosse de condamnation du bloc (Frébutte et al., 1999 ; Toussaint et al., 1998a).

3.3.5. Les menhirs voisins de l’allée couverte de “ Wéris I ” (M19 à M24) Un dessin de 1886 représente une série de blocs de poudingue couchés sur le talus d’un chemin qui longeait alors l’allée couverte de “ Wéris I ”. A la reprise des fouilles sur le site en 1979, F. Hubert dénombra 28 poudingues, la plupart de taille moyenne, qui étaient éparpillés sur une longueur de 15 m depuis le bord de la rue des Dolmens jusqu’à 21 m de l’allée couverte. Ces blocs provenaient-ils des champs avoisinants ou correspondaient-ils à des menhirs ou des fragments de menhirs détruits par l’aménagement du vieux chemin dit “ des Romains ” ? La seconde hypothèse a été corroborée en 1983 par la découverte dans le talus de cette voie d’une fosse dont les dimensions et la forme étaient proches de celles de la base d’un long poudingue qui reposait à proximité (Hubert & Huysecom, 1984). L’excavation a été considérée comme la structure de fondation néolithique dans laquelle le bloc, qui est le menhir M19, se dressait à l’origine et dans laquelle il a été repositionné lors des fouilles de 1983. La fosse contenait une petite dalle également en poudingue apparentée à celles exhumées avec les menhirs M14 et M16.

Seule la moitié orientale de la fosse d’érection était préservée ; l’autre portion a été détruite par la sape précédant l’abattage du menhir. Aménagée en grande partie dans la roche-mère, la structure néolithique contenait quelques pierres de calage en calcaire et en poudingue associées à un sédiment limoneux compacté. Un petit fragment de clavicule d’un jeune enfant a été trouvée au sein de ce remplissage. A une époque non déterminée, le mégalithe a été renversé dans une vaste excavation creusée dans le substrat calcaire. Les vestiges rencontrés ont autorisé le redressement de la pierre qui s’est fait en grande partie “ à la préhistorique ”, dans un but expérimental. Durant la fouille, le bloc avait été couché sur des rondins de bois placés transversalement sur deux rails formés par deux troncs d’arbre. Ce dispositif a permis à sept personnes munies de cordes en chanvre de l’acheminer vers sa fosse primitive. Après avoir glissé dans 102

C. Frébutte et al.: 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique)

cette structure, le menhir a été redressé à la verticale en utilisant le bras de levier d’une chèvre constituée de deux troncs de conifères. A ce stade, la force humaine a dû être complétée par une traction mécanique.

“ Wéris I ” et “ Wéris II ” sont entièrement recouverts par de grandes dalles, deux à “ Wéris I ” et trois à “ Wéris II ”. Lamsoul n’a par contre qu’une seule dalle de couverture, qui ne couvre que le chevet. Faut-il attribuer cette insuffisance de dalles de couverture à l’exploitation du mégalithe pour récupérer des blocs de calcaire à des fins domestiques, par exemple au Moyen Age ? S’agit-il d’un essai de destruction du monument dont la non conservation de la zone d’entrée serait un autre témoignage ? Les Néolithiques ont-ils plutôt choisi de ne pas disposer de dalles de couverture en pierre sur l’essentiel de la chambre, utilisant à la place une structure légère, par exemple en bois ?

3.3.8. Le menhir d’Ozo (M27) En 1999, la Direction de l’Archéologie a étudié un gros bloc de poudingue qui gisait à la surface d’un champ entre les villages d’Izier et d’Ozo (Frébutte et al., 2000a ; Toussaint et al., 1999). Une fosse ovale de 2,50 m sur 80 cm, creusée dans le substrat rocheux, a été mise au jour sous l’extrémité nord du monolithe. D’une profondeur maximale d’une trentaine de centimètres, elle présentait un fond plat recouvert d’un limon très induré. Divers indices ont permis d’identifier cette structure à la fosse d’érection du poudingue, de considérer l’extrémité nord de ce dernier comme sa base et de suggérer que l’abattage du bloc s’était opéré du nord vers le sud. La vaste excavation dans laquelle reposait le poudingue date de cette condamnation ; elle ne contenait pas d’artefacts. Vers le milieu du XXe siècle, un exploitant agricole a tenté de détruire la pierre. Après avoir creusé une tranchée de détourage, il l’a partiellement dynamitée et a provoqué l’arrachement de quelques fragments qui ont ensuite été jetés dans la tranchée.

Des empierrements ont été observés à “ Wéris II ” et Lamsoul, sous la forme d’un chemin dallé qui longe le côté oriental de “ Wéris II ” et d’une sorte de terrasse précédant Lamsoul. 4.2. Les menhirs du “ champ mégalithique de Wéris ” 4.2.1. Répartition et fonction Les menhirs du “ champ mégalithique de Wéris ” se présentent de différentes manières. Ils sont isolés (nos M25 à M27) ou rassemblés par groupe de deux (nos M15-M16), trois (nos M1 à M3) ou cinq (nos M9 à M13). Certains sont disposés à proximité des allées couvertes (nos M9 à M13, M19 à M24), d’autres en sont éloignés.

A l’issue de la fouille, le menhir a été redressé dans sa fosse primitive et les éclats détachés au XXe siècle ont été recollés.

La fonction de ces pierres semble polyvalente. Un rôle funéraire est attesté par les liens qui les unissent aux allées couvertes et par la découverte du fragment d’os humains sur le site du menhir M26. Les sommets angulaires des pierres dressées et leurs alignements évoquent des visées astronomiques. L’existence de telles pratiques serait notamment confirmée par l’alignement du lever du soleil, le jour de l’équinoxe, avec l’allée couverte de “ Wéris I ” et la pierre Haïna, aiguille de poudingue naturelle dominant la crête de Wéris.

4. ANALYSES 4.1. Les allées couvertes A Lamsoul comme à “ Wéris I ” et à “ Wéris II ”, le noyau central que compose la chambre sépulcrale présente une forme sub-rectangulaire avec un chevet et deux longs côtés formés chacun par deux piliers. “ Wéris I ” et “ Wéris II ”, les seuls monuments dont l’avant est conservé, présentent en outre un vestibule, séparé de la chambre par une dallehublot, ainsi qu’une grande dalle située derrière le chevet, peut-être pour le stabiliser (Huysecom, 1981). De tels plans évoquent les allées couvertes du Néolithique récent du Bassin parisien.

4.2.2. Typologie La hauteur moyenne des menhirs varie entre 2,30 et 3,60 m tandis que leur poids oscille entre 2,5 et 8,9 tonnes. Le menhir M25 fait figure d’exception au sein de cet ensemble puisqu’il ne dépasse pas les 60 cm et qu’il ne pèse que quelques centaines de kilos.

L’allée couverte d’Oppagne ou “ Wéris II ” est, comme celle de Lamsoul, installée dans une vaste tranchée d’ implantation, alors que “ Wéris I ” se trouve au niveau du sol.

Une approche typologique préliminaire a analysé les pierres dressées non pas en fonction de leur silhouette générale mais en fonction de la forme de leurs bases, de leurs “ fûts ” et de leurs sommets (Frébutte et al., 2000c). Ces parties ont été apparentées à des solides géométriques distincts et étudiées suivant une application statistique.

Des pierres de calage et de soubassement ont été découvertes sous les orthostates des trois monuments. Elles sont en grès et en calcaire. A Lamsoul et à “ Wéris II ”, des murets de pierres sèches comblent les interstices entre les orthostates.

Le pourcentage prédominant des bases horizontales ou quasi horizontales, soit 76 %, souligne le pragmatisme des constructeurs puisqu’une telle morphologie facilitait l’érection de la pierre, permettait d’économiser les éléments de calage et assurait un maximum de stabilité aux blocs. Pour les fûts, il y a manifestement une recherche de volumes plus

Le chevet de Lamsoul se compose de deux piliers convergents formant un “ V ” bien ouvert alors que les chevets des deux allées couvertes de Wéris sont en “ U ”. A “ Wéris I ”, la dalle sur chant qui fermait la chambre semble cependant avoir disparu. 103

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

ou moins réguliers à très réguliers et symétriques. En attestent les 88 % des fûts qui sont des volumes parallélépipédiques ou polyédriques à face principale trapézoïdale. 70 % des menhirs sont couronnés par une crête dièdre, centrée ou décentrée, ou par deux pans obliques convergents interrompus par un méplat.

humain, à savoir un petit fragment de clavicule de jeune enfant (Toussaint et al., 2000a) ; les trois fragments osseux trouvés en 1906 sur le sites des menhirs M1 à M3 et considérés alors comme humains (de Loë, 1908) appartiennent en fait à un bovidé. Les mobiliers lithique et céramique sont assez pauvres ; ils ont été récoltés sur les sites des allées couvertes, généralement dans les remblais des anciennes fouilles.

Dès lors, la récurrence de certains paramètres morphologiques suggère que le choix des pierres répondait à des critères précis dictés par la fonction de celles-ci et par des contraintes techniques de mise en œuvre. Ces implications sont d’autant plus importantes que les Néolithiques de Wéris, n’étant pas des carriers, devaient composer avec le nombre limité de blocs erratiques disponibles dans la pente de solifluxion.

Les outils en silex les plus représentatifs sont des armatures de flèche. Il s’agit d’une pointe losangique à pédoncule à Lamsoul et de pointes à pédoncule avec ou sans ailerons à “ Wéris I ” et “ Wéris II ” ; une pointe foliacée a été également mise au jour à “ Wéris I ”. Quelques fragments de tessons pourraient évoquer le Néolithique récent.

4.2.3. Vestiges de fondation néolithique et processus de condamnation des menhirs

“ Wéris II ” a fourni des tessons décorés de vases à profil en S de la Civilisation des gobelets (Huysecom, 1981b) ; cette céramique appartient plus précisément au groupe des Allover ornamented beakers, classe 2.II.c de la typologie hollandaise (Lanting & van der Waals, 1976). Ces artefacts, ainsi que des tessons apparentés découverts à “ Wéris I ”, suggèrent une réoccupation des monuments.

La plupart des menhirs ont été élevés dans une excavation aménagée dans la couverture limoneuse ; le substrat rocheux a cependant été entamé sur les sites des menhirs M26 et M27, la couche de sédiment n’étant pas assez épaisse dans ces zones. La forme et les dimensions des excavations témoignent d’une préparation soigneuse et d’une compétence certaine ; elles sont en effet ajustées, parfois très étroitement, aux bases des menhirs. La stabilité des monuments était augmentée par des éléments de calage en calcaire ou en poudingue, l’importance de cette charge dépendant de la régularité des bases.

4.4. Radiocarbone Sept datations radiocarbones par AMS ont été récemment réalisées à partir d’ossements humains découverts dans les mégalithes de la Calestienne, par la Radiocarbon Accelerator Unit, Research Laboratory for Archaeology and the History of Art de l’Université d’Oxford. Après calibration, toutes ces dates s’insèrent dans la première moitié du troisième millénaire avant notre ère. Ci-dessous figure la liste des datations AMS classées par site avec l’indication du contexte de découverte des ossements utilisés.

Les dalles dégagées près des menhirs M14, M16 et M19 demeurent énigmatiques : leur rôle supposé est de renforcer le bord de la fosse sur lequel glissait le bloc lors de son érection (Hubert, 1985 : 22). Des expérimentations futures testeront la validité de cette hypothèse. Seul le menhir M25 et les bases des menhirs fracturés M15 et M21 étaient encore en place dans leur fosse d’érection. Les autres blocs ont été systématiquement déchaussés et basculés dans des excavations voisines.

L’allée couverte de Lamsoul (Jadin et al., 1998) - OxA-6454 = 4115 ± 60 BP. L’échantillon daté est un astragale (ou talus) trouvé dans la rainure d’implantation située sous l’orthostate VII.

Au stade actuel des fouilles et des études, aucune trace certaine de destruction des menhirs par les Néolithiques euxmêmes n’a été démontrée. Quelques tessons attribuables au XVIe ou au XVIIe siècle fournissent un terminus postquem pour l’enfouissement des menhirs M9 à M13 et M15 et M16. Ces agissements résultent probablement de considérations catholiques visant à effacer de la mémoire les témoins d’anciennes croyances. D’autres monuments ont aussi pu être éliminés, comme cela s’est encore produit récemment, pour des raisons économiques, parce qu’ils gênaient les exploitations agricoles.

- OxA-6455 = 4055 ± 55 BP. L’échantillon daté est une phalange en provenance de la base de la chambre sépulcrale. L’allée couverte de “ Wéris I ” (Toussaint & Hubert, 1998c) - OxA-6457 = 4240 ± 65 BP. L’échantillon se composait d’une phalange adulte trouvée en 1983 dans les pierres calcaires formant le soubassement de l’entrée. - OxA-6458 = 4170 ± 60 BP. L’échantillon se composait d’un petit fragment crânien trouvé en 1983, dans les pierres calcaires formant le soubassement de l’entrée.

4.3. Matériel anthropologique et archéologique

L’allée couverte de “ Wéris II ” (Toussaint et al., 2000b)

Les nombreux remaniements qui ont touché les trois allées couvertes expliquent vraisemblablement la rareté des esquilles osseuses humaines qui y ont été exhumées. A “ Wéris II ” et à Lamsoul, quelques ossements démontrent la présence conjointe d’enfants et d’adultes. Le menhir M26 est le seul monument de ce type qui était associé à un reste

- OxA-8939 = 4180 ± 40 BP. L’échantillon se composait d’un cinquième métatarsien gauche adulte. - OxA-8956 = 4240 ± 45 BP. L’échantillon se composait d’un deuxième métacarpien gauche adulte. 104

C. Frébutte et al.: 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique)

Le menhir M26 (Toussaint et al., 2000a)

les distances entre les lieux de récolte des poudingues et les divers sites mégalithiques varient entre quelques centaines de mètres et 4 km. Les pierres de calage et les éléments de dallage, en grès ou en calcaire, sont également d’origine sub-locale.

- OxA-8828 : 4425 ± 45 BP. L’échantillon est un petit fragment de clavicule d’un jeune enfant découvert au sein de la fosse d’érection du menhir. 4.5. Paléoanthropologie et relation avec les sépultures plurielles en milieu karstique

5. PERSPECTIVES La fouille de 1995-1996 à l’allée couverte de Lamsoul n’avait pas permis d’explorer en détail l’arrière et le côté droit du monument. Ces zones feront l’objet de nouvelles investigations dans les années à venir, indispensables pour l’interprétation globale du site.

Au fil du programme de fouilles des mégalithes de Wéris, il a été remarqué que les ossements humains des quelques sépultures collectives du karst régional, datés au radiocarbone, avaient la même ancienneté que ceux des allées couvertes de “ Wéris I ”. C’est le cas au Trou de la PJ (Lv2136 = 4220 ± 80 BP et Lv-2137 = 4130 ± 60 BP), à la grotte du Coléoptère (OxA-3636 = 4695 ± 65 BP) et à l’abri de la Sigillée (IRPA-920 = 4460 ± 60 BP), ainsi qu’à la grotte sépulcrale de Barvaux, ou encore à l’abri de Hottemme (Lv863 = 4050 ± 50 BP). Lorsque ces sites funéraires karstiques ont livré du matériel archéologique, celui-ci est en outre assez similaire à celui des allées couvertes, notamment en ce qui concerne la morphologie des pointes de flèches à pédoncule.

Les perspectives engendrées par les nouvelles recherches à Wéris sont plus vastes. Il y aura d’abord lieu de terminer la rédaction des trois volumes de monographie relatifs aux fouilles de 1979 à 1986 et de 1995 à 2001. Le premier de ces volumes présentera la géologie du secteur de Wéris, replacera le champ mégalithique dans le contexte de la Wallonie et des régions frontalières, synthétisera les données acquises avant la reprise des fouilles en 1979 et analysera en détail l’abondante littérature scientifique consacrée au site. Le deuxième volume décrira en détail le résultat des fouilles exécutées sur le champ mégalithique de 1979 à 2001. Le troisième sera consacré à des essais d’analyses et de synthèses des données accumulées aux cours des fouilles récentes, dans tous les domaines abordés, tant les sciences naturelles (géologie, pétrographie, palynologie, pédologie, paléoanthropologie…) que l’archéologie proprement dite.

Les Néolithiques de la région de Wéris, comme sans doute de celle de Lamsoul, recouraient donc à deux types de pratiques plus ou moins simultanément. Les critères de choix de ces pratiques demeurent problématiques ; ils pourraient refléter une certaine structuration sociale des populations avec, par exemple, des personnages plus importants inhumés dans les monuments construits et la plus grande partie des défunts disposés dans les abris naturels. Seule la découverte de nouveaux ossements dans le champ mégalithique, donc la découverte d’une troisième allée couverte non perturbée, pourrait permettre de tester ces hypothèses, notamment en comparant la stature, l’alimentation et la pathologie des types de défunts exhumés dans ces contextes distincts.

La restauration et la mise en valeur de l’ensemble des monuments connus du champ mégalithique restent également à réaliser, dans le cadre de collaborations avec divers partenaires (Ville de Durbuy, Musée des Mégalithes à Wéris…).

4.6. Pétrographie

Dans la mesure où de nombreux indices montrent qu’un certain nombre de monuments, au minimum des menhirs, sont encore enfouis dans le sous-sol du champ mégalithique, les trois monographies et la mise en valeur du champ mégalithique ne clôtureront cependant pas les investigations encore indispensables à Wéris. Un troisième programme de recherches modernes sur le terrain devra donc être mis sur pied. Il devrait débuter par la réalisation de prospections systématiques - géophysique, photographie aérienne et tranchées exploratoires - de tout le tracé du champ de menhirs et d’allées couvertes, soit une bande d’au moins 8 km de long sur quelques centaines de mètres de large, au minimum entre les trois menhirs d’Oppagne (M1 à M3) et celui d’Ozo (M27). A l’issue des fouilles sur lesquelles ces prospections déboucheront, il devrait être possible, plus d’un siècle après les premières recherches, d’apporter un éclairage encore plus complet à la connaissance du “ champ mégalithique de Wéris ”.

Les stratégies d’acquisition des matériaux utilisés dans la construction des monuments mégalithiques de Lamsoul et de Wéris sont assez uniformes. Les dalles composant l’allée couverte de Lamsoul sont toutes en calcaire givétien (Pirson, 1997). Le substrat géologique se compose par contre de schiste. La construction du monument a donc impliqué le déplacement de blocs pesant souvent plusieurs tonnes sur une distance au moins égale à celle qui sépare Lamsoul des affleurements calcaires les plus proches, soit au minimum 650 m et peut-être près de 2 km, avec franchissement de dénivellations de quelques dizaines de mètres. Les nombreux petits blocs remplissant la tranchée d’implantation et composant l’empierrement antérieur montrent également une origine sub-locale ; ils peuvent être classés en 31 classes lithologiques, toutes disponibles dans un rayon de quelques kilomètres. A Wéris, les piliers et les dalles de couvertures des allées couvertes, comme les divers menhirs, sont tous en poudingue. Ils ont été probablement prélevés parmi les erratiques descendus par solifluxion depuis le bord du plateau à l’est où la roche affleure. En estimant les trajets les moins difficiles,

Remerciements Les auteurs tiennent à adresser leurs remerciements à toutes les personnes et institutions qui les ont secondés au cours 105

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

des opérations archéologiques, particulièrement les techniciens de l’Association wallonne d’Etudes mégalithiques, des membres du personnel de la Direction de l’Archéologie, le Collège des Bourgmestre et Echevins de la Ville de Durbuy et le personnel du Musée des Mégalithes de Wéris.

FRÉBUTTE C., TOUSSAINT M., PIRSON S., ANSLIJN J.-N., HUBERT F. & MASY P., 2001b. Durbuy/Wéris : fouilles à l’allée couverte nord ou “ Wéris I ”, Chronique de l’Archéologie wallonne, 9, p. 154-157. HUBERT F., 1980. Fouilles de contrôle au dolmen nord de Wéris. In : Conspectus MCMLXXIX, Bruxelles (Archaeologia Belgica, 223), p. 41-42. HUBERT F., 1982. Fouilles de contrôle au dolmen nord de Wéris. In : Conspectus MCMLXXXI, Bruxelles (Archaeologia Belgica, 247), p. 18-30.

Adresses des auteurs

HUBERT F., 1985. Recherches aux mégalithes de Wéris, Archaeologia Belgica, I, 2, p. 17-24.

Christian FRÉBUTTE & Michel TOUSSAINT Direction de l’Archéologie du Ministère de la Région wallonne, Rue des Brigades d’Irlande 1, B-5100 Namur (Jambes) BELGIQUE [email protected] ; [email protected]

HUBERT F., 1987. Les menhirs de l’allée couverte II de Wéris (comm. de Durbuy), Archaeologia Belgica, III, 3, p. 77-82. HUBERT F., 2000. Les mégalithes de Wéris avec une boussole. In : 8ème Journée d’Archéologie luxembourgeoise, p. 5-13. HUBERT F. & HUYSECOM E., 1983. Troisième campagne à l’allée couverte I de Wéris. In : Conspectus MCMLXXXII, Bruxelles (Archaeologia Belgica, 253), p. 25-27.

François HUBERT, Philippe MASY & Stéphane PIRSON Association wallonne d’Etudes mégalithiques, Avenue des Tilleuls 62, B-4000 Liège BELGIQUE [email protected] ; [email protected]

HUBERT F. & HUYSECOM E., 1984. Quatrième campagne à l’allée couverte I de Wéris. In : Conspectus MCMLXXXIII, Bruxelles (Archaeologia Belgica, 258), p. 15-19. HUYSECOM E., 1981a. Les allées couvertes de Wéris, Bulletin trimestriel de l’Institut archéologique du Luxembourg, 57 (3-4), p. 63-132. HUYSECOM E., 1981b. A propos de fragments de gobelets “ AOO ” exhumés des allées couvertes de Wéris (Luxembourg), Helinium, XXI, p. 56-60.

Bibliographie DANTHINE H., 1961. Un menhir découvert à Wéris en 1947, Annales de l’Institut archéologique du Luxembourg, 92, p. 2935.

HUYSECOM E., 1983. Examen de l’allée couverte de LamsoulJemelle, Parcs nationaux. Bulletin trimestriel de l’association Ardenne et Gaume, XXXVIII, fasc. 3, p. 132-139.

DE LOË A., 1908. Nos recherches et nos fouilles durant le deuxième semestre de 1906 (suite). Fouilles à Oppagne (province de Luxembourg), Bulletin des Musées royaux des Arts décoratifs et industriels, 2e série, I, p. 52-53, fig. 2. FRANKE W., 1989. Tectonostratigraphic units in the Variscan belt of Central Europe, Geological Society of America, special paper, 230, p. 67-90.

JADIN I., TOUSSAINT M., BECKER A., FRÉBUTTE C., GOFFIOUL C., HUBERT F. & PIRSON S., 1998. Le mégalithisme de Famenne. Approche pluridisciplinaire et perspectives, Anthropologie et Préhistoire, 109, p. 95-119.

FRÉBUTTE C., HUBERT F., TOUSSAINT M. & DEWAMME E., 1998. Durbuy/Wéris : redressement des menhirs de “ Wéris II ”, Chronique de l’Archéologie wallonne, 6, p. 128.

LANTING J.-N. & VAN DER WAALS J.-D., 1976. Beaker Culture Relations in the lower Rhine Basin. In : GlockenbecherSymposium, Oberried 1974, Haarlem, p. 1-80.

FRÉBUTTE C., PIRSON S., TOUSSAINT M., HUBERT F., LAMBERMONT S. & MASY P., 2000a. Durbuy/Izier : fouille et redressement d’un menhir de type “ Zeupire ” à Ozo, Chronique de l’Archéologie wallonne, 8, p. 152-154.

PIRSON S., 1997. Etude pétrographique des monuments mégalithiques de Lamsoul et Wéris II, Notae Praehistoricae, 17, p. 195-197. STAINIER P., 1994. Formation de Hampteau. In : Godefroid J., Blieck A., Bultynck P., Dejonghe L., Gerrienne P., Hance L., Meilliez F., Stainier P. & Steemans P., Les formations du Dévonien inférieur du Massif de la Vesdre, de la Fenêtre de Theux et du Synclinorium de Dinant (Belgique, France), Bruxelles (Mémoires pour servir à l’Explication des Cartes Géologiques et Minières de la Belgique, 38), p. 91-96.

FRÉBUTTE C., PIRSON S., TOUSSAINT M., HUBERT F., LAMBERMONT S. & MASY P., 2000b. Durbuy/Wéris : nouvelles fouilles à l’allée couverte nord “ Wéris I ”, Chronique de l’Archéologie wallonne, 8, p. 156-158. FRÉBUTTE C., TOUSSAINT M. & HUBERT F., 2000c. Essai d’analyse des menhirs du champ mégalithique de Wéris (Durbuy). In : 8ème Journée d’Archéologie luxembourgeoise, p. 14-26.

STAINIER X., 1902. Carte géologique de la Belgique au 1/40.000, no 169 : Durbuy-Mormont. Commission géologique de Belgique. TOUSSAINT M., BECKER A. & HUBERT F., 1996. Fouilles 1996 à l’allée couverte d’Oppagne, ou “ Wéris II ”, à Durbuy (province de Luxembourg), Notae Praehistoricae, 16, p. 197-208.

FRÉBUTTE C., TOUSSAINT M., HUBERT F. & MASY P., 1999. Fouilles au menhir “ å Djèyî ”, une extension septentrionale du champ mégalithique de Wéris, Chronique de l’Archéologie wallonne, 7, p. 123-126.

TOUSSAINT M., FRÉBUTTE C. & HUBERT F., 2000a. Durbuy/ Heyd : datation AMS d’un os humain découvert au menhir “ å Djèyî ”, à Tour, Chronique de l’Archéologie wallonne, 8, p. 151152.

FRÉBUTTE C., TOUSSAINT M., HUBERT F., MASY P. & PIRSON S., 2002. Nouvelle campagne archéologique sur le site des trois menhirs d’Oppagne à Wéris (commune de Durbuy), Bulletin trimestriel de l’Institut archéologique du Luxembourg, 1-2.

TOUSSAINT M., FRÉBUTTE C. & HUBERT F., 2000b. Durbuy/ Wéris : deux datations AMS d’ossements humains découverts à l’allée couverte “ Wéris II ”, Chronique de l’Archéologie wallonne, 8, p. 155-156.

FRÉBUTTE C., TOUSSAINT M., MASY P., PIRSON S. & HUBERT F., 2001a. Campagne archéologique 2001 sur le site du “ champ mégalithique de Wéris ” à Durbuy (province de Luxembourg), Notae Praehistoricae, 21, p. 156-173.

TOUSSAINT M., FRÉBUTTE C., HUBERT F. & DEWAMME E., 1997a. Campagne de fouilles 1997 à l’allée couverte “ Wéris 106

C. Frébutte et al.: 1995-2001 : Sept années de recherches mégalithiques en Calestienne (Belgique) II ” (Durbuy, prov. de Luxembourg), Notae Praehistoricae, 17, p. 185-193.

TOUSSAINT M. & HUBERT F., 1998b. Morville, un nouveau menhir au champ mégalithique de Wéris, Durbuy (province de Luxembourg). Namur (Etudes et Documents, série Archéologie, 5), p. 79-87.

TOUSSAINT M., FRÉBUTTE C., HUBERT F. & MASY P., 1998a. Fouilles 1998 au champ mégalithique de Wéris : le menhir du lieu-dit “ å Djèyî ” à Heyd (Durbuy, province de Luxembourg), Notae Praehistoricae, 18, p. 157-165.

TOUSSAINT M. & HUBERT F., 1998c. Durbuy/Wéris : datations 14 C de l’allée couverte “ Wéris I ”, Chronique de l’Archéologie wallonnne, 6, p. 126.

TOUSSAINT M., FRÉBUTTE C., PIRSON S., ANSLIJN J.-N., HUBERT F., MASY P., LAMBERMONT S., LACROIX P. & BEAUJEAN J.-F., 2000c. Fouilles 1999 et 2000 à l’allée couverte de “ Wéris I ” (Durbuy, province de Luxembourg), Notae Praehistoricae, 20, p. 177-196.

TOUSSAINT M. & JADIN I., 1996. Fouilles 1995-1996 à l’allée couverte de Lamsoul (Jemelle, Rochefort, province de Namur), Notae Praehistoricae, 16, p. 183-195. TOUSSAINT M., JADIN I., HUBERT F. & BECKER A., 1997b. Les monuments mégalithiques de Lamsoul, “ Wéris I ” et “ Wéris II ”. Un essai de comparaison. In : PLUMIER J. & DUHAUT C. (dir.), Cinquième journée d’Archéologie namuroise. Actes 5, Gembloux, p. 31-52.

TOUSSAINT M., FRÉBUTTE C., PIRSON S., HUBERT F., LAMBERMONT S. & MASY P., 1999. Fouilles 1999 au champ mégalithique de Wéris : le menhir d’Ozo (Durbuy, province de Luxembourg), Notae Praehistoricae, 19, p. 143-153.

107

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

108

P. Prieto Martínez: Ceramic Style in Neolithic Societies in Galicia (NW Iberian Peninsula)...

CERAMIC STYLE IN NEOLITHIC SOCIETIES IN GALICIA (NW IBERIAN PENINSULA). SIMILARITIES AND DIFFERENCES IN PATTERNS OF FORMAL REGULARITY M. P. PRIETO MARTÍNEZ

Résumé : Le style céramique aux sociétés néolithiques à la Galicia (Nord-ouest de la Péninsule Ibérique). Patrons de régularité formelle: Similitudes et différences. On présent l’évolution du style céramique au Néolithique galicien au travers des aspects formels en utilisant la Chaîne opératoire comme outil méthodologique et l’Archéologie du Paysage comme cadre théorique. On utilise un enregistrement empirique tout neuf à Galicia: les sites en plein air. On utilise une méthodologie qui nous permet d’étudier de manière systématique la poterie appartenante à différents moments et contextes du Néolithique. On cherche le patron de régularité formelle, c’est à dire, similitudes et différences stylistiques, et à cette fin, on réalise une comparaison à divers degrés de la chaîne opératoire: (a) céramique funéraire et domestique au Néolithique Initial/Moyen, (b) céramique funéraire et domestique au Néolithique Final, (c) céramique tout au long du Néolithique. On croit que la reconstruction de la technologie céramique c’est un autre aspect qui peut contribuer à savoir si les changements formels peuvent être en rapport avec les changements au patron de rationalité des sociétés neolithiques à Galicia. Abstract: We present an evolution of ceramic style in the Galician Neolithic period considered using its formal characteristics. We used the Operative Technological Chain (Chaîne Opératoire) as a methodological tool and applied the theoretical premises of Landscape Archaeology. The archaeological record used is quite innovative in Galicia: open-air pottery sites. This methodology allows us to systematically study pottery from different periods and contexts in the Neolithic period. We were searching for the pattern of formal regularity, stylistic similarities and differences; to do so, a comparison was made at several stages of the Technological Chain: (a) funerary and domestic pottery in the Early/Middle Neolithic, (b) funerary and domestic pottery in the Later Neolithic, and (c) all pottery throughout the Neolithic period. We believe that the reconstruction of pottery technology is an important aspect that helps us to investigate formal and social changes and their relationship with transformations in the rationality of Neolithic societies. Keywords: Operative Sequence, Operative Technological Chain (OTC), Ceramics, Formal Analysis, Style, Patterns of Formal Regularity, Landscape Archaeology, Later Neolithic, Early/Middle Neolithic, Galicia.

1. INTRODUCTION*

about pottery from the Late Neolithic (LN) (2700-2300 BC), referred to in the Iberian Peninsula as the Chalcolithic Period, is much more plentiful, as more sites are known and some of them have been excavated. Several articles have been published about these sites in the last ten years, although the most complete investigation comes from northern Portugal (Jorge O.S. 1986). Today in Galicia, despite a lack of information about sites from the LN, it is still possible to fill in more details. At the end of the 1980’s the first sites from this period were discovered (González 1991, Eguileta 1996). The information which exists about open-air Neolithic sites has still not been brought up to date (the widest range of ceramics studied systematically is detailed in Prieto 2000), partly because the empirical base has grown considerably in the last decade, and partly because it has not been fully published.

The little information which exists about Neolithic ceramics in the Iberian Peninsula is quite irregular and fragmented, and there are very different degrees of awareness about the periods it comprehends, as we shall see. Investigation into the ceramics from the Early/Middle Neolithic (E/MN) (45002700 BC), considered in the Iberian Peninsula as a strict definition of the Neolithic period, is incomplete. In fact, the most recent study dates from the end of the 1980’s, and attempts to present a summary of the information which existed until that point in time (López 1988)1. In Galicia, situated in the north-western corner of the Iberian Peninsula, there is even less information about this period of recent prehistory, as the only studies published to date deal with three little-known open-air sites in O Reiro, A Cunchosa (Vázquez 1988) and O Regueiriño (Suárez and Fábregas 2000). More information is available for funerary contexts, which has led to the publication of two articles (Eguileta 1987, Fábregas and Fuente 1988). The information available *

This paper was presented in the XIVth Congress of the U.I.S.P.P. (Liège, Belgium, 2-8 Sept. 2001). Section: 9. Néolithique au Proche Orient et en Europe / Neolithic in the Near East and Europe.

1

To date the most prominent studies have been presented at congresses, for example at the Congreso del Neolítico in the Gavá Peninsula in 1996, in Valencia in 1999, or in several Peninsular Archaeology congresses, in particular in the year 2000. However, since then there have been no further studies of the period.

2. EMPIRICAL PREMISE The empirical premise of this study is the investigation of ceramic materials recovered from a total of 53 sites (fig. 1), in both domestic and funerary contexts. For the domestic context, our documentary material is based on 40 sites, although we directly worked with 16 open air sites uncovered by archaeological investigations connected with major public works (oil pipelines, gas pipelines, 109

P. Prieto Martínez: Ceramic Style in Neolithic Societies in Galicia (NW Iberian Peninsula)...

Penha-type ceramics have been discovered), in some of which bell-beaker type ceramics have been found.

• The degree of visibility and material complexity are related to the degree of social complexity. • Ceramic production is the result of strict cultural criteria (although these are not predetermined).

There are many archaeological difficulties connected with these sites, as there are very few excavations in domestic contexts, and the little information we do have about them is based on the result of exploratory studies. In funerary contexts, the excavations that do exist are now outdated, there is little documentation, and what does exist is unreliable5. The only real way of comparing these ceramics is through systematic formal analysis, which will help to lay the foundations for the study of ceramics discovered in sites excavated in the future.

• The variety of ceramics corresponds to social fluctuations. • The use of ceramics is differential. • Ceramic production is local. Our main objective is to characterise Neolithic ceramics from Galicia8 and to compare their different features, based on their production process. It will then be possible to define the patterns of formal regularity of societies which have existed in specific areas in particular historical periods. We also attempt to contrast the presence of formal categories within each of these patterns of regularity, with the intention of connecting them with the societies that produced them. This means that it is not only important to analyse the way different formal products are manufactured as a technological feature, but also to characterise the way of creating formal identities using ceramic products as a cultural feature. By identifying the aspects which maintain a certain homogeneity or ‘continuity points’ within and between periods and ‘rupture points’ which indicated a heterogeneous situation, it is possible to detect the changes which occurred within these societies, and to interpret these changes at sociological level.

3. GENERAL PROPOSALS Our study is a proposal based on Landscape Archaeology, which considers archaeological elements and the material culture contained within them not as isolated elements, but instead as forms produced by social action, and which may therefore only be fully understood in relation to the cultural context within which they were created (Criado et al. 1991, Criado 1993). Accordingly, both pottery as a specific material product and the person who makes it are social products, and therefore reproduce and produce the symbolic system of the society to which they belonged. This does not only mean that the finished products are socially significant, but also that the technological knowledge which made a ‘manner of doing’ possible is profoundly involved in the whole system (Cobas and Prieto 2002). The formal models, and the specific ceramic styles within them, are transformed according to the patterns of rationality which governed a particular society6.

At methodological level, we will describe the reconstruction of the operative technical chain (Cobas and Prieto 1998a and 2002) of two important monuments from the Galicia Neolithic period. We use the initial formal deconstruction of each of the groups of elements, of the features related to the morphology of the site, their technical treatment and decoration. Our procedure begins by describing the specific aspects of the ceramics found until arriving at the most general aspects of the global group of elements, although in this particular paper we have only included the results of the investigative process.

The working methodological hypotheses, which have been proposed and explained in greater detail in previous articles (Prieto 1998, 1999 and 2001) are the following: • The degree of visibility7 is an important element in ceramic production, related to other material circumstances, and according to the social criteria which existed when it was made.

We will therefore attempt to firstly characterise the ceramic styles of each type of society using the different formal categories that were documented at the time of describing the materials. ‘Category’ describes the formal group formed by the selection of a series of specific technical choices according to the technological knowledge of these societies (similarities or continuity points). Within each category, it is possible to see stylistic tendencies, which are the varieties or versions of each of the categories; by using them it is possible to observe differences or rupture points, without detracting from the general features observed within the same category. The formal relationships or relationships between categories mean that it is possible for us to globally define the ceramic style of each of these societies, within which there is a pattern of formal regularity, or a ‘range of restricted rules’ (from Lemonnier 1986: 153) that are maintained within a certain

• Ceramic production, and not only decoration, takes place according to the visibility strategies of social action. 1987), Cotogrande 1 and 5 (Abad 1995), Dolmen de Dombate (Bello 1991), Gándaras and San Andrés de Lousada (Criado and Vázquez 1982). 5

For this reason, we have selected so few sites from the Mid-Neolithic, as the precise definition of the ceramics is not clear in the bibliography. Furthermore, the group of sites we have used is sufficient for us to carry out a formal comparison between contexts, despite its small size.

6

It would be interesting to complement this paper with another being presented at this same congress which deals with ceramics from the Bronze Age in Galicia (Prieto 2002).

7

Throughout the text, when dealing with the different characteristics related to the process of manufacturing ceramic pieces, we have also evaluated if these could contribute to the invisibility, visibility, ‘monumentalising’ or concealment of the finished products. We have considered two different aspects: firstly, formal aspects according to the OTC, and secondly (when this has been possible), the contextual aspects, considering how ceramic products are presented within the context in which they are found. This is only possible in the case of excavated sites, meaning that information of this kind will be more limited.

8

111

This study follows in the line of investigation in Material Culture developed by the LACF. Some important results have been obtained by comparing ceramics from the Bronze and Iron Ages (Cobas and Prieto 1998a, 1998b, 1999) and ceramics from recent Prehistory in open-air sites (Prieto 2000), in Galicia.

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

type of technological knowledge or symbolic system. Tendencies or substyles are not types; instead they are variants within a formal category, and do not directly refer to specific types of ceramics, but refer to the use of a series of common and specific technical resources conditioned by the category involved, despite the fact that all this is in turn submerged within a modus operandi which finally refers to style9 (in this case, ceramic style), a formalisation based on technological knowledge. In fact, it is quite common to observe the simultaneous existence of at least two stylistic tendencies corresponding to one or two different categories within the same site.

elements may be equally laden with symbolic significance. We should therefore evaluate all types of ceramic materials, not only those with decorations, as doing so may give us information about some of the cultural norms of the societies we are studying11. With the morphological treatment, although technical simplicity governed the Neolithic period, in the LN a wider variety of resources was used, particular with regard to general morphologies, necks and edges. The morphologies here are similar in both contexts throughout the Neolithic: in the E/MF there was a predominance of simple closed profiles within the funerary context, compared to open-air settlements which offer a wider variety. In the LN simplicity may be seen in both contexts, with a prevalence of ‘longobordos’ – flared neck edges - in the domestic environment, and some carenated ceramics in the funerary context. One particularly outstanding aspect in both periods and contexts is the morphological similarity found in all of the pottery. This scant variety may be seen in several sites.

4. THE FORMAL CHARACTERISATION OF NEOLITHIC CERAMICS IN GALICIA: STUDY We carried out our investigation by comparing the formal features of ceramics from the E/MN and LN using the operative technical chain as our methodological instrument. We indicated the common aspects or ‘continuity points’ and identified aspects which were different, or ‘rupture points’ throughout the Neolithic period and between different contexts, and indicated which aspects made it possible to refer to the degree of visibility of the ceramics in the final product (figures 2 and 3).

With regard to technical treatment, there are several common features in Neolithic ceramics. They are hand-made, with mainly light colours, monochrome oxide firing, finishes both rough and medium-smooth, and there are no technical differences between undecorated and decorated ceramics within each site. In general, ceramic pieces from the Neolithic period have a matt finish, with some rare exceptions. Again, this scant variety may be seen from site to site. It is difficult to define a site when only undecorated ceramics are found, meaning it is necessary to use the global information from the archaeological record to evaluate it. Aspects which may differ slightly are the following:

Although the term ‘visibility’ is applied to studies of landscape archaeology, we believe that it is closely related to all aspects of Material Culture, as “the fact the results of social action itself, whether they are more or less visible, may be related to a ‘desire for visibility’ which may be either conscious and explicit, implicit, or even unintentional” (Criado Boado 1993: 43). We consider ‘visibility’ to be “the way of exhibiting and emphasising the products of Material Culture which reflect the existence of a social group. Considering that the effects of this social process are reflected spatially, we may define the conditions of visibility of the archaeological record by ‘looking over’ the elements which form it, and attempt to work out the whys and wherefores of their visual features” (Criado Boado 1991: 23).

• In the E/MN there was a predominance of fine and thick textures, whereas in the LN they were medium. • There is a wider range of grains (mica, quartz, granite) in the LN than in the E/MN (micas, granite and quartz are rarely seen). • In the LN, some pieces with a glossy finish (with more refined manufacture, micas, dark tones and reduction firing) started to be differentiated from the main body of ceramics. It was an appearance which contributes to differentiating substyles within a formal category, and is rare in pieces from the E/MN.

The concept of visibility may be applied to any element of material culture, and is specifically applied to ceramics by analysing both the different technical choices made in the different phases of manufacture as well as its material consequence or final product10. This means that as well as considering their decorations, it is important to consider aspects related to their morphology, size, grain, finish and colour, as these may be used indifferently as resources for a visibility strategy. However, we should not overlook the invisible elements (n the final product) in the sequence of technological production (compactness of the clay used, modelling techniques, grain, firing technique), as unseen 9

“Considered as one of the mechanisms of the discourse of power, as may be seen in the formal products of a society: the external formalisation of power, in its ‘Foucaultian’ sense (see Foucault’s concept of power in Foucault 1981)” (Prieto 1998, 1999).

10

Sometimes, the same result is not obtained after making the same technical choice, as the final result depends on the different combinations present in each stage of production.

Finally, the decorative treatment used represents the most important differences between ceramics from the E/MN and the LN. Although it is true that a common model was maintained, there are stylistic differences. The aspects which give a sense of continuity in both periods are: the techniques used (incision and stamping) the instruments used (punches and combs), as well as the design of basic decorative 11

112

The degree of visibility, achieved in a given ceramic piece following the choice of a specific technique throughout the technological sequence, acquires a new, more enriching significance if we consider other aspects, such as the different visibility strategies for different contexts known in Recent Galician Prehistory– the domestic and funerary contexts. This aspect was considered in a previous study applied to the Bronze Age (Prieto 1998). A preliminary summary is included in Cobas and Prieto (1998b).

P. Prieto Martínez: Ceramic Style in Neolithic Societies in Galicia (NW Iberian Peninsula)...

Figure 2. Formal characterisation of Early/Middle Neolithic pottery in Galicia (NW Spain) 113

P. Prieto Martínez: Ceramic Style in Neolithic Societies in Galicia (NW Iberian Peninsula)...

elements. The motifs offer a break with the thematic sequence, with important differences in size and the types of stamps applied. Decoration is applied to specific areas, on the upper part of the ceramics; the orientation is both horizontal and vertical, and to ‘read’ the piece properly it must be turned both horizontally and around its axis; the visibility of the decorative element is real (by now there is no possibility of confusing techniques and elements of different types); the visibility of the decoration is based on the way the different decorative instruments were applied, and on the decoration having been applied to the most visible part of the piece. The variety of decorations depends on each site, and it is difficult to find identical decorations in two different places. The aspects which best define the differences for both periods are as follows:

when the decoration is invisible. The ceramic piece is ‘invisible’, as all the technical choices in the different stages of manufacture contribute to achieving a final, matt product. Both decorated and undecorated ceramics are invisible. The morphologies of decorated and undecorated pieces are similar, which contributes to not perceiving the differences which exist between them; technically they have the same characteristics, using clays rich in micas, with light colours and medium finishes (either smoothed or brushed), with a matt finish. The few outstanding pieces are dark, burnished, micaceous, and are not necessarily decorated. There are only three outstanding pieces from the total selection, two of which are burnished and slightly sparkling. One is morphologically important as it is carenated, the other because it is larger (it has a corded section decorated with a pattern made using the fingernails), and the third because it is both large and has a different type of decoration within the funerary context. They are important pieces as they have ‘suggestive’ features which are different from the other ceramics.

• In the LN decoration with combs appeared, together with sharp punch marks, and it appears that marks made with fingernails disappeared, as had been common in the E/MN. • Score marks made by instruments are larger and deeper in the LN than in the E/MN, where at times decorations are barely visible on first inspection.

In the LN, despite the fact that undecorated and decorated ceramics continued to be made using the same manufacturing process both at morphological and technical level, more visible differences appeared with their decorative features. It is therefore possible to refer to the presence of two ceramic categories: undecorated and decorated. There are also differences in the types of decorated ceramics, at formal and contextual level, as it is possible to see differences in the degree of visibility of some pieces compared to others, meaning that within this category we can refer to three stylistic tendencies.

• There is a wider range of marks in the LN, together with more decorative elements and motifs. • The element, motif and scheme are formally the same in the E/MN, whereas in the LN the element (as the smallest object within the design) and scheme (as the decorative whole) are more highly differentiated and complex. • There is no strict positioning of the elements, motifs and schemes in the E/MN. However, there is upper and lateral definition together with a very obvious separation of designs in the LN.

Consequently, there is an undecorated, ‘invisible’ type of ceramics (one category and one stylistic tendency), with a matt finish and quite unassuming, as the clay used has light, gentle tones; the glazes used on the surfaces vary between granite, quartz and mica, and have predominantly rough or medium finishes; there are few pieces with any type of treatment that makes them more conspicuous (smooth or polished finishes, micas or strong colours are the exception).

• The type of rupture in the sequence of motifs is more complex in the LN, combining elements with different directions, or introducing large, undecorated sections which help to break the thematic continuity of the design. In the E/MN the formal break between motifs is very faint, and is limited to the use quite narrow undecorated sections. • Curved lines appeared in the LN, which had not yet been used in the E/MN.

However, there is variability in decorated ceramics, as three formal resources may be seen in their design which makes them ‘visible’: their larger size (stamp marks and other elements) and the interplay of different positions used in their motifs. These differences in the degree of visibility are yet another aspect which confirms the existence of three stylistic tendencies within this ceramic category. These are:

• The situation of decorations in the E/MN is limited to the upper third of the ceramics (from the upper part of the belly to the lip), whereas in the LN this was extended down to the middle part of the belly, occupying the upper two thirds of the piece. • Decorations from the LN offer more possibilities for visibility, and are quite conspicuous compared to decorations from the E/MN, which despite all efforts are still invisible.

• Pottery with decorated borders, which is visible, although not to the extent of metopada ceramics. They have a bright appearance (with dark colours and micas), or may be burnished or be given a rough finished that is not quite as visually appealing. Despite being visible, their decoration may be diluted after the clay is treated, as it was either first decorated and then had the final finish applied to its surface, or to its rim (in the case of pieces with horizontal, less visible rims). To date, this type of pottery has only been recorded in domestic contexts.

As regards the final product obtained, we found differences between the LN and E/MN. In the E/MN, we may refer to a single formal category, considering the morphological and technical similarity present, and which contains two tendencies: undecorated and decorated ceramics. The presence of decoration distinguishes them slightly, although it is a subtle difference

• Simple metopada ceramics have visible decoration, are large, and are made more attractive by the treatment given 115

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

to the clay (using clear and bright colours) . Again, this type of pottery has only been found in domestic contexts.

change in the formal conception of ceramics, and the relationships between different ceramic products and the context in which they appeared (Prieto 2002). These, together with other social areas such as emplacement, contain a range of differential characteristics which are significant enough to reveal a number of strong socio-cultural differences related to the Chalcolithic period. It would therefore appear to be more convenient to use the concept of the LN.

• Complex metopada ceramics have very conspicuous decorations, with the treatment given to the clay contributing to their appeal (using bright, clear colours and mica grains). This type of material is mainly found in funerary contexts12. In general terms, both the undecorated and decorated ceramics have a similar morpho-technical treatment. Their decoration is the most prominent feature when estimating differences in degrees of visibility in the pieces. Their classification is further complicated as we proceed through the Neolithic period, as we go from having one ceramic category with two tendencies in the E/MN, to having two categories and three tendencies within one of these categories in the LN.

We end this text by proposing a working hypothesis for the future, which has been more fully developed for Bronze Age societies (Prieto 1998, 1999 and 2001). In it we consider that ceramics, like any other element of material culture, express in some way both the social relationships and the pattern of rationality of the society that manufactured them. We therefore believe it is particularly important to consider the work of Clastres (1981), when the author characterises and compares primitive/undivided societies with state/divided societies. We believe that Neolithic ceramics express the characteristics of an undivided society, ‘homogenous in its essence’, which seeks dispersion and fragmentation, and is at the same time an autonomous whole, a homogenous unit (Clastres 1981: 112-205).

5. FINAL EVALUATION The features we have described allow us to affirm that although there is a shared formal model in these Neolithic ceramics, there are also two well-differentiated styles, which are visible in the different formal relationships present in both phases, and are particularly visible in the decoration used.

The pottery expresses this pattern through its formal features, revealing a degree of formal diversity between sites (and contexts) and a strong sense of formal similarity within each site. It is true that there are slight differences between the E/MN and the LN, as in the LN a certain internal diversity appeared in decorative features, with the division of decorative features found on ceramic pieces possibly being a sign of the complexity and social division of the LN (the Chalcolithic period in other regions). These features are invisible in the E/MN. It will be necessary to contrast this hypothesis with other elements from these societies’ material culture, such as emplacements, dwellings, art and burial sites, and particularly megalithic elements. It will then be possible to see which elements define internal homogeneity and external diversity, and with what degree of complexity they are expressed within the societies in both periods.

Ceramic material culture was of little importance in Neolithic societies, particularly in its early stages, as the features discovered would suggest simplicity, scarcity, fragmentation and invisibility within the known contexts. Ceramics in the LN became increasingly important in social life, as there was a degree of variability, and it is proportionally more abundant (in known sites). It is visible, and uses decoration as its most important formal resource. This importance would seem to be connected with the funerary world, where the decorations used were more complex and had similar characteristics to the decorations found in Megalithic paintings in Galicia. The formal similarity of ceramics throughout the Neolithic period in Galicia has led us to consider that for the time being we should do away with the term ‘Chalcolithic’. Our study of ceramic materials is not the only reason we have for believing this, as there are other characteristics present in the record: types of fortified settlements in other parts of the Iberian Peninsula which are not found in Galicia (with the north of Portugal having the example closest to Galicia); factors shared with other Neolithic settlements from an earlier date, such as the type of emplacement chosen; finally, the absence of any relationship between ceramics from the LN with bell-beaker pottery, (traditionally considered as being from the Upper Chalcolithic), whose introduction meant a 12

Acknowledgements I would like to thank a F. Criado for correcting the text and his invaluable comments, R. Blanco for his revision of the text, and A Rodríguez, our draughtsman and graphic consultant. Author’s address Mª Pilar PRIETO MARTÍNEZ Laboratorio de Arqueología y Formas Culturales Unidad Asociada del Instituto Padre Sarmiento (CSIC) (Instituto de Investigaciones Tecnológicas, Universidad de Santiago de Compostela) R.U. Monte da Condesa. Campus Universitario Sur 15706 Santiago de Compostela. SPAIN E-mail: [email protected]

It is also recorded in domestic contexts from sites included in the bibliography, although we have not been able to carry out a systematic study, for example in the site of Requeán (González Méndez, personal communication), or in sites from the north of Portugal (S.O. Oliveira 1986). The most important factor in this case is that for these contexts the number of ceramic pieces is truly exceptional.

116

P. Prieto Martínez: Ceramic Style in Neolithic Societies in Galicia (NW Iberian Peninsula)...

Bibliography

FOUCAULT, M., 1981, Diálogos sobre el poder. Madrid: Alianza Editorial.

ABAD GALLEGO, X. C., 1995, Un ejemplo de readaptaciones constructivas en un enterramiento tumular: Cotogrande nº 5. Minius 4, p. 13-30.

GONZÁLEZ MÉNDEZ, M., 1991, Yacimientos del III milenio a.C. entre la problemática del Calcolítico y un pasado huidizo. En Criado Boado, F. (dir.), p.147-72.

BELLO DIÉGUEZ, J. M., 1991, Monumento megalítico de Dombate (Cabana, A Coruña). Arqueoloxía/Informes 2. Campaña 1988. Santiago de Compostela, p. 21-7.

JORGE, S. O., 1986, Povoados da Pré-história recente da regiâo de Chaves-Vª. Pª. de Aguiar (3 vols.). Porto: Cámara Municipal de Chaves.

CLASTRES, P., 1981, Investigaciones en antropología política. Barcelona: Gedisa (1st ed. 1980 París).

LEMONNIER, P., 1986, The study of material culture today: toward an anthropology of technical systems. Journal of Anthropological Archaeology 5, p. 147-86.

COBAS FERNÁNDEZ, M. I. and PRIETO MARTÍNEZ, M.P., 1998a, Regularidades espaciales en la cultura material: la cerámica de la Edad del Bronce y la Edad del Hierro en Galicia. Gallaecia, 17: 151-75.

LÓPEZ, P. (coord.), 1988, El Neolítico en España. Madrid: Cátedra. PATIÑO GÓMEZ; R., 1985a, Excavación de la mámoa 1 de Chan da Cruz (Vilaboa). Pontevedra Arqueológica I, p. 17-44.

