Prehistoric Art As Prehistoric Culture: Studies in Honour of Professor Rodrigo De Balbin-Behrmann: Studies in Honour of Professor Rodrigo de Balbín-Behrmann 1784912220, 9781784912222

Citation preview

Prehistoric Art as Prehistoric Culture Studies in Honour of Professor Rodrigo de Balbín-Behrmann edited by

Primitiva Bueno-Ramírez and Paul G. Bahn

Prehistoric Art as Prehistoric Culture Studies in Honour of Professor Rodrigo de Balbín-Behrmann

edited by

Primitiva Bueno-Ramírez and Paul G. Bahn

Archaeopress Archaeology

Archaeopress Publishing Ltd Gordon House 276 Banbury Road Oxford OX2 7ED

www.archaeopress.com

ISBN 978 1 78491 222 2 ISBN 978 1 78491 223 9 (e-Pdf)

© Archaeopress and the individual authors 2015 Cover: “Galería de los Antropomorfos” Tito Bustillo cave, Asturias, Spain. Photos by Rodrigo de Balbín

All rights reserved. No part of this book may be reproduced or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners.

This book is available direct from Archaeopress or from our website www.archaeopress.com

Contents List of Figures and Tables������������������������������������������������������������������������������������������������������������������������������������� iii List of contributors����������������������������������������������������������������������������������������������������������������������������������������������� v Prehistoric Art as Prehistoric Culture������������������������������������������������������������������������������������������������������������������ vii Primitiva Bueno-Ramírez and Paul Bahn ‘Science’ versus Archaeology: Palaeolithic Rock Art at the beginning of the 21st century��������������������������������������1 José-Javier Alcolea-González and César González-Sainz Raman spectroscopy of prehistoric pictorial materials����������������������������������������������������������������������������������������11 Antonio Hernanz Prehistoric rock art and non-invasive analysis. Rouffignac as a case study����������������������������������������������������������21 Patrick Paillet Reasoning processes in prehistoric art interpretation�����������������������������������������������������������������������������������������25 Sophie A. de Beaune Are hand stencils in European cave art older than we think? An evaluation of the existing data and their potential implications��������������������������������������������������������������������������������������������������������������������������������������������������������31 Paul Pettitt, Pablo Arias, Marcos García-Diez, Dirk Hoffmann, Alfredo Maximiano Castillejo, Roberto Ontañon-Peredo, Alistair Pike and João Zilhão Regional ontologies in the Early Upper Palaeolithic: the place of mammoth and cave lion in the ‘belief world’ (Glaubenswelt) of the Swabian Aurignacian�������������������������������������������������������������������������������������������������������45 Shumon T. Hussain and Harald Floss Aurignacian art in the caves and rock-shelters of Aquitaine (France)�������������������������������������������������������������������59 Brigitte and Gilles Delluc Fuente del Trucho, Huesca (Spain): Reading interaction in Palaeolithic art����������������������������������������������������������69 Pilar Utrilla and Manuel Bea Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon�������������������������������79 Paul G. Bahn Decorated sites and habitat: social appropriation of territories��������������������������������������������������������������������������93 Denis Vialou Deep caves, ritual and graphic expression: a critical review of the archaeological evidence on hypogean human activity during the Upper Palaeolithic/Magdalenian�������������������������������������������������������������������������������������������99 Pablo Arias Magdalenian settlement-subsistence systems in Cantabrian Spain: contributions from El Mirón Cave�������������� 111 Lawrence G. Straus, Manuel González Morales, Ana B. Marín-Arroyo and Lisa M. Fontes The Upper Palaeolithic rock art of Portugal in its Iberian context���������������������������������������������������������������������� 123 André Tomás Santos, Maria de Jesus Sanches and Joana Castro Teixeira

i

Old panels and new readings. La Pileta and pre-Solutrean graphics in Southern Iberia������������������������������������� 135 Miguel Cortés Sánchez, María D. Simón Vallejo, Rubén Parrilla Giráldez, and Lydia Calle Román Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation �������������������������� 145 Valentín Villaverde Small seeds for big debates: Past and present contributions to Palaeoart studies from North-eastern Iberia ��� 157 José María Fullola, Ines Domingo, Didac Román, María Pilar García-Argüelles, Marcos García-Díez and Jorge Nadal Throwing light on the hidden corners. New data on Palaeolithic art from NW Iberia���������������������������������������� 171 Ramón Fábregas Valcarce, Arturo de Lombera-Hermida, Ramón Viñas Vallverdú, Xose Pedro Rodríguez-Álvarez, and Sofia Soares Figueiredo

ii

List of Figures and Tables ‘Science’ versus Archaeology: Palaeolithic Rock Art at the beginning of the 21st century Figure 1. Schematic summary of the main changes�������������������������������������������������������������������������������������������������������������������������� 4

Raman spectroscopy of prehistoric pictorial materials Table 1. Chronological overview of previous work on Raman spectroscopy of prehistoric rock painting materials ���������������������� 13 Figure 1. In situ micro Raman spectrum of a pigmented dot in El Mirón Cave (Cantabria, Spain) ������������������������������������������������� 14 Figure 2. In-lab micro Raman spectrum of a micro-specimen of red pigment from Pruneda Cave (Onís, Asturias, Spain)������������ 14 Figure 3. In-lab micro Raman spectrum of a micro-specimen of red pigment from Pruneda Cave (Onís, Asturias, Spain)������������ 15

Prehistoric rock art and non-invasive analysis. Rouffignac as a case study Figure 1. Raman microspectometry system������������������������������������������������������������������������������������������������������������������������������������ 22

Are hand stencils in European cave art older than we think? An evaluation of the existing data and their potential implications Table 1. Corpus of caves containing hand stencils/prints known to the authors. �������������������������������������������������������������������������� 35 Figure 1. Selection of French and Spanish hand stencils. ��������������������������������������������������������������������������������������������������������������� 36

Regional ontologies in the Early Upper Palaeolithic: the place of mammoth and cave lion in the ‘belief world’ (Glaubenswelt) of the Swabian Aurignacian Figure 1. Map of Central and Southern Germany showing the location of the Swabian Aurignacian sites ����������������������������������� 47 Figure 2. Frequency of different animal representations in the visual art repertoire of the Swabian Aurignacian.����������������������� 47 Figure 3. Hohle Fels, chaîne opératoire of ivory beads in the Aurignacian.������������������������������������������������������������������������������������ 50 Figure 4. Mammoth representations from the Swabian Aurignacian. ������������������������������������������������������������������������������������������� 51 Figure 5. Cave lion and hybrid “lion-man” representations from the Swabian Aurignacian. ��������������������������������������������������������� 53

Aurignacian art in the caves and rock-shelters of Aquitaine (France) Figure 1. Abri Blanchard. 3 vulvas engraved on a block. ���������������������������������������������������������������������������������������������������������������� 60 Figure 2. Abri Blanchard. Portion of a horse, with the belly and limbs, ����������������������������������������������������������������������������������������� 61 Figure 3. La Ferrassie. 2 vulvas engraved on a block. ��������������������������������������������������������������������������������������������������������������������� 62 Figure 4. The cave of La Cavaille: a, the engraved decoration on the left wall (after B. and G. Delluc) ; b, the right mammoth ��� 64 Figure 5. The cave of Pair-non-Pair: a, the engraved decoration of the first panel (after B. and G. Delluc); b, the left horse ������� 66

Fuente del Trucho, Huesca (Spain): Reading interaction in Palaeolithic art Figure 1. Location of the cave of Fuente del Trucho. In the lower image, note how the daylight passes through the natural hole in the rock and illuminates the engravings.���������������������������������������������������������������������������������������������������������������������������������������� 70 Fig. 2. Engravings from the external sanctuary.������������������������������������������������������������������������������������������������������������������������������� 71 Fig. 3. Plan of the ceiling and decorated panels.����������������������������������������������������������������������������������������������������������������������������� 71 Fig. 4. Dots series. 1. Panel VI and VII; 2. Panel XII; 3. Panel XXI; 4. Panel XV. �������������������������������������������������������������������������������� 73 Fig. 5. Depicted horses in Fuente del Trucho. ��������������������������������������������������������������������������������������������������������������������������������� 75

Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon Figure 1a/b. The big horse petroglyph at Shishkino, Siberia. ��������������������������������������������������������������������������������������������������������� 80 Figure 2. The pecked horse at Piedras Blancas . ����������������������������������������������������������������������������������������������������������������������������� 83 Figure 3. Petroglyphs of horses and other motifs, Hunsrück, Germany. ���������������������������������������������������������������������������������������� 85 Figure 4. Petroglyphs of a fish and aurochs, Qurta, Egypt. ������������������������������������������������������������������������������������������������������������� 87 Figure 5. Petroglyph of aurochs, Subeira, Egypt. ���������������������������������������������������������������������������������������������������������������������������� 88

Decorated sites and habitat: social appropriation of territories. Deep caves, ritual and graphic expression: a critical review of the archaeological evidence on hypogean human activity during the Upper Palaeolithic/Magdalenian Figure 1. El Juyo. Late Lower Magdalenian structures ����������������������������������������������������������������������������������������������������������������� 102 Figure 2. Limestone block interpreted as a mixed representation of the head of a human and a feline ������������������������������������� 102 Figure 3. La Garma. Structure IVC.������������������������������������������������������������������������������������������������������������������������������������������������ 103 Figure 4. La Garma, Zone VI. Bison metapodial on the floor of the Lower Gallery.���������������������������������������������������������������������� 104 Figure 5. La Garma, Zone III. Stalagmite fragment in a hole in the wall of the cave.�������������������������������������������������������������������� 105 Figure 6. La Garma, Zone I. Palaeolithic path in a passage above Zone I.������������������������������������������������������������������������������������� 106

iii

Magdalenian settlement-subsistencesystems in Cantabrian Spain: contributions from El Mirón Cave Figure 1. Major Magdalenian Sites and Flint Sources in Cantabrian Spain ����������������������������������������������������������������������������������� 112 Figure 2. Upper Magdalenian harpoon fragments from El Mirón. ����������������������������������������������������������������������������������������������� 117

The Upper Palaeolithic rock art of Portugal in its Iberian context Figure 1. Portuguese Palaeolithic rock art, plus Molino Manzánez ���������������������������������������������������������������������������������������������� 124 Figure 2. Panel 31 of Foz do Tua rockshelter. �������������������������������������������������������������������������������������������������������������������������������� 125 Figure 3. Lejeune’s figure 59 of Escoural. ������������������������������������������������������������������������������������������������������������������������������������� 126 Figure 4. Horse of the left sector of Quinta da Barca 23 (Côa Valley).������������������������������������������������������������������������������������������ 128 Figure 5. Detail of rock 7 of Canada da Moreira-Côa Valley ��������������������������������������������������������������������������������������������������������� 128

Old panels and new readings. La Pileta and pre-Solutrean graphics in Southern Iberia Figure 1. Topography and first artistic horizon of La Pileta ����������������������������������������������������������������������������������������������������������� 136 Figure 2. Rhino Panel �������������������������������������������������������������������������������������������������������������������������������������������������������������������� 138 Figure 3. A) Black serpentiforms, B) Red sinuous sign, C) Positive hand prints, D) Serpentiform, E) Yellow ibex, F) Yellow horse. E-F � 140 Figure 4. EUP and archaeological sites mentioned in the text.����������������������������������������������������������������������������������������������������� 142

Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation Figure 1. Distribution of sites with parietal art in the Mediterranean and the Southern Iberian Peninsula. ������������������������������� 146 Figure 2. Distribution of sites with parietal art in the Mediterranean and Cantabrian regions ��������������������������������������������������� 149 Table 1. Zoomorphs represented in Mediterranean and South Iberian parietal art.�������������������������������������������������������������������� 150 Table 2. Zoomorphs represented in Mediterranean and South Iberian portable art.������������������������������������������������������������������� 151

Small seeds for big debates: Past and present contributions to Palaeoart studies from North-eastern Iberia Figure 1. Timeline of the discoveries in Northeastern Iberia ������������������������������������������������������������������������������������������������������� 158 Figure 2. Examples of portable art from Molí del Salt ������������������������������������������������������������������������������������������������������������������ 160 Figure 3. Other portable art objects from Northeastern Iberia. ��������������������������������������������������������������������������������������������������� 162 Table 1. Inventory of finds, characteristics and subject matter .��������������������������������������������������������������������������������������������������� 164 Figure 4. Geographical distribution of Palaeolithic and Epipalaeolithic art in Northeastern Iberia���������������������������������������������� 165

Throwing light on the hidden corners. New data on Palaeolithic art from NW Iberia Figure 1. Location of the main sites mentioned in the text. ��������������������������������������������������������������������������������������������������������� 172 Table 1. Radiocarbon dates from Cova Eirós.�������������������������������������������������������������������������������������������������������������������������������� 172 Figure 2. Portable art from Galicia. ���������������������������������������������������������������������������������������������������������������������������������������������� 173 Figure 3. Plan of the Cova Eirós cave and location of the panels. ������������������������������������������������������������������������������������������������ 174 Figure 4. Zoomorphic representations from Cova Eirós. �������������������������������������������������������������������������������������������������������������� 175 Figure 5. Representation of A) an ibex (A), B) a horse (B), C) an aurochs from Foz do Medal Terrace. ���������������������������������������� 178

iv

List of contributors José-Javier Alcolea González Área de Prehistoria, Departamento de Historia y Filosofía. Universidad de Alcalá, Spain

María Pilar García-Argüelles Universitat de Barcelona / SERP, Spain Marcos García-Díez Departamento de Geografía, Prehistoria y Arquelogía, Facultad de Letras, Universidad del Pais Vasco, UPV/EHU, c/ Tomás y Valiente s/n, 01006 Vitoria-Gazteiz, Álava, Spain

Pablo Arias The Cantabria Institute for Prehistoric Research, University of Cantabria, Edificio Interfacultativo, Avda. Los Castros s/n, 39005 Santander, Spain Paul G. Bahn Hull, United Kingdom

Manuel González Morales Instituto Internacional de Investigaciones Prehistóricas, Universidad de Cantabria, 39005 Santander, Spain

Manuel Bea “Torres Quevedo” Postdoc Researcher (MINECO+3D Scanner). Group “PPVE”, Spain

Cesar González Sainz Departamento de Ciencias Históricas, Universidad de Cantabria, Spain

Sophie de Beaune Jean Moulin University, Lyon /
“Archéologies et Sciences de l’Antiquité” Research Unit, Nanterre, France

Antonio Hernanz Departamento de Ciencias y Técnicas Fisicoquímicas, Facultad de Ciencias, Universidad Nacional de Educación a Distancia (UNED), Paseo Senda del Rey 9, E-28040 Madrid, Spain

Lydia Calle Román Grupo HUM-949. Tellus. Prehistoria y Arqueología en el sur de Iberia. Universidad de Seville, Spain

Dirk Hoffmann Max Planck Institute for Evolutionary Anthropology Department of Human Evolution, Deutscher Platz 6, 04103 Leipzig, Germany

Miguel Cortés Sánchez Departamento de Prehistoria y Arqueología. Facultad de Geografía e Historia, Universidad de Sevilla, c/. María de Padilla, s/n. 41004. Seville, Spain

Shumon T. Hussain Faculty of Archaeology, Leiden University, the Netherlands

Brigitte & Gilles Delluc Associate researchers in the Department of Prehistory, National Museum of Natural History, Paris, France UMR 7194 du CNRS

Arturo de Lombera-Hermida Grupo de Estudos para a Prehistoria do Noroeste (GEPN), Dpto Historia I, Universidade de Santiago de Compostela, Pz. Universidade nº1, 15782 Santiago de Compostela, Spain

Ines Domingo ICREA at Universitat de Barcelona / SERP, Spain

Ana Belen Marín-Arroyo Instituto Internacional de Investigaciones Prehistóricas, Universidad de Cantabria, 39005 Santander, Spain

Ramón Fábregas Valcarce Grupo de Estudos para a Prehistoria do Noroeste (GEPN), Dpto Historia I, Universidade de Santiago de Compostela, Pz. Universidade nº1, 15782 Santiago de Compostela, Spain

Alfredo Maximiano Castillejo Facultad de Filosofía y Letras UNAM, Circuito Interior. Ciudad Universitaria, s/n. C.P. 04510. Mexico, DF. Mexico Jorge Nadal Universitat de Barcelona / SERP, Spain

Sofia Soares Figueiredo Lab2PT- Landscapes, Heritage and Territory Laboratory. University of Minho, Portugal

Roberto Ontañon Peredo The Cantabria Institute for Prehistoric Research - Cuevas Prehistóricas de Cantabria, Carretera de las Cuevas s/n, 39670 Puente Viesgo, Spain

Harald Floss Department for Early Prehistory and Quaternary Ecology, University of Tübingen, Germany

Patrick Paillet Muséum National d’Histoire Naturelle, Département de Préhistoire, UMR 7194. Musée de l’Homme, 17 place du Trocadéro, 75116 Paris, France

Lisa M. Fontes Department of Anthropology, University of New Mexico, Albuquerque, NM 87131, USA José Maria Fullola Universitat de Barcelona / SERP, Spain

v

Ruben Parrilla Giráldez Grupo HUM-949. Tellus. Prehistoria y Arqueología en el sur de Iberia. Universidad de Seville, Spain

María D. Simón Vallejo Grupo HUM-949. Tellus. Prehistoria y Arqueología en el sur de Iberia. Universidad de Seville, Spain

Paul Pettitt Department of Archaeology, Durham University, South Road, Durham DH1 3LE, United Kingdom

Lawrence G. Straus Department of Anthropology, University of New Mexico, Albuquerque, NM 87131, USA

Alistair Pike Department of Archaeology, University of Southampton, Avenue Campus, Highfield Road, Southampton, SO17 1BF, United Kingdom

Joana Castro Teixeira Researcher of the Transdisciplinary “Culture, Space and Memory” Research Centre (CITCEM), Portugal Pilar Utrilla Area of Prehistory, Group “PPVE”, University of Zaragoza, Spain

Xose Pedro Rodríguez-Alvarez Area de Prehistoria, Universitat Rovira i Virgili (URV). Avinguda de Catalunya 35, 43002 Tarragona, Spain

Denis Vialou Muséum National d’Histoire Naturelle, Paris, France

Didac Román Universitat de Barcelona / SERP, Spain

Valenín Villaverde Universitat de València, Spain

Maria de Jesus Sanches Faculty of Arts and Humanities-University of Porto; Researcher of the Transdisciplinary “Culture, Space and Memory” Research Centre (CITCEM), Portugal

Ramón Viñas Vallverdú Area de Prehistoria, Universitat Rovira i Virgili (URV). Avinguda de Catalunya 35, 43002 Tarragona, Spain

André Tomás Santos Fundação Côa Parque, Portugal

João Zilhão University of Barcelona/ICREA, Departament de Prehistòria, Història Antiga i Arqueologia, “Grup de Recerca” SERP SGR2014-00108, c/ Montalegre 6, 08001 Barcelona, Spain

vi

Prehistoric Art as Prehistoric Culture Primitiva Bueno-Ramírez and Paul Bahn The retirement of Prof. Balbín, who has played an active part in the changes in the approaches that have occurred in Palaeolithic art research in the last 30 years, is an excellent occasion for a reflection on the value of prehistoric art studies as a factor of analysis in the culture of the human groups who produced the art. The symbols on the durable surfaces in caves, on rocks in the open air and on portable artifacts are some of the best ways to approach an understanding of Upper Palaeolithic groups. Their social and territorial value is the basis of an analysis of the position of the hunter-gatherers, their preferred terrains, the topography and the uses made of them. This view, that extends outside caves as the sole containers of Palaeolithic art and as the depositary of functionalities separate from everyday life, is one that has most drastically changed the interpretation of the art. Prof. Balbín is one of the promoters of the new interpretations in which art in the open air and nuances to the generalised function of shrine have been developed theoretically and practically, with convincing results. Our aim is to present an up-dated view of the latest trends in current research. The lines guiding the research of some of the teams that are generating knowledge, and the theoretical and practical proposals they are using, may be deduced through the seventeen papers offered here. A reflection on changes in Palaeolithic art interpretation The profound changes that the study of prehistoric representations has experienced in recent years are not exempt from a large part of the ideology that has governed this kind of work from the first discoveries to the present day (Alcolea & González-Sainz, this volume; de Beaune, this volume). Aspects such as the function of shrine or production by shamans form part of the same set of interpretations that tend to describe a past full of mythology and groups with little cultural cohesion. In contrast, expressions like language, symbolic interaction and graphic markers clash head-on with the opinion reigning in Western European society since the 19th century and which strongly rejected the idea of human beings with a high level of social and symbolic development. If there is one thing that should be highlighted about Palaeolithic art research in the last 30 years, it is that some of the old controversies have reappeared, with a clear impact in the mass media, while new debates have also arisen. It was impossible to include all the studies we would have liked to, but we thought it was interesting to show the various interpretations of some of these controversies. The most significant aspects of the interpretation of Palaeolithic art in recent years possess a vital point of reference in Prof. Balbín’s constant work: chronology, technical applications enabling a reconstruction of the processes involved in the decoration, and the relationship with the territory are three of the fields in which we can divide the new approaches in Palaeolithic art interpretation; each one with its nuances and implications. The generation of prehistorians trained in the 1970s has spearheaded this transformation. The knowledge taken as granted in the framework of French science, with Leroi-Gourhan’s works as its point of reference, has been enriched by the introduction of technical applications, including the systems for the topographic documentation of the surfaces, analyses of the composition of pigments (Hernanz and Paillet, this volume), direct radiocarbon dating and, more recently, dating of calcite (see Alcolea & González-Sainz, this volume). Perfectionism, sometimes exhausting, in documentation in the field and a critical spirit are the tools with which Balbín has succeeded in undermining some of the classic clichés about Palaeolithic art. Above all, the one that restricted these symbols to caves. The confirmation of Palaeolithic art in the open air and its association with the habitat have opened up new lines of research with which to assess the Ice Age hunters (Arias, this volume). The development of his technical and thematic methodology for cave art, his proposals of stylistic chronology now fully confirmed, and his analysis of the connection between vii

representation and territory, open new perspectives for the study of Upper Palaeolithic groups and their immediate heirs (Bahn, this volume). The oldest dates have begun to be confirmed through a programme to date calcite deposits in several caves in northern Spain (Pettitt et al., this volume), and have erased the assumed distance between portable and parietal art (Hussain & Floss, this volume). Two points have been derived from this programme. The first is the evidence that, despite difficulties in funding, the Iberian teams maintain a very high level in Palaeolithic art research, with the collaboration of Anglo-Saxon specialists in the introduction and application of new technologies. The second, and perhaps the most important as regards some of the issues affecting chronological interpretations, has been the re-examination of the issues in radiocarbon dating from a new viewpoint. The dating of thin calcite growths has once again made clear the complex casuistry of radiocarbon dating of figures painted on cave walls, whose specific biography is hard to reconstruct, as Balbín pointed out some time ago. The response of some French teams has not taken long: a clear opposition to the technique as it damages the paintings (they never take samples of the paintings) has left radiocarbon as currently the only viable means to date Palaeolithic art. The ideal situation: a documented archaeological site, a radiocarbon date and dating of calcite, is uncommon but cases like the Passage of the Anthropomorphs in the Cave of Tito Bustillo, where these circumstances converge, provide justified hopes that experimentation with the dating of calcite will be one of the methods for future development. Chronologies of about 40,000 BP were to be expected in the framework of recent research. In this respect, the radiocarbon dates at Grotte Chauvet are not an exception. They are exceptional in that they are used to justify the representation of movement in the oldest phases of Palaeolithic art. It is the style of the figures that is less coherent and which gives rise to discussions. We wanted this touch of controversy to be seen through the contributions by Delluc, Pettitt et al., and Alcolea and González-Sainz. It is to be expected that continuing research in Chauvet will provide an archaeological context, and direct dates for the paintings and calcite deposits that will add high-quality data to the debate. One of the most interesting facets of chronology in recent years is that the oldest phases are very widespread geographically and are not concentrated in the Upper Palaeolithic of southern France. Balbín had already noted this as regards the south and west of the Iberian Peninsula, and new documentation is now available for northern Europe and Africa (Bahn, this volume). The ongoing research in Portugal (Santos et al., this volume) and Andalucia (Cortes et al., this volume), and the discoveries in Galicia (Fábregas et al., this volume) consolidate these new areas for research on Upper Palaeolithic hunters. The symbols are the most visible evidence of human groups in areas where it used to be thought they had not been present. Leroi-Gourhan’s theories seem to have become reduced to characterising an archaic Palaeolithic art and a recent Palaeolithic art (see references in Alcolea & González-Sainz, this volume). Speaking of styles is convincing as a way to overcome the cultural attributions that only function in Europe and only in certain parts of the continent. Some nuances, like the long chronology of archaic symbols, especially the claviforms, hands and venuses, are being confirmed thanks to the studies carried out in the Iberian Peninsula (Pettitt et al., this volume) and other parts of Europe (Hussain & Floss, this volume). The continuance of these during millennia fortified graphic traditions deeply rooted in the collective imagination of Palaeolithic hunters, projecting a picture of greater ideological continuity between the contents of the archaic and recent phases, an idea that also lies in Leroi-Gourhan’s hypotheses. The floruit of Palaeolithic art is the time of the densest human occupation (Vialou, this volume), associating human groups and graphic markers in an interesting point of analysis to assess the parietal ensembles in the framework of their social construction. Together with evidence for the long temporal duration of these systems for marking inhabited spaces (about 30,000 years, which is 150 times longer than the 2000 years of our culture), this should stimulate new reflections on the establishment of consolidated codes among Upper Palaeolithic hunters, the social systems that maintained them over the generations, and the gradual transformation of the codes towards new ways of marking the same spaces. The discussion about the more or less sudden disappearance of this collective imagination is really the debate between a diffusionist history and the perspectives of a social history that locates one of the cultural foundations of the way to production in the tradition and persistence in the sites. This interpretation is in vogue

viii

again with the hypothesis of a Style V dated directly in archaeological contexts in the Iberian Peninsula, attesting longer chronologies in the process of the transformation of the contents of Upper Palaeolithic art. This hypothesis is another of Prof. Balbín’s achievements, as he took part in the development of the arguments supporting the continuation of parietal and portable art in a time after 8000 BP. The definition of a Style V, incorporating Leroi-Gourhan’s stylistic theory, and adopted by Roussot for this phase, is still a cause for discussion among prehistorians. Some of the comments in the papers by Fullola et al. and Santos et al. give an idea of this debate. Its verification in Cova Eirós adds a cave parietal ensemble with direct dates to the site of Cueva Palomera in Ojo Guareña, Burgos, which is well-known and dated directly. In the Iberian Peninsula, it may be noted that, by accepting a Style V, some of the representations in Spanish Levantine art would have been produced at the end of the Ice Age, moving back the chronology of the Neolithic that most researchers have accepted since the second half of the 20th century. It is very likely that, as the direct dating systems known for other European areas are extended to Levantine art, the confirmation of this hypothesis will become more definite. The increasingly well documented sites in the west of the Iberian Peninsula (Fábregas et al., this volume; Santos et al., this volume) confirm a line of research that will contribute high-quality data to the study of the processes of the transformation in the contents of Upper Palaeolithic hunters’ art. Indeed, the Iberian Peninsula is an excellent laboratory for research programmes in this field, as the abundant post-Ice Age art is powerful evidence for the weight of tradition in marking travelled areas ever since the earliest times. In conclusion As stated above, the most important aspects of the interpretation of Palaeolithic art in recent years find a key point of reference in Prof. Balbín’s ceaseless work. They are current topics in a discipline that accepts new approaches with some reticence, but which has finally validated a panorama of Palaeolithic art assemblages in different types of locations, including in the open air. The discovery of open-air Palaeolithic art impacts directly on the location of the hunter-gatherer groups. From the traditional concentration on the plains of Europe to the wider interpretation resulting from chronologies and references to open-air decoration in the west of the Iberian Peninsula, Africa and northern Europe, the present situation requires a reflection on the demography of the Upper Palaeolithic in Europe and its prolongation both in earlier times and in more recent moments. The total visibility, together with proximity to the habitat, are some of the most outstanding facets of the new lines of interpretation that Balbín’s research has opened up. Equally, a better knowledge of pictorial applications and the assessment of radiocarbon chronologies have contributed to the design of interpretations in better accord with a full appraisal of the activities of those groups. Prehistoric art is a basic factor in territorial studies and the reconstruction of exchange networks among hunter-gatherer groups, as it deals with unique documents fixed in the landscape on solid surfaces, which make explicit their widespread presence over large geographical areas, extending beyond the boundaries traditionally accepted. The constant use of the same territories and even the same surfaces, on which techniques in use over very long periods of time depicted symbols on the stones, adds elements to be studied in an analysis of tradition and the learning of the hunter groups’ references among their immediate heirs. No human culture can be understood without the symbols that identify it and characterise it. Without the documents that prehistoric art represents for the overall comprehension of the groups of the past and, above all, without a perception in which this facet of culture occupies the place it deserves, our knowledge of the Upper Palaeolithic would be poorer and our capacity for interpretation would be noticeably smaller. Professor Balbín always encouraged this type of study in the course of his research and his teaching. His colleagues have expressed their appreciation by generously contributing to this tribute on the occasion of his retirement.

ix

x

‘Science’ versus Archaeology: Palaeolithic Rock Art at the beginning of the 21st century José-Javier Alcolea-González1 and César González-Sainz2 1

Área de Prehistoria, Departamento de Historia y Filosofía. Universidad de Alcalá. [email protected] 2 Departamento de Ciencias Históricas, Universidad de Cantabria. [email protected]

Abstract Our main objectives will be (1) explaining –necessarily very briefly– the state-of-the-art concerning some fundamental aspects of Palaeolithic rock art research, such as the chronological debate, the notion of ‘style’, or the relations between graphic expressions and Upper Palaeolithic archaeological contexts, and (2) discussing our positions –not always coincident with each other– with regard to these topics. This panoramic view will be first based on a historiographical reading of Palaeolithic rock art research, aimed at understanding the deep changes that have occurred in this field during the last three decades. Keywords Palaeolitihic art; C14; Style

main one – of those young Spanish archaeologists who started to consider Palaeolithic graphic expressions as archaeological objects, in consonance with LeroiGourhan’s proposals. His work in Tito Bustillo (BalbínBehrmann 1989) or La Pasiega (Balbín-Behrmann & González-Sainz 1993, 1995) included the systematic treatment of depictions by means of complex recording protocols based on descriptive sheets, tracings, precise geo-location and photographs of all kinds of graphic evidence found on the cave walls. This methodological renovation was an important breakthrough for the field of the Spanish Palaeolithic, since it contributed to changing the consideration of graphic assemblages as mere products of a collateral phenomenon, away from the archaeological reality of the Upper Palaeolithic.

Introduction In a recent paper, Rodrigo de Balbín-Behrmann (2014) defended the use of archaeological reasoning in opposition to the interpretative abuses derived from the uncritical application of ‘hard-science’ analytical methods to the study of the Palaeolithic rock art phenomenon. This critique will be used as the common thread of this work, which we present as a humble contribution in tribute to a person who has been our professor, colleague and friend during the last three decades. We thus link our reflections with the scientific work of the honoree, whose proposals are, to different extents, in consonance with ours. The 1980s and 1990s. The crisis of structuralist orthodoxy and the opening of new avenues of research for the study of Palaeolithic rock art

The crisis of the structuralist paradigm broke out in the late 1980s. Paradoxically, this was not an epistemic crisis, but it arose as a consequence of the deepening of the methodological system established by LeroiGourhan, who had been the main supporter of applying archaeological methods to the study of rock art. Thus, the development of new analytical methods applied to the delicate rock art motifs, especially radiocarbon dating (Valladas et al. 1992) and pigment analyses (Clottes et al. 1990), together with some spectacular findings, such as the graphic expressions of Cosquer (Clottes & Courtin 1993) and Chauvet (Clottes 2001) caves, showed results that came into conflict with the chronological system of Leroi-Gourhan. As a consequence of these results, there was a virulent reaction against the system in question, especially within the French discipline (Lorblanchet 1993). In fact, some clarifications to the chronology proposed by Leroi-Gourhan came out earlier, but they questioned only partial aspects of the system. From the early 1990s, critiques became stronger, being especially directed at potentially pre-Magdalenian graphic

At the beginning of the 1980s, research on Upper Palaeolithic rock art was entirely dominated by the structuralist paradigm created by A. Leroi-Gourhan in the 1960s (Leroi-Gourhan 1965). This paradigm reached its culmination in 1984, when the monumental work L’Art des Cavernes was published (Ministèrè de la Culture ed. 1984). This publication presented, as a kind of ‘vademecum’, the state-of-the-art of French Palaeolithic rock art, with its methodological framework completely based on the systematics proposed by LeroiGourhan. The structuralist paradigm was progressively adopted in Spain from the late 1970s onward. Then, a new generation of prehistorians started to adopt the new methods and objectives coming from France, and thus refused to continue using the unsystematic methodology previously established in Spain since the first works of H. Breuil. Rodrigo de Balbín was one – if not the 1

Prehistoric Art as Prehistoric Culture expressions, for which well-dated portable artworks were scarce compared to earlier times.

decorated sites compiled by González-Echegaray (1978), today we recognise a total of 135 (although in 20 of them a post-Palaeolithic chronology cannot be ruled out). Although the proportion of what we currently know with respect to the potential Palaeolithic reality is undoubtedly negligible, and the preservation problems continue to be the same as they were years ago, this increase in the network of sites has led scholars to (1) consider the necessity of abandoning excessively unilinear frameworks for the temporal development of Palaeolithic graphic expressions, and (2) consider the existence of greater graphic variability at the synchronic level than was previously thought.

In previous papers (Alcolea-González & BalbínBehrmann 2007: 442; González-Sainz 1999: 141), we have discussed the fact that the impact of direct chronometric dating, despite its undeniable relevance, causing the abandonment of the chronological framework established by Leroi-Gourhan by most rock art scholars, was probably exaggerated. Yet it is also relevant that the verification of methodological problems in the structuralist chronological framework had an impact on the progressive establishment of contextual analysis as a relevant methodology for studying cave art depictions (Lorblanchet 1995; Moure & González-Morales 1988; Pastoors & Weniger 2011) and, more importantly, on the definitive understanding of graphic expressions as a relevant part of the Upper Palaeolithic archaeological record. Therefore, this new view came as a – probably unwanted – support for one of the methodological bases of Leroi-Gourhan’s systematics: rock art expressions were also archaeological.

The Iberian open-air decorated sites have had an essential impact on the opening-up of new ways of approaching the meaning of Pleistocene graphic expressions. This meaning was traditionally limited to religious behaviour, an interpretation that was recently boosted by the reintroduction of shamanism as a universal mechanism for explaining Palaeolithic graphic activity (Clottes & Lewis-Williams 1996). The work of Rodrigo de Balbín is again of the maximum relevance for this renovation, due to some publications that set the basis for overcoming the exclusively religious interpretations of Palaeolithic graphic expressions (Balbín-Behrmann & AlcoleaGonzález 1999). In this sense, his works propose an open reading of the depictions as communication codes bearing messages of different kinds – that is, a communication system used to codify and transmit information, shared beliefs or even feelings (Bueno-Ramírez et al. 2003), which was not limited to the darkness of the caves, but was also present in the visible – and not very mysterious – rocks of the river shores (Alcolea-González & BalbínBehrmann 2006).

The crisis in the ‘orthodoxy’ led to unprecedented research activity on Palaeolithic graphic expressions during the 1990s and the 2000s. Once the temporal framework of Leroi-Gourhan had been strongly called into question, this activity was mainly centred on the chronological review of the most important graphic assemblages of Southwest Europe, and especially those of the Cantabrian Cornice (Northern Spain). In this area, an important database of direct radiocarbon datings on pigments has been gathered (Valladas et al. 1992; Moure et al. 1996; Fortea 2002; Balbín-Behrmann et al. 2003), although some other chronometric methods have been also applied. The limits of the radiocarbon method, especially when it comes to dating engravings or paintings made with inorganic pigments, but also in the case of dating potentially old motifs, have led scholars to attempt to use methods such as thermoluminiscence (González-Sainz & San Miguel 2001) and, more recently, uranium series (Pike et al. 2012) –- after some specific and preliminary attempts in Covalanas and La Garma caves -– to propose a direct age for graphics in caves.

Many of the above-mentioned reflections, and also those that we are now offering in this paper, began to take shape, always together with Rodrigo de Balbín, in the study of La Pasiega cave, a major and very complex site of the Cantabrian region, in which we worked during several summer campaigns in the 1980s (BalbínBehrmann & González-Sainz 1993, 1995). In La Pasiega we started to observe some contradictions between the data we were collecting and the structuralist orthodoxy. Among these contradictions, the most relevant were the existence of paintings in totally external areas of the two mouths of the cave, a totally unexpected order of figures in the interior spaces, or the recurrent appearance of the Bison/Horse theme throughout the cavity. Concerning the meaning of graphic expressions, a minimum analysis of the interior spaces where Upper Palaeolithic humans chose to produce their depictions prevented us from proposing univocal interpretations. Within the complex topography of La Pasiega we found everything from isolated and peculiar representations to clusters of figures in clear compositions, sometimes showing recurrent iconographic structures that seemed

At the same time, this intensive activity has resulted in an important increase in the Palaeolithic rock art inventories of Southwest Europe, as shown in several recent compilations dealing with the Cantabrian area (ACDPS 2002, 2010; Ríos et al. 2007). More importantly, it has also brought to light some outstanding finds, among which the open-air assemblages of Western Iberia are the most relevant (Alcolea-González & Balbín-Behrmann 2006; Baptista 2008). Thus, the network of rock art sites in Southwest Europe has increased exponentially with respect to the previous 20th-century syntheses by Breuil (1974) and Leroi-Gourhan (1965). For instance, in the Cantabrian region, in comparison with the 55 2

Alcolea-González and González-Sainz: ‘Science’ versus Archaeology to repeat a well-known message directed at the human community. Furthermore, differences in the visibility and formal complexity of figures and compositions, or the great variability in the accessibility and dimensions of decorated spaces, also suggested the need for considering purposes and meanings different from those assumed until then.

based on digital photography has caused a relative decline in the quantity and quality of data recorded by researchers. On the other hand, and quite surprisingly in our view, some of the most prickly problems in the study of Palaeolithic art have been produced due to the blind adoption of results yielded by the new archaeometric analyses used in the recording and study of rock depictions. In the next section we will briefly discuss some of these problems.

Leroi-Gourhan’s interest in syntax, and its expression in underground spaces, led him to focus on animal themes and their topographic location within the caves. Yet he always encountered major difficulties when trying to establish clear and particular topographic categories, given the specific nature of the karstic environments where motifs were depicted – very different, for instance, from those used in historical artistic cycles, in which a strict topographic categorization is easy to define. This emphasis on the topographic location of figures meant that some other aspects that we now tend to consider to be part of the transmitted message were neglected. Thus, for instance, in assemblages that are undoubtedly contemporary with La Pasiega, we noticed important typometric differences depending on the animal themes or the technical procedures, as well as a large variability in formal complexity. These aspects, which we would later confirm in the large open-air site of Siega Verde (Alcolea-González & Balbín-Behrmann 2006: 265-92), and also in some other Cantabrian caves, strongly suggested that Palaeolithic graphic expressions had a large variety of meanings, although they also strengthened some of Leroi-Gourhan’s proposals concerning the spatial hierarchization of animal motifs.

The stylistic debate and its chrono-cultural implications. Back to Culture History? The main problem in current research on European Palaeolithic rock art is of a chronological nature. Furthermore, many other problems under discussion, to some extent, depend on the chronological debate. As a consequence of the general dismissal of the stylistic dating system established by Leroi-Gourhan in Préhistoire de l’Art Occidental (1965), there are now several debates about the general chronology of European Palaeolithic rock art. These discussions can be grouped around three main topics: (1) the chronometric origin of graphic expressions and their initial stylistic nature, (2) the role of stylistic analysis for organizing chronological frameworks, and (3) the possible existence of general evolutionary trends driving the development of graphic expressions. Concerning the first point, it seems that nowadays there is a partial consensus. Once the geographically limited view, in some way French-biased, derived from LeroiGourhan’s Style I, has been abandoned; there is now wide recognition of the multi-regional existence of graphic expressions on rocks from the beginning of the Upper Palaeolithic (Broglio et al. 2009; Fortea et al. 2004; González-Sainz et al. 2013; Sauvet et al. 2007; Pike et al. 2012), and some geo-cultural systematizations have even been attempted (Delluc & Delluc, 1991, 1999; Sauvet et al. 2008). It is also evident that these representations were produced in a variety of locations, including the deepest areas of the caves, and that they respond to a greater technical and stylistic variability than was assumed by Leroi-Gourhan (González-Sainz 2002). However, this generic consensus on the oldest Palaeolithic graphic expressions still features a number of controversies, especially with regard to their stylistic nature.

The final balance of three decades of changes (Fig. 1), based on the review of many previously known graphic assemblages, but also on the study of some recently discovered sites, leads us to a current state-of-the-art in which our knowledge is far greater than in the early 1980s. Furthermore, the new recording and research techniques implemented basically since the 1990s, have also increased the quality of available data. This technological improvement in our techniques has been progressively modifying our working procedures in each research phase (lighting, topography, photography and databases). As a consequence, direct chronometric dating, pigment analysis based on non-invasive techniques such as Raman microscopy (Ospitali et al. 2006; Hernanz et al. 2012), 3D scanning techniques (González-Aguilera et al. 2009; Pastoors & Cantalejo 2014), and advanced methods of digital photography, are now usual in the study of Palaeolithic graphic sites.

Some researchers, in consonance with the first prehistorians who refused Leroi-Gourhan’s legacy en bloc (Lorblanchet 1993, 1995), currently maintain that the Palaeolithic graphic cycle arrived or originated in Europe in an advanced stage of development concerning techniques and style. The main basis for this statement is found in the chronometric results attributed to the black series of Grotte Chauvet (Clottes [ed.] 2001; Clottes & Geneste 2007; Geneste [ed.] 2005), whose complexity

However, although this process of increasing reliance on technology has greatly enlarged our knowledge of Palaeolithic graphic expressions, it can also be related to the appearance of new problems that still remain unsolved. On the one hand, concerning methods, the frequent bad use of graphic design software tools 3

Prehistoric Art as Prehistoric Culture

Graphic activity in the European Upper Palaeolithic Changes in research approaches since 1985

Single artistic cycle throughout time (38–11,5 ka. BP) and space (Europe) It is maintained with minor changes: - Existence of Palaeolithic figurative art outside Europe. - Relevance of the open-air artistic phenomenon. - Increase in the stylistic and iconographic variability.

1985 Model Centre / Periphery

Geographic development. Change of model: Current model Mosaic

- Core regions (Dordogne-Quercy, Pyrenees and Cantabrian Cornice) vs. peripheral areas. - Spread of figurative behaviour from the core regions to the rest of Europe. - Maintenance of the FrancoCantabrian cluster as spreading area throughout the Upper Palaeolithic.

- Aurignacian-age graphic expressions present in different European regions. - Phases of artistic emergence in different regions and periods throughout the Upper Palaeolithic. - Variable degree of inter-regional cultural interactions throughout the Upper Palaeolithic.

Temporal development Change of model: 1985 Model

Current Model

- Evolutionary model based on the idea of progress. From simple to complex. - Linear process oriented to mastery (styles I, II, III, aIV & rIV). - Style is conceived as closely related to chronology.

- General graphic evolution or different trends of temporal development. - “Ancient” vs. “Magdalenian” art: from synthetic and minimalist naturalism to visual naturalism. - Trends of graphic temporal evolution nuanced, to different extents, by intra and inter-regional variability.

Meaning Change of model: 1985 Model

Current Model

- Search for unique and general explanations. - Graphic expressions inextricably associated with the transcendental or religious sphere. - Message is expressed by iconographic structure (A. Leroi-Gourhan)

- Abandonment of unique and general explanations. More diversified meanings and functions, based on: - Impact and analysis of open-air rock art. - Internal analysis of underground contexts (different degrees of compositional diversity and complexity, variability of used spaces). - Assessment of formal treatment ('style') and context of representation as parts of the transmitted message.

Figure 1. Schematic summary of the main changes occurring in the research approaches adopted in the study of Palaeolithic graphic expressions during the last 30 years.

4

Alcolea-González and González-Sainz: ‘Science’ versus Archaeology is taken as a virtually definitive proof. This purported maturity of the graphic expressions since the beginning of the Upper Palaeolithic is also used to support the idea of the possible existence of a large variety of independent graphic traditions in Southwest Europe bearing different stylistic features. This view is undoubtedly close to anthropological particularism, and it has deep implications for the study of the Upper Palaeolithic in Southwest Europe, as we will discuss below.

after a long process based both on the direct inspection of the cave and on the fact that, in his view, the purported Aurignacian age of the black series is no longer an isolated case, but can be integrated into a network of coherent data (including the rock art of L’Aldène – Ambert et al. 2005 – and even that of the upper level of Altxerri – González-Sainz et al. 2013 – the Initial Upper Palaeolithic portable artworks of Central Europe, and some contextual information from Chauvet cave itself), has now accepted an Aurignacian chronology for the black series.

For some other researchers, including ourselves – although we disagree on some minor points – (AlcoleaGonzález & Balbín-Behrmann 2007; González-Sainz 1999), the techno-stylistic definition of the first stages of European Palaeolithic rock art is quite different. Although it is true that variability concerning themes, techniques, style, and even location of depictions, is greater than was assumed in the structuralist paradigm, it is also true that so-called Aurignacian graphic expressions – including those defined not only on radiometric grounds, but also on stylistic comparisons (Delluc y Delluc 1991, 1999; Fortea et al. 2004; Sauvet et al. 2008) – display a notable techno-stylistic homogeneity, which significantly contrasts with the black series of Grotte Chauvet. These techno-stylistic features are very different from those used by Leroi-Gourhan (1965) to define analytical figurative styles, and they are based on the repetition of some specific thematic and stylistic models. The insistence on producing depictions by means of red linear paintings or through deep engravings in the outer areas of caves, the repetitions of perspective models in discord with visual reality, and the generalized absence of complex details and conventions in zoomorphic figures, are all traits more easily related to the other graphic pole defined by Leroi-Gourhan: the synthetic figurative. In our view, the first European graphic expressions do possess a style, a set of defined formal traits which are even present in some thematic features recently proposed, such as the insistence on representing specific animals (felines, carnivores, etc).

It seems evident that our opposing positions on the controversy around the chronology of Grotte Chauvet’s black series make clear how different are our views concerning the relevance that these paintings present for discussing the origins and first development of European Palaeolithic rock art. For one of us (JJAG), Chauvet should be excluded from the discussion, whatever its actual chronology may be, since the rest of Aurignacian art never shows the formal models found in the black series, apart from some very generic thematic parallels in sites such as L’Aldene and Arcy-sur-Cure, which nonetheless present quite different stylistic features. However, for the other (CGS), considering Chauvet’s black series in this debate is not so problematic, since the great formal complexity of the composition, similar to some of the most elaborate ones of the Final Upper Palaeolithic, is produced in very different stylistic terms, and it includes some conventions which are also found in other figures of the same cave whose Aurignacian age is not under discussion. This is not the place for taking sides, again, in one way or the other. However, in our view a relevant corollary of this discussion is that neither radiocarbon nor any other physicochemical analytical method, whatever their quantity, is able, in itself, to solve the chronological problems of current Palaeolithic rock art research. Even in a case on which the two authors of this paper disagree, such as that of Grotte Chauvet, there is a basic consensus between them: problems of chronology (or any other topic) affecting Palaeolithic graphic expressions need to be approached by means of a comprehensive archaeological study including a formal, technical and compositional analysis, and cannot be reduced to a technical debate on the reliability of chronometric data.

It is evident that this debate entirely depends on the discussion of the chronology of Grotte Chauvet’s paintings, and the never-ending diatribe between the supporters of assuming an Aurignacian age for this cave’s black series (Clottes 2001; Clottes & Geneste 2007; Geneste 2005; Sauvet et al. 2008; Petrognani 2013; Von Petzinger & Howell 2014), and those questioning the validity of the results yielded by the direct chronometric dating of the figures (Alcolea-González & BalbínBehrmann 2007; Pettitt & Bahn 2003, 2014, 2015; Pettitt et al. 2009; Züchner 1996, 2014; Combier & Jouve 2012). This controversy is so pronounced that it shows to what extent the interpretation of physicochemical data may interfere in archaeological analysis, affecting even the authors of this paper. One of us (JJAG) strongly disagrees with the ‘official’ reading of the results yielded by Grotte Chauvet’s datings, while the other (CGS),

The discussion on the antiquity of Grotte Chauvet’s representations is, furthermore, a paradigmatic example of the ‘Science versus Archaeology’ conflict. This conflict is expressed through the opposition of an unquestionably poor archaeological context, both local and regional, to the purportedly indisputable evidence yielded by physicochemical analyses. The basis of this conflict actually lies in the way one interprets analytical results, in this case chronometric dates. On the one hand, those researchers who, using a ‘scientific’ approach, take 5

Prehistoric Art as Prehistoric Culture these dates as pivotal and indisputable elements of the discussion, tend to support the Aurignacian age of the black series. On the other hand, researchers defending a more comprehensive archaeological analysis tend to question the dating results obtained in Chauvet. The latter scholars highlight the contamination and error problems that could affect chronometric methods (Pettitt & Bahn 2003: 135-36), and they point out the high number of wrong and contradictory results yielded by them (Moure et al. 2006: 318). Therefore, when trying to solve the chronological problems affecting Palaeolithic rock art, they propose considering chronometric results not as definitive data in themselves, but as a substantial part of a comprehensive archaeological approach to the record.

in some artistic areas, such as the Cantabria-PyreneesPérigord cluster, the techno-stylistic modifications throughout the Upper Palaeolithic were more clearly patterned, in others, such as Central or Western Iberia, formal survivals were more pronounced. Problems with the validity and treatment of the notion of style arise when the above-mentioned evolutionary patterns are denied. Researchers doing so usually also support the old chronology of the black series of Grotte Chauvet (Lorblanchet 1993, 1995). For them, there is only a myriad of different styles responding to the multiple local traditions originating throughout the Upper Palaeolithic. These scholars conceive style in descriptive or archaeographic terms, claiming that it is only useful for describing the graphic activity of a given epoch and place, and that such style is due to the historical and unique nature of that specific tradition. Therefore, according to this view, style would be a totally useless element for dating parietal graphic expressions.

The debate around the stylistic nature of the first Palaeolithic rock art expressions in Europe is directly related to the other topics currently under discussion, as mentioned above: the usefulness of the notion of style for dating, and the existence of an evolutionary process driving the graphic expressions throughout the Upper Palaeolithic. The first of these aspects needs a preliminary clarification: our use of the term ‘style’ is different from the notion of style coined by LeroiGourhan. The latter was not limited to the formal traits of representations, but also included technical, compositional and topographical aspects. The validity of this notion for chronological analysis has already been questioned (Alcolea-González & Balbín-Behrmann 2007: 460). In contrast, we limit the notion of style to the formal appearance of representations, and in this sense we consider it a useful tool for the archaeological analysis of rock art depictions. And this is so because for us, and also for Professor Balbín, it is possible to document evolutionary trends within European rock art throughout the Upper Palaeolithic.

While it is true that the evolutionary nature of LeroiGourhan’s stylistic systematics virtually dismissed the inevitable historical nature of Palaeolithic graphic expressions, it is also true that reclaiming the old and dogmatic methods of Historical particularism to take them into account is another abuse which, furthermore, is in discordance with the archaeological record. We have documented the existence of regional sequences in the Iberian plateau (Alcolea-González & Balbín Behrmann 2003, 2006) or the Cantabrian region (González-Sainz 2002; González-Sainz & San Miguel 2001), which are supported in archaeological data, including both stylistic and radiometric evidence. These sequences admittedly question the stylistic systematics of Leroi-Gourhan, but they also point to the existence of general trends of evolution within the European Palaeolithic graphic expressions, whose acceleration and deceleration mechanisms are, nonetheless, still poorly known. These trends, most probably geographically heterogeneous, show a very generic process in which we detect temporal changes concerning the treatment of some stylistic features, such as perspective, movement, and an increasing naturalism of representations. The existence of this evolutionary process clearly shows that formal traits can be used as diagnostic elements in the process of dating Palaeolithic graphic expressions, even though their accuracy is not as great as one would wish.

However, despite the fact that both of us accept the existence of a graphic evolutionary process, we need to point out that there is some minor discussion between us concerning certain aspects. For instance, we disagree slightly on the convenience of proposing either a general evolutionary process driving the development of Upper Palaeolithic graphic expressions, or just a certain tendency for change from a prevailing synthetic or minimalist naturalism in the first phases to a more visual naturalism fully dominant during the Tardiglacial. Another topic under discussion is the degree of synchronic variability present in the first phases of the Palaeolithic graphic cycle, although we agree that this variability was not an exclusive phenomenon of the advanced phases, as was stated in the classic interpretations. In any case, we both maintain that these tendencies for change or general evolutionary trends were not developed in a gradual and linear way, as was derived from the idea of progress assumed in the classic evolutionist proposals. Also, we maintain that they were not equally developed in the different regions of Southwest Europe, since while

This discussion on the chronological value of stylistic features also presents some implications for the cultural interpretation of rock art expressions. Supporters of the denial of global evolutionary trends within European Pleistocene graphic expressions propose that these expressions must theoretically correspond to an unknowable myriad of local and regional traditions (Lorblanchet 1995: 279). Paradoxically, these scholars have found a simpler way of systematizing 6

Alcolea-González and González-Sainz: ‘Science’ versus Archaeology these traditions: re-assign them, once they have been ‘objectively dated’, to the technocomplexes of the Upper Palaeolithic.

open reading of its possible meanings. The introduction of modern physicochemical analyses to the study of Palaeolithic graphic expressions has caused both an important breakthrough in our understanding of the phenomenon, and an opening up of new debates which will enlarge our knowledge in the coming years.

Again, this ‘new’ tendency must be related to the virulent response to Leroi-Gourhan’s work that occurred in France after his death. The distinction between technological and graphical developments established by Leroi-Gourhan was one of the hardest blows to the prior culture-historical understanding of Upper Palaeolithic cultures, which was then also under attack by proponents of Processual Archaeology. Thus, dismantling the work of the French scholar could not be complete without solving this issue, and this was finally addressed with the help of ‘Science’. In consonance with the archaeographic sense of ‘style’ explained above, and based on the radiometric dating of some graphic expressions, the Upper Palaeolithic graphics have been re-assigned to the Aurignacian, Gravettian, Solutrean or Magdalenian, endowing these concepts with a new cultural substance. In this sense, the current use of frameworks similar to the ancient diagrams of comparative chronology in use at the beginning of the 20th century (Fortea et al. 2004: 173, Table 1), shows this return to Culture History: Style I is now Aurignacian Art, Style III is Badegoulian Art.

However, the main leap forward has been to include the study of Pleistocene graphic expressions in the framework of the archaeological analysis of the Upper Palaeolithic, thus abandoning the ‘metaphysical’ and ‘immaterial’ realm to which they had been relegated throughout the 20th century. Yet, as we pointed out at the beginning of this paper, we have to be wary of the abuses that can be committed in the name of data coming from the hard science, and reclaim a real archaeological treatment of Palaeolithic parietal expressions, in which the images composing those expressions cannot be excluded. That treatment was always taught to us –- and we sincerely hope he will continue to do so -– by Professor Balbín, without a doubt one of the main figures of these three exciting decades of research which have forever changed our knowledge and understanding of European Palaeolithic parietal art.

In our view, this epistemic return to the past is one of the biggest paradoxes to have occurred in the last few decades in the field of Palaeolithic rock art research. In the face of modern and sophisticated data coming from the hard sciences, any reasoning using the notion of ‘style’ has been proscribed, while Palaeolithic graphic expressions have been framed in consonance with old constructs, which are in fact ‘archaeological cultures’ built upon deeper stylistic criteria. Thus, while the validity of concepts internal to the graphic activity, such as ‘stepped mane’, are denied, some others completely external to it, such as ‘split-base antler point’ are – indirectly – used with the same purpose. For us, this might be the most deleterious consequence of the abuses committed in the application of hard-science analyses to the study of Upper Palaeolithic graphic expressions. In fact, results from these analyses are still incipient, since they have been applied to a very reduced number of figures in relation to the large number of depictions currently known. Therefore, the paradigmatic shift proposed by the post-stylistic researchers, which implies rethinking the very nature of Upper Palaeolithic technocomplexes, is not justified.

Translated from Spanish by Manuel Alcaraz-Castaño

References Alcolea-González, J. J. & Balbin-Behrmann, R. de 2003. El Arte Rupestre Paleolítico del interior peninsular. Elementos para el estudio de su variabilidad regional, pp. 223-53 in (R. de Balbín & P. Bueno, eds) El Arte Prehistórico desde los Inicios del Siglo XXI. Primer Symposium Int. de Arte Prehistórico de Ribadesella. Asosiación Cultural Amigos de Ribadesella. Alcolea-González, J. J. & Balbin-Behrmann, R. de 2006. Arte Paleolítico al aire libre. El yacimiento rupestre de Siega Verde, Salamanca. Arqueología en Castilla y León, 16. Valladolid. Alcolea-González, J. J. & Balbin-Behrmann, R. de 2007. C14 et Style. La chronologie de l’art pariétal à l’heure actuelle. L’Anthropologie 111: 435-66. Ambert, P., Guendon, J. L., Galant, P., Quinif, Y., Gruneisen, A., Colomer, A., Dainat, D., Beaumes, B. & Requirand, C. 2005. Attribution des gravures paléolithiques de la grotte d’Aldène (Cesseras, Hérault) à l’Aurignacien par la datation des remplissages géologiques. Comptes Rendus Palevol. 4-3: 275-84. Balbín-Behrmann, R. de 1989. L’Art de la grotte de Tito Bustillo (Ribadesella, Espagne). Une vision de synthèse. L’Anthropologie 93: 435-62. Balbín-Behrmann, R. de 2014. Chauvet, chronologie et archéologie. L’Anthropologie 118: 159-62.

Conclusion Three decades of research on European Palaeolithic art have resulted in a profound revision of our certainties about this subject. Today we possess a richer image of this phenomenon, based on its chronological expansion to the very origins of the Upper Palaeolithic, a larger geographic and spatial variability, and a new and more 7

Prehistoric Art as Prehistoric Culture Balbín-Behrmann, R. de & Alcolea-González, J. J. 1999. Vie quotidienne et vie religieuse. Les sanctuaires dans l’Art Paléolithique. L’Anthropologie 103: 23-49 Balbin, R. de, Alcolea, J. J. & Gonzalez, M. A. 2003. El Macizo de Ardines, Ribadesella, España. Un lugar mayor del Arte Paleolítico Europeo, pp. 91-151 in (R. de Balbín & P. Bueno, eds) El Arte Prehistórico desde los Inicios del Siglo XXI. Primer Symposium Int. de Arte Prehistórico de Ribadesella. Asosiación Cultural Amigos de Ribadesella. Balbin-Behrmann, R. de & Gonzalez-Sainz, C. 1993. Nuevas investigaciones en la cueva de La Pasiega (Puente Viesgo, Cantabria). Boletín del Seminario de Estudios de Arte y Arqueología LIX: 9-38. Balbin-Behrmann, R. de & Gonzalez-Sainz, C. 1995. L’ensemble rupestre paléolithique de ‘La Rotonda’ dans la galerie B de la grotte de La Pasiega (Puente Viesgo, Cantabria). L’Anthropologie 99 (2-3): 296324. Baptista A. M. 2008. O paradigma perdido. O vale do Côa e a Arte Paleolítica de ar livre em Portugal. Edições Afrontamento e Parque Arqueológico do Vale do Côa: Vila Nova de Foz Côa/Porto. Breuil, H. 1974. Quatre cents siècles d’Art Pariétal. Editions Max Fourny: Paris. Broglio, A., De Stefani, M., Gurioli, F., Pallecchi, P., Giachi, G., Higham, T. & Brock, F. 2009. L’art aurignacien dans la décoration de la Grotte de Fumane. L’Anthropologie 113: 753–61. Bueno, P., Balbin-Behrmann, R. de & Alcolea-Gonzalez, J. J. 2003. Prehistoria del lenguaje en las sociedades cazadoras y productoras del sur de Europa, pp. 9-22 in (R. de Balbín & P. Bueno, eds) El Arte Prehistórico desde los Inicios del Siglo XXI. Primer Symposium Int. de Arte Prehistórico de Ribadesella. Asosiación Cultural Amigos de Ribadesella. Clottes, J. (ed.) 2001. La Grotte Chauvet. L’art des origines. Le Seuil: Paris. Clottes, J. & Courtin, J. 1993. La Grotte Cosquer, peintures et gravures de la caverne engloutie. Le Seuil: Paris Clottes, J. & Geneste, J. M. 2007. Le contexte archéologique et la chronologie de la grotte Chauvet, pp. 363-78 in (H. Floss & N. Rouquerol, eds) Les Chemins de l’Art Aurignacien en Europe. Das Aurignacien und die Anfänge der Kunst in Europa. Colloque International–International Fachtagung Aurignac 2005. Editions Musée-forum Aurignac, Cahier 4. Clottes, J. & Lewis-Williams, D. 1996. Les Chamanes de la Préhistoire. Transe et magie dans les grottes ornées. Le Seuil: Paris. Clottes, J., Menu, M. & Walter, P. 1990. La preparation des peintures magdaléniennes des cavernes ariégeoises. Bulletin de la Societé Préhistorique Française 87 (6): 170-92.

Clottes, J., Valladas, H, Cachier, H. & Arnold, M. 1992. Des dates pour Niaux et Gargas. Bulletin de la Société Préhistorique Française 89: 270-74. Combier, J. & Jouve, G. 2012. Chauvet Cave’s art is not Aurignacian: a new examination of the archaeological evidence and dating procedures. Quartär 59: 131–52. Delluc, B. & Delluc, G. 1991. L’art pariétal archaïque en Aquitaine. XVIIIe supplément à Gallia Préhistoire. CNRS: Paris. Delluc, B. & Delluc, G. 1999. El arte paleolítico arcaico en Aquitania. De los orígenes a Lascaux. Edades. Revista de historia 6: 145-65. Fortea, F. J. 2002. Trente-neuf dates C14-SMA pour l’art pariétal paléolithique des Asturies. Bulletin de la Société Préhistorique Ariège-Pyrénées LVII: 7-28. Fortea, F. J., Fritz, C., García, M., Sanchidrián, J. L., Sauvet, G. & Tosello, G. 2004. L’art pariétal paléolithique à l’épreuve du style et du carbone-14, pp. 163-75 in (M. Otte, ed.) La Spiritualité. Actes du Colloque de la Commission 8 de l’UISPP. ERAUL 106: Liège. Geneste, J. M. (ed.) 2005. Recherches pluridisciplinaires dans la grotte Chauvet. S.P.F. Travaux 6: 135-44. González-Aguilera, D. Muñoz, A. L., Lahoz, J. G., Herrero, J. S., Corchón, M. S. & García, E. 2009. Recording and modeling Paleolithic caves through laser scanning. GEOWS 7: 19-26. González-Echegaray, J. 1978. Cuevas con Arte Rupestre en la región cantábrica, pp. 49-77 in Curso de Arte Rupestre. UIMP: Zaragoza. Gonzalez Sainz, C. 1999. Sobre la organización cronológica de las manifestaciones gráficas del Paleolítico superior. Perplejidades y algunos apuntes desde la región cantábrica, pp. 123-44 in (R. Cacho & N. Gálvez) 32.000 BP: Una odisea en el tiempo. Reflexiones sobre la definición cronológica del arte parietal paleolítico. Gonzalez Sainz, C. 2002. Unidad y variedad de la región cantábrica y de sus manifestaciones artísticas paleolíticas, pp. 39-45 in (Various authors) Cuevas con arte paleolítico en Cantabria. ACDPS. Gonzalez-Sainz, C. & San Miguel, C. 2001. Las cuevas del desfiladero. Arte rupestre paleolítico en el valle del río Carranza (Cantabria-Vizcaya). Santander González-Sainz, C., Ruiz-Redondo, A., GárateMaidagán, D. & Iriarte-Avilés, E. 2013. Not only Chauvet: Dating Aurignacian rock art in Altxerri B Cave (northern Spain). Journal of Human Evolution 65: 457-64. Hernanz, A., Gavira-Vallejo, J. M., Ruiz-López, J. F., Martin, S., Maroto-Valiente, A., Balbín-Behrmann, R. de, Menéndez, M., Alcolea-González, J. J. 2012. Spectroscopy of Palaeolithic rock paintings from the Tito Bustillo and El Buxu Caves, Asturias, Spain. Journal of Raman Spectroscopy 43: 1644-50. Leroi-Gourhan, A. 1965. Préhistoire de l’Art Occidental. Mazenod: Paris

8

Alcolea-González and González-Sainz: ‘Science’ versus Archaeology Lorblanchet, M. 1993. From Style to Dates, pp. 61-72 in (M. Lorblanchet & P. G. Bahn, eds) Rock Art Studies: the Post- stylistic Era or Where do we go from here? Oxbow: Oxford. Lorblanchet, M. 1995: Les grottes ornées de la Préhistoire: nouveaux regards. Errance: Paris. Lorblanchet, M. & Bahn, P. G. (eds) 1993. Rock Art. Studies: the Post- stylistic Era or Where do we go from here? Oxbow. Oxford. Ministère de la Culture (ed.) 1984. L’Art des Cavernes. Atlas des grottes ornées paléolithiques françaises. Ministère de la Culture: Paris Moure, A. & González Morales, M. R. 1988. El contexto del arte parietal. La tecnología de los artistas en la cueva de Tito Bustillo (Asturias). Trabajos de Prehistoria 45: 19-49. Moure, J. A., González Sainz, C., Bernaldo de Quirós, F. & Cabrera, V. 1996. Dataciones absolutas de pigmentos en cuevas cantábricas: Altamira, El Castillo, Chimeneas, pp. 295-324 in (J. A. Moure, ed.) El Hombre Fósil 80 años después. Homenaje a Hugo Obermaier. Servicio de Publicaciones, Univ. de Cantabria: Santander. Ospitali, F., Smith, D. C. & Lorblanchet, M. 2006. Preliminary investigations by Raman microscopy of prehistoric pigments in the wall-painted cave at Roucadour, Quercy, France. Journal of Raman Spectroscopy 37 (10): 1063–71. Pastoors, A. & Cantalejo, P. 2014. Aplicaciones con escáner a los grabados prehistóricos, pp. 115-18 in (J. Ramos, G. C. Weniger, P. Cantalejo & M. M. Espejo, eds) Cueva de Ardales. Intervenciones arqueológicas 2011-2014. Ediciones Pinsapar. Junta de Andalucia/ Ayuntamiento de Ardales. Pastoors, A. & Weniger, G. C. 2011. Cave Art in Context: Methods for the Analysis of the Spatial Organization of Cave Sites. Journal of Archaeological Research 21 (4): 377-400. Pettitt, P. & Bahn, P. G. 2003. Current problems in dating Palaeolithic cave art: Candamo and Chauvet. Antiquity 77 (295): 134-41. Pettitt, P. & Bahn, P. G. 2014. Against Chauvetnism. A critique of recent attempts to validate an early chronology for the art of Chauvet Cave. L’Anthropologie 118 (2): 163-82. Pettitt, P. & Bahn, P. G. 2015. An alternative chronology for the art of Chauvet cave. Antiquity 89 (345): 542– 53. Petrognani, S. 2013. De Chauvet à Lascaux. L’art des cavernes, reflet de sociétés préhistoriques en mutation. Editions Errance: Paris Pettitt, P., Bahn, P. G. & Züchner, C. 2009. The Chauvet conundrum: are claims for the birthplace of art premature?, pp. 239-62 in (P. G. Bahn, ed.) An Enquiring Mind. Studies in Honor of Alexander Marshack. Oxbow Books: Oxford. Pike, A., Hoffmann, D., García-Díez, M., Pettitt, P., Alcolea-González, J. J., Balbín-Behrmann, R. de,

González-Sainz, C., Heras, C. de las, Lasheras, J. A., Montes, R. & Zilhao, J. 2012. U-series dating of Paleolithic Art in 11 caves in Spain. Science 336: 1409-13. Ríos, S., García de Castro, C., Rasilla, M. de la & Fortea, J. 2007. Arte rupestre prehistórico del Oriente de Asturias. Consorcio para el desarrollo rural del Oriente de Asturias: Oviedo Sauvet, G., Fritz, C. & Tosello. G. 2008. Emergence et expansion de l’art aurignacien. Préhistoire, Art et Societés. Bulletin de la Société préhistorique AriègePyrénées LXIII: 33-46. Valladas, H., Cachier, H., Maurice, P., Bernaldo de Quirós, F., Clottes, J., Cabrera, V., Uzquiano, P. & Arnold, M. 1992. Direct radiocarbon dates for prehistoric paintings at the Altamira, El Castillo and Niaux caves. Nature 357: 68-70. Various authors 2002. Las cuevas con Arte Paleolítico en Cantabria. Asociación Cántabra para la defensa del Patrimonio Subterráneo (ACDPS) y Cantabria en Imagen: Santander (2nd ed. 2010) Von Petzinger, G. & Howell, A. 2014. A place in time: Situating Chauvet within the long chronology of symbolic behavioral development. Journal of Human Evolution 74: 37-54. Züchner, C. 1996. La Grotte Chauvet: radiocarbone contre archéologie-the Chauvet cave: radiocarbon versus archaeology. INORA 13: 25–27. Züchner, C. 2014. Comments and additional remarks on the paper by Jean Combier and Guy Jouve: New investigations into the cultural and stylistic identity of the Chauvet Cave and its radiocarbon dating. L’Anthropologie 118 (2): 186-89.

9

10

Raman spectroscopy of prehistoric pictorial materials Antonio Hernanz Departamento de Ciencias y Técnicas Fisicoquímicas, Facultad de Ciencias, Universidad Nacional de Educación a Distancia (UNED), Paseo Senda del Rey 9, E-28040 Madrid, Spain. E-mail: [email protected]

Abstract Raman spectroscopy is an appropriate technique to identify the molecular and mineralogical composition of materials used for prehistoric rock paintings. It is non-destructive, it can be used at microscopic scale, it may be applied in situ and the Raman spectra can be considered ‘fingerprints’ of the pictorial materials. Raman microscopes with lateral resolution of up to 1-2 µm enable microstratigraphic studies and the use of micro-specimens. A review of the work carried out in this area is presented here. Rock paintings in caves and rock shelters from the Palaeolithic to the Neolithic in Europe, America, Africa, Asia and Australia have been studied by micro Raman spectroscopy supported by complementary techniques. The characterisation of the materials present in the paintings, the study of deterioration processes that they are suffering and advice about conservation measures have been possible with the information obtained. Keywords Raman spectroscopy; Raman microscopy; prehistoric paintings.

the pictographs could block the Raman photons from the pictorial materials. Removal of micro-specimens is required in these cases. The use of Raman microscopes with lateral resolution of up to 1-2 µm minimises possible damage to the artwork. Microstratigraphic studies of thin cross-sections of these specimens reveal the distribution in layers of the different components of a painted panel, which is very useful information for dating the paintings. Microstratigraphic studies could also be done by deep profiling of relatively transparent materials using confocal Raman microscopes. Raman microscopes are very efficient for in-lab studies. Darkness, no vibrations, high spectral resolution and high sensitivity detectors make in-lab work very attractive. However, the use of portable instruments in situ, especially in open-air sites implies problems. Sunlight, wind, dust, and lack of power lines are some of them. They can be solved or minimised by appropriate operating procedures. A few carefully selected micro-specimen extractions (size 1 ≤ mm2) for in-lab studies combined with meticulous collection of in situ Raman spectra from the different parts of a painted panel is recommended for a comprehensive study of rock art. Micro-specimens may be studied in the lab by other techniques like Fourier transform IR (FTIR) spectroscopy, scanning electronic microscopy combined with energy dispersive Xray spectroscopy (SEM/EDS), X-ray photoelectron spectroscopy (XPS), X-ray diffractometry (XRD), gas chromatography–mass spectrometry (GC/MS) and others (Table 1). These auxiliary techniques provide invaluable help for the identification of the compounds and minerals existing in the specimens. Nowadays, portable and handheld instruments of these techniques are available, and thus the removal of specimens can be reduced to a minimum.

Introduction Raman spectroscopy (RS) has been increasingly applied in research on cultural heritage objects during more than two decades. It is able to characterise the chemical and mineralogical composition of the materials present in these objects, as well as to study their alteration or degradation processes. Several significant advantages make this technique especially appropriate for this purpose. It is a nondestructive technique, portable instruments make in situ studies possible with no sample extractions, and in any case micro-specimens of the size of a few micrometres are enough to obtain the spectroscopic information. Finally, the collected Raman spectra are ‘fingerprints’ of the artwork materials. Thus, they may lead to their unambiguous identification. Nevertheless, the Raman effect is very weak, hence very sensitive detectors must be used; usually refrigerated charge-coupled devices (CCD). This involves some difficulties. Ambient light must be avoided and a darkroom or dark conditions are required, which is simple when studying paintings in caves. Visible and ultraviolet (UV) laser lines (e.g. 200-800 nm) are commonly used for Raman excitation. These photons cause emission of strong fluorescence radiation from some materials (biomolecules, clay… etc.) masking the weak Raman signals, one of the main problems of this technique. The use of near infrared (IR) laser lines (Fourier Transform Raman spectroscopy, FTRS) with longer wavelength (e.g. 1064 nm) to avoid this problem entails the risk of thermal alterations of the samples. A compromise decision must be adopted in these cases (e.g. 785 nm). On the other hand this photon excitation does not penetrate deeply in the sample. Therefore, the information obtained is merely superficial; crusts over 11

Prehistoric Art as Prehistoric Culture

Archaeological site

Dating

Raman technique

Rock shelters: Pecos River Culture (Val Verde, Seminole Valey, Texas, USA),

3000-4200 BP

in-lab MFTRS

Caves: Fieux, Les Merveilles and Pergouset (Quercy, France)

Aurignacian, Magdalenian, 32850±520 BP

in-lab MRS

Rock shelters: Big Bend region (Texas, USA) Rock shelters: Ibíd, Pecos River Culture (Val Verde, Seminole Valey, Texas, USA), Catamarca Cave (Argentina)

Auxiliary techniques

manganese (IV) oxide, organic binders, whewellite, gypsum, calcite, α-quartz haematite, goethite, amorphous carbon, Mn oxide/hydroxide, calcite, α-quartz

in-lab MFTRS 3000-4200 BP

Materials detected

SEM/EDX, FTIR, XRD, GC/MS

in-lab MFTRS

1100-1500 BP

Reference year Edwards 1998 Smith 1999

haematite, manganese(IV)oxide, gypsum, whewellite, calcite, α-quartz

Edwards 1999

haematite, lepidocrocite, pyrolusite, organic binders (limewash, calcite, gypsum), gypsum, calcite, whewellite, aragonite, α-quartz, witherite

Edwards 2000

Rock shelters: Nine Eritrean rock art sites (Eritrea)

3rd millennium BC - 1st millennium AD

in-lab MRS

PIXE

haematite, goethite, carbon, calcite, manganese oxides/hydroxides (bixbyite, haussmanite, groutite)

Zoppi 2002

Minoan and Mycenaean painted plaster (Greece)

Bronze Age Europe

in-lab MRS

XRD

indigo, limonite, cuprorivaite, riebeckite, goethite, haematite, aragonite, calcite, magnetite

Brysbaert 2004

Los Murciélagos Cave (Zuheros, Córdoba, Spain)

4500-5400 BC

in-lab MRS

FTIR

haematite, charcoal, calcite, nitrates

Hernanz 2006a

6th-3rd millennium BC

in-lab MRS

in situ OM, FTIR

haematite, whewellite, weddellite, gypsum, α-quartz, clays, magnetite

Hernanz 2006b

28000-24000 BP

in-lab MRS

haematite, goethite, carbon, manganese oxides, calcite, gypsum, quartz, anatase, rutile

Ospitali 2006

6th-3rd millennium BC

in-lab: MRS, FTRS

in situ OM, PPLM, SEM/EDX

haematite, amorphous iron oxyhydroxides, apatite-calcined bones (a-quartz, anatase, muscovite, illite, apatite), charcoal, α-quartz, whewellite, weddellite, carotenoids, gypsum, baryte, muscovite, calcite, microcline, anatase, rutile, lepidocrocite

Hernanz 2007 Hernanz 2008

Rock shelters: Ukhahlamba Drakensberg Park, (South Africa)

3000-100 BP

in-lab: MRS, RS

FTIR, XRD

(α-quartz, anatase, rutile), haematite, carbon, animal fat, carotenoids, gypsum, anhydrite, anatase, whewellite, weddellite, feldspar

Prinsloo 2008

Wattle-and-daub mortar: Danebury (Leicestershire,UK) La Candelaria Cave, San Fernando del Valle (Catamarca, Argentina)

ca 1500 BC

in-lab: MFTRS, FTRS

limewash putty, calcite

Edwards 2008

Rock shelter: Cueva del Tío Modesto (Cuenca, Spain) Roucadour Cave (Quercy, France)

Rock shelters: Sierra de las Cuerdas (Cuenca, Spain)

ca 500 AD

calcium hydroxide, whewellite, gypsum

Fern Cave, North Queensland (Australia)

22000-17000 BP

in-lab MRS

MFTIR imaging, ATR-FTIR, SEM/ EDX

haematite, clay, rutile, gypsum, calcite, anhydrite, carbon, quartz, calcium phosphate

Goodall 2009

Rock shelter: Hoz de Vicente (Minglanilla, Cuenca, Spain)

6th-3rd millennium BC

in-lab MRS

PPLM, SEM/EDX

haematite, carbon, whewellite, weddellite, gypsum, clay

Hernanz 2010

Rock shelters: Ukhahlamba Drakensberg Park, Eastern Cape (South Africa)

3000-100 BP

in situ MRS

haematite, gypsum, anhydrite, whewellite, anatase, α-quartz, hydromagnesite, carotenoids, nitrates, magnetite, calcite

Tournié 2011

Rock shelters; Hararghe region, (Ethiopia)

3rd-2nd millennium BC

in-lab MRS

ATR-FTIR, LIBS

haematite, goethite, calcite, gypsum, carbon, celadonite, beeswax, whewellite, α-quartz, gypsum, calcite

Caves: Rouffignac-St-Cernin (Dordogne, France)

13500–12000 BP

in situ MRS

XRF, XRD

romanechite, pyrolusite, carbon, calcite, carotenoids

Caves: Tito Bustillo, El Buxu (Asturias, Spain)

11500-14000 BP, 1700014000 BP

in-lab MRS

FTIR, SEM/EDX, XRD, XPS

La Peña Cave (San Roman de Candamo, Asturias, Spain)

25000-11000 BP

in situ MRS

in situ EDXRF

Mesolithic

in situ MRS inlab MRS

SEM/EDX

hematite, goethite, calcite, gypsum, anatase, α-quartz, whewellite, organic binder

Ravindran 2013

6th-3rd millennium BC

in-lab MRS

SEM/EDX

haematite, carbon, paracoquimbite. calcite, dolomite, gypsum, whewellite, weddellite, α-quartz

Iriarte 2013

Rock shelters: Bhimbetka (India)

Rock shelter: Remacha (Villaseca, Segovia, Spain)

12

Lofrumento 2012 Lahlil 2012

haematite, wüstite, carbon, a-quartz, hydroxyapatite, calcite, anatase, clay, calcite

Hernanz 2012

haematite, goethite, quartz, carbon, calcite, metabolites from microorganisms

Olivares 2013

Hernanz: Raman spectroscopy of prehistoric pictorial materials

Archaeological site

Dating

Raman technique

Auxiliary techniques

Five rock shelters: Eastern half of the Iberian Peninsula (Spain)

6th-3rd millennium BC

in situ MRS inlab MRS MRS imaging

in situ: EDXRF DRIFTS in lab: SEM/EDX PPLM

Los Chaparros shelter (Albalate del Arzobispo, Teruel, Spain)

6th-3rd millennium BC

in situ MRS inlab MRS

in situ EDXRF in-lab XRD

Altamira Cave (Cantabria, Spain)

35000-15000 BP

in situ MRS in-lab MRS

Branenez tumulus (Plouezoc’h, Finistère, Brittany, France)

5th-4th millenium BC

in situ MRS in-lab MRS

SEM/EDX

Materials detected

Reference year

haematite, carbon, calcite, dolomite, gypsum, whewellite, α-quartz, clay

Hernanz 2014

haematite, chalcophanite, manganese oxides/hydroxides, dolomite, anhydrite, gypsum, whewellite, calcite, weddellite

Pitarch 2014

haematite, goethite, pyrolusite, carbon, calcite

Edwards 2005 Gázquez 2014

haematite, carbon, pyrochroite, albite, αquartz, muscovite

Bueno 2012, 2015

(*) Paint components in italics, other materials (rocky support, accretions… etc.) in normal fonts. Acronyms: ATR-FTIR, attenuated total reflectance Fourier transform infrared spectroscopy; DRIFTS, diffuse reflectance infrared Fourier transform spectroscopy; EDXRF, energy dispersive X-ray fluorescence; FTIR Fourier transform infrared spectroscopy; FTRS, Fourier transform Raman spectroscopy; GC-MS, gas chromatography combined with mass spectrometry; LIBS, laser induced breakdown spectroscopy; MFTIR, micro Fourier transform infrared spectroscopy; MFTRS, micro Fourier transform Raman spectroscopy, MRS, micro Raman spectroscopy; OM, optical microscopy; PIXE, proton-induced X-ray emission spectroscopy; PPLM, petrographic polarised light microscopy; RS, Raman spectroscopy; SEM/EDX, scanning electronic microscopy combined with energy dispersive X-ray spectrometry; XPS, X-ray photoelectron spectroscopy; XRD, X-ray diffraction.

Table 1. Chronological overview of previous work on Raman spectroscopy of prehistoric rock painting materials

identification of organic components, XRD for minerals, XRF and PIXE for chemical elements, XPS for elements and their chemical bonds (it is especially useful to distinguish between organic and inorganic carbon, an essential step for radiocarbon dating), SEM/EDX combines micromorphology and elemental analysis… etc. These techniques have also been applied to the study of Palaeolithic and post-Palaeolithic paintings without contributions of Raman spectroscopy. In situ work with portable and handheld instruments of auxiliary techniques like EDXRF and DRIFTS are possible, and new advances in the portability of other techniques are expected in the near future. Raman microscopy has also been applied to study Palaeolithic industry (lithic tools and chert artefacts). The increasing interest in this technique to study Palaeolithic (Figs. 1 and 2) and postPalaeolithic pictographs is evident.

Raman spectroscopic studies on prehistoric rock painting materials Representative studies on prehistoric rock paintings by Raman spectroscopy are summarized in Table 1. Materials present in pictographs on the walls of caves, open-air rock shelters and tumulus from America, Europe, Africa, Asia and Australia are identified by this technique since 1998. Paintings since the Upper Palaeolithic to the first millennium AD (Eritrea and Argentina) have been the object of these studies. Initially, specimens extracted from the painted panels were analysed in-lab. However, since 2011, the development of appropriated portable Raman microscopes (Fig. 1) has modified the methodological procedures. As the technique is non-destructive, depictions may be studied in situ exhaustively. Information from a large number of points in the artwork may be obtained without the restrictions of sample removal.

