Technical Systems of Lithic Production in the Lower and Middle Pleistocene of the Iberian Peninsula: Technological variability between north-eastern sites and Sierra de Atapuerca sites 9781841713922, 9781407327501

Table of contents :
CONTENTS
LIST OF FIGURES
LIST OF TABLES
PREFACE
Acknowledgements
THEORETICAL ISSUES AND METHODOLOGY
PROBLEMS AND OBJECTIVES
CONCEPTUAL APPROACHES TO LITHIC TECHNOLOGY
METHODOLOGY
MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF
INTRODUCTION
PUIG D’EN ROCA EXCAVACIÓ (PREX)
CAN GARRIGA
CAU DEL DUC DE TORROELLA DE MONTGRI
NERETS
CLOT DEL BALLESTER
VINYETS
THE ATAPUERCA SITES
INTRODUCTION
LITHIC INDUSTRY OF THE LOWER LEVELS OF GRAN DOLINA
LITHIC INDUSTRY OF THE LEVEL TD10 OF GRAN DOLINA
LITHIC INDUSTRY OF THE LEVEL TD10.1.A OF GRAN DOLINA
RESULTS AND CONCLUSIONS
RESULTS
CONCLUSIONS
THE CONTEXT: LOWER AND MIDDLE PLEISTOCENE OF EUROPE
THE TOPIC OF THE FIRST SETTLEMENT OF EUROPE
THE MIDDLE PLEISTOCENE OF THE IBERIAN PENINSULA
REFERENCES
1323 verso.pdf
British Archaeological Reports
Front Cover
Title Page
Copyright
Dedication
Table of Contents
LIST OF FIGURES
LIST OF TABLES
PREFACE
Acknowledgements
1. THEORETICAL ISSUES AND METHODOLOGY
2. MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA
3. THE ATAPUERCA SITES
4. RESULTS AND CONCLUSIONS
5. THE CONTEXT: LOWER AND MIDDLE PLEISTOCENE OF EUROPE
REFERENCES

Citation preview

BAR S1323 2004  RODRÍGUEZ  LITHIC PRODUCTION IN PLEISTOCENE IBERIA

Technical Systems of Lithic Production in the Lower and Middle Pleistocene of the Iberian Peninsula Technological variability between northeastern sites and Sierra de Atapuerca sites

Xosé Pedro Rodríguez

BAR International Series 1323 B A R

2004

Technical Systems of Lithic Production in the Lower and Middle Pleistocene of the Iberian Peninsula

Technical Systems of Lithic Production in the Lower and Middle Pleistocene of the Iberian Peninsula Technological variability between northeastern sites and Sierra de Atapuerca sites

Xosé Pedro Rodríguez

BAR International Series 1323 2004

ISBN 9781841713922 paperback ISBN 9781407327501 e-format DOI https://doi.org/10.30861/9781841713922 A catalogue record for this book is available from the British Library

BAR

PUBLISHING

To Teolindo, Luz and Marina

CONTENTS

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XI Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XII

1.- THEORETICAL ISSUES AND METHODOLOGY 1.1.- PROBLEMS AND OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2.- CONCEPTUAL APPROACHES TO LITHIC TECHNOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Typology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimentation and lithic technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analytical typology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Logical Analytical System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Current tendencies in the study of chaînes operatoires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cognitive archaeology and lithic technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Our theoretical framework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Artifacts and attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Logical Analytical System and its structural categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Basic concepts of the Logical Analytical System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.- METHODOLOGY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Specific Analysis of lithic remains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Natural Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . First Generation Negative Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Positive Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Second Generation Negative Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fragments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Size of the objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Morphopotential Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Morphofunctional Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identification of the Technical Operational Themes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Morphogenetic Matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5 5 5 6 6 6 6 8 8 8 9

11 11 11 11 12 12 12 12 14 14 14 15

2.- MIDDLE PLEISTOCENE SITES OF THE NORTHEAST OF THE IBERIAN PENINSULA 2.1.- INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.2.- PUIG D’EN ROCA EXCAVACIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Archaeological findings at Puig d’en Roca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Geological and stratigraphical context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fluvial Terraces of the Ter river . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chronology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Stratigraphic Context of Puig d’en Roca Excavació . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lithic industry of Puig d’en Roca Excavació (PREX) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Morphotechnical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for producing Positive Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

I

23 23 23 24 24 25 25 26 26 26 32 34 36

2.3.- CAN GARRIGA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Location and archeological excavations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Stratigraphy and Chronology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Faunal remains and space organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .40 Analysis of the lithic industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 Morphotechnical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 Strategies for producing Positve Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48 2.4.- CAU DEL DUC DE TORROELLA DE MONTGRI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49 Location and geological context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49 History of investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49 Stratigraphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50 Fauna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .50 Others sites of the Montgrí’s massif . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51 Dating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51 Analysis of the lithic industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51 Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .52 Morphotechnical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53 Strategies for producing Positive Bases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .58 The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .60 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .62

2.5.- NERETS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63 Location and geological context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63 Discovery and archeological interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63 Stratigraphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .64 Analysis of the lithic industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .64 Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 Morphotechnical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65 Strategies for producing Positve Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .68 The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .71 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .74 2.6.- CLOT DEL BALLESTER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77 Location and discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77 Analysis of the lithic industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77 Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .77 Morphotechnical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78 Strategies for producing Positive Bases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .79 The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .79 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .81 2.7.- VINYETS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83 Location and archeological interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83 Geological context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83 Stratigraphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .84 Analysis of the lithic industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .84 Raw material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .84 Morphotechnical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 Strategies for producing Positive Bases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .87 The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .88 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .88

II

3.- THE ATAPUERCA SITES 3.1.- GENERAL INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Geological context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Archaeological sites in the Sierra de Atapuerca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Discovery and first archeological interventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 The Galería Complex. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Stratigraphy and chronology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Archaeological remains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Stratigraphy and paleoclimatology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Chronology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Faunal remains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Human remains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 3.2.- LITHIC INDUSTRY OF THE LOWER LEVELS OF GRAN DOLINA (TD4, TD5, TD6 AND TD7) . . . . . .111 The lithic industry of the levels TD4, TD5 and TD7. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .111 Analysis of the Lithic industry of the Aurora Stratum (level TD6) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Morphotechnical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Strategies for producing Positive Bases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 The Configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 3.3.- ANALYSIS OF THE LITHIC INDUSTRY OF THE LEVEL TD10 OF GRAN DOLINA . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Morphotechnical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Natural Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . First Generation Negative Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Positive Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Second Generation Negative Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fragments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for producing Positive Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Indirect Technical Operational Themes (ITOT) in TD10A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Indirect Technical Operational Themes (ITOT) in TD10B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Indirect Technical Operational Themes (ITOT) in TD10C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

123 123 123 124 124 124 126 128 130 130 130 130 131 132 133

3.4.- ANALYSIS OF THE LITHIC INDUSTRY OF TD10.1.a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Morphotechnical Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Natural Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . First Generation Negative Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Positive Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Second Generation Negative Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fragments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies for producing Positive Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The configuration of tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

135 135 135 135 135 135 137 138 139 139 140 141

III

4.- RESULTS AND CONCLUSIONS 4.1.- RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .145 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .145 Utilization of raw materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .146 The natural Bases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .149 The processes of exploitation for producing Positive Bases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .150 The processes of configuration for shaping tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .151 4.2.- CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155 Variability of the Conceptual scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .155 Thesis about the Conceptual scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .157 Variability of the Operational scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .159 Thesis about the Operational scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .161 Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .161

5.- THE CONTEXT: LOWER AND MIDDLE PLEISTOCENE OF EUROPE 5.1.- THE TOPIC OF THE FIRST SETTLEMENT OF EUROPE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .165 5.2.- THE MIDDLE PLEISTOCENE OF THE IBERIAN PENINSULA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .169 The northeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .169 The east and the south . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .171 The interior of the Iberian Peninsula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .172 The North and Northwest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .173 Some reflections about the european middle Pleistocene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .173

REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .175

IV

LIST OF FIGURES

1.- THEORETICAL ISSUES AND METHODOLOGY 1.1.- PROBLEMS AND OBJECTIVES 1.2.- CONCEPTUAL APPROACHES TO LITHIC TECHNOLOGY Figure 1.2.1- Structural categories for the study of lithic remains, following the Logical Analytical System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.3.- METHODOLOGY Figure 1.3.1.- Planes for the description of the objects volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.2.- Extent of the retouched edge of the 1GNB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.3.- Oblicuity (angle of the edge with removals) of the 1GNB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.4.- Extent of removals of the 1GNB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.5.- Frontal edge Morphology and Sagittal edge of the 1GNB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.6.- Some of the attributes for the analysis of 2GNBC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.7.- Geometric models for the morphopotential analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.8.- Some examples of morphopotential analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.9.- Some criteria for the description of Technical Operational Themes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 1.3.10.- Example of a morphogenetic matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12 13 13 13 13 15 16 16 16 17

2.- MIDDLE PLEISTOCENE SITES OF THE NORTHEAST OF THE IBERIAN PENINSULA 2.1.- INTRODUCTION Figure 2.1.1.- Map of the Northeast of the Iberian Peninsula with the situation of the studied sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2.2.- PUIG D’EN ROCA EXCAVACIÓ Figure 2.2.1.- Location of the Puig d’en Roca sites, near the city of Girona . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.2.2.- Excavation of PREX site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.2.3.- Stratigraphic diagram of Puig d’en Roca Excavació . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.2.4.- First Generation Negative Bases of Configuration (1GNBC) found in PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.2.5.- First Generation Negative Bases of Exploitation (cores) found in PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.2.6.- Positive Bases of PREX. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.2.7-. Second Generation Negative Bases of Configuration (2GNBC) found in PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.2.8.- Morphogenetic matrix of the lithic industry of PREX: Indirect Technical Operational Themes (ITOT). . . . . . . . . . . . . . . . . . . . . . Figure 2.2.9.- Morphogenetic matrix of the lithic industry of PREX: Direct Technical Operational Themes (DTOT) . . . . . . . . . . . . . . . . . . . . . .

23 23 25 27 28 30 31 34 35

2.3.- CAN GARRIGA Figure 2.3.1.- Location of Can Garriga site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.2.- View of the Can Garriga site, during the excavation of 1986 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.3.- Plan of the 1991 excavation at Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.4.- Stratigraphy and dates of Can Garriga sequence, indicating the archaeological levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figura 2.3.5.- Distribution of archaeological remains in a sector of the level 1 of Can Garriga. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.6.- First Generation Negative Bases of Configuration (1GNBC) found in level 1 of Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.7.- First Generation Negative Bases of Exploitation (cores) found in Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.8.- Positive Bases of Can Garriga. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.9.- Second Generation Negative Basees of Configuration (2GNBC) found in level 1 of Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.3.10.- Morphogenetic matrix of the lithic industry of Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

39 39 40 40 40 42 42 44 45 46

2.4.- CAU DEL DUC DE TORROELLA DE MONTGRÍ Figure 2.4.1.- Location of Cau del Duc de Torroella de Montgrí site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.2.- View of the Cau del Duc de Torroella de Montgrí site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.3.- First Generation Negative Bases of Configuration (1GNBC) found in CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.4.- First Generation Negative Bases of Configuration (1GNBC) found in CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.5.- First Generation Negative Bases of Configuration (1GNBC) found in CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.6.- First Generation Negative Base of Exploitation (core) found in CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.7.- First Generation Negative Bases of Exploitation (cores) found in CDTM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.8.- Positive Bases of CDTM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.9.- Second Generation Negative Bases of Configuration, from CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.10.- Second Generation Negative Bases of Configuration, from CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.4.11.- Morphogenetic matrix of the lithic industry of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

49 50 53 54 54 54 55 56 57 57 61

2.5.- NERETS Figure 2.5.1.- Location of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.5.2.- General view of the Nerets hill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.5.3.- Topographic map of the Nerets hill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.5.4.- Schematic stratigraphic column of the excavated zone in Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.5.5.- Positive Bases of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.5.6.- First Generation Negative Bases of Exploitation from Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

V

63 64 64 65 67 70

Figure 2.5.7.- First Generation Negative Bases of Configuration (1GNBC) found in Nerets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Figure 2.5.8.- Second Generation Negative Bases of Configuration, from Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Figure 2.5.9.- Morphogenetic matrix of the lithic industry of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

2.6.- CLOT DEL BALLESTER Figure 2.6.1.- Location of Clot del Ballester, near de city of Lleida . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Figure 2.6.2.- Lithic industry of Clot del Ballester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 Figure 2.6.3.- Morphogenetic matrix of the lithic industry of Clot del Ballester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

2.7.- VINYETS Figure 2.7.1.- Location of Vinyets site, near de city of Tarragona . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.7.2.- General view of Vinyets site, during the excavation of 1991 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.7.3.- Schematic stratigraphic column of Vinyets sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.7.4.- First Generation Negative Bases of Configuration (1GNBC) found in Vinyets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.7.5.- Positive Bases of Vinyets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.7.6.- Second Generation Negative Bases of Configuration, from Vinyets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 2.7.7.- Morphogenetic matrix of the lithic industry of Vinyets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

83 83 84 86 86 87 89

3.- THE ATAPUERCA SITES 3.1.- GENERAL INTRODUCTION Figure 3.1.1.- Location of Atapuerca Sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Figure 3.1.2.- The Sierra de Atapuerca, panomaric view from the South . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 Figure 3.1.3.- Location of the sites of Cueva Mayor karstic system , and sites of the abandoned railway trench. . . . . . . . . . . . . . . . . . . . . . . . . . . 94 Figure 3.1.4.- The sites of Gran Dolina and Galería Complex, located in the Trinchera del Ferrocarril . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Figure 3.1.5.- Schematic plan of Gran Dolina comprising the excavated zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Figure 3.1.6.- General view of the biostratigraphic test pit of Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 Figure 3.1.7.- Excavation of the upper part of the level TD10 of Gran Dolina, during 2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 Figure 3.1.8.- Galería Complex, to the beginnings of the decade of 1980, just after beginning the systematic excavations . . . . . . . . . . . . . . . . . 101 Figure 3.1.9.- Stratigraphic cut of Galería, with radiometric ages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Figure 3.1.10.- Lithic industry from Galería (Atapuerca) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Figure 3.1.11.- Lithic industry from Galería (Atapuerca) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Figure 3.1.12.- Lithic industry from Galería (Atapuerca) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Figure 3.1.13.- Lithic industry from Galería (Atapuerca) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Figure 3.1.14.- Gran Dolina stratigraphic section, with the lithostratigraphic levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 Figure 3.1.15.- Stratigraphic column of Gran Dolina, with radiometric and paleomagnetic ages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Figure 3.1.16.- Cranial remains of Homo antecessor, found in the level TD6 of Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .110

3.2.- LITHIC INDUSTRY OF THE LOWER LEVELS OF GRAN DOLINA (TD4, TD5, TD6 AND TD7) Figure 3.2.1.- Lithic industry from the level TD4 of Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .111 Figure 3.2.2.- 1GNB of Exploitation (core) of quartzite from level TD5 of Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .111 Figure 3.2.3.- First Generation Negative Bases from TD6 (“Aurora stratum”, Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113 Figure 3.2.4.- Positive Bases from level TD6 (“Aurora stratum”) of Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115 Figure 3.2.5.- Second Generation Negative Bases of Configuration (2GNBC) from TD6 (Aurora stratum, Gran Dolina) . . . . . . . . . . . . . . . . . . .116 Figure 3.2.6.- 2GNB of Exploitation from TD6 (Aurora stratum, Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .117 Figure 3.2.7.- Morphogenetic matrix of the lithic industry of TD6 (Aurora stratum, Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .118 Figure 3.2.8.- Strategies for producing Positive Bases. 1: Trifacial orthogonal knapping. 2: Longitudinal masive knapping . . . . . . . . . . . . . . . . 119

3.3.- ANALYSIS OF THE LITHIC INDUSTRY OF THE LEVEL TD10 OF GRAN DOLINA Figure 3.3.1.- 1GNBC of quartzite from level TD10C (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.3.2.- 1GNB of Exploitation form TD10 (Gran Dolina). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.3.3.- Postive Bases from level TD10 (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.3.4.- 2GNB of Configuration from TD10 (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.3.5.- Morphogenetic matrix of the lithic industry of TD10A (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.3.6.- Morphogenetic matrix of the lithic industry of TD10B (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.3.7.- Morphogenetic matrix of the lithic industry of TD10C (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

125 126 127 129 131 131 131

3.4.- ANALYSIS OF THE LITHIC INDUSTRY OF TD10.1.a Figure 3.4.1.- 1GNB of Exploitation from level TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.4.2.- Positive Bases from TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.4.3.- 2GNB of Configuration from TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 3.4.4.- Morphogenetic matrix of the lithic industry of TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

136 137 139 140

4.- RESULTS AND CONCLUSIONS 4.1.- RESULTS Figure 4.1.1.- Structural categories of the lithic assemblages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4.1.2.- Raw materials utilized in the lithic assemblages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4.1.3.- Correspondence Analysis of raw materials and lithic assemblages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4.1.4.- Natural Bases, objects related with processes of exploitation, and of configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

VI

146 147 148 150

Figure 4.1.5.- Correspondence Analysis of the butt facets, and the studied sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 Figure 4.1.6.- Morphodynamic capacity of the retouched tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Figure 4.1.7.- Correspondence Analysis between morphopotential models configured in the cutting edges of the instruments, and the lithic assemblages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154

4.2.- CONCLUSIONS Figure 4.2.1.- Configuration of tools (1GNBC and 2GNBC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4.2.2.- Comparison between the percentage of instruments configured on pebbles (1GNBC), and the percentage of handaxes and cleavers, relating to all the configured instruments (1GNBC and 2GNBC) . . . . . . . . . . . . . . . . . . . . . . . . . Figure 4.2.3.- Comparison between the proportion of cores with predetermination of the morphology of the final products; trifacial and multifacial cores; facets of the butts of the PB and 2GNB; and 1GNBC, relating to all the configured instruments . . . . . . . . . . . . Figure 4.2.4.- Criteria that can help to identify the Operational variability of the lithic industry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

156 157 158 159

5.- THE CONTEXT: LOWER AND MIDDLE PLEISTOCENE OF EUROPE 5.1.- THE TOPIC OF THE FIRST SETTLEMENT OF EUROPE Figure 5.1.1.- Location of some of the most interesting Lower and Middle Pleistocene sites in Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

VII

LIST OF TABLES

1.- THEORETICAL ISSUES AND METHODOLOGY 1.1.- PROBLEMS AND OBJECTIVES 1.2.- CONCEPTUAL APPROACHES TO LITHIC TECHNOLOGY 1.3.- METHODOLOGY Table 1.3.1.- Compared terminology between the Logical Analytical System and the most common terms in Anglophone archaeological literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 1.3.2.- Criteria for the analysis of First Generation Negative Bases (1GNB) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 1.3.3.- Criteria for the analysis of Positive Bases (PB) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 1.3.4.- Criteria for the analysis of Second Generation Negative Bases of Configuration (2GNBC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11 13 14 15

2.- MIDDLE PLEISTOCENE SITES OF THE NORTHEAST OF THE IBERIAN PENINSULA 2.1.- INTRODUCTION 2.2.- PUIG D’EN ROCA EXCAVACIÓ Table 2.2.1.- Fluvial terraces of the Ter river, near the city of Girona . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.2.- Chronological framework of the lithic industry of Puig d’en Roca . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.3.- Lithic industry recovered in Puig d’en Roca Excavació (PREX). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.4.- Average size of the First Generation Negative Bases of PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.5.- Average size e of the Positive Bases of PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.6.- Average size of the 2nd Generation Negative Bases of PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.7.- Average size of the Fragments of PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.8.- Negative Bases of Exploitation (NBE) found in Puig d’en Roca Excavació, grouped in terms of the knapping method (Indirect Operational Themes) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.9.- Retouched artifacts of PREX (1GNBC and 2GNBC). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.2.10.- Morphodynamic capacity of the retouched tools of PREX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

24 25 26 27 29 30 32 33 35 36

2.3.- CAN GARRIGA Table 2.3.1.- Structural categories of the lithic industry of Can Garriga (levels 1 and 2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.3.2.- Raw materials of levels 1 and 2 of Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.3.3.- Positive Bases and raw materials of Can Garriga. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.3.4.- Average size of the Positive Bases of Can Garriga. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.3.5.- Average size of the Second Generation Negative Bases of Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.3.6.- Fragments and raw materials of levels 1 and 2 of Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.3.7.- Average size of Fragments of Can Garriga . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.3.8.- Morphodynamic capacity of the retouched tools of Can Garriga. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

41 41 43 43 44 45 45 47

2.4.- CAU DEL DUC DE TORROELLA DE MONTGRÍ Table 2.4.1.- Faunal remains (number of bones) represented in the Paleolithic deposits of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.2.- Lithic materials from CDTM (without fragments and indeterminables), in terms of the Museum where they are preserved, and who has studied them . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.3.- Lithic industry of Cau del Duc de Torroella de Montgrí (CDTM) studied in this work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.4.- Lithic industry of Cau del Duc de Torroella de Montgrí (CDTM) studied in this work, without fragmets neither indeterminable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.5.- Average size of the 1GNB of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.6.- Average size of the Positive Bases of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.7.- Raw materials of the Second Generation Negative Bases of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.8.- Average size of the 2GNB of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.9.- Average size of the Fragments of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.10.- First Generation Negative Bases of Exploitation (1GNBE) of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.11.- Structural categories and raw materials of the configured (or retouched) instruments of CDTM . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.4.12.- Morphodynamic capacity of the retouched tools of CDTM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51 52 52 52 53 55 57 57 58 59 61 62

2.5.- NERETS Table 2.5.1.- Lithic industry of Nerets (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.5.2.- Average size of the 1GNB of Nerets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.5.3.- Average size of the PB of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.5.4.- Average size of the 2GNB of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.5.5.- Average size of the Fragments of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.5.6.- Negative Bases of Exploitation (1GNBE and 2GNBE) of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.5.7.- Morphodynamic capacity of the retouched tools of Nerets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

IX

65 66 66 68 68 69 72

2.6.- CLOT DEL BALLESTER Table 2.6.1.- Lithic industry of Clot del Ballester (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.6.2.- Average size of the 1GNB of Clot del Ballester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.6.3.- Average size of the Positive Bases of Clot del Ballester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.6.4.- Average size of the 2GNB of Clot del Ballester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.6.5.- Average size of the Fragments of Clot del Ballester. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.6.6.- Morphodynamic capacity of the retouched tools of Clot del Ballester . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

78 78 79 79 79 80

2.7.- VINYETS Table 2.7.1.- Lithic industry of all the archaeological levels of Vinyets (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.7.2.- Lithic industry of archaeological levels 1, 2 and 3, and lithic remains without stratigraphic context, from Vinyets . . . . . . . . . . . . . Table 2.7.3.- Average size of the Positive Bases of Vinyets (for each level) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.7.4.- Average size of the Second Generation Negative Bases of Vinyets (for each level). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.7.5.- Average size of Fragments of Vinyets (for each level) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 2.7.6.- Morphodynamic capacity of the retouched tools of Vinyets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

85 85 86 87 87 88

3.- THE ATAPUERCA SITES 3.1.- GENERAL INTRODUCTION Table 3.1.1.- Campaings of excavation in the sites of the Sierra de Atapuerca, from 1976 to 2004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Table 3.1.2.- Main archaeological sites in the Sierra de Atapuerca (with chronology and archaeological remains recovered) . . . . . . . . . . . . . . . . 96 Table 3.1.3.- Radiometric dates for the stratigraphic untis of Galeria Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Table 3.1.4.- Macromammals found in Galería . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Table 3.1.5.- Lithic industry found in Trinchera Galería (TG) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Table 3.1.6.- Macrofauna of Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

3.2.- LITHIC INDUSTRY OF THE LOWER LEVELS OF GRAN DOLINA (TD4, TD5, TD6 AND TD7) Table 3.2.1.- Lithic industry found in the “Aurora stratum” of the level TD6 of Gran Dolina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .112 Table 3.2.2.- Average size of natural Bases from TD6 (“Aurora stratum”) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113 Table 3.2.3.- Average size of 2GNB from TD6 (“Aurora stratum”) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115 Table 3.2.4.- Morphodynamic capacity of the retouched tools of TD6 (Aurora Stratum, Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .118

3.3.- ANALYSIS OF THE LITHIC INDUSTRY OF THE LEVEL TD10 OF GRAN DOLINA Table 3.3.1.- Lithic industry of the level TD10 of Gran Dolina (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.2.- Lithic industry of the sub-level TD10A of Gran Dolina (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.3.- Lithic industry of the sub-level TD10B of Gran Dolina (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.4.- Lithic industry of the sub-level TD10C of Gran Dolina (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.5.- Average size of the natural Bases of TD10 (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.6.- Average size of the 1GNB of TD10 (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.7.- Average size of the Positive Bases of TD10 (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.8.- Average size of the 2GNB of TD10 (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.3.9.- Morphodynamic capacity of the retouched tools of TD10 (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

123 124 124 124 124 125 127 128 133

3.4.- ANALYSIS OF THE LITHIC INDUSTRY OF TD10.1.a Table 3.4.1.- Lithic industry of the level TD10.1.a of Gran Dolina (structural categories and raw materials) . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.4.2.- Average size of the natural Bases of TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.4.3.- Average size of the 1GNB of TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.4.4.- Average size of the Positive Bases of TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.4.5.- Average size of the 2GNB of TD10.1.a (Gran Dolina) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 3.4.6.- Morphodynamic capacity of the retouched tools of TD10.1.a (Gran Dolina). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

135 136 136 137 138 141

4.- RESULTS AND CONCLUSIONS 4.1.- RESULTS Table 4.1.1.- Lithic assemblages studied in this work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 4.1.2.- Structural categories of the lithic assemblages (without fragments neither indeterminables) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 4.1.3.- Raw materials utilized in the lithic assemblages studied in this work (without fragments, but with the indeterminable pieces) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 4.1.4.- Natural Bases, objects related with processes of exploitation, and of configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 4.1.5.- Principal Indirect Technical Operational Themes identified through the study of the cores found in the sites . . . . . . . . . . . . . . . . . Table 4.1.6.- Facets of the butts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 4.1.7.- Configured or retouched tools (1GNBC and 2GNBC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Table 4.1.8.- Morphodynamic capacity of the retouched tools of the studied sites. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

145 145 147 149 151 151 152 153

4.2.- CONCLUSIONS Table 4.2.1.- Technical Mode and type of occupation of the sites studied in this work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

X

PREFACE This book studies Prehistory from the perspective of lithic technology. Its primary objective is to describe the characteristics of the lithic technology at sites in the Northeast of the Iberian Peninsula, and Atapuerca (Burgos). The chronological dimension is extensive, since we study deposits that range from the end of the Lower Pleistocene to the end of the Middle Pleistocene. The problems of our topic, and our objectives and hypotheses make up the first chapter of the book. In the second chapter, we describe the methodology used to study the lithic technology of the sites selected. We also include a brief review of the current state of research into lithic technology. First we present the results of our analysis at each of the sites studied. The structure of all the sections is very similar. After introducing each site (stating its location, the history of research there and, when necessary, its stratigraphy), we go on to describe the data derived from our studies of the lithic record. To be more precise, section 2.1 deals with the lithic record from the Puig d’en Roca Excavació site (Girona); section 2.2 studies material from Can Garriga (Girona); section 2.3 deals with Cau del Duc de Torroella de Montgrí (Girona); section 2.4 examines the lithic industry from Nerets; and section 2.5 is dedicated to the Clot del Ballester site (Lleida). In the final part (chapter 2.6) we analyse the lithic record from Vinyets (Tarragona). Chapter 3 describes the Sierra de Atapuerca sites, their location, geological context, research history and salient features. We pay particular attention to the deposits in the Galería and provide a summary of the lithic industry. Section 3.1 deals with the Gran Dolina site. It begins by presenting the site and goes on to make a detailed survey of the lithic industry of levels TD6 (section 3.2), TD10 (section 3.3), and TD10.1.a (section 3.4) (including stratigraphy and dating). In chapter 4 we describe the results of the work. First we summarize the essential characteristics of the deposits and make comparisons between them. Subsequently we draw our final conclusions (chapter 5). We have included a chapter on the contextualization of the sites examined, in the frame of the Lower and Middle Pleistocene of the Iberian Peninsula and the rest of Europe (chapter 6). It should be noted that this book is an updated version of my Doctoral Thesis, concluded in 1997 at the Universitat Rovira i Virgili (Tarragona, Spain).

XI

Acknowledgements

Fieldwork in Atapuerca was funded by Junta de Castilla y León, and research was funded by Dirección General de Investigación Científica (BOS2003-08938-C03-03). I’m grateful to the Associació Cultural de La Femosa (Artesa de Lleida), and Centre d’Investigacions Arqueològiques de Girona, for their help. In the book we present data that has been obtained as a result of the work of the many people belonging to the Atapuerca research team. In particular, I would like to mention Eudald Carbonell. I would also like to point out that this book could not have been concluded without the support of Marina Lozano.

XII

1 THEORETICAL ISSUES AND METHODOLOGY

THEORETICAL ISSUES AND METHODOLOGY

1.1.- PROBLEMS AND OBJECTIVES

Problems

In the Sierra de Atapuerca sites, however, we find collections of lithic artifacts with datings, stratigraphic context, associated fauna and human fossils, and, therefore, they provide excellent conditions for technological analysis. We decided to concentrate our efforts on two important sites, which we believed would provide us with the information we were looking for: Galería and Gran Dolina. The first one is a deposit that has already been fully excavated, while there is still plenty of work to do in Gran Dolina. The features of Gran Dolina are particularly interesting for our purpose largely because there is a deposit of considerable stratigraphic thickness (18 meters), with various fertile levels from the archaeo-paleontologic point of view. These levels go from the end of the Lower Pleistocene to the end of the Middle Pleistocene. We have opted to synthesize the fundamental features of Galería and we have used level TG11 as a reference. The record from Galería belongs to Mode 2 (Acheulian) and fills in the space left by Dolina for the central Middle Pleistocene.

The chronological period studied here spans the end of the Lower Pleistocene and the Middle Pleistocene. It is a very extensive chronological range, which raises issues of great interest. We will focus on two of them: when and how did the first settlement of Europe take place?; and how did lithic technology evolve during the Middle Pleistocene. At the beginning of the 1990s two models explained how Europe was settled. The defenders of “young chronology” thought that the settlement of Europe took place less than 500,000 years ago (Gamble, 1993; 1994; Roebroeks et al., 1992; Roebroeks, 1994; Roebroeks and van Kolfschoten, 1994; Roebroeks and van Kolfschoten, 1995). However, the findings at Fuente Nueva 3 (Orce, Granada, Spain (Gibert et al., 1998b; Martínez-Navarro et al., 1997; Oms et al., 2000b), Ceprano (Italy) (Ascenzi et al., 1996; Manzi et al. 2001), and above all Gran Dolina level 6 (Carbonell et al., 1995a; Carbonell et al., 1999b), have refuted the arguments of the advocates of Young Europe. The new paradigm considers that the European settlement took place at least one million years ago (Carbonell et al., 1996).

The herogeneous nature of the sites studied here makes it possible to study considerable technological variability. This variability is due to various factors. In any technological assemblage, some components belong to the basics of the Technical System of Lithic Production: the conceptual scheme (Geneste, 1991). The conceptual scheme varies little, because it is determined by the technological tradition. This subjacent scheme determines the general guidelines of tool production. But external factors influence the implementation of these psycho-technical models. An operational scheme is the result of implementing a conceptual scheme that is influenced by these external factors (input). The result is the artifact (output). In this way, final production can be polymorphous, although its conceptual substratum is unique. The complexity of an operational sequence for the production of lithic tools (“chaîne opératoire”) lies essentially in the interaction between the conceptual and the operational schemes (Roche and Texier, 1991).

The evolution (if we can talk about evolution) of lithic technology during the Middle Pleistocene has also caused some interesting debates. The Acheulian (Mode 2 technology) in Europe may have been the consequence of the local evolution of previous industries (pre-Acheulian, Mode 1 technology), or it may have arrived in Europe with a migratory wave from Africa. The truth is that this system of producing tools burst into Europe some 500 Ka ago, although other systems without the Acheulian Operational standards did not disappear. Above all other considerations, we focus on the comparative analysis of technological variability among lithic records from different geographic regions, with a synchronic and diachronic perspective.

Now that we have distinguished these two major domains, we can also distinguish different levels of variability. Structural variability, for example, affects the conceptual scheme, the very concepts of the Technical System. According to Geneste (Geneste, 1991), this kind of variability operates under the influence of the social community, and belongs to the cultural sphere. Therefore, it is impregnated in the Technical System under the weight of tradition. The fundamental choice of the volumetric structures, which presides the conception of lithic production, belongs to this order.

For this work we have selected sites in the Northeast of the Iberian Peninsula, and in the Sierra de Atapuerca (near the city of Burgos, in the centre of the Iberian Peninsula). The northeastern sites have important drawbacks for this kind of research. There are almost never any datings, and the lithic record has no other kind of associated remains, and at times are surface deposits with no stratigraphic context. We are fully aware of these limitations but, unfortunately, in the Iberian Peninsula there are very few sites that do not have these problems. On the Mediterranean coastline, in particular, it is very difficult to find sites with good stratigraphic contexts, the possibility of datings, and associated fauna.

There is a second type of variability. Some variations affect the performance and implementation of Technical Systems at the same level as the operational scheme. 3

Xosé Pedro Rodríguez These variations may be due to technical, environmental and socioeconomic factors, one of which may be the availability of raw material. The individual’s technical capability, in combination with the restrictions of raw material and the immediate need for instruments can influence the final result.

Objectives The general objective of this work is to establish the extent of technological variability among the lithic industries from the sites in the northeast of the Iberian Peninsula, and those from Atapuerca. This objective can be formulated as a series of questions: Is there synchronic and/or diachronic variability among the two groups of sites studied? Should the answer be yes, does this variability affect the conceptual scheme or the operational scheme? If the variability does affect the operational scheme, what factors influence the existence of variability? The answers to these questions will depend on information that we obtain from analysing the lithic industry. So our first step was to acquire this information, and to do so we used the Logical Analytic System. In the following chapter below we will explain our theoretical proposal and our methodology. After establishing the existence (or absence) of variability among the studied sites, we will place our data in the general context of the Lower and Middle Pleistocene of the Iberian Peninsula, and of the rest of Europe.

4

THEORETICAL ISSUES AND METHODOLOGY

1.2.- CONCEPTUAL APPROACHES TO LITHIC TECHNOLOGY

The most lasting remains of the earlier human activities are lithic objects. For this reason, Prehistory has largely been reconstructed by taking information from stone implements. Fortunately, lithic remains are sometimes found in association with other kinds of remains (essentially organic), and our discipline takes into account research fields other than only that of lithic objects. All of these issues contribute to a better understanding of the past. In our work we have tried to consider, when possible, other kinds of information, above and beyond that derived from the lithic industry. However, this study is fundamentally one of lithic technology, and this is where our focus lies. We wish to provide only a general synthesis, describing the most significant conceptual approaches to the field.

contribution was the conceptual element that enabled the experimental knowledges of Bordes, Tixier and Crabtree to be integrated. Lithic knapping, practised for a long time without a scientific approach (Johnson, 1978), began to become an efficient instrument of prehistoric research. In 1964, François Bordes insisted on building a typology on a basis of experimental work. From his point of view, technology is subjected to typology. Tixier tried to combine the typological approach with technological experimentation: “La typologie devra dorénavant tenir compte de la biographie technique pour créer ses nouveaux types et les placer ainsi dans un moment d’une chaîne opérative donnée” (Tixier, 1991). Don Crabtree managed the most experimental approach, which focused on technical and very specific questions (Crabtree, 1975).

Typology But what is most interesting for us is André Leroi-Gourhan’s work. He took the concept of chaîne operatoire (operational sequence) from Ethnology and Social Anthropology (Mauss, 1947), and he applied it to the manufacture of prehistoric implements (Leroi-Gourhan, 1964). LeroiGourhan was the first to draw up a conceptualization of technology in Ethnology and Archaeology. At the same time, Ethnology gave the necessary conceptual raw material to describe and define technical systems (Lemonnier, 1976).