COBAS FERNÁNDEZ, M. I. and PRIETO MARTÍNEZ, M.P., 1998b, Defining Social and Symbolic Changes from the Bronze Age and Iron Age through the Operational Chain in NW Iberian Pottery. En S. Milliken & M. Vidale (eds.). Craft Specialization: Operational Sequences and Beyond. Papers from the EAA Third Annual Meeting at Ravenna 1997. Volume IV. BAR International Series 720: 95-106. Oxford.

PATIÑO GÓMEZ; R., 1985b, Excavación de la mámoa 1 de As Rozas (Campolameiro). Pontevedra Arqueológica I, p. 45-74. PEÑA SANTOS, A., 1985a, Sondeo estratigráfico en el yacimiento de ‘O Regueiriño’ (Moaña). Pontevedra Arqueológica I, p. 8590. PEÑA SANTOS, A., 1985b, El yacimiento de Lavapés (Cangas del Morrazo): Balance de las excavaciones 1981-1982. Pontevedra Arqueológica I, p. 149-178.

COBAS FERNÁNDEZ, M. I. and PRIETO MARTÍNEZ, M.P., 1999, Introducción a la cerámica Prehistórica y Protohistórica de Galicia. Santiago de Compostela: Trabajos en Arqueología del Paisaje, 17.

PRIETO MARTÍNEZ, M.P., 1998, Forma, Estilo y Contexto en la Cultura Material de la Edad del Bronce Gallega: Cerámica Campaniforme y Cerámica No Decorada. (Tesis doctoral). Laboratorio de Arqueoloxía e Formas Culturais, Departamento de Historia I, Facultad de Xeografía e Historia, USC. Santiago de Compostela. Edited by CD-1999.

COBAS FERNÁNDEZ. I. and PRIETO MARTÍNEZ, M.P., 2002, The technological chain as a methodological and theoretical tool from archaeology. XIVth Congress of the U.I.S.P.P. (Liège, Belgique, 2-8 sept. 2001). B.A.R (in press).

PRIETO MARTÍNEZ, M.P., 1999, Caracterización del estilo cerámico de la Edad del Bronce en Galicia: Cerámica campaniforme y cerámica no decorada. Complutum 10, p. 7190.

CRIADO BOADO, F. 1991. Construcción social del espacio y reconstrucción arqueológica del paisaje. Boletín de Antropología Americana 24, p. 5-30. CRIADO BOADO, F., 1993, Visibilidad e interpretación del registro arqueológico. Trabajos de Prehistoria 50, p. 39-56.

PRIETO MARTÍNEZ, M.P., 2000, La cultura material cerámica en la Prehistoria Reciente de Galicia: Yacimientos al aire libre. Santiago de Compostela: Trabajos de Arqueología del Paisaje 20.

CRIADO BOADO, F., 1995, Límites y posibilidades de la arqueología del paisaje. SPAL 2,p. 9-56. CRIADO BOADO, F. (dir.); BONILLA RODRÍGUEZ, A.; CERQUEIRO LANDÍN, D.; DÍAZ VÁZQUEZ, M.; GONZÁLEZ MÉNDEZ, M.; INFANTE ROURA, F.; MÉNDEZ FERNÁNDEZ, F.; PENEDO ROMERO, R.; RODRÍGUEZ PUENTES, E. and VAQUERO LASTRES, J., 1991, Arqueología del Paisaje. El área del Bocelo-Furelos entre los tiempos paleolíticos y medievales. (Campañas de 1987, 1988 y 1989). A Coruña: Arqueoloxía/ Investigación 6.

PRIETO MARTÍNEZ, M.P., 2002, Ceramic style in Bronze Age societies in Galicia (NW of Iberian Peninsula). Similarities y differences in patterns of formal regularity. XIVth Congress of the U.I.S.P.P. (Liège, Belgique, 2-8 sept. 2001). B.A.R (in press). REY GARCÍA, X. M. 1995. Escavación arqueolóxica no illote de Guidoiro Areoso (Vilanova de Arousa, Pontevedra). Arqueoloxía/ Informes 3. Campaña 1989. Santiago de Compostela: Xunta de Galicia, p. 15-8.

CRIADO BOADO, F. and VÁZQUEZ VARELA, J.M., 1982, La cerámica campaniforme en Galicia. A Coruña: Cuadernos do Seminario de Sargadelos, 42. Edicións O Castro.

RODRÍGUEZ CASAL, A.A., 1989, La necrópolis megalítica de Parxubeira. A Coruña: Monografías Urxentes do Museu, 4. SUÁREZ OTERO, J., 1986, A Idade do Bronce en Galicia: Aspectos ceramolóxicos. O Bronce Inicial. Santiago de Compostela: Tesis de Licenciatura inédita. Universidade de Santiago de Compostela: Facultade de Xeografía e Historia.

EGUILETA FRANCO, J. M., 1987, Catálogo dos materiáis ergolóxicos depositados no Museu de Ourense procedentes de túmulos prehistóricos. Boletín Auriense XVII, p. 9-98. EGUILETA FRANCO, J. M., 1996, Yacimientos calcolíticos al aire libre en torno al embalse de As Conchas (Baixa Limia, Orense, Galicia). Minius V, p. 41-64.

SUÁREZ OTERO, J. and FÁBREGAS VALCARCE, R., 2000, O Neolítico en Galicia. Estado da cuestión. Neolitizaçâo e megalitismo da Península Ibérica. Actas do 3º Congreso de Arqueologia Peninsular (septembro 1999, Vila Real Portugal), vol. III: 135-47. Porto: ADECAP.

FÁBREGAS VALCARCE, R. and FUENTE ANDRÉS, F., 1988, Aproximación a la cultura material del megalitismo gallego: la industria lítica pulimentada y el material cerámico. Arqueohistórica, 2. Santiago de Compostela: Tórculo Artes Gráficas.

VÁZQUEZ VARELA, J. M., 1988, El Neolítico en Galicia. En P. López (coord.), p. 329-336.

117

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

118

M.C. De Angelis & A. Moroni Lanfredini: Neolithic Remains in Umbria (Italy)

NEOLITHIC REMAINS IN UMBRIA (ITALY) Maria Cristina DE ANGELIS & Adriana MORONI LANFREDINI

Abstract: The article summarizes the results obtained so far in the revision program and study of the Neolithic sites undertaken over the last few years by the Archeological Office of Umbria. In this region, archeological records related to the Neolithic period are very mixed in nature, spanning a broad chronological period and concerning settlements, burial traditions and other cultural aspects. In this region various cultural traditions followed one another and partially co-existed which can be attributed to the mid-Tyrrhenian Impressed Ware Culture and to the Linear Pottery Culture, as far as the earlier periods are concerned, and to the Ripoli Culture, permeated with Chassey-Lagozza and Diana influences, for the Neolithic last phases. Placing Umbrian Neolithic in its setting is thus problematic: it involves on the one hand the cultural and chronological relationships which run through the variuos facies to be found in the area, and on the other hand the key role that Umbria, with its central position must have played as a link between the Adriatic coast and the Tyrrhenian. Résumé : Le programme de révision et d’étude des sites néolithiques que la Soprintendenza Archeologica dell’Umbria a entrepris ces dernières années a permis d’obtenir Jusqu’ici les résultats suivants. En Ombrie, la documentation archéologique relative au Néolithique se caractérise par son hétérogénéité; elle concerne, en effet, une vaste période et est relative aussi bien l’habitat qu’à la sépulture et aux cultes. L’Ombrie a vu se succéder, et en partie se superposer, des aspects culturels différents, qu’on peut ramener, pour les phases les plus anciennes, au domaine de la céramique imprimée médio-tyrrhénienne et à la céramique linéaire; les plus récents proviennent de la Culture de Ripoli, imprégnée d’influences Chassey-Lagozza et Diana. L’encadrement du Néolithique ombrien présente donc un foisonnement de problématiques relatives d’une part aux rapports chronologiques et culturels qui ont eu cours entre les diverses faciès présentes sur le territoire, de l’autre au rôle clé que cette région, par sa position centrale, doit avoir joué, comme trait d’union entre le versant adriatique et le versant tyrrhénnien de la péninsule.

In 1997 the Archeological Office of Umbria undertook a program of research and review of currently-known Neolithic sites in the region. The decades-long absence of systematic studies and research in Umbria had left a gap in our knowledge of the Neothilic Age in Central Italy in spite of the fact that archeological documentation of this period is, on the whole, substantial. The need to fill this gap is even more urgently felt inasmuch as Umbria had in every epoch played a fundamental role as an area of contact and influence between the Tyrrhenian and Adriatic Coasts. The sites in question are mainly located in the southern part of the Region, whereas in the center-north part the only-known sites are the open settlements of La Lucciola and San Marco di Gubbio (Fig. 1).

4.8% of the industry. Impressions made with various shells, mainly valves of Glycimeris glycimeris, is the most-used technique. The use of Cardium (Fig. 3, n. 1) is rare. Plastic applications such as lugs, cordons and scales are comparatively frequent. Of great interest is the finding of some models of boats (Fig. 3, n.22, 23), all of which unfortunately are rather fragmentary; these objects, together with those of the pile-dwelling of Marmotta on Lake Bracciano, are the only example of the kind from the Italian Neolithic to date. The lithic industry, which is here made up of flint and obsidian, is at the hypermicrolithic and microlithic level; the microburin technique is evidenced by some samples and blade-making technology is well-developed. The assemblage includes end-scrapers (most of them short); truncated blades; geometrics, such as trapezes, segments of circles and rhomboids; borers; side scrapers, retouched blades, denticulated blades and flakes and sickle blades. Some typological characteristics of the ceramics such as the presence of the decorations “sillons d’impressions”, the low incidence of decorated ware and of Cardium impression on their insides, together with some elements of the lithic industry referable to the Mesolithic tradition, would seem to favor the impression that La Lucciola belongs to a rather old phase of Tyrrhenian Impressed Ware. Without the aid of radiometric dating, this assumption needs to be confirmed with the completion of the study of the rest of the material.

La Lucciola is a settlement of the Early Neolithic located on the southwestern bank of Lake Trasimeno near the town Panicarola. It was excavated between the end of the 60s and mid 70s by the Archeological Office. The Neolithic level, which was located below a necropolis from the Late Bronze Age, is composed of a 1-meter thick layer of sand in which it was impossible to recognize any stratigraphical subdivision. The excavation extended for a total of roughly 150 sq. meters. Ceramic and lithic materials were uncovered, but unfortunately no organic remains, not even fauna. Till now only materials coming from the lower part of the stratum were examined. Pottery (Fig. 2, 3) is generally represented by small sherds; it chiefly includes plain troncoconical and hemispherical cups and bowls, with vertical, often tapered, rims and flat bases. The decorated pottery accounts for only

Another open site that has yielded impressed ware, this time the kind found on the Adriatic side of the peninsula, is the village near Gubbio called S. Marco, in which a linear ditch, 119

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1.

along with other man-made structures, was found when it was excavated in 1987. In the pottery assemblage the bulk of material found quite resembles Adriatic Impressed Ware, but elements of the linear pottery styles are also present. Lithics is very rich in blades and bladelets. For this site a range of radiometric dates from seeds and bones covering a span of over 500 radiocarbon years (from 6430 ± 80 to 5750 ± 70 BP not calibrated) is available. Pottery is directly comparable with that from the Marche settlement of Ripabianca di Monterado (which dates back to the end of V Millenium B.C.), also because of the co-existence of Adriatic Impressed Ware and Linear Pottery elements. As for southern Umbria, of particular importance is the Grotta Bella site, located in the Province of Terni in the Amelia territory. The excavations, carried out between 1970 and 1973, brought to light a 6-meter thick stratigraphical sequence, whose lower part is characterized by the presence of a series of levels belonging to the Neolithic; in these levels two different cultural horizons have been identified: the oldest belonging to the Linear Pottery Culture and the most recent related to an aspect of the painted trichrome pottery1. In the contiguous level between the two horizons two C 14 dates were obtained (6350 ± 90 and 6450 ± 90 B P); they establish an ante quem term for the lower strata (i.e. the strata containing Linear Pottery), so that it turns out to be one of the oldest known so far. As for the Painted Pottery Phase, 1

For the qualities of the materials of Grotta Bella, see publication of De Angelis et al. in these Acts.

we do not have any radiometric dates. Nevertheless, some properties of the material, among which there is mainly the typology of the decorative motifs, the presence of consistent elements belonging to the Early Neolithic Fiorano Po Valley Culture and the absence of flat retouched tools, lead one to suppose there was a chronological stage of painted trichrome pottery even prior to the full manifestations of the Ripoli Culture that were documented in the village of Ripoli itself. The presence of Fiorano elements also brings to the foreground the problem of the role of this culture in the genesis of the aspects of painted trichrome pottery of central Italy. Not far from Grotta Bella (roughly 12 km) there is another important site: Pozzi della Piana, which is a labyrinthine system of karstic galleries. The system opens up onto the right bank of the Tiber River in the Forello gorge. These galleries were frequented during the Neolithic Age, most likely for ritualistic purposes; inside of them, apart from some accumulations of ochre, numerous ceramic containers (belonging to Linear Pottery), lithic and bone industry were found on the floor of the cave. The pottery (Fig. 4, 5), which is very well preserved, includes globular flasks with four handles and troncoconical neck with small handles on the rim; hemispheric, carinated and troncoconical cups of high quality, which were often decorated with small lugs, lobs or knobs; and troncoconical and convex bowls. Two of the cups are specimens of impressed decoration that, together with the finding of San Marco di Gubbio, constitute the only examples of Adriatic Impressed Ware currently known of in Umbria. Other significant findings are two containers, a vase 120

M.C. De Angelis & A. Moroni Lanfredini: Neolithic Remains in Umbria (Italy)

Figure 2. 121

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 3. 122

M.C. De Angelis & A. Moroni Lanfredini: Neolithic Remains in Umbria (Italy)

Figure 4. 123

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 5.

and a cup painted with the “in negativo” (Fig. 4, n. 4, Fig. 5) technique, and a carinated cup decorated with incised lines. The lithic industry, barely represented, is very laminar and contains burins, truncated blades, trapezes, retouched blades, side scrapers and small polished adzes. Even though the lack

of stratigraphical references makes the association of the various findings from Pozzi della Piana not altogether trustworthy, the technological and typological properties of the materials reveal a remarkable homogeneity. Moreover, it is worth pointing out that the pottery from Pozzi della 124

M.C. De Angelis & A. Moroni Lanfredini: Neolithic Remains in Umbria (Italy)

Piana, on the whole, presents merely generic analogies with the pottery coming from the lower horizon of Grotta Bella; on the other hand, we find some typical elements of the upper horizon, such as the raised lobs and handles on the rim. It is therefore possible that slight resemblance to the lower horizon of Grotta Bella – taking into consideration as well the geographical proximity between the two sites – is not due to cultural factors, but to chronological reasons. The Pozzi della Piana could therefore be attributed to an intermediate moment between the two horizons, roughly contemporaneous to the settlement of S. Marco di Gubbio, with which it has in common the carinated cups with convex sides and lug or lob on the carina and the presence of Adriatic Impressed Ware.

After a chronological hiatus lasting a few centuries, the first evidence is the Grotta del Lago, which is still being excavated by the Archeological Office. In the cave, which is of a sepulchral nature, a particular cultural aspect characterized by pottery with spool-shaped handles of the “Diana” style was identified. This aspect, which has only recently been individuated, is so far known (apart from in Umbria) in Abruzzo, Lazio and the Marche, where as well as the spoolshaped handles can also be found elements of the southern Serra d’Alto culture, and dates back to roughly the mid IV Millenium B.C. according to radiometric dating. Three dates are available for the Grotta del Lago: 5732 ± 76, 5413 ± 88 and 5427 ± 99. The first of the dates is however probably too early. In the pottery plain shapes, generally troncoconic bowls and cups; on the bowls, a decoration of lines of impressed points on the inside of the rim that also appears in the advanced phases of the Ripoli Culture, is frequent. The lithic industry is mostly made up of flakes; in the assemblage are present a high percentage of flat - retouched tools.

The Grotta dei Cocci, currently being excavated by the Archeological Office of Umbria, opens up in a steep rocky wall on the River Nera. In the Neolithic levels, located below a Bronze Age horizon, have been found secondary burials and circles of stones that might indicate that the site had a ritual function. The complex is attributable to the Abbruzzo Catignano Culture because of the presence of painted pottery decorated with incisions filled with ochre. The lithic industry, similar to that uncovered in Pozzi della Piana and in the lower horizon of Grotta Bella, but more numerous, has greater evidence of blade technology. The highly specialized tool assemblage is largely made up of geometrical (trapezoidal and semicrescent) truncated and retouched blades. There are also some sickle blades and polished adzes. In Grotta dei Cocci, a particular aspect of the Neolithic – heretofore unknown in Umbria - seems to be represented. This aspect is possibly comparable, from the chronological point of view, to the lower levels of the upper horizon of Grotta Bella. For a more complete picture, it is necessary to wait for the results of the radiometric analyses and study of the pottery.

The open sites of Acciaierie di Terni, of Capanna di Norcia and Capanna del Torbidone near Norcia are attributable to the final phase of the Neolithic. Unfortunately, they are old excavations from the early 1900s and are poorly documented. These sites belong to a sole cultural aspect in which preponderant elements from the later Ripoli culture meld with Chassey-Lagozza and Diana Culture elements. The pottery (Fig. 6) includes vases (also made of purified paste), cups and carinated and troncoconic bowls. Their characteristics are handles with lob decorations. Important, though rare, is the scratched decoration that is a commonly-held taste in Late Neolithic periods of the peninsula by the ChasseyLagozza and Diana Cultures. In the lithic industry, known especially from Capanna di Norcia, can be found end scrapers, flat - retouched blades and a small campignano tool. Blade technology is well developed.

Sporadic findings that are attributable to the Neolithic Linear Pottery have also come from the Tane del Diavolo, near Orvieto, and from the Bettona zone in central Umbria.

As has been seen, the picture that emerges from the examination of the documentation currently available is still missing many pieces. There are gaps from the chronological point of view, because of the paucity of radiocarbon dates, as well as from the cultural point of view, given the total absence of paleoeconomic and paleoenvironmental studies. The hope is that other more complete and in-depth research and studies follow this first revision, as well as a further use of the data available, for example through the study of raw material provenance (e.g. obsidian) and pottery production methods.

If we exclude the settlement of La Lucciola, which represents a cultural aspect that for now is isolated in Umbria, clearly projected toward the world of Middle Tyhrrenian Impressed Ware, the rest of the settlements so far described make up a rather homogenous nucleus characterized by the presence, albeit in differing percentages, of elements of Linear Pottery, accompanied by a bichrome painted ware during its more advanced stages, and subsequently by trichrome painted ware. According to the unfortunately rare radiometric references, these cultural aspects may have developed starting in the first half of the V Millenium B.C., as would be attested to by the dating of the lower horizon of Grotta Bella; for the upper chronological limit, to be located in the first centuries of the IV Millenium B.C., on the other hand, the only currently available term is the dating 5630 ± 80 BP relative to the initial phase of the Abruzzo village Ripoli, which, as we have said, seems less old than the upper horizon of Grotta Bella.

Authors’ addresses Maria Cristina DE ANGELIS Soprintendenza Archeologica per l’Umbria Piazza Giordano Bruno 10 06100 Perugia, ITALIA Adriana MORONI LANFREDINI Dipartimento di Archeologia e Storia delle Arti dell’Università di Siena Sezione di Preistoria Via delle Cerchia 5, 53100 Siena, ITALIA

In subsequent periods, the archeological sources exclusively concentrated in southwestern Umbria become less frequent. 125

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 6. 126

M.C. De Angelis & A. Moroni Lanfredini: Neolithic Remains in Umbria (Italy)

Bibliography CALZONI, U., 1939, Un fondo di capanna scoperto presso Norcia. Bullettino di Paletnologia Italiana III, p. 37 – 50.

GUERRESCHI, G., CATALANI, P., LONGO, G. & IANNONE A., 1981 – 1992, Grotta Bella (Terni). Una sequenza stratigrafica dal Neolitico inferiore all’età imperiale. I livelli preistorici. Bullettino di Paletnologia Italiana I, p. 143 – 228.

CORRIDI, C. & MORONI, A., 1993, I materiali della Capanna di Norcia conservati al Museo Archeologico di Perugia: industria litica, ossea e resti faunistici. Bullettino di Paletnologia Italiana 84, p. 381 – 434.

GUERZONI, R.P., 1984 – 1985, Materiali ceramici della Capanna di Norcia conservati nel Museo Archeologico di Perugia (scavo U. Calzoni). Annali della Facoltà di Lettere e Filosofia dell’Università degli Studi di Perugia XXII, p. 141 – 180.

DE ANGELIS, M. C. & TALIANA D., 1998, La Grotta del Lago di Triponzo nel comune di Cerreto di Spoleto (Valnerina, Umbria), nota preliminare. Rassegna di Archeologia 15, p. 55-76.

LONGO MARZIANI, G., IANNONE, A. & GUERRESCHI, G., 1989, Reconstitution du climat d’un site prèhistorique en Italie centrale d’après l’analyse des fragments de charbons. L’Anthropologie 93, n. 1, p. 279 – 263.

GUERRESCHI, G., CATALANI P., LONGO G. & IANNONE, A., 1987, Grotta Bella (Terni). Una sequenza culturale dal Neolitico inferiore all’età imperiale. I livelli preistorici. Atti della XXVI Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria II, p. 631 – 644.

MALONE, C. & STODDART, S., 1992, The Neolithic Site of San Marco, Gubbio (Perugia), Umbria: Survey Excavation 1985 – 1987. Papers of the British School at Rome LX.

127

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

128

H. Collet: Les mines néolithiques de Spiennes : État des connaissances et perspectives de recherche

LES MINES NÉOLITHIQUES DE SPIENNES : ÉTAT DES CONNAISSANCES ET PERSPECTIVES DE RECHERCHE Hélène COLLET

Résumé : Le site néolithique de Spiennes (Hainaut, Belgique) est l’un des plus vastes centre d’extraction et de taille du silex néolithique d’Europe. Entre 4400/4200 et 3000/2450 avant notre ère, le silex extrait dans des exploitations à ciel ouvert ainsi que dans des structures d’extraction souterraines pouvant atteindre 16 m de profondeur fournit la matière première nécessaire à la production de haches et de lames. La présente contribution rassemble les données disponibles sur le site(étendue, structures, chronologie,...) tout en donnant un aperçu des recherches en cours . Summary: The site of Spiennes (Hainaut, Belgium) is one of the broadest flint excavation and knapping centre of the European Neolithic. From 4400/4200 to 3000/2450 before Christ, flint has been extracted from open quarries as well as from shafts that could reach up to 16 meters depth in order to be used for an axes and blades production. This text gathers up the available data on this topic and gives an overview of the ongoing investigations.

EXTENSION DU SITE

bande de terrain, large d’environ 200 m, parallèle à la Trouille, soit environ 14 hectares. Au nord de la ligne de chemin de fer, son extension est connue tant par les déchets de taille en surface que par les nombreuses structures d’extraction qui y ont été repérées. Au sud, par contre, seule la répartition des déchets de taille permet de circonscrire son étendue potenielle. Une troisième zone minière, pratiquement inexplorée, pourrait couvrir, d’après différents sondages, toute la crête dominant la Wampe (Hubert, 1976a).

Les mines néolithiques de Spiennes, situées à 6 km au sud est de Mons (Hainaut), s’étendent sur environ 100 hectares sur deux plateaux de part et d’autre de la Trouille, affluent de la Haine. Celui de Petit-Spiennes, localisé sur la rive gauche, culmine à 77 m d’altitude. Au sud, à l’est et à l’ouest, il s’interrompt brusquement et forme une sorte de promontoire qui domine d’environ 35 m les vallées de la Wampe et de la Trouille. Au nord par contre, la déclivité du terrain est plus insensible et le plateau descend en pente douce vers l’actuel village de Spiennes. Le plateau de la rive opposée, dit du Camp-à-Cayaux (le Champ-aux-Cailloux), présente la même topographie ; ses flancs sont escarpés à l’ouest, à l’est et au sud tandis que le plateau, qui atteint une altitude de 92 m, s’abaisse de manière plus progressive vers le nord.

La densité des structures d’extraction du silex est telle dans les zones déjà explorées —un puits tous les 4 ou 6 m dans de nombreux secteurs— que le nombre total de puits néolithiques pourrait avoisiner 20 à 30 000 unités si les 100 hectares précités ont été uniformément exploités. ÉTAT DE LA DOCUMENTATION

L’étendue des différentes zones minières est connue actuellement uniquement par la répartition des déchets de taille en surface et les structures d’extraction repérées en un peu plus d’un siècle d’investigations (figure 1). La zone minière du Camp-à-Cayaux s’étend, de manière certaine, depuis les flancs sud-ouest et ouest jusqu’au-delà du chemin reliant Spiennes à Harmignies. Ces terrains abondamment sondés, notamment durant la première moitié du 20ème siècle, ont tous livré des structures d’extraction (de Loë & Rahir, 1929 ; Verheyleweghen, 1963). Au nord, des puits et des fosses furent identifiés récemment à hauteur des dernières maisons de la rue d’Harmignies (Van Assche et al., 1993). Les limites nord-est et sud-est sont, elles, plus imprécises. D’après les déchets de taille en surface et quelques anomalies géologiques, la zone minière pourrait s’étendre jusqu’à proximité de la chaussée de Beaumont et la courbe de niveau des 75 m (Centre archéologique de Recherche minière, 1990). Elle couvrirait donc au total environ 65 hectares. La zone minière de Petit-Spiennes, bien plus modeste, occupe une

Bien que le site soit connu depuis près d’un siècle et demi, seule une douzaine de structures d’extraction ont été intégralement fouillées à ce jour. Parmi celles-ci, plus de la moitié l’ont été durant ces dernières années au Camp-àCayaux comme à Petit-Spiennes et sont encore en cours d’étude (Bonenfant, 1993 ; Collet & Van Neer, à paraître ; Société de recherche préhistorique en Hainaut, 1997). L’exploitation des données issues de ces recherches devrait enrichir considérablement, dans le futur, notre connaissance des modes d’extraction, des remblaiements, des outils miniers et de la production lithique. En fait, la plupart des investigations menées à Spiennes se sont limitées à la fouille, dans la zone minière du Camp-àCayaux, de parties supérieures de puits, qui furent pour certaines considérées, au début du 20ème siècle, comme des « fonds de cabane ». Elles ont livré un mobilier considérable conservé principalement aux Musées royaux d’Art et 129

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. Etendue des zones minières.

d’Histoire (collections de Loë & Rahir et Verheyleweghen) et à l’Institut royal des Sciences naturelles de Belgique (fonds Stevens). Ces collections constituent la source principale pour l’étude du mobilier mis au jour à Spiennes. L’homogénéité de ces parties supérieures de puits et leur contemporanéité avec l’exploitation minière fait l’objet de discussions. Certains auteurs supposent que le comblement de ces têtes de puits peut être très largement diachronique (de Heinzelin et al., 1993) tandis que d’autres marquent une relative confiance quant à leur homogénéité et considèrent que bien que ces comblements terminaux ne soient pas strictement contemporains du remblaiement des structures, ils ne sont pas non plus franchement postérieurs (Gautier et al., 1993). Les arguments avancés de part et d’autre sont à prendre en considération. D’une part, les brusques tassements de puits invoqués par J. de Heinzelin -parfois sur deux mètres de profondeur- peuvent encore aujourd’hui être observés à Spiennes en de rares occasions, sans que l’on sache toujours si on a affaire à une structure néolithique ou à un de ces puits creusés dans la seconde moitié du 19ème siècle pour les besoins en silex des faïenceries. De plus, de la céramique des Ages des métaux a été quelques fois mise au jour dans ces parties supérieures de puits (Colman, 1957). Néanmoins, la faune qui en est issue paraît homogène et s’accorde bien avec l’époque d’exploitation en raison de la grande taille des bovidés domestiques et de la présence d’ossements d’animaux sauvages tels que l’élan ou le chat sauvage qui sont très rares durant les périodes postérieures (Gautier et

al., 1993 : 90). Il semble donc que chaque comblement doit être évalué au cas par cas même si globalement ces ensembles sont probablement fiables. Il est enfin à noter que ces « fonds » ne reflètent probablement pas fidèlement la réalité du terrain puisque l’on sait que, dans certains cas au moins, seuls les ensembles les plus riches ont été fouillés tandis que des structures jugées trop pauvres ont été systématiquement délaissées après sondage (de Loë et Rahir, 1929 : 55). Une découverte majeure effectuée à Petit-Spiennes est celle d’un camp fortifié Michelsberg, contemporain d’une partie au moins de l’exploitation minière dont le tracé a pu être reconnu grâce à des fouilles systématiques menées durant les années 60-70. Une part importante de la surface interne demeure, hélas, inexplorée. CHRONOLOGIE Les dates radiocarbones, encore fort peu nombreuses (fig. 2), montrent que le site a été exploité entre 5510 et 4230 BP soit entre 4470/4240 et 3350/2450 avant notre ère sans que l’on puisse dire si cette exploitation est continue ou non. L’activité minière s’inscrit principalement dans les limites chronologiques retenues pour l’occupation Michelsberg ou d’affinité Michelsberg dans nos régions mais en déborde 130

H. Collet: Les mines néolithiques de Spiennes : État des connaissances et perspectives de recherche 5510 ± 55 BP

4470-4240 av. J.-C.

4460-4040 av. J.-C.

Camp-à-Cayaux

KN-I.16

5110 ± 40 BP

4000-3780 av. J.-C.

Spiennes

Lv-1598

5100 ± 65 BP

4040-3710 av. J.-C.

Petit-Spiennes

Lv-1566 GrN-4674

5420 ± 75 BP

Petit-Spiennes

OxA-3196

4830 ± 80 BP

3780-3370 av. J.-C.

Spiennes (Petit-Spiennes ?)

Bêta-110683

4500 ± 50 BP

3360-3030 av. J.-C.

Petit-Spiennes

Lv-1599

4490 ± 100 BP

3360-2938 av. J.-C.

Petit-Spiennes

BM-289

4230 ± 130 BP

3350-2450 av. J.-C.

Spiennes

Figure 2. Datations absolues.

partiellement. La céramique découverte jusqu’à présent paraît confirmer le fait que le site a été principalement exploité durant le Néolithique moyen II (culture de Michelsberg) et sans doute plus faiblement au Néolithique final (Seine-OiseMarne ?). Les fonds conservés à l’IRSNB montrent que ceux dont le contenu paraît homogène sont aux trois quarts Michelsberg (Gosselin in Gautier et al., 1993 : 82). Bien qu’une distinction soit souvent établie entre les zones minières du Camp-à-Cayaux et de Petit-Spiennes, les datations comme la céramique montrent que toutes deux furent fréquentées si pas exploitées- aux mêmes époques. Plusieurs découvertes montrent également que le site connaît des occupations à diverses reprises durant les Ages des Métaux sans que l’on puisse clairement en établir la nature (Colman, 1957 ; Hubert, 1976c).

la craie. Des puits proches peuvent exploiter des veines distinctes comme le montre, par exemple, la coupe réalisée à Petit-Spiennes en 1867 à l’occasion de la construction d’une ligne de chemin de fer (Briart et al., 1872). Plus récemment, des recherches effectuées au Camp-à-Cayaux ont mis au jour un ensemble de puits profonds de 3 à 3,5 m (Bonenfant, 1993) sur la même parcelle que celle où furent fouillés les fameux puits de 15 et 16 m de profondeur (Hubert, 1978). Dans les premiers, des rognons digités ont été récoltés dans la partie supérieure de la craie tandis que dans les seconds, les puits d’accès traversent une quinzaine de bancs de silex avant d’atteindre les dalles de silex qui seront extraites. Comme ces deux types de structures coexistent sans se recouper et que de la céramique Michelsberg a été découverte dans les deux, il n’est pas impossible que ces différentes exploitations soient contemporaines.

EXPLOITATION MINIÈRE ET COMPLEXITÉ DE L’OCCUPATION

A Petit-Spiennes, différentes fouilles indiquent que des occupations successives se superposent et, dans certains cas, il semble même que l’existence de mines antérieures ait été ignorée. A proximité de la voie de chemin de fer, par exemple, des recoupements de puits et de fosses d’extraction correspondant à trois périodes distinctes (Hubert, 1976b) tandis qu’ailleurs des puits aboutissent dans des salles déjà exploitées (Gosselin, 1986).

Comme dans la plupart des sites miniers, les méthodes d’extraction sont fortement conditionnées par l’accessibilité à la matière première. A Spiennes, sur les flancs de la vallée, là où le silex affleure, de courtes galeries, voire de petites carrières, ont suffi. De telles exploitations furent découvertes et fouillées au Camp-à-Cayaux mais sont relativement mal documentées (De Pauw & Van Overloop, 1889-1890 ; Verheyleweghen, 1953 ; Lefort, 1954). Sur le rebord des plateaux, à Petit-Spiennes comme au Camp-à-Cayaux, seule une mince couche de limon recouvre la craie si bien que le silex a pu être exploité grâce à de simples fosses de 3 à 4 m de profondeur (Briart et al., 1872 ; Hubert, 1976b). Plus avant sur les plateaux, les terrains à traverser pouvant atteindre 6 m d’épaisseur, la technique d’exploitation la plus économe a consisté à recourir à des exploitations souterraines composées d’un puits d’accès vertical cylindrique d’environ un mètre de diamètre à la base duquel se développent plusieurs galeries d’extraction -hautes ou basses suivant le nombre de bancs exploités- sur des longueurs de 3 ou 6 m, rarement plus.

La manière dont les puits ont été comblés est variable. Jusqu’à présent, l’accent a été mis sur le fait que les puits étaient systématiquement rebouchés une fois l’exploitation achevée comme semble l’indiquer la bonne préservation de la forme initiale des structures et le remontage de pièces trouvées à différentes profondeurs (Gosselin, 1986). Plusieurs puits en cours de fouille à Petit-Spiennes montrent cependant que dans certains cas le comblement s’étale sur un long laps de temps sans que l’on soit actuellement en mesure de préciser s’il s’agit d’arrêts saisonniers ou de périodes plus longues pouvant correspondre aux différentes époques d’exploitation notées dans ce secteur. La localisation de ces structures en bordure de la zone minière pourrait également avoir joué un rôle déterminant. Pour une des structures, les étapes de comblement sont séparées par deux phases d’arrêt qui correspondent au minimum à plusieurs mois (Collet & Van Neer, à paraître). Une seconde structure montre un puits dont le comblement en grande partie naturel pourrait s’étaler sur un temps encore bien plus long. Sur la même parcelle, de

La craie de Spiennes, qui suite à une révision récente remonte au Campanien supérieur (Robaszynski & Christensen, 1989), a pour particularité de renfermer de nombreux bancs de silex si bien que la profondeur des structures d’extraction n’est pas uniquement tributaire de la profondeur à laquelle apparaît 131

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

nombreuses structures non fouillées se caractérisant par leur très grandes dimensions pourraient également correspondre à des structures dont le colmatage fut lent. Elles voisinent avec d’autres puits dont le comblement a dû être rapide comme semble l’indiquer la très bonne préservation de la forme puisqu’ils n’excèdent pas 1 m de diamètre en surface.

l’activité minière ne constitue qu’une des activités d’une communauté paysanne qui pratique par ailleurs l’agriculture et l’élevage (Gautier et al., 1993). Les âges d’abattage montrent, par exemple, que les ruminants ne sont pas uniquement consommés mais fournissent également des produits secondaires tel que le lait et peu-être une force de traction (Gautier et al., 1993). La découverte récente de quelques éléments de foetus d’animaux domestiques dans le comblement d’un puits à Petit-Spiennes pourrait constituer un indice supplémentaire de la pratique de l’élevage (Collet & Van Neer, à paraître) tout comme les indices palynologiques évoqués plus haut.

HABITAT ET ÉCONOMIE Le mobilier archéologique découvert au sein des zones minières pourrait laisser supposer qu’un ou plusieurs habitats étaient localisés dans un périmètre relativement proche des mines, en tout cas à certaines époques, comme l’ont proposé différents auteurs (Clason, 1971 ; Gautier et al., 1993) sans pour autant se confondre avec celles-ci. Les têtes de puits fouillées anciennement au Camp-à-Cayaux ont livré une quantité parfois impressionnante de faune pour un site minier même si des réserves doivent être émises quant à la représentativité de l’échantillonnage (cf. supra). La collection conservée à l’Institut royal des Sciences naturelles de Belgique comprend, par exemple, plus de 13 000 ossements issus de la fouille de 73 têtes de puits soit en moyenne 178 ossements par fonds. Parmi ceux-ci, environ 97 % sont des restes d’animaux domestiques. La fouille récente d’un puits à Spiennes a, elle, livré environ 80 ossements d’animaux domestiques répartis sur toute la hauteur de la structure. Etant donné le comblement lent de ce puits, les animaux intrusifs y sont cependant bien plus nombreux (Collet & Van Neer, à paraître). Du mobilier à caractère plutôt domestique (poinçons et lissoirs en os, peignes en bois de cerf, meules...), inhabituel pour un site minier, a été mis au jour au début du siècle comme dans les fouilles récentes (Collet, 2000). La présence d’assez nombreux ossements humains (cf. infra) devrait peut-être être envisagée dans cette perspective. Une catégorie d’objet particulière est constituée par d’assez nombreux fragments de polissoirs mis au jour dans le comblement de puits ou dans des ateliers de taille alors que ceux-ci sont eux aussi rares ou absents sur la plupart des sites miniers (Collet, 2000). Dans ce cas il est difficile de dire si ces fragments doivent être considérés comme des réserves de matières premières acheminées depuis l’habitat ou s’il s’agit de polissoirs utilisés dans le cadre de la production puis rebutés.

OSSEMENTS HUMAINS Bien que quelques sépultures aient été découvertes sur des sites miniers en Europe, les mines de Spiennes se singularisent par la quantité d’ossements humains qui y auraient été mis au jour puisque, d’après la littérature, ce sont près de 50 découvertes successives qui ont été signalées. Il s’agit d’environ 10 squelettes plus ou moins complets qui auraient été découverts dans des galeries ou dans des puits, principalement dans la seconde moitié du 19ème siècle, et d’environ 40 ossements humains isolés découverts dans des têtes de puits, au début du 20ème siècle (Colman, 1957 ; de Heinzelin et al., 1993). Si la découverte de squelettes complets n’a pas suscité de théorie particulière, il n’en fut pas de même pour les restes isolés, il est vrai malaisé à expliquer, qui tantôt furent considérés comme des sépultures du second degré (de Loë & Rahir, 1929) tantôt comme des indices possibles de cannibalisme (Clason, 1971). Bien que ces ossements aient été souvent découverts en association avec de la faune, il faut noter qu’aucune marque caractéristique de découpe ou de décarnisation n’a pu être identifiée sur différents ossements récemment étudiés (Gautier et al., 1993 : 90). Le caractère rituel et intentionnel des dépôts n’est pas non plus établi, loin s’en faut. Curieusement, il semble que les fouilleurs aient uniquement pris en considération, tandis que les restes postcrâniens découverts étaient négligés (de Heinzelin et al., 1993). Le mobilier présenté comme funéraire est constitué d’ébauches, de quelques tessons, d’un peu de faune ou encore d’outils miniers brisés tel que n’importe quelle tête de puits peut en livrer à Spiennes (Colman, 1957 : 251).

Le camp fortifié découvert à Petit-Spiennes, certainement contemporain d’une partie de l’exploitation minière, peut être l’un de ces habitats. Les pollens recueillis à la base des fossés indiquent, en tout cas, la présence de champs et de pâtures à proximité de celui-ci (Heim, 1971).

L’appartenance au Néolithique de ces ossements humains fut récemment remise en cause dans le cadre d’une étude qui, par ailleurs, démontrait clairement que les fameux mineurs néolithiques d’Obourg et de Strépy étaient des faux (de Heinzelin et al., 1993). Des fouilles menées à PetitSpiennes en 1997 ont, cependant, permis la découverte d’un squelette humain adulte ainsi que d’ossements appartenant à un péri-natal dans le comblement d’un puits d’extraction entre 4 et 5 m de profondeur (Toussaint et al, 1997 ; Collet & Van Neer, à paraître). La datation obtenue à cette occasion -4500 ± 50 BP (Bêta 110 683)- montre que l’attribution de ces anciennes découvertes au Néolithique n’est pas complètement improbable et doit être réexaminée au cas par cas notamment par des datations d’autant plus qu’un squelette humain provenant de Spiennes, mais dont le contexte de découverte

L’intensité de l’activité minière comme de la production est, en l’état de nos connaissances, impossible à déterminer puisque l’on ne connaît pas avec certitude l’étendue du site, la durée exacte de son exploitation et la continuité ou non de cette dernière. Pourtant l’ampleur du site de Spiennes, les techniques utilisées ou encore la qualité de la production ont fait supposer à différents auteurs que le travail minier était une activité hautement spécialisée (cf. par exemple Verheyleweghen, 1966). Bien qu’il ne fasse aucun doute que les lames et les haches produites à Spiennes aient circulé et aient été échangées, il est pour d’autres plus probable que 132

H. Collet: Les mines néolithiques de Spiennes : État des connaissances et perspectives de recherche

est inconnu, fut lui aussi daté du Néolithique (de Heinzelin et al., 1993).

HEIM, J., 1971, Datation par la palynologie d’un fond de fossé de la civilisation de Michelsberg à Petit-Spiennes. In F. Hubert, Fossés néolithiques à Spiennes. Premier rapport avec annexe de J. Heim. Archaeologia Belgica 111, Bruxelles : Service national des Fouilles, p. 62-66.

Adresse de l’auteur

HUBERT, F., 1971, Fossés néolithiques à Spiennes. Premier rapport. Avec annexe de J. Heim Archaeologia Belgica 136. Bruxelles : Service national des Fouilles.

Hélène COLLET Société de Recherche préhistorique en Hainaut 16 chaussée du Roeulx B-7000 Mons BELGIQUE

HUBERT, F., 1976a, Une nouvelle zone minière à Spiennes. Archaeologia Belgica Conspectus MCMLXXV 186, p. 12-15. HUBERT, F., 1976b, Puits de mines à la tranchée du chemin de fer à Spiennes. Archaeologia Belgica Conspectus MCMLXXV 186, p. 9-11.

Bibliographie

HUBERT, F., 1976c, Atelier de taille du silex de l’Âge du Bronze à Spiennes. Archaeologia Belgica Conspectus MCMLXXIX 186, p.16-20.

BONENFANT, P.-P., 1993, Spiennes (Mons, Hainaut) : mines néolithiques de silex. Annales d’Histoire de l’Art et d’Archéologie de l’Université Libre de Bruxelles 15, p. 123-125.

HUBERT, F., 1978, Une minière néolithique à silex au Camp-àCayaux de Spiennes. Archaeologia Belgica 210, Bruxelles : Service national des Fouilles.

BRIART, A., CORNET, F. & HOUZEAU DE LEHAIE, A., 1872, Sur l’âge de la pierre polie et les exploitations préhistoriques de silex dans la province de Hainaut. In Compte-rendu du Congrès international d’Anthropologie et d’Archéologie préhistoriques. 6ème session. Bruxelles, p. 279-294.

LEFORT, M., 1954, Les cahiers de Spiennes 3 (fascicule polycopié). ROBASZYNSKI, F. & CHRISTENSEN, K., 1989, The Upper Campanian-Lower Maastrichtian chalks of the Mons basin, Belgium : a preliminary study of belemnites and foraminifera in the Harmignies and Ciply areas. Geologie en Mijnbouw 68, p. 391-408.

CENTRE ARCHÉOLOGIQUE DE RECHERCHE MINIÈRE, 1990, Etude préalable à la sauvegarde et à la mise en valeur du site archéologique de Spiennes - Mons, Spiennes.

SOCIÉTÉ DE RECHERCHES PRÉHISTORIQUES EN HAINAUT, 1997, Minières néolithiques à Spiennes (PetitSpiennes). Mons : S.R.P.H.

CLASON, A. T., 1971.,The flint-mine workers of Spiennes and Rijckholt – St. Geertruid and their animals. Helinium 11 : 3-33. COLLET, H., 2000, Fouille de nouveaux puits d’extraction à PetitSpiennes et découverte de fragments de polissoir. Notae Praehistoricae 20, p.163-170.

TOUSSAINT, M., COLLET, H. & VANDER LINDEN, M., 1997, Découverte d’un squelette humain dans le puits de mine néolithique ST11 de Petit-Spiennes (Hainaut). Notae Praehistoricae 17, p. 213-219.

COLLET, H. & VAN NEER, W., (à paraître), Stratigraphie et faune d’un puits d’extraction de silex néolithique à Petit-Spiennes. Anthropologie et Préhistoire 113.

VAN ASSCHE, M., FOURNY, M., CLARYS, B., FECHNER, K., LAURENT, C., MATHIEU, S. & HENTON, A., 1993, Néolithique ancien rubané, Néolithique moyen et autres traces préhistoriques dans la tranchée du gazoduc Flobecq-Quévy (Hainaut) : prospection et découvertes “hors des sentiers battus”. Notae Praehistoricae 12, p. 131-136.

COLMAN, P., 1957, Le Néolithique et ses prolongements à Spiennes. Bulletin de la Société royale belge d’Études géologiques et archéologiques «Les Chercheurs de la Wallonie» 16, p. 226290. DE HEINZELIN, J., ORBAN, R., ROELS, D. & HURT, V., 1993, Ossements humains dits néolithiques de la région de Mons (Belgique), une évaluation. Bulletin de l’Institut royal des Sciences naturelles de Belgique, Sciences de la Terre 63, p. 311-336.

VERHEYLEWEGHEN, J., 1953, Découverte sur le territoire de Spiennes d’une phase d’occupation des hommes de Michelsberg, antérieure à celle du plateau du “Camp-à-Cayaux” (Compte-rendu des fouilles entreprises en 1953). Bulletin de la Société royale belge d’Anthropologie et de Préhistoire 64, p. 141-162.

DE LOË, A. & RAHIR, E., 1929, Notice sur les fouilles exécutées à Spiennes en 1925 et en 1928. Bulletin de la Société d’Anthropologie de Bruxelles 54, p. 52 - 69.

VERHEYLEWEGHEN, J., 1963, Évolution chronologique du Néolithique au «Camp-à-Cayaux» de Spiennes (Hainaut) avec une étude statistique dimensionnelle et caractérielle du matériel lithique. Helinium 3 (1), p. 3-38.

DE PAUW, L. & VAN OVERLOOP, E., 1889-1890, Les ateliers préhistoriques de Spiennes. Bulletin de la Société d’Anthropologie de Bruxelles 8, p. 28-39.

VERHEYLEWEGHEN, J., 1966, Le Néolithique minier belge, son origine et ses relations culturelles. Second Colloque. In Palaeohistoria, 12 . Acta et Communicationes Instituti BioArchaeologici universitatis Groningae. Proceedings of the Second Atlantic Colloquium (Groningen 6-11 April 1964), p. 529-57.

GAUTIER, A. & BIONDI, S., 1993, Forgotten Faunal Collection from the Prehistoric Flint Mines at Spiennes (Province of Hainaut, Belgium). Helinium 1 (28), p. 80-98. GOSSELIN, F., 1986, Un site d’exploitation du silex à Spiennes (Hainaut), au lieu-dit «Petit-Spiennes». Vie archéologique 22, p. 33-160.

133

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

134

J. Ramos et al.: Le gisement Néolithique de “El Retamar”. Un exemple des communautés du VIe millénaire...

LE GISEMENT NÉOLITHIQUE DE “EL RETAMAR”. UN EXEMPLE DES COMMUNAUTÉS DU VIe MILLÉNAIRE AVANT NOTRE ÈRE DANS LA BAIE DE CADIX (ESPAGNE) José RAMOS, María LAZARICH, Nuria HERRERO, Vicente CASTAÑEDA, Manuela PÉREZ, Salvador DOMÍNGUEZ-BELLA, Mª Eugenia GARCÍA & Isabel CÁCERES

Résumé : Il s’agit de l’étude des structures archéologiques du gisement « El Retamar » : au total 98 structures (62 foyers, 10 escargotières, 2 sépultures et 24 concentrations de pierres), liées aux activités sociales et de production qui ont été développées par les communautés qui ont habité le gisement au VIe millénaire a.n.e. Nous présentons l’étude microspatiale, avec l’analyse des produits, des structures et des aires d’activités, et définissons le mode de production et le mode de vie de ces communautés qui ont une structure économique basée sur la pêche et le ramassage des coquillages, avec également la présence d’animaux domestiques. Nous apportons une nouvelle perspective pour l’étude du Néolithique dans le sud-ouest de la Péninsule Ibérique, qui est basée sur une industrie lithique géométrique, avec une céramique cardiale. Nous posons la question de l’accès à l’économie de production dans les communautés autochtones des chasseurs-cueilleurs. Abstract: Archaeological structures from El Retamar site (Puerto Real, Cádiz, Spain) have been studied. In this archaeological site 98 structures appear (62 fireplaces, 10 shell accumulations, 2 burials and 24 stone concentrations). These structures are related with social and production activities, which were developed by the communities that occupied this settlement at VI millennium B.C. We present the microspatial study, with an analysis of the archaeological products, structures and activity areas. We determine the production mode and life mode of these communities which have a economic structure based in the fishing and sea-shell collecting, with the presence of domestic animals as well. We present a new perspective for the study of the Neolithic in the South-West of the Iberian Peninsula, based in a geometric lithic industry and a cardial ceramic. The production economy access after the autochthonous hunter-gatherer communities is a proposed question in this work.

INTRODUCTION ET MÉTHODOLOGIE

explications diffusionnistes en direction est-ouest pour le transit vers l’économie de production (Bernabeu, Aura, et Badal, 1995). Comme nous le voyons « El Retamar » remet en question ce modèle.

Les théories substantives conforment des projets conceptuels d´explications du monde. Les modèles historiques traditionnels, fonctionnels, structuraux, postmodernes ou matérialistes se fondent sur des idéologies très précises et définies. Les projections vers les sociétés préhistoriques prétendent expliquer la dynamique du changement. Nous prétendons nous éloigner du relativisme et de l’historicisme, et nous assumons une position théorique (Gándara, 1993), très définie dans l’Archéologie Sociale. L’étude des sociétés préhistoriques implique l’étude de leurs modes de production et de reproduction sociale, leurs modes de vie et de travail (Vargas, 1987; Bate, 1998). Nous prétendons expliquer la technologie, dans sa relation au travail et définir les relations de la base économique avec l’idéologie. Nous privilégions la priorité de l´étude des formations sociales sur les manifestations culturelles (Arteaga, 1992: 181), aspirant à une conception critique de l’Archéologie au service de l’Histoire et valorons les aspects sociaux et économiques, dans les analyses du processus historique (Ramos, 1999, 2000).

Face à ces modèles on peut documenter une particularité du milieu naturel au caractère Atlantique-Méditerranéen (Arteaga et Hoffmmann, 1999), qu´on peut assigner au SudOuest de la Péninsule Ibérique (Calado 2000), à la Vallée du Guadalquivir (Arteaga et Cruz-Auñón, 1999), côte Sud Méditerranée et Nord de l’Afrique (Mikdad et Eiwanger, 2000), avec des caractéristiques climatiques très définies depuis le Pléistocène (Ruiz Bustos, 2000). Tout ceci supose une potentialité du milieu en ressources naturelles. Les espèces végétales autochtones (olivier sauvage, chêne, légumineux…) et les animaux qui ont été documentés depuis les registres paléolithiques du Pléistocène (bœufs, chèvres, chevaux…), permettent de poser des modèles locaux de transitions vers l’économie de production, à partir des groupes régionaux de chasseurs et cueilleurs. Il est donc nécessaire d’intégrer les bases économiques et idéologiques des processus économiques depuis le peuplement autochtone. Nous avons observé que la technologie des derniers chasseurs-cueilleurs a une présence substantielle dans la transition vers les premières communautés de bergers et de paysans (Ramos et alii, 2000; Castañeda, 2000).

Les modèles qui ont prédominé dans les explications du Néolithique dans l´historiographie au sud de la Péninsule Ibérique ont été basés sur des critères de diffusion et d’acculturation. Une ordination tripartite du Néolithique a dominé dans la recherche. Parallèlement aux explications diffusionnistes on a surtout travaillé avec les styles de la poterie (Pellicer et Acosta, 1986; Acosta, 1987, 1995). Le modèle de « front d´avance » (Ammerman y Cavalli Sforza, 1984; Ammerman, 2000) a renforcé de nouveau les

LE MILIEU NATUREL “ El Retamar ” est situé dans le Sud de la Péninsule Ibérique, dans le département de Cadix, commune de Puerto Real. Le 135

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

El Retamar

N

Figure 1. Carte de localisation du gisement de “El Retamar”.

X= 7,00 Y=10,00

X= 4,40 Y= 10,00

Foyer 13

A A’

Foyer 14 Foyer 15 A



A

Foyer 9

Foyer 12

A

A’

Foyer 11 X= 7,00 Y= 7,75

X= 4,40 Y= 7,75

0

1 m.

Plante et profil du foyer 9 dessous l’escargotière 4

0

C. de pierres 15. Plante et profil.

10 m.