Haematite is the most common pigment detected (Table 1) in prehistoric paintings. It has an attractive red colour, it is an abundant mineral that may be found easily, it is very stable and not photodegradable like organic pigments. The size of the haematite particles in the paint can be used as an indication of the technology used by the prehistoric artists. In addition, the dark haematite colour lightens in the powdered mineral on diminishing the granular size. At the same time, the half band width of the Raman bands of haematite is a measure of their crystallinity. Narrow Raman bands are observed in well-crystallised haematite, whereas band broadening is observed in impure, amorphous or altered haematite containing other iron oxides and oxyhydroxides. These changes in the haematite crystal structure could be natural or the result of anthropic manipulation. The almost

Generalisation of the results is therefore supported better. However, Raman spectra collected in situ normally exhibit lower signal-to-noise ratio than the spectra obtained in-lab with benchtop Raman microscopes (Figs. 2 and 3). In addition, the use of portable instruments in situ involves some problems. Hence, as mentioned previously, carefully selected micro-specimen extractions are an invaluable help for further in-lab work (e.g. microstratigraphy) and the use of auxiliary techniques. Elemental microstratigraphic analyses by laser ablation-inductively coupled plasma mass spectrometry (LA-ICPMS) is a curious application, FTIR spectroscopy is particularly efficient for complementary 13

Prehistoric Art as Prehistoric Culture constant presence of haematite in prehistoric and ancient decorations has provoked a pronounced interest in the study of ochre, earth pigments and their provenance. Other iron oxides or oxyhydroxides like goethite, wüstite and lepidocrocite (Table 1) have been detected mixed with haematite in prehistoric paintings. The addition of other materials like carbon, apatite (calcined bones)… etc. suggests a pictorial recipe to make the paint. Amorphous carbon (charcoal or soot) and manganese oxides and oxyhydroxides are the principal components found in black pictographs. The presence of amorphous carbon is detected easily by Raman spectroscopy. Nevertheless, identification of manganese oxides and oxyhydroxides by Raman spectroscopy is not exempt from difficulties. White paint is not so frequent as red and black paint in prehistoric paintings. Components like α-quartz, anatase, apatite, muscovite, gypsum and illite were found in white lines of a bicolour figure. Calcite, gypsum and α-quartz were detected in other prehistoric white pictographs. A very simple colour palette was used by prehistoric artists. It was not until the European Bronze Age that natural organic compounds like indigo added new colours to the prehistoric palette. An essential point in the analyses of pictorial materials is to distinguish the components of the paint from those forming part of the support and possible crusts. This is sometimes extremely difficult, e.g. calcite as a pigment component in a pictograph over calcareous rock with possible calcite layers deposited above the painting. In these cases Raman microstratigraphy of a cross section of the painting is of valuable assistance to determine the in-depth distribution of the different materials.

Figure 1. In situ MRS of Palaeolithic pigments in El Mirón Cave (Cantabria, Spain) (Photo: Manuel R. González Morales).

Figure 2. In situ micro Raman spectrum of a pigmented point from El Mirón Cave (Cantabria, Spain) obtained with a portable Raman microscope B&W TEK InnoRam 785H, Figure 1. Labels: h, haematite; c, calcite.

14

Hernanz: Raman spectroscopy of prehistoric pictorial materials

Figure 3. In-lab micro Raman spectrum of a micro-specimen of red pigment from Pruneda Cave (Onís, Asturias, Spain). Haematite Raman bands with high signalto-noise ratio have been obtained with a Raman microscope Jobin Yvon HORIBA LabRam-IR HR-800 UV.

To date rock paintings is a fundamental archaeological objective. Pictographs containing amorphous carbon (charcoal, soot or bone black) can be dated by the radiocarbon dating technique. Nevertheless, most prehistoric rock paintings were made with inorganic pigments (Table 1). Hence, archaeologists are very interested in the detection of possible organic binders in the paint, as they could be used for radiocarbon dating. Natural carbohydrates, lipids and proteins could have been used as binders. But as these compounds are basic nutrients of microorganisms, it is unlikely they would have remained on the rock surface since prehistoric times. Discovery of possible organic binders has been possible in some cases. Inorganic binders (clay minerals) have also been detected. On the other hand, many events could have left traces of organic substances over the rock surface since the pictorial event (insects and other invertebrates moving over the painted panels, plants and animals leaving their metabolites on the pictographs, colonies of microorganisms, fungi, lichens, soot from fires, anthropic activities… etc.). As a consequence, rock paintings are among the most difficult archaeological artefacts to date. A remarkable difference may be observed in the materials identified in rock paintings found in caves (no light) and over the walls of open-air rock shelters: the presence of different forms of hydrated calcium oxalate (whewellite and weddellite) in the latter case (Table 1). These organic compounds are frequently found on the surface of rocks exposed to light as a result of the metabolic activity of microorganisms, fungi and lichens living on the rock surface. These

accretions resist weathering during thousands of years. Consequently, dates of pictographs made with inorganic materials can be established from radiocarbon dating of layers of whewellite and weddellite related with the paint layer. Gypsum is another frequent accretion detected in painted panels in open-air rock shelters. The presence of gypsum accretions in pores and external layers of the rock supporting the paintings is related to deterioration processes observed in flaking and spallation areas of the panels. Concluding remarks and future prospects Micro Raman spectroscopy has proven to be particularly suitable to identify materials used in prehistoric rock paintings, i.e. to characterise their molecular and mineralogical composition, as well as their microstratigraphic distribution. This is a nondestructive technique, it may be applied in situ and it provides spectra that can be considered ‘fingerprints’ of the studied materials. Complementary data from other techniques like FTIR, SEM/EDX, XRD… etc. (Table 1) are appreciated to ascertain an unequivocal identification and obtain additional information. Extensive in situ work on the painted panels followed by in-lab studies on carefully selected micro-specimens is recommended to obtain the maximum amount of information with the minimum micro-damage to the artwork. Removal of micro-specimens has the advantages of the in-lab techniques, and in some cases microstratigraphic studies have revealed previous sketching of the pictographs. 15

Prehistoric Art as Prehistoric Culture manganese oxides. Physical Chemistry Chemical Physics, 1, 185–190. Bueno-Ramirez, P., de Balbín-Behrmann, R., Laporte, L., Gouezin, P., Barroso, R., Hernanz, A., GaviraVallejo, J.M., Iriarte, M. 2012. Paintings in Atlantic Megalithic Art: Barnenez. Trabajos de Prehistoria, 69, 123-132. Bueno-Ramirez, P., de Balbín-Behrmann, R., Laporte, L., Gouezin, P., Cousseau, F. ,Barroso, R., Hernanz, A., Iriarte., M., Quesnel, L. 2015. Natural and artificial colours: the megalithic monuments of Brittany. Antiquity, 89, 55-71. Capel Ferrón, C., Jorge Villar, S.E., Medianero Soto, F.J., López Navarrete, J.T., Hernández ,V. 2014. Applications of Raman and Infrared Spectroscopies to the research and conservation of subterranean cultural heritage. In The Conservation of Subterranean Cultural Heritage (C. Saiz-Jimenez ed.), CRC Press, Boca Raton, FL, USA, pp. 281–292. Clottes, J., Menu, M., Walter, P. 1990. La préparation des peintures magdaléniennes des cavernes ariégeoises. Bulletin de la Société Prehistorique Française, 87, 170-191. Colomban, P. 2012. The on-site/remote Raman analysis with mobile instruments: a review of drawbacks and success in cultural heritage studies and other associated fields. Journal of Raman Spectroscopy, 43, 1529-1535. Collado Giraldo, H., Rosina, P., García Arranz, J.J., Gomes, H., da Silva Nobre, L.F., Domínguez García, I.M., Duque Espino, D., Fernández Valdés, J.M., Blasco Laffón, E. , Torrado Cárdeno, J.M., Rodríguez Dorado, L., Rivera Rubio, E., Nacarino de los Santos, M., Capilla Nicolás, J.E., Pérez Romero, S. 2014. El arte rupestre esquemático del Arroyo Barbaón (Parque Nacional de Monfragüe, Cáceres): Contextualización arqueológica y caracterización de pigmentos Zephyrus, 2014; LXXIV, 15-39. Chalmin, E., Menu, M., Vignaud, C. 2003. Analysis of rock art painting and technology of Palaeolithic painters. Measurement Science And Technology, 14, 1590–1597. Chalmin, E., Vignaud, C., Menu, M. 2004. Palaeolithic painting matter: natural or heat-treated pigment?. Applied Physics, A 79, 187–191. Chalmin, E., Vignaud, C., Farges, F., Menu, M. 2008. Heating effect on manganese oxihydroxides used as black Palaeolithic pigment Phase Transitions, 81, 179–203. Edwards, H.G.M. 2005. Case Study: Prehistoric Art. In Raman Spectroscopy in Archaeology and Art History, Edwards , H.G.M. Chalmers, J.M. (eds.). The Royal Society of Chemistry, UK, Ch. 5, pp. 84-96. Edwards, H.G.M., Drummond, L., Russ, J. 1998. Fouriertransform Raman spectroscopic study of pigments in native American Indian rock art: Seminole Canyon. Spectrochimica Acta, 54A, 1849–1856.

The results obtained are fundamental to investigate deterioration processes and to offer appropriate advice for the conservation of the paintings. Enhanced portable Raman microscopes with broader spectral range, higher spectral resolution, more sensitive detectors, different laser lines in the same instrument and improved focusing and stabilisation systems are expected in the near future. At the same time, better portable instruments of the auxiliary techniques (Table 1) will be developed. The portability of traditional in-lab techniques (SEM/EDX, XRD… etc.) has increased in recent years and it is presumed that new instruments will be of great assistance in the study of prehistoric rock paintings. Terahertz imaging is a very promising technique for this purpose. Occult pictographs by crusts, accretions or layers of nonpictorial materials could be discovered using this technique. Neolithic wall paintings in Çatalhöyük, Turkey, have been revealed by this method. Acknowledgements I acknowledge the support of my colleagues who were associated with this spectroscopic work: Jose María Gavira Vallejo, Howell G.M. Edwards, Juan Manuel Madariaga, Santiago Martín, Egor Gavrilenko, Ángel Maroto Valiente and Ramiro Alloza Izquierdo. I would like to thank the following archaeologists and rock art specialists who collaborated with us: Juan Francisco Ruiz López, Rodrigo de Balbín Behrmann, Primitiva Bueno Ramírez, Mario Menéndez, Manuel Ramón González Morales, Martí Mas, Ramón Viñas Vallverdú, Albert Rubio i Mora, Vicente Baldellou Martínez, José J. Alcolea González, Beatríz Gavilán, Rosa Barroso Bermejo and Luz Cardito Rollán. Financial support from UNED, the European Regional Development Fund, Ministerio de Educación y Ciencia (project CTQ200508959) Ministerio de Ciencia e Innovación (project CTQ2009-12489BQU) and Universidad de Cantabria is also acknowledged. I also thank the Consejerías de Cultura of the Spanish Autonomous Communities for permission to take photographs and micro-specimens of the paintings and substrata of the archaeological sites.

References Bersani, D., Madariaga, J.M. 2012. Applications of Raman spectroscopy in art and archaeology. Journal of Raman Spectroscopy, 43, 1523–1528. Brysbaert, A., Vandenabeele, P. 2004. Bronze Age painted plaster in Mycenaean Greece:a pilot study on the testing and application of micro-Raman spectroscopy. Journal of Raman Spectroscopy, 35, 686–693. Buciuman, F., Patcas, F., Cracium. R., Zahn, D. 1999. Vibrational spectroscopy of bulk and supported 16

Hernanz: Raman spectroscopy of prehistoric pictorial materials Edwards, H.G.M., Drummond, L., Russ, J. 1999. Fourier Transform Raman Spectroscopic Study of Prehistoric Rock Paintings from the Big Bend Region, Texas. Journal of Raman Spectroscopy, 30, 421–428. Edwards H.G.M., Farwell, D.W. 2008. The conservational heritage of wall paintings and buildings: an FTRaman spectroscopic study of prehistoric, Roman, mediaeval and Renaissance lime substrates and mortars. Journal of Raman Spectroscopy, 39, 985– 992. Edwards, H.G.M., Newton, E.M., Russ, J. 2000. Raman spectroscopic analysis of pigments and substrata in prehistoric rock art. Journal of Molecular Structure, 550-551, 245–256 Froment, F., Tournié, A., Colomban, P. 2008. Raman identification of natural red to yellow pigments: ochre and iron-containing ores Journal of Raman Spectroscopy, 39, 560-568. Gay, M., Alfeld, M., Menu, M., Laval, E., Arias, P., Ontañón, R., Reiche, I. 2015. Palaeolithic paint palettes used at La Garma Cave (Cantabria, Spain) investigated by means of combined in situ and synchrotron X-ray analyticalmethods. Journal of Analytical Atomic Spectrometry, 30, 767–776. Gázquez, F., Rull, F., Calaforra, J.M., Guirado, E., Sanz, A., Medina, J., de las Heras, C., Prada, A., Lasheras, J.A. 2014. Análisis no destructivo e in situ de minerales y pigmentos en cuevas mediante espectroscopia Raman. In Cuevatur 2014 / Iberoamérica Subterránea, J.M. Calaforra and J.J. Durán (eds.), pp. 297-306. Genestar, C., Pons, C. 2005. Earth pigments in painting: characterisation and differentiation by means of FTIR spectroscopy and SEM-EDS microanalysis. Analytical and Bioanalytical Chemistry, 382, 269-274. Goodall, R.A., David, B., Kershaw, P., Fredericks, P.M. 2009. Prehistoric hand stencils at Fern Cave, North Queensland (Australia): environmental and chronological implications of Raman spectroscopy and FT-IR imaging results. Journal of Archaeological Science, 36, 2617–2624. Gómez Merino, G., Sarró, M.I., García Diez, M., Vaquero, M., Vallverdú i Poch, J. 2005. Colouring material and deterioration agents of the engrabed palaeolithic plaques from Molí del Salt site (Vimbodí, Conca de Barberà, Tarragona). Avances en Arqueometría, (VI Congreso Ibérico de Arqueometría), Universitat de Girona, Spain, pp. 251–261. González-Morales, M.R., Hernanz, A., Ruiz-López, J.F. In situ µ-Raman spectroscopy of Palaeolithic pigments in El Mirón Cave (Cantabria, Spain) paper in preparation. Guineau, B., Lorblanchet, M., Gratuze, B., Dulin, L., Roger, P., Akrich, R., Muller, F., 2001. Manganese black pigments in prehistoric paintings: The case of the black frieze of Pech Merle (France). Archaeometry, 43, 211-225.

Hernández, V., Jorge-Villar, S., Capel Ferrón, C., Medianero, F.J., Ramos, J., Weniger, G.-C., Domínguez-Bella, S., Linstaedter, J., Cantalejo, P., Espejo, M., Durán Valsero, J.J., 2012. Raman spectroscopy analysis of Palaeolithic industry from Guadalteba terrace river, Campillos (Guadalteba county, Southern of Iberian Peninsula). Journal of Raman Spectroscopy, 43, 1651–1657. Hernanz, A., Gavira-Vallejo, J.M., Ruiz-López, J.F 2006b. Introduction to Raman microscopy of prehistoric rock paintings from the Sierra de las Cuerdas, Cuenca, Spain. Journal of Raman Spectroscopy, 37, 1054–1062. Hernanz, A., Gavira-Vallejo, J.M., Ruiz-López, J.F. 2007. Calcium oxalates and prehistoric paintings. The usefulness of these biomaterials. Journal of Optoelectronics and Advanced Materials, 9, 512521. Hernanz, A., Gavira-Vallejo, J.M., Ruiz-López, J.F., Edwards H.G.M. 2008. A comprehensive microRaman spectroscopic study of prehistoric rock paintings from the Sierra de las Cuerdas, Cuenca, Spain. Journal of Raman Spectroscopy, 39, 972–984. Hernanz, A., Gavira-Vallejo, J.M., Ruiz-López, J.F., Martin, S., Maroto-Valiente, Á. , Balbín-Behrmann, R. de, Menéndez, M., Alcolea-González, J.J. 2012. Spectroscopy of Palaeolithic rock paintings from the Tito Bustillo and El Buxu Caves, Asturias, Spain. Journal of Raman Spectroscopy, 43, 1644–1650. Hernanz, A., Mas, M., Gavilán, B., Hernández, B. 2006a. Raman microscopy and IR spectroscopy of prehistoric paintings from Los Murciélagos cave (Zuheros, Córdoba, Spain). Journal of Raman Spectroscopy, 37, 492–497. Hernanz, A., Ruiz-López, J.F., Gavira-Vallejo, J.M., Martin, S., Gavrilenko, E. 2010. Raman microscopy of prehistoric rock paintings from the Hoz de Vicente, Minglanilla, Cuenca, Spain. Journal of Raman Spectroscopy, 41, 1394–1399. Hernanz, A., Ruiz-López, J.F., Madariaga, J.M. , Gavrilenko, E., Maguregui, M., Fdez-Ortiz de Vallejuelo, S., Martínez-Arkarazo, I., AllozaIzquierdo, R., Baldellou-Martínez, V. , ViñasVallverdú, R., Rubio i Mora, A., Pitarch, A., Giakoumaki, A. 2014. Spectroscopic characterisation of crusts interstratified with prehistoric paintings preserved in open-air rock art shelters. Journal of Raman Spectroscopy, 45, 1236–1243. Iriarte, E., Foyo, A., Sánchez, M.A., Tomillo, C., Setién, J. 2009. The origin and geochemical characterization of red ochres from the Tito Bustillo and Monte Castillo caves (Northern Spain). Archaeometry, 51, 231–251. Iriarte, M., Hernanz, A., Ruiz-López, J.F., Martín, S. 2013. μ-Raman spectroscopy of prehistoric paintings from the Abrigo Remacha rock shelter (Villaseca, Segovia, Spain). Journal of Raman Spectroscopy, 44, 1557–1562. 17

Prehistoric Art as Prehistoric Culture Jezequel, P., Wille, G., Beny, C., Delorme, F., JeanProst, V., Cottier, R., Breton, J., Dure, F., Despriee, J. 2011. Characterization and origin of black and red Magdalenian pigments from Grottes de la Garenne (Vallée moyenne de la Creuse-France): A mineralogical and geochemical approach of the study of prehistorical paintings. Journal of Archaeological Science, 38, 1165-1172. Julien, C.M., Massot, M., Poinsignon, C. 2004. Lattice vibrations of manganese oxides. Part I. Periodic structures. Spectrochimica Acta, 60A, 689–700. Kim, H.S., Stair, P.C. 2004. Bacterially Produced Manganese Oxide and Todorokite: UV Raman Spectroscopic Comparison. The Journal of Physical Chemistry B, 108, 17019-17026. Lahlil, S., Lebon, M., Beck, L., Rousselière, H., Vignaud, C., Reiche, I., Menu, M., Paillet, P., Plassard, F. 2012. The first in situ micro-Raman spectroscopic analysis of prehistoric cave art of Rouffignac StCernin, France. Journal of Raman Spectroscopy, 43, 1637–1643. Lofrumento, C., Ricci, M., Bachechi, L., De Feo, D., Castellucci, E.M. 2012. The first spectroscopic analysis of Ethiopian prehistoric rock painting. Journal of Raman Spectroscopy, 43, 809–816. López-Montalvo, E., Villaverde, V., Roldán, C., Murcia, S., Badal, E., 2014. An approximation to the study of black pigments in Cova Remigia (Castellón, Spain). Technical and cultural assessments of the use of carbon-based black pigments in Spanish Levantine Rock Art. Journal of Archaeological Science, 52, 535-545. Menéndez, M., Martínez-Villa, A., Hernanz, A. µ– Raman spectroscopy of prehistoric paintings from Pruneda and Molín caves (Onís, Asturias, Spain) paper in preparation. Menu, M., Walter, P. 1996. Les rythmes de l’art prehistorique. Technè, 3, 11-23.

2014. In situ characterization by Raman and X-ray fluorescence spectroscopy of post-Paleolithic blackish pictographs exposed to the open air in Los Chaparros shelter (Albalate del Arzobispo, Teruel, Spain). Analytical Methods, 6, 6641-6650. Prinsloo, L.C., Barnard, W., I. Meiklejohn, I., Hall, K. 2008. The first Raman spectroscopic study of San rock art in the Ukhahlamba Drakensberg Park, South Africa. Journal of Raman Spectroscopy, 37, 646– 654. Ravindran, T.R., Arora, A.K., Singh, M., Ota, S.B. 2013. On- and off-site Raman study of rock-shelter paintings at world-heritage site of Bhimbetka. Journal of Raman Spectroscopy, 44, 108-113. Resano, M., García-Ruiz, E., Alloza, R., Marzo, M.P., Vandenabeele, P., Vanhaecke, F. 2007. Laser AblationInductively Coupled Plasma Mass Spectrometry for the Characterization of Pigments in Prehistoric Rock Art. Analytical Chemistry, 79, 8947-8955. Roldán, C., Murcia-Mascarós, S. Ferrero, J., Villaverde, V., López, E., Domingo, I., Martínez, R., Guillem, P.M. 2010. Application of field portable EDXRFspectrometry to analysis of pigments of Levantine rock art. X-Ray Spectrometry, 39, 243– 250. Ropret, P., Madariaga, J.M. 2014. Applications of Raman spectroscopy in art and archaeology. Journal of Raman Spectroscopy, 45, 985-992. Rowe, M.R.W. 2009. Radiocarbon dating of ancient rock paintings. Analytical Chemistry, 81, 1728-1735. Rowe, M.R.W, Steelman, K.L. 2003. Comment on “some evidence of a date of first humans to arrive in Brazil”. Journal of Archaeological Science, 30, 1349–1351. Ruiz, J.F., Mas, M., Hernanz, A., Rowe, M.W., Steelman, K.L., Gavira, J.M. 2006. First radiocarbon dating of oxalate crusts over Spanish prehistoric rock art. International Newsletter on Rock Art, 46, 1-5. Ruiz-López, J.F., Rowe, M.W., Hernanz, A., GaviraVallejo, J.M., Viñas-Vallverdú, R. A. Rubio i Mora, A. 2009. Cronología del arte rupestre Postpaleolítico y datación absoluta de pátinas de oxalato cálcico. Primeras experiencias en Castilla-La Mancha (20042007). In El Arte Rupestre del ArcoMediterráneo de la Península Ibérica. 10 Años en la Lista del Patrimonio Mundial de la UNESCO, J.A. López Mira, R. Martínez Valle and C. Matamoros de Villa (eds.), Actas del IV Congreso, Generalitat Valenciana, Valencia, Spain, 303-316. Ruiz-López, J.F., Hernanz, A., Armitage, R.A., ViñasVallverdú, R., Gavira-Vallejo, J.M., Rowe, M.W., Rubio i Mora, A. 2012. Calcium oxalate AMS 14C dating and chronology of post-Palaeolithic rock paintings in the Iberian Peninsula. Two dates from Abrigo de los Oculados (Henarejos, Cuenca, Spain). Journal of Archaeological Science, 39, 2655-2667. Smith, D.C., Bouchard, M., Lorblanchet, M., 1999. An Initial Raman Microscopic Investigation of Prehistoric Rock Art in Caves of the Quercy District,

Mori, F., Ponti, R., Messina, A., Flieger, M., Havlicek, V., Sinibaldi, M. 2006. Chemical characterization and AMS radiocarbon dating of the binder of a prehistoric rock pictograph at Tadrart Acacus, southern west Libya. Journal of Cultural Heritage, 7, 344–349. Olivares, M., Castro, K., Corchón, M.S., Gárate, D., Murelaga, X., Sarmiento, A., Etxebarria, N. 2013. Non-invasive portable instrumentation to study Palaeolithic rock paintings: the case of La Peña Cave in San Roman de Candamo (Asturias, Spain). Journal of Archaeological Science, 40, 1354–1360 Ospitali, F., Smith, D.C., Lorblanchet, M. 2006. Preliminary investigations by Raman microscopy of prehistoric pigments in the wall-painted cave at Roucadour, Quercy, France. Journal of Raman Spectroscopy, 37, 1063–107.1 Pitarch, Á., Ruiz, J.F., Fdez-Ortiz de Vallejuelo,S., Hernanz, A., Maguregui, M., Madariaga, J.M., 18

Hernanz: Raman spectroscopy of prehistoric pictorial materials S. W. France. Journal of Raman Spectroscopy, 30, 347–354. Tournié, A., Prinsloo, L.C., Paris, C., Colomban, P., Smith, B. 2011. The first in situ Raman spectroscopic study of San rock art in South Africa: procedures and preliminary results. Journal of Raman Spectroscopy, 42, 399-406. Valladas, H., Cachier, H., Maurice, P., Bernaldo de Quiros, F., Clottes, J., Cabrera Valdés, V., Uzquiano, P., Arnold, M. 1992. Direct radiocarbon dates for prehistoric paintings at the Altamira, El Castillo and Niaux caves. Nature, 357, 68 – 70. Vandenabeele, P., Edwards, H.G.M., Moens, L. 2007. A Decade of Raman Spectroscopy in Art and Archaeology. Chemical Review, 107, 675-686. Walker, G.C., Bowen, J.W., Matthews, W., Roychowdhury, S., Labaune, J., Mourou, G., Menu, M., Hodder, I., Jackson, J.B. 2013. Subsurface terahertz imaging through uneven surfaces: visualizing Neolithic wall paintings in Çatalhöyük. Optics Express, 21, 8126-8134. Watchman, A., O’Connor, S., Jones, R. 2005. Dating oxalate minerals 20–45 ka. Journal of Archaeological Science, 32, 369–374. Zoppi, G.F., Signorini, G.F., Lucarelli, F., Bachechi, L. 2002. Characterisation of painting materials from Eritrea rock art site with non-destructive spectroscopic techniques. Journal of Cultural Heritage, 3, 299–308.

19

20

Prehistoric rock art and non-invasive analysis Rouffignac as a case study Patrick Paillet Muséum National d’Histoire Naturelle, Département de Préhistoire, UMR 7194. Musée de l’Homme, 17 place du Trocadéro, 75116 Paris, France

Abstract The development of non-destructive methods for analyzing pigment composition, in the laboratory or using portable systems, is a positive outcome of discussions concerning conservation, and a compromise between necessary research and heritage preservation. These methods now have great potential for a respectful study of the conservation of prehistoric art. As part of the ANR-MADAPCA programme of research and experimental development (2007-2012), we have undertaken micro-Raman analysis in the cave of Rouffignac. Keywords Palaeolithic art; non-destructive methods; Rouffignac

In recent years the study of prehistoric art has been enhanced by methods of analysis which have become both more sophisticated and less invasive. They have been developed by physicists and chemists to question the material directly, especially the pigments in parietal and portable drawings and paintings.1

production of the depictions, as well as of the acquisition and treatment of raw materials. Some of them lead to the detection of the presence of organic carbon, and thus open up the broader field of radiocarbon dates. In the laboratory, the study of pigments and prehistoric objects is done with X-spectrometry techniques such as X-ray diffraction (XRD), which provides information on the crystalline structure of the minerals that constitute the pigment. A micro-XRD, developed in C2RMF (Louvre, Paris), makes possible the characterization of pigments on entire objects or samples without any preparation. The elementary composition of the pigment, that is to say the nature and quantity of its chemical elements, is made accessible through X-ray emission techniques. In the case of ion beams, the technique used is Pixe (Particle Induced X-Ray Emission).

The development of non-destructive methods, in the laboratory or using portable systems, is a positive outcome of discussions concerning conservation, and a compromise between necessary research and heritage preservation. These methods now have a great potential for a respectful study of the conservation of prehistoric art. It is now possible to investigate raw materials, without any preparation or invasive sampling. Through their portability, which is constantly being improved, and their reliability, these devices enter the caves and shelters, allowing direct study in situ. The prehistoric drawings and paintings are now accessible safely in their chemical and mineral signature. These methods share the use of different sources of non-destructive radiation such as X-rays, infrared or visible light lasers. They have been successfully tested on several rock paintings in South Africa, in the Sahara and in Europe (Grotte Margot, Mayenne; Villars and Rouffignac, Dordogne; etc.).

Studies of decorated caves by in situ analysis Within the framework of a scientific programme of experimental development, the MADAPCA programme (2007-2012), funded by the French National Research Agency and coordinated by the author, various noninvasive analyses were tested for the first time in complex karst environments (humidity, temperature, accessibility). A campaign of measurements was undertaken in the cave of Villars (Dordogne) with a small X-ray fluorescence device. This led to the determination of the main constituents of pigments, and indicated the presence of carbon, even though it did not make it possible to carry out a quantitative analysis or to obtain information on the mineral phases. In the cave of Rouffignac (Dordogne) several instruments were used, which had already been used in museums analysis and which had been made portable, in particular a device combining fluorescence (XRF) and X-ray diffraction

In the field, X-ray fluorescence (XRF), X-ray diffraction (XRD) and of course Raman spectroscopy are the best adapted to observing and characterizing the molecular composition and external structure of pigments. They make possible a better understanding of the technical 1^ Maître de conférences. Muséum national d’Histoire naturelle, Département de Préhistoire, UMR 7194. Musée de l’Homme – 17 place du Trocadéro – 75116 Paris. Mail : [email protected]

21

Prehistoric Art as Prehistoric Culture

Figure 1. after Beck et al. 2014: 63-74 ; (in) Paillet (ed.), 2014 1. Raman microspectometry system; 2. map of cave of Rouffignac; 3. Raman spectra obtained from mammoth No. 66 on the ‘Great Ceiling’ and mammoth No.180 from the confronted mammoths; 4. Raman spectra obtained from mammoth No. 199, showing the presence of manganese oxide and carotenoids.

(XRD) – the Raman microspectrometer and infrared reflectography system.

pigments. This has been carried out on a dozen drawings in Rouffignac. Their chemical composition combines mainly manganese, iron and barium. Rarer items are also associated with the pigment, such as potassium, titanium, chromium, zinc and silicon. The mineralogical phases determined by XRD show the presence of three different manganese oxides: pyrolusite, romanechite

By combining X-ray fluorescence and diffraction in a portable system one can simultaneously gain access to the chemical composition (XRF) and to the identification of the mineralogical phases (XRD) of the

22

Paillet: Prehistoric rock art and non-invasive analysis. Rouffignac as a case study and hollandite. Analyses indicate that the compounds are variable in nature, depending on the figures and their location in the cave. Some sectors have pictorial compositions that are chemically and stylistically homogeneous, while others, such as the Grand Plafond, are characterized by a greater variety.

this assessment, although the chronology is not based on any direct dating. The Raman analysis campaign therefore focused on finding carbon elements in several of the cave’s drawings. Ten figures were analysed : 4 mammoths and 1 rhinoceros from the two famous friezes of the ‘Henri Breuil’ gallery, 3 drawings from the ‘Great Ceiling’ and 2 isolated mammoth heads. The Raman spectrum obtained on the black line crossing the head of mammoth No. 66 of the ‘Great Ceiling’ shows that it is charcoal (lack of phosphates), but was possibly added in a more recent period. A 14C date, made possible by the presence of charcoal, could confirm or invalidate this hypothesis.

The first Raman microspectrometry in a cave was undertaken on twelve black drawings in Rouffignac. Despite some problems with the laser fluorescence which limited the spectra obtained, the portable Raman proved to be perfectly suitable for the non-destructive analysis of drawings in complex environmental conditions. It is more portable than the XRF / XRD devices. The results of the Raman analyses have been compared with those obtained by XRF / XRD. They merge perfectly. The presence of black carbon in the drawings of Rouffignac was identified, as well as different manganese oxide phases. Therefore, the use of Raman in detecting samples for 14C dating shows great promise for the future. The rapid detection of carbon on the walls is also possible through infrared reflectography, which is similarly easy to use. The Raman was also used to identify carotenoid substances produced by photosynthetic bacteria, algae and fungi. These light elements contained in the organic materials are not detected by X-ray fluorescence.

The Raman spectrum obtained on the eyebrow line of mammoth No. 180 is characteristic of manganese oxide. It also showed the presence of amorphous carbon such as charcoal, in small dispersed particles along the contour line. The presence of amorphous carbon particles mixed with other pigments has already been noted on red and black Palaeolithic paintings. They could come from soot from lamps or torches used by prehistoric people for light, from organic matter contained in the original pigments, or from more recent retouching. Unfortunately these particles are too localized and too rare to allow 14C dating.

Although XRF / XRD is more effective than the Raman technique for characterizing different mineral phases, it is now recommended to use these portable and complementary fluorescence / XRD and Raman spectrometry instruments in karst environments.

Analyses of the mammoths and rhinoceros of the ‘Henri Breuil’ gallery show the presence of manganese oxides and, in localized areas and in very small quantities, of organic material of carotenoid type. This material is very widespread in nature and has been discovered in several prehistoric sites. Carotenoids can come from red pigments of biological origin, for example shells, or from the activity of photosynthetic bacteria. Again, the small quantities and the uncertain origin of the material do not make possible any 14C dating.

Micro-Raman spectroscopy in the cave of Rouffignac (Dordogne, France) Where prehistoric paintings are concerned, MicroRaman analyses have been carried out on samples in the laboratory, and more recently in situ on San red and white rock paintings (South Africa), dating to about 3000 years BP. Raman spectroscopy is also very useful for detecting the presence of some organic materials, especially carbon, in the drawings. As part of the ANRMADAPCA programme of research and experimental development (2007-2012), we have carried out microRaman analysis in the cave of Rouffignac, using a commercial spectrometer (HE532-HORIBA Scientific, Jobin Yvon Technology).

Micro-Raman spectroscopy is very effective for the detection of carbon in prehistoric drawings and paintings. It may be associated with infrared reflectography which also makes it possible, in several testing phases, to differentiate charcoal and manganese. Conclusion These portable systems, whose two main qualities are that they are absolutely non-destructive and easy to use on site, do not exclude the various studies on pigments based on iron oxides or manganese conducted in the laboratory. Diffraction techniques X-ray (XRD), infrared (IR) spectroscopy, Raman spectroscopy, transmission electron microscopy (TEM), ICP (Inductively Coupled Plasma), neutron activation analysis (NAA) and X-ray induced charged particle emission (Pixe) are the basis of studies of archaeological materials in the laboratory. We identify materials, we look for ‘recipes’ or provenances, according to protocols that remain sometimes invasive

Rouffignac, where drawings were scientifically discovered in 1956, is one of the main Palaeolithic caves of the Franco-Iberian region. It presents over 250 animal figures, mainly drawn in manganese (black) and engraved. The mammoth is the dominant animal (with over 160 individuals). The parietal art is dated to the Middle Magdalenian on the basis of style. The presence of tectiform signs, typologically identical to those from Font-de-Gaume, Les Combarelles and Bernifal, confirms 23

Prehistoric Art as Prehistoric Culture and are therefore non-applicable on rare and valuable objects. But we can also work on whole objects, without transforming them, to identify pigments and explore their origin by detecting trace elements as markers of a geological origin. The contribution of non-destructive analysis is now well established in the study of cultural heritage materials. These techniques not only preserve the integrity of the objects, but they also allow the multiplication of measurements. In situ analysis in the caves has only just begun. All these techniques contribute to the establishment of a non-invasive analytical strategy for a more comprehensive approach to prehistoric art, from object to site and from cave to laboratory.

References Lahlil, S., Lebon, M., Beck, L., Rousselière, H., Vignaud, C., Reiche, I., Menu, M., Paillet, P. & Plassard, F. 2012. The first in situ micro-Raman spectroscopic analysis of prehistoric cave art of Rouffignac StCernin, France. Journal of Raman Spectroscopy 43 (11), November: 1637-43. Paillet, P. (ed.), 2014. Les Arts de la Préhistoire: microanalyses, mises en contextes et conservation. Numéro Spécial de la revue PALEO. Actes du colloque ‘Microanalyses et datations de l’art préhistorique dans son contexte archéologique’, MADAPCA, Paris, 16-18 novembre 2011.

24

Reasoning processes in prehistoric art interpretation Sophie A. de Beaune Jean Moulin University, Lyon / ‘Archéologies et Sciences de l’Antiquité’ Research Unit, Nanterre

I have no data yet. It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. A. Conan Doyle, A Scandal in Bohemia, The Adventures of Sherlock Holmes, 1891. Abstract Since the 19th century, the authors of suggested explanations to figure out prehistoric art have tried to convince the scientific community of the legitimacy of their interpretation. To these ends, they have resorted to different types of reasoning — deduction, induction and abduction— that we analyse here on the basis of some famous examples. We will see how the intellectual approach has changed since the beginning of prehistoric archaeology, although the reasoning modes have remained the same. Résumé Les modes de raisonnement en jeu dans l’interprétation de l’art préhistorique Depuis le xixe siècle, les auteurs des explications proposées pour cerner l’art préhistorique ont essayé de convaincre la communauté scientifique du bien-fondé de leur interprétation. Pour ce faire, ils ont eu recours à différents types de raisonnement – la déduction, l’induction et l’abduction – que nous analysons ici à partir de quelques exemples fameux. Nous verrons en quoi la démarche intellectuelle a changé depuis les débuts de la préhistoire, même si les types de raisonnement sont eux restés les mêmes.

The various interpretations that have been made for more than a century in prehistory in general, and more specifically in parietal art, were influenced by the intellectual environment in which their authors were immersed. Thus, the theory of art for art ‘s sake, supported by Gabriel de Mortillet, represented the anti-religious and anticlerical wave from the end of the 19th century. It then became tempting to link together all magical-religious interpretations to what we thought we knew of the ‘primitive’ populations that were starting to be studied, and to the works of LévyBruhl concerning primitive mentality, of Durkheim concerning elementary forms of religious life ,and of Frazer who in 1887 came up with the very fashionable idea of totemism. More recently, Clottes and LewisWilliams were influenced, consciously or not, by the neo-shamanism movement that was very trendy in the USA in the 1990s.

The first consists of starting from a general assertion in order to infer a specific one. Socrates is a man, men are mortal, therefore etc. In Prehistory, an example of deduction would consist of saying: ‘if I found numerous scrapers in this site and if scrapers were used to prepare skins, then skins have been prepared in this site.’ Or ‘if I found numerous cores and flint debris in a specific area of a site and if these elements were produced during flint knapping, then it is a flint workshop area.’ The second, on the other hand, consists of assuming that a specific phenomenon illustrates a more general law. It naturally has hypothetical characteristics and must be handled with tact, its purpose being to be able to feel whether the risk taken can lead to any heuristics. ‘Induction reasoning occurs when having put forward a property of one or few individuals of a same class of beings or events, one considers that all other individuals of this class show the same property’. The accumulation of corroborative facts as well as the absence of counterexamples enable the law’s plausibility level to increase until one decides to consider it a near-certainty. Therefore, the repetition of a phenomenon is what increases the likelihood of it happening again. That is how Marcel Otte considers that prehistoric men had a religion, for there is no such thing as a people without religion. Another example would be the fact that all gallery graves that were not in acid soil have yielded human bones, making it plausible for those gallery graves to be funerary monuments, even when no human remains are found, due to a lack of good preservation conditions. Results from an induction are nevertheless tenuous, since the unexpected appearance of a counterexample is enough to make them invalid.

I will not study their assumptions themselves, which have nowadays been hit by the transience to which all our productions are doomed. I will rather take interest in the arguments of their authors. For as dependent as they were on the ideas of their time, they nevertheless went through the effort of suggesting types of reasoning worthy of being looked into. My work will therefore be applied to a specific case: a reflection on the reasoning style at work in prehistory. Reasoning in prehistory A word first about the two types of reasoning from which a prehistorian — like his colleagues in other subjects — cannot be exempt: deduction and induction. 25

Prehistoric Art as Prehistoric Culture A third type of reasoning, abduction, needs to be added. It consists of going from a clue to a hypothesis, or even from a piece of data to a theory. This type of reasoning starts with the principle that to a given effect, a corresponding cause can be found. It makes it possible to go back from the trace to the fact, as Zadig does in Voltaire’s tale: confronted by a few traces, uncertain clues, he proves himself able to describe Babylon’s sovereigns’ dog and horse, although he saw neither one nor the other. Zadig carries out deductions as well as inductions, for he must establish a general hypothesis concerning dogs’ and horses’ behaviour. But he applies them less to proposals (‘Socrates is a man’) than to clues, from which he imagines a scenario that his experience and his acute use of induction and deduction make plausible.

of prehistoric age. We know that Don Marcelino Sanz de Sautuola did not succeed in convincing others of the antiquity of the polychrome paintings on Altamira’s ceiling, and that he was even accused of forgery. The general idea was that man necessarily went through gradual evolutionary stages and that, consequently, Renaissance art was necessarily superior to that of Prehistoric men. Following the classification of societies by Lewis Morgan in Ancient Society in 1877, we could identify six stages — the three lower stages of ‘savagery’ and the three upper stages of ‘barbarism’ — through which human societies had to pass before reaching, for some of them, the seventh stage: civilization. As newly discovered ‘primitive’ societies, prehistoric societies were placed in the savagery stage, and it was thus unconceivable to connect them to an art as technically and stylistically evolved as the one we could observe at the stage of civilization. That was a deductive reasoning, although based on false premises.

Moreover, Arthur Koestler speaks of reasoning in reverse with regard to this process. It is also more the detective’s method —it was popularized by Conan Doyle with his hero, Sherlock Holmes— than that of the prehistorian. It must be said that what the prehistorian has in common with the detective —and with Zadig— is that he is in the presence of traces, often unintentional, and that he must, from those, imagine something about the men who left them behind.

When it became necessary to submit to the obvious, and admit that certain objects’ decorations were made by prehistoric men contemporary with extinct animal species, a similar argument was used to explain this art: it was an art searching no reward, recreational, for these people were still unable to develop any religious feeling. It was the theory of art for art’s sake. According to Moro Abadia and González Morales, it was difficult to deny the aesthetic character of these objects, but at the same time one could not consider their authors as real artists. As a result, they were interpreted as solely decorative manifestations, assimilating them to decorative art and opposing them to major art forms or fine arts. Here again deduction was used, although it was not formally clarified. But this hypothesis was also based on induction since parietal art was considered a result of an artistic need common to all humanity.

The most famous examples in prehistory may be those of Lewis Binford and André Leroi-Gourhan, who attempted to infer the organisation of habitation structures from analysis of the spatial distribution of remains on the ground. The episode that remained famous amongst Camp Millie’s prehistorians shows just how delicate this type of reasoning can be. In actual fact, every scientific construction (as well as every police investigation) is made up of abduction, induction and deduction. But one has to admit that the scientific approach is essentially abductive. Whether in ‘hard’ sciences or in human sciences, the researcher offers from what he observes — and what he observes is often reduced to very little — an explanatory principle that is as plausible as possible.

Once the antiquity of cave art was definitely accepted, several concurrent hypotheses were developed at the same time. Let us briefly focus on the one concerning prehistoric totemism. Expressed since the end of the 19th century, this hypothesis was promoted by Salomon Reinach, under the strong influence of the work of Baldwin Spencer and Francis Gillen, published in 1899, concerning Australian Aborigines. These authors wrote that, according to their totemic religion, Australians made cave paintings during grand ceremonies that gathered the community together regularly. Totemism, then considered the most archaic form of religion, was seen as the cult of a sacred animal that every clan worshipped and used as a rallying symbol for its members. The animals represented in caves — bison, horse, rhinoceros… — would have represented totems of different prehistoric clans. We can see here that besides very frequently resorting to analogy at that time, Salomon Reinach tried to apply abduction based on animal representation. This theory was nevertheless quickly abandoned, because

Modes of reasoning in prehistoric art Let us now see what occurs in the study of cave art. Without going through an exhaustive historical background of all the explanatory hypotheses suggested for more than a century, let us examine on what type of reasoning the most famous were based. Before the recognition of the Palaeolithic age of this art, these manifestations were believed to be attributable to shepherds from the Middle Ages, on the grounds that prehistoric men were unable to achieve such art. One could question the reasoning that led most archaeologists of these times to deny that artistic manifestations were 26

de Beaune: Reasoning processes in prehistoric art interpretation since these animals were found in all the caves, it was impossible to precisely delineate the clans. Furthermore, the fact that some animals are outlined with strokes goes against this hypothesis: a totemic animal cannot be both game and worshipped ancestor at the same time.

that only the making of the works mattered, and that it was done during magical rites, the result being after all of lesser importance. The presence of hands and sexual representations were the reflection of rituals related to fertility, as a usual complement of hunting magic, meant to facilitate the multiplication and abundance of game.

Another explanatory hypothesis, which largely replaced the others during the first half of the 20th century, was that concerning the magic of hunting and bewitchment, notably defended by Salomon Reinach and the abbé Breuil and regularly maintained with a few variables in the 1970s. It was based on the idea that the production of an animal representation on cave walls was supposed to encourage game multiplication, to ensure the success of the coming hunt, to keep danger away, or to seek the animal’s forgiveness and consent before its sacrifice. During those ‘magical’ or ‘magico-religious’ rites, animals would have been symbolically killed on the wall. We could see here an implicit deduction starting with the premise that the represented animals were exclusively those that this hunting people ate. Salomon Reinach, one of the first to support this hypothesis, was on the other hand quite dishonest, for he claimed that other represented animals, those that were not game, like hyenas or snakes, were misinterpreted. Later on, this hypothesis was reinforced by the discovery of other caves in which strokes that could be interpreted as arrows or assegais were represented. The geometrical signs were also interpreted by some, like Henri Bégouën, as representations of traps.

At least four arguments were put forward to reject the hunting magic hypothesis: the animal species represented on the walls are not those that were hunted, as proved by the presence of their bones in habitation sites; the signs that were interpreted as throwing weapons only concerned a small proportion of animals; some of these alleged weapons were drawn on the wall before the animal they were supposed to symbolically kill; and lastly, this hypothesis does not account for the existence of many abstract figures. We see that the opponents to this hypothesis also largely resorted to abduction, since they noted that, in order to justify the hypothesis, there is a lack of evidence which could enable one to go from traces to facts. Or, in other words, they pointed out the mistake of their opponents’ abduction. In the 1960s, Leroi-Gourhan analysed the relationship between the figures and their distribution in the inner space of the cave. He looked for recurring elements that could highlight the caves’ systematic structuring. This approach was qualified as ‘structuralist’ — even though Leroi-Gourhan always refused to use the term — since it focused more on the layout — the ‘structure’ — of the elements that make up a cave’s decoration, than on the elements themselves. He then practised induction, trying to pass from specifics to the general in order to identify general laws, invariants. He believed that there existed an identical structure in all the FrancoCantabrian area’s decorated caves. That was probably the weakness in his analysis, although the idea that figures are not randomly placed and that we have to try to understand the connections between them as well as with the cavities’ inner topography was nevertheless a definite and innovative contribution. If the existence of a preconceived ideal plan that could be applied to all caves has proved false, or subsequently at least unprovable, it is now widely acknowledged that the cave decorations were not done randomly, and that there really is an internal cave organization, possibly regionally speaking. We can also note that he did not use abduction: indeed he was not trying to build a plausible scenario that would help to explain the clues around him, he was simply trying to see how these clues fitted together. Induction came next, when he assumed that the arrangements he brought to light were the illustration of a general law.

This hypothesis was based on the methodological principle stated by Georges-Henri Luquet in 1931: ‘a human action whose motive is unknown by us can have been produced by the same motive that we know to have produced the same action in other circumstances, and this possibility is all the more likely from the same actions and the same motives meeting each other in numerous and various environments.’ Which means that the same causes produce the same effects, and that if we know what actions led to the representation of certain patterns elsewhere, we can deduce what patterns were uppermost in the representation of prehistoric patterns. This is a deduction, but an induction as well, since we go from particular to general. And finally it is also an abduction, since all these theories are built on the representations themselves, which are supposed to prove these magical doings. Thus the representations themselves were seen as proof of these practices: the represented animals were the animals hunted for their meat — bison and horses — but also because they were dangerous, which could explain the representations of felines or bears. The absence of animals such as reindeer or salmon — that were nevertheless on these Palaeolithic hunters’ menu — was explained by the fact that this game, very abundant and quite stupid, did not need any specific magic, being so easy to catch. Moreover, the destruction and the superimposition of numerous figures proved

At the end of the 20th century, Jean Clottes and David Lewis-Williams attempted to prove that parietal art had been produced by shamans who represented on the cave walls the visions they had access to while in trance, either during or after seeing the visions. Their first 27

Prehistoric Art as Prehistoric Culture hypothesis was based on the fact that, according to one of them, a link could be drawn with Southern African Bushman art, which was supposedly the work of shamans who came to draw their visions. By reasoning through deduction, these authors were putting together a dangerous syllogism: trance is universal, shamans go into trance, so prehistoric people were shamans. There has been a great deal of criticism of this hypothesis, and this is not the place to revisit the matter. We can bear in mind that their deduction is wrong on several points. They maintain that trance is universal and that it goes through several stages, which are also universal. However, no neurologist confirms the existence of such universal stages of trance. Moreover, it is known that Siberian shamans often only simulate trance. Conversely, trance is to be found in many practices, not necessarily ritual or religious, such as in dance. In other words, trance is neither ‘necessary nor sufficient’ to characterize shamanism.

Even among those who use abduction, approaches can be very different from one another. One consists of starting from a fully made model, and then applying archeological data to it. It is the approach followed by all the first prehistorians, from Salomon Reinach to René Nougier in the 1970s, who had in mind one or several models borrowed from ethnology — totemism, hunting magic, bewitchment… — and who tried to insert their observations into those models. They were all trying to include data in prior explanatory settings: beyond the figures themselves, there were other archaeological clues such as footprints, etc, that had to fit the mould. They freely dipped into travellers’ stories and the first ethnographic monographs in order to find suitable examples. This approach, consisting of integrating these traces into a pre-established model, can be called top-down. It is not used much nowadays, at least in the artistic field, although Clottes’ and Lewis-Williams’ approach was close to it, even though they denied it. Several prehistorians recently rose up against the practice of starting from a model, especially in the social field, concerning the classification of societies.

In order to reach their goal, they also used abduction, as they had to build up scenarios, imagine shamans going through their rituals and drawing their visions on the wall: from their point of view, incomplete animal figures reveal the shaman’s will to represent animals coming out of the wall, that is to say, going from one world to another. They also returned to the old hypothesis concerning the (very rare) figures of therianthropic half-human half-animal beings, who were themselves shamans at a certain point of the trance. Let us merely specify that, if one had to take this to its logical conclusion, one would have to call upon shamanism every time one was faced with an art using geometrical patterns as well as partly human and partly bestial representations. So there would have to be Greek shamanism, shamanism in Roman art, etc. It shows here that their abductive reasoning lacks plausibility, for it is obvious that the — in this case incomplete – animal figurations can be explained by other causes then those they call upon.

The opposite approach consists of starting with the archaeological data themselves in order to try and find their meaning. It is nowadays the most common approach. It is for instance the case of Rodrigo de Balbín Behrmann when he tries to bring to light a ‘geography’ of art in the Sella valley from detailed examination of the depicted themes. It is also Randall White’s case when he looks at the skills involved in the making of adornments that can — incidentally — tell us something about the society’s organization and the interaction between different human groups. Leroi-Gourhan’s approach, regardless of its specificity, is also similar, and can be called bottom-up. Even if the archaeological clues are always used as validation of evidence, one can see that there is obviously an evolution in the type of reasoning that is practised. Nowadays, no one practises prehistory as they did at the turn of the 20th century, by trying to stick what we thought we knew about ‘savage’ peoples onto prehistoric men, and this is not only a question of the analytical means to which researchers have access. If reasonings — induction, deduction, abduction — have not changed, simply because we do not have an infinite number of them at our disposal, the approach is now globally more cautious, more reasonable: we are no longer looking for grand explanations with universal validity, which probably reflects a more humble attitude on the part of researchers.

In conclusion Whatever their interpretation, prehistorians search walls for traces of their assertions, as a manifestation of the evidence they submit. In doing so, they all apply abduction with varying degrees of success. What Breuil searched for in figures was the proof that thrown weapons were used to symbolically kill animals on walls; Clottes looked for proof of shamanism in incomplete figures and therianthropic beings. This can seem obvious, as a prehistorian has no other option than to start from remains. He can only start with the trace in order to understand its origin. The only one not to have used abduction, as I have said, is Leroi-Gourhan: he did not go from clues to something else, he stayed at the stage of clues. The approach is specific, and it is probably why it was the most fertile as far as heuristics are concerned.

References Balbín Behrmann, R. de 2014. Los caminos más antiguos de la imagen: el Sella, pp. 65-91 in (R. de 28

de Beaune: Reasoning processes in prehistoric art interpretation Balbín Berhmann et al., eds) Expresión simbólica y territorial: los cursos fluviales y el arte paleolítico en Asturias. Real Instituto de Estudios Asturianos: Oviedo. Beaune, S. A. de 1999. [1995] Les Hommes au temps de Lascaux. 40 000 - 10 000 avant J.C.. Hachette, La Vie Quotidienne: Paris. Beaune, S. A. de 2016. Qu’est-ce que la préhistoire? Gallimard, Folio Histoire : Paris. Bégouën, H. 1939. Les bases magiques de l’art préhistorique. Scientia LXV (CCCXXIV), série IV, vol. XXXII: 206-16. Bégouën, H. 1943. De la mentalité spiritualiste des premiers hommes. Académie des Jeux floraux, Toulouse, 27 novembre 1942. Douladoure Frères : Toulouse. Bonnichsen, R. 1973. Millie’s camp: An experiment in archaeology. World Archaeology 4 (3): 277-91. Breuil, H. 1952. Quatre cent siècles d’art pariétal, les cavernes ornées de l’Âge du Renne. Centre d’Etudes et de Documentation Préhistoriques : Montiognac. Butterlin, P. 2012. Archéologie et sociologie: le cas de l’Orient ancien, pp. 184-93 in (S. A. de Beaune & Henri-Paul Francfort, eds) L’Archéologie à Découvert. CNRS Éditions: Paris. Clottes, J. & Lewis-Williams, D. 1996. Les Chamanes de la Préhistoire. Transe et magie dans les grottes ornées. Le Seuil: Paris. Denis, G. 2008. L’analogie dans les sciences du végétal : à propos des positions de F. Fontana et d’A.-P. de Candolle sur les maladies des plantes,, pp. 237-81 in (M,-J. Durand-Richard et al.,eds) L’analogie dans la démarche scientifique. Perspective historique. L’Harmattan: Paris. Groenen, M. 1994. Pour une histoire de la préhistoire. Le Paléolithique. Jérôme Millon: Grenoble. Hamayon, R. 1995. Pour en finir avec l’extase et la transe dans l’étude du chamanisme. Études mongoles et sibériennes 26: 155-90. Helvenston, P. A. & Bahn, P. G. 2002. Desperately seeking Trance Plants: Testing the ‘Three Stages of Trance’ model. RJ communications LLC: New York. Koestler, A. 1964. The Act of Creation. Hutchinson & Co.: London. Laming-Emperaire, A. 1962. La Signification de l’Art Rupestre Paléolithique. Picard: Paris. Leroi-Gourhan, A. 1965. Préhistoire de l’Art Occidental. Lucien Mazenod: Paris. Lorblanchet, M. et al. (eds) 2006. Chamanisme et Arts Préhistoriques. Vision critique. Errance: Paris. Luquet, G.-H. 1931. La magie dans l’art paléolithique. Journal de Psychologie normale et pathologique 28e année: 390-427. Mainage, T. 1921. Les Religions de la Préhistoire. L’Âge Paléolithique. Desclée de Brouwer: Paris. Moro Abadia, O. & González Morales, M. R. 2005. L’analogie et la représentation de l’art primitif à la fin du XIXe siècle. L’Anthropologie 109 (4): 703-21.

Reinach, S. 1903. L’art et la magie. À propos des peintures et des gravures de l’Âge du Renne. L’Anthropologie XIV: 257-66. Richard, N. 1993. De l’art ludique à l’art magique. Interprétations de l’art pariétal au xixe siècle. Bulletin de la Société Préhistorique française 90 (1-2): 60-68. White, R. 2001. Personal ornaments from the Grotte du Renne at Arcy-sur-Cure. Athena Review 2 (4): 41-46. Yar, B. & Dubois, P. 1996. Les structures d’habitat au Paléolithique inférieur et moyen en France : entre réalité et imaginaire. Bulletin de la Société Préhistorique française 93 (2): 149-63.

29

30

Are hand stencils in European cave art older than we think? An evaluation of the existing data and their potential implications Paul Pettitt,1 Pablo Arias,2 Marcos García-Diez,3 Dirk Hoffmann,4 Alfredo Maximiano Castillejo,5 Roberto Ontañon-Peredo,6 Alistair Pike7 and João Zilhão8 Department of Archaeology, Durham University, South Road, Durham DH1 3LE, United Kingdom. The Cantabria Institute for Prehistoric Research, University of Cantabria, Edificio Interfacultativo, Avda. Los Castros s/n, 39005 Santander, Spain. 3 Departamento de Geografía, Prehistoria y Arquelogía, Facultad de Letras, University of the Basque Country UPV/EHU, c/ Tomás y Valiente s/n, 01006 Vitoria-Gazteiz, Álava, Spain. 4 Max Planck Institute for Evolutionary Anthropology, Department of Human Evolution, Deutscher Platz 6, 04103 Leipzig, Germany. 5 Facultad de Filosofía y Letras UNAM, Circuito Interior. Ciudad Universitaria, s/n. C.P. 04510. México, DF. México. 6 The Cantabria Institute for Prehistoric Research - Cuevas Prehistóricas de Cantabria, Carretera de las Cuevas s/n, 39670 Puente Viesgo, Spain. 7 Department of Archaeology, University of Southampton, Avenue Campus, Highfield Road, Southampton, SO17 1BF, UK. 8 University of Barcelona/ICREA, Departament de Prehistòria, Història Antiga i Arqueologia, ‘Grup de Recerca’ SERP SGR201400108, c/ Montalegre 6, 08001 Barcelona, Spain. 1

2

Abstract Since their discovery at the turn of the 20th century, hand stencils and prints have usually been assumed to belong largely, or entirely to the Mid Upper Palaeolithic (Gravettian sensu lato). These assumptions – rarely backed up by stringent consideration of chronometric data – derive from the pioneering chronological schemes of Breuil and Leroi-Gourhan. Here, we compile an exhaustive list of European Upper Palaeolithic caves which contain stencils and prints, and critically assess the available chronometric data. We conclude that in all or almost all known cases they are probably older, belonging to the Early or Initial Upper Palaeolithic, and for some examples one cannot rule out an older date. Keywords Upper Palaeolithic; cave art; Aurignacian; Gravettian; chronometry; hand stencils

has been published about these ostensibly intimate but intellectually ambiguous images, but the scholarly community has reached little understanding about their meaning and function in a century of research. Today, hand stencils (and far less commonly, positive prints) are known (to our understanding) in 37 caves: France: 26 caves = 70.3% of the total; Spain: 10 caves = 27%; and Italy: 1 cave = 2.7%. This estimate is based on a critical assessment of claims known to us (Table 1), and supersedes that in Pettitt et al. 2014. In most cases single caves contain only one or two stencils: more rarely they contain 5-15, and larger numbers are found only in La Garma (at least 39), Fuente del Trucho (at least 40), Cosquer (about 46), El Castillo (at least 85), Maltravieso (at least 71) and, most famously, Gargas (at least 212). Production usually involved the projection of a wet pigment – primarily red but occasionally black – via a tube or occasionally directly from the mouth, although other methods are known, such as the rubbing of pigments around the hand at Roucadour (Table 1).

Introduction Among his many meticulous publications on Spanish Upper Palaeolithic art, Rodrigo de Balbín Behrmann has documented many examples of the application of red pigments to cave walls directly by the fingers or hand, such as washes of red, paired or multiple lines, and finger dots (e.g. in La Lloseta [Balbín et al. 2005] and in Tito Bustillo [Balbín 1989; Balbín et al. 2002]). Perhaps the most iconic form of such interactions between the hand, pigments and cave walls are hand stencils, which are perhaps best contextualised as the most obvious extreme on a continuum of hand markings on walls. Given this, and as the chronology of cave art has been at the heart of his interests (e.g. Alcolea González and Balbín Behrmann 2007) we address here the question of the age of hand stencils as our homage to Rodrigo. Hand stencils in European Palaeolithic ‘cave art’ Since the first major discovery of hand stencils in Gargas in 1906 these have become a familiar component of the corpus of European Palaeolithic ‘cave art’. From the pioneering work of Breuil onwards, much

Meticulous documentation of multiple hand stencils exists only for Gargas (Groenen 1988; Barrière and Suères 1993; Sahly 1966; Foucher and Rumeau 2007), 31

Prehistoric Art as Prehistoric Culture Site

Notes

France (N=26) Abri du 1 black stencil. Poisson

Dating (chronometric) /

Assumed dating (associations &c)

References

No.

Assumed to be Gravettian on the basis of wider comparisons. Close proximity to engraved salmonid. Archaeological levels contain Aurignacian, Gravettian (Noaillian) and Solutrean levels. Prints are located away from the cave’s figurative art (thought to be Solutrean) and finger tracings. Dating of cave’s figurative art – amidst which the prints are located – is unclear: possibly Solutrean. Parietal art includes numerous black and red dots: figurative art of Middle and Late Magdalenian. Stencil found in close proximity to mammoth of same colour which (like the rest of the cave’s figurative art) is thought to be Magdalenian. Breuil saw the hand/s as Aurignacian.

Roussot 1984a. Delluc & Delluc 1991.

1

Leroi-Gourhan 1968. Drouot 1984a.

2

Leroi-Gourhan 1968. Drouot 1984b.

3

Gailli et al. 1984.Gailli 2006, 99-100.

4

Leroi-Gourhan 1968. Roussot 1984b.

5

BaumeLatrone

5 differing red prints.

/

Bayol (de Collias)

6 prints (5 adult, 1 child) of reddish clay

/

Bédeilhac

2 black prints, each with a red thumb.

/

Bernifal

1 brown/black stencil; 2 or 3 other possible engraved hands opposite this (alternatively these could be motifs arborescents). 2 black stencils.

/

/

/

Roussot 1984c.

6

1 brown/red stencil.

/

Lorblanchet 1984a.

7

Chauvet

11 in red (6 prints and 5 stencils).

/

Clottes 2003.

8

Les Combarelles (Section 1)

1 black stencil.

/

Three finger traces of the same colour 10cm from the stencil. Assumed to be early on the basis of associations & the wider reconstruction of the cave’s chronology. Breuil thought the stencil Aurignacian: Combarelles 1 engravings are early and Middle Magdalenian. Assumed to be Gravettian on the basis of wider regional parallels. Black punctuations, animal outlines in black.

Barrière 1984a.

9

Lorblanchet 2010, 390-2.

10

Grotte du Bison Bourgnetou

Combe-Nègre 1 stencil in black, not 1 blown but produced by a wash (badigeon) possibly similar to those of Roucadour (see below). Cosquer 65 stencils in red (21) and black (44).

Erberua (Isturitz Inférieur) Les Fieux

3 stencils (2 red, 1 black) (in cave’s 7th ensemble) 1 black. 14 stencils (12 red, 2 black) in two groups.

/

At least six AMS radiocarbon measurements on three hand stencils: MR7 (27,110 ± 430; 27,110 ± 400; 26,180 ± 370); M12 (24,840 ± 340; 23,150 ± 620) and M19 (27,740 ± 410). / /

11 Clottes et al. 2007. Assumed to be Gravettian on the basis of associations and the (Clottes & Courtin1996 direct radiocarbon measurements. for an earlier publication with lower count of stencils). Rouillon 2006.

Ensemble VII contains Magdalenian engravings as with the other of the cave’s ensembles. Assumed to be Gravettian or earlier on the grounds of associations and wider regional parallels e.g. Pech-Merle. Red punctuations and lines, animal outline engravings.

32

Larribau & Prudhomme 1984.

12

Lorblanchet 2010, 323-7.

13

Pettitt et al: Are hand stencils in European cave art older than we think? Site Font de Gaume

Notes 4 black stencils.

Dating (chronometric)

Assumed dating (associations &c)

/

Cave’s archaeology contains Mousterian, Aurignacian, Gravettian, Solutrean and Magdalenian levels. Figurative art is Magdalenian: Breuil thought the stencils Aurignacian. Les Garennes 1 black stencil. / Assumed to be early Gravettian (Vilhonneur) on the basis of proximity of the stencil to absolutely dated human remains from the cave floor. Art includes red dots, black bars and other traces of colour. / Engraved animal outlines, finger Gargas At least 212 stencils traces. Assumed to be Gravettian (Groenen), possibly after Breuil; on the basis of one 231 (Barrière) in red, radiocarbon measurement (see maroon, black and text); the basis of the cave’s white. 137 cluster archaeology, and probably closure together in Salle 1. shortly after the late Gravettian. Grand Grotte 8 stencils and 1 print / Assumed to be Aurignacian at Arcy-surin red or Gravettian on the basis of Cure the cave’s archaeology. AMS radiocarbon measurement of 26,700 ± 410 BP on bone found below panel including a partial stencil. Labattut (or 1 black stencil on / Stratigraphically earlier than the Labatut) detached ceiling block. upper level of Perigordian V with Noailles Burins (Noaillian) = early Gravettian or older. / Assumed to be Late Magdalenian Grotte (à) 2 black hand stencils on the basis of the cave’s Margot (one with attenuated archaeology, figurative engravings fingers). 2 positive with similarities to other regional brown prints. examples of Magdalenian art, and lack of Gravettian in the region, but the cave’s Aurignacian is more abundant that its Magdalenian. Moulin de 2 black stencils. / Assumed to be Gravettian on the Laguenay basis of wider regional parallels, and presumed association with hearth dated to ~26-27 ka (uncal) BP. Les Merveilles 6 hand stencils (four / Red dots, animal outlines in (Rocamadour) red, 2 black). red and black. Assumed to be Gravettian on the basis of wider regional parallels. Art of the cave’s (earliest) Radiocarbon Pech Merle 11 stencils and 1 ‘Sanctuaire A’ art phase including positive print in black measurement of black stencils of the spotted horse 24,640 ± 390 BP and red, 6 of which panel; dots and red hand stencils belong to the Spotted (Gif A 95357) on charcoal from right of the ‘femmes-bisons’ sector. Horse Panel. hand horse in the Spotted Horse panel of which hand stencils are part. / Assumed to be Gravettian on the Roucadour 13 stencils in red and basis of wider regional parallels. black in six panels (the second richest in the Quercy after Les Fieux). These were, however, produced by a method as yet unknown elsewhere,

33

References

No.

Leroi-Gourhan 1968. Roussot 1984d.

14

Henry-Gambier et al. 2007.

15

Sahly 1966. Pradel 1975. Barrière 1976. 1984b Barrière & Suères 1993. Groenen 1988. Foucher et al. 2007.

16

Baffier & Girard 2007.

17

Delluc & Delluc 1984.

18

Pigeaud et al. 2006. Jaubert & Feruglio 2007.

19

Lorblanchet 2010, 399. 20 Mélard et al. 2010.

Lorblanchet 2010. Leroi-Gourhan 1968.

21

Leroi-Gourhan 1968. Lorblanchet 1984b. Valladas et al. 1990. Lorblanchet 2010, 12227.

22

Lorblanchet 1984c. 2010, 351-2; 363.

23

Prehistoric Art as Prehistoric Culture Site

Tibiran

Trois-Frères

Roc de Vezac Spain (N=10) Altamira

Ardales Askondo El Castillo

Cudón

Fuente del Salín

Fuente del Trucho

Notes

Dating (chronometric)

notably the rubbing/ washing of red pigment across an elaborate area of fine incisions; as a result they should be viewed as representations not reproductions of the outline of hands. At least 11 (possibly / 18) stencils in red and grey, clustered in two panels. 5 red stencils. /

2 juxtaposed stencils (1 black, 1 red).

/

2 red prints and 4 violet stencils.

/

9 hands: 2 stencils (black) and 7 prints (red) 1 red print. At least 85 stencils in red.

/

1 stencil in red; the only one in Cantabria with attenuated fingers. 14 stencils in red and black.

~40 stencils of adults and infants clustered in 2 zones; 37 red, 3 black. This is probably an underestimate as more may be revealed with future cleaning: it has been conjectured that as many as 100 may eventually be revealed.

Assumed dating (associations &c)

References

No.

Contains finger engravings. Figurative art is Middle Magdalenian.

Leroi-Gourhan 1968. 24 Clot 1984. Pradel 1975.

Associated with numerous red points and traces. Breuil thought the stencils Aurignacian: cave’s figurative art is Middle Magdalenian. Unclear.

Bégouën & Clottes 1984.

25

Aujoulat 1984.

26

Assumed to be Aurignacian (or earlier) based on U-series minimum ages obtained for other red dots and images. Cave’s figurative art is Magdalenian.

Saura Ramos 1999. García-Diez et al. 2013 (dating).

27

28 Espejo Herrerías & Cantalejo Duarte 2006. Mijares 2011. Garate & Rios 2012. 29 30 Leroi-Gourhan 1968. Pike et al. 2012 (dating). Groenen 2012.

/ U-series dating of stalactite overlying stencil of the Panel de las Manos provides minimum age of ~37 ka cal BP. Similar for a red disk on the panel provides minimum age of ~40 ka cal BP. /

Probably Palaeolithic. The cave’s art probably relates to several periods: ongoing research is showing that the hand stencils and red dots are at least early Gravettian and probably older.

/

/

31

Direct AMS radiocarbon Measurement of 18,200 ± 70 BP on stencil (GX-27757-AMS) with incomplete pretreatment. U-series dating of stalactites stratified above one stencil indicate a minimum age of 27,000 (cal) BP.

AMS radiocarbon measurements of 22,580 ± 100 BP, 23,190 ± 900 BP and 22,340 +510/-480 BP on charcoal from hearths below stencils.

Bohigas et al. 1985. Moure & González Morales 1992. Moure et al. 1985. González Morales & Moure 2008.

32

Assumed to be early on the basis of superimpositioning of later figurative art on stencils; probably pre-solutrean.

Utrilla et al. 2013. 2014. Utrilla & Bea, this volume.

33

34

Pettitt et al: Are hand stencils in European cave art older than we think? Site La Garma Maltravieso

Tito Bustillo Italy (N=1) Paglicci

Notes

Dating (chronometric)

At least 39 stencils in / red (24) and yellow (15). At least 71 red stencils. /

1 red stencil. Possibly a / second. / At least 3 stencils. Colour is unclear: this appears red but could be due to the rock; some white colourant is visible (M. Mussi pers. comm.)

Assumed dating (associations &c)

References

No.

Assumed to belong to the cave’s older phase, not its Magdalenian art. Assumed to be Gravettian on the basis of wider parallels, e.g. Gargas. Unclear associations: possibly red triangles, meanders. Unclear: potentially Early Upper Palaeolithic.

González-Sainz 2003.

34

Ripoll López et al. 1999a. 1999b.

35

Saura Ramos & PérezSeoane 2007.

36

Usually assumed to be Gravettian due to parallels with stencils elsewhere, or Solutrean on the basis of the style of the cave’s horse depictions.

Zorzi 1962. Mussi 2000, 264-5.

37

Table 1. Corpus of caves containing hand stencils/prints known to the authors. Note that some counts of hand stencils/prints include an example from Cougnac (e.g. Ripoll López et al. 1999, Figure 115). This is actually a main essuyée/frottée (a ‘wiped’ or ‘rubbed’ hand) produced by dragging fingers covered in black pigment down the wall (Lorblanchet 2010, 274-5; see also Lorblanchet 2009 for a wider discussion of these). This is not a depiction of a hand, and in fact is much closer to finger tracings than to hand stencils, and for this reason we omit it from our quantification. Similarly, a circle of 5 finger dots from the cave (ibid., 257) is excluded. We also omit the main frottée in the Grotte du Cantal, Lot (Lorblanchet 2010, 394), and possible engravings of hands in Bara-Bahau and Ebbou, the former of which was suggested by the Abbé Glory but it is debatable, and the latter probably a natural stain (Paul Bahn pers. comm.). We also omit caves which have from time to time been reported informally as having hand stencils but which do not, i.e. Le Portel, El Pindal (actually a red disk - González-Pumariega Solís 2011), Oxocelhaya, Grotte du Cheval (these are actually all finger tracings); we also omit sites for which a possible stencil has been suggested but which nevertheless remains unclear (one between two bovids in Gallery B of La Pasiega: Balbín and González Sainz 1992; González Sainz and Balbín 2000); and omit three stencils in La Lastrilla for which a Palaeolithic ascription is not certain.

Cosquer (Clottes et al. 1992; Clottes and Courtin 1996, 69-79) and Maltravieso caves (Ripoll López et al. 1999a, b). Although these account for a large sample of known stencils, on a site-by-site basis the literature is poor, but this lack of an overall corpus of data on stencils from the 38 sites has not prevented the accumulation of a relatively large literature on their production and possible meaning. One notable exception is the chronological review of García-Diez et al. (2015). Other than the onsite study of stencils in context, the literature typically reflects research focussed entirely on the identity of hand stencils rather than their physical context, i.e.

neutral) which is usually taken to mean either missing/mutilated or bent back (e.g.Breuil 1952; Janssens 1957; Sahly 1966; Leroi-Gourhan 1967; Pradel 1975; Hooper 1980; Wildgoose et al. 1982; Barrière and Suères 1993; Ripoll López et al. 1999a, b; Guthrie 2005, 114-32; Rouillon 2006.). We should not be too focussed on these as they occur in only a small number of caves that contain hand stencils (notably Gargas, Tibiran, Cosquer and Fuente del Trucho) and need not be central to the understanding of stencils and prints as a whole. •• The possibility that stencils/prints were ‘signatures for those who were responsible for the art on the walls’ (Gregg 2008, 380 our emphasis; see also Taçon et al. 2014).

•• The gender and age of the people whose hands were depicted (e.g. Manning et al. 1998; Gunn 2006; Snow 2006). •• Whether left or right hands were depicted (e.g. Barrière 1976; Groenen 1997; Faurie and Raymond 2004; Frayer et al. 2007; Steele and Uomini 2009), usually from the perspective of handedness and its evolution among hominins. •• Why in some caves fingers or parts of them appear attenuated (a term we prefer instead of ‘missing’ or ‘mutilated’ as it is interpretatively

To summarise the results of research in these areas, it would probably be fair to say that most researchers agree that the left hand was overwhelmingly stencilled (presumably because 80% of the time the right hand was the active one and thus held the materials necessary for stencilling of the passive left hand); that taken at face value finger ratios and lengths are often (but not always) 35

Prehistoric Art as Prehistoric Culture

Figure 1. Selection of French and Spanish hand stencils. Clockwise from top left: El Castillo (placed in concave depression); La Garma (small group); Ardales (on stalactite); Pech Merle (with red discs). Photo credits: Gobierno de Cantabria (La Garma and El Castillo), Pedro Cantalejo Duarte (Ardales) and Paul Bahn/Jean Vertut (Pech Merle).

consistent with female hands; that in the few cases where attenuated fingers are present these are probably the result of deliberate bending rather than disease, frost bite or accident; and that there is no reason to assume that surviving stencils represent more than a single or small number of individuals in each cave. There has been relatively little interest in the physical context of stencils, although a recent study of such in La Garma and El Castillo caves has demonstrated how stencils were commonly associated with cracks in the cave walls, and with subtle concavities and bosses, revealing an interest in the small-scale scrutiny of the cave walls (Pettitt et al. 2014).

stratigraphie, le contexte, ou les superpositions (Fuente del Salín, Altamira, Castillo, La Garma en Espagne), (Gargas, Hautes-Pyrénées), Cosquer (Bouches-duRhône), Labattut et l’Abri du Poisson (Dordogne), Moulin de Laguenay (Corrèze), Vilhonneur (Charente) se situent au Gravettien, entre 22,000 et 28,000 ans avec une plus grande fréquence entre 25,000 et 28,000 ans BP’ (our emphasis).How robust are such conclusions? We review critically the existing data pertinent to the age of hand stencils on which such a long-standing consensus is based, and conclude that they are almost certainly older than has been previously thought. We then consider the ramifications of this conclusion.

Here, we are not concerned with the production and ‘function/s’ of hand stencils or the identity of the stencilled, but with their antiquity. It is universally assumed that they are of Mid Upper Palaeolithic age, i.e. that they are culturally Gravettian. As Lorblanchet (2010, 221), for example, has noted, ‘toutes les mains négatives paléolithiques datées par le radiocarbone, la

Relative schemes and artistic associations from Breuil onwards Breuil (1952, 38) assigned hand prints and stencils to his earliest (Aurignacian-Perigordian) art cycle on the basis of their preceding stratigraphically ‘all other paintings’ and their apparent lack of association with anything other 36

Pettitt et al: Are hand stencils in European cave art older than we think? than ‘rare spots, lines of discs in series, and sometimes timid attempts at line drawing.’ During the next decade Leroi-Gourhan acknowledged, however, that the dating of hand stencils was ambiguous, although a close reading of his statements makes it clear that he was aware that the little data available were not inconsistent with Breuil’s notion of a relatively early age. Thus ‘the [dating of] hands present one of the problems still needing clarification. The Abbé Breuil regarded them as very archaic, and in several cases they do seem to belong to an early phase of cave decoration’ (1968, 199). LeroiGourhan used the association of art attributable to one or more of his stylistic phases – assuming that the association was meaningful – in order to assign hand stencils to one of his four great phases of cave art. Thus, he argued ‘at Gargas, the cave contains only figures in Style II and Style III; at Pech Merle, the hands occur in the vicinity of figures in the earliest Style III; at Bernifal, we find them in the first chamber, opposite painted figures that are in an indefinable style, but are a priori earlier than the engravings in the remote part. In a few cases, such as Les Combarelles, Font-de-Gaume, and El Castillo, it was hard to place the hands chronologically in relation to a group that is predominantly style III’ (ibid., 199). From this it is clear to infer that he thought that the examples of hand stencils in these caves belonged to his early Style III or earlier, thus to the Solutrean/Early Magdalenian – although only in one case did he explicitly state this (Tibiran; 1968, 321). Today we may be more critical of Leroi-Gourhan’s assumption that the perceived style of art in relatively close proximity to hand stencils is a reliable indication of their age, although as we shall see below this assumption is still made and still can form the basis of assumptions about the Mid Upper Palaeolithic age of stencils.

age, although are contradicted by Clottes and Courtin (1996, 167) and Clottes (1998, 114-5) who thought the oldest examples were of Gravettian age. Snow (1996) recognised that some might be older than the Gravettian; Davidson (1997, 148) assumed that they are the ‘earliest figures in Upper Palaeolithic cave art’ although referred to the stencils of Cosquer Cave as Gravettian; and Gárate (2008, 24) saw them as part of a set of human themes including human outlines and vulvae which was ‘significant until the Solutrean’. Bahn and Vertut (1988, 135) saw the issue as open, noting that they may span the entirety of the Upper Palaeolithic on the basis of the lack of evidence to the contrary. The age of hand stencils and prints Recently, García-Diez et al. (2015) critically reviewed the chronology of hand stencils in the context of new U-series minimum ages for stencils in El Castillo, concluding on the basis of production technique and colour and of a critical consideration of available chronological data, that the stencils can broadly be viewed as a diachronic phenomenon, probably an initial and non-figurative phase (Aurignacian or earlier) of European Palaeolithic cave art, of which the youngest examples were created around 27,000 cal BP. Here, we have assembled what we hope to be the most comprehensive catalogue of Upper Palaeolithic stencils (and the less frequent prints), and we assess how their age has been ascertained and conclude that in most (or all) cases they are likely to be early Gravettian at youngest, and probably much older. As the following discussion shows, direct dates on hand stencils (AMS radiocarbon on charcoal) are remarkably rare, and where they exist may be underestimates given how long ago the dates were produced and given that pretreatment techniques have improved considerably since. Stratigraphic associations (such as when fragments of cave ceiling bearing stencils have fallen into dated contexts) are even rarer. Much ‘dating’ of stencils/prints tends to be based on perceived spatial associations, either between the art of concern and dated archaeology, dated bones stuffed into cracks in the cave wall, or stylistically dated art. Such associations may be illusory. Most ‘dating’ of stencils simply reflects the dogma that they are ‘Gravettian’. As we shall see, when Occam’s razor is applied to cut out questionable ‘dating’ the results are consistent with a relatively old age for the stencils/prints for whom reliable information exists.

Breuil’s view - which at least partly overlapped with that of Leroi-Gourhan - prevailed, but subsequent researchers to the present day have come to view hand stencils as largely or entirely Gravettian, whether explicitly or implicity (e.g. Barrière and Suères 1993, 49; Clottes 1998; Clottes and Courtin 1996, 166-7; Foucher et al. 2007, 83; Lawson 2012, 318; Lorblanchet 1995, 245-6; 2010, 224; Ripoll López et al. 1999b, 13; Von Petzinger and Nowell 2010. White 1993, 69). Thus although Breuil assigned stencils and prints to a phase that spanned both the Aurignacian and Gravettian, subsequent publications have come to associate them only with the Gravettian, although in no published case, however, is it clear why an earlier age has apparently been ruled out. A few exceptions exist. Sahly (1966, 276) viewed them as Aurignacian although did not explain why; a broader Aurignacian/Gravettian age was suggested by Bernaldo de Quirós and Cabrera (1994, 268) and by Lorblanchet (2007, 211), views that seem to be echoed by von Petzinger and Nowell (2011, 1178-80) in their critique of stylistic dating of cave art. Clottes and Lewis-Williams (1998, 45) also suggest a broad Aurignacian/Gravettian

Dating: one stratified example Ucko and Rosenfeld (1967, 67) were critical of a supposed stencil on a block recovered from between two Perigordian levels in the Labattut rockshelter (Dordogne), although its context is well recorded and the stencil is clear on a photo published by Delluc and Delluc (1991). It can be taken as a clear indication that the fragment of 37

Prehistoric Art as Prehistoric Culture cave wall/ceiling on which the stencil was created fell during the Gravettian, the context of which therefore provides a minimum age for the creation of the stencil itself. This is perhaps not surprising given the general similarity of the Labattut art with Aurignacian rock art from shelters in the vicinity (cf. Delluc and Delluc 1991); it could be Gravettian, it may well be older.

association only; they are conjectural, and should not become dogma. The same caution must be applied to the hand stencil found several metres from human remains radiocarbon dated to 27,110 ± 210 BP and 26,790 ± 190 BP in Les Garennes cave, Vilhonneur, France (Henry-Gambier et al. 2007). Once again, while the measurements presumably constitute reliable evidence of the death of this individual during the Gravettian, an association between the two is conjecture and as it has not been demonstrated should be removed from consideration.

Absolute dating: radiocarbon Independent verification of the supposed age of stencils/prints in the form of absolute dates is very rare. Despite this rarity, the consensus has been built up that existing radiocarbon measurements support the notion of a Gravettian age, and thus stencils and prints have, like ‘Venus figurines’ come to be seen as icons of the European Mid Upper Palaeolithic (e.g. Foucher and San Juan-Foucher 2007. Jaubert 2008. Ucko and Rosenfeld 1967, 72).

Grotte à Margot in Mayenne is assumed to be Late Magdalenian in age but is not directly dated (Pigeaud et al. 2006). In addition to its Magdalenian archaeology the cave has yielded Aurignacian material (actually more abundant than the cave’s scarce Late Magdalenian), thus while it seems to have no Gravettian activity one cannot rule out an EUP age for its four stencils; once again we would urge caution against arguing from the basis of the cave’s archaeology.