Typological classification is usually the first step in the study of lithic objects. From this point of view, not all lithic objects have the same typological value; for example, some have a special value because, per se, they can characterize a “culture”. This is the “master fossil” concept in Prehistory. However, cultures are characterized by the relative proportions of different types of tools (Bordes, 1961). For a long time, the finished lithic tool was most interesting for researchers, and they paid no attention to the production processes: “The morphological classification of lithic assemblages has long led to a simplified division of lithic production into three major stages: the manufacture of bifaces, the production of flakes, and the production of blades. Later classifications were nothing more than variations on this same theme” (Karlin and Julien, 1994).

The application of the chaîne operatoire concept to the lithic industry involved introducing the spatial-temporal order to the analysis of the production processes: “La notion de ‘chaîne operatoire’ est destinée à la compréhension chronologique des divers objects constitutifs d’une industrie. Organiser une chaîne operatoire consiste à définir sur des bases expérimentales les étapes chronologiques de la fabrication d’un outillage. Chaque étape théorique pouvant être carácterisée par un ou une série de produits, de déchets, de débris, et de produits bruts destinées à subir d’autres transformations techniques au cours d’étapes ultériures” (Geneste, 1985).

Morphology is the most important classificatory criteria. In this way, a considerable number of studies enumerate different kinds of lithic objects, according to their morphological characteristics. Some of these criteria have been in common use until today. Two examples are Bordes’ typology of the Lower and Middle Paleolithic (Bordes, 1961), and, in a different context, the typology drawn up by Mary Leakey to classify Olduvai’s tools (Leakey, 1971). According to Bisson, “Bordes’ type definitions are inadequate for use in modern quantitatively and technologically oriented studies of lithics because they are overly subjective and are an uncontrolled mixture of technological and functional variables” (Bisson, 2000). Below we shall explain why we reject Bordes’ typological vision.

In the 1980s the chaîne operatoire concept in lithic technology analysis quickly spread throughout France. The appearance in the mid 1980s of several doctoral theses that use this new methodology marked a new inflection point (Boëda, 1986; Geneste, 1985; Pelegrin, 1986; Perles, 1985). It is no chance that the great majority of these studies were concluded at the same University: Paris X. It is not mere chance that other Paris University (Paris VI) finished another series of doctoral theses about lithic technology at the same time (Carbonell, 1985; Guilbaud, 1985; Sahnouni, 1985). These studies were the direct heirs of the analytical typology of Georges Laplace

Experimentation and lithic technology During the second half of last century, considerable interest in acquiring practical knowledge of the manufacture of lithic tools developed and experimentation was initiated as a source of this knowledge. This was all due to Leroi-Gourhan, Bordes, Tixier and Crabtree. André Leroi-Gourhan’s 5

Xosé Pedro Rodríguez

Analytical typology

The techno-psychological approach aims to determine the knowledge required to manufacture implements; it studies the abstract cognitive operations and the psycho-motor operations that are part of the technical process.

Between the late sixties and the beginning of the seventies, the analytical and structural typology of Georges Laplace arose in response to the empirical typological tradition (Laplace, 1972; 1974). His theoretical approach (more developed than empirical typology) was based on the dialectical character of historical processes, and on the structural conception of implements, under the influence of structuralism. Laplace’s classificatory system hoped to break down the morphotechnical structures of lithic implements into a series of significant attributes, on the basis of which a typology could be established. The subjectivity in classification was reduced and an open typology was created, in opposition to Borde’s closed typology (Bordes, 1961). This approach currently is little used. At present, this approach is only used by a few Italian and Spanish researchers.

Jean-Michel Geneste is a good exponent of the technoeconomical approach in the study of technical systems. In fact, lithic technology studies focusing on the supply and processing of raw materials have been very usual in recent years (Collina-Girard, 1975; Jaubert, 1995; Perlès, 1991; Tavoso, 1984; Turq, 1992). On the other hand, most of Boeda’s work is an example of the techno-psychological approach. Jacques Pelegrin (Pelegrin, 1986), Pierre-Jean Texier (Texier, 1995), and Hélène Roche (Roche, 1980), belong to the same research tradition, although their work is considerably influenced by experimentation (linked to Jacques Tixier). One of the most important researchers of the techno-psychological approach is Michel Guilbaud, whose approach is more analytical and less techno-typological (Guilbaud, 1987; Guilbaud, 1996).

The Logical Analytical System The Logical Analytical System appeared at the beginning of the eighties as a non-typological derivation of Laplace’s approach (Carbonell et al., 1983). It preserved the analytical and structural aim, but eliminated the typological factor. David L. Clarke’s analytical and systemic approach also influenced the theoretical formulation of this new system, which rejected radically empirical typology. From a theoretical point of view, there were three inspirational sources: Georges Laplace’s analytic typology (Laplace, 1972), David L. Clarke’s analytic archaeology (Clarke, 1968), and Thompson’s historical logic (Thompson, 1981). On a practical level, the development of lithic technology, and the experimentation when the Logical Analytic System appeared in France helped give definitive shape to this theoretical and methodological approach.

These kinds of study are cognitive and sometimes introduce social elements. Therefore, we can talk about a techno-sociological approach, which was directly inspired by André Leroi-Gourhan and was related to ethnology: “Or la technologie contient plus que l’étude morphologique des activités materielles, elle aboutit à la définition du comportement technique de l’homme, à une «sociologie des techniques» dans laquelle la préhistoire prend progressivement la position de préface ... (...) Ainsi donc il paraît imposible de disjoindre le passé et le présent, l’homme et ses gestes, la pensée et les actes, le matériel et le spirituel, la technique et le social” (Leroi-Gourhan, 1952). Most of the studies about Pincevent or Etiolles belong to this research tradition (Karlin and Julien, 1994; Pigeot, 1987; Ploux, 1991; Roux, 1991).

First, an analytical method was developed to classify lithic industry, in particular lithic assemblages of pebble tools (Carbonell et al., 1983). Then the system was adapted slightly to avoid the danger of an excessively mechanistic approach that was more quantitative than qualitative (Carbonell et al., 1992b). The new conceptual framework followed the same theoretical line, but it did not stop at merely analysing implements because it wants reach the synthesis. At the same time, efforts were made to simplify formulations of the system, which not only made it easier to apply but also easier to explain.

Although these approaches have given fresh impetus to the study of lithic implements, most research work that studies technical systems and chaînes operatoires still continue to use the traditional, empirical and descriptive typology. However, the majority take into account that preeminence belong clearly to the technological aspect “un objet technique tel que le biface doit donc être défini pas sa genèse et non comme un simple utensile [...]. En prenant en compte que l’aboutissement, ou ce qui est considéré comme la phase terminale des opérations techniques, la typologie est incapable de rendre compte de l’ensemble des connaissances mises en jeu pour aboutir à l’objet” (Boëda, 1991).

Current tendencies in the study of chaînes operatoires At present, stone-tool production is studied from two complementary perspectives: the techno-economical and techno-psychological (Boëda et al., 1990).

Cognitive archaeology and lithic technology Technological complexity can provide clues about cognitive capacities, and is related to the linguistic function and other intellectual functions (Gibson, 1996). Some authors think that there may be a structural similarity between linguistic behaviour and implement manufacture (LeroiGourhan, 1964). According to Wynn, the operational concepts that govern human spatial organisation developed in the late Acheulian. In his opinion, there are big differences

Techno-economy attempts to analyze the technical hominid behaviour from an economical perspective, which consists of studying the sources of raw materials, and paying particular attention to the mobility of human groups (spatial dimension of the chaîne operatoire).

6

THEORETICAL ISSUES AND METHODOLOGY between the “pre-operational” Oldowan structures, and the geometric structures of the Acheulian (the bilateral symmetry of the handaxe, for example) (Wynn, 1979). Wynn explains that the different phases of mental development in children are similar to the different phases of the cognitive evolution of our early relatives. This kind of approach is based on child psychology, developed by Jean Piaget (Piaget, 1971).

For Steven Mithen, the human mind is a general mechanism of intentional knowledge, formed by series of relatively independent mental modules (“psychological mechanisms”), dedicated to specific jobs or concrete behavioural spheres (Mithen, 1994; 1996). Human cognitive evolution consists of an increasing accessibility between mental modules. This mechanism allows for generalised intelligence. A modular architecture, in which each mental module has specific knowledge and capabilities, is present early in human evolution. In particular, the cognitive capabilities involved in tool manufacture are evident in the Acheulian, and possibly even in the Oldowan. Nevertheless, the difference is in the level of interconnection between different cognitive spheres. These interconnections began in the Upper Paleolithic, “the hypothesis that the behavioural changes associated with the Middle/Upper Palaeolithic transition may be due to increased accessibility between mental modules is not dependent upon a change in hominid species or population replacement” (Mithen, 1994: 36).

Other research groups believe that language appeared in the Upper Paleolithic, with Homo sapiens. Davidson and Noble defend the late origin of language, and support their theory with two arguments that are indicative of the complex behaviour that language implies (Davidson and Noble, 1993): the colonization of Australia, and the appearance of art. These two events happened, according to these authors, between 30-40 Ka. The technical complexity of the Acheulian handaxe could be a fallacy, because it is a nonintentional object that is only the result of core exhaustion. Davidson and Noble also think that predetermined “Levallois” knapping does not exist. Belfer-Cohen and GorenInbar stated that these affirmations show a complete ignorance of lithic knapping (Belfer-Cohen and Goren-Inbar, 1994). In fact, Davidson and Noble’s arguments are not valid because Australia was colonised ca. 50-60 ka (Bowler et al., 2003; O’Connell and Allen, 2004; Thorne et al., 1999), and symbolic manifestations may have appeared before it is thought, as indicated by the Berekhat Ram figurine, dated to ca. 240 ka (d’Errico and Nowell, 2000; Marshack, 1997). Belfer-Cohen and Goren-Inbar believe that there is information about cultural complexity in early human groups. The lithic technology from the Ubeidiya and Gesher Benot Ya’aqov sites (Israel) shows that human behaviour in the Lower and Middle Pleistocene is neither simple nor primitive. Lithic evidence proves that predetermination exists, and that the activity sequences developed were selected intentionally. These authors agree with Wynn when he concludes that knapping reveals the cognitive capacity of Lower Paleolithic hominids. Belfer-Cohen and Goren-Inbar considered that the eighties’ approach to Homo erectus should be revised: “It seems prudent in the present state of Paleolithic research to re-examine the assumptions that were ‘in vogue’ in the 1980s as regards the ‘bestiality’ of Homo erectus” (Belfer-Cohen and GorenInbar, 1994). In this respect, they retrieve Isaac’s proposal about reconstructing the social structure and behaviour of early humans (Isaac, 1978). . This vision moves away from primates and closer to anatomically modern humans “...we can assume that both human social structure and human intellectual capabilities appeared quite early in the history of humankind” (Belfer-Cohen and Goren-Inbar, 1994).

However, other researchers think that these types of inferences are hazardous and speculative (Atran, 1982). In this respect, Binford condemns any attempt to reconstruct the intellectual capacity of early hominids through the study of lithic tools. Gamble also shows his scepticism about this issue (Gamble, 1993). The participants in this discussion about the relation between lithic technology and the cognitive capabilities of early humans are basically Anglo-Saxon researchers. However, Anglo-Saxons, and North-Americans in particular, have avoided using the chaîne operatoire approach. Geneste shows that its use is restricted to European researchers, specially in France (Geneste, 1991). North-American archaeologists study lithic industry primarily from a processual perspective. Practically nobody has adopted the chaîne operatoire approach. Functionalist archaeology began with the so-called “New Archaeology” in the 1960s (Binford, 1962; Binford, 1965). The study of lithic remains was never the basic objective of “New Archaeology”; they were not a relevant aspect of their theoretical framework. One of the most interesting aspects of “New Archaeology” is that it applies systemic approaches to archaeology. In this respect, the work of David L. Clarke is fundamental (Clarke, 1968). His systemic and processual approach follows K. Flannery’s work (Flannery, 1967). However, Clarke was inspired by cybernetic and New Geography; and Flannery by Ecology. Neither of them came out radically in favour of formulating universal laws explaining archaeology. Clarke’s systemic approach is especially interesting in its application to the study of lithic technology.

In his recent papers, Wynn does not directly relate cognitive capabilities derived from the manufacture of particular objects to the appearance of language. He is less ambitious about the possibility of making inferences about the mental capability of early humans. Now, his approach is to show intelligence as a modular phenomenon (Wynn, 1993; 1995).

Most of the papers on lithic technology published in the United States are about knapping experiments and functional studies. For example, Dibble and Rolland believe 7

Xosé Pedro Rodríguez that the variability and availability of raw material, and occupation intensity are the principal explaining factors for Mousterian variability (Dibble and Rolland, 1992). They do not study chaînes operatoires, but focus on retouched objects. In Kuhn’s work, the functionalism approach is clear. He states that the Mousterian and its variability must be understood as an adaptative response to spatial use, mobility and survivor strategy, and the availability of raw materials. These variables changed in different sites and environments, so each site is unique, and cannot be extrapolated. The variability observed in faunal and lithic remains is the hominid’s response to diversity in the environment in which they carry out their activities....”the limitations imposed and opportunities offered by differing biotic and raw material environments produce different kinds of strategic responses” (Kuhn, 1993).

but the more informative attributes need to be selected. An archaeological attribute is an independent variable within a specific referential framework (Clarke, 1968). The anthropic object is manufactured within a technical process. This technical process progressively organises inorganic raw material, and turns it into an extension of the human body towards the environment. The object is an instrument, which makes it possible to act on the surrounding environment. So the artifact is the consequence of a process of selection and interaction. We study every artifact by relating it to other artifacts from the same lithic assemblage. Traditionally, these relationships are governed by typology. This approach does not take production processes into account. A system acquires a temporal dimension when it is moving, and archaeological material is always the result of movement, which gives a sequence and an order to its elements. The movement is permanent and directional (Chang, 1967).

Experimentation occupies an important place in the research of North-American prehistorians. Nicholas Toth is the best exponent of Glynn Isaac’s research trajectory. His approach is close to the study of technical systems and chaînes operatoires (Toth 1982; 1987).

The Logical Analytical System and its structural categories The Logical Analytical System is a systematic and processual reading of the lithic record, and conceives it as an association of characters (Carbonell et al., 1992b). These characters are the product of three interrelating components: morphotechnical, morphopotential, and morphofunctional. The morphotechnical component is a group of technical characteristics generated during the production process, which are reflected in the final morphology of the artifact. The morphopotential component provides information about the theoretical potential capacity of action of a particular lithic morphology (Airvaux, 1987). Morphofunctional analysis studies the usewear of the lithic implements in order to establish their specific use.

Our theoretical framework Our initial aim is to obtain information by studying different samples of objects. We want these samples to make it possible for us to describe the technology and culture of specific human groups. In a broad sense we can define technology as “an interactive complex of social organization patterns, which are involved in the production and use of artifacts and in the management of resources” (González García et al., 1994). A technique can be defined as a “system of intentional actions aimed at transforming specific objects to get a useful result” (Quintanilla, 1988). According to Clarke (Clarke, 1968), the material culture subsystem or technological subsystem is part of the human group cultural system.

The Logical Analytical System defines structural categories (not types) so that each object can be positioned in the technical process. Consecutive phases of the knapping process result in different structural categories (Table 1.3.1.). When a natural object (natural Base, nB) is modified by a human being, it is transformed. At the first moment, two Natural Bases must be selected, one to be used as a hammerstone and the other to be the matrix (Figure 1.2.1). As a consequence of this first intervention (Time 1), two or more objects are produced from the initial nB. One of them, the initial matrix (that is, the flaked piece), has one or several negative scars that correspond to the removals, which in turn are the corresponding ‘‘positives’’ (that is, flakes or detached pieces). In this way, we speak of First Generation Negative Bases (1GNB, the matrix) and First Generation Positive Bases (1GPB, the flakes). As work on the 1GNB progresses, new 1GPBs will arise, but if any of the 1GPBs are taken up again and transformed, a new stage of the process will begin (Time 2). By this transformation, the previous 1GPB becomes a Second Generation Negative Base (2GNB), thus beginning the production of Second Generation Positive Bases (2GPB). The process can continue to produce a third generation of objects (Car-

We assume that the archaeological record is a preserved structure of part of an ancient dynamic system. So there is a logic that can relate every piece of a particular record, and no piece can be interpreted alone, only in context with the other components (Carbonell et al., 1992b). We do not want to limit our work to a simple enumeration of technical characteristics. Without a theoretical model that structures information so that it can be understood and explained, we cannot generate knowledge.

Artifacts and attributes The object, manufactured and used by hominids, is our basic unit of study, “Archaeologist’s facts are artifacts and information extracted from their contextual and specific attributes” (Clarke, 1968). We will use the word “artifact” as a synonym of anthropic object; that is, any object modified by a human. To obtain information from this artifact, and then from who manufactured and used it, it is necessary to analyse the attributes and features that result from human intervention. Many features can be studied, 8

THEORETICAL ISSUES AND METHODOLOGY bonell et al., 1999b; Carbonell et al., 1992b; Carbonell and Rodríguez, 1994).

Bases to be used as tools. In this case, we speak of First Generation Negative Bases of Configuration (1GNBC). For example, a cobble that is knapped to configure a chopper is a 1GNBC, and it is in the framework of a Direct Technical Operational Theme. Indirect Technical Operational Themes aim to produce Positive Bases; that is to say, they are strategies for producing flakes. Therefore, First Generation Negative Bases used as cores for producing flakes are First Generation Negative Bases of Exploitation (1GNBE), and they belong to Indirect Technical Operational Themes.

Parallel to the development of this system, other scholars have also shown the need to use a logical language that is more appropriate to defining the technical processes used in the most ancient industries. For example, Glynn Isaac applied the concept of “flaked pieces” to the matrices from which flakes are detached, “any piece from which significant flakes or chips have been removed. This includes raw lumps of stone from which flakes have been detached, and flakes or fragments which have had flakes struck off them, after their separation from a parent block”. This concept is equivalent to the Negative Bases of the Logical Analytical System. “Detached pieces” are “... flakes, flake fragments and shattered flaking products” (that is, our Positive Bases) (Isaac et al., 1981; Isaac, 1984). In this sense, for Geneste “une séquence de production peut être lue en positif (les objets eux-mêmes) et en négatif (les cicatrices des enlèvements de matière antérieure” (Geneste, 1991: 11).

In consequence, we can make a broad distinction among Negative Bases in terms of the objective of knapping: Exploitation (when the matrix is used as a core to produce flakes), and Configuration (when the matrix is configured to be used directly as a tool). There can be first- or second-generation Negative Bases of Exploitation (or cores): that is to say, cores from the exploitation of a cobble or a nodule, or cores from the exploitation of a flake, respectively. Negative Bases of Configuration can also be 1st or 2nd generation. First-generation Negative Bases of Configuration are knapped cobbles, while second-generation bases are retouched flakes.

Basic concepts of the Logical Analytical System Structural categories allow us to place each object in its production sequence. Nevertheless, a conceptual system must be developed that allows us to organize the different levels of anthropic action. This system departs from the concepts of selection and interaction, which characterize the Technical Operational Unit (TOU). Whenever there is a selection and an interaction aimed at achieving a specific objective, with no change in structural category, we talk of Technical Operational Unit. An accumulation of Technical Operational Units with the same general purpose gives rise to the Technical Operational Theme (TOT). We define this concept as the enchainment of actions that have the same final objective. We can talk of Direct Technical Operational Themes (DTOT), and Indirect Technical Operational Themes (ITOT). The objective of Direct Themes is to configure (that is, to shape) First Generation Negative

The concept of Technical Operational Sequence (TOS) is on a higher level than the concept of Technical Operational Theme. In addition to the initial selection of the raw material and its transformation into artifacts (with its different phases, or Operational Units, and the sum of these, Operational Themes), it includes the use and subsequent abandonment of the artifacts. Technical Operational Systems (TOSY) are the group of psychomaterial human activities that select, interact with and transform the environment by means of practical models (Operational schemes) that are the expression of mental (or conceptual) schemes.

Figure 1.2.1: Structural categories for the study of lithic remains, following the Logical Analytical System (Carbonell et al. 1995b)

9

THEORETICAL ISSUES AND METHODOLOGY

1.3.- METHODOLOGY

Our work is based on the analysis of over 9,000 lithic objects from various archaeological sites. The process of analysis began by determining the raw material. Next, we classified each object by following the proposals of the Logical Analytical System. The first classificatory step is to determine the structural category of every object, and the object is then analyzed in accordance with the specific attributes of each structural category (Table 1.3.1). To these categories, the Fragments (FRAG), and the indeterminable objects (INDET) must be added.

their morphological structure necessarily having to be altered. At times particular traces can be seen that suggest that these objects were used as hammers; because of the existence of traces or marks, more or less evidents (nBb), or because to have fractures (nBc). However, these objects may not show, apparent traces (nBa) of percussion, in which case we will only be able to say that they have been imported to the site by hominids. Fundamentally, we have studied the raw material and the size of the natural Bases. First Generation Negative Bases (1GNB) 1GNB preserve the negatives that are the consequence of a knapping sequence. Reduction processes can be addressed to configure (that is to shape) the object (in these cases we will talk about a 1GNB of Configuration, 1GNBC), or to exploit it to produce Positive Bases (in this case we will talk about a 1GNB of Exploitation, 1GNBE).

Specific Analysis of lithic remains Each object should be analyzed from three different perspectives: morphotechnical, morphopotential, and morphofunctional (Carbonell, 1987; Carbonell et al., 1992b). Below we shall give a detailed description of how we carried out our analysis, which focused on the morphotechnical aspect. The morphopotential analysis was made in a very generic way, without going into details, and a functional analysis was not made. Other papers have studied the use wear of some of the materials that we have examined (Ollé, 2002; Sala, 1997; Vergès, 1996). We will include information on these analyses, when it is possible.

The objective of the analysis is to acquire the maximum amount of information from the fewest attributes possible. For the 1GNB, we take into account six attributes. However, before the analysis is begun, the object must be orientated in relation to the mininal rectangle, with the most knapped edge in the distal part, and the most knapped face on the horizontal superior face. In order to describe objects volumetrically we distinguished three planes: horizontal,

Natural Bases (nB) Natural Bases (nB) are objects that can be used without

Structural Category, according to the Logical Analytical System

Abbreviation

Conventional term equivalent

Natural Base without anthropic marks

nBa

Cobbles, pebbles or blocks selected in order to flake them or use them as hammers

Natural Base witht anthropic marks

nBb

Cobbles, pebbles or blocks with percussion marks

Natural Base with fracture (s)

nBc

Cobbles, pebbles or blocks with anthropic fracture(s)

First Generation Negative Base with Configuration

First Generation Negative Base with Exploitation

1GNBC

Cobbles, pebbles or blocks once flaked. They show the scars of the flakes detached from their surfaces. The objective of the flaking process is to configure (that is, to shape) a tool

1GNBE

Cobbles, pebbles or blocks once flaked. They show the scars of the flakes detached from their surfaces. The objective of the flaking process is to produce flakes systematically

Positive Base

PB

Flake, detached from a Negative Base

Second Generation Negative Base with Configuration

2GNBC

Retouched flakes whose blanks were 1GPB. That is, flakes (PB)—detached from the 1GNB—that have been retouched. The common retouched flakes, such as denticulates, notches, side-scrapers,etc.

Second Generation Negative Base with Exploitation

2GNBE

Flake (Positive Base) knapped as a core, for the systematic production of flakes

Table 1.3.1.- Compared terminology between the Logical Analytical System and the most common terms in Anglophone archaeological literature (Carbonell et al. 1999b)

11

Xosé Pedro Rodríguez the shape of the cutting edge: incurvated, sinuous and straight. In addition, the sagittal edge can be symmetrical or non-symmetrical, in relation to the piece’s theoretical bisector plane (Figure 1.3.5).

transverse and sagittal (Figure 1.3.1). At the start of the analysis, the position of the removals must be indicated. We have to observe if there are removals from the left-hand side and then continue clockwise. When there is more than one series of removals, symbols must be used to determine their relative position (for more information on these symbols, see Carbonell et al. 1983: 41-43). The analytical criteria (attributes) applied to the morphotechnical study of First Generation Negative Bases are the following (Carbonell et al., 1983) (Figures 1.3.21.3.5) (Table 1.3.2):

Positive Bases (PB) A Positive Base is a detached piece, removed from a Negative Base. Positive Bases have a dorsal face, a ventral face, and a butt. In the ventral face, the bulb can sometimes be observed. The butt is the part of the percussion platform that has come off with the Positive Base when striking the matrix (Negative Base). The Positive Bases are oriented taking the butt as a reference, placing it in the proximal part (technical orientation of the object). The butt and bulb can be absent, in which case the flake will only be oriented if the other fracture marks are visible on the lower (or ventral) face: percussion or pressure ripples, hackles (Inizan et al., 1992). A series of attributes help us to define the most important features of the PB’s three faces (dorsal, ventral and butt) (Table 1.3.3).

1. Facial attribute. Indicates the number of faces knapped. 2. Extent of the retouched edge. Indicates what part of the object’s periphery is knapped (proportion between knapped zone, and unknapped zone). There are five possibilities, depending on the extent of the knapped zone. (Figure 1.3.2).

Second Generation Negative Bases (2GNB) Second Generation Negative Bases are flakes that have been retouched. The 2GNB were first examined with the Positive Base methodology, since they were PB before they were retouched. However, this analysis is not always complete, since at times retouch eliminates some of the characteristics of the original flake. The orientation of the 2GNB follows the same criterion as the Positive Bases; that is, the butt is in the proximal part.

3. Obliquity. Indicates the angle of the edge with removals (Figure 1.3.3).

The first basic step is to differentiate between the 2GNB of Configuration (that is, artifacts shaped to be used as tools), and the 2GNB of Exploitation (matrices for PB production; that is, cores or nuclei on flakes). We applied Georges Laplace’s analytical criteria for the study of retouching to the 2GNBC (Laplace, 1972; 1987). However, we present the typology on a general level, and do not go into details. From the analysis of 1GNB by the Logical Analytical System, we selected three new attributes and added them to Laplace’s analytical criteria: facial attribute, extent of the retouched edge, and extent of the removal negatives (the amount of surface area of the object that is occupied by the retouching) (Figure 1.3.6) (Table 1.3.4). Figure 1.3.1.- Planes for the description of the objects volume

The same criteria of analysis described above for the 1GNB were applied to the 2GNB of Exploitation.

4.- Extent of the removals. This character describes the amount of surface area of the object that is covered by the removals, taking into account the relative length of the deepest removal relating to the cortex (Figure 1.3.4).

Fragments Fragments are objects whose morphology does not make it possible for them to be classified into one particular structural category. Hence, a morphotechnical study provides little information. We measured their size and examined the amount of cortex. The variables were the same as the ones that we used for the PB: that is, Noncortical (NCO), Noncortical dominant (50% cortical, CO(NCO)), and Totally

5. Frontal edge morphology. Indicates the shape of the cutting edge of the piece, from the perspective of the horizontal superior face (Figure 1.3.5). 6.- Sagittal edge attributes. Observes the form and the symmetry of the sagittal edge. Three characters describe 12

THEORETICAL ISSUES AND METHODOLOGY First Generation Negative Bases Facial attribute Unifacial (U) Bifacial (B) Trifacial (T) Multifacial (M) Extent of the retouched edge retouched zone equivalent to less than 1/8 of the edge (NC) retouched zone between 1/8 and 3/8 of the edge (C) retouched zone between 3/8 and 5/8 of the edge (2C) retouched zone between 5/8 and 7/8 of the edge (3C) All edge is occupied by retouches (4C) Oblicuity (angle of the edge with removals)

Figure 1.3.2.- Extent of the retouched edge of the 1GNB

Shallow-acute or plain angle, between 0º and 15º (P) Semi-acute or semi-plain angle, between 15º and 35º (SP) Acute or simple angle, between 35º and 55º (S) Medium or semi-abrupt angle, between 55º and 75º (SA) Steep or abrupt angle, between 75º and 90º (A) Extent of removals Very marginal (mm) Marginal (m) Extensive (p) Very extensive (mp) Total (t) Frontal edge Morphology Convex (cx) Circular or semicircular (c) Oval or semioval (o) Convergent or Uniangular (1a)

Figure 1.3.3.- Oblicuity (angle of the edge with removals) of the 1GNB

Biangular (2a) Triangular (3a) Quadrangular (4a) Straight (rt) Concave (cc) Sagittal edge Delineation

Symmetry

Straight (r)

Symmetrical (si)

Incurvated (inc)

No symmetrical (nsi)

Sinuous (sin) Table 1.3.2.- Criteria for the analysis of First Generation Negative Bases (1GNB)

cortical (CO). Figure 1.3.4.- Extent of removals of the 1GNB

Size of the objects The objects were measured using the criteria of the Logical Analytical System. When we examined each site, we counted the number of Positive Bases less than 20 mm long. Most these PB were the result of shaping instruments. We refer to the size of objects in four different ways. The analytical parameters depend on hand anatomy, and the metric characteristics of the objects (Carbonell et al., 1999c): Figure 1.3.5.- Frontal edge Morphology and Sagittal edge of the 1GNB

1. Large: objects over 100mm long. Prehension includes 13

Xosé Pedro Rodríguez Criteria for the analysis of Positive Bases Ventral or lower face Type of bulb Delineation Marked (M) Straight (RT) Diffuse (D) Convex (CX) Concave (CC) Sinuous (SIN) Butt Corticality Type of butt Morphology Facets Delineation Cortical (CO) Platform (PLA) Triangular (TRG) No faceted (NF) Straight (RT) No cortical (NCO) Linear (LIN) Quadrangular (CDG) Unifaceted (UF) (single facet, or plain) Convex (CX) Punctiform (PUN) Trapezoidal (TRP) Bifaceted (BF) (two facets) Concave (CC) Pentagonal (PTG) Multifaceted (MF) (more than two facets) Sinuous (SIN) Poligonal (PLG) Convergent or Uniangular (1a) Circular (CIR) Biangular (2a) Oval (OV) Dorsal or upper face Amount of cortex Scars Arrises Delineation Noncortical (NCO) 1 scar (1) 1 arris (1) Straight (RT) Totally cortical (CO) 2 scars (2) 2 arrises (2) Convex (CX) Noncortical dominant (50% cortical) (CO(NCO)) 4 scars (4) 4 arrises (4) Sinuous (SIN) 5 scars (5) 5 arrises (5) Convergent or Uniangular (1a) 6 scars (6) 6 arrises (6) Biangular (2a) More than 6 scars (+6) More than 6 arrises (+6) General morphology of the Positive Base Frontal Morphology Sagittal Morphology Transversal Morphology Quadrangular (CDG) Quadrangular (CDG) Quadrangular (CDG) Trapezoidal (TRP) Trapezoidal (TRP) Trapezoidal (TRP) Pentagonal (PTG) Pentagonal (PTG) Pentagonal (PTG) Poligonal (PLG) Poligonal (PLG) Poligonal (PLG) Circular (CIR)

Circular (CIR)

Circular (CIR)

Oval (OV)

Oval (OV)

Oval (OV) Table 1.3.3.- Criteria for the analysis of Positive Bases (PB)

the digital zones and the whole of the palm.

intersection et chaque réunion représente une potentialité d’interaction” (Airvaux, 1987: 26) (Figures 1.3.7 and 1.3.8). The potentiality of the objects is quantified by measuring the angles that make up these edges. (Airvaux, 1987; 1994; Carbonell et al., 1992b).

2. Medium: objects between 61 and 100mm long. The zone of contact with the hand includes the distal and medial phalanges of the index finger, heart finger, ring finger and little finger, and the distal zone of the palm.

Morphofunctional Analysis

3. Small: objects between 21 and 60 mm long. The prehension of these objects is precise and involves the distal and medial phalanges of the index finger, heart finger, ring finger and little finger, and the proximal phalanx of the thumb.

As we have said above, we did not make a functional analysis of the objects. This analysis is not feasible when working with materials found on the surface. Therefore, in most of the sites in the Northeast of the Iberian Peninsula studied here, it is impossible to make the functional analysis. On the other hand, in the Atapuerca sites, other colleagues from our research team were able to make a functional analysis of the instruments. The information derived from these analyses will be mentioned when we refer to the Atapuerca sites.

4. Very small: objects less than 20mm long. The prehension of these objects is precise and involves the distal phalanges of the thumbs, index and heart fingers.

Morphopotential Analysis The objective of morphopotential analysis is to establish the theoretical capability of action of an artefact. Some geometric models have been drawn up to evaluate the different use potentialities: dihedron, trihedron, semitrihedron and pyramid. The structures of the objects’ edges are classified in any one of these geometric models, “chaque

Identification of the Technical Operational Themes One of the fundamental objectives of the study of lithic tools is to identify the Technical Operational Themes. As we have pointed out above, the Technical Operational Themes are the strategies that produce retouched tools and 14

THEORETICAL ISSUES AND METHODOLOGY Criteria for the analysis of Second Generation Negative Bases of Configuration Facial attribute Unifacial (U) Bifacial (B) Trifacial (T) Multifacial (M) Extent of the retouched edge retouched zone equivalent to less than 1/8 of the edge (NC) retouched zone between 1/8 and 3/8 of the edge (C) retouched zone between 3/8 and 5/8 of the edge (2C) retouched zone between 5/8 and 7/8 of the edge (3C) All edge is occupied by the retouches (4C) Angle of the retouched edge Acute or Simple (S) Steep or Abrupt (A) Shallow-acute or Plain (P) Depth of the retouch Marginal (m) Very marginal (mm) Deep (p) Very deep (mp) Extent of the scars originated because of the retouch Very marginal (mm) Marginal (m) Extensive (p) Very extensive (mp) Total (t) Position or Direction of the retouch Direct (d) Inverse or indirect (i) Alternate (a) Alternating (al) Bifacial (b) Delineation of the retouch Continuous (c) No continuous (nc) Notch (e) Denticulate (dent)

Figure 1.3.6.- Some of the attributes for the analysis of 2GNBC

flakes.

Morphology of the retouched edge

Of course, this work is easier when the lithic record is extensive, and the objects are from a variety of structural categories, particularly if there are refittings. Nevertheless, identifying Indirect Technical Operational Themes (whose objective is the exploitation of Negative Bases, for the production of Positive Bases) is sometimes difficult. In general, this is because they might not be remains of one of the production phases. For example, it is difficult to identify Technical Operational Themes if no Negative Bases of Exploitation (cores) are found. Identifying one of these Themes through PB and 2GNB is not easy.

Straight (rect) Convex (cx) Concave (cc) Sinuous (sin) Table 1.3.4.- Criteria for the analysis of Second Generation Negative Bases of Configuration (2GNBC)

trix, a tool that explains the principal technological characteristics of the lithic record (Figure 1.3.10). We can define the Morphogenetic Matrix as the graphic interpretation of the processes of lithic production, which gives information about the genetic relations among the objects, and the production process. Each one of these processes, or technical strategies, is a Technical Operational Theme (Airvaux, 1987; 1994; Carbonell et al., 1992b).

To define the Indirect Technical Operational Themes, we have taken into account the number of faces knapped in the core, the volumetric structure, and the direction and arrangement of the removals (Figure 1.3.9).

The Morphogenetic Matrix can be read vertically and horizontally. Vertical reading reveals the different knapping phases (Technical Operational Units, TOU) involved in object production. This Technical Operational Units can be of Configuration (C) or of Exploitation (E). Horizontal

The Morphogenetic Matrix To facilitate the description of the Technical Operational Themes, and to show the production processes and tool configuration graphically, we used the Morphogenetic Ma15

Xosé Pedro Rodríguez

Figure 1.3.7.- Geometric models for the morphopotential analysis

Figue 1.3.8.- Some examples of morphopotential analysis

Figure 1.3.9.- Some criteria for the description of Technical Operational Themes

16

THEORETICAL ISSUES AND METHODOLOGY reading provides information about the relations among objects that are part of the same production phase, but which belong to different strategies (Operational Themes). As we have mentioned above, the Operational Themes are Direct (DTOT) when the objective of the technical process is to modify the matrix (1GNB) so that it can be used as an instrument. The objective of Indirect Technical Operational Themes (ITOT) is to produce flakes by means of matrix exploitation. Various reduction strategies can be put into practice to exploit the matrix.

been identified in the archaeological record.

The drawing of lithic production processes follows a pattern. The knowledge of these standards is necessary to interpret correctly the Morphogenetic Matrix (Figure 1.3.10):

4. The arrows drawn with a continuous line indicate a morphogenetic relation between structural categories.

2. Abbreviations of structural categories that have a straight line on top but no drawing indicate that this kind of object has not been found in the record. 3. Abbreviations of structural categories that are not accompanied by a drawing, and with no line on top indicate that this type of Base has not been identified in the record, although there is the possibility that it is there.