A’

A

Plante et profil de l’escargotière 3. Figure 2. Plantes et profils des diverses structures. 136

J. Ramos et al.: Le gisement Néolithique de “El Retamar”. Un exemple des communautés du VIe millénaire...

gisement se trouve près de la côte à l’extrémité Nord-Est de la baie de Cadix, sur un relief de collines, à une altitude de 18 m sur l’actuel niveau de la mer. Pendant l’Holocène le milieu naturel était très différencié, avec une plus haute élévation du niveau de la mer, ce qui approcherait la côte du gisement (Gracia, et al., sous presse). “ El Retamar ” a été localisé sur une dune, proche de dépôts de plages, dans la zone de Manchon de Mora. Donc la côte était proche de l’emplacement, et a facilité les activités de la pêche et du ramassage des coquillages. La datation absolue du gisement, en 6370 + 80 années B.P. (5.025 cal.) reste localisé immédiatement aux environs de la transgression Flandrienne (Zazo et al, 1994). Comme on l’a indiqué (Gracia et al, sous presse) cette élévation de la mer a uni “ El Retamar ” avec une baie interne ouverte sur la mer.

comportement en ce qui concerne les us et les coutumes. Ils sont plus petits que les escargotières, avec une prédominance de ceux qui mesurent dans les 60-90 cm, qui occupent un espace de 0,10-0,50 m2 et un volume inférieur à 0,50 m3, et des profondeurs entre 21-30 cm. Les escargotières sont des concentrations spatiales conformées par l’accumulation de mollusques, à côté de restes de l’icthyofaune et de faune terrestre. Des roches brûlés et des produits lithiques y sont aussi intégrés. Ils correspondent à des aires de consommation et accumulation de gaspillage. Ils présentent un rapport immédiat avec les structures de foyers. Ils s´associent aux activités de la productionconsommation. Les escargotières ont des dimensions plus grandes, plus de volumes et d’épaisseurs que les foyers: 80% d’entre elles présentent une plus grande extension et surpassent 120 cm dans plante; 60% surpasse les 150 m3 de volume et 90% oscille entre 10 et 30 cm de profondeur.

FOUILLE ET STRATIGRAPHIE Nous avons réalisé une fouille à l´automne 1995 et pendant l´hiver 1996. Quelques travaux de démantèlement en vue d’une urbanisation ont laissé une extension de sables, qui ont permis de documenter beaucoup de produits archéologiques se rapportant à des structures de foyers et d’escargotières. Nous avons fouillé un espace de 871 m2 en 5 coupes stratigraphiques. Nous avons délimité les structures et avons fait un registre microspatial de produits lithiques, poteries et restes de faune, de coquillages et poissons. Nous étudions les relations des produits avec les structures pour déduire les aires d’activité, et ainsi procéder à une reconstruction des processus de production et de consommation.

Les concentrations de pierres sont en relation avec des actions et réparations entre les foyers, avec les tâches de l’aménagement et même comme zone de dépôts d’ordures. Quelques-unes peuvent être associées aux structures pour la préparation de la nourriture et comme système de protection du vent. Leurs dimensions sont inférieures aux foyers, pas très profondes, avec prédominance des 10-15 cm d’épaisseur. On remarque essentiellement celles de dimensions inférieures à 60 cm, et ils occupent un espace qui oscille entre 0,10-0,50 m2. Sur le volume ce sont les structures documentées de dimension inférieure, qui restent comprise entre 0,1 m3 et 0,10 m3.

La stratigraphie du gisement est constituée par des marges du Tertiaire avec des biocalcarenites qui constituent le substrat géologique. Sur ceux-ci il y a un niveau du glacis Pléistocène Supérieur, et un peu plus haut une dune Holocène où se trouve le niveau d´occupation, qui est constitué par des sables noirs et bruns avec des structures. Par dessus continuent les sables qui se superposent par des sols de couleurs marron foncé.

LA CHRONOLOGIE Nous avons une datation qui vient du foyer 18 (coupe 1, quadrillage F, complexe 3: Re-91-1-FI (3) coquille) en analyse radiométrique standard:- Beta-90122. L’âge du radiocarbone conventionnel: 6780 + 80 BP.; cal BC 5025 (Le Radiocarbone Analytique de Bêta Laboratoire, Miami, Floride, USA).

LES STRUCTURES DISTRIBUTION SPATIALE DE PRODUITS Nous avons documenté 62 foyers, 10 escargotières, 2 sépultures et 24 concentrations de pierres, qui comptabilisent un total de 98 structures. Les structures proviennent de diverses aires d’activité réalisées dans le gisement. Ce sont des zones de production, de consommation et de gaspillage.

La distribution spatiale des structures, leurs associations avec les produits archéologiques et avec les restes osseux et malacologiques nous informent sur les activités sociales, domestiques ou productives. Nous avons travaillé avec un système de trois dimensions de mesures pour localiser les produits dans les coupes et les quadrillages. Nous établissons une relation des produits avec les structures et obtenons d’intéressants indices d’aires d’activité. Nous voulons indiquer qu’une fouille envisagée avec des prémisses méthodologiques définis et avec des questions appropriées peut apporter de l’information sur une société concrète, ses modes de vie et sur les aspects de la production et des relations sociales de production et de reproduction (Bate, 1998). Nous travaillons avec l’idée que les produits dans leurs distributions spatiales sont les conséquences d’activités et de relations sociales (Wünsch, 1989; Vila et Wünsch, 1990). Les aires

Les foyers sont des structures subcirculaires composées de grès et de roches subvolcaniques thermoaltérés. A l’intérieur il y a du sable avec une altération thermique et une accumulation de matière organique. En rapport avec ces derniers il y a des restes de taille et de produits lithiques avec des retouches, ainsi qu’une présence de malacofaune, ichtyofaune et faune terrestre, qui sont liés au préparatif de repas pour la consommation directe. Ils ont une forme de cuve en «U» et «V», avec des sables thermoaltérés du type SYR 2.5/1 (Munsell, 1975). Les patrons de régularité morphologique montrent des règles et modèles de 137

Figure 3. Plan general et stratigraphiedes structures des coupes 1, 3 et 5.

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

138

J. Ramos et al.: Le gisement Néolithique de “El Retamar”. Un exemple des communautés du VIe millénaire...

d’activité nous infèrent des modèles de comportement social qui se déduisent de la régularité des dimensions. A partir de ces associations (produits, structures régulières, indice d´occupation saisonnière, évidences d’aire de activité) nous formulons des hypothèses en rapport avec les relations sociales de production et avec la division du travail.

dans le gisement. Nous avons analysé les processus de travail développés en vue de la configuration d’outils. Le calcul général de restes de taille et des produits retouchés nous donne une idée considérable du processus du travail et de la production de l’industrie lithique dans “ El Retamar ”, comme production d’outils orientée vers la pêche et le ramassage de coquillages, l’élaboration de feux et la consommation de poisson et de mollusques.

La distribution de l’espace des produits lithiques est variée en fonction de leur présence dans les structures trouvées. Nous avons étudié leur distribution pour les coupes, les quadrillages et les structures. Nous présentons ici quelques données quantitatives dans leurs associations avec les structures. Dans les foyers ont été documentés 2.079 exemplaires, 1.759 produits lithiques taillés et 320 supports. Dans les escargotières il y a 1.378 exemplaires qui correspondent à 1.205 produits lithiques taillés et 123 supports. Dans les concentrations de pierres apparaîssent 447 exemplaires distribués en 428 produits lithiques taillés et 19 supports. Le calcul total de l’industrie lithique documentée dans le gisement montre 23.593 produits, 16.070 exemplaires de la campagne de fouille et 7.723 du démantèlement initial.

MATIÈRES PREMIÈRES DE L’INDUSTRIE LITHIQUE Les matières premières étudiées constituent un groupe de 744 fragments lithiques, sélectionnés parmi un total de 23.593 registres. La caractérisation minéralogique et pétrologique des différentes lithologies a été réalisée au moyen de l’étude à visu et grâce à des techniques physiques-chimiques, comme les rayons X, la diffraction et la microscopie optique polarisant. Après l’étude de ces matières il est possible qu’aparaissent dans cet emplacement 19 lithologies différentes: silex (massif, oolitiques, avec grand ooides, poreux, à rayures, etc.) et radiolarites, jusqu’à des rocs sédimentaires documentés dans la géologie locale (Gutiérrez, et al., 1991; Baena et al., 1987) comme le sont les grès de El Aljibe, calcaires et marges dont l’âge est compris entre le Triasique et le Tertiaire, et finalement des roches ignées typiques de la région comme le sont les dolerites. Statistiquement il y a une prédominance dans les industries taillés du silex (80%). Quartzites (8,3%), radiolarites (6,7%), et le reste des types lithologiques sont très minoritaires et en général chacun ne dépasse pas 1% du total. Les matières lithiques qui viennent des foyers ou qui montrent des traces de thermoaltérations sont: les grès, les dolerites et les marges. Dans toutes les structures dominent les grès.

Une synthèse de l’association des structures avec les produits nous informe que parmi les 62 foyers, 43 (69,35%) ont entre 1-100 produits lithiques taillés, 13 foyers (20,97%) ne contiennent pas du tout de produit lithique et seulement 6 foyers (9,68%) comptent plus de 100 registres. Dans les 24 concentrations de pierres il y en a 12 (50%) qui ont moins de 100 produits, 11 (45,83%) n´ont aucun produit lithique taillé et seulement 1 présente plus de 100 registres. Les 10 escargotières montrent des valeurs différentes des structures mentionnées, parce que proportionnellement ils ont plus de produits lithiques; pour 7 on a documenté plus de 100 produits et 3 seulement en ont moins de 100. Dans toutes les escargotières il y a des produits lithiques. Les 2 sépultures comptent moins de 50 produits lithiques. Il y a des structures et des aires unies à des zones de production lithique. Il y a seulement 6 structures (3 foyers, 2 escargotières et 1 concentration de pierres) qui ont plus de 100 produits taillés. Entre 101 et 150 produits taillés, seulement 3 foyers, 2 escargotières et 1 concentration de pierres. Un foyer et 3 escargotières comptent une moyenne de 151-200 produits. Comptant plus de 200 produits il y a seulement 2 foyers. Dans les foyers 11 et 12 sont documentés des processus d’élaboration et de débitage des produits lithiques. Ils constituent vraiment des lieux de production. La documentation remarquable de nucléus, les éclats et autres restes de la taille confirment ces processus de travail.

Des études antérieures sur les deux grands groupes de matières (l’industrie lithique taillée et les rocs de foyers, escargotières et accumulations) nous pouvons déduire des perspectives sur ce quil’a pu être l’origine de ces ressources. A propos de l’industrie lithique, les matières qui prédominent fondamentalement ont été les siliceuses de lithologies, basés sur le silex et le radiolarite. Dans les deux cas, la présence de nucléus taillés et restes de taille démontre qu’ils ont pu être élaborés in situ sur des matières premières de petits galets de silex et radiolarite fondamentalement. L’origine de ces galets peut se situer à quelque endroit des terrasses du lit bas de la rivière Guadalete, relativement proche de l’emplacement, bien qu’on ne puisse rejeter l’idée que quelques-unes de ces matières premières ont été transportées de plus loin. Il s’agit en général de matières siliceuses d’âge Jurassique et Crétacé, associées à des rocs calcaires des chaînes de montagnes d’Andalousie, de la région de la Sierra de Cadix. Le quartzite qui apparaît dans l’industrie lithique, est d’origine locale et a été associé à des dépôts de glacis Pléistocène (Zazo, 1980; Baena et al., 1987) où les galets lissés de cette matière sont abondants, et ils ont été fabriqués in situ.

LA TECHNOLOGIE LITHIQUE COMME INSTRUMENTS DU TRAVAIL La technologie lithique est caractéristique d’une communauté de pêcheurs, avec une tendance répandue à l’obtention de nucléus pour lames. Pour ce qui est fabriqué, au moyen de la technique du microburin, il s’agit de microlithes géométriques, surtout des trapèzes. La grande quantité de produits et leur variété technologique nous a permis de connaître le processus technologique de la production et une certaine distribution et organisation des produits lithiques

À propos des matières premières des pierres de foyers, escargotières, et concentrations de pierres, nous pouvons 139

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe Table 1. Balance générale de produits lithiques par structures. 0

ESTRUCTURES

1-100

> 100

TOTAL



%



%



%



%

FOYERS

13

20,97

43

69,35

6

9,68

62

100

C. DE PIERRES

11

45,83

12

50,00

1

4,17

24

100

ESCARGOTIÈRES

-

0,00

5

50,00

5

50,00

10

100

ENTERRAMENTS

-

0,00

2

100

-

0,00

2

100

24

24,49

62

63,27

12

12,24

98

100

TOTAL

V

T

U

R

S

Q

P

Ñ

O

M

N

L

LL

K

J

H

I

F

G

D

E

C

B

A

10

V 8 IV 6 III 4 II 2 I 48

46

44

40

42

38

36

32

34

30

28

26

24

22

20

18

16

14

12

8

10

6

4

2

Figure 4. Numéro de produits lithiques retouchés pour carrées dans coupes 1 et 3.

1 2

3

5

6

7

4

9

10

8

11 12

Figure 5. Produits lithiques retouchés du gisement du “El Retamar”. 140

13

0

J. Ramos et al.: Le gisement Néolithique de “El Retamar”. Un exemple des communautés du VIe millénaire...

séparer les lithologies différentes par ordre d’importance quantitative. Le grès de l’Aljibe est un matériau de la région, mais il n’y a pas d’affleurement d’importance jusqu’à plusieurs kilomètres de « El Retamar ». La présence d’une grande quantité de galets de cette matière dans les structures peut être due à un ramassage sélectif de ceux-ci dans les environs de l’emplacement, sur traînage fluvial. Les biocalcarenites proviennent de dépôts tertiaires localisés approximativement à 500 mètres au Nord-est du gisement où cette lithologie apparaît. Ce sont des matières premières locales par conséquent, la même chose qui a lieu avec les marnes de tons différents et quelques-uns des calcaires qui sont apparus. Les fragments de dolerites peuvent avoir leur origine dans des traînages superficiels, depuis leur affleurement, et ensuite, localisés au Nord, dans le ruisseau Salado et de là jusqu’à l’ancienne baie du Guadalete sur la berge duquel apparaissent des matières plio-quaternaires et quaternaires qui peuvent les contenir.

Les produits retouchés ont été analysés au moyen des ordinations de Fortea (1973) et Laplace (1973, 1986): 47 grattoirs ont été documentés (4,49%), 2 perçoirs (0,19%), 11 burins (1,05%), 60 lames à dos abattu (5,73%), 242 encoches et denticulés (22,49%), 33 fractures retouchés (3,15%), 107 microlithes géométriques (11,22%), 86 microburins (8,22%), 444 lames avec retouches d´usage (42,45%) et 14 divers (1,34%). La séquence structurelle est: A/2 S/1 SE/ B. Il est évident qu’il existe une syntonie historique de ce groupe avec les niveaux XVI-XIV et 13-12 de la caverne de Nerja (Aura, 1986; Jordá, 1986; González-Tablas, 1986), aussi bien qu’avec les niveaux appelés Tardenoisien et Capsien de « Hoyo de la Mina » (Such, 1920; Fortea, 1973). Le contexte immédiat avec la rivière Palmones dans la baie d’Algeciras est évident (Ramos, et al., 1995, 2001); ainsi que la continuité technologique avec des groupes régionaux de la tradition Épipaléolithique. Cela reflète la continuité de la population et les processus autochtones vers la fixation du nouveau mode de production des sociétés tribales (Sanoja et Vargas, 1995).

VALORISATION TECHNOLOGIQUE ET HISTORIQUE DE L’INDUSTRIE LITHIQUE

On apprécie ainsi une continuité technologique importante, la base du travail, d’une communauté qui a développé des formes de travail de la pêche et du ramassage de coquillages. C’est une communauté qui, ayant atteint de nouvelles formes dans le domaine de la production, avec l’élevage, maintient des formes traditionnelles de travail, comme complément économique. Les rapports et ressemblances contiennent beaucoup plus que ces syntonies techniques, étant donné la fréquence qu’ont ces instruments comme outils de travail et pour la conformation en vue de la production (Bate, 1998). Nous pensons que la marque de la continuité de la population et des processus autochtones vers la fixation de la nouvelle manière de production des sociétés tribales est très significative.

Il y a une présence considérable de galets et de fragments de silex et quartzites, qui ont été apportés du ruisseau de la Quijada, tout proche. Si nous évaluons les 16.070 produits qui viennent de la fouille, 1.053 nucléus sont documentés (7,28%), 6.602 éclats et lames (45,67%), 5755 autres restes de taille (esquilles, tablettes de ravivage, résidus) (39,81%). Les produits retouchés constituent 1046 exemplaires (7,24%). Il y a une moyenne de débitage de 6,27 éclats et lames par nucléus et cela confirme un épuisement de quelques nucléus (prismatiques et à lames), avec une présence remarquable d’esquilles et de résidus qui prouvent la production lithique dans le gisement. Les nucléus polyédriques prédominent avec 451 exemplaires (42,84%), suivis des 334 prismatiques (31,72%). Ceux-ci sont en relation avec les 41 nucléus pour lames (3,89%). La bonne présence de nucléus prismatiques et à lames permet une importante documentation de lames. Il y a aussi 105 nucléus Levallois (9,96%). Il y a 66 nucléus du commencement du débitage (6,30%) et dans une plus petite mesure 11 nucléus sont documentés avec un plan de frappe (1,04%), 5 sur éclats (0,47), 3 avec deux plans de frappes (0,28% ) et 37 divers (3,50%).

L´étude archéologique de l’espace, de la typologie, et le rapport avec les matières premières, permettent d’exposer des hypothèses au sujet de la propre considération morphologique-utilitaire et son imbrication dans les processus du travail: présence considérable d’outils à caractère domestique (grattoirs, burins, perforateurs), traditionnellement reliés au travail du bois, à celui des peaux et en général de la vie journalière d’appui et de réintégration de pratiques productives pour la consommation (Vila, 1987; Pie y Vila, 1991). Une prédominance d’éléments en relation avec les activités de la pêche et de ramassage des coquillages (tous les produits liés à la production de microlithes géométriques: microburins, fractures retouchés, trapèzes, triangles). Et aussi des outils se rapportant au travail du poisson et des mollusques à l’intérieur du gisement dans les foyers et les escargotières (lames à bord abattu, lames avec retouches d´usures).

Les 6.602 éclats et lames offrent une correspondance claire avec les nucléus obtenus. Le calcul des éclats et des lames parmi les groupes représentent 5.198 éclats (78,74%) et 1.404 lames (21,26%). Une analyse technologique des groupes permet d’observer une présence considérable d’éclats qui vient des commencements de la taille, avec 559 éclats d´amorçage (8,46%) et 1246 de décorticage (18,87%). Il y a aussi une prédominance quantitative d’éclats de taille interne, avec 3.143 exemplaires (47,62%) ; une présence modérée des éclats de la technique levallois, avec 243 exemplaires(3,68%) ; un bon échantillon de produits de taille à pression, avec 165 lames de crêtes (2,49%) et une grande proportion de lames: 1.216 exemplaires (18,42%).

De tout ce qui vient d’être mentionné on peut déduire l’importance de la technologie comme force productive, suggestive dans la définition du travail en relation avec un mode de vie basé sur la pêche, aussi bien pour la conformation de projectiles et de harpons que pour le travail en soi à l’intérieur du gisement. L’implication socio-économique de l’industrie lithique est que les modes de travail des chasseurscueilleurs sont insérés dans la même société qui déjà pratique 141

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

LE REGISTRE DE COQUILLAGES ET DE POISSONS

des formes agricoles dans les campagnes environnantes (La Mesa) (Ramos, et al., 1999). C’est la même communauté qui de façon saisonnière pratique la pêche et qui a déjà atteint un mode de vie villageois avec le développement de l´économie de production (premiers bergers et paysans).

Le restes malacologiques et ichtyologiques ont une grande importance dans l’alimentation de « Le Retamar » (Soriguer, Zabala et Hernández, sous presse). Les restes malacologiques ont une représentation qui permet de supposer un poids approximatif d’un peu plus de 100 Kg.

LA CÉRAMIQUE COMME PRODUIT POUR LA PRÉPARATION DE NOURRITURES ET CONSOMMATION

Si nous étudions la distribution de l’espace des 32 espèces de faune marine localisée dans l’établissement, nous observons que le groupe du Molusca est le plus représenté avec 25 d’entre elles, dont 10 bivalve, 13 gastéropodes et 2 crustacés. Parmi ces espèces on peut signaler numériquement les bivalves et parmi eux, ce sont les Solen marginatus et les Tapes decussatus les mieux représentés. Une plus petite présence ont les espèces Scrobicularia plana, Crassostrea angulata, Cerastoderma edulis. Le reste des bivalves a une représentation réduite Chlamys sp., Pecten sp., Glycimeris sp., Mytilus edulis, Acanthocardia sp. . À l’intérieur du groupe des gastéropodes la meilleure espèce représentée est le Trunculariopsis trunculus et le Murex Brandaris. Un total inférieur possèdent les espèces Cerithium vulgatum, Monodonta lineata, Hydrobia ulvae, Nassarius reticulatus, Patella vulgata, Astrea rugosa, Monodonta turbinata, Calliostomo zizyphinum, Omalogyra sp. et Helix aspersa.

Les découvertes de céramiques lors de la fouille n’ont pas été très nombreuses (118 registres). Dans tous les cas ce sont des fragments de faible proportion qui correspondent à des bols et récipients globulaires de petites dimensions. Les caractéristiques technologiques et morphologiques externes présentent une totale uniformité dans toutes les découvertes. En même temps, les analyses de caractérisation, au moyen de la microscopie électronique de balayage, à laquelle ont été soumis des échantillons, indiquent une origine probablement locale de ces récipients. La distribution spatiale des découvertes montre une plus grande prépondérance de céramique dans les foyers ou dans les zones environnantes. Parmi eux il faut signaler ceux qui sont localisés dans la coupe 3, où apparaît aussi une plus grande concentration de céramique décorée (cardiales et imprimés digités), en ce qui concerne les foyers et quelques concentrations de pierres. Ce qui peut-être pourrait être mis en relation avec une cabane ou, au moins, avec un lieu de repos et de refuge réalisé avec un système de parevents. Les caractéristiques techniques et morphologiques que présentent ces récipients nous permettent d’inférer des fonctions orientées vers la production et la consommation de nourritures à caractère domestique. Le lien existant entre ces fragments et les foyers indique un usage préférentiel.

La distribution de restes malacologiques des découvertes dans le gisement signale un plus grande pourcentage de découvertes dans la Coupe 1: cette présence de la majorité de mollusques est dû au nombre élevé de structures (7 escargotières et 54 foyers). Un élément plus considérable est la représentation du poids en grammes des échantillons choisis dans la Coupe 1 avec un poids total de 90.536,30 gr . La distribution ichtyologique des découvertes pour les coupes est semblable aux restes malacologiques bien qu’avec de légères variations. C’est la coupe 1 qui a fourni le plus grand nombre d’échantillons (1.373 inscriptions d’un calcul total de 1.756 découvertes). Il est aussi nécessaire de signaler que c’est dans cette coupe qu’a eu lieu la localisation diversifiée en espèces, parce qu’elle possède 5 des 7 espéces signalées, qui sont les Sparus aurata, Dentex sp., Diplodus sp., Lithognathus mormyrus, Argyrosomus regius, Thunnus thynnus et Galeorhynus galeus. Les restes de poisson correspondent à la famille des Sparus, et les échantillons les plus nombreux appartiennent à Sparus aurata pour la plupart.

BILAN DE LA FAUNE Concernant la faune sont présents les taxones suivants: Bos Taurus, Equus sp, Cervus elaphus, Capra hircus, Ovis aries, Ovis/Capra, Sus domesticus, Canis familiaris, Oryctolagus cuniculus, Lepus capensis et Alectoris rufa. La distribution dans l’espace des restes osseux dans le gisement est complètement variée et en rapport avec les diferentes structures. La plus grande concentration de restes est distribuée dans les coupes 1 et 3, autant en quantité qu’en fréquence. L’espèce la plus représentée est le lapin (30%), suivi par le cerf (7,4%), chèvre/mouton (5,8%), cheval/bœuf (6,8%), oiseau (2,1), cheval (1,1%), chèvre (1,05%), bœuf (0,1%), chien (0,1%), caille (0,06%).

Si nous conduisons l’analyse en fonction des structures où ils apparaissent nous constatons que c’est dans les foyers ou dans les régions adjacentes qu’il y a un plus grand nombre de découvertes. Nous avons aussi observé une différence dans la présence de certaines parties de restes de poisson dans quelques structures. Il existe quelques foyers où se concentrent un plus grand nombre de vertèbres et moins de fragments correspondent aux têtes (mâchoires, molaires, otolithes).

Dans l´aire localisée à l’est du gisement, dans la coupe 1 sont localisés principalement les secteurs consacrés à l’activité productive et à la préparation pour la consommation de la faune marine et terrestre. Autour des foyers sont distribuées les aires de la production pour la consommation. Quelques aires sont consacrées au travail de la viande (désarticulation, décharnement...), aires de consommation et d’accumulation de résidus.

La consommation de mollusques trouvée dans le gisement pourrait être réalisée d’une forme directe, c’est-à-dire une consommation crue. Cependant, les nombreuses structures thermoaltérées localisées dans le gisement où leurs restes 142

J. Ramos et al.: Le gisement Néolithique de “El Retamar”. Un exemple des communautés du VIe millénaire...

apparaissent nous conduisent à considérer que ceux-ci ont été cuits, soit à l’intérieur de fosses culinaires, soit directement sur les cendres ardentes des foyers ou introduits dans les poteries. La préparation du poisson serait encore plus simple. Ils pouvaient être placés sur les cendres ardentes, peut-être seulement vidés de leurs viscères, ou séchés et salés.

communauté d’autoconsommation, qui ne compte pas d’objets obtenus par la distribution ou l’échange. Les processus de production et de préparation de nourritures documentent des évidences de consommation immédiate de coquillages. Une partie de ces poissons pouvait être emportée vers les communautés villageoises de l’intérieur. Une force de travail intense a été documentée au moyen d’instruments élaborés avec une technologie très précise. Il reste des évidences des processus de pêche au moyen de captures avec les harpons munis de microlithes géométriques.

MODES DE PRODUCTION ET CATÉGORIES DE L’ANALYSE POUR UNE ÉVALUATION DE LA FORMATION ÉCONOMIQUE SOCIALE TRIBALE

La propriété dans ces sociétés est collective en ce qui concerne les objets, les outils de travail et les produits obtenus. C’est une société remarquablement équitable, mais nous ne pouvons pas écarter l’hypothèse d’une division technique du travail. Comme dans toutes les sociétés historiques, les rapports entre forces productives et rapports sociaux sont réglés par les formes de propriété. La communauté qui s’est établie à « El Retamar » devait s’approprier ce territoire d’une manière efficace. Cela reste documenté dans la régularité des fréquentations, les sépultures et la normalisation dans les dimensions des structures. Ces caractéristiques peuvent être évaluées comme modèles sociaux de comportement. Le registre archéologique montre des pratiques de travail mixte et de consommation sans élément de prestige, ni aucun qui révèle une origine externe. Dans « El Retamar » est présente une communauté tribale qui développe une forme collective de propriété sur la pêche. Cela se perçoit dans les modèles du comportement en rapport avec les récurrences, morphologies et dimensions des structures archéologiques, aussi bien que dans les sépultures. Toutes ces caractéristiques montrent différentes fréquentations. Les données apportées par la pêche de la daurade (Sparus auratus) posent l’hypothèse d’une occupation saisonnière entre octobre et décembre. Tout permet d’esquisser un contenu d’une propiété véritable et efficace sur les moyens naturels de production. Dans ce cas, pour le développement de ces travaux saisonniers il y aurait un déplacement vers le territoire de pêche aux environs de « El Retamar ».

L’analyse technologique, la distribution dans l’espace et l’association des produits organiques de faune terrestre et marine permettent de déduire l’importance du travail en rapport avec la pêche. Ceci reste documenté par la conformation de projectiles et de harpons, et par le travail à l’intérieur du gisement. Nous travaillons avec l’hypothèse que “ El Retamar ” est un établissement fréquenté par une communauté tribale qui développe un mode de production défini par une exploitation des moyens naturels côtiers, sous la forme de pêche et de ramassage de coquillages. L’économie de la production a été confirmée dans la domestication des bœuf, chèvre, mouton et porc. Nous avons indiqué l’importance du registre de malacofaune et icthyofaune dans la consommation de l’établissement. L’étude de la faune a confirmé la consommation d’espèces animales avec documentation considérable de la chasse et prédominance claire de faune sauvage sur le faune domestique. Ils ont chassé les cerfs, les lapins et les oiseaux surtout. Il y a des évidences de domestication du bœuf, chèvre, mouton, porc et cheval. Les indices de l’agriculture n’ont pas été relevés. Il y a des évidences d’espèces sauvages, comme l’olive sauvage et de Quercus (Uzquiano et Arnanz, s.p.). Cette société a disposé d´une grande diversité de ressources qui ont été utilisées dans la production et reproduction sociale. Les ressources nutritives ont été variées, aussi bien la faune marine que la faune terrestre, et les ressources des végétaux sauvages. “ El Retamar ” constitue un exemple de communauté qui comptent déjà des pratiques de domestication animale, au commencement de l’économie de production. Aussi bien avec une base importante de pêche, chasse et cueillette de végétaux. Le développement de la production a été accompli par la conformation d’une technologie précise, comme instruments de travail. Etant donné son milieu côtier, « El Retamar » occupe un lieu d’économie diversifiée. Il possède une aire de pêche, avec des possibilité de ressources de chasse (cerfs, oiseaux, lapins...). Le milieu facilite aussi la provision des ressources lithiques pour l´élaboration d´outils de travail, et aussi des ressources minéralogiques, d’argiles et de composants pour l’ élaboration des poteries.

« El Retamar » est en rapport avec des gisements néolithiques de la Baie de Cadix (région de San Fernando (Ramos, et al., 1994), Chiclana de la Frontera (Ramos, et al., 1997, 1999)) qui témoignent communautés qui ont développé des modes de vie basés sur la pêche. Ces communautés ont en même temps, dans les campagnes de l’intérieur, un mode de production basé sur les formes agricoles. Cela prouve que le territoire des campagnes de l’intérieur a développé un mode de vie villageois équitable, alors que sur les côtes et îles de la Baie de Cadix s’est développé le travail de pêche et ramassage de coquillage. Nous travaillons actuellement à l’analyse de ces processus de travail entre communautés de la côte et de l’intérieur (Schumacher et Weniger, 1995), et prétendons connaître les synchronies et leurs rapports sociaux.

Des objets ou produits reçus par les mécanismes de distribution ou d’échange n’ont pas été découverts. La technologie lithique est totalement locale, aussi bien pour les produits lithiques taillés, dans une variété locale de silex, comme les pierres utilisés dans les foyers et les structures (dolerites, grès, roches subvolcaniques...). Il s´agit d’une

Dans les sociétés tribales il y a un rapport clair entre la production et l’idéologie. Dans « El Retamar» il y a une correspondance claire entre propriété de la communauté et accès aux moyens de production. Il n’y a pas du tout de témoignage qui marque une contradiction sociale. Les sépultures sont simples, contenant des produits lithiques et 143

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

organiques quotidiens de la vie de ces communautés. La documentation des sépultures est en relation avec une propriété d’un territoire comme objet du travail (Bate, 1998:93). C’est une forme idéologique qui donne un aspect institutionnel à la formation sociale et se reproduit comme système de valeurs dans la société.

Mediterránea. Historia Universal Prehistoria 4. Editorial Síntesis. Madrid. CALADO, D., 2000. « Poblados con Menhires del extremo Suroeste Peninsular. Notas para su cronología y economía.». Revista Atlántica Mediterránea de Prehistoria y Arqueología Social III, pp. 47-99. Universidad de Cádiz CASTAÑEDA, V., 2000: Las sociedades cazadoras-recolectoras especializadas en Andalucía. Tesis Doctoral. Universidad de Cádiz.

Adresses des auteurs

FORTEA, J., 1973: Los complejos microlaminares y geométricos del Epipaleolítico mediterráneo español. Salamanca.

José RAMOS, María LAZARICH, Nuria HERRERO, Vicente CASTAÑEDA, Manuela PÉREZ, Mª Eugenia GARCÍA, Isabel CÁCERES Área de Prehistoria Facultad de Filosofía y Letras Universidad de Cádiz Avda. Gómez Ulla s.n. 11003 Cádiz, ESPAGNE

GÁNDARA, M, 1993: «El análisis de posiciones teóricas: aplicaciones a la arqueología social» Boletín de Antropología Americana 27, pp. 5-20. México. GONZÁLEZ-TABLAS, F. J., 1986: “La ocupación postmagdaleniense de la Cueva de Nerja (la Sala de la Mina)”. En JORDÁ, F.,: La Prehistoria de la Cueva de Nerja (Málaga). En Trabajos de la Cueva de Nerja 1, pp. 270-282. Málaga.

Salvador DOMÍNGUEZ-BELLA Área de Cristalografía y Mineralogía Facultad de Ciencias Universidad de Cádiz Campus Río San Pedro Puerto Real Cádiz, ESPAGNE

GRACIA, J., BENAVENTE, J. et MARTÍNEZ, J.A., sous presse: “Geomorfología y emplazamiento. Enmarque Holoceno de ‘El Retamar’”. En RAMOS, J. et LAZARICH, M. (Édition) sous presse: El asentamiento… GUTIÉRREZ, J .M.; MARTÍN, A.; DOMÍNGUEZ-BELLA, S. et MORAL, J. P., 1991: Introducción a la geología de la provincia de Cádiz. Universidad de Cádiz, 315 pp. JORDÁ, F., 1986: La Prehistoria de la Cueva de Nerja (Málaga). Primera Parte. Paleolítico Superior y Epipaleolítico En Trabajos de la Cueva de Nerja 1. Málaga.

Bibliographie ACOSTA, P., 1987: “El Neolítico Antiguo en el Suroeste español. La Cueva de la Dehesilla (Cádiz)”. Premières Communautés Paysannes en Méditerranée Occidentale, pp. 653-659. CNRS. Paris.

LAZARICH, M., RAMOS, J., CASTAÑEDA, V., PÉREZ, M., HERRERO, N., LOZANO, J., GARCÍA, E., AGUILAR, S., MONTAÑÉS, M. et BLANES, C., 1997: “El Retamar (Puerto Real, Cádiz). Un asentamiento Neolítico especializado en la pesca y el marisqueo”, II Congreso de Arqueología Peninsular. T. II, pp. 49-58. Zamora.

ACOSTA, P., 1995: “Las culturas del Neolítico y Calcolítico en Andalucía Occidental”. Espacio, Tiempo y Forma. Serie I. Prehistoria y Arqueología, pp. 33-80. UNED. Madrid.

LAZARICH, M., RAMOS, J., CASTAÑEDA, V., PÉREZ, M., HERRERO, N., BLANES, C., MONTAÑÉS, M., LOZANO, J., GARCÍA, E., AGUILAR, S. et NÚÑEZ, M., 1998: “El Retamar: un asentamiento de pescadores del VIº milenio a.n.e. en la Bahía de Cádiz”. V Jornadas de Historia de Puerto Real, pp. 56-73. Puerto Real.

AMMERMAN, A., 2000: “Retorno a la transición neolítica en Europa”. El paisaje en el Neol. Mediterráneo. Valencia. AMMERMAN, A.J. et CAVALLI-SFORZA, L., 1984: The Neolithic Transition and the Population Genetics in Europe. Princeton University Press.

LAZARICH, M., RAMOS, J. CASTAÑEDA, V., BLANES, C., MONTAÑÉS, M., HERRERO, PÉREZ, M., LOZANO, J., et GARCIA, E., 1999: “Excavación de urgencia en el asentamiento “El Retamar” (Puerto Real, Cádiz). Informe preliminar”. Anuario Arqueológico de Andalucía, 1995. II, pp. 67-73. Junta de Andalucía. Sevilla.

ARTEAGA, O., 1992: “Tribalización, jerarquización y estado en el territorio de El Argar”. Spal 1, pp. 179-208. Universidad de Sevilla. ARTEAGA, O. et CRUZ AUÑÓN, R., 1999: “El asentamiento al aire libre de ‘Los Álamos’ (Fuentes de Andalucía, Sevilla). Excavaciones de urgencia de 1995”. Anuario Arqueológico de Andalucía.1995. II, pp. 559-566. Sevilla.

MIKDAD, A. et EIWANGER, J. 2000: “Recherches préhistoiques et protohistoriques dans le Rif oriental (Maroc). Rapports préliminaires” Beiträge Zur Allgemeine und Vergleichende Archäologie. Band 20, pp. 109-167. Bonn.

ARTEAGA, O. et HOFFMANN, G., 1999: “Dialéctica del proceso natural y sociohistórico en las costas mediterráneas de Andalucía”. Revista Atlántica-Mediterránea de Prehistoria y Arqueología Social II, pp. 13-121. Univ. de Cádiz. AURA, J.E., 1986: “La ocupación magdaleniense de la Cueva de Nerja (La Sala de la Mina)”. En JORDÁ, F.: La Prehistoria de la Cueva de Nerja (Málaga). Trabajos sobre la Cueva de Nerja 1, pp. 207-267. Málaga.

MUNSELL, A.H., 1975: Soil Color Chart PELLICER, M. et ACOSTA, P., 1986: “El Neolítico y Calcolítico de la Cueva de Nerja”. En Trabajos de la Cueva de Nerja 1, pp. 339-450. Patronato de la Cueva de Nerja. Málaga. PIE, J. et VILA, A., 1991: « Relaciones entre objetivos y métodos en el estudio de la industria lítica ». En Tecnología y cadenas operativas líticas. Treballs d’Arqueologia I, pp. 271-278. Barcelona.

BAENA, J.; ZAZO, C.; GOY, J.L.; DABRIO, C.; LEYVA, F et RUIZ, P., 1987: “Memoria y mapa geológico de Paterna de Rivera. Hoja nº 1.062”. Instituto Geológico y Minero de España, 54 pp. Madrid. BATE, L. F., 1998: El proceso de investigación en Arqueología. Editorial Crítica. Barcelona.

RAMOS, J., coord., 1995: El Paleolítico Superior Final del río Palmones (Algeciras, Cádiz). Instituto de Estudios Campogibraltareños. Algeciras.

BERNABEU, J., AURA, J. E. et BADAL, E., 1995: Al Oeste del Edén. Las primeras sociedades agrícolas en la Europa

RAMOS, J., 1999: Europa prehistórica. Cazadores-recolectores. Editorial Sílex. Madrid. 144

J. Ramos et al.: Le gisement Néolithique de “El Retamar”. Un exemple des communautés du VIe millénaire... RAMOS, J, 2000: «Las formaciones sociales son mucho más que adaptación ecológica», Revista Atlántica Mediterránea de Prehistoria y Arqueología Social III, pp. 29-46. Universidad de Cádiz.

Carihuela. Síntesis ambiental del Würm mediterráneo en la Cordillera Bética. I.A.C.T. Granada. SANOJA, M. et VARGAS, I., 1995: Gente de la canoa. Economía política de la antigua sociedad apropiadora del Noreste de Venezuela. Fondo Editorial Tropykos/Comisión de Estudios de Postgrado. Caracas.

RAMOS, J. et LAZARICH, M. (Édition) sous presse: El asentamiento de « El Retamar ». Contribución al estudio de la formación social tribal y a los inicios de la economía de producción en la Bahía de Cádiz. Universidad de Cádiz.

SCHUMACHER, T.X. et WENIGER, G.C., 1995: « Continuidad y cambio. Problemas de la Neolitización en el este de la Península Ibérica ». Trabajos de Prehistoria 52, 2, pp. 83-97. Madrid.

RAMOS, J., CASTAÑEDA, V., PÉREZ, M., LAZARICH, M. et MONTAÑÉS, M., 2000: « Contributions to the study of the specialized hunter-gatherer production mode and to the beginning of the production economy in the Atlantic coast of Cadiz (Southern Spain). En FINLAYSON et al (Ed): Gibraltar during the Quaternary, pp. 135-158. Gibraltar.

SORIGUER, M., ZABALA, C. et HERNANDO, J. sous presse: «Características biológicas de la fauna marina del yacimiento `El Retamar’», en RAMOS, J. et LAZARICH, M. (Édition): El asentamiento… Universidad de Cádiz. SUCH, M., 1920: Avance al estudio de la caverna “Hoyo de la Mina”. Sociedad Malagueña de Ciencias. Málaga.

RAMOS, J., GARCÍA, M.E., CASTAÑEDA, V., JURADO, G., SÁNCHEZ, M., DOMÍNGUEZ, S., GRACIA, J. et MONCAYO, F., 2001: « Primeros resultados de la campaña de excavaciones desarrollada en el asentamiento de cazadores-recolectores del Embarcadero del río Palmones (Algeciras, Cádiz) ». Almoraima 25, pp. 81-90. Algeciras.

VARGAS, I., 1987: “La formación económico social tribal”. Bol.Ant.Am. 15, pp. 15-26. México. UZQUIANO, P., et ARNANZ, A., sous presse : « La evidencia arqueobotánica. Los macrorrestos carbonizados del yacimiento de ‘El Retamar’ ». En RAMOS, J. et LAZARICH, M. (Édition) sous presse: El asentamiento…

RAMOS, J., LAZARICH, M., CASTAÑEDA, V., PÉREZ, M., MONTAÑÉS, M., HERRERO, N., et GARCÍA, M.E.., 1997: “Los inicios de la economía de producción en la Bahía de Cádiz”. En RODRÍGUEZ, A. (Edt.): o Neolítico Atlántico e as Orixes do Megalitismo, pp. 677-689. Universidad de Santiago de Compostela.

VILA, A., 1987: Introducció a l’estudi de les eines lithiques prehistoriques. Institut Jaume Almera, CSIC. Barcelona. VILA, A. et WÜNSCH, G., 1990: “Un pequeño paso antes del gran salto. Buscando como preguntar (o investigación teóricometodológica en Tierra del Fuego)”. Xábiga, 6, pp. 19-30. Xábia.

RAMOS, J., MONTAÑÉS, M., PÉREZ, M., CASTAÑEDA, V., HERRERO, N. GARCÍA, M.E., et CACERES, I., 1999: Excavaciones arqueológicas en La Mesa (Chiclana de la Frontera, Cádiz). Campaña de 1998. Aproximación al estudio del proceso histórico de su ocupación. Ayuntamiento de Chiclana. Fundación Vipren. Universidad de Cádiz.

WUNSCH, G., 1989: “La organización interna de los asentamientos de comunidades cazadorasrecolectoras: el análisis de las interrelaciones espaciales de los elementos arqueológicos”. Trabajos de Prehistoria, 46, pp. 13-33. Madrid. ZAZO, C., 1980: El cuaternario marino-continental y el límite PlioPleistoceno en el litoral de Cádiz. Ph. D. Thesis, Univ. Complutense, Madrid (Spain).

RAMOS, J., SÁEZ, A., CASTAÑEDA, V. et PÉREZ, M. (coord..), 1994: Aproximación a la Prehistoria de San Fernando. Un modelo de poblamiento periférico en la Banda Atlántica de Cádiz. Ayuntamiento de San Fernando.

ZAZO, C., GOY, J.L., SOMOZA, L., DABRIO, C., BELLUOMINI, G., IMPROTA, J., LARIO, J., BARDAJÍ, T. et SILVA, P., 1994: « Holocene sequence of sea-level fluctuations in relation to climatic trends in the Atlantic-Mediterranean linkage coast ». Journal of Coastal Research, 10, pp. 933-945. Florida.

RUIZ BUSTOS, A., 2000: Estudio paleoecológico de los sedimentos con presencia del Hombre de Neandertal en la Cueva de la

145

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

146

L. Larsson: Interacting Cosmology. Ritual Use Of Fire during Two Phases of the Southern Scandinavian Neolithic

INTERACTING COSMOLOGY: RITUAL USE OF FIRE DURING TWO PHASES OF THE SOUTHERN SCANDINAVIAN NEOLITHIC Lars LARSSON

Résumé : L’utilisation du feu pour des buts rituels ou cérémoniels est un phénomène bien connu dans certains contextes préhistoriques, et complètement inconnu dans d’autres. Pourtant, l’utilisation du feu dans des cérémonies liées aux pratiques mortuaires, ont pu avoir un plus grand utilisation que nous avons l’habitude de penser. C’est tout aussi vrai pour des cérémonies liées aux sacrifices. Cet article propose deux exemples de ces phénomènes venant de la Suède méridionale. Dans deux sites Néolithiques du sud-est de la Scanie, la province la plus méridionale de la Suède, nous avons pu constater une destruction massive par le feu d’artefacts de silex et de pierre. Au cours des années 1980s, le site LBK de Svartskylle a été trouvé. Le site date de la transition entre le Néolithique ancien et le Néolithique Moyen, à environ 3400 av J. Chr. Sur ce site un grand nombre d’haches ont été fragmentées par un traitement de feu. A une distance de 17 km vers l’est de Svartskylle, il y a un autre site contenant un grand nombre de fragments d’objets en silex détruits par le feu. Ici plus d’un centaine d’haches, de burins et un grand nombre de pointes de flèches ont été endommagés par le feu. Aussi des objets en pierre, comme des haches de diverses types ont subi le même traitement. Parmi les artefacts on trouve aussi bien des types régionaux que des objets importés de loin. Les artefacts sont issus de la culture régionale appelée «Battle Axe Culture,» et ils datent d’environ 2300 av. J. Chr. Ce que semble exprimé à ces sites paraît différent par rapport à la destruction d’artefacts isolés qu’on retrouve à travers le Néolithique moyen dans les sites d’habitation, dans les mégalithes, et dans les sites de palissades. Dans certains occasions on a l’impression que la destruction d’objets en silex a crée un effet imposant. La consommation par le feu est un événement public, direct, évocatif et magique. Le silex se transforme et change couleur de noir ou gris au rouge et finalement au blanc. Ce changement ressemble à l’incinération d’un corps humain. La destruction de la culture matérielle a du être très évident, et la richesse ainsi exprimée dans le nombre d’outils et d’objets exotiques a du être considérable. Abstract: Fire for ritual and ceremonial purposes during prehistory is well known in certain respects yet almost unknown in others. Yet fire in different ceremonies connected with mortuary practice had a much wider use than is often considered. The same is true of the ceremonies related to sacrifices. Two examples from southern most part of Sweden are provided. Massive demolition of flint and stone artefacts by fire has been found at two Neolithic sites in south-eastern Scania, the southernmost county of Sweden. During the late 1980s, Svartskylle was found, a site dated to the Funnel Beaker Culture at the transition from the Early Neolithic to the Middle Neolithic at about 3400 BC. At the site fragments showed that several thin-butted flint axes as well as a small number of thin-bladed axes had been heavily fragmented by fire. Another site with fragments of flint artefacts destroyed by fire was found about 17 km to the east of Svartskylle. More than a hundred flint axes and chisels as well as a large number of arrowheads were damaged by fire. Non-flint tools such as axes and battle-axes had also been affected by fire. The material includes regional types as well as long-distance items. They are dated to the Battle Axe Culture at c. 2300 BC. What is expressed at this site differs from the destruction of single artefacts by fire, which is evident at settlement sites, megalithic tombs and causewayed enclosures throughout the Middle Neolithic. On certain occasions destruction of flint tools seems to achieve an imposing effect. Ritual burning has a public, direct, evocative and magical appearance. Flint undergoes a colour transformation from natural black or grey, sometimes to red and finally to white. This change is similar to the cremation of a human body. The destruction of material culture must have been very obvious, and the wealth represented by the number of tools and exotics included must have been considerable.

INTRODUCTION

Intentional destruction of flint by fire, however, seems to be a much later phenomenon.

In many societies earth, water, air and fire constituted the elements of which the universe was formed. Fire was intentional used already during the early Palaeolithic. It was first given a function in ceremonials in the Upper Palaeolithic for such purposes as cremations at Lake Mungo in southeastern Australia (Barbetti & Allen 1972). The habit of destroying or rather transforming artefacts by fire can also be found during the Upper Palaeolithic. In Dolni Vestonice in the Czech Republic, clay figures cracked in pieces when exposed to fire (Soffer et al. 1993). The number of fragmented clay figures is a strong indication of intentional destruction.

Scandinavian archaeologists have long interpreted collections of artefacts in water or wetlands as results of ritual acts (Larsson 1998, 2001), but they have had a more indifferent attitude to sacral destruction by burning. This has not received the same attention as offerings in wetlands. For prehistoric people, fire meant warmth, cooking and protection. Fire had important implications, including rebirth and cleansing. Prehistoric ritual burning was much more common than we usually suspect. 147

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. Southern Scandinavia (top) and Scania – southernmost part of Sweden (bottom) – with the sites mentioned in the text.

THE USE OF FIRE IN MASSIVE DEMOLITIONS In south-eastern Scania, in the southernmost part of Sweden, two Neolithic sites have evidence of massive intentional demolition by fire. During surveys for Neolithic settlement, as part of research involving the cultural landscape in the 1980s, one such site was found at Svartskylle (Fig. 1). Several thin-butted flint axes as well as a small number of thin-bladed axes, flake scrapers and blades, along with some ordinary flakes, had been heavily fragmented by fire. The axe types date the finds to the later part of the Early Neolithic, c. 3500 cal. BC (Larsson 1989). The burnt flints lay in four

concentrations on a hill with a dominant position in the landscape and partly surrounded by wetlands. Svartskylle is situated 12 km from the coast. During the Early Neolithic this area as well as the coastal area bore evidence of well-established settlement (Larsson, M. 1992). The area in between was sparsely settled and probably used as hunting grounds and pastureland (Berglund 1991). During the extensive revised survey of monuments carried out in Scania during 1990s, fire-damaged flint artefacts were spotted in a field at Kverrestad about 17 km east of the 148

L. Larsson: Interacting Cosmology. Ritual Use Of Fire during Two Phases of the Southern Scandinavian Neolithic

previously mentioned site and at about the same distance about 15 km - from the sea (Fig. 1). Just like the site at Svartskylle, the other site is situated in a specific setting. A river valley at Kverrestad runs from east to west, delimited to the south by a marked ridge. The site is situated on a small plateau on the southern side of the river with marked slopes on three sides and with a wetland below the site. The bedrock at the site is made up of slate. In the main part of the excavated area only a thin layer of clayish sand, which equals the plough zone, covers the bedrock.

These marked differences indicate an intentional selection of which type of tool to put in fire and which not to. That the offerings at Svartskylle and Kverrestad are the result of short-term activities indicates that they should be regarded as completed depositions of mass material. These seem to have been of exceptional size and intended to impress humans as well as metaphysical beings. The destruction of material culture should have been very obvious and the wealth represented by the number of tools and exotics included must have been considerable.

Fire-damaged flints could be found on the plateau within an area of approximately 70 x 70 m (Larsson 2000). Excavations revealed a number of pits of varying size and depth in which damaged flint and stone artefacts had been deposited together with a considerable amount of fragmentary pottery. The largest pit was about 4 m long, the shortest less than 0.5 m. The depth varied from more than a metre to less than ten centimetres below the topsoil. Finds were made throughout the filling, which show that the artefacts were deposited during the entire process of filling in the pits.

With the exception of Kverrestad and Svartskylle, the finds from just one other site in southern Scandinavia indicate the mass destruction by fire. More than one hundred fire damaged fragments from axes, chisels and scrapers have been found at Strandby, southern Funen (Andersen 2000). The pottery from the site dates the deposition to the earliest part of the Middle Neolithic. This type of public sacrifice of rare objects and with a direct effect may have been practised on special occasions, probably in combination with external or internal threats. In addition this could be an act, which was primarily meant to legitimate power by impressing representatives of another community.

Fragments from about a hundred thick-butted hollow-ground axes and chisels have been found, as well as a small number of thin-bladed axes. There is a variety in the degree of final polishing of the axes. Among the arrowheads, tanged arrowheads have been identified, as well as pressure-flaked projectile points. These types of points have not been identified in southern Sweden before. However, a small number have been found in Denmark (Ebbesen 1980). They are present in the Battle Axe culture of the upper Oder area. The finds from Kverrestad make up the largest collection of such points in Scandinavia. Fragments of a prototype of daggers, known as feeding-knives (Nielsen 1976), have been identified as well. Flake scrapers and large blades have also been damaged by fire.

BURIALS AND FIRE Although the time difference between Svartskylle and Kverrestad is more than a millennium, there are other finds in the neighbourhood, which might partly bridge the time gap. Destruction of tools by fire is relatively frequent at sites throughout the Neolithic (Karsten 1994). The destruction of flint artefacts by fire seems to be less common in burial contexts dated to the Early and Middle Neolithic. However, during a late part of the Funnel Beaker culture axes destroyed by fire are found outside the entrances to megalithic tombs.

Non-flint tools such as thick-butted axes and battle-axes have been exposed to fire. Among the finds there are battle axes of regional types as well as of a late type from the West Danish Single Grave culture. The stone axes have not only been broken to pieces, like the flint axes they have also been subjected to demolition, such as destruction of the edge.

Outside two megalithic tombs in south-eastern Scania, Trollasten and Ramshög (Fig. 1), features have been excavated which included flint tools exposed to fire (Strömberg 1968, 1971). Concentrations of burnt human bones, including axes and chisels damaged or destroyed by fire, were found on the stone setting in front of a dolmen. The axes as well as other finds date the features to the late Funnel Beaker culture.

Clay vessels, mainly of a semicircular shape or rounded with no marked transition between the belly and the neck, decorated with large zigzags on the belly and horizontal impressions just below the rim, also occur. Burnt human bones, intentionally cracked into small pieces, were also found.

Alteration of flint by fire is practised in other circumstances. Large quantities of unworked flint are often found in connection with megalithic tombs. Burnt flint is found mixed with clay, which covers the orthostats and sometimes the entire chamber. In these cases the flint has been interpreted as constituting a material with a special quality, namely, the ability to absorb moisture in order to keep the chamber dry (Strömberg 1971). Another purpose relates to the aesthetic quality of burnt flint - the white colour. A thick layer of burnt flint covered the surface along the southern side of a passage grave, Kong Svends Høj, on the Danish island of Lolland

The dating to the latest part of the Battle Age Culture at about 3900 BP, equal to about 2300 cal. BC, is in good agreement with all the finds. Of special interest is the degree of exposure to fire shown by the different tool types. While more than 90% of the axe finds display changes by fire, just one-third of the arrowheads made by pressure-flaking has the same alteration by fire. 149

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

(Fig. 1). Based on the amount of finds from some test pits, between 4 and 5 tons of burnt flint was used (Dehn et al. 1995).

In the Funnel Beaker culture cremated human bones as well as axes exposed to fire appear in several different contexts such as settlement sites, graves, causewayed enclosures and wetland offerings. Bones and axes might have acted within a common cosmology but were used as agents in different rituals.