A very few AMS radiocarbon measurements exist which are cited as constituting chronological evidence of the antiquity of stencils. Most of these are not without problems, however. In fact these few measurements take the form of:

A clearer association can be observed in Le Moulin de Laguenay cave, Corrèze. Here, a radiocarbon measurement of 26,770 ± 380 BP (Lyon-3361 Poz) was obtained on charcoal from a hearth in a thin horizon directly atop bedrock that contained fragments of spalled roof on which pigments are visible, immediately below two ceiling stencils (Mélard et al. 2010). The lack of any evidence for activity belonging to any other periods in the cave, and general scarcity of archaeological material strengthens the notion that these data pertain to the same period, but this is not unequivocally demonstrated. If such an association is valid then the measurement may only provide a minimum age for the stencils, given that it would be the spalling of art on the part of the cave’s ceiling on which they were produced – not necessarily their production per se – that occurred in the same broad period that the hearth was lit.

•• Measurements on objects found close to hand stencils in caves, for which a meaningful association between the two is assumed but not demonstrated beyond doubt. •• Measurements on objects found close to hand stencils in caves for which a meaningful association between the two is probably but not completely unequivocal. •• Measurements on charcoal from cave art apparently in clear association with hand stencils and thus meaningfully associated. •• Measurements on charcoal taken directly from hand stencils. Three results of 22,580 ± 100 BP, 23,190 ± 900 BP and 22,340 +510/-480 BP from Fuente del Salín (Moure Romanillo and González Morales 1992) actually measure charcoal taken from hearths close to the stencils of interest, although a direct measurement of 18,200 ± 70 BP, if correct (see below), suggests caution in the use of such apparently spatially ‘associated’ dates, and a measurement of 26,860 ± 460 BP from Gargas is actually on a bone splinter wedged into a crack near the Great Panel of Hands (Foucher and Rumeau 2007, 83). These are not clear associations, and while they demonstrate close to the location of hand stencils the burning of hearths and the insertion into a crack of the bone of an animal that died during the Gravettian (although the insertion could of course have occurred later), and are thus not inconsistent with Gravettian ages for them, they are not necessarily relevant to the stencils’ age. One should be cautious of these age assignments based on

A measurement of 24,640 ± 390 BP (Gif A 95357) was obtained on charcoal from the chest area of the right of the two dappled horses1 of Pech-Merle, which do appear to be meaningfully associated with six hand stencils in this panel on the basis of both its complex compositional phases and of pigment analysis of several elements including two stencils and the horses themselves (Lorblanchet 1995. 2010, 122-35). This has been interpreted in the light of the regional style of art in several caves of the Quercy, which is seen as fairly homogeneously Gravettian (e.g. Jaubert 2008) albeit of several phases (Lorblanchet 2010). Pech-Merle 1 By using the normal means of reference to these, we do not mean to imply that they depict horses with dappled pelage. As Lorblanchet (2010, 105) has argued, the presence of punctuations outside the drawn outline of these animals argues against this; instead one is dealing with a complex interplay of animal outlines, punctuations, hand stencils and other signs, which may or may not reflect true pelage.

38

Pettitt et al: Are hand stencils in European cave art older than we think? using the previously routine acid-base-acid pretreatment for charcoal; re-measuring several samples from the Grotta di Fumane using the more rigorous ABOx-SC method ages were obtained that were typically 2-4kyr older than the previous measurements (and in some cases more). We would expect that the minuscule samples of charcoal removed from the cave art samples of concern here would compound the problem even further. With regard to the remaining measurements from Fuente del Salín, the lack of explicit published information on pretreatment and measurement precludes independent assessment of the accuracy of the measurements.

does in fact present a clear warning about the dangers of assuming the age of art on a cave’s walls on the basis of radiometric dates on materials found in close proximity. A metacarpal of reindeer recovered from Sondage 1 beneath the Panel of Dappled Horses yielded a radiocarbon measurement of 18,400 ± 350 BP and a charcoal fragment 11,380 ± 390 BP (Valladas et al. 1990. Lorblanchet 2010, 18), which are clearly of much younger ages than that of the charcoal that went into the production of the dappled horses which are presumed to belong to the cave’s oldest phase of art (Lorblanchet 2010, 220-5). Similarly, a charcoal fragment from Sondage VII beneath the Gravettian Frise Noir yielded an age of 11,200 ± 800 BP. In the Grande Grotte at Arcy-sur-Cure, a measurement of 26,700 ± 410 BP was obtained on a bone recovered at the foot of a panel which included a partial hand stencil (Baffier and Girard 2007), and measurements of 26,360 ± 290 BP and 26,250 ± 280 BP were obtained on charcoal from the floor of the Gallery of Dots in the Grotte Chauvet. Why these should pertain to the art is unclear. The dangers of assuming associations between art panels and objects immediately below them on the cave’s floor should be obvious.

What are we to make of such a poor database? First, that consensus can emerge among archaeologists on the basis of relatively poor data; when we critically examine the database on which our assumptions are made it becomes clear how unsound some of our conclusions can be. Secondly, that the very few measurements that can be taken as at all reliable suggest that the hand stencils of concern are at least of Gravettian age but in fact could be considerably older. One should of course put this in perspective: almost all hand stencils known to us have no direct dates, i.e. the assumption on the basis of stylistic associations that they are of Mid Upper Palaeolithic age has not been independently verified by reliable radiometric dating. Viewed from this perspective we regard the issue of the age of hand stencils as open.

To our knowledge AMS 14C measurements directly on the charcoal of a hand stencil derive only from two caves: Grotte Cosquer (Clottes et al. 1992) and Fuente del Salín (González Morales and Moure 2008). Publication of the dates from Cosquer has not been consistent but we identify at least six measurements on three hand stencils: MR7 (27,110 ± 430; 27,110 ± 400; 26,180 ± 370); M12 (24,840 ± 340; 23,150 ± 620) and M19 (27,740 ± 410) although the lack of supporting information renders it impossible to evaluate these independently. A direct AMS radiocarbon measurement of 18,200 ± 70 BP on a stencil from Fuente del Salín (González Morales and Moure 2008) was measured at Geochron without full pretreatment, so this must be regarded as questionable. The lesson with these direct dates is not to publish AMS measurements resulting from samples that have been incompletely pretreated; how can one be confident that all contaminating carbon has been removed?

Absolute dating: U-series Recent U-series dating of stalagmites overlying two stencils in El Castillo has provided clearer indications of their minimum ages, in this case of ~24,000 and ~37,000 years ago (Pike et al. 2012). U-series dating of calcite deposits has several advantages over 14C dating of charcoal pigments in that it doesn’t require the presence of organic pigments, nor suffers from the ‘old wood’ effect, and can be verified by stratigraphic consistency of dates along the growth axis of the calcite. These new results provide independent verification of the early age of stencils as suspected by Breuil, and in the case of the oldest measurement clearly a pre-Gravettian cultural context. They are part of a suite of dates on various motifs, including disks and hand stencils, from several caves that show that red non-figurative painting dates back at least to the Aurignacian in Northern Spain.

Thus we are left with only two sites where direct dates on stencils exist, and one (Pech Merle) where a plausible relationship exists between dated art and stencils that seem to be part of the same panel: Pech-Merle and Cosquer. These were, however, measured two decades ago; available samples sizes for measurement of these would be problematically small at the time, and modern pretreatment methods for charcoal which have been proven to be more successful removing contamination would not have been available, thus for these reasons specialists today would presumably view these as inaccurate (probably minimum) ages. Higham (2011) has, for example, demonstrated considerable problems with the accuracy of measurements on charcoal for samples older than 20,000 14C BP that were produced

The age of hand stencils and some possible implications Overall, the reliable chronometric data available at present are consistent with the notion that stencils and prints belong to an early, largely non-figurative phase of cave art, prior to a subsequent rise to dominance of animal figures that began in the Gravettian and culminated in the Magdalenian (Ripoll López et al. 1999, 73. Gárate 2008; García-Diez et al. 2015). As Breuil noted artistic associations of hand stencils are typically with disks 39

Prehistoric Art as Prehistoric Culture (‘ponctuations’) usually produced by a similar method of pigment projection, and possibly with animal outlines assumed to be early Gravettian in age (although this needs verification). Some simple conclusions clarify the issue somewhat:

hands or fingers through impressions) and the artificial creation of (one might say representation) of the outline of the hands using the projection or rubbing of pigments. In a sense hand stencils are both figurative (in that they ‘depict’ a human hand) and non-figurative (in that they are not conscious drawings of the hand but an attempt to fix the outline of the hand in place). Is it possible that their very nature at the borders of the figurative and non-figurative, and their apparent appearance just as figurative art is emerging in European caves, suggests they played a role in the recognition that things could be figured in art? If the hand could be represented in outline, then why not animals?

•• Artistic associations, where demonstrable, support Breuil’s view that hand stencils belong to a relatively early (or indeed the earliest) artistic period. •• By contrast, caves that seem to contain parietal art of exclusively Magdalenian age – e.g. several in the valley of the Lot river in Quercy (Lorblanchet 2010, 406-27) – do not contain hand stencils. There are, therefore, no associations between hand stencils and post-Gravettian art. •• Radiocarbon measurements have indicated an early to late Gravettian age for a very few stencils, but these were produced a long time ago with previous laboratory methods and are almost certainly inaccurate underestimates. Even if they are chronometrically reliable they probably indicate minimum ages. •• Preliminary U-series measurements attest a Gravettian age as a minimum, and in one case a clearly pre-Gravettian age; a date — older than 39.9 ka —falling clearly in pre-Gravettian times has also been obtained for the one example of a hand stencil outside Europe (the Leang Timpuseng cave in Sulawesi) where U-series was applied to overlying calcite (Aubert et al. 2014). In Europe and Sulawesi artistic associations place hand stencils in the context of broader nonfigurative art. •• If the early age of stencils is borne out by further analyses it may be of interpretive importance, given that they fall into a conceptual space between non-figurative and figurative art, and it may be no coincidence that their creation forms an outline (of a hand) in the same period as simple animal outlines were emerging in parietal art.

The apparently older age of hand stencils also raises the question of their authorship. It is important to recognise that the chronology we have for them at present is poor, and is entirely comprised of minimum ages. While these may belong to Aurignacian or Protoaurignacian cultural contexts – and thus presumably indicate that the stencilled and stencillers were Homo sapiens, can we eliminate the possibility that they were made by, and depict Neanderthals? Further minimum ages for hand stencils should at least be able to test this hypothesis.

Acknowledgements We are grateful to Pilar Utrilla, Margherita Mussi and Paul Bahn for information about hand stencils in Spanish, Italian and French caves, and to Paul Bahn, Pedro Cantalejo Duarte, and the Gobierno de Cantabria for photographs. References Alcolea González, J. J. and Balbín Behrmann, R. de. 2007. 14C and style. La chronologie de l’art parietal à l’heure actuelle. L’Anthropologie 111, 435-666. Arias, P., Laval, E.,Menu, M., González-Sainz, C. and Ontañón, R. 2011. Les colorants dans l’art parietal et mobilier paléolithique de La Garma (Cantabrie, Espagne). L’Anthropologie 115, 425-45. Aubert, M., Brumm, A., Ramli, R., Sutikna, T., Sapromo, E. W., Hakim, B., Morwood, M. J., Van Den Berg, G. D., Kinsley, L. and Dosseto, A. Pleistocene cave art from Sulawesi, Indonesia. Nature 514, 223-7. Aujoulat, N. 1984.Grotte du Roc de Vézac.In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises.Paris: Ministère de la Culture, 242-4. Aujoulat, N. and Dauriac, N. 1984. Grotte de BaraBahau. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 92-95. Baffier, D. and Girard, M. 2007. La Grand Grotte d’Arcysur-Cure. Les Dossiers d’Archéologie 324, 74-85.

If, then, hand stencils belong to an early – perhaps the earliest – phase of European cave art, one should view them in the context of the emergence of the evolution of art. What exactly are hand stencils: figures or signs, or something in-between? Might they have played a role in the evolution of figurative art in Europe? Stencils form part of a continuum of marks on the walls, ceilings and floors of caves created by direct contact with parts of the body, from foot and hand prints (Lorblanchet 2009) and finger meanders (Sharpe and Van Gelder 2006), through positive palm prints (Clottes and Courtin 1996), finger and hand ‘rubbing’ (Lorblanchet 2010) to the projected pigment hand stencils and positive prints that are of concern here. A conceptual continuity runs through this set of examples, from ‘natural’ markings (which one might conceive of as the reproduction of the outline of the 40

Pettitt et al: Are hand stencils in European cave art older than we think? Balbín Behrmann, R. de. 1989. L’art de la Grotte de Tito Bustillo (Ribadasella, Espagne). Une vision de synthèse. L’Anthropologie 93, 435-62. Balbín Behrmann, R. de.,Alcolea González, J. J., González Pereda, M. A. and Moure Romanillo, A. 2002. Recherches dans le massif d’Ardines : nouvelles galeries ornées de la grotte de Tito Bustillo. L’Anthropologie 106, 565-602. Balbín Behrmann, R. de.,Alcolea González, J. J. and González Pereda, M. A. 2005. La Lloseta : une grotte importante et presque méconnue dans l’ensemble de Ardines, Ribadesella. L’Anthropologie 109, 641-701. Balbín, R. de & González Sainz, C., 1992: La Pasiega. Monte de El Castillo, Puente Viesgo, Cantabria. En VV. AA. El nacimiento del arte en Europa: 239-241 (Catálogo de la exposición). Paris: Unión Latina. Barrière, C. 1976. L’Art Pariétal de la Grotte de Gargas. Oxford: British Archaeological Reports International Series 409. Barrière, C. 1984a. Grotte des Combarelles 1. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 109-13. Barrière, C. 1984b. Grotte de Gargas. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 514-22. Barrière, C. and Suères, M. 1993. Les mains de Gargas. Dossiers d’Archeologie 178 (La Main dans la Préhistoire), 46-54. Bégouën, R. and Clottes, J. 1984. Grotte des TroisFrères. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 400-9. Bohigas, R., Sarabia, P., Brígido, B., Ibáñez, L., 1985. Informe sobre el santuario rupestre paleolítico de la Fuente del Salín (Muñorrodero, Val de San Vicente, Cantabria). Boletín Cántabro de Espeleología 6, 8198. Breuil, H. 1952. Four Hundred Centuries of Cave Art. Paris: Sappho. Clot, A. 1984. Grotte de Tibiran. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 536-9. Clottes, J. 1998. The ‘three Cs’: fresh avenues towards European Palaeolithic art. In Chippindale, C. and Taçon, P. S. The Archaeology of Rock Art. Cambridge: Cambridge University Press, 112-29. Clottes, J. 2003. Return to Chauvet Cave. Excavating the Birthplace of Art: the First Report. London: Thames and Hudson. Clottes, J. and Courtin, J. 1996. The Cave Beneath the Sea. Palaeolithic Images at Cosquer. New York: Harry N. Abrams. Clottes, J., Courtin, J., Valladas, H., Cachier, H., Mercier, N. and Arnold, M. 1992. La Grotte Cosquer datée. Bulletin de la Société Préhistorique Française 89, 230-4.

Clottes, J., Courtin, J. et Vanrell, L. 2007. La Grotte Cosquer à Marseille. Les Dossiers d’Archéologie 324, 38-45. Combier, J. 1984.Grotted’Ebbou. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 609-16. Delluc, B. and Delluc, G. 1984. Abri Labattut. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 220-1. Delluc, B. and Delluc, G. 1991. L’Art Pariétal Archaïque en Aquitaine. Paris: Gallia Préhistoire Supplement 28. Drouot, E. 1984a. Grotte de la Baume Latrone. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 333-9. Drouot, E. 1984b. Grotte Bayol. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 323-6. Espejo Herrerías, M. and Cantalejo Duarte, P. 2006. Cueva de Ardales (Malaga). Reproducción Digital del Arte Rupestre Paleolítico. Comarca del Guadalteba. Faurie, C. and Raymond, M. 2004. Handedness frequency over more than ten thousand years. Proceedings of the Royal Society of London Series B 271, S43-S45. Foucher, P. and Rumeau, Y. 2007. Les galleries ornées de Gargas. In Foucher, P., San Juan-Foucher, C. and Rumeau, Y. (eds.) La Grotte de Gargas. Un Siècle de Découvertes. Communauté de Communes du Canton de Saint-Laurant de Neste, 61-86. Foucher, P. and San Juan-Foucher, C. 2007. Gargas dans le context Gravettien Européen. In Foucher, P., San Juan-Foucher, C. and Rumeau, Y. (eds.) La Grotte de Gargas. Un Siècle de Découvertes. Communauté de Communes du Canton de Saint-Laurant de Neste, 113-6. Foucher, P., San Juan-Foucher, C. and Rumeau, Y. (eds.) 2007. La Grotte de Gargas. Un Siècle de Découvertes. Communauté de Communes du Canton de Saint-Laurant de Neste. Frayer, D. W., Lozano, M., Bermúdez de Castro, J. M., Carbonell, E., Arsuaga, J. L. Radovčić, J., Fiore, I. and Bondioli, L. 2007. More than 500,000 years of right-handedness in Europe. Laterality 1, 1-19. Gailli, R. 2006. La Grotte de Bédeilhac. Préhistoire Histoire et Histoires. Toulouse: Editions Larrey. Gaillie, R., Pailhaugue, N. and Rouzaud, F. 1984. Grotte de Bédeilhac. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 369-75. Garate, D. 2008. The continuation of graphic traditions in Cantabrian pre-Magdalenian parietal art. International Newsletter on Rock Art 50, 18-25. Garate, D. and Ríos, J. 2012. La Cueva de Askondo (Mañaria, Bizkaia). Arte Parietal y Occupación humana Durante la Preistoria. Bilbao: Diputación Foral de Bizkaia. 41

Prehistoric Art as Prehistoric Culture García-Diez, M., Hoffman, D. L., Zilhão, J., de las Heras, C., Lasheras, J. A., Montes, R. and Pike, A. W. G. 2013. Uranium-series dating reveals a long sequence of rock art at Altamira Cave (Santillana del Mar, Cantabria). Journal of Archaeological Science 40, 4098-16. García-Diez, M., Garrido, D., Hoffmann. D. L., Pettitt, P. B., Pike, A. W. G. and Zilhão, J. 2015. The chronology of hand stencils in European Palaeolithic rock art: implications of new U-series results from El Castillo Cave (Cantabria, Spain). Journal of Anthropological Sciences 93, 1-18. González Morales, M. R. & Moure, A. 2008. Excavaciones y estudio de arte rupestre en la cueva de la Fuente del Salín (Muñorrodero, Val de San Vicente). Campaña de 2000. In: Excavaciones Arqueológicas en Cantabria 2000-2003, pp. 7982. Consejería de Cultura, Turismo y Deporte del Gobierno de Cantabria, Santander.  González-Pumariega Solís, M. 2011. La cueva de El Pindal 1911-2011, Estudio de su arte rupestre cien años después de Les cavernes de la région cantabrique. Gijón: Ménsula Ediciones. González Sainz, C. 2003. El conjunto parietal paleolítico de la Galería inferior de La Garma (Cantabria): avance de su organización interna. In Balbín Behrmann, R. and Bueno Ramírez, P. (eds.) El Arte Prehistórico desde los Inicios del Siglo XXI: Primer Symposium Internacional de Arte Prehistórico di Ribadasella. Ribadasella: Associación Cultural Amigos de Ribadasella, 201-22. González Sainz, C. & Balbín, R. de, 2000: Revisión de las representaciones rupestres paleolíticas de la cueva de La Pasiega en el conjunto del monte Castillo. Topografía y documentación artística. En R. Ontañón (coord.): Actuaciones Arqueológicas en Cantabria 1984-1999. Santander: Consejería de Cultura y Deporte, Gobierno de Cantabria, 69-73. Gregg, J. M. 2008. Ancients inspire modern memory. Nature Nanotechnology 3, 380-81. Groenen, M. 1988. Les representations de mains negatives dans le grottes de Gargas et de Tibiran (Hautes-Pyrénées). Approche méthodologique. Bulletin de la Société Royale Belge d’Anthropologie et de Préhistoire 99, 81-113. Groenen, M. 1997. Ombre et Lumières dans l’Art des Grottes. Brussels: U. L. B. Cahiers d’études VI. Groenen, M. 2012. Recorridos por la cueva de El Castillo. En busca de la mirada del Paleolítico. In: VV.AA. Arte sin artistas. Una mirada al Paleolítico [Catálogo de exposición]. Alcalá de Henares: Museo Arqueológico de la Comunidad de Madrid, 372-393. Gunn, R.G. 2006. Hand sizes in rock art: interpreting the measurements of hand stencils and prints. Rock Art Research 23, 97–112. Guthrie, R. D. 2005. The Nature of Paleolithic Art. Chicago: University of Chicago Press.

Henry-Gambier, D., Beauval, C., Airvaux, J., Aujoulat, N., Baratin, J. F. and Buisson-Catil, J. 2007. New hominid remains associated with gravettian parietal art (Les Garennes, Vilhonneur, France). Journal of Human Evolution 53, 747-50. Hooper, A. 1980.Further information on the prehistoric representations of human hands in the cave of Gargas. Medical History 24, 214-6. Janssens, P. A. 1957. Medical views on prehistoric representations of human hands. Medical History 1, 318-22. Jaubert, J. 2008. L’art pariétal gravettien en France: éléments pour un bilan chronologique. Paleo 20, 205-37. Jaubert, J. and Feruglio, V. 2007. Les grottes ornées de Combe-Nègre. Les Dossiers d’Archéologie 24, 6873. Larribau, J. D. and Prudhomme, S. 1984. Grotte d’Erberua. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 275-9. Leroi-Gourhan, A. 1967. Les mains de Gargas. Essaie pour une étude d’ensemble. Bulletin de la Sociétè Préhistorique Française 64, 107-22. Leroi-Gourhan, A. 1968. The Art of Prehistoric Man in Western Europe.London: Thames and Hudson. Lorblanchet, M. 1984a. Grotte du Bournetou. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 488-9. Lorblanchet, M. 1984b. Grotte du Pech-Merle. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 467-74. Lorblanchet, M. 1984c. Grotte de Roucadour. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 511-13. Lorblanchet, M. 1995. Les Grottes Ornées de la Préhistoire: Nouveaux Regards. Paris: Editions Errance. Lorblanchet, M. 2009. Claw marks and ritual traces in the Paleolithic sanctuaries of the Quercy. In Bahn, P. (ed.) An Enquiring Mind: Studies in Honor of Alexander Marshack. Oxford: Oxbow and Cambridge MA: American School of Prehistoric Research Monograph Series, 166-70. Lorblanchet, M. 2010. Art Pariétal. Grottes Ornées du Quercy. Parc-Saint-Joseph: Éditions Rouergue. Manning, J. T., Scutt, D., Wilson, J. and Lewis-Jones, D. I. 1998. The ratio of 2nd to 4th digit length: a predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen. Human Reproduction 13, 3000-4. Mélard, N., Pigeaud, R., Primault, J. and Rodet, J. 2010. Gravettian painting and associated activity at Le Moulin de Laguenay (Lissac-sur-Couze, Corréze). Antiquity 84, 666-80. Mijares, R. M. 2011. Arte Prehistórico en las Tierras de Antequera. Junta de Andalucía. 42

Pettitt et al: Are hand stencils in European cave art older than we think? Moure Romanillo, A. and González Morales, M.R., 1992. Datatión 14 C d’une zone décorée de la grotte Fuente del Salín en Espagne. International Newsletter on Rock Art (INORA) 3, 1-2. Moure, A., González Morales, M.R. and González Sainz, C. 1985. Las pinturas paleolíticas de la cueva de la Fuente del Salín (Muñorrodero, Cantabria). Ars Praehistórica 3-4, 13-23. Mussi, M. 2000. Earliest Italy. An Overview of the Italian Paleolithic and Mesolithic. New York: Kluwer and Plenum. Pettitt, P. B., Maximiano-Castillejo, A., Arias, P., Ontañón-Peredo, R. and Harrison, R. 2014. New views on old hands: the context of stencils in El Castillo and La Garma Caves (Cantabria, Spain). Antiquity 88, 47-63. Pigeaud, R., Rodet, J., Devièse, T., Dufayet, C., TrelohanChauve, E., Betton, J.-P. and Bonic, P. 2006. Palaeolithic cave art in west France: an exceptional discovery: the Margot Cave (Mayenne). Antiquity 80 project gallery: http://antiquity.ac.uk/projgall/ pigeaud309. Last accessed 26.5.2015. Pike, A. W. G., Hoffman, D. L., García-Diez, M., Pettitt, P. B., Alcolea, J., González-Sainz, C., de las Heras, C., Lasheras, J. A., Montes, R. and Zilhão, J. 2012. Uranium-series dating of Upper Palaeolithic art in Spanish caves. Science 336, 1409-13. Pradel, L. 1975. Les mains incomplètes de Gargas, Tibiran et Maltravieso. Quartär 26, 159-166. Ripoll López, S., Ripoll Perelló, E. and Collado Giraldo, H. 1999a. Maltravieso. El Santuario Extremeño de las Manos. Cáceres: Museo de Cáceres. Ripoll López, S., Ripoll Perelló, E., Collado Giraldo, H., Mas Cornellá, M. and Jordá Pardo, J. 1999b. Maltravieso. El santuario extremeño de las manos. Trabajos de Prehistoria 56, 59-84. Rouillon, A. 2006. Au Gravettien, dans la Grotte Cosquer (Marseille, Bouches-du-Rhône), l’homme a-t-il compté sur ses doigts? L’Anthropologie 110, 500-9. Roussot, A. 1984a. Abri du Poisson. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 154-6. Roussot, A. 1984b. Grotte de Bernifal. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 170-74. Roussot, A. 1984c. Grotte du Bison. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 175-7. Roussot, A. 1984d. Grotte de Font-de-Gaume. In L’Art des Cavernes. Atlas des Grottes Ornées Paléolithiques Françaises. Paris: Ministère de la Culture, 129-34. Sahly, A. 1966. Les Mains Mutilées dans l’Art Préhistorique. Toulouse: Privat. Saura Ramos, P. A. 1999. The Cave of Altamira. New York: Harry N. Abrams.

Saura Ramos, P. A. and Pérez-Seoane, M. M. 2007. Arte Paleolítico de Asturias: Ocho Santuarios Subterráneos. Oviedo: CajAstur. Snow, D.R. (2006) Sexual dimorphism in Upper Palaeolithic hand stencils. Antiquity 80, 390-404. Steele, J. and Uomini, N., 2009. Can the archaeology of manual specialization tell us anything about language evolution? a survey of the state of play. In: Malafouris, L. and Renfrew, C. (eds.) Cambridge Archaeological Journal 19(1), 97–110. Taçon, P., Hidalgo Tan, N., O’Connor, S., Xueping, J., Gang, L., Curnoe, D., Bulbeck, D., Hakim, B., Sumantri, I., Than, H., Sokrithy, I., Chia, S., KhunNeay, K. and Kong, S. 2014. The global implications of the early surviving rock art of greater Southeast Indonesia. Antiquity 88, 1050-64. Ucko, P. and Rosenfeld, A. 1967. Palaeolithic Cave Art. London: Weidenfeld and Nicolson. Utrilla, P. Baldellou, V., Bea, M. and Viñas, R. 2013. La cueva de la Fuente del Trucho (Asque-Colungo, Huesca). Una cueva mayor del arte gravetiense. In Pensando el Gravetiense: Nuevos datos para la Región Cantábrica en su Contexto Peninsular y Pirenaico. Altamira: Monografías del Museo Nacional y Centro de Investigación de Altamira, 52637. Utrilla, P. Baldellou, V., Bea, M., Montes, L. and Domingo, R. 2014. La Fuente del Trucho. Ocupación, estilo y cronología. In Soledad Corchón, M. and Menéndez, M. (eds.) Cien años de arte rupestre paleolítico: Centenario del descubrimiento de la cueva de la Peña de Candamo(1914-2014). Salamanca: Ediciones Universidad de Salamanca (Acta Salmanticensia. Estudios históricos y geográficos 160), 119-132. Valladas, H., Cachier, H. and Arnold, M. 1990. Application de la datation C14 en specrométrie de masse par accélérateur aux grottes ornées de Cougnac et du Pech-Merle (Lot). In Étude des pigments des grottes ornées paléolithiques du Quercy. Bulletin de la Société des Études du Lot, 2e fascicule, 93-143. Von Petzinger, G. and Nowell, A. 2011. A question of style: reconsidering stylistic approaches to dating Palaeolithic parietal art in France. Antiquity 85, 1165-83. White, R. 1993. Préhistoire. Bordeaux: SudOuest. Wildgoose, M., Hadingham, E. and Hooper, A. 1982. The prehistoric hand pictures at Gargas: attempts at simulation. Medical History 26, 205-7. Zorzi F. 1962. Pitture parietali e oggetti d’arte mobiliare del Paleolitico scoperti nella grotta Paglicci presso Rignano garganico. Rivista di Scienze Preistoriche 17, 123-137.

43

44

Regional ontologies in the Early Upper Palaeolithic: the place of mammoth and cave lion in the ‘belief world’ (Glaubenswelt) of the Swabian Aurignacian Shumon T. Hussain1 and Harald Floss2 Faculty of Archaeology, Leiden University, the Netherlands Department for Early Prehistory and Quaternary Ecology, University of Tübingen, Germany 1

2

Abstract The treatment of animals and their representation in systems of personal ornamentation and figurative art in the Early Upper Palaeolithic are traditionally seen as sources of ecological information or as a general expression of symbolic cognition and social identity. Here, we suggest that a non-Cartesian and effectively relational account of human-animal interactions in glacial landscapes opens up interpretive avenues that provide fresh and even more persuasive perspectives on the issue. While discussing the case of the Swabian Aurignacian, we argue that mammoths and cave lions are not only the most frequently represented beings in the material culture repertoire, but also that their encounters with humans have probably been highly significant, on both cognitive and eco-behavioural grounds. We therefore propose that the treatment of both animals in the Swabian Aurignacian mirrors their social relevance and thereby reflects a mode of policing the animal-human interface that assigns them a crucial place in regional and deeply animistic ontologies. At the same time, differences in the eco-behavioural matrix of both animals can account for differences in cultural practices that relate to them. We conclude that behaviourally and ecologically salient keystone species are also cultural keystone species in the Swabian Aurignacian which, in turn, points to the importance of naturally inherent interaction dynamics in constructing the animal-human interface in this period. Keywords Aurignacian; early Upper Palaeolithic; figurative art; personal ornaments; mammoth; cave lion; animal-human interface; keystone species; regional ontology

Introduction In recent years, scrutinizing the historical and crosscultural validity of the animal-human boundary has become a common point of departure for academic inquiry (Corbey 2005; Kalof 2007; DeMello 2012; Corbey & Lanjouw 2013; Ogden et al. 2013). Such work adds to an emerging awareness that the Cartesian, ‘scientific’ and often essentialist conception of the world that the ‘West’ – and therefore most of academia – subscribes to, is actually a historically highly exceptional case (e.g. Descola 1992, 2011; Willerslev 2007; Kohn 2013). How we see, experience, linguistically and conceptually ‘dissect’, and ultimately relate to the world is a sociocultural construction in itself, and therefore nothing that can be taken for granted (or even as a solid starting point) when we try to analyse past human ways of life. At the same time, the fundamental tempospatial plasticity of the human-world interface, opens up plenty of opportunities for archaeologists to better understand these alien social units and the ‘modes of being’ that underpin their material and non-material lifestyle performances (e.g. Robb & Harris 2013). Each classificatory boundary we draw today – for example the classical Cartesian dualities of body vs. mind, human vs. object, living vs. non-living and sentient vs. non-sentient – can then be understood as a problematic interface whose construction can vary through time and 45

can thereby inform us about fundamentals of ‘being-inthe-world’ that underlie human lifestyles in the past. In other words: how different human-nonhuman interfaces are policed and negotiated in different past sociocultural settings can tell us something about past ontologies and past ways of knowing and relating to the world that are very different from our own (David 2006; Gosden 2008; Hill 2012). The delineation of the animal-human boundary is a salient expression of such a mode of relating to the world because it epitomises assumptions about human exceptionality and human supremacy, as well as reflecting an advancing separation between animal and human worlds in modern societies. Hence, it offers a perfect case study for exploring the nature and role (if any) of the animal-human divide in humanity’s deep past. Animal-human interfaces in the Palaeolithic Most westerners today live in almost entirely crafted and artificially fabricated environments, for example in cities, industrial hubs or even metropolises – they are essentially ‘city-dwellers’ that live in human-dominated environments. In most places, this situation has critically reduced animal presence by dissecting and destroying their habitats and has repulsed other-than-human beings (sensu Whatmore 2006) to the ‘wilderness’. Animal and human existences – at least where animal life

Prehistoric Art as Prehistoric Culture could not adapt to the new situation – have therefore become critically disconnected, both experientially and spatially. This (human-made) worldly configuration can be contrasted with the situation that Palaeolithic forager groups encountered when they navigated Pleistocene tundra-steppe environments almost 40,000 years ago.

landscape dwellers’ that lived in mammalian, otherthan-human-dominated environments. This matrix of interaction preconditions clearly makes a difference and has sweeping consequences for the nature and role of animal-human relationships in the Palaeolithic, which, in turn, can inform us about some basic features of Palaeolithic world-views and ontologies.

Non-human animals were not a rarity and not even an uncommon feature of these landscapes, but rather the dominant entities, both in population densities and in terms of ecological impact (Hofreiter & Steward 2009; Zimov et al. 2012). Additionally, living in a populated and in fact rather crowded landscape critically reduces the proximity between humans and animals, and at times even crucially intersects animal and human ‘taskscapes’ (sensu Ingold 2000). The glacial mammoth steppe environments of the last Ice Age are essentially animal landscapes that alone afford significance to animal agents (cf. Reed 1988). Humans directly engage with many animal species on a daily basis, animals preoccupy the visual space of human groups, and the latter often rely heavily on their resources for survival and cultural efflorescence (cf. Mithen 1999; Shipman 2010). As indicated by many ethnographically documented communities that have retained a hunter-gatherer mode of existence (e.g. Bird-David 1999; 2006; Kuriyan 2002; DeMello 2012: 33-35), such a phenomenology of the landscape invokes notions of co-inhabitation and sharing, notions that have the power to undermine the construction of the Cartesian animal-human boundary in the first place, and rather afford policing the animalhuman interface in more intimate and socially significant ways.

Keystone animals in the Swabian Aurignacian The specific mode of being-in-the-world as mobile hunter-gatherers in open mammoth steppe environments suggests that the focus of most European Upper Palaeolithic communities on the representation of animals in their material culture, either directly in figurative form, or indirectly by transforming animal parts into tools and personal ornaments (e.g. Mithen 1999), is not only an abstract expression of the sociocultural fabric underpinning these units but also refers to specific ways of modulating the animal-human interface in this period. Material culture, in particular in Palaeolithic settings where animals, one way or another, nearly always figure prominently, thus opens up the excellent possibility of exploring human-animal relationships and their bearing on wider sociocultural organisation. If animals invoke meaning and significance beyond Cartesian economics, it is important to analyse the ‘handling’ of these animals in relation to better defined regional groupings since different local communities can develop quite different regional ontologies, although their general sociocultural performance remains quite similar – a phenomenon that Alfred Kroeber (1939) and Franz Boas tried to capture with the term ‘cultural area’ (Harris 1968: 237). In the case of the Early Upper Palaeolithic, such an internally structured sociocultural landscape is indeed indicated by regionally discrete traditions of personal ornament making and use (Vanhaeren & d’Errico 2006). We therefore use the example of the Swabian Aurignacian (ca. 42-30 ka cal. BP) and its ‘handling’ of the animal topic to demonstrate the utility of discussing the Palaeolithic material record through the lenses of animal-human relations on a regional scale (Fig. 1). Since the preconditions of animal-human interactions that shape encounters and experiences are dependent not only on variables pertaining to the overall configuration of the landscape and to more general modes of lifestyle (e.g. ‘forager-hood’), but also on the specifics of animal appearance and behaviour, it seems crucial to also confront the ethology and socioecology of animals in these landscapes with the characteristics of animal-related material culture signatures in our regional sample. This approach is exemplified here by the contextualisation of information on the behavioural matrix of woolly mammoth (Mammuthus primigenius) and cave lion (Panthera spelaea) and the sociocultural treatment of these two animals in the Swabian Aurignacian. Both are keystone species (Cottee-Jones & Whittaker 2012) and have a salient effect on their environment, while at the

The natural centrality of animal life to mammoth steppe environments cannot be overstated (e.g. Van Königswald 2004). A significantly reduced tree-cover, coupled with glacier- and loess-flattened surroundings, renders dwelling animals the main experience of Pleistocene landscape inhabitation. A small but potentially critical aspect of this ‘worldly architecture’ is that animals are clearly one of the most dynamic and ‘active’ landscape features in these rather generic settings. Attention is therefore naturally channelled towards animal life and the peculiarities and effects of animal behaviour. The significance of animals for human subsistence and other cultural practices – for the realm that cultural ecologist Julian Steward (1955: 36-39) has called ‘cultural core’ – further fosters their focality (sensu Schelling 1981) from a mobile human forager perspective. In this sense, animals can be considered to open up a space of naturally inherent social significance. Hence, the construction of the animal-human interface in the Palaeolithic is framed by radically different and rather unique encounter and experience conditions from those of today, and is generally underlined by proximity and intimacy rather than by deprivation and seclusion. In contrast to most westerners today, Pleistocene hunter-gatherer groups were ‘animal46

Hussain and Floss: Regional ontologies in the Early Upper Palaeolithic

Figure 1. Map of Central and Southern Germany showing the location of the Swabian Aurignacian sites discussed in the text. 1 Geissenklösterle, 2 Sirgenstein, 3 Hohle Fels, 4 Brillenhöhle, 5 Börslingen, 6 Bockstein, 7 Hohlenstein-Stadel, 8 Vogelherd. All figurine sites are cave sites.

Figure 2. Frequency of different animal representations in the visual art repertoire of the Swabian Aurignacian.

same time being the most frequently depicted animals in the Aurignacian figurative art sample from the Swabian Jura (Fig. 2). What follows is thus a comparison of natural relevances and significances that are already laid out in mammoth and cave lion behaviour on the one hand, and the role of both animals in figurative art making, personal ornamentation and subsistence practices of the Swabian Aurignacian on the other hand. This framework will then ultimately allow us, in the last step, to shed new

light on the construction of the animal-human interface in Southwestern Germany during the Aurignacian. Mammoth-human interfaces a. Eco-behavioural matrix of the mammoth The woolly mammoth (Mammuthus primigenius) is one of the most iconic inhabitants of Pleistocene Eurasian 47

Prehistoric Art as Prehistoric Culture tundra-steppe environments (e.g. Van Königswald 2004). It is essentially a cold-adapted elephant that was part of the ‘megaherbivore’ guild of the mammoth steppe, once being distributed from Western Eurasia to Alaska (Hofreiter & Lister 1999; Sukumar 2003; Bocherens et al. 2014). Several factors of mammoth spatial presence, ecology and behaviour render the largest mammal of the last Ice Age both a keystone species of its environment and a very significant encounter for human foragers. To begin with, mammoths are highly visible animals – not only because they are almost as big as African or Asian elephants, but also because they are expected to have moved around in social units (Hofreiter & Lister 1999) – and this visibility is further enhanced by reduced vertical vegetation and flattened glacial landscapes. The visio-spatial dominance of mammoths is also supported by preliminary calculations of population densities for mammalian communities in the mammoth steppe superbiome, reaching numbers of over 300,000 individuals for the glacier-free areas of Germany alone (calculations per km² after Zimov et al. 2012). Like today’s elephants, they can further be expected – at least occasionally – to dwell near human camps, thereby further enhancing an experiential matrix of mammoth omnipresence in the landscape.

bushes, for example, they rip down entire branches and often leave behind only tree-carcasses, creating scatters of easy collectable wood that can be used for fire making and other purposes. Elephant behaviour also generates dispersed and spatially patched natural bone and ivory stocks in the wider landscape. These examples show that elephants – and by analogy mammoths – have a strong impact on their surroundings and on how other animals, including humans, behave and organise their presence. For human foragers, this opens up a co-adaptive interface since mammoth and human ‘taskscapes’ can easily become inextricably interlocked (see Locke 2013). Generally speaking, mammoth behaviour and presence alone can be an effective heuristic (sensu Gigerenzer & Gaissmaier 2011) to operate in the landscape, and the keystone function of these animals renders the open glacial environments of Eurasia effectively as mammothscapes (Haynes 2006, 2013). These landscapes are virtually and even literally ‘stained in mammoth’, and mammoth behaviour and its clues are both behaviourally significant and visio-spatially omnipresent. This matrix structurally intersects the cognitive and behavioural space of mammoth and humans and affords a ‘thinking with mammoths’ (sensu Lévi-Strauss 2007).

The visio-spatial salience of elephants in open landscapes is only one side of the medal, however; behavioural prominence and eco-systemic significance is the other (e.g. Putshkov 2003). Elephants in general are ‘eco-system engineers’ (Haynes 2013) and occupy unique and integral positions in their wider ecosystem, and even critically help to ensure its functioning (Haynes 2006; Zimov et al. 2012). Yet, elephantid ecological significance is not only abstract and purely ecological in nature, but can be translated in tangible affordances that help human foragers to navigate and exploit the landscapes they co-inhabit with these animals (Haynes 2006, 2013). From this perspective, mammoth ecological and behavioural prominence is not merely something that can be experienced, but something that naturally and quite ‘automatically’ reveals itself as behaviourally significant. A few examples will suffice here to review this well-documented dimension of elephant spatial presence.

We would argue that the embodied proximity between mammoths and humans laid out so far indicates not only a naturally inherent predisposition of mammoth presence and activity to become socially significant for humans, but also that this proximity is in fact strongly underscored by the social similarity between elephant ‘societies’ and human ‘societies’ (Bradshaw 2009). In this light, the argument would be that the mammothhuman intersection in the Late Pleistocene mammoth steppe is characterised by at least partial intersection of human and mammoth ‘taskscapes’, inseparability between landscape knowledge and experience and mammoth behaviour and close phenomenological proximity between human and mammoth sociality. The latter would add to interaction preconditions that easily ‘entrap’ people in seeing and experiencing mammoths as social agents, even granting them the status of ‘persons’ at times. Elephant social organisation is known to be complex, social relationships can be very intimate and elephant socially embedded behaviour is at times very idiosyncratic and, that way, suggests personality and individuality (Moss 2000; Sukumar 2003: 170-184; Bradshaw et al. 2005; Wittemyer et al. 2005). These animals display behaviours across the entire emotive spectrum, from weeping over intimate trunk-touching to aggressive display (Masson & McCarthy 1996: 91-110) and therefore engage with humans in different situations in many different ways. Together with their environmentshaping impact, they are therefore easily perceived as bearers of agency, intentionality and sentience – in particular when humans do not conceptualise worldly affairs along Cartesian dichotomies.

Elephants are large-scale ‘earth-movers’ and thereby shape their surroundings rather extensively (cf. Redmond 1982; Haynes 2006). For example, they dig waterholes or extend water patches and even streams, expose salt-stones and -sources, and even ‘construct’ trails as they walk the landscape in a recurrent pattern. Elephant presence, hence, signals the availability of water and other resources, and their ‘road networks’ afford and channel movement. Also critical for Palaeolithic foragers seems to be the fact that elephantinhabited landscapes incorporate a unique raw material potential. When elephants feed on trees and bigger 48

Hussain and Floss: Regional ontologies in the Early Upper Palaeolithic Personhood is in many ethnographically documented hunter-gatherer communities in fact not at all reserved for human agents, but can rather be a crucial property of many living and even some non-living entities in the world (Bird-David 1999; Descola 2011; Santos-Granero 2012; Kohn 2013). It is often more a consequence of relational engagements and thus of how people encounter and experience other-than-human entities than a preconceived fixation of essential properties in specific phenomena (e.g. Bird-David & Naveh 2008; Hill 2013). For the reasons presented above, elephants seem to be predisposed, being regarded as such meaningful and socially significant encounters – the structural and ecological preconditions of human-mammoth interactions thus stipulate inclusion of elephantids in the social grid. A selective reading of the ethnographic literature seems to support this view – the motif of an ‘elephant-person’ figures quite prominently there (cf. Locke 2013; Barua 2014). We give two short examples here to indicate the wealth and richness of elephant-human relations in the ethnographic record before we turn to the archaeological evidence that might support the view that has been sketched so far. For both the Samburu of Northern Kenya (Kahindi 2001; Kuriyan 2002) and the Nayaka of Southern India (Bird-David 2006; Bird-David & Naveh 2008) elephants are important social agents that occupy salient positions in the world-web of these people. Both groups use anthropomorphic terminology to refer to specific elephants and their behaviour, and directly compare elephant groups and human groups. Elephants are also deeply entrenched in ritual and social practices, and both communities feel a constant need to re-negotiate their relationship with these animals. Although elephants remain ambivalent for those people in one aspect or another, the general relationship with these animals is considered to be truly positive. This interaction matrix climaxes in the Samburu case in sociocultural techniques that aim at sustaining this positive bond by, for example, prohibiting (tabooing) the killing and eating of elephants (Kahindi 2001: 36-37). Both examples ultimately illustrate how behavioural and ecological significance regularly merges into social and even cultural relevance. Keeping this observation in mind, we can now turn to mammoth-related characteristics of the material record of the Swabian Aurignacian.

personal ornaments of the Swabian Aurignacian are predominantly manufactured from mammoth ivory (Wolf 2015; Fig. 3-4). Mammoth is therefore omnipresent, either in its distinctive shape or indirectly in its ‘matter’. This already indicates the social significance of the mammoth and the interweaving of mammoth and human domains. One could even say that the social nature of daily interactions with mammoths is mirrored in the strongly patterned chaînes opératoires of ivory beads and pendants which are embedded in everyday practices (Niven 2006; Wolf 2015). Use-wear and abrasion traces, moreover, demonstrate that these items have been attached to clothing and accessories (Wolf 2015: 29193), and have therefore probably been a salient feature of the public and visual space of the Swabian Aurignacian. The omnipresence of the mammoth-reference thus seems to reproduce the omnipresence of the living mammoth in glacial mammothscapes (that are ‘stained in mammoth’). We know that many traditional societies do not reduce the characteristics of specific entities to specific ‘matter states’, for example certain behavioural traits to a specific living animal, but rather believe in the transposition and genuinely animated nature of various such states that sometimes relate only very loosely to ‘matter states’ that bring forth these characteristics in a causal sense. Yupik Inuit groups and many others, for example, believe in the evocation of animal personalities and attributes by means of attaching amulets that relate to these animals – examples are perforated teeth, ivory carvings and animal figurines (Hill 2012). The cementation of the mammoth-theme in the sociocultural architecture of the Swabian Aurignacian speaks for the blurring of boundaries between mammoths and humans, for example in the sense of ‘mammoths are humanpersons and humans are mammoth-persons’. Carving mammoth figurines and wearing mammoth ivory beads and pendants could thus reflect cultural techniques to constantly (and performatively) re-establish the human-mammoth bond and concurrently express the close social proximity of man and mammoth. Such a process could even be interpreted from a gift-giving perspective where wearing mammoth-derived material culture would constitute the co-presence of the human individual and the mammoth, since materials that are endowed or ‘gifted’ are often regarded as conveying or embodying a part of the giver’s immaterial personality (Malinowski 1920, 1922; Därmann 2010). Seen from this light, the omnipresence and predominance of mammothrelated material culture in the Swabian Aurignacian is potentially indicative of exchange and the social nature of the mammoth-human relationship. It at least points to the fact that systems of personal ornamentation are poorly addressed as purely decorative elements, but have a rich social life and often reflect complex social relationships at the human-animal interface (e.g. Hill 2012, 2013). The salience of the mammoth-theme, then, at least indicates the social importance of the mammoth in the ontology of the Swabian Aurignacian.

b. Personal ornament making and figurative art as a mirror of mammoth significance The mammoth-theme is predominant in the craftsmanship of the Swabian Aurignacian. It penetrates the realms of personal ornamentation and figurative art-making in a double sense. Mammoths are either figuratively carved out of mammoth ivory or the latter is transformed into various types of beads, pendants and/or amulets (Hahn 1986; Floss 2007; Conard et al. 2015; Wolf 2015). Mammoth figurines represent the most frequent animal species in the sample (see Fig. 2), and the diagnostic 49

Prehistoric Art as Prehistoric Culture

Figure 3. Hohle Fels, chaîne opératoire of ivory beads in the Aurignacian. Photographs: S. Wolf and H. Jensen. After Wolf 2015 (Abb. 86). Courtesy of Sibylle Wolf.

sometimes to be overrepresented (Münzel 2001) is also consistent with the view that mammoth remains were collected from the landscape and imported to the sites independently of killing and/or hunting these animals since elephant-landscapes are known to produce a wealth of elephant carcasses and young individuals are more likely to die of natural causes (Haynes 2006). In this context, it is interesting to note that several pieces of the ivory assemblage from the Swabian Aurignacian clearly made their way as rotten specimens to the sites, and rotten ivory even seems to be the raw material for some animal figurines from these sites (Steguweit 2015). Differential weathering of mammoth bones and ivory from the Aurignacian layers of Vogelherd (Niven 2003) further supports the argument that mammoth material was predominantly ‘gathered’ from the landscape rather than retrieved from intentionally killed animals. The splitting and percussive techniques that underlie Aurignacian ivory processing in the Swabian Jura (Heckel & Wolf 2014) would at least be consistent with this ‘embedded’ perspective on ivory procurement.

The individualistic representation of the mammoth in figurative art (compare Floss 2007; Fig. 4) adds to this picture and supports the conceptualisation of mammoths at the fringe of personhood. c. Ivory and bone management as a reflection of mammothscapes Claude Lévi-Strauss (2007) famously claimed that – at least in totemic worldviews – animals are either ‘good to eat’ or ‘good to think’ but rarely both at the same time. The faunal record of the Swabian Aurignacian sites is rather suspicious in this respect (see Münzel & Conard 2004; Niven 2006, 2007). Whereas the subsistence- and consumption-related exploitation of reindeer and horse is well evidenced (Niven 2007), the treatment of mammoth remains is highly contested and ambiguous (Münzel 2001; Wolf 2015: 287-89). In Vogelherd cave, for example, there are in fact almost no cut-marks on mammoth bones and hardly any signs of marrow extraction or the like (Niven 2006). The fact that younger individuals seem 50

Hussain and Floss: Regional ontologies in the Early Upper Palaeolithic

Figure 4. Mammoth representations from the Swabian Aurignacian. 1-4, 6: Vogelherd, 5: Geissenklösterle. Unscaled. Photographs: J. Lipták and H. Jensen, © University of Tübingen.

hunting behaviours and partly extensive hunting ranges in these animals. Although cave lions were probably not as rare as sometimes assumed (Zimov et al. 2012), their ‘taskscapes’ – in contrast to the structural dynamics of mammoth-human interactions – are crucially separated from those of human foragers. Like other large felids today, these animals are essentially adapted to night vision (Bradshaw 2013) and can thus be expected to have primarily been nocturnal hunters (e.g. Seidenstricker et al. 1973). The comparison with other felids and their morphology strongly suggests that they were solitary predators – African lions mark a clear anomaly in this respect (cf. Packer 2010) – and very effective ‘stalkers’ (MacDonald & Loveridge 2010). Therefore, the cave lion appears as a keystone predator in the Pleistocene mammoth steppe that critically controlled ungulate biomass, while at the same time being almost elusive and scarcely perceptible for humans. Stalkers usually try to avoid open areas and consequently use the veil of darkness as an additional camouflage to approach their prey (Laing & Lindzey 1991; Parker et al. 2011).

One way or another, the evidence indicates that mammoths are unlikely to fall into the category of ordinary prey species – an observation that is very interesting in light of the likely social significance of mammoths for Swabian Aurignacian people. In the Lévi-Straussian sense, this can be read as suggestive of mammoths being ‘not good to eat’ and therefore ‘good to think’ – a perspective that could even be interpreted as an expression of latent hunting prohibitions in this period. Lion-human interfaces a. Eco-behavioural matrix of the cave lion The cave lion (Panthera spelaea) was probably the prime predator of the Eurasian mammoth steppe and was once – like the mammoth – dispersed from Western Eurasia to Northwest America (Stuart & Lister 2011; Yeakel et al. 2013). Isotopic fingerprinting points to highly individualistic diets in this taxon (Bocherens et al. 2011; Bocherens 2015) which, in turn, indicates idiosyncratic 51

Prehistoric Art as Prehistoric Culture These behavioural variables effectively reduce the visibility of the cave lion and disconnect the animal from the human ‘taskscape’. The fact that lions usually do not generate characteristic large-scale communication soundscapes, but rely more on olfactory clues – as, for example, exemplified in the North American mountain lion (Seidenstricker et al. 1973) – adds to this picture and further reduces their entanglement with human foragers.

highlighting representations, clearly situates the cave lion as the most salient animal in the figurine sample. The unusual blending of human and lion characteristics that constitutes the hybrid motif also shows that cave lions have been conceptualised in close reference to humans. Three figurines from the Swabian Aurignacian incorporate a ‘mixture’ of lion and human traits: the iconic ‘lion man’ from Hohlenstein-Stadel in the Lone valley, the ‘small lion man’ from Hohle Fels, and the ‘adorant’ from Geißenklösterle, the latter two sites being located in the Ach valley (Hahn 1986; Floss 2007; Kind et al. 2014; Conard et al. 2015; see Fig. 5). All three representations have been described as anthropomorphic in stature and posture with a ‘felid-morphic’ head as well as less defined, ‘transitional’ features (Kind et al. 2014; Wolf 2015: 252-53). This representational logic points to the blurring of boundaries between the ‘humane’ and the ‘lione’ and consequently to an (ontological) fluidity between these two domains (Mithen 1999; Porr 2015). The fusion of lion-ness and human-ness shows that some properties have been regarded as interchangeable across the human-lion intersection by Aurignacian people, while the difference between being a ‘lion player’ and being a ‘human player’ was still unbridgeable. This configuration reproduces the entanglement of human and lion affairs as hunters and the disentanglement of their ‘taskscapes’ as well as the deep ambiguity and tension inherent in lion-human interactions in Late Pleistocene settings. As such the ‘treatment’ of the cave lion in the Swabian Aurignacian figurative art indicates both social significance and deep ambivalence, and therefore marks the figuration of a liminal being whose relations to the human world had to be re-negotiated time and again.

At the same time, the cave lion emerges as a ‘contester’ for ‘man the hunter’ in these landscapes, and competes with humans for certain prey species – a structural interaction precondition that is deeply ambivalent in nature. As versatile predators, cave lions evoke comparison and similarity with human hunters while concurrently appearing as deadly and dangerous antagonists. This matrix of interaction conditions indicates the natural significance of the cave lion, but – in contrast to the mammoth – not so much as a ‘social partner’ than as an ambiguous and liminal being that ‘walks between human and other-than-human worlds’. Such an ambivalent stance towards lions is for example welldocumented within lion-Maasai relationships in Kenya and Tanzania (Lichtenberg 2005; Goldman et al. 2010). These semi-nomadic people see lions as respectable and human-like but also dangerous animals – lion attacks on humans are not uncommon there (Packer et al. 2011). In comparison to other savannah-grassland carnivores, however, the lion is considered the only ‘smart’ hunter that is regularly associated with attributes like ‘awe’, ‘bravery’ and ‘strength’, and that way elicits positive and negative sentiments at the same time (Goldman et al. 2010) – human-lion interactions are therefore essentially characterised by tension. This, again, shows how ecobehavioural relations between lions and humans merge into ways of relating to these animals socioculturally. The modalities of representing the lion in the figurine sample of the Swabian Aurignacian is a perfect case for exploring this blending of eco-behavioural and sociocultural variables that signify the lion-human interface at this time.

Interpreting the animal-human interface in the Swabian Aurignacian The ‘ontological turn’ in French philosophical anthropology draws attention to the fact that humanity – in history and prehistory – has always been characterised by a multiplicity of worldviews grounded in different ‘modes of being’ which, in turn, usually express themselves in corresponding ontologies. How the world is conceptually ‘dissected’ and finally conceived and experienced is not a given but arises from the conditions under which other worldly entities such as objects, animals and even landscapes are encountered and socioculturally processed – the Cartesian conception of worldly order being a mere historical exception in this respect. Palaeolithic archaeology is hence in a unique position to shed new light on how humans in the deep past – in the ‘cradle of humanity’ – have accessed and seen the world, and thereby to push the history of ‘human ontology’ far back in time. Methodologically, it is useful, therefore, to re-cast prominent Cartesian dichotomies as problematic ‘interfaces’ and analyse their construction in specific spatio-temporal settings. Our contribution is a stepping stone on this path,

b. Figuration of the lion-man as a reflection of cave lion liminality Alongside the mammoth, the cave lion is the most frequently depicted animal in the figurine sample from the Swabian Aurignacian (Floss 2007; Conard et al. 2015; see Fig. 2). Lions are either represented naturalistically and with a high degree of artistic variation, again pointing to an individualistic emphasis, or appear as mixed anthropomorphic figures (Fig. 5). Depending on whether these two groups are merged together or not, the cave lion emerges as either the dominant, or the second-most frequently realised motif – after the mammoth – in the Aurignacian figurative art of the Swabian Jura. However, the high degree of general artistic variation as well as the internal fragmentation in animality- vs. hybridity52

Hussain and Floss: Regional ontologies in the Early Upper Palaeolithic

Figure 5. Cave lion and hybrid “lion-man” representations from the Swabian Aurignacian. 1, 2, 4-5: Vogelherd, 3: Hohle Fels, 6: Geissenklösterle, 7: Hohlenstein-Stadel. Unscaled. Photographs: J. Lipták and H. Jensen (University of Tübingen), Y. Mühleis (Landesamt für Denkmahlpflege BadenWürttemberg, Regierungspräsidium Stuttgart), © University of Tübingen and Museum Ulm. Courtesy of Kurt Wehrberger.

Strikingly, the most extensively portrayed animals in the Swabian Aurignacian material record – mammoth and cave lion – are also animals that display behaviours that are relevant and significant for humans in one way or another. This can be interpreted as a strong pattern for re-casting eco-behavioural keystone species as cultural keystone species in the Swabian Aurignacian (for the term ‘cultural keystone species’, see e.g. Garibaldi & Turner 2004). At the same time, different trajectories of eco-behavioural significance are reflected in different patterns of sociocultural treatment, suggesting different albeit socially meaningful roles for both non-human animals in the ontological fabric of Swabian Aurignacian communities. Both species, in other words, play a crucial role in the ‘symbolic ecology’ of the time (cf. Descola 1992; Betts et al. 2015).

and focuses on the animal-human interface and its peculiarities in the Swabian Jura region during the Early Upper Palaeolithic about 40,000 years ago. We suggest that the structural preconditions of humananimal interactions emerging from the characteristics of mammoth steppe ecology and physiography as well as hunter-gatherer lifestyle modalities invoke proximity and intimacy as well as notions of co-inhabitation rather than an experiential disclosure from the animal kingdom as today. The comparison of mammoth (Mammuthus primigenius) and cave lion (Panthera spelaea) ecobehavioural characteristics and the ‘management’ of these two animals in the material culture of the Swabian Aurignacian supports this view, and demonstrates how naturally inherent interaction dynamics converge with sociocultural modalities of constructing the animalhuman interface at these two intersections.

The mammoth pre-occupies the visual space of the wider landscape, impacts and manipulates its surroundings in 53

Prehistoric Art as Prehistoric Culture profound ways, and thereby creates critical affordances for human foragers, ultimately unclosing a co-adaptive mammoth-human interface, and displays a set of behaviours evoking social similarity to human groups as well as autonomy, intentionality and individualised personality that helps to empathise and to develop socially significant relationships with them (Hussain & Breyer, in press). These structural characteristics of the human-mammoth intersection in the mammoth steppe environments of the last glacial are mirrored by the permeation of mammoth-related material culture (ivory beads, pendants and amulets) directly associated with the human body in the sociocultural substrate of the Swabian Aurignacian, thereby blurring the boundary between mammoth-ness and person-ness. The cognitive implications of living in a mammothscape in which human and mammoth worlds are inextricably interwoven are, moreover, indicated by the fact that mammoth remains are often a common feature of Aurignacian cave-sites, although there is little evidence for active and regular hunting of these animals, potentially suggesting a certain proscription to harm them. The significance of human-mammoth relationships, lastly, is reflected in the importance and individualistic representation of the mammoth in the figurative art of the Swabian Aurignacian and points to the central role of this species in the ‘belief world’ (Glaubenswelt) of these people.

These findings demonstrate that Aurignacian groups that inhabited the Swabian Jura about 40,000 years ago were closely attached and sensitive to the peculiarities of their surroundings and had clearly non-Cartesian, relational and deeply animistic ontologies. The modalities of constructing the mammoth-human and the lion-human interface in this spatio-temporal setting exemplifies the blurring of Cartesian boundaries and thus illustrates the necessity and fruitfulness of exploring the ontological bearing of Palaeolithic material culture across various interfaces at the human-world intersection. We hope that our results also demonstrate the ability of Palaeolithic archaeologists to contribute to the growing body of interdisciplinary research in Animal-Human Studies and the importance of extending this perspective into early stages of human evolution. Finally, two important perspectives emerge from the findings presented here. First, the relationship between eco-behavioural keystone species and cultural keystone species might be an interesting avenue for revealing some fundamental differences between regional socio-historical constellations in the Palaeolithic, for example between the Swabian Aurignacian and the Magdalenian of the Upper Danube region. Second, the argument that personal ornament and figurine making is mainly a reflection of certain modalities to construct the animal-human interface opens up new (and bold) perspectives on interpreting the Middle to Upper Palaeolithic transition. The emergence, proliferation and systematisation of personal ornaments and figurative art across the transition might then simply reflect the radical re-organisation of the sociocultural realm (cf. Floss & Hussain 2015) including underlying ontological fabrics and the ways people relate to the world.

The cave lion, on the other hand, is critically disconnected from human ‘taskscapes’ and, as an idiosyncratic and nocturnal hunter, emerges as the prime antagonist for human foragers in the eco-web of the Pleistocene mammoth steppe. As formidable and deadly stalkers that are almost elusive in the wider landscape, cave lions are predisposed to evoke deeply ambivalent attitudes. Their predatory behaviour elicits both respect and proximity to human hunters, that way displaying a certain human-likeness, and their status as ‘contester’ and true competitors that even prey on humans at times simultaneously provokes negative sentiments and tension. This ambivalent and highly problematic interaction condition manifests itself in an archaeological signature that highlights the transitory and liminal character of the cave lion and the integral need to police the human-lion interface. The presence of both individualised cave lion figurines and lion-human hybrids demonstrates the double status of these animals as well as the sociocultural necessity to negotiate and thematise the lion-human interface. This, in turn, indicates the fluidity, permeability and even arbitrariness of the human-lion divide. The liminality of cave lions, therefore, assigns them a salient place in Swabian Aurignacian worldviews that is well-reflected in the importance of the lion motif in the figurine repertoire, but in contrast to the salience of mammoths, cave lion prominence is rooted in deeply agonistic human-lion interaction dynamics.

Acknowledgements These reflections are dedicated to Rodrigo de Balbín Behrmann. We would like to thank Mimi Bueno and Paul Bahn for the opportunity to contribute to this volume and their eminent support in preparing the manuscript. STH acknowledges enduring financial support of the Studienstiftung des Deutschen Volkes since 2010. All remaining shortcomings and errors remain ours. References Barua, M. 2014. Bio-geo-graphy: landscape, dwelling, and the political ecology of human-elephant relations. Environment and Planning D: Society and Space 32: 915-34. Betts, M. W., Hardenberg, M. & Stirling, I. 2015. How animals create human history: relational ecology and the Dorset-polar bear connection. American Antiquity 80 (1): 89-112. 54

Hussain and Floss: Regional ontologies in the Early Upper Palaeolithic ed.) Archaeology of Oceania: Australia and the Pacific Islands. Blackwell: Malden. Descola, P. 1992. Societies of Nature and the Nature of Society, pp. 107-26 in (A. Kuper, ed.) Conceptualizing Society. Routledge: New York. Descola, P. 2011. Jenseits von Natur und Kultur. Suhrkamp: Berlin. Därmann, I. 2010. Theorien der Gabe – zur Einführung. Junius: Hamburg. Garibaldi, A. & Turner, N. 2004. Cultural keystone species: implications for ecological conservation and restoration. Ecology and Society 9 (3): 1. Gigerenzer, G. & Gaissmaier, W. 2011. Heuristic decision making. Annual Review in Psychology 2011: 451-82. Goldman, M. J., Roque De Pinho, J. & Perry, J. 2010. Maintaining complex relations with large cats: Maasai and lions in Kenya and Tanzania. Human Dimensions of Wildlife 15: 332-46. Gosden, C. 2008. Social ontologies. Philosophical Transactions of the Royal Society London B 363: 2003-10. Floss, H. 2007. Die Kleinkunst des Aurignacien auf der Schwäbischen Alb und ihre Stellung in der paläolithischen Kunst, pp. 295-316 in (H. Floss & N. Rouquerol, eds) Les Chemins de l’Art aurignacien en Europe – Das Aurignacien und die Anfänge der Kunst in Europa. Colloque international Aurignac, 16-18 septembre 2005. Éditions Musée-forum Aurignac: Aurignac. Floss, H. & Hussain, S. T. 2015. Substantiating the saltationist view of Aurignacian emergence in Central and Western Europe: a reassessment of qualitative and quantitative arguments. Paper presented at annual 57th meeting of the Hugo Obermaier society, 7.-11 April 2015, Heidenheim. Abstract in accompanying booklet, p. 33. Hahn, J. 1986. Kraft und Aggression. Die Botschaft der Eiszeitkunst im Aurignacian Süddeuschlands? Verlag Archaeologica Venatoria: Tübingen. Harris, M. 1968. The Rise of Anthropological Theory. A History of Theories of Culture. Harper & Row: New York. Haynes, G. 2006. Mammoth landscapes: good country for hunter-gatherers. Quaternary International 142143: 20-29. Haynes, G. 2013. Elephants (and extinct relatives) as earth-movers and ecosystem engineers. Geomorphology 157-158: 99-107. Heckel, C. E. & Wolf, S. 2014. Ivory debitage by fracture in the Aurignacian: experimental and archaeological examples. Journal of Archaeological Science 42: 1-14. Hill, E. 2012. The nonempirical past: enculturated landscapes and other-than-human persons in Southwest Alaska. Arctic Anthropology 49 (2): 4157.

Bird-David, N. 1999. ‘Animism’ revisited: personhood, environment, and relational epistemology. Current Anthropology 40: 67-91. Bird-David, N. 2006. Animistic epistemology: why do some hunter-gatherers not depict animals? Ethnos 71 (1): 33-50. Bird-David, N. & Naveh, D. 2008. Relational epistemology, immediacy, and conservation: Or, what do the Nayaka try to conserve? Journal for the Study of Religion: Nature & Culture 2 (1): 55-73. Bocherens, H. 2015. Isotope tracking of large carnivore palaeoecology in the mammoth steppe. Quaternary Science Reviews 117: 42-71. Bocherens, H. et al. 2011. Isotopic evidence for dietary ecology of cave lion (Panthera spelaea) in NorthWestern Europe: Prey choice, competition and implications for extinction. Quaternary International 245: 249-61. Bocherens, H., Drucker, D. & Wißing, C. 2014. Die Mammutsteppe: Isotopenuntersuchungen in einem vergangenen Ökosystem. Senckenberg: Natur – Forschung – Museum 144 (7/8): 226-30. Bradshaw, G. A. 2009. Elephants on the Edge: What Animals Teach Us about Humanity. Yale University Press: Nw Haven & London. Bradshaw, G. A., Schore, A. N., Brown, J. L., Poole, J. H. & Moss, C. J. 2005. Elephant breakdown. Social trauma: early disruption of attachment can affect the physiology, behaviour and culture of animals and humans over generations. Nature 433: 807. Bradshaw, J. 2013. Cat Sense: How the New Feline Science Can Make You a Better Friend to Your Pet. Basic Books: New York. Conard, N. J &, Münzel, S. C. 2004. Change and continuity in subsistence during the Middle and Upper Palaeolithic in the Ach Valley of Swabia (South-west Germany). International Journal of Osteoarchaeology 14: 225-43. Conard, N. J., Bolus, M., Dutkiewicz, E. & Wolf, S. 2015. Eiszeitarchäologie auf der Schwäbischen Alb: Die Fundstellen im Ach- und Lonetal und in ihrer Umgebung. Tübingen Publications in Prehistory. Kerns Verlag: Tübingen. Cottee-Jones, H. E. W. & Whittaker, R. J. 2012. The keystone species concept: a critical appraisal. Frontiers of Biogeography 4 (3): 117-27. Corbey, R. H. A. 2005. The Metaphysics of Apes: Negotiating the Animal-Human boundary. Cambridge University Press: Cambridge. Corbey, R. H. A. & Lanjouw, A. 2013. The Politics of Species: Reshaping our Relationships with Other Animals. Cambridge University Press: Cambridge. DeMello, M. 2012. Animals and Society. An Introduction to Human-Animal Studies. Columbia University Press: New York. David, B. 2006. Archaeology and the dreaming: towards an archaeology of ontology, pp. 46-68 in (I. Lilley,

55

Prehistoric Art as Prehistoric Culture Hill, E. 2013. Archaeology and animal persons. Toward a prehistory of human-animal relations. Environment and Society: Advances in Research 4: 117-36. Hofreiter, M. & Lister, A. 1999. Mammoths. Current Biology 9 (9): R347. Hofreiter, M. & Stewart, J. 2009. Ecological change, range fluctuations and population dynamics during the Pleistocene. Current Biology 19: R584-R594. Hussain, S. T. & Breyer, T. (in press) Menschwerdung, Verkörperung und Empathie: Perspektiven im Schnittfeld von Philosophie, Anthropologie und Paläolitharchäologie, in (G. Etzelmüller, ed.) Verkörperung. Eine neue interdisziplinäre Anthropologie. De Gruyter. Ingold, T. 2000. The Perception of the Environment: Essays on Livelihood, Perception, Dwelling and Skill. Routledge: London and New York. Kalof, L. 2007. Looking at Animals in Human History. Reaktion Books: London. Kind, C.-J., Ebinger-Rist, N., Wolf, S., Beutelsbacher, T. & Wehrberger, K. 2014. The Smile of the Lion Man. Recent Excavations in Stadel Cave (BadenWürttemberg, south-western Germany) and the Restauration of the Famous Upper Palaeolithic Figurine. Quartär 61: 129-45. Kohn, E. 2013. How Forests Think. Toward an Anthropology Beyond the Human. University of California Press: Berkeley. Kroeber, A. L. 1939. Cultural and Natural Areas of Native North America. University of California Press: Berkeley. Kahindi, O. 2001. Cultural perceptions of elephants by the Samburu people in northern Kenya. Unpublished Masters Dissertation, University of Strathclyde. Kuriyan, R. 2002. Linking local perceptions of elephants and conservation: Samburu pastoralists in Northern Kenya. Society and Natural Resources 15: 949-57. Laing, S. P. & Lindzey, F. G. 1991. Cougar habitat selection in south-central Utah, pp. 27-37 in (C. S. Braun, ed.) Proceedings of the Mountain LionHuman Interaction Symposium and Workshop. Colorado Division of Wildlife: Denver. Lévi-Strauss, C. 2007. The totemic illusion, pp. 26269 in (L. Kalof & A. Fitzgerald, eds) The Animals Reader: The Essential Classic and Contemporary Writings. Bloomsbury: London. Lichtenfeld, L. L. 2005. Our Shared Kingdom At Risk: Human-Lion Relationships in the 21st Century. Unpublished Ph.D. Dissertation, University of Yale. Locke, P. 2013. Explorations in ethnoelephantology. Social, historical, and ecological intersections between Asian elephants and humans. Environment and Society: Advances in Research 4: 79-97. MacDonald, D. W., Loveridge, A. J. & Nowell, K. 2010. Dramatis personae: an introduction to the wild felids, pp. 3-58 in (D. W. MacDonald & A. J. Loveridge, eds) Biology and Conservation of Wild Felids. Oxford University Press: Oxford and New York.

Malinowski, B. 1920. Kula: The circulating exchange of valuables in the Archipelagoes of Eastern New Guinea. Man 20: 97-105. Malinowski, B. 1922. Argonauts of the Western Pacific: An Account of Native Enterprise and Adventure in the Archipelagoes of Melanesian New Guinea. Routledge: London. Masson, J. F. & McCarthy, S. 1996. When Elephants Weep: The Emotional Lives of Animals. Delta Book: New York. Mithen, S. 1999. The hunter-gatherer prehistory of human-animal interactions. Antrozoos 12 (4): 195204. Moss, C. 2000. Elephant Memories: Thirteen Years in the Life of an Elephant Family. University of Chicago Press: Chicago. Münzel, S. C. 2001. Seasonal hunting of mammoth in the Ach-Valley of the Swabian Jura, pp. 318-22 in (G. Cavarretta, P. Gioia, M. Mussi & M. R. Palombo, eds) The World of Elephants. Proceedings of the 1st International Congress, Consiglio Nazionale delle Ricerche, Rome 16.-20. October. CNR: Rome. Niven, L. 2003. Patterns of subsistence and settlement during the Aurignacian of the Swabian Jura, Germany, pp. 199-211 in (O. Bar-Yosef & J. Zilhão, eds) Toward a Definition of the Aurignacian. Trabalhos de Arqueologia 45: Lisbon. Niven, L. 2006. The Palaeolithic Occupation of Vogelherd Cave. Implications for the subsistence behavior of late Neanderthals and Early Modern Humans. Tübingen Publications in Prehistory. Kerns Verlag: Tübingen. Niven, L. 2007. From carcass to cave: Large mammal exploitation during the Aurignacian at Vogelherd, Germany. Journal of Human Evolution 53: 362-82. Ogden, L. A., Hall, B. & Tanita, K. 2013. Animals, plants, people, and things. A review of multispecies ethnography. Environment and Society: Advances in Research 4: 5-24. Packer, C. 2010. Lions. Current Biology 20 (14): R590. Packer, C., Swanson, A., Ikanda, D. & Kushnir, H. 2011. Fear of darkness, the full moon, and the nocturnal ecology of African lions. PLoS ONE 6 (7): e22285. Porr, M. 2015. Beyond animality and humanity. Landscape, metaphor and identity in the Early Upper Palaeolithic of Central Europe, pp. 54-74 in (F. Coward, R. Hosfield, M. Pope & F. WenbanSmith, eds) Settlement, Society and Cognition in Human Evolution: Landscapes in Mind. Cambridge University Press: New York. Putshkov, P. V. 2003. The impact of mammoth on their biome: clash of two paradigms. Deinsea 9: 365-80. Redmond, I. 1982. Salt mining elephants of Mount Elgon. Swara 5: 28-31. Reed, E. S. 1988. The affordances of the animate environment: social science from the ecological point of view, pp. 110-26 in (T. Ingold, ed.) What is an Animal? Routledge: London and New York. 56

Hussain and Floss: Regional ontologies in the Early Upper Palaeolithic Robb, J. & Harris, O. J. T. 2013. The Body in History. Europe from the Palaeolithic to the Future. Cambridge University Press: Cambridge. Santos-Granero, F. 2012. Beinghood and people-making in native Amazonia. HAU: Journal of Ethnographic Theory 2 (1): 181-211. Schelling, T. C. 1981. A Strategy of Conflict. Harvard University Press: Harvard. Seidenstricker, J., Hornocker, M. G., Wiles, W. V. & Messick, J. P. 1973. Mountain lion social organization in the Idaho Primitive Area. Wildlife Monograph 35. Wiley-Blackwell: New York. Shipman, P. 2010. The animal connection and human evolution. Current Anthropology 51 (4): 519-38. Steguweit, L. 2015. Rotten ivory as raw material source in European Upper Paleolithic. Quaternary International 361: 313-18. Steward, J. H. 1955. Theory of Culture Change: the Methodology of Multilinear Evolution. University of Illinois Press: Urbana. Stuart, A. J. & Lister, A. M. 2011. Extinction chronology of the cave lion Panthera spelaea. Quaternary Science Reviews 30 (17-18): 2329-40. Sukumar, R. 2003. The Living Elephants: Evolutionary Ecology, Behavior, and Conservation. Oxford University Press: Oxford. Vanhaeren, M. & d’Errico, F. 2006. Aurignacian ethnolinguistic geography revealed by personal ornaments. Journal of Archaeological Science 33 (8): 1105-28. Von Königswald, W. 2004. Klima und Tierwelt im Eiszeitalter Mitteleuropas: Das Quartär. Biologie unserer Zeit 34 (3): 151-58. Whatmore, S. 2006. Materialist returns: practising cultural geography in and for a more-than-human world. Cultural Geographies 13: 600-09. Wittemyer, G., Douglas-Hamilton, I. & Getz, W. M. 2005. The socioecology of elephants: analysis of the processes creating multitiered social structures. Animal Behaviour 69: 1357-71. Willerslev, R. 2007. Soul Hunters: Hunting, Animism, and Personhood among the Siberian Yukaghirs. California University Press: Berkeley. Wolf, S. 2015. Schmuckstücke: Die Elfenbeinbearbeitung im Schwäbischen Aurignacien. Tübinger Monographien zur Urgeschichte. Kerns Verlag: Tübingen. Yaekel, J. D., Guimarães Jr., P. R., Bocherens, H. & Koch, P. L. 2013. The impact of climate change on the structure of Pleistocene food webs across the mammoth steppe. Proceedings of the Royal Society of London B 280: 20130239. Zimov, S. A., Zimov, N. S., Tikhonov, A. N. & Chapin III, F. S. 2012. Mammoth steppe: a high productivity phenomenon. Quaternary Science Reviews 57: 2645.

57

58

Aurignacian art in the caves and rock-shelters of Aquitaine (France) Brigitte and Gilles Delluc1 Ph.D.s in Prehistory, Associate researchers in the Department of Prehistory National Museum of Natural History, Paris.UMR 7194 du CNRS [email protected]

1

Abstract Archaic remains of artistic activities in the rock-shelters and caves of Aquitaine (France) were discovered a century ago, but the conditions of discovery were not sufficiently precise. New research in the abri Blanchard-abri Castanet has shown that this site was an exceptional place in the emergence of figurative art more than 36 000 years ago (uncalibrated), and this has revived interest in other decorated Aurignacian rock-shelters and caves in Aquitaine. Keywords Palaeolithic art; Aurignacian; Aquitaine; parietal; rock engraving; painting; sculpture

Once ‘Modern Man’s’ Aurignacian culture emerged in Europe, all the graphic expressions of engraving, painting and sculpture appeared. For a long time, prehistorians only knew some fragments of rock-shelter walls destroyed by severe frost breakage (cryoclastic erosion) during Würm III, with only bits of engraved or painted lines surviving. They considered that this was the emergence of a rudimentary art.

Gourhan’s Style I, which he considered Aurignacian, was seen as too narrow and largely outdated: Style I was limited to some stone blocks decorated with vulvas and fragmentary animals, and neglected the small decorated caves (Leroi-Gourhan 1965: 242). In the Dordogne, six rock-shelters, now collapsed, conserve remains of rock paintings and engravings attributable to the Aurignacian period. They are all located around Les Eyzies, in the Vézère valley or along its tributaries: abri Blanchard, Castanet, Lartet, Le Poisson, La Ferrassie, Pataud. However, Aurignacian rock paintings and engravings are not very spectacular. Only one fragment of the abri Blanchard’s vault conserves a large part of a polychrome painting (black lines with a red background): a portion of a big herbivore with its abdomen and limbs.

Our works Les manifestations graphiques aurignaciennes sur support rocheux des environs des Eyzies (Delluc 1978) and L’art pariétal archaïque en Aquitaine (Delluc 1991) highlighted the fact that Aurignacian parietal art on blocks and walls was a reality in Aquitaine, southwest France. Recent research has confirmed those conclusions. The discovery of the exceptional Chauvet cave (Ardèche), with its well preserved wealth of paintings and engravings in a deep cave – an outlier from the geographical point of view – has produced spectacular evidence for early Upper Palaeolithic art. Even though scientific research and discussions continue, trying to establish exactly when the decoration of this cave took place (Combier & Jouve 2014, and all the articles in L’Anthropologie, vol. 118, n° 2, 2014), everybody acknowledges its antiquity. Miraculously preserved under the earth, this decorated cave provides remarkable proof of what the art of the first Modern Men could be.

Moreover, Aurignacian parietal art is not exclusively French, as was thought for a long time. The first proof of this was the discovery in Spain of the entrances of the caves of La Vina (Fortea Perez 1990/1992) and El Conde (Fortea Perez 2002: 15; Barandiarán and García Diez 2007: 216): they feature Aurignacian engraved parietal lines, without figurative representations. Then fragments of a collapsed rock-shelter, decorated with Aurignacian paintings, were discovered at Fumane (PreAlps of Venetia, Italy): they contain unquestionable remains of red backgrounds and large red lines. A figurative interpretation of some of these remains has been proposed: e.g. quadruped, anthropomorph (Broglio et al. 2007; Broglio 2009).

That discovery made it possible for the remains of paintings, engravings and sculptures in the Aurignacian rock-shelters of the Les Eyzies region or the engraved decoration on complex panels in the caves of Pair-nonPair (Gironde), La Cavaille, Les Bernous and La Croze à Gontran (Dordogne) to be taken into consideration. Thus, even before the discovery of Chauvet cave, André Leroi-

It is a fact that today we know only a small proportion of Aurignacian productions on blocks and walls in Europe. The aim of this communication is to present an inventory of what is known today in Aquitaine. 59

Prehistoric Art as Prehistoric Culture 1. Aurignacian art on broken pieces from collapsed rock-shelters and from decorated blocks extracted from these rock-shelters, i.e. ‘Aurignacian openair rock art’. 2. Aurignacian art in decorated caves, i.e. underground ‘parietal Aurignacian art’. I Aurignacian open-air rock art in Aquitaine At present this art is only known in the Les Eyzies region, Dordogne. 1 The abri Blanchard (Sergeac, Dordogne) It is located on the right bank of the stream of Les Roches, a small tributary of the Vézère. It was a dwelling for people at the beginning of the Upper Palaeolithic. Its vault has been destroyed by cryoclastic erosion. Its sediments were excavated in 1910 by Marcel Castanet, directed by Louis Didon, and the results were published the following year (Didon 1911). According to Louis Didon, the site contained two very rich Aurignacian archaeological levels, and its ceiling collapsed at the end of this period. This site has generated a very rich lithic and bone industry, as well as numerous decorative objects. A bovid ankle bone carved into a phallus was discovered at the base of the lower level, next to a fireplace – the oldest figurative object in France! Figure 1. Abri Blanchard. 3 vulvas engraved on a block. Photo: Delluc

In addition, around 15 stone blocks of a very large size (between 100 x 50 x 50 cm and 26 x 25 x 25 cm) are decorated with deep engravings, with outlines of bas-relief sculptures. Their decoration, often well preserved, is very elaborate and rich from a symbolic and thematic point of view. One can see, either isolated or in association:

by C14 at 32 400 + 480 BP’ (White et al. 2014: 63-64). The study of this site’s bone industry in the USA has shown the massive importance of the tools of the early Aurignacian. It has also been possible to find tools from the evolved Aurignacian: they come from the upper level of abri Blanchard (Tartar et al. 2014).