5. The arrows drawn with a discontinuous line indicate that there may be a morphogenetic relation between structural categories.

1. Abbreviations of structural categories accompanied by the drawing of an object indicate that this kind of Base has

Figure 1.3.10.- Example of a morphogenetic matrix

17

2 MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

2.1.- INTRODUCTION

Here we study the lithic industry of six sites in the Northeast of the Iberian Peninsula: Puig d’en Roca Excavació (PREX, in the province of Girona), Can Garriga (CG, Girona), Cau del Duc de Torroella de Montgrí (CDTM, Girona), Nerets (NER, province of Lleida), Clot del Ballester (CB, Lleida), and Vinyets (VIN, province of Tarragona) (Figure 2.1.1). We have tried to study materials with stratigraphic context , but this has not always been possible.

Unfortunately, the study of the Lower and early Middle Paleolithic in the Northeast of the Iberian Peninsula has been conditioned by various factors (Rodríguez and Lozano, 1999; 2000; (Rodríguez et al., 2004). In the first place, most of the materials that probably belong to these periods have neither a stratigraphic context or associated fauna. As a result, analysis and interpretation of the lithic industry is often the only resource for obtaining information. Secondly, we have very few radiometric dates. And finally, the first stages of the settlement of this zone has only been systematically studied since the beginning of the 1970s.

Figure 2.1.1.- Map of the Northeast of the Iberian Peninsula with the situation of the studied sites. PREX= Puig d’en Roca Excavació, CG= Can Garriga, CDTM= Cau del Duc de Torroella de Montgrí, NER= Nerets, CB= Clot del Ballester, VIN= Vinyets

21

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

2.2.- PUIG D’EN ROCA EXCAVACIÓ (PREX)

aged Josep Canal to believe that there may be Lower Paleolithic sites in Catalonia, and particularly around Girona. Canal thought that no deposits of this chronology had been found in Catalonia because of the lack of investigations. On the other hand, to the North of the Pyrenees there were numerous Middle Pleistocene sites. For this reason, he was particularly interested in the existence of a Paleolithic deposit at Puig d’en Roca, and in 1972 he showed some pieces to the archaeologist Josep Corominas, who believed they belonged to the Lower Paleolithic.

Location Puig d’en Roca is a hill near the city of Girona (Catalonia) (Figure 2.2.1), where four archaeological assemblages have been identified: Puig d’en Roca I-II (PRI-II), Puig d’en Roca III (PRIII), Puig d’en Roca IV (PRIV), and Puig d’en Roca Excavació (PREX). PRI-II is a site with archaeological remains on the surface, close to the highest part of the hill. PREX is a few meters from PRI-II (2º 48’ 40’’ E, 42º 00’ 00’’ N), and its material comes from several archaeological excavations carried out between the end of the 1970s and the middle of the 1980s (Figure 2.2.2). Surface lithic remains have also been collected from this zone; PRIV consists of this surface lithic remains. Finally, PRIII is East of PRI-II, at the promontory known as the “Turó de la Bateria”, at a height of 100 meters a.s.l. A small excavation was carried out here in 1985 ( 2º 49’ 10’’ E, 41º 59’ 51’’ N).

The founding of the “Associació Arqueològica de Girona” (Archaeological Association of Girona) in 1972 made it possible to carry out prospections at Puig d’en Roca. From the very first, the surface prospected was divided into zones (Figure 2.2.1). The lithic material appeared on this little hill’s slopes after it had been dragged along by erosive agents and dispersed over a relatively wide surface area. In order to determine whether these artifacts belonged to the Lower Paleolithic, Josep Canal got in touch with Henry de Lumley. In 1974, Lumley visited Puig d’en Roca and recognized the antiquity of its lithic industry. In 1976, the participants in the UISPP Congress, held in Nice, visited the Puig d’en Roca sites. Between this moment and the present day, several exhibitions have been held and

Archaeological findings at Puig d’en Roca In his publication about the Middle Paleolithic in Catalonia, Ripoll and Lumley point out that in 1961 M. Oliva found at the Ter’s terrace in the Puig d’en Roca promontory, close to the city of Girona, a Paleolithic deposit that provided large crudely knapped cobbles (Ripoll and Lumley, 1965: 22). In a subsequent publication, Lumley again mentioned the existence of knapped cobbles at Puig d’en Roca (Lumley, 1971: 330).This latter publication encour-

Figure 2.2.1.- Location of the Puig d’en Roca sites, near the city of Girona

Figure 2.2.2.- Excavation of PREX site

23

Xosé Pedro Rodríguez several texts published (Canal and Carbonell, 1979).

Guilleries” and the “Baixa Garrotxa”. The paleozoic block caved in was covered by lacustrine Pliocene sedimentary materials, which in turn were coated with quaternary fluvial and alluvial materials. The peneplain, very dismantled and fossilized, appear at the place where the sediments have been eroded, and next excavated by fluvial valleys originating small hills (approximately 200 meters of heigh).

In 1977, surveys were made along the road that goes to Can Faixeda and the radio station (Figure 2.2.1). A small excavation revealed that the archaeological material appeared some 50 cms below the surface clays. The objects found were small and knapped on porphyry, quartz and quartzite, like the objects recovered on the surface. According to Carbonell et al. (Carbonell et al. 1988a; Carbonell et al. 1988b), these archaeological interventions demonstrated that the material did not come from the detritic level of the Ter’s high terraces, but from the clays that covered these terraces. These deposits are floors that developed after the deposition and alteration of the terraces. As a consequence of these surveys, it was decided to carry out a campaign of systematic excavation (Figure 2.2.2.).

Fluvial Terraces of the Ter river Several researchers have described four terraces on the left bank of the river Ter, near the city of Girona (Pallí, 1976; 1982; Solé Sabarís and Ribera, 1949). Lluís Pallí, however, subdivides the first, the second and the third terraces (TI-TI’, TII-TII’, TIII-TIII’) (Table 2.2.1). According to him, in the sector of Puig d’en Roca, terrace IV is located on the hill that is 145 meters above sea level, and 85 meters above the present-day level of the river (Figure 2.2.1). This terrace, which has a base of fine-grained sandstones, has been almost totally dismantled by erosion, so only part of it stands out from the conglomerate formation. The silty-sandy matrix contains cobbles, the average size of which is between 4 and 7 cm. They are water-rolled and have encircling carbonate crusts of calcium a few millimeters thick. The acid plutonic rocks are frequently broken. The dominant lithologic elements are slates, mica and calcareous schists, quartz, and granite and the acid plutonic rocks. According to Pallí, the third terrace is 104 m above sea level, and 45 above the present-day level of the river (Pallí, 1976) (Table 2.2.2 ).

The first excavation took place in 1979 and a surface of 12m2 was excavated. Two archaeological levels appeared: the first at a depth of about 30 cm, and the second at 1 meter. However, between 1978 and 1979, a fortuitous fire in the Turó de la Bateria enabled systematic prospections to be carried out at the site called Puig d’en Roca III. Over 300 pieces were obtained (Serra et al., 1981). The fine calcareous coating suggested that these materials were probably related to the Ter’s third terrace and forced us to think than they were younger than the formation of this terrace (Carbonell et al., 1988b). The second campaign of excavations in PREX took place in 1982. On this occasion 44 m2 was excavated to the west of the 1979 excavation. The excavation continued in 1984, and all the material recovered was from a single archaeological level. Work was resumed in 1985, when 9m2 was excavated close to the zone excavated in 1982. Soundings were also carried out in the Turó de la Batería (or Puig d’en Roca III).

On this terrace Pallí describes five horizons. Horizon A is a conglomerate formation 4 meters thick, consisting of cobbles that are more rounded than those in terrace IV. The average size of the cobbles is between 7 and 12 cm. The lithologic composition is dominated by granite and acid like rocks. According to Pallí, locally and immediately 20 cm on top of the conglomerates we can observe rough sands and badly gravels rolled without alteration, which have solifluction structures and would be correspond to lateral contributions of periglacial type.

In general, the zones excavated are on a steep slope, so erosion has dispersed the lithic objects through little holes in big blocks of sandstone, and these objects were mixed with clays (Figure 2.2.2.).

Horizon B is a yellow, silty-sandy formation, 80 cm thick, with no signs of stratification. It contains calcareous dispersed nodules, although they are more abundant in the upper part. Sometimes this formation moves laterally or it gets mixed up with rows of angular erosions, of local origin, dragged from the versant by the water flow.

Geological and stratigraphical context The Girona plain is part of the Coastal Catalonian Basin (which is known as the “Depressió de la Selva” in this area). At this point, the Coastal Basin is delimited on the west by “Les Gavarres” Mountains, and boxed in by “Les Fluvial Terraces Altitude Relative altitude (Ter river) a.s.l.(m) (Pallí, 1982) Actual bed 52 0 TI 55 +3 T I’ 57 +5 T II 60 +8 T II’ 72 +20 T III 97-100 +45 T III’ 112 +60 T IV 130-132 +80

Sediments (Pallí, 1982) Gravels, sands and silts Gravels, sands and silts, with pebbles Sands and pebbles, “caliche” nodules Limestones and brown silts, few consolidates, with pebbles Pebbles, sands and silts Pebbles and yellow-reddish silts. Calcification zones Partially cemented conglomerates Pebbles with basal cementation

Table 2.2.1.- Fluvial terraces of the Ter river, near the city of Girona

24

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA Horizons C, D and E are a cyclic formation, about 4 meters thick, the basal part of which is made up of clayey red silts. Later the clayey yellow silts overlap. Both are separated by zones of calcification. Calcareous crusts 35 cm thick form above the yellow silts. The yellow, sandy formation could be the result of aeolian deposits (Mediterranean-type loess). This formation was deposited in drought conditions and without vegetation (dry and cold climate). Red silts, on the other hand, would imply the existence of relatively abundant rains, a warmer climate than now, and the ground protected by vegetation (humid and warm climate). The calcareous crusts and caliche nodules would have formed in conditions between those described; that is to say, in conditions of not very extreme aridity (semiarid and warm climate).

Geology Fluvial Terrace IV Reddish silty clay colluvium of Puig d’en Roca Fluvial Terrace III, with basaltic elements Yellow colluvium and calcareous crusts Fluvial Terrace II Fluvial Terrace I

Archaeological Sites

Vulcanism

Relative Chronology

PRI-II

Mindel

Puig d’en Roca IV Pui d’en Roca Excavació

Final Mindel, or Mindel-Riss Volcanic phase

PR III

Riss Interestadial of the Riss, or Riss-Würm Würm Würm

Table 2.2.2.- Chronological framework of the lithic industry of Puig d’en Roca

Puig d’en Roca I-II’s archaeological material is dismembered and dragged down from the Ter’s fourth terrace, while the remains at Puig d’en Roca III come from the third terrace. The Puig d’en Roca IV material was found among the red clays, retained among big blocks of sandstone, immediately under the fourth terrace (TIV). The material recovered during the excavation of the PREX site comes from the same place, since it appeared in a colluvium of clays, slightly after the formation of the fourth terrace. However, the Puig d’en Roca I-II material is slightly older, since it is related to the fourth terrace.

Stratigraphic context of Puig d’en Roca Excavació The stratigraphy in Puig d’en Roca Excavació is quite homogeneous (Carbonell et al., 1988a; Carbonell et al., 1988b). Over a yellowishness substratum of thin materials (Horizon E), we find the “Formació Rocacorba”, consisting of big sandstone blocks and plates (Horizon C). The little channels between these blocks guide the sedimentation back. The position of the archaeological material is conditioned by the presence of these blocks (Figure 2.2.3). The sedimentary deposit, which was subdivided into horizons A, B and D, consists of various levels of clay with few stratigraphic or sedimentological differences. Only horizon B is different because of the presence of detrital materials, coming from terrace IV, which is at a higher altitude. In this horizon a considerable number of archaeological objects were located. On the lithological level, we cannot establish differences among the archaeological remains and materials from the same terrace, the dominant elements of which are quartz cobbles, cobbles and fragments of porphyry.

Chronology As far as the chronology of the terraces is concerned, Pallí emphasizes the absence of basalt cobbles on the highest terrace (TIV), which would suggest that it was deposited before the volcanic phenomena in the northeast of Catalonia took place (dated at 110 Ka) (Pallí, 1976). The fact that the archaeological material appears on the surface and also in the stratigraphy (although it is part of a colluvium, and therefore in a secondary position), makes it very difficult to determine the chronology. Besides, the absence of paleontological remains is yet another obstacle to obtaining a reliable chronological orientation. Nevertheless, by comparing the older levels of the fluvial terraces (from other fluvial streams), and with the help of lithic analysis, Carbonell placed these sites in the Mindel and Mindel-Riss phases (Carbonell et al., 1988b: 51) (Table 2.2.2).

Horizon A is the highest part of the stratigraphy. The last five centimeters of the stratigraphy are characterized by a vegetable floor of present-day formation.

Figure 2.2.3.- Stratigraphic diagram of Horizons A through E, squares L501 through L510 (Puig d’en Roca Excavació). A, superficial clays ; B, clays with detrital elements,; C, sandstone boulders; D, clays without detrital elements; E, yellow loams (Carbonell et al 1988b)

25

Xosé Pedro Rodríguez in Puig d’en Roca Excavació (less than 5%). Its presence in each structural category, however, is different. Although it is over 5% in PB and 2GNB, it is hardly used in nB and Fragments. This suggests that this raw material is largely used to manufacture flakes. The same can be said of hornfels (2.7%). The fourth most commonly used raw material is quartzite (2.6 %), which is frequently found as a natural Base (7% of all nB are quartzite). It is also used more than the average in the category of 1GNB (4.1%).

The lithic industry of Puig d’en Roca Excavació (PREX) The material that we study here comes from the excavations carried out in 1979, 1982, 1984 and 1985, and from a sounding made in 1978. In total, we have examined 3305 objects (Table 2.2.3). Positive Bases make up the largest structural category (39%), followed by a considerable number of fragments (24%). First Generation Negative bases are also numerous (565 objects, which is 17 % of the material).

Morphotechnical analysis Natural Bases In PREX there are 186 natural Bases (nB), which is 5% of the lithic industry (Table 2.2.3). A total of 88% of the nB are quartz, 7% quartzite, and 2% sandstone. Almost all nB show some fractures or the negative of one removal (that is type “c”, nBc ) (n=181). The fact that there is only one extraction may be the result of testing the quality and the suitability of the pebble. Once the raw material was seen to be not good enough, the pebble would have been abandoned. It is also possible that a fragment became detached when a pebble was used as a hammer. In total, 23 nB have the negative of one extraction (12.4 % of all the nB).

Raw materials Of the raw materials in PREX, quartz is the most commonly used (85.6%). The other materials have hardly been knapped. Only porphyry (4.7%), hornfels (2.8%), quartzite (2.7%), and sandstone (2%) exceed 1% (Table 2.2.3). The fourth terrace of the Ter river provided the quartz necessary for toolmaking. According to Pallí, the lithological composition of this terrace contains 26% of quartz. Given the proximity of the supply, the procurement of raw material for knapping should not have been a problem. The quartz pebbles selected are generally quite small. According to Carbonell et al. (Carbonell et al., 1988b), the selection of materials such as porphyry and quartzite, which are more difficult to locate in the detrital levels of the Ter’s terraces, can be explained by the need for tools of greater morphodynamic potential.

The average dimensions of the natural Bases with fractures or with a negative are 43x34x24 mm. The dimensions of pebbles of the different raw materials vary little. However, the differences are greater if we consider the dimensions of only the nBc that show a negative of extraction. These measure 58x43x30 mm on average, clearly larger than the group of nBc as a whole.

Quartz is particularly dominant among the fragments (nearly 94%), while it is used a little less in the PB and the 2GNB. This suggests that quartz cores produced less PB than the cores of other materials because they are exploited less, or because they are smaller and therefore provide fewer PB.

This is logical if we consider that the single removal may be a test of knapping skill. For this reason, these pebbles are a little larger than the ones that are used only as hammers. First Generation Negative Bases We analyzed 565 1GBN (17% of all the objects). At times it was difficult to discern the 1GNB of Configuration (peb-

Porphyry is the second most commonly used raw material Natural Bases

First Generation Negative Bases 1GNBC

Quartz

164

6,2% 211

Porphyry

4

2,6%

7

4,5%

Hornfels

0

0,0%

4

4,4%

Quartzite Sandstone

13 14,8%

11 12,5%

3,0%

7 10,4%

2

1GNBE

5,8% 174

7,5%

Positive Bases

1GNB Indet. 98

Second Generation Negative Bases 2GNBC

INDET

TOTAL

2GNBE

3,5% 1076 38,0% 345 12,2% 8

0,3% 751 26,5% 2 0,1% 2829 85,60%

20 12,9%

2

1,3%

75 48,4%

28 18,1% 2

1,3%

3,3%

3

3,3%

53 58,9%

19 21,1% 0

0,0%

3

FRAGS

16 10,3% 1 0,6% 155

4,69%

4

4,4% 4 4,4%

2,72%

90

11 12,5%

1

1,1%

31 35,2%

16 18,2% 0

0,0%

5

5,7% 0 0,0%

88

2,66%

1,5%

0

0,0%

33 49,3%

14 20,9% 0

0,0%

8 11,9% 2 3,0%

67

2,03%

1

Indeternitate

1

3,8%

3 11,5%

2

7,7%

1

3,8%

9 34,6%

5 19,2% 0

0,0%

3 11,5% 2 7,7%

26

0,79%

Limestone

0

0,0%

0

0,0%

0

0,0%

1

5,0%

13 65,0%

3 15,0% 0

0,0%

3 15,0% 0 0,0%

20

0,61%

Lydite

1 11,1%

1 11,1%

1 11,1%

1 11,1%

0 0,0%

1 11,1% 0

0,0%

4 44,4% 0 0,0%

9

0,27%

Schist

0

0,0%

1 12,5%

0

0,0%

0

0,0%

4 50,0%

2 25,0% 0

0,0%

1 12,5% 0 0,0%

8

0,24%

Slate

1 14,3%

0

0,0%

0

0,0%

0

0,0%

2 28,6%

2 28,6% 0

0,0%

2 28,6% 0 0,0%

7

0,21%

Flint

0

0,0%

0

0,0%

1 20,0%

0

0,0%

3 60,0%

0

0,0% 0

0,0%

1 20,0% 0 0,0%

5

0,15%

Granite

0

0,0%

0

0,0%

0

0,0%

0

0,0%

0 0,0%

0

0,0% 0

0,0%

1

100% 0 0,0%

1

0,03%

Total

186

5,6% 208

6,3% 250

7,6% 107

3,2% 1299 39,3% 435 13,2% 10

0,3% 799 24,2% 11 0,3% 3305

Table 2.2.3.- Lithic industry recovered in Puig d’en Roca Excavació (PREX)

26

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA bles configured as instruments) from the BN1G of Exploitation (pebbles exploited as cores). When we were unsure, we preferred not to classify the objects in either of the two categories. For this reason, the group of indeterminable 1GNB is relatively numerous.

three removals are frequent. These removals do not follow a pattern. More than a third of the unidentifiable 1GNB belong to this model (34.5%). The size of these objects (average 41 x 35 x 24 mm) is under the average of all the indeterminable 1GNB. Some of them may have been used to test the material, only to be abandoned if the quality was not sufficient or the morphology not suitable for toolmaking.

Most 1GNB are of Exploitation (cores) (n=250, 44.2% of the 1GNB). We identified 208 1GNBC (36.8% of all the 1GNB) (Table 2.2.3). A total of 85.5% of the 1GNB are of quartz. However, we should point out some interesting aspects of the use of the other raw materials. Porphyry is fundamentally used for 1GNB of Exploitation (69%), while quartzite and hornfels distribute his efectives between the 1GNBC (slightly more abundant) and the 1GNBE. In the case of sandstone, the predominance of 1GNBC is very clear. These data can be interpreted as a preference to use porphyry for processes of exploitation, and sandstone for shaping instruments. Quartzite and hornfels would have been used indistinctly for both technical processes.

1GNB of Configuration We have identified 208 1GNBC. Unifacial objects are prevalent, mainly within the 1GNBC of quartz. Half or less than half of the perimeter of 82% of the 1GNBC is knapped (2C or less than 2C). Very few pieces have the whole perimeter configured (4C). The most frequent angle of removals is the medium (32.9%), followed by the steep (26.8%). It is interesting to point out that the objects made out of quartzite have removals with less steep angles.

In general, 1GNB are of small size, with an average of 46x44x28mm. 1GNBE are a little larger than 1GNBC (Table 2.2.4).

The removals are fundamentally extensive (33.5%), and marginal (32%). Very few removals have negatives that

The characteristics of the most commonly used raw material (small rolled pebbles of quartz) determine the general dimensions of 1GNB. It should be pointed out that the dimensions of the quartzite and porphyry objects are clearly greater than the average (Table 2.2.4). This phenomenon is specially evident among the 1GNBC.

Average size of the 1GNB (mm) Length Width 1GNBC 48,1 43,6 1GNBC quartz 44,3 41,1 1GNBC quartzite 80,0 69,0 1GNBC porphyry 72,6 55,3 1GNBC sandstone 46,4 42,4 1GNBC hornfels 46,5 50,9 1GNBE 50,5 47,5 1GNBE quartz 48,9 45,9 1GNBE quartzite 69,1 61,8 1GNBE porphyry 53,8 53,1 1GNB indeterm. 44,8 39,7 Global average 48,6 44,3

Thickness 25,4 24,9 37,6 30,0 17,3 17,4 32,4 31,7 40,5 36,1 26,2 28,5

Table 2.2.4.- Average size of the First Generation Negative Bases

Indeterminable 1GNB Unfortunately there are a significant number of 1GNB that could not be classified within the configuration or exploitation processes. The notable similarities between these objects and the 1GNBC suggests that most of them could fit in this category. However, because we were unsure, we preferred to describe them as indeterminable 1GNB. These objects are fundamentally unifacial.

Figure 2.2.4.-First Generation Negative Bases of Configuration (1GNBC) found in PREX. All artefacts were knapped with quartz pebbles. 1 and 2: small pebbles with straigh dihedral edges, and with bifacial retouch. 3: Unifacial denticulate. 4: Pointed Unifacial. 5: Unifacial with concave dihedral edge. 6: Unifacial with convex dihedral edge. 7: Unifacial with straigh dihedral edge

Among the 1GNB that have not been classified as configuration or exploitation, small quartz pebbles with two or 27

Xosé Pedro Rodríguez occupy the entire face of the object (3.9%). These general data are slightly different in the quartzite 1GNBC, the removals of which are very marginal (35.3%), or extensive (29.4%). In other materials removals are mainly marginal (30.6%).

cases these are total (38%) or very extensive (21%). This characteristic is more significant in the porphyry 1GNBE. The sagittal edges are habitually sinuous, specially in porphyry and materials other than quartz. The number of objects that have symmetrical or non-symmetrical edges is approximately the same.

There are few sagittal sinuous edges. Fundamentally the edges are straight or incurvated. The sagittal edge of most objects is symmetrical.

Centripetal cores are particularly important among the 250 1GNBE. There is also a considerable number of 1GNBE with longitudinal unipolar or bipolar knapping. These 1GNBE have removals that were executed by striking on the horizontal or transversal planes, making good use of the morphology of the pebbles. We have also identified 25 1GNBE trifacials and 18 multifacials, as well as a large group of 1GNBE with orthogonal, opposed and linear knapping, adapted to the morphology of pebbles. Finally, some 1GNBE have a cylindrical morphology, and linear or bipolar opposed knapping. The Objective is to extract elongated PB.

1GNB of Exploitation We have identified 250 1GNBE, most of which are manufactured from quartz (84.4%). Porphyry is the second most common raw material (8% of the 1GNBE) and quartzite the third (4.4%). The analytical study of the material reveals that there is a majority of bifacial pieces (57%), although the presence of trifacials and multifacials is important (8% each group). The percentage of bifacials in the group of porphyry and other raw materials is greater than in the group of objects made from quartz.

Metric analysis suggests that the smallest 1GNBE are trifacials, multifacials and unifacials with multipolar centripetal knapping. None of these cores reach a maximum length of 48 mm. Trifacial cores have an approximate average maximum length of 40mm (42x37x29 mm). Cores with longitudinal unipolar linear knapping are larger (60x59x41 mm on average).

In 65 % of the 1GNBE, more than three quarters of the object’s perimeter has been knapped (3C and 4C). This percentage is even greater in the porphyry objects. The angle of removals is generally steep (41%), although medium angled removals are relatively abundant (24%). The study of the extent of removals suggests that in over half the

Figure 2.2.5.- First Generation Negative Bases of Exploitation (cores) found in PREX. 1: Unifacial centripetal core of quartz. 2: Quartz core with bipolar opposed kanapping. 3: Centripetal bifacial core of quartz. 4: Centripetal bifacial core of porphyry, with hierarchization of one of his faces, destined to the production an specific type of PB. 5: Bifacial core of porphyr, with bipolar opposed knapping. 6: Trifacial core of quartz, with orthogonal knapping.

28

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA The average size of the centripetal cores (without the more clearly predetermined cores) is 50x44x30 mm. The cores with centripetal predetermined knapping are slightly larger (53x48x28 mm of average).

ity of cortical butts. This phenomenon also appears in the 2GNB, with a very similar percentages. If we apply this analysis to each raw material we observe that quartz PB are the ones that have the greatest percentage of cortical butts, while PB made from hornfels and, in particular, porphyry, mainly have non-cortical butts. These data are applicable for the 2GNB of each one of these raw materials.

In general, the 1GNBE of Puig d’en Roca Excavació are small. Only 13 % of the 1GNBE are 70mm or longer. In fact, nearly 30 % of these cores are shorter than 40mm

The platform butt is the most habitual. There is a high proportion of non-faceted butts. Next in number are unifaceted butts, while bifaceted and multifaceted butts are infrequent. Again the percentages are very similar in the 2GNB. Also in this attribute there are differences in the raw material. For example, non-faceted butts dominate in the PB of quartz, but not in the other raw materials: unifaceted butts are more numerous in the PB of porphyry and hornfels, followed by non-faceted butts. The PB of porphyry are the ones that have least cortex. More than 30% of the PB of this raw material have bifaceted or multifaceted butts.

Positive Bases Positive Bases are the structural category which contains most objects: 1299 elements (39.3% of all the lithic industry). Quartz is the raw material that is most commonly used to make PB (82.8% of Positive Bases are quartz). The percentages of PB made from other materials are considerably lower: porphyry (5.8%), hornfels (4.1%), sandstone (2.5%), and quartzite (2.4%). The Positive Bases recovered at this site are small. The overall average is 31x29x14 mm (Table 2.2.5). On average, the quartz PB are smaller than the PB of other raw materials. The average PB made of porphyry, hornfels or quartzite is larger. We have found some PB that are clearly larger than the average (with a maximum length of over 60mm). These objects (and in particular the ones that are longer than 80mm) often have cortex on their dorsal face. They probably belong to the initial stages of the Operational Sequence.

In 2GNB, the data are similar: quartz has the highest percentage of non-faceted butts, while the other materials mainly have unifaceted butts. It should be pointed out that the 2GNB of porphyry have hardly any bifaceted and multifaceted butts, in obvious opposition to the PB of this material. Hornfels 2GNB have the highest percentage of bifaceted and multifaceted butts (over 28%).

We used the Laplace criterion to calculate the PB’s laminar index. This means that we consider as blades those PBs whose length divided by its width gives a result equal or greater than 1.6. In PREX, 124 PB complied with this condition (9.5%). If we use a stricter criterion, and we consider as blades only those PBs whose lengths are greater than twice their widths (that is, an index equal or greater than 2), there are only 28 blades.

Finally we examined the cortex of the dorsal face of the PB and 2GNB. We did not find an obvious preponderance of a single variable, but the percentage of completely non-cortical dorsal faces was greater. Totally cortical, noncortical dominant, and cortical dominant faces have very similar percentages (about 20%). The data for 2GNB are similar, with the exception that completely cortical dorsal faces are the second most numerous, with a significant 27.8%, not far behind completely non-cortical faces (35%). The 2GNB of porphyry are also the ones that have least cortex on their dorsal face. The dorsal faces of hornfels 2GNB have particularly high amounts of cortex.

Of the 1299 PB there are 140 with a length or width equal or less than 20mm (10.8%). These PB are probably the result of the retouching of 2GNBC. We have analyzed all the PB’s technical attributes and whenever possible also those of the 2GNB, because before being retouched a 2GNB was also a PB. So we will have a wider and more detailed vision of the morphotechnical features of all the flakes produced.

PB’s have various different morphologies. Centripetal products are frequent in quartz, porphyry and hornfels. The morphology of some of the PBs we found suggest predetermination in the knapping. Longitudinal unipolar recurrent knapping products are also abundant in quartz, quartzite, hornfels and porphyry. The PB that have unipolar linear removals and bipolar opposite removals on their dorsal face are habitual.

The morphotechnical analysis of PB shows a slight majorAverage size of the PB (mm) Length Width PB of quartz 30,5 28,3 PB of porphyry 36,6 34,1 PB of hornfels 36,9 35,1 PB of sandstone 33,5 31,0 PB of quartzite 37,6 33,8 Global average 31,5 29,3

Thickness 14,3 13,7 13,0 11,3 14,0 14,1

The principal objective of the production of Positive Bases was to procure artifacts with convex dihedral edges or straight dihedral edges. Second Generation Negative Bases We analyzed 445 2GNB, most of which were knapped with quartz (79.3%). Porphyry (6.7%), hornfels (4.3%),

Table 2.2.5.- Average size of the Positive Bases of PREX

29

Xosé Pedro Rodríguez quartzite (3.6%), and sandstone (3.1%) are found in small numbers among the 2GNB. The average size of these objects is a little larger than the average PB. There are no great differences between raw materials, although sandstone and quartzite 2GNB are a little larger than the 2GNB of other raw materials. The quartz objects are the smallest. In any event, the average size of all these negative bases can be considered to be small (between 20 and 60mm) (Table 2.2.6).

retouch is more habitual in quartz, and much less so in porphyry and hornfels. In these two materials, alternate retouch is quite common (about 20%). Generally, retouch is continuous (52.2%), although denticulated retouch is also frequent (19.9%), most of all in hornfels 2GNB (31%). Finally, the most frequent delineation is convex (33.1%), followed very closely by straight (26.7%), and concave (21.2%).

2GNB of Configuration Of the total number of 2GNB found, 435 were 2GNB of Configuration (97.8%). The retouched PB had been produced from all kinds of knapping strategies. We found 2GNB of Configuration made from centripetal products, longitudinal unipolar linear knapping products with a cortical back, multipolar orthogonal knapping products, etc. There were also 2GNB on medium-sized knapped blanks (some over 80mm in length). The analysis of 2GNBC retouch shows that unifacial objects (68.2%) predominate over bifacials (31.8%). This predominance is particularly obvious in quartz 2GNBC, of which 27.6% are retouched on both faces. In the other raw materials there are equal quantities of unifacial and bifacial objects. There is even a slight predominance of bifacials among the porphyry 2GNB (55.2%). Most 2GNBC have been retouched on half or less than half of their periphery (NC+C+2C= 81.3%), and for a considerable number of objects less than a quarter of the perimeter has been retouched (NC+C= 45.2%). Quartz and porphyry 2GNBC are the ones that have the smallest retouched zones, while hornfels and other materials have the largest zones. The angle of retouch is fundamentally acute or simple (45.8%), followed by steep (27.6%). The percentage of steep retouching is significantly greater in quartz (32%) than in the other rocks. The retouching is generally deep with regard to the border (40.3%), although marginal retouchings are also very numerous (30.5%). Very deep retouching is only found with any real frequency among the quartz 2GNBC (26%). The width or extent of the retouching negatives is almost always marginal or very marginal.

Figure 2.2.6.- Positive Bases of PREX. 1, 4, 5, 7, 9 and 10: PB of hornfels. 2 and 6: PB of porphyry. 3 and 8 PB of quartz.

As far as the direction of the retouch is concerned, in more than half of the 2GNBC direct retouch can be found (53.2%), and almost a third have some indirect retouch (30.3%). Next in importance is alternate retouch (8.7%). Alternating and bifacial retouch are very scarce. Direct Average size of the 2GNB (mm) Length Width 2GNB of quartz 36,8 32,7 2GNB of porphyry 38,3 35,2 2GNB of hornfels 44,7 41,7 2GNB of sandstone 49,0 46,1 2GNB of quartzite 50,9 43,2 Global average 38,0 34,1

2GNB of Exploitation Only ten of the 445 2GNB were knapped as cores for PB production. Eight of these ten 2GNB are quartz and two porphyry. The average size is 55.4 x 56.2 x 34.4mm. This average is slightly larger than the average of the 1GNB of exploitation.

Thickness 16,5 15,4 13,1 18,7 18,5 16,2

Among the 10 2GNBE there is a trifacial, two unifacials, and seven bifacials. Two of these cores have longitudinal unipolar linear knapping. This strategy consists of making removals from horizontal planes, which affect the sagittal and make good use of the thickness of the object. There are six 2GNB of exploitation with predominantly centripetal removals. Of these, three have bifacial centripetal knap-

Table 2.2.6.- Average size of the 2nd Generation Negative Bases of PREX

30

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA ping on both faces, and one is an unifacial centripetal. We found a unifacial object with orthogonal knapping, but with a centripetal tendency. Centripetal and orthogonal knapping can also be observed in the trifacial 2GNB.

The knapping methods for the 2GNBE are not more innovative than those for the 1GNBE. Fragments Fragments are the second largest category in PREX (24.2%). A total of 94% of these fragments are quartz. The average size is small, and the maximum length is not a great deal more than 30mm. Cortex is less frequent among the fragments than among the Positive Bases. The percent-

One of the porphyry objects has unifacial multipolar centripetal extractions on its ventral face, and posterior retouch on its dorsal face, which form a dihedral edge. This suggests that the 2GNB was first used as a core, and was subsequently converted into a configured instrument.

Figure 2.2.7-. 2nd Generation Negative Bases of Configuration (2GNBC) found in PREX. 1 and 2: 2GNBC of quartz (n. 1) and sandstone (n. 2), with dihedral edges (“cleavers”). 3: Unifacial of sandstone, with convex dihedral edge. 4: Bifacial uniangular (convergent) of quartz. 5 and 8: Unifacial uniangulars (convergents) of quartz. 6: Unifacial of quartz, with lateral transversal dihedral edge. 7: Unifacial of porphyry, with transversal convex dihedral edge. 9:Unifacial of quartz, with concave dihedral edge. 10: Unifacial denticule of quartz. 11: Unifacial of porphyry, with lateral dihedral edge. 12: Bifacial of hornfels, with lateral-transversal dihedral edge.

31

Xosé Pedro Rodríguez age of fragments with dominant cortex is 17%, and with total cortex 15%.

are concerned, we have identified predetermined PB but no 2GNB. In total there are 100 Negative Bases of Exploitation with centripetal knapping, 38.5% of the cores.

Average size of the Fragments (mm) Length Width Thickness Fragments of quartz 32,0 24,1 16,2 Fragments of porphyry 39,9 29,7 15,5 Fragments of sandstone 40,0 33,2 14,5 Global average 32,4 24,4 16,1

Another class of cores has longitudinal massive knapping, with removals that are unipolar linear or bipolar opposed. These cores have removals at steep or medium angles, starting from the horizontal or the transversal plane, whose negatives can be observed in the sagittal planes. The core need not be previously prepared to put this strategy into practice, because only the natural bases with a suitable morphology are chosen. We have located PB and 2GNBC that belong to this production strategy. We have identified 38 such cores, of which 36 are 1GNBE, and two are 2GNBE. In consequence, we have a good representation of the various Operational Units. This Indirect Operational Theme is the second most important, with 14.6% of the cores.

Table 2.2.7.- Average size of the Fragments of PREX

Strategies for producing Positive Bases The processes of exploitation include the 1GNBE, 2GNBE and PB, as well as the nB used as hammerstones. Adding up all the Negative Bases of Exploitation (of 1st and 2nd generation) we obtain a total of 260 cores. PREX’s lithic industry reveals the existence of 10 strategies for producing PB (Table 2.2.8, and Figure 2.2.8). The most common Indirect Operational Themes are the centripetals, of which there are more than 100 out of the 260 Negative Bases of Exploitation. The most common centripetal Operational Themes are unifacial centripetal knapping and centripetal bifacial. This last has centripetal knapping in one of the faces, associate in the other face to linear knapping, bipolar opposed or else orthogonal knapping. We have identified this strategy of PB production in fifty-nine 1GNBE and three 2GNBE. Some PBs are also probably the result of this strategy. Therefore, all the stages of the operational sequence (chaîne operatoire) are present in the lithic record. The cores with this production strategy are 23.8% of the total number of cores. Quartz is the prevailing raw material among this kind of cores.