Destruction by fire is not common in the context of burials from the Battle Axe culture (Hansen 1917; Malmer 1962; Särlvik & Jonsäter 1974). However, recent excavations have shown that cremation as well as other uses of fire in the mortuary practices are more common than previously thought.

Axes might have also been regarded as agent for supernatural forces and treated in a way similar to some of the human members of the society. The burning of axes should be understood as a kind of cremation.

Despite the presence of cremated human bones in Kverrestad, the deposition cannot be regarded as a true grave deposition. No other grave-good depositions show any similarities to the number of tools and vessels found.

FIRE – DESTROYER AND CREATOR There is a very marked difference in the attitude towards burning compared with offering in wetlands which was very frequent in southern Scandinavia (Karsten 1994). In wetlands the deposition of artefacts does not change their function. In the former, the destruction of the artefact is easily visible at the point when the practical function of the tool ceases to exist. Fire is the destroyer. Destruction by fire had a direct visual impact. The artefact undergoes remarkable and rapid changes during the act. A colour transformation takes place from natural black or grey to white. Some changes are similar to the cremation of a human body when the colour of the bones changes to white. Ritual burning has a public, direct, evocative and even more magical appearance.

Cremated human bones do appear in many other circumstances. One example is the observations made during the excavation of the initial causewayed enclosure at Sarup on south-western Funen, Denmark, dated to an early part of the Funnel Beaker culture. Burnt human bones were found close to the palisade of the first enclosure at the same site (Andersen 1997). Flint axes exposed to fire are common in connection to causewayed enclosures in southern Scandinavia as well as henge structures dated to the transition from the Funnel Beaker culture to the Battle Axe culture (Nielsen 1997). In cases like Sarup and Kverrestad, parts of humans seem to have been an active component of the ritual use of fire as well as the deposition. Parts of the same human body might have been deposited during different activities and at different sites related to the society. Through deposition and different media such as soil, water and fire, the human body was transferred to the supernatural world. In such a way the deposition of some people might have enclosed the entire space in which they played an active part during their lifetime.

Fire as a medium for transformations connected to rites de passage has mainly been applied to the mortuary practice but has also been used in many other circumstances (van Gennep 1909; Douglas 1966; Block 1971; Perlès 1977; Huntington & Metcalf 1979). Fire was the destroyer but at the same time the creator. It is also conceivable that transformation by fire changed character because of purely concrete circumstances. For example, there are major differences as to burning of the forest during the Early and Middle Neolithic. During the former period the trees were cut and burnt – slash-and-burn management. Later, however, parts like the root shoots were protected in order to produce leaf-fodder (Göransson 1996). The significance of fire thereby changed from a medium of total destruction to an element that prepared the way for the budding of new life.

AXES AND FIRE That flint axes represent the majority of finds at Kverrestad as well as Svartskylle is by no means remarkable. Axes seem to be the most used component in many activities of ritual relations during the Neolithic. In the large number of votive depositions in wetland of southern Scandinavia, axes make up the majority of finds (Karsten 1994; Nielsen 1985). Burnt axes rarely appear in wetland depositions.

Burning was destructive, but in some cases it had a conserving, preserving result. Posts were singed so that they would last longer in the earth. This was known as early as the Stone Age.

During the late part of the Middle Neolithic, finds from the Pitted Ware culture contemporaneous with the Battle Age culture provide examples of destruction involving flint axes. Of the axe fragments from the late Pitted Ware culture site at Jonstorp RÄ in north-western Scania (Fig. 1), about twothirds had been exposed to fire (Malmer 1969). Of the axe fragments from the excavation, 6% have no polish. But in the surface collection from the same site 21% of the axe fragments are not polished. The interpretation is that offerings of unpolished axes in which fire was involved were performed in the neighbourhood of the settlement, and later spread by ploughing.

The deposition of artefacts recorded at Kverrestad is a ritual practice, which seems to be a part of a process with several analogies during the Neolithic as well as the Mesolithic. During the Mesolithic, use of fire in rituals is mainly related to cemeteries from the Late Mesolithic. At the cemeteries at Skateholm I, southernmost part of Scania, a small number of cremation graves were found (Larsson 1993). Cremation graves are forthcoming at Mesolithic sites related to the 150

L. Larsson: Interacting Cosmology. Ritual Use Of Fire during Two Phases of the Southern Scandinavian Neolithic

former lagoon at Vedbæk, eastern Zealand (Kannegaard Nielsen & Brinch Petersen 1993).

Sandgatan 1, SE-223 50 Lund, SWEDEN Email: [email protected]

However, fire can also be related to ritual without cremation. Traces of wooden structures raised over the grave-pit were found at Skateholm. These had been burnt down prior to the refilling of the grave. On of several dog graves were found just below a fireplace. The fire has not affected the dog.

Bibliography ANDERSEN, N. H., 1997, The Sarup Enclosures. The Funnel Beaker Culture of the Sarup site including two causewayed camps compared to the contemporary settlements in the area and other European enclosures. Jutland Archaeological Society Publications XXXIII:1. Aarhus: Aarhus University Press.

In Western Europe cemeteries are found in and below shell middens, such as Téviec and Hoïdic on the southern shore of Brittany (Pequart et al. 1937; Pequart & Pequart 1954; Schulting 1996). The use of fire in the mortuary practice without effecting the interred is evident.

ANDERSEN, N. H., 2000. Kult og ritualer i den ældre bondestenalder. Kuml 2000, p. 13–57. BARBETTI, N. & ALLEN, H., 1972, Prehistoric Man at Lake Mungo, Australia, by 32,000 BP. Nature 240, p. 46–48.

The earliest Neolithic analogies in southern Scandinavia are probably the destruction by fire of the palisade of posts in front of the grave which was then covered with soil – the first stage in the construction of earthen long barrows during the earliest part of the Neolithic (Larsson 2002). There are examples of firing around some of the earliest stone chambers in long dolmens, which were totally covered by earth (Ebbesen 1990). There are several examples in megalithic tombs where layers of stones covers the deposition of intentionally cracked pottery and other representatives of the material culture some of which combined with fire. After the transitional stage when the artefacts were transferred from this to a supernatural world, they might have been regarded as dangerous or even evil to the existing society. The simplest way to get rid of this material was to cover it order to prevent contact with members of the society. This practice is not much different from the mortuary practice related to dead members of the society. So in a way the artefacts at Kverrestad and Svartskylle were actually buried.

Berglund, B. E, 1991, The Early Neolithic landscape. Landscape, land-use, and vegetation. In The cultural landscape during 6000 years in southern Sweden - the Ystad Project, edited by B. E. Berglund. Ecological Bulletins 41. Copenhagen: Munksgaard, p. 69–70. BLOCH, M. 1971, Placing the Dead. Tombs, Ancestral Villages, and Kinship Organization in Madagascar. London: Seminar Press. DEHN, T., HANSEN, S., & KAUL, F., 1995, Kong Svends Høj. Restaureringer og undersøgelser på Lolland 1991. Stenaldergrave i Danmark 1. København: Nationalmuseet. DOUGLAS, M., 1966, Purity and danger. An analysis of the concepts of pollution and taboo. London: Routledge. EBBESEN, K., 1980, En Oldtidshøj ved Systofte. Lolland-Falsters Stiftsmuseums årsskrift 1980, p. 13–33. EBBESEN, K., 1990. The Long Dolmen at Grøfte, South-West Zealand. Journal of Danish Archaeology 7, p. 53–69. GÖRANSSON, H., 1996. Om skottskogsbruk och utfodring med kvistar under mellanneolitisk tid och om skogutnyttjandet under mesolitisk tid. In Lövtäkt och stubbskottsbruk. Människans förändring av landskapet – boskapsskötsel och åkerbruk med hjälp av skogen, edited by H. Slotte & H. Göransson. Stockholm: Kungl. Lantbruksakademien, p. 409–452.

CONCLUSION What is expressed at Svartskylle and Kverrestad with massive destruction by fire differs from the destruction of single artefacts by fire evidenced at settlement sites and graves. This type of offering including axes is found throughout the entire Neolithic but on certain occasions the act achieves an imposing effect. The cosmology, which dictated burning, just like that related to wetland offerings, was active throughout most of the Neolithic.

HANSEN, F., 1917, Några enmansgravar från stenålder. Fornvännen 1917, p. 67–89.

The change of colour of the flint artefacts from the natural black or grey to white might be connected to a rite de passage. The colour change could be linked to the process when a human being is cremated. It all ends up with some white bones. In that sense the use of fire on axes could be regarded as the cremation of these flint objects and thereby the mortuary practice as well as rules related to votive depositions might have been interrelated.

KARSTEN, P., 1994, Att kasta yxan i sjön. En studie över rituell tradition och förändring utifrån skånska neolitiska offerfynd. Acta Archaeologica Lundensia, Series in 8o, No. 23. Stockholm: Almqvist & Wiksell International.

Author’s address

LARSSON, L., 1998, Prehistoric Wetland Sites in Sweden. In Hidden Dimensions: The Cultural Significance of Wetland Archaeology, edited by K. Bernick. Vancover: UBCPress, p. 64– 82.

HUNTINGTON, R., & METCALF, P., 1979, Celebration of Death. The anthropology of mortuary ritual. Cambridge: Cambridge University Press. KANNEGAARD NIELSEN, E., & BRINCH PETERSEN, E., 1993, Burials, People and Dogs. In Digging into the Past, editerd by S. Hvass & B. Storgaard. København: Aarhus University Press, p. 76–80.

LARSSON, L., 1989, Brandopfer. Der frühneolithische Fundplatz Svartskylle im südlichen Schonen, Schweden. Acta Archaeologica 59, p. 143-153. LARSSON, L., 1993, The Skateholm Project: Late Mesolithic Coastal Settlement in Southern Sweden. In Case Studies in European Prehistory, edited by P. Bogucki. Ann Arbor: CRC Press, p. 31–62

Lars LARSSON Department of Archaeology and Ancient History 151

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe LARSSON, L., 2000, The passage of axes: fire transformation of flint objects in the Neolithic of southern Sweden. Antiquity 74, No. 285, p. 602–610.

PEQUART, M., & PEQUART, S-J., 1954, Hoïdic. Deuxime stationnécropole du mésolithique cotier Armoricain. Anvers: De Sikkel. PEQUART, M., PEQUART, S-J., BOULE, M., & VALLOIS, H., 1937, Téviec. Station-nécropole mésolithique du Morbihan. Archives de l´Institut de Paléontologie Humaine 18. Paris.

LARSSON, L., 2001, South Scandinavian wetland sites and finds from the Mesolithic and the Neolithic. In The Environmental and Cultural Heritage of Wetlands, edited by B.A. Purdy. Oxford: Oxbow Books, p. 158–171.

PERLÈS, C., 1977, Préhistoire du feu. Paris: Masson. SCHULTING, R.J., 1996, Antlers, Bone Pins and Flintblades: the Mesolithic Cemeteries of Téviec and HoÎdic, Brittany. Antiquity 70, no. 268, p. 335–350.

LARSSON, L., (ed.) 2002, Monumentala gravanläggningar i det äldsta bondesamhället. Institute of Archaeology, University of Lund Report Series 83. Lund: Institute of Archaeology. In press.

SOFFER, O., VANDIVER, P., KLIMA, B., & SVOBODA, J. 1993, The Pyrotechnology of Performance Art: Moravian Venuses and Wolverines. In Before Lascaux. The Complex Record of the Early Upper Paleolithic, edited by H. Knecht, A. Pike-Tay & R. White. Boca Raton: CRC Press, p. 259–275.

LARSSON, M., 1992, The Early and Middle Neolithic Funnel Beaker Culture in the Ystad area (Southern Scania). Economic and social change, 3100–2300 BC. In The Archaeology of the Cultural Landscape. Field Work and Research in a South Swedish Rural Region, editerd by L. Larsson, J. Callmer & B. Stjernquist. Acta Archaeologica Lundensia, series in 4o, No. 19. Stockholm: Almqvist & Wiksell International, p. 17–90.

STRÖMBERG, M., 1968, Der Dolmen Trollasten in St. Köpinge, Schonen. Acta Archaeologica Lundensia, series in 8o, No. 7. Lund: Greerup.

MALMER. M.P., 1962, Jungneolithische Studien. Acta Archaeologica Lundensia, series in 8o, No.2. Lund: Gleerup.

STRÖMBERG, M., 1971. Die Megalithgräber von Hagestad. Zur Problematik von Grabbauten und Grabriten. Acta Archaeologica Lundensia, series in 8o, No. 9. Lund: Gleerup.

MALMER. M.P., 1969, Gropkeramikboplatsen Jonstorp RÄ. Antikvariskt Arkiv 36. Stockholm: Almqvist & Wiksell.

SÄRLVIK, I., & JONSÄTER, M., 1974, Prästgårdskulle i Veddinge. Undersökning av en förhistorisk gravhög. Varbergs Museums årsbok 1974, p. 85–94.

NIELSEN, P. O., 1976, De såkaldte ‘madknive’ af flint. Aarbøger for nordisk Oldkyndighed og Historie 1974, p. 104–118. MALMER. M.P., 1985, Neolithic Hoards from Denmark. In Archaeological Formation Processes. The representativit of archaeological remains from Danish prehistory, edited by K. Kristiansen. København: Arnold Busch, p. 102–109.

VAN GENNEP, A., 1909, Les Rites de Passage. Paris: Emily Nourry.

152

M.O. Baldia: The Oldest Stone Rampart: Enclosures and Megalithic Tombs of the Funnel Beaker Culture...

THE OLDEST STONE RAMPART: ENCLOSURES AND MEGALITHIC TOMBS OF THE FUNNEL BEAKER CULTURE (4100-2800 CAL BC) IN NORTH AND CENTRAL EUROPE Maximilian O. BALDIA

Résumé : La culture des Vases à Entonnoir (TRB) est connue dans différentes parties de dix pays. Des groupes régionaux variés ont existé, mais il y a des similarités identifiables dans la modalité de disposition des fortifications, villages et tombes. Quelques-unes de ces plus anciennes modalités semblent correspondrent aux emplacements fortifiés de hauteurs en République Tchèque. En tout cas, le rempart de pierre de la phase Baalberge (4000/3800 – 3600/3400) appartenant à l’emplacement fortifié de Rmíz est unique en Europe centrale et septentrionale. À l’intérieur de cet emplacement fortifié, des sols d’occupation ou de maisons indiquent différentes périodes d’utilisation. Après Baalberge, le remaniement et la construction d’un nouveau terre-plein à fossé sont évidents. En même temps, un nouveau village TRB de plaine avec les traces de production de cuivre se développe sous Rmíz. D’autres fortifications de pierre apparaissent. À proximité, se trouvent des groupements de tombes avec des douzaines de « long-barrows » et de tumuli, livrant du cuivre, des fragments de textiles et des indications de changements dans les pratiques funéraires à travers le temps. Ce type de fortifications et de tombes apparaît aussi dans le sud de la Scandinavie à une époque où le cuivre devient accessible pour l’essentiel de la culture des Vases en Entonnoir et des chars sont attestés par l’archéologie. Ceci suggère une vaste sphère d’interaction maintenue par des liens de communication. Les tombes mégalithiques belges ne semblent pourtant pas appartenir à ce réseau de communication, suscitant des questionnements sur l’origine de la culture des Vases à Entonnoir et des tombes mégalithiques du début de la culture de Wartberg en France. Abstract: The Funnel Beaker culture (TRB) is found in parts of ten countries. Different regional groups existed, but there are recognizable similarities in site patterning of enclosures, villages, and tombs. Some of the earliest representatives of this pattern seem to be the hilltop enclosures in the Czech Republic. However, the stone Rampart of the Baalberge phase (4000/3800 – 3600/3400) at the Rmíz hilltop enclosure is unique in Central and Northern Europe. Inside the enclosure, living or house floors indicate different periods of occupation. Following Baalberge, remodeling and construction of a new earth wall with ditch is evident. At the same time, a large lowland TRB village with traces of copper production develops below Rmíz. Other stone enclosures appear. Nearby are tomb clusters with dozens of longbarrows and tumuli, yielding copper, textile fragments, and evidence of changes in burial practices though time. This pattern of enclosures and tombs also occurs in Southern Scandinavia at a time when copper becomes accessible throughout most of the TRB and wagons are archaeologically attested. It suggests a broad interaction sphere tied together by communication links. However, the Belgian megalithic tombs appear not to be part of this communication network, raising questions about the traditional derivation of TRB and Early Wartberg culture megalithic tombs from France.

DIVERSITY WITH UNITY: THE CASE OF THE FUNNEL BEAKER (TRB) INTERACTION SPHERE

gatherers. The interaction yielded shared cultural patterns, based on plow agriculture and animal husbandry, punctuated by shifts to greater hunting, gathering and fishing. Eventually the system was sustained through the production and exchange of copper artifacts, the use of wheeled vehicles, construction of enclosed central sites, and monumental tomb architecture. Specific pottery shapes, such as funnel beakers, lugged jars or amphorae, collared flasks, and pedestal bowls, along with certain decorative patterns, also helped to create varying degrees of unity over large distances.

The TRB ranges from southern Norway and Sweden to Austria, and from the Netherlands to Ukraine, covering parts of ten countries. Lasting for more than a millennium, it belongs to the Neolithic in Northern Europe and the Copper Age in Central Europe. Therefore, the genesis of the TRB cannot be understood in traditional terms. Archaeological evidence for the TRB indicates that it developed through intensified interaction between autochthonous people with different socio-economic systems. Its roots must be sought between 4500 – 3800 cal BC, when escalating climatic fluctuations caused increased foraging, coupled with frequent relocation of relatively large, permanent “lake dwellings” in Southern Germany and Northern Switzerland (SCHIBLER et al. 1997, SHENNAN 2001, STRAHM 2001). Simultaneously, the decedents of the Linearbandkeramik farmers, collectively known as “EpiRössen” in West Central Europe and “Epi-Lengyel” in East Central Europe intensified mutually beneficial interaction with their neighbors – the northern Mesolithic hunter-

In spite of the shared pattern, differences remained. Environmental condition in parts of Scandinavia allowed the traditional foraging economy to persist, and yet TRB pottery, stone tools, and even megalithic tombs came into use. Although enclosed sites, constructed of earth and wood developed throughout much of the TRB, stone walls were only built in parts of the traditionally agricultural south. Similarly, earthen long-barrows and tumuli are distributed widely, but megalithic tombs appear primarily in the north and west. Extended burials in supine position occur in southern Scandinavia, Poland, North and Northwestern Germany. In Central Germany and the Czech Republic, flexed 153

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

burials, resting on their side (Hocker), were initially the norm. Later, cremation became dominant, eventually also occurring (sporadically) in the north and west. The pottery styles indicate regional variation and complex diachronic changes. Consequently, the TRB is traditionally divided into five geographic groups: North, West, East, South, and Southeast. Even so, schools of thought exist that prefer to see certain assemblages or phases, such as the South Group’s Baalberge and Bernburg, as discrete cultures. Clearly, the TRB was never a uniform, nor a static culture. In fact, the differences are sufficiently pronounced, that I have proposed to view it not as a single culture, but rather as an interaction sphere (BALDIA 1982a, 1982b). To demonstrate the TRB’s regional variability and overall cohesion, the focus will be on the uniqueness of the TRB South Group. Its relative chronology is presented in light of new excavation results. These results are compared with examples form others regions. In addition, the mechanisms, which unified the interaction sphere, are delineated.

POTTERY TYPOLOGY AND CHRONOLOGY OF THE SOUTHERN TRB The South Group is located primarily in Central Germany, the Czech Republic, and Austria. It belongs to the Copper Age of Central Europe, at a time when the more northerly reaches are transitioning from the Mesolithic to the Neolithic. The research area is centered on the districts of Olomouc and Prostejov in Central Moravia, Czech Republic. It encompasses the eastern part of Drahana Highlands and the adjacent Hana Valley, west of the Morava River. Here the Czech-American Research Program (CARPRO) 1 has surveyed and excavated stone built enclosures and associated burial mounds starting in 1998.2 The precise beginning of the southern TRB is difficult to pin down. Baalberge pottery is viewed as its first manifestation in the Czech Republic and Austria, where it is thought to correlate with the Michelsberg culture (LÜNING’s MK II), but there are subtle differences in the typology and dating (LÜNING 1998, RUTKAY & MAYER 1995, ZÁPOTOCKÝ 1991, 1998, 2000). The Austrian typology is not exactly the same as LICHARDUS A and B used in 1

2

I thank Dr. Miroslav ŠMÍD, Ústav Archeologické Památkové Péče Brno, Pracoviště Prostějov, Prostějov, Czech Republic, for making the research possible; Douglas FRINK (Director, Archaeological Consulting Team, Vermont) for his meticulous attention the microstratigraphy and depositional history in the field and lab; Prof. Dr. John STAECK (Director, College of DuPage Field School) for furthering the CARPRO research aims; OCR Carbon Dating, Inc. Essex Junction, Vt., USA, for soil analysis; Prof. Dr. Kathryn JAKES and Christel BALDIA (Ohio State University) for SEM, EDX and macroscopic textile analysis; Prof. Dr. Curt BECK (Vassar College) for material analysis of pitch; Dr. Sue MILLER, Idaho Falls, ID, for faunal analysis; Rebecca HILL (Soil Systems, Inc. and Arizona State University) for analysis of cremation remains; Prof. Dr. Garth SAMPSON (Anthropology Department) and the Institute for the Study of Earth and Man (Southern Methodis University) for continued support of my research).

Central Germany (RUTKAY & MAYER 1995, p. 129) and the Bohemian classification differs slightly form the Moravian. Also, new Central Moravian finds of MK-like pottery correlate with LÜNING III and IV (ŠMÍD 2001, p. 290). I focus mostly on the Central Moravian typology (ŠMÍD 1994, 1995, 2001), which sees a coexistence of Epi-Lengyel (late AustrianMoravian Painted Ware; Jordanów or Jordansmühle), Baalberge, and MK. In ŠMÍD’s relative chronology, TRB I A has MK attributes. TRB I B does not. TRB II is characterized by a decline in TRB attributes, as Baden-Boleráz traits increase. Radiocarbon dates do not always confirm pottery typology sequences, perhaps because the number and quality of absolute dates is still less than desirable. For example, the oldest Polish TRB dates have been questioned (BAKKER et al. 1969, GRYGIEL & BOGUCKI 1991) and Baalberge da ing is being revised downward.3 In the south, Austrian analysis sums the calibrated dates for Epi-Lenyel (4250 – 3950 cal BC) and Baalberge (3780 – 3350 cal BC; STADLER 1995). Baalberge has a similar range in Germany (MÜLLER 2001a, 2001b; RAETZEL-FABIAN 2000). This leaves a ca. 200year gap between Epi-Lengyel and Baalberge in the south. Luckily, in North Germany, this gap is filled by new dates, placing the transition from Ertebølle to TRB around 4100 cal BC (LÜBKE 2000). Likewise, the formative Central German or Central Elbe/Saale region’s TRB 1 with early funnel beakers, simple amphorae and evidence of Jordanów influence is dated 4100 – 3800 cal BC (MÜLLER 2001a, p. 77; 2001b, p.83-96). Since ŠMÍD also correlates TRB I with Jordanów, a similar early Moravian stage can be postulated. Furthermore, my standard calibration of twenty Baalberge C14 dates does not preclude a beginning around 4000/3900 cal BC. This falls in line with MK III, starting ca. 4000 cal BC and MK IV starting perhaps at 3900 cal BC (RAETZEL-FABIAN personal communication. 2002). The end of Baalberge (TRB I B) and the beginning of TRB II is perhaps even more problematic. My calibration suggests an end at 3600/3500 cal BC, although an end around 3400 is possible. This leaves an imprecise starting date for TRB II A. Calibration of the five Baden-Boleráz dates from Hlinsko, Moravia (FORENBAHER 1993, p. 238-239), yield ca. 3600 - 3400 cal BC, failing to improve precision. The calibration results overlap with MÜLLER’s Central German TRB III (3500 – 3350 cal BC), consisting of Baalberge and Salzmünde A. Its end coinsides with the end of Hlinsko’s dates and may mark the beginning of TRB II B. Therefore, TRB II B could coincide with the Central German TRB IV (3350 – 3100 cal BC; Salzmünde A and B). This may receive support in my calibrations, which include Austria’s Classic Baden (Ossarn I and II) dates and the early Baden-like pottery of Bronocice IV, Poland (KRUK et al. 1996). However, the Bronocice dates could have a systematically late bias (BAKKER et al. 1999) and a direct comparison of the pottery is needed. The demise of the TRB in Moravia must be sought 3

Since 2000, CARPRO has been joined by the College of DuPage Field School, Prof. J. STAECK, Director.

154

The oldest dated Baalberge amphora from Grube-Rosenhof, North Germany, has been re-dated to 5000±60 BP, based on the pitch that glues the pot together (MÜLLER 2001b, p. 86 Footnote 14). OxCal v3.5 calibration: 22.0% = 3940 – 3870 cal BC, 46.2% = 3810 – 3700 cal BC).

M.O. Baldia: The Oldest Stone Rampart: Enclosures and Megalithic Tombs of the Funnel Beaker Culture...

around 3200/3100 cal BC, because TRB II B is followed by the Jevišovice culture in Moravia, Řivnač in Bohemia, Cham in Bavaria, and Bernburg in Central Germany. In Central Germany this is MÜLLER’s TRB V (3100 – 2750 cal BC; Salzmünde C, Bernburg, Globular Amphora culture, etc.).

tentative results indicate that clearing and habitation of Rmíz began sometime between 4500 and 4130 BC. This is congruent with occasional finds of Epi-Lenyel pottery. More intense occupation probably existed ca. 4100 – 3990 BC. It may correspond with the early phase of monumental enclosure construction of MK I/II (ca. 4200 – 4000 cal BC, RAETZEL-FABIAN 1999, p. 93-95). However, limited evidence could also place the Rampart III construction around 3870 BC (early MK III). Either way, it seems, that after a ca. 100-year hiatus reoccupation occurs (ca. 3870 – 3670 BC), roughly corresponding to the earliest C14 date from the geometric enclosure of Makotřasy, Bohemia (PLESLOVÁŠTIKOVÁ 1985, p. 183). Around 3610 BC a house or living surface near Rampart III was still associated with Baalberge pottery. By contrast, the later, stratigraphically higher house floor or living surface dates to ca. 3250 BC. This and other evidence implies a lengthy hiatus before TRB II at Rmíz. ŠMÍD concluded that Rampart II was constructed at this time. Limited stratigraphic observation suggests that the inner of the two walls of Rampart III and the posts (palisade) were probably built around 3350 BC. The TRB II occupation at Rmíz (ca. 3350/3300 – 3100 BC) may thus equate with the Central German TRB IV (Salzmünde A and B).

In spite of numerous attempts, precise correlation with the TRB of South Scandinavia and North Germany remains problematic. It may be that ŠMÍD’s TRB I A (earliest Baalberge with MK and Epi-Lengyel traits) in the south is roughly equivalent to the transition between the Mesolithic and the beginning of the Nordic Early Neolithic I (ca. 4100/ 3900 – 3700 cal BC). TRB I B (3800/3700 – 3600/3500 cal BC) may coincide with the later part of the Nordic Early Neolithic I. TRB II A could theoretically correlate with Early Neolithic II (3500 – 3300 cal BC). TRB II B may coincide with the Nordic Middle Neolithic A I (3300 – 3100 cal BC), while Nordic TRB A II – IV (3100 – 2800 cal BC) may equate with the Jevišovice culture, Řivnač, Cham, Bernburg, Salzmünde C, etc. Given this less than optimal chronological situation of the TRB South Group, CARPRO sees the establishment of a solid absolute chronology as one of its primary objectives. Improvements are sought through a thorough understanding of the depositional soil record, including microstratigraphic association of artifacts and dates. The first results of this Czech-American venture are producing tantalizing new insights about the settlement history, construction phases of enclosures, changes in burial practices, regional and interregional communication, and possible linkages to climate change.

Although microstratigraphic correlation of the pottery at Rmíz is still under study, it is possible to draw some early conclusions. At the lower levels, there is polished pottery, including black “graphite” slipped, variously colored, and (Epi-Lengyel?) unpainted pottery. Overlapping at a higher level are (possibly unburnished) “graphite” slip, sliproughened, and spicula-slipped sherds. The last pottery is rare (10%). Based on two OCR test columns, it starts at ca. 3900 BC and fades out ca. 3650 BC. ŠMÍD could classify at least one of the sherds as TRB I B. There are also numerous trampled house or living surfaces, waddle and daub, spindle whorls, well preserved human and animal bones (some exposed to fire), and a small quantity of lithics made of Polish Œwieciechów, Banded, and Chocolate flint, as well as querns and quern stones. Most curious are three chunks of a coallike substance from widely separated test pits near Rampart III and IV. Preliminary analysis the 0.4 g to 35.8 g chunks shows them to be resin from a deciduous tree (birch?). Inside Rampart IV the first copper object (wire) from Rmíz was recovered in association with TRB II pottery.

NEW INSIGHTS FROM MORAVIAN ENCLOSURES AND THEIR STONE RAMPARTS CARPRO has tested and mapped parts of the enclosure of Rmíz u Lašcov, the small enclosure of Hřad u Bílovice, and the newly discovered tomb cluster near Ludeřov. Previous excavations (ŠMÍD 1994, 1995) concluded that the 17.5 ha site of Rmíz is the oldest and largest enclosure in the immediate research area. The multiphase site consists of four ramparts. Rampart I (the outermost enclosure) is a palisade. Rampart II is a later earthen enclosure. Rampart III is a complex structure. CARPRO research shows that Rampart III consists of two walls, 50 cm apart, plus several outside ditches. Two post molds, implying an inner palisade, were discovered inside Rampart III, near the postulated causeway. According to ŠMÍD Rampart III probably coexisted with Rampart I, while Rampart IV, enclosing the highest point of Rmíz Hill, is a Hallstatt fortification.

Other nearby sites include Laškov, a large lowland TRB village. This TRB II A site, at the foot of Rmíz Hill, is a horizontal distance of only 400 m from the enclosure. Here traces of copper production, including slag, were found (ŠMÍD 1998). Some 7 km south-southwest of Rmíz is Hřad u Bílovice. This 1.5 ha hilltop enclosure with stone wall and deep ditches, was excavated in 2001. Before this excavation, research (e.g. GOTTWALD 1937) indicated that it was also built during TRB II A. Preliminary analysis of the wall indicates three stone wall elements. The main structure appears to have had a core wall (one course thick), which may have been supported by logs. There is considerable variation in depositional events, perhaps suggesting that the Rampart was built and enlarged at different times, but this needs verification. The site harbors much stone tool material, functioning as a quarry and manufacturing site of querns from

So far, CARPRO has focused on the site history near Rampart III as well as the center of the site (inside Rampart IV), using careful, time-consuming excavations with meticulous attention to stratigraphy, rather than the customary opening of large areas. While funding is sought for costly C14 dating, the newly developed Oxidizable Carbon Ratio (OCR) analysis of the soils is employed (FRINK & DORN 2002). Based on a composite of four test unit soil profiles, 155

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

the quartzite bedrock at an early stage. The querns recovered at Rmíz are of similar material as that identified at Hřad u Bílovice.

recently discovered burial cluster near Ludeřov is less than 2 km southeast of Rmíz. Mound 1 was excavated in 2000, providing more insight into this cremation ritual. It contained three separate urn graves, each with intact content of at least one individual. The bodies were cremated at temperatures above 700°- 800° C shortly after death. While in the fire, they may have been maneuvered to insure thorough incineration. Afterwards the remains were probably crushed into small fragments. Only a fraction of the individual’s remains was placed into the urn. Grave 3 had the largest urn, containing the greatest amount of bone. Moreover, it included an animal long bone, burnt in defleshed or dry condition, suggesting feasting prior to cremation. From this and other excavations, we also know that urns are associated with finely spun linen textiles and possible offerings of grain (C. BALDIA this volume), pendants, ground stone axes, and additional pots, including a collared flask (ŠMÍD 1990).

Although conclusions are very tentative, the discovery of recurring cycles of abandonment and reoccupation appear to coincide with climatic fluctuations and may indicate the use of hilltop enclosures during optimum conditions. Similar observations have been reported for the MK enclosure of Bruchsal “Aue 1” in Germany, where over two meters of erosion ended the early MK occupation (BEHRENDS 1998, p. 117-119). This event (STEPPAN 1998, p. 148 Fig. 3) seems to coincide with the first abandonment of Rmíz. STEPPAN (ibid.) also indicates a reoccupation at Bruchsal (“Aue 2”) during MK III/IV, roughly corresponding to the second occupation of Rmíz. At Rmíz this is followed by another settlement interruption, similar to that in Swiss lake dwellings. There it has been is associated with the rise of lake levels due to a climatic downturn, possibly caused by two volcanic eruptions (SCHIBLER et al. 1977, p. 337-342). Brief settlement recurs at Rmíz around 3610 BC. The later TRB II reoccupation may have parallels with nearby Hlinsko, where a hiatus was observed during Baden III, a reoccupation during Baden IVa and a final abandonment after Baden IVb/Bošáka 1a (Pavelčík 2001, p. 73). Therefore, it appears that the demonstrated recurring abandonment of wetland sites in South Germany and North Switzerland also occurred in upland villages of adjacent regions. However, regional and interregional parallels between settlement phases will require further specialized pedogenic analysis of non-wetland site formation.

In the final stage of monumental burial construction (TRB II B), stone disappears from the mounds entirely, except for the occasional smattering used to cover some individual urn burials. Cremations remain the norm. Still undated, it is likely that Monumental mound construction in Moravia ends by 3200 cal BC, coinciding with the end of megalithic tomb construction at Sarup, Denmark (N. ANDERSEN 1997, p. 99) and probably the remainder of the TRB. Comparison with other TRB areas leads to the conclusion that monumental long-mounds occur in most places, but megalithic tombs do not. Moravia must have had thousands of mounds, but used only sub-megalithic stones for TRB I and II A tombs. Simultaneously, Rmíz Rampart III incorporated some megalithic-like stones. Thus, the concept of monumental architecture was shared throughout the TRB, but not the specific architectural expression. This implies a relatively loosely organized communication network with evolving centralization at the local or regional level.

NEW INSIGHTS FROM MORAVIAN MONUMENTAL BURIAL MOUND CLUSTERS There are well-preserved burial mounds in the research area, perhaps numbering in the hundreds. Křemela I is 800 m and Křemela II is merely 200 m north of Rampart I at Rmíz. The mounds include 2 m high earthen long-barrows. Recently ŠMÍD discovered two more previously unknown burial clusters at greater distance from Rmíz. The earliest longmounds (TRB I B) contain small Baalberge stone chambers (cists) with flexed (hocker) inhumations, suggesting a link with the Elbe River in Bohemia and Central Germany to ŠMÍD (1990, 1992). The tiny chambers, made of stone slabs, contain single individuals, resting on their side in a flexed position.

THE LOCAL SETTLEMENT PATTERN AND COMMUNICATION Understanding the degree of centralization requires an understanding of the time and space relationship of enclosures, tombs, and other sites. During TRB I there are five village sites 4 km or more from the central site of Rmíz, but there is only one cluster of burial mounds. ŠMÍD dates these first Baalberge long-mounds, containing stone cists, to TRB I B. The tombs are located at Slatinki, 5 km south southeast of Rmíz. This period must relate to the Central German Phase II (3800-3500 cal BC), when “richly” furnished burials and large enclosures are in use (MÜLLER 2001a, 2001b). During TRB II, when there is renewed building activity at Rmíz, Slatinki appears to become a walled, double ditched hilltop enclosure, joining Hřad u Bílovice and two or three other sites as candidates for central sites in the immediate research area. The number of apparently unenclosed lowland sites increases to seven. Concomitantly, there is a vast increase of burial mound clusters. Several of them appear to be directly associated with enclosed sites.

The second stage of burial construction is associated with ŠMÍD’s TRB II A. The mound’s stone cover was largely replaced with a sub-megalithic, slightly trapezoidal drywall structure that is open at one end. Within its confines, cremation urns take the place of stone cists. It remains to be seen if this switch is a wider trend, related to the development of Central German masonry chambers or the trapezoidal Wartberg culture tombs (RAETZEL-FABIAN 2000, p. 215, Fig. 139). The same holds true for the increase of cremation in other parts of the late TRB and Wartberg culture. After all, the change could signal the advent of a more widespread, common ritual development than previously anticipated. The 156

M.O. Baldia: The Oldest Stone Rampart: Enclosures and Megalithic Tombs of the Funnel Beaker Culture...

Two alternative models of communication could explain this changing pattern. The TRB intrusion model portrays the initial appearance of Rmíz as an early, fortified central site. If so, it could have harbored a relatively large population and all its social activities. This Czech and Polish model is attractive, if one expects a population movement of Baalberge from the Elbe into Moravia, still occupied by a local EpiLengyel population. Indeed, Moravian excavations suggest coexistence of the different pottery styles. Alternatively, Rmíz functioned as the focal point of a disperse population. In this case, the Danish model, exemplified by the Sarup enclosure (e.g. N. Andersen 1997) should apply. It suggests that the first central sites were constructed by a dispersed community to facilitate social interaction. The primary purpose was the ritual disposal of the dead. The corpses may have been kept at the enclosures to be interred later, in part, in the numerous monumental tombs nearby. Later the centralized ritual aspect of the enclosure was replaced a central village, replacing the dispersed settlement pattern.

interaction sphere during periods of intensified interaction. These periods seemingly correlate with the Rmíz settlement phases, climatic conditions, and the exchange of special artifacts with the Western Baltic region. The earliest interregional communication is documented through the export of Central European amphibolite shafthole axes to the Mesolithic Ertebølle found as far north as Sweden. At Krusminde, Denmark, an ax was found in Middle Ertebølle context (ca. 4800-4300 cal. BC; PEDERSEN et al. 1997:201205). Amphibolite is reported from the Moravian/Slovakian border. Around the same time, the Nordic Copper Import Period 1 of the southern Baltic region exhibits artifacts from non-arsenical copper sources. These too are likely to be from West Slovakia (KLASSEN 1997, KLASSEN & PERNITCKA 1998). However, a slightly later date (42003800 cal. BC) has also been considered, implying Jordanów copper production (MÜLLER 2001:410-417). Either way, Central Moravia would have been part of the communication network.

At Rmíz there are certainly human bones, some scorched by fire, just inside the western end of Rampart III. However, the earliest mounds (TRB I B) are over 7 km away. Although Rmíz, like Sarup, was constructed in several phases, its purpose may have been exactly the reverse of Sarup’s. Hypothetically speaking, Rmíz may first have been more like a fortified village where all social, economic, and religious activities took place. Local communication would have been easily facilitated. But during TRB II, a segment of Rmíz could have become a Sarup-like ritual center, where cremation of the deceased bodies took place, before parts were placed in an urn for burial in one of the many nearby long-mounds. Local communication would have required a more structured organization, i.e. greater centralization, which may have been housed at the highest portion of Rmíz. Only careful excavation, analysis, and dating can ultimately provide certainty for this scenario. Nevertheless, it provides a basis for understanding the regional and interregional communication pattern.

During the second Scandinavian Import Phase, starting at 3800/3700 cal BC and ending with the third phase around 3300 cal BC, arsenical “Mondsee copper” was placed into Scandinavian non-megalithic tombs and a copper hoard may be associated with the Sarup-type enclosure of Årupgård. The Nordic phases overlap with copper production in Moravia, where copper was placed into nearby submegalithic and non-megalithic long-mounds. The Hlinsko hilltop enclosure contained a copper hoard. Rmíz itself yielded a copper wire dating to TRB II and there was copper smelting activity at Laškov. Concurrently communication with the Austrian Mondsee culture is likely, since its occasionally incised pottery is akin to the Retz style found in Baalberge context in Austria and Moravia. Animal figurines plus knob- and round-butted axes occurred at Hlinsko, Mondsee sites, and Strachów, Poland. This indicates a wide spread communication network linked with Scandinavia. Nevertheless, the metallurgical footprint of the Rmíz area copper (ŠMÍD 1998) is not identical with the Mondsee source (Pernicka personal communication). Furthermore, KALSSEN’s dating for the later Nordic import periods, as well as the dates from Hlinsko, are not entirely reconcilable with the copper import “crisis” in South and Central Germany (ca. 3600/3500 – 3350 cal. BC), while the possible temporary abandonment of Rmíz appears to coincide with this “crisis”. It is also difficult to explain the lack of copper in Scandinavia after 3300 cal BC, since the German copper “crisis” is said to end before that time. This makes it currently difficult to correlate the interaction pattern satisfactorily.

REGIONAL AND INTERREGIONAL INTERACTION Theoretically, regional and interregional communication is reflected in the distribution pattern of sites. The regional site patterning reveals that Rmíz is the northwestern-most link in a long chain of a dozen or so similar sites. They are distributed on the western and northern edge of the known TRB settlement area in Moravia (PODBORSKÝ et al. 1993 Fig. 4). The topography suggests that they functioned as central sites, usually located on promontories, perhaps to control the communication lines. Rmiz is strategically located within sight of the Moravian Gate. Hlinkso is positioned right in this mountain gap, which has traditionally been viewed as part of the elusive “Amber Road,” connecting Southeastern Europe with Scandinavia. Both Rmíz and Hlinsko are close to the source of the northwest flowing Elbe, the north flowing Odra, and Wisła (Vistula), as well as the south flowing Morava River, a tributary to the Danube. Rmíz (and later Hlinksko as well) may have been corner a stone of the communication network, facilitating the existence of the TRB

In Central Germany the copper “crisis” is followed by a renewed rise in construction of small enclosures and the use of local copper sources (MÜLLER 2001 ibid.). If this is concomitant with the proliferation of smaller enclosures, such as Hřad u Bílovice, and “richly” furnished burial mounds in Central Moravia, we may expect a grater local self-reliance, including the local exploitation and manufacture of copper. This occurs twhen the climate seems became more stable. However, another factor may also be at work, changing the 157

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

means of communication and facilitating a qualitatively different kind of interaction.

alignment, suggesting a river crossing, was (inadvertently?) illustrated near the TRB village of Sachsenwaldau by SCHIRREN (1996). From here, the tomb alignment can be followed across the Elbe via the Lüneburg Heath and Haldesleben, where in 1938 FISCHER (1956, p. 84) first proposed a connection with Neolithic roads. The communication line continued southeast, probably aiming at the massive TRB enclosure of Halle-Dölauer Heide (BALDIA 1995 Fig 17.33). Theoretically, this line can be extended to the Elbe gorge, leading through the mountains between Germany and the Czech Republic. Near the border, an ancient trade route connected Prina with Lovocice (DRESLEROVÁ & SALAČ 2000). Unfortunately, the connection cannot be demonstrated by tomb alignments. Nonetheless, these and other examples may indicate overland routes combined with waterways, connecting central sites in Southern Scandinavia, Germany, Poland, and Netherland. A similar situation may have existed in Moravia, but the tight tomb clusters do not lend themselves readily to this analysis. Even so, the topography and the site distribution patterns suggest the possibility of prehistoric ways. Obviously, much more research is needed to provide tangible evidence of Neolithic/Copper Age ways, especially in Moravia.

Technology to facilitate communication started with corduroy roads (ca. 4600 and 4000 cal BC) in the Große Moor of the Dümmer Lake in Germany (METZLER 1997). Direct evidence of TRB wheeled vehicles begins at 3630 – 3370 cal BC in Poland and North Germany (e.g. BAKKER et al. 1999, MILISAUSKAS 1978, VOSTEEN 1999, ZICH 1992, 1994, 1995). This may explain why a 1.6 m wide stone pavement-like way was constructed at Hlinsko, 45 km to the east of Rmíz at this time (PODBORSKÝ et al. 1993, p. 185). The way led into the center of its enclosure. Simultaneously, the first clay cups, representing four-wheeled wagons, occur in the Baden-Boleráz phase of Lower Austria and adjacent regions (RUTKAY & MAYER 1995, p. 156). The wagon models increase in frequency during the remainder of Baden and exist in the TRB’s Bernburg phase of Central Germany (ANDERSEN 1997, p. 336 Footnote 231). Numerous wooden wheels are known from Corded Ware sites. Unfortunately, the evidence of ways used by wagons is largely circumstantial. 4 For the TRB prehistoric ways may be associated with megalithic tombs, as proposed by BAKKER (1976, 1980, 1991). JAGER (1985) supported this, pointing also to the juxtaposition of a prehistoric way with the TRB enclosure of Anloo, Netherland. Similarly, RAETZELFABIAN (1999, p. 97-101) has argued for a connection between MK enclosures and ways, based on the proximity of the sites to medieval roads. Traces of a way were discovered at the MK hilltop enclosure of Untergrombach, Germany (BEHRENDS 1998, p. 116, Fig. 2). RAETZELFABIAN (2000 Fig. 142) also projects a way between the Late MK enclosure, the nearby Early Wartberg Gallery-grave II, and several later tumuli at Calden, Germany, showing a connection between enclosures and tombs similar to that at Anloo. The proposition receives additional support by the fact that two Wartberg gallery-graves (Lohne-Züschen 1 and Warburg I) exhibit symbols of several single-axle, wheeled ox-drawn carts.

LIMITS OF INTERACTION Even if one rejects the notion of actual Neolithic roads, the distribution of nearly 5000 long-mounds and megalithic tombs from Poland to Netherland and from Oslo Bay to Central Germany shows a complex network of east-west and north-south alignments indicative of communication lines (BALDIA 1995 Fig. 16.2, 16.3). The huge number of Moravian long-barrows and enclosed sites must now be added to round out this distribution pattern. However, the geographic and chronological extent of this interaction sphere is limited, in spite of the lack of strict boundaries. Due to space requirements, I can only focus on the western limit, confining myself to the TRB and Wartberg culture tombs and their relationship with those west of the Rhein (Rhine River). Nonetheless, this issue is very important, because it bears on the traditional conjecture that the TRB and Wartberg culture tombs are derived from France via Belgium.

Spatial analysis of ca. 5000 megalithic tombs supports the idea that they were juxtaposed with prehistoric communication lines. The tombs form dendritic linear distributions and cluster near enclosed sites, which are usually located on hilltops and/or near rivers (BALDIA 1994, 1995). For example, the megalithic tombs near the TRB enclosure of Büdelsdorf, North Germany, form an intersecting alignment, leading past the enclosure on two sides. One line continues west along the Schlei River, perhaps aiming at the Dieksknöll enclosure near Albersdorf. The other suggests a river crossing southeast towards the Ostenfeld tombs, possibly meeting up with the minimally 4 km long tomb alignment at Flintbek. Here the oldest recorded wagon tracks, buried beneath a long-barrow (ZICH 1992, 1994, 1995), must date to ca. 3500/3400 cal BC. Near Hamburg another tomb 4

Long-mounds and megalithic tombs are rare between France and the Rhein, making direct geographic and chronological connections problematic at best (BALDIA 1994, 1995, RAETZEL-FABIAN 2000, p. 142-144, 199-201). Potentially early tombs include possible Belgian MK long-mounds originally described by De Laet (e.g. BAKKER et al. 1969 Fig. 13); the Belgian abris used as MK sepulchres (TOUSSAINT 1997b); a megalithic-like reconstructed chamber at the Langer Stein menhir at Einselthum; Germany (BOSINSKI 1961, p. 179-184); the stone cists of Kruft near the MK enclosure of Urmitz, containing possible MK pottery (RAETZEL-FABIAN 2000 No. 133, 135). Given the emphasis on causewayed enclosures in the MK, tomb information is insufficient to confirm a communication link between the French/Belgian MK on the one hand and Baalberge on the other. Following MK (3600/3500 cal BC), the Belgian sepulchres, including rare megalithic tombs with

Although the earliest recognition of a juxtaposition of ways and tombs occurred in Denmark 100 years ago and spawned the notion of the “Amber Road”, the idea is now rejected out of hand there and in North Germany. Remaining highly controversial, the notion is nonetheless accepted in Netherland and adjacent parts of Germany.

158

M.O. Baldia: The Oldest Stone Rampart: Enclosures and Megalithic Tombs of the Funnel Beaker Culture...

a largely destroyed inventory, seem to be associated with Seine-Oise-Marne (SOM) artifacts, although Corded Ware, Bell Beaker and other associations are also observed (DE LAET 1981, HUBERT and HUYSECOM 1980a – 1984b, HUYSECOM 1981a – 1982b, TOUSSAINT 1997a, 1997b, FRÉUTTE et al. 2001). Not surprisingly, the available absolute dates imply that most tombs were constructed after 3000 cal BC. (TOUSSAINT 1997b). In nearby Limburg, Netherland, the tomb of Stein probably dates no earlier than to 3470 – 3370 cal BC, but its cultural affiliation is difficult to assess. In adjacent Germany the peculiarly constructed Schankweiler stone cist, as well as those of Kruft, Mayen, and Urmitz contain ill defined pottery, providing only the faintest hint at an association with Late Wartberg, Globular Amphora, and/or Corded Ware. Furthermore, the TRB and Wartberg tomb distribution does not indicate any direct and sustained communication lines between the builders of the French and Belgian tombs, and those in Germany. Only after about 3000 cal BC is a reorientation or widening of the communication network plausible, but by that time TRB megalithic tomb construction had ceased.

Baden interaction via Bulgaria and northern Greece to the Aegean (HÄUSLER 1992). Indeed, Baden handled jugs and cups may indicate a formal relationship with Coţofeni, Ezero and Cernavoda pottery. Likewise, Baden pedestal bowls may have spawned similar TRB pottery as far north as Scandinavia (MADSEN 1976). However, the distribution and dating of monumental burial architecture makes it clear that the Belgian megalithic tombs are not directly linked with those of the TRB and the Wartberg culture. This means that intensified interaction with areas west of the Rhein River probably developed only after ca. 3000 cal. BC and French tombs are unlikely to be the precursors to of the TRB megalithic tombs and the gallery-graves of the Wartberg culture.

Author’s address Maximilian O. BALDIA, Director Czech/American Research Program (CARPRO) 3616 Dinsmore Castle Dr. Columbus, OH 43221-4410, USA [email protected] Research Associate Institute for the Study of Earth and Man Southern Methodist University 3225 Daniel Avenue Dallas, Texas 75275-0274, USA

CONCLUSION Excavations in Moravia have not only yielded the oldest, very complex TRB stone Rampart at the hilltop enclosure of Rmíz, they also show that it leads to a local tradition of stone wall enclosures. Rmíz contained numerous house or living surfaces. Soil analysis indicates that there were several occupation sequences, possibly interrupted during climatic downturns, when the surviving population moved elsewhere. The enclosures are part of a settlement pattern that includes large clusters of nearby burial mounds. The long-barrows indicate changes in architecture and ritual. Individual inhumation in stone cists is supplanted by individual cremation in urns. Only a fraction of the human remains is found in the urns. Animals, grain, copper, and finely spun linen were part the burial ritual.

Bibliography BAKKER, J. A., 1976, On the Possibility of Reconstructing Roads from the TRB Period. Berichten van de Rijksdienst voor het Oudheidkundig Bodemonderzoek 26, p. 63-91. BAKKER, J. A., 1980, Book review: E. Jørgensen. HagebrogårdVroue-Koldkur: Neolithische Gräberfelder aus Nordwest-Jütland. 1977, In Helenium XX/1, p. 80-82. BAKKER, J. A., 1991, Prehistoric Long-Distance Roads in NorthWest Europe. In Die Kupferzeit als historische Epoche, edited by J. LICHARDUS. Bonn: Habelt, p. 505-528.

Comparison of the research results with other regions shows that the TRB was not a static culture with easily definable chronological and geographic boundaries, but rather an interaction sphere, driven by different local traditions. It was held together by a dynamic communication network in which enclosures functioned as the communication nodes and the long-barrows and megalithic tombs were largely juxtaposed with the lines of communication. The system afforded contact with neighboring cultures, such as Michelsberg and Wartberg. It also provided sufficient freedom for different populations to cope with their specific local and regional conditions by developing their own socio-economic and political subsystems. This fostered innovations and resulted in the expansion of the communication network, especially during favorable climatic conditions. Thus, it is tempting to see the expansion of the TRB interaction sphere as resulting from the documented use of wheeled vehicles and the exchange of copper. This may have started during TRB II/ Baden-Boleráz (3600/3400 call BC), if not in an earlier Michelsberg/Baalberge context. It may have been part of a wider trend, because wheeled vehicles probably increased

BAKKER, J. A., KRUK, J., LANTING, A. E., & MILISAUSKAS, S., 1999, The Earliest Evidence of Wheeled Vehicles in Europe and the Near East. Antiquity 73, p. 778-790. BAKKER, J. A., VOGEL, J. C., & WIŚLAŃSKI, T., 1969, TRB and other C-14 Dates from Poland (c.4350-1350BC and 800900AD). Helenium IX/3, p. 209-238. BALDIA, M. O., 1982a, Environment and Cultural Evolution: Towards an Interdisciplinary Theory of Human Behavior. In Environment and Evolution, edited by V. J. A. Novak & J. Mlikovsky. Praha: Czechoslovak Academy of Sciences, p. 829844. BALDIA, M. O., 1982b, Archaeology, Evolution, Culture, and Communication: A Biologically Based Explanation of Cultural Evolution. In Man and his Origins, edited by J. Jelinek. Brno: Anthropos, p. 5-12. BALDIA, M. O., 1994, Megalithic Tombs and Interregional Communication. Paper presented at the international symposium: Megaliths and Social Geography, 13-17 May 1994, Falköping, Sweden. (Updated paper published by the Comparative Archaeology WEB, http://www.comp-archaeology.org/ Falbytxt.htm) 159

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe HUBERT, F. & HUYSECOM, É., 1984b, L’allée couverte “Wéris I:” Fouilles de contrôle de 1979 a 1983. Notae Praehistoricae 4, p. 125-130.

BALDIA, M. O., 1995, A Spatial Analysis of Megalithic Tombs. Vol. 1-2. Ph. D. Dissertation. Southern Methodist University. (Updated excerpts of the dissertation are published by the Comparative Archaeology WEB, http://www.comp-archaeology. org/010_Intro_Dis_SPATIAL_ANALYSIS_MEGT.htm).