–– 3 animals (heads of horse, ibex and bear) –– 6 representations of complete and very detailed vulvas (fig. 1) –– 9 ogival, hoop and oval lines –– 1 phallus –– 11 exceptional images of engraved bear’s paws. –– 2 groups of cupules organised concentrically (Delluc 1978: 221-61).

In addition, in the middle of the lower level, the excavations unearthed ‘a huge piece of calcareous stone; the bottom of the stone exhibited a splendid engraving of aurochs with several cupule lines’ (White et al.2014: 63-64).

New excavations, directed by Randall White in 2011 and 2012, have mad it possible to confirm the Aurignacian date of this exceptional site. They have brought to light important data about the site’s stratigraphy and the decorated blocks discovered in the previous excavations. Some intact fragments of the archaeological levels, under Mr. Castanet’s spoilheap remained in the northern part of the site. A portion of the upper level was found the first year: it is considered to be an evolved Aurignacian. In 2012, the excavations focused on a portion of 8 sq m of the lower level, which was very homogeneous and free of contamination, in contact with the bedrock: the tools discovered are those of the early Aurignacian, ‘practically identical to those of the Castanet-sud, dated

Until recently, by referring only to the documents found in the former excavations, it was not possible to go beyond an attribution of the blocks to Aurignacian, without precise dating. Today we can confirm that at least some of the blocks from Blanchard come from the lower level , the early Aurignacian. In this site the first excavations of Marcel Castanet and Louis Didon discovered the remains of an elaborate decoration of the rock-shelter’s ceiling: a large fragment of it, in two pieces (respectively 128 x 70 x 28 cm and 38 x 28 x 22 cm) displayed black lines on a red background, representing two pairs of rigid limbs next to each other, without perspective, and terminated by round hoofs, and 60

Delluc and Delluc: Aurignacian art in the caves and rock-shelters of Aquitaine (France)

Figure 2. Abri Blanchard. Portion of a horse, with the belly and limbs, painted on a fragment of the rock-shelter’s vault. Photo: Delluc

the very convex belly line of a horse. These two blocks had fallen, next to each other, with the decorated part lying on the surface of the upper level (Fig. 2).

four vulvas (two of them with double lines), a phallus (next to a ring), and one image representing a bear pawprint (Delluc 1978: 261-77).

Spalling of the ceiling started slowly during the first occupations; it continued, and was over by the end of the Aurignacian. Therefore the rock-shelter was certainly painted during the second period of occupation. It cannot be ruled out that decoration started at the beginning of the Aurignacian (Delluc 1978: 221-61; 1991: 121-30).

In addition, the rock-shelter’s ceiling – or at least part of it – was decorated with paintings. The first excavations unearthed some painted stones. On one (45 x 32 x 10 cm), a black piriform line, interrupted by the fracture of the block, and two parallel red lines, were interpreted by André Leroi-Gourhan as the remains of a schematic vulva image associated with a sign made of two parallel lines. These stones, which resulted from spalling, confirm, as in Blanchard, the very early painting of the rock-shelter’s ceiling (Delluc 1978: 261-77; 1991: 131-34).

2 The abri Castanet (Sergeac, Dordogne) Located right next to the abri Blanchard, this seems to be its prolongation. It too had collapsed completely. For the first two excavators (Marcel Castanet and Denis Peyrony), the two stratigraphies are similar (Peyrony 1935). The checks, carried out recently by a team of excavators directed by Randall White, confirmed that the ‘the archaeological level (Early Aurignacian), in contact with the bedrock, spreads from the northern part to the southern part of abri Castanet, sealed by the collapse of the rock-shelter’s ceiling’ (White et al. 2014). However, the upper Aurignacian level in the abri Castanet could not be found: ‘the second level doesn’t exist’ in the southern portion of the site, where the new excavations too place (White et al. 2011: 51).

The current excavations, undertaken to understand the rock-shelter’s evolution, have demonstrated that its vault collapsed directly onto the archaeological layers, and led to the discovery of an engraved drawing attributed to the Early Aurignacian. It is on a huge block (K) (unfortunately cut into several pieces during its extraction), decorated on its lower face. The reassembling of the fragments revealed: ‘an embossed oval shape, an underlying broken ring, other suggestive embossed shapes […] and traces of red colour’ (White 2009: 60-61; Mensan et al. 2012). The oval shape has been identified as a vulva. Six bone fragments (one of which was found stuck to the decorated surface of block K) have been dated by C14. ‘Perfectly consistent between them, these dates give an average age of 32 400 BP’. The northern and southern sites have provided around 20 C14 dates, corresponding to a relatively short period of occupation, with dates using ‘Bayesian’ modelling of between 36 940 and 36 510’ (Mensan et al. 2012: 179-80).

In the archaeological level, among the inhabitants’ remains were found a huge block of limestone (80 x 57 x 28 cm), with two rings drilled into it, and with two series of small cupules (in the lower level of the Early Aurignacian); and above all, four stone blocks (between 75 x 35 x 14 cm and 25 x 30 x 11 cm) with deep engravings, essentially recti-curvilinear, representing 61

Prehistoric Art as Prehistoric Culture shelter at the beginning of the Aurignacian: one can see two deeply engraved ibex horns and several images of engraved vulvas. Painted plaquettes are the result of spalling, and some were discovered in all Aurignacian levels: four of them have been reassembled together (54 x 30 x 17.5 cm); three others have also been fitted together (34 x 28 x 6 cm). The rock-shelter’s ceiling was not only carved and engraved, but painted as well.

3 The big rock-shelter of La Ferrassie (Savignac-deMiremont, Dordogne) The westernmost of the Aurignacian sites around Les Eyzies, it was excavated by Denis Peyrony during the early part of the 20th century (Peyrony 1934). Its study was recommenced in 1968, under the leadership of Henri Delporte (Delporte 1969). Its stratigraphy, with different Mousterian levels at the base, followed by 14 Aurignacian and 3 middle Gravettian levels, is used as a reference for this period.

4 The abri Cellier (Tursac, close to Le Moustier, Dordogne) It contained an Aurignacian level (Aurignacian I and doubtless a more evolved Aurignacian) and a Middle Gravettian level. Five stone blocks (between 74 x 31 x 14 cm and 36 x 25 x 8 cm) come from the Aurignacian level. They were found during G. L. Collie’s excavations: their decorated face was against the earth, and showed deeply engraved lines representing animals, but limited to the depiction of the head (1 horse and 2 Alpine ibex with curved horns), 6 perineal vulvas, 2 large cupules and 3 series of parallel lines (Peyrony 1946).

Among the everyday remains, this site has also yielded around 20 decorated blocks, perfectly dated to the different periods of the Aurignacian. These blocks (between 57 x 45 x 24 cm and 17.5 x 12.5 x 8.5 cm) are decorated with deeply engraved lines, with some bas-relief outlines. Most of the blocks are very well preserved. Their decorations are complex and varied: fragmentary animals (a body and limbs of an herbivore, a rhinoceros head), 10 vulvas (Fig. 3), 3 images of curved lines, a badge, a phallus, two series of aligned dots and two groups of dots (Delluc 1978: 277-325).

Two huge pyramidal shaped stone blocks (respectively 60 x 60 x 45 cm for one and 75 x 67 x 50 cm for the second one) have been discovered at the upper limit of the Aurignacian level and decorated at this time. Their bases were deeply rooted in this layer and their upper face deeply engraved with very realistic representations of vulvas (the medial fissure illustrated by a very deep incision). One of these two blocks has five vulvas, one

In addition, a huge fragment of the rock-shelter’s ceiling (broken into pieces of different dimensions and today reassembled [99 x 50 cm]) came from the Aurignacian II level. It has deeply engraved and carved drawings, sometimes interrupted by the breakage of the rock which prove the complex parietal decoration of the rock-

Figure 3. La Ferrassie. 2 vulvas engraved on a block. Photo: Delluc

62

Delluc and Delluc: Aurignacian art in the caves and rock-shelters of Aquitaine (France) of which is partly engraved (bas-relief technique); the second block has the remains of two other vulvas of the same type (Delluc, 1978).

fallen from the vault. The rock-shelter’s ceiling, today completely destroyed, was decorated in many different periods – starting at the beginning of the Aurignacian – with paintings, engravings and sculptures (Delluc 2004).

5 The abri du Poisson (small valley of Gorge d’Enfer, Les Eyzies, Dordogne)

The study of these fragments was recommenced recently (Chiotti et al. 2007), particularly those of the Aurignacian period. One fragment (c. 25 x 14 cm), covered with red pigment, with less visible traces of black pigment, comes from level 10/11. It corresponds to a decoration of the rock-shelter’s vault in Aurignacian I. Nine fragments, of different dimensions (between 15 x 8 cm and 2 x 2 cm), covered with red pigment, rather uniform, correspond to the remains of a painting dating to the end of the Aurignacian: the decoration probably lasted through several occupations of the rock-shelter because, in level 8, there are abundant remains of red pigment at the base of the rock-shelter, and it seems that the destruction of this decoration occurred during the formation of level 6.

The shelter had already been excavated at the end of the 19th century, but a sufficient portion of the archaeological site remained to permit Denis Peyrony to undertake its study in 1917-1918: the stratigraphy includes an Aurignacian level surmounted by a Middle Gravettian level (Peyrony 1932). The parietal decoration, still in place today on the rock-shelter’s ceiling, dates back to the latter period (a bas-relief salmon, the remains of a large red tempera, some finely engraved lines and a negative hand surrounded with black paint). In fact, the ceiling was also decorated at the time of the Aurignacians, but that decoration was destroyed by cryoclastic erosion. All that is left are about ten thin plaquettes of about ten centimetres per side each. They retain traces of red pigment with, sometimes, previous engraved lines which are fragmented and unreadable. A large, thick fragment of the collapsed ceiling (about 150 x 120 cm), conserved in situ, displays part of the engraved decoration: the remains of an engraved vulva, interrupted by the fracture of the rock (Delluc 1991: 212-25).

8 The abri Belcayre (Thonac, Dordogne) This rock-shelter has not yielded remains of Aurignacian rock art. But excavations, which started at the beginning of the 20th century, led to the discovery in this exclusively Aurignacian site of a a large flat limestone plaque engraved with a complete figure of an ibex, with short horns,which remarkable from a technological point of view (Delage 1935). The outline of the animal is made with a broad dotted line with some highly regulated segments. The animal’s two horns encroach onto a lateral surface. The representation of the animal’s back is particularly detailed (Delluc 1978: 325-32).

The site also yielded two engraved limestone blocks: a perineal vulva, and an engraved image of a bear’s paw. These blocks probably come from the Aurignacian level. However, the conditions of discovery were such that it is not possible to exclude an attribution to the Gravettian level (Delluc 1978: 377-81).

II Parietal art of the underground Aurignacian world in Aquitaine

6 The abri Lartet (small valley of Gorge d’Enfer, Les Eyzies, Dordogne)

A Dordogne

Upstream of the abri du Poisson and located just next to it, it is today completely filled up. Like Le Poisson, it was explored by different excavators in the 19th century. D. Peyrony recommenced the study in 1918 (Peyrony 1932). He described remains of Châtelperronian and Aurignacian I, and a large plaquette of parietal origin, bearing the remains of painted decoration with traces of red and black pigments without any discernible layout (Delluc 1978; 1991: 166-67).

1 The cave of La Cavaille (Couze-et-Saint-Front, Dordogne) It is a small horizontal cavity, about 20 m long. Its entrance was illuminated by daylight during prehistoric times.This entrance, the first part of the cave, is decorated with engravings that we discovered in 1988 (Delluc 1988). On the left, the decoration starts with a big bovid below a big mammoth; then comes a series of 3 mammoths, with a horse inserted between them (Fig. 4). On the right, the panel is dominated by an enormous, schematised and complex vulva image. It is completed with a group of marginal small animals (doubtless another mammoth, and an acephalous animal), framed by two panels covered with thin parallel lines. Some archaic stylistic details of the animals’ shapes deserve attention: the big bovid’s horns are facing the front; the animals are limited to outlines; the mammoths’ bellies are arched, a classical representation for the

7 The abri Pataud (Les Eyzies, Dordogne) It is a reference site for the beginning of the Upper Palaeolithic (early and evolved Aurignacian, early, middle and evolved Gravettian, Solutrean). It was excavated between 1953 and 1964 under the directorship of Hallam Movius (Movius 1977). Its ceiling quickly collapsed during Würm III. The everyday remains from different periods were regularly covered by fragments

63

Prehistoric Art as Prehistoric Culture

Figure 4. The cave of La Cavaille: a, the engraved decoration on the left wall (after B. and G. Delluc) ; b, the right mammoth (photo: Delluc)

64

Delluc and Delluc: Aurignacian art in the caves and rock-shelters of Aquitaine (France) beginning of the Upper Palaeolithic. The decoration of La Cavaille, despite having deteriorated through time, is very important because it can be attributed to the beginning of the Upper Palaeolithic. In fact, the previous excavations had only found Mousterian, Châtelperronian and Magdalenian (Delluc 1991: 11017). And the engravings of La Cavaille definitely do not correspond to the Magdalenian.

B Gironde The cave of Pair-non-Pair (Prignac-et-Marcamps, the valley of Girondine Dordogne) The present cave is only about 20 metres long, its first part having collapsed (Cheynier & Breuil 1963). At the beginning of the 1980s, we recommenced the study of its decoration (Delluc 1991: 55-110). In prehistoric times the decorated zone, located at the crossroads of galleries, was illuminated by daylight coming trough a hole in the ceiling. We have recorded 36 animals, complete and incomplete. They are deeply engraved in a soft fossiliferous limestone. In one case, only the belly of the horse has been put in evidence with the original technique of bas-relief.

2 The cave of La Croze à Gontran (Les Eyzies, Dordogne) It is a small gallery of about 50 m that penetrates a limestone massif dominating the Vézère river. The walls and the ceiling in the half-light zone, and especially in the dark zone of this dry narrow meander, are decorated with fine engravings of animals, both complete or incomplete (mammoths, horse, bison, ibex), in three panels separated by series of parallel lines. The outline of the horse is very characteristic : its cephalic extremity is long, the end of the nose turns down, the mane is outlined as a crest; the belly is bloated; the outline is simple, without details, without animation, without any consideration of perspective (Delluc 1991: 35-55).

We can observe, among the complete figures: –– 8 ibex, including 7 Alpine, –– 5 horses, one of which, gigantic, measures more than 2 m long and touches the ceiling, –– 3 mammoths, –– 2 stags and a megaloceros,

This engraved decoration, difficult to read and poorly preserved, is most probably Aurignacian. At the beginning of the 20th century, the cave’s floor yielded tools identified by Henri Breuil as dating back to the ‘early Aurignacian’ (Capitan et al. 1914).

among the incomplete figures: –– ibex heads, –– mammoths and bovids, –– a couple of animals’ bodies from unidentified species.

3 The cave of Les Bernous (Bourdeilles, Dordogne)

Some details are visible : the muzzle and mane of the horses, the megaloceros’s hump, the stags’ throat, the ibexes’ tails, two juxtaposed legs terminated by a cloven hoof. Three animals are particularly lively: two horses turn their head like an Agnus dei, a mammoth’s tusk is curved. Perspective is generally non-existent: the outlines are rigid; the bovids’ horns face the front, and those of the ibex are represented on top of each other.

This cave is no more than about 20 m long. A wall at the entrance, on the left, and today eroded, had been decorated with a deeply engraved series of animals: a mammoth, a rhinoceros and a bear (Peyrony 1932: 5-9). There are no details, except a cervical line for the two last animals and the horn of the rhinoceros. The first animal is an engraved mammoth, making use of the natural rounded contours of the rock-wall for the back. The surface with the rhinoceros’ head seems to have been regularised and levelled (Delluc 1991: 2735).

These engravings are displayed in four panels, under the shaft of light, more or less limited by the natural relief of the rock. They are all laid out horizontally, juxtaposed in a frieze, or one on top of another, or overlapping completely or partly, with animals face to face or head to foot, which creates an impression of a complex organisation with some symmetrical layout (Fig. 5). One horse seems to follow another. Finally there are two stone rings, one of which is incomplet next to the ibex, but no geometric symbols, contrary to the erroneous claims of Henri Breuil, who had made the first study of the site (Cheynier & Breuil 1963).

A new study of the cave of Les Bernous has brought new data to light. There seems to be a second mammoth, back to back with the first, though its forequarters are difficult to read. Traces of red pigment have been identified on the two walls (Petrognani et al. 2014). On the floor, Denis Peyrony found only some pieces of flint: according to him, some of them had a Mousterian appearance, whereas others seemed to be attributable to Aurignacian. It is highly probable that the decoration of this small cave of the Dronne valley dates back to this period (Delluc 1991: 27-35). A new study of these tools confirm this conclusion (Petrognani et al. 2014).

These exceptional and very well preserved decorated panels, discovered in 1896 by François Daleau, were buried under Aurignacian levels, surmounted by levels of the Middle Gravettian, in this cave which was practically

65

Prehistoric Art as Prehistoric Culture

Figure 5. The cave of Pair-non-Pair: a, the engraved decoration of the first panel (after B. and G. Delluc); b, the left horse (photo: Delluc)

66

Delluc and Delluc: Aurignacian art in the caves and rock-shelters of Aquitaine (France) filled-up with archaeological deposits at the time of its discovery. A recent multidisciplinary study concluded that these engravings date back to the Aurignacian (Lenoir et al. 2006). But for a long time, these very elaborate engravings were attributed to the Gravettian. This ambiguity explains the confusion of styles I and II of André Leroi-Gourhan (Leroi-Gourhan 1971: 246).

The revival of studies concerning the Aurignacian rockshelters in Dordogne (especially Blanchard and Castanet) have made it possible to confirm that these places were the theatre of the birth of Homo sapiens figurative art and that, right from its origins, this art already exhibited a technology and a symbolic conceptual capacity that were very modern.

Conclusion From the beginning, Aurignacian art on blocks and cave walls was very diversified (painting, engraving and sculpture). It was identified at the very beginning of the 20th century, in the rock-shelters of south-west France, particularly in the area of Les Eyzies, and more recently in Spain and Italy as well. The Aurignacians also decorated some caves in the same area, that of Pair-nonPair in Gironde being the most important. In Aquitaine rock art, sexual representations are particularly numerous and explicit (vulvas and phalluses, isolated or grouped). The animal representations form part of the classical inventory of animals in Upper Palaeolithic art, but with specific elements such as the abundance of mammoths, with arched bellies.

Acknowledgements We thank Christine Rachèle Guidi for her translation of this text into English. References Barandiaran, I. & Garcia Diez, M. 2007.  Les débuts du graphisme paléolithique dans le nord de la péninsule ibérique, pp. 209-22 in (H. Floss & N. Rouquerol, eds) Les chemins de l’art aurignacien en Europe . Editions Musée-forum Aurignac, Cahier 4. Broglio, A. 2009. L’art aurignacien dans la décoration de la grotte de Fumane. L’Anthropologie 113: 753-61. Broglio, A., Giachi, G., Gurioli, F. & Pallecchi, P. 2007. Les peintures aurignaciennes de la grotte de Fumane (Italie), pp. 157-70 in (H. Floss & N. Rouquerol, eds) Les chemins de l’art aurignacien en Europe . Editions Musée-forum Aurignac, Cahier 4. Capitan, L., Breuil, H. & Peyrony, D. 1914. La Croze à Gontran (Tayac), grotte à dessins aurignaciens. Revue anthropologique XXIV (7-8): 277-80, 4 fig. Cheynier, A. & Breuil, H. 1963. La caverne de Pairnon-Pair, Gironde. Fouilles de François Daleau. Documents et industries, Société archéologique de Bordeaux, Documents d’Aquitaine, III, 236 pp., 70 fig., 23 pl. Chiotti, L., Delluc, B. & Delluc, G. 2007. Art et parure aurignaciens de l’abri Pataud (Les Eyzies-de-Tayac, Dordogne, France), dans le contexte aurignacien du Périgord, pp. 171-86 in (H. Floss & N. Rouquerol, eds) Les chemins de l’art aurignacien en Europe . Editions Musée-forum Aurignac, Cahier 4. Clottes, J. (ed.) 2001. La Grotte Chauvet. L’art des origines. Editions du Seuil: Paris. Collective work (edited by J.-M. Geneste) 2005. Recherches pluridisciplinaires dans la grotte Chauvet. Journées SPF, Lyon, 11-12 octobre 2003, Société préhistorique française, Travaux 6. Combier, J. & Jouve, G. 2014. Nouvelles recherches sur l’identité culturelle et stylistique de la grotte Chauvet et sur sa datation par la méthode du C14. L’Anthropologie. 118 (2): 115-51, 12 fig. Delage, F. 1935. Gravure aurignacienne de Belcayre (Dordogne), pp. 388-92 in Congrès préhistorique de FrancePérigueux 1934. Delluc, B. & Delluc, G. 1978. Les manifestations graphiques aurignaciennes sur support rocheux des

Aurignacian decorated objects are known, especially in Baden-Württemberg (Germany), which represent humans and animals. In France, the most spectacular object we have that was discovered next to a fire-place dating back to the early Aurignacian, is the phallus of Blanchard, engraved on a bovid ankle bone. The art of Chauvet Cave (Ardèche) is an exceptional case. It displays in one place all the graphic specificities of arts before Lascaux: Aurignacian, Gravettian and Solutrean. For the time being, we have C14 dates for some painted lay-outs: they date back to the Aurignacian in the opinion of some and to the Gravettian in the opinion of others (Combier & Jouve 2014; and all the articles in L’Anthropologie, vol. 118, n° 2, 2014). The multidisciplinary studies carried out in this cave are starting to shed light not only on the art on its walls, but also on the whole environment (Clottes 2001; Collective. 2005). Since excavations are forbidden, we impatiently await the results of the study of the artists’ culture and of the Palaeolithic visitors to this cave, which will enable us to really define the culture(s) of these people. The beginning of the Upper Palaeolithic occurred in a period of particularly tough climatic conditions. It is therefore true that we know only a tiny part of the artistic production of this period. The walls of the rockshelters were mostly destroyed by cryoclastic erosion. It is a miracle that any remains could be recovered. The discovery of decorated caves always depends on luck. It is highly probable that many of them are still sleeping under thick layers of sediment.

67

Prehistoric Art as Prehistoric Culture environs des Eyzies (Dordogne). Gallia Préhistoire 21: 213-438. Delluc, B. & Delluc, G. 1988. Les gravures de la grotte de la Cavaille à Couze (Couze-et-Saint-Front, Dordogne). Bulletin de la Société historique et archéologique du Périgord 115: 111-23, 9 fig. Delluc, B. & Delluc, G. 1991. L’Art pariétal archaïque en Aquitaine. XXVIIIe suppl. à Gallia Préhistoire. Editions du CNRS: Paris. Delluc, B. & Delluc, G. 2004. L’art à l’abri Pataud (Les Eyzies, Dordogne), pp. 87-94 in (M. Lejeune & A-C. Welté, eds) L’art du Paléolithique supérieur. Actes du congrès de l’UISPP à Liège 2001. ERAUL 107, Liège. Delporte, H. 1969. Les fouilles du Musée des Antiquités Nationales à la Ferrassie. Antiquités Nationales 1: 15-28, 2 fig. Didon, L. 1911. L’abri Blanchard des Roches (commune de Sergeac). Gisement aurignacien moyen. Bulletin de la Société historique et archéologique du Périgord 38: 246-61 & 321-45, 8 fig., 8 pl. Fortea Perez, F. J. 1990/1992. Abrigo de la Viña. Informe de las campanas 1980-1986 y 1987-1990. Excavaciones arqueologicas en Asturias. Oviedo Fortea Pérez, F. J. 2002. Trente neuf dates C14-SMA pour l’art pariétal paléolithique des Asturies. Préhistoire. Art et Sociétés LVII: 7-28. L’Anthropologie, vol. 118, n° 2, 2014, devoted to discussions of the dating of Chauvet Cave, with articles by about twenty specialists. Lenoir, M., Roussot, A., Delluc, B., Delluc, G., Martinez, M., Loiseau, S. & Mémoire, N. 2006. La grotte de Pair-non-Pair à Prignac-et-Marcamps (Gironde). Editions de la Société Archéologique de Bordeaux (Collection Mémoires, volume 5). Leroi-Gourhan, A. 1965. (2nd ed. in 1971). Préhistoire de l’Art Occidental. Editions Mazenod: Paris (revised and augmented in 1991 by B. & G. Delluc). Mensan, R., Bourrillon, R., Crétin, C., White, R., Gardère, P., Chiotti, L., Sisk, M., Clark, A., Higham, T. & Tartar, E. 2012. Une nouvelle découverte d’art pariétal aurignacien in situ à l’abri Castanet (Dordogne, France) : contexte et datation. Paléo, pp. 171-87, 8 fig. Movius, H. L. Jr. 1977. Excavation of the Abri Pataud, Les Eyzies (Dordogne). Stratigraphy. Bulletin of the American School of Prehistoric Research, 31. Peabody Museum, Harvard University. 165 pp., ill., plans. Petrognani, S., Robert, E., Boche, E., Cailhol, D., Lucas, C. & Lesvignes, E. 2014. Bourdeilles. La grotte ornée des Bernous, pp. 18-19 in DRAC d’Aquitaine, Bilan scientifique de la région Aquitaine 2012, Ministère de la Culture et de la communication. Peyrony, D. 1924. La Ferrassie. Préhistoire 3: 1-92, 89 Peyrony, D. 1932. Les gisements préhistoriques de Bourdeilles (Dordogne). Archives de l’IPH, mémoire n° 10, 95 pp., 60 fig., 11 pl.

Peyrony, D. 1935. Le gisement Castanet, Vallon de Castelmerle (commune de Sergeac), Aurignacien I et II. Bulletin de la Société historique et archéologique du Périgord 32: 418-43, 22 fig. Peyrony, D. 1946. Le gisement préhistorique de l’abri Cellier, au Ruth, commune de Tursac (Dordogne). Gallia Préhistoire 4: 294-301, 6 fig. Peyrony, D. 1932. Les abris Lartet et du Poisson. L’Anthropologie 42: 241-68, 11 fig. Tartar, E., White, R., Chiotti, L., Cretin, C. & Mensan, R. 2014. Quel(s) Aurignacien(s) à l’abri Blanchard (Sergeac, Dordogne, France). Données des collections d’industrie osseuse conservées aux EtatsUnis et retour sur le terrain. Paléo n° 25: 309-28, 7 fig., 4 tables. White, R. 2009. Sergeac. Abri Castanet, pp. 60-61 in DRAC d’Aquitaine, Bilan scientifique de la région Aquitaine 2007, Ministère de la Culture et de la communication. White, R., Mensan, R., Sisk, M. & Clark, A. 2011. Abri Blanchard (Aurignacien ancien et récent) et Abri Castanet, secteur central (Aurignacien ancien), pp. 51-53 in DRAC d’Aquitaine, Bilan scientifique de la région Aquitaine 2009, Ministère de la Culture et de la communication. White, R., Mensan, R., Bourrillon, R., Clark, A., Ranlett, S., Chiotti, L. & Cretin, C. 2014. Abri Blanchard (Aurignacien ancien et récent) et Abri Castanet, secteur central (Aurignacien ancien), pp. 62-64 in DRAC d’Aquitaine, Bilan scientifique de la région Aquitaine 2012, Ministère de la Culture et de la communication.

68

Fuente del Trucho, Huesca (Spain): Reading interaction in Palaeolithic art Pilar Utrilla1 and Manuel Bea2 Area of Prehistory. University of Zaragoza (Spain). Group ‘PPVE’ ‘Torres Quevedo’ Postdoc Researcher (MINECO+3D Scanner). Group ‘PPVE’ 1

2

Abstract We present a brief look at the Palaeolithic rock art of the cave of Fuente del Trucho (Asque-Colungo, Huesca, Spain). There are two different decorated areas: the outside part of the cave, with its engravings sanctuary; and the inner part, with 21 decorated panels. A stylistic and thematic analysis of this site allows us to determine different points in common with some other caves in the Franco-Cantabrian coastal region, the northern side of the Pyrenees and the south of the Iberian Peninsula. Because of the large number of motifs, the themes (hand stencils, complex series of dots, horses, trefoil signs…) and the U/Th dating carried out on some of them, this site can be considered as a major and ancient Palaeolithic sanctuary. Keywords Palaeolithic rock art; hand stencils; geographical interactions; pre-Pyrenees

the rock canvas, a splendid white support, which would make it possible to highlight the hundreds of paintings on the cave’s ceiling and walls.

Historiography The decorated cave of Fuente del Trucho was discovered in 1978 by a team from the Museum of Huesca and the University of Zaragoza, led by V. Baldellou. The study of its rock art was published for the first time in 1981, in the Altamira Symposium (Beltrán & Baldellou 1981). Despite the great importance of this cave, there were only a few general studies and references to its rock art and it was twenty years after its discovery that an initial description of the 21 painted panels was released (Ripoll et al. 2001). At the congress on the Gravettian (held in 2011 at Altamira), the cave’s two main rock art groups were published: the ceiling (panel XV) and the frieze (panel V-VII) (Utrilla et al. 2013). At the XVIIth UISPP Congress (held in September 2014 at Burgos), in a session called ‘The chronology of Palaeolithic cave art: new data, new debates’, we were able to present the U/Th dating results obtained from the cleanest crusts that cover hands stencils, series of dots, horse figures and trefoil signs (Hoffman et al. 2014).

The Cave of Fuente del Trucho is located in Asque-Colungo (Huesca, Spain), in the Arpán ravine, a tributary of the Vero river on the left side, next to the natural spring that gives the site its name. It cannot be defined as a real cave, being much more like a deep shelter, with a large entrance facing the southeast that gives access to a wide chamber 24 m deep, divided into two dissymmetrical lobes. The smaller one, to the south, has a spherical dome blackened by organic matter and smoke, and an oval ‘window’ (a natural oval hole) through which sunlight enters. The outside part of the oval hole had been surrounded by a hundred red dots. These motifs could belong either to Palaeolithic art or to Schematic art, the latter being documented in the rock art of the Arpán shelter, located in the same ravine. The floor of the southern part of the cave falls into an oblique calcite flow, where several deeply carved engravings were made. The second lobe, in semidarkness, presents on its walls and ceiling a large number of paintings, most of them in red. There are 21 decorated panels where so far there have been documented a minimum of 44 hand stencils, 9 horses, 3 bears, 2 cervids, 1 small ibex, 4 trefoil signs and a large number of series of dots. Those are the four main themes of the paintings and engravings. In what follows we are going to focus on the relationship between all these matters with those from Franco-Cantabrian and Iberian rock art.

The archaeological excavations of the Mousterian levels of the external area, directed by V. Baldellou and A. Mir, involved five different campaigns from 1979 onwards (Mir 1987). In 2005 P. Utrilla and L. Montes recommenced the archaeological survey of the Upper Palaeolithic levels, focusing on two areas: at the base of the outdoor engravings and at the foot of the trefoil signs, in the right inner area of the cave (Montes et al. 2006). In the International Congress ‘Cien Años de Arte Rupestre Paleolítico’ (Oviedo, July 2014) we focused on the relations between the archaeological site and the paintings (Utrilla et al. 2014). The final study, devoted to the paintings, is still awaiting the resolution of an integral project for the restoration, preservation and cleaning of

The engravings panel The engraved motifs occupy a 5 m long panel. A great motif of a bear (1.6 m long), in a hibernating position, 69

Prehistoric Art as Prehistoric Culture

Figure 1. Location of the cave of Fuente del Trucho. In the lower image, note how the daylight passes through the natural hole in the rock and illuminates the engravings.

appears in a preferential location, receiving direct sunlight from the oval hole in the morning (Fig. 1). It was shaped by a technique that combines excision for the body and incision for the head. A paw, also excised, is located above the bear’s body. On the right, there is also an engraved head of a second bear (Utrilla et al. 2013, fig. 2, numbers 1 to 4). On the right part of the

panel, three herbivore heads can be distinguished. From top to bottom: a horse and a deer facing left and a second horse, with a rectangular muzzle, looking right. It is quite difficult to identify the deer’s species: the start of antlers pointing forwards and next to the forehead excludes a red deer and instead suggests a reindeer (Fig. 2).

70

Utrilla and Bea: Fuente del Trucho, Huesca (Spain): Reading interaction in Palaeolithic art

Figure 2. Engravings from the external sanctuary.

Figure 3. Plan of the ceiling and decorated panels.

71

Prehistoric Art as Prehistoric Culture As a whole this is clearly an atypical ensemble, both for the techniques (the excision on the bear and the paw) and also because of the animal themes in external engraved sanctuaries in Cantabrian caves. Indeed, there are neither female deer with a tri-lineal head (Nalon group, Chufin) nor headless bison (Santo Adriano, Murciélagos, La Luz, Chufin, Venta Laperra) nor headless horses (Hornos de la Peña). However, the bear and its paw have parallels in the engraved paws in Niaux (Man-Estier 2011) or in external sanctuaries such as Venta La Perra (Arias et al. 1998).

However, the most interesting decorated core is placed in the centre of the ceiling (panel XV). Here there are three child-sized black hand stencils with incomplete fingers together with two adult red hand stencils related to the complex series of red dots organised in a radial distribution. In all cases the hand stencils on the ceiling are orientated inwards. It is important to highlight the apparent tendency of the hand stencils to be grouped together in trios, as shown in panels XX, XXI, XIII and IX, although there are red patches that could potentially belong to a fourth stencil in all of them. The hand motifs have, in many cases, incomplete phalanges, in a clear relationship to those represented in the caves of Gargas and Tibiran, located 110 km away as the crow flies, but with the difficult obstacle of the Pyrenees in-between.

The horse with a rectangular muzzle and pointed ears was interpreted as a feline by Beltrán (1993), but the lions in many Palaeolithic art sites (as Trois Frères, Chauvet or Vogelherd) have rounded ears and the long and rectangular muzzle appears, for example, on an Aurignacian block from Abri Cellier, interpreted as a horse head.

Strange or singular elements are documented for the following cases: 1) in panel X there is a forearm stencil, as in the caves of Maltravieso, Fuente del Salín or El Castillo. It is also associated with five discs; 2) in panel XIII, to the left of the three black hand stencils, two probable representations of bent fingers have been documented, similar to those from Gargas and Pech Merle; 3) In panel VII the fingers are so short and separate that they look like a bear’s paw; 4) child-sized hand stencils are quite common, some of them even very tiny. For example, in panel I, in the deepest part of the cave, there is a very small hand stencil that seems to be baby-sized, younger than 1 year old, as has also been documented in Gargas (Sahly 1974). In panel XV two of the black hands seem to be a 4-year old child, and the same dimensions have been documented for another hand in panel XIX and three others in panel XIII; 5) in panel XV, close to the cave’s entrance, there is a red hand stencil with the thumb bent to the inside part of the motif, in a similar position to one observed in Les Garennes (Airvaux et al. 2006) or even hands G4-2 and G1-2 in Les Trois Frères (Bégoüen et al. 2014: 76 & 88), Pech Merle (Lorblanchet 2010: 142) or Moulin de Laguenay (Lorblanchet 2010: 399, fig. 21), in these last three cases the little finger is the one bent.

The paintings The paintings are concentrated in the great lobe, and they are located in three main areas: the north wall, the west and the ceiling, although it is possible that most of its surface could have been decorated, but currently it is masked by the black soot or lichen. Nevertheless, the natural support of the cave seems to be white, so the decorated panels in Palaeolithic times would have been a truly spectacular sight (Fig. 3). There is a right axis in the centre of the cave where one finds the most outstanding panels: the frieze with horses associated with linear series of dots, and a trefoil (panel VI) and, on the ceiling, black and red hand stencils also associated with complex series of dots and a horse motif (panel XV). At the entrance, panel XXI is the most visible and can be observed with natural light, being located in front of the exterior engraved sanctuary. The hands Hand stencils constitute the cave’s most abundant and characteristic iconography. At least fourteen panels with representations of hand stencils have been documented, and about fifty hands are clearly recognizable. However, it is possible that the total could reach a hundred of these motifs once the walls and ceiling have been cleaned. Most of the hand stencils were painted in red, but three are black (panel XV) and some others yellow (panels VII and XIII). Eight of these panels are located on the north (XXI, XIX, XVIII, XIV, X) and west (III, II, I) walls, and six other panels on the ceiling, associated with series of dots and horses (XX, XV, XIII, IX, VIII, VII). The placing of the hands, in the deep central area of the cavity, is repeated in La Garma, likewise with dots or paired marks. Some other common characteristics are also documented between Fuente del Trucho and La Garma: they occupy large spaces; they are not hidden; they adapt to natural limits; and they are easily visible (González-Sáinz 2003).

At least 13 are left hands and 6 are right hands. Seen as a whole, the stencils from Fuente del Trucho present two peculiar features which make them different from the others already known on the Cantabrian coast and, on the contrary, resemble those from the north side of the Pyrenees: 1) there are black stencils and 2) most of them are incomplete, with the third phalange missing. In fact, among the hundred hands represented in the Cantabrian caves (El Castillo, La Garma, Fuente del Salín, Altamira, Tito Bustillo and Cudón) only the black stencils from Cudón (1), Fuente del Salín (1) and Altamira (2, on the Great Ceiling) present incomplete fingers. In the south of the Iberian Peninsula we find incomplete black hands in Ardales (Málaga), and some with the little finger altered, in this case with red colour, in Maltravieso (Cáceres). 72

Utrilla and Bea: Fuente del Trucho, Huesca (Spain): Reading interaction in Palaeolithic art with the U/Th result from Fuente del Trucho. A similar date to the one in Fuente del Trucho is that obtained from the charcoals found at the base of the decorated panel in Fuente del Salín (around 27,000 once calibrated) (García Díez & Garrido, 2013). As for the hand stencils of El Castillo, where the same team have used the same dating technique, the oldest dates reach 37,630 for sample 0-82 (Pike et al. 2012).

However, on the north side of the Pyrenees (France) black stencils are predominant and, at the same time, most of the incomplete ones are located in Gargas and Tibiran, on a vertical axis with Fuente del Trucho, on the south side of the Pyrenees. This fact allows us to suggest, merely as a hypothesis, that there may have been a real loss of fingers due to frostbite while crossing the central part of the Pyrenees (the highest part), injuries that could have had a higher incidence in weaker people (children) (Utrilla 2005). In the case of Fuente del Trucho it is not possible to support Leroi-Gourhan’s classic interpretation of a hunting code with bent fingers, because most of the incomplete fingers appear on child-sized hands. Incomplete black stencils also appear in some other French caves, such as Cosquer, Les Combarelles, Font de Gaume or Moulin de Laguenay.

The series of dots are represented in eight panels. In fact, series of red dots are the second theme, after the hand stencils, in the iconography of Fuente del Trucho. There are three interesting patterns: 1) Alignments of series of dots in three, four or five rows, occupying the main places and with quite good visibility. This theme appears as a background scene in the frieze of panel VI that is almost 6 m long. It seems to join a horse, located at the left part (panel V), with two others in panel VI, with a large head of an undetermined animal (bear, horse, bovid) in an intermediate position, right in the middle (Fig. 4.1).

Calcite crusts overlap several hand stencils, so it was possible to date six cases by U/Th, establishing that the most ancient date is over 27.500 years, in the same date range as all the Gravettian hand stencils (Utrilla et al. 2014). The Gravettian date of the crust over the stencils is a little bit younger than those around 26.000 from the cave of Cosquer or the one from a bone in Gargas (26,860±460), because once calibrated (cal BP) they give dates around 32,000, and these should be compared

Another similar case is documented in panel XXI where at least two series of dots series in four and five rows are aligned some cm below an ibex figure. They are also

Figure 4. Dots series. 1. Panel VI and VII; 2. Panel XII; 3. Panel XXI; 4. Panel XV. Photographs have been treated with D-Stretch®.

73

Prehistoric Art as Prehistoric Culture associated with several trefoil signs. These series of dots can be seen in daylight, being close to the entrance (Fig. 4.3). With regard to the series of dots on the vertical frieze, in bands of 4 and 5 lines, the same theme has also been documented in some other Levantine or Schematic rock art shelters, for example at Arpán, in the same ravine as Fuente del Trucho; and Huerto de las Tajadas (Bezas).

1 doubtful), VIII, XII (2 specimens) and XV. All of them seem to be associated with series of dots (panels V, VI and XII), hand stencils or finger flutings (VIII) (Fig. 5). The frieze, a vertical rocky ledge, was the place chosen to present five of these motifs in a horizontal development, like a scene (panels V, VI and VIII). In panel VI, at one of the edges of the series of dots, there are two horses with erect manes and striped long hair looking to the left. At the other edge, 6 m away, there is a horse head orientated to the right (panel V), and a fourth doubtful head in the centre of the frieze. These details could date them to an evolved Solutrean, as could be indicated in the mane on the neck of an engraved horse from the Petite Grotte de Bize, classified in the Upper Solutrean by Sacchi (1986). However. according to Djindjian (2013) that decorated pebble comes from an ancient excavation (Genson) with mixed materials, so it could belong to either Solutrean or Gravettian.

2) Radial and curvilinear motifs. In this group are integrated the angular signs that accompany the two opposing horses of panel (Fig. 4.2) as well as the complex series of dots(that seem to form geometrical motifs on panel XV associated with three black and two red hand stencils (Fig. 4.4). All of them are clearly visible in the centre of the ceiling. With regard to the complex series of dots, Beltrán proposed in 1993 that they probably represent an image of a starry sky. Another hypothesis was proposed by Utrilla (2005) who pointed out the existence of routes through the Pyrenean passes. These routes could have established a communication network between both sides of the Pyrenees (Gargas and Fuente del Trucho). The difficulty of the route could have impressed the travellers, who perhaps got lost on the way, making detours which are represented by the complexity of some motifs.

A fifth horse motif (panel IV), with a ‘M’ shape, is placed after panel V. This specimen appears symmetrical to a male deer (panel VII) placed at the other end of the frieze, both of them with the same ‘M’ shape. Two other horses, placed back to back, are associated with series of dots in panel XII. Their similarity to several horses at Siega Verde, classified in the so-called Style III, is clear (Alcolea & Balbín 2006, 2007).

3) Complex zoomorphic dot motifs. In the final part of the frieze and below a deer motif (panel VII) the series of dots seem to form an animal motif with a long muzzle, looking to the right, or maybe a curvilinear ideomorph (Fig. 4.1). In the upper part of panel XV the dots seem to follow the profile of a possible horse looking left (Fig. 4.4). In both cases there are yellow motifs around them: one hand stencil in panel VII and yellow opposing rows of dots in panel XV. The U/Th date obtained for the crust that covers the series of dots in panel VII, more than 31.000 years (the most ancient date obtained from the paintings in the cave) places this theme in the Aurignacian-Gravettian transition (Utrilla et al. 2014).

In panel VIII, associated with five finger marks, another horse head with a long nape and nostril was represented. A calcite crust on its back has been dated to around 29,000, so it is claimed to be Gravettian. On the other hand, this figure’s close similarity to one represented in sector C2 of La Pasiega is surprising: the same long and fallen nape, a forward-curved bow in the horse-hair and a double line on the back (Utrilla et al. 2013: fig. 7). That figure is associated with two curved series of dots and a triangular sign (González Sáinz & Balbín 2002). A similar horse motif is also documented in the cave of La Haza.

Palaeolithic parallels are found in many caves of the Cantabrian coast, most of them with an early chronology (Candamo, Llonín, Pindal, Mazaculos II, La Meaza, Chufin, Porquerizo, El Castillo, La Pasiega, Altamira, La Garma, Cueva Aura or Cullalvera), the French Pyrenees (Niaux, Marsoulas, Les Trois Frères, Bedeilhac); withe Quercy, associated to hand stencils (Pech Merle-Galerie du Combel, Les Fieux, Moulin de Laguenay) or the south of the Iberian Peninsula, in this case associated with horses (Las Palomas, Atlanterra, El Moro de Tarifa).

In panel XII, on the ceiling, were depicted two horses, back to back. One of them seems to be hurt by a lance or spear. The lack of details in the interior of the motif, the pronounced cervical-dorsal curvature and the open parallel line-shaped legs indicate that it should be attributed to the Middle Solutrean, according to the Parpalló sequence (Villaverde 1994). The other motif, only represented by a small elongated head and a long curved neck, would fit better in the Gravettian (Utrilla et al. 2013: fig. 8).

The horses

The little ibex

Seven red specimens have been documented, together with another two which are more doubtful. They are placed in panels IV, V (doubtful), VI (2 clear motifs and

In panel XXI there is a little motif of an ibex with its muzzle up and open, and two small parallel horns (Fig. 4.3). This figure is framed by some natural fissures and 74

Utrilla and Bea: Fuente del Trucho, Huesca (Spain): Reading interaction in Palaeolithic art

Figure 5. Depicted horses in Fuente del Trucho. 1,2,4,7: Fuente del Trucho; 3: Bize (after Sacchi); 5, 8: La Pasiega (after González Sáinz & Balbín); 6: Parpalló (after Villaverde); 9: La Haza (after García-Díez & Eguizabal)

also seems to be associated with hand stencils and four trefoil signs.

Nevertheless, these data are very different from the general norm for the Cantabrian full signs (quadrilateral and oval) which were depicted in diverticules, lateral niches or at the edges of the main compositions, showing an effort for concealment that contrasts with the visibility of the animal figures (González Sáinz 2003). In Fuente del Trucho, all large examples are located in dominant and visible places. There are some – albeit inexact – typological parallels in the vulva-shaped signs of El Castillo, Roucadour, El Pindal or Tito Bustillo, and the triangular motifs in La Pasiega, Le Portel or Maltravieso.

This ibex looks very similar to a zoomorphic figure in Nerja. For that motif it has been proposed, according to remains of lighting, a pre-Solutrean chronology, with dates of 24,130±140 BP and 20,980±100 BP (Medina et al. 2010-2011). That would be in line with the TL and U/Th dates proposed for similar ibex motifs in panel IV/6 in La Garma, around 26,000 BP (González Sainz 2003: 214) or the tri-lineal headed hind of Antoliña, represented on a pebble from a Gravettian level dated to 27,390±320 (Aguirre 2007). There is also another similar ibex in Lloseta (Balbín et al. 2005), and another from the Solutrean-Gravettian of Parpalló (plaque 18100), although it corresponds to the TTN (Triple Naturalistic Trace) type, which is much more frequent at the beginning of the Solutrean (Villaverde 1994: 100 & 105).

Conclusions 1. U/Th dates obtained from the crusts over series of dots, hand stencils, trefoil signs and some of the horse motifs date the rock art of Fuente del Trucho to the Gravetian at the latest. 2. The black hand stencils of panel XV are more recent than the others, postdating 22,000, because in Altamira this theme is superimposed on one of the red rampant horses of that chronology. The incomplete black hand, dated by AMS, of Fuente del Salín gave 18,200±70BP (García-Díez et al. 2015). 3. The horse with the long neck and small head in panel XII and the little ibex in panel XXI are also stylistically archaic (Gravettian or Lower Solutrean), similar to other dated animals in Antoliña, Nerja or Parpalló. 4. The typology of some horses (such as the one with parallel open legs in panel XII) possibly

The trefoil signs There are five trefoil signs distributed in two panels: one of them (panel VI), with a pointed-arch shape and associated with the striped-neck horse, was interpreted as a vulva by Beltrán (1993). There are four other signs, with a semi-circular leaf-like shape, in panel XXI, close to the cave entrance and next to the ibex motif; they are also associated with two hand stencils and series of dots. There is a similar motif in the lower galleries of the cave of Candamo, also next to the entrance but hidden from the observer. 75

Prehistoric Art as Prehistoric Culture

5.

6.

7.

8.

points to a Lower-Middle Solutrean chronology, based on criteria fixed for Parpalló. The two striped horses of the frieze (panel VI) could be assigned to the Upper Solutrean, if the chronology for the controversial horse of the Petite Grotte de Bize is accepted. The ‘M’ shape of the horse (panel IV) and deer (panel VII) allows us to propose, at least, their classification in Style III, according to the chronology for the engraved horses of Siega Verde. We cannot rule out the possibility that all the depicted motifs, as a whole, could be contemporaneous, all of them included in an early phase, at the border between the Gravettian and the Solutrean, if one believes that Chauvet cave has destroyed all stylistic criteria. Lithic remains found at the base of the figures confirm the Gravettian or Solutrean chronology.

Beltrán, A. & Baldellou, V. 1981. Avance al estudio de las cuevas pintadas del Barranco de Villacantal (Huesca), pp. 131-40 in Altamira Symposium. Ministerio de Cultura: Madrid. Djindjian, F. 2013. L’apport des données de l’art solutréen dans les problématiques de circulations des chasseurs cueilleurs au Maximum Glaciaire en Europe occidentale. Supplément à la Revue Archéologique du Centre de la France 47: 275-96. García Díez, M. & Eguizabal, J. 2007. Los dibujos rojos de estilo paleolítico de la Cueva de La Haza (Ramales de la Victoria, Cantabria): estudio monográfico. Munibe 58: 177-222. García Díez, M. & Garrido, D. 2013. La cronología de las manos parietales en el arte paleolítico, pp. 492-500 in (C. de las Heras, J. A. Lasheras, A. Arrizabalaga & M. de la Rasilla, eds) Pensando el Gravetiense: nuevos datos para la región cantábrica en su contexto peninsular y pirenaico. Monografías del Museo Nacional y Centro de Investigación de Altamira, 23: Madrid. García Díez, M., Garrido, D., Hoffman, D., Pettitt, P., Pike, A. W. G. & Zilhão, J. 2015.The chronology of hand stencils in European Palaeolithic rock art: implications of new U-series results from El Castillo Cave (Cantabria, Spain). Journal of Anthropological Sciences 93: 1-18. González-Sáinz, C. 2003. El conjunto parietal paleolítico de La Galería Inferior de La Garma (Cantabria). Avance de su organización interna, pp. 201-22 in (R. de Balbín & P. Bueno, eds) El arte prehistórico desde los inicios del siglo XXI. Primer Symposium Int. de Arte Prehistórico de Ribadesella. Asosiación Cultural Amigos de Ribadesella. González-Sáinz, C. & Balbín, R. 2002. La Pasiega, pp. 165-78 in Las Cuevas con Arte Paleolítico en Cantabria. A.C.D.P.S.: Santander. Hoffman, D., Utrilla, P., Pike, A. W., Bea, M., Baldellou, V. García-Díez, M. & Zilhão, J. 2014. Chronology of Palaeolithic rock art at Fuente del Trucho: style, U-series dates and comparison with Cantabrian sites. In Preactas Congreso UISPP, Burgos 2014, session A11a. Lorblanchet, M. 2010. Art pariétal, grottes ornées du Quercy. Le Rouergue: Arles-Rodez. Man-Estier, H. 2011. Les ursidés au naturel et au figuré pendant la préhistoire. ERAUL: Liège. Medina, Mª. A., Cristo, A., Romero, A. & Sanchidrián, J. L. 2010-2011. Otro punto de luz. Iluminación estática en los santuarios paleolíticos: el ejemplo de la cueva de Nerja (Málaga, España), pp. 105-22 in (J. Clottes, ed.) L’art pléistocène dans le monde. Actes du Congrés IFRAO, Préhistoire, Art et Sociétés LXV-LXVI. Mir, A. 1987. Memoria de la quinta campaña de excavaciones en el yacimiento de la Cueva de la Fuente del Trucho. Asque-Colungo (Huesca). Arqueología Aragonesa 1985: 19-21.

Acknowledgements Project HAR2014-59042-P and Research Group Primeros Pobladores del Valle del Ebro (H-07). References Aguirre, M. 2007. Antoliñako Koba (Gautegiz-Arteaga). Arkeoikuska, 6: 121-24. Airvaux, J. J., Aujoulat, N., Baratin, J.-F., Beauval, C. & Henry-Gambier, D. 2006. Découverte d’un réseau karstique orné au lieu-dit Les Garennes, commune de Vilhonneur, Charente. Bulletin Préhistorique du SudOuest 13: 25-35. Alcolea, J. J. & de Balbin, R. 2006. Arte Paleolítico al aire libre. El Yacimiento rupestre de Siega Verde, Salamanca. Arqueología en Castilla y León, 16: Valladolid. Alcolea, J. J. & de Balbín, R. 2007. Le gisement rupestre de Siega Verde, Salamanque. Une vision de synthèse. L’Anthropologie 111: 501–48. Arias, P., Calderón, T., González-Sáinz; C., Millán, A., Moure,A., Ontañón, R. & Ruiz, R. 1998. Dataciones absolutas para el arte rupestre paleolítico de Venta La Perra (Carranza, Bizkaia). Kobie (Paleoantropología) XXV: 85-92. Balbín, R., Alcolea, J. J. & González, M. A. 2005. La Lloseta: une grotte importante et presque méconnue dans l’ensemble de Ardines, Ribadesella. L’Anthropologie 109: 641-701. Bégouën, R., Clottes, J., Feruglio, V. & Pastoors, A. 2014. La caverne des Trois Frères. Anthologie d’un exceptionnel sanctuaire préhistorique. Somogy Éditons d’Art: Paris. Beltrán, A. 1993. Arte prehistórico en Aragón. Zaragoza.

76

Utrilla and Bea: Fuente del Trucho, Huesca (Spain): Reading interaction in Palaeolithic art Montes, L., Utrilla, P. & Martínez-Bea, M. 2006. Trabajos recientes en yacimientos musterienses de Aragón: una revisión de la transición Paleolítico Medio/Superior en el Valle del Ebro, pp. 214-33 in (J. M. Maillo & E. Baquedano, eds) Miscelánea en homenaje a Victoria Cabrera. Zona Arqueológica 7 (1). Pike, A. W. G., Hoffman, D. L., García-Díez, M., Pettitt, P. B. Alcolea, J., de Balbín, R., González-Sáinz, C., de las Heras, C., Lasheras, J. A., Montes, R.& Zilhão, J. 2012. U-Series Dating of Paleolithic Art, in 11 Caves in Spain. Science 336 (6087): 1409-13. Ripoll, S., Baldellou, V., Muñoz, F. J. & Ayuso, P. 2001. La Fuente del Trucho (Asque-Colungo, Huesca). Bolskan 18: 211-24. Sacchi, D. 1986. Le Paléolithique Supérieur du Languedoc Occidental et du Roussillon. XXI Supplément à Gallia Prehistoire, CNRS: París. Sahly, A. 1974: La grotte préhistorique de Gargas. Imp. J. Cucuron: Rieumes. Utrilla, P. 2005. El Arte Rupestre en Aragón. 100 Años después de Calapatá, pp. 341-77 in (M. S. Hernández & J. Soler, eds) Arte Rupestre en la España Mediterránea. MARQ: Alicante. Utrilla, P., Baldellou, V., Bea, M. & Viñas, R. 2013. La cueva de la Fuente del Trucho (Asque-Colungo, Huesca): Una cueva mayor del arte gravetiense, pp. 526-37 in Congreso Internacional: El Gravetiense Cantábrico, estado de la cuestión. Monografías del Museo Nacional y Centro de Investigación de Altamira, 23: Madrid Utrilla, P., Baldellou, V., Bea, M., Montes, L & Domingo, R. 2014. La Fuente del Trucho. Ocupación, estilo y cronología, pp. 119-32 in (M. S. Corchón & M. Menéndez, eds) Cien años de arte rupestre paleolítico. Ediciones Universidad de Salamanca: Salamanca. Villaverde, V. 1994. Arte paleolítico de la Cova del Parpalló. Estudio de la colección de plaquetas y cantos grabados y pintados. SIP: Valencia.

77

78

Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon Paul G. Bahn Hull, United Kingdom

Abstract The paper presents an account of the gradual realisation that Palaeolithic people not only produced rock art outside caves and rock-shelters, but that it could survive in exceptional circumstances. Keywords open-air art; Palaeolithic; Shishkino; Domingo García; Siega Verde; Côa Valley

This paper is dedicated to my friend Rodrigo de Balbín who has contributed so much to the study of Ice Age imagery, including that of open-air Ice Age art. Since Ice Age people spent almost their whole lives in the great outdoors, it has always been assumed that they must also have produced art outside the caves, but that it has not survived the millennia of weathering and erosion. Indeed, Cartailhac and Breuil were the first to make this claim, at the very beginning of cave art studies: ‘If there were – and this is obviously probable – rock surfaces in the open air that were covered with paintings and engravings, it was not possible for them to survive a long time. However, the engravings could have lasted longer, depending on the rock and the depth of the lines.....they were made by hand and outside the caves, in open-air works whose existence one can now suspect but which are forever lost.....How many similar works, made in the open air, have disappeared’ (Cartailhac & Breuil 1906: 33, 77, 142; my translation).

a considerable distance from the two adjacent horses, he considered these three figures to form a separate group, the oldest figures at Shishkino, on the basis of their size, their technique and style, and the treatment of their front legs (Okladnikov & Zaporozhskaya 1959: 89). He also felt that, since horses and bovids dominate Upper Palaeolithic art, this was a further supporting factor, as was the existence of Palaeolithic settlements in the area. He claimed that horses and wild bovids died out in the area at the end of the Palaeolithic. He saw the figures’ style as being Palaeolithic, and hence also their meaning. In particular, the small oval shape beneath the big horses’s belly was compared with the so-called ‘vulva signs’ in European Palaeolithic art, and the existence of this sign next to a stallion with an erection suggested to him some link with hunting magic (this was before Leroi-Gourhan’s theories had become widely known), while the zigzag sign under the big horse’s tail was seen as an image of water (Okladnikov & Zaporozhskaya 1959: 90) [Fig. 1a/b].

It is certainly true that paintings could not survive in the open-air: in a simple experiment using haematite and charcoal and limonite, all mixed with water, some slabs of limestone and sandstone were painted on both surfaces, and then left leaning against walls in the open air. The upper side was thus exposed to wind, rain and sun, while the underside was protected against everything but frost. After four months, almost all traces had been washed off the upper sides, as one would expect, but even the underside paintings also deteriorated in that time, and some disappeared completely (Barham et al. 1999: 8485).

Needless to say, none of these factors can really withstand scrutiny, and few scholars at the time took them seriously (Bahn 2001). The presence of Palaeolithic settlements in the area in no way provides a date for these images. The front leg of the smaller horse overlaps with a small anthropomorph, probably of the Bronze Age, but the leg is so lightly abraded and so much damage has been done to this panel that it is impossible to tell which figure overlies which. The bovid – which unfortunately was painted white in Okladnikov’s time or shortly after so as to stand out in photographs – does not bear the slightest resemblance to the known Palaeolithic bovid figures of western Europe. The horses have some features that could pass as Palaeolithic, though the head of the big one resembles that of a moose. The problem here is that, unlike that of Europe, Siberian Palaeolithic portable art consists almost entirely of human figurines, so that we simply do not know what the local styles of animal depiction were in this part of the world in the Ice Age, and hence

There were occasional claims that open-air figures were of Palaeolithic age. The first major claim was made by Okladnikov in his work at the Siberian site of Shishkino by the river Lena, which has hundreds of animal depictions over a distance of about 3 km. He believed that some of the bigger petroglyphs here dated to the Palaeolithic (Okladnikov 1959; Okladnikov & Zaporozhskaya 1959) – a huge horse, a smaller horse nearby and a bovid. Unfortunately, his reasoning in this case was extremely poor. First, despite the bovid being 79

Prehistoric Art as Prehistoric Culture

Figure 1.a/b: The big horse petroglyph at Shishkino, Siberia. Photo P. Bahn; drawing after Okladnikov.

we have nothing with which to compare the Shishkino figures. The large horse is especially troubling, being by far the biggest figure not only at this site but in the whole of Siberian rock art. However, the most crucial factor against there being any surviving Palaeolithic figures at Shishkino is that the rock face here is a somewhat friable sandstone, which is constantly deteriorating, and none of its surfaces could possibly have withstood ten thousand Siberian winters intact.

All of Okladnikov’s claims were soon opposed, justifiably, by Formozov (1969), who pointed out that horse bones had been found by Okladnikov himself at the site of Kullaty in Yakutia, dating to the 3rd/2nd millennia BC, while at another site of the same period he found bovid bones which he assumed to be domesticated – but of course we do not know if the big bovid image at Shishkino is wild or not, and it is surrounded by other images of similar size and style which Okladnikov attributed to the Bronze Age. Formozov also pointed out that the stylistic analogies claimed by Okladnikov 80

Bahn: Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon between the three Shishkino figures and Palaeolithic images are far from decisive; better analogies can be made between the bovid and images of the Bronze or Iron Age, while there is no equivalent known in Palaeolithic art for the way the giant horse’s hooves are depicted. Moreover, only the giant horse is really big – 2.8 m long – while the smaller horse is only 1.2 m and the bovid 1.1 m. Many other figures of this smaller size were attributed by Okladnikov to later periods (Formozov 1969: 93-101).

shallow Cueva del Moro (southern Spain) (Martínez García 2008: 245-47). So it is highly probable that many or most open-air petroglyphs were originally painted, which would certainly have enhanced their visibility from a distance. The discovery of Palaeolithic depictions in the open air was arguably the most important development in the study of Ice Age imagery since the authentication of cave art at the start of the 20th century. Indeed, there are some notable parallels between the two events, and in the 1990s we heard remarkable echoes of the past from those who could not believe the open-air figures to be so ancient. Just as the paintings of Altamira were rejected because they looked too fresh to be ancient, and that the cave was too humid and the rock too friable to have preserved art for so long, it was claimed that the open air figures must be post-Palaeolithic, since erosion and weathering would long ago have destroyed any Palaeolithic images made outside caves. This is true for paintings; but in the Alps, for example, there are innumerable striations on rocks, left by the retreating glaciers at the end of the Ice Age, and many of them remain clear and even sharp-edged despite the passing of so many millennia.

What began as a simple supposition by Okladnikov eventually became a conviction and he and his followers began to attribute figures at other sites in Siberia and even Mongolia to the Palaeolithic, simply because the Shishkino figures existed; and other people also started to make claims for Palaeolithic open-air art in other parts of the USSR (Bahn & Vertut 1997: 215, note 90) – including even open-air paintings in Yakutia! For the most part, the reasons given were the large size of some figures, or some unusual feature, or the species involved (rhino, horse, bovid), or simply that the figures ‘looked archaic’. The exercise became self-perpetuating, and as more images were attributed to the Palaeolithic, the more it appeared to be true. The phenomenon came to a head in the 1970s when it was the subject of fierce debate in publications by several leading specialists. Since then, more images that could conceivably be Palaeolithic have been found in other regions, and in particular some Russian specialists harbour a cautious suspicion that what has been dubbed the naturalistic ‘Minusinsk style’ petroglyphs in the open-air rock art of the Yenisei region and the Altai may indeed belong to the end of the Ice Age. As yet, they have no proof of this, but they have found comfort in the discoveries in western Europe.

It was also objected that many of the open-air figures – particularly at Siega Verde and Foz Côa – are pecked, and that this technique is not known in Palaeolithic art. However, this ignores the fact that, on many of the engraved blocks of the Aurignacian period in France, the incisions were clearly made by crudely joining together rows of cupmarks or hammer-blows; the basrelief sculptures at a whole series of French rock-shelters were produced by robust percussion tools, judging by the traces of impact still visible on and around many of them; and, of course, some massive battered picks have been recovered from Angles-sur-l’Anglin and other sites. In Les Fieux, there is an ibex figure pecked into a stalagmite (see Lorblanchet 1999: 228; 2010: 323, 332); and pecked figures also exist in other caves such as Venta de la Perra, Cueva del Reno, etc. Even without these examples, it should be self-evident from the wide range of techniques mastered by the Palaeolithic artists in their portable and cave art that adaptation to pecking on openair rocks would have posed no problem whatsoever. Cave limestone is very different from schists.

In the past 30 years, a series of important finds in western Europe has finally brought the proof that Palaeolithic people did produce art in the open air, and that it can survive in exceptional circumstances (Bahn 1992, 1995; Sacchi 2002; de Balbín 2008; de Balbín & Bueno 2009). Since one of the original arguments against the authenticity of both the Altamira ceiling and the painted pebbles of the Azilian was that parietal art could not possibly survive from such a remote age even inside a cave, it goes without saying that virtually nobody was prepared to entertain the possibility that Ice Age depictions in the open air could have survived; and, as already mentioned, certainly no Ice Age painting could have survived outside – all of the sites in question consist of figures engraved and pecked into rocks. However, some of these engravings were originally coloured, like Palaeolithic bas-reliefs and much portable art. Some traces of black and red pigments have been reported at Siega Verde (de Balbín 2008a: 39; de Balbín & Alcolea 2009); while the petroglyphs of aurochs heads at Faia (Portugal), in a sheltered overhang, have preserved traces of red paint in their grooves (Baptista 1999: 15457, 2009: 72), like the deeply engraved horses in the

Perhaps the most ironic of the objections was the claim that the animals depicted are not those of the Pleistocene, but instead are modern species such as domestic goats rather than ibex. This is a precise echo of the belief by Edouard Martel, the great speleologist, who foolishly claimed in 1905 that the Altamira figures were Neolithic at best, because their paint was so fresh, and they were better preserved than some Etruscan paintings; he also believed that the cave only had depictions of presentday or ‘Neolithic’ fauna because there were no images of extinct or emigrated species such as rhino, mammoth 81

Prehistoric Art as Prehistoric Culture or reindeer (unlike the Dordogne caves, whose art he accepted as being Palaeolithic). He also saw the animals of Niaux as Neolithic, with only modern species, and he interpreted its ‘claviform signs’ as depictions of hafted Neolithic axes (Martel 1905: 679, 1906: 120; 1930: 19697. See also Cartailhac & Breuil 1906: 234; Delluc & Delluc 1998; Bahn 2001: 159). Similarly, Angel de los Ríos y Ríos declared – on the basis of Sanz de Sautuola’s sketch – that the Altamira bison were modern bulls, and the deer was a wolf (Madariaga 2002: 95)!

– especially since there is still no consensus about the meaning of cave art after more than a century of study, with new caves like Cussac, Cosquer and Chauvet constantly bringing surprises and modifying our knowledge. There are some similarities – the recognisable figures are primarily adult animals drawn in profile, with stylistic traits that correspond to those of known portable and cave art; they are dominated by horses and bovids; there are a few ‘signs’ and apparently abstract motifs; there are no scenes, virtually no humans, and no ground-lines; the art seems to cluster in ‘panels’ (i.e. separate rocks); and there is a frequent use of cracks and rock-edges in the placing of animal figures.

Likewise, Elie Massénat, a French prehistorian, claimed at the 1902 Congress of the French Association for the Advancement of Science that the animal depictions in the caves of La Mouthe, Font de Gaume and Les Combarelles were recent works made by people hiding in these caves, perhaps during the religious wars! He thus believed that the bison, reindeer and mammoths in these caves were bulls, red deer and modern elephants. He considered that the ‘grossières ébauches’ (crude sketches) in these caves, these ‘formes apocalyptiques et grotesques’, were utterly different from the animals in the remarkable portable figures of the Magdalenian. The modern fugitives who drew them were merely depicting the animals of the surrounding countryside – bulls, horses, goats and sheep – and for some reason added red deer and elephants! Moreover, he wondered how soft limestone walls could possibly have preserved such figures for very long, in view of the frequent visits to the caves, with generations of people brushing against them. In short, he saw all the cave depictions as historical, even recent, works, and refused to accept the attribution of these ‘caricatures d’animaux modernes’ to the Reindeer Age. The analogies with the debate over Foz Côa, 95 years later, are striking (obviously the species depicted are the very ones present in these regions in the late Ice Age; there were no mammoths, reindeer, etc, this far south; and it is the excellent micro-climate that has helped preserve the figures).

The first such find occurred at the Spanish open-air site of Domingo García (Segovia) in 1970, when a team of aficionados found the first pecked horse figure, hammered into a rocky outcrop, 960 m above sea level, and saw it as ‘of Palaeolithic style’, but this was not taken seriously till the 1980s (Ripoll & Municio 1999: 18). Later, a number of rock-shelters and caves (including Murciélagos) in the Nalón Valley, Asturias, were found to have some eroded lines and animal figures deeply engraved in their exterior areas (Fortea 1981); nevertheless, these engravings were still in or near shelters and cave-mouths, rather like the sculptured friezes and carved blocks of France. The next discoveries, however, were at truly open-air sites. In 1981, three animal figures, including a fine horse, 62 cm long and 37.5 cm high, were found on a rock-face on the right bank of the Albagueira, a little tributary of the river Douro, at Mazouco in northeast Portugal, at an altitude of 210 m above sea level (Jorge et al. 1981, 1982; Jorge 1987; Jorge et al. 1981/2.); they had survived thanks to a position which protected them from the elements. They were hammered out, though the marks were subsequently scored into continuous lines. Since its discovery, the horse has been badly damaged by chalking, scoring and painting. Around the same time, at Domingo García, the pecked horse figure, almost a metre in length, was studied more closely (Martín & Moure 1981; de Balbín & Moure 1988). In style, it resembles the engraved horses in the cave of La Griega, Segovia, in the same region (Sauvet 1983). A crude engraving of a different style and period is superimposed on its outline. Since then, a meticulous examination of this rock and other rocks in the region (some of them up to 15 km away) as well as a removal of lichens revealed at least 150 figures in eight areas; most of them are fine-line engravings, and there are many hundreds of unidentified lines or fragments of figures (Ripoll & Municio 1992, 1999; Ripoll et al. 1994). Stylistically, most of the figures have been ascribed to the end of the Solutrean and to the early Magdalenian They are dominated by horses, but cervids and caprids are also well represented. Bovids are comparatively rare. It was found that the big horse (actually the best

As Emile Cartailhac said in reply to Massénat, his opposition would serve only to underline the merit of those who drew attention to the cave depictions; a little opposition never hurt discoveries, and did not prevent facts from being certain. After all, he himself, twenty years before, had doubted the Palaeolithic age of Altamira’s figures! Gustave Chauvet went further, pointing out to Massénat that, if the images had indeed been made by recent refugees from the religious wars or from the Revolution, one might expect their motifs to reflect the preoccupations of the artists – i.e. religious or political depictions, rather than bison, reindeer or horses. Henri Breuil, in a later publication, alluded to the stupid position of the ‘vieux Massénat’ (Roussot 1990: 170; Hurel 2011: 88, 98). It is hard to say, at this relatively early stage of the investigations, how open-air art relates to cave art 82

Bahn: Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon Personally, I was astonished and baffled by the general apathy that greeted these first discoveries, since I considered them the opening of a whole new chapter in Palaeolithic art studies. I placed a summary of the Fornols Haut engravings in Nature (Bahn 1985), and included illustrations and a description of all three sites in my book on Ice Age imagery (Bahn & Vertut 1988). As far as I am aware, this was the first book on the subject to include an account of the phenomenon.

preserved in a group of three) was originally a fine-line engraving, which was pecked later (Ripoll & Municio 1999: 70). The discoveries at Domingo García and Mazouco were major events in Palaeolithic art studies, yet they received very little recognition at the time, let alone aroused any display of interest on the part of specialists. The same thing happened when a series of fine incisions were found at Fornols-Haut, Campôme, in the eastern French Pyrenees, on a huge block of schist located at an altitude of 750 m on a mountainside (Sacchi 1987, 2008; Sacchi, Abelanet & Brulé 1987; Sacchi et al. 1988; Bahn 1985). The rock has been greatly weathered by wind but, because the eastern face is sheltered, its engravings, although eroded, are clearly visible. The face is covered in engravings, drawn in all directions, and comprising about ten small animals – none complete – as well as signs and zigzags. The finest figures include the head of an isard, 7.5 cm high, and that of an ibex. Stylistically they seem to belong to the Magdalenian. Although published in some major journals, Fornols was still accorded little importance by most specialists; and until a special conference (with an excursion to the rock) was held at Tautavel in October 1999, very few had bothered to try and visit this unique site for themselves, or even the cast of its engraved surfaces kept in Carcassonne (Sacchi, pers. comm..).

Subsequently, two further sites of the same nature were found in Spain. In Almería, an area previously bereft of Palaeolithic art, a horse figure was discovered at Piedras Blancas (Martínez García 1986/7, 1992, 2008). Situated on an inclined block, at an altitude of 1400 m near the town of Escúllar, the horse is made with multiple deeply incised lines. Stylistically it has been ascribed to the final Gravettian or the Solutrean, through comparisons with engraved plaquettes from the cave of Parpalló. In view of developments at Domingo García, it was always likely that more figures would eventually be found in the vicinity, and sure enough a few more have indeed been discovered recently (de Balbín 2008a: 26-27) [Fig. 2]. This is certainly what occurred at Siega Verde, near Ciudad Rodrigo, Spain, about 60 km south of Mazouco. Here, in rocks along the left bank of the river Agueda (another tributary of the Douro), at its best crossing-

Figure 2. The pecked horse at Piedras Blancas . Photo P. Bahn.

83

Prehistoric Art as Prehistoric Culture point (a major ford), what were at first thought to be a few hammered-out figures, discovered in 1989, turned out to be a minimum of 540 pecked and incised images (including 244 animals), on 91 decorated surfaces, most of them located within a 1300 m stretch, though 75% are concentrated within 400 m (Alcolea & de Balbín 2006; de Balbín et al. 1995; de Balbín & Alcolea 1994); no less than half of the identifiable figures are horses, with bovids and cervids in second place. Like the other sites of the region, the style has been attributed to the final Solutrean/early Magdalenian.

Vila Nova de Foz Côa, was planned for the valley, and work began on it in 1992. Had it been completed, the engravings would have been irrevocably lost under 100 metres of water.

Yet even now, after these new finds, plus the dramatic increase in the number of known figures at Domingo García (Ripoll & Municio 1999), most specialists still remained unimpressed; amazingly, some major volumes published on Palaeolithic art in France (e.g. Vialou 1991), Italy (Leonardi 1989) and elsewhere during these years made absolutely no mention of the phenomenon. They were still fixated entirely on cave art and portable art.

Since then, more such sites have continued to be discovered, primarily in Portugal, where the microclimate and the schist rocks seem to have favoured their preservation. Most of them – like Mazouco, Fornols and Piedras Blancas – comprise one or a few figures or rock-faces, so our extremely limited knowledge of this newly discovered phenomenon really rests primarily on the Côa Valley, Siega Verde and Domingo García. Most recently, another possible example of the phenomenon – two definite horses plus two or three other animal figures, pecked into a schist block – has come to light in the Hunsrück region of Germany (Welker 2014, 2015) [Fig. 3], while a horse figure in Valcamonica, Italy, has also been claimed to be a candidate (Martini et al. 2009).

An international campaign which lasted exactly a year finally led to victory for the art; thanks to a change of government in Portugal in October 1995, work on the dam was stopped, and an archaeological park was created in the Côa Valley, which shortly afterwards became a Unesco World Heritage site.

What finally changed this situation and forced all specialists to acknowledge the existence of open-air Ice Age imagery was by far the most important discovery – first made public in November 1994. It occurred in the Côa valley in northeast Portugal, yet another tributary of the Douro. Here, there are at least 2000 pecked figures and engravings from the Ice Age, on more than 500 schist outcrops at 45 sites spread over about 20 km (Fernandes 2010; A. M. Baptista, pers. comm.). They depict horses, ibex, deer and, especially, aurochs (wild oxen), plus occasional fish or humans; they measure from 5 cm to over 2.5 metres in size. From their style, it was first estimated that they were probably Solutrean, though, as will be seen below, at least two different periods are represented. They include many very fine images, particularly two large horses with overlapping heads. The vast majority of known figures are the visible, pecked examples, but the area also contains great quantities of exceedingly fine, almost invisible engravings similar to those of Domingo García and Fornols.

But is this really Palaeolithic art? The sites were originally assigned to this period simply on the basis of the style of their pecked or engraved figures; on the other hand, the same is true of the vast majority of Palaeolithic cave art, and it is safe to say that if most of these figures had been found inside caves they would have been classed as Palaeolithic without hesitation. All parietal art is notoriously difficult to date, open-air examples especially so since they normally have no context whatsoever. However, there is nothing remotely similar to these figures in the Middle Stone Age and later art of eastern Spain, or in the schematic art of more recent periods. Every single European Ice Age art specialist who visited these sites or even saw photographs of them unhesitatingly attributed some of the figures to the Upper Palaeolithic, giving age estimates varying from 20,000 to 10,000 years (Bahn 1995; Lorblanchet 1995, 1995a; Zilhão 1995).

The current total of known figures is clearly minimal, as many rocks and images have been destroyed through natural factors, as well as through human exploitation of the area: roads, tracks, the railway line and constructions have all taken their toll, and vestiges of images have been found in walls. Many decorated rocks are doubtless masked by sediments (as at Fariseu, see below) or by vegetation. Moreover the Pocinho dam, 5 km away, drowned unknown quantities of art under 12 m of water decades ago – 70 rocks are known underwater at 5 sites, but there are probably many more – one estimate is that there may originally have been between 3500 and 5000 figures (Baptista & Reis 2008: 154; Baptista 2009: 130). Fluctuations in water-level have exposed them several times since the discovery, so that they are still more or less accessible. A new hydro-electric dam, that of

In 1995, an attempt was made by the electricity company building the Côa dam to have a few of the valley’s figures dated directly by four dating specialists using different methods, all of them highly experimental, such as AMS dating of organic deposits and the microerosion technique. Unfortunately, some of the panels chosen had already been much affected by chalking and other damage. The results were mixed, to say the least, with two specialists concluding that the art was indeed Palaeolithic in age, the third claiming they were probably only a few thousand years old, and the fourth insisting they dated back only a couple of centuries (Bednarik 84

Bahn: Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon

Figure 3. Petroglyphs of horses and other motifs, Hunsrück, Germany. Photo P. Bahn.

1995a, 1995b; Dorn 1997; Phillips et al 1997; Watchman 1995, 1996)! Although some saw this as the last nail in the coffin of stylistic dating, others pointed out the many flaws and uncertainties in the particular methods used (especially the ‘micro-erosion method’), and dismissed the results as useless (Bednarik 1995b; Zilhão 1995a, 1995b).

soon as investigations began, after the discovery of the rock art, some major and important open-air Palaeolithic living sites were found in close proximity: for example, the site of Olga Grande contained quartzite picks in Gravettian levels (dating to between 31,000 and 26,800 bp), and use-wear analysis showed that they had been used for pecking rocks. Ochre and manganese pigments were also found in this site (Aubry 2002; Aubry & Sampaio 2008).

Stylistic comparisons are always somewhat subjective, unless the two kinds of art are from the same site, but the same applies to any typological scheme and, as in anything, including direct dating, one ultimately has to rely on the experience and expertise of the practitioner. The Palaeolithic art corpus is uniquely fortunate in world rock art in that, alongside the parietal figures, we have many thousands of well-dated pieces of portable art with which comparisons can be made regarding style, content, technique, etc. Hence the similarities of some Côa figures – and figures from the other open-air sites – to solidly dated drawings from Parpalló (Villaverde 1994), La Tête du Lion and elsewhere were so striking and incontrovertible that there could be absolutely no doubt of their Pleistocene age.

Doubters also claimed that some of the fine-line engravings had been made with metal tools. Consequently, some experiments were carried out to differentiate the grooves made with 30 stone and 30 metal points. Most of the criteria that had been used to attribute some Côa engravings at Ribeira do Piscos to metal points proved to be invalid. Metal marks are not necessarily finer than those made by stone. Other factors such as the orientation of the tool, the nature of the support, and the pressure applied also play a role. At Fornols, Côa and Siega Verde the figures were unquestionably made with stone, as is proved by the existence of an occasional double groove (d’Errico, Sacchi & Vanhaeren 2002). Final proof (which was not really needed!) of the Pleistocene age of the Côa figures came at the site of Fariseu, where some pristine panels of pecked figures were masked by in situ stratified Palaeolithic deposits

One objection raised by the doubters was that no Palaeolithic occupation was known in this region of Portugal. However, this was simply due to the fact that little prehistoric research had ever been done here. As 85

Prehistoric Art as Prehistoric Culture which not only could be dated by radiocarbon and thermoluminescence, but which also contained more than 60 engraved plaquettes in the 15 square metres excavated of the Final Magdalenian layer at the top. These are not fallen pieces of engraved panels, but plaquettes which were engraved after falling (Aubry & Sampaio 2008, 2008a; García & Aubry 2002). Recently, the open-air living site of Foz do Medal in the Sabor valley has yielded more than 1250 fragments of engraved plaquettes (Figueiredo et al. 2014). It is worth noting that one large rock along the river, Piscos rock 24, is naturally shattered into small panels, 32 of which bear small engravings, so one could consider these to be ‘parietal plaquettes’, a new concept in Ice Age art.

In addition, just as moving the light source in caves can cause tremendous effects of light and shadow, and make figures appear and disappear, so in the open-air sites it is certain the daily movements of the sun and the moon will have likewise caused variations in visibility. Finally, a pioneering geomorphological study of the location of the pre-Magdalenian figures of the Côa has led to a predictive model of the areas favourable to the preservation of rock art panels and prehistoric human occupation (Aubry et al. 2010, 2012.). One can already speculate that much open-air art seems to be linked to territorial marking, passes or animal migration routes, while some is clearly linked to dwelling areas. Open-air Palaeolithic petroglyphs have now also been found and dated (to a minimum of 15,000 years) at sites along the Nile in Egypt (Huyge et al. 2011; Kelany 2012, 2014) [Figs. 4 & 5]. Recently, a claim has been made for possible Palaeolithic petroglyphs in Mongolia, based largely on style and archaic appearance, including two probable mammoths and a possible rhinoceros (JacobsonTepfer 2013); and claims – so far unsubstantiated – have also been made that some engravings of bovids and humans at Gobustan, in Azerbaijan, may also date to the late Palaeolithic (Farajova 2012).

At present, the consensus view is that the Côa’s pecked figures are Gravettian in age, while its fine-line engravings date to the Magdalenian. The pecked figures tend to be highly visible – some were clearly meant to be seen from a distance, such as the huge life-size aurochs at the confluence of the Piscos and Côa rivers (obviously intended to be seen from the opposite bank), and, as mentioned above, when coloured they would have been even more visible. The fine engravings, on the other hand, are far smaller and could only be seen from close-by. So one could see the Gravettian figures as public, and the Magdalenian as private. There are other contrasts between them: the pecked figures are mostly down by the water, on large schist panels along the river bank or on adjacent slopes; while the fine engravings are generally higher up on slopes, and/or up tributaries. The Gravettian figures are primarily horses, bovids and ibex (with few deer), whereas the Magdalenians also drew many red deer. The pecked figures have one leg per pair, no hooves, and no internal details; while the fine engravings have multiple lines, four legs, and lots of infill (Baptista 1999, 2009; Baptista & Reis 2008; Luis 2008. For a major study of the Gravettian images of the Côa, see Baptista et al. 2008).

To sum up, a series of major discoveries over the past 30 years have transformed our conception of the parietal art of the last Ice Age in Europe, confirming what had long been suspected by some researchers – that the wellknown art surviving in at least 400 caves and shelters in western Europe is unrepresentative and uncharacteristic of the period, owing its apparent predominance in the archaeological record to a taphonomic fluke. In reality we have no idea how important or frequent the decoration of caves was in Ice Age Europe, but it is extremely probable that the vast majority of that period’s rock art was produced in the open air. This in turn has very profound implications for our interpretation of the possible functions of Palaeolithic art as a whole. It was by no means limited to dark, mysterious caves – they are merely the places where it has been best preserved.