In PREX, the trifacial cores are relatively numerous (26 objects, (10%)). Most of these cores are quartz (only one is porphyry). Twenty-five cores are 1GNBE, and one 2GNBE. The most frequent reduction strategy among the trifacial cores is bipolar (particularly bipolar opposed). Twenty-one trifacial cores have bipolar knapping (opposed or orthogonal) on one of their faces. We have also observed that 15 cores have multipolar knapping (centripetal or orthogonal). Another large group of Negative Bases of Exploitation consists of unifacial and bifacial objects, with orthogonal multipolar knapping on one of their faces. More specifically, there are 13 cores (5%) with orthogonal multipolar unifacial knapping, and bifacial cores with orthogonal multipolar knapping on one of their faces and unipolar linear knapping on the other. Among them there is a 2GNBE with unifacial multipolar orthogonal knapping. The cores with orthogonal multipolar knapping on one face, and bipolar opposed or orthogonal knapping on the other are less frequent (7 objects (2.8%)). Finally, we have found three cores with orthogonal multipolar knapping on both faces. In total this group of cores contains 23 objects (8.8% of the NBE), 19 of which are quartz, and 4 are other raw materials (2 porphyries, 1 hornfels, 1 quartzite, and 1 flint).

A second centripetal group consists of 31 bifacial cores with centripetal knapping on both faces. Twnty-nine of these cores are 1GNBE and two are 2GNBE. Six of the 1GNBE show a biconical (or biconvex) model, so we could classify them as discoid (2.3% of all cores). A third of the centripetal bifacial cores are not of quartz. There is a third centripetal group of cores, with even greater technical complexity. Some centripetal cores have predetermined knapping. They have an obvious hierarchization of one of their faces, destined to produce a specific type of PB, while the other face is for preparing percussion surfaces for executing the removals of choice. The knapping strategy of these cores involves obvious premeditation (similar to the volumetric conception and premeditation of the Levallois cores). In Puig d’en Roca Excavació, we have located seven such cores (2.7 % of the Negative Bases of Exploitation). All these cores are 1GNBE. It should be pointed out that none of these cores was manufactured with quartz: three of porphyry, two of quartzite, one of hornfels and one of sandstone. We consider this to be highly significant, because it demonstrates that the raw material was selected, that different rocks were put to different uses, depending on their quality. As far as products

The following group of cores has bipolar-opposed or orthogonal knapping with little structuring, and almost always with few removals. The cores were unifacial with bipolar knapping (8 objects), bifacial with unipolar linear knapping on the other face (8 objects), and bifacial with bipolar opposed knapping on both faces (5 objects). One of the cores is a porphyry bifacial 2GNBE, with unipolar linear knapping on the second face. There are 21 of these cores (8.1%), and the quartz is by far the most common raw material (there are only 3 porphyry cores). The last significant group of cores consist of multifacials. 32

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA some 1GNBE classified in different IOT could belong to the same general strategy, but at different moments of the exploitation. After all, the bipolar-opposed and unipolar linear strategies are not very far from the multipolar orthogonals, which can lead to multipolar centripetal strategies.

Quartz is also the most used material in this case (6 of the 18 cores). Three of the multifacial cores have a tendency to trifacial and two have a tendency to bifacial. Eight of these objects have orthogonal multipolar knapping, seven have bipolar-opposed knapping, and four unipolar linear. It is interesting to point out the existence of cores with a cylindrical morphology, and bipolar-opposed knapping. These cores would be able to give long flakes. There are only three cores of this kind, two of which are porphyry. Most laminar Positive Bases (blades) come from this Operational Theme, although we have not located any 2GNBC.

The raw material depended on the exploitation strategy. Generally quartz was sufficient for their objectives, but hominids sometimes looked for rocks of greater quality. For example, when the objective is centripetal knapping, with predetermined final products, they chose not quartz but other rocks (porphyry, quartzite, hornfels, and sandstone). When they required a centripetal bifacial strategy without predetermination, they used quartz, although not as often as when they required orthogonal, opposed or linear knapping.

Finally we found four quartz cores with very simple knapping: unipolar linear unifacial. In this list of various types of NBE we have not included 27 objects (10.4% of the total of cores), because we have not been able to identify the knapping method.

It is hardly surprising that the products are small, since the raw material is generally small. Positive Bases with cortical butts have an important role, mainly when they are made of quartz. This is because of the small dimensions of this raw material’s Negative Bases. Cortical butts are not common enough among the PB of porphyry and horn-

As a result, we can deduce that there is a great variability of strategies. However, some apparently different Indirect Operational Themes (IOT) may be related. For example,

Faciality Unifacial Bifacial Bifacial Bifacial Bifacial

Bifacial

Unifacial or Bifacial Trifacial Unifacial Bifacial Bifacial Bifacial Unifacial Bifacial Bifacial Multifacial

Unifacial Bifacial

NEGATIVE BASES OF EXPLOITATION (CORES) Knapping Method Raw material multip. Centripetal (2 2GNBE) quartz: 56 (90,3%) multip. centripetal/ unip. linear quartzite: 2 (3,2%) multip. centripetal/ bipol. opposed indet: 2 (3,2%) multip. centripetal/ multip. orthogonal (1 2GNBE) porphyry: 2 (3,2%) multip. centripetal/ Multip. centripetal (2 2GNBE) quartz: 21 (67,7%) porphyry: 5 (16,1%) quartzite: 2 (6,4%) hornfels: 2 (6,4%) lydite: 1 (3,2%) multip. centripetal with hierarchization and predetermination porphyry: 3 (42,9%) quartzite: 2 (28,6%) hornfels : 1 (14,3%) sandstone: 1 (14,3%) Total centripetal cores (5 2GNBE) longitudinal, unipolar or bipolar, massive quartz: 35 (92,1%) (2 2GNBE) porphyry: 2 (5,3%) quartzite:1 (2,6%) linear, opposed, orthogonal (1 2GNBE) quartz: 25 (96,2%) porphyry: 1 (3,8%) multipolar orthogonal (1 2GNBE) quartz: 19 (82,6%) mult. orthogonal/ unip. linear porphyry: 2 (8,7%) mult. orthogonal/ bip. opposed quartzite: 1 (4,3%) mult. orthogonal/ mult. orthogonal flint: 1 (4,3%) bip. opposed or bip. orthogonal quartz: 18 (85,7%) bip. opposed / unip. Linear (1 2GNBE) porphyry: 3 (14,3%) bip. opposed / bip. opposed orthogonal quartz: 16 (88,9%) porphyry: 1 (5,6%) quartzite: 1 (5,6%) unipolar linear quartz: 4 (100%) bipolar opposed, with cylindrical morphology porphyry: 2 (66,7%) quartz: 1 (33,3%) Not identified Total Negative Bases of Exploitation

62

Objects 23,8%

31

11,9%

7

2,7%

100 38

38,5% 14,6%

26

10%

23

8,8%

21

8,1%

18

6,9%

4 3

1,5% 1,2%

27 260

10,4%

Table 2.2.8.- Negative Bases of Exploitation (NBE) found in Puig d’en Roca Excavació, grouped in terms of the knapping method (Indirect Operational Themes). We point out the raw materials utilized for each group of cores

33

Xosé Pedro Rodríguez

Figure 2.2.8.- Morphogenetic matrix of the lithic industry of Puig d’en Roca Excavació: Indirect Technical Operational Themes (ITOT)

fels, which have around 30% of bifaceted or multifaceted butts. It seems quite obvious that these raw materials were chosen for a more complex, or at least a more intense, exploitation.

more than 7% have been shaped out of quartzite pebbles. Also highly significant is the existence of pebbles shaped with uniangular (convergent) morphology, the objective of which is to make a distal trihedral. This is the case of 49 objects (23.6%), most of which are unifacials (36, while only 13 are bifacials). A total of 89.8% of the objects of this DOT were shaped out of quartz pebbles.

The operational sequence of exploitation is well documented, from the selection of the raw material to the production of flakes. And this is demonstrated by the fact that we have found Negative Bases of Exploitation and products from all the Operational Themes.

The third DOT consists of concave dihedrons (notches) shaped out of small pebbles. We have identified 17 notches on pebbles, which is approximately 8% of the 1GNBC. The presence of quartz is very important in this case too (70.6%), although some tools are shaped out of sandstone, hornfels, quartzite (one artifact each one), and porphyry (two artifacts).

The configuration of tools In PREX we have identified 643 objects that have been shaped as tools. The instruments knapped directly with pebbles (1GNBC) are less numerous than those knapped on flakes (2GNBC). More precisely, 67.7% of the instruments were shaped by retouching flakes, and 32.3% by retouching pebbles. This proportion is not the same in all raw materials. Pebbles are used more frequently when the raw material is quartzite. On the other hand, porphyry, hornfels and sandstone are preferred to retouch flakes (2GNBC) (Table 2.2.9).

A total of 79.3% of the 1GNBC belong to one of the DOT mentioned so far. There are also three DOT with very few representatives, but their presence is significant. First, some small pebbles have continuous and/or not very deep denticulated retouch which form lateral cutting edges. We have identified thirteen such objects (ten of them on quartz). There are 13 1GNBC (6.3%) with lateral dihedral edges, which create objects that are similar to side scrapers, although on small pebbles.

We have identified six strategies for shaping tools on pebbles (that is, Direct Operational Themes, DOT) (Figure 2.2.9). Most of these objects are framed in a DOT whose objective is to shape straight or lightly convex cutting edges in the transverse distal zone. The fundamental objective is to form a dihedral edge in the transverse distal zone of the objects (Table 2.2.10). Most of these instruments are unifacials, and they often have few removals. To be precise, we have identified 67 such objects that are unifacials, and 32 that are bifacials. Between them they represent 47.6% of all the 1GNB of Configuration. Approximately 84% have been shaped out of quartz pebbles, while slightly

There are few other morphotypes, but some of them are particularly significant. For example, there are four 1GNBC with denticulated edges. We also found two bifacial pebbles with straight dihedral edges in the distal zone. The morphology of these objects (both of quartz) is similar to that of cleavers. There were also two bifacial pebbles that from the typological point of view, could be classified as handaxes. One of them was shaped out of quartz and the 34

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA other out of schist rock. The existence of a pebble with a bevelled edge at the distal end is interesting. Its morphopotential capability is similar to that of a burin.

Quartz Porphyry Quartzite Hornfels Sandstone Other TOTAL

To sum up, the diversity of DOT is considerable, although most of them can be classified as one of the three types described at the beginning. Quartz is always predominant, but some objects of other raw materials such as sandstone, quartzite and hornfels also exist.

1GNBC 174 33,5% 7 20,0% 11 40,7% 4 17,4% 7 33,3% 5 27,8% 208 32,3%

2GNBC 345 66,5% 28 80,0% 16 59,3% 19 82,6% 14 66,7% 13 72,2% 435 67,7%

TOTAL 519 80,7% 35 5,4% 27 4,2% 23 3,6% 21 3,3% 18 2,8% 643

Table 2.2.9.- Retouched artifacts of PREX. 1GNBC are retouched pebbles, and 2GNBC are retouched flakes (retouched PB).

We have identified 435 retouched flakes (2GNBC). From the morphopotential point of view, the primary objective of shaping sequences is to form dihedral edges with straight or convex delineation; 133 objects have this pattern. More specifically, lateral dihedrons predominate (n=46), followed by steep angled dihedrons (n=30), simple transverse dihedrons (n=31), and lateral-transverse dihedrons (n=26). Particularly interesting are two 2GNBC that are larger than the rest. Excellent straight dihedrons have been configured in the transverse distaland the lateral zones of these objects. From a typological point of view they are similar to cleavers.

ness. These are nearly always straight angled removals in the ventral face. Laplace’s analytical typology shows a predominance of denticulates (26.7%). Then, in order of importance are the side scrapers (19.1%), the acute-angled points (10.8%), and the steep angled continuous objects (8.7%). More specifically, notches with retouch of acute angle (D1) are 13% (n= 56) of the total, while the ones with steep angle (A11) are 4.6% (n=20). In total, the 76 notches (D1 and A11) represent 17.5% of all the retouched flakes. There are 43 simple scrapers (R1), and 43 denticulated scrapers (D3), 9.9% of the total. A considerable number of objects are points: 47 have been produced with retouches of acute angle (P1 and P2), and 19 with steep angled retouches (A24). In total, 66 elements are points (15.2%).

On other occasions, concave dihedral edges have been created (76 objects) (Figure 2.2.7.9). Denticulated edges appear in 69 objects (Figure 2.2.7.10), and trihedrons in 67 (Figure 2.2.7.4, and 2.2.7.8). Denticulated edges are in some cases associated with dihedral edges (2 cases). Two objects had a dihedral edge, and a trihedron. Finally, three artifacts have dihedral concave edges in association with convex or straight edges.

It should be pointed out that there are seven burins (B11 and B12) (1.6%), and eleven end scrapers of acute angle (G11 and G12) (2.5%). We have also identified seven steep angled end scrapers (A15) (1.6%). Therefore, in total there are 18 end scrapers (4.1%). A borer was also present. The typology of 19.5% of the 2GNBC has not been determined,

It is interesting to point out that several objects (n= 9) only show removals whose objective is to reduce their thick-

Figure 2.2.9.- Morphogenetic matrix of the lithic industry of Puig d’en Roca Excavació: Direct Technical Operational Themes (DTOT)

35

Xosé Pedro Rodríguez Morphodynamic capacity of the retouched tools Transversal dihedral edge, with straight-convex morphology Lateral dihedral edge, with straight-convex morphology Transversal and Lateral dihedral edge, with straight-convex morphology Transversal and Lateral dihedral edge, with straight morphology (“cleaver”) Concave dihedral edge (“notch”) Convex straight dihedral edge + Concave dihedral edge Convex straight dihedral edge + Denticulate edge Concave dihedral edge + Trihedral edge Bilateral dihedral edge + distal Trihedral edge (“handaxe”) Distal Trihedral edge (“pick or point”) Straight-convex denticulate edge Uniangular (convergent) denticulate edge (“denticualte point”) Total identified

1GNBC 99 53,2% 13 7,0% 0 0,0% 2 1,1% 17 9,1% 0 0,0% 0 0,0% 0 0,0% 2 1,1% 49 26,3% 4 2,2% 0 0,0% 186

2GNBC 38 10,8% 68 19,3% 24 6,8% 2 0,6% 77 21,9% 3 0,9% 2 0,6% 2 0,6% 0 0,0% 67 19,0% 64 18,2% 5 1,4% 352

TOTAL 137 25,5% 81 15,1% 24 4,5% 4 0,7% 94 17,5% 3 0,6% 2 0,4% 2 0,4% 2 0,4% 116 21,6% 68 12,6% 5 0,9% 538

Table 2.2.10.- Morphodynamic capacity of the retouched tools of Puig d’en Roca Excavació

because these pieces have discontinuous retouchings. The generic purpose of these retouchings is to configure dihedral edges.

• Porphyry is preferred for exploitation processes of some technical complexity. • Hornfels is preferred for exploitation processes, and for shaping instruments on pebbles (1GNBC). • Quartzite is preferentially used as a natural Base and in processes of configuration (1GNBC and 2GNBC). • Sandstone is preferentially used for producing 1GNBC. • Quartz is used in shaping and expoitation processes. However, when complex methods of exploitation were put into practice, with specific preparation of the core, these hominids prefer better quality raw materials.

In total, 27.1% of the initial flakes (PB of more than 20mm of maximum length + 2GNB) were subsequently retouched. This percentage is similar in quartz (26.6%), porphyry (26.9%), honrfels (27.5%), and sandstone (29.8%). This index of configuration (shaping index) is higher in quartzite (36.4%). To sum up, the objective was often to create dihedral cutting edges. In the 1GNBC, there are 131 dihedral cutting edges (70%). Trihedrons are 26%, and denticulated artifacts only 2%. There are also two objects with lateral dihedrons associated with a distal trihedron, that present bilateral symmetry (1%). Morphologically they are similar to handaxes. From the morphopotential point of view, 1GNBC are mainly shaped to create dihedrons. However, greater variability is observed in 2GNBC.

4) There are a wide variety of both Direct and Indirect Operational Themes • The principal DOTs are those that produce transverse dihedrons through unifacial or bifacial removals, and the ones that produce distal trihedrons. • The principal IOT is centripetal knapping, sometimes with predetermination of the final products.

In general, between the two categories 1GNBC and 2GNBC, there are 340 dihedral cutting edges (63.2%), 116 trihedrons (21.6%), and 73 denticulates (13.6%). The concave dihedrons (notches), and the straight or convex dihedrons in the transverse distal zone are the most frequent. Four objects can be classified from the typological point of view as cleavers, which is 0.7% of the objects analysed. There are only two handaxes, which is 0.4%. These percentages are even lower if we take the total number of shaped instruments into account.

5) All phases of the operational sequences are well represented in the record.

Conclusions

The lithic industry from PREX is of no great complexity, particularly in the configuration of most of the instruments on pebbles (1GNBC). However, some elements do lend it a certain complexity. From the point of view of the production processes, the presence of cleavers and handaxes is minimal but significant. What is more, we have identified some complex strategies for exploiting cores, particularly bifacial knapping, the objective of which was to produce PB with a predetermined morphology (“Levallois”). This strategy is used very little in PREX, but the fact that it is present at all is significant. The technical features we have

6) In general this lithic industry is of small size. We deduce that the lithic record recovered in PREX was produced by occupations that both exploited and configured objects, making good use of the raw material provided by the 4th terrace of the Ter river. Tools were made at the site.

We can summarize the conclusions from our study of the lithic industry at the PREX site as follows: 1) Raw materials were procured from a place that was very near to the site (terrace IV of the river Ter). 2) Quartz was the main raw material. 3) The raw materials were processed depending on the objective of the Operational Theme: 36

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA studied mean that the lithic industry at the PREX site can be classified as technical Mode 2, although handaxes and cleavers are scarce. Taking into account the location of the site, the high terraces of the Ter river, and the characteristics of the lithic industry, we believe that the chronology of PREX corresponds to the central Middle Pleistocene.

37

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

2.3.- CAN GARRIGA

Location and archeological excavations Can Garriga is an open air site located in Sant Julià de Ramis, near the city of Girona (42º 01’ 48” North, 2º 50’ 00” East). This site is situated to few meters of the left margin of the river Ter, at the slope of a small hill, seventy meters above the sea level, and 22 meters above the Ter river. Very close to Can Garriga the river crosses the “Congost de Sant Julià de Ramis”, a narrow pass among two small hills (Figures 2.3.1 and 2.3.2). At the moment of his discovery, in 1986, the deposit was next to the arterial highway II, what facilitated the destruction of part of the archeological record (Mora et al., 1987). The same year of the discovery an excavation took place, directed by Eudald Carbonell (Figure 2.3.2). In the spring of 1991 another excavation was accomplished, because of the imminent destruction of the archaeological deposit, due to the remaking of the arterial highway II, and to his connection with the Expressway 7 (Prats, 1991; Rodríguez et al., 1995) (Figure 2.3.3).

Figure 2.3.2.- View of the Can Garriga site, during the excavation of 1986 (at the back the Ter flows) (©Narcís Soler)

main groups, consequence of diverse occupations, close to the riverside of the Ter. Human occupation took place in a moment of travertine regression. Concretely, the sedimentary dynamics of the deposit would be explained in the following way: A travertine was formed on the terrace II’ of the river Ter (Pallí, 1982), dated in his downside in 128.8±6.5 Ka, and in his topside in 112.2±7.5 Ka (all Can Garriaga’s dates were achieved by James L. Bischoff, in cooperation with Ramon Julià of Institut de Ciències de la Terra Jaume Almera (Barcelona), utilizing the Uranium Series)

Stratigraphy and Chronology The sedimentary dynamics of the Can Garriga sequence is characterized by the alternation of glacis deposits, putting sands, and periods of stability, whereon little rafts appeared and generated travertines. These travertines present a very important lateral variation. The artifacts from Can Garriga can be divided into four

Figure 2.3.1.- Location of Can Garriga site

39

Xosé Pedro Rodríguez

Figure 2.3.3.- Plan of the 1991 excavation at Can Garriga (Prats 1991) Figure 2.3.4.- Stratigraphy and dates of Can Garriga sequence, indicating the archaeological levels. On the left the stratigraphic sequence drawn in 1986 (Mora et al. 1987), and on the right the new stratigraphy drawn in 1991 (Giralt et al. 1995). In 1991 column do not appear the level 1 neither the travertine 1, indicated in 1986 column. Legend (for 1991 stratigraphy): 1: sands, 2: sands with travertine nodules, 3: travertine, 4: sands with volcanic particles.

Next appear a tractive level, or of alluvion, formed as a consequence of superficial streams of the Garriga’s hill. In this level lithic material appeared in secondary position: five objects recovered in 1991, and 90 in 1986 (archaeological level 4). On top a travertine was localized, dated in 107.6 Ka (Figure 2.3.4). Almost simultaneously we find an archeological level in pedogenic clays (paleosoil), with lithic industry quite dispersed (24 objects recovered in 1991, level 3). On this level there were volcanic particles, whose analysis has offered a magnetic inverse polarity, that corresponds to the Blake episode (dated in 118 Ka, according to Valet and Meynadier (1993). Above this level of volcanic particles another archeological level (level 2) was deposited, in a deposit of clays and sands. At the top of this second anthropic level, a travertine was dated in 103,5±3,2 Ka. Above these we observed the archeological level 1, also in clays and sands. Finally a travertinic level dated in 87,7±2,5 Ka appeared. Anthropic levels 1 and 2 provided scarce archaeological remains in the campaign of 1986, but in the campaign of 1991, we find 300 and 130 artifacts respectively. In general we can correlate the site of Can Garriga with the Oxygen Isotope Episode 5.

Faunal remains and space organization The unique paleontological remains found in Can Garriga are two indeterminable and burned fragments of diaphysis, and a reduced record of malacofauna. In the level 1 natural Bases of limestone and travertine plaques appeared, that may have been situated intentionally (Figure 2.3.5). The main association consist of a speleothem of 270 x 190 x 30 mm, six travertine’s fragments,

Figura 2.3.5.- Distribution of archaeological remains in a sector of the level 1 of Can Garriga (Rodríguez et al. 1995)

40

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA nB Level 1 Level 2 TOTAL

22 5 27

7,3% 3,8% 6,3%

B1GNB 1GNBC 1GNBE 1GNB Ind. 2 0,7% 4 1,3% 0 0,0% 1 0,8% 8 6,2% 1 0,8% 3 0,7% 12 2,8% 1 0,2%

PB 93 41 134

31,0% 31,5% 31,2%

2GNB 2GNB-C 2GNB-E 29 9,7% 1 0,3% 11 8,5% 1 0,8% 40 9,3% 2 0,5%

FRAG 149 62 211

49,7% 47,7% 49,1%

TOTAL 300 130 430

Table 2.3.1.- Structural categories of the lithic industry of Can Garriga (levels 1 and 2)

The nB fractured are in general of small size, with an average of 49,5x36,5x23,9mm. Also the nBa (cobbles without obvious marks of percussion) are of reduced size, with an average of 39x31,5x13 mm. On the other hand, the nBb (only two objects) are very much larger in size (with an average of 207,5x138x70 mm). As a whole the natural Bases of the level 1 have an average size of 91,7x70,3x41,2 mm.

and seven cobbles (six of them of limestone, and one of granite). This accumulation of blocks and cobbles could not be originated by geological phenomena. In the level 1 we also found a group of syenite cobbles, with marks of percussion, that would indicate an utilization as anvils in activities of breakage, possibly of bones. The absence of faunal remains handicaps to verify this hypothesis (Rodríguez et al., 1995).

The notable presence of natural Bases in the level 1 (specially of large size) would be related with the existence of an area dedicated to the treatment of faunal remains (Figure 2.3.5). Nevertheless, the absence of this kind of record in Can Garriga frustrates the contrastation of this hypothesis.

Analysis of the lithic industry In this work we study the artifacts found in the levels 1 and 2, that were discovered during the 1991 fieldworks. These levels provided the greater number of objects. The level 1 provided 300 objects, and 130 the level two (Table 2.3.1). Both levels have a notable similitude in the representation of diverse structural categories. Only there are obvious differences in the role of the natural Bases (more numerous in the level 1), and in the First Generation Negative Bases (more frequent in the level 2). Both levels are well dated, between 107 and 87 ka.

In the level 2 the number of natural Bases is lesser (only 5 objects). The most common are nB fractured (nBc). The nBc have two very differentiated dimensions: there are two objects with dimensions between 77/71 mm x 51/57 mm, and 17/28 mm; and other two between 39/43 mm x 36/39mm x 29/35 mm. Also there are a granite nBb of small size.

Raw materials Raw materials were collected at the terraces of the Ter river, very close to the site. As we already have said previously, Pallí recognized in 1976 the existence of four terraces of the Ter at the surroundings of Girona’s city (Table 2.2.1). Can Garriga site is directly related with the terrace II’ (Pallí, 1982). Furthermore, in the higher part of the hill where is located the site, remains of the terrace III have been found. Quartz is the dominant raw material in the levels 1 and 2. Quartz is the second more frequent rock at the terraces IV and III, while its frequency is much minor at the terrace II. The rest of raw materials also could be found in the terraces of the Ter river.

First Generation Negative Bases In the level 1 of Can Garriga six 1GNB were recovered, that is the 2% of all the objects. Each 1GNB was manufactured with a different raw material. There are 1GNB of unrepresentative raw materials, like carbonated lutite or lydite. The scarcity of 1GNB of quartz is surprising, because this is the dominant raw material. Four 1GNB are of exploitation and two of configuration. In the level 2 the number of 1GNB is more elevated (n=10), and also its representativeness regarding the material recovered in this level (7.7 %, that becomes a 14.7 % if we

In the level 1 of Can Garriga quartz, quartzite, hornfels, and porphyry represent almost the 90% of all the raw materials utilized. The utilization of the raw materials is similar in the level 2: quartz is still further preponderant (71%), followed by the porphyry (10%), quartzite (6%), and hornfels (5%) (Table 2.3.2).

Quartz Quartzite Porphyry Hornfels Granite Syenite Sandstone Limestone Basalt Lydite Other Indeter. TOTAL

Morphotechnical analysis Natural Bases The natural Bases have an important presence among the lithic material of the level 1 (n= 22). Most of these Bases are fractured (54.5%). A notable diversity of raw materials exist. Quartz and syenite stand out, with eleven of the 22 natural Bases. Syenite and granite were utilized basically as natural Bases.

Level 1 183 61,00% 32 10,67% 22 7,33% 24 8,00% 9 3,00% 8 2,67% 6 2,00% 4 1,33% 4 1,33% 3 1,00% 3 1,00% 2 0,67% 300

Level 2 93 71,54% 9 6,92% 13 10,00% 7 5,38% 1 0,77% 2 1,54% 2 1,54% 1 0,77% 0 0 0 2 1,54% 130

TOTAL 276 64,18% 41 9,53% 35 8,14% 31 7,21% 10 2,33% 10 2,33% 8 1,86% 5 1,16% 4 0,93% 3 0,70% 3 0,70% 4 0,93% 430

Table 2.3.2.- Raw materials of levels 1 and 2 of Can Garriga

41

Xosé Pedro Rodríguez not have into account Fragments). Also in this level there is a significant diversity of raw materials utilized to manufacture 1GNB: six raw materials for 10 objects. There is not an obvious preponderance of any rock, although quartz outshoots. Among the ten 1GNB only one is of configuration, the rest are cores (1GNBE). 1GNB of Configuration Two 1GNB of level 1 are objects configured for their utilization. One of them, manufactured with hornfels, have a transverse distal dihedron configured with bifacial retouching (Figure 2.3.6.1). The other, manufactured with a pebble of quartzite, has scarce unifacial removals that configure a transverse distal convex cutting edge, with tendency to uniangular or convergent (Figure 2.3.6.2). The 1GNBC of hornfels has 81x87x26 mm, and the 1GNBC of quartzite 143x74x49 mm. In the level 2 we only have identified a 1GNBC: a hornfels of large size (134x120x55 mm), with a transverse distal cutting edge, configured with bifacial removals. This object has certain similitude with the 1GNBC of hornfels found in level 1.

Figure 2.3.6.-First Generation Negative Bases of Configuration (1GNBC) found in level 1 of Can Garriga. 1: transverse distal dihedron configured on a pebble of hornfels, with bifacial retouching. 2: unifacial of quartzite, with transverse distal convex cutting edge, with tendency to uniangular (convergent)

1GNB of Exploitation In the level 1 there are four 1GNBE, three of them are bifacials and one is unifacial. The dimensions of three 1GNBE are very similar. They measure between 46 and 56 mm of length, between 40 and 51 mm of width, and between 25 and 34 mm of thickness. Only a 1GNBE of porphyry comes out of this pattern, with clearly different characteristics, so much for his size like for the strategy of knapping. This object measures 157 mm of length and has a unipolar linear recurrent exploitation. The four 1GNBE show some remnant of cortex, although everyone, except the object of porphyry, have small size. However, exploitation never reaches to eliminate the cortex completely. In no case a preconfiguration of 1GNBE was observed. As to the reduction strategies, two objects have centripetal multipolar knapping in a face, and bipolar orthogonal in the other (Figure 2.3.7.3). Also an unifacial object has bipolar orthogonal knapping. The fourth object is the 1GNBE of porphyry with unipolar linear exploitation. In the level 2 there are eight 1GNBE. Among these objects a nucleus of quartz, almost exhausted, has multifacial removals. Therefore, it is difficult to ascribe it to a specific Indirect Operational Theme. We have recovered also a unifacial 1GNBE of syenite, in an initial phase of knapping, with scarce bipolar opposed removals. Among the rest it is necessary to quote two unifacial objects with multipolar centripetal knapping. One of them is a largesized quartzite, whose edges were after reconfigured for its utilization as instrument. The other is a quartz core of small size (48x44x30mm).

Figure 2.3.7.- First Generation Negative Bases of Exploitation (cores) found in Can Garriga. 1: Bifacial centripetal of hornfels, from the level 2, with hierarchically organized faces. 2: Bifacial core of porphyry (level 2), with orthogonal knapping. 3: Bifacial core of lydite (level 1), with orthogonal knapping. It is possible that, after its exploitation, this object be retouched with the objective of configuring a distal trihedron

There are four bifacial cores aside from the multifacial and 42

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA the 2GNB. Among the PB stand out the unifaceted butts (39.1%), while among the 2GNB stand out the no faceted butts (40.7%). In the level 2 the PB and the 2GNB agree in the predominance of no faceted butts (40.5%, and 37.5% respectively). The dorsal faces completely no cortical are very frequent in both levels (68.8 % in the level 1, and 68.3 % in level 2).

the three unifacials. Three of them show a clear orthogonality, and at times opposition in his removals. The three cores have in common the presence of cortical zones. Two objects are of porphyry (of similar dimensions, the larger with 66x52x45mm ) (Figure 2.3.7.2), and the other of limestone (with larger dimensions, 95x70x42mm). Two of these objects show a face much more knapped than the other. Finally it is necessary to point out the existence of a core of hornfels, clearly preconfigured (Figure 2.3.7.1). Its knapping strategy is bifacial centripetal, with two hierarchically organized surfaces, so that one of them becomes of preparation, and the other for Positive Bases extraction, with a predetermined morphology. This core measures 72 x 46 x 32 mm. In general in this level we can identify three knapping strategies: unifacial centripetal, orthogonal bifacial opposed, and bifacial with predetermination of the PB’s final morphology. Later we will keep on talking about this issue.

The fractures are frequent in the PB and 2GNB of Can Garriga. In the level 1 almost a third part of the PB show some kind of fracture (29%). This percentage is minor in the 2GNB (13%). As a whole the 25 % of the products show some kind of fracture. In the level 2 the fractured PB suppose the 26.8 %, while this percentage gets to the 50% among the 2GNB. As a whole the 32% of the products of level 2 show some fracture. A detailed analysis of the PB and 2GNB has been accomplished, identifying 6 types of products :

Positive Bases We have recovered 93 PB in the level 1 of Can Garriga, and 41 in level two (Figure 2.3.8). The analysis of these objects, suggest a great similitude between the two archeological levels. The percentage of PB, with regard to all the industry, coincide in both levels (31%), and are very similar if we deduct Fragments (61.6 % in the level 1, and 60.3 % in level 2). In both levels the quartz is the raw material with more PB. However, also it is necessary to indicate the presence of other materials: quartzite, hornfels, and porphyry (Table 2.3.3).

1.- Products with cortical butt, and dorsal face completely cortical or with dominant cortex (type A) 2.- products with cortical butt, and dorsal face with no dominant cortex (type B) 3.- products with cortical butt, and dorsal face no cortical (type C) 4.- products with no cortical butt, and dorsal face completely cortical or with dominant cortex (type D) 5.- products with no cortical butt, and dorsal face with no dominant cortex (type E) 6.- products with no cortical butt, and dorsal face no cortical (type F)

The PB of both levels are small sized, although the PB of level 2 are lightly further large than the ones belonging to level 1 (Table 2.3.4). In both levels the PB of quartz are the smaller. The average of dimensions of the PB not manufactured with quartz of level 2 is notably larger.In the level 1 PB of hornfels have the larger average size. Only there are four laminar PB (with laminar index greater than 1.60) in the level 1 (4.3%), and 3 in the level 2 (7.3%).

Positive Bases of Can Garriga Level 1 Level 2 Quartz 46 49,5% 30 73,2% Porphyry 13 6 14,0% 14,6% Quartzite 15 3 16,1% 7,3% Hornfels 14 2 15,1% 4,9% Granite 2 2,2% Sandstone 1 1,1% Metamorphic rock 1 1,1% Indeter. 1 1,1% TOTAL 93 41

In the level 1 there are a 16.1% of PB with length and width equal or less than 20 mm (n=15), and in the level 2 a 14.6% (n=6). These data demonstrate the development of sequences of configuration (retouching) of tools at the site.

TOTAL 76 19 18 16 2 1 1 1 134

Table 2.3.3.- Positive Bases and raw materials of Can Garriga

The technical features of the PB of levels 1 and 2 are very similar. In both levels the butts are generally no cortical (63.2 % in level 1, and 62.5 % in level 2), although the presence of cortical butts is significant. Almost all the butts are of platform type (97.7 % in level 1, and 95 % in level 2). If we study the same features in the retouched flakes (2GNBC) results are very similar.

Average size of the PB (mm) Level 1 Raw material Length Width Quartz 27,5 24,8 Porphyry 30,7 28,9 Hornfels 34,6 31,9 Quartzite 27,7 28,3 Global average 29,2 26,8 Level 2 Quartz 27,9 28,3 Other materials 46,6 40,0 Global average 33,2 32,2

The flakes without retouch (PB), and retouched flakes (2GNBC) of levels 1 and 2 offer a predominance of no faceted butts, followed by the unifaceted butts. A fourth part of the butts are bifaceted or multifaceted. However, in the level 1 we observed a difference between the PB and

Thickness 11,5 9,5 10,5 8,1 11,2 13,9 14,3 13,8

Table 2.3.4.- Average size of the Positive Bases of Can Garriga

43

Xosé Pedro Rodríguez

Most of the 2GNB are of configuration, we have identified only a 2GNB of Exploitation in level 1, and another one in the level 2. 2GNB of Configuration In the level 1 the 82.8 % of the 2GNBC are of quartz (24, with regard to a total of 29). Also there are 3 2GNBC of quartzite, one of hornfels, and another of limestone. In level 2 seven 2GNBC are of quartz. Also, there are two 2GNBC of porphyry, and two of quartzite. The analysis of the retouch shows coincidences between the two levels. There are an equilibrium between the 2GNBC with unifacial and bifacial retouch, although with little predominance of the first (51.7 % in the level 1, and 54.5 % in the 2). In the level 1 nearly the 50% of the 2GNBC have 3/8 or less than 3/8 of his edge retouched (NC+C), while in the level 2 this percentage reaches a 80%. In the level 1 the retouch with acute angle predominate (55.3 %), followed by the shallow-acute retouch (21.3 %). In the level 2 the predominance of acute retouching is greater (71.4 %). Steep retouching is less habitual (6.4 % in level 1, and 14.3 % in level 2).