HUYSECOM, É., 1981a, Les Allées couvertes de Wéris. Buletin trimestriel d l’Institut Archéologique du Luxembourg, 57, p. 63131.

BEHRENDS, R.-H., 1998, Neue Forschungen zur Michelsberger Kultur im Kraichgau. In Die Michelsberger Kultur und ihre Randgebiete: Probleme der Entstehung, Chronologie und des Siedlungswesens, edited by J. Biel, H. Schlichtherle, M. Strobel & A. Zee. Stuttgart: Theiss, p. 115-119.

HUYSECOM, É., 1981b, À Propos de Fragments de gobelets “AOO” Exumés des allées couvertes de Wéris (Luxembourg). Helenium 21, p. 5-60. HUYSECOM, É., 1982a, Les sépultures mégalithiques en Belgique: Inventaire et essai de synthèse. Bulletin de Societe Royale Belge d’Anthropologie et de Prehistoire 93, p. 63-85.

BOSINSKI, G., 1961, „Der Lange Stein“ bei Einselthum (Pfalz): Ein Menhir der Rössener Kultur?“ Germania 39, 1/2, p. 171185.

HUYSECOM, É., 1982b, Le monument mégalithique de Laviô (Province de Luxembourg). Bulletin de la Société Préhistorique Française 79/9, p. 279-283.

DE LAET, S., 1981, Megalithic Graves in Belgium – A Status Quaestionis. In Antiquity of Man: Essays in honour of Glyn Daniel, edited by J. D. Evans, B. Cunliffe and C. Renfrew. London: Thames and Hudson, p. 155-168.

HUYSECOM, É., 1982c, Le dolmen de Jambes. Annales de la Société Archéologique de Namur 62, p. 274-280.

DRESLEROVÁ, D. & SALAČ, V., 2000, Elbe-Archäologie in Böhmen. In Schutz des Kulturerbes unter Wasser, Veränderungen europäischer Lebenskultur durch Fluß- und Seehandel, Beiträge zum Internationalen Kongreß für Unterwasserarchäologie (IKUWA ´99), Lübstorf, Germany: Beiträge zur Ur- und Frühgeschichte Meklenburg-Vorpommerns 35, p. 351-355.

JAGER, S. W., 1985, A prehistoric Route and Ancient Cart-Tracks in the Gemeente Anloo (Province Derenthe). Paleohistoria 27, p. 185-202. KLASSEN, L., 1997, Die Kupferfunde der Nordgruppe der Trichterbecherkultur. Archaeologische Informationen 20/1, p. 189-193.

FISCHER, U., 1956, Die Gräber der Steinzeit im Saalegebiet: Studien über Neolithische und Frühbronzezeitliche Grab- und Bestattungsformen in Sachsen-Thüringen. Berlin: De Gruyer & Co.

KLASSEN, L. & PERNICKA E., 1998, Eine kreuzschneidige Axthacke aus Südskandinavien: Ein Beispiel für die Anwendungsmöglichkeiten der Stuttgarter Analysendatenbank. Archäologisches Korresbondenzblatt 28/1, p. 35-45.

FRÉBUTTE, C., M., TOUSSAINT, S. PIRSON, J.-N. ANSLIJN, & F. HUBERT, 2001. Durby/Wéris: fouilles à l’allée couverte nord ou « Wéris ». Chronique de l’Archéologie wallone 9, p. 158161.

KRUK, J., MILISAUSKAS, S., ALEXANDROWICZ, W., & ŚNIESZKO, Z., 1996, Osadnictwo i zmiany srodowiska naturalego wyżyn lessowych: Studium archeologiczne i paleogeograficzne nad neolitem w dorzeczu Nidzicy. Kraków: Instytut Archeologii i Etnologii Polskiej Akademii Nauk.

FRINK, D. S., & DORN, R. I., 2002, Beyond Taphonomy: Pedogenic Transformations of the Archaeological Record in Monumental Earthworks. Journal of the Arizona-Nevada Academy of Science 33/3, p. 24-43.

LÜBKE, H., 2000, Timmendorf-Nordmole und Jäckelberg-Nord. Erste Untersuchungsergebnisse zu submarinen Siedlungsplätzen der endmesolithischen Ertebølle-Kultur in der Wismar-Bucht, Mecklenburg-Vorpommern. Nachrichtenblatt Arbeitskreis Unterwasserarchäologie 7, (http://www.jkoeninger.de/seite5.htm accessed 23 December 2001 and http://www.abc.se/~pa/publ/ wismarbu.htm accessed 24 December 2001).

FORENBAHER, S., 1993, Radiocarbon Dates and Absolute Chronology of the Central European Early Bronze Age. Antiquity 67, p. 218-20, 235-256. GOTTWALD, A., 1937, Hřad u Bílovic a nordická sídlište na výšinách na Prostějovsku. Časopis vlasteneckého muzejního spolku v Olomouci, p. 187-188 (3-12).

LÜNING, J., 1998, Betrachtungen über die Michelsberg Kultur. In Die Michelsberger Kultur und ihre Randgebiete: Probleme der Entstehung, Chronologie und des Siedlungswesens, edited by J. Biel, H. Schlichtherle, M. Strobel & A. Zee. Stuttgart: Theiss, p. 277-290.

GRYGIEL, R. & BOGUCKI, P., 1991, A Settlement of the Funnel Beaker culture at Nowy Młyn, Site 6 (Kujavia, Poland) - Initial Results. In: Die Trichterbecherkultur: Neue Forschungen und Hypothesen II, edited by D. Jankowska. Poznan: Instyitut Prahistorii Universitetu im. Adam Mickiewicza w Poznaniu Zakład Archeologii Wielkopolski IHKM PAN w Poznaniu, p. 127-131.

MADSEN, T., 1976, Jættestuen Hørret Skov I: Et nyt fund af fodskåle med massive midtdel. Kuml, p. 65- 94. METZLER, A., 1997, Aktuelles zur Moorarchäologie in Niedersachsen. Urgeschichts-Abteilung des Niedersächsischen, Landesmuseums Hannover. (http://www.urgeschichte.de/artikel/ moorarch.htm, accessed 17 March 2002).

HÄUSLER, A., 1992, Der Ursprung des Wagens in der Diskussion der Gegenwart. Archäologische Mitteilungen aus Nordwestdeutschland 15, p. 179-190.

MILISAUSKAS, S., 1978, European Prehistory. New York: Academic Press.

HUBERT, F. & HUYSECOM, É., 1980, Les mégalithes de la région de Marche-en-Famenne. Marche-en-Famenne, son passé et son avenir, p. 26.

MÜLLER, J., 2001a, Zur Radiokarbondatierung des Jung- bis Endneolithkums und der Frühbronzezeit im Mittelelbe-SaaleGebiet (4100-1500 v. Chr.). Berichte der Römisch-Germanischen Kommission 80, p. 31-90.

HUBERT, F. & HUYSECOM, É., 1982a, L’ensamble mégalithique de Wéris: Fouilles anciennes et récentes. Notae Praehistoricae 2, p. 57-62.

HUBERT, F. & HUYSECOM, É., 1983, Troisème campagne á l’allée couverte I de Wéris. Conspectus, p. 25-27.

MÜLLER, J. 2001b, Soziochronologische Studien zum Jung- und Spätneolithikum im Mittelelbe-Saale-Gebiet (4100-2700 v.Chr.): Eine sozialhistorische Interpretation prähistorischer Quellen. Vorgeschichtliche Forschungen 21. Rahden/Westf.: Verlag Marie Leidorf.

HUBERT, F. & HUYSECOM, É., 1984a, Quatrème campagne á l’allée couverte I de Wéris. Conspectus , p. 15-19.

PEDERSEN, L, A. FISCHER & B. AABY, 1997, The Danish Storebælt since the Ice Age: Man, Sea and Forest. København:

HUBERT, F. & HUYSECOM, É., 1982b, Fouilles de contrôle au dolmen nord de Wéris. Conspectus, p. 28-30.

160

M.O. Baldia: The Oldest Stone Rampart: Enclosures and Megalithic Tombs of the Funnel Beaker Culture... A/S Storebælt Fixed Link, National Forest and Nature Agency, National Museum of Denmark.

ŠMÍD, M., 1998, Sídliště kultury s nálevkovitými poháry u Laškova, okr. Přostejov. Pravěk, Nova řada 6, p. 97-138.

PLESLOVÁ-ŠTIKOVÁ, E. et al., 1985, Makotřasy: A TRB Site in Bohemia. Fontes Archaeologici Pragenses 17. Praha: Museum Nationale Pragae.

ŠMÍD, M., 2001, Nálezy kultury s nálevkovitými poháry z Kostelce na Hane a typologie keramiky straší fáze I. Stupne KNP na Moravě. Pravěk Supplementum 8, p. 276-298.

PODBORSKÝ, V., et al., 1993, Praveké Dejiny Moravy. Vlastiveda Moravská Zeme a Lid, Nová Rada 3. Brno: Muzejní a vlastivedna spolecnost.

STADLER, P., 1995, Ein Beitrag zur Absolutchronologie des Neolithikums in Österreich auf Grund der 14C-Daten. In Jungsteinzeit im Osten Österreichs, edited by E. Lenneis, C. Neugebauer-Maresch & E. Ruttkay. Forschungsberichte zur Uru. Frühgeschichte 17. St. Pölten – Wien: Niederösterreichisches Pressehaus u. Verlagsgesellschaft, p. 210-224.

PAVELČÍK, J., 2001, Erkenntnisse bezüglich auf die Bevölkerung der Höhensiedlung der Badener Kultur in Hlinsko bei Lipník nad Beèvou (Mährisches Tor). In Mensch und Umwelt während des Neolithikums und der Frühbronzezeit in Mitteleuropa, edited by A. Lippert, M. Schultz, S. Shennan, & M. Teschler-Nicola. Rahden/Westf.: Verlag Marie Leidorf, p. 73-82.

STEPPAN, K., 1998, Archäologische Untersuchungen der Säugetiere aus den Graben der Michelsberger‚ Erdwerke’ in Bruchsal, Landkreis Karlsruhe. In Die Michelsberger Kultur und ihre Randgebiete: Probleme der Entstehung, Chronologie und des Siedlungswesens, edited by J. Biel, H. Schlichtherle, M. Strobel & A. Zee. Stuttgart: Theiss, p. 143-158.

RAETZEL-FABIAN, D., 1999, Der umhegte Raum – Funktionale Aspekte jungneolithischer Monumental-Erdwerke. Jahresschrift für mitteldeutsche Vorgeschichte 81, p. 81-117.

STRAHM, C., 2001, Klimaentwicklung und Besiedlungsgeschichte im nordalpinen Raum. In Mensch und Umwelt während des Neolithikums und der Frühbronzezeit in Mitteleuropa, edited by A. Lippert, M. Schultz, S. Shennan, & M. Teschler-Nicola. Rahden/Westf.: Verlag Marie Leidorf, p. 17-19.

RAETZEL-FABIAN, D., 2000, Calden: Erdwerk und Bestattungsplätze des Jungneolithikums; Architektur – Ritual – Chronologie. Bonn: Habelt. RUTTKAY, E. & MAYER, C. 1995, Spätneolitikum. In Jungsteinzeit im Osten Österreichs, edited by E. Lenneis, C. Neugebauer-Maresch & E. Ruttkay. Forschungsberichte zur Ur- u. Frühgeschichte 17. St. Pölten – Wien: Niederösterreichisches Pressehaus u. Verlagsgesellschaft, p. 108209.

TOUSSAINT, M., 1997a, Les Sépultures Mégalithiques du Bassin Mosan Wallon. Le Secret des Dolmens. Wéris-Durby: Musée des Megalithes, p. 53-63. TOUSSAINT, M., 1997b, Les sépultures néolithiques dans les grottes et les abris sous roche de Wallonie. Le Secret des Dolmens. 1997, p. 65-75.

SCHIBLER, J, HÜSTER-PLOGMANN, H., JACOMET, S., et al., 1997, Ökonomie und ökologie neolithischer und Bronzezeitlicher Ufersiedlungen am Zürichsee. Zürich: Zürich und Egg.

ZÁPOTOCKÝ, M., 1991, Michelsberg und die relative Chronologie des Frühen und Älteren Äneolithikums in Böhmen. In: Die Trichterbecherkultur: Neue Forschungen und Hypothesen II, edited by D. Jankowska. Poznań: Instyitut Prahistorii Universitetu im. Adam Mickiewicza w Poznaniu Zakład Archeologii Wielkopolski IHKM PAN w Poznaniu, p. 127-131.

SCHIRREN, M., 1996, Archäologische Flußfahrt längs der Elbe. Archäologie in Deutschland 2, p. 20-23. SHENNAN, S., 2001, Trends in der Bevölkerungszahl in Mitteleuropa 4000-1500 v. Chr. und ihre Bedeutung. In Mensch und Umwelt während des Neolithikums und der Frühbronzezeit in Mitteleuropa, edited by A. Lippert, M. Schultz, S. Shennan, & M. Teschler-Nicola. Rahden/Westf.: Verlag Marie Leidorf, p. 97103.

ZÁPOTOCKÝ, M., 1998, Zum Forschungsstand über die relative Chronologie des frühen Äneolithikums in Böhmen. In Die Michelsberger Kultur und ihre Randgebiete: Probleme der Entstehung, Chronologie und des Siedlungswesens, edited by J. Biel, H. Schlichtherle, M. Strobel & A. Zee. Stuttgart: Theiss, p. 303-302.

ŠMÍD, M., 1990, Přispevek k poznání eneolitických mohylových pohřebišť na střední Moravě. Pravěké a slovanské osídlení Moravy, p. 67-89.

ZÁPOTOCKÝ, M., 2000, Keramika s Brázděným Vpichem a synchronizace Čech, Morav a Středního Poduynají ve Starším Eneolitu. Archeologické Rozhledy 52, p. 595-622.

ŠMÍD, M., 1992, Druhé eneolitické mohylové pohřebiště na katastru obce Náměšť na Hané, okr. Olomouc. Pravěk, Nova řada 1, p. 44-65.

ZICH, B., 1992, Ausgrabungen auf dem stein- und bronzezeitlichen Grabhügelfeld von Flintbek, Kreis Rendsburg-Eckernförde. Ein Vorbericht. Archäologische Nachrichten aus Schleswig-Holstein 3, p. 6-21.

ŠMÍD, M., 1991, Nové nálezy kultury s moravskou malovaou keramikou na Přostejovsku. AR XLIII, p. 185-205. ŠMÍD, M., 1993, Nástin periodizace kultury s nálevkovitými poháry na Moravé. Pravěk, Nova řada 2, p. 131-157.

ZICH, B., 1994, Die Ausgrabungen chronisch gefährdeter Hügelgräber der Stein- und Bronzezeit in Flintbek, Kreis Rendsburg-Eckernförde: Ein Vorbericht. Offa 49/50, p. 15-31.

ŠMÍD, M., 1994, Ein Burgwall mit steinerner Stirnmauer aus der älteren Stufe der Trichterbecherkultur auf dem Burgwall Rmíz bei Laškov im Kataster der Gemeinde Náměšť na Hané, Kreis Olomouc, Land Mähren. Jahresschrift für mitteldeutsche Vorgeschichte 76, p. 201-230.

ZICH, B., 1995, Drei Jahrtausende Siedlungsverlauf und Landausbau. Archäologie in Deutschland, 2, p. 6-11.

ŠMÍD, M., 1995, Výsledky zjišťovacího výzkumu na eneolitickém Hřadisku Rmíz u Laškova. Pravěk, Nova řada 3, p. 19-77.

161

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

162

E. Gatto: La place de la crémation dans le traitement des défunts à la fin du Néolithique en France

LA PLACE DE LA CRÉMATION DANS LE TRAITEMENT DES DÉFUNTS À LA FIN DU NÉOLITHIQUE EN FRANCE Esther GATTO

Résumé : Parmi les milliers de sépultures collectives connues en France, peu de sites (près de 100) témoignent de la crémation dans le traitement des cadavres. Pourquoi cette pratique originale ? La crémation a été employée dans des lieux de dépôt identiques à ceux communément utilisés au Néolithique récent-final, elle a été associée à tous les types de dépôts rencontrés à l’époque (primaires, secondaires, simultanés, successifs). La crémation paraît s’intégrer parfaitement aux pratiques funéraires de la fin du Néolithique et ne pas être réservée à une partie de la population. Pour cette dernière recherche cependant, les analyses envisageables sont limitées du fait de la mauvaise conservation des os brûlés. Actuellement, on peut seulement noter que la répartition des sépultures plurielles à crémation semblent s’organiser autour de deux pôles, le Bassin parisien et le sud-sud-est de la France. Cependant on peut se demander s’il s’agit réellement de deux pôles d’influence ou du reflet des activités de recherches publiées dans le domaine qui nous intéresse. Abstract: Among the thousands of the late Neolithic collective burial sites known in France, few sites (near one hundred) display the cremation in the treatment of corpses. Why this particular practice? Cremation was used in the same places (dolmen, cave, grave, hypogeum…) as the other sepulchres, was present during all the terminal Neolithic and was associated with all the types of known deposits (primary deposits, secondary ones, simultaneous deposits, successive deposits). Cremation seems to be perfectly integrated in the funerary practices of the terminal Neolithic and not limited to a specific segment of the population, but, for this last research, the realizable analyses are restricted because of the bad preservation of burnt bones. Only one fact can be noticed, the geographical distribution of cremated collective burial sites gets settled around two poles, the Bassin parisien and the south-south-east of France. Are there two poles of influence or is there a reflection of the differences in bringing-out of the research activities on our topic?

La fin du Néolithique voit se généraliser l’utilisation des sépultures plurielles. J’ai recours ici à la terminologie proposée par J. Leclerc (d’après Chambon, 1999), lorsque l’on ne peut trancher entre le caractère simultané ou successif des dépôts des défunts dans la sépulture, ou lorsque l’on souhaite évoquer ces deux types de geste. Dans le cas de dépôts simultanés, on parle de sépulture multiple, s’il s’agit de dépôts successifs, la sépulture est dite collective (Duday, 1987, Duday et al., 1990).

Sur la carte présentée en figure 1, qui montre la répartition des 84 autres sites étudiés, nous pouvons observer deux concentrations distinctes : une première au niveau du Bassin parisien et de la Marne, une seconde dans le sud de la France (Lot, Tarn-et-Garonne, Aveyron, Gard, Var). Je ne sais encore comment l’interpréter ; s’agit-il effectivement de deux régions dans lesquelles la crémation a été plus utilisée ou ces concentrations sont-elles davantage représentatives des travaux de synthèse ou de l’intérêt que les archéologues ont manifesté aux restes crémés retrouvés dans les sépultures plurielles dans ces zones géographiques ?

Le nombre de sépultures plurielles se chiffre en milliers sur le territoire français (d’après Chambon 1999). La crémation, quant à elle, est une pratique peu fréquente. Actuellement, j’ai pu dénombrer 101 sites (liste 1) pour lesquels le feu semble avoir joué un rôle dans le traitement du défunt ou dans celui du sépulcre.

Les restes de crémation retrouvés sur ces gisements peuvent résulter de différents types d’ustion. Dans le cas de sépulture à dépôts primaires (les os sont retrouvés à l’endroit où le corps s’est décharné ; Duday, 1978) la crémationa été pratiquée sur le lieu même du dépôt. Elle peut intervenir :

En effet, dans ce corpus j’ai pris en compte : - 2 sites dont la surface sur laquelle les corps ont été déposés est rubéfiée, alors qu’aucun os brûlé n’a été retrouvé ;

- de manière répétée dans le cas de sépultures collectives (au fur et à mesure de leur arrivée, les corps sont brûlés) comme à la grotte du Gardon (Ain ; Gatto et Buquet, 2000) ;

- 4 sites pour lesquels le feu a eu une fonction autre que brûler des cadavres ces foyers annexes sont localisés dans la chambre funéraire au centre ou en périphérie du dépôt sépulcral, ils ont pu être allumés à plusieurs reprises… seuls les os en contact du foyer portent des traces de combustion ;

- de manière unique dans le cas de sépulture multiple (il y a eu un feu unique pour l’ensemble des corps) ainsi que cela s’est passé à Reichstett-Mundolsheim-Souffelweyersheim (Bas-Rhin ; Blaizot, 2001) ;

- enfin, il y a 11 sites pour lesquels le feu ou des produits de combustion tels que la cendre, la chaux ou le calcaire chaulé interviennent parmi les gestes qui clôturent l’utilisation du lieu funéraire.

- ou de manière exclusive pour un défunt, dont les restes sont souvent retrouvés en marge du dépôt funéraire ou dans le vestibule du monument. 163

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe Liste 1 – Liste des sites intégrés au corpus de l’étude Crémation comme traitement des défunts 1 - Gardon, Ambérieu-en-Bugey (Ain), grotte 2 - Ribemont (Aisne), fosse 3 - Le Coulet du Stramousse, Cabris (Alpes-Maritimes), dolmen 4 - Les Cascades 1, Creissels (Aveyron) 5 - Peyrolebade, Espalion (Aveyron), dolmen 6 - Rosier 1, Martiel (Aveyron), dolmen 7 - Rosier 3, Martiel (Aveyron), dolmen 8 - Les Treilles, Saint-Jean et Saint-Paul (Aveyron), grotte 9 - Les Truels, Millau (Aveyron), grotte 10 - Reichstett-Mundelsheim-Souffelweyersheim (Bas-Rhin), fosse 11 - Les Abeilles, Châteauneuf-les-Martigues (Bouches-du-Rhône), grotte 12 - La Bastide Blanche, Peyrolles (Bouches-du-Rhône) 13 - Baume d’Onze Heures, Trets (Bouches-du-Rhône), grotte 14 - Enco-de-Bote, Allauch (Bouches-du-Rhône), tholos 15 - La Marane, Châteauneuf-les-Martigues (Bouches-du-Rhône), grotte 16 - Le Pilon du Roy, Allauch (Bouches-du-Rhône), grotte D du vallon 17 - Pitchoun-Ome, Allauch (Bouches-du-Rhône), grotte 18 - La Plaine, Meyrargues (Bouches-du-Rhône), dolmen 19 - Pomeiret, Saint-Vallier (Bouches-du-Rhône), dolmen 20 - Le Rove (Bouches-du-Rhône), grotte 21 - Terrevaine (La Ciotat), Bouches-du-Rhône, grotte 22 - Tholos 1 de Sainte-Anne, Saint-Vallier (Bouches-du-Rhône), tholos 23 - Tholos 4 de Sainte-Anne, Saint-Vallier (Bouches-du-Rhône), tholos 24 - La Hoguette, (Calvados), fosse et éléments en matière organique 25 - Les Barbilloux, Saint-Aquilin (Dordogne), grotte 26 - Perpétairi, Mollans (Drôme), hypogée 27 - La Clause, Carnas (Gard), aire de combustion 28 - Macassargue, Corconne (Gard), grotte 29 – Rossi, Corconne (Gard), grotte 30 - La Coste, Saint-Haon (Haute-Loire), abri 31 - Mont-Jonnet, Coubon (Haute-Loire), abri 32 - Le Rond-du-Lévrier, Salette (Haute-Loire), abri, 33 - La Roquette, Saint-Pargoire (Hérault), sépulture mégalithique 34 - Les Marais de Pouline, Villerable (Loir-et-Cher), fosse sous dalle 35 - Les Canas, Varennes-Changy (Loiret), fosses 36 - Maison-Rouge, Montigny (Loiret), fosse 37 - Barthes 1, Floirac (Lot), dolmen 38 - Barthes 2, Floirac (Lot), dolmen 39 - Bout de la Combe, Marcilhac (Lot), dolmen 40 - Causse de Montvalent, Montvalent (Lot), dolmen 41 - Les Garroustes, Sérignac (Lot), dolmen 42 - Hauteurs de Lugagnac (Lot), dolmen 43 - Pech Curet, Salviac (Lot), dolmen 44 - Pourquayré, Marcilhac (Lot), dolmen 45 - Dignas, Sainte-Enimie (Lozère), tumulus 46 - Mont-Saint-Michel, Mont-Saint-Michel (Manche), tertre tumulaire 47 - Le Chemin des Breton, Aulnay-aux-Planches (Marne), allée sépulcrale 48 - Chiozé, Lignon (Marne), fosse 50 - Les Mournouards 3, Mesnil-sur-Oger (Marne), hypogée 51 - Oyes (Marne), 2 hypogées 52 - La Pierre-Michelot, Villevenard (Marne), hypogée 53 - Les Ronces, Villevenard (Marne), grotte 54 - Vert-la-Gravelle, canton de Vertus (Marne), hypogée 55 - Mané-Lud, Locmariaquer (Morbihan), dolmen 56 - Belleville, Vendrest (Oise), allée sépulcrale 57 - Hérouval, Montjavoult (Oise), allée sépulcrale 58 - Layon-Beurrefrais, Peroy-les-Gombries (Oise), fosse 59 - Ecuelles, Moret-sur-Loing, près de Saint-Mammés, (Seine-et-Marne) allée sépulcrale 60 - Les Boutards, Saint-Hilaire (Seine-et-Oise), fosse sous bloc de grès 61 - Dampont, Us (Seine-et-Oise), allée sépulcrale 62 - Pech 1, Bruniquel (Tarn-et-Garonne), dolmen, dépôts secondaires, feu à l’extérieur 63 - Pech 5, Bruniquel (Tarn-et-Garonne), dolmen, dépôts secondaires, feu à l’extérieur 64 - Pech 9, Bruniquel (Tarn-et-Garonne), dolmen, dépôts secondaires, feu à l’extérieur 65 - Parmain 2, Parmain, Val-d’Oise, fosse, dépôts secondaires, bûcher hors du site 66 - Villeneuve-Saint-Georges, Varenne-Saint-Hilaire (Val-de-Marne), fosse 164

E. Gatto: La place de la crémation dans le traitement des défunts à la fin du Néolithique en France 67 - La Bouissière, Cabasse (Var), dolmen 68 - La Briande, Ramatuelle (Var), dolmen 69 - Les Cèdres, Le Plan-d’Aups (Var), grotte, ossements fortement brûlés, charbons et cendres 70 - Gonfaron (Var), grotte 71 - Gotteaubry, La Londe-les-Maures (Var), dolmen 72 - La Haute-Suane, Grimaud/Sainte-Maxime (Var), dolmen 73 - La Lauve, Salernes (Var), grotte 74 - L’Olivier, Les Adrets-de-l’Esterel (Var), grotte 75 - Roque-d’Aille, Lorgues (Var), dolmen 76 - Saint-Jeaume, Lorgues (Var), dolmen 77 - San Sébastien 1, Plan de la Tour/Sainte-Maxime (Var), dolmen 78 - San sébastien 2, Plan de la Tour/Sainte-Maxime (Var), dolmen 79 - La Verrerie-Vieille Saint-Paul-les-Fayence (Var), dolmen 80 - Les Echaffins, Cairanne (Vaucluse), hypogée 81 - Cheny (Yonne), monument en bois 82 - Monéteau (Yonne), fosses, dépôts secondaires 83 - Villeneuve-sur-Yonne (Yonne), plusieurs corps brûlés partiellement 84 - Vinneuf (Yonne), fosse, dallage de fond Feu présent en tant que condamamnation ou incendie Le-Champ-à-Fosses, Montigny-sous-Crécy (Aisne), fosse Le Collet-Sausset (Bouches-du-Rhône), fosse Le Bas-Berlue, Pocancy (Marne), fosse Le Collet-Sausset (Bouches-du-Rhône), fosse Le Bas-Berlue, Pocancy (Marne), fosse Lacroix-Saint-Ouen, Lacroix-Saint-Ouen (Oise) Houssoy, Troissereux (Oise), fosse La Pierre-Plate, Presle (Seine-et-Oise), allée sépulcrale Montigny-Esbly, Lesche (Seine-et-Marne), allée sépulcrale La Chaussée-Tirancourt (Somme, allée sépulcrale Les Crottes, Roaix (Vaucluse), hypogée Surface de dépôt rubéfié Les Montièvres, Saint-Mammés (Seine-et-Marne), caveau en pierre St Mammés (Seine-et-Marne) Foyers annexes Dessus-le-Bois-de-Thizy, Montigny-l’Engrain (Aisne), allée sépulcrale Aven de la Boucle, Corconne (Gard), grotte Devezas, Saint-Maurice de Navace (Hérault), dolmen La Justice, Epone, canton de Mantes (Seine-et-Oise), allée sépulcrale

Dans le cas des dépôts secondaires à crémation, les corps ou les restes ont été brûlés sur un champ de combustion distinct du lieu de dépôt final. Le ramassage des os sur le champ de combustion a pu être plus ou moins exhaustif. Deux champs de combustion ont été observés à La Bouissière, à Cabasse dans le département du Var (Bérard, 1954, Roudil et Bérard, 1981), et dans le quartier de la Clauze, à Carnas dans le département du Gard (Louis et Peyrolle, 1934).

inhumations, ils peuvent alors faire partie du remplissage comme à Peyrolebade (Aveyron ; Allard, 1984) ou à La Coste (Gisclon, 1983 et 1986). Ainsi, il n’est pas possible d’affecter une tranche chronologique à cette pratique au sein du Néolithique récent et final.

Pour un grand nombre de sites, les données publiées sont insuffisantes pour déterminer si les os ou les corps ont été brûlés à l’endroit même de leur découverte ou si les restes ont été déposés dans un second temps, après avoir été ramassés sur le champ de combustion.

Sur la carte présentée en figure 2, nous pouvons observer la variété des monuments et des milieux qui abritent les restes de crémation. Cette diversité est conforme à celle rencontrée à cette époque et respecte les spécificités régionales (allées couvertes du Bassin parisien, plus de grottes et de dolmens dans le sud…). On peut cependant noter un grand nombre de fosses aménagées, ou non, dans le Bassin parisien et sa périphérie.

Il est à noter que les dépôts à crémation peuvent être rencontrés au-dessus de couches funéraires non brûlées (il ne faut pas alors les confondre avec des condamnations) ; ils peuvent aussi être recouverts par des restes ostéologiques non brûlés comme au Rond-du-Lévrier (Haute-Loire ; Morel et Perrot, 1976), voire réduits pour laisser place à des

A présent, la question que l’on se pose est « qu’est-ce qui a été brûlé : un corps ou des os ? ». Par cette interrogation, est évoqué le problème que soulève la reconnaissance de l’os brûlé à l’état frais de celui brûlé à l’état sec. De nombreux articles existent sur le sujet et qui s’appuient sur l’observation des os. Les critères retenus sont fréquemment la couleur et le 165

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. La crémation dans les pratiques funéraires de 3300 à 2300 avant J.-C.

Figure 2. Différents types d’aménagements contenant des restes de crémation. 166

E. Gatto: La place de la crémation dans le traitement des défunts à la fin du Néolithique en France

type de fissuration de l’os (Buikstra et Swegle, 1989, Eckert, James et Katchis, 1988, Franchet, 1933, Thurman et Willmore, 1981). Le premier peut être écarté d’emblée à cause de la subjectivité des marqueurs qui ne tiennent pas compte de l’influence du sédiment. Le second n’est guère plus rigoureux, car il s’appuie sur des expérimentations qui ne tiennent pas compte des conditions archéologiques ou repose en général sur des échantillons peu nombreux et peu variés.

forme d’une grotte. Au fond du porche se trouvait un amas d’os enchevêtrés tous brûlés, qui reposait sur un dallage, visant vraisemblablement à niveler le sol. Le sédiment encaissant était rubéfié. La première option a été de voir s’il y avait une cohérence dans le désordre apparent, c’est-à-dire dégager intégralement les os avant de les prélever, suivre les os qui plongent dans la masse et noter leur rupture, afin de se donner toutes les chances de percevoir des connexions. Un grand nombre de connexions a pu ainsi être repéré dont des connexions labiles (figure 3) ; ces dernières, nombreuses, nous ont permis de déduire qu’il y avait eu des dépôts primaires suivis rapidement d’une crémation. Le caractère primaire des dépôts est appuyé par le profil ostéologique (figure 4) qui ressemble fort à celui obtenu, à l’hypogée II des Mournouards où des squelettes en connexion ont été mis au jour par A. Leroi-Gourhan et son équipe (Leroi-Gourhan et al. 1962). Ainsi le feu a brûlé des corps complets.

Il paraît donc utopique et peu rigoureux de vouloir expliquer la nature d’un dépôt à partir de la seule observation des vestiges osseux. En revanche, la qualité des données de terrain et les données de fouilles sont déterminantes pour la compréhension du fonctionnement sépulcral. Nous allons voir ce qu’il est possible d’obtenir comme résultats à partir de l’étude qui a été menée sur la zone sépulcrale du Gardon (Ain) dont la fouille a été achevée il y a un an, par Cécile Buquet et moi-même dans le cadre d’un chantier dirigé par J.-L. Voruz. Le site se présentait sous la

Par ailleurs, la plupart des connexions montrent une flexion des membres. Donc les corps étaient en position repliée à la

bloc cranio-facial mandibule atlas axis sternum scapula clavicule humérus radius ulna capitatum hamatum triquetrum pisiforme lunatum trapèze trapézoïde scaphoïde métacarpien 1

Os

métacarpien 2 métacarpien 3 métacarpien 4 métacarpien 5 sacrum os coxal fémur patella tibia fibula talus calcanéus naviculaire cuboïde cunéiforme latéral cunéiforme intermédiaire cunéiforme médial métatarsien 1 métatarsien 2 métatarsien 3 métatarsien 4 métatarsien 5

0

10

20

30

40

50

60

70

80

90

Nombre d'individus (pour 100)

Mournouards

Gardon

Figure 3. La grotte du Gardon (Ain) : exemple de connexion labile avec ce pied droit immature. La partie distale a été perturbée par un terrier. Photographie J.-L. Voruz. 167

100

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 4. Comparaison des NMI de fréquence obtenus à l’hypogée II des Mournouards (Marne ; Leroi-Gourhan et al., 1962) et à la grotte du Gardon (Ain). Les valeurs sont présentées pour un NMI maximal égal à 100 afin de faciliter la lecture, les valeurs réelles sont les suivantes : NMI aux Mournouards = 60, NMI au Gardon = 43 ; NMI : Nombre minimal d’individus

Figure 5. La grotte du Gardon (Ain) : projection verticale des collages réalisés dans le mètre carré J20.

fin de la combustion. Il est impossible de préciser si c’était le cas avant la crémation, car l’une des premières phases de la combustion d’un corps consiste en la déshydratation des tissus, ce qui entraîne une rétraction des muscles et donc l’attitude pugiliste du défunt (Wahl, 1981).

des bandes qui ne s’interpénètrent pas, donnant l’impression de couches successives. Parallèlement, sur le terrain, des couches de cendres ont été observées. Elles découpent la couche funéraire en 4. Cette donnée conforte l’aspect lité dégagé par la projection verticale des liaisons ostéologiques. Il y a eu au minimum 4 phases de dépôt suivis de crémation sur ce site.

Tous les os ayant été relevés sur plan, il a été possible de procéder à une analyse spatiale des os et fragments d’os. Les déplacements de surface sont de faibles amplitudes et toujours inscriptibles dans un volume corporel. J’ai ensuite réalisé une étude verticale des liaisons ostéologiques (figure 5) ; les déplacements sont faibles également et l’on peut distinguer

Il était alors intéressant de voir si, à travers l’étude ostéologique, cette pratique touchait une catégorie de population perceptible, c’est-à-dire certaines catégories d’âge. Le quotient de mortalité des sujets de moins de 20 est 168

E. Gatto: La place de la crémation dans le traitement des défunts à la fin du Néolithique en France

Figure 6. La grotte du Gardon (Ain) : distribution des 16 sujets immatures, après la ventilation des individus à cheval sur plusieurs classes d’âge, suivant le principe de minimalisation des anomalies (Sellier, 1996).

compatible avec une population naturelle (ayant une espérance de vie comprise entre 25 et 35 ans ; Ledermann, 1969) et la distribution des sujets immatures1 s’inscrit dans un schéma de mortalité naturelle (figure 6 ; Sellier, 1996). Nous voyons qu’il n’y a pas d’anomalie perceptible qui laisserait présager une sélection par l’âge de la population crémée.

Est-elle en relation avec deux foyers d’influence ? Cette pratique de l’accélération de la décomposition a-t-elle connu une « vogue » à un moment donné ? Concernait-elle une population particulière ? Les réponses de ma recherche actuelle sont pour l’instant limitées. Les études des sépultures plurielles à crémations doivent, pour la majorité, encore être approfondies, voire même être réalisées afin de mieux cerner la diversité des gestes propres à l’intervention du feu, ce à quoi je m’emploie pour quelques uns de ces sites. Par ailleurs, si les concentrations se confirment dans le sud-sud-est de la France et dans le Bassin parisien avec une extension vers l’est, il serait intéressant de voir si l’on retrouve cet usage du côté de l’Italie et de l’Allemagne.

L’étude montre ainsi qu’au moins 43 défunts ont été déposés à la grotte du Gardon. Il y a eu au minimum 4 phases de dépôt suivi rapidement à chaque fois d’une crémation. Ce rituel ne semble pas avoir été appliqué à une tranche d’âge particulière. Aucun acte de rangement, de réduction ou de prélèvement des os n’a été observé. Nous voyons qu’il est possible de comprendre assez précisément le fonctionnement funéraire d’un site et l’importance que jouent à cette fin la stratégie de fouille, l’observation sur le terrain et la précision de l’enregistrement. Cependant, compte tenu de son isolement géographique, la compréhension du rituel funéraire utilisé à la grotte du Gardon ne nous apporte pas d’éclairage sur les pratiques qui ont pu être adoptées dans les deux zones de concentration observées.

Adresse de auteur Esther GATTO Laboratoire d’Anthropologie Université Bordeaux I Avenue des Facultés 33405 Talence Cedex france [email protected]

CONCLUSION Bibliographie

La crémation apparaît comme un élément qui s’intègre parfaitement au fonctionnement des sépultures plurielles de la fin du Néolithique en France dans l’état actuel de nos connaissances. Elle est implantée dans les mêmes monuments ou milieux, elle couvre la même fourchette chronologique, elle s’associe à tous les types de dépôts rencontrés à cette époque. 1

ALLARD, M., 1984. Site de Peyrolebade à Espalion (Aveyron). Etude des tertres n° 1, 2 et 3. Rapport de sauvetage urgent, 45 p. BERARD, G., 1954. Nouvelles découvertes au dolmen de la «Bouissière», Cabasse (Var), Bulletin de la Société Préhistorique Française, LI, p. 666-667.

L’âge a été estimé à partir des germes dentaires (Moorrees, Fanning et Hunt, 1963) et de la maturation osseuse (Birkner, 1980).

169

BIRKNER, R., 1980. L’image radiologique typique du squelette, Paris, Maloine, traduit de l’Allemand par M. Bléry, édition

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe originale : 1977 (Das typische Röntgenbild des Skeletts, München, Urban Scharztenberg).

GISCLON, J.-L., 1986. Saint-Haon (Haute-Loire). Le gisement funéraire néolithique de la Coste 2, Revue archéologique du Centre de la France, 25, 1, p. 101-102.

BLAIZOT F. 2001, Premières données sur le traitement des corps humains à la transition du Néolithique récent et du Néolithique final dans le Bas-Rhin, Dimensions culturelles, Gallia Préhistoire, 43, p. 175-225

LEDERMANN, S., 1969. Nouvelles tables-type de mortalité, Presse Universitaires de France (Travaux et documents ; 53). LEROI-GOURHAN, A., BAILLOUD, G., BREZILLON, M., et al., 1962. L’hypogée II des Mournouards (Mesnil-sur-Oger, Marne), Gallia Préhistoire, V, 1, p. 23-133.

BUIKSTRA, J.E. and SWEGLE, M., 1989. Bone modification due to burning: Experimental Evidence. In Bone Modification, edited by R. Bonnichsen and M.H. Sorg. Center of Study of the First American, p.247-258.

LOUIS, M., et PEYROLLE, D., 1934. Aire à incinérations de l’époque néolithique, Congrès Préhistorique de France. Compterendu de la XIe session, Périgueux, p. 142-144.

CHAMBON, P., 1999. Du cadavre aux ossements. La gestion des sépultures collectives dans la France néolithique. Thèse de nouveau doctorat d’Ethnologie – Anthropologie – Préhistoire. Université de Paris I. 2 vol.

MOORREES, C.F.A., FANNING, E.A., and HUNT, E.E., 1963a. Formation and resorption of three deciduous teeth in children, American Journal of Physical Anthropology, 21, p. 205-213.

DUDAY, H., 1978. Archéologie funéraire et Anthropologie. Application des relevés et de l’étude ostéologiques à l’interprétation de quelques sépultures pré- et protohistoriques du Midi de la France, Cahiers d’anthropologie, 1, p. 55-101.

MOORREES, C.F.A., FANNING, E.A., and HUNT, E.E., 1963b. Age variation of formation stages for ten permanent teeth, Journal of Dentist Research, 42, 6, p. 1490-1502. MOREL, P., et PERROT, R.., 1976. Anthropobiologie d’un abrisous-roche prhistorique, Le Rond-du-Lévrier (Haute-Loire), Travaux et documents de paléoanthropologie et paléopathologie, 3, Lyon : C.N.R.S., 261 p.

DUDAY, H., COURTAUD, P., CRUBEZY, E., SELLIER, P., et TILLIER, A.-M., 1990. L’anthropologie « de terrain » : reconnaissance et interprétation de gestes funéraires, Bulletins et Mémoires de la Société d’Anthropologie de Paris, 2, n.s., p. 2950.

ROUDIL, O. et BERARD, G., 1981. Les sépulture mégalithique du Var, CNRS, Paris.

ECKERT, W.G., JAMES, J. and KATCHIS, S., 1988. Investigation of Cremation and Severely Burned Bodies, The American Journal of Forensic Medicine and Pathology, p. 188-200.

SELLIER, P., 1996. La mise en évidence d’anomalies démographiques et leur interprétation : population, recrutement et pratiques funéraires du tumulus de Courtesoult. In Nécropole et société au premier Âge du Fer : le tumulus de Courtesoult (Haute-Saône), edited by J.-F. Piningre. Document d’Archéologie Française, 54, Paris, La maison des sciences de l’Home, p. 188212.

FRANCHET, L., 1933. La coloration des os dans le sol. Le bouillage des cadavres au Moyen-Âge : l’incinération et ses phénomènes, Revue scientifique, 15, p. 483-495 et 520-532. GATTO, E. et BUQUET, C., 2000. La structure plurielle à crémations de la grotte du Gardon (Ain) : une pratique funéraire originale du Néolithique récent ?, p. 303-332.

THURMAN, M.D., and WILLMORE, L.J., 1981. A replicative cremation experiment, North American Archaeologist, 3, 4, 19801981, p. 275-283.

GISCLON, J.-L., 1983. Les sépultures collectives sous-abri de la Coste 2, Saint-Haon (Haute-Loire), Bulletin de la Société d’Anthropologie du Sud-Ouest, XVIII, 3, Bordeaux, p. 107-116.

WAHL, J., 1981. Beobachtungen zur Verbrennung Mensclicher Leichname, Archäologisches Korrespondenzblatt, 11, p. 271-279.

170

L. Guzzardi et al.: Archaeological and Anthropological Research on Neolithic Burials from South-Eastern Sicily...

ARCHEOLOGICAL AND ANTHROPOLOGICAL RESEARCH ON NEOLITHIC BURIALS FROM SOUTH-EASTERN SICILY: VULPIGLIA (PACHINO, SIRACUSA) Lorenzo GUZZARDI, Maria Rosa IOVINO, Annalisa RIVOLI & John ROBB

Abstract: Three burials, (tombs 1, 2, and 3) have been found at the Neolithic settlement of Vulpiglia, located in the coastal area between Marzamemi and Portopalo, where Stentinello phase occupation is well documented. On the basis of their context characteristics and gravegoods they can be ascribed to a later Neolithic phase that will be clarified by C14 analysis. The occurrence of undisturbed single inhumation in tombs 2 and 3 will provide information concerning mortuary practice, osteology, paleonutrition, gender/age differentiation, state of health and any disease. Résumé : À l’intérieur du village néolithique de Vulpiglia, situé sur la côte entre Marzamemi et Portopalo, où un établissement de la phase de Stentinello est attesté, trois sépultures ont été trouvées (tombeaux 1, 2, 3). Ces sépultures appartiennent à une époque néolithique plus récente que la période de vie du village. Les tombeaux 2 et 3 sont en connexion. Les inhumés sont probablement de sexe féminin, l’un d’un âge compris entre 15 et 20 ans, l’autre entre 20 et 30 ans. L’inhumé du tombeau 3 est accompagné d’un trousseau funéraire. L’objectif de cette recherche est celui de contribuer à la connaissance des rituels funèbres et des caractéristiques anthropologiques des groupes humains néolithiques de la Sicile orientale.

The site of Vulpiglia is on the coast between Marzamemi and Portopalo, a little further south than the Corruggi and Calafarina caves (fig.1).

identified (Guzzardi et alii 2000). Three tombs (fig. 2) were found in the same area within the boundaries of the village, and date from a Neolithic phase which followed the period of activity in the village.

The presence of Neolithic remains had already been noted at the end of the Sixties (G. Kapitàn 1991). However, a stone quarry and illegal constructions have devastated the archaeological deposits, destroying the central part of the settlement. Holes for piles can still be seen in some areas of the quarry walls, others can be found south and east of the quarry, and to the south of the beach of Morghella. The latter may relate to the outskirts of the settlement. The presence of holes in such a vast area seems to indicate a settlement of about 15.000 mq (Guzzardi and Basile 1993).The position of the village, similar to other Neolithic sites along the coast of Syracuse, must have been determined by strategic and economic considerations. In fact, it lies on a promontory which dominates the bay of Morghella and the area of the marsh with the same name, thus controlling both the coast and the territory immediately behind it. Thanks also to the presence of the marsh, the area was suitable for various economic activities.

Tomb 1 was in disarray and contained only a few bonefragments which are insufficient for a thorough anthropological study. Tomb 2 (fig. 3) contained a skeleton, oriented east-west, lying huddled up on its right side. There were medium-sized stones around and over the skeleton. Two animal teeth and a jawbone were found next to the body, and may be funerary objects. The skeletal remains are still being studied, but indicate that we are dealing with an adult. Tomb 3 (fig. 4) also contained a skeleton, oriented east-west, lying huddled up on its right side. Here too there were stones around and over the skeleton. The skeletal remains are very fragmentary, but indicate an adult, and on the basis of the morphology of the skull, probably female. The teeth show a medium grade of wear of the first molars and light wear on the third molars; two hypoplastic cracks in the enamel of the first upper right incisive are the first indicators of infantile malnutrition. The study of the wear of the teeth indicates an age of between twenty and thirty. A macroscopic examination did not reveal evidence of any pathologies. Other chemical analysis will supply confirmation in future.

The excavation was carried out in 1991 in an area not damaged by heavy vehicles, excavators, etc., and directed by Dr. L. Guzzardi on behalf of the Soprintendenza ai Beni Culturali e Ambientali di Siracusa (Guzzardi 1993). The settlement is a Stentinello-style nucleus of a certain importance, distinguished by a truly Neolithic economy. The analysis of the information obtained during the excavation, from the study of the ceramics and the lithic industry provides an idea of the organisation of the village of Vulpiglia. In the zone under examination, an area dedicated to working activities and an area of waste linked to these activities were

This last tomb also contained funerary objects including a vase (fig. 5, 6), a flint blade, a flake, an obsidian point and a bladelet (fig. 7), 10 beads for necklaces or bracelets made from fossil shells of the Spondylus type (fig. 8). The lithic objects were studied with functional analysis in order to 171

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. Map of south-eastern Sicily showing the Vulpiglia location.

understand whether they had been created specially for the dead person. The objects in obsidian have had the surfaces altered through contact with a heat source, while the flint blade shows some spots of white patina, but still it has clear signs of use (fig. 9) relating to the working of wood. This could imply that the blade was an object of everyday use, probably a personal tool buried with its owner.

Figure 2. Vulpiglia. Tombs n.1, n.2 e n.3.

The chronology of this last tomb can be deduced from the vase found, dating from the middle Neolithic, Serra d’Alto style. This shape (globular body, rim made by a vertical band, unpierced trumpet-like handles on the shoulders, heel foot with a flat base) appears in a late period of the Serra d’Alto style and prefigures a typology which will become typical of the following facies of Diana. In the Serra d’Alto area, the vase of Vulpiglia can be compared with a vase found in Basilicata in the hut of Fondo Gravela (AA.VV.1976), in the site of the same name as the facies, which however, has a decorative motif of wavy lines in brown on a light ground (fig. 10 ). Similar typologies (fig. 11) have also been found in Lipari (Bernabo Brea and Cavalier 1980), in the area of the Castello (a vase with painted decoration where the body has a similar shape but the handles are different) and in contrada Diana (fig. 12 ) where the base of the vase is missing but the upper part and handles are very similar to that one from Vulpiglia. Vase with a globular shape and unpierced trumpet172

L. Guzzardi et al.: Archaeological and Anthropological Research on Neolithic Burials from South-Eastern Sicily...

Figure 3. Vulpiglia. Tomb n.2.

Figure 4. Vulpiglia. Tomb n.3.

Figure 6. Lateral view of the olla from tomb 3.

Figure 5. Frontal view of the olla from tomb 3.

like handles are also documented in the Serra d’Alto facies in central Puglia (Geniola 1986, 1987).

Figure 7. Lithic implements belonging to the burials from tomb. 3.

It should be noted that the surface of the vase of Vulpiglia is partially covered by limestone encrustations and partially eroded; however, in some parts one can make out a reddish colour (slides), although it is not possible to say anything about any decoration or treatment with pigment.

right sides, huddled up in a position similar to the foetal position, looking towards the north. Both tombs are characterised by the presence of stones around and over the body ( as if to cover or to “check” it).

The finding of these tombs offers some interesting information about Neolithic funerary rites in Eastern Sicily. The presence of two tombs with the same characteristics implies that there were precise funeral rites; tombs 2 and 3 are both oriented east-west, both bodies are lying on their

Unfortunately, knowledge of the funerary rites of the Neolithic period in Sicily is very limited. There is information about Neolithic burials around Siracusa (Tusa 1992) – Gisira (facies di Stentinello), Matrensa, Megara Hyblea (facies di 173

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 8. Ten shell beads belonging to the burials from tomb 3.

Figure 9. Micro wear trace observed on the lateral edge of the flint tool from tomb 3.

Figure 11. Serra D’alto style from Lipari Castello, Lipari, Isole Eolie.

Figure 10. Serra D’alto style from Fondo Gravela, (Serra d’Alto), Basilicata.

Figure 12. Serra D’alto style from Lipari C.da Diana, Lipari, Isole Eolie.

Diana) – around Catania – Fontanazza (Recami et al. 1983) (facies di Stentinello), Biancavilla (Tusa 1992) (facies di Diana) – around Ragusa (Tusa 1992) – Calaforno, Paolina (facies di Stentinello) – around Messina (Tusa 1992) – torrente Boccetta (facies Serra d’Alto) – around Trapani (Di Salvo and Germana 1997) – Stretto Partanna ( middle Neolithic). In some of these cases, stones or slabs have been documented around the grave, but there is little or no information as to the alignment of the body or the position

of the skeletons ( the huddled up position is documented at Matrensa, Megara and near torrente Boccetta). The above information will be completed by the analyses currently underway to establish an absolute chronology, paleonutrition, paleopathology, causes of death and, where possible given the fragmentary nature of the remains, any physical stress caused by particular activities carried out during the individual’s lifespan. 174

L. Guzzardi et al.: Archaeological and Anthropological Research on Neolithic Burials from South-Eastern Sicily...

Acknowledgements

Bibliography

Our thanks to the Soprintendente Dr. Giuseppe Voza who authorised and made this research project possible and to Dr. Beatrice Basile for her help with the study of the materials. Thanks also go to Dr. E. Rivoli for improving the graphics, to Dr. S. Rivoli for the translation into French, and to Dr. S. Kimbell for the translation into English.

AA.VV., 1976, Il Museo Nazionale Ridola di Matera. Matera, pp. 57-61. BERNABO’ BREA, L., CAVALIER, M., 1980, Meligunìs Lipára IV. L’acropoli di Lipari nella preistoria. Palermo, pp.454-475. COCCHI GENICK, D., 1994, Manuale di Preistoria. Neolitico. Viareggio. DI SALVO, R., GERMANÀ, F., 1997, Frequentazione umana nella Sicilia preistorica. In TUSA, S., (a cura di) Prima Sicilia. Palermo, 1997, pp.

A special thought goes to the memory of Dr. Efisio G. Picone who carried out the preliminary studies of the materials.

GENIOLA, A., 1986, Il Neolitico nella Puglia centrale. Atti della XXV Riunione Scientifica IIPP, pp. 55-86.

Authors’ addresses

GENIOLA, A., 1987, La Cultura di Serra d’Alto nella Puglia centrale. Atti della XXVI Riunione Scientifica IIPP, pp.771-781.

Lorenzo GUZZARDI Soprintendenza ai BB. CC. AA, Di Enna via Orfanatrofio, 15 94100, Enna, ITALIE

GUZZARDI, L., 1993-1994, Ricerche archeologiche nel Siracusano. Kokalos, XXXIX-XL, II, 2, pp. 1299-1303.

Maria Rosa IOVINO Centro Internazionale di Sperimentazione, di Documentazione, e di Studio per la Preistoria e l’Etnografia dei popoli primitivi via Targia, 2 I-96100, Siracusa, Sicile, ITALIE

GUZZARDI, L., BASILE, B.,1996, Il Capo Pachino nell’Antichità. In AA. VV. 1996, La Magna Grecia e il mare. Studi di storia marittima, Istituto per la storia e l’archeologia della Magna Grecia, Taranto, pp. 191-95.

GUZZARDI, L., Il Capo Pachino nell’Antichità. Atti e Memorie ISVNA, Noto, in corso di stampa.

GUZZARDI, L., IOVINO, M.R., RIVOLI, A., 2000, L’organizzazione del villaggio neolitico di Vulpiglia presso Pachino (Siracusa). Atti della XXXV Riunione Scientifica IIPP, in corso di stampa.