Studies of the Côa and other sites are starting to reveal some interesting details of location. At a basic level some sites are along rivers, whereas others (Fornols, Piedras Blancas, Domingo García) are in uplands and dominate the landscape. It will be recalled that, at Siega Verde, almost all the hundreds of figures are on the left bank of the river, with only one on the right. In the Côa, only a quarter of sites (9) are on the right banks of the Côa (6) and Douro (3). More than half (299) of the outcrops bearing Pleistocene images are at or very close to the foot of slopes, between 110 and 200 m in altitude. Because so many of the sites are on the left banks, they overwhelmingly face east, and the same applies at Siega Verde – perhaps this predominance of southeast-facing panels is due to the fact that they accumulate heat during the day (Fernandes 2010).

It is understandable that many people, having devoted their lives or careers to studying the art inside caves, and to conjuring up all kinds of functional or mystical explanations for its location, are loath to accept this view. One can never prove whether the one art-form was more important than the other – we do not know if we have now discovered almost all of the decorated caves or 0.001% of them; and it is highly likely that most outdoor art, including all paintings, will indeed have vanished for ever. However, in view of the ubiquity of open-air rock art in every part of the world, and in so many postPalaeolithic cultures, it is surely sensible to suppose that, along with the portable art which must primarily have been produced out of doors, this was likewise the norm in the Ice Age. That does not necessarily mean that 86

Bahn: Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon

Figure 4. Petroglyphs of a fish and aurochs, Qurta, Egypt. Photo P. Bahn.

87

Prehistoric Art as Prehistoric Culture

Figure 5. Petroglyph of aurochs, Subeira, Egypt. Photo P. Bahn.

88

Bahn: Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon outdoor art was less mystic or religious or powerful than the art inside caves – dark caves certainly add an aura of mystery and dread, but we know from ethnography in Australia, for example, that open-air art can be just as full of taboos and power, just as it can also be simply decorative or narrative.

Bahn, P. G. 1985. Ice Age drawings on open rock faces in the Pyrenees. Nature 313, 530-31. Bahn, P. G. 1992. Open air rock art in the Palaeolithic, pp. 395-400 in (M. Lorblanchet, ed.) Rock Art in the Old World. Indira Gandhi National Centre for the Arts: New Delhi. Bahn, P. G. 1995. Cave art without the caves. Antiquity 69, 231-37. Bahn, P. G. 2001. Palaeolithic open-air art: the impact and implications of a ‘new phenomenon’, pp. 15560 in Actes du Colloque de la Commission VIII de l’UISPP 1998. Trabalhos de Arqueologia 17 (J. Zilhão, T. Aubry & A. Faustino Carvalho eds), Instituto Português de Arqueologia. Bahn, P. G. & Vertut, J. 1988. Images of the Ice Age. Windward: London / Facts on File: New York. de Balbín Behrmann, R. (ed.) 2008. Arte Prehistórico al aire libre en el Sur de Europa. Actas. Junta de Castilla y León: Salamanca. de Balbín Behrmann, R. 2008a. El arte rupestre paleolítico al aire libre en la Peninsula Ibérica, pp. 19-56 in (R. de Balbín, ed.) Arte Prehistórico al aire libre en el Sur de Europa. Actas. Junta de Castilla y León: Salamanca. de Balbín Behrmann, R. & Alcolea González, J. J. 1994. Arte paleolítico de la meseta española. Complutum 5, 97-138. de Balbín Behrmann, R. & Alcolea González, J. J. 2009. Les colorants de l’art paléolithique dans les grottes et en plein air. L’Anthropologie 113, 559-601. de Balbín, R. & Bueno, P. 2009. Altamira, un siècle après: art paléolithique en plein air. L’Anthropologie 113, 602-28. de Balbín Behrmann, R. & Moure Romanillo, J. A. 1988. El arte rupestre en Domingo García (Segovia). Revista de Arqueología 87, 16-24. de Balbín Behrmann, R. et al. 1995. El yacimiento rupestre paleolítico al aire libre de Siega Verde (Salamanca, España): una visión de conjunto. Actas VII, 11 Congresso de Arqueol. Peninsular. Trabalhos de Antropologia e Etnologia (Porto) 35, 73-102. Baptista, A. M. 1999. No Tempo Sem Tempo. A arte dos caçadores paleolíticos do Vale do Côa. Centro Nacional de Arte Rupestre: Vila Nova de Foz Côa. Baptista, A. M. 2009. O Paradigma Perdido. O Vale do Côa e a Arte Paleolítica de Ar Livre em Portugal. Edições Afrontamento e Parque Arqueológico do Vale do Côa: Vila Nova de Foz Côa/Porto. Baptista, A. M. & Reis, M. 2008. Prospecção da arte rupestre no Vale do Côa e Alto Douro portugés: ponto da situação em julho de 2006, pp. 145-92 in (R. de Balbín, ed.) Arte Prehistórico al aire libre en el Sur de Europa. Actas. Junta de Castilla y León: Salamanca. Baptista, A. M. et al. 2008. O santuário arcaico do Vale do Côa: novas pistas para o compreensão da estruturação do bestiário Gravettense e/ou gravettosolutrense, pp. 89-144 in (R. de Balbín, ed.) Arte

It may be concluded that henceforth it might be more useful simply to divide Ice Age rock art into that of daylight, semi-darkness and total darkness. It can now be argued very plausibly that cave art, albeit striking and important, was not the normal or characteristic art of the Ice Age, but rather owes its prominence to a freak of survival. It is far more likely – though this can never be proved, thanks to lack of preservation – that art in shelters, cave-mouths and especially in the open air was overwhelmingly dominant in the period. If this art was constantly to be seen all around, in the landscape (as well as on tents, clothing and bodies) rather than hidden away in caves, this would certainly help to explain how the artistic canons, styles and subjects were transmitted over such vast distances and over such huge spans of time.

References Alcolea González, J. J. & de Balbín Behrmann, R. 2006. Arte Paleolítico al Aire Libre. El Yacimiento Rupestre de Siega Verde, Salamanca. Arqueología en Castilla y León, Memoria 16. Junta de Castilla y León: Valladolid. Aubry, T. 2002. Le contexte archéologique de l’art paléolithique à l’air libre de la vallée du Côa, pp. 25-38 in (D. Sacchi, ed.), L’Art Paléolithique à l’Air Libre. Le paysage modifié par l’image. TautavelCampôme, 7-9 octobre 1999. GAEP & GEOPRE: Carcassonne. Aubry, T. & Sampaio, J. D. 2008. Chronologie et contexte archéologique des gravures paléolithiques de plein air de la vallée du Côa (Portugal), pp. 211-23 in (R. de Balbín, ed.) Arte Prehistórico al aire libre en el Sur de Europa. Actas. Junta de Castilla y León: Salamanca. Aubry, T & Sampaio, J. 2008a. Fariseu: new chronological evidence for open-air Palaeolithic art in the Côa valley (Portugal). Antiquity 82 http:// antiquity.ac.uk/ProjGall/aubry/index.html Aubry, T. et al. 2010. Paleolithic engravings and sedimentary environments in the Côa river valley (Portugal): implications for the detection, interpretation and dating of open-air rock art. Journal of Archaeological Science 37, 3306-19. Aubry, T. et al. 2012. Nature vs. Culture: present-day spatial distribution and preservation of open-air rock art in the Côa and Douro River Valleys (Portugal). Journal of Archaeological Science 39, 848-66. 89

Prehistoric Art as Prehistoric Culture Prehistórico al aire libre en el Sur de Europa. Actas. Junta de Castilla y León: Salamanca. Barham, L., Priestley, P. & Targett, A. 1999. In Search of Cheddar Man. Tempus: Stroud. Bednarik, R. G. 1995a. The age of the Côa Valley petroglyphs in Portugal. Rock Art Research 12, 86103. Bednarik, R. G. 1995b. The Côa petroglyphs: an obituary to the stylistic dating of palaeolithic rockart. Antiquity 69, 877-83. Cartailhac, E. & Breuil, H. 1906. La Caverne d’Altamira à Santillane, près Santander (Espagne). Imp. de Monaco. Delluc, B. & Delluc, G. 1998. Les mésaventures du spéléologue Edouard-Alfred Martel dans la préhistoire en Périgord. Bull. Société Historique et Archéologique du Périgord 125, 627-55. Dorn, R. I. 1997. Constraining the age of the Côa valley (Portugal) engravings with radiocarbon dating. Antiquity 71, 105-15. d’Errico, F., Sacchi, D. & Vanhaeren, M. 2002. L’analyse technique de l’art gravé de Fornols-Haut, Campôme, France. Implications dans la datation des représentations de style paléolithique des sites de plein air, pp. 75-86 in (D. Sacchi, ed.) L’Art Paléolithique à l’Air Libre. Le paysage modifié par l’image. Tautavel-Campôme, 7-9 octobre 1999. GAEP & GEOPRE: Carcassonne. Farajova, M. 2012. Pleistocene art in Azerbaijan, in (J. Clottes, ed.) L’art pléistocène dans le monde / Pleistocene art of the world / Arte pleistoceno en el mundo, Actes du Congrès IFRAO, Tarasconsur-Ariège, septembre 2010, Symposium ‘ Art pléistocène en Asie ‘. N° spécial de Préhistoire, Art et Sociétés, Bulletin de la Société Préhistorique AriègePyrénées, LXV-LXVI, 2010-2011, CD, 929-42. Fernandes, A. P. Batarda 2010. Slope orientation of rock art sites in the Côa Valley, Portugal: a case study in the spatial distribution of open-air Upper Palaeolithic rock art. IFRAO Congress, September 2010 – Symposium: Pleistocene art in Europe (Pre-Acts). Figueiredo, S. S. et al. 2014. Foz do Medal terrace – an open-air settlement with Paleolithic portable art. INORA 68, 12-20. Formozov, A. A. 1969. Ocherki po Pervobytnomu Isskustvu. Nauka: Moscow. Fortea, J. 1981. Investigaciones en la cuenca media del Nalón, Asturias (España). Noticia y primeros resultados. Zephyrus 32/33, 5-16. García Diez, M. & Aubry, T. 2002. Grafismo mueble en el Valle de Côa (Vila Nova de Foz Côa, Portugal): la estación arqueológica de Fariseu. Zephyrus 55, 15782. Hurel, A. 2011. L’Abbé Breuil. Un Préhistorien dans le Siècle. CNRS Editions: Paris. Huyge, D. et al. 2011. First evidence of Pleistocene rock art in North Africa: securing the age of the Qurta

petroglyphs (Egypt) through OSL dating. Antiquity 85, 1184-93. Jacobson-Tepfer, E. 2013. Late Pleistocene and early Holocene rock art from the Mongolian Altai: the material and it cultural implications. Arts 2, 151-81. Jorge, S. O. et al. 1981. Gravuras rupestres de Mazouco (Freixo de Espada à Cinta). Arqueologia (Porto) 3, 3-12. Jorge, S. O. et al. 1982. Descoberta de gravuras rupestres em Mazouco, Freixo de Espada-à-Cinta (Portugal). Zephyrus 34/35, 65-70. Jorge, V. O. 1987. Arte rupestre en Portugal. Revista de Arqueología 76, 10-19. Jorge, V. O. et al. 1981/2. Mazouco (Freixo de Espada à Cinta). Nótula arqueológica. Portugalia, nova serie II/III, 143-145. Kelany, A. 2012. More Late Palaeolithic rock art at Wadi Abu Subeira, Upper Egypt. Bull. des Musées royaux d’art et d’histoire (Brussels) 83, 5-21. Kelany, A. 2014. Late Palaeolithic rock art sites at Wadi Abu Subeira and el-’Aqaba el-Saghira, Upper Egypt. Cahiers de l’AARS 17, 105-15. Leonardi, P. 1989. Sacralità Arte e Grafia Paleolitiche – Splendori e Problemi. Museo Civico di Storia Naturale di Trieste. Lorblanchet, M. 1995. Les Grottes Ornées de la Préhistoire. Nouveaux Regards. Editions Errance: Paris. Lorblanchet, M. 1995a. Les gravures rupestres de la vallée de Côa (Portugal). L’Archéologue, Archéologie Nouvelle No. 17, Dec. 1995/Jan. 1996, 38-41. Lorblanchet, M. 1999. La Naissance de l’Art. Genèse de l’art préhistorique. Errance: Paris. Lorblanchet, M. 2010. Art Pariétal. Grottes Ornées du Quercy. Rouergue: Rodez. Luis, L. 2008. A Arte e os Artistas do Vale do Côa. Parque Arqueológico do Vale do Côa: Vila Nova da Foz Côa. Madariaga de la Campa, B. (ed.) 2002. Escritos de Marcelino Sanz de Sautuola y Primeras Noticias sobre la Cueva de Altamira. Gobierno de Cantabria: Santander. Martel, E-A. 1905. La spéléologie au XXe siècle (Revue et bibliographie des recherches souterraines de 1901 à 1905). Spelunca VI. Martel, E-A. 1906. Réflexions sur Altamira. L’âge des gravures et peintures des cavernes. Congrès préhistorique de France, pp. 112-36. Martel, E-A. 1930. La France ignorée (des Ardennes aux Pyrénées). Delagrave: Paris. Martín Santamaria, E. & Moure Romanillo, J. A. 1981. El grabado de estilo paleolítico de Domingo García (Segovia). Trabajos de Prehistoria 38, 97-105, 2 pl. Martínez García, J. 1986/7. Un grabado paleolítico al aire libre en Piedras Blancas (Escúllar, Almería). Ars Praehistorica V/VI, 49-58. Martínez García, J. 1992. Arte paleolítico en Almería. Los primeros documentos. Revista de Arqueología 130, 24-33. 90

Bahn: Open-air Ice Age art: the history and reluctant acceptance of an unexpected phenomenon Martínez García, J. 2008. Arte paleolítico al aire libre en el sur de la Península Ibérica: Andalucía, pp. 237-58 in (R. de Balbín, ed.) Arte Prehistórico al aire libre en el Sur de Europa. Actas. Junta de Castilla y León: Salamanca. Martini, F. et al. 2009. Alle origini dell’arte rupestre camuna, pp. 183-96 in La Valle delle Incisioni. 19092009 cento anni di scoperte. 1979-2009 trente anni con l’Unesco in Valle Camonica. Catalogo della mostra, Brescia 21 marzo-10 maggio 2009. Brescia. Okladnikov, A. P. 1959. Shishkinskie Pisantsy. Irkutsk. Okladnikov, A. P. & Zaporozhskaya, V. D. 1959. Lenskie Pisanitsy. Nauka: Moscow-Leningrad. Phillips. F. M. et al. 1997. Maximum ages of the Côa valley (Portugal) engravings measured with Chlorine-36. Antiquity 71, 100-4. Ripoll López, S. & Municio González, L. 1992, Las representaciones de estilo paleolítico en el conjunto de Domingo García (Segovia). Espacio, Tiempo y Forma, Serie 1, Prehistoria y Arqueología, V, Madrid, pp. 107-38. Ripoll López, S. & Municio González, L. J. 1999. Domingo García. Arte Rupestre Paleolítico al aire libre en la meseta castellana. Arqueología en Castilla y León 8. Junta de Castilla y León: Salamanca. Ripoll López, S. et al. 1994. Un conjunto excepcional del arte paleolítico. El Cerro de San Isidro en Domingo García. Nuevos descubrimientos. Revista de Arqueología 157, 12-21. Roussot, A. 1990. Daleau et la découverte des grottes ornées: l’excursion historique d’août 1902 aux Eyzies, pp. 168-79 in (A. Coffyn, ed.) Aux Origines de l’Archéologie en Gironde. François Daleau (1845-1927). Société Archéologique d Bordeaux: Bordeaux. Sacchi, D. 1987. L’art paléolithique des Pyrénées roussillonnaises, pp. 47-52 in Etudes Roussillonnaises offertes à Pierre Ponsich, Perpignan. Sacchi, D. (ed.) 2002. L’Art Paléolithique à l’Air Libre. Le paysage modifié par l’image. Tautavel-Campôme, 7-9 octobre 1999. GAEP & GEOPRE: Carcassonne. Sacchi, D. 2008. Le rocher gravé de Fornols vingt trois ans après sa découverte, pp. 193-209 in (R. de Balbín, ed.) Arte Prehistórico al aire libre en el Sur de Europa. Actas. Junta de Castilla y León: Salamanca. Sacchi, D., Abelanet, J. & Brûlé, J-L. 1987. Le rocher gravé de Fornols-Haut. Archéologia 225, juin, 52-57. Sacchi, D. et al. 1988. Les gravures rupestres de Fornols-Haut, Pyrénées-Orientales. L’Anthropologie 92, 87-100. Sauvet, G. 1983. Les représentations d’équidés paléolithiques de la grotte de la Griega (Pedraza, Segovia). A propos d’une nouvelle découverte. Ars Praehistorica 2, 49-59. Vialou, D. 1991. La Préhistoire. Gallimard: Paris. Villaverde Bonilla, V. 1994. Arte Paleolítico de la Cova del Parpalló. 2 vols. Servei d’Investigació Prehistòrica: Valencia.

Watchman, A. 1995. Recent petroglyphs, Foz Côa, Portugal. Rock Art Research 12, 86-103. Watchman, A. 1996. A review of the theory and assumptions in the AMS dating of the Foz Côa petroglyphs, Portugal. Rock Art Research 13, 21-30. Welker, W. 2014. Felsbilder im Hunsrück. Erste paläolithische Felskunst in Deutschland. Berichte zur Archäologie am Mittelrhein und Mosel (Koblenz) 20, 9-33. Welker, W. 2015. First Palaeolithic rock art in Germany. INORA 71, 1-7. Zilhão, J. 1995. L’art rupestre paléolithique de plein air. Dossiers d’Archéologie 209, Dec. 1995/Jan. 1996, 106-17. Zilhão, J. 1995a. The age of the Côa Valley (Portugal) rock-art: Validation of archaeological dating to the Palaeolithic and refutation of ‘scientific’ dating to historic or protohistoric times. Antiquity 69, 883901. Zilhão, J. 1995b. The stylistically Palaeolithic petroglyphs of the Côa Valley (Portugal) are of Palaeolithic age. A refutation of their ‘direct dating’ to recent times. Trabalhos Antropol. Etnol. (Porto) 35, 3-49.

91

92

Decorated sites and habitat: social appropriation of territories Denis Vialou Muséum National d’Histoire Naturelle, Paris, France

Abstract The mobility of Palaeolithic societies took place between regions and territories that are considered cultural, according to the analyses in use. Numerous long-distance movements are reconstructed through figurative and abstract graphic symbolism and chrono-stylistic relationships. Archaeological contexts defined by habitats and chronological schemes obtained by dating the habitats and the representations allow territories to be delimited. Their economic or technical appropriations are discriminated from the symbolic appropriations. These are marked by spatio-temporal dynamics but also by spatio-temporal permanences. Symbology materialises the socialisation of spaces, their implacable originality. They generate inter-populational contacts and inter-regional movements. Keywords Palaeolithic societies; Magdalenian; France

on varied scales such as the Solutrean and Magdalenian in the Iberian Peninsula and even the scarce Solutrean in the Ebro valley compared with the widespread Solutrean in Andalucia. Spaces are circumscribed without taking into account the real distances covered by different groups in a given time or moment.

Introduction Modern political ideologies and the scientific concepts in western cultural systems that are derived from them play a bigger role in prehistoric research than is generally thought. Therefore prehistorians, specialists in the last Ice Age in Europe, grant importance to chronological, palaeoenvironmental and palaeo-climatic data to situate the Palaeolithic societies they study on the largest possible comparative scale (Jaubert & Barbaza 2005; Jaubert et al. 2013; Mangado 2010). The monographs and papers of my colleague and friend Rodrigo de Balbín Behrmann (e.g. 2008) perfectly reach that level of analysis of population, habitats and decorated sites, renewing their interpretation. To access our analysis of the appropriation, by Upper Palaeolithic groups, of their everyday spaces, a methodological compilation characterising space and time is necessary.

The invocation of radiocarbon determinations, particularly direct AMS dates for parietal art, sometimes exceeds the limits of what is reasonable or, at least, of what is objectively controllable. Dates and their accumulated lists make and unmake cultural divisions, occasionally without including a rigorous critical examination of the comparative validity of the dates and the origin and nature of the samples, or indeed of their precise correspondence with the archaeological data themselves. Additionally, the dates do not possess a temporal extension: they are instants like a photograph and not time, like a film. The oldest dates at Chauvet provide some ages related with the Aurignacian and more precisely with the production of the rhinoceroses, but they do not date the parietal ensemble. In contrast, they have been used to say too often that Chauvet is Aurignacian, which if anything it will be partially. When direct dates accumulate on a ceiling like at Altamira or on the panels at El Castillo, the chronologies are drawn with a certain clarity, but do not put complete precise and definitive order in the parietal history, whether it is long or short, continuous or discontinuous.

First notes: cartographic illusions, chronological illusions On the 200-300 cm2 covered by a full page map, prehistorians encompass tens of thousands of square kilometres when they want to show a defined geography, such as Cantabria (over 5300 km2) or the Pyrenees (a little under 116,000 km2). The Iberian Solutrean, in its whole geographical extension, covers 582,000 km2 and the European dimension of Palaeolithic art is surrealistic at the scale of a simple map. Prehistorians fill their maps with dots to mark the prehistoric sites or to suggest relationships between two or more sites within a given geographical area, although some issues deserve greater precision in the distribution of the dots or their clustering. For example, the Middle Magdalenian habitats in the Périgord, with a high density on the axis of the River Vézère and their dispersal towards the north of the region. Other important examples are the comparisons of the demographic-cultural densities in geo-cultural spaces or

The physical objectivity of the dates supports parallels between them. But archaeological data or assemblages cannot always be compared. They may be disassociated culturally. We should ask ourselves about the archaeological validity of the relationship between Chaves and La Saltpêtrière, or between cultural regions manifestly independent from one another, like the Cantabrian Solutrean, spread across a large region with 93

Prehistoric Art as Prehistoric Culture different landscapes, and the Andalucian Solutrean in a smaller, more ecologically-heterogeneous area.

Mobile habitats and symbols In the last Ice Age, whereas some regions were densely populated, others were depopulated temporarily or for longer periods of time. Maps locating the sites attest these variations in time and in the occupied space. Environmental and climate factors are increasingly used to explain these variations. Thus, the late glacial maximum (LGM) is regarded as the cause for the disappearance (or no development) of Solutrean populations north of the Pyrenees and the intensification (or no regression) south of the mountain range. Regional studies highlight particular areas, like Andalucia and Mediterranean Iberia. Subsistence is linked to environmental factors, especially in the use of floral and faunal resources.

Dynamics of Palaeolithic societies Sedentism and nomadism supported ethnographicethnological approaches to primitive societies in the 19th century and the first half of the 20th century. This theoretical dichotomy was accompanied by another, equally poorly founded: the opposition between hunter-gatherer groups and farming societies. Current prehistoric research, whether or not it is accompanied by population maps and tables with radiocarbon dates, attests movements of Palaeolithic societies though archaeological and palaeoenvironmental data. In fact, they lived in variable geographical areas that correspond to natural geographic and eco-climatic boundaries. They built their habitats in the open air, in rock-shelters and cave entrances, on hills or on riverbanks. Their territories or everyday space corresponded to a seasonal, annual or pluri-annual subsistence sedentism. Thus we can distinguish geo-culturally and chrono-climatically, in the late Ice Age, and possibly coetaneous: the Magdalenian group in the Paris Basin (Pincevent, Etoilles, …), the Magdalenian group in the Rhineland (Gönnersdorf), the one in the Périgord (Laugerie-Basse, Font-Bargeix) and another in Cantabrian Spain. Different socio-cultural histories correspond to spatial appropriations in each of these regions, related to their use of different rock types, river or pelagic fish, snails, or reindeer, red deer, ibex, etc. They clearly maintained relations or contacts, sometimes over considerable distances; it is enough to consider the Mediterranean shell adornments in the Magdalenian levels at Etoilles. Influences or contacts of a social, psycho-cultural or socio-symbolic kind confirm power relations and even conflict between different social groups. One of the best examples of relationships in the late or transitional phases of the Magdalenian is the depiction of stylised female profiles in portable and parietal art. Their distribution in central Europe materialises the same symbolic figurative emblem in several Magdalenian regions, some of them in the Swiss mountains. However, other Magdalenian provinces, with graphic-symbolic relationships (Central Pyrenees, Basque Country and Cantabria-Asturias for the classic claviforms) do not possess this emblem. These similarities and disparities raise questions about the occasional expansionist modes that can be appraised in the figurative and abstract representations not circumscribed to frontiers or everyday territorial boundaries. The most detailed comparison between late Magdalenian groups in the Paris Basin and the Périgord reveals technological similarities in some tools and arrow heads, as well as in the procurement of raw materials and in subsistence, but cannot obviate the cultural distance that separates them: the total absence of representations in the former region (with a few isolated exceptions) and the abundance in the latter.

These environmental factors are less clear when a region contains diverse micro-climates. This is the case of Périgord and Cantabria (Arias et al. 2012); the former is located between the Massif Central with its rigorous climate and the Atlantic plain of the River Garonne; Cantabria between the coast and the rugged mountain range. All the Upper Palaeolithic periods, Aurignacian, Gravettian, Solutrean and Magdalenian are well represented in both regions. Maps certainly show increasingly large populations, with demographic expansion in the Magdalenian compared with the modest Solutrean populations. Surrounding areas share the same climate and environmental conditions but appear not to have experienced the population growth. For example, the Magdalenian population in Cantabria, the Basque Country and Asturias is not seen in Galicia. The Magdalenian demographic growth must have been based on more favourable palaeo-climate and subsistence conditions. An origin for this Magdalenian ‘baby boom’ should be sought. Thriving habitats like La Madeleine, Laugerie-Basse and El Castillo allowed large family groups to gather under the same roof (enormous rock-shelters and porches) and therefore create new social relationships, accompanied by dynamic cultural and symbolic behaviour. The size of these habitats and the succession of their occupations revealed by excavations indicate that many generations belonging to several families lived in those places, sometimes called ‘aggregation sites’. A multi-generational temporal dimension, still not measured objectively, characterised them and discriminated them radically from places occupied briefly, such as hunting camps and workshops where lithic resources were extracted and prepared. Good examples of the differences between the sites have recently been found at Solutrean sites on the banks of the River Creuse, a tributary of the Loire (Société d’Etudes 2013): habitats in a karst area next to the river – the Fritsch rock-shelter and a little further away the Monthaud rock-shelter; workshops on the outcrops of Turonian flint, like the open-air site of Maitreaux on a tributary of the River Creuse; and a seasonal and 94

Vialou: Decorated Sites and Habitat: Social Appropriation of Territories isolated hunting camp in the first foothills of the Massif Central, Fressignes. The diachrony of the habitats is fundamental to understanding the spatial structuring of Palaeolithic population in the region. It results in occupations lasting years or even centuries, as we know occurred in Cantabria or in Périgord in the Magdalenian. The occupied space establishes a socio-cultural time that is perpetuated from generation to generation, like a history shared economically and, on the evidence of the representations, linguistically, by the society.

other from Bédeilhac, less than a day’s walk from Le Mas d’Azil. The similarity between the two objects is so great that it might be wondered whether or not they were made by the same craftsperson. This resemblance is inscribed in or derived from the great similarity between the two sites: prolonged habitatation in areas of huge caves reached by daylight, very similar lithic and bone assemblages, a profusion of portable depictions with some of the animal figures comparable in their technique and theme, and a division of the parietal ensembles between chambers and passages with easy access and hidden zones that are difficult to reach. These two caves located in different environments are twins within the overall Pyrenean Magdalenian in Ariège. They differ in the parietal ensembles with different distributions from the topographic-morphological and thematic-figurative points of view and in their symbolic construction. The effective mobility of portable symbolism participates in these relationships with prolonged habitatations, sharing technical activities (industry) and subsistence tasks (procurement of minerals, animals and plants).

Portable representations, which are abundant in certain occupations, and parietal depictions like at El Castillo, or sites with both modes, for example Le Mas d’Azil in the Pyrenees, generate static and pluri-generational habitats, a dimension between groups and regional populations, which are fundamental to understanding cultures and their expansion. The symbolism of the portable representations is partly associated with the manufacture of utilitarian objects, tools and weapons (spearpoints, rods, perforated batons...) made from bone and other hard animal substances. A triple alliance is revealed by this association between symbolism and substance: direct relationships between hunting to obtain the carcasses, technical activities to process them, and the symbolic production. Other close relationships are included in this alliance orientated towards individual and collective behaviour: thus body ornaments and funerary adornments which involved gathering and working hard substances like ivory, antler and teeth of rarely-hunted animals, particularly the canines of carnivores and dangerous animals: wolves, foxes, bears and felines. These chains of specialised activities involve members of socio-family groups in the habitats, the central points of the activities carried out in the territory and the distributors of some of these symbolic objects like circular pendants, pierced hyoid bones and contours découpés. The numerous excavations and studies lead to a profusion of archaeological data revealing these dynamics linked to the planned and chosen symbolism in portable art. We can measure the amplitude of its semantic function, as well as the vigour of its sociocultural function and, naturally, its expansion across one or several territories. Hafts and heads of spear-throwers made from reindeer antler, carefully sculpted with animal figures, attest visions where the imaginary and the real are mixed together. The case of the spear-throwers with ibex at Le Mas d’Azil and Las Caldas is particularly illustrative, even ignoring their real contemporaneity or the origin of their manufacture. Some Magdalenian occupants of the Cave of Le Mas d’Azil (several occupation levels) must have moved (‘travelled’) with certain ease, as shown by a further two unquestionable types of contact. First, the technical similarity (carving spear-throwers from reindeer antler) and the figurative resemblance (sculpted young herbivore) and its symbolism (head turned backwards and defecating) in two spear-throwers, one from Le Mas d’Azil and the

A second type of long-distance contact of the Pyrenean Magdalenian groups at Le Mas d’Azil, attesting their active mobility, is seen in the parietal representation of two sea fish: flat fish of the Pleuronectes genus, which live in the benthic zones of the Atlantic and Mediterranean coast. These fish were observed by the Magdalenian groups at Le Mas d’Azil either on the Atlantic shore, 230 km away, or in the Mediterranean, about 150 km away, or perhaps at the site itself if the fish were taken there. The pendant depicting two kids carved from a sperm whale tooth is further evidence of direct contact with the marine world, either Atlantic or Mediterranean. However, an Atlantic provenance is more likely, given the large Magdalenian population on the whole Gascony coast. Parietal and portable symbolisms are closely related, but not chronologically (the exact dates are not known). At Le Mas d’Azil these clear long-distance relations imply inter-ethnic, regional and ultra-regional contacts. These belong to or result from the social and economic dynamics inherent in a lasting multi-family or multi-ethnic habitat. Centrifugal orientations appear to dominate. However, if contacts between sites in the Pyrenean foothills and the Atlantic coast had been numerous and repeated, centripetal relationships from the maritime provinces would also be acceptable. Technical, economic and symbolic dynamics form large cultural groups distributed in such different environments as the limestone foothills on the north side of the Pyrenees and the marine landscape close to the mountain ranges (Barbaza & Fritz 2000; Cazals et al. 2007). Symbolic immobility of decorated sites In regions with a high population density in the Palaeolithic (mainly in the Magdalenian), the decorated sites are in 95

Prehistoric Art as Prehistoric Culture territory-habitats according to the characteristics and properties of the geographic settings. The environmental micro-scale largely explains reciprocal locations. Thus, the decorated caves in Monte Castillo, in a prominent conical hill, associate habitat and parietal ensembles in a setting of valleys and areas of easy transit. In Périgord, decorated caves like Rouffignac, La Mouthe and Font de Gaume are equally located on communication routes in the valley of the River Vézère. These different sites are reflected in certain differences in the activities carried out in the proximity of the decorated sites and those performed in complex larger habitats. For example, food procurement activities may be more favourable for gathering plants and small animals on high ground, or more favourable for fishing or hunting large herd animals in the valley.

the decorated sites and the habitats in a totally isolated karst region, distant from the main south-north axis of population. Contrasting with the analysis of this remote assemblage in the west of the Paris Basin, we might wonder about the archaeological basis for cultural attributions to the Gravettian or Solutrean, based on the direct dates of the representations in Cosquer Cave, as these cultures are not found on the Mediterranean coast. Did the artists come from neighbouring provinces, such as the Rhône corridor to the west, or from the Epigravettian province to the east? Its immersion under the sea possibly explains this isolation from habitats or other decorated sites, which may be submerged. However, it should be pointed out that the existence of a decorated site (or perhaps more) in this Mediterranean region lies outside (in the current state of our knowledge) the framework of the population that produced the symbolic representations known to the west of the Rhone.

In regions with a low population density in the Palaeolithic, such as the northern Spanish Plateau during severe climatic episodes such as the LGM, habitations and decorated sites are not in close spatial relations to each other, susceptible of forming clear social and economic networks. The decorated caves are undoubtedly isolated (or sometimes accompanied by another site) and are allotted a geo-cultural function that often remains unverified. Fuente del Trucho and the nearby Chaves sites, in the Ebro Basin near the central Pyrenees, are usually cited as points of union with the symbolic and cultural relationship of the French decorated sites of Gargas and Tibiran (particularly because of the stencilled hands). However, the issue of cultural attributions, especially the direct dates of these parietal ensembles, is still unresolved. The same applies to the geographical position of the site of Isturitz, theoretically (but not proven) a hub for Gravettian, Solutrean and Magdalenian populations on the axis of south-west France – Pyrenees – northern Spain.

The spectacular sites of Côa (Baptista 2009) and Siega Verde (Alcolea & Balbín 2006) in the Douro valley have considerably altered the question of Palaeolithic decorated sites in the open-air that a series of sites like Mazouco had raised (Balbín 2008). Their chronology has been based on archaeological occupations, the themes, with cold climate fauna like rhinoceros, bison and felines at Siega Verde, and decorated caves in the Spanish Plateau (El Reno, La Hoz) (Balbín & Alcolea 2005). As regards techno-stylistic comparisons, the immense iconography at Parpalló, on the other side of the Plateau, dominates and orientates the approximations. These sites are included within a long transcultural duration, covering the Gravettian, Solutrean and Magdalenian (lato sensu) (Balbín & Alcolea 2002). This iconographic monumentality and durability associated with occupations beside the river, in a region next to the Plateau and open to the extreme south-west of Europe in the Palaeolithic, make possible the deduction of a series of general characteristics that are relevant for the analysis of territorial appropriations.

Long-isolated parietal sites now possess data of their association with other parietal sites. The case of the cave of Mayenne-Sciences in the west of the Paris Basin is particularly revealing of these geo-cultural associations in small territories. Margot Cave, in the same small karst valley of the River Erve, contains even more representations (research still in progress) than the cave of Mayenne-Sciences. The visual homogeneity of the ensemble in Mayenne-Sciences contrasts with the heterogeneity in Margot Cave. The population in the small Erve valley is mainly Solutrean, as shown by finds of Solutrean artifacts since the 19th century and, above all, by the occupation in Rochefort Cave (excavation in progress) and Chèvre Cave. This indicates profound geographic isolation from the northernmost Solutrean population, in the valley of the River Creuse. Curiously, the direct dates and stylistic attributions of the representations in these two caves do not correspond to the Solutrean but to the Gravettian and Magdalenian. There must be partial cultural discordance between

The open-air decorated sites, especially in the Douro valley, are located on the edge of the classic French territories of habitats and decorated sites, first moderately Aurignacian and Gravettian and increasingly Solutrean and Magdalenian. They are not the result of brief territorial expeditions. On the contrary, they mark territorial emplacements on the side of other cultural territories where there is no open-air rock art. They define their own territory, inaugurated in the Gravettian and concluded in the Magdalenian. Stylistic and partially thematic parallels with the iconographic ensemble at Parpalló confirm their links, which were quite intense at certain times, with this Mediterranean province in the Iberian Peninsula. Parpalló itself reflects successive cultural movements from the north, the Pyrenees and 96

Vialou: Decorated Sites and Habitat: Social Appropriation of Territories even further away, in south-west France. Isolated engraved rocks, as at Fornols Haut in the Mediterranean Pyrenees, demonstrate another cultural dimension, with no equal in the engraved and sometimes painted open-air sites in the Iberian Peninsula. This Iberian open-air rock art may be considered an autonomous and original spatial and social entity in the framework of Upper Palaeolithic cultures – a spatial permanence, defined by an open-air rock art symbology, independent from the underground symbology in other proximate or distant Palaeolithic provinces. Balbín and Alcolea’s interpretation (2003) of the relationship between caves and the open-air recognises the creative power of these symbols, capable of questioning the relations and economic order of territories.

The appropriations of territories reflect multiple social formations appreciable at two extremes: the habitats, connected with where these groups go or return, and the symbolic realm, which crosses space-time with its possibilities of communication-comprehension (Vialou 2003). They form the history of societies in an experienced and conceived space-time.

References Alcolea González, J. & Balbín Behrmann, R. de 2006. Arte Paleolítico al aire libre. El yacimiento rupestre de Siega Verde, Salamanca. Arqueología en Castilla y León. Memoria n°16, Junta de Castilla y León: Valladolid. Arias Cabal, P., Corchón Rodríguez, M. S., Menéndez Fernández, M. & Rodríguez Asensio, J. A. (eds) 2012. El Paleolítico Superior Cantábrico. Actas de la Primera Mesa Redonda, San Román de Candamo (Asturias) 26-28 de abril de 2007. Universidad de Cantabria: Santander. Balbín Behrmann, R. de (ed.) 2008. Arte Prehistórico al aire libre en el Sur de Europa. Documentos Pahis, Junta de Castilla y León: Salamanca. Balbín Behrmann, R. de & Alcolea González, J. 2002. L’art rupestre paléolithique de l’intérieur péninsulaire ibérique: une vision chronoculturelle d’ensemble, pp. 139-57 in (D. Sacchi, ed.) L’art Paléolithique à l’Air Libre. Le Paysage Modifié par l’Image, Tautavel Campôme, 7-9 octobre 1999. GAEP & GÉOPRÉ: Tautavel. Balbín Behrmann, R. de & Alcolea González, J., 2003. El Arte Rupestre Paleolítico del interior peninsular. Elementos para el estudio de su variabilidad regional, pp. 223-53 in (R. de Balbín Behrmann & P. Bueno Ramirez, eds) El Arte Prehistórico desde los Inicios del Siglo XXI. Primer Symposium Int. de Arte Prehistórico de Ribadesella. Asosiación Cultural Amigos de Ribadesella: Ribadesella. Balbín Behrmann, R. de & Alcolea González, J. 2005. Testigos del frío. La fauna en el Arte Rupestre Paleolítico del interior peninsular, pp. 547-66 in Geoarqueología y Patrimonio en la Península Ibérica y el entorno mediterráneo. Adema: Soria. Baptista A. M. 2009 . O paradigma perdido: O Vale doCôa e a arte paleolítica de ar livre em Portugal. Edições Afrontamento: Porto; Parque arqueológico do Vale do Côa. Barbaza, M. & Fritz, C. 2000. Art et habitats magdaléniens dans les Pyrénées centrales: un siècle et demi de recherches en Comminges. Revue de Comminges 116 (3): 291-320. Cazals, N., González Urquijo, J. & Terradas, X. (eds) 2007. Fronteras naturales y fronteras culturales en los Pirinéos prehistóricos. Monografías del Instituto Internacional de Investigaciones Prehistóricas de

The thematic range of parietal representations in rockshelters and porches reached by daylight, normally in sites near decorated caves in the Franco-Cantabrian area in general, evoke the parietal iconography of open-air sites. The animal figures form the main group in this iconography, while abstract and geometric representations are scarce, not very varied and absent from many sites. The symbolic choice of rock art in the open-air or underground is partly different, as we know. But a close relationship exists in the conception of these parietal ensembles, as if the ‘naturally delimited vertical wall’ were primordial in the symbolic conception of the ensembles. None of this occurs at decorated sites in the open air, with their rocks distributed over a wide area, with their symbolic labyrinths that include thematic sectors, or even chronological sectors, as at Siega Verde. Modern interpretative trends attempt to assess various contacts, and exchanges between proximate or distant provinces, based on the particularity of portable representations, whose symbolism and themes, especially in body ornaments (circular pendants, contours découpés) or for arrows (sagaies) are unique. The stylistic parallels in animal representations that are used are not always demonstrable. It seems that the symbolic territories of parietal ensembles become implacably original, as most complex signs suggest. Palaeolithic decorated sites can define symbolic parietal provinces: for example, Cantabrian caves with complex linear or dotted signs, or the Perigordian caves with their tectiforms. But each site possesses its own symbolic personality that transmits a permanence in space, in the territory and in time: a cultural phase. Finally, the numerous contacts and evidence of exchanges, attested by technical, economic and cultural dynamics of Palaeolithic societies, are combined with the unyielding independence of symbolic parietal dynamics, both in rock art and in the space-time of the history of each society. The force of the symbols resurged unceasingly in their free trajectories, at the same time as their enduring implementations. 97

Prehistoric Art as Prehistoric Culture Cantabria, 2. Universidad de Cantabria: Santander. PubliCan-Ediciones: Santander Jaubert, J. & Barbaza, M. (eds) 2005. Territoires, déplacements, mobilité, échanges pendant la préhistoire. Terres et hommes du Sud. CTHS : Paris. Jaubert, J., Fourment, N. & Depaepe, P. (eds) 2013. Transitions, rupture et continuité en préhistoire. 2. Paléolithique et mésolithique. XXVIIe Congrès Préhistorique de France, Bordeaux-Les Eyzies, 2010. SPF, vol. 106. Mangado X. (ed.). 2010. El paleolítico superior peninsular. Novedades del siglo XXI. Homenaje al Professor Javier Fortea. Monografie 8. SERP: Barcelona. Société d’Etudes et de Recherches Archéologiques sur le Paléolithique de la Vallée de la Claise 2013. Le Solutréen… 40 ans après Smith 66. Actes du Colloque Preuilly-sur-Claise, 2007, 47e Supplément à la Revue Archéologique du Centre de la France, ARCHEAFERACF: Tours. Vialou, D. (ed.) 2003. La Préhistoire. Histoire et Dictionnaire, collection ‘ Bouquins ‘ Robert-Laffont: Paris.

98

Deep caves, ritual and graphic expression: a critical review of the archaeological evidence on hypogean human activity during the Upper Palaeolithic/Magdalenian Pablo Arias The Cantabria Institute for Prehistoric Research, University of Cantabria, Edificio Interfacultativo, Avda. Los Castros s/n, 39005 Santander, Spain

Abstract Palaeolithic rock art has frequently been related to ritual activity, especially when it is located in the inner part of karst systems. Yet evidence of rites in the vicinity of the Palaeolithic paintings and engravings is scarce. In this paper in honour of Professor Rodrigo de Balbín we attempt to explore this issue, focusing on the relationship between areas where reasonable evidence of Palaeolithic ritual activity has been found and Palaeolithic graphic expression. Particular attention is given to hypogean Middle Magdalenian structures in the Pyrenees and northern Spain that have been interpreted as ritual spaces. Although current evidence is not conclusive, it appears that there is no clear relationship between these areas and coterminous rock art. Keywords Palaeolithic rock art; prehistoric ritual; prehistoric religion; Pyrenean Magdalenian; Cantabrian Magdalenian

undemonstrated lieux communs in Palaeolithic cave art studies. His deep first-hand knowledge of prehistoric rock art (not only the Palaeolithic!) and his independent, often heterodox, views have been most stimulating for many researchers who, like myself, have followed his magisterial example. I am pleased to offer him, as a modest contribution to his homage, a revision and critical discussion of the evidence of Palaeolithic ritual activities and their relations with graphic expression.

Introduction The topic of this paper is a classic issue in rock art studies. Since the discovery of the paintings and engravings of Altamira and other caves in northern Spain and southwestern France in the late 19th and early 20th centuries, scholars have showed their concern for the location of Palaeolithic art inside the caves, often in completely dark – and sometimes in very deep – areas. As a matter of fact, until the relatively recent discovery of cases of Palaeolithic graphic expression in open-air places (Jorge et al. 1981; Martín & Moure 1981; Alcolea & Balbín 2000; Balbín & Bueno 2009), ‘rock art’ and ‘cave art’ were, for the Palaeolithic, perfectly synonymous terms. As everyday activities of the Palaeolithic huntergatherers are located in either the open air or the entrance of caves and rock shelters, the presence of paintings and engravings in deep areas tended to be attributed to ritual causes (without much discussion on what this exactly meant). Therefore, there is an abundant literature suggesting the performance of ritual activities related to rock art inside the caves or hypotheses on the progressive colonization of the deepest part of the karst in the course of the Palaeolithic (Leroi-Gourhan 1965: 114). However, the common treatment of this issue lacks the required critical approach. As we have suggested above, frequently it is just the location of graphic expression elements in odd, difficult places which is claimed as evidence of their ritual nature, without further discussion of the contexts or of the concept of ritual itself. Professor Rodrigo de Balbín has been one of the few relevant scholars in the field of rock art studies who have shown a critical position on this issue. He has showed his concern about the uncritical assumption that rock art was to be considered necessarily as a religious phenomenon, the conception of the caves as ‘sanctuaries’, and many other

What do we mean by ‘ritual’? As we have discussed in some detail elsewhere (Arias 2009), ritual is a complicated concept even for disciplines such as Social Anthropology which can observe human behaviour directly (Cazeneuve 1971; Huxley 1971; Izard & Smith 1979; Segalen 1998), not to mention for Archaeology. Even though interest in ritual has increased significantly in the last thirty years (Bradley 2005; Rakita & Buikstra 2008), the actual practice of archaeologists rarely goes beyond the rough approach that we have criticized above: the classification as ritual of anything that the observer considers odd or that cannot be understood (Demoule 2001; Richards & Thomas 1984; Bahn 1989: 69; Insoll 2004). In practice, little has been done to overcome the traditional, scarcely convincing, criteria proposed by A. Leroi-Gourhan for recognising religious behaviour in Prehistory as something which displays ‘concerns which appear to go beyond the material realm’ (Leroi-Gourhan 1986: 5). As we have argued before, this approach can hardly be applied to pre-industrial societies (if to any, according to Feldtkeller 2006), in which the practical and symbolical parts of social life are inextricably intermingled. The traditional, durkheimian distinction between sacred and profane (Durkheim 1912) has little meaning in prehistoric societies. Moreover, the simple statement that what we 99

Prehistoric Art as Prehistoric Culture cannot understand is the result of ritual activity contributes little to understanding the societies that we are studying. In general, the conception of ritual as an ideological manifestation of a non-practical or irrational nature usually leads to a blind alley. Once a particular archaeological item or context is classified as ‘ritual’ the enquiry ends, without further research into ideological or social aspects of the communities being studied.

(particularly in the European Palaeolithic, see for example Straus 1979), caves are not necessarily the best option for human activity. It is quite likely that their relevance within archaeological research is connected with their excellent conditions for the preservation and discovery of remains, rather than with their true importance as the location of the activities of past human groups. Without doubt, caves do have certain advantages, such as providing a readymade shelter from inclement weather, which could well have been a significant factor in colder phases of the Pleistocene. However, these advantages are counteracted by the disadvantages, such as the darkness, humidity or spatial constrictions that characterise the inner parts of caves. Nor can certain cultural factors be ignored: caves are frequently associated with a feeling of mystery, which often makes them the setting for myths and legends. Ii is not by chance that, in many cultures, caves are the home of worrisome beings, like fairies and goblins, or terrifying creatures (think of the terrible Cyclops Polyphemus in the Odyssey, or the dragon Fafnir in Germanic mythology). It is likely that this is connected with one of the most characteristic traits of caves: darkness. Our species, a diurnal primate, defenceless against its enemies in the night, may associate caves with terrible ancestral fears. In fact, in contrast with the image of prehistoric man as a caveman, so widespread in popular culture, we might wonder about the reasons why Palaeolithic hunter-gatherers chose to enter these kinds of places, and occasionally to visit them repeatedly. This is particularly true for the inner parts of the caves, which require permanent lighting, and are often relatively difficult for human transit. However, archaeological research has demonstrated that humans did explore very remote part of the caves since the Palaeolithic (Rouzaud 1978, 1996, 1997a, 1997b; Colomer et al. 2001; Moyes 2013).

However, some important theoretical and methodological contributions have been made in recent years. Let us highlight the work of J. Brück (1999) and R. Bradley (2003, 2005). Both of them, largely influenced by Catherine Bell’s approach (Bell 1992, 1997), reject the pertinence of distinguishing between practical and ritual realms, or the utilitarian nature of the former and the supposed irrationality of the latter. As these scholars observe, in pre-industrial societies, pragmatic and symbolic aspects of reality are interrelated inextricably, so it would be a meaningless effort to attempt to isolate objects, structures or ritual deposits. Thus, we are not obliged to classify archaeological items as either ritual or functional; they can be (and frequently are) both. As Bradley states, ‘Ritual and domestic life were not two halves of a single phenomenon, to be picked apart by the archaeologist. Instead they formed two layers that seem to have been precisely superimposed.’ According to this point of view, the classic question of how to identify ritual practice in archaeology is irrelevant and stops us tackling a much more important topic: the study of prehistoric rationality. If all the actions of the past make sense within specific coordinates of rationality, their archaeological testimony should allow us to attempt their study. A static and simple definition of ritual is probably meaningless. Rather, we should try a dynamic approach, assuming that it is likely that the social process of ritualization frequently took place in Palaeolithic societies, even in what we would define as normal, practical activities such as hunting, cooking or knapping flint. Rituals were constructed, as Bradley suggests, out of the materials of domestic life (Bradley 2005: 119). Elements taken from everyday activities seem to have been emphasised and acted out in the past. Our task, as archaeologists, is to analyse the material remains of the past in a way that permits us to reveal the ritual aspect of prehistoric activity. And since we have lost every link with the meaning of the symbols that were acting there, our only resource is contextual analysis. The key does not lie in the ‘irrational’ character (i.e. foreign to our rationality or incomprehensible to us) of the archaeological evidence, but rather other factors such as the stereotypical repetition of gestures or activities, their spatial location, the variations between coetaneous communities, etc. Caves as a human environment

Deep karst provides another advantage for the study of ritual. It usually makes it possible to overcome one of the main problems of recognising ritual activity, derived from the very nature of the archaeological record. A typical Palaeolithic layer is the result of the accumulation of varied residues of human activity during a quite long period of time (often several centuries), combined with post-depositional processes and other taphonomic phenomena. Furthermore, they can correspond to very different groups. As F. Bordes remarked long ago (Bordes 1975), in most cases it is illusory to reconstruct actual areas of activity in such sites, let alone individual actions. However, the inner part of the caves gives us a unique chance to isolate the ritualised part of human activity. Inside the karst, the density of archaeological items is usually much lower than in the entrance, and the cave itself provides a physical limit which can be judged to be identical or at least similar to the environment of the Palaeolithic hunter-gatherers.

It is an interesting paradox that, although karst environments have played a vital role in prehistoric research

We will use as a particularly interesting case study an unusual space-temporal concentration of possible 100

Arias: Deep caves, ritual and graphic expression Palaeolithic ritual evidence: Middle Magdalenian innercave contexts in the Pyrenees and Cantabrian Spain, which also share a high density of roughly coterminous rock art.

researchers working on the topic of Palaeolithic religion and ritual (Clottes 1989; Insoll 2004: 55-57; Arias 2009). The area interpreted as a sanctuary is located in the back part of a large chamber near the entrance of the cave. There, L. G. Freeman and J. González Echegaray (1981) observed two complex structures (Fig. 1) consisting, among other minor details, of large excavated areas filled with a succession of layers of sand and archaeological items (marine molluscs, pigments, bones in anatomical connection….) and covered with an earthen mound, formed by a succession of layers of offerings. One of them was covered by a huge limestone slab supported by large flat stones.

Ritual areas during the Magdalenian: the state-of-the-art As we have presented in more detail in a previous paper (Arias 2009), several contexts dating between ca. 17,500 and 14,000 cal BC (but the most important in the 15th millennium), and corresponding to the late Lower Magdalenian and the Middle Magdalenian of southern France and northern Spain, can reasonably be considered to be the remains of ritual areas. They share several traits that might indicate some kind of ritualised behaviour. The most striking is the presence of some kinds of nondwelling stone structures (drystone walls, upright stones, pavements), sometimes complemented by the excavation of ditches or large pits. They have been found in parts of the cave which are separated and often distant from the main occupation areas. Frequently the contexts are located in interior chambers of the respective cave systems, where darkness, spatial constrictions or ponding limited the possible activities of the human groups. Evidence has also been found suggesting that most of them were used during a short period, making the isolation of the Palaeolithic activities much more feasible.

However the most outstanding feature was a large block of limestone and sandstone (probably from an outcrop nearby) found in the middle of the southeastern side of structure 2, facing in. The excavators read a series of natural features and human modifications of that block as the deliberate representation of a head, with a half human, half feline face (Fig. 2). The archaeological content of these structures showed a substantial difference with that of the rest of the Lower Magdalenian layer of the site, and it also included very particular items of portable art, like a rib on which a hind was represented using a technique similar to the bone cutouts of the Pyrenean Middle Magdalenian (Freeman & González Echegaray 1984: 164).

Frequently, these anomalous structures or assemblages were deposited when the cave (or at least that part of the cave) was being occupied for the first time, or when it was re-occupied after a long period of abandonment. In most cases, the structures were located directly on top of thick sterile layers, suggesting a relationship with foundational rites (Clottes 1989; Corchón 1994).

Another site that has provided most interesting evidence of this kind is Las Caldas, in Asturias (Corchón 1992: 36, 43-44; 1994b: 249). In the furthest part from the cave entrance, at the end of Chamber II, a very peculiar deposit was observed at the base of the Middle Magdalenian, directly above a thick bed of sterile silt. Taphonomic studies carried out by S. Corchón’s team were able to infer that, when the material in question was deposited, this part of the cave was flooded, as suggested by the unusual patina seen on the surface of the bones (Corchón 1995: 48). In comparison with other Middle Magdalenian sectors in the cave, this assemblage was characterised by the presence of a small number of quite unusual and well preserved objects. They consisted of horse hemimandibles, some deer hemi-mandibles, an aurochs horn, bear teeth, and numerous large flint cores, some of which had hardly been used, and very few implements, although of great technical quality, manufactured on blades: scrapers and burins, retouched blades and thick retouched laminar flakes, some bifaces and a cleaver with double patina (evidence of re-knapping), equid incisors engraved with angular motifs and series of lines, perforated longitudinally, as well as engraved sandstone plaquettes, which in some cases had been burnt, with very peculiar types of animals, consisting of steppe-type fauna (reindeer, mammoth, rhinoceros, equids), and numerous anthropomorphs. Corchón emphasises that there were hardly any of the most common elements in Palaeolithic

The contents of those particular areas display some interesting regularities. At some sites (Praile Aitz, La Garma and Las Caldas) certain types of archaeological materials that are usually associated with everyday activities, such as lithic implements or hearths, prove scarce. Faunal assemblages also show unusual species composition and anatomical parts, with an intriguing overrepresentation of the horse, and of rare parts such as skulls or hemi-mandibles (see Bahn 1982 for a discussion of the role of the horse in the Pyrenean Magdalenian). We must also mention the presence – and sometimes high density – of portable art or objects of adornment, including evidence that some objects were manufactured there. It is worth mentioning that the graphic representations often include anthropomorphs and mixed beings with some human features. The most complex and elaborate case is that of El Juyo, in northern Spain. As a matter of fact it has been argued that it may have acted as a sanctuary (Freeman & González Echegaray 1981), a hypothesis that, although it cannot be said to be proved, has been considered by several 101

Prehistoric Art as Prehistoric Culture

Figure 1. El Juyo. Late Lower Magdalenian structures (from Freeman & González Echegaray 1981).

Figure 2. Limestone block interpreted as a mixed representation of the head of a human and a feline (from Freeman & González Echegaray 1981).

102

Arias: Deep caves, ritual and graphic expression Three stone structures are found next to the western wall of the chamber. They consist of roughly oblong areas demarcated by speleothems and large slabs, forming drystone walls, or placed vertically (Fig. 3). The analysis of the archaeological content of that area shows several unusual features. The distribution of archaeological items is strongly biased towards faunal remains with some strange features for the regional Magdalenian, such as the predominance of horse (Equus caballus), or the presence of the skull of an equid whose dome had been removed, or the appearance of two nearly complete skeletons of shelduck (Tadorna tadorna L.) whose bones do not present butchery marks or burnt surfaces, suggesting that the birds were left as whole bodies on the cave floor.

assemblages, such as knapping waste, lithic or bone implements, or bone splinters. The singularity of the objects that were found and their location in a flooded part of the cave led Corchón to interpret this assemblage as the result of the intentional deposit of selected items in the water. In her opinion, the assemblage went beyond the boundaries of ‘the merely utilitarian or functional’ (Corchón 1992: 44), and could be compared with the deposits at Erralla and some Middle Magdalenian sites in the Pyrenees (e.g. Enlène) with intentional deposits at the base or commencement of the occupation (Corchón 1994a, 1994b). Roughly coterminous with the previous site is the Lower Gallery of La Garma, a large and incredibly preserved Middle Magdalenian site (Arias & Ontañón 2012) where it is possible to observe an extraordinarily rich deposit, including several stone structures (Ontañón 2003; Arias et al. 2011). The evidence that we are going to discuss here is located in Zone IV, an area with a low ceiling some 130m from the entrance of the cave. The Magdalenian remains are concentrated there. The archaeological objects (mainly mammal bones, but also lithics, shells, charcoal, red ochre and portable art items) are distributed over an area of about 55 sq. m.

In contrast with the abundance of bones, the scarcity of the lithic assemblage is quite striking. Other outstanding traits are the high indices of objects of adornment and portable art items (bone or antler sculptures, engraved bones, and engraved slabs of limestone or calcite). Some evidence suggests that at least some of these were made in that part of the cave. Finally, in Erberua, the Lowest Gallery of the Isturitz karst, four small stone circular structures were identified in an

Figure 3. La Garma. Structure IVC.

103

Prehistoric Art as Prehistoric Culture on a Littorina shell recently found on the floor of a corridor above Zone III at La Garma. In others, it seems likely that we are facing a deliberate gesture, as with the bison leg bones found in Zone VI at La Garma (Fig. 4) (Arias et al. 2007-2008), or the Magdalenian cutouts from a deep sector of the Tito Bustillo karst (Balbín et al. 2003). Whether they are the remains of a deposit or the accidental result of human roaming inside the caves is hard to say, although the relative exceptionality of the latter appears to suggest that the first option is the more likely. However, we agree with Professor Balbín and colleagues when they state that they cannot be automatically considered to be the result of ritual activity.

area that was in total darkness during the Palaeolithic (Larribau & Prudhomme 1983, 1989). One of them, 50 cm in diameter, had a horse’s tooth inside it and there were ribs in the other (or others, as it is not made clear in the publications). Some bone and flint remains were found in the proximity of the structures. As in other galleries of the system (Labarge et al. 2015), numerous bone splinters were stuck into the walls. It is interesting to note that not a single piece of charcoal was found inside these constructions, which appears to discard the alternative interpretation of these structures as hearths, also discussed by Larribau and Prudhomme (1989). Other contexts dated in the same time span, such as Erralla (Altuna et al. 1985) and Praile Aitz I (Peñalver 2014) in the Basque Country, or Enlène (Bégouën et al. 1996), Le Mas d’Azil (Péquart & Péquart 1963), Fontanet (Clottes 1989), Le Portel (Baills 1997) and Labastide (Omnès 1982) in the French Pyrenees, share some traits with the sites described above (see Arias 2009 for details).

A particular case, which appears to be more clearly related to deliberate actions is the placing of objects in cracks and fissures in the wall, which has been described for numerous caves with Upper Palaeolithic (usually Magdalenian) occupations or cave art, such as Enlène, Erberua, Bédeilhac, Montespan, Labastide or Gargas (Bégouën & Clottes 1981: 158; Larribau & Prudhomme 1983: 283-84; Bégouën et al. 1996) and particularly Isturitz (Labarge et al. 2015). Evidence of that kind has also been found in several caves in Cantabrian Spain,

A problematic issue: ‘Palaeolithic spelunkers’ and ritual As mentioned above, evidence of a human presence in deep areas of karst is relatively abundant (Rouzaud 1978). Moreover, it is likely that the available information is just a small fraction of the actual documents that once existed. Remains of the Palaeolithic explorations were rarely observed in the early 20th century, when most of the large sites were discovered, or more recently in the course of sport caving. As a matter of fact, many of the cases of early recording and preservation of that kind of archaeological items are probably more related to the special sensitivity of some scholars and landowners such as Count Bégouën than to a much higher density of evidence. Among the features that help us document Palaeolithic incursions into the deepest parts of caves, we can highlight footprints, which are fairly frequent in southern France (Bédeilhac, Gargas, Massat, Labouiche, Tuc d’Audoubert, Fontanet, Niaux-Réseau Clastres, Montespan, Chauvet) and which have also been recorded in Palaeolithic contexts in northern Spain, such as in La Garma. They provide a very vivid, direct evidence of human gestures. Yet the interpretation of their meaning is particularly difficult. A long tradition in research tends to relate them to ritual, especially when they include some relatively odd traits such as evidence of people walking on their heels, or the presence of children, but there is no convincing evidence of that (see Owens & Hayden 1997 for a reappraisal of this issue). We can describe as indirect evidence of a human presence inside the cave the appearance of isolated archaeological items in deep zones of the karst. In some cases these might be attributed just to the accidental loss of objects by people exploring the cave, as with an isolated pendant

Figure 4. La Garma, Zone VI. Bison metapodial on the floor of the Lower Gallery.

104

Arias: Deep caves, ritual and graphic expression in the same cave, where a horse engraved in the clay was covered with supposed spear holes, were the most famous cases, but ritual was central to the interpretation of parietal art as a form of sympathetic magic.

such as Altxerri, and especially La Garma, where the chronology of these deposits might be attributed to the Middle Magdalenian. These deliberate gestures, which are concentrated in the vicinity of the Middle Magdalenian structures, include the setting in place of flakes, bones and speleothems, such as the recently observed example that we show in Fig. 5, a distal fragment of a stalagmite that was put in a narrow hole in a small chamber above Zone III with many Palaeolithic paintings and engravings. The phallic appearance of this object might have contributed to its selection.

The structuralist approach developed by researchers like A. Laming-Emperaire (1962) and A. Leroi-Gourhan (1965, 1986) abandoned the field of ritual, despite the latter’s interest in the study of prehistoric religion. Palaeolithic art remained within the realm of religious behaviour, but there was a shift from ritual to the symbolic meaning of the syntax of the images, i.e. from action to concept. From the notion of isolated ritual acts in certain parts of the cave, the cave itself came to be considered as a ‘sanctuary’, as a space where the interaction of the morphology of the passages and their walls with the cave’s art was endowed with a religious meaning (see Balbín & Alcolea 1999 for criticism of this concept).

In a similar vein, we might mention the deliberate breakage of speleothems (Le Portel), sculptures (Enlène, Isturitz) or plaquettes (Isturitz, Labastide, Massat, Le Tuc d’Audoubert, Les Trois-Frères, Le Mas d’Azil, Le Portel, Bédeilhac, Espalungue, Les Espélugues, Lortet, Gourdan, Enlène, the Spugo de Ganties, Roquecourbère, Labouiche, La Vache) (Clottes 1989: 311-13), or the incisions in clay at Le Tuc d’Audoubert.

An aspect of Leroi-Gourhan’s conception that deserves to be explored in greater detail is the notion that the cave itself is a relevant actor in the symbolic contents in which wall art also participates. From this viewpoint, graphic expression is considered only as part of a wider and more complex phenomenon. The profound meaning of an image would not be fully understood in isolation, but through its articulation with the shape of the wall, the play of light and shadows (Groenen 1997), the sounds and the movement of the people taking part in the rites. The cave, from this perspective, becomes, in LeroiGourhan’s words (1965), a ‘caverne participante’, not only because its forms are integrated with the graphic expression, but also as a theatre, a space which holds and configures the set of relationships of all the aspects involved. Archaeological research should therefore attempt to understand the role these spatial relationships played in the Palaeolithic, including dimensions which are particularly difficult to analyse from an archaeological perspective, such as sound (Reznikoff 2012; Till 2014). As a possible case from one of the sites that we are discussing, let us remember our suggestion that at La Garma there might be a relationship between some graphemes, such as a group of five hand stencils, and the deeply disturbing sound of the river in another level of the karst, which can only be heard from the place where those paintings are located (Arias et al. 2003).

And what happens, finally, with ‘art’? A favourite topic of the first decades in cave art research was the supposed relationship of the graphics to rites performed to propitiate successful hunting (Cartailhac & Breuil 1906). The clay figure, probably representing a bear, at Montespan, or the so-called ‘scène de chasse’

However, little evidence has been found of the performance of ritual activities in the vicinity of rock art. It is even unclear whether people had regular access to the areas where the paintings and engravings in the inner part of the karst are located. Among the scarce data on that issue, we have already mentioned the possible relationship between a Palaeolithic path and panels of rock art in an elevated passage above Zone I at La Garma (Arias et al. 2003) (Fig. 6). This suggests a relatively intense frequentation of that place, in spite of its relatively difficult location.

Figure 5. La Garma, Zone III. Stalagmite fragment in a hole in the wall of the cave.

105

Prehistoric Art as Prehistoric Culture

Figure 6. La Garma, Zone I. Palaeolithic path in a passage above Zone I.

shelters and the entrance of caves, close to sectors where everyday activities were performed. As Professor Balbín and colleagues have remarked (Balbín et al. 2003), we cannot uncritically assume a relationship of the rock art either with the religious realm or with ritual. This is an issue that must remain in the agenda of rock art analysts in the coming years!

From another point of view, no consistent relationship can be established between the Middle Magdalenian ritual areas described above and rock. In spite of the high density of Magdalenian rock art both in Cantabrian Spain and in the French Pyrenees, in only a few cases does it seem likely that coterminous paintings or engravings are associated with them. The only case of a convincing association is La Garma, where clearly Magdalenian engravings, representing horses, bisons and hinds, can be observed on the ceiling above the southern part of structure IV-B. Some claims have been made for the cases of Erberua and Mas d’Azil, but it appears that, generally, there is no association between that particular kind of ritual space and rock art. This appears to be significant because, on the one hand, during the Cantabrian Middle Magdalenian a statistical association exists between the densities of portable art and parietal art (Arias & Ontañón 2004), and, on the other, portable art is a constituent feature of those ritual areas.

Final remarks In this paper we have tried to demonstrate that the analysis of ritual in the Palaeolithic requires the development of a specific research programme. The existence of something ‘strange’ or simply beyond our understanding is not enough to propose the existence of ritual activity. In the case of rock art, a detailed analysis of the spatial relationship between the graphic elements on the walls and other kinds of archaeological evidence is strongly required. That includes direct and indirect evidence of the human activity in the surroundings of the ‘decorated’ panels, but also of the topography, at macro and microscales of the cave itself.

Of course, that does not mean that Magdalenian rock art has no relationship at all with ritual. It is simply that the issue is much more complicated. Rock art is located, certainly, in deep, sometimes hidden, areas inside the karst, but is also in public areas, such as rock

We have also tried to show that the inner parts of caves provide an isolated environment, particularly suitable 106

Arias: Deep caves, ritual and graphic expression l’Adour: Royaume du Cheval. Musée Pyrénéen: Lourdes. Bahn, P. G. 1989. Bluff your way in Archaeology. Ravette Books: Horsham. Baills, H. 1997. La relation sanctuaire-habitat: le cas de la grotte du Portel (Ariège). Quaternaire 8 (2-3): 225-32. Balbín, R. de & Alcolea, J. J. 1999. Vie quotidienne et vie religieuse. Les sanctuaires dans l’art paléolithique. L’Anthropologie 103: 23-49. Balbín, R. de, Alcolea, J. J. & González Pereda, M. A. 2003. El macizo de Ardines, Ribadesella, España. Un lugar mayor del arte paleolítico europeo, pp. 91-151 in (R. de Balbín & P. Bueno, eds) Arte prehistórico desde los inicios del s. XXI. Primer Symposium Internacional de Arte Prehistórico de Ribadesella. Asociación Cultural Amigos de Ribadesella: Ribadesella. Balbín, R. de & Bueno, P. 2009. Altamira, un siècle après: art paléolithique en plein air. L’Anthropologie 113: 602-28. Bégouën, R. & Clottes, J. 1981. Apports mobiliers dans les Cavernes du Volp (Enlène, les Trois-Frères, le Tuc d’Audoubert), pp. 157-88 in Altamira Symposium. Ministerio de Cultura: Madrid. Bégouën, R., Clottes, J., Giraud, J. P. & Rouzaud, F. 1996. Os plantés et peintures rupestres dans la caverne d’Enlène, pp. 283-306 in (H. Delporte & J. Clottes, eds) Pyrénées Préhistoriques. Arts et Sociétés. 118e congrès national des sociétés historiques est scientifiques. Pau 1993. CTHS: Paris. Bell, C. 1992. Ritual Theory, Ritual Practice. Oxford University Press: Oxford. Bell, C. 1997. Ritual: Perspectives and Dimensions. Oxford University Press: Oxford. Bordes, F. 1975. Sur la notion de sol d’habitat en préhistoire paléolithique. Bulletin de la Société Préhistorique Française 72 (5): 139-44. Bradley, R. 2003. A life less ordinary: the ritualization of the domestic sphere in later prehistoric Europe. Cambridge Archaeological Journal 13: 5-23. Bradley, R. 2005. Ritual and domestic life in Prehistoric Europe. Routledge: London. Brück, J. 1999. Ritual and rationality: some problems of interpretation in European archaeology. European Journal of Archaeology 2 (3): 313-44. Cartailhac, E. & Breuil, H. 1906. La Caverne d’Altamira à Santillane près Santander (Espagne). Imprimerie de Monaco: Monaco. Cazeneuve, J. 1971. Sociologie du Rite. Presses Universitaires de France: Paris. Clottes, J. 1989. Le Magdalénien des Pyrénées, pp. 281-357 in (J-P. Rigaud ed.) Le Magdalénien en Europe. ‘La structuration du Magdalénien’. Actes du Colloque de Mayence 1987. Université de Liège: Liège. Colomer, A., Galant, P. & Ambert, P. 2001. Incursions spéléologiques mésolithiques dans la grotte

for delimiting many of the practical problems arising in the study of ritual among Palaeolithic societies. Only through the conjunction of high-quality empirical documentation and a correct methodological approach will it be possible to make significant progress in the research on such a complex topic.

References Alcolea, J. J. & Balbín, R. de 2000. Arte paleolítico al aire libre. El yacimiento rupestre de Siega Verde, Salamanca. Junta de Castilla y León: Valladolid. Altuna, J., Baldeón, A. & Mariezkurrena, K. 1985. Cazadores magdalenienses en la cueva de Erralla (Cestona, País Vasco). Sociedad de Ciencias Aranzadi: San Sebastián. Arias, P. 2009. Rites in the dark? An evaluation of the current evidence for ritual areas at Magdalenian cave sites. World Archaeology 41 (2): 262-94. Arias, P., González Sainz, C., Moure, A. & Ontañón, R. 2003. Unterirdischer Raum, Wandkunst und paläolithische Strukturen. Einige Beispiele der Höhle La Garma (Spanien), pp. 29-46 in (A. Pastoors & G. Weniger, eds) Höhlenkunst und Raum: Archäeologische und architektonische Perspektiven. Neanderthal Museum: Mettmann. Arias, P. & Ontañón, R. 2004. El contexto del arte mobiliar paleolítico en la región Cantábrica, pp. 3752 in (P. Arias & R. Ontañon, eds) La Materia del Lenguaje Prehistórico. El arte mueble paleolítico de Cantabria en su contexto. Consejería de Cultura, Turismo y Deporte del Gobierno de Cantabria: Santander. Arias, P. & Ontañón, R. 2012. La Garma (Spain): Longterm human activity in a karst system, pp. 101-17 in (K. A. Bergsvik & R. Skeates, eds) Caves in Context: the cultural significance of caves and rockshelters in Europe. Oxbow: Oxford. Arias, P., Ontañón, R., Álvarez Fernández, E., Cueto, M., Elorza, M., García-Moncó, C., Güth, A., Iriarte, M.J., Teira, L.C. & Zurro, D. 2011. Magdalenian floors in the Lower Gallery of La Garma. A preliminary report, pp. 31-51 in (S. Gaudzinski-Windheuser, O. Jöris, M. Sensburg, M. Street & E. Turner, eds) Site-internal spatial organization of hunter-gatherer societies: case studies from the European Palaeolithic and Mesolithic. Verlag des Römisch-Germanischen Zentralmuseums: Mainz. Arias, P., Ontañón, R., Álvarez Fernández, E., Cueto, M., García-Moncó, C. & Teira, L. C. 2007-2008. Falange grabada de la Galería Inferior de La Garma: aportación al estudio del arte mobiliar del Magdaleniense Medio, pp. 97-129 in (J. Fernández Eraso & J. Santos, eds) Homenaje a Ignacio Barandiarán Maestu. Servicio editorial de la Universidad del País Vasco: Vitoria. Bahn, P. G. 1982. Homme et cheval dans le Quaternaire des Pays de l’Adour, pp. 21-26 in Les Pays de 107

Prehistoric Art as Prehistoric Culture d’Aldène (Cesseras, Hérault). Bulletin de la Société Préhistorique Française 98 (3): 497-503. Corchón, M. S. 1992. La cueva de Las Caldas (Priorio, Oviedo). II Investigaciones efectuadas entre 1987 y 1990, pp. 33-47 in Excavaciones arqueológicas en Asturias 1987-90. Servicio de Publicaciones del Principado de Asturias: Oviedo. Corchón, M. S. 1994a. Arte mobiliar e industria ósea solutrense en la Cornisa Cantábrica. Férvedes 1: 13148. Corchón, M. S. 1994b. Últimos hallazgos y nuevas interpretaciones del arte mueble paleolítico en el Occidente asturiano. Complutum 5: 235-64. Corchón, M. S. 1995. La cueva de Las Caldas (Priorio, Oviedo). III. Resultados preliminares de las excavaciones (campañas 1991-1994), pp. 45-60 in Excavaciones arqueológicas en Asturias 1991-94. Servicio de Publicaciones del Principado de Asturias: Oviedo. Demoule, J. 2001. Archaeology of cult and religion: a comment, or how to study irrationality rationally, pp. 279-84 in (P. F. Biehl & F. Bertemes, eds) The Archaeology of Cult and Religion. Archaeolingua: Budapest. Durkheim, E. 1912. Les formes élementaires de la vie religieuse: le système totémique en Australie. Alcan: Paris. Feldtkeller, A. 2006. Warum denn Religion? Eine Begründung. Gütersloher Verlagshaus: Gütersloh. Freeman, L. G. & González Echegaray, J. 1981. El Juyo: a 14,000-year-old sanctuary from northern Spain. History of Religions 21 (1): 1-19. Freeman, L. G. & González Echegaray, J. 1984. Magdalenian structures and sanctuary from the cave of El Juyo (Igollo, Cantabria, Spain), pp. 39-49 in ( J. Berke, J. Hahn & C .J. Kind, eds) Jungpaläolithische Siedlungstrukturen in Europa. Institut für Urgeschichte: Tübingen. Groenen, M. 1997. Ombre et lumière dans l’art des grottes. Université Libre de Bruxelles: Brussels. Huxley, J. 1971. Le comportement rituel chez l’homme et l’animal. Gallimard: Paris. Insoll, T. 2004. Archaeology, Ritual, Religion. Routledge: London. Izard, M. & Smith, P. (eds) 1979. La Fonction symbolique. Gallimard: Paris. Jorge, S. O., Jorge, V. O., Almeida, C. A. F., Sanches, M. d. J. & Soeiro, M. T. 1981. Gravuras rupestres de Maxouco (Freixo da Espada a Cinta). Arqueologia, Porto 3: 3-12. Labarge, A., Rivero, O., Barhay-Szmidt, C., Normand, C. & Garate, D. 2015. Dépôts en paroi dans la grotte d’Isturitz (Pyrénées-Atlantiques): vers une définition des procédures d’une démarche singulière. Arkeos 37: 495-98. Laming-Emperaire, A. 1962. La signification de l’art rupestre paléolithique: méthodes et applications. Picard: Paris.

Larribau, J. & Prudhomme, S. 1983. La grotte ornée d’Erberua (Pyrénées-Atlantiques). Note préliminaire. Bulletin de la Société Préhistorique Française 80 (9): 280-84. Larribau, J. & Prudhomme, S. 1989. La grotte d’Erberua (Pyrénées-Atlantiques), pp. 65-67 in L’Art pariétal paléolithique: Étude et conservation. Colloque International. Périgueux-Le Thot 19-22 novembre 1984. Ministère de la Culture, de la Communication, des Grands Travaux et du Bicentenaire: Paris. Leroi-Gourhan, A. 1965. Préhistoire de l’Art Occidental. Lucien Mazenod: Paris. Leroi-Gourhan, A. 1986. Les Religions de la Préhistoire. 5th ed. Presses Universitaires de France: Paris. Martín, E. & Moure, A. 1981. El grabado de estilo paleolítico de Domingo García (Segovia). Trabajos de Prehistoria 38: 97-105. Moyes, H. (ed.) 2013. Sacred Darkness: a global perspective on the ritual use of caves. University Press of Colorado: Boulder. Omnès, J. 1982. La grotte ornée de Labastide (HautesPyrénées). Omnès: Lourdes. Ontañón, R. 2003. Sols et structures d’habitat du Paléolithique supérieur, nouvelles données depuis les Cantabres: la Galerie Inférieure de La Garma (Cantabrie, Espagne). L´Anthropologie 107: 333-63. Owens, D. & Hayden, B. 1997. Prehistoric rites of passage: a comparative study of transegalitarian hunter-gatherers. Journal of Anthropological Archaeology 16 (2): 121-61. Peñalver, X. 2014. Praileaitz I haitzuloa = la cueva de Parileaitz I. Gipuzkoako Foru Aldundia: Donostia/ San Sebastián. Péquart, M. & Péquart, S. 1963. Grotte du Mas d’Azil (Ariège): une nouvelle galerie magdalénienne. Annales de Paléontologie XLIX: 3-98. Rakita, G. F. M. & Buikstra, J. E. 2008. Feather waving or the numinous?: archaeological perspectives on ritual, religion, and ideology, pp. 1-17 in (G. F. M. Rakita & J. E. Buikstra, eds) An Archaeological Perspective on Ritual, Religion, and Ideology from American Antiquity and Latin American Antiquity. The Society for American Archaeology: Washington. Reznikoff, I. 2012. L’existence de signes sonores et leur signification dans les grottes paléolithiques, pp. 300-301 in (J. Clottes, ed.) L’art pléistocène dans le monde / Pleistocene art of the world / Arte pleistoceno en el mundo. Actes du Congrès IFRAO, Tarascon-sur-Ariège, septembre 2010. Société Préhistorique Ariège-Pyrénées, Tarascon-sur-Ariège; CD: 1741-47. Richards, C. & Thomas, J. 1984. Ritual activity and structures deposition in later Neolithic Wessex, pp. 189-218 in (R. Bradley & J. Gardiner, eds) Neolithic Studies: a review of some recent work. British Archaeological Reports: Oxford. Rouzaud, F. 1978. La Paléospéléologie. L’homme et le milieu souterrain pyrénéen au Paléolithique 108

Arias: Deep caves, ritual and graphic expression supérieur. École des Hautes Études en Sciences Sociales: Toulouse. Rouzaud, F. 1996. La paléospéléologie: une méthode d’étude des grottes préhistoriques et paléontologiques, pp. 143-48 in (H. Delporte & J. Clottes, eds) Pyrénées Préhistoriques. Arts et Sociétés. C.T.H.S.: Paris. Rouzaud, F. 1997a. La paléospéléologie, pp. 49-52 in Proceedings of the 12th International Congress of Speleology. La Chaux de Fonds, 10-17 August 1997. Speleo Projects: Basel. Rouzaud, F. 1997b. La paléospéléologie ou : l’approche globale des documents anthropiques et paléontologiques conservés dans le karst profond. Quaternaire 8 (2-3): 257-65. Segalen, M. 1998. Rites et Rituels Contemporains. Nathan: Paris. Straus, L. G. 1979. Caves: a palaeoanthropological resource. World Archaeology 10 (3): 331-39. Till, R. 2014. Sound archaeology: terminology, Palaeolithic cave art and the soundscape. World Archaeology 46 (3): 292-304.

109

110

Magdalenian settlement-subsistence systems in Cantabrian Spain: contributions from El Mirón Cave Lawrence G. Straus,¹ ² Manuel González Morales,² Ana B. Marín-Arroyo² and Lisa M. Fontes¹ Department of Anthropology, University of New Mexico, Albuquerque, NM 87131, USA Instituto Internacional de Investigaciones Prehistóricas de Cantabria, Universidad de Cantabria, 39005 Santander, Spain 1

2

Abstract This paper describes aspects of human adaptations to late Last Glacial environments in Cantabrian Spain based on excavations in El Mirón Cave in the Cantabrian Cordillera, and on analyses of data from other sites in the Asón basin of Cantabria, as well as from several recent excavations of Magdalenian (20-13k cal. BP) sites throughout this narrow, high-relief, coastal region. Keywords Magdalenian; Late Glacial; Cantabrian Spain; settlement-subsistence systems

Upper Magdalenian occupation in the montane zone (Ruiz 2007).