Figure 2.3.8.- Positive Bases of Can Garriga. 1: Predetermined PB of metamorphic rock (level 1). 2: PB of hornfels (level 2). 3: PB of quartzite with unipolar linear removals (level 2). 4: PB of hornfels (level 1). 5: PB of quartz (Level 2). 6: PB of quartz (level 1). 7: PB of quartzite (level 2)

The retouchings are predominantly deep regarding the object’s edge (48.9 % in level 1, and 42.8 % in level 2). Retouch is fundamentally indirect in the two levels (42.5% in level 1, and 42.8 % in level 2), even though there is also an important presence of direct retouch (38.3% in level 1, and 28.6% in level 2), and also a significant presence of alternating retouch (12.8% in level 1, and 21.4 % in level 2). There is few bifacial retouches, and there is no alternate retouch. In both levels most of the retouches are continuous (59.8% in level 1, and 50% in level 2), and with convex delineation (42.5% in level 1, and 50% in level 2).

Results suggest that the flakes of type F predominate, that is without cortex remains in the butt neither in the dorsal face. Nevertheless, this is a relative majority (37%), in front of the significant presence of flakes of type A (9.7%), and of type B (10.4 %), that would correspond to the beginning of the knapping process (initialization phase). Also the presence of flakes with cortical butts, and no cortical dorsal faces is important (type C). In the levels 1 and 2 of Can Garriga we have located PB whose dorsal faces show negatives of centripetal removals. Also there are PB with unipolar linear removals.

In the level 1 the basic objective of the configuration was the haping of dihedral cutting edges (21 objects). We point out the configuration of concave dihedrons (notches), in 10 artifacts (that is a third part) (Figure 2.3.9.1 and 2.3.9.2). Follow in number the lateral dihedrons with straight delineation, or else convex (7 objects) (Figure 2.3.9.3). Also there are transverse dihedrons (n=3), and lateral-transverse

In terms of the traits that we have described, we can affirm than in Can Garriga all phases of knapping are present, from the first (with cores and Positive Bases with complete or dominant cortex), to the last phases, with retouched flakes and debris.

Average size of the 2GNB (mm) Level 1 Raw material Length Width Quartz 35,9 37,5 Other materials 53,6 40,8 Global average 38,6 38 Level 2 Quartz 39,5 37,9 Other materials 66,5 43,5 Global average 41,9 39,4

Second Generation Negative Bases The 2GNB are in the level 1 of Can Garriga a 10 % of all the objects (n=30), and in level 2 a 9.2 % (n=12). Eliminating Fragments the percentage of 2GNB is 19.9 % in level 1, and 17.6 % in level 2. The majority of 2GNBG are of quartz. The size of 2GNB of the two levels are very similar (Table 2.3.5). In both levels the BN2G not manufactured with quartz have a larger size than the ones of quartz.

Thickness 16,1 17,5 16,3 18,1 20,5 18,9

Table 2.3.5.- Average size of the Second Generation Negative Bases of Can Garriga

44

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA strategy is multipolar orthogonal, and bipolar orthogonal . The sizes of these two cores are similar: 46 x 46 x 39 mm (level 1), and 37 x 31 x 22 mm (level 2). Possibly they are two exhausted cores. Fragments We have found 211 fragments, adding up the objects of levels 1 and 2. Fragments are mainly of quartz (Table 2.3.6). The dimensions of the fragments are practically equal in the two levels (Table 2.3.7). In general the objects are of very small size. Only the granite objects have larger dimensions (between 60 and 85 mm of length). In the level 2 only a fragment of sandstone and another one of hornfels surpass the 60 mm. The fragments are basically no cortical. However, in both levels a fourth part show some rests of cortex. In the level 1 the amount of cortex is lightly bigger among the fragments that are not of quartz.

Quartz Quartzite Hornfels Porphyry Granite Syenite Sandstone Basalt Limestone Lydite Other Indterm. TOTAL

Figure 2.3.9.- Second Generation Negative Basees of Configuration (2GNBC) found in level 1 of Can Garriga. 1: Quartzite with a concave dihedron (notch, D1). 2: Quartz with concave dihedrons in both sides (D1+D1). 3: Quartz with a lateral dihedron with straight delineation (R1). 4: Quartz with a lateral-transverse dihedron (R3)

(n=1) (Figure 2.3.9.4). The configuration of denticulated cutting edges affects six objects (20%). We have not found trihedrons.

Fragments Level 1 Level 2 105 70,5% 51 82,3% 13 2 8,7% 3,2% 7 3 4,7% 4,8% 5 3 3,4% 4,8% 3 2,0% 0,0% 3 2,0% 0,0% 3 2 2,0% 3,2% 4 2,7% 0,0% 2 1,3% 0,0% 2 1,3% 0,0% 1 0,7% 0,0% 1 1 0,7% 1,6% 149 62

Total 156 73,9% 15 7,1% 10 4,7% 8 3,8% 3 1,4% 3 1,4% 5 2,4% 4 1,9% 2 0,9% 2 0,9% 1 0,5% 2 0,9% 211

Table 2.3.6.- Fragments and raw materials of levels 1 and 2 of Can Garriga

In the level 2 the type of configuration is similar. The majority of the eleven instruments configured show dihedral cutting edges (7 objects), in front of the denticulated edges (n=3). In an object shallow-acute angled retouches reduce their thickness. Neither we have found trihedrons in this level.

Average size of the Fragments (mm) Level 1 Raw material Length Width Thickness Quartz 19,8 15,2 8,5 Other materials 27,5 22,4 10,3 Global average 22,1 17,3 9,0 Level 2 Quartz 21,4 16,0 9,1 Other materials 30,7 24,2 11,4 Global average 23,1 17,4 9,5

From a typological point of view, in the level 1 a third part of the retouches (10 objects) configure notches (D1), while lateral side scrapers are a fifth part (R1) (6 objects). Denticulated side scrapers (D3) barely surpass the 13% (4 objects). Globally nearly a half of 2GNBC are denticulates, and a third part are scrapers. In level 2 the more frequent types are the D3 or denticulated side scraper, and the R1 or lateral side scraper, with three objects each one. Also there is a notch (D1).

Table 2.3.7.- Average size of Fragments of Can Garriga

Strategies for producing Positive Bases In the level 1 of Can Garriga we have found five cores: Four 1GNBE and one 2GNBE (nearly the 3% of the lithic record, not taking into account the fragments). In the level 2 there are 9 cores (8 1GNBE, and 1 2GNBE ), that is the 13% of the lithic industry (without fragments). The higher percentage of cores in the level 2 is the more significant difference between the two levels. However, PB’s percent-

2GNB of Exploitation We have located only two 2GNBE of quartz, one in each level. The 2GNBE of level 1 is multifacial, with cubic morphology, and orthogonal multipolar knapping. The 2GNBE of level 2 has also cubic morphology, with multifacial knapping with trifacial tendency. The reduction 45

Xosé Pedro Rodríguez age is very similar in both levels (61% in level 1, and 60% in level 2 ). In spite of the differences in the percentages of cores, the systems of exploitation are similar in both levels.

recovered in level 2. This 1GNBE would be consequence of a test to evaluate the quality of the raw material, since there are no PB neither 2GNB of this raw material in the two levels. In the level 1 the syenite is utilized exclusively as natural Base.

Six strategies for the production of flakes have been recognized in Can Garriga (Figure 2.3.10). In the first place we have identified an Operational Theme with unipolar linear recurrent bifacial knapping. Utilizing this strategy the volume of the core is profited by to extract BP. Removals were accomplished fundamentally in the transversal distal plane, striking on the horizontal planes. Only one 1GNBE of porphyry attests the existence of this strategy. The thickness of the object (85 mm) facilitates the obtaining of products. Also we have found PB of quartzite that would be able to correspond to this Operational Theme. We have not identified retouched flakes (2GNBE), that come from this reduction strategy.

We have identified unifacial centripetal cores in both levels. In level 1 we find a core of carbonated lutite, and in the level 2 two unifacial cores knapped with quartzite and quartz. The core of quartzite stand out for his large size (246x112x90mm). This core was later configured to be used as instrument (1GNBC). There is another group of similar cores. These cores are bifacials with multipolar centripetal removals in a face, and bipolar orthogonal in the other. In no case there is specific preparation of the cores. In the level 2 there are two bifacial 1GNBE of lydite and quartz, that correspond to this strategy (Figure 2.3.7.3).

The second system of exploitation is bifacial preconfigured, with multipolar centripetal knapping. One of the core’s faces was used to prepare the removals, that take effect in the other face. This Operational Theme is represented by a 1GNBE of hornfels in level 2 (Figure 2.3.7.1), and by a PB of volcanic rock in level 1 (Figure 2.3.8.1). No 2GNBE could have been ascribed to this Operational Theme.

Finally we have identified a strategy with bifacial orthogonal multipolar removals in a face, and bipolar opposed or else orthogonal in the other. This Operational Theme is represented by 3 1GNBE (two of porphyry, and one of limestone), and by two 2GNBE of quartz. We have found products relating to the porphyry’s operational sequence, but we have not discovered by-products of the knapping of the limestone core.

The third strategy of exploitation is bipolar unifacial opposed, and only it is represented for a 1GNBE of syenite,

Pebbles of small size were utilized to put into practice the three last Operational Themes (the ones that count with

Figure 2.3.10.- Morphogenetic matrix of the lithic industry of Can Garriga

46

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA ments, while in the level 2 there are 13. The proportion of instruments in relation to all the lithic industry (without fragments) is very similar in both levels: 20.5% in level 1, and 19.1% in level 2. Also there is coincidence in the abundance of objects configured on flakes (flake tools), and in the scarcity of instruments on pebbles (two in level 1, and one in level 2).

more cores). The orthogonal, opposed or centripetal knapping was used in order to obtain the best performance of the raw material, but without arriving to exhaust the scarce possibilities of cores. Thus the constant presence of cortical zones in the 1GNBE suggests it. Probably the abundance of these pebbles (fundamentally of quartz) did not do necessary an exhaustive exploitation of the cores. The knapping follows a strategy of simple conception, but very operational for this kind of raw material.

Nevertheless, there are some differences, caused principally because of the differential utilization of the raw materials. In level 1 the retouched instruments of quartz constitute the 30.8% of all the objects of this raw material (without fragments); in level 2 involve the 16,7%. On the contrary, in level 1 the instruments manufactured with quartzite, hornfels and porphyry involve the 11.3%, whereas in the level 2 involve the 23.8%. The conclusion would be that in the level 1 hominids prefer to configure instruments with quartz, but in level 2 they prefer to shape artifacts with quartzite, hornfels and porphyry.

Most of the knapping products of levels 1 and 2 correspond to these three Operational Themes, because in his dorsal face they have negatives of bipolar, orthogonal or else centripetal removals. The presence of a high percentage of products with cortex in the dorsal face, or in the butt is logical, because generally small pebbles were knapping. This circumstance suggests a knapping in situ, at least for this kind of Operational Themes. Also points in this direction the presence of all the phases of the Operational sequence, included small debris. However, we do not have enough elements to extend these affirmations to the others three strategies of exploitation.

Only three instruments on pebble were found in Can Garriga’s excavation (Direct Operational Technical Themes): two 1GNBC of hornfels, and one of quartzite. Hornfels’s 1GNBC have a convex cutting edge, shaped with bifacial knapping. The pebble of quartzite is elongated, and show a convex cutting edge, with tendency to uniangular (convergent), achieved through unifacial extractions (Figure 2.3.6.2).

Although these three Operational Themes are the more frequent, the presence of elements that indicate the existence of an Operational Theme with preconfiguration of PB is very significant. In order to develop this Operational Theme they utilized materials of greater quality, like the hornfels and a volcanic rock.

If we calculated the percentage of retouched flakes or flake tools (2GNBC) in relation to the initial quantity of products derived from the exploitation of cores (PB over 20mm + 2NBG), we will verify the same phenomenon. The general data of the two levels are similar, but there is difference in the configuration of the raw materials. In the level 1, 26.9% of all of the products were configured as instruments, whereas in level 2 the 23.4% were configured. However, the calculation by raw materials displays obvious differences: In the level one, 42.1% of the quartz products, and 9.8% of the other rocks were retouched; In the level two, 21.9% of the quartz products, and 26.7% of the products of another rocks were configured. While in the level 2 quartzite, hornfels and porphyry frequently are utilized to configure instruments, in the level 1 principally they were utilized to produce Positive Bases. These

The essential difference between the two archeological levels studied does not consist in the strategies put into practice, but in the relation between cores and products: In the level 1 cores are very scarce, while than in the level 2 are more frequent. This circumstance can be caused because we have a small, and possibly fragmentary record, since the total surface of the site was not excavated. That would explain the lack of by-products of the knapping of lutite and lydite cores in the level 1, thus as of limestone and syenite in the level 2. However, another possibility is that this type of materials was not knapped at the site, but carried from another place. The configuration of tools In the level 1 we find 31 specifically configured instru-

Morphodynamic capacity of the retouched tools Level 1 TOTAL 1GNBC 2GNBC Transversal dihedral edge, with straight-convex morphology Lateral dihedral edge, with straight-convex morphology Transversal and Lateral dihedral edge, with straight-convex morphology Concave dihedral edge (“notch”) Straight-convex denticulate edge Total identified

Level 2 1GNBC

2

100%

3 11,1%

5

17,2%

1

0

0,0%

7 25,9%

7

24,1%

0

0,0%

1

3,7%

1

0 0 2

0,0% 0,0% 6,9%

10 37% 6 22,2% 27 93,1%

10 6 29

100%

TOTAL

1

10%

2

18,2%

0

5

50%

5

45,5%

3,4%

0

0

0%

0

0%

34,5% 20,7%

0 0 1

1 3 10

10% 30% 90,9%

1 3 11

9,1% 27,3%

9,1%

Table 2.3.8.- Morphodynamic capacity of the retouched tools of Can Garriga

47

2GNBC

Xosé Pedro Rodríguez PB were potentially further effective than the PB obtained by means of the exploitation of small pebbles of quartz, for the greater quality of another raw materials. This fact would be related to the existence of a processing area of fauna in the level 1.

phyry, were retouched, while in the level 1 these flakes few times were retouched, shaping with more frequency the quartz products. This fact would suggest a selection of the raw material of better quality for the production of flakes with good dihedral cutting edges, suitable in the profiting of faunal remains. Frequently cortical PB were configured, perhaps because they had not been usable without retouch.

The most habitual configuration consists in the shaping of dihedral cutting edges. In the level 1 the configuration of concave dihedrons, and of lateral dihedrons stands out, whereas in the level 2 the lateral dihedrons stand out (Table 2.3.8). The configuration of denticulated cutting edges affects the 20.7% of the configured artifacts of the level 1, and the 27.3% of the level 2. We have not identified configuration of trihedrons.

Obviously the absence of faunal remains, as a consequence of the characteristics of the deposit, makes it impossible to contrast this hypothesis. The fauna may have been dragged or deteriorated because of postdepositional phenomena, that also may have affected the lithic industry.

From the typological point of view, in the level 1 the presence of 10 notches (D1) (37% of the classifiable 2GNBC) stands out, followed by the lateral side scrapers (R1, with 10 pieces, 26%). In the level 2, the half (n= 5) of the classified 2GNBC are lateral side scrapers (R1).

The lithic Operational Chain is more complete in the level 2, since in the level 1 there are very few cores. In spite of this, there are evidences of Positive Bases production, and of their configuration at the site. The activity of lithic production is demonstrated, from the initial phases (with notable presence of cortical products), to the final stages (with a core of quartz practically exhausted). The scarcity of quartz cores in the level 1 can be due to that some cores were carried away from the site. Also it can be due to that we only have excavated this level partially.

Conclusions Between 107 and 90 Ka ago the human occupations of the levels 1, and 2 of Can Garriga took place. The Ter river terraces supplied raw material to manufacture instruments; principally pebbles of quartz, quartzite, hornfels, and porphyry. There is a notable diversity of raw materials (11 in level 1, and 9 in level 2 ), although most of these rocks have few pieces.

In synthesis, we proposed that the record of level 1 is fruit of occupations no very extended, with realization of activities of knapping (most of all of configuration, but also of exploitation). Probably also faunal resources were profited, with a simple organization of the space. The level 2 would be the result of less habitual occupations (because of the scarce record found), but also with realization of processes of exploitation and configuration. Both levels have the same technical conceptual scheme, that can correspond to the Mode 3.

The lithic industry of the levels 1 and 2 of Can Garriga has similar features, with some differences that would be related to the functionality of the occupations. The methods of exploitation are very similar in both levels, including the centripetal knapping with preconfiguration of the flakes. The lesser number of cores in the level 1 would be related to the scarce processes of exploitation developed in this level. Also there are similitudes in the type of retouched instruments. We proposed the hypothesis of that the differences between the level 1 and the level 2 are consequence of the different functionality of the occupations. Level 1 would work as a place for the configuration of instruments, and primarily for Positive Bases production, destined to the profiting of fauna. These are our arguments: 1) Strategic situation of Can Garriga, at the Ter riverside, and close to a river’s strait. 2) High quantity of natural Bases, some large-sized, with characteristic marks that would be to indicate that they were used as anvils. 3) Existence of a space organization, because the situation of the large natural Bases 4) High number of Positive Bases in the two levels; but in the level 2 frequently the PB of hornfels, quartzite and por48

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

2.4.- CAU DEL DUC DE TORROELLA DE MONTGRI

Location and geological context

of 20 meters, and the height of the entrance is of 11 meters. This small cave’s surface is of approximately 175 m2.

The Cau del Duc site is located in the Montgrí’s Massif, very close to the city of Torroella de Montgrí (Girona), at 3º07’50 E, and 42º03’12 N. The Montgrí’s Massif is a relief of scarce altitude (barely surpasses the 300 m a.s.l.), emergent in the Empordà zone (Figure 2.4.1). It descends softly toward the Alt Empordà at the north. To the south, this massif falls with relative brusqueness on the plain where the final stretch of the river Ter flows. This massif has two units of relief clearly differentiated. The occidental sector is constituted fundamentally by three elevations, named, from West to East: Muntanya d’Ullà, Mont Pla (or of Santa Caterina, and Muntanya del Montgrí). The highest one of these three peaks is the central with 309 meters. The Oriental sector, known as Muntanya Gran, has a less abrupt morphology. A small depression, at present covered with dunes, separates these two sectors.

History of investigations In 1917 Rossell i Vilà searched for the first time traces of prehistoric occupation in CDTM. In the summer of 1922 Pericot and Miquel Pallarés achieved a systematic exploration of the cave. This exploration allowed discovering 50 objects that, according to these researches, had a similar appearance to the Asturiense (an epipaleolithic lithic industry, located in some sites of the north of the Iberian Pensula) (Pericot, 1923). This ascription was confirmed by H. Obermaier (Obermaier, 1925: 383). Nevertheless, Pallarès and Pericot (Pallarès and Pericot, 1931: 38) indicated the similitude of these pieces with the Mousterian, and even they talked about a technique that would have his origin in the Lower Paleolithic.

The Montgrí’s Massif is a formation of secondary epoch, that consists of cretacic limestones. Its calcareous nature has facilitated the formation of dolines, rockshelters and caves, typical of a karstic system.The possibility of that in interglacial epoch the Montgrí be an island has been proposed, because of the penetration of the Mediterranean sea at the Baix Emporda’s coastal plain (Pallarès and Pericot, 1931: 27; Canal and Carbonell, 1989: 179).

In 1963 Ripoll and Lumley studied the archeological material deposited by Pericot and Pallarés in 1922 at the Archaeological Museum of Barcelona. Ripoll and Lumley considered that this material belonged to the typical Mousterian, and could be placed at the beginning of the Middle Paleolithic (Ripoll and Lumley, 1965). Around that time Pericot reconsidered the antiquity initially given to the Cau del Duc de Torroella de Montgrí (Ripoll and Lumley, 1965). When Lumley published in 1971 his Doctoral Thesis, he indicated the possibility that the lithic industry of CDTM dip his roots in the Lower Paleolithic (Lumley, 1971).

Cau del Duc de Torroella de Montgrí (CDTM) is at some 200 meters of height above the sea level, at the central hill of the three that constitute the occidental sector of the Massif of the Montgrí (Figure 2.4.2). This cave has a depth

Figure 2.4.1.- Location of Cau del Duc de Torroella de Montgí site

49

Xosé Pedro Rodríguez

Figure 2.4.2.- View of the Cau del Duc de Torroella de Montgrí site (the arrow indicates the entrance of the cave)

Lumley visited the site in January 1975 and studied new materials, collected between 1972 and 1974 by local amateurs, and members of the Associació Arqueológica de Girona. According to Lumley, these materials would be able to belong to the Lower Paleolithic.

the infilling. However, there are a series of data that permit to formulate some hypothesis about the deposit, by analogy with the sedimentary infilling of Cau del Duc d’Ulla site, located very close. According to Carbonell, both cavities had a process of similar infilling. Nevertheless, the diagenesis was bigger in CDTM becasue its morphology, its width and its slope toward the exterior. The Cau del Duc d’Ullà site did not suffer a fast emptying of the infilling, because of its most reduced dimensions, and for the protection of the conduit because of the structure of the cave. There were falls of blocks for gravity that changed the morphology of the floor of both caves.

Narcís Soler conduced in 1976 and 1977 excavations at this cave. The material recovered was in secondary position, as a consequence of the dismantling of the sedimentary deposit. From that moment, publications about materials recovered during the seventies appeared (Canal and Carbonell, 1976; Canal, 1977; Canal and Carbonell, 1978; 1980; Canal and Maroto, 1982; Canal, 1983; Carbonell and Mora, 1985; 1986; Vert, 1976; Vert et al., 1977). The most profound study accomplished to the moment is Eudald Carbonell’s Doctoral Thesis (Carbonell, 1985).

Fauna According to Estévez, (Canal and Carbonell, 1989) CDTM’s fauna corresponds to several epoches (Table 2.4.1): Lower Paleolithic (very rolled, and fossilized remains, with stains of manganese); Middle or Lower Paleolithic (fossilized remains, and with manganese stains, sometimes encrusted in calcareous concretion, and fragmented); Upper Paleolithic and postglaciar epoches (subfossil material, with many burned remains of dark coloration, with dark and compact sediment between the spaces of the spongy tissue, not rolled and quite enough entire); and subactual material (mostly remains of microfauna, small animals, domestic animals little fossilized and clearcolor bones).

Stratigraphy Very few remains of the stratigraphy of the sedimentary infilling have remained in CDTM. According to Eudald Carbonell (Carbonell, 1985), Pericot and Pallarés first reports (Pallarès and Pericot, 1931) already indicate that there was a great diagenesis, that destroyed the sedimentary infilling. For this reason, when the first excavations were carried out the archeological material already was found in secondary position. When Narcís Soler directed in 1976 a campaign of excavation at the site, he observed the existence of three levels. In the first place was a cape of sediment deposited in a recent epoch; underneath a cape of blackish sediment appeared, containing materials of various epoches. Finally, a fine sandy, reddish-color layer appeared, that contained in his upside prehistoric lithic industry, and some faunal remains. According to Soler, this last layer only existed in narrow fissures that were left among the blocks of the base of the cave (Soler, 1982: 31).

Among the remains of Paleolithic epoch, the animals of grassland stand out over the animals of forest. The horses’s dimensions place this site in a later epoch in relation to Arago (France). It is probable that part of the sediment had deposited in the Riss period, while another part would be able to have formed in the Wurm’s first phases. The animal with more remains is the horse, followed by the goat. The rabbit is not abundant. Relative abundance of young animals would demonstrate, according to Estévez, the importance of the opportunistic hunting of the easiest

Eudald Carbonell indicates in his Doctoral Thesis, that it is impossible to accomplish a stratigraphic reconstruction of 50

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA Taxon Felis (Lynx) spelaea Vulpes vulpes Dicerorhinus sp. Equus caballus sp. Equus sp. Bos primigenius Elephas merid. or antiquus Cervus elaphus Sus scrofa Capra pyrenaica Capra sp. Rupricapra rupricapra Oryctolagus cuniculus Lepus capensis Avian fauna Emys caspica

Ascribed to the Lower Paleolithic 0 0 0 3 0 1 1 0 1 5 2 3 27 0 0 0

In the Montgrí, other Paleolithic sites with scarce archeological remains have been located. In 1975 five objects were located in Mas Blanc: among them a partial bifacial pebble (chopping tool), two unifacials (choppers), and a side scraper. Most of the pieces show patina, and partial concretion. Seven objects were found in Tossal Gros in 1977: a partial bifacial (chopping tool), two partial unifacials (choppers), two Positive Bases, and two 2GNB. In Sobrestany in 1976 a partial bifacial (chopping tool) of small size was found. In Palloses in 1976 an unifacial (chopper) on a pebble fragment was discovered. Finally, in Palau (at the north zone of the Massif of the Montgrí), various quartz objects were localized in 1980, standing out an unifacial centripetal (epannelée), and various PB.

Paleol. Indeterm. 6 1 1 14 133 1 1 10 1 14 20 9 309 1 14 6

Dating Samples of the bottom stalagmite of CDTM were recovered, in order to make analysis by Uranium series. The result was more than 350 ka (Rodríguez et al., 2004). This cavity’s archeological record is posterior to this chronology. We have a date, also by Uranium series, of the stalagmitic layer that cover the stratigraphic deposit of CDU: 135+10/-9 ka (Tissoux, 1999). Therefore, the Paleolithic assemblages of these two caves would be able to extend throughout a chronological range from 350 to 135 ka ago (OIS 6-9).

Table 2.4.1.- Faunal remains (number of bones) represented in the Paleolithic deposits of CDTM (according to Canal & Carbonell, 1989)

animals to obtain.

Others sites of the Montgrí’s massif There are another Paleolithic sites in the Montgrí’s Massif, aside from the CDTM.The most important is the Cau del Duc D’Ullà (CDU). This site is located to the Northwest of the CDTM (3º 07’ 20’’ East and 42º 03’ 26’’ North). The first excavations were accomplished by Pericot and Pallarés in 1922. They found as a result of his investigations an eneolithic site, and underneath a level with lithic industry and fauna similar to CDTM’s materials. Lumley and Ripoll studied this materials in 1963, and they classified them as Middle Paleolithic (Ripoll and Lumley, 1965). Between 1974 and 1977 the cave was cleaned, and the sediments deposited at the entrance, as a consequence of Pericot and Pallarés’s interventions, were sieved. Nevertheless, the great hardness of the fosilíferous breccia handicapped the continuation of archaeological works.

Analysis of the lithic industry The archeological material of CDTM is deposited in the Museu de Torroella de Montgrí, and in the Centre d’Investigacions Arqueològiques de Girona. The materials preserved in Torroella Museum come from various surveys accomplished by local amateurs during the decade of 1970. This collection was studied by Eudald Carbonell in its Doctoral Thesis (Carbonell, 1985). We have examined the lithic industry deposited in the Centre d’Investigacions Arqueològiques de Girona. These materials come basically from the excavations of 1976 and 1977, and from previous prospections. In order to know the representativeness of the materials that we have studied, we have compared the general data (raw materials and structural categories) obtained by Carbonell, and our data, obtained in this work (Table 2.4.2). Carbonell indicates that there are a coherence between the materials of the various collections in spite of his diverse origin (amateurs’ prospections, professionals’ prospections, excavations). The systematic study of all the lithic material is feasible, treating it as an unitary stratigraphic deposit, that was dispersed by various diagenetic processes. According to Carbonell, the different conservation of the artifacts can be attributed to differential diagenesis phenomena.

The entrance to this small cave has approximately four meters of width, his length is of 12 meters, and his maximum height of 4 meters. Canal and Carbonell distinguish three stratigraphic horizons: • An upper part, named horizon A, where they find postPaleolithic materials. • The horizon B is a compact breccia that contains abundant remains of fauna, and quartz industry. • The horizon C is a detrital level within a clayely matrix, enough altered, where also faunal remains and lithic industry were found. Aside from carnivores’ presence, logic taking into account the CDU’s good conditions for these animals, remains fon Capra ibex highlight. It, accrete to the characteristics of the lithic record, have induced to interpret CDU as a place utilized in hunting expeditions, probably to obtain goats (Canal and Carbonell, 1989).

We have disregarded the fragments to achieve the comparison, since Carbonell does not take this type of objects into consideration. The collection of Museu de Torroella de Montgrí has 3524 objects, and 1052 the collection of Centre d’Investigacions Arqueològiques de Girona. In the Table 2.4.2 we observed that the similitude is noto51

Xosé Pedro Rodríguez Lithic material preserved at the Museu de Torroella de Montgrí, studied by Eudald Carbonell (1985) 1GNB PB 2GNB TOTAL Quartz 1366 0 1231 40,27% 135 68,18% 38,76% Other materials 2158 269 1826 59,73% 63 100,00% 31,82% 61,24% Total 269 3057 198 3524 7,63% 86,75% 5,62% Lithic material preserved at the Centre d’Investigacions Arqueològiques de Girona, studied in this work 1GNB PB 2GNB TOTAL Quartz 444 15 395 46,04% 34 13,76% 40,00% 42,21% Other materials 608 94 463 53,96% 51 86,24% 60,00% 57,79% Total 109 858 85 1052 10,36% 81,56% 8,08% Table 2.4.2.- Lithic materials from CDTM (without fragments, natural Bases and indeterminables), in terms of the Museum where they are preserved, and who has studied them

nB Quartz Hornfels Quartzite Porphyry Limestone Sandstone Flint Lydite Other Total

5 2 5 0 1 1 0 0 2 16

0,3% 0,7% 3,1% 0,0% 1,4% 2,4% 0,0% 0,0% 4,5% 0,6%

1GNBC 0 0,0% 6 2,2% 9 5,6% 2 1,8% 1 1,4% 4 9,5% 0 0,0% 0 0,0% 2 4,5% 24 1,0%

1GNB 1GNBE 1GNB IND. 15 0,8% 0 0,0% 24 8,7% 3 1,1% 16 9,9% 2 1,2% 10 8,8% 1 0,9% 0 0,0% 0 0,0% 3 7,1% 3 7,1% 1 5,6% 0 0,0% 0 0,0% 0 0,0% 3 6,8% 4 9,1% 72 2,9% 13 0,5%

PB 395 179 97 79 48 23 9 3 25 858

22,2% 64,9% 59,9% 69,3% 68,6% 54,8% 50,0% 20,0% 56,8% 34,1%

2GNB FRAG INDET TOTAL 2GNBC 34 1,9% 1327 74,7% 1 0,1% 1777 70,57% 16 43 15,6% 3 1,1% 276 10,96% 5,8% 11 22 13,6% 0 0,0% 162 6,8% 6,43% 13 11,4% 9 7,9% 0 0,0% 114 4,53% 3 17 24,3% 0 0,0% 70 4,3% 2,78% 1 5 11,9% 2 4,8% 42 2,4% 1,67% 2 11,1% 6 33,3% 0 0,0% 18 0,71% 3 20,0% 9 60,0% 0 0,0% 15 0,60% 2 6 13,6% 0 0,0% 44 4,5% 1,75% 85 3,4% 1444 57,3% 6 0,2% 2518

Table 2.4.3.- Lithic industry of Cau del Duc de Torroella de Montgrí (CDTM) studied in this work nB Quartz Hornfels Quartzite Porphyry Limestone Sandstone Flint Lydite Other Total

5 2 5 0 1 1 0 0 2 16

1,1% 0,9% 3,6% 0,0% 1,9% 2,9% 0,0% 0,0% 5,3% 1,5%

1GNBC 0 0,0% 6 2,6% 9 6,4% 2 1,9% 1 1,9% 4 11,4% 0 0,0% 0 0,0% 2 5,3% 24 2,2%

1GNB 1GNBE 15 3,3% 24 10,4% 16 11,4% 10 9,5% 0 0,0% 3 8,6% 1 8,3% 0 0,0% 3 7,9% 72 6,7%

PB 1GNB IND. 0 0,0% 3 1,3% 2 1,4% 1 1,0% 0 0,0% 3 8,6% 0 0,0% 0 0,0% 4 10,5% 13 1,2%

395 179 97 79 48 23 9 3 25 858

88,0% 77,8% 69,3% 75,2% 90,6% 65,7% 75,0% 50,0% 65,8% 80,3%

2GNB 2GNBC 34 7,6% 16 7,0% 11 7,9% 13 12,4% 3 5,7% 1 2,9% 2 16,7% 3 50,0% 2 5,3% 85 8,0%

TOTAL 449 230 140 105 53 35 12 6 38 1068

42,04% 21,54% 13,11% 9,83% 4,96% 3,28% 1,12% 0,56% 3,56%

Table 2.4.4.- Lithic industry of Cau del Duc de Torroella de Montgrí (CDTM) studied in this work, without fragmets neither indeterminable

rious, so much in the structural categories like in the raw materials. The PB represent over 80% of the lithic industry, while the 2GNB are the less habitual artifacts in both collections. As to the raw materials, the global data indicate a predominance of the artifacts that were not made with quartz. Quartz is very scarce among the 1GNB, while it is more frequent among the PB and the 2GNB. The only difference is that quartz is the most frequent raw material among 2GNB of Museu de Torroella de Montgrí , and does not happen the same in the Centre d’Investigacions Arqueològiques de Girona. In general there is an obvious similarity between the two collections. Hence we thought that the data that we expose in this work are representative, in spite of the fact that the collection that we have studied is less numerous than the sample studied by Carbonell. Otherwise, if we added the fragments on the aforementioned

artifacts, in total we have examined 2518 objects. Raw materials Quartz is the most utilized raw material in CDTM (Table 2.4.3). This preponderance is consequence of the high number of quartz fragments. Hornfels is the second most commonly used raw material, followed by quartzite, porphyry, limestone and sandstone. The rest of raw materials do not reach the 1%. Quartz barely is utilized for the knapping of 1GNB. The vast majority of the quartz objects are fragments. The Positive Bases are the second most frequent structural category among the objets of quartz. However, hornfels, quartzite, porphyry, and sandstone are used mainly as PB and as 1GNB. Furthermore, the role of the porphyry among the 2GNB is outstanding. 52

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA Average size of the 1GNB (mm) Raw material Length Width All 1GNBC 112,7 78,8 Quartz 109,4 79,6 Hornfels 109,7 70,9 Sandstones 92,4 77,3 Other raw materials 138,3 87,9 All 1GNBE 86,2 74,0 Hornfels 90,1 79,2 Quartzite 88,2 79,1 Quartz 53,0 49,3 Pophyry 110,6 81,1 1GNB indeterm. 99,0 77,0 Global average 93,5 75,4

The high number of fragments would be able to misrepresent the role of each structural category, hence we have elaborated a table that do not show this type of objects (Table 2.4.4). This new calculation includes 1068 objects, among which the high presence of Positive Bases (80%) stands out; the 1GNB constitute the 10%, and the 2GNB the 8%. The most utilized raw materials continue to be the same, but the difference between quartz and the rest decreases considerably: there is a 42% of quartz, 21% of hornfels, and 13% of quartzite. The river Ter and his terraces provided the necessary raw materials for the tool knapping. At present the Ter flows to 1750 meters of the site. His proximity made the supply of rocks easy.

Thickness 53,1 56,0 52,4 46,6 54,4 52,5 52,0 55,2 39,0 64,9 44,8 51,8

Table 2.4.5.- Average size of the 1GNB of CDTM

Morphotechnical analysis Natural bases The pieces included in this category are not enough (16 objects). The most habitual raw materials are quartz and quartzite. The fractured natural Bases are more frequent (7 pieces) than the ones that show marks, but do not fractures (4 pieces). We have classified a pebble of hornfels as nBa. Finally, four natural Bases could not be ascribed to any sub-category. The average of dimensions is similiar in all the natural Bases. The general average is of 88x70x53 mm. First Generation Negative Bases We presented in this work the data obtained from the analysis of 109 1GNB, that represent 4.3% of all of the studied objects (the 10.2% without the fragments, and the indeterminable objects). More than half of the 1GNB were knapped with honrfels and quartzite, while the rest were knapped fundamentally with quartz, porphyry, and sandstone. The 1GNB of exploitation are majority (66%), front to the 1GNB of configuration (22%), and the no determinables (12%). Quartz was used exclusively for the exploitation of cores (1GNBE). Honrfels and porphyry were used fundamentally, but not exclusively, for the 1GNB of exploitation. Quartzite also was used mainly for exploitation, but obtains significant percentages among the 1GNB of configuration. The objects of sandstone, very scarce, are distributed among the three sub-categories. The average of the dimensions of the 1GNB is 93x75x51 mm, although the 1GNBC have some larger dimensions (Table 2.4.5), and the 1GNBE are smaller. We have observed that the 1GNBC of quartzite and of hornfels have very similar dimensions. The larger dimensions correspond to 1GNBC of other materials, like limestone or porphyry. Among the 1GNBE, porphyry objects have larger size, followed by the 1GNBE of hornfels and quartzite (with very similar size).