Annalisa RIVOLI viale Teracati, 51/h I-96100, Siracusa, ITALIE

KAPITÄN G. 1991, Maritime aspects of early neolithic coastal settlements in southeast Sicily, Atti Thracia Pontica (Sophia) IV, pp. 399-426.

John ROBB Department of Archaeology University of Southampton Southampton, SO17 1BF UNITED KINGDOM

RECAMI, E., MIGNOSA, C., BALDINI, L.R., 1983, Nuovo contributo sulla preistoria della Sicilia. Sicilia Archeologica, 5253, pp. 45-82. TUSA, S., 1992, La Sicilia nella Preistoria, Palermo.

175

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

176

E. Kashina: Human-Elk Images in Northwestern Russian Neolithic Art

HUMAN-ELK IMAGES IN NORTHWESTERN RUSSIAN NEOLITHIC ART Ekaterina KASHINA

Résumé : Il s’agit des représentations de l’homme-élan, l’être qui contient simultanément les traits de l’homme et de l’animal. C’est un image unique et expressive de la créature composite, élaborée en manière concrète et spécifique. Leur dispersion sur la large territoire de la région forestière de la Russie Européenne en troisième millénaire avant l’ère chrétienne peut-être signifie la large extension et l’importance considérable pour les sociétés des chasseurs néolithique. Évidemment, ces représentations présentent un certain héros mythique, dout le rôle symbolique et les fonctions pouvaient être différents. Abstract: The discussed human-elk depictions represent a creature containing both human and animal traits. It is the unique and expressive image of a compound creature patterned in the concrete framework. The dispersion over large territories in the European Russian forest zone in the 3rd millennium BC probably means its wide extension and notable importance among the hunter-gatherers’ societies. Obviously, these depictions demonstrate the particular mythical hero, whose functions and symbolic meaning could be different.

INTRODUCTION

materials for tools, weapons and sculpture. Today, about thirty pieces of the Mesolithic and Neolithic elk sculpture, made of bone, antler, amber, wood and flint, are known in the forest zone of the European Russia. This image was closely connected with hunt and cosmogonic beliefs (Kosarev 1988).

The territory of the European Russian forest zone in the Late Neolithic (3rd millennium BC) was inhabited by huntergatherers. Several archaeological cultures of that period yielded a large amount of findings, mainly presented by small sculpture and rock art.

The human imagery is even richer. About 150 representations made of the same raw materials (mentioned earlier) are known. And finally, both images – elk and human – were widely represented in the Northern Russian Late Neolithic rock art.

At that time there existed two main kinds of images - animal and human. Actually, they were most important ones in primitive mentality and ideology all over the world during several epochs: from the Late Paleolithic to contemporary primitive peoples.

HUMAN-ELK DEPICTIONS

The epoch of the Late Neolithic in the European Russian forest zone yielded highly developed small sculpture depicting humans and animals and possessing its own particular stylistic features. Lots of raw materials were successfully used for sculpture making: flint, bone, antler, fired clay, amber, wood and stone.

The deep connection between human and elk in the mentality of the Neolithic inhabitants obviously influenced the creation of a particular compound image – the “human-elk”. Three human-elk representations of the North-West of Russia are connected with the Neolithic sites on the lakes Onega, Lacha and Peno and fall within the range of 2500-2000 BC. Apparently, they are dispersed over a large territory (Fig. 1; Fig. 2; 1-3). The first piece is a petroglyph (15 cm high) pecked in rock. The second is an antler figurine (9 cm high). The third piece is the flint figurine (6.2 cm high). So, they are made in different techniques and of various materials. Nevertheless, they possess the same morphological structure which characterizes, in fact, a particular compound creature: front-facing human figure, oblong torso (in two cases (Fig. 2; 1-2) without arms), straight spread legs with heart-shaped hooves (in one case the hoof is made in realistic style (Fig. 2; 2)).

Human depictions are rather schematical, without any particular details. In most cases even the sex is not recognizable, only the main body parts are present: head, torso, arms and legs. The face (if it is depicted) is also made in a simplified manner: eyes and nose, sometimes no mouth at all. The diverse animal imagery includes bears, elks, beavers, birds and snakes. Unlike schematical human depictions, those of the animals seem to be more realistic. In some cases, even individual species of birds can be determined in their depictions. The elk image was probably the most popular. Its representations were spread through the whole forest zone. The elk was a much desired hunting pray and also might be a sacred animal. Elk bone and antler were the common raw

Another group of specimens is spread over a much larger territory (Fig. 1), including the lake Onega region and the region of Upper and Middle Volga river. There are also three 177

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Figure 1. The territory of the human-elk figurines expansion.

representations falling supposedly within the range of 25002000 BC, approximately simultaneous to the latter group of images. All of them (Fig. 2; 4-6) are figurines, made of flint and depicted in a simpler manner. Their heights are from 3.5 to 6 cm. The details remind those of the first group: frontfacing figure, oblong torso, straight spread legs (but without hooves). One figurine has no arms (Fig. 2; 6), another has something like horns or animal ears on the head (Fig. 2; 5). Despite such a small amount of human-elk representations, their body forms (construction, pose) reflect the evident similarity. Their spreading on large territory reveals their functional and symbolic importance for the inhabitants of different regions located very far from each other. Unfortunately, the discussed representations have no radiocarbon dates and may be approximately dated by the second half of the 3rd millennium BC. They could be connected with the sites of the Volosovo archaeological culture (approximately the 3 rd – beginning of the 2 nd millennium BC) and some neighboring and approximately simultaneous cultures. To my mind, the two studied groups of images represent a particular type of a mythical hero, possessing the features of both human and elk, who probably possessed – as ancient hunters probably believed – special power and qualities. I agree with the supposition of Russian

researcher Svetlana Studzitskaya (who was the first to reveal the human-elk image as a result of studying two figurines (Fig. 2; 2-3)) that the human-elk appeared approximately in 2500-2000 BC (Studzitskaya 1994). THE FUNCTIONAL USE OF DEPICTIONS The fact, that the figurines have almost the same height, may witness their similar functional use. Their small size probably means that they could be a personal property. Let us discuss them individually. Two of the flint figurines are stray findings and their context is unknown (Fig. 2; 3, 4). One flint figurine (Fig. 2; 6) was found inside a dwelling (Ovchinnikova 2000). Another flint figurine (Fig. 2; 5) was found on the territory of a graveyard (Utkin & Kostyleva 1996). The antler figurine (Fig 2; 2) was found in a burial (probably female, because its inventory was presented by two bone awls); the stretched body lying on the belly, face down. The figurine was also (!) on its belly, face down and situated near the hip of the skeleton (Foss 1938). The choice of this 178

E. Kashina: Human-Elk Images in Northwestern Russian Neolithic Art

Figure 2. The human-elk figurines. Illustrations: 1 - Ravdonikas 1936; 2 – Oshibkina 1992; 3 – Studzitskaya 1994; 4, 5 – Utkin & Kostyleva 1996; 6 – Ovchinnikova 2000.

THE SYMBOLIC MEANING

particular kind of figurines may signify that it was undoubtedly very important, for example, as a certain protector of a person after death of as a guide to the underground world.

We think that the image represented in these depictions may be interpreted as a particular mythical hero – an animal master in general, or the master of elks, who is well known for example in mythology of Urals and Syberian huntergatherers. The presence of elk features in a human figure reflected the purity and the sacred nature of the represented object (Kosarev 1988). Another possible interpretation could be the representation of an ancestor (connected with totemism) or a zoomorphic spirit as a manifestation of animism (Studzitskaya 1994). Thus, the context of humanelk depiction findings reflects possibly their extended symbolic meaning: from providing lucky hunt (in the role of an animal master) to accompanying a person after death (in the role of a mythical ancestor).

And finally the petroglyph (Fig. 2; 1) was pecked on the shore rock of the Onega lake (Kladovets cape) and was surrounded by a large amount of other depictions of different creatures and objects, such as elks, humans, boats with elk heads on its nose, birds (probably gooses or swans) and numerous sun and moon symbols (Ravdonikas 1936). So, the discussed depiction was included in the sophisticated system reflecting the mentality of local hunter-gatherers, which unfortunately can not be fully deciphered. The rather small height of the discussed petroglyph (compared to many others around it) suggests that it has the same nature as the flint figurines, not only morphologically, but also symbolically and probably functionally. Such similarity of small human depictions in petroglyphs and flint figurines was already pointed by Russian researchers (Zamyatnin 1948; Stoliar 1994).

Two more specific traits attract our attention. First, all figurines have straight spread legs. This trait is specific for anthropomorphic depictions connected with hunt magic rites of numerous primitive societies (Virsaladze 1976). Second, the majority of specimens possess the deliberately modeled rounded torso (Fig. 2; 2, 4, 5, 6). This feature allows the assumption that they represent the female nature of a creature, which is possibly depicted pregnant or giving birth.

Thus, we can suppose, that there might exist several functional uses of human-elk depictions. They could be used in hunt magic, funeral rites and in the system backing up a certain world order. 179

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

CONCLUSIONS

OKLADNIKOVA, E., 1996, Naskalnoe iskusstvo aborigenov Sibiri i problema proiskhozhdeniya idei sviazi shamana i vselennoi. In Systemnye Issledovaniya Vzaimosvyazi Drevnich Kultur Sibiri i Severnoi Ameriki 3. (The rock art of Syberian aborigens and the problem of the origins of the following idea: ties between the shaman and the Universe. In Studying the Contacts Between Ancient Cultures of Syberia and North America 3.) St. Petersbourgh: The Museum of Anthropology and Ethnography.

Apparently, in comparison with the number of elk and human representations of the 3rd millennium BC in the European Russian forest zone, the given image is significantly rare. So, we can suppose that the concrete framework of the given image was probably new (or just beginning to be used). Maybe it was the true predecessor of later, numerous humananimal images, which were more specific for shamanism as the religious system and reflected in Bronze and Iron Age rock art and traditions of primitive societies in Fennoscandia, Urals and Syberia (Nunez, 1995; Okladnikova 1996). The cult of human-animal spirits and the system of rites, together with the habit of putting on animal skins and horns were more suited for shamanism, so the iconography of the humanelk was further developed there.

OSHIBKINA, S., 1992, Sever Vostochnoi Evropy. Iskusstvo v neolite i eneolite. In: Iskusstvo Kamennogo Veka: Lesnaya Zona Vostochnoi Evropy. (The North of Western Europe. Neolithic and Aeneolithic art. In Stone Age Art: The European Russian Forest Zone.) Moscow. OVCHINNIKOVA, N., 2000, Volosovskiye Drevnosti Yuga Lesostepnogo Povolzhya. Twerskoi Arkheologicheskiy Sbornik IV. (Volosovo Antiquities in the Forest-Steppe of the Middle Volga River Region. Twer Archaeological Revue IV.) Twer. RAVDONIKAS, V., 1936, Naskalniye Izobrazheniya Onezhskogo Ozera. (Onega Lake Rock Art.) Moscow-St. Petersbourgh.

Author’s address

STOLIAR, A., 1994, Transliatsia idey kak mnogoobraznaya arkheologicheskaya realnost. In Kulturniye Transliatsii i Istoricheskiy Protses. (The translation of ideas as the wide archaeological reality. In Cultural Translations and the Way of History.) St. Petersbourgh.

Ekaterina KASHINA Archaeological Department State Historical Museum 103012, Krasnaya square, ½ Moscow, RUSSIA

STUDZITSKAYA, S., 1994, Osobennosti Duchovnoi Kulturi Volosovskich Plemyon. Drevnosti Oki. Trudy GIM 85. (The Volosovo Inhabitants’ Mental Features. The Oka River Antiquities. Studies of the State Historical Museum 85.) Moscow. UTKIN, A. & KOSTYLEVA, E., 1996, Antropomorphniye Izobrazheniya Volosovskoi Kulturi. Twerskoi Arkheologicheskiy Sbornik II. (Volosovo Anthropomorphic Depictions. Twer Archaeological Revue II.) Twer.

Bibliography FOSS, M., 1938, Pogrebeniya na Stoyanke Kubenino. Trudy GIM 8. (Burials of the Kubenino Site. Studies of the State Historical Museum 8.) Moscow.

VIRSALADZE, E., 1976, Gruzinskiy Okhotnichiy Mif I Poeziya. (Georgian Hunt Myths and Poetry.) Moscow.

KOSAREV, M., 1988, Chelovek i zhivaya priroda v svete sibirskich etnographicheskich i arkheologicheskich materialov. In Nekotoriye Problemy Sibirskoi Archeologii. (Man and nature according to archaeological and ethnographical data. In Some Problems of Syberian Archaeology.) Moscow.

ZAMYATNIN, S., 1948, Miniaturniye Kremneviye Sculpturi v Neolite Severo-Vostochnoi Evropy. Sovetskaya Arkheologiya X. (The Neolithic Flint Sculpture of the Northeastern Europe. Soviet Archaeology X.) Moscow.

NUNEZ, M., 1995, Reflections on Finnish Rock Art and Ethnohistorical Data. Fennoscandia Archaeologica XII. Helsinki: The Archaeological Society of Finland.

180

B. Tripković: The Role of Obsidian in the Neolithic: A Symbolic Expression of Human Domestication?

THE ROLE OF OBSIDIAN IN THE NEOLITHIC: A SYMBOLIC EXPRESSION OF HUMAN DOMESTICATION? Boban TRIPKOVIĆ

Résumé : La distribution de l’obsidienne sur le continent européen dépasse de loin la signification de sa fonction primaire de l’outil taillé, et ses sources dans les locations difficilement accessibles, sur les îles et les montagnes, confirme d’avantage cette impression. L’obsidienne a été utilisée en néolithique d’Europe presque exclusivement en tant que l’outil, comme objet dont l’emploi utilitaire est déterminé par la qualité et les propriétés mécaniques de la matières première. Cependant, le mode de son utilisation n’indique pas les différences importantes par rapport au reste de l’industrie de la pierre taillée, et les grandes quantités de l’obsidienne dans certaines parties de l’Europe, ainsi que les énormes distances qu’elle traversait jusqu’à sa destination finale n’ont pas d’explication logique uniquement dans la sphère matérielle. L’idée qui s’impose d’elle même, est qu’il existaient des motifs supplémentaires qui reliaient, par la distribution de l’obsidienne, des territoires éloignés, dans une pratique commune et compréhensible par tous. La définition de ses motifs, pourtant, exige l’utilisation d’une approche méthodologique qui se différencie de la simple classification typologique et de la recherche la plus appropriée du processus d’échange. En annexe, l’obsidienne a été examinée dans le contexte de la néolithisation de l’espace européen lorsque, d’après l’auteur, elle a acquiert une valeur supplémentaire qui la distingue de l’emploi exclusivement utilitaire chez les sociétés de chasseurs-ramasseurs. Abstract: The distribution of obsidian in Europe goes far beyond its primary utilitarian significance to function as a knapping raw material, whilst its sources at remote and hardly accessible locations, such as islands and mountains, additionally confirm such an impression. Obsidian was used almost exclusively as a tool in the Neolithic Europe, as an object that is valued according to the qualities and mechanic properties inherent in such a raw material. However, the mode of utilization does not indicate any difference in relation to the rest of the stone knapped industry and it remains difficult to restrict our explanation in any logical way to the material sphere, having in mind large quantities of obsidian in particular parts of Europe and large distances it traversed to the place of its final deposition. Henceforth there must have existed additional reasons based on mutually dependent and motivated practices, connecting distant territories via obsidian distribution. In order to account for these motives, it would be necessary to employ methodological approaches that differ from simple typological classification and to search for the most appropriate explanation of the exchange process. In this paper obsidian is examined in the context of the Neolithisation of Europe and the author assumes that at precisely this time obsidian acquires an ‘additional’ value, which distinguishes its circulation in the Neolithic from merely utilitarian use it assumed in the context of hunter-gatherer societies.

INTRODUCTION: THE PROBLEM

of subsequent research projects on the distribution of obsidian in Europe. Differences in the geographical position of Anatolian and European obsidian sources triggered important theoretical maneuver in proposing plausible ways to explain its distribution (Torrence 1986; Karimali 1994).

Despite the fact that sources of obsidian are restricted to only few locations, it is important to note that obsidian has not been understood as having a culture specific trait in some previous studies. For instance, sources of obsidian in Europe are restricted to particular islands of the Mediterranean or were detected in the continental mountainous regions, such as the Zemplen Mountain (Tykot 1996; Willms 1983; Biro 1998). Hence it is unusual that the distribution of obsidian goes far beyond its primary utilitarian function as a knapped tool, whilst its sources at remote and hardly accessible locations, such as islands and mountains, additionally confirm such an impression. However, any attempt at a thematic review of obsidian has been confined to two basic research agendas (cf. Torrence 1986; Karimali 1994). The first one is of typological–technological character, relating obsidian to the chipped-stone industry of a site, region or cultural tradition. Here obsidian is equated with flint tools - the same attribute analyses that are applied to other chipped-stone tools are equally applied to obsidian. The second research agenda aims at indicating mechanisms of long-distance exchange in the distribution of obsidian. Renfrew initiated such studies in relation to the regions of the Near East, Anatolia and Europe (Renfrew, Cann & Dixon 1965; ibid 1966; ibid 1968; Renfrew & Dixon 1976), and his results have been used as the basis

Recently, Renfrew (1993) has made another step forward in order to rule out a simple equation of artifacts and their utilitarian value, in this way overcoming general statements in relation to distribution models. In an attempt to understand the role of trade in cultural systems, he moves away from the sphere of practical activities to the cognitive salience of such actions. Among other objectives, he examines reasons to participate in the exchange in the first place. This approach can be backed by numerous ethnographic instances, such as the Kula exchange system as probably the best example (cf. Malinowski 1979). It would be unwise to propose a similar solution in relation to obsidian automatically, especially in regard to untheorized exchange dynamics and a lack of nuanced model that would account for such processes. There are numerous ethnographic examples from Americas or Australia that could usefully be applied, yet one should bear in mind the limits of ethnography as “individual memories may reveal more about ideal rather than real behaviour” (Torrence 1993: 467). Thus at this stage of research the least one can accept is a generalization that stems from the 181

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

archaeological context, indicating that “exchange system ... is merely material relic of patterns of interaction which had symbolic dimensions of a much wider kind” (Renfrew 1993: 14). It follows that one should postulate additional motives underlying the connection of distant territories via obsidian distribution as part of mutually dependent and motivated practices. In order to account for these motives, it would be necessary to employ methodological approaches that differ from simple typological classification and to search for the most appropriate explanation of the exchange process. One of the questions that deserves special attention relates to obsidian as a prestigious good. The view that obsidian had no prestigious value in the context of Anatolia and the Near East is largely based on the fact that this raw material was intensively exploited and almost exclusively utilized for tool manufacturing in this regional context (Özdogan 1994: 424). On the other hand, in European prehistory obsidian has been figured as a prestigious object (Ammerman and Polglase 1993: 101; Renfrew and Bahn 1991; Sherrat 1982), raw material or an artefact difficult to obtain and therefore possibly providing its owner with some status in the community. Thus, the idea of prestige is connected with its availability, i.e. limitations in acquiring this material. This view maintains obsidian’s exotic value, whilst it becomes classified together with other materials of similar symbolic potential, such as gold, amber, copper, Spondylus shell, etc. (Renfrew & Bahn 1991: 312). However, this kind of interpretation with a tendency to lump different materials under the same interpretive category neglects particularities and significant differences of cultural attitudes toward these materials. One can assume that these materials acquired their respective values out of different socio–political circumstances, while their use varies significantly. Similarly, the intensity of their reoccurrence coincides neither chronologically nor spatially. It seems that the generalizations regarding a uniform meaning of various exotic goods remains unsatisfactory. The dilemma whether obsidian is to be treated as a prestigious good in European archaeology is nonexistent since no particular attention has been drawn on such question. However, it seems important to answer this question as this would provide an understanding of what motivated the very process of exchange. Certainly, in order to solve the problem it is necessary to provide the social context for the occurrence of obsidian, the relationship between the quantity of obsidian and the necessity of its acquisition, establish definition of ‘prestigious’ within a certain cultural context and define a chronological framework of our analysis. In brief overview my intention is to underline specific differences between areas with obsidian outcrops in Europe, and to point out the significance of chronological distinctions in the study of obsidian. TIME AND CONTEXT: A VERY SHORT REGIONAL SURVEY Aegean. In the Aegean region obsidian sources are situated at the islands of Melos, Antiparos and Giali (Renfrew, Cann & Dixon 1965: 225; Torrence 1984: 49). Obsidian from

Antiparos is of low quality and was never exploited, while the obsidian from the island of Giali due to its lower quality was used for manufacturing tools used in the immediate vicinity of the island, where an immediate replacement of this brittle material was possible. On the other hand, this obsidian is interlaced with white spherules and achieves a specific aesthetic impression. Due to this factor, it was used during the Bronze Age for manufacturing of seals and vessels, which were exported as far as the island of Crete (Betancourt 1997). At Melos, on of the western Cycladic islands, there are outcrops of very good quality obsidian and precisely this served as the main source for acquisition of obsidian in the Aegean. In this area, the oldest obsidian is recorded in the Palaeolithic and Mesolithic layers of the Franchti Cave of the Peloponesos, while the actual ‘revolution’ in its use started only in the Early Neolithic (Perles 1989). At this time, at a large number of sites obsidian is a dominant raw material, frequently accounting for c. 70-80% of the chipped-stone industry and the technique of its manufacturing was of significantly greater quality if compared to other raw materials (Perles 1989). It is interesting to note that the quantities of obsidian at particular sites do not decrease in proportion to a greater distance from the source. This was the reason to postulate a scenario that the distribution of obsidian in the Agean comes as a consequence of the existing institution of specialized tradesmen (Perles 1989; Whittle 1996: 62). During this time, similar to the preceding periods, obsidian is solely used for manufacturing tools, while its contextual position within the cultural layer of a settlement does not imply any symbolic significance. However, the highest degree of obsidian distribution from Melos starts with the Bronze Age, and this is backed with the evidence from the Asia Minor (Torrence 1986). Obsidian continues to be used for manufacturing tools, although one can to note that with the start of metal use obsidian acquires a new dimension - it is frequently found in burials as a grave offering. Later during the Bronze Age, when metal objects become widely utilized, the quantity of obsidian in the Aegean is decreasing and it fades away from the historic scene (Torrence 1986). Central and western Mediterranean. Obsidian outcrops are found on the islands of Sardinia, Lipari, Palmarola and Penteleria, while its distribution, apart from the islands of the central Mediterranean and the Apenine peninsula, includes southern France, parts of northern Africa and most likely the eastern Adriatic coast (Willms 1983; Tykot 1996). Extensive characterization of finds from this vast area has indicated that the distribution of highest quality obsidian (from Lipari) encompasses the whole region, while lesser quality obsidian from other sources is distributed within zones of several hundred kilometers from the source at the most (Tykot 1996: 42-43). In addition, the obsidian from Lipari was imported only in the form of end-products while the obsidian from other sources was also found in the form of debitage, indicating its manufacturing at the settlement sites (Hallam, Warren and Renfrew 1976: 103). It has been suggested that this kind of contradictions can be explained by reference to several mechanisms of exchange (Ammerman and Polglase 1993: 102). In this region, similar to the eastern Mediterranean, all sources of obsidian are found at islands and it is surely possible to make analogies based on modes of transport and exchange, 182

B. Tripković: The Role of Obsidian in the Neolithic: A Symbolic Expression of Human Domestication?

although in contrast to Melos, these outcrops in the western Mediterranean were not exploited before the start of the Neolithic (Tykot 1996). With the appearance of production economy, at sites in the vicinity of outcrops obsidian becomes a dominant raw material in the chipped stone industry, whilst its quantity exponentially decreases with the greater distance from the source (Tykot 1996). During the Neolithic in the area of the central and western Mediterranean there are no contexts that would indicate an additional symbolic component of obsidian, implying a different meaning or role of this material. For the whole duration of the Neolithic use of obsidian is restricted to its utilitarian function, stemming from its functional and technical qualities. At the end of the Neolithic and during the Copper Age, obsidian is significantly less utilized while the zone of its distribution shrinks. At some sites at Sardinia and Corsica, obsidian is relatively often found in burials in the form of grave offering (Tykot 1996: 70).

alternative sources in the vicinity etc.) surely had the major influence on the start of obsidian utilization. There is a consensus in relation to the question about the start of intensive acquisition of obsidian that becomes an object of exchange found in significant quantities even at those sites that are on a considerable distance from the sources of this raw material. This happens with the start of the Neolithic. The start of the Neolithic varies considerably across Europe, and it seems that the obsidian follows these oscillations. Obsidian always occurs at the start of the Neolithic, which implies that humans relate to obsidian in a different manner at this period. This specific link justifies our intention to look at the evidence from this period in its own right, separated from other periods. Obsidian continues to be used in the utilitarian manner although the attitude of prehistoric communities toward obsidian becomes different; its value increases and only with the start of the Neolithic, and in relation to specific conditions connected with this period, it acquires attributes of an exotic material.

Carpathian Basin. In this area obsidian outcrops are restricted to southeastern Slovakia (Karpatski 1) and northeastern Hungary (Karpatski 2), while it was used during a long period, starting from the Middle Palaeolithic up to the Bronze Age (Williams-Thorpe, Warren and Nandris 1984). The source of a good-quality obsidian lies in Slovakia and this one was more intensively utilized although the zones of the distribution of the two types overlap (Biro 1998). Having in mind great chronological differences, it is most definite that the intensity of obsidian use highly varies through time. It was significantly utilized also during the Upper Palaeolithic (Gravettien) and Mesolithic, although these finds are mostly restricted to the immediate vicinity of outcrops. This differs sharply from the Neolithic on the basis of its quantity and distribution zones (Biro, Pozsgai & Vladar 1986: 278). During the Neolithic, the exploitation of obsidian is the most intensive and its dispersion is at maximum. However, directions of its distribution indicate some oscillations since these are mostly directed toward those cultural groups that, one assumes, were controlling exploitation of obsidian during this particular period (Biro 1998: 33). These fundamental changes during the Neolithic in the distribution of obsidian if compared with the patterns of its use during the preceding periods are not very noticeable on the matter of its immediate usage. The use of obsidian is still within its utilitarian role and it serves for the same range of functions, similar to other chipped stone tools. However during later periods, coinciding with the use of metal, it is possible to note significant changes. The quantity of obsidian significantly decreases, while new meanings are transposed from the domestic context into the domain of mortuary practice (Bognar-Kutzian 1972: 137138). One of the best examples of such a process is the Tiszapolgar-Basatanya cemetery, where obsidian appears as a grave offering in c. one-sixth of 155 burials. Here it appears as a tool, weapon (arrow-heads), nucleus, and even in the form of flakes (Bognar-Kutzian 1963). All of this points to a wide range of possible solutions and interpretations. Still, from this chronological point, the quantity of obsidian visibly decreases and it gradually disappears in the course of the Bronze Age.

On the other hand, it is interesting to note that the decline of the role of obsidian starts in all areas with the appearance of metal artifacts and although its usage continues for some time its role and meanings acquire a new dimension, which is nonexistent in the previous period. In contrast to the Neolithic, occurrence of obsidian in burials as a grave offering is relatively frequent. While the use of obsidian in the settlement context is possibly only practical, once it appears in burials its meanings become attached to a different sphere. It is possible that the meaning is transferred from the communal level to individuals, importantly in relation to individual identity. It is not my intention to specify in more detail those interpretative problems connected to the meaning of obsidian in the Copper Age that would warrant a more detailed study of contextual data in relation to buried individuals. Here it suffices to emphasize the obvious difference in relation to the Neolithic. Taking into account all of the aforementioned concerning the relationship between historical and contextual data, it is possible to distinguish phases of obsidian use in European Prehistory grouped in three major periods: 1. Period of hunter-gatherer communities when obsidian is mainly found in those areas around outcrops. The use of obsidian is particularly pronounced in the Upper Palaeolithic and Mesolithic of central Europe, while the Mediterranean sources were yet unknown, remained unexploited or were used on a small scale. During this period the reasons for the obsidian exploitation were primarily dependent on the vicinity of outcrops and the raw material quality; 2. The Neolithic period when the intensive use of obsidian starts with its wide distribution and significant increase in the relative proportion of other raw materials within assemblages of chipped stone industry. This wide distribution of obsidian objects stems actually from established inter-cultural network of exchange in which obsidian played a significant role. On the other hand, there are no direct indicators about obsidian’s prestigious connotations as it usually appears in the settlement context

In all the aforementioned instances, local conditions (availability of outcrops, settlement pattern, existence of 183

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

(so-called archaeological ‘cultural’ layer), without being related to specific contextual features (e.g. burials);

only but can be seen as a way to connect Neolithic communities through a unique practice. The case of the Mediterranean sites is the best example. At the island of Sardinia, for instance, there are a number of obsidian outcrops that were not exploited before the start of the Neolithic although the island was inhabited in the previous period, and although, as it was assumed, these outcrops could not have stayed unnoticed by the local inhabitants (Tykot 1996). Simply during this period there was no interest for obsidian and the local needs for chipped stone industry were regulated in the exploitation of alternative raw materials. With the start of the Neolithic within the wider region, obsidian is frequently found at the Sardinian sites consisting 100% of chipped stone industry or alternatively it highly dominated the assemblage. Also, via exchange it gradually reaches continental areas (Tykot 1996). It remains unclear what was the reason that obsidian was not exploited in the previous period or why the exchange of this and other types of obsidian starts only in the Neolithic. Exchange is an economic category, which assumes that exchange goods possess particular value. How is it possible to determine the value of obsidian in this instance?

3. Period of metal use, from early phases of the Copper Age up to the end of the prehistoric periods. Here obsidian is not present in significant quantities (with the exception of the Aegean) although patterns of obsidian distribution indicate its potential value for these communities. It is frequently found in burials, sometimes with some other valuable offerings, implying a similar value system. It is possible to assume that it served as one of the means of expressing prestige. On the basis of the previous discussion it is evident that chronological and social dimensions play a significant role in such a division, and existing cultural and social milieus expressed different approaches to obsidian. Regarding the fact that only in the Neolithic obsidian becomes an item of exchange, i.e., that only with the appearance of Neolithic cultures its use becomes significant, it seems logical to start the pursuit of its meaning and role with this period. This Neolithic milieu, when obsidian becomes widely distributed, represents a cross-road in all aspects of the development of the human society: mobile communities are replaced by sedentary, new technologies are advanced (food production, pottery), a demographic growth is attested and new models of social organization are established. However, the change occurs in the realm of ideology too, with new elements and material practices that are at this time adopted in relation to a ‘new reality’ and different style of life. Under such conditions of the early Neolithisation process in Europe obsidian becomes a new way of life and part of new value system. A possible answer to the question what obsidian meant for Neolithic communities is likely related to the question what the Neolithisation process meant for the pre-Neolithic communities of Europe. Therefore it is necessary to go back once again to the culture-historic context of the discussion, related to the appearance of obsidian, and I shall point out specific details related to its distribution, which in a more direct way imply possible explanations of the process.

The Aegean region also provides an interesting case of the connection between obsidian and Neolithic populations. Acquisition of obsidian, according to evidence from the Franchti cave, started significantly before the start of the Neolithic and possibly this took place across small islands, connecting Melos with the mainland (Pantelidou-Gofas 1996: 13). Thus during this period the acquisition of obsidian was based on a direct access to obsidian sources, without evidence of existing exchange. Only with the start of the Neolithic quantities of obsidian in the Aegean significantly increase, and at the same time its distribution leapfrogs deep into the continental hinterlands (Perles 1989). Having in mind that the quantities of obsidian in settlements reach an average of c. 80% of the chipped stone industry, and since this quantity does not decrease in proportion to the distance from the original source it was concluded that the so-called ‘downthe-line’ system of exchange cannot be postulated (Whittle 1996). On the other hand, some detailed analyses of obsidian artifacts have clearly indicated significant differences that exist in relation to the rest of the chipped stone industry. The type of manufacture and large distances obsidian traveled forced C. Perles (1989) to suppose the existence of specialists for obsidian acquisition, manufacture and distribution. If we accept this interpretation it would imply certain knowledge that not everyone could have obtained, as well as special attributes that were connected with obsidian. What kind of knowledge and attributes are in question?

IN PURSUIT FOR MEANINGS: NEOLITHIC EUROPE AND OBSIDIAN ARTIFACTS In the material record of the European Neolithic obsidian was almost exclusively used for manufacturing tools; its utilitarian function is determined by its quality and mechanic properties of the material. Although the mode of its utilization does not differ from the rest of the chipped stone industry, great quantities of obsidian in certain parts of Europe and large distances it traveled to site of its final deposition cannot be logically explained by reference to the material utility only. In almost every region where obsidian occurs as a consequence of long-distance exchange local sources of alternative raw material are found as well, and a possibility of exploiting the local raw materials could have negated any need for the acquisition of obsidian (Williams-Thorpe, Warren & Nandris 1984: 198). Despite this logic obsidian was acquired and according to Whittle (1996: 62) motives of such behavior are not to be found in the utilitarian domain

In the Carpathian region when compared to the Mediterranean, the situation is completely different. Obsidian widely utilized raw material and with the beginning of the Neolithic its dispersion starts. In the first wave of the Neolithisation of the Carpathian Basin only the middle Danube region was neolithized while the sources of obsidian in the north remained out of this initial Neolithic zone (Makkay 1996). This is the reason that the quantities of obsidian at this time at the sites of the middle Danube region are not very massive, although it is still present. However, the second wave of Neolithisation, encompassing the area of 184

B. Tripković: The Role of Obsidian in the Neolithic: A Symbolic Expression of Human Domestication?

central Europe included obsidian sources. It coincides with the appearance of the Linear Pottery culture (Biro 1998). Obsidian is present in larger quantities and, as it was already noted, its large distribution corresponds with the distribution of the LBK settlements, more precisely with its early and middle phases (Williams-Thorpe Warren & Nandris 1984: 186). However, the need for the raw material in this case might have been solved in different ways. The acquisition of obsidian was not necessary and it might have had some less utilitarian motives. What were the motives?

logical that the highest concentration of obsidian is recorded in the immediate zone around the sources of the raw material, while sets of circumstances are implicated in the overlap of cultural complexes. Such a pattern of distribution of obsidian cannot be considered coincidental, and it should be connected with the expansion of the Neolithic in these continental areas. Additional support for this interpretation may account for the fact that the obsidian from the territory of one cultural complex is not archaeologically attested on the territories of other complexes. The exceptions are very rare and almost always located in the contact zones or in those areas of the three complexes where cultural distinctions are blurred. The site of Obre 1 in central Bosnia is an extraordinary example of this last situation. In the earliest horizon of the settlement, which is characterized by the Starčevo-impresso material culture, several examples of obsidian of unknown origin were found. However, during the later phase of this settlement, characterized by the cultural label of Proto-Kakanj and Kakanj phases of the central Bosnian Neolithic, the presence of obsidian derived from Sardinia is mentioned (Greif 1995:85). Although the sources of the Carpathians are much closer to central Bosnia and while probably a part of the obsidian found here actually originated from the Carpathian sources, the presence of the Mediterranean obsidian is not surprising and it can be considered as a logical consequence of the mixture of influences in this geographic zone from both the Danubian complex characterized with barbotine pottery and the Mediterranean complex characterized with impresso pottery (Brukner 1977) .

It is obvious that these questions about reasons for obsidian acquisition are connected with the question about the way the first Neolithic communities were established in Europe. A suggestion that a wide distribution of obsidian in the Neolithic is a direct consequence of a large number of existing settlements during the previous period (Williams – Thorpe, Warren & Nandris 1984: 187) cannot account for the question posed and explain how come that more mobile communities were characterized with a low-level of obsidian distribution. A possible answer would be to argue that the economy of hunter–gatherer communities, without a productive quality, could not have the premises of organized trade. However, some contacts and limited exchange must have happened having in mind that the Carpathian obsidian appears, for example, in the Epipaleolithic context of the Djerdap gorge, few hundred kilometers away from its source; still the quantities of obsidian here are small and its significance rates secondary in comparison with the rest of chipped stone industry (Radovanović 1992). Thus mechanisms of acquisition existed even during this phase, without a necessity or a motive to obtain obsidian in any large quantities. Hence we can go back to the question of the Neolithisation of Europe. The main assumption here is that the changing role of obsidian was one component of wider changes that came about at the beginning of the Neolithic; it was spatially defined simultaneously with the spread of the Neolithic across Europe. Was this really a coincidence?

There is an additional important point here. I have said that each cultural complex was connected with its own sources of obsidian and that the distribution of obsidian was limited by cultural factors. However, the Danubian, i.e. the StarčevoKörös-Criș complex had no outcrops of obsidian on its own territory since in the course of the first wave of the Neolithisation these sources remained out of its reach further to the North, at territories where local Mesolithic communities were still at large (Kertesz 1996). Nevertheless, the Danubian complex was the closest Neolithic complex to the outcrops situated in Hungary and Slovakia, and these communities acquired obsidian, either in a direct way, or, as previously supposed, through contacts with Mesolithic communities in the North (Makkay 1996). These limitations in the accessibility of sources must have certainly affected the proliferation of obsidian, and within this cultural complex it was present in smaller quantities when compared with two other complexes. It is necessary to mention this need of the Neolithic communities to possess obsidian as part of their cultural (material) inventory. On the other hand, I have already mentioned that within the subsequent Neolithisation of the whole Carpathian Basin, including the outcrops of obsidian, this raw material was significantly exploited with a trend on territorial expansion that almost coincided with the complex of the Linear Pottery culture as the first Neolithic culture to encompass the whole of the Carpathian Basin (Williams – Thorpe, Warren & Nandris 1984:186). In relation to possible reasons that might have caused such a distribution, Tykot (1996:64) supposes that the limits of dispersion were probably equal with the movement of other materials and ideas, coinciding with cultural and ethnic boundaries.

The Neolithisation process, i.e. spread of the Neolithic in Europe, is one of the most complex issues within European prehistory. This may be confirmed if one looks back on the more than a half century long debate with a whole history of different hypotheses regarding this process. My intention is not to deepen the discussion in connection to this load of hypotheses. However, I need to mention that the establishment of the Early Neolithic cultures in Europe was organized according to the principle of cultural unity within wider regional frameworks. These units are usually described as cultural complexes (Garašanin 1979) or as cultural– formative zones (Özdogan 1998; Nikolov 1998), and this is based on the principle of regional interaction. This is a good example to show the formation and expansion of the Neolithic in Europe. It was proposed that three cultural complexes can be defined in the Early Neolithic - Balkan–Anatolian, WestMediterranean and the Danubian complex (Garašanin 1979). It is interesting to note that the expansion and regional distribution of obsidian absolutely match the confines of these cultural units (Willms 1983, Tykot 1996). Partly, the distribution of obsidian was naturally imposed since the resources were limited within these areas. Therefore it is only 185

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Obviously the communication between these communities involved a number of different aspects, for instance, common economic interests, (possibly) language, similar genetic basis, etc.; all these might have produced the cultural unity (Tripković 2001). The exchange of obsidian possibly had the connecting role and might have been one of the mechanisms of the Neolithisation process, and this might have caused the created cultural unity. In order to fully accept this idea, it would require detailed analyses of a number of Early Neolithic sites. One should also provide an explanation for the question why obsidian does not appear in some parts of the Balkan–Anatolian complex, such as Thrace or southern Serbia. One may assume that the distance from the source was too great and this raw material did not become a component of the exchange mechanisms. In some attempts this was explained by reference to social and economic barriers (Perles 1989). However, this kind of interpretations would require a more comprehensive analysis of exchange dynamics in relation to other goods. However, leaving aside these dilemmas regarding the role of obsidian in the first wave of the Neolithisation of Europe, it is clear that the Carpathian area plays an important role. Neolithic communities of the Starčevo-Körös-Criș culture complex acquired obsidian through the contact with the northern Mesolithic communities (Makkay 1996). At the same time, they probably conveyed and transmitted some of the basic ideas, principles and values characterizing the Neolithic way of life. This exchange was probably of a limited scope; however, it was probably based on the reciprocity and its contribution to the Neolithisation of the Carpathians must be acknowledged.

of general principles that characterize both Anatolia and larger parts of Europe over the course of the Neolithic. The content of these principles is culturally attractive and structured in a simple form, hence their long potency and survival of basic meanings despite possible changes in the way of expression. It can be supposed that meanings of these symbolic structures were defined and negotiated at the time of sharp social, cultural and economic contradictions, such as the process of the Mesolithic–Neolithic transition. Mesolithic communities in Europe were not able to compete with Neolithic cultures for equal rights in the exploitation of the same ecosystems (Starnini 2000:215). It would mean that at the time of transition the choice was either to adapt to a new situation or disappear (Efstratiou 1995). Furthermore, the adoption of the Neolithic for local populations might have been achieved as a specific innovation that allowed future survival of these communities. One can observe obvious changes that Neolithic brought about, such as permanent settlements, elaborate architecture, food production, new technologies (pottery, polished stone tools, etc.), as well as a new impetus for the existing prehistoric communities. However the question arises concerning the way these manifestations related to mental structures of Neolithic people. I hypothesize that the use of obsidian was adopted from the areas of Anatolia and the Near East, together with some fundamental Neolithic innovations, providing principles for new and different behavior that also involved a strong mental projection of these new values materialized through symbolic expressions in the production of figurines, house models or elaboration of sacrificial altars. Similarly, obsidian probably represented one of the basic metaphors while its acquisition in this context might have symbolized for an individual and/ or a community “both knowledge of broader horizons, which had after all brought the novelties of pottery, sheep, and goats and cereals, and participation in networks of contact and exchange” (Whittle 1996:311).

A similar model was previously postulated for the area of the Near East and Anatolia. Here obsidian was a part of exchange networks much earlier then in Europe. It was sometimes found thousand kilometers away from the source of its original exploitation (Moore 1985; Renfrew & Bahn 1991). Runnels and van Andel (1998; Runnels 1989) emphasize the significant role of trade for the process of formation and consequent spread of the Neolithic in regions of Anatolia and the Near East. The basic role of exchange in such processes provided mechanisms that were used to distribute product surpluses toward neighboring communities, i.e. toward those territories where ecological conditions did not satisfy required conditions of production. The site of Çatalhöyük is the case in point for contacts and exchange of raw materials between different ecological zones, providing a sharp contrast between locally domesticated cattle and abundant raw materials, including obsidian, imported from the highlands. According to A. Sherratt, this settlement accumulated its treasury right in the phase of agricultural expansion by supplying the highland sites with domestic cattle and taking from them various highland resources (1982:254).

The affirmative answer to the key question regarding the Mesolithic–Neolithic transition in Europe as a specific manifestation of both economic and material, and cognitive aspects, may solve the dilemma whether obsidian might have had some other non-utilitarian function during the Neolithic. I am inclined to ascribe to the obsidian, together with figurines or sacrificial altars etc., a similar role of a Neolithic metaphor. This seems as the only way to explain large quantities of obsidian throughout diverse Neolithic cultures. According to this hypothesis, metaphoric features assigned to obsidian were generated primarily through a direct contact with Neolithic populations arriving from the Near East and Anatolia. This took place in the process of introducing basic principles and values of a new way of life; obsidian was an integral part in this process. In this way, for European communities, obsidian might have had both cultural and important symbolic connotations, being a result of specific individual histories in relation to the use of this material (Renfrew 1993: 15). A subsequent diffusion of similar ideas was triggered on one side by constantly moving populations, interregional contacts and exchange of goods and information, and on the other by narrative forms (such as myths, legends, stories, etc.) that might have transmitted social memory of communities and ideas about a “new

Although all constitutive aspects of the Neolithisation process can only be comprehended in the proper way via focused regional-level approach, it is obvious that larger, more generalized trends can be attested that justify the chronological and cultural determination of a specific ‘Neolithic world’. Independent from explicitly material expression, Hodder (1990; 1998) singles out the existence 186

B. Tripković: The Role of Obsidian in the Neolithic: A Symbolic Expression of Human Domestication?

beginning”. Due to the complex mechanisms of transmission of these concepts, it can be assumed that the basic idea expressed in the acquisition of obsidian might have been a common one, whilst the ways it was manifested might have been many.

of chipped stone tools. Since new copper tools were an alternative not for chipped stone industry but for polished stone tools, it is likely that the absence of obsidian rather reflects new social constellations primarily related to the fact that one prestigious material (obsidian) was replaced with the other (copper). However, a sharp contrast in the amount of obsidian at sites of the Vinča culture is in proportion with a contrast between those settlements with all principles of the Neolithic way of life (stable economy, long occupation of a place, connection with a particular territory, etc.), delivering large quantities of obsidian, on one side and those settlements of much smaller size, occupied over shorter periods of time, frequently without or with only small quantities of obsidian present, on the other. The former type of sites represented regional centers although these did not function in organizing a redistribution of obsidian toward regionally marginal settlements. These differences are obvious at those sites that must have been interdependent, and the presence or absence of obsidian at particular sites can only be explained with the fact that only the participation in the network could enable obsidian acquisition. For instance, those sites without obsidian that possessed necessary means of exchange show that material conditions were not the only requirement. It is possible that there was a specific system where settlements with a long history, strong tradition and communal memory (multilayer sites or tells) held some kind of monopoly over the acquisition and use of obsidian. It seems that these Neolithic centers possessed specific knowledge about the symbolic ‘power’ of obsidian, connecting its origin with the past, distant and mystic areas. In addition, through obsidian acquisition, these communities might have expressed their right to structure and re-shape the knowledge about the past in this specific region.

A LOCAL EXAMPLE: THE CENTRAL BALKAN AREA In order to test the validity of this interpretation I look at the context of obsidian use in the central Balkan Neolithic, i.e. primarily referring to the territory of present-day Serbia. In culture history terms, Neolithic communities on this territory are characterized by the Starčevo and Vinča cultures, spanning the period c. 6200-4000 cal BC (Gläser 1996; Nikolova 1998). There are around 800 sites attributed to these cultures, and c. 200 sites was excavated (Chapman 1981; Srejović 1988). Many sites provided evidence about different kinds of the so-called ‘exotic’ objects with considerable differences between different sites and chronological phases. At the beginning I have mentioned the significance of the Kula type model of exchange as a way to evoke ideas about the motives and mechanisms of object or raw material acquisition from distant and unknown areas. However, in the central Balkans the applicability of this type of exchange system have already been tested and it showed that sociocultural patterns of the Neolithic cultures in this region cannot fulfill the required criteria of this model, and that reasons for the acquisition of exotic materials should be sought elsewhere (Greenfield 1991). Following previously mentioned principles about the role of obsidian as a metaphor of the Neolithic world, a new analysis for the central Balkan region indicates the following (Tripković 2001): - During the Early Neolithic, obsidian was acquired and due to non-existent local sources its acquisition was primarily connected with the Tokay region that was during this phase out of reach of the first Neolithic communities in the region, since in this area mobile groups of hunter-gatherers were still at large. The appearance of obsidian at the sites of the Starčevo culture unequivocally points to the existing contact between Mesolithic and Neolithic communities, indicating one of possible mechanisms for the Neolithisation of the Mesolithic communities. Non-existent local sources account for small quantities of obsidian at the sites of the Starčevo culture. However, there are considerable variations in the presence of obsidian at Early Neolithic sites – obsidian is both present and absent in the assemblages of chipped stone tools. According to the suggested interpretation this imbalance is most likely due to ideological contradictions between Mesolithic and Neolithic (Starčevo culture) communities and proportional to the imbalance of their conceptions of the world; despite the existence of a strong inter-cultural network of exchange and obvious cultural homogeneity, obsidian is completely absent or present only in small quantities in those areas and at sites with a strong Mesolithic tradition.

It is difficult to make a comparison with different regional traditions and cultures in this kind of analysis since the relationship of the Mesolithic and the Neolithic worlds should be conceptualized as a formation of “information spaces in ways that they have not been before” (Thomas 1998: 154). Hence, there can be considerable variations in the way Neolithic communities of Europe accepted obsidian and expressed what it meant in specific contexts. One striking example comes from Cres, an island at the Croatian part of the eastern Adriatic region. Along the relatively unexplored eastern Adriatic cost both Mesolithic and Neolithic sites were attested. From one of these sites, Ustrin, comes a so-called »hunter medallion« manufactured on obsidian and only approximately dated to the Neolithic period. Although it has not been published on the basis of its description in some publications it seems that it bears a representation of a hunting scene (Greif 1995: 87). Without more information on the condition of this find it can be interpreted in different ways. One of the possible explanations (Tripković 2001) directly supports the previous discussion about the role of obsidian in the Neolithic, with an additional indication about the Neolithisation process in the eastern Adriatic region. In this particular case it is possible to assume that by a symbiosis of a Mesolithic (hunting scene) and Neolithic (obsidian) traits a principle of unity was achieved through an existential

- Obsidian in the central Balkans is used more intensively up to the start of copper exploitation; once this process starts obsidian disappears completely from the assemblages 187

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

conjoining of local and newcomer traditions. In the time to come the unity will become one of the basic legacies of these prehistoric communities.

BETANCOURT, P. P., 1997, The Trade Route for Ghyali Opsidian. AEGAEUM 16, p. 171-175.

CONCLUSION

BIRO, K. T., POZSGAI, I., & VLADAR, A., 1986, Electron beam microanalyses of obsidian samples from geological and archaeological sites. Acta Archaeologica Scientiarum Hungaricae 38, p. 257-278

BIRO, K., 1998, Lithic Implements and the Circulation of Raw Materials in the Great Hungarian Plain during the Late Neolithic Period. Budapest: Hungarian National Museum.

On the basis of the previous discussion, obsidian both had a practical and very important communicational role during the Neolithic. It conveyed a specific message that was understandable for past communities. The expression of this message comes from a need to acquire obsidian, shifting its meaning from the raw material with no specific symbolic connotation among hunter-gatherer communities to a raw material with a specific value, history, content and meaning. Being connected with some major human innovations, first in the context of the Near East and subsequently in Europe, obsidian provides an additional possibility to approach and better understand basic structures that held the Neolithic world together. It further indicates that parallel with the food production and animal and plant domestication, another less visible and understandable process was taking place – the domestication of humans. On the basis of present evidence in the central Balkans this process occurred only in the Late Neolithic, i.e. with the start of the Vinča culture. The previous period (the Starčevo culture) that bears the name of the Early Neolithic one can rather see as only a transitional phase toward a typical Neolithic culture. The Mesolithic component is still strong during this early phase, which is obvious having in mind small quantities of obsidian. This is also related to all other aspects of the Starčevo culture. It is still unclear whether the meaning of obsidian starts to be transformed again with the formation of Copper Age cultures. Although later transition is not that radical when compared to the start of the Neolithic, the ‘metalization’ of Neolithic cultures occurs in the social context of a transition and, similarly to the Neolithic, this process is related to the expansion of new resources, techniques and ideas (Pluciennik 1998: 78). Hence it is plausible that the relationship between obsidian and humans takes a new form in a new cultural milieu.

BOGNAR-KUTZIAN, I., 1963, The Copper Age Cemetery of Tiszapolgar/Basatanya. Budapest: Akademiai Kiado. BOGNAR-KUTZIAN, I., 1972, The Early Copper Age Tiszapolgar Culture in the Carpathian Basin. Budapest: Akademiai Kiado. BRUKNER, B., 1977, Boundary Line between the WestMediterranean and Balkan-Anatolian Complex. Balcanica 8, p. 17-24. CANN, J. R., DIXON, J. E., & RENFREW, C., 1969, Obsidian Analysis and the Obsidian Trade. In Science in Archaeology, edited by D. Brothwell & E. Higgs. London: Thames and Hudson, p. 578-591. CHAPMAN, J., 1981, The Vinča Culture of South-East Europe: Studies in Chronology, Economy and Society. Oxford: British Archaeological Reports. EFSTRATIOU, N., 1995, The Neolithisation of Europe: The Greek Expirience. Poročilo o raziskovanju paleolitika, neolitika in eneolitika v Sloveniji 22, p. 63-82. GARAŠANIN, M., 1979, Centralnobalkanska zona. In Praistorija jugoslovenskih zemalja 2, edited by A. Benac. Sarajevo: ANU BiH, p. 79-212. GLÄSER, R., 1996, Zur absoluten Datierung der Vinča-Kultur anhand von C-Daten. In The Vinča Culture, its Role and Cultural Connections, edited by F. Drasovean. Timisoara: The Museum of Banat, p. 141-162. GREENFIELD, H., 1991, A Kula Ring in Prehistoric Europe? A Consideration of Local and Interregional Exchange during the Late Neolithic of the Central Balkans. In Between Band and States, edited by S. Gregg. Carbondale: Southern Ilinois University, p. 287-308. GREIF, T., 1995, Obsidian v neolitiku na območju med centralnim Balkanom in vzhodno jadransko obalo. Balkanica 26, p. 75-108. HALLAM, B. R., WARREN, S. E., & RENFREW, C., 1976, Obsidian in the Western Mediterranean: Characterisation by Neutron Activation Analysis and Optical Emission Spectroscopy. Proceedings of the Prehistoric Society 42, p. 85-110.

Author’s address

HODDER, I., 1990, The domestication of Europe. Oxford: Blackwell.

Boban TRIPKOVIĆ Faculty of Philosophy Department of Archaeology Čika Ljubina 18-20 11 000 Belgrade YUGOSLAVIA e-mail: [email protected]

HODDER, I., 1998, The Domus: Some Problems Reconsidered. In Understanding the Neolithic of North-Western Europe, edited by M. Edmonds & C. Richards. Glasgow: Cruithne Press, p. 84101. KARIMALI, L., 1994, The Neolithic Mode of Production and Exchange Reconsidered: Lithic Production and Exchange Patterns in Thessally, Greece. Ph. D. Thesis, Boston University. KERTESZ, R., 1996, The Mesolithic in the Great Hungarian Plain: A Survey of the Evidence. In At the Fringes of Three Worlds. Hunter-gatherers and Farmers in the Middle Tisza Valley, edited by L. Talas, Szolnok: Damjanich Museum Press, p. 5-34.

Bibliography AMMERMAN, A. A., & POLGLASE, C., 1993, New Evidence on the Exchange of Obsidian in Italy. In Trade and Exchange in Prehistoric Europe, edited by C. Scarre & F. Healy. Oxford: Oxbow Books, p. 101-107.