The Magdalenian of the Río Asón Valley The Asón River is located at the eastern end of the Region of Cantabria (Figure 1). At 40 km, it is one of the longer rivers in the region and (under Holocene interglacial conditions) has one of the largest estuaries. Equidistant between Santander and Bilbao, the Asón was relatively far from centres of archaeological activity and thus, despite several very early 20th-century discoveries (Covalanas, La Haza and El Mirón caves – all in the upper valley) and even an excavation (El Valle – in the middle valley), the drainage was significantly under-researched. Two exceptions were the Magdalenian sites of La Chora and El Otero caves on the coastal plain of the lower valley, excavated by J. González Echegaray et al. (1963, 1966) in the early 1960s, with then-new interdisciplinary analyses of fauna and pollen. The drainage basin once again fell into a state of archeological abandonment until González Morales (2000; González Morales et al. 2000a, b) undertook a research project around the Asón estuary in the early 1990s, including small-scale Azilian and Magdalenian excavations in El Perro Rockshelter and La Fragua Cave (Marín-Arroyo & González Morales 2007). Subsequent to the beginning of the El Mirón Project, test pits were dug in El Valle (García-Gelabert & Talavera 2004) and a small excavation was conducted in El Horno Cave, at the foot of the cliff in the opposite side of the mountain in which El Mirón, La Haza and Covalanas are located (Fano 2005; Costamagno & Fano 2006); both sites have Upper Magdalenian and Azilian levels. Earlier overviews of the late Upper Palaeolithic human settlement of the Asón drainage basin can be found in Straus et al. (2002a, b). During testing of the site of El Polvorín Cave just inside Vizcaya in the Río Carranza gorge (a major Asón tributary), two antler harpoons and a sagaie were found, indicating yet another

The Magdalenian occupations of El Mirón Cave El Mirón is a large, west-facing cave, located on a cliff at 260 m above present sea level and 100 m above the valley floor of the Ríos Calera and Gándara, tributaries of the upper Asón at Ramales. It is on the edge of the Cantabrian Cordillera, surrounded by summits ≥1000 m. The site is at a crossroads between important routes North-South between the coast at the mouth of the Asón and the northern meseta of Old Castile (via the 900 m-high Los Tornos Pass) and East-West between Vizcaya and central Cantabria via the 675 m-high Alisas Pass. El Mirón is 20 km from the present shore and would have been about 25-30 km from the Late Glacial littoral. It is surrounded by ibex habitat (steep rocky slopes), but is also at the edge of the Asón’s broad Ruesga Valley. whose floor and lower slopes would have been excellent red deer habitat, while the river was rich in salmon. Magdalenian deposits have been uncovered in excavations since 1996 throughout the vestibule (30 m deep x 8-16 m wide x 13-20 m high). In the outer vestibule, vestibule rear and mid-vestibule trench occupation layers (and a human burial [Straus et al. 2011, 2015a]) have been excavated over a total of about 21 m², although not continuously. Separate stratigraphic designations have thus been assigned to levels in the various excavation areas. The Magdalenian (and Azilian) levels have been dated by 56 radiocarbon assays. The Late Glacial levels are divided among Initial (17-16 uncal. kya/20.2-19.3 cal kya), Lower (16-14.5 uncal kya/19.3-17.5 cal kya), Middle (14.5-13 uncal kya/17.5-15.6 cal kya) and Upper Magdalenian (13-11.5 uncal kya/15.6-13.7 cal kya) 111

Figure 1. Major Magdalenian Sites and Flint Sources in Cantabrian Spain (L. G. Straus & R. Stauber) Sites: 1. Isturitz; 2. Aitzbitarte; 3. La Fragua, El Perro, El Otero, La Chora; 4. Ekain, Amalda; 5. Ermittia, Urtiaga, Praile Aitz; 6. Lumentxa, Santa Catalina; 7. Santimamiñe, Antoliñako; 8. Bolinkoba; 9. Arlanpe; 10. El Mirón, El Horno, El Valle; 11. La Garma; 12. El Rascaño; 13. Morín; 14. El Juyo, El Pendo; 15. El Castillo; 16. La Pila; 17. Altamira, Cualventi, Linar; 18. La Riera, Cueto de la Mina, Balmori; 19. Tito Bustillo, El Cierro, Cova Rosa; 20. Coimbre, Llonín; 21. La Guëlga; 22. La Paloma; 23. Las Caldas; 24. La Viña. Flint sources: C: Chalosse; SB: Salies de Béarn; B: Bidache*; BZ: Biarritz*; G: Gaintxurizketa*; U: Urbasa; L: Loza; T: Treviño; BK: Barrika (Soplana/Kurtzia)*; S: Sonabia; LL: Llaranza; MP: Monte Picota; P: Piloña; P: Piedramuella. Based on the work of P. Sarabia, A. Tarriño and J. Rissetto. (*=flysch)

Prehistoric Art as Prehistoric Culture

112

Straus et al: Magdalenian Settlement-Subsistence Systems in Cantabrian Spain and Azilian (11.5-10 uncal kya/13.7-12 cal kya) (see Straus & González Morales 2012; Straus et al. 2015b). The Initial, Lower and Middle Magdalenian levels (119.3-108 for this whole sequence in the vestibule rear and 21-13 in the outer vestibule; 312-309 for the Lower and Middle Magdalenian in the mid-vestibule test pit; 505-503.1 for the Lower Magdalenian in the burial area of the vestibule rear) are dense, rich habitation layers unbroken by either sterile or even lower-density layers. They are clearly palimpsests formed by frequently repeated (but necessarily not continuously, as suggested by the formation of manganese oxide coating on bones in the vestibule rear [Marín-Arroyo et al. 2008]), largescale, multi-functional occupations of the cave with pits, hearths, charcoal from fires, fire-cracked rocks (Nakazawa et al. 2009) and other ‘manuports’ such as hammers and anvils, patches of ochre, abundant, diverse chipped stone and osseous artifact and faunal assemblages. In contrast, the Upper Magdalenian and Azilian levels (106-102.1 in the vestibule rear; 308305 in the mid-vestibule; 12-11 in the outer vestibule) are poor in artifacts, manuports, features and fauna, attesting to short, limited-function visits to the cave. Clearly there was a shift in the role of El Mirón within the territory of the valley between Oldest Dryas-BøllingOlder Dryas (i.e., Greenland Stadial 2 and Greenland Interstadial 1e-d) times and Allerød-Younger Dryas (i.e., Greenland Interstadial 1c-a and Greenland Stadial 1) times. Interestingly, two valley-floor sites in the middle and upper Asón drainage, one large (El Valle) and the other small (El Horno), have major Upper Magdalenian and Azilian deposits.

perforators, continuously retouched blades, etc., but also sidescrapers, denticulates and notches on large flakes of local, non-flint stones, as in the Initial Magdalenian. The abundance of ‘archaic’ or ‘Mousteroid’ flake tools in many Cantabrian Upper Palaeolithic assemblages is not exclusive to the Initial Magdalenian, but is widespread, especially in areas with little or no local high-quality flint (Asturias and central-western Cantabria); these artifact types are not markers of any particular industry or cultural phase of Upper Palaeolithic age, but do reflect on both activities and access to lithic raw materials at certain sites. The antler points in the El Mirón Lower Magdalenian levels are often quadrangular in cross-section and frequently have geometric engraved decorations that are characteristic of this period in the region. There is an antler spear-thrower that is similar to ones from sites of similar age in SW France (Roc de Marcamps and Le Placard). Perforated teeth and shells are abundant and there are red deer scapulae with engravings, including one whole one with a striated image of a red deer hind absolutely typical of the Lower Magdalenian of Cantabria and eastern Asturias (González Morales & Straus 2009; Straus et al. 2008). Much of the same holds true for the Middle Magdalenian levels, but without the temporally restricted engraved scapulae, while the Upper Magdalenian ones emphasize the small flint bladelet tools and micropoints and endscrapers. This tendency also characterizes the Azilian levels, with backed micropoints and thumbnail and other small endscrapers and generally numerically poor, lowdiversity assemblages. Bone tools are rare in these levels, with only one round-section harpoon fragment in Upper Magdalenian Level 12 (13.0 uncal. kya) at the vestibule front, plus a harpoon barb from a small, loose, remnant deposit near the 1996 ground surface at the vestibule rear–Level 502 (Figures 2a & 2b), and no harpoons in the Azilian ones (González Morales & Straus 2012). An ochre-stained cobble in Level 11.1 would seem to mark this as an Azilian level by definition, at an average C14 age of c. 11.3 uncal. kya.

The Initial Magdalenian levels of El Mirón are characterized by ‘bi-focal’ lithic artifact assemblages: backed and retouched bladelets (and other classic, generally small Upper Palaeolithic tools) on highquality, non-local flints and large, flake-based ‘archaic’ tools — denticulates, notches and sidescrapers — made on local non-flint raw materials [mudstone, quartzite, limestone]). Osseous artifacts include large, often undecorated, round-section sagaies (spear-points) and needles, as well as some perforated shells and teeth (Straus et al. 2014). Badegoulian diagnostic artifacts (both lithic and osseous) are lacking (or nearly so).

Lithic Raw Materials Many types of lithic raw materials have been identified at El Mirón, including mudstones, limestones, quartzes, quartzites, calcites, and flints. Mudstone, quartzite and limestone are available locally — the former two in the form of cobbles in the bedloads of the rivers below the cave and probably transported from ancient alluvial terraces in the Cordillera. As to the third material, it may be even more strictly local since, of course, the cave is developed in the hard Lower Cretaceous limestone of Monte Pando. On the other hand, the Magdalenian levels are replete with debris and tools made on flints that are nothing like the small quantities of poor-quality ones that can be found locally in the limestone of the Gándara and Carranza gorges.

The Lower Magdalenian levels are enormously rich in unretouched and backed bladelets fabricated on small pyramidal or prismatic cores of excellent-quality nonlocal flint and there are consistently small numbers of microlithic triangles (all presumably weapon tips, barbs or cutting edge elements set into antler points – some of which have longitudinal grooves made for that purpose). Many of the bladelet cores seem to have been regularized along one or more platform edges possibly for use as scrapers – a clear diagnostic type for the Cantabrian Lower Magdalenian. There are also other kinds of endscrapers on blades and flakes, simple burins on breaks, some simple 113

Prehistoric Art as Prehistoric Culture With an emphasis on at least initial ‘splitting’, we empirically created 44 flint types. These were defined on the bases of colour, texture, inclusions, banding, fracture characteristics, translucence/opacity, cortex type, etc., and are represented by samples in our lithic comparative collection. Many of these types are very distinctive but extremely rare, and may occur in only one level and/or area of the site (perhaps the result of just one or a few exotic nodules having been brought to the site and then knapped); others are more widespread within the assemblages, but not very common; still others (pertaining to two large type-groups: A and B) are of excellent quality and are very abundant among all the Magdalenian levels. Most bladelet cores and bladelets are made of these materials (especially flints A, which come in various shades of gray, with or without banding, generally opaque or only slightly translucent and with few, small inclusions).

30 km from El Mirón. While the Asón valley may have been the centre of their territory (an area of some 300400 km², including the then-exposed continental shelf), these flint outcrops (and perhaps others as yet unknown and possibly under-water off the mouth of the river) were very important places at the edges of the habitually utilized ‘local’ catchment of the Mirón ‘Magdalenians’. The quantities of non-local flint groups A and B at the site are so large that they were almost certainly brought to the site by the inhabitants — either directly or in the course of successive movements between the coastal zone and the upper Asón — not traded ‘down-the-line’ from band to band. The regular presence of small numbers of both perforated and non-perforated marine mollusc shells in the Magdalenian (and Solutrean) levels attest to regular human movements from the shore to the site. Although the cores and chunks (angular debris/core remnants) are generally rather small (attesting to their ‘exhaustion’), they are very abundant and their products (including cortical flakes and blade[let]s) are overwhelmingly common, again suggesting direct procurement. There are clear knapping areas, such as at the rear of the vestibule where, in the 2.5 m² area of the Lower Magdalenian burial in Level 504, we identified masses of small, standardized bladelets on excellent-quality Barrika flint — some backed, but most unretouched — associated with bladelet cores (Fontes et al. 2015). However, not all flints come from within the ‘local’ catchment zone around El Mirón. Some flints, including from Treviño (≥90 km southeast of the site across the Cordillera), Chalosse (~250 km away in the southwest corner of France), and Urbasa (~110 km from Mirón in northern Navarra on the southern flank of the Cordillera) reached the site as a result of either longdistance trips taken by individual residents or perhaps down-the-line exchange among neighbouring bands. Such exotic flints define a world different from the subsistence territory of a local band, but rather a social territory in which networks of relationships are made manifest archaeologically by stones, while in life they may have been important in the procurement of mates, alliances, information and a safety net. Additionally, these non-local flints circulated throughout the Cantabrian region during the Lower Magdalenian; Fontes has identified them in assemblages from Altamira, El Juyo, and El Rascaño, three other significant LCM palimpsests. The widespread presence of these exotic flints testifies to the importance of inter-group contacts — locally and distantly — in maintaining the ‘Magdalenian world’.

What is known about the procurement of lithic raw materials at El Mirón is the result of prospecting by project members both locally and more widely, and then the dissertation research by John Rissetto (2009). Pedro Sarabia and Andoni Tarriño, specialists on the flints of Cantabria and the Basque Country respectively, have also helped match identified flints with their outcrops. Fontes has also recently compared these visually distinct flints with geological samples collected by Rissetto (2009) and a similar reference collection created by Tarriño (2012) for Aitzbitarte III (Guipúzcoa), geographically attributing approximately half of the visually distinct flints to outcrops in northern Spain and southwest France. Group A flints come from the huge Upper Cretaceous flysch formation along the Atlantic coast due north of Bilbao at Barrika. This is the famous outcrop at Soplana, where there are open-air quarry-workshop sites dating back to the Mousterian and Early Upper Palaeolithic periods. Today there is a cliff that is beaten by the waves of the ocean; in Magdalenian times it would have been a cliff that dominated a relatively narrow band of emerged continental shelf. Barrika and other coastal outcrops of flysch along the Basque coast up to Biarritz in France and inland up the Gaves river to Bidache were clearly major places in the Palaeolithic landscape of both Gascony and Vasco-Cantabria (see Tarriño 2006). Other good quality flints in eastern Cantabria include Upper Cretaceous outcrops at Sonabia Point east of the mouth of the Asón and at the foot of sea cliffs at Llaranza (Group B flints) west of the Asón. Magdalenian occupants of El Mirón collected flints from both of these outcrops. In the case of Barrika, the alternative routes would have been north via the Asón valley and then east along the now-drowned continental shelf (a distance of some 65-70 km) or eastward up the Carranza valley into the territory of Vizcaya and then northeastwardly via the Cadagua or Barbadún rivers to the Nervión at the area of Bilbao and on to Barrika (a distance of 50-60 km). Llaranza is about 45 km and Sonabia about

Osseous artifacts and portable ‘art’ It is interesting to note that at certain times and in certain aspects (and with the caveat that we have excavated only a small percentage of the vast habitable area of the vestibule), El Mirón dwellers seem to have participated in regional or perhaps even extra-regional social networks, as manifested by designs on antler points and by portable art styles, while at other times they seemingly did less or 114

Straus et al: Magdalenian Settlement-Subsistence Systems in Cantabrian Spain not. In the case of the Initial Magdalenian, there are few points of comparison in the region, as modern-quality excavations of deposits independently dated to this period are very scarce, although cultural materials of this age were probably found in several old excavations where they had not been separated from overlying Lower Magdalenian ones (e.g., El Castillo, Urtiaga) and a few new sites with levels of the relevant age are currently being excavated (e.g. Arlanpe, Coimbre). There are some similarities between large, round-section antler points in the Initial Magdalenian of El Mirón and in Level 5 of El Rascaño (16.4 kya uncal BP) (Barandiarán 1981) and especially 1.45-2.0 m-thick (!) basal Level 8 (Magdalenian β) in El Castillo dug by Obermaier in 1911 (Cabrera 1984, e.g. Fig. 149) , where such a sagaie has been directly dated to 16.85 kya uncal BP (Barandiarán 1988). The latter object has a kind of a chevron or ‘wheat-sheaf’ engraved design, reminiscent of the most spectacular large, round-section sagaie from Level 119 in El Mirón, as well as one from Cova Rosa in eastern Asturias (Barandiarán 1973, Plate 26.5; Utrilla 1981: 177). There are wheat-sheaf motifs on objects from other sites (e.g. Lumentxa in Vizcaya), but their dating is uncertain. There are other large, round-section sagaies from El Castillo’s Magdalenian β horizon, but it is not known whether they too may have come from the basal part of this horizon that also includes many classic Lower Magdalenian diagnostic artifacts (e.g. scapulae with striation-engraved images of hinds, quadrangular section sagaies with tectiform engraved designs). El Mirón, El Rascaño and El Castillo are located in three contiguous valleys (the Asón, Miera and Pas); El Mirón and El Castillo are about 50 km from one another. One could imagine a social territory consisting of these three valleys in central-eastern Cantabria within which people may have preferentially cooperated in group hunts, conducted rituals, exchanged information, sought mates, swapped band membership, etc., but this is only an hypothesis at this stage of our knowledge (see Sauvet [2014]).

Most of these objects come from the valleys of the Asón, Miera, Pas and Saja. Radiocarbon dates from Altamira, El Mirón and El Juyo indicate that they were made between about 15.5-14.5 uncal kya. Images very similar to the striation engravings of hinds and other ungulates on shoulder blades are found on the walls of such Cantabrian caves as Altamira, El Castillo, La Garma and Los Emboscados (near El Mirón), and further west in the Asturian sites of Llonín and Candamo. This is not an artifact type or style commonly found in the Spanish Basque Country or in France. Associated with the scapulae and also in many other sites in Cantabria and Asturias are many quadrangular cross-section sagaies with geometric or ‘tectiform’ (triangles, zigzags) engravings and nucleiform endscrapers. These are very frequent in El Mirón, some being nearly identical in design motifs to classic ones on sagaies from El Castillo and Altamira (González Morales & Straus 2005). The distribution of these objects and design styles may correspond to the territory of a regional band — a collection of individuals, families and local bands with a density of social connections and relations more frequent than those that may have occurred between the Cantabro-Asturian bands and those of Euskadi and SW France. Although the El Mirón people of the Lower Magdalenian seem to have procured much of their best flint from the Barrika outcrop in western Bizcaya, there is no evidence of sites with the aforementioned osseous diagnostics east of the Asón. On the other hand, it is possible, based on the morphological and metrical characteristics of the spear-thrower from El Mirón Level 17, that contacts existed with groups in SW France, where very similar atl-atls have been found (González Morales & Straus 2009). The presence of perforated Homalopoma sanguineum shells (a perhaps strictly Mediterranean species of univalve mollusc identified by E. Alvarez) in Level 17 is indicative (along with Treviño and Urbasa flints) of very far-flung connections on the part of El Mirón inhabitants. Such shells are also known in the Upper Magdalenian of Tito Bustillo and even El Horno and as far from the Mediterranean as the Magdalenian sites of Gönnersdorf and Andernach in the German Rhineland (Alvarez 2002, 2005). Another originally Mediterranean mollusc, Cyclope sp., has been identified in the Magdalenian of El Mirón (Level 503), as well as in La Garma and Tito Bustillo, in the coastal zones of central Cantabria and eastern Asturias, respectively (Gutiérrez-Zugasti & Cuenca-Solana 2015).

It is the Lower Magdalenian of El Mirón that is most distinctive and well-defined by the presence of artifacts that link the occupations of this cave with other contemporaneous ones in sites in Cantabria and eastern Asturias. The presence in outer vestibule Level 17 of a complete red deer stag scapula with the striation engraving of a hind (and an outline engraving of a bovine forequarter) plus other fragments of similar pieces from both Level 17 (average of 5 dates: c.15.5 uncal. kya) and chronologically correlated Level 116 in the vestibule rear, make El Mirón the easternmost of a series of sites with almost identical objects (notably El Castillo and Altamira, but also El Juyo, El Pendo, El Rascaño in central Cantabria, plus El Cierro in eastern Asturias) (González Morales et al. 2006; González Morales & Straus 2009). Another such virtually identical engraved scapula has recently been found in an excavation outside the present mouth of Altamira cave (Heras et al. 2012).

But in terms of vast networks, it was in the Middle Magdalenian that the situation changed noticeably for the region. Contours découpés — small effigies of horse and caprine heads cut out of horse hyoid bones and usually perforated — have their centre of distribution in the French Pyrenees at sites like Isturitz and Labastide. Not known from Cantabrian Spain until J. Fortea began his Nalón Valley project, several contours have been found in Middle Magdalenian-age levels from Ekain in 115

Prehistoric Art as Prehistoric Culture Guipúzcoa (the only one in the Spanish Basque Region and atypical to boot, being a cut-out of a bird) to La Viña and Las Caldas in central Asturias (Corchón et al. 2012). An apparent cache of four horse head effigies has recently been discovered deep within Tito Bustillo Cave (Balbín et al. 2003). A possible imitation (an ibex head effigy not cut out of a hyoid bone) has been found in a level of either late Middle or early Upper Magdalenian age in Tito Bustillo Cave (eastern coastal Asturias) (see Arias & Ontañón 2005 for a catalogue). Also typical of the French Pyrenean and Aquitanian Middle (and Upper) Magdalenian are discs cut out of flat bones (shoulder blades) and engraved with animal images and/or geometric designs and often perforated. These too have been found in recent years in five Middle Magdalenian Asturian and Cantabrian sites (but not in Euskadi) (Schwendler 2005; Corchón & Rivero 2008). Another peculiar type of portable art object of the Middle Magdalenian is edge-ticked hyoid bones, prominent in the La Guëlga and Tito Bustillo sites of the Sella valley in eastern Asturias, but also found in sites of Navarra and west-central France (García-Sánchez et al. 2014). Proto-harpoons (with stubby barbs that are more like bumps along the antler shaft) had earlier been known from the Basque area site of Ermittia, but have now also appeared as far west as Las Caldas, reinforcing the idea of widespread human contacts with SW France over the whole length of the Cantabrian region. However, none of these temporally and trans-regionally diagnostic artifact types has been found in the Middle Magdalenianage levels at El Mirón, either as a question of sampling ‘bad luck’ or because the inhabitants had not acquired or imitated any of these extra-territorial objects. (A short, grooved, single bevel-base sagaie from Level 116 [González Morales & Straus 2005: Fig. 8.3] resembles a Lussac-Angles point, supposedly characteristic of the Middle Magdalenian in France, but found at El Mirón in a layer containing a fragmentary striation-engraved scapula, which is a Cantabrian Lower Magdalenian diagnostic.) For the region as a whole (though not specifically El Mirón or the Asón valley during the Middle Magdalenian), contacts with SW France and the upper Ebro drainage seem to have been relatively important, as exemplified by the recent identification of small quantities of flints from Chalosse and Bidache (Les Landes and Basse-Navarre) and from Urbasa and Treviño, as well as from Barrika in the Middle Magdalenian of Las Caldas. Works of portable art (and/ or ideas thereof) and flints make manifest a network of band-to-band, individual-to-individual relationships extending along the Cantabrian strip, a distance of up to 550 km (Chalosse to Las Caldas) (Corchón et al. 2009). While these flints were circulating the region during the Lower Magdalenian period, the large-scale diffusion of the osseous industries and art objects in the Middle and Upper phases indicates that perhaps these inter-group relationships intensified as the Magdalenian period progressed.

Even less can be said about indicators of wider social territories for the Upper Magdalenian of El Mirón, since there are only one harpoon shaft fragment, one harpoon barb and few sagaies in the radiometrically corresponding levels. In contrast, El Valle and El Horno have works of portable art that clearly link them with the Cantabrian (and Pyrenean) style zone during this Tardiglacial period. For example, the well-known, stylized and strictly Cantabrian engraved frontal views of ibex heads on sagaies and other osseous objects of Upper Magdalenian age (Barandiarán et al. 2013) are absent from the excavated areas of El Mirón. The minor Azilian occupations displayed their relationship to the wider Pyrenean and Cantabrian sociocultural worlds by the typology of their small retouched lithic tools and the presence of an ochre-stained pebble, generally imitating the style of artifact first identified in the rich Azilian layers of Le Mas d’Azil and weakly represented throughout the Cantabrian region. Magdalenian Subsistence in the Asón Valley Analyses of ungulate faunal remains from El Mirón have only been completed for the Middle and Upper Magdalenian (and Azilian) levels (Marín-Arroyo 2009, 2010), as well as for the Lower Magdalenian human burial deposit (Level 504) at the southeastern corner of the vestibule (Marín-Arroyo & Geiling 2015), while work on major Lower Magdalenian Level 17 (plus levels 16 and 15) is currently underway (by J-M. Geiling). A. B. Marín-Arroyo has also studied the mammals from modern excavations in the small La Fragua Cave (including those of 13 uncal kya Upper Magdalenian Level 4) at the mouth of the Asón (Marín-Arroyo 2004; Marín-Arroyo & González Morales 2007). S. Costamagno has analyzed the c. 12.3-12.5 uncal kya Upper Magdalenian faunal assemblages from El Horno Cave below El Mirón (Costamagno & Fano 2006). The collections of (undated) Magdalenian fauna from the 1960s excavations in La Chora and El Otero caves in the coastal zone of the lower Asón have recently been reanalyzed by J. Yravedra (2010). The evidence summarized here stems from the cited sources. El Miron The Lower, Middle and Upper Magdalenian (and Solutrean [Straus et al. 2012]) levels analyzed so far are co-dominated by Cervus elaphus and Capra pyrenaica, with a fairly constant (but modest) presence of Rupicapra rupicapra. Red deer generally outnumbers ibex in terms of NISP, MNI and bone weight, but not always. The ibex skeletons are generally well-represented, suggesting nearby kills, which is logical given the location of the site on a steep, rocky cliff. Red deer skeletons are more partially represented (more axial than appendicular), suggesting field butchery at and differential transport from kill sites, which were presumably on the valley floor and lower slopes. Long bones of both species were intensively 116

Straus et al: Magdalenian Settlement-Subsistence Systems in Cantabrian Spain

Figure 2. Upper Magdalenian harpoon fragments from El Mirón. A. Level 12 (L. Agudo), B. Level 502; length of barb = 13.0 mm (L. Teira).

117

Prehistoric Art as Prehistoric Culture and systematically broken for extraction of marrow and grease. Equus caballus is found in trace quantities in a few of the levels (mainly the older ones) and bovines (Bos/Bison) are almost absent. Capreolus capreolus and Sus scrofa are also rare, but a bit more numerous in the latest (Final Magdalenian and Azilian) levels analyzed by Marín-Arroyo, as is logical given on-going reforestation of the region. Carnivores and lagomorphs are found in very small quantities. Carnivore tooth marks on ungulate bones (and on one human tibia from the burial) are very rare. However the cave was probably used at times by bone-breaking bearded vultures (Marín-Arroyo et al. 2009). Usually in Magdalenian levels, the evidence of seasonality from young deer and ibex teeth and bone fusion indicates that kills took place in late summer and early summer, with only one additional fall occupation in an Azilian layer. There is no evidence for cold-season Magdalenian occupations, except for the small burial area during the Lower Magdalenian, which has winter-spring kills, but no summer ones (Marín-Arroyo & Geiling 2015). The pattern of warm season kills of ibex and red deer is also true of the Solutrean occupations according to preliminary analyses by A. B. Marín (Straus et al. 2012). Our current hypothesis is that the bulk of the Lower and Initial Magdalenian levels (also overwhelmingly dominated by red deer and ibex) will turn out to have been the results of warm season occupations, but on a scale much larger and of longer duration than either the underor over-lying levels. Ibex seems to increase with depth in terms of numbers of remains and individuals, perhaps in relation to the colder conditions of early Oldest Dryas and the Last Glacial Maximum, although, on average for each level, red deer would always have provided the largest amount of meat.

and ibex, followed by roe deer and boar (indicative of the presence of some woods) and traces of bovines and chamois. The ibex were killed in the fall and the red deer in late summer-early fall, according to Marín-Arroyo. El Otero, located on the floor of the coastal plain in the lower Asón basin, has a small Upper Magdalenian faunal assemblage in Level 2, dominated by red deer, with traces of roe deer, bovines, chamois, ibex, horse, boar. There is some indication of carnivore activity on chamois bones, according to Yravedra. That author also reanalyzed a small collection of Upper Magdalenian fauna from nearby La Chora Cave, which is also dominated by red deer with traces of all the other ungulates except bovines. Unfortunately there are no seasonal indicators. Finally, in El Valle Cave in the middle Asón valley, there are indications of very specialized red deer hunting in all the Upper and Final Magdalenian and Azilian levels sampled by several small (c. 1x1 m), isolated test pits excavated by García-Gelabert and Talavera (2004). It is difficult to summarize and correlate all the various faunal reports done by A. Morales and associates as reported in the monograph, but the specialization in red deer is glaring and the archaeozoologists are quoted (pp. 2312) as concluding that many were hunted in FebruaryMarch, i.e. winter. The scarcity of ibex is interesting, since El Valle is at the foot of a very steep slope on the north face of the first foothill range of the Cordillera (El Mirón and El Horno are in the second range). By way of comparison, the small cave of El Rascaño on the side of a steep, rocky gorge of the Río Miera, 21 linear km west of El Mirón, was overwhelmingly specialized in ibex hunting and the animals were killed at different seasons of the year, according to Jesús Altuna (1981). In the levels with several kills whose season could be determined (Initial and Lower Magdalenian and Azilian), spring-early summer, winter and possibly fall are represented in the first; spring-early summer, fall-winter in the second; and spring-summer, winter and possibly fall in the third. In short, hunting parties seem to have gone up to this strategic, but incommodious site at different times of the year, fundamentally to kill ibex and perhaps the occasional deer (unless joints of meat of the latter were brought there as ‘trail food’). They were probably based (at least in winter) at residential sites on the nearby rolling coastal plain (places like La Garma in the lower Miera basin and others perhaps now under the waters of Santander Bay). Unfortunately there is virtually no seasonality information from sites on the coastal plain of central Cantabria — just some indication (unworn neonatal or foetal teeth) of spring hunting of red deer (presumably of hinds with their fawns) in the main Lower Magdalenian levels at El Juyo, although Klein and Cruz-Uribe (1987: 115) do not rule out the possibility of occupations of the site at all seasons of the year (though not necessarily continuously). El Juyo is extremely rich

El Horno The small Upper Magdalenian assemblages from El Horno (especially Level 2) are dominated by ibex, followed distantly by red deer plus traces of horse, boar, roe deer, chamois and a lagomorph. The few ibex and red deer for which Costamagno could determine seasonality were killed in late winter/early spring for the former and early spring for the latter. Field-butchered parts of carcasses were transported to the cave and then heavily fractured for marrow and grease extraction. Thus there is a hint of a difference in the use of this small cave versus the nearby, far larger, but higher El Mirón. The montane zone seems to have been used in both cold and warm seasons, but in rather different ways during the waning centuries of the Late Glacial. La Fragua, El Otero, La Chora and El Valle Level 4 in La Fragua, on the flank of steep, rocky, 380 m-high Monte Buciero which overlooks the present northern edge of the coastal plain, Asón river mouth and the ocean, has a diverse fauna, co-dominated by red deer 118

Straus et al: Magdalenian Settlement-Subsistence Systems in Cantabrian Spain in marine molluscs (like its contemporary, Altamira) attesting to the importance of this food resource in Lower Magdalenian times. Freeman (1973) highlighted what he called the ‘wild harvesting’ of red deer and limpets.

levels have red deer (the overwhelmingly dominant species) killed in summer (and late spring), fall and winter – i.e. multi-season (Altuna 1976). At La Riera scanty evidence from Upper Magdalenian Level 24 shows at least spring-early summer occupation, and warm season occupations are also suggested for earlier (Upper) Magdalenian levels 21-23, all based on young red deer (Altuna 1986). Tentatively one could argue that the coastal zone of Asturias was used in all seasons of the year, and certainly not only in winter. Seasonality data from the montane interior Magdalenian sites of La Güelga, Coimbre and Llonín are anxiously awaited. Both Tito Bustillo and La Riera (like other sites in eastern Asturias, such as El Cierro and Cova Rosa) have abundant marine molluscs, especially limpets, which had become a not insignificant part of the human diet, at least seasonally or situationally.

A recent overview of the Initial and Lower Magdalenian occupations of the Deva and adjacent Urola valleys in west-central Guipúzcoa (Ermittia, Praile Aitz, Erralla, Ekain and Amalda caves) concludes (from faunal analyses principally by J. Altuna) that these took place almost entirely in the warm season (Mujika & Peñalver 2012). This is an area of steep, chaotic relief, where the mountains extend northward all the way to the shore, with no continuous east-west coastal plain or other easy avenue of communication. Higher montane Magdalenian sites are not currently known from this sector, and the authors speculate that the cold season sites may lie on the narrow, now-flooded continental shelf. Each of the sites seems to have been more or less specialized in the hunting of different ungulates: ibex at Ermittia and Erralla; red deer at Ekain and Urtiaga; chamois at Amalda, suggesting that people chose them for the kind of game that were habitually found locally and could best be procured near each cave.

Conclusions Given the steep, but locally variable relief and the proximity of the coastal zone to very high mountains, it is perhaps not surprising that Magdalenian bands, centred in territories probably corresponding to the short valleys of the Cantabrian regions, could use a variety of strategies for the exploitation of ungulate species whose migratory patterns were limited and altitudinal in character: red deer and ibex (plus chamois, horse and, in colder periods, reindeer, and, in warmer ones, boar and roe deer). Under some circumstances families/bands could residentially move between the coastal zone, where they exploited marine molluscs, and the edge of the Cordillera – preferentially in summer for the latter. However, they could also send hunting parties into the montane zone from base camps in the coastal zone at any season of the year, since the distances are short. Likewise, parties from base camps in the mountains or mountain edge could go to the coast for the procurement of flint, especially in the Basque and eastern Cantabrian sectors. Salmon and trout/sea trout were fished both in the estuaries and along the rivers, up to the mountains (such as in the Asón at El Mirón). Food evidently could be procured normally within relatively small valleycentred territories, exploiting the montane interior, the lower valleys/coastal zone and littoral.

At the other end of the Cantabrian region an extensive analysis of seasonality was conduced by A. Mateos (2002) on the relatively diversified ungulate faunal spectrum from Las Caldas Cave, Level VIII (one of the richest layers, early Middle Magdalenian, 13.6 uncal kya). Las Caldas is relatively far into the interior of central Asturias, but on the edge of the major Nalón valley near the point where its gorge skirts the montane zone and opens out onto the broad, but rolling coastal plain. Based on mandibles of red deer, chamois, ibex and horse (in descending order of abundance), Mateos concluded that kills occurred in late winter-spring, summer and fall, with a strong emphasis on the first. Las Caldas could be seen as a major residential site that was preferentially used during the cold season, but visited at other times as well due to its strategic location for hunting on a steepsided, cul-de-sac valley on the edge of the Nalón gorge, a major axis of north-south and east-west communication. Unfortunately there are no montane sites in this sector, and seasonality data are not yet available from the lower Nalón or littoral of central Asturias. That the Magdalenian inhabitants of Las Caldas visited the Late Glacial shore (40-50 km to the north) is attested by the presence of marine molluscs, fragments of crustaceans, and a few (presumably scavenged) marine mammal (dolphin, whale and seal) teeth (one with an engraved whale image). In the territory between the site and the coast, the inhabitants also collected amber, jet and lignite and some flints (Corchón et al. 2008).

However, for the purposes of maintaining social relations (to seek mates, to assure outlets in the event of local disputes and the need to ‘move in’ with neighbouring bands, to get prized materials such as particular flints or exotic objects of art or adornment, to obtain information on the location and condition of herd game and perhaps seasonal help with its collective hunting, to conduct rituals and feasts–-possibly related to such hunts), people kept contact with individuals outside their own bands of the moment, usually in adjacent valleys. This led to down-the-line movement of objects and ideas, most intensive within the geographic confines of the

From eastern Asturias, the coastal zone sites of Tito Bustillo and La Riera do have seasonality data for the Magdalenian. In the former, the Upper Magdalenian 119

Prehistoric Art as Prehistoric Culture Cantabrian strip (or, at times, preferentially within sectors thereof, namely central-eastern Asturias and Cantabria versus Spanish and French Basque Country). Indirect contacts with the western French Pyrenees and beyond would have been easier and more fluid at the eastern (Basque) end of the region, but at some times (and for reasons we do not yet understand) — notably the Middle Magdalenian — the chain of band-to-band relations extended rather spectacularly all the way west to the Nalón, as manifested by the finds in La Viña and Las Caldas. Less well known, but equally certain is the fact that there were trans-Cordilleran contacts (both near [Treviño and Urbasa flints] and far [Cyclope, possibly Homalopoma]), meaning that Ebro Valley, Levantine and Cantabrian Magdalenian bands were ultimately part of the same larger socio-cultural world. Magdalenian hunter-gatherers operated at several different territorial scales: food, neighbours and the oikoumene (known world). The latter extended from the caves of the deep, narrow valleys, high mountains and steep coast of Vasco-Cantabria up to the vast plains, major river basins and lake shores of northern France, western Switzerland, England, Low Countries and Central Europe. The result was a Magdalenian European Union, made up of local territories, distinctive cultural regions and a vast network of relationships that transcended different environments and ‘food areas’ (see Schwendler 2012). El Mirón was 1) part of a local band exploitation territory centred on the Asón River valley, 2) part of a regional band social territory corresponding roughly to the modern province of Cantabria, 3) to varying degrees across time a participant in the distinctive Magdalenian network of relatively dense human contacts spanning the French Pyrenees and greater Vasco-Cantabria, and 4) linked to the wider Magdalenian world that ultimately stretched from southern Iberia to Central Europe.

References Altuna, J. 1976. Los mamíferos del yacimiento prehistórico de Tito Bustillo, pp. 149-94 in (J. A. Moure & N. Cano, eds) Excavaciones en la Cueva de Tito Bustillo. Trabajos de 1975. Instituto de Estudios Asturianos: Oviedo. Altuna, J. 1981. Restos óseos del yacimiento prehistórico del Rascaño, pp. 221-69 i n (J. González Echegaray & I. Barandiarán, eds) El Paleolítico Superior de la Cueva del Rascaño. Monografías del Centro de Investigación y Museo de Altamira 3, Santander. Altuna, J. 1986. The mammalian faunas from the prehistoric site of La Riera, pp. 237-74 & 421-79 in (L. G. Straus & G. A. Clark, eds) La Riera Cave. Anthropological Research Papers 36, Tempe. Alvarez, E. 2002. Perforated Homalopoma sanguineum from Tito Bustillo (Asturias): mobility of Magdalenian groups in northern Spain. Antiquity 76: 641-46. Alvarez, E. 2005. Eloignés, mais pas isolés: la parure hors de la ‘frontière française’ pendant le Magdalénien, pp. 25-38 in (V. Dujardin, ed.) Industrie Osseuse et Parures du Solutréen au Magdalénien en Europe. Mémoires de la Société Préhistorique Française 39. Paris. Arias, P. & Ontañón, R. (eds) 2005. La Materia del Lenguaje Prehistórico. Gobierno de Cantabria: Santander. de Balbín, R., Alcolea, J. & González Pereda, M. 2003. El macizo de Ardines, un lugar mayor del arte paleolítico europeo, pp. 91-151 in (R. de Balbín & P. Bueno, eds) El Arte Prehistórico desde los Inicios del Siglo XXI. Asociación Cultural Amigos de Ribadesella: Ribadesella. Barandiarán, I. 1973. Arte Mueble del Paleolítico Cantábrico. Universidad de Zaragoza: Zaragoza. Barandiarán, I. 1981. Industria ósea, pp. 95-164 in (J. González Echegaray & I. Barandiarán, eds) El Paleolítico Superior de la Cueva del Rascaño. Monografías del Centro de Investigación y Museo de Altamira 3, Santander. Barandiarán, I. 1988. Datation C14 de l’art mobilier magdalénien cantabrique. Bulletin de la Société Préhistorique Ariège-Pyrénées 43: 63-84. Barandiarán, I., Cava, A. & Gundín, E. 2013. La cabra alerta: marcador gráfico del Magdaleniense cantábrico avanzado, pp. 263-86 in (M. de la Rasilla, ed.) F. Javier Fortea Pérez. Ménsula: Oviedo. Cabrera, V. 1984. El Yacimiento de la Cueva de ‘El Castillo’. Bibliotheca Praehistorica Hispana 22, Madrid. Corchón, M. S. & Rivero, O. 2008. Los rodetes del Magdaleniense medio cántabro-pirenaico. Zephyrus 61: 61-84. Corchón, M. S., Mateos, A., Alvarez, E., Peñalver, E., Delclòs, X. & van der Made, J. 2008. Ressources complémentaires dans le Magdalénien cantabrique.

Acknowledgements Excavations in El Mirón Cave since 1996 have been directed by González Morales and Straus, as authorized and partially funded by the Gobierno de Cantabria. Additional financing has been provided by the U.S. National Science Foundation, Fundación Marcelino Botín, L. S. B. Leakey Foundation, Ministerio de Educación y Ciencia, National Geographic Society, University of New Mexico and the Fund for Stone Age Research (Jean and Ray Auel, principal donors). Material support has been provided by the Universidad de Cantabria and Town of Ramales de la Victoria. Paul Bahn and ‘Mimi’ Bueno kindly invited us to contribute to this well-deserved Libro Homenaje for our friend and colleague Rodrigo de Balbín in acknowledgement of his decades of research into the Upper Palaeolithic of Spain — especially its Magdalenian rock art and the importance of territories such as that of the Sella River in Asturias, anchored in the great art and habitation site of Tito Bustillo. 120

Straus et al: Magdalenian Settlement-Subsistence Systems in Cantabrian Spain Nouvelles données sur les mammifères marins, les crustacés, les mollusques et les roches organogènes de la Grotte de Las Caldas. L’Anthropologie 112: 284-327. Corchón, M. S., Martínez, J. & Tarriño, A. 2009. Mobilité, territoires et relations culturelles au début du Magdalénien moyen cantabrique: nouvelles perspectives, pp. 217-30 in (F. Djindjian, J. Kozłowski & N. Bicho, eds) Le Concept de Territoires dans le Paléolithique Supérieur Européen. British Archaeological Reports S-1938: Oxford. Corchón, M. S., Alvarez, E. & Rivero, O. 2012. Contactos extra-cantábricos en el Magdaleniense medio: nuevos datos de la Cueva de Las Caldas, pp. 113-27 in (P. Arias, M. S. Corchón, M. Menéndez & A. Rodríguez, eds) El Paleolítico Superior Cantábrico. PUbliCan: Santander. Costamagno, S. & Fano, M. 2006. Pratiques cynégétiques et exploitation des ressources animales dans les niveaux du Magdalénien supérieur-final de El Horno. Paléo 17: 31-56. Fano, M. 2005. El final del Magdaleniense en la cuenca del Río Asón. Nuevos datos procedentes de la Cueva de El Horno, pp. 109-22 in (N. Bicho, ed.) O Paleolítico. Promontoria Monográfica 2. Fontes, L. M., Straus, L. G. & González Morales, M. R. 2015. Lithic and osseous artifacts from the Lower Magdalenian human burial deposit in El Mirón Cave, Cantabria, Spain. Journal of Archaeological Science 60: 99-111. Fortea, J. 1981. Investigaciones en la cuenca média del Nalón, Asturias (España): Noticia y primeros resultados. Zephyrus 32-33: 2-16. Freeman, L. G. 1973. The significance of mammalian faunas from Paleolithic occupations in Cantabrian Spain. American Antiquity 38: 3-44. García-Gelabert, M. P. & Talavera, J. 2004. La Cueva del Valle, Rasines, Cantabria, España. British Archaeological Reports S-1252: Oxford. García-Sánchez, E., Menéndez, M., Alvarez, D., Andres, M. de, Quesada, J. M. & Rojo, J. 2014. Los hiodes decorados del Magdaleniense de la Cueva de La Guëlga, pp. 333-47 in (M. S. Corchón & M. Menéndez, eds) Cien Años de Arte Rupestre Paleolítico. Universidad de Salamanca: Salamanca. González Echegaray, J., García Guinea, M., Begines, A. & Madariaga, B. 1963. Cueva de La Chora. Excavaciones Arqueológicas en España 26, Madrid. González Echegaray, J., García Guinea, M., Begines, A. & Madariaga, B. 1966. Cueva del Otero. Excavaciones Arqueológicas en España 53, Madrid. González Morales, M. 2000. La prehistoria de las Marismas: excavaciones en la Cueva del La Fragua, pp. 177-80 in (R. Ontañón, ed.) Actuaciones Arqueológicas en Cantabria, 1984-1999. Gobierno de Cantabria: Santander. González Morales, M. & Díaz, Y. 2000. La prehistoria de las Marismas: excavaciones arqueológicas en los

abrigos de la Peña del Perro, pp. 93-98 in (R. Ontañón, ed.) Actuaciones Arqueológicas en Cantabria, 19841999. Gobierno de Cantabria: Santander. González Morales, M., Yudego, C. & Ituarte, C. 2000b. La prehistoria de las Marismas: prospeccion arqueológica de la zona del bajo Asón y marismas de Santoña y toma de muestras en los yacimientos de las cuevas del Otero, La Chora y El Valle, pp. 15152 in (R. Ontañón, ed.) Actuaciones Arqueológicas en Cantabria, 1984-1999. Gobierno de Cantabria: Santander. González Morales, M. & Straus, L. G. 2005. The Magdalenian sequence of El Mirón Cave (Cantabria, Spain): an approach to the problems of definition of the Lower Magdalenian in Cantabrian Spain, pp. 20919 in (V. Dujardin, ed.) Industrie Osseuse et Parures du Solutréen au Magdalénien en Europe. Mémoire de la Société Préhistorique Française 39, Paris. González Morales, M. & Straus, L. G. 2009. Extraordinary Early Magdalenian finds from El Mirón Cave, Cantabria (Spain). Antiquity 83: 26781. González Morales, M. & Straus, L. G. 2012. Terminal Magdalenian/Azilian at El Mirón Cave (Ramales de la Victoria, Cantabria) and the Río Asón valley, pp. 189-215 in (J. Muñiz, ed.) Ad Orientem. Ménsula: Oviedo. González Morales, M. & Straus, L. G. & Marín-Arroyo, A. B. 2006. Los omóplatos magdalenienses de la Cueva del Mirón (Ramales de la Victoria, Cantabria) y su relación con la Cueva del Castillo, Altamira y El Juyo, pp. 482-95 in (J. M. Maillo & E. Baquedano, eds) Miscelánea en Homenaje a Victoria Cabrera, vol. 1, Zona Arqueológica 7. Gutiérrez-Zugasti, I. & Cuenca-Solana, D. 2015. Ornaments from the Magdalenian burial area in El Mirón Cave (Cantabria, northern Spain): were they grave goods? Journal of Archaeological Science 60: 112-24. Heras, C. de las, Lasheras, J. A., Rasines, P., Montes, R., Fatás, P., Prada, A. & Muñoz, E. 2012. Datation et contexte archéologique de la nouvelle omoplate gravée découverte à Altamira, pp. 1571-88 in (J. Clottes, ed.) L’Art Pléistocène dans le Monde. Bulletin de la Société Préhistorique Ariège-Pyrénées 55-56: 270-271+CD. Klein, R. & Cruz-Uribe, K. 1987. La fauna mamifera del yacimiento de la Cueva de ‘El Juyo’. Campañas de 1978 y 1979, pp. 97-120 in (I. Barandiarán, L. G. Freeman, J. GonzálezEchegaray & R. Klein, eds) Excavaciones en la Cueva del Juyo. Monografías del Centro de Investigación y Museo de Altamira 14, Madrid. Marín-Arroyo, A. B. 2004. Análisis Arqueozoológico, Tafonómicon y de Distribución Espacial de la Fauna de Mamíferos de la Cueva de La Fragua. Ediciones TGD: Santander.

121

Prehistoric Art as Prehistoric Culture Marín-Arroyo, A. B. 2009. Exploitation of the montane zone of Cantabrian Spain during the Late Glacial: faunal evidence from El Mirón Cave. Journal of Anthropological Research 65: 69-102. Marín-Arroyo, A. B. 2010. Arqueozoología en el Cantábrico Oriental durante la Transición Pleistoceno/Holoceno: La Cueva del Mirón. PubliCan: Santander. Marín-Arroyo, A. B., Fosse, P. & Vigne, J.-D. 2009. Probable evidence of bone accumulation by Pleistocene bearded vulture at the archaeological site of El Mirón Cave, Cantabrian Spain. Journal of Archaeological Science 35: 801-13. Marín-Arroyo, A. B. & Geiling, J. M. 2015. Archaeozoological study of the macromammal remains associated with the human Magdalenian burial of El Mirón Cave. Journal of Archaeological Science 60: 75-83. Marín-Arroyo, A. B. & González Morales, M. 2007. La Fragua Cave, a seasonal hunting camp in the lower Asón Valley (Cantabria, Spain) at the PleistoceneHolocene transition. Anthropozoologica 42: 61-84. Marín-Arroyo, A. B., Landete, M., Vidal, G., Seva, R., González Morales, M. & Straus, L. G. 2008. Archaeological implications of human-derived manganese coatings: a study of blackened bones in El Mirón Cave, Cantabrian Spain. Journal of Archaeological Science 35: 801-13. Mateos, A. 2002. Apuntes sobre estacionalidad y subsistencia de los grupos humanos delCantábrico occidental en torno al 13000 BP. Trabajos de Prehistoria 59: 27-41. Mujika, J. & Peñalver, X. 2012. La ocupación de la cuenca del Deba (Gipuzkoa) durante el Magdaleniense inferior, pp. 97-112 in (P. Arias, M. S. Corchón, M. Menéndez & A. Rodríguez, eds) El Paleolítico Superior Cantábrico. PubliCan: Santander. Nakazawa, Y., Straus, L. G., González Morales, M., Cuenca, D. & Caro, J. 2009. On stone-boiling in the Upper Paleolithic: behavioral implications from an Early Magdalenian hearth in El Mirón Cave, Cantabria, Spain. Journal of Archaeological Science 36: 684-93. Rissetto, J. 2006. Late Pleistocene Hunter-Gatherer Mobility Patterns and Lithic Exploitation in Eastern Cantabria ( Spain). Ph.D. dissertation, University of New Mexico. Ruiz Idarraga, R. 2007. Cueva de El Polvorín. Arkeoikuska 2006: 136-37. Vitoria-Gasteiz. Sauvet, G. 2014. Histoires de chasseurs. Chronique des temps paléolithiques, pp. 15-30 in (M. S. Corchón & M. Menéndez, eds) Cien Años de Arte Rupestre Paleolítico. Universidad de Salamanca: Salamanca. Schwendler, R. 2005. Magdalenian perforated disks in geographic and social context, pp. 73-84 in (V. Dujardin, ed.) Industrie Osseuse et Parures du Solutréen au Magdalénien en Europe. Mémoire de la Société Préhistorique Française 39, Paris.

Schwendler, R 2012. Diversity in social organization across Magdalenian Western Europe, ca. 17-12,000 BP, pp. 333-53 in (L. G. Straus, T. Terberger & D. Leesch, eds) The Magdalenian Settlement of Europe. Quaternary International 272-273. Straus, L. G. & González Morales, M. 2012. The Magdalenian settlement of the Cantabrian region (northern Spain): the view from El Mirón Cave, pp. 111-24 in (L. G. Straus, T. Terberger & D. Leesch, eds) The Magdalenian Settlement of Europe. Quaternary International 272-273. Straus, L. G., González Morales, M., Fano, M. & GarcíaGelabert, M. P. 2002a. Last Glacial human settlement in eastern Cantabria. Journal of Archaeological Science 29: 1403-14. Straus, L. G., González Morales, M., García-Gelabert, M. P. & Fano, M. 2002b. The late Quaternary human uses of a natural territory: the case of the Río Asón drainage. Journal of Iberian Archaeology 4: 21-61. Straus, L. G., González Morales, M. & Stewart, E. 2008. Early Magdalenian variability: new evidence from El Mirón Cave, Cantabria, Spain. Journal of Field Archaeology 33: 197-218, 367-69. Straus, L. G., González Morales, M. & Fontes, L. 2014. Initial Magdalenian artifact assemblages in El Mirón Cave (Ramales de la Victoria, Cantabria, Spain): a preliminary report. Zephyrus 73: 45-65. Straus, L. G., González Morales, M. R., Marín-Arroyo, A. B. & Iriarte, M. J. 2012 (2014) The human occupations of El Mirón Cave (Ramales de la Victoria, Cantabria, Spain) during the Last Glacial Maximum/ Solutrean period. Espacio, Tiempo y Forma, Serie I, Nueva Epoca, Prehistoria y Arqueología 3: 413-26. Straus, L. G., González Morales, M. & Carretero, J. M. (eds) 2015a. The Red Lady of El Mirón Cave: Lower Magdalenian Human Burial in Cantabrian Spain. Special issue of Journal of Archaeological Science, vol. 60. Straus, L. G., González Morales, M., Higham, T., Richard, M. & Talamo, S. 2015b. Radiocarbon dating the late Upper Paleolithic of Cantabrian Spain: El Mirón Cave date list IV. Radiocarbon 57: 183-88. Tarriño, A. 2006. El Sílex en la Cuenca Vasco-Cantábrica y Pirineo Navarro. Monografías del Museo Nacional y Centro de Investigación de Altamira 20, Madrid. Utrilla, P. 1981. El Magdaleniense Inferior y Medio en la Costa Cantábrica. Monografías del Centro de Investigación y Museo de Altamira 4, Santander. Yravedra, J. 2010. Estudio zooarqueológico y tafonómico del yacimiento del Otero. Espacio, Tiempo y Forma. Serie 1. Prehistoria y Arqueología. Nueva Epoca 3: 21-38.

122

The Upper Palaeolithic rock art of Portugal in its Iberian context André Tomás Santos,1 Maria de Jesus Sanches2 and Joana Castro Teixeira3 Fundação Côa Parque; e-mail: [email protected] Faculty of Arts and Humanities-University of Porto; Researcher of the Transdisciplinary ‘Culture, Space and Memory’ Research Centre (CITCEM): [email protected] 3 Researcher of the Transdisciplinary ‘Culture, Space and Memory’ Research Centre (CITCEM): [email protected] 1

2

Abstract This text presents a synthesis of what is known of Palaeolithic rock art in Portugal. We observe, in Portugal, a great graphic homogeneity in pre-Magdalenian rock art and considerable differences between the southern and northern rock art during the Magdalenian. However, during the Pleistocene/Holocene transition, we recognize a new process of homogenization of Portuguese rock art. This evolution is, nevertheless, perfectly explained when we take into account the wider Iberian context of Upper Palaeolithic rock art. Keywords Portugal; Upper Palaeolithic; rock art; Iberian Peninsula

Foreword Our aim in this text is to present a general overview of the Upper Palaeolithic rock art of Portugal. Texts about the same subject have appeared recently (Gomes 2006; Bicho et al. 2007: 117-31; Baptista 2012). However, those publications had very specific purposes and, for this reason, the Iberian context of Portuguese Palaeolithic rock-art (Fig. 1) was disregarded. The most prominent characteristic of Portuguese Palaeolithic rock art is its vast number of open-air sites in comparison with its cave art, which is only to be found in a single cavern. Although open-air sites have some specific features (e.g. Balbín 2008; Balbín & Bueno 2009), to fully comprehend them we must take into account the cave art of the same period (Balbín, Bueno & Alcolea 2012) and refrain from creating artificial divisions that are meaningful only to us — prehistorians of the 21st century. Consequently, to understand Portuguese Palaeolithic rock art we must consider not only other Iberian open-air sites but also its cave art. Therefore, we will try to present the Upper Palaeolithic rock art of Portugal against the background of its Iberian context. Something of this sort has already been presented in texts by Balbín and Alcolea (e.g. Balbín & Alcolea 2002; Alcolea & Balbín 2012), in which the Portuguese sites are integrated into the chronological framework those authors have developed for the Upper Palaeolithic art of the Iberian interior. This chronological framework raises several important questions, such as: what happened in the Côa valley while Siega Verde was being intensely engraved? This is the kind of issue we will try to explore while taking a bird’s eye view of Portuguese Palaeolithic rock art and its Iberian context. 123

1. Common ground: the pre-Magdalenian phase of the rock art of western Iberia We know that during the Gravettian, people were already engraving in the Côa valley. This inference is possible because of the quartzite picks that were found in the Gravettian level of Olga Grande 4, which were, most probably, used to peck and abrade the traces that define a large number of figures of the Côa valley (e.g. Aubry, Sampaio & Luís 2011). The vast majority of these figures are stylistically very homogeneous, with parallels found in caves such as Pairnon-Pair, La Tête du Lion, La Croze à Gontran, PechMerle, Nerja, Pileta, Escoural, etc (e. g. Guy 2000). Some of these sites have radiocarbon-dated figures from the Gravettian period, namely Mayenne-Sciences (Pigeaud 2004: 127) or Pech-Merle (Lorblanchet, Cachier & Valladas 1995). Nevertheless, paintings from other sites, such as La Pileta or Nerja, have been dated to the Solutrean period (Sanchidrían et al. 2001). Moreover, the Solutrean portable art of the aforementioned region has figures that resemble those of which we are speaking, such as the plaquette of Vale Boi (Algarve) (Simón, Cortés & Bicho 2012) or the Solutrean plaquettes of Parpalló (Villaverde 1994). Other sites, such as Pairnon-Pair, can be dated to the Aurignacian (Lenoir et al. 2006). Consequently, for the sake of precision, this art should be classified as pre-Magdalenian. The Magdalenian as a terminus ante quem for this kind of figures is, in fact, well established in the Côa valley, thanks to the excavation of Fariseu (Aubry 2009a: 36671). Here, the study of the stratigraphic sequence of the art on rock 1, in conjunction with the study of the archaeological stratigraphy that covers it, has shown

Figure 1. Portuguese Palaeolithic rock art, plus Molino Manzánez (located in Spain, it borders Portugal on the opposite side of the Guadiana river).

Prehistoric Art as Prehistoric Culture

124

Santos, de Jesus Sanches and Castro Teixeira: The Upper Palaeolithic rock art of Portugal

Figure 2. Panel 31 of Foz do Tua rockshelter.

125

Prehistoric Art as Prehistoric Culture that the rock was engraved around 14,500 BP (Aubry, Santos & Luís 2014). Moreover, an engraved fragment of the rock was found inside layer 8, from the top of which comes an OSL date of 18,400 BP (Aubry, 2009a: 368). Hence, we can assume that, around 18,400 BP, the rock was already engraved and facing a process of degradation. On the other hand, the geoarchaeological analysis of the site (Aubry, Santos & Luís 2014) suggests that the soil contemporary to the engraving of the rock must have been eroded by fluvial action in a moment previous to the accumulation of layer 8. Probable remains of this ancient soil can be found some metres away from the rock panel, where a layer radiocarbondated to 19,020+80 BP (GrA-40167) was identified (ibidem: 265). Pre-Magdalenian zoomorphic figures are, grosso modo, characterized by prominent and convex bellies, rounded hips, naturalistic heads with few or no anatomical details. With few exceptions, only one leg per pair is represented, most of the time without hoofs; usually only one horn is observed, but if both were portrayed they are depicted in a straight or oblique bi-angular profile; the tails are depicted in a very formalized manner, obeying very strict rules that are in accordance with each species; the same goes for the dorsal lines, which, as a common aspect, have their anatomical features (withers, back and croups) depicted in a very prominent fashion. Figures of this kind are extremely common in the Côa valley, defining what researchers unanimously accept as the oldest phase of engraving in the region (e.g. Balbín & Alcolea 2002: 150; Zilhão 2003; Baptista 2012; Santos 2012). We also find figures of this kind in other sites of western Iberia, such as in the cave of Escoural (e.g. Glory, Vaultier & Santos 1965; Santos, Gomes & Monteiro 1981; Lejeune 1996; Gomes 2002; García et al. 2000; Silva 2011; Baptista 2012: 308-13), and in open-air sites like Siega Verde (Alcolea & Balbín 2006: 334) and Redor do Porco in the Águeda valley (Baptista & Reis 2011), Ribeira da Sardinha, Fraga Escrevida, Sampaio and Pousadouro in the Sabor valley (Baptista 2009: 196207), Ocreza near the mouth of the eponymous river in the Tagus valley (Baptista 2009: 208-11), La Grajera 2 in the Spanish Extremadura (Bueno et al. 2010), Mazouco (e.g. Jorge et al. 1981; Balbín & Alcolea 1992: 436-41; Gomes 1994; Baptista 2009: 194-95) and the Foz do Tua rock-shelter (Sanches & Teixeira 2013; Valdez-Tullett 2013). The monographic study of the Foz do do rock-shelter is still in progress, but it is imperative to clarify some observations that have been made about its panel 31 (Baptista 2012: 326). Forty-eight panels were found at the site, forty-six of which are located inside the shelter and are characterized mainly by the presence of deep

Figure 3. Lejeune’s Figure 5.9 of Escoural. From top to bottom: original photo; photo worked on by DStretch; our interpretation of the painted motifs.

126

Santos, de Jesus Sanches and Castro Teixeira: The Upper Palaeolithic rock art of Portugal and light incisions that are dated to the beginning of the Holocene, although a slightly earlier chronology should not be ruled out (Sanches & Teixeira 2013). Schematic red paintings were found outside the shelter on a panel to the right of the opening. These paintings date to the Neolithic or Chalcolithic (ibid.: 62). A panel with possible Palaeolithic engravings was also found to the right of the big opening (panel 2) and a second one, clearly displaying Palaeolithic engravings, was found to the left of the opening (panel 31) (Fig. 2). We have identified, on this last panel, two red deer in a line. The one to the right is a little more complex, because a second neck was added to its body. This fact, per se, is not uncommon in Palaeolithic art, especially in the Côa valley where several examples of animation have been identified (Luís 2012). However, at the end of this second neck we do not see a red deer head, but the head of a second animal under the head of a third. In short, we have one body, two necks and three different heads! In fact, a close inspection of the relevant sector of the panel reveals that the head of an aurochs is transformed into that of a horse. This transformation is achieved by the reconfiguration of the original head and the addition of a mane, which converts the horns of the aurochs into the ears of the horse (Fig. 2). Pecking and abrasion define all the figures on the panel, with incision also noticeable in the muzzle and mouth of the second/third head.

We find, however, another type of incised animals in rock 3 at Canada do Inferno, namely the horses and the aurochs that are engraved above the bigger animals that are depicted in the bottom third of the rock. These animals, particularly the horses, have striking differences from those we have previously dealt with. On the other hand, they bear a remarkable resemblance to some figures that are usually dated to the late Solutrean or early Magdalenian… 2. Breaking ground: Magdalenian art In fact, the most similar parallels to the horses of Canada do Inferno 3 can be found in the cave of El Buxu, in Asturias, which dates to the late Solutrean / early Magdalenian (Obermaier & Vega del Sella 1918; Menéndez 1984). The horses of group number VI of gallery B (Obermaier & Vega del Sella 1918: pl. IV and V) are among the best parallels to the figures of Canada do Inferno 3. Moreover, if we look closely, we find in the Asturian cave a lot of motifs that greatly resemble other engravings of the Côa valley. Good examples from the cave include horse number 1 of gallery A (ibid.: 13), ibex g of group XII of gallery C (ibid.: 13) or horse A of group XIII of gallery D (ibid.: pl. XIII). The respective parallels can be seen in the horses of the left sector of Quinta da Barca 23 (Fig. 4), the ibexes of Vale de Cabrões 5 (Baptista 2009: 166) and Quinta da Barca 56 (Santos 2012: 51) or the horse of Vale de Cabrões 7 (Santos 2012: 58).

It is very interesting to note that the ibex is absent at the Foz do Tua site, whereas it is the third most represented animal in the Côa valley, after aurochs and horses.

This connection with northern Spain, particularly with Asturias, is something that is observed in the Côa valley throughout the Magdalenian period, a period during which no southern influence can be identified (Santos 2012: 46). In fact, although style IV, in its classic form (Leroi-Gourhan, Delluc & Delluc 1995: 283-89), cannot be identified in the Mediterranean region of the Peninsula (Villaverde 1994: 333-34), it is easily recognized in the Côa valley (Santos 2012: 46). Vale de José Esteves 4 (Santos, Aubry & Walter 2014: 45), Vale de Cabrões 32 (Santos 2012: 60), Fariseu 8 (Baptista 2009: 106-7), Canada da Moreira 7 (Fig. 5) or Canada do Inferno 41 (Baptista 2009: 188-89), are among the best examples of ‘classic’ style IV in the Côa valley. Parallels to these figures are impossible to find in southern Iberia, but are easily spotted in northern Spain. The plaquette of La Güelga (Menéndez & Martínez 1991-1992), the ‘wall of engravings’ of La Peña de Candamo (Hernández 1919: 46-48; Moure 1981), the ibexes of phase V of Llonín (Fortea, Rasilla & Rodríguez 2004: 22), the engraved panel of La Loja (Alcalde, Breuil & Sierra 1911: 5359), or some horses of Pindal (e.g. ibid.: 73-74) are remarkable examples of parallels to the figures we have mentioned above.

In Escoural, by contrast, both red deer and ibex appear to be rare or absent during this period. In fact, figures that are identified as ibex have probably been misinterpreted. Take for instance Lejeune’s figure 59, which is usually interpreted as one or several ibexes, but we think that it could depict a horse, and the head of another horse or of a female red deer (Fig. 3). However, besides animals, we also find signs during this phase. These signs are mainly incised and very simple in shape, consisting of single lines, pairs of lines, packs of parallel lines, angles, wavy lines, etc. In the Côa valley there is an apparent tendency to concentrate these signs on certain rocks, like Canada do Inferno 1 (Baptista & Gomes 1997: 264), Penascosa 3 (ibid.: 380) or Fariseu 1 (Baptista 2009: 66-67). Pecking, abrasion and red (Escoural, and Faia in the Côa valley) and black painting (Escoural) are the main techniques used to depict animals during this phase. However, in the Côa valley we also find some incised animals, on rocks such as Fariseu 1 or Canada do Inferno 1 (Baptista & Gomes 1997: 268), that are depicted according to the same formal principles applied to the pecked and abraded ones.

Rocks 2, 3, 5, 6, 7 (Baptista & Gomes 1997: 319-26) or 24 (Baptista 2009: 94-101) of Ribeira de Piscos are 127

Prehistoric Art as Prehistoric Culture

Figure 4. Horse of the left sector of Quinta da Barca 23 (Côa Valley).

Figure 5. Detail of rock 7 of Canada da Moreira-Côa Valley (drawing by Fernando Barbosa).

other important examples of this period. Engravings of this phase are mainly produced by incision, but they can also be pecked, or pecked and abraded. Rego de Vide 1 and 6 (Baptista & Gomes 1997: 298, 303) or Quinta da Barca 3 (Baptista 2009: 148-49) are among the best examples.

The greater part of the Siega Verde rock-art is also dated to the late Solutrean/early Magdalenian (e.g. Alcolea & Balbín 2006) and, indeed, very strong similarities can be found between the pecked and abraded rock art of Siega Verde and engravings from such rocks as Rego de Vide 1 and 6, Fariseu 3 and 5, or Canada do Inferno 128

Santos, de Jesus Sanches and Castro Teixeira: The Upper Palaeolithic rock art of Portugal 12 and 22 (Baptista & Gomes 1997: 274-76). However, the high number of engravings of this ‘type’ in Siega Verde is quite astonishing in comparison with the low numbers found in the Côa valley. This probably has a geoarchaeological explanation.

Côa valley. Moreover, a lot of rocks that were engraved during this phase are probably under the sediments of the valley, and these buried rocks most probably belong to the monumental ‘variant’ of this phase: a kind of rock art that can be seen nowadays in Siega Verde and only here and there in the Côa valley (Rego da Vide, rocks 12 and 22 of Canada do Inferno or rocks 3 and 5 of Fariseu).

In fact, the work of Aubry et al. (2010) showed the existence of several phases of sedimentation and erosion in the Côa valley. The localization of rocks 12 and 22 of Canada do Inferno, which have Siega Verde-like figures, at a much lower altitude than the nearby panels with engravings of older style (rocks 16 and 17 of Canada do Inferno), can be explained by the probable existence of a phase of erosion between the engraving of the older panels and that of the Siega Verde-like panels. This phase of erosion would have destroyed the ancient pre-Magdalenian soil and ‘uncovered’ new panels that were used by Magdalenian people to engrave new motifs. Rocks 16 and 17, as a consequence, became out of reach of a man’s arm, as they are today. This last fact was usually interpreted as evidence of use of scaffolding, but it should most probably be seen as the result of this process of erosion. However, in most parts of the Côa valley, subsequent phases of sedimentation were also identified, such as at Penascosa, where 5 m of deposits were identified in front of rocks 4 and 5 (Almeida 1997). This permits us to hypothesize the existence of Siega Verde-like art in the Côa valley under heavy deposits of sediment.

The identification of portable art in well established contexts is a crucial element in clarifying stylistic evolution during the Magdalenian. Unfortunately, portable art from the period is only represented in the Côa valley by the plaquette of Cardina, on which there are only non-figurative incisions (García 2009: 377). Outside the Côa valley, the known portable art is not, likewise, of much help in studying the Magdalenian sequence. In fact, some of the known examples are either very fragmented (e.g. Buraca Grande [Aubry & Moura 1993]) or engraved only with geometric motifs (e.g. Palha [Braz & Gaspar 2003] and Caldeirão [Zilhão 1988]). In the Sabor valley, a huge series of decorated plaquettes has been found in the site of Medal (Figueiredo et al. 2014). The zoomorphic motifs depicted on these pieces are undoubtedly very similar to some of the Côa valley and stylistically integrated within style IV. The exhaustive publication of the plaquettes and of their archaeological context is, therefore, awaited with great expectation. Regarding the portable art of western Iberia we must also mention the Magdalenian pebble engraved with angular designs that was discovered on the site of Xarez, in Alentejo (Gomes et al. 2000: 97).

Late Solutrean / early Magdalenian or Magdalenian engravings are also identified at Mazouco in the Douro valley, Poço do Caldeirão and Costalta in the Zêzere valley (Baptista 2009: 216-22), Gardete and Fratel in the Tagus valley (Gomes 2010: 476), Molino Manzánez (Collado 2006) and Porto Portel III (Baptista & Santos 2013: 220-26) on the Guadiana river, and in the cave of Escoural. We only find geometric figures in the sites of the Tagus valley. We also find figurative motifs in Molino Manzánez and Porto Portel III. These figures are much simpler than those found in the Côa valley. Parallels for these last figures must be sought in southern Iberia (e.g. Bicho et al. 2007: 95-116, 119).

Although portable art cannot help us to clarify the Magdalenian rock art sequence, it can help to determine the end of the Magdalenian phase, at least in the Côa valley. 3. Late Magdalenian / early Azilian rock-art We know that between 12,000 and 10,000 BP the kind of Magdalenian art which we have discussed above was no longer being made. The plaquettes that were discovered in layer 4 of Fariseu, in Quinta da Barca Sul (Aubry 2009b; García 2009: 376-77) and recently in the azilian layer of Cardina (Aubry et al. in press) are proof of that fact.

In the Côa valley this phase has not been as profoundly studied as the pre-Magdalenian. Apparently, the most represented animals are the same as those of the previous period. It appears, however, that the proportion of aurochs tends to become closer to that of the horses, and the same happens with ibex and red deer. The appearance of human figures should also be stressed. The signs of this phase are more complex than the older ones. We observe an increase of wavy lines, and the emergence of reticulates, triangles, tree-like signs and other complex motifs such as the one of Vale de Cabrões 32 (Santos 2012: 60).

What we see in these plaquettes is the type of art that was defined by Roussot as style V (1990: 199-201). Bueno, Balbín & Alcolea (2007) had already provided evidence for the presence of this style in the Iberian Peninsula, namely in the Douro Basin. The animals from this period have very geometricised bodies, with usually two legs per pair that are depicted according to an oblique or straight bi-angular perspective; the torsos are long, oval or rectangular in shape; the legs are usually triangular in shape; no anatomical details are visible, especially in the heads.

Incision becomes the standard engraving technique. Pecking and abrasion are not, however, absent in the 129

Prehistoric Art as Prehistoric Culture Most importantly, they are mainly incised and their interior is filled with lines. This feature led researchers to draw parallels between this kind of engraving and engraved figures in Cantabria daing to the late Solutrean / early Magdalenian that are, nevertheless, very different stylistically; this happened in the Côa valley (e.g. Baptista 2009: 170), at Domingo García (e.g. Ripoll & Municio 1999: 232) and in the cave of Escoural (e.g. Santos, Gomes & Monteiro 1981: 235). Today, both the plaquettes of Fariseu and the horizontal and vertical graphic stratigraphy of several rocks of the Côa valley (Santos 2012: 45) do not enable us to maintain that idea. This phase is the most widespread in the Côa valley (Santos 2012: 44), being identified on 70% of the 533 known rocks that were engraved in the region during the Pleistocene (Reis 2014: 33). Therefore, Collado was partially right when he defended the need to re-evaluate the chronology of this type of figures in the Côa valley (Collado 2006: 370). It is debatable if these figures should be dated to the Epipalaeolithic (Bueno, Balbín & Alcolea 2007: 553), but it is undeniable that they are more recent than late Solutrean / early Magdalenian.

milder after the end of the Pleniglacial (Santos 2012: 46). This led to a regionalization of contacts, and the previous homogeneity of the rock art is no longer found. Consequently, during the Magdalenian we continue to identify in the Côa valley and in Siega Verde strong relations with the Cantabrian region but not with southern Iberia. On the other hand, in Escoural or Guadiana we find graphic entities that are more closely related to the Magdalenian rock art of southern and eastern Iberia. However, the beginning of the late Dryas and the reappearance of very cold conditions triggered a new need for social interactions, and thus long-distance contacts returned and rock art became more homogeneous once again, as is seen in the advent of ‘style V’ in several distant parts of Iberia. This is obviously a very schematic view, imposed by the space limitations of this text. Reality, as always, was certainly more complex. For instance, if pre-Magdalenian times were undoubtedly a period of great formal homogeneity, this does not necessarily mean that local traditions did not exist and privileged relations between specific regions were not established. Not only have substantial differences already been noticed between the sites of the western and eastern Meseta (e.g. Alcolea & Balbín 2012), but also a factor analysis of the horses of La Griega showed a stronger connection of this site with southeastern Iberia than with the Cantabrian region or western Iberia (Corchón et al. 2012). On the other hand, this privileged relationship between the eastern Meseta and southeastern Iberia is no longer evident during the Magdalenian, when style IV is attested at Domingo García (e.g. rock 6 of Las Canteras [Ripoll & Municio 1999: 151-52]).

Besides incision, other techniques are known, such as pecking (e.g. Canada do Inferno 34 [Baptista & Gomes 1997: 296]) or red painting (e.g. Faia 1 [Baptista 2009: 47]). In the Côa valley, the bestiary of this period is dominated by the red deer, followed by ibex and horses, leaving aurochs as a residual species. In Escoural, however, ibexes are not known and horses are much more represented. This phase is also represented in Guadiana, at least at Moinhola 30 (Baptista & Santos 2013: 14749).

On the other hand, it should be stressed that there are two other known Portuguese Palaeolithic sites that we have not taken into account in this narrative. One is the site of Fraga do Gato in the Douro valley; although most probably Palaeolithic (Baptista 2009: 227), its date is hard to determine with any accuracy; the other is Monte de Góios in the Minho valley (Gomes 2007: 121) of which very little is currently known. The location of this last site strongly brings to mind a rock identified at the site of Vinhas, also in the Minho province, with a probable horse depiction that was compared to the horses of PechMerle and Penascosa (Cruz & Cardoso 2011: 262, 270; Cardoso 2014: 236-38, 270-71). Unfortunately, this rock was destroyed shortly after it was discovered, and all we have of it is a photo that lacks the required detail. After a thorough examination of that photo, we are not even sure if a horse is depicted there; for this reason we think that it should not be included among Portuguese Palaeolithic rock art sites. Last but not least, we must underline the need to further investigate the Erges valley where some clues that point to the existence of Palaeolithic rock art were found (Gomes 2010: 476; Henriques et al. 2011).

4. Some final remarks The common features and dissimilarities that we observe in Portuguese Palaeolithic rock art must be explained within the Iberian context of which they are part. Portuguese rock art of pre-Magdalenian times is very homogeneous. Likewise, rock art with the same morphological features is observed in several other sites of France and Spain (Guy 2000), which suggests a strong social interaction between people living in south-western Europe at the time (Zilhão 2003), as well as a very strict codification of the ways of depicting animals during this period (Guy 2003). Zilhão (2003) defended the idea that this phenomenon was due to the cold conditions of the Pleniglacial, which promoted a greater need for social interaction. We believe that the ‘upgrading’ of that reasoning can help us to explain the remaining sequence of the Palaeolithic rock art of western Iberia. In fact, the need for social interaction decreased when the climate became 130

Santos, de Jesus Sanches and Castro Teixeira: The Upper Palaeolithic rock art of Portugal Balbin, R. de 2009. El Arte Rupestre Paleolítico al aire libre en la Península Ibérica, pp. 19-56 in (R. de Balbín, ed.) Arte Prehistórico al aire libre en el Sur de Europa. .Actas. PAHIS. Junta de Castilla y León: Salamanca. Balbín, R. de & Alcolea, J. J. 1992. La grotte de Los Casares et l’art paléolithique de la Meseta espagnole. L’Anthropologie 96 (2-3): 397-452. Balbín, R. de & Alcolea, J. J. 2002. L’art rupestre paléolithique de l’intérieur péninsulaire ibérique: une vision chronoculturelle d’ensemble, pp. 139-57 in (D. Sacchi, ed.) L’art Paléolithique à l’Air Libre. Le Paysage Modifié par l’Image, Tautavel - Campôme, 7-9 octobre 1999. Tautavel: GAEP & GÉOPRÉ. Balbin, R. de & Bueno, P. 2009 Altamira, un siècle après: art paléolithique en plein air. L’Anthropologie 113: 602-28. Balbín, R. de, Bueno, P. & Alcolea, J. J. 2012. Técnicas, estilo y cronología en el arte paleolítico del sur de Europa: cuevas y aire libre, pp. 105-24 in (M. de J. Sanches, ed.) Iª Mesa Redonda ‘Artes Rupestres da Pré-história e da Proto-história: paradigmas e metodologias de registo. Lisbon: DGPC [Trabalhos de Arqueologia, 54]. Baptista, A. M. 2009. O Paradigma Perdido: O Vale do Côa e a arte paleolítica de ar livre em Portugal. Porto/ Vila Nova de Foz Côa: Edições Afrontamento; Parque Arqueológico do Vale do Côa. Baptista, A. M. 2012. El arte paleolítico en Portugal, pp. 305-37 in (S. Ripoll, ed.) Arte Sin Artistas. Una Mirada al Paleolítico. Madrid: Museo Arqueológico Regional. Baptista, A. M. & Gomes, M. V. 1997. Arte rupestre, pp. 211-406 in (J. Zilhão, ed.) Arte Rupestre e Préhistória do Vale do Côa. Lisbon: Ministério da Cultura. Baptista, A. M. & Reis, M. 2011. A rocha gravada de Redor do Porco. Um novo sítio com arte paleolítica de ar livre no rio Águeda (Escalhão, Figueira de Castelo Rodrigo). Côavisão 13: 15-20. Baptista, A. M. & Santos, A. T. 2013. A Arte Rupestre do Guadiana Português na área de influência do Alqueva, s. l.: Edia & Drcalen [Memórias d’Odiana, nova série, 1]. Bicho, N. F., Carvalho, A. F., González, C., Sanchidrian, J. L., Villaverde, V. & Straus, L. G. 2007. The Upper Paleolithic rock art of Iberia. Journal of Archaeological Method and Theory 14 (1): 81-151. Braz, A. F. & Gaspar, R. 2003. Intervenção de emergência no vale da Ribeira das Chitas. O caso de dois abrigos com Pré-história antiga. Al-madan, IIª série, 12: 186. Bueno, P., Balbín, R. de & Alcolea, J. J. 2007. Style V dans le bassin du Douro. Tradition et changement dans les graphies des chasseurs du Paléolithique supérieur européen. L’Anthropologie 111: 549-89. Bueno, P., Balbín, R. de, Barroso, R., Carrera, F., Alfonso, J., Alonso, J., Barbado, J. J., Berzas, G., Martín, M. A. & Salgado, P. 2010. Secuencias

References Alcalde, H.; Breuil, H. & Sierra, L. 1911. Les Cavernes de la Région Cantabrique (Espagne). Monaco: Imprimerie Vve A. Chêne. Alcolea, J. J. & Balbín, R. de 2006. Arte Paleolítico al Aire Libre. El Yacimiento Rupestre de Siega Verde. Salamanca: Junta de Castilla y León [Arqueología de Castilla y León, 16]. Alcolea, J. J. & Balbín, R. de 2012. El arte rupestre paleolítico del interior peninsular, pp. 185-208 in (S. Ripoll, ed.) Arte Sin Artistas. Una Mirada al Paleolítico. Madrid: Museo Arqueológico Regional. Almeida, F. 1997. Prospecção geofísica dos depósitos quaternários, pp. 55-73 in (J. Zilhão, ed.) Arte Rupestre e Pré-história do Vale do Côa. Lisbon: Ministério da Cultura. Aubry, T. 2009a. Datação indirecta da arte do Vale do Côa; estratigrafia, arte rupestre e móvel, pp. 361-73 in (T. Aubry, ed.) 200 Séculos de História do Vale do Côa: Incursões na vida quotidiana dos caçadoresartistas do Paleolítico. Lisbon: IGESPAR, I. P. [Trabalhos de Arqueologia, 52]. Aubry, T. 2009b. Actualisation des données sur les vestiges d’art paléolithique sur support mobilier de la Vallée du Côa, pp. 382-95 in (T. Aubry, ed.) 200 Séculos de História do Vale do Côa: Incursões na vida quotidiana dos caçadores-artistas do Paleolítico. Lisbon: IGESPAR, I. P. [Trabalhos de Arqueologia, 52]. Aubry, T. & Moura, M. H. 1993. Plaquinha de xisto com gravuras paleolíticas. Boletim da Associação de Defesa do Património Cultural de Pombal, s. n. , pp. 13-17. Aubry, T., Dimuccio, L. A., Bergadà, M. M., Sampaio, J. D. & Sellami, F. 2010. Palaeolithic engravings and sedimentary environments in the Côa River Valley (Portugal): implications for the detection, interpretation and dating of open-air rock art. Journal of Archaeological Science 37: 3306-19. Aubry, T., Sampaio, J. D. & Luís, L. 2011. Approche expérimentale appliquée à l’étude des vestiges du Paléolithique supérieur de la Vallée du Côa (Portugal), pp. 87-96 in (A. Morgado, J. Baena Preysler & D. García, eds) La Investigáción Experimental Aplicada a la Arqueología. Vol. 1: Tecnología y Traceología Lítica Prehistórica y su Experimentación. Granada: Universidad de Granada. Aubry, T., Santos, A. T. & Luís, L. 2014. Stratigraphies du panneau 1 de Fariseu: analyse structurelle d’un système graphique paléolithique à l’air libre de la vallée du Côa (Portugal), pp. 259-70 in (P. Paillet, ed.) Les Arts de la Préhistoire: Micro-analyses, Mises en Contextes et Conservation. Actes du colloque ‘Microanalyses et datations de l’art préhistorique dans son contexte archéologique’, MADAPCA — Paris, 16-18 novembre 2011. Les Eyzies: SAMRA [Paleo, numéro spécial]. 131

Prehistoric Art as Prehistoric Culture gráficas Paleolítico-Postpaleolítico en la Sierra de San Pedro. Tajo Internacional. Cáceres. Trabajos de Prehistoria 67 (1): 197-209. Cardoso, D. D. F. 2014. A Arte Atlântica do Monte de S. Romão (Guimarães) no contexto da arte rupestre póspaleolítica da bacia do Ave — Noroeste Português. Vila Real: UTAD [PhD Thesis, unpublished]. Collado, H. 2006. Arte Rupestre en la Cuenca del Guadiana: el conjunto de grabados del Molino Manzánez (Alconchel—Cheles). Beja: Edia [Memórias d’Odiana—Estudos arqueológicos do Alqueva, 4]. Corchón, M. S., Hernando, C., Rivero, O., Gárate, D. & Ortega, P. 2012. La cueva de La Griega (Pedraza, Segovia, España) en la encrucijada ibérica: nuevos análisis del arte parietal paaleolítico a través del análisis factorial de corresondencias. Espacio, Tiempo y Forma, Serie I, Nueva época, 5: 527-42. Cruz, G. P. C. da & Cardoso, D. D. F. 2011. Arte rupestre de Briteiros. Investigação e possível musealização, pp. 255-71 in (M. A. Rodrigues, A. C. LIma & A. T. Santos, eds) V Congresso de Arqueologia — Interior Norte e Centro de Portugal. Caleidoscópio & Direcção Regional de Cultura do Norte. Figueiredo, S. C. S. de, Nobre, L., Gaspar, R., Carrondo, J., Cristo, A., Ferreira, J., Silva, M. J. D. & Molina, F. J. 2014. Foz do Medal terrace — an open-air settlement with paleolithic portable art. International Newsletter on Rock Art 68: 12-19. Fortea, J., Rasilla, M. de la & Rodríguez, V. 2004. L’art pariétal et la séquence archéologique paléolithique de la grotte de Llonín (Peñamellera Alta, Asturies, Espagne). Préhistoire, Art et Sociétés 59: 7-29. García, M. 2009. Grafismo mueble: las estaciones de Fariseu, Quinta da Barca Sul y Cardina I, pp. 37382 in (T. Aubry, ed.) 200 Séculos de História do Vale do Côa: Incursões na vida quotidiana dos caçadores-artistas do Paleolítico. Lisbon: Igespar, I. P. [Trabalhos de Arqueologia, 52]. García, M., Baptista, A. M., Almeida, M., Barbosa, F. & Félix, J. 2000. Observaciones en torno a las grafías de estilo paleolítico de la Gruta de Escoural y su conservación (Santiago de Escoural, Montemor-oNovo, Évora). Revista Portuguesa de Arqueologia 3 (2): 5-14. Glory, A., Vaultier, M. & Santos, M. F. D. 1965. La grotte ornée d’Escoural (Portugal). Bulletin de la Société Préhistorique Française. Études et travaux 62 (1): 110-17. Gomes, M. V. 1994. Escoural et Mazouco. Deux sanctuaires paléolithiques du Portugal. Les Dossiers d’Archéologie 198: 4-9. Gomes, M. V. 2002. Gruta do Escoural. Arte Parietal. Lisbon: IPPAR. Gomes, M. V. 2006. Catálogo del arte prehistórico de la Península Ibérica y de la España Insular. Arte Paleolítico II. Catálogo de Portugal. Série Arqueológica, Varia IV, pp. 85-162.