Figure 2.4.3.- First Generation Negative Bases of Configuration (1GNBC) found in CDTM. 1: Unifacial of indeterminable fine- grain rock, with transverse distal convex cutting edge. 2: Unifacial of hornfels, with transverse distal straight-convex cutting edge

1GNB of Configuration We have identified 24 1GNBC, knapped utilizing quartzite, hornfels and sandstone. The average of the dimensions of this objects (112x78x53mm) is notably larger than the rest of 1GNB (Table 2.4.5). 53

Xosé Pedro Rodríguez

Figure 2.4.5.- First Generation Negative Bases of Configuration (1GNBC) found in CDTM. 1: Unifacial of quartzite, with transverse distal convex cutting edge. 2: Bifacial tool of sandstone, with transverse distal trihedron, and lateral dihedrons. 3: Bifacial of sandstone, with lateral and transversal dihedrons

Figure 2.4.4.- First Generation Negative Bases of Configuration (1GNBC) found in CDTM. 1: Unifacial of limestone, with transverse distal convex cutting edge. 2: Unifacial of quartzite, with transverse distal straight-convex cutting edge

The majority of the 1GNB are unifacial objects, with a scarce part of their perimeter shaped with acute angled removals fundamentally, and in minor measure, medium angled retouches. The extent of these removals generally is marginal or very marginal, although the extensive extractions also are relatively frequent; in no case they occupy all of the face where they have taken effect. The sagittal edge is for the most part incurvated, and symmetric with regard to the horizontal plane. Significant differences between the 1GNBC of various raw materials do not exist. 1GNB of Exploitation We have analyzed 72 1GNBE. The most used raw material for the exploitation of 1GNB was hornfels (33.3%), followed by quartzite (22.2%), quartz (20.8%), and porphyry (13.9%). The bifacial cores predominate, although the presence of a 8% of trifacial cores, and a 30% of unifacial cores stand out. The bifacial attribute is clearly dominant among hornfels’s objects.

Figure 2.4.6.- First Generation Negative Base of Exploitation (core) found in CDTM. The raw material is porphyry. Orthogonal removals affects the transverse distal plane

54

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA Average size of the PB (mm) Raw material Length Width Quartz 19,8 19,5 Hornfels 30,5 32,5 Quartzite 28,2 29,0 Porphyry 36,1 32,2 Limestone 30,9 27,9 Sandstone 31,2 33,8 Global average 25,5 25,3

Thickness 8,1 9,0 7,9 11,3 8,8 9,0 8,6

Table 2.4.6.- Average size of the Positive Bases of CDTM

porphyry objects have a larger size (36x32x11 mm). The laminar Positive Bases are very scarce. Only 51 PB have a laminar index equal or greater than 1.6 (5.9% of the total), and among these only 15 have an index equal or greater than 2 (1.7%). A 29.4% of the PB have a length and a width less or equal to 20 mm (n=252). This percentage is higher among the PB of quartz (47.8%), in front of the PB of quartzite (17.5%), hornfels (10.1%), and porphyry (5.1%). Figure 2.4.7.- First Generation Negative Bases of Exploitation (cores) found in CDTM. 1: Bifacial core of hornfels, with centripetal knapping. 2: Trifacial core of flint, with orthogonal knapping, and with some hierarchization of the faces

A 69% of the analyzed PB have no cortical butt, while this percentage descends to the 53% among the 2GNB. Adding up PB and 2GNB no cortical butts obtain a 67.9%. The analysis in terms of the raw material indicates that the PB of quartz have higher percentages of cortical butts (42%), while the PB of quartzite obtain only a 14.5%. We observed also a high percentage of cortical butts among the 2GNB of quartz (42%). The rest of raw materials have a small quantity of objects, that offer results enough balanced between cortical butts and no cortical (with the exception of the 2GNB of hornfels, with few cortical butts).

There is not a dominant pattern relating to the extension of the knapped zone of the perimeter of the object. The angles of extractions are frequently steep (37.9%), or medium (27.3%). In the quartz cores almost half of all the extractions have steep angle. The shallow-acute or semiacute angled extractions are very scarce. Nearly a third part of extractions are extensive (29.2%), and a fourth part marginal (24.8%). Also there is a significant percentage of total extractions (18.6%).

The most common butts are of platform type. The same happens among the 2GNB. The linear butts barely reach a 3% among the PB and the 2GNB. In this sense, we can highlight a 5% of linear butts among the PB of quartzite, and the 7.7% among the 2GNB of porphyry.

The sagittal edges of the 1GNBE use to be sinuous or else incurvated. The straight edges are not much habitual. We observed a distribution of artifacts between the symmetric edges (46.7%), and no symmetric (49.3%). Among quartzite, quartz and porphyry no symmetric edges predominate, while the symmetric edges dominate in hornfels, and in other raw materials.

The unifaceted butts are majority among the PB (39.4%). However, there is an important percentage of no faceted butts (30.5%). The presence of bifaceted (16%) and multifaceted butts (11.9%) also is outstanding. These percentages are something different among the 2GNB, since exist a predominance of no faceted butts (43%), while the unifaceted reach a 35.3%. The bifaceted and multifaceted butts are less frequent among the 2GNB than among the PB. In addition, there is an obvious difference between the PB of quartz, and the rest of raw materials. Among the PB of quartz there is a preponderance of no faceted butts (44%), followed very closely for the unifaceted (41%). In the rest of rocks the no faceted butts are small in number, corresponding the predominance to the unifaceted butts. Also the role of bifaceted and multifaceted butts in the rest of materials is very important, specially among the PB of quartzite (with a 45.8%).

Positive Bases We have examined 858 Positve Bases for this study, that is the 34.1 % of all the lithic industry (Table 2.4.3). However, without the fragments and the indeterminable objects, the Positive Bases reach 80 % of all the lithic record (Table 2.4.4). The 46% of the PB are of quartz. After quartz, the most habitual raw materials are the hornfels (20.9%), quartzite (11.3%), and porphyry (9.2%). In general, the PB’s size is very small. The average of the dimensions of all the PB is 25x25x8 mm (Table 2.4.6). The PB of quartz are the smallest. On the contrary, the

Finally, the study of the dorsal face of the PB and 2GNB 55

Xosé Pedro Rodríguez evidences a majority of no cortical dorsal faces (69.1%). In general the PB conserve less cortex than the 2GNB (71.9% of no cortical dorsal faces in the PB, in front the 27.8% in the 2GNB). Only 3.3% of the PB have their dorsal face completely cortical, while there is a 16.4% of 2GNB with dorsal face completely cortical. Globally, PB and 2GNB have a 4.1% of dorsal faces completely cortical. The amount of cortex is greater among the PB of quartzite, and lesser among quartz PB. Results are different among the 2GNB: the principal amount of cortex appears in the quartz artifacts, and the minor in the artifacts of porphyry. In general, we can say that the PB have dorsal faces predominantly no cortical, and the 2GNB have an elevated percentage of dorsal faces with >50% of cortex (45.5%). The combined study of the amount of cortex in the butt, and in the dorsal face has confirmed the existence of more cortex in the 2GNB. Among the 821 PB whose butt and dorsal face are analyzable, 57 preserve some rest of cortex in his butt, and also in his dorsal face (6.9%). Specifically, only 8 PB have butt and dorsal face completely cortical (roughly the 1%), while 16 have cortical butt and dorsal face with dominant cortex (1.9%). A notable similitude between the PB various raw materials exists. The amount of cortex is greater in the 2GNB. Among the 69 analyzable objects, 24 preserve some rests of cortex in both faces, butt and dorsal (34.8%). Concretely, there are seven 2GNB with dorsal face and butt completely cortical (10.1%), while 10 show cortex on the butt, and dominant cortex in the dorsal face (14.5%). The principal amount of cortex appears among the 2GNB of quartzite (60 % with cortical butt and dorsal face with some rest of cortex), and of hornfels (36.4 %). These data seem to indicate a preference for the cortical blanks to be retouched and transformed in 2GNB, specially in raw materials like quartzite and hornfels. Adding up the PB’s data and the 2GNB we obtain 81 objects with cortex on the butt and in the dorsal face, referring to a total of 890 pieces in that these variables could be analyzed. This implies that a 9.1% of the initial knapping products have an important amount of cortex. It is significative that 24 of those 81 initial products had been retouched (that is, the 29.6%).

Figure 2.4.8.- Positive Bases of CDTM. 1, 6: Positive Bases of hornfels, with scars that denote an unipolar linear knapping, and with cortex remains. 2: Quartzite, with unipolar linear knapping. 3, 5 and 7: Centripetal Positive Bases of quartzite. 4: Preconfigured PB of quartzite. 8: Centripetal PB of porphyry

tified PB with a morphology that corresponds to multipolar centripetal predetermined knapping (Figures 2.4.8.3, 2.4.8.5, and 2.4.8.7). Also there is an object with uniangular (convergent) morphology clearly predetermined (Figure 2.4.8.4). The Positive Bases of hornfels show an ample spectrum of strategies. We have identified unipolar linear, bipolar opposed, bipolar orthogonal, multipolar orthogonal and multipolar centripetal knapping. Also in this case some PB retain a perimeter of cortex, and unipolar linear or bipolar opposed scars, that can indicate an Operational Theme of longitudinal knapping, with systematic reduction of the volume of the matrix (Figures 2.4.8.1, and 2.4.8.6). There are multipolar centripetal PB of hornfels, whose morphology can suggest predetermination.

The PB of quartz indicate the existence of unipolar linear, bipolar orthogonal and multipolar centripetal knapping. Some PB with multipolar orthogonal knapping, and with bipolar orthogonal knapping would be able to fit in strategies of exploitation of the sagittal and transverse planes of the 1GNBE, reducing the volume of the cores systematically. There is also an object whose triangular morphology seems predetermined.

The PB of porphyry show also great diversity of knapping strategies. Unipolar linear, bipolar opposed, bipolar orthogonal, multipolar orthogonal, and multipolar centripetal knapping exists. The longitudinal knapping is demonstrated for PB whose dorsal faces evidence a cortical perimeter, and unipolar linear removals. Also exist centripetal PB with a preconfigured morphology (Figure 2.4.8.8). In this sense, some PB that belong to a first phase of preparation of the flaked surface of the 1GNE have been found.

We have confirmed the existence of PB of quartzite with unipolar linear, bipolar opposed, bipolar orthogonal and multipolar centripetal knapping (Figure 2.4.8). One part of these PB belong to the longitudinal knapping that is practiced on the transverse and/or sagittal planes; these PB retain a perimeter of cortex, and show unipolar linear or bipolar opposed scars (Figure 2.4.8.2). Also we have iden56

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA We have identified unipolar linear, bipolar opposed, bipolar orthogonal, multipolar orthogonal, and multipolar centripetal knapping among the PB of sandstone. The longitudinal knapping with volumetric reduction of the transverse, and/or sagittal planes is manifested through unipolar linear PB. Also there are centripetal PB, whose morphology would suggest predetermination.

lows affirming that for the most part only a face of the object was retouched, while the configuration affects two faces in the 32%. However, the bifacial 2GNB are majority among the objects manufactured with hornfels. 2GNB of CDTM Quartz 34 Hornfels 16 Porphyry 13 Quartzite 11 Limestone 3 Lydite 3 Flint 2 Sandstone 1 Schist 1 Slate 1 Total 85

Second Generation Negative Bases The collection of lithic artifacts from CDTM deposited in the Centre d’Investigacions Arqueològiques de Girona has 85 2GNB. This number of pieces represents a 3.4 % of the total (Table 2.4.3), although the percentage increases to the 8% without the fragments, and the indeterminable (Table 2.4.4). The presence of quartz (40%), hornfels (18.8%), porphyry (15.3%), and quartzite (12.9%) stands out (Table 2.4.7).

40,0% 18,8% 15,3% 12,9% 3,5% 3,5% 2,4% 1,2% 1,2% 1,2%

Table 2.4.7.- Raw materials of the Second Generation Negative Bases of CDTM

The average of the dimensions of all the 2GNB is of 44x41x15mm, what implies a small size. The 2GNB of larger size are of quartzite, while the smallest are of quartz (Table 2.4.8). If we compared the averages of the 2GNB (Table 2.4.8) to the PB’s averages (Table 2.4.6), we observed that the dimensions of the 2GNB are larger than the BP’s dimensions (25,5x25,3x8,6mm). Quartzite is the raw material with more difference between PB’s size (28x29x8mm), and 2GNB size (78x69x25mm).

Average size of the 2GNB (mm) Raw material Length Width Quartz 28,4 27,8 Hornfels 52,5 46,5 Porphyry 49,9 47,3 Quartzite 78,1 69,5 Other 47,2 40,5 Global average 44,8 41,0

Thickness 14,0 15,4 15,5 25,4 13,4 15,7

Table 2.4.8.- Average size of the 2GNB of CDTM

The analysis of the technical attributes of the 2GNBG al-

Figure 2.4.10.- Second Generation Negative Bases of Configuration, from CDTM. 1: Quartzite with convex lateral dihedrons. Retouch is unifacial in the left side, and bifacial in the right side. There is bilateral symmetry, but no sagittal. General morphology is similar to a handaxe. 2: Uniangular (convergent) of quartzite, with a distal trihedron. Lateral retouches are denticulate

Figure 2.4.9.- Second Generation Negative Bases of Configuration, from CDTM. 1: Hornfels with a lateral dihedron with straight delineation. 2: Uniangular (convergent 2GNBC) of Porphyry. 3: Denticulate of hornfels. 4: Distal transverse dihedron of quartzite. 5: Denticulate of quartzite. 6: Concave denticulate of quartz (notch)

57

Xosé Pedro Rodríguez Generally the retouches extend for a fourth part of the perimeter of the object. The 2GNB with retouches that occupy all of the perimeter (4C) are very scarce. The angle of extractions is habitually acute (64%); the shallow-acute (14%), and steep angled retouches (11%) are not enough. The 2GNB of quartz have a greater presence of steep retouches (20%); there are no steep retouches among porphyry, but shallow-acute retouches reach the 21%.

Average size of the Fragments (mm) Raw material Length Width Thickness Quartz 16,5 12,5 6,8 Hornfels 26,1 20,2 8,6 Quartzite 26,4 20,1 8,1 Limestone 24,5 19,4 7,4 Porphyry 28,5 18,7 10,7 Lydite 20,3 17,9 8,4 Other 28,1 20,0 9,5

An equilibrium in the presence of deep (37%), and marginal retouches (36%) concerning the edge exists. The deep retouches overcome the marginal largely in quartz. Regarding the face where the negatives of extractions appear, the majority of the retouches are marginal (49%), although also the very marginal are frequent (28%).

Global average

17,2

13,1

7,0

Table 2.4.9.- Average size of the Fragments of CDTM

the rest of fragments (about 24-26 mm of length, 17-20mm of width, and 7-10 mm of thickness). Most of the fragments do not preserve rests of cortex. The completely cortical fragments are very few (3.3%), although some raw materials present a percentage superior to the global average. The greater amount of cortex appears among the fragments of hornfels and quartzite, while the less cortical fragments are of quartz and porphyry.

Retouches are direct in 49% of the cases, and inverse in the 34%. The quartzite objects are the only ones that do not fit in this pattern, since they have inverse retouches fundamentally. Generally, retouches are continuous, although there are 18% of denticulated retouches. Finally, the morphology of the retouched edge is fundmentally straight (37%), or else convex (23%), except in the 2GNB of porphyry, with predominance of the sinuous delineation.

Strategies for producing Positive Bases Several systems of production were put into practice in order to get Positive Bases, always from 1GNBE (that is, all the cores are 1GNBE) (Table 2.4.10). We have identified seven strategies for producing flakes.

We already have indicated previously that the amount of cortex is quite greater among the 2GNB than among the PB. Let’s remember that 34.8% of the 2GNB show remains of cortex so much on the butt like in the dorsal face. A 10% of the 2GNB were elaborated with completely cortical blanks, and a 14.5% on blanks with predominance of the cortex. These data would indicate a cortical PB’s selection to be retouched.

The majority of the cores are characterized by a knapping that affects the transverse distal plane (Figure 2.4.10), utilizing elongated and thick pebbles of medium or large size, basically of hornfels and quartzite. The exploitation reduces the volume of the object at this zone by means of removals of diverse direction (generally orthogonal). In other cases extractions were accomplished departing from the horizontal planes, and affecting the sagittal planes; that is, extractions of steep or medium angle that exploit the laterals of the 1GNBE. This reduction strategy has been observed in a 40% of the studied 1GNBE (concretely in 29 objects) (Table 2.4.10). This is a longitudinal unipolar or bipolar knapping. This method of knapping also is known as “salami slice technique” or “parallel slice” (Bourlon, 1907). Usually PB retain a perimeter of cortex. There are 1GNB, Positive Bases, and 2GNB, that correspond to this reduction strategy.

As to the strategies of knapping, among the quartz objects, there are blanks with unipolar linear, and bipolar opposed knapping fundamentally. We have identified unipolar linear, bipolar opposed and multipolar orthogonal (with tendency to centripetal) knapping among the 2GNB of hornfels. Also we have located 2GNB coming from a longitudinal Operational system with systematic reduction of volume, through extractions in the transverse, and/or sagittal planes. The majority of the 2GNB of porphyry show scars of unipolar linear type in his dorsal face. Among the 2GNB of quartzite a greater diversity exists, with strategies of longitudinal type with systematic reduction of the volume, and also bipolar opposed. The existence of a 2GNB of large size on a fundamentally cortical blank stands out.

These cores have the most larger dimensions: the cores with extractions fundamentally in the transverse distal face have 110x91x63mm of average, while the ones that concentrate the exploitation on one of his sides (sagittal planes) have 96x67x47mm of average. As to the raw materials, the hornfels and quartzite are the most frequent raw materials, with 10 and 8 cores respectively. Also there are 1GNBE of this type knapped with porphyry (n=5), and with quartz (n=3).

Fragments Fragments are the most numerous objects in the studied sample. The vast majority of the fragments are of quartz (91.9%). There are few fragments of the rest of raw materials. The average of the dimensions of all the fragments is of 17x13x7mm (Table 2.4.9). Nevertheless, some differences depending of the raw material exist. The smaller fragments are of quartz (16x12x6 mm); something larger size present

The centripetal knapping also is frequent (Figure 2.4.7): 21 cores present centripetal removals in any one of their faces (29.2 % of the 1GNBE, whose strategy of knapping

58

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

Faciality Unifacial Bifacial Trifacial

Bifacial

Bifacial

Bifacial Bifacial Unifacial

Trifacial

Bifacial

Bifacial Bifacial Unifacial

FIRST GENERATION NEGATIVE BASES OF EXPLOITATION (CORES) Knapping Method Raw material hornfels: 10 (34,5%) quartzite: 8 (27,9%) longitudinal, unipolar or bipolar, masive, with removals porphyry: 5 (17,2%) in the transverse and/or sagital planes quartz: 3 (10,3%) indet.: 3 (10,3%) hornfels: 4 (50%) quartz: 2 (25%) centripetal /linear, opposed or orthogonal quartzite: 1 (12,5%) porphyry: 1 (12,5%) quartz: 2 (33,3%) sandstone: 1 (16,7%) centripetal / centripetal quartzite: 1 (16,7%) hornfels: 1 (16,7%) hornfels: 1(50%) centripetal with a conical volume or troncoconical porphyry: 1 (50%) hornfels: 2 (50%) centripetal with hierarchization and predetermination quartzite:1 (25%) sandstone: 1 (25%) quartz: 1 (50%) centripetal quartzite: 1 (50%) Total centripetal cores quartzite: 2 (40%) hornfels: 1 (20%) linear, opposed, orthogonal quartz: 1 (20%) flint: 1 (20%) quartz: 3 (50%) multip. orthogonal or opposed / quartzite: 1 (16,7%) multip. orthogonal or opposed, linear hornfels: 1 (16,7%) porphyry: 1 (16,7%) hornfels: 2 (66,7%) unip. linear / unip. linear porphyry: 1 (33,3%) bip. opposed with removals accomplished from the two hornfels: 2 (66,7%) distal extremes quartzite: 1 (33,3%) quartz: 2 (50%) unip. linear or bipolar orthogonal sandstone: 1 (25%) porphyry: 1 (25%) Not identified Total Negative Bases of Exploitation (All of First Generation)

Objects

29

40,3%

8

11,1%

5

6,9%

2

2,8%

4

5,6%

2

2,8%

21

29,2%

5

6,9%

6

8,3%

3

4,2%

3

4,2%

4

5,6%

1 72

1,4%

Table 2.4.10.- First Generation Negative Bases of Exploitation (1GNBE) of CDTM, grouped in terms of the knapping method (Indirect Operational Themes). We point out the raw materials utilized for each group of cores

could have been identified). The majority have centripetal removals in one of their faces, and extractions of orthogonal type, opposite or linear in the other face (8 objects, the half of hornfels). This strategy, with 1GNB, PB, and 2GNB, does not require a specific preparation of the matrix. Among the centripetal cores also there are two unifacials (of quartz and quartzite).

or troncoconical (consequence of removals of medium or steep angle) (Figure 2.4.7.1). The half of these cores (n=3) are of hornfels, and the rest of quartzite, porphyry and sandstone. Among these six objects there are four with predetermination and hierarchization of the faces of the core: a face is of exploitation (flaking surface), and the other of preparation. This is a reduction strategy with predetermination of the final morphology of the flakes (similar to the Levallois Method). These cores with preconfiguration are of hornfels (n=2), of quartzite, and of sandstone. Quartz was not used for this Operational Theme. This reduction strategy also shows a complete Operational Sequence, since there are 1GNB, PB, and 2GNB.

The centripetal knapping in both faces appear in five cores, distributed among various raw materials (Table 2.4.10). At least two of these objects have a biconical (biconvex) structure, similar to the discoid cores. The 1GNB, and the PB have allowed identifying this strategy, however it’s possible that retouched products ascribed to this Operational Theme exist also.

There are 5 trifacial cores with dominant orthogonal knapping, associate to bipolar and linear knapping (6.9%): two of quartzite, one of quartz, one of silex (Figure 2.4.7.2), and another one of hornfels. Their size is smaller than the rest of 1GNBE: 67x61x52mm. It is difficult to ascribe to this strategy of knapping PB and 2GNB, although it seems probable that to the less exist PB.

There are six 1GNBE with centripetal knapping that shows a plano-convex morphology, with one of their faces more knapped than the other. The most knapped face always is plano-centripetal, while the other can be centripetal, orthogonal and even linear, but always with a conical volume 59

Xosé Pedro Rodríguez The first group consists of 15 artifacts with unifacial extractions (choppers), that configure transverse dihedral cutting edges. These cutting edges have a straight, or lightly convex delineation (Figures 2.4.3.1, 2.4.3.2, 2.4.4.1, 2.4.4.2, and 2.4.5.1). From a traditional typological viewpoint this objects would fit in the choppers’s group. Nine of these objects are of quartzite, three of hornfels, two of sandstone, one of limestone, one of porphyry, and another one of an indeterminable rock.

A group of bifacial cores shows orthogonal or opposed knapping in one of their faces, and linear, orthogonal or opposed in the other face. These cores do not have preconfiguration, and preserve some cortical zones. Their exploitation never gets to be total. Three of the six objects that fit in this pattern are of quartz. The knapping of three bifacial cores with linear extractions in both faces is less complex. Also there are three cores with bipolar opposed knapping, with removals accomplished from the two distal extremes of generally elongated pebbles.

The second group of 1GNBC is composed by seven artifacts with similar characteristics to the previous, but with a bifacial configuration (chopping tools) (Figure 2.4.5.3). These objects are of quartzite (n=2), hornfels (n=2), sandstone (n=1), porphyry ( n=1), and a not identified rock (n=1).

Finally, we have identified four unifacial 1GNBE with linear knapping or else orthogonal. The 1GNBE preserve remains of cortex (Table 2.4.10). We verified the existence of PB and 2GNB correspondent to this strategy.

Another Direct Operational Theme consists in the shaping of unifacial uniangular (convergent) objects (Figure 2.4.5.2). From a typological point of view this artifacts could be classified as picks. These 1GNBC are of hornfels and of sandstone.

It is necessary to emphasize that an interrelation between all the centripetal reduction strategies can exist. Also it is interesting to point out the difficulty to identify to what Operational Theme the 2GNB belong, because of the similitude that exists between some Operational Themes byproducts.

The average of the dimensions of the unifacial objects with distal transverse cutting edges (straight or convex), is of 118x78x53 mm, while the bifacials with the same type of cutting edges are slightly less lengthy, but some more wide and thick (107x84x56 mm). Therefore, these are objects of large size. As to the uniangulars, their size is reduced as compared with the others two groups (87x58x42 mm), so that we would be able to catalogue them as artifacts of medium size.

The configuration of tools We have identified 109 instruments that have been retouched; 24 are 1GNBC, and 85 are 2GNBC. The raw material with more shaped instruments is quartz (n=34), followed by hornfels (n=22), quartzite (n=20), and porphyry (n=15) (Table 2.4.11). If we compared the number of shaped objects with the total of pieces out of every raw material (without fragments, neither indeterminable) we observed that the 34 shaped artifacts of quartz only suppose a 7.6% of all the quartz objects, but quartzite, porphyry and sandstone instruments suppose the 14.3%. It seems evident that there was not a special interest to retouch quartz instruments. In fact there is no 1GNBC of quartz; all quartz instruments are 2GNBC. The raw material with the most elevated percentage of shaped tools on 1GNB is the sandstone, followed by the quartzite. However, when working with hornfels and with porphyry there is preference to retouch flakes, producing 2GNB. Globally, the 109 shaped artifacts suppose the 10.2% of the determinable lithic industry (without fragments).

The Operational Sequence is represented in both cases by totally cortical Positive Bases, or PB with cortex’s predominance. It’s possible that any one of these products had been retouched later. Flake tools The PB of more than 20mm of maximum length, and all the 2GNB have been accumulated on in order to calculate the initial number of flakes, next we have calculated the percentage of 2GNBC relating to this number. The result suggests that a 12.3% of the initial products were retouched. This percentage varies slightly in terms of the raw material. The 12.1% of the initial quartz flakes, 9% of hornfels, 12.1% of quartzite, and 14.8% of porphyry were retouched. The basic objective in the case of quartz was the production of PB and no its later configuration.

Pebble tools Quartzite and hornfels were the most utilized rocks to shape tools on pebble. We have identified three types of configuration, that can be included in two main Direct Operational Themes.

An important fact is that a significant part of the 2GNBC preserve cortex in the butt, and in the dorsal face (nearly the 35%). This circumstance would suggest a selection of cortical flakes for their configuration like retouched instruments. But how are the configured instruments?

The first Direct Technical Operational Theme (DTOT) has as objective to produce unifacial (choppers) or bifacial instruments (chopping tools), with transverse straight or lightly convex cutting edge.

The primary objective of the configuration of 2GNB is to form dihedral cutting edges. In total 45 artifacts present this morfopotential model as a result of the retouches (62.5%) 60

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

Figure 2.4.11.- Morphogenetic matrix of the lithic industry of CDTM

(Table 2.4.12). The straight or convex dihedrons placed in a lateral are more frequent (n=25) (Figure 2.4.9.1) than the concave dihedrons (n=8) (Figure 2.4.9.6). Also there are straight or convex transverse dihedrons (5 objects) (Figure 2.4.9.4), and lateral-transverse dihedrons (7 objects).

on pebble, dihedral cutting edges at the transverse distal zone were created, by means of unifacial (choppers) or bifacial extractions (chopping tools). Sometimes also trihedrons (picks) were configured. The configuration of 2GNB does not have the same objectives. Fundamentally dihedral straight cutting edges or else convex were configured in the lateral, although also the shaping of straight or convex denticulated cutting edges is frequent. The generation of straight or convex dihedrons affects 51.4% of the 2GNBC, while the denticulation is present at the 26.4% (we included the uniangular or convergent denticulate here). The creation of trihedrons appears in 6.9 % of the 2GNBC.

The configuration of trihedrons appears in five 2GNB (6.9%) (Figure 2.4.9.2). In addition, lateral dihedrons associated to a distal trihedron were created in three objects (3.5%), maintaining a bilateral symmetry (Figure 2.4.10.1). The result is a morphology similar to a not much elaborated handaxe (lacking sagittal symmetry). The shaping of denticulated cutting edges has been observed in 19 artifacts (26.4%) (Figure 2.4.9.3, and 2.4.9.5). The denticulated straight or convex cutting edges (n=16) are more abundant than the uniangular (convergent) (n=3). Finally, three objects have shallow-acute angled retouches in the ventral face, whose objective is to reduce their thickness.

It is necessary to quote the existence of 3 2GNBC with lateral straight-convex dihedrons, associated to distal trihedrons. The morphology of these artifacts, with bilateral symmetry, is related with handaxes. Adding up the 1GNBC and 2GNBC, we arrived to the conclusion of that the creation of dihedral cutting edges at the transverse distal zone, and in the lateral of the objects has been the basic objective of the configuration of instruments (with a 61.5%). Follow in importance the shaping of denticulated cutting edges (19.8%), and of concave dihedrons (8.3%). The configuration of trihedrons implicates 7.6% of all the instruments. Finally, the bifacial instruments with bilateral symmetry suppose a 3.1% (Table 2.4.12).

From a typological point of view, we could have classified 69 of the 85 2GNB according to Laplace’s analytical typology (81.2%). The group of the side scrapers is the most numerous, with 30 objects (43.5% of classified 2GNB). The denticulates are 25 (36.2%), and the steep angled 2GNB are four (5.8%). There are also three end scrapers (4.3%), two points (2.9%), and three objects with a morphology similar to handaxes. The specific type most habitual is the lateral side scraper (R1) with 23 objects, followed by the denticulate side scraper (D3) with 11, and for the notch (D1) with 7. Lateral-transverse side scrapers (R3), and denticulate of type “épine” (D2) complete the group of the most numerous instruments (5 and 4 objects respectively).

Quartz Hornfels Quartzite Porphyry Sandstone Other TOTAL

In the Table 2.4.12 we have summarized the morphodynamic capacity of all the retouched tools, shaped on 1GNB and 2GNB. When the objective was to shape instruments

0 6 9 2 4 3 24

1GNBC 0,0% 27,3% 45,0% 13,3% 80,0% 23,1% 22,0%

2GNBC 34 100,0% 16 72,7% 11 55,0% 13 86,7% 1 20,0% 10 76,9% 85 78,0%

TOTAL 34 31,2% 22 20,2% 20 18,3% 15 13,8% 5 4,6% 13 11,9% 109

Table 2.4.11.- Structural categories and raw materials of the configured (or retouched) instruments of CDTM

61

Xosé Pedro Rodríguez Morphodynamic capacity of the retouched tools Transversal dihedral edge, with straight-convex morphology Lateral dihedral edge, with straight-convex morphology Transversal and Lateral dihedral edge, with straight-convex morphology Concave dihedral edge (“notch”) Bilateral dihedral edge + distal Trihedral edge (“handaxe”) Distal Trihedral edge Straight-convex denticulate edge Uniangular (convergent) denticulate edge (“denticualte point”) Total identified

1GNBC 22 91,7% 0 0,0% 0 0,0% 0 0,0% 0 0,0% 2 8,3% 0 0,0% 0 0,0% 24 25%

2GNBC 5 6,9% 25 34,7% 7 9,7% 8 11,1% 3 4,2% 5 6,9% 16 22,2% 3 4,2% 72 75%

TOTAL 27 28,1% 25 26,0% 7 7,6% 8 8,3% 3 3,1% 7 7,6% 16 16,7% 3 3,1% 96

Table 2.4.12.- Morphodynamic capacity of the retouched tools of CDTM

Conclusions

for the systematic production of PB. Some scientists have interpreted CDTM like a Middle Pleistocene site dedicated basically to the hunting of horses and to the lithic knapping (Canal and Carbonell, 1989).

The lithic industry of Cau del Duc de Torroella de Montgrí (CDTM) is characterized for a high number of fragments of quartz. Excluding fragments, the Positive Bases are the most most habitual objects (80% of the objects fit in this structural category). It seems quite obvious that the principal objective was PB’s production. The quartz was used as basic raw material, although also the hornfels, quartzite, porphyry and other rocks were used. The supply of raw material was accomplished close to the site, making good use of the Ter river bed (to less than 2 km from the site). Several methods of knapping were put into practice for producing Positive Bases. The longitudinal massive knapping, the centripetal knapping (predetermined knapping included), and the linear/opposed/orthogonal knapping stands out among them. Instruments explicitly shaped with retouches (1GNBC and 2GNBC) are scarce, and frequently cortical blanks were retouched. This would indicate a secondary role of the retouch of flakes, always as an alternative to PB’s systematic production. On the oher hand, there is a selective utilization of the raw materials: • Quartz was used almost always in processes of production. Quartz objects rarely were configured as instruments. • Hornfels also was used preferentially in processes of exploitation. • Quartzite, porphyry and sandstone were used in processes of production of flakes, most of all when the elected strategy requires a raw material of greater quality than quartz. Preferably the cortical products of these raw materials were retouched. Processes of exploitation and configuration are well documented in CDTM’s lithic record. We have analyzed objects of the various phases (Operational Units) of the Operational Sequence; from the beginning of the knapping (with an outstanding number of cortical products) to the configuration of instruments with the retouch of flakes (2GNBC, accompanied by elevated number of PB’s with less than 20 mm). This site has some features of Mode 2 lithic technology, but also there are traits more related with Mode 3, specially the significant presence of predetermined knapping 62

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

2.5.- NERETS

made easy that some erosive agents act on the paleo-relief and its dismantling began. Exokarstic morphologies (karren) at present quite developed appeared, because there was a substratum of sandstones with great quantity of carbonates.

Location and geological context Nerets is located in the municipality of Talarn (province of Lleida, Catalonia) (0º 55’ E, 42º 10’ N) (Figure 2.5.1). This site was discovered in 1989, when lithic instruments were found accidentally at the slopes of a hill, in the surface. A rescue excavation was achieved in 1995 (Figure 2.5.2). Nerets’s hill has a maximum height of 625 meters a.s.l., with a good visibility on the valley of the river Noguera Pallaresa.

During the Pleistocene a series of soils were developed on the surface of Nerets. At a later time, in the colddest periods of the upper Pleistocene and Holocene, the vegetation was less abundant. This circumstance facilitated the progressive erosion of the ancient soil of the Middle Pleistocene, and the reactivation of the system of karren and streams. As a result, the remains of the activities of the hominids left dispersed for all the slope. It is very possible that in the bed of the ancient ravines lithic industry in primary context keep in good condition. However, after the soundings realized during the excavation of 1995 we could verify that the reforestation of Nerets’s mountain in modern epoch distorted the surface record enormously, and removed it from the original context. As a result, the lithic implements appear in surface across all the hill’s slope, particularly at the small ravines formed by erosion (Rodríguez and Rosell, 1993).

Nerets is located at the “Conca de Tremp” (Basin of Tremp), one of the structural units of the occidental sector of the Catalan pre-Pyrenees. The geological formation named “Formació Nerets” consists of a kind of sandstone with a great contents of flint and cemented quartz. The presence of these materials in the sandstones, as well as their small size and their grade of erosion, evidence the aeolian formation of the rock. Nerets is attached to an ancient mesozoic relief, previous to the formation of Tremp’s Basin, pertaining to an inland. Although this coastal paleo-relief is at present deformed by the Alpine foldings, the large accumulations of aeolian material are related to the existence of paleodunes, to few meters of an ancient marine coast. Under this substratum we can recognize, in some places where erosion has been intenser, a stratum of very cemented sandstones of a reddish color. In this stratum there are numerous fossils of dinosaurs.