MAKKAY, J., 1996, Theories about the Origin, the Distribution and the End of the Körös Culture. In At the Fringes of Three Worlds. Hunter-Gatherers and Farmers in the Middle Tisza Valley, edited by L. Talas, Szolnok: Damjanich Museum Press, p. 35-49.

VAN ANDEL, T. H., & RUNNELS, C. 1988, An Essay on the Emergence of Civilization in the Aegean World. Antiquity 62, p. 234-247.

MALINOWSKI, B., 1979, Argonauti zapadnog Pacifika. Beograd: BIGZ. 188

B. Tripković: The Role of Obsidian in the Neolithic: A Symbolic Expression of Human Domestication? MOORE, A., M., T., 1985, The Development of Neolithic Societies in the Near East, Advances in World Archaeology 4, p. 1-69. NIKOLOV, V., 1998, The Circumpontic Cultural Zone during the Neolithic Period. Archaeologia Bulgarica 2, p. 1-9.

SCHIER, W., 1996, The Relative and Absolute Chronology of Vinča: New Evidence from the Type Site. In The Vinča Culture, its Role and Cultural Connections, edited by F. Drasovean, Timisoara: The Museum of Banat, p. 141-162.

NIKOLOVA, L., 1998, Neolithic Sequence: the Upper Stryama Valley in Western Thrace (with an Apendix: Radiocarbon Dating of the Balkan Neolithic). Documenta Praehistorica 25, p. 99131.

SEFERIADES, M. L., 1995, Spondylus Gaederopus: The Earliest European Long Distance Exchange System. Poročilo o raziskovanju paleolitika, neolitika in eneolitika v Sloveniji 22, p. 238-256.

ÖZDOGAN, M., 1994, Obsidian in Anatolia: an Archaeological Perspective on the Status of Research. Archaeometry 94, p. 423431.

SHERRAT, A., 1982, Mobile Resources: Settlement and Exchange in Early Agricultural Europe. In Economy and Society in Prehistoric Europe. Changing Perspectives, edited by A. Sherratt, Edinburgh: Edinburgh University Press, p. 252-269.

PANTELIDOU GOFAS, M., 1996, Ties Across the Sea in Neolithic Times: Attica and the Cyclades. The Colloquia of the XIII International Congress of Prehistoric and Protohistoric Sciences. Section 9 – The Neolithic in the Near East and Europe, Colloquium XVII, p. 11-20.

SREJOVIĆ, D., (ed.) 1988, The Neolithic of Serbia. Belgrade: Centar for Archaeological Research. STARNINI, E., 2000, Stone Industries of the Early Neolithic Cultures in Hungary and their Relationships with the Mesolithic Background. Quaderno 8, p. 207-219.

PERLES, C., 1989, From Stone Procurement to Neolithic Society in Greece. The David Scomp Distinguished Lectures in Anthropology, Indiana University, Indiana.

ŠARIĆ, Ј., 1999, Кремена индустрија најстаријих земљорадничких култура на тлу Србије. Ph. D. Dissertation. Belgrade: Faculty of Philosophy.

PLUCIENNIK, M. Z., 1997, Historical, geographical and anthropological imaginations: early ceramics in southern Italy. In Not so much a pot, more a way of life, edited by C. G. Cumberpatch & P. W. Blinkhorn. Oxford: Oxbow Books, p. 3756.

THOMAS, J., 1998, Some Problems with the Notion of External Symbolic Storage, and the Case of Neolithic Material Culture in Britain. In Cognition and Material Culture: the Archaeology of Symbolic Storage, edited by C. Renfrew & C. Scarre. Cambridge: The University of Cambridge, p. 149-156.

RADOVANOVIĆ, I., 1992 Mezolit Ðerdapa. Ph. D. Dissertation. Belgrade: Faculty of Philosophy.

TORRENCE, R 1984 Monopoly or direct access? Industrial organization at the Melos obsidian quarries. In Prehistoric quarries and lithic production, edited by J. Ericson & B. Purdy. Cambridge: Cambridge University Press, p. 49-64.

RENFREW, C., 1993, Trade Beyond the Material. In Trade and Exchange in Prehistoric Europe, edited by C. Scarre & and F. Healy. Oxford: Oxbow Books, p. 5-16

TORRENCE, R., 1986, Production and Exchange of Stone Tools, Cambridge: Cambridge University Press.

RENFREW, C., CANN, J. R., & DIXON, J. E., 1965, Obsidian in the Aegean. The Annual of the British School at Athens 60, p. 225-247.

TORRENCE, R., 1993, Ethnoarchaeology, museum collections and prehistoric exchange: obsidian-tipped artifacts from the Admirality Islands. World Archaeology 24 (3), p. 467-481.

RENFREW, C., CANN, J. R., & DIXON, J. E., 1966, Obsidian and early cultural contact in the Near East. Proceedings of the Prehistoric Society 32, p. 30-72.

TRIPKOVIĆ, B., 2001, Uloga opsidijana u neolitu: utilitarni predmeti ili sredstvo prestiža. Glasnik Srpskog Arheološkog društva 17, p. 21-42.

RENFREW, C., CANN, J. R., & DIXON, J. E., 19688, Further analysis of Near Eastern obsidians. Proceedings of the Prehistoric Society 34, p. 319-331.

TYKOT, R.H., 1996, Obsidian Procurement and Distribution in the Central and Western Mediterranean. Journal of Mediterranean Archaeology 9.1, p. 39-82.

RENFREW, C., & DIXON, J. E., Obsidian in Western Asia: A review. In Problems in economic and social archaeology, edited by G. de G. Sieveking, I. H. Longworth & K. E. Wilson. London: Duckworth.

WHITTLE, A., 1996, Europe in the Neolithic. The Creation of New Worlds. Cambridge: Cambridge University Press.

RENFREW, C. & BAHN, P., 1991, Archaeology. Theories, Methods and Practice, London: Thames and Hudson Ltd.

WILLIAMS THORPE, O., WARREN, S. E., & NANDRIS, J. G., 1984, The Distribution and Provenance of Archaeological Obsidian in Central and Eastern Europe. Journal of Archaeological Science 11, 183-212.

RUNNELS, C., 1989, Trade Models in the Study of Agricultural Origins and Dispersals. Journal of Mediterranean Archaeology 2/1, p. 149-156.

WILLMS, C., 1983, Obsidian im Neolithicum und Äneolithicum Europas, Germania 61(2), p. 327-352.

RUNNELS, C., & VAN ANDEL, T. H., 1988, Trade and the Origins of Agriculture in the Eastern Mediterranean. Journal of Mediterranean Archaeology 1/1, p. 83-109.

189

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

190

N. Syriamina: Prehistoric Art of the Neolithic Bronze Age

PREHISTORIC ART OF THE NEOLITHIC BRONZE AGE Natalie SYRIAMINA

Résumé : Cette article présente la méthode novatrice de l’analyse des tendances de naissance des cultures préhistoriques du Néolithique et de l’Âge du Bronze sur les différents territoires. L’auteur a utilisé et a traité non traditionnellement de nombreuses oeuvres des philosophes antiques connus, l’ouvrage d’Inde ancienne Rigueveda. L’auteur a trouvé l’occasion opportune d’utiliser et comparer les progrès de la science et de la technique moderne. Les résultats des recherches de l’astronome indien connu Dzhoyante Narlikare utilisés dans cet article sont toujours actuels. Le point de vue non traditionnel sur certaines sphère d’activité et de développement historique de la société primitive donne la possibilité d’amener à son achèvement logique l’analyse de nombreuses comparaisons et parallèles existant dans la mythologie comparatif. Abstract: This article presents a new method of analysis of the tendency for conception of the artistic cultures for the Neolithic and Bronze Age in different territories. Survey of the artistic cultures and of comparative mythology displays the conception is permitted to make the interesting, non-perceived and original conclusions. Numerous nontraditional and known works by antique philosophers were used, as well as the ancient Indian memorial Rignvedas. Use and comparison the achievements of the modern sciences, such as, for example, the research of the known Indian astronomer Jayant Narlikar, are up to date. A nontraditional view of the certain sphere of the action and historical development of the prehistoric society permits to complete absolutely the analysis of the numerous comparisons and parallels are in the comparative mythology.

SYSTEMATIC SURVEY OF CONCEPTION AND DEVELOPMENT

uniform cosmogonic concept for all territories. The newest numerous discoveries and achievements of archaeology (end of ХІХ-ХХ cent.) also confirmed and fixed the scientific concept, that Mythology - this is symbolical reflection of the real world.

The Neolithic epoch is connected with stabilization of all natural processes of the last Interglacials (Ruddiman & McIntyre, 1981). The characteristic of the advanced Neolithic Age (4000-2000 B.C.) have the tendency of occurrence and development constant with the most bright pronounced generations of ethnic cultures in different regions.

So, on the basis of the data given in tables 1,2 it is not difficult to make the comparative analysis of development Neolithic Artistic culture and to systematize conceptual reasons on the basis of which there was this process.

Most authentically, the process of formation of ethnic attributes in Neolithic artistic cultures began much earlier, and to 4 thousand B.C., as the object archaeology set testifies to it, has achieved the certain level. Nevertheless, objects of a classic archaeology, isolated and insufficiently filled, do not give an opportunity to look after character and motives of tendencies of process of conception and development of ethnic attributes of Neolithic artistic cultures. Probably, process of formation of ethnic generations is necessary to connect not so much with social and economic changes, which certainly took place, but, first of all, with transdemographic consequences, which in turn, are connected to global both local catastrophic geological and climatic transformations.

We shall note some most important characteristics of Neolith for a history of development of mankind. In the Neolithic, beginning from 7,4 thousand to BC (depended on territories) build the first stone’s constructions – megaliths, which can be attributed to initial monumental constructions. Megalith phenomenon (about which it is necessary to speak separately) will play the important role in becoming and formation of spiritual development of mankind, it becomes initial school of modern architecture, base on the basis of which the Modern Art of building arises. Types of megalithic constructions in all territories repeat and their initial cult assignment, despite of following utilitarian use, also are similar in all territories.

Tendencies and circuits of development Neolithic artistic cultures in different territories are shown in table 1. The review, are carried out on territories of Northern Hemisphere, in part grasps Indonesia and North-East Africa.

The second important attribute, describing the Neolithic, is resettlements legendary Aryan tribe (the Indo-European language group, or rests of it) in news territories. To the theory of this tribe’s origin from Northern territory the majority of the Indian scientists (Tilak 1893) is declined.

Neolithic artistic cultures have extremely symbolical conceptual sense, therefore sources of formation it’s necessary to search in mentality of Prehistoric world. Namely, in creation of mythology, that with later ethnic interpretations, has, apparently from table 2, the common roots and the

Scientific theories, that are based on the stylistic analysis to ancient Indian literary - mythological memorial Reignvedas, are declined to that resettlements was held by three waves in 191

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe Table 1. Territorial characteristic of artistic culture at Neolithic-Chalcolithic.

I. Territories with constant development and with the exceptional culture. Period 12-4 thousand B.C.

Paleolithic Mesolithic 10-7 thousand B.C. Neolithic 6-3 thousand B.C. Bronze Age End of 3-1 thousand B.C.

Name Egypt (kinghome Upper and Law Egypt; joint Egypt – 341dynasty11341B.C.) Russia Space of valleys by rivers Volgo-DonneBaikale Forest space of NorthEast Europe.

Difference Monumental building: pyramids, sphinx.

Characteristic Agricultural culture.

Mobile and Rock Art.

Finnish-Ugor community

Building of frame-work. Settlements are reinforced by banks and ditches; Wood’s and stone’s anthropho-zoomorphic minor plastic; petroglyphs. Art “Style of Beast”; frescos (village Shipovo, Saratov-region); water-pipe and severage.

Cultures jamna and of tumulus. Cultures of Skiphs and Sarmaths

Finland Building of frame-work. Mesolithic Wood’s and stone’s anthropho-zoomorphic Neolithic minor plastic 6-3 thousand B.C. Art “Style of Beast”; Bronze Age Middle of 3-2 thousand B.C. India (Bcharat) Ancient civilizations at the slopes of rivers Inda Paleolithic and Ganga Neolithic Stone’s and clay’s sculpture Middle of 3-2 thousand B.C. Northern India Oval clay’s dwellings Middle of 2-1 Ancient literary memorial-Reignvedas thousand B.C. Rock building: chajtija 4-2 cent.B.C. High sculpture’s Art 2-1 cent.B.C. Syria Rock Art Paleolithic Stone’s and clay’s anthropomorphic figguring East of Mediterranean, Neolithic 9-4 thousand B.C. Lebanon, North and South of Chalcolithic City-state Turkey 3-end of 2 thousand B.C. Jordan Rock Art of Kil’va Paleolithicic Western Asia City-state Jerichon Neolithic Megaliths for died, cult’s temples with Art graphic 5-3 thousand B.C. Clay’s dwellings Chalcolithic I I. Territories, that have not archaeological confirmation before Middle Asia Rock Art in grotto Shachta Mesolithic Tajikistan Rock Art in Zaraut-Tsaj Neolithic Primitive craft of pottery making 6-5 thousand B.C. Uzbekistan Petroglyphs Stone’s dwellings with frescos (Djeitun) Iraq Protourban dwellings: composite Mesolithic architectural assemblages Neolithic End of 8-5 thousand B.C. Japan’s islands String’s pots and clay’s figurines of idols – non Mesolithic analogy Neolithic Anthropomorphic masks “domen” – non 8-6 thousand B.C. analogy 5-middle of 2 thousand B.C. Chalcolithic Migration of new tribes End of 2 thousand B.C. Indonesia Rock Art Mesolithic Neolithic Schematic zoomorphic drawings in caves 7-5 thousand B.C. Menhirs; step’s tumulus; guigant’s statues; end of 5-2 stone’s graves thousand B.C. 192

Finnish- Ugor community Culture jamna.

Charapp’s civilization

Veda’s civilization State of Maur’ja Culture of Djabrude Samara’s culture

Gissar’s culture

Culture’s of Kchassun and Chalaf

Djemon’s culture Domen’s culture

State of Jamato

N. Syriamina: Prehistoric Art of the Neolithic Bronze Age

World of Egej Protourban architectural assemblages; Mesolithic Monumental sculptures; synthesis of Art Neolithic 7-big.3 thousand B.C. Phoenicia Primary city’s settling Mesolithic (Lebanon, north-west of Neolithic 7-6 thousand B.C. Syria) Temple of obelisks in Bible 5-4 thousand B.C. Asia Minor Sity-state Jerichon Mesolithic Clay’s dwellings, underground’s temples Palestine Neolithic 7-6 thousand B.C. Jewelry Art; urban settlements; undeground Chalcolithic sette -Barrbelite 4-3 thousand B.C. Cyprus Round clay’s dwellings Mesolithic 7-6 thousand B.C. Differents shapes of stone’s pots; zoomorphic Neolithic pot sec.part 3 thousand B.C. City,temples; large bows with anthropoBronze Age zoomorphic painting 14-11 cent. B.C. Central Asia Neolithic Primitive craft of potery 6-5 thousand B.C. Turkmenistan Many-rooms and raw-brick’s dwellings; 4-3 thousand B.C. Protourban large settlements Elam Neolithic Craft of potery; koreplastic; glypthic 4-rirst part of 3 (North-west of Iran, valleys of rivers thousand B.C. Temples, zikkurats five-level Charush-Cherche – 3 thousand populations-migrants) 5cent. B.C. China Clay’s dwellings; differents exceptional craft of Neolithic pottery; 4- 3 thousand B.C. Craft of pottery from white clay Bronze Age Origin bronze life’s and cult’s vessels with bas2 thousand B.C. relief – non analogy; end of 2 thousand Come of analogies states B.C. ІІІ. High level Art on the territories with law previous level Neolithic Afganistan Clay’s figurines of womans and beasts End of 4- 3 thousand B.C. Pakistan Primitive settlement; Neolithic City-state Kcharappa 3-big.2 thousand B.C. Iran Clay’s dwellings with wall’s painting; Neolithic Vasseles with “Style of Beast”; 5- 4 thousand Anthropo-zoomorphical bone’s and clay’s B.C. figurines Ukraine Round clay’s dmellings are united in Neolithic (Dnipr-Dniester region) settlements are reinforcemented by numeros 4- 3 thousand banks and ditches B.C. Breatten Irelands Dwellings – frame-works; Neolithic Round non-living settlements are 4- 2 thousand reinforcemented a way banks and ditches; B.C. Monumental stone’s building – Hendge’s, tumulus, dead stone’s gaves. Archipelago of Malta Neolithic Megalithic building of differents directions: for living, for worships, for dead (islands Malta, Gozo, 4,3-2 thousand Gjantija) B.C. Ireland Megalithic constructions: dolmens, chromlex, Bronze Age from middle tumulus with dead gaves 1thousand B.C. ІV. Territories with civilisations. Egypt; Shumere; NeolithicAkkade; Babilone; Chalcolithic 4-2 thousand B.C. Indian; China; Indonesia; SouthWestern Asia; Azia Minor; Cyprus; Greece.

193

Tirr, Ugaritt, Sidone, Bible. Natuph culture

Syntesis of Crete’s and Chette’s directions in Art

Important influence on the civilization of Shumere

Culture of Janshao Culture Lun’shane State of Shane-Ine

Earlier agricultural culture Earlier agricultural culture Culture of Kcharappa Agricultural culture

Trippil culture Agricultural culture of tribes – migrants from Northen East

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

territories of Syria, Palestine, Northern India and East Iran on an extent with II thousand up to 9 cent. to BC (Reignvedas 1972).

this process has considerably amplified. Result became that in territories which marked a joint of two different classes of cultures, in Bronze Age (IV-II thousand B.C.) all known initial civilizations (table I, part IV) were formed.

Such dating is considerably conditional. At significant continuance of time the most possible date of creation RV, is date 6-4 thousand B.C., to which separate scientists (Tilak 1893; Tchelysheva 1986) are declined. Comparisons of some data: antique sources (Dіodorі. Biblioteca Historica), last archaeological researches (5), and also the data given in table 1 (part I, II, III) resolves make corrections the dates of several waves of resettlement of a legendary tribe which, certainly, has made the significant contribution to formation of Neolithic Artistic cultures, in the last resort, of their spiritual filling.

The short characteristic of mentality of classes of Neolithic Artistic culture the subsequent. “Northern” culture is more rational, stylized (that is speaking mathematical language coded), constructed on understanding of Universal cosmogonic unity of all processes, however is spiritually filled. “Southern” culture considerably practical, but constructed on sensitive principals (it is possible, influence by one more, not existing by later culture). As it is possible to see from the scheme, the territory with a “southern” class Neolithic Art belong to one geographical breadth: Egypt, India, China (Southern and Minor). This list includes Mexico, which is not present on the scheme (figure I), but its art, taking into account the latest archaeological research, also belong to same class of Neolithic Art. Probably, all mentioned territories in earlier period were association considerably smaller distances. The second more southern zone its are - Brazil, Southern Africa, Indonesia has similar characteristics. The modern situation on a world map brings to a conclusion, that all three territories in Prehistoric were of one, and the centre of cataclysm, owing to which they have received modern distances, was in Southern Africa. So, the important conclusion, which can be made on the basis of the systematic survey of the data given in table 1 the following. The Population, who from 9 thousand B.C. had the constant and enough advanced level of Art (see part I), it became, most authentic, the successor of the previous considerably advanced societies, which for that period did not exist (were perished?). Concentrations of such societies were at least two - in the North and in the South (scheme-figure I).

Making the systematic survey of data in table 1, one may make the conclusion, that most early and mass was resettlements of tribes, that may belong to the Indo-European language group at the territory of the western coasts of West Asia (Jordan, Palestine), to the North to Iran, to Iraq, and also at the islands of Mediterranean – Crete, Sardajgne, Archipelago of Malta – 9-4 mill. B.C. The second wave of resettlement of this group became arrival of agricultural, more advanced tribes from North-East to British isles - 4- off 3 mill. B.C. (Wood 1981). Most authentically, that to 2 mill. B.C.in territories of East Iran and Northern India, where advanced civilizations were formed, there was a further development of tribes, that have moved in considerably earlier period, whether came from the next territories, were settled with them in previous times. This process has found a reflection in stylistic formation of RV which becomes by then written memorial, as has created a basis for the most widespread version of dating of memorial2 mill. B.C.

The Myths given in table 2, were generated and written down (or copied) when modern ethnic formations were already formed. In the table they are placed under the name of ethnoses, which have inherited them from sources existing up to them. In table 2 not included mythology of territories with the advanced civilizations. The civilisations have arisen on a joint of territories where there passed process of diffusion of two classes of Neolithic Art. (see above): namely Shumir, Babylon, Akkad, Jewish civilization. Mythology of these societies as tells about similar events, but covers later period of development (II-I thousand B.C.) local territories and, consequently, is confirmed with numerous archaeological researches.

The data, mentioned in table 1, are based on the objects of the academic archaeology. So, the comparative territorial analysis of development of Neolithic Artistic cultures shown in table 1, allows to create more objective picture of process in time and spacious. The further survey of data is given in table 1 allows add to the theory of resettlement of a legendary tribe at the new territories, formulated on the basis of stylistic research of RV-memorial. On the basis of the systematized data in table 1 are reflected graphically on the scheme (figure 1), becomes obvious, that Neolithic Art was formed on the basis of two classes of cultures which had different stylistic directions and spiritual filling.

I shall try to find confirmation to the conclusion made above on the basis of the survey of mythologies in different territories, which are resulted in table 2. The basic and indisputable statement which follows from ordering the data given in table 2 is the following. Myths of the basic territories with the oldest and constant periods of development of Neolithic artistic cultures, tell about the same events of the earliest period of development of all mankind. It may mean, that at them uniform considerably older source was.

Two classes of Neolithic cultures (we shall name their “northern” and “southern”) the long constant period, starting approximately with 9 thousand B.C., developed on same, all over again the limited territories it is isolated (see table 1, part I). Process of diffusion eventually began is caused, probably, demographic tendencies. To V-IV thousand B.C. 194

N. Syriamina: Prehistoric Art of the Neolithic Bronze Age Table 2. The comparative characteristic of spiritual culture -Mythology of Neolithic – Bronze Age.

The name of territory-owner of Myths Egypt** Prescription of existence of a civilization in territory of Egypt proves to be true several indirect sources: 1. The basrelief of a ceiling in a temple of goddess Khator (the Valley of Pharaohs) represents accommodation of stars, which could be observed from the Earth many thous.ago. 2.Simpsyllij has written down, that Egyptian priests some hundreds of thousand years have data on astronomical supervision on an extent. India Similar (with a temple of goddess Khator) images can be seen in temples of Northern India of the period of Paleolithic civilizations.

The name of written source “Aegytica” Manepho (the main priest of Geliopol temple300 years BC) – Herodoth wrote

Greece** Classical archaeology fixes on their territory, beginning from 12-8 thous. BC—Mesolith.

Gerodoth-“The work and days”, Ovidiy“Metamorphoses”, Plato - “The state”.

RV collection of hymns, written down on the territory of North India – 2 – the middle of 1 thous. BC

The characteristic of the main conception Mythology concept-- about creation of the mankind in the Earth: There were 8 gods (the first among them was Hephest), that banished to the Earth and have created 12 demigods (among them Herakle). Gods prevailed Earthmen 13900 years. Demigods having based on the Earth of 12 human tribes and civilizations. After them kings – 1817 years dominated; a dynasty from 30 kings from Memphis – 1790 years; a dynasty from 10 kings from Memphis – 350years; " Spirits died " – 5813 years then dominated. After that kings from people - pharaons (them recollects Herodoth) dominated – 341 dynasties of pharaons --this period is fixed in written sources and confirmed by archaeologicts of differents countries (15-10 thous. BC - age of Cheopth pyramid). This date put forward recently by paleoastronomer Robert B'juvelom,. Mythology concept – cosmogonic concept about creation of the World (the Earth? The Universe?): The cosmogonic RV concept covers earlier period of a history of the Earth, than it is submitted in the Egyptian mythological chronology. It is possible to divide into three stages: І - the Earth in structure of the Universe; ІІ - the Earth is the center of the Universe; ІІІ - a history of development of the Earth, as planet. First of all hymns sing of some initial phenomenon which exist before formation of the terrestrial world. Hence, the first stage can be identified with evolution of development of Space - birth, development and destruction of Stars. All these processes, that occur irrespective of external forces but only on the basis transformations of high energy. It is extremely accessible described known Indian astrophysicist Jajant Narlikar (6). The second stage in hymns begins with occurrence Іndra. This stage is marked by mountain, that floats at Ocean. Іndra letting out dual positive forces: Fire - Augni-Syrja-Sun and Water – ComaRiver. Since, Fire becomes the Sun. Water turns to four rivers, which cross wisely flow in four directions. This period can be treated doubly: 1. The hill in water is a process of formation of the Earth, as planet and creation of its constant site in the Galaxy. The dragon is internal force, which due to nuclear reactions turns to gravitational force, which provides a site of the Earth in the Galaxy. Four rivers are four magnetic streams, which have designated poles and cardinal points of the Earth. 2. The hill is an primary continent, the division of which in the future on four continents was predetermined to four primary rivers are placed as the Cross views. The third stage of Vedas hymns reflected events, which are similar to Egyptian mythology, that is creation of mankind The mythological concept - as well as the Egyptian chronicle, reflects historical process of concept and development of mankind on the Earth (it is authentic, is the continuer of Egypt’s mythology). All authors submit five stages of development. Plato’s interpretation reflected processes not only in historical, but also in a social foreshortening. The I stage - Gods and "those who born of Earth" lived. Thay were not ancestors of alive Earthmen. Alls perished. The ІІ stage - there are original Greeks (?). But they perished also. Those survived those, have based the first dynastic state. (Cities-states). The ІІІ stage - " the second state " - occurrence of aristocracy, the kings - rulers, the new cities - states, terraced agriculture, the first stationary ground dwellings, cult constructions. Territory of moving - mountains, foothills. (the Neolithic - see table 1). The ІV stage - " the third state " - territory of moving – valleys. Establishment Troya and its destruction with arrival of other tribes. The V stage - " the fourth state”. 195

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe The name of territory-owner of Myths Japan

The name of written source The literary written memorandum (712 year), “Kojiki” written on the base of chronicle “Nichoni”.

China** There were fixed about 100 pyramids at the beginning of XX cent. in the Sichuan valleys.

Chronicle “ShaoKin”.

The characteristic of the main conception Japanese mythology, which is based on legends, that enter the name only in the beginning from 8 cent., reflects, nevertheless, not the oldest period of human development and as a matter of fact is similar to mythologies of other territories (see above). The Godin Japanese mythology has no precise anthro-pomorphic type. His symbol is the nature, all world, which surrounds. Generalization of these postulates and together with initial laws of a Universe, has found reflection in the plastic form of ritual vessels by culture Djemon. Result of a Japanese mythology became occurrence of religious system Synto -“ Way of Gods" in epoch of the state Jamato. This religia has received an embodiment in temples in the end of I thous. B.C. In the chronicle of the Chinese emperors, is dated 5 thous. BC, records about pyramids, that was built by "Sons of the Sky ". The Chinese emperors were kept their descents.

As we see, chronology of Egypt’s Mythology, which antiquity unfortunately is not confirmed, not only archaeology, but also other modern joint and coordination sciences, nevertheless, does not maintain comparison to events represented in myths on which was under construction ancient Indian literary RVmemorial. Taking into account events are reflected in myths which became the basis for creation RV, it Myths is not only the most ancient, but also becomes base for forming of Mythology of other territories. You see not only the most antiquity chronology of events about the creation of the Universe represented in RV which, certainly, precedes creation of mankind (Egypt, Greece), but also for symbolism submitted in RV knowledge which very successfully represents real similar processes (see table 2 - India), speaks well this idea. Symbolism of RV-memorial - the basic difference from mythology other territories which, as a matter of fact, are primitive interpretations of the events reflected in Vedas hymns. Shall try to compare some events and dates, that are reflected in Egyptian and Greek mythology with achievements of such sciences, as archaeology, anthropology, geology, paleoastronomy and others.

archaeology. It necessary the comparison of the data of different modern sciences, as more, as possible. Such analysis given in table 2 carried out also submitted in article. On the basis of last researches of adjacent and coordination directions the modern science owns following information: 1. In a geological history of development of the Earth existed global and local catastrophes: -change of a direction of a revolution of the Terrestrial axis that has resulted in replacement of poles and has taken place approximately 26-23 thous. B.C. Accordingly, territories which are placed most close to old and to new-form poles must have been ruined more: - the catastrophic rise of the World Ocean level (global “flood”), that owing to Space catastrophe, has taken place, approximately, 15-10 thous. B.C. Territories, that have most suffered, are in a stage of definition; - local “floods” occurred approximately in the period 52 thous. B.C. - territories known from the Greek mythology and are determined by last archaeological research: South-Western Asia (Syria, Iraq, Palestine, Jordan), North-East Caucase, South-East Ukraine and South-West Russia, Central Asia.

Known modern historian M. Tchmychov, (Tchmychov 1992) makes interesting comparisons of results of researches in archaeology with Greek mythology. Its certificates about IV stages of development of mankind (see table. 2). On M. Tchmychov each period in Greek mythology corresponds to the periods of Prehistoric development (on the basis of archaeological research), and begins from Mesolithic (X thousand B.C.). Comparisons are really interesting, at first sight, coincide, and are confirmed subjects of archaeology. But, in the Greek mythology speech not necessarily about territory of Greece (see table 2: Greece - II, III stages). As we see from table 1 (p. II), the first cities - states (III stage of the Greek mythology) arose not necessarily in territory of Greece.

2. The anthropological history of development of mankind owns the following the data: - Neanderthal man, as the species (Homo sapiens) who appeared, approximately, 300-200 thous. B.C. ; - Cro-Magnon Man, as the species (Homo sapiens) who appeared, approximately, 40-35 thous. B.C. Approximate for 10 thousand years of coexistence of the Neanderthal Man and Cromagnon Man, the Neanderthal Man, as the type, disappears - approximately 30-25 thous. B.C. At least, comparison of two pairs dates in a geological and anthropological history can be identified with Greek and Egypt’s mythology:

Obviously, ripened objective necessity to determine an ancient history of mankind not only on the basis of subject 196

N. Syriamina: Prehistoric Art of the Neolithic Bronze Age

The I pair (25-26 thous. BC)

The change of direction of rotation of the Earth

Disappearance of Neanderthal man type

The II pair (15-10 thous. BC)

Catastrophic rise of the world Ocean level

The beginning of the jump-kind progressing development of the Neolithic Artistic cultures

carried the concept of spirituality of most earlier period. Because, types of mass Megalithic building from 7-4thous. B.C. in most of territories, are carries this spiritual concept, that was filled up a society, who perished, accidentally 1510 thous. B.C. Hence, the carried out systematic analysis of the data given in table 1,2 allows to draw the following final conclusions: 1. Development of Neolithic artistics cultures occured a progressing jump, all over again in separate territories. Up to II thous. B.C. it covers significant territories to the North from Equator. This process began, approximately 10 thous. B.C., and was connected, in turn, with progressing tranthdemographic changes in a society, that have taken place owing to global catastroph.

The first (I) pair of dates is identified with I stage of the Greek mythology. The second (II) pair dates is identified with II stage - archaeological confirmations of the first settlers in the Greek territory (12-8 mill.B.C). Egypt’s mythology “illustrates” pairs of dates are connecting geological and anthropological histories of development even more precisely. The rule of “human” kings - pharaohs began approximately 13-10 thous. B.C., that is after global accident. Accordingly, the Egyptian civilization arisen on base of existing up, more advanced to “flood” (“demigods” - see table 2). About it’s mentioned in the Greek mythology (destruction of Atlantide - by Plato). Addition of all figures, that above mentioned in the Egyptian mythology (table 2), we shall receive number 37340. But it – the Age of approximate appearance the new Human type on the Earth 40-35 mill. B.C.

2. Myths, on the basis of which ancient Indian’s RVmemorial was created (6-4 thous. B.C.) are origin knowledge and a source for Mythology of all territories, where were formed advanced Neolithic artistics cultures. 3. Conceptual bases of origin knowledge were reflected, all over again, in subjects of a life, in graphic Art and in applied Art, but since 7-4 thous. B.C. - in building of stone’s constructions with cult assignment (8). Author’s address Dr. Natalia SYRIAMINA Naukova str. 64/19 79060-Lviv, UKRAINE

On the basis of these comparisons becomes obvious, that civilizations of Greece and Egypt arose almost simultaneously. Distinction is in some thousands. But, as Egypt’s mythology covers older period, it is possible to make the fellows assumptions. The territory of Egypt was origin territory, that was occupied after catastrophe by representatives of a society, who were carriers of initial knowledge. If to take into account, that archaeological confirmation of settling of the Greek territory, begins from 12-8 thous. B.C., and that the top border of date of probable cataclysm (15 thous. B.C.) is approximate time of settling of territories of Egypt, Greece - is one of secondary territories, that were occupied after Egypt.

Bibliography DJ. NARLIKAR. 1985. The violent phenomena in the Universe: Moskow. REIGNVEDAS. 1972. Selected Gymns. By T.J.Jelizarova: Moskow. RUDDIMAN W.F., MCINTYRE A., 1981. The North Atlantic Ocean during the last glasiation, Palaeogeogr., Palaeoclim., Palaeoecol. V.35: p.145-69. SYRIAMINA N. 2000. Megalithic phenomenon: Lviv-Complex Tchortova Skelia, The Ethnology Notebooks 4(32)// Institut NAS of Ukraine. Lviv: p.681-7.

Still it is necessary to add, that in territories of the East (China, Japan, Indonesia) Mythology is under construction on the basis of the same, common source for other territories. Nevertheless, as Eastern mythology is written down in later period, it’s more nature, and is considered philosophically. The Japanese mythology most successfully embodied and reflected not only in a subject set of Neolithic artistic culture, but in types of cult’s constructions also. Temple’s compositions, have begun to arise in territory of Japan in the end I thous. B.C., but their design and a spatial composition

TCHELYSHEVA I.P. 1986. The Ethnical ideas in world-out-look by Vivekanda, B.G.Tilak and Aurobindo Tchosh: Moskow. TCHMYCHOV M. 1992. The eight thousands of the archaeological cosmology in Space: Kiev. TILAK B.G., 1893. The Origin or Researches into the Antiquity of the Vedas: Bombay. WOOD D.J. 1981. Sun, Month and ancients stones: Moskow.

197

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

198

R. Grifoni Cremonesi: Le rituel funéraire au cours du Néolithique italien

LE RITUEL FUNÉRAIRE AU COURS DU NÉOLITHIQUE ITALIEN Renata GRIFONI CREMONESI

Résumé : On donne un cadre d’ensemble sur les problèmes concernant les changements du rituel funéraire au cours du néolithique italien, et l’utilisation de sites d’habitats ou de grottes par rapport aux changements de l’organisation des sociétés et par rapport aux aspects cultuels. Abstract: A general overview of the Neolithic burial in Italy is presented. The aim is to point out the change in the ritual practices, the development of cemeteries and the use of various features in settlements and in the caves, as connected in the social organisation during the Neolithic.

Ces aspects ont été étudiés en Italie2 en tenant toujours compte de la réalité de la donnée archéologique: à présent le problème des cultes dans la préhistoire, auparavant considéré comme peu intéressant, est souvent abordé avec des méthodologies qui préfèrent utiliser des modèles strictement théoriques sans prêter vraiment attention aux données de fouille ou à la littérature.

Le Néolithique italien nous offre une documentation suffisamment riche en ce qui concerne les idéologies funéraires mais, malheureusement, cette relative richesse des données contraste avec une certaine pauvreté des connaissances, surtout pour les vieilles fouilles, concernant tous les éléments nécessaires à la reconstruction des phénomènes liés aux rites et aux implications de type sociologique: il nous manque souvent les données relatives aux structures tombales, à la positions des squelettes, à la compositions du mobilier funéraire, aux données anthropologiques (sexe, age de mort, pathologies, etc.) (Bagolini et Grifoni Cremonesi 1994; Mordacci et Grifoni Cremonesi en presse).

J’ai toujours soutenu que la question des cultes et des rites funéraires, quand elle est abordée sans précaution et sans une rigoureuse analyse des données, risque de nous conduire à des reconstructions de type socio- économique et de type religieux fondées sur un très simpliste comparativisme ethnographique ou, souvent, à une pure fantaisie. Pour cela j’ai été accusée de scepticisme pragmatique (ou pragmatisme sceptique?) ou bien encore de pessimisme qui empêcheraient le développement de la recherche: je voudrais seulement souligner qu’on peut et doit certes avancer des hypothèse et essayer de reconstruire de quelque façon les sociétés du passé, mais on ne peut pas le faire en ignorant presque totalement les problèmes réels de fouille et de méthodologie (Grifoni Cremonesi 2000).

Malgré l’extrême complexité du néolithique italien, avec ses différences entre Nord, centre et Sud de la péninsule, et entre côtes thyrreniennes et côtes adriatiques et les aspects des îles, il est néanmoins possible de reconnaître dans les phénomènes funéraires des constantes et des variables qui nous permettent de reconstruire, bien qu’en ligne générale, un tableau général de référence.1 Il faut dire, avant tout, que, bien que l’on possède un certain nombre de sépultures en grotte et dans les villages, il n’est pas possible de distinguer toujours avec certitude l’existence d’espaces bien définis et séparés pour les rites sépulcraux et que le rapport entre espace des morts et espace des vivants dans les diverses sociétés n’est pas encore très clair (Grifoni Cremonesi en presse).

En tout cas, la connaissance du néolithique italien présente bien des difficultés, pour la position et la conformation géographique et pour les divers contacts que les régions italiennes du Nord, centre et Sud ont eu avec d’autres cultures (danubiennes, balkaniques, franco ibériques, etc.) constituant donc une mosaïque très complexe dont on ne connaît pas encore bien les processus de formation.

Un autre problème se présente souvent dans le néolithique italien, concernant les phénomènes cultuels liés de quelque façon aux rites funéraires et généralement considérés comme faisant partie d’une idéologie liée à l’agriculture et au cycle de la végétation avec les mystères de la mort et de la résurrection (Grifoni Cremonesi 1986,1994, Di Fraia et Grifoni Cremonesi 1996).

Pour le néolithique ancien on a une assez bonne documentation surtout pour les régions centro méridionales (Pouilles, Basilicata, Abruzzes et Marche) tandis qu’on sait très peu de choses pour les régions orientales et septentrionales où cette période est quand même très bien documentée.

1

2

Pour une synthèse sur le néolithique italien cfr., Atlas du Néolithique européen, vol. 2A, J. Guilaine ed., Liège.

199

Radmilli 1963, 1975; Cremonesi 1976, 1985-90, Grifoni Cremonesi 1969, 1986, 1994, 1996, 2000.

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Dans les Pouilles, où le néolithique apparait vers 6000 b.C. avec la «ceramica impressa», on connaît différentes typologies de structures tombales:il s’agit généralement de simples fosses ovales creusées dans le terrain ou quelque fois dans le rocher, en proximité ou à l’intérieur des habitats; je rappelle seulement Rendina (Cipolloni Sampò 1972-77) et Favella de Sibari (Tinè 1987), comme les plus connus): à Rendina I un enfant était enseveli dans une structure délimitée par des trous de poteaux au centre du village. Le plus souvent on trouve un seul individu, mais il y a aussi des femmes avec enfant, à Ripa Tetta (Tozzi e Verola 1990) il y avait une fosse avec deux squelettes et à Gandoli il y en avait quatre (Gorgoglione 1987). Recentement on a découvert des sépultures, avec adultes sur le coté droit, en fosses entourées et couvertes de pierres dans l’habitat de Balsignano (Bari) où il y avait aussi, à coté d’une grande cabane, une petite cuvette contenant les fragments du crane d’un adulte, incinérés (Fiorentino et Alii 2000). La couverture de la fosse avec des pierres plates se trouve aussi à Torre Sabea (Lecce), un peu éloignée de la structure du Néolithique ancien: on y avait déposé un adulte male à coté duquel était une petite fosse contenant du blé et de l’orge (Grifoni Cremonesi et Alii c.s.). Peut être qu’une des sépultures du Pulo de Molfetta, la n. 30, qui contenait un vase avec 27 petits galets et était couverte de pierres plates, appartient au néolithique ancien (Radina 1999). Un phénomène très particulier est signalé à Samari, à proximité de Torre Sabea: il s’agit de fosses creusées dans le rocher sans aucun rapport avec un habitat; trois d’entre elles étaient entourées et couvertes de pierres comme à Balsignano et deux d’entre elles contenaient un mâle, une autre un adulte et un jeune Il y avait en outre des amas d’ossements (un avec les restes de six individus et un autre avec trois), des pavements de galets et des foyers avec des cendres blanches. Un des inhumés avait une meule à coté du crane (Orlando 1997). Des amas d’ossements, dus à une déposition secondaire, sont connus aussi à Santa Tecchia (Foggia) (Cassano et alii 1987). On a interprété ce phénomène comme un rite d’éloignement ou de séparation On connaît enfin des sépultures dans des niches creusées dans les parois des grands fossés du Tavoliere et quelques sépultures dans des grottes. Très particulière est la récente découverte de restes incinérés déposés dans des fosses à coté d’une structure d’habitat à Modugno (Bari) (Fiorentino et alii 2000), qui confirme, avec le crâne de Balsignano, l’existence de ce rite dans le néolithique ancien, auparavant connu seulement à la Grotta Continenza dans les Abruzzes (Grifoni Cremonesi 1987, Barra et alii 1989-90). Par rapport à l’Italie du Sud, où les sépultures dans les zones des habitats sont plus fréquentes, au centre d’Italie (Abruzzes et Marche) on a plus de données concernant des grottes: pour les habitats on connaît des ensevelissements en fosses à Lama dei Peligni (Manzi et Macchiarelli 1989) et à Maddalena di Muccia (deux femmes dont une brachycéphale) (Corrain et Capitanio 1968) mais les données sur les structures tombales sont très incertaines (les relations

parlent de fonds de cabanes ou de fosses) sauf que pour la fosse entourée de pierres de Lama dei Peligni. A Ripabianca (Corrain et Capitanio1968) un homme et une femme étaient couchés sur le dos en dehors d’une structure et, un peu éloignés, se trouvaient les restes d’un autre homme, d’une femme et d’un enfant. Pour le grottes on a une bonne documentation de Grotta Continenza (Barra et alii 1989-90), Grotta S. Angelo (Di Fraia et Grifoni Cremonesi 1996) et Grotta dei Piccioni (Cremonesi 1976), surtout en ce qui concerne les aspects du rituel. A Grotta S.Angelo, située dans un vallon à la limite entre Marche et Abruzzes, on a trouvé les restes de cinq adultes, trois jeunes et six enfants, dispersés par les néolithiques mêmes qui ont creusé des fosses à offrandes et jeté les restes dans des foyers: les analyses ont démontré qu’il ne s’agit absolument pas d’incinération intentionnelle car les os ont subi l’action du feu quand ils étaient depuis longtemps à état squelettique accompli. Dans la grotte la fréquence des fosses, la présence de vases qui n’apparaissent jamais dans les habitats, des offrandes de poires et pommes sauvages et d’une seule meule dans une fosse sont très particuliers. Il faut ici rappeler la reconstruction des cérémonies faite par R. Skeates (qui ne connaît évidemment pas la littérature relative à la grotte) et sa reconstruction d’une société qui ensevelissait les jeunes dans les grottes et les adultes dans les villages pour maintenir le pouvoir des anciens (Skeates 1991, 1995, 1997). Il faut alors se rappeler qu’à la Grotta Continenza (déjà connue pour le premier cas d’incinération au néolithique ancien) on a trouvé les restes de 36 individus (19 adultes et 17 entre jeunes et enfants) inhumées le long des parois de l’abri et malheureusement dispersés soit par les néolithiques mêmes avec leur coutume de creuser des fosses, soit par les amateurs. On a pu néanmoins observer les sépultures de deux femmes avec enfants (une d’elles avec un fœtus était déposée dans une niche de la paroi et l’autre dans la grotte interne); de deux enfants de deux ans, l’un déposé à coté de la paroi et protégé par des blocs et l’autre déposé dans une petite fosse ovalaire, à coté de laquelle était une sorte de pavement en pierres plates avec un grand foyer: tous les deux étaient dépourvus de crâne. Dans la grotte on a trouvé aussi des fosses contenant les restes de chiens, porcs et brebis avec traces de boucherie et de fosses contenant des morceaux de vases avec ocre rouge ou argile jaune. Les vases du complexe à incinération sont du même type que ceux qu’on a trouvé à Grotta S.Angelo et l’un d’eux était enduit d’argile jaune, la même avec laquelle on avait préparé un pavement et revêtu d’autres fosses en série. A Grotta dei Piccioni un squelette d’enfant était déposé à la base du dépôt (Radmilli et Alii 1978). Très pauvre est enfin la documentation pour l’Italie centrale thyrrenienne, occupée par la céramique cardiale: on connaît à présent un crâne d’enfant dans un cercle de pierres et avec une meule à Grotta Settecannelle (100 km au Nord de Rome) (Ucelli Gnesutta et Mallegni 1985) et une sépulture dont il reste un crâne posé sur une pierre plate et un fémur sur deux pierres dans une fosse de l’habitat de Pienza (Calvi Rezia 1973). 200

R. Grifoni Cremonesi: Le rituel funéraire au cours du Néolithique italien

En Ligurie à la Caverna delle Arene Candide on a une sépulture en fosse avec de l’ocre (Tinè 1976) et une sépulture d’enfant en fosse, avec tessons et coquilles (cerastoderma) provenant du site de Piancada en Friuli, occupé par les gens du néolithique ancien à céramique à lignes incisées (Pessina 1997).

l’empreinte d’un arc, et des meules brisées intentionnellement (Patrizi et aliii 1954, Grifoni Cremonesi et Radmilli en presse). D’autres sépultures étaient dans la grotte, accompagnées de vases, sagaies en os, éléments de parure et offrandes de brebis, porcs, lièvres, dont il existe seulement les quartiers antérieurs ou postérieurs, et surtout des meules, souvent brisées ou exposées au feu.

En considérant toutes les sépultures que l’on connaît pour cette période, on peut souligner avant tout la présence dans les habitats mais aussi la fréquentation des grottes, la simplicité des structures en fosse, la presque totale constance de la position sur le coté gauche et la complète absence de mobilier et de parure. Il y a quand même la présence d’offrandes, constituées par des céréales ou des meules ou des vases particuliers.

Une autre grotte qui a donné des restes humains, malheureusement pas en connexion, mais pourvus de riches mobiliers, est Grotta dell’Orso di Sarteano : on y a trouvé plusieurs vases décorés, un vase à double col, meules et coquillages avec ocre rouge, parures, harpons et sagaies en os, anneaux - bracelets en pierre ou en os (Grifoni 1967). Cette richesse dans la composition du mobilier nous signale sûrement un changement dans l’idéologie de ces sociétés, qui néanmoins ne présentent aucune rupture évidente avec les gens de l’impressa, au niveau du choix des sites, de la tradition de l’industrie lithique et osseuse. En particulier pour la culture de Catignano et pour les céramiques à lignes incisées on peut bien constater la continuité culturelle (Cremonesi et Tozzi 1987).

On pourrait relier à la sphère des cultes la présence d’un crane dans une structure du site de Marcianese (Geniola et Mallegni 1975) ou le petit vase en forme d’animal déposé dans une fosse aux marges d’une cabane à Santo Stefano (Radi 1991), près de la Grotta Continenza. Même si on a très peu d’éléments pour envisager l’existence d’espaces des morts différenciés par âge ou sexe ou état social, on peut bien reconnaître l’homogénéité du rituel dans toutes les cultures faisant partie de la ceramica impressa et on peut aussi envisager une connexion entre le rituel funéraire et une idéologie (ou bien religion) liée à l’agriculture.

Mais c’est avec les cultures à céramiques trichromiques (Ripoli et Serra d’Alto pour le centre et le Sud) et avec les vases à bouche carrée du Nord d’Italie, qu’on voit apparaître de véritables nécropoles organisées, situées en proximité des habitats ou dans les zones occupées par les cabanes.

Cette homogénéité semble subir des changements avec l’apparition des céramiques peintes dans le Centre Sud et des céramiques à lignes incisées dans la région toscan latiale, datées de 5200 à 4500 b.C. A ce moment là on voit paraître des structures sépulcrales diversifiées et la présence de vases et d’objets particuliers accompagnant les morts. On a des fosses ovalaires ou rectangulaires, parfois avec dalles de couverture. Pour la culture des céramiques à bandes rouges je rappelle seulement les sépultures les plus importantes, toutes dans les habitats: Catignano (de 5400 a 5000 b.C.) où il y a une fosse à coté d’une cabane avec une femme qui avait subi la trépanation du crane et une autre fosse avec un enfant accompagné de vases (Robb et alii 1991), Villa Badessa (près de Catignano), fosse avec deux femmes et une quantité de vases peints, de passage entre Catignano et la culture de Ripoli (Germanà et alii 1990), Passo di Corvo (Tinè 1983) avec plusieurs sépultures en fosse, Trasano (deux fosses, une avec femme au crâne trépané) (Guilaine et Cremonesi 1992) et les villages près de Matera (Tirlecchia, Murgecchia, Murgia Timone) où les sépultures étaient aussi bien en dehors que à l’intérieur des habitats, ou bien à coté des fossés, en fosses rectangulaires creusées dans le rocher (Lo Porto1992-93, 1998). En Italie du Sud il faut aussi rappeler le niveau d’incinérés dans la Grotta Pavolella (Calabre) (Carancini et Guerzoni 1987). La position des squelettes est aussi bien sur le coté droit que gauche que sur le dos, tandis qu’au néolithique ancien le coté gauche est plus fréquent.

Pour le Nord d’Italie, où il y a un bon nombre de sépultures en fosses ou en caissons de dalles, les données les plus importantes nous viennent du site de Chiozza en Emilia, où l’on constate l’existence de fosses rectangulaires disposées en rangées régulières, avec inhumations individuelles et mobilier consistant en vases, haches polies: une femme avait aux pieds le squelette d’un chien (De Buoi 1940). En Ligurie on connaît les sépulture en caisson de dalles de la Caverna des Arene Candide (Bernabò Brea 1956) et d’autres grottes. Dans la vallée de l’Adige près de Trento on a la petite nécropole de La Vela, avec fosses ovalaires entourées de pierres et mobilier consistant en vases et objets d’importation (vases de la culture de Serra d’Alto et herminettes type Hinkelstein) (Bagolini 1990). Des vases type Serra d’Alto proviennent aussi de la nécropole de Gaione près de Parma en Emilie (M.Bernabò Brea et alii 1988) et témoignent donc d’une vivacité d’échanges entre le Nord et le Sud de la péninsule. Toujours au Nord, en Emilie près de Faenza on a le village de Fornace Cappuccini: dans le fossé du néolithique ancien, au cours du Néolithique récent furent déposés 33 individus, couchés sur le dos ou accroupis, sans mobilier: il y avait quand même deux chiens à proximité d’une femme, un enfant assis aux pieds d’un adulte, une grande quantité d’ossements de bœuf, brebis, porc et cerf à coté des sépultures et un crane de loup déposé dans un cercle de pierres avec de l’ocre et des tessons (Antoniazzi et alii 1990).

Pour les céramiques à lignes incisées je rappelle la riche sépulture de Grotta Patrizi al Sasso di Furbara (Rome) d’un homme avec de nombreuses pathologies et le crâne trépané, accompagné d’un important nombre de vases décorés,

Pour le centre d’Italie, où, entre 4500 et 3600 b.C., la culture de Ripoli (Cremonesi 1965, 1976, 1978, 1998) est dominante, 201

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

on connaît des restes humains dans les grottes mais on a aussi une grande nécropole dans le village éponyme dans les Abruzzes: malheureusement il s’agit d’une vieille fouille de 1910 et les mobiliers furent mélangés et ensuite perdus (guerre, tremblements de terre etc.). On a seulement, à ce propos, une publication de U. Rellini (1934) qui parle de fosses ovalaires, rectangulaires, elliptiques, circulaires, triangulaires et d’une fosse longue 15 mètres avec 14 squelettes. On sait sûrement que la nécropole se trouvait au centre du village et qu’une des fosses contenait une femme avec un chien, mais il est impossible de savoir si les tombeaux appartenaient à une seule phase de la culture (qui s’étale de 5500 a 4700 BP) ou seulement à un moment de son développement.

(Guilaine et Cremonesi 1992, Bianco 1976, Ingravallo 1976). Dans le site éponyme de Serra d’Alto près de Matera un tombeau était signalé par une sorte de stèle et souvent on a deux pierres triangulaires à coté des squelettes. A Trasano un crâne de bœuf accompagnait les restes d’un homme, d’une femme et d’un enfant. La typologie des dépositions est très variée (coté droit ou gauche, sur le dos, accroupie). On connaît aussi des sépultures déposées sur des lits de pierres dans des grottes, avec foyers et restes d’animaux (Striccoli 1988; Geniola et Tunzi 1980). Il est quand même difficile d’établir un rapport entre nécropoles et habitats car, comme nous l’avons déjà dit, les sépultures de Serra d’Alto se trouvent presque toujours en correspondance avec les vieux sites de l’impressa et des bandes rouges et très rares sont à présent (?) les villages de Serra d’Alto. Il faut remarquer un phénomène très important: les sépultures en silos ou les niches dans les parois des fossés peuvent être considérées comme un prélude à l’hypogéisme funéraire qui aura le maximum de sa diffusion au cours du III millénaire. Avec Serra d’Alto fait son apparition l’hypogeisme cultuel bien connu par l’Hypogée Manfredi de Polignano (près de Bari) (Geniola 1987). Dans les grottes Cala Scizzo (Geniola et Tunzi 1980) et Grotta Pacelli (Striccoli ) on a de grands autels en pierres , des stèles et des statuettes féminines très élaborées.