Gomes, M. V. 2007. Arte rupestre em Portugal. Os últimos 25 anos. Al-madan II série, 15: 120-24. Gomes, M. V. 2010. Arte rupestre do Vale do Tejo. Um ciclo artístico-cultural pré e proto-histórico. Lisbon: FCSH [PhD Thesis, unpublished]. Gomes, M. V., Almeida, F., Carvalho, A. M. F. D., Antunes, M. T., Silva, Z. C. G., Simão, J. A. R. & Pais, J. 2000. Cromeleque do Xarez. A ordenação do caos, pp. 17-190 in (A. C. Silva, ed.) Das Pedras do Xerez às novas terras da Luz. Beja: Edia [Memórias d’Odiana, 2]. Guy, E. 2000. Le style des figurations paléolithiques piquetées de la vallée du Côa (Portugal): premier essai de caractérisation. L’Anthropologie 104 (3): 415-26. Guy, E. 2003. Esthétique et préhistoire. Pour une anthropologie du style. L’Homme 165: 283-90. Henriques, F., Caninas, J. C. & Cardoso, J. L. 2011. Grafismos rupestres pré-históricos no Baixo Erges (Idanha-a-Nova, Portugal), pp. 199-217 in (P. Bueno, E. Cerrilo & A. González, eds) From the origins: The Prehistory of the Inner Tagus Region. Oxford: Archaeopress, [BAR International Series, 2219]. Hernández-Pacheco, E. 1919. La Caverna de la Peña de Candamo. Madrid: Museo Nacional de Ciencias Naturales [Comisión de Investigaciones Paleontológicas y Prehistóricas, Memoria número 24]. Jorge, S. O., Almeida, C. A. F. de, Jorge, V. O., Sanches, M. de J. & Soeiro, M. T. 1981. Gravuras rupestres de Mazouco (Freixo de Espada à Cinta). Arqueologia 3: 3-12. Lejeune, M. 1996. L’art pariétal de la grotte d’Escoural (Portugal): analyse critique, comparaisons et problèmes, pp. 137-240 in (M. Otte & A. C. Silva, eds) Recherches Préhistoriques à la Grotte d’Escoural (Portugal). Liège: Université de Liège [Eraul, 65]. Lenoir, M., Roussot, A., Delluc, B. & Delluc, G. 2006. La Grotte de Pair-non-Pair à Prignac-et-Marcamps (Gironde). Bordeaux: Société Archéologique de Bordeaux. Leroi-Gourhan, A., Delluc, B. & Delluc, G. 1995. Préhistoire de l’Art Occidental. Nouvelle édition revue et augmentée. Paris: Citadelles & Mazenod [L’Art et les Grands Civilisations, 1]. Lorblanchet, M., Cachier, H. & Valladas, H. 1995. Datations des chevaux ponctués du Pech-Merle. International Newsletter on Rock Art 12: 2-3. Luís, L. 2012. Desenhos animados! Uma gramática do movimento para a arte paleolítica do vale do Côa, pp. 69-80 in (M. de J. Sanches, ed.) Iª Mesa Redonda ‘Artes Rupestres da Pré-história e da Proto-história: paradigmas e metodologias de registo. Lisbon: Dgpc [Trabalhos de Arqueologia, 54]. Menéndez, M. 1984. La cueva del Buxu. Estudio del yacimiento arqueológico y de las manifestaciones artísticas. Boletín del Real Instituto de Estudios Asturianos 111: 143-85. 132

Santos, de Jesus Sanches and Castro Teixeira: The Upper Palaeolithic rock art of Portugal de Altamira (1879-1979), Madrid: Ministerio de Cultura. Silva, A. C. 2011. Uma Gruta Pré-histórica no Alentejo — Escoural. Évora: DRCALEN. Simón, M. D., Cortés, M. & Bicho, N. F. 2012. Primeras evidencias de arte mueble paleolítico en el sur de Portugal. Trabajos de Prehistoria 69 (1): 7-20. Valdez-Tullet, J. 2013. O abrigo de Foz Tua. A ampla diacronia de um espaço significante, pp. 355-66 in (J. C. Sastre, R. Catalán & P. Fuentes, eds) Arqueología en el Valle del Duero. Del Neolítico a la Antigüedad Tardía: Nuevas persectivas. Madrid: Ediciones de La Ergástula [Colección Simposia, 4]. Villaverde, V. 1994. Arte mueble de la España mediterránea: breve síntesis y algunas consideraciones teóricas. Complutum 5: 139-62. Zilhão, J. 1988. Plaquette gravée du Solutréen supérieur de la Gruta do Caldeirão (Tomar, Portugal). Bulletin de la Société Préhistorique Française 85 (4): 105-09. Zilhão, J. 2003. Vers une chronologie lus fine de l’art paléolithique de la Côa: quelques hypothèses de travail, pp. 75-90 in (R. de Balbín & P. Bueno, eds) Primer symposium internacional de arte prehistórico de Ribadesella. El Arte prehistórico desde los inicios del siglo XXI. Ribadesella: Asociación Cultural Amigos de Ribadesella.

Menéndez Fernández, M. & Martínez Villa, A. 19911992. Una tibia con ciervas grabadas de la cueva de La Güelga. Cangas de Onís, Asturias. Zehpyrus 4445: 65-75. Moure, A. 1981. Algunas consideraciones sobre el ‘mundo de los grabados’ de San Roman de Candamo (Asturias), pp. 339-52 in Altamira Symposium. Actas del Symposium Internacional sobre Arte Prehistórico celebrado en conmemoración del primer centenario del descubrimiento de las pinturas de Altamira (1879-1979), Madrid: Ministerio de Cultura. Obermaier, H. & Vega del Sella, C. de la 1918. La Cueva del Buxu (Asturias). Madrid: Museo Nacional de Ciencias Naturales [Comisión de Investigaciones Paleontológicas y Prehistóricas, Memoria número 20]. Pigeaud, R. 2004. La grotte ornée Mayenne-Sciences (Thorigné-en-Charnie, Mayenne). Gallia Préhistoire 46: 1-154. Reis, M. 2014. ‘Mil rochas e tal...!’: Inventário dos sítios da arte rupestre do Vale do Côa (conclusão). Portvgalia 35: 17-59. Ripoll, S. & Municio, L. J. 1999. Domingo García: arte rupestre paleolítico al aire libre en la Meseta. Valladolid: Junta de Castilla y León [Arqueología en Castilla y León, 8]. Roussot, A. 1990. Art mobilier et pariétal du Périgord et de la Gironde: compairaisons stylistiques, pp. 188-202 in (J. Clottes, ed.) L’Art des Objets au Paléolithique. Colloque international. Foix—La Mas-d’Azil. 16-21 novembre 1987, 1: L’art mobilier et son contexte. Paris: Ministére de la Culture, de la Communication, des Grands Travaux et du Bicentenaire. Sanches, M. de J. & Teixeira, J. C. 2013. An interpretative approach to ‘devil claw’ carvings: the case of river Tua Mouth Shelter (Alijó, Trás-os-Montes, Northeast Portugal), pp. 59-68 in (E. Anati, ed.) XXV Valcamonica Symposium 2013. Art as a source of History. Capo di Ponte: Centro Camuno di Studi Preistorici. Sanchidrían, J. L., Márquez, A. M., Valladas, H. & Tisnérat-Laborde, N. 2001. Dates directes pour l’art rupestre d’Andalousie (Espagne). International Newsletter on Rock Art 29: 15-19. Santos, A. T. 2012. Reflexões sobre a arte paleolítica do Côa: a propósito de uma persistente dicotomia conceptual, pp. 39-67 in (M. de J. Sanches, ed.) Iª Mesa Redonda ‘Artes Rupestres da Pré-história e da Proto-história: paradigmas e metodologias de registo. Lisbon: DGPC [Trabalhos de Arqueologia, 54]. Santos, M. F. dos, Gomes, M. V. & Monteiro, J. P. 1981. Descobertas de arte rupestre na Gruta do Escoural (Évora, Portugal), pp. 205-42 in Altamira Symposium. Actas del Symposium Internacional sobre Arte Prehistórico celebrado en conmemoración del primer centenario del descubrimiento de las pinturas

133

134

Old panels and new readings. La Pileta and pre-Solutrean graphics in Southern Iberia Miguel Cortés Sánchez,1-2-3 María D. Simón Vallejo,3-4 Rubén Parrilla Giráldez,3 and Lydia Calle Román3 Departamento de Prehistoria y Arqueología. Facultad de Geografía e Historia, Universidad de Sevilla, c/. María de Padilla, s/n. 41004. Sevilla (Spain). orcid: 0000-0001-9093-3338, e-mail: [email protected] 2 Interdisciplinary Center for Archaeology and Evolution of Human Behaviour, Universidadedo Algarve (Portugal). 3 Grupo HUM-949. Tellus. Prehistoria y Arqueología en el sur de Iberia. Universidad de Sevilla (Spain) orcid: 0000-0002-9894512X, e-mail: [email protected]; orcid: 0000-0003-2125-0765, e-mail: [email protected] 4 Museo Arqueológico of Frigiliana. c/. Cuesta del Apero, 10. 29788-Frigiliana, Málaga (Spain). orcid: 0000-0002-6885-1464,e-mail: [email protected]

1

Abstract La Pileta was the first Palaeolithic art site to be discovered in Southern Iberia, but it has never been published in detail. The existence of pre-Solutrean graphic horizons was barely mentioned in earlier works, so all the Pleistocene art in La Pileta is usually associated with the Solutrean and Magdalenian. Subsequently, direct 14C-AMS dating of an aurochs in Horizon-C of Sanchidrián et al. indirectly raised the possibility of pre-Solutrean rock art in the site. In this work, we will try to re-evaluate the presence of a pre-Solutrean artistic cycle in La Pileta and, by extension, in Southern Iberia. Keywords Palaeolithic art; south of Iberian Peninsula; Presolutrean chronology

In memory of a ‘... rude mais passionnante campagne...’ Breuil et al. 1915 of the 20th century, the existence of pre-Solutrean graphic horizons was barely mentioned in later works (Dams 1978); so all the Pleistocene art in La Pileta is usually associated with the Solutrean and Magdalenian (Sanchidrián 1997). Subsequently, direct 14C-AMS dating of an aurochs in Horizon-C of Sanchidrián et al. (2001) indirectly raised the possibility of pre-Solutrean rock art in the site. Finally, Fortea (2005) suggested this possibility, based on techno-stylistic parallels.

Introduction Professor R. de Balbín Berhman has approached one of the most fascinating topics in the study of human symbolism, Palaeolithic art, and contributed works of great significance regarding issues such as its origin; change and survival in its manifestations; extending its expression to open-air sites; and working on some of the most representative sites. The result has left a singular imprint on the historiography of this topic.

In this work, we will try to re-evaluate the presence of an pre-Solutrean artistic cycle in La Pileta and, by extension, in Southern Iberia.

Coinciding with the tribute in this volume, a century has gone by since the first scientific publication on La Pileta (Breuil et al. 1915), the first Palaeolithic art site to be discovered in Southern Iberia. This cave remains one of the essential places for studying some issues related to the earliest stages and the processes of innovation and survival of Palaeolithic graphics in the south-western corner of Europe.

A new reading Archaeology In 1942 La Pileta´s main excavation was undertaken in the Murciélagos Chamber (Fig. 1). A stratigraphy with a 7 m sedimentary fillwas documented, but its archaeological sequence was never published properly. Thus the site’s cultural sequence had to wait for recent reviews (Cortés & Simón 2008) to assess the presence of UP-levels. It is also important to remember that the existence of an Early Upper Palaeolithic (EUP) sequence in southern Iberia was only confirmed recently (Aura et al. 2010, with references; Cortés 2007; Cortés et al. 2012; etc.).

La Pileta has been the subject of important archaeological excavations, but the findings have never been published in detail; so the structure of its Pleistocene art has been analyzed almost exclusively from a techno-stylistic perspective, and it has always been linked to the historical vicissitudes of knowledge about the Upper Palaeolithic (UP) in southern Iberia. Accordingly, except for the proposal by Breuil et al. (1915), based on parallels with ‘Franco-Cantabrian’ sites discovered at the beginning 135

Figure 1. Topography and first artistic horizon of La Pileta; B/ (Nº Breuil Panel). Overlap of the proposal by Breuil et al. (1915) and Sanchidrián (1997). Zoomorphic tracing after Sanchidrián 1997, except B/13, 25, 29 (Breuil et al. 1915).

Prehistoric Art as Prehistoric Culture

136

Cortés Sánchez et al: Old panels and new readings. La Pileta and pre-Solutrean graphics in Southern Iberia Along these lines, the potential archaeological evidence of EUP in La Pileta could be a couple of projectile points (with a long and flattened section), one Gravette point (Cortés & Simón 2008) and, above all, a lamp, whose 14 C-AMS date would be chronologically concordant with EUP (Cortés et al. in press).

existence of a hind, whose conventions are reminiscent of Cantabrian ‘flathead hinds’ (Fig. 1A), associated with the Gravettian-Early Solutrean. However, Horizon-B was finally classified between Horizon-A and C, both Early Solutrean (Sanchidrián & Márquez 2003: 282, fig. 3); and so, from a graphical cross-referencing point of view, Horizon-B would also be Solutrean. As for the Rhino panel, this author reduced it to two zoomorphic figures and various black vertical strokes.

Old panels. La Pileta Review of interpretations

Later, Fortea (2005) produced a review of the earliest evidence for Palaeolithic art in Iberia, and focused on the Rhino Panel and the yellow serpentiforms horizon in La Pileta (Fig. 1E and 3E). This was not an accidental choice but a ‘Gordian knot’ in the graphic sequence. This researcher travelled to the site on several occasions and, after analyzing the different panels, identified some of the yellow horizon’s particularities, made by a ‘slimy’ pigment applied with the fingers in large lumps, which delineate some zoomorphs and meandering forms. Likewise, and partially matching Dams (1978: 38, Fig. 31), the panel is completed by four positive hand prints (Fig. 2A). The parallels and chronology attributed to this technique in some French Mediterranean sites (i.e. Baume-Latrone, Azéma et al. 2012, with references) led him to propose a pre-Solutrean age for the panel and the horizon (Fortea 2005). However, the administrative circumstances affecting La Pileta prevented Fortea from obtaining the necessary documentation to come up with a more detailed presentation.

We do not know how much time was devoted to each of the 59 panels/areas that Breuil catalogued during the 27 days of his campaign (20March - 15 April 1912), but the work was carried out in difficult conditions (logistics, speleological or lighting difficulties), including exploration, surveying, prospecting surface remains and the art, its documentation, and finally the excavation of two areas in the site, Vacas and Murciélagos (Fig. 1) (Breuil et al. 1915). Recordings were subsequently made by tracing from B/W photographs, focusing, above all, on the main panels with Palaeolithic art. Panel 19 is an example, limited to two zoomorphics (Breuil et al. 1915: Pl-III). In other cases, there are no photographs or tracings, perhaps due to the fact the shots were wrong. As a result, although one acknowledges the huge work that was carried out, it cannot be used as an authoritative reference to rule out new findings or reinterpretations of the panels. Breuil talks about the existence of three artistic Palaeolithic cycles (Breuil et al. 1915). The earliest, designated by yellow pigments, would be in an ‘Aurignaco-Perigordian’ chronology and would include several panels (Fig. 1), among them one located at the end of the Cabras Gallery (no. 19: Breuil et al. 1915), now known as the Rhino Panel (Fig. 1D and 2).

Knowing of our work on the archaeological material and our files on La Pileta (Cortés & Simón 2008), Fortea asked us to try to locate documentation on the paintings. His sudden death made it impossible to continue this line of work with our collective documentation. This unexpected situation made us return to the previously collected information and reread the Rhino panel.

In the 1970s, Dams (1978) undertook three campaigns in La Pileta and expanded Breuil’s inventory. For example, the Rhino Panel (20A=19-Breuil) was enlarged in size and number of motifs, as it includes hands, curved signs and different marks. However, its graphic documentation (topography, tracings and photographs) is qualitatively poor, and, at times, their interpretations are questionable. These circumstances have limited the assessment of the work done and hindered the use of her tracings for later historiography.

Rhino panel The Bullón family, owners of La Pileta, kindly provided us with their personal files and photographic footage, which are of unquestionable interest. We can present this information here thanks to their authorization and generosity. The new material enables us to produce a very consistent reading of the panel, providing its true dimensions and identifying the presence of two animals, four positive hand prints, sinuous motifs and very faded signs of various pigments (Fig. 2).

Later, Sanchidrián (1997) carried out a brief campaign (November 1985) with better lighting and analogue photography. This author catalogued 872 motifs and grouped them into ten techno-stylistic blocks assigned to the Solutrean and Magdalenian. Subsequently, from a direct AMS-date (20,130±350 BP, Sanchidrián et al. 2001) from the pigment of an Horizon-C bovid, this author alluded to a possible ageing of phases A and B (Sanchidrián & Márquez 2003), as well as the

a) Animals The interpretation of the panel (Fig. 2) has been controversial because the zoomorphic figures have been classified, from left to right, as a rhino and aurochs (Breuil et al. 1915), two aurochs (Dams 1978), aurochs and cervid (Sanchidrián 1997), or rhino and wild goat (Fortea 2005). 137

Prehistoric Art as Prehistoric Culture

Figure 2. Rhino Panel: A) Hands, B) Rhino and aurochs, C) Serpentiforms. Tracing after Sanchidrián 1997 (A, E and, partially, D); Breuil et al. 1915 (B-C).

138

Cortés Sánchez et al: Old panels and new readings. La Pileta and pre-Solutrean graphics in Southern Iberia Where the rhino is concerned, some authors had reservations about Breuil’s view, mentioned at the bottom of a figure (‘Rhinocéros?’) (Breuil et al. 1915: Pl. V.). Breuil described it as ‘...un vrai rhinocéros’ in the original work (Breuil et al. 1915: 19), in lectures (University of Seville, Hazañas y La Rúa, 1918) and in subsequent publications (Breuil & Obermaier 1984: 127). However, these references have remained unnoticed until now.

Four clearer positive hands (Fig. 2A) are located at more than 2 m above the floor of the gallery, so they must have been made by standing on a ledge a few centimeters wide in the gallery wall. This ledge is located quite close to the bottom of the rhino. In this sector, the width of the gallery is more than 3 m and there is no hint of any speleothem, gallery or duct to encourage exploration, which might have led to an accidental application of hands on the wall. These hands cannot be associated with contemplation of the zoomorphic form on the Rhino panel, because, once one is up on this minimal cornice, the motifs are well below one’s visual field, and the face is too close to the wall, making it impossible to see.

Techno-stylistic analysis shows that the strokes interpreted by Breuil as a horn are part of the animal representation (Fig. 2). Thus, the main line forms an open mouth made by strokes defining the nose and the jaw, outlining them and adapting to the previous design, a few millimeters from both, following a pattern already seen, for example, on some portable art in Vale Boi (Simón et al. 2012).

The palms of the hands were impregnated with pigment, and then pressed onto the wall carefully, leaving a very clear outline of the naked hand. In contrast, there are no hand slides caused by accidental displacement, thus showing that the application could have been made with a ‘tampon’.

In addition, the extremities of the lines that define the zoomorphic figures always end in inflections which try to shy away from a straight line. This trend is also followed by the sign inside the rhino (an inverted U) as well as the meanders. So the horn, if interpreted in isolation from the rhino, would constitute an anomalous element in the set.

Direct observation of these motifs reveals the use of a very homogeneous organic pigment, possibly made with grey-black ash, and probably applied with a thick liquid to the palms of the hands; its composition is not related to the clays around the La Pileta area, a fact that could be further investigated in the future.

Likewise, the inventory of rhinos in European Pleistocene art in the mid-20th century (Nougier & Robert 1957) indicated great variability in the representation of this species. In La Pileta, the presence of a single horn allows us to identify it as a steppe rhino (Stephanorhinus hemitoechus).

c) Other symbols and vestiges The area where these vestiges are located appears to be dotted with faded pigment remnants, which, if processed with an appropriate imaging technique, may reveal new evidence. Among these, we can highlight some faded remnants of red pigment, i.e. just above the two central hands of the group, several groups of two or three black strokes above the rhino. There are other vestiges whose interpretation is unclear, like a reddish curved sign, already mentioned by Dams (1978: fig. 31) and a serpentiform on a blackened area of wall close to the conduit that leads to the Serpientes Chamber (Fig. 2A-B).

Finally, it should also be remembered that the rhino has been found in archaeological Palaeolithic sites near La Pileta, such as Gorham’s Cave during the Solutrean (Riquelme et al. 2011), and there are even more recent graphic representations of these animals in Iberia (i.e. Los Casares: Balbín & Alcolea 1992), so this extends the motif to the southernmost areas of Iberia. Another animal represented is the front third of an aurochs, showing a body in profile and frontal U-shaped horns; the left horn is partially lost (Fig. 2).

The Rhino Panel’s extent is therefore much greater than was claimed by Breuil et al. (1915: Pl-III) or Sanchidrián (1997), and quite different from Dams (1978).

b) Hands

Other pre-Solutrean panels

The positive hand prints were observed and approved by Fortea (2005), and earlier by Dams (1978). They are located in a darkened area 90 cm above the top of Breuil’s rhino-aurochs 19-group (Fig. 2A). On the right hand side of this group, one finds two right hands. The outermost is on an upper level, above the previous fingermarks. The innermost overlaps another hand, possibly a left hand, while at the far left of the group, at the same height as the previous ones, another left hand print appears. Dams (1978) identified a fifth mark, not verifiable with the available documentation.

The techno-stylistic characteristics of the Rhino panel also appear in other areas of La Pileta. Following the nomenclature of Breuil et al. (1915), we find: Panel-2, 10, 16 and 35: five horses painted in yellow (Fig. 1). Panel-29: engraved horse with multiple strokes, with an inner sign similar to the rhinoceros´s interior (Rhino Panel), which is superimposed by a yellow equine (Fig. 1B and 3F). 139

Prehistoric Art as Prehistoric Culture

Figure 3. A) Black serpentiforms, B) Red sinuous sign, C) Positive hand prints, D) Serpentiform, E) Yellow ibex, F) Yellow horse. E-F, tracing after Breuil et al. 1915.

140

Cortés Sánchez et al: Old panels and new readings. La Pileta and pre-Solutrean graphics in Southern Iberia Panel-13 (Fig. 1C and 3E): a wild goat (Capra pyrenaica) and serpentiform lines in yellow, with a similar execution to the Rhino Panel.

upwards and few anatomical details; some sinuous signs are linked to them.

Likewise, some of the zoomorphic figures identified by Sanchidrián (1997) in Horizon A-B could fit into this group.

Moreover, the serpentiform-sinuous signs are not exclusive to the Magdalenian period: one can also find them in early artistic contexts in Gargas and Hornos de la Peña (Breuil et al. 1915: 18), Baume-Latrone or Llonín (Fortea 2005). Examples can also be found in Gravettian contexts, for example, engraved motifs on blocks in the Abri Pataud, recovered in Middle Gravettian levels (vid. Jaubert 2008).

Chronology

Columnas Chamber

One of the fundamental interpretative problems for the Rhino Panel is chrono-stylistic. Thus, Breuil, Dams and Fortea included these yellow graphics in the Cabras and Serpientes chambers (Fig. 2, 3C) in the Early European Palaeolithic art cycle, whereas others thought they could belong to the Magdalenian (Horizon-G, Sanchidrián 1997). The latter author based his view on some portable parallel meanders (a plaquette from Nerja and several plaquettes from Parpalló, Upper Madalenian) and the superimposition of three yellow-brown parallel digit strokes, very faded, from Horizon-G in the Serpientes Chamber on a washed red slick spot associated with matched strokes (Horizon-E, Evolved Solutrean) (Sanchidrián 1997: 426, Fig. 11: 5).

There are others positive hand prints in La Pileta, located in the deepest part of the cave (discovered in 1933), which were associated with four human skeletal remains on the surface, located in different areas in the Nuevas Galerías (Fig. 3C, Pérez & Maura 1936). These authors created a ‘narrative’ which includes an accidental episode during the Bronze Age and the execution of accidental hand stains impregnated with clay by some of these unfortunate individuals.

Panels-24-25 and 27: serpentiform motifs with 2-3-5 yellow waves with the same number of fingers (Fig. 1E and 3D).

The positive hand prints correspond to two hands, possibly to a single adult. The left print is located at a higher level, quite separate from the other, which entailed a difficult position for the individual. Both prints comprise a larger spot at the bottom, probably corresponding to the palm of the hand impregnated with diluted soaked clay material. Above these, there are five prints probably made with the fingers and a bent carpal, but separate from each other. The reverse of the metacarpal was leaning on the wall, impregnated with the same matter. The left thumb print is bigger and corresponds to a fingertip, while the right one has longer strokes. The final result (Fig. 3C) evokes the shape of a paw. Thus, there was a dense and homogeneous impregnation of matter onto the palm and reverse of the fingers. The forced posture, the planning and careful execution already evoke common practices like the ‘mutilated hand stencils’ of Pleistocene art (i.e. Gargas or Maltravieso).

However, this overlap is a very weak argument, as it omits eight cases of yellow motifs underneath figures in other colours in La Pileta (Breuil et al. 1915: 16). In addition, data from a lamp in this cave shows how the pre-Solutrean palette was not monochrome but included yellow, red and, perhaps, black (Cortés et al., in preparation). As we have already said, the Rhino Panel is a crucial part of the graphic sequence of La Pileta. In this sense, Breuil´s and Fortea´s proposal seems to better fit reality; and, accordingly, in this panel, the serpentiform-meanders, made with a plastic paste, and some zoomorphic forms would belong to the earliest cycle of La Pileta rather than the Late Magdalenian.

Concerning the possible link between the human remains and the positive hand prints (Fig. 1F), the closest ones are located at least 10 m away, but we have no direct chronology (no human remains have been dated) or relative chronology (there are no archaeological remains in the Nuevas Galerías), and the skeletons in a primary context (more or less) were in an extended position, with no goods,a rare funerary pattern during the Bronze Age in southern Iberia. Moreover, we cannot forget the link between hands and human remains found in some archaeological sites attributed to the Gravettian (i.e, Cussac or Vilhonneur: Jaubert 2008, with references).

The same can be said of other motifs which have characteristics of Early Palaeolithic art, such as a ‘yellow disc’ underneath red and black strokes (PI, IX-2, Breuil et al. 1915) and some of the faded stains. Similarly, some features display zoomorphic conventions which are usually pre-Solutrean graphic indicators (Fig. 1). Thus, the techno-stylistic analogies and themes (rhinoceroshand prints) of the Rhino Panel horizon are to be found elsewhere, for example, in La Baume-Latrone (Azéma et al. 2012). One can mention the use of thick pigments applied with fingers (three fingers); the animals depicted tend to be rhinos and wild goat in large format (75 cm in La Pileta), with dorsal curves that curve obliquely

Finally, it is worth noting the human hemi-mandible found in a crack in the niche where a yellow horse of the 141

Prehistoric Art as Prehistoric Culture

Figure 4. EUP and archaeological sites mentioned in the text.

old graphic cycle can also be found (Fig. 1B, Breuil et al. 1915: 10); no attention has yet been paid to this, but it should be reviewed.

with a fully developed symbolism during this UP stage in this geographical area. Hands in the South of Iberia

In conclusion, since there are no more data, these positive hand prints and the anthropological remains found in the Nuevas Galerías (Fig. 1F and 3C) should not be intuitively attributed to Recent Prehistory, since the hand prints could possibly belong to the Pleistocene.

During the 21st century, the identification of hands in the rock art of southern Iberia is experiencing significant growth. Thus, there are two negative and several doubtful positive hands in Ardales, one positive in Higuerón and another in Victoria (Cantalejo et al. 2006; 2007), and there are others still unpublished in Gorham. In addition, we should add the positive hand prints in rock-shelters (Fig. 4), such as Ladrones or Pretina I (Mas 2006); these are closer to being open-air art (so profusely documented in Iberia in recent years, vid. Balbín 2008) than to the deep cave art of La Pileta. However, there are parallels between both, which makes us think that they probably just represent different variants of the same reality, expressed by the hunter-gatherers at the western end of the Baetic System’s subsistence territory, adapting ad hoc to the available possibilities and to the geomorphological constraints of the landscape.

Reflections Early Upper Palaeolithic symbolism The EUP in southern Iberia begins with the Aurignacian (Bajondillo), while the number of sites increased considerably during the Gravettian (vid. Cortés et al. 2012, with references). If this technocomplex seems to represent a moment of consolidation of the population during the UP in this geographical area (Fig. 4), it is logical to think that these communities brought not only a technological culture but a different cultural heritage.

Other pre-Solutrean rock art candidates

Thus, the symbolic world seems to have been expressed in manufacturing and using pigments and personal ornaments on malacological supports (Nerja, Gorham and Vale Boi) or portable art (Vale Boi) (Cortés et al. 2012, with references); so, to match other areas in Western Europe, we only need to find the rock art. In this sense, taking into account the relationships between the Gravettian of the western end of Europe and the other northern regions or the proliferation of pre-Solutrean art in the Cantabrian corniche (i.e. Fortea 2005; Pike et al. 2014) as in the Spanish Levante (Villaverde 1994), the absence of Pre-Solutrean graphics in southern Iberia would be an abnormal phenomenon. Thus, we consider the techno-stylistic parallels in La Pileta’s rock art to be consistent with the existence of an early horizon, and

This work enables us to consider all the ‘candidates’ that can be integrated into pre-Solutrean Palaeolithic Art (Fig. 4). However, we need to mention some other sites where one can find signs attributable to this stage (i. e. Ardales and, maybe, some motifs in Nerja, not including the ‘pisciforms’, Simón & Cortés 2011) and others that have been already mentioned by several authors (i.e. Moro, Piedras Blancas, Toro or Victoria), which must be re-assessed in the future. Conclusion La Pileta is a landmark for the symbolic manifestations of prehistoric communities in southern Iberia, particularly 142

Cortés Sánchez et al: Old panels and new readings. La Pileta and pre-Solutrean graphics in Southern Iberia during the UP. The revision of its excavated material and surface finds has enabled us to point to a frequentation of the cave during the EUP, perhaps in the Aurignacian at times (possibly some spear points) and, more strongly, in the Gravettian period.

Balbín Behrmann, R. de & Alcolea González, J. J. 1992. La Grotte de Los Casares et l’Art paléolithique de la Meseta Espagnole. L’Anthropologie 96 (2-3): 397– 442. Breuil, H. & Obermaier, H. 1984. La Cueva de Altamira en Santillana del Mar. El Viso: Madrid. Breuil, H., Obermaier, H. & Verner, W. 1915. La Pileta à Benaoján (Málaga)(Espagne). Institut de Paléontologie Humaine: Monaco. Cantalejo, P., Maura, R., Espejo, M.-M., Ramos, J., Medianero, J., Aranda, A. & Durán, J. J. 2006. La Cueva de Ardales: Arte prehistórico y ocupación en el Paleolítico Superior. Estudios, 1985-2005. CEDMA: Málaga. Cantalejo, P., Maura, R., Aranda, A. & Espejo, M.M. 2007. Prehistoria en las cuevas del Cantal. La Serranía: Málaga. Cortés-Sánchez, M. 2007. El Paleolítico Medio y Superior en el sector central de Andalucía (Córdoba y Málaga). Monografías Museo de Altamira 22. Ministerio de Cultura: Madrid. Cortés-Sánchez, M., Marreiros, J. M., Simón-Vallejo, M. D., Gibaja-Bao, J. F. & Bicho, N. 2012. Reevaluación del Gravetiense en el sur de Iberia’, pp. 73-85 in (C. de las Heras et al., eds) Pensando el Gravetiense: nuevos datos para la región cantábrica en su contexto peninsular y pirenaico. Monografías del Museo de Altamira 23. Museo Nacional y Centro de Investigación de Altamira: Madrid. Cortés-Sánchez, M. & Simón-Vallejo, M. D. 2008. La Pileta (Benaoján, Málaga) cien años después. Aportaciones al conocimiento de su secuencia arqueológica. Saguntum 40: 45-64. Dams, L. 1978. La Pileta. Akademische Druck-u Verlagsanstalt: Graz. Fortea, J. 2005. La plus ancienne production artistique du Paléolithique ibérique, pp. 89-99 in (A. Broglio & G. Dalmeri, eds) Actas del Simposio Pitture paleolitiche nelle Prealpivenete: Grotta di Fumane e Riparo Dalmieri. Memorie del Museo Civico di Storia Naturale de Verona. Preistoria Alpina, Nr. speciale: Verona. Hazañas y La Rúa, J. 1918. Apuntes de dos conferencias dadas por el Abate H. Breuil en la Universidad de Sevilla (20-21 de Marzo de 1918). Sobrinos de Izquierdo: Seville. Jaubert, J. 2008. L´Art pariétal gravettien en France: éléments pour un bilan chronologique. Paléo 20: 439-74. Mas, M. 2006. Las cuevas de los Ladrones o Pretinas (Benalup-Casas Viejas, Cádiz): de la icnología al arte, una auténtica ‘Action Painting’ de un grupo de cazadores recolectores. Zona arqueológica 7 (2): 7584. Nougier, L. R. & Robert, R. 1957. Le rhinocéros dans l’art franco-cantabrique occidental. Bulletin de la Société Préhistorique de l’Ariège XII: 15-52.

Most of the Palaeolithic graphics seem to be well linked to the Solutrean and Magdalenian periods, but also, in accordance with the proposal by Breuil, Dams and Fortea, there is also a pictorial style-horizon in the early cycle of European Palaeolithic art. This is characterized by meandering and zoomorphic figures (rhino, aurochs, wild goat and horses, Fig. 1), of great size (75 cm), designated by a yellowish pigment applied with a style and technique that have clear parallels in other sites in Western Europe, and with the addition of positive hand prints. This horizon highlights the importance of La Pileta, and, by extension, southern Iberia which was, since ancient times, a piece of the puzzle of techno-cultural complexity and symbolic implementation during the EUP in the extreme west of Europe. A hundred years after the first findings, La Pileta still needs a comprehensive programme of study to document adequately its different occupations in the UP and the chrono-stylistic evolution of its artistic manifestations.

Acknowledgements This work is a contribution to the investigative projects HAR2013-44269-P (Ministerio de Economía y Competitividad), HUM949-Group (University of Seville), ICAEFB (University of Algarve). Study of material from La Pileta, Benaoján, deposited in the Museum of Málaga (IDPH/JT/18/05/PU/MA) has been authorised by the Ministry of Culture of the Junta de Andalucía to one of the authors (MCS). References Aura, J. E., Jordá, J. F., Pérez, M., Badal, E., Morales, J. V., Avezuela, B., Tiffagom, M. & Jardón, P. 2010. Treinta años de investigación sobre el Paleolítico superior de Andalucía: la Cueva de Nerja, pp. 17398 in (X. Mangado, ed.) El Paleolítico superior peninsular. Novedades del siglo XXI. Homenaje al profesor Javier Fortea. Monografies SERP 8. Universidad de Barcelona: Barcelona. Azéma, M., Gély, B., Bourrillon, R. & Galant, P. 2012. L´art paleólithique de la Baume Latrone (France, Gard): Nouveaux éléments de datation. INORA 64: 6-12. Balbín Behrmann, R. de (ed.) 2008. El Arte Rupestre Paleolítico al aire libre en la Península Ibérica. Junta de Castilla y León. 143

Prehistoric Art as Prehistoric Culture Pérez de Barradas, J. & Maura Salas, M. 1936. Nuevos descubrimientos en la Cueva de la Pileta. Extracto de Notas y Comunicaciones del Instituto Geológico y Minero de España: Madrid. Pike, A. W. G., Hoffman, D. L., García, M., Pettitt, P. B., Alcolea, J.et al. 2012. U-Series Dating of Palaeolithic Art in 11 Caves in Spain. Science 336: 1409-13. Doi:10.1126/science.1219957 Riquelme, J. A., Finlayson, C., Giles, F., Rodríguez, J. & Santiago, A. 2011. La fauna de mamíferos solutrense de Gorham´s Cave, Gibraltar, pp. 16178 in (J. J. Fernández Caro & T. Baena Escudero, eds) Arqueología, paleontología y geomorfología del Cuaternario en España: XX Aniversario del seminario Francisco Sousa (La Rinconada, Sevilla). Ayuntamiento de La Rinconada: Seville. Sanchidrián, J. L. 1997. Propuesta de la secuencia figurativa en la Cueva de La Pileta, pp. 411-30 in El món mediterrani després del Pleniglacial (18.000-12.000 B.P.).Centre d´Investigacions Arqueològiques, Girona. Sèrie Monogràfica 17. Museo d´Arqueologia de Catalunya: Gerona. Sanchidrián, J. L. & Márquez, A. M. 2003. Radiodataciones y sus repercusiones en el arte prehistórico malagueño. Mainake 25: 275-92. Sanchidrián Torti, J. L., Márquez Alcántara, A., Valladas, H. & Tisnerat, N. 2001. Dates directes pour l’art rupestre d´Andalousie (Espagne). INORA 29: 15-19. Simón Vallejo, M. D., Bergadà Zapata, M. M., Gibaja Bao, J. & Cortés-Sánchez, M. 2011. El Solutrense meridional ibérico: el núcleo de la provincia de Málaga. Spal 20: 67-80. Simón-Vallejo, M. D., Cortés-Sánchez, M. & Bicho, N. F. 2012. Primeras evidencias de arte mueble paleolítico en el sur de Portugal. Trabajos de Prehistoria 69 (1): 7-20. Doi: 10.3989/tp.2012.12076 Villaverde, V. 1994. Arte paleolítico de la Cova de Parpalló. Estudio de la colección de plaquetas y cantos grabados y pintados. SIP-Diputación de Valencia: Valencia.

144

Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation Valentín Villaverde Universitat de València

To Rodrigo de Balbín, whose reflections have enlightened our view of Palaeolithic Art Abstract The purpose of this study is to present a general approximation of Mediterranean Palaeolithic art, paying special attention to its distribution, technical characteristics and themes, as well as some sequential and chronological aspects. In this brief synthesis we will not consider the non-figurative decorated bone instruments which are quantitatively unimportant except in a very few sites. Keywords Palaeolithic art; Mediterranean Iberian Peninsula

throughout much of the Upper Palaeolithic in the Iberian Mediterranean region, and these traits extend beyond the strict Mediterranean coast, reaching the southern Iberian Atlantic.

Introduction The number of finds of portable and parietal Palaeolithic art have increased considerably in recent decades in the Iberian Mediterranean region. Although some of these finds have been published, only a few of the supraregional analyses have proposed an updated synthesis of Palaeolithic art from this wide geographical region. These include the work carried out by Fortea (1978) which, according to the publication dates, predates much of the current inventory of caves and known objects; a general approximation of the sequence produced by Villaverde (2005); and the more specific works focused on Andalucia, presented by Sanchidrián (1994), Martínez (2009) and Cortés (2010). In addition to these studies, one should mention the joint assessment of the art in the Serranía de Ronda (Cantalejo et al. 2006) and the recent revision of the art in the Campo de Gibraltar (Ruiz et al. 2014).

In other words, the material culture of the different stages of the Upper Palaeolithic presents enough similarities for us to consider the existence of a cultural community throughout this area, irrespective of the groups or territorial units that undoubtedly existed in a territory that covers a coastal route of over 1300 km from Southern Catalonia to the Portuguese Estremadura. In terms of artistic analysis, the Iberian Mediterranean region meets at least the same requirements as the Cantabrian region to qualify for a study of Palaeolithic art in regional terms: the means and resources are similar, the rhythm of industrial change has a similar chronological pace, and the industries have a common style (Fullola et al. 2005). In my opinion this statement does not contradict the possible existence of territorial variations associated not only with the possible differences in the raw materials available or the particularity of local food resources, but also with the existence of territorial units of a group nature, due to the geographical extent of the analysed region. The strongest justification for this circumstance is in the existence of certain concentrations of archaeological sites with evidence of important occupation in some periods, and with a certain number of sets of rock art associated with them.

The purpose of this study is to present a general approximation of Mediterranean Palaeolithic art, paying special attention to its distribution, technical characteristics and themes, as well as some sequential and chronological aspects. In this brief synthesis we will not consider the non-figurative decorated bone instruments which are quantitatively unimportant except in a very few sites. Addressing the study of Mediterranean Palaeolithic art is a task that could even be questioned in advance, because the areas under analysis must present common culture traits in aspects other than artistic expression – traits that indicate the existence of contacts and social networks of some extent and continuity, which justify a comparative study of the graphic expressions. In this regard, it can be said that various components of material culture, both lithic and bone related, exhibit remarkable coincidences

On the other hand, it is known that for some time the Palaeolithic art of the Iberian Mediterranean region was regarded as a peripheral and atypical region with respect to the Franco-Cantabrian art (Leroi-Gourhan 1965), as the graphic expression of a region closely related to Italy, especially its southern part (Graziosi 1956, 1964), at the end of the Palaeolithic artistic cycle. In contradiction 145

Prehistoric Art as Prehistoric Culture

Figure 1. Distribution of sites with parietal art in the Mediterranean and the Southern Iberian Peninsula. The circles indicate the main areas of concentration. 1. Bora Gran; 2. Tut de Fustanyà; 3 Cova de la Taverna; 4. Molí del Salt; 5. Hort de la Boquera; 6. Sant Gregori; 7. Cova de la Moleta; 8. Abric de Melià; 9. Cova de Bovalar; 10. Abric del Cingle del Barranc d l’Espigolar; 11. Cova Matutano; 12. Cova dels Blaus; 13. Roca Hernando; 14. Abrigo de los Morenos; 15. Cova de les Malladetes; 16. Cova del Parpalló; 17. Cova de les Meravelles; 18. Cova del Comte; 19. Abric del Tossal de la Roca; 20. Cova Fosca; 21. Cova de Reinós; 22. Cueva del Niño; 23. Cueva de Jorge; 24. Cueva de las Cabras; 25. Arco I y II; 26. Cueva de Ambrosio; 27. Cueva de Almaceta; 28. Piedras Blancas; 29. Cueva del Morrón; 30. Cueva de Malalmuerzo; 31. Cueva de la Ermita del Calvario; 32. Cueva de Nerja; 33. Cueva del Higuerón; 34. Cueva Victoria; 35. Cueva del Tesoro; 36. Cueva del Toro; 37. Cueva de Ardales; 38. Cueva de la Pileta; 39. Cueva del Gato; 40 Cueva de El Pirulejo; 41. Cueva Horadada; 42. Gorham’s Cave; 43. El Ciervo; 44. Las Palomas I: 45. Cueva del Vencejo Moro; 46-49. Atlanterra, El Realillo, la Jarsa I and Caminate; 50. Cueva de la Motilla; 51 Vale Boi; 52 Grotta do Escoural.

of this view, which led to the distinction between an artistic Mediterranean province and a Franco-Cantabrian province, a series of criticisms have been developed in recent years that incorporate the current knowledge of Palaeolithic art in the regions involved into the original proposal by Graziosi. These criticisms draw attention to the difficulties that must be taken into consideration, at least in the terms that they were formulated at that time: the lack of consistency when formulated in terms that encompass the entire Palaeolithic sequence, or even some of the main evolutionary phases; the chronological differences between some of the compared thematic and stylistic elements; and how problematic it is to establish a cultural relationship between remote regions –- the Iberian Mediterranean region and the South of the Italian Peninsula and Sicily — they are separated by major geographical areas that lack precisely the elements of material culture that would justify the existence of contacts (Villaverde 2004, 2005a, 2005b).

Sites with art: distribution and patterns It is necessary to point out at the start that it would be more appropriate, when referring to the art in the region to which this study is devoted, to talk about art from the Mediterranean region and Southern Iberia, given that it is possible to establish traits of cultural units from Southern Catalonia right to Andalucia’s Atlantic coast and the Portuguese Estremadura. This is something that we do not do outright, only because there is a certain tradition in the use of the Mediterranean term in a loose sense, to encompass these other areas. On the other hand, it is logical to suggest that this vast territory has undergone changes over time and presents certain extensions to inland areas in the southern half of Iberia. However, except in very few specific cases, we have not incorporated in this region the sites with art whose inclusion would be based only on the consideration of stylistic traits, which could be the result of wide-scale 146

Villaverde: Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation contacts but of a limited nature in chronological terms, and that are not accompanied by a common industrial context capable of sustaining cultural similarity between the compared elements. It is unthinkable in this regard to propose the existence of a strictly littoral Mediterranean and South Atlantic region disconnected from the hinterlands, in the same way as it is impossible to establish a Mediterranean region isolated or disconnected from the important centres of Palaeolithic occupation found in the Cantabrian region or the Pyrenees and the South of France. But while the contacts and influences in these regions could be sporadic or limited to certain industrial phases, and much of the time indirect, the identity of industrial processes in the Mediterranean and South Atlantic regions refers to the existence of contacts with some continuity, capable of contributing to the general unity of styles in the artistic representations and in the components of the material culture.

with engraved figurative motifs; in the other two, the northernmost sites with portable art, had a plaquette with lines painted in red, a doubtful engraving of a figurative motif and the representation of some engraved cervid heads; this scant evidence is concentrated mainly in southern areas and also around the mouth of the Ebro River, and relatively close to the group of finds in northern Castellon. In this other area the parietal expressions appear to be of late chronology: the Abric Melià (Serra d’en Galcerán), the Bovalar Cave (Culla) and the Abric del Cingle del Barranc d l’Espigolar (Sarratella), and probably Mas de Serra Amporta (Culla), the Marfullada II and the Abric del Mas de la Vall (Ares del Maestre) (Martínez et al. 2009). Evidence of portable art comes from the Matutano Cave (Olaria 2008) and the Blaus Cave (Casabó 2005) with objects of Magdalenian date, and a small engraved plaquette with no chronological context from the Cabàs Cave (Casabó & Salvador 2004) which, due to its simplicity, adds nothing to the characterization of the portable art in the region.

There are currently 45 sites that have provided evidence of Palaeolithic cave art (Fig. 1). With regard to portable art, objects have been discovered in 15 sites. This total does not include the sites that have provided only engraved non-figurative motifs related to the bone industry (points, spears or rods), thus limiting the count to figurative portable art. On the other hand, only four of the sites include both portable and parietal art (Parpalló, Ambrosio, Nerja and Gorham). In the parietal art total, some locations have been left out due to problems with their authentication or the recognition of the motifs depicted, either through doubts raised by their chronology, or because they are under study and nothing more than the existence of Palaeolithic art is known.

In short, in this part of the Mediterranean region there is abundant artistic evidence of Magdalenian or Epimagdalenian chronology (Villaverde et al. 2012) in close relation to other evidence from occupations of this same period, linked to the same sites in the case of portable art, or to the regional environment in the case of parietal art. Also worth emphasizing, on the other hand, is the domain of the parietal supports on the walls of small caves and the importance of the engravings, in themes that alternate between animal forms and symbols. To the west, at a distance of more than 50 km, there is an engraved block at Roca Hernando (Cabra de Mora, Teruel) (Utrilla et al. 2001), of uncertain chronology and with no defined figurative motifs.

In any case, the first thing that can be observed in geographical terms from the data available is that the distribution is uneven, with certain concentrations and gaps. A quick review of the information reveals the problems that may arise from the very definition of the Iberian Mediterranean region – in particular the shortage of Palaeolithic art in the northern area and the difficulty of establishing a gap between the Mediterranean and the Atlantic regions of Andalucia, or between the latter and the Portuguese Estremadura.

Further south, after a gap in documented finds ranging from the valley of the Millars River to the Xúquer, a distance of more than 100 km, there is a new concentration of sites, linked to the central areas of the Valencia region. These are sites with parietal art, including Parpalló (Beltrán 2002) and Meravelles (Gandia, Valencia) (Villaverde et al. 2005, 2009), the Comte Cave (Pedreguer, Alicante) (Casabo et al. 2014) and Fosca Cave and Reinós Cave (La Vall d’Ebo, Alicante) (Hernández et al. 1988). Finds of portable art either coincide with this distribution – as is the case with the abundant collection of plaquettes at Parpalló (Pericot 1942; Villaverde 1994) or the more reduced batch of objects in the Malladetes Cave (Barx, Valencia) (Villaverde et al. 2007), the Tossal de la Roca (la Vall d’Alcalà, Alicante) and the Cave at the Barranc del Infern (Laguart Valley, Alicante) (Cacho & Ripoll 1987) – or are linked to archaeological sites in neighbouring regions, which is the case of the objects identified in

To the north of the Ebro Valley the only evidence of parietal art is in the Taverna Cave, where a male deer has been identified, engraved partially using the relief of the wall (Fullola & Viñas 1985), and in the (now disappeared) La Moleta Cave in Cartagena (Ripoll 1965), with a painting of a vertical aurochs; whereas evidence of portable art is limited to the finds in the Molí del Salt (García & Vaquero, 2006), Sant Gregori (Fullola et al. 1990; García-Arguelles et al. 2002-2003), Hort de la Boquera (Garcia-Argüelles et al. 2014) and Tut de Fustanyà (Carbonell & Marcet 1976), as well as a bone with engravings at Bora Gran (Pericot & Maluquer 1951). The first two yielded sets of 4 and 2 plaquettes 147

Prehistoric Art as Prehistoric Culture the Volcán del Faro Cave (Cullera, Valencia) (Aparicio 1977) and in the Beneito Cave (Agres, Alicante) (Iturbe et al. 1994). In these two cases, the portable art should be considered with caution, because it is limited to the presence of some pebbles or plaquettes with only some traces of pigment.

stratigraphic sequences in Almeria (again, Ambrosio Cave, La Zájara II and the Serrón) or the Pirulejo site (Priego de Córdoba) with some items of portable art (Cortés & Simón 2008), located around 33 km as the crow flies from Malalmuerzo and some 22 km from the Ermita del Calvario, the lack of references to human occupation during the Upper Palaeolithic is the most characteristic feature of these artistic sites. The decorative techniques are varied, with documentation of painting and engraving in Ambrosio, engraving in Piedras Blancas and painting in Malalmuerzo, Morrón and Almaceta.

Parietal art is still dominated by the engraving technique, but it is now linked to caves of greater or lesser development: in the first case the engravings in Fosca Cave, Comte Cave – where some painted motifs have also been identified — and Meravelles Cave; and in the second case the engravings in Parpalló Cave and the paintings in Reinós Cave.

The following core concentration of sites and caves containing art is located on the Málaga coast between Nerja and the Bay of Málaga, including Nerja Cave (Nerja) (Sanchidrián 1986), the Cantal Caves, Victoria Cave (Rincón de la Victoria), the Tesoro Caves, Higuerón, and Navarro Cave (Cala del Moral) (Sanchidrián 1981; Cantalejo et al. 2007), Toro or Calamoro Cave (Benalmádena) (Fortea & Giménez 1972/73), and the Ardales Cave in the Guadalteba region (Cantalejo et al. 2006); and more inland in the Serrania de Ronda, the Pileta Caves (Sanchidrián 1997) and the Gato (Benaoján) (Cantalejo et al. 2006). This is a geographical area in which evidence of the Upper Palaeolithic, apart from Nerja Cave itself, extends to other locations such as Bajondillo and Complejo del Humo.

Further inland, and in a relatively remote position compared to the area we have just discussed, a painted horse has been discovered in a small cave in the Barranco de los Morenos (Requena, Valencia) (Martínez-Valle et al. 2014), attributed to the Magdalenian style. Its presence proposes a possible line of contact with the southern sub-plateau, with a certain proximity to the sites of similar chronology documented in Cuenca (de la Rasilla et al. 1996). In the Valencia region the importance of the parietal and portable art has a clear correlation with the importance in this area of sites with archaeological evidence of the Upper Palaeolithic. And there is an important geographical gap, again surpassing 100 km, between this region and the next, in which we could group the evidence found around Cieza and the Segura Valley, and Ayna and the Sierra del Segura.

Extensively developed caves predominate here, including several sites with a number of animal forms, mainly engravings in Ardales and paintings in the other sites. Farther south and west, near the Strait, with sites situated in both the Mediterranean and the Atlantic, is the large concentration of settlements known as Southern Art, linked to Cadiz and including the renowned art in Gibraltar.

The sites found in this other region – Niño Cave (Ayna, Albacete) (Almagro 1971; Garate & García Moreno 2011), Jorge Cave, Las Cabras Cave and Arco Cave (Cieza, Murcia) (Salmerón & Lomba 1996), separated by about 50 km – are in turn surrounded by various sites from the Upper Palaeolithic, providing evidence of the importance of settlements in these areas. This is the case in Molino de Vadico and El Palomar in the Sierra del Segura (Vega 1993; Vega & Martín 2006), and in the recent finds in the Rambla Perea, Mula (Zilhão et al. 2010).

In total, from east to west, 10 sites with rock art have been recorded: Horadada Cave (San Roque), Gorham (Gibraltar), Ciervo Cave (Los Barrios), La Jara I Cave (Tarifa), Palomas I Cave (Facinas), Moro or Vencejo Moro Cave (Tarifa), El Realillo (Tarifa), Caminante Cave (Tarifa), Atlanterra Cave (Zahara de los Atunes) and Motillas Cave (Jerez de la Frontera) (Bergmann 2009; Martínez 2009; Simón et al. 2009; Ruiz et al. 2014).

Following a route from north to south, the following parietal sites are located in a distinctly mountainous, inland environment, linked to the Andalucian system, displaying a relatively dispersed distribution pattern with distances ranging from 40 to 130 km. These sites include Morrón Cave (Jimena, Jaén) (Sanchidrián 1982), Ambrosio Cave (Vélez Blanco, Almería) (Ripoll et al. 2006), Almaceta Cave (Lucar, Almería) (Martínez 1992), Piedras Blancas (Escullar, Almería) (Martínez 1986/87), Malalmuerzo Cave (Moclín, Granada) (Cantalejo 1983) and Ermita del Calvario Cave (Cabra, Córdoba) (Asquerino et al. 2005). Except for the small batch of sites with Upper Palaeolithic material associated with

Many of these locations are actually rock-shelters or small caves, including mainly sites with few figurative representations, some engraved but in most cases painted. Upper Palaeolithic deposits have been studied in Gorham, with some portable art objects (Simón et al. 2009), in La Paja Cave, the Levante Caves and certain enclaves in the Lagoon of La Janda.

148

Villaverde: Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation

35  

30   25   20   15   10   5   0  

MEDITERRANEAN  

CANTABRIAN  

Figure 2. Distribution of sites with parietal art in the Mediterranean and Cantabrian regions (González Sainz, 2004) considering the number of figurative representations (abscissa axis)

 

Horadada Cave, the Motillas and, especially, Gorham. In the first because there is insufficient and provisional published data, in the second because some of the figures are difficult to identify and classify, and in the third because some of the identified themes are doubtful, mainly due to the poor condition of the pigment. In any case, the quantifications would only change appreciably at Gorham, where we have decided to consider only 8 of the zoomorphic figures for their clear representation, one in the Gallery (Simón et al. 2009) and seven on the north interior wall, furthest from the mouth of the cave. Other researchers (Balbín et al. 2000) have identified 17 representations in this location.

At the southern tip of Portugal, around 200 km from the heart of Cadiz, is the site of Vale Boi, which yielded a plaquette with figurative animal motifs of Solutrean date (Simon et al. 2012), and a bit farther north, the site of Escoural with parietal art, where conventions and techniques have been identified showing some similarity to the sequence at Parpalló, but also with their own specific traits (García et al. 2000). The proximity between this enclave and the art at Molino Manzanes in Badajox (Collado 1999) is a feature that must not be overlooked, as well as the similarity of certain figures found in the two sites. Some caves have been left out of this list, which includes a total of 45 sites with parietal art, for various reasons. In a few cases because they have only been cited and are lacking publication or details of the artistic motifs, as is the case in Pecho Redondo (Marbella, Malaga), the Vaca or Tajo del Jorox Caves (Alozaina, Malaga), Cholones Cave (Zagrilla, Córdoba), Saint Martin (Gibraltar) and Estrellas Cave (Castellar, Cádiz); and in other cases because the motifs are difficult to identify, as in VR15 Cave (Villaluenga del Rosario, Cádiz), or the Levante Caves (Jerez de la Frontera, Cádiz), the Tajo de las Figuras and Arco Cave (Benalup, Cádiz), and Buitre II Cave (Tarifa, Cádiz) (Martínez 2009).

Five of the 45 sites have no zoomorphic depictions. Considering the number of zoomorphic representations, the distribution of the Mediterranean parietal sites presents some interesting differences compared with the Cantabrian area (González Sainz 2004). The sites characterized by very few representations clearly dominate the Mediterranean region: 92 % of the sites contain only a few zoomorphic figures (between 1 and 20), and 73.3 % have very few representations in total (between 1 and 5 motifs). These values for the Cantabrian region are 70 and 41.4 % respectively. As a result of this, the number of parietal sites with a medium number of representations (between 21 and 60) is much lower in the Mediterranean region (2.6 % which corresponds to just one set, compared with 21.4 % corresponding to 15 sites in the Cantabrian region). Finally, the parietal sites with a high number of figures (more than 61) are scarce in both areas, especially in the Mediterranean where there are

General characteristics of Mediterranean Palaeolithic art The total number of zoomorphic representations from these 45 parietal art sites is 363. Some doubts have arisen in this total about certain sites, as is the case at 149

Prehistoric Art as Prehistoric Culture Site

Cova Moleta de Roca de la Cartagena Hernando Taverna

Nº animals 1 Male deer 1 Female deer Horse Aurochs Ibex Other/ Indeter.

1

1

Abric d’en AAbric del Melià Cingle del Barranc de l’Espigolar 14 12 1 4   2 1 1 9 8

Site

Cueva Piedras Cueva Cueva del Ambrosio Blancas de El Malalmuerzo Morrón Nº animals 12 2 2 2 Male deer   Female deer   Horse 10 1 1 Aurochs 1 1 Ibex 2 Other/ 1 1 Indeter.

Site

Cueva del El Moro o del Caminate Vencejo Moro 8 1 1

Nº animals Male deer Female deer Horse 7 Aurochs Ibex Other/ Indeter.

1

Cueva de Nerja 38 1 7 9 1 6 15

Cova Cova del Cova de les de Parpalló Meravelles Bovalar 4

1     1

1 1 1 1

20 + 1 1 3 6 2+1 5 3

Cueva  Cueva Abrigo Navarro Victoria de los Morenos 1 1 1 1

Cova del Cova de Cova Cueva Arco Arco Cueva Cueva Cueva de Comte Reinós Fosca de I II de las de El Almaceta Jorge Cabras Niño 9

1

1 1

1

Cueva  Cueva del del Higuerón Tesoro 1 2 1 1

1

1

El Cueva de Horadada Realillo Atlanterra I

Cueva de las Motillas

Groham’s Cave

Ermita del Total Calvario

1

3

2?

1

8 1   2   1 4

1

1 1 1

2?

2 2

1

1

17   5 8 1 1 2

2 5

363 17 92 104 33 59 86

1     1

3

2

2 1

Cueva del Toro o Calamorro 1

Cueva de Ardales 97

Cueva de la Pileta 98

1

63 25 2 3 4

8 13 19 29 29

2

4       1 1 2

Cueva del Gato 1 1

13 3 2 2 1 3 2

1     1

Cueva de las Palomas I 1     1

  Cueva de El Ciervo 1 1

La Jara I 1 1

Table 1. Zoomorphs represented in Mediterranean and South Iberian parietal art.

On the other hand, the fact that the majority of the representations are concentrated in only a few caves introduces an undeniable regional bias in the interpretation of this aspect. We must not forget that, on establishing the totals distinguishing pre-Magdalenian and Magdalenian parietal art, after deducting the indeterminate zoomorphic figures, only three sites of presumably Magdalenian or Epimagdalenian date account for more than ten representations (Cova Fosca, Ardales and Pileta), whereas in the pre-Magdalenian period only five surpass this number (Meravelles, Ambrosio, Nerja, Ardales and Pileta). With a certain number of zoomorphs, but a high proportion of them indeterminable, Melià and Espigolar would be in the first category, and in the second, El Niño. In the latter, although pre-Magdalenian chronology has recently been proposed for the whole set (Garate & García 2010), there are sufficient elements (Fortea 1978) to make us believe that some zoomorphic representations are from the Magdalenian period, meaning that their quantification by phases is subject to interpretation. With regard to the determination of species, the identification of snakes does not seem proven, since the serpentine motifs documented there show clear parallels with the signs at Parpalló.

only two caves, both in the Málaga area, which represent 5.3 %, compared with the 6 cavities in the Cantabrian region which represent 8.6 % of the total (Fig. 2). The degree of thematic diversity of the zoomorphic figures increases with the number of representations, as is the case in the Cantabrian region. In the Mediterranean region 20 sites have just one species depicted, and significantly the female deer has not been documented in any of them. It is however possible to observe a certain predilection for the representation of deer and equines in the heart of the Campo de Gibraltar (Bergmann 2009; Ruiz et al. 2014). If we consider the information provided by the sequence of portable art at Parpalló (Table 1) and the small number of animals represented in many of the caves, the little variation in the themes of the Iberian Mediterranean and Southern Atlantic art, and the simplification and low naturalism that characterize not only the pre-Magdalenian but also much of the Magdalenian art, it is difficult to count and characterize the themes. In this regard, we can mention the significant problems that arise due to the lack of the head, which is normally essential for the identification of the species. This explains the high number of unidentified animals that appear in the totals (21.95 % of the inventoried animals), and this percentage is even higher when focusing on the Magdalenian cave art (31.49 %). The same is true of the portable art (39.96 %), where the data are defined by the totals at Parpalló, but in the other known sites the percentage of undefined animals is 25.71 %.

With the exception of Parpalló, the other sites with portable art – mostly of Magdalenian or Epimagdalenian date – are characterized by the small size of the collections and, thus, the reduced number of identifiable zoomorphs. The site with the most is Molí del Salt with seven identified animals, while the other sites have between one and five specimens (Table 2).

150

Villaverde: Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation Site

Cova del ParpallóPM Animals 386 Male deer 6 Female deer 49 Horse 75 Aurochs 22 Ibex 74 Other/indet. 160

Cova del Mallaetes- G Hort Parpalló-M de la Boquera 202 1 1 16 16 36 19 1 35 80 3

Filador -M Sant Molí Matutano Tossal Gregori del de la Salt Roca 4 1 12 14 6 2 1 1

1

4 2 1 5

1 3

1 2

1 9

1 1

Barranc Cendres Ambrosio Nerja Pirulejo del Infern 1 2 1 2 5 1 2 1 1 1 1

2 2

Gorham Total 1 1?

639 23 76 122 45 113 261

Table 2. Zoomorphs represented in Mediterranean and South Iberian portable art.

With these data, the only variation that we observed in the parietal art is the prominence of male deer in the Magdalenian period, in clear contrast with the greater importance of female deer in the pre-Magdalenian. Considering the data collected from Parpalló, it could even be said that the number of female deer decreases throughout the Magdalenian sequence. Tthis trait seems to be limited to the strict Upper Magdalenian, because the pieces of portable art from the end of the Palaeolithic artistic cycle, relatable to the Final Magdalenian and Epimagdalenian, have a considerable presence of female deer. This is, for example, the species with most depictions in both Filador and Molí del Salt.

example of the richness and complexity of the signs in Andalucian parietal art. Considering only this trait, the art from the Málaga region can be distinguished from the rest of the Andalucian region, because, although the markings are present in almost all of the sites with paintings and engravings, the quantifications are a long way off the high percentages (close to or exceeding 90% of representations) that are reached in caves such as Navarro, Nerja and Toro, or the Malalmuerzo series in Granada (Medina & Sanchidrián 2014). The link between the paintings and the apparent correlation with pre-Magdalenian horizons clarify the importance of ideomorphs in certain southern territories. This circumstance leads to a differentiation with the preMagdalenian parietal sites in the Murcia area, or in the Valencia region.

Another typical thematic trait of the Magdalenian is the presence of a greater number of naturalistically represented birds. There is also greater thematic diversity in the herbivores considering the data collected from Parpalló, where wild boar, chamois, a mustelid, some canids and two different species of bird are documented.

Considering the techniques used, engraving was predominant in Catalonia and the Valencia region, where the only paintings are a bovid at the Moleta in Cartagena, a female deer at Bovalar and an ibex at Reinós. In contrast, paintings are hegemonic in the rest or the Mediterranean region, although there are some sites with the presence of only engravings or a combination of both paintings and engravings: Ambrosio, Piedras Blancas, Ardales, Moro Cave, Horadada Cave and the Motillas.

The limited variation in the species documented and the absence of cold species are well known features of Mediterranean art. As for the above-mentioned lack of thematic variation, suffice it to note that most of the identifiable zoomorphs can be included in the four basic thematic groups of deer, equines, ibex and aurochs. In total, 298 representations fall into these groups, of which 171 correspond to pre-Magdalenian art and 127 to Magdalenian art, with the latter representations concentrated mainly in the parietal art from Melià, Espigolar, Fosca, Pileta and Ardales.

Although paint offers a variety of colours, it is lineal and only some partial flat colours are documented in ‘Block 2’ of Pileta (Sanchidrián 1997). With regard to the positions of the decorated panels, the large number of sites in which the motifs are directly illuminated by sunlight is significant, in rock-shelters or cavities that could be classed as small caves. Besides Piedras Blancas, where there are engravings on blocks in the open air that could be from the pre-Magdalenian period, this characteristic includes all of the parietal sites from the end of the Palaeolithic artistic cycle known to date in the Maestrazgo in Castellón (Melià, Bovalar, Espigolar, etc), or the small cave of Morenos, attributed to the Magdalenian period. This is also the case in some sites clearly of pre-Magdalenian date from the rest of the Mediterranean region, such as Parpalló, Jorge and Ambrosio, as well as many of the sites from the southern core, although in this case regardless of their chronology. So this choice of shallow spaces, sometimes linked to places of occupation, can be considered a trait that

A detailed study of the signs and ideomorphs in parietal art is very difficult to carry out, mainly due to the lack of common classification criteria in several publications and the limited detail in the studies published about some caves. While there is an abundance of reticulated marks, zigzags and bundles of lines in the art from the Castellón region, in the painted pre-Magdalenian art the series of markings and strokes are especially significant, documented in many of the Málaga sites, with excellent examples in caves such as Pileta, Nerja and Navarro, to name the most representative. The thematic association of the markings in the Andalucian parietal sites has been studied closely by Sanchidrián (1994), and the careful study of the compositions in Pileta’s E horizon (Medina & Sanchidrián, 2014) has revealed a clear 151

Prehistoric Art as Prehistoric Culture seems quite common in Mediterranean Palaeolithic art, especially north and south of the regional distribution, characterising the heart of Málaga, Jaen and Almeria for the existence of decoration located in deep areas of fairly well-developed caves.

this territory, especially since the Solutrean. The regional breadth and variation of the Solutrean industries in France and the Iberian Peninsula constitute the strongest evidence that contacts were produced throughout this area, but the regional traditions formed the population frame that eased the flow of ideas and contacts.

Evaluation of the stylistic similarities and differences in the main graphic horizons

A more careful assessment of the similarities between, for example, the heart of Málaga and the central Valencia region, is hampered by the differences in techniques (engraving versus painting), the differences in the support (portable or parietal) and the validity and accuracy of the chronological criteria used when ordering the parietal art, as ideally the oldest phases of pre-Magdalenian art should be differentially compared with the most recent, as they seem to appear in the Parpalló sequence (Villaverde et al. 2005) – a differentiation not easy to establish given the information available today.

As is common in other parts of south-west Europe (Petrognani 2013), pre-Magdalenian art presents numerous unifying traits throughout this area, especially if we consider criteria such as perspective, proportions, attention to the outline, and lack of peripheral and internal details. The presence of horses with ‘duck-bill’ muzzles, the scant attention to the hooves, the frequent use of bi-angular perspectives, the disproportion and the lack of internal detail, constitute a common stylistic ‘background’ that can only be understood through the existence of extensive exchange networks and high mobility linking large areas of south-west Europe. In the Cantabrian region these similarities also extend to the thematic field, with the importance of female deer depictions and the coincidences that can be observed in many of the markings associated with zoomorphic figures.

It is, however, possible to mark the breaking of these contacts over broad territorial dimensions, external to the Mediterranean and the South Atlantic coastal areas, after the Magdalenian. From these dates, the data collected at Parpalló show little differentiation with respect to pre-Magdalenian art and, similarly, a reduced or often non-existent influence from the Cantabrian or Pyrenean regions. It is true that some technical and stylistic traits, and above all the themes of the signs, make it possible to differentiate the art from those levels at Parpalló from the earlier ones. But the differences in the art in these regions is very noticeable. There are few graphic elements that match the figurative analytical concept, and only some representations show any detail of peripheral or internal elements and perspectives common in the Magdalenian art in these areas.

These similarities span much of the pre-Magdalenian art, and also reach the artistic production at the end of the Solutrean period in a very particular way. Again, the indicative elements of contacts and relationships go beyond the Mediterranean and southern Iberian setting to include south-east France, where the modes of execution of some of the zoomorphic figures and certain signs at Cosquer Cave provide clear confirmation, and the Cantabrian region (Villaverde et al. 2005a). These stylistic coincidences have also been identified in some deposits in the central system (Hernando 2014), and particularly in relation to the parietal representations at La Griega (Corchón et al. 2012), as well as at the open-air sites in the Douro Valley (Baptista 1999; Ripoll & Municio 1999) and the Guadiana (Collado 1998), especially from the presence of striated engravings as a technique for filling in the body.

It is interesting to observe that the same happens in the rest of the geographical area that we are analysing. Those traits, brought about by Pyrenean and Aquitaine influences, are not observed either in Murcia or Andalucia, and much less in the Portuguese Atlantic region. This suggests the continuity of the contacts, relationships and influences determined in the previous phase. There are two issues that, in my opinion, are especially important when addressing greater territorial detail during the Magdalenian: the concentration of sites in shallow caves, almost shelters, with few representations in the Campo de Gibraltar; and the concentration of engravings in rock-shelters in the Castellón area. The latter have stylistic traits in the portable art that can be related to the Epimagdalenian. In the former, naturalism is associated with only a few details and the theme registers an insignificance of female deer depictions (with exceptions at Ardales and Fosca). At Fosca, the figures tend towards stylization and little naturalism,

However, assessing the extent of these similarity traits in terms of cultural unity is much more difficult, precisely because these traits are spread over a territory that is so broad, and they present sufficient internal variation elements to make it very difficult to link the similarity to a common cultural tradition, which is the result of ongoing relationships and contacts supported by other common material culture traits. It seems more that the interpretation of these similarities is to be made in relation to the broad interrelations that were produced discontinuously, although repeatedly, in 152

Villaverde: Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation Cantalejo, P., Maura, R. & Becerra, M. 2006. Arte rupestre prehistórico en la Serranía de Ronda. Valles del Guadiaro, Turón y Guadalteba. Edit. La Serranía: Ronda. Cantalejo, P., Maura, R., Espejo, M. M., Ramos, J. F., Medianero, J., Aranda, A. & Durán, J. J. 2006. La Cueva de Ardales: Arte prehistórico y ocupación en el Paleolítico Superior. Diputación Provincial: Málaga. Carbonell, E. & Marcet, R. 1976. El Tut de Fustanyá. III Assemblea d’estudis sobre el Comtat de Besalá pp. 73-86. Casabó, J. 2005. Paleolítico superior final y Epipaleolítico en la Comunidad valenciana. Marq, Serie Mayor, 3. Casabo, J., de Dios, J., Costa, P., Esquembre, M. A.& Bolufer, J. 2014. Cova del Comte (PedreguerAlicante), nuevo yacimiento con arte parietal paleolítico en el litoral mediterráneo, pp. 285-99 in (M. S. Corchón & M. Menéndez, eds) Cien años de arte rupestre paleolítico. Centenario del Descubrimiento de la Cueva de la Peña de Candamo (1914-2014). Ediciones Universidad de Salamanca: Salamanca. Casabó, J. & Salvador, L. 2004. Hallazgo de una plaqueta paleolítica con decoración incisa en la Cova d’En Cabàs (Torreblanca, Castelló). Saguntum-PLAV, 36: 149-52. Collado, H. 1998. Arte rupestre en la Cuenca del Guadian. El conjunto de grabados del Molino Manzánez (Alconchel-Cheles). Estudos Arqueológicos do Alqueva. Corchón, M. S., Hernando, C., Rivero, O., Garate, D. & Ortega, P. 2012. La cueva de la Griega (Pedraza, Segovia, España) en la encrucijada ibérica: nuevos análisis del arte parietal paleolítico a través del análisis factorial de correspondencias. Espacio, Tiempo y Forma 5: 527-42. Cortés, M. 2010. El Paleolítico superior en el sur de la Península Ibérica. Un punto de partida a comienzos del siglo XXI, pp. 173-97 in (X. Mangado, ed.) El Paleolítico superior peninsular. Novedades del siglo XXI. Monografíes del SERP 8: Barcelona. Cortés, M. & Simón, M. D. 2008. Manifestaciones simbólicas, pp. 185-92 in El Pirulejo (Priego de Córdoba): Cazadores recolectores del Paleolítico superior en la sierra Bubbética. Antiqvitas 20. Fortea, F. J. 1978. Arte paleolítico del Mediterráneo español. Trabajos de Prehistoria 35: 99-149. Fortea, F. J. & Giménez, M. 1972/73. La cueva del Toro. Nueva estación malagueña con Arte paleolítico. Zephyrus 23-24: 5-17. Fullola, J. M., Villaverde, V., Sanchidrián, J. L., Aura, J. E., Fortea, J. & Soler, N. 2005. El Paleolítico superior mediterráneo ibérico, pp. 192-211 in La Cuenca Mediterránea durante el Paleolítico superior. IV Simposio de Prehistoria Cueva de Nerja.

with conventions that find their parallel in the contexts of portable art from the end of the Upper Magdalenian. The variety of parietal supports is a trait that seems common throughout the entire geographical setting and throughout the whole sequence, as rock-shelters, small caves and deep caves are documented in all periods. Acknowledgements Research carried out for this paper has benefited from the following contributions: HAR 200804273-E/HIST (MCI), HAR2011-24878 (MCI) and PROMETEOII/2013/016 (GV). The text was translated into English by Véronique Petit and Beth Duffield.

References Almagro Gorbea, M. 1971. La cueva del Niño (Albacete) y la cueva de la Griega (Segovia). Trabajos de Prehistoria 28: 9-47. Aparicio, J. 1977. La cueva del Volcán del Faro (Cullera, Valencia). Diputación de Valencia. Asquerino, M. D., Sanchidrián, J. L. & Márquez, A. M. 2005. Presentación del arte rupestre de la cueva de la Ermita de El Calvario (Cabra, Córdoba, España), pp. 248-55 in (J. L. Sanchidrián, A. M. Márquez & J. M. Fullola, eds) La Cuenca Mediterránea durante el Paleolítico superior: 38.000-10.000. IV Simposio de Prehistoria Cueva de Nerja. Balbín, R. de, Bueno, P., Alcolea, J. J., Barroso, R., Aldecoa, A., Giles Pacheco, F., Finlayson, C. & Santiago, A. 2001. The engravings and Palaeolithic paintings from Gorham’s Cave, pp. 179-96 in (C. Finlayson & D. Fa, eds) Gibraltar during the Quaternary. Gibraltar Government Heritage Publications Monographs 1: Gibraltar. Baptista, A. M. 1999. No tempo sem tempo. A arte dos caçadores paleolíticos do Vale do Côa. Centro Nacional de Arte Rupestre: Vila Nova de Foz Côa.. Beltrán, A. 2002. Art rupestre dans la Grotte du Parpalló (Gandía, Valence, Espagne). International Newsletter on Rock Art 33: 7-11. Bergmann, L. 2009. El Arte rupestre paleolítico del extremo sur de la Península Ibérica y la problemática de su conservación. Almoraima 39: 45-65. Cacho, C. & Ripoll López, S. 1987. Nuevas piezas de arte mueble en el Mediterráneo Español. Trabajos de Prehistoria 44: 35-62. Cantalejo, P. 1983. La cueva del Malalmuerzo (Moclín, Granada): Nueva estación con arte rupestre paleolítico en el área mediterránea. Antropología y Paleoecología Humana 3: 59-99. Cantalejo, P., Maura, R., Aranda, A. & Espejo, M. M. 2007. Prehistoria en las cuevas del Cantal. Málaga.

153

Prehistoric Art as Prehistoric Culture Leroi-Gourhan, A. 1965. Préhistoire de l’Art Occidental. Mazenod: Paris. Martínez, J. 1986/87. Un grabado paleolítico al aire libre en Piedras Blancas (Escullar, Almería). Ars Praehistorica V-VI: 49-58. Martínez, J. 1992. El arte paleolítico en Almería: Los primeros documentos. Revista de Arqueología 130: 24-33. Martínez, J. 2009. Arte paleolítico al aire libre en el sur de la Península Ibérica: Andalucía, pp. 237-58 in (R.de Balbín, ed.) Arte prehistórico al aire libre en el Sur de Europa. Junta de Castilla y León: Salamanca. Martínez-Valle, R., Guillem, P. M. & Villaverde, V. 2009. Figuras grabadas de estilo paleolítico en la provincia de Castellón, pp. 225-36 in (R. de Balbín, ed.) Arte prehistórico al aire libre en el Sur de Europa. Junta de Castilla y León: Salamanca. Martínez-Valle, R., Villaverde, V., Guillem, P. M., Lerma, J. L., Roldán, C. & Murcia, S. 2014. El abrigo de los Morenos (Requena, Valencia) y su valoración en el context del arte rupestre paleolítico del Mediterráneo ibérico, pp. 195-208 in (M. S. Corchón & M. Menéndez, eds) Cien años de arte rupestre paleolítico. Centenario del Descubrimiento de la Cueva de la Peña de Candamo (1914-2014). Ediciones Universidad de Salamanca: Salamanca. Medina, M. A. & Sanchidrián, J. L. 2014. Los signos integrados de Pileta-E: Análisis a diferentes profundidades de campo, pp. 116-29 in (M. A. Medina, A. J. Romero, R. Ruiz & J. L. Sanchidrián, eds) Sobre Rocas y Huesos: las Sociedades Prehistóricas y sus Manifestaciones Plásticas. Universidad de Córdoba: Córdoba. Pericot, L. 1942. La Cueva del Parpalló (Gandía). Consejo Superior de Investigaciones Científicas. Instituto Diego Velázquez: Madrid. Pericot, L. & Maluquer de Motes, J. 1951. La colección de Bosóms. Monografías del Instituto de Estudios Pirenaicos: 5-76. Petrognani, S. 2013. De Chauvet à Lascaux. L’art des cavernes reflect de sociétés préhistoriques en mutation. Editions Errance: Paris. Rasilla, M. de la, Hoyos, M. & Cañaveras, J. C. 1996. El abrigo de Verdelpino (Cuenca). Revisión de su evolución sedimentaria y arqueológica. Complutum Extra 6 (1): 75-82. Ripoll, E. 1965. Una pintura de tipo paleolítico en la sierra de Montsiá (Tarragona) y su posible relación con los orígenes del arte levantino, pp. 297-302 in (E. Ripoll, ed.) Miscelánea en homenaje al Abate H. Breuil, t. II. Diputación Provincial de Barcelona: Barcelona. Ripoll, S. & Municio, J. (eds) 1999. Domingo García. Arte rupestre paleolítico al aire libre en la meseta castellana. Arqueología en Castilla y León, 8. Ripoll, S., Muñoz, F. J. & Fernández, J. L. 2006. Nuevos datos para el arte rupestre paleolítico de la Cueva de Ambrosio (Vélez-Blanco, Almería), pp. 573-88 in

Fullola, J. M. & Viñas, R. 1985. El primer grabado parietal naturalista en cueva de Cataluña: la Cova de la Taverna (Margalef de Montsant, Priorat, Tarragona). Caesaraugusta 61-62: 67-78. Fullola, J. M., Viñas, R. & García-Argüelles, P. 1990. La nouvelle plaquette gravée de Sant Gregori (Catalogne, Espagne), pp. 279-86 in L’Art des Objets au Paléolithique. Colloque international d’art mobilier paléolithique, vol. 1. Ministère de la Culture: Paris. Garate, D. & García Moreno, A. 2011. Revisión crítica y contextualización espacio-temporal del arte parietal paleolítico de la cueva de El Niño (Ayna, Albacete), Zephyrus 68: 15-39. García, M., Baptista, A. M., Almeida, M., Barbosa, F. & Félix, J. 2000. Observaciones en torno a las grafías de estilo paleolítico de la Gruta de Escoural y su conservación (Santigao de Escoural, Montemor-oNovo, Évora). Revista Portuguesa de Arqueologia 3: 5-14. García, M. & Vaquero, M. 2006. La variabilité graphique du Molí del Salt (Vimbodí, Catalogne, Espagne) et l’art mobilier de la fin du Paléolithique supérieur à l’est de la Péninsule Ibérique. L’Anthropologie 110: 453-81. García-Argüelles, M. P., Fontanals, M. & Zaragoza, J. 2002-2003. Dues noves peces gravades del jaciment de Sant gregori (Falset, Tarragona): La col.lecció Ramón Rodón del Museu Municipal d’Alcover (Alt Camp). Pyrenae 33-34: 165-74. González Sainz, C. 2004. Arte parietal en la region cantábrica: centros y peculiaridades regionals. Kobie 8: 403-24. Graziosi, P. 1956. L´arte dell´antica età della pietra. Florence. Graziosi, P. 1964. L’art paléolithique de la «province artistique mediterranéenne» et ses influences dans les temps postpaléolithiques, pp. 35-46 in (L. Pericot & E. Ripoll, eds) Prehistoric Art of Western Mediterranean and the Sahara. Viking Fund Publications in Anthropology No. 39. Wenner-Gren Foundation: New York. Hernández, M. S., Ferrer, P. & Catalá, E. 1988. Arte rupestre en Alicante. Fundación Banco Exterior de España. Hernando, C. 2014. Cruzar la frontera: contactos e interrelaciones de las sociedades paleolíticas en la Península Ibérica a través de su arte (ca. 33000-16000 BP), pp. 40-55 in (M. A. Medina, A. J. Romero, R. Ruiz & J. L. Sanchidrián, eds) Sobre Rocas y Huesos: las Sociedades Prehistóricas y sus Manifestaciones Plásticas. Universidad de Córdoba: Córdoba. Iturbe, G., Fumanal, M. P., Carrión, J. S., Cortell, E., Martínez, R., Guillem, P. M., Garralda, M. D. & Vandermeersch, B. 1993. Cova Beneito (Muro, Alicante): una perspectiva interdisciplinar. Recerques del Museu d’Alcoi II: 23-88.