Discovery and archeological interventions Nerets’s site was discovered fortuitously in 1989 by a local amateur (Rosell and Rodríguez, 1991). The administrative difficulties impeded the realization of archeological interventions until 1995 (conducted by Jordi Rosell). The objective was to accomplish a systematic survey and to make soundings in the places where there was a greater concentration of lithic material. Quadrants of 125x125meters were established in order to carried out the prospection,

After the Alpine movements that elevated the Pyrenees, Nerets’s original paleodunes also experienced a vertical uprising, that conferred a bigger slope to this relief. This

Figure 2.5.1.- Location of Nerets

63

Xosé Pedro Rodríguez

Figure 2.5.2.- General view of the Nerets hill. Lithic artifacts were found at the slopes of this hill. The arrow indicates the place where an excavation took place in 1995

taking like reference a topographical map at scale 1:5.000. A surface of approximately 85 hectares was prospected during a week (Figures 2.5.2 and 2.5.3). The quadrants with more material were located in the higher part of the hill. The richest quadrant provided 110 pieces, but the majority offered between 20 and 40 objects. The quadrants of the middle part of the hill provided very little material. The density of vegetation made the prospection difficult at this sector. In the downside of the hill at a little ravine, an excavation was accomplished, with the intention of finding material in stratigraphic context. At first the excavation affected six square meters, that at a later time were enlarged to 16 m2.

Stratigraphy From bottom to top, the stratigraphic column consists of a package of conglomerates with big heterometric pebbles, polygenics and very rounded, with a matrix of silts and very fine sands (Figure 2.5.4). The lithic industry was found in this deposit. On top was a centimetric package of sands, fine to middle-sized, with clays. Part of the lithic industry also was in this level. The top was the present-day soil. The thickness of the stratigraphic succession was of 60 cm.

Analysis of the lithic industry We analyzed 1009 lithic objects (Table 2.5.1). This material come from no systematic prospections accomplished in 1989, and fundamentally from a systematic prospection, and an excavation accomplished in 1995. In spite of his diverse procedence, all material presents a great homogeneity. Hence we have not distinguished between the material of surface, and the one belonging to the excavation.

Figure 2.5.3.- Topographic map of the Nerets hill (contour lines each 5 meters), with the location of prospected zones. Each quadrant has 125x125 meters. The color of the quadrants indicates the quantity of artifacts that were found. Also we have located the place where an excavation was accomplished (quadrant E10)

64

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA We have observed that hornfels, sandstone and quartz have his implements distributed among the nB, the 1GNB, the PB, and Fragments. However, the quartzite concentrates over half of their objects in the Positive Bases. The percentage of quartzites utilized like natural Bases is comparatively lesser than the percentage of hornfels, sandstone and quartz. Also the percentages of hornfels and sandstones in the category of 1GNB are notable (fundamentally 1GNBC). A first hypothesis, propose the preferential utilization of the quartzite for the production of Positives Bases, and of the hornfels and the sandstone to shape 1GNBC (that is pebble tools). This hypothesis also would be validated because of the less proportion of Positive Bases among hornfels and sandstone, in front of the greater number of quartzite PB’s . Morphotechnical Analysis Natural Bases Natural Bases are near the 10% of all lithic industry (this percentage increases to the 12.1% without Fragments, neither the indeterminable objects). It would be able to seem imprudent to identify natural Bases in a surface site, however assuming that the utilized raw materials do not appear spontaneously in all of the Nerets’s hill, we could have inferred that they were transported that far by the humans.

Figure 2.5.4.- Schematic stratigraphic column of the excavated zone in Nerets. 1: boulders, 2: sands and clays, 3: silts and sands, 4: presentday soil. The drawing indicates where the lithic industry appeared

The majority of the natural Bases are of type “c” (nBc), that is fractured pebbles (67.8%). The natural Bases without marks (nBa) are the 10%, while the nB with marks of percussion (nBb) barely surpass the 7%. There are a 15% of objects that could not be ascribed to a sub-category, fundamentally because of their degradation by erosion, or else to be covered by concretion.

Raw materials We have found lithic industry knapped with various raw materials (Table 2.5.1). However, there is a predominance of quartzite (almost 80 % of all the material). Also the utilization of hornfels, sandstone and quartz stands out. The rest of raw materials do not get to the 1%, except flint, that obtains this percentage. Nevertheless, the presence of instruments of flint can be consequence of a different dynamics. The characteristics of the flint objects differ from the artifacts manufactured with other raw materials.

Most of the nautral Bases are of quartzite, although nB only supposes a 6% of all of the objects of this rock. Natural Bases of hornfels, and of sandstone are scarce in number, but significant in proportion for each rock. Concretely, the 28% of the pebbles of sandstone, and the 23% of hornfels were not knapped, but utilized like natural Bases. It’s evident that these hominids preferred to utilize the quartzite to produce another type of objects, while the raw materials of less quality were more utilized like nB.

In general these raw materials appear in the river Noguera Pallaresa present-day bed, or else in any one of his ancient terraces, close to the site. However, at Nerets’s hill, in where is the site, no one of the utilized rocks appear spontaneously. This criterion has helped us to identify the existence of pebbles taken there by the humans, although not knapped (natural Bases). nB Quartzite Hornfels Sandstone Quartz Schist Limestone Slate Porphyry Flint Indet Total

52 17 14 6 0 1 1 2 0 3 96

6,5% 18,7% 28,6% 23,1% 0% 20% 12,5% 100% 0% 20% 9,5%

1GNBC 41 5,2% 14 15,4% 9 18,4% 0 0% 0 0% 1 20% 0 0% 0 0% 0 0% 2 13,3% 67 6,6%

1GNB PB 1GNBE 1GNB Ind. 53 6,7% 3 0,4% 424 53,3% 3 3,3% 2 2,2% 24 26,4% 1 1 9 18,4% 2% 2% 2 7,7% 0 8 30,8% 0% 0 0 3 42,9% 0% 0% 0 0 2 0% 0% 40% 0 0 4 0% 0% 50% 0 0 0 0% 0% 0% 1 9,1% 0 4 36,4% 0% 1 6,7% 0 3 0% 20% 61 6,0% 6 0,6% 481 47,7%

2GNB FRAG INDET TOTAL 2GNBC 2GNBE 67 8,4% 5 0,6% 146 18,4% 4 0,5% 795 78,79% 91 5 5,5% 0 0% 22 24,2% 4 4,4% 9,02% 49 2 4,1% 0 0% 12 24,5% 1 2% 4,86% 26 0 9 34,6% 1 3,8% 0% 0 0% 2,58% 7 4 57,1% 0 0 0% 0,69% 0% 0 0% 5 0 0 0% 0 0% 0% 1 20% 0,50% 8 0 3 37,5% 0 0% 0 0% 0% 0,79% 2 0 0 0% 0 0% 0% 0 0% 0,20% 11 4 36,4% 0 2 18,2% 0 0% 0% 1,09% 15 0 4 26,7% 2 13,3% 0% 0 0% 1,49% 78 7,7% 5 0,5% 202 20,0% 13 1,3% 1009

Table 2.5.1.- Lithic industry of Nerets (structural categories and raw materials)

65

Xosé Pedro Rodríguez als (fundamentally hornfels and sandstone), and we have not appreciated significant differences.

The natural Bases of larger size are of type “b” (if we excepted the unclassifiable nB), with an average of 110.1 x 79.1 x 40.1 mm. Follow in size the fractured Bases (105.9 x 76.1 x 43.7 mm), while the ones that do not show marks are the smallest (80.9 x 58.1 x 34.1 mm).

Positive Bases Nearly a 50 % of the objects recovered in Nerets are flakes. The 88.1% of the PB were knapped with quartzite, and the 5% with hornfels. The rest is distributed between various rocks.

First Generation Negative Bases The 134 1GNB recovered in Nerets suppose a 13.3% of all the objects (the 16.9% without Fragments, neither indeterminable objects) (Table 2.5.1). Most of Nerets’s 1GNB are of quartzite, although also the utilization of hornfels, and of sandstone stands out. 1GNB of Configuration are lightly further numerous (n=67) than 1GNB of Explotation (n=61). Six 1GNB could not be identified as cores (1GNBE) neither as pebble tools (1GNBC) (Table 2.5.1).

The average of the size of the PB is 46,9x44,0x16,3mm (Table 2.5.3). The size of the PB of hornfels is considerably larger than the PB of quartzite. The explanation of this difference is that the majority of the PB of hornfels come from 1GNBC of large size, while the PB of quartzite proceed so much from 1GNBC of large size as from 1GNBE. In general, it is pertinent to highlight the scarce number of PB of very small size: only there are 5 PB with length and width of less than 20mm (barely the 1%). This circumstance is related to the scarcity of debris derived from processes of configuration of 2GNB. On the contrary, 31.1% of the PB have a length or a width of more than 60mm.

We have observed an obvious difference in the utilization of the quartzite regarding the hornfels and the sandstone. The quartzite is used in similar percentages to retouch pebbles (1GNBC), and to elaborate cores 1GNBE), in order to the production of PB; The hornfels and the sandstone were used preferentially to shape 1GNBC. Very rarely the 1GNB of sandstone and of hornfels were exploited as cores.

Average size of the PB (mm) Raw material Length Width Quartzite 45,6 43,6 Hornfels 66,8 59,8 Sandstone 47,6 39,0 Global average 46,9 44,0

The size of the 1GNBC is greater than the 1GNBE (Table 2.5.2). The instruments configured on pebble use to have a large size. Among the 1GNBC, the larger size corresponds to the artifacts of hornfels. Exactly the same happens with the 1GNBE: the hornfels is always the raw material with the objects of largest dimensions.

Table 2.5.3.- Average size of the PB of Nerets

Morphotechnical analysis indicates that 1GNB of Exploitation were knapped generally throughout their perimeter (4C), with extractions of acute or medium angle. The removals frequently occupy the face where they have taken effect totally. The sagittal sinuous edges, and no symmetric regarding the horizontal plane, predominate.

We have identified 59 flakes with a laminar index greater than 1.60, what supposes the 13.6% of all PB. The 18.7% of the PB have a cortical and no faceted butt. This percentage is greater in the PB of hornfels, with 70% of cortical butts. The percentage decreases to a 14% in quartzite. If we included the 2GNB in this analysis, the global percentage of cortical butts roughly changes (20.1%). However, in the 2GNB the presence of cortical butts is very important, with a 31.9%. Its influence in the global calculation is not significative because the number of 2GNB is scarce, compared with the number of Positive Bases.

The analysis of the morphotechnical attributes indicates that the majority of the 1GNBC are unifacial. The perimeter of the objects is scarcely knapped (between ¼ and ½). A preferential angle to execute removals does not exist. As to the extent of removals, the extensive and the marginal are the most frequent. Rarely extractions are total, and eliminate all of the cortex of the configured face. The sagittal edge is by majority incurvated or else straight. Finally, the sagittal edge is generally symmetric regarding the object’s horizontal plane. We have compared the technical characters of the quartzite objects, and other materiAverage size of the 1GNB (mm) Raw material Length Width All 1GNBC 125,4 99,7 1GNBC of quartzite 121,4 103,2 1GNBC of hornfels 151,4 98,3 All 1GNBE 71,4 64,8 1GNBE of quartzite 68,8 65,0 1GNBE of hornfels 123,0 70,8 Global average 101,1 83,4

Thickness 16,3 17,8 15,9 16,3

The presence of bifaceted and multifaceted butts stands out particularly in the PB of quartzite (32.7%). However, there are no Positive Bases of hornfels with bifaceted or multifaceted butts. These data indicate that the hornfels is not used to produce PB by means of reduction strategies that imply a specific preparation of the cores, so that it happens with quartzite. As a whole, a 30% of Nerets Positive Bases have bifaceted or multifaceted butts. The PB that later were converted in 2GNB have a slightly greater percentage (31.9%). The percentage of bifaceted plus multifaceted butts in all the initial products of knapping (PB+2GNB) is of 30.2%. The vast majority of the butts are of platform type.

Thickness 49,3 50,2 47,6 37,6 36,9 44,2 44,2

Table 2.5.2.- Average size of the 1GNB of Nerets

66

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA We have observed that nearly a fifth part of the PB (19.9%) has total or dominant cortex in his dorsal face, while this percentage is of 29.5% in the 2GNB. Globally, 21.1% of the initial products have dorsal faces with total cortex or dominant. If furthermore we added the products with no dominant cortex, we got to 40% of PB and 42.6% of 2GNB. Adding up PB and 2GNB we arrived to the conclusion that 40.2% of the products of knapping have cortex’s remains in their dorsal face. Cortex’s remains are very frequent among the PB of hornfels (60.9% of these PB have some corex’s rest). The fact that the amount of cortex be more important among the PB of hornfels reinforces the hypothesis about the procedence of these products, that would be the result nearly exclusively of the shaping of 1GNB. The quartzite products also present significant cortex’s percentages, but it fit well with the fact that aforementioned products come so much from the configuration such as from the exploitation of cores. There are 6 PB with butt and dorsal face completely cortical, relating to 399 analyzable in what relative to these attributes (1.5%). This percentage is greater among the 2GNB, 7.3% (three completely cortical objects relating to 41 analyzable). Adding up the PB’s data and 2GNB we have 9 products of knapping completely cortical, relating to a total of 440 analyzable, that is a 2%. The initial products (PB + 2GNB) that have cortical butt and dorsal face with dominant cortex are the 5.9%. The 11.8% of PB and the 14.5% of 2GNB have some kind of break. The percentage of initial products of knapping (PB + BN2G) with some kind of break is 11.9%.

Figure 2.5.5.- Positive Bases of Nerets. 1, 2, 3, 6: Quartzite with predetermined morphology. 4: Quartzite with remains of cortex. 5: Centripetal PB of quartzite

There are 57 BP with predetermined morphology, what means 11.8% of the PB (Figures 2.5.5.1, 2.5.5.2, 2.5.5.3, and 2.5.5.6). Also there are 8 2GNB of this type (9.6%). Globally we have identified 65 predetermined products (Levallois), relating to a total of 564 (11.5%). Seven of these 65 products have triangular morphology. The study of the size of the products with predetermined morphology indicates that the 47.7% have a maximum dimension between 40 and 54 mm.

straight cutting edges. In synthesis, it seems that the principal objective was the systematic production of PB of medium size, with dihedral cutting edges, utilizing quartzite most of all. We have located in Nerets few PB of very small size, that would correspond to the retouch of the 2GNB. Nevertheless, these objects are difficult to locate in surface prospections, and in addition they should not have been excessively abundant, since there are few 2GNB in Nerets (only 8 % of all the lithic industry).

In Nerets there are PB correspondent to different phases of the Operational Sequence, from preparation and/or configuration of the 1GNB (in quartzite and hornfels) to their exploitation (fundamentally in quartzite).

Second Generation Negative Bases The 83 objects classified as 2GNB constitute around the 8% of Nerets’s lithic industry (without Fragments they are the 10.5%). The majority of Nerets’s 2GNB are of quartzite (86.7%). Also there are 2GNB of hornfels (6%), sandstone (2.4%), and flint (4.8%).

Also we have located PB originating from centripetal Operational Themes (Figure 2.5.5.5), and from strategies of laminar production (three objects). In this sense two PB of preparation of cores destined to the production of PB of laminar tendency have been found. The presence of PB of large size with very operational cutting edges stands out. Concretely, there are two PB similar to cleavers, and six with a morphology less clearly related to the cleaver standard, but with an identical morphopotential structure, that is with excellent lateral- transverse

Of the total number of 2GNB found, 78 were 2GNB of Configuration, and 5 2GNB of Exploitation. The analysis of the dimensions of the 2GNB shows

67

Xosé Pedro Rodríguez Average size of the 2GNB (mm) Raw material Length Width Hornfels 82,5 74,0 Quartzite 72,3 62,6 Flint 26,8 21,4 Global average 71,4 63,1

Fragments Fragments are the 20% of the lithic industry recovered in Nerets. Most of the fragments are of quartzite (72.3%). In general, fragments are of small and medium size, with an average of 59,1x39,9x23,2mm (Table 2.5.5).

Thickness 18,1 26,0 8,5 25,3

The quartzite fragments generally are smaller than the fragments of hornfels and sandstone. The fragments of hornfels have the largest size.

Table 2.5.4.- Average size of the 2GNB of Nerets

some interesting issues. The average dimensions of the 2GNB (once the fractured pieces were excluded) are of 71.4x63.1x25.3mm (Table 2.5.4). This average is clearly greater than the average of PB (Table 2.5.3). The 2GNB of hornfels are larger than 2GNB of quartzite. This phenomenon also has been observed in the PB. These data would indicate that the products of medium size were selected to be retouched. The 2GNB of flint are of very small size, and have characteristics different to the rest. These objects do not fit well with the rest of lithic industry, and probably stem from another type of human occupations.

More than 50% of the fragments have some cortex’s rest. A 27.2% of the fragments have dominant cortex, and among these there is a 11.8% with total cortex. The amount of cortex is therefore important, such as it happened in the PB and the 2GNB. Average size of the Fragments (mm) Raw material Length Width Thickness Quartzite 56,9 38,5 22,6 Hornfels 76,9 45,4 25,1 Sandstone 60,6 46,3 24,4 Global average 59,1 39,9 23,1

The study of the technical attibutes of 2GNB shows an important presence of cortical butts (nearly a third part of the total), greater than the percentage observed in PB. At the same time there is an outstanding role of multifaceted butts (25.5%). So then, there are similar percentages of no faceted, unifaceted and multifaceted butts.

Table 2.5.5.- Average size of the Fragments of Nerets

Strategies for producing Positive Bases The processes of exploitation includes 66 cores for producing flakes (61 1GNBE, and 5 2GNBE). Nearly the 90% of the 1GNBE are quartzite objects. The utilization of hornfels and another materials is scarce. In general, the 1GNBE have medium and small size.

The percentage of completely cortical dorsal faces is greater among the 2GNB than among the PB (19.7%, in front of 7% of the PB). Furthermore, as to the products with dominant cortex, there are a 20% among the PB, and a 29.5% among the 2GNB. Other datum reinforces the general impression of a greater amount of cortex among the 2GNB: three of the nine initial products completely cortical (butt and dorsal face completely cortical), were retouched and converted in 2GNB.

Five of the cores are 2GNBE (cores on flake) of quartzite. Four have maximum dimensions between 69 and 77 mm. The fifth object is smaller, and is practically exhausted (46x44x22mm). We could have ascribed 55 1GNBE and 5 2GNBE to 10 different knapping strategies (Table 2.5.6). There are only six cores that do not fit well in no one of these strategies.

These data indicate that the cortical products also were selected to be retouched, just like some predetermined PB (there is a high percentage of 2GNB with multifaceted butts). We have identified at the very least 8 2GNB shaped on predetermined blank. This means that 12.3% of the predetermined products were retouched. This type of objects suppose the 9.6% of all the 2GNB.

1) Indirect Technical Operational Theme of preconfiguration of flakes (Levallois) The most habitual strategy consists in the preconfiguration of the final morphology of the flakes by means of a bifacial centripetal exploitation. This strategy has been identified in 24 BN1GE (Table 2.5.6). A hierarchization of the faces exists. Extractions that had like objective to prepare platforms of percussion were realized in one of the faces. The removals of second face were carried out from these prepared platforms. The principal objective is to arrange the morphology of this second face to extract PB with predetermined characteristics. Generally the chosen face for the systematic extraction of Positive Bases was completely knapped, while the face of preparation only has peripheral extractions (there are cortical zones). However, in 8 cores the two faces are completely knapped, and there are not cortex’s remains (Figure 2.5.6.1). We have found objects of medium and small size among

A selection that affects the size of the products seems to exist, so that the average size of the 2GNB are clearly larger than the average of PB. The breaks have affected 12 2GNB (14.5%). This proportion is similar in the PB (11.4%). The majority of 2GNBC have removals in the two faces. Generally the extractions affect to half or less than half of the perimeter of the 2GNB (58.1%). The angles use to be acute (in nearly half of the pieces), or else steep (in a fifth part). The 50% of the retouches are deep regarding the object’s edge, and a fourth part are marginal. The direct retouches surpass the indirect (47% in front of 31%). In addition, the majority of the retouches are continuous, and with convex delineation. 68

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA this type of 1GNBE, depending on the status more or less advanced of the knapping. In this sense, there are 8 exhausted cores or nearly exhausted, with scarce possibilities to continue with the knapping (4 have less than 20mm of thickness) (Figure 2.5.6.2). The rest have lengths and widths that almost always are between the 60-80mm (6 objects), and between the 80-100 mm (5 objects).

3) Bifacial centripetal biconical strategy (discoid) We have identified six bifacial, multipolar centripetal biconical cores (discoid). In this case there is no hierarchization (Table 2.5.6) although at times one of the faces is not completely knapped. Almost all extractions have an acute or medium angle, independently of the face where they appear. The length and the width of these cores oscillate between the 60 and the 80 mm, and their thickness between 29 and 47 mm. Five of the six cores are of quartzite.

From a morphotechnical point of view, these cores have a face with a morphology that tends to be conical, with extractions of medium or acute angle, while the other face is plano-convex, with extractions of shallow-acute angle or semi-acute. The edges that delimit these two hierarchically organized surfaces (plane of intersection) are almost always sinuous and symmetric.

4) Unifacial centripetal strategy We have studied two unifacial 1GNBE with centripetal knapping (Figure 2.5.6.3). Both are of quartzite of scarce quality (his grain is coarse). However, all the perimeter (4C) is knapped, with extractions of acute angle. The edge is sinuous and symmetric. Also there is an unifacial centripetal 2GNBE (Table 2.5.6).

We have identified so much PB like 2GNBC correspondent to this strategy of production.

5) Bifacial strategy, with centripetal, linear extractions or else orthogonal There are 3 bifacial 1GNBE with ¾ parts of his perimeter (3C) knapped. There are no differences between the strategy followed in the two faces. The proximal zone of these three cores is not knapped, and their morphology is oval with tendency to circular. These cores, manufactured with quartzite, have centripetal knapping generally, although also we observed bipolar orthogonal and unipolar linear knapping (Table 2.5.6). Their length and their width oscillate between the 84 and the 123mm, while their minimal thickness is of 50mm, and the maximum of 71mm, therefore these cores are quite thick. It is necessary to add up a 2GNBE of quartzite, with centripetal extractions in both faces without hierarchization, neither volumetric predetermined structure.

2) Bifacial centripetal plano-conical strategy There is other group of cores (concretely 10) with a similar volumetric conception to the described before. Apparently these cores follow the same strategy, although with some variations (Table 2.5.6). Generally the face that was chosen to the systematic extraction of PB is not plano-convex but tends to be conical. The other face -of preparation- is also quite thick. The basic difference of these cores regarding the previous consists in that there is not a so evident hierarchization of the faces. At times these objects show one lateral affected for a fracture, and that forces to adapt the original idea to the circumstances of the knapping process. In any event, extractions also are multipolar centripetal in both faces. Almost always they are cores with knapping in all their perimeter (4C), and with extractions of acute angle, semi-acute or even medium angle. There are no significant differences in the size of these cores regarding the previous group.

Faciality Bifacial Bifacial Bifacial Unifacial Bifacial Unifacial Bifacial Trifacial Multifacial Bifacial Bifacial Unifacial

6) Unifacial or bifacial strategy with unipolar/bipolar extractions that exploit the transverse planes This group consists of five 1GNBE of hornfels (n=3), and

NEGATIVE BASES OF EXPLOITATION (CORES) Knapping Method Raw material centripetal with hierarchization and predetermination quartzite: 24 Centripetal / centripetal flat-conical quartzite: 10 Centripetal / centripetal quartzite: 5 biconic indet. 1 centripetal quartzite: 3 (1 2GNBE) centripetall/linear or orthogonal quartzite: 4 (1 2GNBE) longitudinal, unipolar or bipolar, masive, with removals hornfels: 3 in the transverse planes quartzite: 2 quartzite: 1 orthogonal, opposed, linear flint: 1 orthogonal quartzite: 1 unipolar linear, and bipolar opposed to obtain laminar quartzite: 2 products Multipolar orthogonal Quartzite: 1 (2GNBE) bipolar opposed, or unipolar linear Quartzite: 2 (both 2GNBE) Not identified Total Negative Bases of Exploitation

24 10

Objects 36,4% 15,2%

6

9,1%

3 4

4,5% 6,1%

5

7,6%

2

3,0%

1

1,6%

2

3,0%

1 2 6 66

1,6% 3,0% 9,1%

Table 2.5.6.- Negative Bases of Exploitation (1GNBE and 2GNBE) of Nerets, grouped in terms of the knapping method (Indirect Operational Themes). We point out the raw materials utilized for each group of cores

69

Xosé Pedro Rodríguez quartzite (n=2). Taking like starting point rolled pebbles with cylindrical morphology, extractions mainly in the transverse planes, although also in a case in the horizontal plane were realized. The removals cut off the extremes of the pebble, following a direction (unipolar linear), or two directions (bipolar orthogonal, and also bipolar opposed). The elongated and thicken morphology of the pebble, with cylindrical tendency, was taken the opportunity to accomplish extractions in the transverse planes. These extractions systematically reduce the volume of the core. This is not a too elaborate strategy, but rather simple. The angle of extractions is generally steep or medium. Extensive cortical zones remain; In fact only a part of the pebble changes. This reduction strategy is also known as “salami slice technique” or “parallel slice technique”. These objects have large size, with dimensions of their longer axis that oscillate between 107 and 178 mm. The thickness oscillates between 45 and 69 mm.

10) Unifacial bipolar opposed, or unipolar linear strategy There are two unifacial cores on flake (2GNBE) with few extractions. These removals are bipolar opposed in a case, and unipolar linear in the other. Both cores are of quartzite.

7) Trifacial and multifacial strategies Far less representative, although very significant is the presence of two trifacial cores and one multifacial (Table 2.5.6). All of them are of small size, and practically exhausted. The direction of extractions is variable, with predominance of the orthogonality, that allows taking advantage of the cores when leave not much raw material to exploit. It is pertinent to emphasize that one of the trifacial cores is of flint, an infrequent raw material among Nerets’s artifacts.

The blanks produced by means of all these strategies are PB of medium size, with dihedral cutting edges, or with trihedrons generally. All Operational Units implicated in processes of production are present in the most representative Indirect Technical Operational Themes. The existence of cortical blanks and also of PB of preparation of some

In synthesis, the bifacial centripetal knapping dominates (40 cores, the 60.6 %), particularly when it implies predetermination of the final morphology of the flakes (group 1). In this sense, maybe the 10 bifacial centripetal cores that apparently do not present predetermination (group 2) constitute an intermediate stage between the centripetal biconical discoid cores (group 3), and the centripetal planoconical with predetermination (group 1). No one of the 2GNBE follows the dominant guideline of the 1GNBE: There are no cores on flakes with predetermination of the final morphology of the products.

8) Bifacial strategy to obtain laminar products This group consists of two quartzite objects, with an obvious predetermination in the strategy of knapping. The objective was to extract elongated and narrow, laminartendency Positive Bases. In order to get this objective, striking platforms that were offering a good angle, relating to the face elected to extract the PB, were prepared. The knapping is longitudinal, making good use of the major axis of the core, and maintaining the slight convexity of the flaked face to obtain long products, and that they not break. In one of the cores there are bipolar opposed extractions; that means that two opposed striking platforms are used to accomplish extractions, although the majority follow an unique direction (Figure 2.5.6.4). In the other core one of the striking platforms stands out (with fundamentally unipolar linear extractions). This is a core of reduced dimensions (65x35x28mm), while the other is slightly larger (64x70x43mm). At the very least there are three products (one PB, and two 2GNB) that come from this reduction strategy. Furthermore two PB of reconfiguration or resharpening of this type of cores have been identified. 9) Bifacial multipolar orthogonal strategy This strategy is represented only by a 2GNBE of quartzite. This core have multipolar removals in both faces, with orthogonal structure.

Figure 2.5.6.- First Generation Negative Bases of Exploitation from Nerets. 1: Bifacial core of quartzite, with hierarchization of the faces and preconfiguration of the final products. 2: Exhausted core of quartzite, also with hierarchization of the faces and preconfiguration of the final products. 3: Unifacial centripetal core of quartzite. 4: Core of quartzite, with a bifacial strategy to obtain laminar products

70

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA NBE (initialization phase) confirm this affirmation.

maximum length of less than 100mm.

In addition, we have recovered cores that were abandoned in different phases of exploitation: from the initial stages to practically exhausted cores. Everything it allows affirming that, as general pattern, the exploitation was developed in the site.

Among these artifacts with distal dihedral cutting edge we have included an unifacial pebble of limestone, with cylindrical morphology, that in principle was used as core, and next its distal and proximal extremes were configured to create dihedral cutting edges of 55° and 78° respectively. This is the unique case identified of 1GNBE reconverted in 1GNBC.

The configuration of tools The shaped artifacts (1GNBC and 2GNBC) constitute the 18.3% of Nerets lithic industry (excluded Fragments). This percentage changes of important manner according to the raw material: the 16% of the quartzite objects, and the 29% of hornfels artifacts were retouched to be used as instruments. Therefore, the processes of configuration affected comparatively more the objects of hornfels. Two data confirm this hypothesis: three fourth parts of the 1GNB of hornfels are configured instruments (1GNBC), while among the quartzite objects the 1GNBE are more numerous than the 1GNBC.

2) Another important DTOT has like objective the configuration of uniangular (convergent) morphologies by means of unifacial extractions. Objects classified generally as picks are the result. These objects, manufactured utilizing quartzite (n=8), hornfels (n=3), and sandstone (n=3), have only between the 25% and 50 % of his perimeter affected by removals. These extractions configure an uniangular

Pebble tools We have identified 67 instruments on pebble, or pebble tools. Quartzite, hornfels and sandstone are the raw materials utilized to make 1GNBC. Although the majority of the objects are of quartzite, the presence of objects of hornfels and sandstone is significant. These two raw materials are selected specially to shape this type of objects. There is a difference of size between the 1GNBC of quartzite and hornfels. The 1GNBC of hornfels are larger (average: 151x98x47mm) than 1GNBC of quartzite (average: 121x103x50mm ) (Table 2.5.2). We have identified four principal strategies for shaping tools on pebble (Direct Technical Operational Themes, DTOT) (Table 2.5.7): 1) The most abundant DTOT consists of pebbles with a transverse cutting edge lightly convex or straight, and a morphology that tends to be oval. These objects were configured almost always with unifacial retouches (n=15), and sometime bifacial (n=4). Most of these artifacts can be classified, from a typological point of view, as choopers and chopping tools. The majority of these objects are of quartzite (n=12) (Figure 2.5.7.2). Also there are 3 of sandstone, 2 of hornfels, 1 of limestone, and 1 of not determined raw material. The objective is to shape an extensive transverse distal cutting edge, convex or convex-straight. These 1GNBC are not much configured (all except three have the 50% or less than 50% of the perimeter affected by removals), with medium angled extractions, acute or even shallow-acute (4 objects). Some removals can be extensive, but also marginal or very marginal extractions to complete the configuration of the cutting edge are common. Edges generally are incurvated and symmetric, or else straight and symmetric. Two of the bifacial pebbles have sinuous and symmetric edges. Almost half of these artifacts have a maximum length between 128 and 154mm. Only 4 have a

Figure 2.5.7.- First Generation Negative Bases of Configuration (1GNBC) found in Nerets. 1: Unifacial uniangular (convergent) of hornfels, with distal trihedron and lateral dihedrons (pick). 2: Unifacial of quartzite, with a transverse dihedron (chopper)

71

Xosé Pedro Rodríguez tion that do not fit well exactly with these four principal strategies. These objects have some intermediate features between the four previous groups. The majority are pebbles with lateral straight or lightly convex cutting edges, and transverse cutting edges, also straight or convex.

(convergent) distal extreme, or convex with tendency to uniangular (Figure 2.5.7.1). Generally extractions are acute or medium angled (in some case steep), and extensive. The sagittal edges have a delineation preferentially incurvated (8 of the 14 objects), and they are generally symmetric regarding the object’s horizontal plane (9 of the 14 objects). The size of this kind of 1GNBC are large. The larger picks reaches the 232 mm of length, and the majority oscillate between the 127 and 189 mm. Also there are two picks with less than 90 mm.

For example, a group of pebbles has some characteristics than place them between the strategy 1 (choppers + chopping tools group), and strategy 2 (configuration of picks). Concretely there is 4 objects with an elongated morphology, with a distal convex dihedron (with tendency to become trihedron), associate to a lateral dihedron. Three of these objects are of quartzite and one of hornfels. All of them have large size, between 135 and 192 mm of maximum length. Their technical traits correspond in general with the unifacial uniangulars (picks), although one of them is bifacial.

3) The DTOT that has like objective to shape ample lateral-transverse cutting edges of straight delineation, that give as a result objects similar to cleavers is less frequent. There are three 1GNBC of large size in this group (between 125 and 156mm of maximum length), with ample straight cutting edges in the transverse distal zone, and in one of his lateral sides. Two of these objects are of quartzite (with bifacial configuration), and one of hornfels (unifacial). The object of hornfels has an excellent transverse distal cutting edge with straight delineation and no symmetric, with an angle of 45°. One of the quartzite artifacts, the more knapped, has an excellent dihedral cutting edge in the lateral right side, of between 29º and 32º.

Also we have studied six objects (four of them unifacial) with similar characteristics to the ones that we have described before. These 1GNBC have extensive lateral and transverse distal cutting edges, that affect three fourth parts of the object. The proximal zone preserves rets of cortex. The fundamental difference with regard to the previous series lies in that these objects have extractions in both lateral sides, and the previous in only one. The transverse cutting edges have convex delineation, and the lateral are straight. Four of these artifacts are of quartzite, and two of hornfels. The quartzite artifacts show a certain equilibrium between length and width, with a tendency to the narrowing at the distal zone, with straight or lightly convex delineation. The two objects of hornfels have a rectangular morphology, so that their width is notably larger than his length. In general the maximum dimensions surpass the 100mm.

4) Finally we identified a DTOT whose objective was the manufacture of objects typologically classifiable as handaxes. In these objects the objective is to configure dihedral bilateral convex cutting edges, that converge in one of the extremes. One of the objects is of hornfels, and two of quartzite. The bifacial of hornfels has some differentiated features. In the first place its size are large (196x111x54mm), and is less shaped. One of its faces has very few removals. The final result is an uniangular (convergent) bifacial artifact, quite thick, but with a good dihedral cutting edge of 66° in the lateral left zone. The sagittal edge is sinuous and no symmetric. Others two bifacial implements are smaller (about 90mm of length, 70 of width and 30 of thickness). Their morphology is oval and they are completely knapped, without cortex remains. The cutting edges have dihedrons that range from 50° to 76°, although preferentially they are between the 55° and 65°. The sagittal edges are sinuous and symmetric in one of the objects, and no symmetric in the other.

Seven objects have a configuration that approximates them to the morphology of the cleavers. These pebbles have dihedral straight or convex cutting edge at the transverse distal sector, and also in one lateral side (and not in both lateral sides like the six previous objects). Therefore these instruments have an intermediate configuration between the previous series and the group of the cleavers. Six of these seven objects are of quartzite, and one of hornfels. Two of them have small-size (less than 60mm of maximum length).