Il faut à ce propos rappeler encore une fois R. Skeates (1995) qui imagine un fin dramatique pour cette culture, avec des cadavres qu’on a laissés se décomposer pour être ensuite désarticulés et ensevelis au cours de fêtes et de repas cérémoniels. Je voudrais seulement rappeler que la culture de Ripoli se développe très tranquillement jusqu’au début du III millénaire avec une évolution interne qui, des vases peints, arrive au faciès de Paterno Macchia a Mare, comme il a été démontré par G. Cremonesi à partir de 1966, et qui occupe, dans les phases finales, tout le centre de l’Italie et une partie du Sud (Cremonesi 1984, Bagolini et alii 1987). En outre, rien dans le rapport Rellini ne peut nous faire envisager des désarticulations et des ensevelissements secondaires: des photos d’époque nous montrent, il est vrai, d’épouvantables amas de terrain de la fouille 1910, mais les squelettes visibles sont bien disposées et celui de la femme avec chien est parfaitement conservé au Musée de Chieti. On sait que les squelettes étaient couchées sur le dos, sur le coté droit ou gauche, repliés. Encore une fois la presque totale absence de considération de la réalité des données archéologiques (et de la littérature) conduit à des reconstructions historiques un peu trop fantaisistes!

Dans le contexte des cultes nous avons aussi les peintures de Grotta dei Cervi à Porto Badisco (Graziosi 1980), où se manifeste un monde idéologique complexe, dont une des expressions, commune à Ripoli et à Lagozza, sont les cercles de pierres que l’on trouve également dans la Grotta Cosma, qui fait partie du même système carsique que Grotta dei Cervi. La diversité des structures funéraires augmente pendant la phase suivante du néolithique, caractérisée par la culture de Diana qui se superpose et se mélange avec Serra d’Alto et Ripoli et qui occupe toute la péninsule jusqu’à la Toscane et à l’Emile, expansion due à la grande diffusion de l’obsidienne de Lipari (Bernabò Brea et Cavalier 1980).

Pour la culture de Ripoli et les céramiques trichromiques il faut aussi rappeler les enfants dans les cercles de pierres à Grotta dei Piccioni, avec offrandes et objets particuliers (Cremonesi 1976) et, pour la côte thyrrénienne, les sépultures de Grotta delle Felci à Capri, avec une grande quantité de meules et meulettes couvertes d’ocre rouge (Buchner 195455).

Nous connaissons bien les sépultures de cette culture, mais presque rien n’est connu des rapports avec habitats (ou : concernant les habitats ?): les structures connues comprennent aussi bien des fosses ovalaires simples ou entourées de pierres, que des cistes lithiques ou des petits puits. Souvent on a pu reconnaître, comme aux îles Tremiti (Palma di Cesnola 1967), Scoglio del Tonno et Masseria Bellavista (Quagliati 1906), des dépositions secondaires avec des amas d’os longs ou de crânes. On a aussi des fosses surmontées par un tumulus de pierres (dans un cas avec une meule au sommet) ou bien des cistes très grandes avec inhumations multiples. Dans les grottes, déjà utilisées par les gens de Serra d’Alto continuent les sépultures de Diana, quelquefois très particulières comme à Cala Colombo (De Lucia et alii 1977) avec un des squelette enveloppé d’argile rouge, phénomène qui se trouve aussi dans la grande ciste de Madonna delle Grazie (Damato 1993).

Très riche est la documentation pour la culture de Serra d’Alto (Geniola 1987, Lo Porto 1989) qui occupe le Sud, les îles éoliennes, la Sicile et qui arrive jusqu’au centre d’Italie et pour laquelle on connaît une extrême variété d’aspects funéraires. On a de véritables nécropoles (par exemple Pulo di Molfetta) mais aussi des sépultures en grottes ou des tombeaux dans des villages. Les structures les plus communes sont des fosses ovalaires ou circulaires creusées dans le terrain, souvent entourées de pierres, avec inhumations individuelles, mais aussi souvent doubles, et mobilier de vases peints, lames en silex et obsidienne. Il y a aussi des petites niches creusées dans les parois des fossés du Néolithique ancien (Matera et Tavoliere) ou bien des silos, comme à Trasano, ou des cistes lithiques

On reconnaît aussi des phénomènes tels la fragmentation intentionnelle de vases, la présence de vases particuliers, les 202

R. Grifoni Cremonesi: Le rituel funéraire au cours du Néolithique italien

vases renversés, les meules (une, renversée, accompagnait un crâne à Porto Badisco)

BAGOLINI B. 1990, Cultura dei vasi a bocca quadrata. Il sepolcreto neolitico de La Vela di Trento, in Die ersten Bauern, 2, Zürich, p.225-231.

Le type de sépulture le plus intéressant est enfin le véritable petit hypogée à four d’Arnesano près de Taranto, dans lequel gisait un jeune homme accompagné d’un vase typique de Diana et d’une petite statue stèle (Lo Porto 1972).

BAGOLINI B., BARFIELD L.H., CREMONESI G. 1987, La fine del neolitico, in Atti XXVI Riun.Sc Ist. Ital.Preist.Protost., Firenze, p.79-88. BAGOLINI B., GRIFONI CREMONESI R. 1994, Il neolitico italiano., facies culturali e manifestazioni funerarie, Bull.Palet.Ital. 85, p.139-170.

De ce rapide excursus on peut certes constater la diversité des manifestations funéraires au cours du néolithique italien et les différences relatives au structures dans une même culture, mais on peut constater aussi une homogénéité de base, surtout en ce qui concerne les aspects du rite (fosses à offrandes, absence ou présence de mobilier, objets particuliers, statuettes féminines, etc.) qui nous font apercevoir des entités supra- culturelles qui occupent de vastes régions dans une sorte de koinè qui outrepasse les différences locales.

BARRA A. GRIFONI CREMONESI R. MALLEGNI F. PIANCASTELLI M. VITIELLO A. WILKENS B. 1989-90, La Grotta Continenza di Trasacco. I livelli a ceramiche, Riv.Sc.Preist., XLII, p.31-100. BERNABO’BREA L. CAVALIER M. 1980, Meligunìs Lipàra, IV, Palermo. BERNABO’BREA L. 1946, 1956 La Caverna delle Arene Candide, I e II, Bordighera. BERNABO’BREA M., GHIRETTI A., POLGLASE C., VISCONTI V., 1988, I siti neolitici lungo il torrente Cinghia (Parma), Preistoria Alpina, 24, p.103-164.

J’ai dit que nous disposons de peu d’éléments pour tenter des reconstructions de type socio -économique ou de type idéologique: on peut quand même apercevoir des particularités dans la composition du mobilier ou bien des anomalies qui mettent en évidence l’existence de personnages importants, sans pourtant impliquer l’existence de sociétés stratifiées. Je pense par exemple au mâle de Grotta Patrizi, avec son riche mobilier et tout seul dans une vaste chambre de la grotte, ou aux objets d’importation de La Vela, ou bien à quelques sépultures plus riches ou aux femmes avec chiens. Il faut considérer aussi quelques inhumations particulières d’enfants, seuls ou avec la mère.

BIANCO S. 1976, Matinelle di Malvezzi, in Il Museo Nazionale Ridola di Matera, Matera, p.71-73. BUCHNER G. 1954-1955, La stratigrafia dei livelli a ceramica ed i ciottoli con dipinti schematici antropomorfi della Grotta delle Felci, Bull.Palet.Ital., 64,p.107-135. CALVI REZIA G. 1972, I resti dell’insediamento neolitico di Pienza, Atti XIV Riun. Sc.Ist.Ital.Preist.Protost., Firenze, p.285-300. CARANCINI G.L., GUERZONI R.P. 1987, Gli scavi nella Grotta Pavolella presso Cassano allo Ionio, Atti XXVI Riun.Sc.Ist.Ital. Preist.Protost., p.783-792. CASSANO S.M. , MANFREDINI A. 1983, Studi sul Neolitico del Tavoliere della Puglia, British Arhaeological Reports S.160, Oxford.

D’autres indications nous viennent, pour les fouilles plus récentes, des résultats de l’analyse spatiale: à La Vela, par exemple, il y a une distribution différente des femmes par rapport aux hommes, mais généralement on ne trouve pas de grandes différences liées au sexe ou à l’age, en ce qui concerne le type de structure ou le mobilier.

CASSANO S.M., CAZZELLA A., MANFREDINI S. MOSCOLONI M. 1987, Coppa Nevigata e il suo territorio. Testimonianze archeologiche dal VII al II millennio a.C., Roma. CASSANO S.M., MANFREDINI A. 1991 , Rinvenimento di una sepoltura Serra d’Alto a Masseria Candelaro, Scavo 1990 , Atti XII Convegno Preist.Protost.Storia Daunia, Sansevero, p.3136.

Il y a donc encore beaucoup à travailler à ce sujet: une révision des sépultures néolithiques italiennes a montré qu’il y a, surtout pour les vieilles fouilles, un manque de données qui nous oblige à ne pas trop tenir compte de certains sites, qu’on pourrait définir «historiques» dans la littérature, mais qui ne nous offrent aucune aide pour une reconstruction satisfaisante des anciennes sociétés.

CIPOLLONI M. 1977-82, Scavi nel villaggio neolitico di Rendina (1970 - 1976). Relazione preliminare, Origini, XI, p.183-323. CORRAIN C., CAPITANIO M. 1968, I resti scheletrici dei depositi neolitici di Maddalena di Muccia e di Ripabianca di Monterado nelle Marche, Riv.Sc.Preist., XXIII, p.223-244. COSTANTINI L., TOZZI C. 1987, Un gisement à céramique imprimée dans le Subapennin de la Daunia (Lucera -Foggia): le village de Ripatetta. Economie et culture matérielle, in Prémières communautés paysannes de la Méditerranée occidentale, Paris, p. 387- 394.

Adresse de l’auteur

CREMONESI G. 1965, Il villaggio di Ripoli alla luce dei recenti scavi, Riv.Sc.Preist. X, p. 88-55.

Renata GRIFONI CREMONESI Dipartimento di ScienzeArcheologiche Université de Pise, ITALIE

CREMONESI G.1974, Osservazioni sulla cultura di Ripoli, Annali Univ.Lecce, Facoltà di Lettere e Filosofia, VI, p.81-103.

Bibliographie

CREMONESI G. 1976, La Grotta dei Piccioni di Bolognano nel quadro delle culture dal neolitico all’età del bronzo in Abruzzo, Pisa.

ANTONIAZZI A., BERMOND MONTANARI G., GIUSBERTI G., MASSI PASI M., MENGOLI D., MORICO G., PRATI L., 1990 - Lo scavo preistorico a Fornace Cappuccini, in Archeologia a Faenza, Bologna, p.23-59.

CREMONESI G. 1985-86, Samari, RivSc.Preist., XL, Not. p.424425.

CREMONESI G.1984, Osservazioni su alcuni aspetti dell’eneolitico del versante adriatico, Atti 3° Convegno Preist.Protost.Storia Daunia, Sansevero, p.131-148.

203

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe GRIFONI CREMONESI R. 2000, Sull’interpretazione di alcuni aspetti funerari e cultuali nel neolitico abruzzese, in P. BIAGI a cura di, Studi sul Paleolitico, Mesolitico e Neolitico del bacino adriatico in ricordo di A.M.Radmilli, Società Preist.Protost.Friuli Venezia Giulia, quad.8, p.127-139.

CREMONESI G. 1985-90, Su alcune manifestazioni di culto del neolitico abruzzese, Abruzzo, XXII- XXVII, p. 465-474. CREMONESI G. 1998, La culture de Ripoli, in GUILAINE J. a cura di, Atlas du Néolithique Européen, 2A, Liège p.180-183. CREMONESI G., GUILAINE J. 1987, L’habitat de Torre Sabea (Gallipoli, Puglia) dans le cadre du Néolithique ancien de l’Italie du Sud Est, in Prémières communautés paysannes en Méditerranée occidentale, Paris, p.377-386.

GRIFONI CREMONESI R., MALLEGNI F. 1978, Testimonianze di un culto ad incinerazione nel livello a ceramica impressa della grotta, riparo Continenza di Trasacco (L’Aquila) e studio dei resti umani cremati, Atti Soc. Tosc. Sc. Nat. Mem. Serie A, LXXXV, p.253-279.

CREMONESI G.,TOZZI C. 1987, Il Neolitico dell’Abruzzo, Atti XXVI Riun.Sc.Ist.Ital.Preist.Protost., Firenze, p.21-30

GRIFONI CREMONESI R., MALLEGNI F., TRAMONTI A. c.s., La sepoltura neolitica di Torre Sabea, in GUILAINE J., CREMONESI G.(a cura di), Le site néolithique de Torre Sabea (Gallipoli, Lecce), Tolosa. GRIFONI CREMONESI R., RADI G. 1999, Le Néolithique de l’Italie centrale adriatique, in Le Néolithique du Nord Ouest méditerranéen, Carcassonne, p.39-50. GRIFONI CREMONESI R., RADMILLI A.M. c.s. La Grotta Patrizi al Sasso di Furbara, Bull.Palet.Ital. GUILAINE J. CREMONESI G. 1987, L’habitat néolithique de Trasano (Matera, Basilicata). Prémiers résultats, Atti XXVI Riun.Sc Ist. Ital.Preist.Protost., Firenze, p.707-720.

DAMATO A. 1993, Gli insediamenti neolitici di Rutigliano, Fasano. DE BUOI L. 1940, Nuovi ritrovamenti nel sepolcreto neolitico di Chiozza di Scandiano (Reggio Emilia), Atti Società Naturalisti e Matematici di Modena, 71, p.3-9. DE LUCIA A. FERRI D. GENIOLA A. GIOVE C. MAGGIORE M. MELONE N. PESCE DELFINO V. PIERI P. SCATTARELLA V. 1977, La comunità neolitica di Cala Colombo presso Torre a Mare (Bari), Bari. DI FRAIA T., GRIFONI CREMONESI R. 1996, La Grotta Sant’Angelo sulla Montagna dei Fiori (Teramo) e il problema delle frequentazioni cultuali in grotta, Pisa. FIORENTINO G., MUNTONI I.M., RADINA F. 2000, La neolitizzazione delle Murge Baresi: ambienti, insediamenti e attività produttive, in “ PESSINA A., MUSCIO G. a cura di, La neolitizzazione tra Oriente e Occidente, Udine, p.381-412.

GENIOLA A., TUNZI A.M. 1980, Espressioni cultuali e d’arte nella Grotta di Cala Scizzo presso Torre a Mare (Bari), Riv.Sc. Preist., XXXV, p.125-146.

GUILAINE J. CREMONESI G. 1992, Trasano (Matera): site du néolithique et de l’âge du bronze, Mélanges de l’Ecole Française de Rome, 104, p.518-523. GUILAINE J. CREMONESI G. RADI G. 1992, Trasano et la céramique gravée materane, in Autour de Jean Arnal, Montpellier, p.123-137. INGRAVALLO E. 1976, San Martino, in Il Museo Nazionale Ridola di Matera, Matera, p.67-70. LOLLINI D.G. 1965, Il neolitico delle Marche alla luce delle recenti scoperte, Atti VI Congresso Intern.Sc.Preist.Protost., Firenze, vol.II, p.309-315. LO PORTO F. 1972, La tomba neolitica con idolo in pietra di Arnesano (Lecce), Riv.Sc.Preist., XXVII, p.357-372.

GERMANA’ F., FORNACIARI G. 1992, Trapanazioni, craniotomie e traumi cranici in Italia dalla preistoria all’età moderna, Pisa.

LO PORTO F.G. 1989, L’insediamento neolitico di Serra d’Alto nel Materano, MonAnt, serie monografica, LIII.

GERMANA’F., MALLEGNI F., DE POMPEIS C., RONCO D., 1990, Il villaggio neolitico di Villa Badessa (Pescara), aspetti paletnologici, antropologici e paleopatologici, Atti Soc.Tosc.Sc. Nat. Mem., ser.A, 97, p.271-310.

LO PORTO F.G. 1992-93, Matera. Vecchi e nuovi scavi nell’ insediamento neolitico di Tirlecchia, Not.Sc., serie IX, vol.III-IV.

GENIOLA A. 1982, Marcianese, Lanciano. GENIOLA A. 1987, La cultura di Serra d’Alto nella Puglia centrale, Atti XXVI Riun.Sc.Ist.Ital.Preist.Protost. Firenze, p.771-781. GENIOLA A., MALLEGNI F. 1975, Il calvario neolitico di Lanciano (Chieti), note paletnologiche e studio antropologico, Atti Soc. Tosc. Sc. Nat. Mem. ser.A, LXXXII, 237- 253.

LO PORTO F.G. 1998, Murgia Timone. Murgecchia, Mon.Ant., serie monografica, vol.V.

GNESUTTA UCELLI P. MALLEGNI F. 1988, Note preliminari sullo scavo della Grotta delle Settecannelle (Ischia di Castro, Viterbo), Atti Soc.Tosc.Sc.Nat. Mem., ser.A, XCV, p.303-323.

MALLEGNI F., RONCO D. 1996, I resti scheletrici umani di Grotta S.Angelo, in DI FRAIA T., GRIFONI CREMONESI R. La Grotta S.Angelo sulla Montagna dei Fiori (Teramo) e il problema delle frequentazioni cultuali in grotta, Pisa, p.263-275.

GORGOGLIONE M.A. 1987, Manduria (Taranto), Taras, X, p.282283. GRAZIOSI P. 1980, Le pitture preistoriche di Porto Badisco, Firenze.

MANZI G. MACCHIARELLI R. 1989, L’ ”Uomo pre-neolitico” della Maiella (Lama dei Peligni, Abruzzo), Riv. Antrop. LXVII, p.325-328.

GRIFONI R. 1967 La Grotta dell’Orso di Sarteano. I. I livelli neolitici, Origini I.

MORDACCI A., GRIFONI CREMONESI R., c.s, Manifestazioni funerarie nel neolitico italiano. Firenze.

GRIFONI CREMONESI R. 1986, Alcuni dati relativi a fenomeni funerari con implicazioni cultuali nella preistoria e problemi di interpretazione, Dialoghi di Archeologia, 2, p.265- 269.

MOSSO A. 1910, La necropoli neolitica di Molfetta, MonAnt, XX, cc 237-252. ORLANDO M.A. 1997, Samari, in INGRAVALLO E. a cura di,. La Passione dell’Origine. Giuliano Cremonesi e la ricerca preistorica nel Salento, Lecce, p.122-134.

GRIFONI CREMONESI R. 1987, Témoignages de cultes dans le néolithique ancien des Abruzzes, in “Prémières communautés paysannes de la Méditerranée occidentale”, Paris, p. 413418.

PALMA di CESNOLA A. 1967, Il neolitico medio - superiore a San Domino (Arcipelago delle Tremiti), Riv.Sc.Preist. XXII, p.349-392.

GRIFONI CREMONESI R. 1994, Observations on the problems related to certain cult phenomena during the Neolithic in the Italian peninsula, Journal of European Archaeology. 2.2, p.179-197.

PARENTI R. 1957, Lo scheletro umano della “tomba del cane” a Ripoli, Arch.Antrop.Etnol., LXXXVII, p.5-95.

GRIFONI CREMONESI R.1996, Considerazioni su alcuni aspetti del rituale funerario nel Neolitico antico in Forme e tempi della neolitizzazione in Italia meridionale e in Sicilia, p. 416-418.

PATRIZI S., RADMILLI A.M., MANGILI G. 1954, Sepoltura ad inumazione con cranio trapanato nella Grotta Patrizi al Sasso di Furbara, Riv.Antrop., XLI, p. 33-67. 204

R. Grifoni Cremonesi: Le rituel funéraire au cours du Néolithique italien PESSINA A., FERRARI A., FONTANA A. 1997, Le prime popolazioni agricole del Friuli.

SKEATES R. 1995, Transformations in mortuary practice and meaning in the Neolithic and Copper Age of Lowland East, Central Italy, in Ritual, Rites and Religion, IIIrd DEYA International Conference of Prehistory, vol. I, BAR International Series, n.611, p.212-238.

PESSINA A., MUSCIO G., Settemila anni fa il primo pane, Udine, p.133-146. QUAGLIATI Q. 1906, Tombe neolitiche in Taranto e nel suo territorio, Bull.Palet.Ital.,32, p. 17-49.

SKEATES R. 1997, The Human Uses of Caves in East, Central Italy during the Mesolithic, Neolithic and Copper Age, in C. Bonsall, C. Tolan, Smith, The Human Use of Caves, BAR International Series n.667, p.79-86.

RADI G. 1991, Il villaggio neolitico di Santo Stefano (Ortucchio). Atti Conv. “Il Fucino e le aree limitrofe nell’antichità”, Roma, p.110-121.

STRICCOLI R. 1988, Le culture preistoriche di Grotta Pacelli (Castellana Grotte - Bari), Fasano.

RADMILLI A.M., MALLEGNI F., FORNACIARI G., 1978, Recenti scavi nella Grotta dei Piccioni di Bolognano (Pescara) e riesame dei resti scheletrici umani provenienti dai circoli, in Atti SocToscScNat Mem. ser.A, LXXXV, p.175-198.

TINE’ S. 1976, Arene Candide, in Archeologia in Liguria.Scavi e scoperte1967-1975, p.149-158. TINE’ S. 1987, Il Neolitico, in La Calabria antica (a cura di S. Settis), Roma, p. 41-62.

RELLINI U. 1934, La più antica ceramica dipinta in Italia, Roma. RELLINI U. 1923, La Grotta delle Felci a Capri, Mon.Ant., XXIX.

TINÈ S., 1983, Passo di Corvo e la Civiltà neolitica del Tavoliere, Genova.

RIDOLA D. 1924-26, Le grandi trincee preistoriche di Matera, Bull. Palet.Ital., 44, 45,46.

TOZZI C. 1978, Un aspetto della corrente culturale della ceramica dipinta in Abruzzo, il villaggio di Catignano (Pescara), Quaderni de La Ricerca Scientifica, 100, p. 95-111.

ROBB J., MALLEGNI F. 1994, Anthropology and paleopathology of Neolithic humans from Catignano (Pescara), Riv. Antrop., 72, p.197-224.

TOZZI C., 1998, Culture de Catignano, Scaloria Bassa. in GUILAINE J. a cura di, Atlas du Néolithique Européen, 2A, Liège, p. 178-180.

ROBB J., MALLEGNI F., RONCO D. 1991, New human remains from the Southern Italian Neolithic: Ripa Tetta and Latronico, Riv. Antrop., 69, p.125-164.

TOZZI C., VEROLA L.,1990, La campagna di scavo 1990 a Ripa Tetta (Lucera, Foggia), Atti 12° Convegno di Preistoria, Protostoria e Storia della Daunia, Sansevero, p.37-48.

SKEATES R. 1991, Caves, Cult and Children in Neolithic Abruzzo, Central Italy, in Sacred and Profane, Oxford Committee for Archaeology, Mon.n.32, p.122-134.

TUNZI SISTO A.M. 1996b, Le origini di Foggia, in TUNZI SISTO A.M. a cura di 1996, Ipogei della Daunia, Foggia, p.99-101.

205

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

206

P. Kieselbach: Lithic Procurement and Production in the Late Neolithic of Southwestern Germany

LITHIC PROCUREMENT AND PRODUCTION IN THE LATE NEOLITHIC OF SOUTHWEST GERMANY Petra KIESELBACH

Abstract: Lithic artifacts from 14 late Neolithic settlements of Southwestern Germany were examined on the basis of technological and typological criteria. The main interest was to investigate the production system for the different lithic assemblages, which can be assigned to different late Neolithic ceramic groups. According to that clues can be given to the raw material procurement, manufacture and use of lithic artifacts. The raw material analysis revealed a different use of local, regional and supraregional sources. It was determined by the location of the settlements relative to the raw material sources. The lithic assemblages of the different ceramic groups show an opportunistic behavior with regards to the supply of raw materials that was adapted to a large extent to the natural conditions of the environment. The results of the analysis of the production system suggest that no group-specific utility system existed during the late Neolithic. Rather, a supply network linked all groups together in the same way. The extent of the supply network for the individual settlements and its dependence on neighboring settlements were primarily determined by the local resource situation. Résumé : Les industries lithiques de 14 sites du Néolithique tardif de l’Allemagne du sud-ouest ont été examinées en prenant pour base des critères technologiques et typologiques. L’intérêt principal était d’étudier le système de production des différents ensembles lithiques, qui peuvent être attribues à différents groupes céramiques du Néolithique tardif. Conformément à cela, des indices peuvent être fournis sur l’acquisition, la production et l’utilisation des matériaux lithiques. L’analyse de la matière première a révélé une utilisation différente de sources locales, régionales et plus lointains, qui était déterminée par l’emplacement des sites relatifs aux gisements. Les différents ensembles lithiques font preuve d’un comportement opportuniste en ce qui concerne l’approvisionnement en matières premières qui était en grande partie adapté aux conditions naturelles de l’environnement. Les résultats du système de production montrent qu’il n’existait pas d’approvisionnement spécifique a chaque groupe lors du Néolithique tardif. Il s’agirait plutôt d’un réseau d’approvisionnement reliant tous les groupes de la même manière. L’étendue du réseau d’approvisionnement des sites individuels et sa dépendance envers des sites voisins ont été déterminés principalement par la situation des ressources locales.

and use of lithic artifacts and afford an insight into the economy of the sites. On the other hand there is the raw material and its place of origin that might afford an insight into the communication systems of the sites (Biro1998).

INTRODUCTION In the Late Neolithic Period (4400-3500 B.C.) people of different regional groups characterized by special ceramics populated Southwest Germany. All of these coexisting and successive ceramic groups were influenced by each other or developed from a previous one (Schlichtherle 1988). In the following I will discuss the lithic procurement and production of different Late Neolithic sites belonging to these ceramic groups. It is based on the results of my PhD project (Kieselbach 2000). The main questions are, how was the lithic supply of the different sites organized, and was there a chronological and/or regional change in the procurement, manufacture and use of lithic material at the different Late Neolithic ceramic groups of Southwest Germany? To answer these questions I analyzed the lithic assemblages of 14 sites that can be distinguished on the basis of ceramic findings as well as radiocarbon and dendrochronological dates in chronological and regional order (fig. 1 and 2).

RAW MATERIAL AND PROCUREMENT The raw material of the 14 lithic assemblages came from different raw material sources (fig. 3). Since for most of the raw material types the exact location of their sources cannot be determined, three zones of origin are distinguished: 1. Local raw material that is found within 30 km as the crow flies, 2. Regional raw material that is found 30 to 100 km away from the site, 3. Supraregional raw material that has its origin more than 100 km away from the site. According to the three zones of origin the procurement of the raw materials occurred differently in the 14 sites (fig. 4). It turned out that the procurement was not homogenous within the different culture groups, but was rather affected by the spatial relationship between the sites and the raw material sources. Because Jurassic chert built the predominant raw material for almost all sites, the distance to the Swabian Alb Mountains determine, whether local or regional and supraregional raw material make the most part of the lithic assemblages. According to that, the sites located in the

Seven of the sites are situated in the north alpine foreland. These are all damp sites on lakeshores of the lake Constance and the Federsee bog or on the riverbank of the Danube. The other seven sites are situated north of the Swabian Alb Mountains, on mineral soils along the Neckar river valley. Figure 2 shows, which ceramic groups the 14 sites represent. For analyzing lithic artifacts and the lithic supply two sources of information exist. On the one hand there are technological and typological attributes that tell us about the manufacture 207

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe B.C.

Lake of Constance

Upper Swabian, Federsee bog

Middle Neckar valley

Lower Neckar valley

3500 3600 3700

3800

3900 4000 4100 4200 4300 4400

Ödenahlen KS III 3710-3688 B.C. 3600 B.C. Reute| Schorrenried E | 3700-3740 B.C. h | Wangenr Hinterhorn KS Taubried I e I 3900-3800 B.C. n 3825 B.C. s Alleshausent Hartöschle ei 3920 B.C. n Aichbühl 4260 B.C.

IV III II Ib 3955 B.C.

| | | | | | LeonbergHöfingen 4400-3800 B.C. | |

| | | | | Eberdingen -Hochdorf 4300-3800 B.C. | |

? | | LudwigsburgSchlößlesfeld 4300-4000 B.C.

? | | RemseckAldingen ?

? | HeilbronnKlingenberg | | | | | | Ilsfeld-Ebene 4400- 3790 B.C. | Neckargart.Hetzenberg 5270+/-40 BP, 5080+/-50 BP | |

MK V

MK IV

MK III

MK II

Figure 1. Chronological and regional order of the Late Neolithic sites.

Figure 2. Location of the Late Neolithic sites in Southwest Germany

vicinity of the Swabian Alb Mountains like the sites near the Danube, the Federsee bog or at lake Constance are predominated by local raw materials. The sites that are situated in greater distance to the Swabian Alb Mountains like all the sites along the Neckar River valley as well as Reute-Schorrenried in the north alpine foreland are predominated by regional and supraregional raw materials. The local raw materials of the north alpine foreland –

Radiolarian chert, Quartz and Quartzite – and the local raw materials north of the Swabian Alb mountain range – Shelly limestone chert and Keuper chert – weren’t used very often because of their inferior quality. The lithic artifacts from supraregional raw material sources indicate widespread contacts for the different sites. Especially the tabular chert of the Franconia Alb Mountains was an 208

P. Kieselbach: Lithic Procurement and Production in the Late Neolithic of Southwestern Germany

Figure 3. Location of the raw materials used in the lithic assemblages.

regional and supraregional raw materials? An analysis of the production system helps answering these questions.

attractive raw material – particularly for the 3 sites of the Michelsberger group. In contrast to these east-relations, contacts to the north, south and west were generally smaller. It turned out that lithic artifacts made of south alpine and inner alpine raw materials were found almost only in the sites of the north alpine foreland. Lithic artifacts made of French, Belgian, Dutch and Baltic flint were found almost exclusively in the sites along the Neckar River. Although the communication between the sites north and south of the Swabian Alb Mountains wasn’t very pronounced, there are a few indications of contacts. For example the chalk flint artifacts of the Maas region were also found in sites south of the Swabian Alb Mountains.

ANALYSIS OF THE PRODUCTION SYSTEM

In summary the lithic supply was managed opportunistically and was dependent on the local lithic resource situation. Especially the identified supraregional raw materials of the 14 lithic assemblages indicate spacious supply thinking and a widespread communication network for the different culture groups of Southwest Germany. This suggests that the lithic supply of the different culture groups wasn’t territorially restricted.

Jean-Michel Geneste (1985) defined a production system that describes the systematic and chronological reduction sequences of the chaîne opératoire. It contains 6 reduction phases that a lithic tool passes through from the extraction till it’s discarded (fig. 5). Specific products are built in each phase, representing waste products or alternatively, target products that will be used or transformed in the following phase. In the analyzed lithic assemblages 23 artifact types could be distinguished and divided into the different reduction phases (fig. 6). Because refitting could only be managed for a few artifacts in the lithic assemblages the reconstruction of the complete chaîne opératoire and its conceptual operative scheme wasn’t feasible. However the division of the lithic assemblages into the different reduction phases allows reconstructing the technological processes that happened in the sites.

New questions arose from these results: 1. How were the raw materials transported to the sites? 2. Were there different mechanisms for the procurement and production of local,

The lithic assemblages of the different culture groups indicate varying production processes (fig. 7-10): In all lithic assemblages most artifacts represent Phase 2, whereas raw 209

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Aichbühl

Taubried I 34% 5% 1%

60%

5%

n=195

95%

AlleshausenHartöschle

n=113

Ehrenstein

2% 5% 0.7% 0.6%

3% 2% 2% 93%

93% n=60

n=903

LeonbergHöfingen

LudwigsburgSchlößlesfeld

9%

2% 3% 3%

1% 11%

92%

79%

EberdingenHochdorf

n=171

Remseck-Aldingen, Halden II

n=111

local regional supraregional - east supraregional - north/northwest supraregional - south

3% 16%

17%

3% 3% 2%

6%

72%

n=88

78%

HetzenbergNeckargartach

n=104

HeilbronnKlingenberg

Ilsfeld-Ebene 36%

52%

78% 7%

1% 5% n=138

41%

2% 0,9% 2%

6%

17%

51%

WangenHinterhorn

n=1752

n=243

ReuteSchorrenried

Ödenahlen

28% 100%

6% 2% 0,5% 1%

91%

n=212

n=518

2% 0,5% 6% n=257

Figure 4. Percentage of local, regional and supraregional raw material in the different Late Neolithic sites. 210

P. Kieselbach: Lithic Procurement and Production in the Late Neolithic of Southwestern Germany Acquisition

Production Reduction

Phase 0 extraction, testing

Phase 1 shaping the core, decortification, preparation of striking platform ? trimming flakes >2/3 cortex, striking platform flakes, blank cores, crested blades

? complete or tested rough outs

Consumption

Discard

Use

Discard

Retouching Phase 2 production of blanks

Phase 3 retouch or direct use

Phase 4 use, resharpening, recycling, transformation ? ? ? flakes and blades retouching debris, retouching debris, 2/3 cortex blades > 2/3, cortex flakes or blades > 2/3 crested blades or flakes, trimming flakes with cortex preshaped tabular chert cortex flakes < 2/3 cortex blades < 2/3 cortex flakes or blades < 2/3 flakes without cortex blades without cortex flakes or blades without cortex trimming flakes, striking platform flakes without cortex cores shatter „siret“ debris chips primary retouching debris, modification debris like burin spall, micro burin tabular chert tools secondary retouching and modification debris heat or frost shatter, natural shatter unknown type Figure 6. Artifact types of the different reduction phases.

chunks of phase 0, preparation products of phase 1 or retouch and modification debris of phase 3 and 4 are represented only by a few artifacts.

few retouch debris in the lithic assemblages of Aichbühl and Ehrenstein. The percentage of tools lies in all 4 assemblages between 39% and 48%. It is probable that these high amounts in the early excavated sites of Aichbühl, Taubried and Ehrenstein are caused by the lost of small flakes and chips.

In the 4 lithic assemblages of the Aichbühl group and Schussenried-south group (fig. 7) the high numbers of cores, trimming flakes and shatter (no. 6, 14, 15, 16) as well as flakes (no 3, 8, 11), which obtain more than 60 %, indicate extensive knapping. Because most of the flakes have cortex, it would suggest that complete raw material chunks were worked in the sites. Blades (no. 4, 9, 12) come up to 26%. Here Ehrenstein has the lowest percentage with 10%. It is likely that in Aichbühl and Ehrenstein tabular cherts (no. 2, 7, 20) were processed, too. The production, use and resharpening of tools (no. 19, 21) is documented with a

In the 4 assemblages of the Schwieberdinger and Schussenried-north group (fig. 8) cores, trimming flakes and shatter (no. 6, 14, 15, 16) are found always in small numbers and never exceeded 5%. Flakes (no. 3, 8, 11) only come up to 35%, whereas the percentage of blades (no. 4, 9, 12) is always more than 30 % and the amount of tools is more than 50%. It seems that knapping was insignificant and that mainly blades and tools were imported to the sites. The documented 211

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Aichbühl

Taubried I

25

30

25

20

Häufigkeit [%]

Häufigkeit [%]

20 15

modified (n=102) unmodified (n=110) 10

15

modified (n=53) unmodified (n=68)

10

5

5

0

2

1

0

3

0

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

3

4 Div.

Ehrenstein

AlleshausenHartöschle

25

2

1

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

30

25

20

modified (n=25) unmodified (n=39)

10

Häufigkeit [%]

Häufigkeit [%]

20 15

5

0

modified (n=779) unmodified (n=1225)

15

10

5

0

1

2

3

0

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Figure 7. Percentages of artifact types of the different reduction phases in the lithic assemblages of the Aichbühl and Schussenried-south group.

LudwigsburgSchlößlesfeld

Leonberg-Höfingen 35 25 30 20

20

modified (n=135) unmodified (n=70)

15 10

Häufigkeit [%]

Häufigkeit [%]

25

modified (n=82) unmodified (n=60)

10

5

5 0

15

0

1

2

3

0

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

1

2

3

4 Div.

Remseck-Aldingen

EberdingenHochdorf

30

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

25

25

20

15

modified (n=62) unmodified (n=51)

10

Häufigkeit [%]

Häufigkeit [%]

20

10

modified (n=69) unmodified (n=54)

5

5

0

15

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

0

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Figure 8. Percentages of artifact types of the different reduction phases in the lithic assemblages of the Schwieberdinger and Schussenried-north group. 212

P. Kieselbach: Lithic Procurement and Production in the Late Neolithic of Southwestern Germany

Ilsfeld-Ebene

NeckargartachHetzenberg

30

30

25

25

20

modified (n=49) unmodified (n=103)

15

Häufigkeit [%]

Häufigkeit [%]

20

10

10

5

5

0

modified (n=522) unmodified (n=1922)

15

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

0

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

HeilbronnKlingenberg

25

Häufigkeit [%]

20

modified (n=125) unmodified (n=178)

15

10

5

0

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Figure 9. Percentages of artifact types of the different reduction phases in the lithic assemblages of the Michelsberg group.

raw material chunks in Ludwigburg-Schlösslesfeld weren’t used for knapping but as hammer stones.

there is evidence for the processing of tabular chert (no. 2, 7, 20), too. The production, use and resharpening of tools is documented in all 3 assemblages with modification debris (no. 19, 21). Especially in Reute-Schorrenried the retouching of tools is demonstrated by a high number of chips (no. 18).

In the 3 assemblages of the Michelsberg group (fig. 9) cores, trimming flakes and shatter (no. 6, 14, 15, 16) lie between 7% and 12%. The amount of flakes (no. 3, 8, 11) is always low and reaches the highest percentage in Ilsfeld-Ebene with 36%. Blades (no. 4, 9, 12) obtain only 13%. The percentage of tools varies in the assemblages between 20% and 40%. In all 3 assemblages unworked tabular chert (no 2) as well as semi-finished and finished tools (no 7, 20) made of tabular chert are presented in high percentages. They reach 36% in Neckargartach-Hetzenberg. It seems that in all 3 assemblages the processing of tabular chert was more important than knapping of nodular raw materials.

In summary, the analysis of the production systems suggests a non-uniform behavior within the sites. For the sites that depend mainly on local raw materials like the sites near the Federsee bog and lake Constance, the whole chaîne opératoire is documented. For the sites that are located in greater distance to the Swabian Alb Mountains knapping of local and regional raw materials occurs too, but merely in low quantity. An additional import of blades and tools as well as blanks or tools of tabular cherts took place. So, the extent of local knapping and the quantity of imported semi-finished and finished tools is not so much tied to a culture group but is rather determined by the distance to the raw material sources.

In the 3 assemblages of the Pfyn group and Pfyn-Altheim group (fig. 10) cores, trimming flakes and shatter (no. 6, 14, 15, 16) lie between 12 and 14%. The percentages of flakes are in Wangen-Hinterhorn and Ödenahlen very high with up to 60%. In contrast Reute-Schorrenried has only 45% flakes. The amount of blades is always low and reaches only 15% in Reute-Schorrenried. It seems that in Wangen-Hinterhorn and Ödenahlen extensive knapping of complete raw material chunks took place. In Reute-Schorrenried knapping wasn’t that much. Maybe the lower amount of cortex flakes in ReuteSchorrenried results from an import of decortified raw materials. In Ödenahlen and especially in Reute-Schorrenried

Although the lithic supply presumably didn’t play a very big role in the economic system of the settlements, a constant and secure as well as less time consuming supply can be supposed. For the sites located close to the Swabian Alb Mountains mainly a direct procurement with complete raw material chunks is probable. For the sites that are in greater distance to the raw material sources of the Swabian Alb Mountains the higher material exploitation and additional import of blanks and tools suggest a different lithic supply: 213

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

Ödenahlen

Wangen-Hinterhorn 35

25

30 20

modified (n=75) unmodified (n=255)

10

Häufigkeit [%]

Häufigkeit [%]

25 15

20

modified (n=122) unmodified (n=411)

15 10

5

0

5

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

0

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Reute-Schorrenried 30

25

Häufigkeit [%]

20

15

modified (n=54) unmodified (n=116)

10

5

0

0

1

2

3

4 Div.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Figure 10. Percentages of artifact types of the different reduction phases in the lithic assemblages of the Pfyn and Pfyn-Altheim group.

For the sites of the Schwieberdinger and Schussenried-north group presumably a good supply system to the raw material sources of the Swabian Alb Mountains existed. However it is not clear, whether the supply network is similar as H.-C. Strien (1990) suspected for the former LBK settlements. Namely, a centralized exchange system for all settlements of the area, where people of one settlement procure and manufacture the lithic material and subsequently provide the neighboring settlements with semi-finished and finished tools. Because there is no evidence for such a ‘central place’ yet, an independent supply for each site must be considered. For the lithic assemblages of the Schussenried-south group, the higher amount of blades in Taubried I and AlleshausenHartöschle and the low amount of blades in Ehrenstein may also indicate an import of blades for the former or rather an additional export of blades for the latter. The same can be suggested for the Pfyn-Altheim assemblages. Here ReuteSchorrenried, which is located in greater distance to the Swabian Alb Mountains than Ödenahlen has a higher amount of blades and lower amount of cortex flakes. An import of blank cores or blades might occur.

supraregional raw material that presumably wasn’t transmitted by the usual supply system of the sites. It is more likely that these pieces are by-products of a long distance exchange of unknown goods and thus reflect primarily social contacts. In this connection the people of the Michelsberg group might transmitted the lithic artifacts made of chalk flint to the sites of the Schussenried-north and south group. It seems that the Michelsberg group, which expand over the whole Central Europe in the course of the Late Neolithic Period (Lüning 1998), maintained their contacts to the quarries of the Maas region. Especially blades and blade tools made of chalk flint that were found not only in lithic assemblages of the Michelsberg group but also of the Aichbühl, Schwieberdinger and Schussenried-south group suggest a special exchange and might characterized prestige goods. Because all 3 Michelsberg sites were strong defended enclosures, they might have played a special role for the exchange of lithic material.

The lithic artifacts made of supraregional raw material indicate a different supply. On the one hand there are the high amounts of tabular chert in the 3 assemblages of the Michelsberg group that suggest a continuous exchange system to the Franconian Alb Mountains. On the other hand there are in almost all sites single artifacts made of

Petra KIESELBACH M.A. Univ. Tübingen, Inst. f. Ur- und Frühgesch. u. Arch. d. Mittelalters Abt. f. Ältere Urgeschichte und Quartärökologie Burgsteige 11, Schloß Hohentübingen 72070 Tübingen GERMANY

Author’s address

214

P. Kieselbach: Lithic Procurement and Production in the Late Neolithic of Southwestern Germany

Bibliography

LÜNING, J. 1998, Betrachtungen über die Michelsberger Kultur. In: J. Biel, H. Schlichtherle, M. Strobel, A. Zeeb (Hrsg.), Die Michelsberger Kultur und ihre Randgebiete - Probleme der Entstehung, Chronologie und des Siedlungswesens. Kolloquium Hemmenhofen 1997. Materialh. Arch. Baden-Württemberg 43 (Stuttgart 1998) 277-289.

BIRÓ, K. 1998, Stones, Numbers - History? The Utilization of Lithic Raw Materials in the Middle and Late Neolithic of Hungary. Jour. Anthrop. Arch. 17, 1998, 1-18. GENESTE, J.-M. 1985, Analyse lithic d’industries Moustériennes du Périgord: une approche technologique du comportement des groupes humains au Paléolithique moyen (Dissertation, Université Bordeaux 1, 1985).

SCHLICHTHERLE, H. 1988, Das Jung- und Endneolithikum in Baden-Württemberg. Zum Stand der Forschung Aus siedlungsarchäologischer Sicht. In: D. Planck (Hrsg.), Archäologie in Württemberg. Ergebnisse und Perspektiven archäologischer Forschung von der Altsteinzeit bis zur Neuzeit (Stuttgart 1988) 91-110.

KIESELBACH, P. 2000, Metamorphose des Steins - vom Rohmaterial zum Kulturgut. Versorgungsaspekte und technische Prozesse der Silexverarbeitung von jungneolithischen Silexinventaren aus Südwestdeutschland. (Dissertation Universität Tübingen 2000).

STRIEN, H.C. 1990, Untersuchungen zur Bandkeramik in Württemberg (Dissertation Universität Frankfurt a.M. 1990).

215

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe

216

C. Sestier et al.: Débitage et façonnage de la métadolérite de type A (Plussulien)...

DÉBITAGE ET FAÇONNAGE DE LA MÉTADOLÉRITE DE TYPE A (PLUSSULIEN) : RÉALISATION DE TESTS ET IMPLICATIONS ARCHÉOLOGIQUES Claude SESTIER* & Charles-Tanguy LE ROUX**, avec la collaboration de Bertrand FAUCILLON**

Résumé : Le gîte primaire de Métadolérite de type A de Seledin (Plussulien, Côtes d’Armor, France) a été reconnu comme une source de matière première exploitée intensivement au cours du néolithique, approvisionnant un réseau d’échanges étendu à tout l’Ouest de la France. La mise au point d’une technique de débitage et façonnage de la métadolérite nous a permis de reproduire une chaîne opératoire complète pour les deux modes de façonnage connus des ébauches de haches. Ces observations ont été confrontées à l’analyse granulométrique des déchets d’exploitation et de façonnage des ébauches et comparées aux données archéologiques. Les informations ainsi obtenues permettent de proposer une nouvelle interprétation de certaines structures observées sur le site et soulignent l’intérêt de la méthode granulométrique pour l’étude des minières au néolithique. Abstract: The primary source of type A metadolerite of Seledin (Plussulien, Côtes d’Armor, France) has been recognised as a raw material source intensively exploited during the Neolithic, providing an exchange network extending across western France. The clarification of the reduction technique and tool preparation on metadolerite has permitted the reconstruction of the complete chaîne opératoire for the two preparation modes of hatchet preforms. These observations were compared with granulometric analysis of the reduction debris for preforms and archaeological data. The data obtained allow a new interpretation to be proposed for some structures observed on the site and highlight the potential of the granulometric method for the study of wastes found on Neolithic open quarries.

INTRODUCTION

haches en roches dans la péninsule Armoricaine, alors que d’abondantes sources de Dolérites et autres matériaux microcristallins existent. L’identification pétrographique reste la technique de choix dans l’étude de diffusion de ces matériaux, pouvant cependant être complétée par des études géochimiques afin d’en préciser l’origine régionale (à paraître, Chancerelle, SRA, Caen ). Des métadolérites sont aussi rencontrées en marge du Massif central (Vuillat 1995, Santallier 1986), cependant à plus de 500 km de distance, et en toute vraissemblance n’ont pas atteint les zones de diffusion de la métadolérite de Bretagne. Ces dernières métadolérites gardent d’ailleurs des différences de faciès pétrographique.

La diffusion de haches en Métadolérite de type A (MTDA) dans l’Ouest de la France a été reconnue très tôt par les néolithiciens (Giot 1952), cependant que le gîte primaire a finalement été découvert à la suite de prospections intensives (Le Roux 1965, 1971, 1990). Bien que les Dolérites et Métadolérites soient relativement fréquentes dans la zône Armoricaine, le faciès corresondant au type A de Giot n’a été trouvé en surface que sur le gîte de Roc’h-Pol à SeledinQuelfenec. Ce site culmine en un promontoire rocheux 70 mètres plus haut que la rivière locale (le Daoulas, 193 mètres d’altitude), le massif rocheux actuellement visible mesure 35 mètres sur 15 mètres, la zone couverte de déchets archéologiques s’étend sur plusieurs dizaines d’hectares. La zone proche du sommet de l’affleurement a été fouillée par l’équipe de C-T Leroux (Le-Roux 1999), les datations C14 montrent une exploitation 4200-2200 BC (Tableau 1a). Des découvertes régionales de haches de MTDA en contexte funéraire confirment que le gite a été exploité pendant une période très longue, et permettent de dater la durée des échanges (l’essentiel des découvertes de haches importées étant faite hors stratigraphie). La métadolérite du type A présente une très grande résistance mécanique, comparée à d’autres Dolérites régionales ou aux matériaux siliçeux exploités au Néolithique (Tableau 1b). Ainsi, les résistances à la compression, à la flexion et à la traction sont deux fois celles des faciès grossiers de Dolérites, et l’énergie nécessaire pour détacher un éclat est estimée 3,5 fois celle nécessaire pour le silex dans les mêmes conditions (Le Roux 1999). Ce faciès extrêmement tenace a été très apprécié des néolithiques, puisqu’il représente de 40 à 80% de l’effectif connu des

Les caractéristiques mécaniques remarquables de cette roche, ainsi que son exploitation spécifique en font un élément précieux pour la compréhension des logiques de production et d’échange au néolithique, période voyant s’intensifier et se concentrer l’exploitation de ressources lithiques régionales. Les modèles de structuration des terroirs néolithiques tenant compte de la localisation de ces ressources et des particularités topographiques et écologiques, nécessitent pour leur validation bien plus que l’analogie ethnologique. Pour le gîte de Plussulien, des ateliers « secondaires » ont probablement été identifiés (Le Roux 1999). L’énorme volume de déchets produit sur la colline de Roc’h-Pol (dont seule une petite partie a été étudiée) et son étude typologique n’a cependant pas permis de distinguer de façon certaine de simples déchets d’abattage de « carrière » (le feu ayant été utilisé), de déchets provenant de façonnage d’ébauches de haches et même de bouchardage. Les connaissances acquises pour le silex (Stahle 1982) sont en effet d’une utilité limitée, et leur transposition à 217

Le Néolithique au proche orient et en Europe / The Neolithic in the Near East and Europe Tableau 1a : Repères chronoculturels (d’après Le Roux 1999). Sites où ont été trouvées des haches en métadolérite A (d'après Le Roux 1999) (Côtes-d'Armor)

PLUSSULIEN, Quelfennec

An.BC Cal. 21 dates entre GIF 1539 et GIF 1877

4450-3700 4450-3500

Chasséen

(Oise)

JONCQUIERES, Mont d'Huette

LY 2918, 2919, 2970

Chasséen/MK

(Pas-de-Calais)

LIEVIN, rue Clovis

GIF 4022, 4023

4460-3700

Chasséen

(Maine-et-Loire)

LE LION D'ANGERS, L'île-Briand

GIF 3818

4000-3300

NMII Groupe Kerugou

(Morbihan) (LoireAtlantique)

ARZON, le Petit-Monnt II

GIF 6842, 6843, 6845 GIF 4936, 4937, 4951, 8106

3950-3100

Artenac

(Vendée)

XANTON-CHASSENON, Pierre-virante

GIF 3762

2900-2200

(Côtes d'Armor)

PLELAUFF, Kerivoelen

GIF 3762

2900-2200

(Côtes-d'Armor)

PLUSSULIEN, Quelfennec

Campaniforme

MACHECOUL, Les Prises

3800-2570

2900-2200

Tableau 1b : Caractéristiques mécaniques de la métadolérite du type A comparée à d’autres roches. Pour la flexion, Tf = 3M/b3, avec M = Cm (entre les couteaux)xN (force appliquée), b(Cm) = dimension de l’éprouvette prismatique de section carrée

Compression 6 2 10 N/m

Flexion 4 2 10 N/m

Traction 6 2 10 N/m

Eclatement J

Origine Plussulien plussulien Plouagat Guipel Fibrolite Coat-meal "Tuff" de Great-Langdale, Cumbria (GB) (Bradley 1992)

226 103 135 140 112 nd nd

75-98 28-35 16-20 62-77 38 nd nd

287 149 152 186 152 34±3 42±2

4,4-4,6 2,5-2,8 3,9-4,4 2,9-3,4 1,5-1,7 nd nd

Silex Grand pressigny (F) Silex grand pressigny Jaspe de Fontmaure (F)

nd nd nd

nd nd nd

nd 68 38

1,3 nd nd

Marbre (Carrare)

nd

nd

7à8

nd

Métadolérite Faciès fin (type A) Facièes grossier Dolérites

(Pradel 1965) (Pradel 1965) (Lama 1978)

Tableau 1. (a) Repères chronologiques, pour l’exploitation de la métadolérite de type A; (b) Caractéristiques mécaniques de la métadolérite de type A et d’autres matériaux.

l’étude de ce matériau reste à valider. Pour ce qui concerne le façonnage des haches, deux stratégies ont été clairement identifiées : l’une utilise des bloc naturels produits par les diaclases (« briquettes »), l’autre utilise des gros éclats décimétriques voir pluridécimétriques, les ébauches portant alors une face lisse (face d’éclatement) et une face retouchée (Fig.1e). Nous avons par ailleurs mis en évidence des preuves certaines d’un « savoir-faire » classique pour le néolithique, comme l’utilisation du façonnage bifacial, et la réalisation d’enlèvements alternés, formant des « crêtes » (Photo.1a,1b). Cette évidence d’un niveau technique équivalent à ce qu’on peut attendre d’un producteur de haches en silex nous a permis de relativiser l’impression subjective de maladresse ressentie pour une forte proportion d’ébauches inachevées. La pratique du débitage et façonnage de la MTDA nous ont confirmé la difficulté inhérente à ce produit, tout en nous permettant d’évaluer les limites du possible en fonction des techniques que nous connaissons. Après avoir mis au point une méthode de débitage et façonnage de la métadolérite donnant des produits compatibles avec le matériel archéologique, la méthode granulométrique a été utilisée de façon à comparer

la production de déchets et éclats, archéologiques et expérimentaux. Cette méthode a été testée pour pouvoir ultérieurement être appliquée à des cas similaires. Ce préalable était en effet nécessaire, les résultats de granulométrie ayant été interprétés (bien que sous réserves très prudentes) sur la base des connaissances acquises pour la taille du silex. MATÉRIEL ET MÉTHODES Méthode granulométrique : le détail est publié dans Le Roux (1999, chapitre XII). En bref, les fractions « grossières » ont été analysées par les méthodes de la granulométrie sédimentologique classique (Riviere 1977), avec certaines adaptations en raison des rapports longueur/largeur des éclats. Les échantillons archéologiques étudiés proviennent de divers niveaux stratigraphiques, chacun pèse entre 10 à 30 Kg. Les fractions fines de 10 mm à 50 µm ont été traitées sur une colonne aux normes AFNOR. Les classes granulométriques sont (µm) : refus > 1000, fractions 800-630-500-400-315250-200-160-125-100-80-63-50-40 et