154

Villaverde: Palaeolithic art in the Iberian Mediterranean region. Characteristics and territorial variation (J. Martínez & M. S. Hernández, eds) Arte Rupestre Esquemático en la Península Ibérica. Ruiz, A., Gomar, A. M. & Lazarich, M 2015. Síntesis de las manifestaciones gráficas paleolíticas en cavidades poco profundas del Campo de Gibraltar (Cádiz), pp. 152-69 in (M. A. Medina, A. J. Romero, R. Ruiz & J. L. Sanchidrián, eds) Sobre Rocas y Huesos: las Sociedades Prehistóricas y sus Manifestaciones Plásticas. Universidad de Córdoba: Córdoba. Salmerón, J. & Lomba, J. 1996. El Arte Rupestre Paleolítico, pp. 71-89 in (J. Lomba, M. Martínez, R. Montés & J. Salmerón, eds) Historia de Cieza, Vol. I. Cieza Prehistórica. De la depredación al mundo urbano. Ayuntamiento de Cieza, Murcia. Sanchidrián, J. L. 1981. Cueva Navarro (Cala del Moral, Málaga). Corpus Artis Rupestris, I. Paleolithica, Vol. I: Salamanca. Sanchidrián, J. L. 1982. La cueva del Morrón (Jimena, Jaén). Zephyrus XXXIV-XXXV: 5-12. Sanchidrián, J. L. 1986. Arte prehistórico de la Cueva de Nerja, pp. 283-330 in Trabajos sobre la cueva de Nerja, 1. La Prehistoria de la Cueva de Nerja (Málaga). Sanchidrián, J. L. 1994. Arte paleolítico de la zona meridional de la Península Ibérica. Complutum 5. Arte paleolítico (T. Chapa & M. Menéndez, eds), pp. 163-95. Sanchidrián, J. L. 1997. Propuesta de la secuencia figurativa en la Cueva de la Pileta, pp. 411-33 in (J. M. Fullola & N. Soler, eds) El món mediterraní després del Plenigalcial (18.000-12.000 BP). Museu d’Arqueologia de Catalunya-Girona, Serie Monográfica, 17. Simón, M. D., Cortés, M. & Bicho, N. 2012. Primeras evidencias de arte mueble paleolítico en el sur de Portugal. Trabajos de Prehistoria 69 (1): 7-20. Simon, M. D., Cortes, M., Finlayson, C., Giles-Pacheco, F. & Rodríguez, J. 2009. Arte paleolítico en Gorham’s Cave (Gibraltar). Saguntum-PLAV 41: 9-22. Utrilla, P., Villaverde, V. & Martínez, R. 2001. Les gravures rupestres de Roca Hernando (Cabra de Mora, Teruel), pp. 161-74 in (J. Zilhão, T. Aubry & A. F. Carvalho, eds) Les premiers hommes modernes de la Péninsule Ibérique. Trabalhos de Arqueologia 17. Vega, L. G. 1993. Excavaciones en el abrigo del Molino del Vadico (Yeste, albacete). El final del Paleolítico y los inicios del Neolítico en la Sierra Alta del Segura, pp. 17-32 in Jornadas de Arqueología albacetense en la Universidad autónoma de Madrid. Vega, L. G. & Martín, P. 2006. Análisis preliminar de las cadenas operativas en el material lítico procedente del nivel IV del Abrigo del Palomar (Yeste, Albacete), pp. 396-405 in (J. M. Maillo & E. Baquedano, ed) Miscelánea en homenaje a Victoria Cabrera. Zona Arqueológica 7. Museo Arqueológico Regional: Alcalá de Henares.

Villaverde, V. 1994. Arte paleolítico de la Cova del Parpalló. Estudio de la colección de plaquetas y cantos con grabados y pinturas. Diputación Provincial de Valencia: Valencia. Villaverde, V. 2004. Arte mueble paleolítico en el Mediterráneo occidental.: contexto y diversidad regional, pp. 67-84 in (P. Arias & R. Ontañon, eds) La Materia del Lenguaje Prehistórico. El arte mueble paleolítico de Cantabría en su contexto. Santander. Villaverde, V. 2005a. Arte Paleolítico mediterráneo: de la Cueva de la Pileta a la Cova de les Meravelles, pp. 17-44 in (M. S. Hernández & J. A. Soler, eds) Arte Rupestre en la España Mediterránea. Alicante. Villaverde, V. 2005b. Art paléolithique de la Méditerranée espagnole: arguments contre son rattachement à une “Province artistique méditerranéenne”, pp. 16376 in (D. Vialou, J. Renault-Miskovsky & M. H. Patou-Mathis, eds) Comportements des hommes du Paléolithique moyen et superieur en Europe: territoires et milieux. ERAUL 111: Liège. Villaverde, V., Cardona, J. & Martínez Valle, R. 2005. Noticia de los grabados paleolíticos de la Cova de les Meravelles (Gandia, Valencia): la importancia del arte solutrense en la región mediterránea ibérica, pp. 214-25 in (J. L. Sanchidrián, A. M. Marquéz & J. M. Fullola, eds) La Cuenca mediterránea durante el Paleolítico superior: 28.000-10.000 años. Fundación de la Cueva de Nerja. Villaverde, V., Cardona, J. & Martínez Valle, R. 2009. L’art pariétal de la grotte Les Meravelles. Vers une caractérisation de l’art paléolithique pré-magdalénien du versant méditerranéen de la Péninsule Ibérique. L’Anthropologie 113: 762-93. Villaverde, V. Martínez-Valle, R., Roman, D. PérezRipoll, M. & Iborra, P. 2007. El Gravetiense de la vertiente mediterránea ibérica: reflexiones a partir de la secuencia de la Cova de les Cendres (Moraira, Alicante). Veleia 24-25: 445-68. Villaverde, V., Roman, D., Pérez Ripoll, M., Bergadà, M. & Real, C. 2012. The End of the Upper Palaeolithic in the Mediterranean Basin of the Iberian Peninsula. Quaternary International 272-273: 17-32. Zilhão, J., Angelucci, D., Badal, E., Lucena, A., Martín, S. Martínez, S., Villaverde, V. & Zapata, J. 2010. Dos abrigos del Paleolítico superior en Rambla Perea (Mula, Murcia), pp. 137-48 in (X. Mangado, ed.) El Paleolítico superior peninsular. Novedades del siglo XXI. Monografíes del SERP 8.

155

156

Small seeds for big debates: Past and present contributions to Palaeoart studies from North-eastern Iberia José María Fullola,1 Ines Domingo,2 Didac Román,1 María Pilar García-Argüelles,1 Marcos García-Díez3 and Jorge Nadal1 Universitat de Barcelona / SERP ICREA at Universitat de Barcelona / SERP 3 Universidad del País Vasco 1

2

Abstract This paper offers an updated summary and a state of the art of Palaeolithic and Epipalaeolithic rock and portable art produced on surfaces with non-utilitarian use in Northeastern Iberia. This review will also include a few historiographical references to Levantine rock art, as it was initially ascribed to the Palaeolithic. As is well known, this region breaks into debates on Prehistoric art early in the 20th century with the discovery of la Roca dels Moros de Cogul (Lleida) in 1908, which was then considered Palaeolithic. Ever since it has maintained a small but continuous presence in the history of research on prehistoric art. This paper summarizes the contributions to this facet of prehistoric research in this geographic area, to identify when and to which prehistoric traditions the finds in the region are ascribed. We will especially focus on the key role of some Catalan portable art objects with figurative motifs dating to the Late Upper Palaeolithic and the Epimagdalenian to support some relative dates for the latest finds of very fine rock art engravings located in open-air rock shelters in Castellón and southern Tarragona. We will also reflect on their implications in discussions on the continuity or rupture between Final Palaeolithic traditions and Levantine rock art. This is where these territories have provided most of the known sites in Mediterranean Iberia so far. Keywords rock art; portable art; Late Upper Magdalenian; Epipalaeolithic

rock art. Vidal’s discussions on the chronology and the alleged presence of a bison were later reiterated by other researchers (Duran 1923; Hernández-Pacheco 1924; Almagro 1952, to name a few). Nevertheless Breuil and Obermaier (Obermaier 1916; Ripoll 1970) continued supporting the Palaeolithic chronology of Levantine rock art at least until the 1950s, suggesting that the differences between Cantabrian and Levantine art were due to the African or Capsian connections of the cultures in Eastern Iberia. Today the Palaeolithic chronology of Levantine rock art has been completely discarded to focus the debates on the potential attribution of this tradition to either the Epipalaeolithic (Alonso & Grimal 1999; Mateo 2008; Olaria 2008; Viñas et al. 2011) or the Neolithic (Martí & Hernandez, 1988; Martí & Juan Cabanilles 2002; Molina et al. 2003 to name a few) (for a more complete discussion on this topic see for example Villaverde et al. 2012).

Introduction Northeastern Iberia breaks into Palaeolithic rock art research soon after the discovery of Altamira and, more specifically, once the sceptic Emile Cartailhac (1902) and the scientific community accepted that this milestone of prehistory was a masterpiece from the Palaeolithic (Fig. 1). There is an earlier reference to rock art engravings and drawings in the region in 1830, which described motifs, today included in the Schematic tradition, at Portell de les Lletres site (Rojals, Montblanc) ( Viñas et al. 2011; Hernández & Hernández, 2013). However, the first sample of rock art directly ascribed to the Palaeolithic dates to 1908, with the discovery of la Roca dels Moros de Cogul (Lleida). At the time, Breuil described this first sample of Levantine rock art in this region and the second in the Iberian Peninsula as a regional variant of Franco-Cantabrian Palaeolithic rock art (Breuil 1908). He based the ascription of the site on the identification of a Quaternary species (a bison) in the panel at Cogul (Breuil 1908: 11). Interestingly, barely a year after the discovery J. Mª Vidal from the Institut d’Estudis Catalans not only discarded the presence of a bison (Vidal 1909: 548), but also suggested an alternative chronology for the site: the Neolithic. His paper marks the beginning of the current debates on the chronology of Levantine

Thus, once Levantine rock art is discarded as a Palaeolithic tradition, we can state that the first Catalan find ascribed to the Epipalaeolithic took place in the 1930s with the discovery of a slab with an engraved hind at the site of Sant Gregori (Falset, Tarragona) (Vilaseca 1934), while we have to wait until the 1960s for a full Palaeolithic find: a bull depicted in black in an upright ascending position identified at la Cova del Tendo (Moleta de 157

Prehistoric Art as Prehistoric Culture

Figure 1. Timeline of the discoveries in Northeastern Iberia (none of the figures is to scale).

158

Fullola et al: Small seeds for big debates Cartagena, Sant Carles de la Rápita, Tarragona) in 1964 (Ripoll 1965). This find was also the first Palaeolithic rock art site discovered on the eastern side of the Iberian Peninsula, but unfortunately the site was later destroyed, preventing any further research (Ripoll 1965; Grimal & Alonso 1989).

references, so at the moment it can only be described as a slab with some engraved lines ascribable to an unspecified period of the Upper Palaeolithic (García et al. 2003: 167). Nevertheless, some researchers still refer to it as a figurative depiction (Mateo 2012: 3). Similarly, the slab from Cova del Parco (Alós de Balaguer, Lleida) published by Maluquer (1983: 224-26) as a possible carving of an animal has also been rejected as such (García-Díez et al. 2003: 167) (Fig. 1).

Nevertheless the history of Palaeolithic discoveries in the region may have to be rewritten in the next few years with a complete direct reanalysis of the engravings at Cova dels Moros de Cogul, which Almagro described as the earliest representations at the site (Almagro 1952: 39). These engravings have been overlooked for a long time. However, recent discoveries of Levantine engravings at Barranco Hondo (Castellote, Teruel) (Utrilla & Villaverde 2004) and presumably at Llavería P-IV (Capçanes) (Viñas & Sarriá 2011; Viñas 2012) as well as potential Epimagdalenian engravings at different sites in Castellón (Abric d’en Melià, Cingle del Barranc de l’Espigolar, la Belladona, Mas de la Vall, la Marfullada II, Mas de Serra Amporta and Cova del Bovalar) (Guillem et al. 2001; Martínez et al. 2009; Guillem & Martínez 2009) urge a reassessment of this early find, as already suggested by Hernández and Hernández (2013: 14).

Despite the limited quantity of evidence, it is interesting to note that Catalonia has provided the only examples of portable art known so far in Eastern Iberia for the Epimagdalenian (Picamoixons, Sant Gregori and Molí del Salt), the Sauveterrian (Filador) and the Notches and Denticulates Mesolithic (Balma Guilanyà) traditions, and two of the very few examples in the Late Upper Magdalenian (Molí del Salt and l’Hort de la Boquera), with only two other examples in Castellón (Matutano) and Valencia (Parpalló). In the following section we will describe the finds attributed to each of these periods, to identify and summarize the main features of these artistic traditions, and their contributions to our understanding of the different Palaeolithic and Epipalaeolithic traditions in the region.

Today, the assemblage of Palaeolithic and Epipalaeolithic samples of art accepted as such is still exiguous, but includes some of the few examples of prehistoric art for some of the final phases of the Palaeolithic and the transition to the Neolithic on the eastern coast of Spain.

2. Prehistoric art sequence in northeastern Iberia The Palaeolithic and Epipalaeolithic art of Northeastern Iberia includes two types of media (rock art and portable art) and a very small number of techniques (mainly fine engravings and a few paintings in black and red). These characteristics are in line with the few examples located on the Eastern side of Iberia (except for the substantial collection from Parpalló Cave, Gandía, Valencia) (Pericot 1942; Villaverde 1994), when compared to the Franco-Cantabrian and the Southern Iberian sites.

The body of rock art is limited to two sites, each of them containing a very reduced number of figurative motifs and techniques (Table 1): Cova del Tendo (Sant Carles de la Ràpita, Tarragona) and Cova de la Taverna (Margalef de Montsant, Tarragona). The number of examples of portable art is slightly bigger, although equally concentrated in a small number of sites: 4 schist slabs with engraved figurative and geometric motifs at Sant Gregori (Vilaseca 1934; García-Argüelles et al. 2002), 5 slate plates (3 of them fragments of the same piece) with engraved geometric motifs and 1 pebble with red lines at Filador (Vilaseca 1949, 1968; Fullola & Couraud 1984; Fullola et al. 1986; Garcia-Argüelles et al. 2005), 1 slab with red lines at Picamoixons (GarcíaDíez et al. 1997), several slabs with figurative and nonfigurative motifs at Molí del Salt (with only 4 of them published so far) (García-Díez 2004; García-Díez & Vaquero 2006), 1 slab with non-figurative engravings at Balma Guilanyà (Martínez et al. 2011) and 1 calcareous slab with figurative engravings at Hort de la Boquera (Garcia-Argüelles et al. 2014).

2.1. Evidence of rock art Two examples of rock art have been identified in the Province of Tarragona, and repeat the most classic patterns of location of Palaeolithic art, i.e. inside caves and away from direct sunlight. The number of figures is certainly reduced, and so is the number of decorative techniques documented (one engraved motif and a couple of figures in black). At Cova del Tendo only one black aurochs and a controversial representation of a Levantine human with a bow or alternatively a representation of the floor, were identified before the site was destroyed (Ripoll 1965, 1970). Considering the main stylistic traits of the aurochs, a simple outline with no internal details, legs with no indication of the hooves and so forth, it was ascribed to an early stage within the Upper Palaeolithic, and more specifically to a time that would precede the Upper Solutrean, considering the collection of portable

The supposed figurative representation of a goat and several red lines identified on a slab at Tut de Fustanyà (Queralbs, Girona) (Carbonell & Marcet 1976) has already been discarded as such (Villaverde 1994: 140) (Fig. 1). Furthermore it lacks reliable stratigraphic 159

Prehistoric Art as Prehistoric Culture

Figure 2. Examples of portable art from Molí del Salt (drawings by García, 2004).

art from Parpalló and an AMS date obtained for a bull from Pileta Cave (Sanchidrián 2001). The unfortunate destruction of the site and the problematic and scarce documentation of the paintings preserved today (García et al. 2002) prevent any further progress in refining the chrono-cultural ascription of the site.

In the summer of 1983 an engraved red deer was located at Cova de la Taverna (Fullola & Viñas 1985; Fullola 1987). The deer is partially engraved and partially insinuated by the natural shape of the rock surface. The chronology is also uncertain with an initial attribution to Leroi-Gourhan’s Style III (Fullola & Viñas 1985:

160

Fullola et al: Small seeds for big debates 76-77), and later a less precise ascription to the Upper Palaeolithic (Fullola & Viñas 1988: 126). Thus, a direct review of this engraving considering an updated understanding of the sequence is required.

absence of the secondary anatomy and scarce presence of infill. But interestingly, in this sample of portable art the central figure representing a crane is almost complete, and non-figurative motifs are missing.

2.2. Evidence of portable art

2.2.2. Epimagdalenian

The cultural ascription of the examples of portable art is less problematic, since most of the pieces have been recovered in well-defined stratigraphic contexts. Thus, it can be noted that almost all the objects known so far are strictly from the final stages of the Palaeolithic sequence, and more specifically from the final moments of the Magdalenian and the Epipalaeolithic.

For this period 7 samples of portable art have been located at three different archaeological sites (Picamoixons, Molí del Salt [Assemblage Asup] and Sant Gregori) (Figs. 2 and 3). Unfortunately, three of the pieces from Sant Gregori and the one from Picamoixons lack archaeological context. Despite the risk of relying on art forms without real archaeological context to define the main artistic trends of this time period, the analysis of both the stratigraphy and the stylistic features of the pieces supports their ascription to this artistic phase.

2.2.1. Late Upper Magdalenian. For a few years, the only evidence in the region ascribable to this chronology was two quartzite schist plaquettes, with both figurative and non-figurative motifs, recovered at Molí del Salt (assemblage B) (Vimbodí, Tarragona), (García-Díez et al. 2002; García-Díez & Vaquero 2006) (Fig. 2). But the recent discovery of a limestone block with figurative motifs including humans and birds at L’Hort de la Boquera (García-Argüelles et al. 2014) has increased the artistic repertoire for the period (Fig. 3.10).

At Picamoixons (Alt Camp, Tarragona), a limestone slab with 7 curved and linear lines depicted in red was recovered (García-Díez et al. 1997). The red colouring material was directly applied over the rock surface using the raw material as a pencil or crayon, instead of transformed into paint. This piece of portable art was found out of context and a priori could be related to any of the human occupations identified at this site so far. According to the most recent studies (Garcia et al. 2009) the site has two archaeological levels, with Level IIA dating to between 9170±80 BP (10,46010,260 cal. BP) and 9570±50 B.P. (11,050-10,790 cal. BP), and Level IIB, dating to 11,055±90 BP (13,02012,820 cal. BP). Level IIB is clearly Epimagdalenian, as suggested by the authors. But the cultural ascription of Level IIA, which the authors interpret as belonging to the Notches and Denticulates Mesolithic, is far more complex. In our opinion both the absolute dates and the industries published seem more likely to belong to the Epimagdalenian or the Sauveterrian traditions. Interestingly, the stylistic analysis of the decorated piece also points to one of these two periods. The subject matter shows a certain equivalence with some French Azilian assemblages (Couraud 1985), pointing towards these Epimagdalenian and Sauveterrian traditions; and certainly it would not look out of place if compared to the painted pebble from Filador, ascribed to the Sauveterrian.

Molí del Salt assemblages B1 and B2 date to between 12,510 ± 100 BP (15,129-14,230 cal. BP) and 11,940 ± 100 BP (14,052-13,553 cal. BP). Only two quartzite schist plaquettes from this assemblage have been published so far. The type of raw material employed to produce these two samples of portable art reflect a local procurement. Animals predominate among the figurative motifs, with one aurochs, one possible hind and three undefined zoomorphic figures. But one anthropomorphic figure has been identified as well. The non-figurative motifs include a beam of rectilinear and curvilinear lines with a symmetric trend, among another unidentified set of strokes. Almost all of the motifs were engraved, with a preference for the simple outline technique, although the outline of one of the figures was produced with multiple orderly lines. Overall, the figures show the following morpho-stylistic features: incomplete anatomical formats, absence of secondary anatomy, scarce presence of infilling and internal lines, absolute profile perspective combined with straight biangular perspective, and a synthetic figurative formal conception.

Two of the slabs from Molí del Salt published so far belong to Assemblage Asup, an archaeological level dating to between 10,990±50 BP (12,995-12,729 cal. BP) and 10,840±50 BP (12,801-12,682 cal. BP) (GarcíaDíez et al. 2002; García-Díez & Vaquero 2006). The quartzite schist plaquettes used to produce these artworks were locally sourced, a practice already documented in the Late Upper Magdalenian levels of the same site. The figurative component is restricted to zoomorphic figures, including 2 hinds, 2 possible cervidae (potentially hinds as well), 2 horses and two indeterminate animals. All

The discovery from Hort de la Boquera (García-Argüelles et al. 2014) confirms some of the features already deduced for the period from the find at Molí del Salt, such as the dominance of engravings, local procurement of the raw materials, as well as the coexistence of animal (birds this time) and human figures. Some of the morpho-stylistic features are also replicated, with some incomplete anatomical formats (especially for humans),

161

Prehistoric Art as Prehistoric Culture

Figure 3. Other portable art objects from Northeastern Iberia. 1-3: Filador (drawings by Fullola & Viñas); 4-7: Sant Gregori (drawings by Vilaseca, 1934; Fullola et al. 1990); 8: Balma Guilanyà (drawings by M. López in Martínez et al. 2011); 9: Picamoixons (drawings by Garcia et al. 1997); 10: Hort de la Boquera (3D drawings by I. Domingo and GIFLE). None of the figures is scaled.

of them were engraved. The dominant technique is the simple outline, but the so-called ‘barbed wire’ technique (i.e. series of short oblique strokes attached to the simple outline) is also documented.

At Sant Gregori del Falset (Falset, Tarragona) three of the four art works known so far lack archaeological context. Once again engraving is the only technique used in the art production. Only the piece with a complete and finely engraved hind recovered by Vilaseca in 1932 was located in situ, in what he defined as Level 2 (Vilaseca 1934). This Epimagdalenian level includes endscrapers, backed bladelets and points. In the 1980s a new piece, from the Aznar private collection, including a possible horse, a hind and a bull, as well as a series of lines including a potential zig-zag, was studied and published by some

The morpho-stylistic features of these motifs include: complete anatomical formats, moderate presence of secondary anatomy, infilling and lines; absolute profile perspective, sometimes combined with oblique biangular perspective; and analytical figurative formal conception, combined with certain degree of stylization. 162

Fullola et al: Small seeds for big debates of the present authors (Fullola et al. 1990a: 279-86). Finally, García-Díez, Fontanals and Zaragoza published the last two finds in 2003: two pieces with linear motifs belonging to the Rodón collection, stored at Museu Municipal d’Alcover. Information on the provenance of these pieces is missing (García-Diez et al. 2003). Despite the lack of context of these last three artworks, they can be openly ascribed to the Epimagdalenian, since both the review of the lithic industry recovered by Vilaseca and the new excavations support a global affiliation of the archaeological deposit to this time period.

The slabs located by Vilaseca, making a single artwork, include an engraved reticulated motif covering the entire surface. In the other two pieces, the artistic intent includes not only a large number of engraved lines covering the surfaces, but also the deliberate retouching of the edges, probably designed to prevent delamination. Some reticulated motifs are visible among the strokes (Garcia-Argüelles et al. 2005: 73). These pieces are currently under study. But a quick description of one of them offers an idea of the imagery prevailing in this period. The larger piece measures 364 mm long, between 200 and 100 mm wide, and between 9 and 10 mm thick. This piece is made of slate and it is trapezoidal in shape. The engravings are all located on one of the surfaces. Overlapping strokes of various thickness and depth cover the surface, drawing at least one reticule and a series of parallel horizontal and diagonal lines. Stains of red pigment are still visible inside some of the engraved strokes (Fullola et al. 1986).

A global analysis of the artworks of this period indicates that both figurative (and more specifically zoomorphic) and non-figurative depictions coexist in this time frame. The animals depicted show a trend towards stylization, despite corresponding to a certain range of modes of representation. 2.2.3. Sauveterrian The site of Filador (Margalef de Montsant, Tarragona) yielded the only artistic evidence ascribable today to the Sauveterrian (Fig. 3.1-2). The site has provided 5 slate slabs with geometric and abstract engravings and 1 pebble with red lines. Three of the slabs were originally part of the same piece, but Vilaseca recovered them in two different field seasons. The largest fragment was found in 1948, in what he describes as a ‘hollow in Level III’ (Vilaseca 1949). This level corresponds to the Sauveterrian according to the industry (endscrapers, backed bladelets, segments with abrupt retouch, and a significant number of microburins) (Vilaseca 1949: 35253). The other two fragments were recovered in Level II during a 4-day excavation in 1963 (Vilaseca 1968). Once at the Museum, Vilaseca realised that the three fragments fitted together. This observation led to a series of misunderstandings on the chronology of these fragments, based on their recovery in two different archaeological levels. After discovering the two new fragments Vilaseca attributed the three pieces to Level II (Notches and Denticulates Mesolithic), discarding the initial allocation of the first find to Level III (Sauveterrian). It is unclear in either his publications or his field diary what led him to change his mind. This new chronological attribution has been repeated by other researchers (Martínez et al. 2011).

The other significant find at this site is a painted pebble located in 1982, in Square 7D, very close to the wall of the rock shelter. In fact this piece formed part of a torrential sediment flow from the Southwestern area of the site, which transported archaeological remains located in the upper part of Level 4, such as the painted pebble. It is a small and oval calcareous pebble measuring 57x40x18 mm, although it is partially broken. Interestingly, the surface was prepared with a very pale red layer before being decorated with a series of red bands: four on the best-preserved side, one on the edge and the last one on the fractured face. The first four bands are parallel, while the upper two join at one end. The use of some sort of brush can be deduced through the visual analysis of the bands (Fullola & Couraud 1984).

The reanalysis of the stratigraphy and the discovery of new slabs in the 1980s during the field seasons conducted by S.E.R.P. (University of Barcelona) provided new data for the discussion. The new finds (a further two engraved slate plaquettes recovered in 1986 and 1987) have similar decorations to the previous slabs. Furthermore, they are clearly located in current Level 4 (Vilaseca’s Level III), thus suggesting that the first three finds should also be ascribed to Vilaseca’s Level III, as initially published (Vilaseca 1949).

The only site with evidence for this period so far is Balma Guilanyà (Navés, Lleida), (Martínez et al. 2011) (Fig. 3.8). It is described as a large lutite block or slab, with non-figurative (geometric and abstract) engraved motifs, including a reticulated grid and an x-shaped form with a minimum of 8 arms. It combines two types of engravings, very fine strokes for most of the remains (including the reticulated grid), and deeper grooves for the x-shaped form, which was produced after the reticulate (Martínez et al. 2011: 167). The slab was

To sum up, the pieces from Filador define a period with geometric and abstract motifs dominating the collective imagination. These sorts of motif are either engraved or painted on two different sorts of rocks: slate plaquettes for the engravings and a calcareous pebble for the paintings. For some researchers, this difference in raw materials stems from the search for softer surfaces to facilitate engraving (García-Díez et al. 2003: 170). 2.2.4. Notches and Denticulates Mesolithic

163

Site

164

– – –

Tut de Fustanyà Cova del Parco Sant Gregori

Discarded

1 1 1

1 lutite block x – x

x

x

–

x

x

x

x

x

–

Engraving

Technique

– x –

–

–

–

–

–

–

Carving

–

– –

–

–

–

–

–

–

–

–

–

x –

–

x

x crayon

–

–

–

–

–

Figurative Painting Black Red x –

– – –

–

–

–

2

4

1?

–

–

–

Hind

– – –

–

–

–

–

–

–

–

1

–

Deer

– – –

–

–

–

1

–

1

–

–

1

Auroch

1 – –

–

–

–

–

–

–

–

–

–

Goat

– – –

–

–

–

1

2

–

–

–

–

Horse

– – –

–

–

–

–

–

–

1-2?

–

–

Bird

– – –

–

–

–

–

–

1

2-3?

–

-

Human

Subject matter Non-figurative

– – –

–

–

–

–

1

3

–

–

–

– – –

x

–

–

–

–

–

–

–

–

Indet. X-shape animal form

x – x

x

x

x

x

x

x

–

–

–

Lines

Figure 4. Inventory of finds, characteristics and subject matter documented in the Palaeolithic and Epipalaeolithic artistic assemblages of Northeastern Iberia .

–

–

–

Balma Guilanyà

Filador

Picamoixons

–

–

Molí del Salt Asup

Sant Gregori

–

Molí del Salt B1/B2

1 limestone block 2 quartzite schist plates 2 quartzite schist plates 4 slate plates 1 limestone slab 3 slate plates and 1 calcareous pebble

–

x

–

–

x

Portable art

Medium

Rock art

Notches and denticulated Mesolithic

Sauveterrian

Epimagdalenian

Late Upper Magdalenian

Hort de la Boquera

Prior to Solutrean Cova del Tendo Cova de la Magdalenian? Taverna

Chrono-cultural ascription

– – –

x

x

–

–

–

–

–

–

–

Reticulated

Prehistoric Art as Prehistoric Culture

Fullola et al: Small seeds for big debates periods. While cave art belongs to different phases within the Upper Palaeolithic (prior to the Upper Solutrean for Cova del Tendo and potentially Magdalenian for la Taverna), portable art works are confined to the Late Upper Magdalenian and the Epipalaeolithic (Table 1).

recovered in an archaeological layer dated between 8680 and 7320 BP (9790-8000 cal. BP), and more probably between 9800-9000 cal. BP (for further discussions see Martinez et al. 2011: 165). The authors discarded their potential interpretation as functional lines, rather than art forms, based on the irregular morphology of the block, the presence of curved lines and the identification of two individualized motifs (the reticulated grid and the x-shaped form). Whereas these researchers advocate similarities between Vilaseca’s find at Filador and this piece, regarding chronology and decoration, a direct review of the stratigraphy and Vilaseca’s field diaries, and the new finds at Filador in the 1980s support the initial ascription of the piece from Filador to the Sauveterrian, as discussed in the previous section.

While the examples of rock art provide little input to discussions on Palaeolithic art, the significance of the portable art assemblage lies in its comprising the only existing examples of prehistoric art known so far for most of the Epipalaeolithic (and specifically the Epimagdalenian, Sauveterrian and the Notches and Denticulates Mesolithic), and some of the very few known for the Late Upper Magdalenian in Eastern Iberia, with only two other examples in Castellón (Matutano) and Valencia (Parpalló).

Analysis and implications of the finds in north-eastern Iberia

Thus in this section we will mainly reflect on the portable art assemblage, summarizing the main characteristics and the evolutionary patterns of the art in the time periods represented. Nevertheless we are aware of the limitations of this analysis, based on a small number of samples, and it will have to be redefined in the future in the light of new discoveries.

The inventory of parietal and portable representations dating to the Palaeolithic and the Epipalaeolithic in Northeastern Iberia is still meagre today, especially when compared with the Cantabrian and the southern Spanish regions. Only two sites (Cova del Tendo and Cova de la Taverna) have evidence of rock art, with one motif each; while another six sites have provided the 15 pieces of portable art published so far in the region (Sant Gregori, Filador, Picamoixons, Molí del Salt, Balma Guilanyà and Hort de la Boquera). Interestingly each type of medium (cave art or portable art) is linked to different time

From the geographical point of view it is interesting to note that almost all the sites are concentrated in the same region, the Province of Tarragona, with only one example from further north, in the pre-Pyrenees of Lleida (Balma Guilanyà) (Fig. 4). This example is the only one

Figure 5. Geographical distribution of Palaeolithic and Epipalaeolithic art in Northeastern Iberia (white circle: rock art; black circle: portable art). 1-Moleta Cartagena, 2-Filador, 3-Hort de la Boquera, 4-Taverna, 5-Sant Gregori, 6-Picamoixons, 7-Molí del Salt, 8-Balma Guilanyà.

165

Prehistoric Art as Prehistoric Culture known for the last period with artworks, the Notches and Denticulates Mesolithic.

tradition was defined from the engraved plates with geometric decorations recovered at Cocina Cave (Fortea 1975, 1976). This disappearance of the figurative component, which is replaced by a geometric and abstract imagery, does not seem to support the hypothesis of continuity between Palaeolithic and Levantine rock art advocated by some researchers (Mateo 2012; Viñas 2012, among others), who suggest a connection between the Final Palaeolithic engravings from Castellón (described by Villaverde et al. 2006; Martínez et al. 2009; Guillem & Martínez 2009) and the Levantine engravings from Barranco Hondo (described by Utrilla & Villaverde 2004) and Llavería P-IV (Viñas & Sarriá 2011; Viñas 2012). But this is a separate topic deserving a more thorough analysis that is beyond the scope of this paper.

Regarding techniques, while engravings are produced throughout the entire sequence (from the late Upper Magdalenian to the Mesolithic) and over a range of rock surfaces of different sizes (slate plaquettes, quartzite schist plaquettes and limestone and lutite blocks), paintings are scarcely present, with only two examples produced in the central part of the sequence (Epimagdalenian and/or Sauveterrian) (Table 1). The analysis of the subject matter is of especial interest to explore the evolutionary patterns of prehistoric art in the region. While non-figurative motifs, especially linear and geometric forms, are present throughout the sequence, except at Hort de la Boquera, figurative motifs are restricted to the Late Upper Magdalenian and the Epimagdalenian.

Hence, to summarize, the portable art objects from Northeastern Iberia are not only of interest for describing the artistic assemblage between the Late Upper Magdalenian and the Mesolithic in Eastern Iberia. They also offer a unique and quite well-dated comparative framework to infer the potential chronology of the most recent finds of fine rock engravings discovered in Castellón.

Broadly speaking, a diachronic analysis of the figurative component reveals a slightly greater variability in subject matter in the Late Upper Magdalenian. The list of animal species is equivalent, with hind and aurochs in both periods, birds in the former and horses in the latter. But humans are exclusive to the Late Upper Magdalenian and remain absent until the emergence of Levantine rock art (Table 1).

The first figurative rock engravings located in Castellón were discovered in 2001 (Guillem et al. 2001) and include two types of animal depictions: more naturalistic forms, with elongated necks; and those with more geometric morphology. Further discoveries include the sites of Mas de les Llometes, el Abric de la Belladona, la Cova del Bobalar, el barranc de la Marfullada, el Coll de Cabres, el Mas de Serra Emporta, el Barranc del Gentisclar (Villaverde et al. 2006; Martínez et al. 2009) and l’abric de l’Espigolar (Guillem & Martínez 2009) in Castellón. The Sant Gregori plaquette with a hind has been used as one of the main references to infer the chronological and cultural adscription of the new finds (Guillem et al. 2001; Viñas 2012; Mateo 2012). A discussion, in which Northeastern Iberian sites have played and will continue to play a significant role, has only recently started and will probably last for a long time.

Some of the morpho-stylistic differences between these two periods have been defined from the finds at Molí del Salt (García-Díez & Vaquero 2006). These include: more incomplete anatomical formats in the Late Upper Magdalenian, more internal infilling and lines in the Epimagdalenian, less attention to the secondary anatomy in the Late Upper Magdalenian, preference for simple outline engravings, combined with multiple orderly strokes applied to the external outline in the Magdalenian and restricted to specific anatomical features in the Epimagdalenian; the ‘barbed wire’ technique also used in the Epimagdalenian; general predisposition to simplified forms in both periods even though more stylized in the Epimagdalenian. Some of these general traits are also recurrent in the pieces from Hort de la Boquera and Sant Gregori, though both sites offer some peculiarities as well, as discussed in previous sections.

Conclusions The artistic assemblage of Northeastern Iberia is limited but of great significance in prehistoric art research for several reasons.

From that moment on, there is a significant shift in subject matter, with a radical abandonment of the figurative component, which had persisted until the Epimagdalenian (10,000 BP, 11,500 cal. BP), which was replaced by non-figurative motifs (linear, geometric and abstract forms). This abstract geometric world, more structured during the Late Upper Magdalenian, the Epimagdalenian and the Sauveterrian (Filador) and with a less visible structure in the Notches and Denticulates Mesolithic (Balma Guilanyà), has continuity in the so-called Lineal-Geometric art in Eastern Iberia. This

First of all, it offered the first fully Palaeolithic discovery of rock art in Eastern Iberia (the disappeared black bull from Moleta de Cartagena). Second, the current portable art assemblage includes the only samples known so far for Epimagdalenian, Sauveterrian and the Notches and Denticulates Mesolithic, and two of the few examples known for 166

Fullola et al: Small seeds for big debates the Late Upper Magdalenian in Eastern Iberia. Thus they are unique for defining the artistic trends for these prehistoric time periods.

Fullola, J. Mª. 1987. Primera noticia de la troballa d’un gravat paleolític a la vall del Montsant (Priorat). Cypsela VI: 211-14. Fullola, J. M. & Couraud, C. 1984. Le galet peint de l’abri du Filador (Catalogne, Espagne). L’Anthropologie 83 (1): 119-23. Fullola, J. M. & Viñas, R. 1985. El primer grabado parietal naturalista en cueva de Cataluña: la Cova de la Taverna (Margalef de Montsant, Priorat, Tarragona). Caesaraugusta 61-62: 67-78. Fullola, J. M. & Viñas, R. 1988. Dernières découvertes dans l’art préhistorique de Catalogne (Espagne). L’Anthropologie 92 (1): 123-32. Fullola, J. M., Viñas, R. & García-Argüelles, P. 1986. La plaque en ardoise gravée de l’abri du Filador (Catalogne, Espagne). Cahiers Ligures de Préhistoire et de Prototohistoire, nouvelle série, nº3, Institut International d’Études Ligures: 145-56. Fullola, J. M., Viñas, R. & Garcia-Argüelles, P. 1990, La nouvelle plaquette gravée de Sant Gregori (Catalogne, Espagne), pp. 279-86 in L’art des objets au Paléolithique, Tome I, L’art mobilier et son contexte. Direction du Patrimoine: París. García-Argüelles, P., Adserias, M., Bartrolí, R., Bergadà, M .M., Cebrià, A., Doce, R., Fullola, J. M., Nadal, J., Ribé, G., Rodon, T. & Viñas, R. 1992. Síntesis de los primeros resultados del programa sobre Epipaleolítico de la Cataluña central y meridional, pp. 269-84 in Aragón / litoral mediterráneo. Intercambios culturales durante la Prehistoria. Inst. Fernando el Católico: Zaragoza. Garcia-Argüelles, P., Nadal, J. & Fullola, J. M., 2002. Vint anys d’excavacions a l’abric del Filador (Margalef de Montsant, Priorat, Tarragona). Tribuna d’Arqueologia 1998-99, ed. Servei d’Arqueologia, Generalitat de Catalunya, pp. 71-95. García-Argüelles, P., Nadal, J. & Fullola, J. M. 2005. El abrigo del Filador (Margalef de Montsant, Tarragona) y su contextualización cultural y cronológica en el nordeste peninsular. Trabajos de Prehistoria 62 (1): 65-83. Garcia-Argüelles, P., Nadal, J., Fullola, J. M., Bergadà, M., Domingo, I., Alluè, E. & Lloveras, L. L. 2014. Nuevas interpretaciones del Paleolítico Superior Final de la Cataluña meridional: el yacimiento de L’ Hort de la Boquera (Priorat, Tarragona). Trabajos de Prehistoria 71 (2): 242-60. Garcia Catalan, S., Vaquero, M., Pérez, I., Menéndez, B., Peña, L., Blasco, R., Mancha, E., Moreno, E. & Munoz, L. 2009. Palimpsestos y cambios culturales en el límite Pleistoceno-Holoceno: el conjunto lítico de Picamoixons (Alt Camp, Tarragona). Trabajos de Prehistoria 66 (2): 61-76. García-Diez, M. 2004. El grafisme moble del Molí del Salt i la figuració moble durant el Tardiglaciar en el vessant mediterrani de la Península Ibèrica, pp. 211-63 in (M., Vaquero, ed.) Els darrers caça-dorsrecol·lectors de la Conca de Barberà: el jaciment del

These sites have also provided an almost unique source for the relative dating of the most recent rock art finds, including the finely engraved animal figures discovered in Castellón. Finally they offer unquestionable graphic evidence of the lack of continuity between Palaeolithic and Levantine rock art. As previously discussed, the portable art assemblage in this region shows a significant shift in the artistic sequence with the disappearance of human figures in the Epimagdalenian and of animal figures from the Sauveterrian onwards. Thus any discussions of the final stages of the Palaeolithic and the Epipalaeolithic in Iberia have to refer to the sites in Northeastern Iberia, which makes them unique and irreplaceable, despite the small number of sites and representations known so far.

References Almagro, M. 1952, El covacho con pinturas rupestres de Cogul (Lérida). Instituto de Estudios Ilerdenses / Diputación Provincial de Lérida. Alonso, A. & Grimal, A. 1999. El Arte Levantino: una manifestación pictórica del epipaleolítico peninsular, pp. 43-76 in (A. Alonso, J. Aparicio, A. Beltrán, A. Grimal & G. Morote, eds) Cronología del Arte Rupestre Levantino. Ed. Real Academia de Cultura Valenciana, Serie Arqueológica, 17: Valencia. Breuil, H. 1908. Les pintures quaternàries de la Roca del Cogul. Butlletí del Centre Excursionista de Lleyda 10: 10-14. Carbonell, E. & Marcet, B. 1976. El Tut de Fustanya. Ill Assemblea d’Amics del Comtat de Besalu pp. 73- 86. Cartailhac, E. 1902, La grotte d’Altamira, Espagne. Mea culpa d’un sceptique. L’Anthropologie 13: 348–54. Couraud, C. 1985. L’Art Azilien. Origine et Survivance. XXème Supplement a Gallia Préhistoire, Paris. Duran, A. 1923. Exploració arqueològica del Barranc de la Valltorta (provincia de Castelló). Anuari de l’institut d’Estudis Catalans, vol. VI, 1915-1920: 451-54. Fortea, J. 1975. En tomo a la cronología relativa del Arte Levantino (Avance sobre las plaquetas de La Cocina), pp. 185-97 in El Aniversario de la Fundación del Laboratorio de Arqueología de Valencia, 1924-1974, Papeles 11, Valencia. Fortea, J. 1976. El arte parietaI epipaleolítico del 6.’ al 5.’ milenio y su sustitución por el arte levantino, pp. 12131 in IX Congrès Union International des Sciences Préhistoriques et Protohistoriques, Colloque XIX Pretirage, Nice.. 167

Prehistoric Art as Prehistoric Culture Molí del Salt (Vimbodí). Excavacions 1999-2003, Museu-Arxiu de Montblanc i Comarca: Montblanc. García-Diez, M. & Vaquero, M. 2006. La variabilité graphique du Molí del Salt (Vimbodí, Catalogne, Espagne) et l’art mobilier de la fin du Paléolithique supérieur à l’est de la Péninsule Ibérique. L’Anthropologie 110 (4): 453-81. García Diez, M., Martín, J., Gené, J. M. & Vaquero, M. 2003. La plaqueta gravada del Molí del Salt (Vimbodí, Conca de Barberà) i el grafisme Paleolític/ Epipaleolític a Catalunya. Cypsela 14: 159-73. García-Diez, M., Rosell, J., Vallverdú, J. & Vergès, J. M. 1997. La plaqueta pintada del yacimiento epipaleolítico de Picamioxons (Alt Camp, Tarragona): aproximación al estudio de la cadena operativa. Pyrenae 28: 25-40. García-Diez, M., Fontanals, M. & Zaragoza, J. 2003. Dues noves peces gravades del jaciment de Sant Gregori (Falset, Tarragona): la ‘col·lecció Ramon Rodón’ del Museu Municipal d’ Alcover (Alt Camp). Pyrenae 33-34: 165-74. Grimal, A. & Alonso, A. 1989. Sobre la figura de tipo levantino en la Cova del Tendo. Moleta de Cartagena (Montsià-Tarragona). Boletín de la Asociación Española de Arte Rupestre, Barcelona, 2: 18-20. Guillem, P. & Martínez, R. 2009. Arte rupestre en el Cingle del Barranc de l’Espigolar (La Serratella, Castelló), pp. 35-48 in (J. A. López, R. Martínez & C. Matamoros, eds) El arte rupestre del Arco Mediterráneo de la Península Ibérica: 10 años en la lista del Patrimonio Mundial de la UNESCO. Valencia. Guillem, P., Martínez, R. & Melià, F. 2001. Hallazgo de grabados rupestres de estilo paleolítico en el norte de la provincia de Castellón: el Abric d’en Melià (Serra d’en Galceran). Saguntum-PLAV 33: 133-40. Hernández, G. & Hernández, M. S. 2013. Art rupestre a l’arc mediterrani de la península Ibèrica. Del Cogul a Kyoto. Catalan Historical Review 6: 129-46. Hernández Pacheco, E. 1924. Las pinturas prehistóricas de las cuevas de la Araña (Valencia). Evolución del arte rupestre de España. Com. de Inv. Paleontológicas y Preh. Mem. núm. 34, Madrid. Maluquer de Motes, J. 1983. Un jaciment Paleolític a la comarca de la Noguera. Pyrenae 19-20: 215-33. Martí, B. & Hernández, M. S. 1988. El Neolític valencià. Art rupestre i cultura material. Diputación de Valencia: Valencia. Martí, B. & Cabanilles, J. 2002. La decoració de les ceràmiques neolítiques i la seua relación amb les pintures rupestres del abrics de La Sarga, pp. 147-70 in (M. S. Hernández & J. M. Segura, eds) La Sarga. Arte rupestre y territorio. Alcoi. Martinez, R., Guillem, P. M.  & Villaverde, V. 2009. Grabados rupestres de estilo paleolítico en el norte de Castellón, pp. 225-36 in (R. de Balbín, ed.) Arte Prehistórico al aire libre en el Sur de Europa. Junta de Castilla y León: Salamanca.

Martínez, J., Villaverde, V. & Mora, R. 2011. La placa grabada de Balma Guilanyà (Prepirineo de Lleida) y las manifestaciones artísticas del Mesolítico de la Península Ibérica. Trabajos de Prehistoria 68 (1): 159-73. Mateo, M. A. 2012. Defining Neolithic art. Levantine, Macroschematic and Schematic art in the Mediterranean Arc of the Iberian Peninsula, pp. 16786 in (J. J. García, H. Collado & G. Nash, eds) The Levantine Question. Archaeolingua: Budapest. Mateo, M. A. 2008. La cronología neolítica del arte rupestre levantino: ¿realidad o deseo?’ Quaderns de Prehistòria i Arqueologia de Castelló 26: 7-27. Molina, L. L., Garcia, O. & García, R. 2003. Apuntes al marco crono-cultural del arte levantino: Neolítico vs neolitización. Saguntum-PLAV 35: 51-67, Obermaier, H. 1916. El Hombre Fósil. Mem. núm. 9 de la Com. De Inv. Paleontológicas y Prehistóricas. Olaria, C. 2008. Cova Matutano (Vilafamés, Castellón). Grafismo mobiliar magdaleniense en el contexto del Mediterráneo peninsular. Monografies de Prehistòria i Arqueologia Castellonenques 7. Pericot, L. 1942. La Cueva del Parpalló (Gandía). CSIC, Inst. Diego Velázquez: Madrid. Ripoll, E. 1965. Una pintura de tipo paleolítico en la Sierra del Montsiá (Tarragona) y su posible relación con los orígenes del arte levantino, pp. 297-305 in Miscelánea en Homenaje al Abate Henri Breuil. Barcelona. Ripoll, E. 1970, Acerca del problema de los orígenes del Arte Levantino, pp. 57-67 in Valcamonica, Simposium. Centro Camuno di Studi Preistorici: Capo di Ponti. Sanchidrián, J. L. 2001. Manual de Arte Prehistórico. Ariel: Barcelona. Utrilla, P. & Villaverde, V. 2004. Los grabados levantinos del Barranco Hondo. Castellote (Teruel). Gobierno de Aragón. Departamento de Educación, Cultura y Deporte. Vidal, J. M. 1909. Les pintures rupestres de Cogul. Anuari de l’Institut d’Estudis Catalans MCMVIII: 544-50. Vilaseca, S. 1934. L’Estació-taller de sílex de Sant Gregori. Memoria de la Academia de Ciencias y Arte de Barcelona 23 (21): 415-39. Vilaseca, S. 1936. La indústria del sílex a Catalunya. Les estacions tallers del Priorat i extensions. Llibreria Nacional i Estrangera: Reus. Vilaseca, S. 1949. Avance al estudio de la Cueva del Filador, de Margalef (provincia de Tarragona). Archivo Español de Arqueología XXII: 347-61. Vilaseca, S. 1953. Las industrias del sílex tarraconenses. C.S.I.C., Instituto Rodrigo Caro: Madrid. Vilaseca, S. 1968. Cuatro días en la cova del Filador, (Margalef), pp. 476-90 in La Préhistoire, problèmes et tendances. CNRS: París. Vilaseca, S. 1973. Reus y su entorno en la Prehistoria. Rosa de Reus 48 & 49: 1-2, 282, Reus. 168

Fullola et al: Small seeds for big debates Villaverde, V. 1994. Arte mueble de la España mediterránea: breve síntesis y algunas consideraciones teóricas. Complutum 5: 139-62. Villaverde, V., Guillem, P. M. & Martínez, R. 2006. El horizonte gráfico centelles y su posición en la secuencia del Arte Levantino del Maestrazgo. Zephyrus 59: 181-98. Villaverde, V., Martínez, R., Guillem, P. M., López, E. & Domingo, I. 2012. What do we mean by Levantine rock art?, pp. 81-115 in (J. J. García, H. Collado & G. Nash, eds) The Levantine Question. Archaeolingua: Budapest. Viñas, R. 2012. Superimposition in Spanish Levantine Rock Art: previous proposals and new evidence for a reassessment, pp. 55-80 in (J. J. García, H. Collado & G. Nash, eds) The Levantine Question. Archaeolingua: Budapest. Viñas, R. & Sarrià, E. 2011. Documentació dels nous conjunts d’art rupestre del Priorat (Tarragona). Tribuna d’Arqueología 25 (2009-2010): 53-84.

169

170

Throwing light on the hidden corners. New data on Palaeolithic art from NW Iberia Ramón Fábregas Valcarce,1 Arturo de Lombera-Hermida,1,2,3 Ramón Viñas Vallverdú,2,3 Xose Pedro Rodríguez-Álvarez,2,3 and Sofia Soares Figueiredo4 Grupo de Estudos para a Prehistoria do Noroeste (GEPN), Dpto Historia I, Universidade de Santiago de Compostela, Pz. Universidade nº1, 15782 Santiago de Compostela. E-mail: ramó[email protected]; [email protected] 2 IPHES, Institut Català de Paleoecologia Humana i Evolució Social, C/ Marcel.lí Domingo s/n- Campus Sescelades URV (Edifici W3) 43007 Tarragona, Spain. E-mail: [email protected]; [email protected] 3 Area de Prehistoria, Universitat Rovira i Virgili (URV). Avinguda de Catalunya 35, 43002 Tarragona, Spain. 4 Lab2PT- Landscapes, Heritage and Territory Laboratory. University of Minho, Portugal. E-mail: [email protected]

1

Abstract NW Iberia has often been described as a nearly empty space when considering the Palaeolithic settlement of the Iberian Peninsula. This fact has been especially patent with respect to the Palaeolithic art. The discovery at the end of the 20th century of the open-air Palaeolithic rock art in the Côa Valley abruptly changed that perspective. Systematic research projects on the limestone formations of Eastern Galicia and the intense archaeological surveys linked to the hydroelectric projects of the northern tributaries of the Douro river have contributed to finding new examples of Palaeolithic rock art and portable art. Stylistic studies point to a convergence with the observed trends in other Iberian regions such as, for instance, the high number of sites belonging to the late Upper Palaeolithic and the early Holocene. Keywords Northwest Iberia; cave art; open-air art; portable art; Magdalenian; Epipalaeolithic

was the discovery of the Côa rock-art which led to an important effort to survey the whole area; this in turn led to the finding of numerous decorated panels, engraved plaquettettes and also unassailable evidence of the human groups that were responsible for the carvings (Aubry 2009). Subsequently, the research momentum that this generated, together with the implementation of rescue archaeology linked to hydroelectric developments in NE Portugal, has produced a steady Drang nach Norden of finds of Palaeolithic art, in areas like the valleys of the Tua and Sabor (Figure 1).

1. Introduction In spite of relatively early and auspicious beginnings during the last third of the 19th century, institutional, economic and even geological circumstances led to NW Iberia remaining outside of Palaeolithic research until, almost, the last third of the 20th century. Even so, the limited advances taking place were marred by the limited academic support and continuity of the research, notwithstanding the interesting appraisals of questions such as the settlement during the last phases of the Upper Palaeolithic in the NE rock-shelters of Galicia. It is precisely in the latter area that the first example of portable art was reported: a small stone pendant ascribed to the Magdalenian was recovered in the site of Férvedes-II (Ramil & Vázquez, 1983). In addition, by 1981 an isolated panel with animal carvings had been found in Tras-os-Montes, that of Mazouco (Jorge et al. 1981, 1982).

Another circumstance, which particularly affected the Galician region, involved the growth of systematic research projects in several areas, like the Baixo Miño or – more relevant to our present concern – in SE Lugo, one of the few areas with a significant presence of karstic formations (de Lombera & Fábregas, 2011). It is precisely in the latter region where an increasing number of finds under the label of art has been reported since the beginning of the present century. They are concentrated in just 2 caves (Valdavara and Cova Eirós) and encompass categories such as pendants, Dentalium beads and decorated sagaies. The identification of painted and engraved figures in Cova Eirós in September 2011 meant the first discovery of cave art in NW Iberia, ending for good the anomaly of the void in Palaeolithic cave art in Galicia (de Lombera & Fábregas 2013).

When the awesome artistic assemblage at Foz Côa came to light in the mid 1990s, the Iberian Northwest was still a blank area with respect to the distribution of Palaeolithic art (Zilhão & Fábregas, 1996), a rather surprising fact given the long-known decorated caves of neighbouring Asturias and the regular discovery of openair carvings in the Western Meseta. Two main factors were to produce a turn-around in this situation: the first 171

Prehistoric Art as Prehistoric Culture

Figure 1. Location of the main sites mentioned in the text. A) Cova Eirós. B) Cova de Valdavara. C) Férvedes-II. D) Distribution of the sites mentioned in the Trás-os-Montes region (map by Ana Rita Ferreira).

carvings or paintings, as has happened in the Cantabrian region. The discovery of cave art in Cova Eirós led to a thorough review of the accessible galleries, as well as a series of analyses aimed to confirm its Palaeolithic attribution.

2. The cave art of Cova Eirós The quest for cave art in Galicia has been severely limited by the dominance of a Paleozoic geological substratum where limestone formations are restricted to narrow strips in the Eastern Sierras of Galicia (de Lombera 2011). The increased activity of Galician speleologists in recent decades might have brought to light examples of cave art but, since no systematic, archaeologicallysupervised, effort has been undertaken, we cannot rule out the appearance of Palaeolithic art even in caves known from old, perhaps containing hardly visible

Procedence

Level 2

Level B

Cova Eirós is located in the eastern Sierras of Galicia (Triacastela, Lugo) at 780 m asl. The cave is currently 104 metres long and the entrance narrows after the first 7 metres into a 15 m long neck, followed by the cave’s largest space, the ‘Main- or Mammoth- Hall’, 15 m long, 6 m wide at its broadest and up to about 5 metres high. Since

Passage

Entrance (UA6)

hearth nner test-pit

IPanel XI

Ref. Lab

Beta - 254280

Beta - 308859

Ua-38121

Beta - 308578

Beta - 333971

Beta - 345400

Method

C14-AMS

C14-AMS

C14-AMS

C14-AMS

C14-AMS

C14-AMS

Sample

Animal bone

Charcoal

Human phalanx

Charcoal

Charcoal

Charcoal

Years BP

31690 ± 240

12040 ± 50

3151 ± 31

1040 ± 30

1020 ± 30

1050 ± 30

Reference

Rodríguez et al. 2011

Fábregas et al. 2012

Vidal et al. 2010

Fábregas et al. 2012

Unpublished

Unpublished

Table 1. Radiocarbon dates from Cova Eirós.

172

Fábregas Valcarce et al: Throwing light on the hidden corners. New data on Palaeolithic art from NW Iberia 2008 systematic archaeological work has been carried out at the cave’s entrance, yielding several Upper Palaeolithic occupations ranging from the Aurignacian (Level 2) and Gravettian (Level 1) to the Final Magdalenian (Level B) (Table 1) (Rodríguez et al. 2011). Several items of portalle art have been found amidst the lithic and faunal remains of the Upper Palaeolithic levels (Fábregas et al. 2012). A small pendant in a perforated canine, probably from a fox (Vulpes vulpes), occurred in the Gravettian level (Level 1). Several bone items were also retrieved, including a double-pointed sagaie from the interfaces of an Upper Palaeolithic level (1C). Both sides are decorated with a zigzag pattern composed of several parallel, discontinuous lines, whose closest equivalent is to be found in other Magdalenian sites from the Cantabrian Coast (Fábregas et al. 2012) (Figure 2). 2.1. Rock art finds Since 2011 thirteen decorated panels have been identified (de Lombera & Fábregas 2013). They are concentrated in the Hall of the Mammoth, the largest space of the cavity (Figure 3): on its walls there are numerous painted and engraved images at an intermediate or low height, never more than 2.5 m above the current soil surface. Here more complex motifs and associations are found, including the co-occurrence of black paintings and engravings, especially in Panel III. Other graphic examples are found in different alcoves and in the south-east gallery, but in lesser densities than in the Main Chamber. Some surface finds and a test-pit dug inside the cave have shown intensive occupations of the cave during late prehistoric (Neolithic and Bronze Age) and historic times. A large bonfire discovered just beneath Panel III and some charcoals recovered from a fissure in Panel XI were dated to AD 900-1000, contemporary with the storage pits recorded at the entrance (Teira et al. 2012) (Table 1). These pieces of evidence point to intense activity during medieval times, both at the entrance and inside the cave. We have inventoried a total of 93 Graphic Units, half of them painted, followed by engravings (46%). The motifs are usually small, constrained by the limited availability of space on surfaces that are very cracked and weathered, but they seem to reflect a stylistic choice as well. Another feature – and one of the main obstacles facing their study – is the poor state of preservation, either by the washing-off of the paintings (which may explain some of the isolated spots or lines) or due to the blackish film that covers part of the tracings, as shown in Panel III. Many modern graffiti, some dating back to the early 20th century, have altered or nearly destroyed many figures. Thus, the analysis of rock art in Cova Eirós requires a thorough and systematic survey of the walls, as the visibility of the graphics is very poor.

Figure 2. Portable art from Galicia. A) sagaie and C) perforated canine from Cova Eirós. B) Dentalium beads from Cova de Valdavara.

173

Figure 3. Plan of the Cova Eirós cave and location of the panels.

Prehistoric Art as Prehistoric Culture

174

Fábregas Valcarce et al: Throwing light on the hidden corners. New data on Palaeolithic art from NW Iberia

Figure 4. Zoomorphic representations from Cova Eirós. A) Bovid and cervid (Panel I-P5 & P6). B) Example of superimpositioning of the painted and engraved motifs in Panel III (Panel III-P1 & G1). C) Bovid (Panel IV-G3). D) Fusiform signs (Panel II-G2; Panel III-G4, Panel IV-G6).

along with a motif defined by a double series of black dots (PIII-P2).

The type of support is very relevant to the study of engravings since many of them were done on the softened walls of the cave, and the superficial clay layer was susceptible to subsequent removal by water runoff, while engravings executed directly on the bedrock are more resilient. This differential erosion may explain, in turn, the common finds of zoomorphic anatomical segments or partial motifs.

Dots or simple paint marks, isolated or grouped, are usually placed in the lower reaches of the panels all over the cave. Depending on their location, this kind of marks or tracings can be interpreted as intentional or, instead, as evidence of reviving a torch, accidental scratches or topographic signals, whose execution could reach historic times (García & Gonzalez, 2003). In this sense, we must bear in mind the intense activity inside the cave during Medieval times. Also frequent are fine-line engravings, isolated or in groups, and often appearing as a maze that covers much of the surface. Panel VIII is a good example, displaying a dense array of lines, among which both abstract and figurative images are present.

As to themes, three categories have been identified: 1) Figurative motifs, especially incomplete zoomorphs, such as limbs, dorsal lines and heads; 2) some dots and paint marks applied to natural reliefs, perhaps underlining presumed animal features; and 3) signs, abstract motifs, dots and dispersed lines. In the figurative section bovids, horses, deer, and pisciforms have been documented. Some of these motifs are complete (PI-P5, PI-P6 and Panel XI-G1), but incomplete contours of animal figures are the most common, due to the aforementioned preservation conditions of the panels. Among the third group several abstract motifs and well-defined fusiform and ramiform signs have been recorded (Figure 4d),

Among the paintings the black pigment is clearly dominant for tracing the contours of the animals; in Panel II only some possible red spots departs from the seemingly general use of charcoal. The FT-Raman spectrum obtained on one sample from Panel III (CE11-

175

Prehistoric Art as Prehistoric Culture MC-1ab, PIII-P1) clearly shows broad bands at 1340 cm-1 (D band) and 1597 cm-1 (G band), related to amorphous carbon of organic origin (Hernanz et al. 2010). The engravings are usually thin and shallow, at times forming dense clusters, but occasionally the grooves have a wider U-section, made with a blunt engraver. For the figured animals, in only one case is the body of the animal filled with longitudinal lines (PIII-G1).

would be consistent with the chronology of the Upper Palaeolithic levels reported at the cave entrance, but the actual execution of the graphic representations further inside could have a complex sequence, not necessarily or always related to the site’s settlement sequence. 3. Northeast Portugal As for Northern Portugal, before the astounding discoveries in the area of Foz Côa, an open-air rock art site with Palaeolithic engravings had already been detected and published in the 1980s. In the region on which we are focusing in this text, called Trás-os-Montes, located north of the Côa valley, in the small locality of Mazouco (Freixo-de-Espada à Cinta), a horse figure was identified on schist on the right bank of the Douro river (Jorge et al. 1981, 1982). Further to the south, at Fraga do Gato in the Ribeira do Mosteiro valley, another site was also identified in the 1980s, although it was only later recognized as a Palaeolithic panel containing one of the few examples of open-air painted rock art with a unique representation of a black owl and a red otter (Figure 5d) (Baptista 2009; Figueiredo et al. 2011).

Technical and stylistic studies are still going on, but some techno-morphological traits allow us to put forward a working hypothesis about the chrono-cultural framework for these artistic manifestations. The overlapping of several engravings (i.e. Panel VIII) and, especially, the superimposition of two painted motifs on two engravings in Panel III point to the existence of several episodes in the configuration of the artistic representations in Cova Eirós. With the available data, we might distinguish two main assemblages. Some painted motifs, such as the double series of black dots, and certain multiple and crossed carved lines (‘V’ and ‘X’ shaped) might suggest an older chronology (Gravettian/Solutrean?). However, the animal figures provide more clues about their stylistic attribution: the zoomorphs with elongated and schematic bodies, a simplified representation of the limbs and filled-in bodies (i.e. the bovid from Panel I-P5, or the engraved zoomorph from Panel III-G1) (Figures 4 a and b), together with the small size of the figures seemingly point to the final moments of the Magdalenian or the transition to the Epipalaeolithic (Bueno et al. 2009; Viñas et al. 2010). These traits are also found on some plaquettettes from Rock #1 at Fariseu, dated on stylistic and archaeo-stratigraphic grounds to 11,500-9,500 BP, or engravings at Penascosa (Rock 10) or Canada do Inferno (Rock 14) in the Côa Valley (García & Aubry 2002; Aubry & Sampaio 2009). Similar figures with fine-line engravings and bodies filled-in with striations, superimposed on older motifs, are also found in the open-air site of Siega Verde (Northern Meseta) (Alcolea & Balbín 2006), Abrigo do Passadeiro (Miranda do Douro, North Portugal) (Sanches & Teixeira 2014) and on portable art with geometric carvings belonging to the final Magdalenian/early Epipalaeolithic from several locations in the Northern Meseta (Estebanvela, La Dehesa, La Uña) (Bueno et al. 2009; García Díez, 2013). In caves, too, the painted figures from Sala de las Pinturas at Cueva Palomera (Burgos), dated to 11,470 ± 110 BP and 10,950 ± 100 BP (Corchón et al. 1996), and some motifs from Escoural (Portugal) can be related to this style.

In the 1990s and in the 21st century, during the preparation of the construction of two large hydroelectric dams in the Trás-os-Montes region – one in the Sabor river valley and the other in the Tua river valley, –intense field surveys began to be carried out. As a result, five new rock art sites with Palaeolithic engravings were discovered, four in the Sabor valley and one in the Tua valley. The Abrigo Rupestre de Foz Tua comprises two adjacent schist shelters, where fifty panels have been recognized (Valdez-Tullett 2013; Teixeira et al. 2010). Among schematic paintings and abstract or geometric engraved figures, in panel 31 of shelter A, the archaeological team responsible for its study identified naturalistic representations of animals in Palaeolithic style. Here, we find the depiction of a caprid, and a much more complex figure where the same body seems to have three different animal heads, representing an aurochs, a horse and a red deer (Valdez-Tullett 2013). Moving further east, to the Sabor valley, four open-air rock art sites were found before the beginning of the construction of the Sabor dam. The waters of this major dam will submerge one of them. Starting from the north and moving south, the first outcrop, Sampaio, displays three (probably four) aurochs in three different panels (Baptista 2009). The Pousadouro site is formed by a small shelter and, among the four Sabor valley openair Palaeolithic rock art sites, is the only one that is not mono-thematic. In fact, in the first panel one can observe an interesting group formed by horses and red deer and, in a second one, a horse looking back can be discerned. The third rock, Pedra Escrevida, is an immense outcrop where one finds a virtually full-scale representation of an aurochs. The final site, further to the south, will be

The presence of charcoal in the pigments and certain engravings recovered by calcitic crusts may make it feasible to obtain radiometric dates that will help to specify the chrono-cultural attribution of the rock art at Cova Eirós. All in all, the parallels considered above 176

Fábregas Valcarce et al: Throwing light on the hidden corners. New data on Palaeolithic art from NW Iberia submerged and it is named Ribeira da Sardinha. This schist rock also displays an aurochs figure, although the size of the figure is much smaller than the one previously described.

others is the horse. It is also interesting to note that there are two species that never occur together, the ibex and the red deer.

The Sabor hydroelectric project had a huge impact on the landscape and submerged an area of 3000 hectares. In order to minimize the impact, a Plan for Heritage Protection was created, which began in 2010 and has now come to an end. The excavations at Foz do Medal Terrace took place from 2011 until 2013, revealing one of the most significant finds of the last decade regarding Palaeolithic art. In this site, 1511 fragments of engraved plaquettes were identified during the archaeological excavations (Figueiredo et al. 2014, 2015). As we have seen, all the open-air rock art sites were included in the oldest phase of the Côa Valley; but at the Medal terrace, at least 1257 fragments were found in a layer with remains from the Magdalenian period. Unfortunately, this deposit, layer 1055, suffered a post-depositional shift. Although the level moved from its original position, the integrity of the artifact collection is still assured.

Without the support of direct dating, establishing acceptable chronologies for this collection is rather challenging. Taking possible analogies between the Foz do Medal Terrace and other Portuguese sites containing portable art into account is not easy at this stage. First, because of the fact that in two of the considered sites only two plaquettes were detected: one in Quinta da Barca Sul and the other in Chancudo 3 (Pereira 2010; Aubry 2009). Secondly, because a complete study of the Fariseu plaquettes has not yet been published. Nevertheless, regarding this last important site and its portable art collection, from the Late Magdalenian period, situated in chronological terms between 12,000 and 10,000 BP (Santos 2012), we observe that the zoomorphic figures, mostly ibex and deer, are characterized by almost quadrangular bodies, filled with multiple thin lines (Aubry 2009; Baptista 2009). Considering the Magdalenian zoomorphic figures from the Foz do Medal Terrace, the figures change substantially. Here the shapes are rounded, more naturalistic, and the animals are drawn with a well-defined contour line. But moving from the Fariseu portable art assemblage to Foz Côa’s rock art, we find similarities between some of the Medal representations and the recently defined Phase 2 of Côa’s open-air rock art (Santos 2012). This Phase is linked to Leroi-Gourhan’s Style IV. Further investigation of the Medal collection will certainly bring more insights.

When studying and classifying the representations found on the Magdalenian plaquettes, we considered four different categories: figurative, geometric, abstract and indeterminate motifs (Figueiredo et al., in press.). Among the 1257 fragments analysed, abstract figures are dominant, accounting for almost half of the representations. They are followed by geometric patterns, indeterminate figures and, finally, figurative themes. However, when analysing the collection, it is the latter group that provides us with better clues about styles, making it possible to produce a more precise chronology.

Ending our brief summary of finds of Palaeolithic rock art in the Trás-os-Montes region, the richness of this area has become clear (Figure 1), and it is most likely that new discoveries will take place, especially further to the north in both the Sabor and the Tua basins, as well as in their tributaries. In the future, issues such as the relationship between the isolated sites along the valley and their connection to the Côa valley must be addressed, as well as the role of open-air sites such as the Foz do Medal Terrace, where the high number of decorated plaquettes discovered has already surpassed the largest collections of this kind in at the whole of Europe.

In order of importance, the animals depicted are ibex, horses, aurochs and red deer (Figure 5). In the ibex representations, the style ranges from schematic to a more naturalistic form, where great attention is given to details. The same trend is observed in the representation of horses, where different styles are found in the same theme. Regarding the depiction of aurochs, we see that the representations are more homogeneous than the previous cases. Finally, red deer are very poorly represented in the collection, with only one or two clear cases showing more naturalistic features. Regarding the associations between different species in the same operative space, the animal that is the most related to the

4. Conclusions

With the exception of Fraga do Gato, where the lack of similar occurrences in the region does not make it possible to indicate a precise chronology, the themes and styles present in the seven sites we have briefly described are included in the first known phase of the Côa valley, which is the Gravettian-Solutrean phase (Baptista 2009). Until the summer of 2011, this was the known panorama for Palaeolithic rock art in the Trás-os-Montes region.

The discoveries in the Côa Valley (North Portugal) not only meant the cataloguing of new rock art sites away from traditional research zones, but also highlighted the importance of the open-air art in Palaeolithic societies. Despite its utmost importance, some years would pass by before new finds occurred in other regions of NW Iberia. The first decade of the present century saw the implementation of research programmes in the limestone areas of the north-west, together with a massive development of Rescue Archaeology in the northern tributaries of the river Douro. Progress in these 177

Prehistoric Art as Prehistoric Culture

Figure 5. Representation of A) an ibex (A), B) a horse (B), C) an aurochs from Foz do Medal Terrace (Drawing by Figueiredo et al.. in press; photo by Adriano Ferreira Borges). D) Representation of a black owl and a red otter from Fraga do Gato. Drawing by Baptista, 2009.

178

Fábregas Valcarce et al: Throwing light on the hidden corners. New data on Palaeolithic art from NW Iberia investigations has yielded a significant amount of rock art both in caves and in open-air locations and, at Foz do Medal Terrace, an impressive collection of portable art.

Aubry, T. & Sampaio, J. 2009. Chronologie et contexte archéologique des gravures paléolithiques de plein air de la Vellée du Côa (Portugal),pp. 211-23 in (R. de Balbín Berhmann, ed.) Arte prehistórico al aire libre en el sur de Europa. Junta de Castilla y León, Consejería de Cultura y Turismo: Valladolid. Baptista, A. M. 2009. O Paradigma Perdido: O Vale do Côa e a Arte Paleolítica de Ar Livre em Portugal. Edições Afrontamento e Parque Arqueológico do Vale do Côa. Bueno Ramírez, P., de Balbín Berhmann, R. & Alcolea González, J. J. 2007. Style V dans le bassin du Douro. Tradition et changement dans les graphies des chasseurs du Paléolithique Supérieur européen. L’ Anthropologie 111: 549-89. Bueno Ramírez, P., de Balbín Berhmann, R. & Alcolea González, J. J. 2009. Estilo V en el ámbito del Duero: Cazadores finiglaciares en Siega Verde (Salamanca), pp. 259-86 in (R. de Balbín Berhmann, ed.) Arte Prehistórico al aire libre en el Sur de Europa. Junta de Castilla y León. Consejería de Cultura y Turismo: Valladolid. Corchón, M. S., Valladas, H., Bécares, J., Arnold, M., Tisnerat, N. & Cachier, H. 1996. Datación de las pinturas y revisión del arte paleolítico de Cueva Palomera (Ojo Guareña, Burgos, España). Zephyrus 49: 37-60. Fábregas Valcarce, R., de Lombera Hermida, A., Serna González, M. R., Vaquero Rodríguez, M., Pérez Rama, M., Grandal D´Anglade, A., Rodríguez Álvarez, X. P., Alonso Fernández, S. & Ameijenda Iglesias, A. 2012. Ocupacións prehistóricas e históricas nas cavidades das Serras Orientais galegas. As covas de Eirós (Triacastela) e Valdavara (Becerreá). Gallaecia 31: 19-46. Figueiredo, S. S., Gaspar, R. & Xavier, P. 2011. Cruzando ocupações pré-históricas e arte rupestre no vale da Ribeira do Mosteiro: dados da primeira campanha, pp. 125-59 in Actas do V Congresso de Arqueologia - Interior Norte e Centro de Portugal. Caleidoscópio, Direcção Regional de Cultura do Norte: Vila Real. Figueiredo, S. S., Nobre, L., Gaspar, R., Carrondo, J., Cristo Ropero, A., Ferreira, J., Silva, M. J. & Molinna, F. J. 2014. Foz do Medal Terrace - an openair settlement with palaeolithic portable art. INORAInternational Newsletter on Rock Art 68: 12–20. Figueiredo, S. S., Nobre, L., Cristo Ropero, A., Xavier, P., Gaspar, R.& Carrondo, J. 2015. Reassembly Methodology in Palaeolithic Engraved plaquettes from Foz do Medal Terrace (Trás-os-Montes, Portugal), pp. 428-39 in (M. Á. Medina-Alcaide, A. J. Romero Alonso, R. M. Ruiz-Márquez & J. L. Sanchidrián, eds) Sobre rocas y huesos: las sociedades pré-históricas y sus manifestaciones plásticas. Torti. Figueiredo, S. S., Xavier, P. & Nobre, L. (in press) Placas móveis com grafismos rupestres paleolíticos do Terraço do Medal (Nordeste, Portugal): uma primeira

The recent discoveries open the door to new finds in traditional niches such as the karstic caves, which have been little researched, and at open-air sites in the geological domains of schist and quartzite rocks. As to the latter, the proximity of the recent discoveries in the Tras-os-Montes region to the Galician-Portuguese border heralds the presumed existence of Palaeolithic open-air sites in south-east Galicia, taking into account not only the good natural communications but also the lithological similarities with the Trasmontane districts. Our present knowledge is consistent with the reported intensity of settlement during the later phases of the Upper Palaeolithic in our study area. From the archaeological point of view, this previously unknown corner of Iberia corner follows a similar trend to that of other Iberian regions and may play a key role in the communication between the Cantabrian rim and the Atlantic Façade. Although some of our samples can be attributed to the Gravettian and Solutrean stages, equivalent to the first artistic phase identified in the Côa Valley, the more culturally diagnostic belong to the Magdalenian and even later periods. More systematic surveys and dates, in both known and unexplored regions of NW Iberia, must be carried out in order to enlarge the Upper Palaeolithic artistic corpus.

Acknowledgements Fieldwork and research at Cova Eirós were funded by the Spanish Ministerio de Economía y Competitividad (MINECO), project HAR2010-21786/HIST. A. de L-H was awarded a pre-doctoral grant from the Atapuerca Foundation. The Plan of Heritage Protection is part of the Baixo Sabor Hydroelectric Exploitation, promoted by EDP Production, and whose implementation is the responsibility of Baixo Sabor, ACE- ODEBRECHT/ Bento Pedroso Constructions and Lena Engineering. References Alcolea González, J. J. & de Balbín Berhmann, R. 2006. Arte paleolítico al aire libre. El yacimiento rupestre de Siega Verde, Salamanca. Junta de Castilla y León: Salamanca. Aubry, T. 2009. Actualisation des données sur les vestiges d’art paléolithique sur support mobilier de la Vallée du Côa, pp. 382-95 in (T. Aubry, ed.) 200 séculos da história do vale do Côa: incursões na vida quotidiana dos caçadores artistas do Paleolítico. IGESPAR: Lisbon. 179

Prehistoric Art as Prehistoric Culture análise a temas e estilos. In ARKEOS: Perspectivas em diálogo - Proceedings from the XIX International Rock Art Conference IFRAO 2015. García Díez, M. 2013. La expresión gráfica en La Peña de Estebanvela (Segovia) en el contexto de los últimos grupos de cazadores-recolectores europeos, pp. 472515 in (C. Cacho, ed.) Ocupaciones magdalenienses en el interior de la Península Ibérica. La Peña de Estebanvela (Ayllón, Segovia). Junta de Castilla y León. CSIC. García Díez, M. & Aubry, T. 2002. Grafismo mueble en el Valle del Côa (Vila Nova de Foz Côa, Portugal): La estación arqueológica de Fariseu. Zephyrus 55: 157-82. García Díez, M. & González Morales, I. 2003. En torno al llamado ‘arte Esquemático-abstracto’: A propósito de unas fechas de Covalanas (Ramales de la Victoria, Cantabria). Veleia 20: 227-41. Hernánz, A., Ruiz-López, J. F., Gavira-Vallejo, J. M., Martin, S. & Gavrilenko, E. 2010. Raman microscopy of prehistoric rock paintings from the Hoz de Vicente, Minglanilla, Cuenca, Spain. Journal of Raman Spectrometry 41: 1104–09. Jorge, S. O., Jorge, V. O., Almeida, C. A., Sanches, M. J. & Soeiro, M. T. 1981. Gravuras rupestres de Mazouco (Freixo de Espada à Cinta). Arqueologia 3: 3-12. Jorge, S. O., Jorge, V. O., Almeida, C. A., Sanches, M. J. & Soeiro, M. T. 1982. Descoberta de gravuras rupestres em Mazouco, Freixo de Espada à Cinta (Portugal). Zephyrvs 34-35: 65–70. de Lombera Hermida, A. 2011. Caves and people. Archaeological research at the eastern margins of NW Iberia, pp. 111-22 in (A. de Lombera Hermida & R. Fábregas Valcarce, eds) To the West of Spanish Cantabria: the Palaeolithic Settlement of Galicia, BAR 2283. Archaeopress: Oxford. de Lombera Hermida, A. & Fábregas Valcarce, R. (eds) 2011. To the West of Spanish Cantabria: the Palaeolithic Settlement of Galicia. British Archaeological Reports (BAR). 2283. Archaeopress: Oxford. de Lombera Hermida, A. & Fábregas Valcarce, R. (eds) 2013. Cova Eirós. Primeras evidencias de arte rupestre Paleolítico en el Noroeste Peninsular. Andavira Editora SL: Santiago de Compostela. Pereira, T. 2010. A exploração do quartzito na Faixa Atlântica Peninsular do Pleistocénico. Ph.D. Thesis. Universidade do Algarve. Ramil Soneira, J. & Vázquez Varela, J. M. 1983. Primer hallazgo de arte mueble paleolítico en Galicia. Ars Praehistórica 2: 191-93. Rodríguez Álvarez, X. P., de Lombera Hermida, A., Fábregas Valcarce, R. & Lazuén Fernández, T. 2011. The Upper Pleistocene site of Cova Eirós (Triacastela, Lugo, Spain), pp. 123-33 in (A. de Lombera Hermida & R. Fábregas Valcarce, eds) To the West of Spanish Cantabria: the Palaeolithic Settlement of Galicia.

British Archaeological Reports. Archaeopress: Oxford. Sanches, M. J. & Teixeira, J. 2014. O Abrigo do Passadeiro, Palaçoulo (Miranda do Douro). Um caso de estudo de gravuras rupestres dos inícios do Holocénico no Nordeste de Portugal. Portvgalia. Nova Serie, 35: 61-75. Santos, A. T. 2012. Reflexões sobre a arte paleolítica do Côa: a propósito da superação de uma persistente dicotomia conceptual. Trabalhos de Arqueologia 54: 39–67. Teira Brión, A., Martín Seijo, M., de Lombera Hermida, A., Fábregas Valcarce, R. & Rodríguez Álvarez, X. P.. 2012. Forest resources management during Roman and Medieval cave occupations in the Northwest of the Iberian Peninsula: Cova do Xato and Cova Eirós (Galicia, Spain). Wood and charcoal. Evidence for human and natural history. Sagvntvm extra-13: 15966. Teixeira, J. C., Valdez, J. & Sanches, M. J. 2010. O Abrigo da Foz do rio Tua - Alijó (Trás-os-Montes, Portugal): Identificação e Estudo Preliminar. In I Mesa Redonda: Artes Rupestres da Pré-história e da Proto-história. Valdez-Tullet, J. 2013. O Abrigo Rupestre de Foz Tua: a Ampla Diacronia de um Espaço Significante, pp. 355-66 in (J. Sastre Blanco, R. Catalán Ramos & P. Fuentes Melgar, eds) Arqueología en el Valle del Duero. Del Neolítico a la Antiguedad Tardía: nuevas perspectivas. Actas de las primeras jornadas de jóvenes investigadores en el valle del Duero. La Ergastula. Vidal Romaní, J. R., Sanjurjo Sánchez, J., Grandal D´Anglade, A., Vaqueiro Rodríguez, M. & Fernández Mosquera, D. 2010. Geocaracterización de yacimientos arqueológicos en medio sedimentario: cronología absoluta y relativa, pp. 7-19 in (A. J. López Díaz & E. Ramil Rego, eds) Arqueoloxía: Ciencia e Restauración. Museo de Prehistoria e Arqueoloxía de Vilalba: Vilalba, Monografía 4. Viñas R., Rubio, A. & Ruiz, J. F. 2010. La técnica paleolítica del trazo fino y estriado entre los orígenes del estilo levantino de la Península Ibérica, in (J. Clottes, ed.) Evidencias para una reflexión. Congreso Mundial de Arte Pleistoceno, IFRAO, Tarascon-surAriège. Zilhão, J. & Fábregas Valcarce, R. 1996. Os Gravados de Foz Côa e a arte paleolítica do Noroeste, pp. 147-63 in (R. Fábregas Valcarce, ed.) Os primeiros poboadores de Galicia: O Paleolítico. Edicións do Castro: Sada, 73.

180