Furthermore, there is a series of objects with a configuraMorphodynamic capacity of the retouched tools Transversal dihedral edge, with straight-convex morphology Lateral dihedral edge, with straight-convex morphology Transversal and Lateral dihedral edge, with straight-convex morphology Transversal and Lateral dihedral edge, with straight morphology (“cleaver”) Bilateral dihedral edge + distal Trihedral edge (“handaxe”) Concave dihedral edge (“notch”) Distal Trihedral edge (“picks” and “points”) Straight-convex denticulate edge Uniangular (convergent) denticulate edge (“denticualte point”) Total identified

1GNBC 19 33,3% 1 1,8% 17 29,8% 3 5,3% 3 5,3% 0 0,0% 14 24,6% 0 0,0% 0 0,0% 57 49,1%

2GNBC 15 25,4% 12 20,3% 5 8,5% 5 8,5% 1 1,7% 5 8,5% 8 13,6% 7 11,9% 1 1,7% 59 50,9%

Table 2.5.7.- Morphodynamic capacity of the retouched tools of Nerets

72

TOTAL 34 29,3% 13 11,2% 22 19,0% 8 6,9% 4 3,4% 5 4,3% 22 19,0% 7 6,0% 1 0,9% 116

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA Finally there is a large object of hornfels with 208 mm of length. This instrument has an excellent lateral straight, and symmetric cutting edge (with tendency to incurvate). In reality this diversity of morphologies is reduced to three principal morphopotential structures with several variants: trihedrons (24.5%), transverse dihedrons (33.3%), and lateral - transverse dihedrons (42.1%). Most of the cortical PB recovered in Nerets probably come from these Direct Operational Themes, particularly cortical PB of hornfels. Also some cortical 2GNB come from any one of these Direct Operational Themes for shaping pebble tools. Flake tools There are 78 retouched flakes or 2GNBC (94% of all the 2GNB). Concerning to the instruments configured on flake, we have calculated the percentage of initial products that later were retouched (shaping index). There were 553 initial products of >20mm of maximum length in Nerets; among these, 78 were chosen (14.1%), to be shaped as instruments, and converted in 2GNBC. The proportion of retouched products relating to the initial flakes is of 13.5% for quartzite (67 2GNBC relating to 497 initial products), while for the hornfels is of 20.8% (5 2GNBC relating to 24 initial products). It does not seem to have a selection of a specific morphology of PB to be retouched. So much flakes with cortical butts (of quartzite and hornfels) as the flakes with predetermined morphology, and flakes with bifaceted or multifaceted butts (of quartzite) were retouched. However, there was some kind of selection in the size of the blanks. The dimensions of the 2GNBC are, on the average, 24mm more long than the PB. This phenomenon is valid so much for the quartzite as well as for the hornfels. In general the instruments of hornfels (so much 1GNBC like 2GNBC) have a larger size than quartzite objects. This phenomenon is clearly visible among the 1GNBC.

Figure 2.5.8.- Second Generation Negative Bases of Configuration, from Nerets. 1: Bifacial of quartzite, with bilateral and sagittal symmetry (handaxe). 2: Quartzite with a lateral dihedron (side scraper). 3: Quartzite, with a concave dihedron in the right side (D1), and a denticulate in the left side. 4: Quartzite, with a distal trihedron, making good use of a lateral fracture and retouching the other lateral side. 5: Quartzite, with a straight dihedral cutting edge in the left side (R1), and a concave dihedron in the right side (D1). 6: Quartzite, with a bifacial straight dihedral cutting edge in the transverse zone

The configuration of 2GNB searches primarily the creation of dihedral cutting edges at the artifact’s transverse sector, or sometimes in a lateral side (Table 2.5.7) (Figures 2.5.8.2, 2.5.8.5, 2.5.8.6): there are objects of medium or large size, assimilable, because of their morphology, to the “cleaver” operational standard. On the other hand, five objects have concave dihedrons (Figure 2.5.8.3, 2.5.8.5).

2.5.8.4). Also we have localized configuration of denticulated cutting edges (8 artifacts); almost always with straight or convex delineation, although also there is an uniangular (convergent) denticulate. The typological classification shows the predominance of the group of side scrapers, with 13 implements. Among the side scrapers the most frequent types are R1 and R2 (each one with 6 objects). There are 11 denticulates, standing out notches (D1), with 5 objects. There are 5 steep angled pieces: two are denticulated and three with continuous retouches. Also we have identified three end scrapers (a G12 and two G11), two truncations, and a burin (B22).

The presence of an uniangular (convergent) object with dihedral lateral cutting edges stands out. The bilateral symmetry of this object allows to classify it as handaxe (Figure 2.5.8.1). In eight objects trihedral edges were configured. The trihedron was shaped in two cases making good use of a lateral fracture and retouching the other lateral side (Figure 73

Xosé Pedro Rodríguez

Figure 2.5.9.- Morphogenetic matrix of the lithic industry of Nerets

Objects derived from processes of production of flakes and of configuration of instruments have been found . Also many natural Bases were found.

Finally let’s remember the existence of five cleavers and one handaxe. In summary, the dihedral cutting edges of straight delineation or else convex were configured in 68% of the 1GNBC and 2GNBC (Table 2.5.7). Adding the concave dihedrons (notches) the percentage of dihedral cutting edges reach the 72%. The configuration of transverse dihedrons, so much in the 1GNBC such as in the 1GNBC stands out specially. In nearly a fifth part of the instruments distal trihedrons were shaped. The denticulated cutting edges are not very numerous, since only they appear in the 6.9% of artifacts. There are 4 objects with bilateral dihedral cutting edges and symmetric morphology, assimilable to the Operational standard handaxe (3.4% of the retouched tools). There are 8 artifacts (6.9%) with ample dihedral cutting edges in the lateral, and at the distal transverse sector, and with straight delineation. These artifacts have a rectangular morphology, that fits well in the cleaver morphotype.

We have observed a selective management of the raw material. Quartzite is used so much in processes of production as well as for the configuration of instruments. The processes of exploitation consist in the systematic production of Positive Bases of medium size, with excellent dihedral straight or straight-convex cutting edges, and trihedral edges. Several strategies were put into practice in order to obtain PB: the bifacial centripetal Operational Theme with preconfiguration of the PB’s final morphology stands out. On the contrary, the hornfels is used mostly for shaping instruments of large size on pebble (1GNBC). This rock practically was not used in processes of exploitation, since he does not offer so good capacities like the quartzite. In general, there is a notable variety of strategies of knapping, so much destined to the configuration of instruments on pebble (DTOT), like to producing flakes (ITOT). Among the Direct Technical Operational Themes the shaping of unifacial uniangular (convergent) objects, and unifacial objects with distal convex cutting edges highlights. Among the ITOT, the most important reduction strategy is the one that predetermines the final morphology of the products. Objects of all the phases of these two Operational Sequences appeared in Nerets.

Conclusions Nerets is an open air site, with lithic artifacts in surface and in stratigrapic context. Faunal remains have not been recovered at this site. The lithic industry appears dispersed across an ample surface, as a consequence of its remobilization because of erosion. Nerets’s situation is strategic: At a hill that dominate a narrow valley, across where the river Noguera Pallaresa flows. Furthermore, there are rock shelters at the hill’s most abrupt slope. It is possible that in the Paleolithic these rock shelters can be utilized by the hominids. Unfortunately, the access to these rock shelters is nowadays practically impossible from the top of the mountain, and neither it is possible from his downside because Sant Antoni’s swamp impedes it.

The scarcity of knapping debris (PB of less than 20 mm) can be consequence of that the heft of the material was collected in surface, during prospections. Furthermore, the small debris of knapping derived from processes of configuration probably scattered more than the rest of objects. On the other hand, the percentage of 2GNBC is not very high (9.8% without fragments), and in consequence the PB of less than 20 mm neither would be very numerous. In short, we thought that the processes of exploitation and also of configuration, were realized at the site.

The lithic industry was made basically with quartzite, and hornfels pebbles, collected at the terraces of the Noguera Pallaresa river in a place close to the site.

74

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA As we already have said, the absence of faunal remains handicaps an interpretation of the function of this site. With the lithic record that we dispose we can venture the hypothesis that Nerets was a place that, for his strategic situation, was visited frequently by hominids to develop processes of production and of configuration of artifacts. These artifacts may have been utilized in food procurement strategies. The lithic industry of Nerets, fit well with an initial moment of the Mode 3, most of all for the outstanding presence of strategies of exploitation with preconfiguration of the morphology of the knapping products. However, “archaic” technical elements subsist, like for example the notable presence of pebble tools (1GNBC). Unfortunately there are few criteria, aside from the morphotechnical, that allow us ascribing this site to a specific chronology. In terms of the aforementioned criteria, Nerets could be correspond to the end of the Middle Pleistocene.

75

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

2.6.- CLOT DEL BALLESTER

(Puigverd de Lleida), la Gravera de l’Eugeni (Artesa de Lleida), Mimferri (Juneda), El Secanet (Artesa de LleidaPuigverd de Lleida), and the Mousterian site of Les Fonts (Artesa de Lleida) (Canal and Carbonell, 1989; Carbonell et al., 1987b; Carbonell et al., 1993; Grup de Recerques Arqueològiques de La Femosa, 1976).

Location and discovery Clot del Ballester is an open air site, with lithic artifacts in surface. This site is located approximately two kilometers to the Southwest of the village of Artesa de Lleida (41º32’ N y 0º42’ E) (Figure 2.6.1). The archeological materials studied in this work were collected by members of the Grup de Recerques Arqueològiques La Femosa. The first materials were recovered in 1979 at a quite extensive terrace of the stream of La Femosa (Carbonell et al., 1993). This little river is born near Borges Blanques (Lleida) and flows into the river Segre, outside of the town of Lleida.

Analysis of the lithic industry We have examined 124 objects in this work, deposited at the establishments of Grup de Recerques Arqueològiques La Femosa (Artesa de Lleida). As we already have pointed out, this materials were collected in surface. The material was recovered during many years, thanks to repeated visits to the site. Not any type of material was discarded, neither was selection of the collected objects. Therefore, we considered that this is a representative sample.

The Valley of La Femosa, that is part of the Ebro’s Basin, is ample and is surrounded by small plateaus, with some hills that do not surpass the 300 meters of height above the sea level. Several Paleolithic sites were located at the terraces of La Femosa from the decade of 1970 (Fgure 2.6.1). The majority of the lithic materials comes from the gravels of the lowermost terraces, between 15 and 20 meters above the present-day level of the river. Nevertheless, the materials of the Clot del Ballester come from a higher terrace, placed between 40 and 45 meters of height above the level of the river, and 240 meters above sea level. Another interesting sites have been discovered aside from Clot del Ballestert: La Serreta (Puigverd de Lleida), Secots

The majority of the pieces of Clot del Ballester are Positive Bases (38.7%). However, the quantity of 1GNB (30%) stands out. Also the percentage of 2GNB is important. Finally, there are 15 fragments and a fractured natural Base. Raw materials The lithic industry recovered in Clot del Ballestar was knapped most of all with quartzite and hornfels (Table 2.6.1). Limestone and flint are secondary materials. There are no quartzite neither hornfels at the terraces of the river

Figure 2.6.1.- Location of Clot del Ballester, near de city of Lleida

77

Xosé Pedro Rodríguez nB Quartzite Hornfels Limestone Flint TOTAL

1 0 0 0 1

1,2% 0,0% 0,0% 0,0% 0,8%

1GNBC 7 8,2% 8 26,7% 1 16,7% 0 0,0% 16 12,9%

1GNB 1GNBE 11 12,9% 0 0,0% 2 33,3% 0 0,0% 13 10,5%

PB 1GNB Ind. 4 4,7% 5 16,7% 0 0,0% 0 0,0% 9 7,3%

37 7 3 1 48

43,5% 23,3% 50,0% 33,3% 38,7%

2GNB 2GNBC 14 16,5% 6 20,0% 0 0,0% 2 66,7% 22 17,7%

FRAG 11 4 0 0 15

TOTAL

12,9% 85 13,3% 30 6 0,0% 3 0,0% 12,1% 124

68,5% 24,2% 4,8% 2,4%

Table 2.6.1.- Lithic industry of Clot del Ballester (structural categories and raw materials)

La Femosa, so that the supply of these rocks could be accomplished at the Segre’s terraces, to less than five kilometers of distance.

BN1GC only a fourth part of the perimeter of the object was normally knapped. The 1GNB have extractions of shallow-acute or medium angle generally. The medium and acute removals are frequent among the 1GNBC. Concerning to the extent of removals, there is a high percentage of total and extensive removals. The total extractions dominate mainly in the 1GNBE. The sagittal edges of the 1GNB are very frequently sinuous. There is a similar number of objects with sagittal symmetric edges and no symmetric. The symmetric edges are habitual in the 1GNBE, while among the 1GNBC the no symmetric edges dominate .

Quartzite was used most of all for the production of PB (43.5 % of quartzites are PB), while the hornfels has a high percentage of 1GNB (43.3%). Morphotechnical Analysis Natural bases The one and only natural Base of the record is a fractured pebble of quartzite (nBc), with 69x57x40 mm. First Generation Negative Bases The 1GNB have an important role in this site, with the 30.6% of all the lithic industry (n=38). The 1GNBC are more numerous (n=18) than the 1GNBE (n=13), although also some 1GNB are undeterminable (n=7). The 1GNBC are of quartzite and hornfels. The majority of the 1GNBE are of quartzite. On the contrary there are no cores of hornfels. It’s evident that there is a selection of the raw materials. When the objective is the configuration of instruments, this hominids utilize quartzite and hornfels. However, when the objective is the systematic production of PB they select quartzite.

1GNB of Configuration When the objective is to shape an instrument directly on a pebble, hornfels and quartzite were utilized indistinctly. These objects can be unifacial or bifacial. When the implement is bifacial normally a fourth part of the perimeter is knapped. In the bifacial objects is frequent that at least one of his faces have all the perimeter knapped (4C). Extractions are preferentially medium angled, although also there are numerous of acute angle and shallow-acute. Habitually their extent is total. For the most part the 1GNBC have sagittal sinuous edges, and no symmetric.

The global average of the size of the 1GNB is 89x71x42mm (Table 2.6.2). The 1GNB of larger size are of hornfels. In addition, the 1GNBC have larger size than the 1GNBE.

1GNB of Exploitation Quartzite is utilized basically for producing PB, although also we have located two limestone 1GNBE. Generally cores are bifacial, with the perimeter totally knapped, by means of extractions of steep angle (and acute at times), that affect all of the face where the removals appear. The sagittal edges are sinuous, and by majority symmetric.

Average size of the 1GNB (mm) Raw material Length Width Thickness Quartzite 77,1 67,7 36,3 Hornfels 110,5 75,6 50,1 Limestone 87,0 81,3 48,0 Type of 1GNB Length Width Thickness 1GNB Indet. 76,6 62,3 37,9 1GNBC 106,8 82,4 48,2 1GNBE 72,0 61,3 35,5 Global average 89,3 71,5 41,9 Table 2.6.2.- Average size of the 1GNB of Clot del Ballester

Positive Bases With 48 pieces Positive Bases constitute the structural category better represented at this site. Nearly 80% of the PB are of quartzite (37 objects), while hornfels, limestone and flint have scarce PB. No PB has length or width equal or inferior to 20 mm. Probably this fact is related to the scarce number of pieces, and with their collection in surface.

The results of the analysis of all the 1GNB show that the bifacial objects predominate. However, there are differences between the 1GNBE and the 1GNBC. All 1GNBE are bifacial, but among the 1GNBC the unifacial objects surpass bifacials slightly. All the perimeter of the object was knapped in the majority of the occasions. This phenomenon appears most of all in the 1GNBE, while in the

The size of the PB of quartzite is lesser than the rest of raw materials (Table 2.6.3). These data agree with the dimensions of the 1GNB. Let’s remember that the 1GNB of quartzite have a minor size than hornfels and limestone. Globally PB have an average of 39 mm of length, 38 mm of width, and 15 mm of thickness. There are 7 PB with a length and/or width of more than 60mm (14.6% of the BP). Only two of the 48 PB have a laminar index greater 78

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA Average size of the PB (mm) Raw material Length Width Quartzite 35,5 36,9 Other raw material 50,5 42,8 Global average 39,7 38,5

The majority are fragments of quartzite (eleven), although also there are some of hornfels (n=4).

Thickness 14,1 16,9 15,0

Like the PB and the 2GNB, the quartzite fragments are smaller than the fragments of hornfels (Table 2.6.5). The global average of the dimensions of the fragments is similar to the PB and the 2GNB. There are no fragments of very small size.

Table 2.6.3.- Average size of the Positive Bases of Clot del Ballester

than 1.6 (4.2%).

Most of the fragments do not have remains of cortex. Nevertheless, there are a 26,6% of fragments with dorsal face completely cortical. In this sense, three of the four fragments of hornfels are cortical, while only one of the eleven of quartzite has total cortex.

The cortical butts are lightly further numerous than no cortical. Nevertheless, the 2GNB have almost always no cortical butts. Adding up both categories we get a slight predominance of no cortical butts. The vast majority of the butts are of platform type. The no faceted butts are more frequent, most of all among the PB. Secondly, multifaceted butts are frequent so much among the PB like among the 2GNB. The dorsal face of the PB and of the 2GNB uses to be completely no cortical. We have not located products of knapping with dorsal face completely cortical.

Average size of the Fragments (mm) Raw material Length Width Thickness Quartzite 48,2 32,3 18,4 Hornfels 57,5 31,8 18,3 Global average 50,7 32,1 18,3 Table 2.6.5.- Average size of the Fragments of Clot del Ballester

The Positive Bases show excellent dihedral cutting edges, generally in the lateral although also there are PB with trihedrons.

Strategies for producing Positive Bases All the cores of Clot del Ballester are 1GNBE (n=13). This implies that we have not found cores on flake (2GNBE).

The majority of the PB of this site come from Indirect Operational Themes of multipolar centripetal knapping (Figure 2.6.2.3 and 2.6.2.4). A PB with multifaceted butt corresponds to a knapping strategy clearly predetermined, for producing flakes with a standardized morphology.

We have identified two reduction strategies, that coincide in the multipolar centripetal knapping. In the first place there is an Indirect Technical Operational Theme (ITOT) that consists in the systematic production of PB by means of a strategy clearly predetermined. This Operational Theme has been identified due to the existence of a 1GNBE and also of a PB, and a 2GNB.

Second Generation Negative Bases The 22 artifacts framed in this category suppose the 17.7% of the lithic industry of CB (without fragments the percentage is 20.2%). Quartzite is the most frequent raw material (72.7%), followed by the hornfels (27.3%), and flint (9.1%). All 2GNB are retouched artifacts; there is no 2GNBE.

Another centripetal strategy exists besides, although without predetermination of the morphology of the final products. This Operational Theme could have been identified by the existence of 12 1GNBE, and of several PB and 2GNB.

The size of the 2GNB of quartzite are minor than hornfels and flint (Table 2.6.4). The global average is of 45 x 39 x 16 mm, lightly larger than the average of the Positive Bases. Average size of the 2GNB (mm) Raw material Length Width Quartzite 41,6 38,7 Other raw material 49,8 41,8 Global average 45,4 39,5

The configuration of tools We have identified 18 shaped pebbles (1GNBC), and 22 flake tools (2GNBC). In consequence there are 40 configured artifacts (Table 2.6.6). Pebble tools Eight 1GNBC have dihedral straight convex cutting edges at the distal transverse zone, accomplished with unifacial extractions (Table 2.6.6) (typologically choppers). Three of these artifacts are of quartzite, and five of hornfels. With the same type of morphopotential and delineation, but with bifacial extractions there are five objects (chopping tools) (three of hornfels, one of quartzite, and another one of limestone).

Thickness 15,6 16,8 16,1

Table 2.6.4.- Average size of the 2GNB of Clot del Ballester

We already have indicated that the 2GNB show a lesser amount of cortex than the PB. In general, the butt is no cortical, and of platform type, unifaceted or multifaceted. The dorsal face does not have almost cortex.

Furthermore there are two instruments of hornfels with a distal trihedron (picks) (Figure 2.6.2.1). These pebble tools are unifacial uniangular (convergent). In addition, we

Fragments There are fifteen fragments among the studied material. 79

Xosé Pedro Rodríguez

Figure 2.6.2.- Lithic industry of Clot del Ballester. 1: 1GNBC of quartzite, with a distal trihedron. 2: 1GNBE of quartzite, with centripetal knapping. 3 and 4: Centripetal Positive Bases of quartzite

Flake tools All 2GNB are retouched flakes (that is, 2GNBC). The processes of configuration of flake tools have an important role. One out of every three initial products was retouched, and transformed in 2GNBC (22 relating to 70 initial products). The shaping index of the products is greater in the blanks of hornfels than in quartzite. Nearly half of the 13 initial PB of hornfels were transformed in 2GNBC

have identified a concave dihedron (notch) on a quartzite pebble. Finally, there are two quartzite artifacts with lateral dihedrons that converge in a distal trihedron; bifacial removals configure a morphology that has bilateral symmetry. We can classify this objects as handaxes.

Morphodynamic capacity of the retouched tools Transversal dihedral edge, with straight-convex morphology Lateral dihedral edge, with straight-convex morphology Bilateral dihedral edge + distal Trihedral edge (“handaxe”) Concave dihedral edge (“notch”) Distal Trihedral edge (“picks” and “points”) Straight-convex denticulate edge Uniangular (convergent) denticulate edge (“denticualte point”) Total identified

1GNBC 13 72,2% 0 0,0% 2 11,1% 1 5,6% 2 11,1% 0 0,0% 0 0,0% 18 45%

2GNBC 3 13,6% 5 22,7% 0 0,0% 1 4,5% 2 9,1% 8 36,4% 3 13,6% 22 55%

Table 2.6.6.- Morphodynamic capacity of the retouched tools of Clot del Ballester

80

TOTAL 16 40% 5 12,5% 2 5% 2 5% 4 10% 8 20% 3 7,5% 40

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

Figure 2.6.3.- Morphogenetic matrix of the lithic industry of Clot del Ballester

es of 2GNB retouch, is an important feature of this sample. These small objects may have been disappeared from the record, since we are examining a surface site. On the other hand, the retouch or configuration of the flakes could be accomplished in another place.

(46.2%), while 27.5% of the initial quartzite products were retouched. In the majority of the occasions only one of the faces of the object was retouched . The angle of extractions uses to be acute, while the depth concerning the edge is almost always deep. The direction of the retouch is most of all direct, although also there is significant presence of indirect extractions. An equilibrium between the presence of continuous retouches and denticulated retouches exist. Finally, the delineation uses to be convex.

Taking the results of the study of the 1GNBC and 2GNBC into account we can affirm that the basic objective is the shaping of dihedral cutting edges located at the transverse distal zone of the objects (basically (1GNBC). Follow in importance the configuration of denticulated cutting edges (in 2GNBC), of lateral dihedrons (in 2GNBC), and trihedrons (so much in 1GNBC like in 2GNBC).

The configuration of denticulated cutting edges appears in more than a third part of the 2GNBC (8 objects). There are also three artifacts with denticulated convergent cutting edges. In two pieces trihedrons were configured. The rest of the 2GNBC have various types of dihedral cutting edges. Concretely there are five artifacts with dihedral lateral cutting edges, three with dihedrons at the transverse distal sector, and one with a concave dihedron (Table 2.6.6).

Conclusions Clot del Ballester is an open air site, with artifacts in surface. Although the recovered record is scarce (124 objects), we can outline roughly the characteristics of this lithic assemblage. The raw material, eroded pebbles of quartzite and of hornfels, was obtained at the terraces of the Segre river, that flow at present to less than five kilometers of the site. A specific selection of the raw materials exists: quartzite was used for the exploitation of cores in order to produce PB, and also for the configuration of tools. On the contrary, the hornfels was used basically to shape large instruments on pebble. Without the fragments, the 11.9% of the studied objects are 1GNBE; the 44% are PB, and the 36.7% configured instruments (1GNBC and 2GNBC). These percentages differ notably according to the raw material. A 14.9% of the quartzite objects, and a 7.1% of hornfels are cores; the PB sum 50% of the quartzite objects, and only 25% of hornfels objects. The 28.4% of the quartzite objects are

From the typologic point of view, denticulates are majority, with 12 of the 22 2GNBC. Eight D3 (denticulated side scraper), and three D4 (denticulated point) stands out. Also, among the 2GNBC there are 6 side scrapers (five are R1), two points, an end scraper (G11), and a steep angled piece with continuous retouch. It is interesting to point out that one of the retouched flakes (a quartzite side scraper) comes from a strategy of predetermined knapping, since its morphology is clearly preconfigured. The lack of PB with very small size, related with sequenc81

Xosé Pedro Rodríguez instruments explicitly configured. This percentage raises the 57.1% among the objects of hornfels. Exploitation was accomplished through bifacial centripetal Operational Themes. One of these strategies consists in the preparation of the 1GNBE to obtain products with a predetermined morphology. This type of exploitation was performed with quartzite. The configuration of instruments on pebble pursues two objectives: The generation of transverse straight or convex cutting edges, and the creation of trihedrons (Table 2.6.6). The flake tools (2GNBC) use to have denticulated cutting edges, or else dihedral straight convex cutting edges, placed in the transverse distal, or in one of the lateral sides of the instrument (Table 2.6.6). The Operational Sequence shows some outstanding absences: we have found very few debris of knapping (PB of less than 20 mm), by-products of the configuration of the 2GNBC. Also we would be able to wait for a greater number of PB, taking into account the quantity of recovered cores. These absences can be consequence that we worked of with a surface site, and hence the collect of material may have been selective. However, the prospections in Clot del Ballester were accomplished systematically during years. In addition, the people that collected the material have the sufficient experience and knowledge like for not achieving a previous selection. The lithic material of Clot del Ballester has a great technological homogeneity, that allow us discarding any type of mixture of materials of very different chronologies. Besides, its traits make sense absolutely with the ones that have other sites analyzed in this work, and can be classified as initial Mode 3. We discuss the hypothesis that the lithic industry corresponds to an occupation of scarce intensity. Apparently the Technical Operational System of Clot del Ballester has similarities with Nerets, although its Operational scheme shows some differences. These differences can be a consequence of the low intensity of the occupation, or of a different function. We will go back to discuss this issue at the end of this work.

82

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

2.7.- VINYETS

Location and archeological interventions

Geological context

Vinyets is located at the municipality of El Catllar, some 10 km Northeast of the city of Tarragona (Figure 2.7.1). As a consequence of the extraction of clays for the construction of a dam, a little ravine with a slope orientated SE-NW was left. In this place Josep Zaragoza located lithic industry, in a stratigraphic section (Figure 2.7.2).

The main structural units that configure the relief of the “Camp de Tarragona”, where Vinyets is, are the VallsReus Basin, the great Horst of the Pre-littoral Mountain Range, and Bonastre’s Massif with their peripheral arch (Benzaquen et al., 1973).

The site is placed to some 280 meters from the river Gaià, and to 20 meters of height above the present-day river bed. At present the sea is about 7 kilometers to the East, and the site is 70 meters a.s.l.. The topographic coordinates of the place where the lithic industry was found are 1º 18’ 40’ E, and 41º 12’ 00’’ N (Figure 2.7.1).

The Valls-Reus Basin is a graben with almost 60 km of length, that is greater if we regarded it as a continuation of the Valls-Penedés Basin. The Valls-Reus Basin is placed between the Littoral Mountain Range (Bloc del Garraf and Alts de Barcelona), and the Pre-littoral Mountain Range (Bloc Priorat-Gaià-Terrassa). The municipal district of El Catllar is located at this Basin’s Southwest sector.

We achieved the first rescue excavation in June 1991. This excavation affected a surface of 17 square meters (Figure 2.7.2). A second excavation was undertaken during November 1995, that enlarged the sector excavated in 1991 in 8 square meters. Also we excavated a test pit some 50 meters to the West of the main area excavated in 1991, with poor results. Only lithic industry appeared during these two campaigns of excavation.

El Catllar’s geological substratum consists of deposits of Tertiary sedimentation, correspondents to the upper Miocene. The most abundant materials are calcareous conglomerates, sandstones, clays and silts. These miocenic levels consist basically of materials eroded from the immediate massifs: The Horst of the Gaià (longitudinal prolongation of the block raised from the Pre-Littoral Mountain Range), and Bonastre’s Massif, differentiated of the block of the Gaià for a set of longitudinal faults starting from Montmell. The great Horst of the Pre-Littoral Mountain Range consists of a set of blocks raised during the Alpine movement, separating the Coastal Basin from the Ebro Basin. We can differentiate three major structures divided by two transverse faults: The Massif of the Priorat, the Massif of the Gaià, and the Horst of Terrassa. Bonastre’s Massif forms a mesozoic block raised by a translational movement of the paleozoic socle toward the SE. The North and West fringe that surrounds this massif is known as “Peripheral Arch of Bonastre”. This unit

Figure 2.7.2.- General view of Vinyets site, during the excavation of 1991

Figure 2.7.1.- Location of Vinyets site, near de city of Tarragona

83

Xosé Pedro Rodríguez has been submitted to lateral and internal pressures by the formation of Bonastre’s Massif, that have forced the mesozoic coverture to form a synclinal fold and failing, causing structures of great complexity. The river Gaià grows on the high plateau of Santa Coloma de Queralt, close to Aguiló, and crosses the great Pre-littoral Horst, fitted in during several kilometers in the Gaià Massif (Bloc del Gaià).The river follows its destination crossing the Valls-Reus Basin for his easternmost zone, and bordering Bonastre’s Massif. In this place the river is inserted between the mountains, forming complicated meanders. Next the river continues through the Coastal Basin of Tarragona-Vendrell, and flows in the Mediterranean sea, between the localities of Tamarit and Altafulla.

Stratigraphy Vinyets’s quaternary formation has been described by Vallverdú (Vallverdú, 1993). From top to bottom, the sequence is the following (Figure 2.7.3): 0-1,70 meters: Calcareous crust, lixiviated horizon and paleosoil (BT1) of diffuse and gradual limit of clays and silts. 1,70-3,70 meters: Lixiviated horizon with pseudomycelium of 3 to 5 cms, calcareous nodules. With diffuse limit but rectilinear. 3,70-4,80 meters: Paleosoil (BT2). Clayey horizon illuvial of clear and rectilinear limit. Clay of prismatic structure. 4,80-6,20 meters: Clays crust, silts and fine sands and very cemented pseudomycelium, bottom of the BT2. Diffuse and gradual limit forming a very cemented bar.

Figure 2.7.3.- Schematic stratigraphic column of Vinyets sequence. 1: Calcareous crust. 2: Paleosoil. 3: silts and fine sands with pseudomycelium. 4: Fine sands and silts of fluvial and aeolian origin. 5: Archaeological levels. 6: Colluvium of gravels

6,20-11,20 meters: Fine sands and silts of fluvial and aeolian origin. Diffuse and sinuous limit, with laminar organized structure.

12,60-14,00 m. Colluvium of gravels (with blocks of 0.5 to 20 cm). Can be a river terrace.

11,20-12,60 meters: Paleosoil BT3. Clayey horizon illuvial. Three archeological levels appear in colluviums’s bags. Upper limit diffuse and sinuous. The archeological levels and the matrix where they appear are the following:

Analysis of the lithic industry The excavations of 1991 and 1995 gave 318 lithic objects (Table 2.7.1). The majority of these artifacts were recovered in stratigraphic context, framed in 3 archeological levels. The level 2 has more implements than the rest (n=196). Also we recovered 37 objects without stratigraphic context. These objects can not be ascribed to any level (Table 2.7.2.). The lithic industry of the 3 levels has a great similitude, hence we have accomplished an unitary study.

11,20-11,70: Clay in diffuse horizon forming tongues. Colluvium of 20 cms of thickness. Squamate-laminar structure of mid-size, bioturbated. Archeological level 1. 11,70-12,00: Clay in diffuse horizon, in tongue. Colluvium of 10 cm of thickness. Squamate polyhedral structure. Archeological level 2.

Raw material Flint is the most utilized raw material to manufacture the lithic industry of Vinyets. The rest of raw materials almost reaches the 2%. The percentage of raw materials in the levels with more industry is very similar. Globally, flint reaches the 92%. All the levels agree in the presence of quartzite, quartz and limestone like secondary raw materi-

12,00-12,30: Clay in diffuse horizon, in tongue. Colluvium of 5 cms. Squamate polyhedral structure. Archeological level 3. 12,30-12,60: Clay in diffuse horizon, in tongue. Colluvium in the bottom of the horizon, on calcareous crust. Without artifacts. 84

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA nB

1GNB PB 1GNBC 1GNBE 1GNB Ind. Flint 5 1,7% 2 0,7% 7 2,4% 1 0,3% 102 34,8% Quartzite 3 42,9% Quartz 1 14,3% 1 14,3% Limestone 1 16,7% 4 66,7% Oher raw materials 1 20% Total 6 1,9% 3 0,9% 7 2,2% 1 0,3% 111 34,9%

2GNB 2GNBC 2GNBE 73 24,9% 2 0,7% 3 42,9%

FRAG

99 1 5 1 2 1 40% 78 24,5% 2 0,6% 107

INDET

TOTAL

33,8% 2 0,7% 293 92,1% 7 2,2% 14,3% 7 2,2% 71,4% 6 1,9% 16,7% 5 1,6% 20% 1 20% 33,6% 3 0,9% 318

Table 2.7.1.- Lithic industry of all the archaeological levels of Vinyets (structural categories and raw materials) Level 1 Flint Quartzite Quartz Limestone Oher raw materials TOTAL

62 2 1 2 0 67

92.5% 3.0% 1.5% 3.0% 0.0%

Level 2 182 3 3 3 5 196

Without stratigraphic TOTAL context 32 94.4% 86.5% 293 92,1% 7 2 5.4% 2,2% 7 3 8.1% 2,2% 6 5.6% 1,9% 5 1,6% 37 318

Level 3

92.9% 1.5% 1.5% 1.5% 2.6%

17

1 18

Table 2.7.2.- Lithic industry of archaeological levels 1, 2 and 3, and lithic remains without stratigraphic context, from Vinyets

als (Table 2.7.2). We highlighted the finding of a 1GNBC of quartz. The rest of objects not knapped with flint are basically PB and Fragments.

First Generation Negative Bases Eleven objects are included in this category: Three are clearly 1GNBC, and also three are 1GNBE. Besides, there are three objects that in the beginning seem to have been exploited to produce PB, but that at the end they were shaped to be used as instruments. Finally there are two objects whose analysis is more difficult. One of them is a fragment of 1GNB, very likely of exploitation, while the other is practically unclassifiable, mainly because of the concretion that covers it.

Most of the flint objects show a yellowish patina, and at times also rolling. Some pieces are crusted. The flint utilized in Vinyets was collected in the Gaià river bed. This river carries away in its bed small flint nodules. This flint is of poor quality for the knapping. Furthermore, the reduced dimensions of the nodules impose limitations, and they condition the lithic production. There are only three objects among the 318 implements with a length of more than 100 mm, and two of them are natural Bases. The 1GNB of configuration, that are not much knapped, have an average size of 79 x 89 x 56.5 mm. The rest of the artifacts are smaller. The PB have an average of 23.5 x 21.5 x 10.5 mm; the 2GNB 31x28,5x13,5 mm; and Fragments 21,5x16x10 mm. Therefore, Vinyets’s lithic industry is of small size. In the Gaià river bed, or in his terraces, we can find the rest of raw materials.

Vinyets’s 1GNB are by majority bifacial, and frequently the knapping affects to more than half of their perimeter. The removals with acute and medium angle dominate. The intense anthropization of the Bases is evident when analyzing the extent of removals: no one is very marginal, and the majority are very extensive or total. These objects have sagittal sinuous edges, distributed between the symmetric, and the no symmetric regarding the horizontal plane. 1GNB of Configuration Three objects have been identified like instruments shaped directly on pebble: two on rolled pebbles of flint, and one on a small pebble of quartz. An ample transverse cutting edge, with sinuous delineation, denticulate, and with tendency to uniangular (convergent) was configured in the object of larger size (132,5x127x83,5mm) (Figure 2.7.4.2). The other flint object has an uniangular morphology at its transverse distal zone, with a trihedron (Figure 2.7.4.1). The quartz instrument is scarcely configured (it is unifacial, while the others two are bifacial), but the objective also was to create a distal trihedron. Therefore, the three artifacts have distal trihedrons.

Morphotechnical analysis Natural Bases We have identified six natural Bases: five of flint and one of limestone. There are two nBa, that is without apparent marks, and four nBc. We observed an obvious difference between the dimensions of the various structural categories. The nBa are of larger size, with an average of 166 x 133.5 x 93.5 mm. Precisely its large size and its morphology allow us to think that they were anvils on which some kind of work was made. Only human transport may explain the apparition of these large-sized objects. The object of larger size is a limestone nBa, with 177 x 147 x 92 mm.

1GNB of Exploitation We analyzed three 1GNB exploited as cores. The size of the three cores is similar: length between 41 and 42 mm, width between 36 and 41mm, and thickness between 25

The fractured natural Bases (nBc) could be used as hammers. His average size is clearly lesser than the nBa, with 66 x 58 x 44 mm. 85

Xosé Pedro Rodríguez cortex were left in the three implements. It seems that the possibilities of these objects as cores failed, although two of them are small, and they do not offer many possibilities for PB’s extraction.

and 27mm. The morphology of the three 1GNBE tends to be cubic. The three cores are practically exhausted, and offer scarce possibilities to continue achieving removals. As to the strategy of exploitation, the three cores show orthogonal extractions, with predominance of the multipolar orthogonal, and unipolar linear knapping. These cores are bifacial with tendency to multifacial. It became apparent that the features of the raw material (rolled pebbles of flint, with scarce quality), have influence and condition the knapping , provoking the election of a flexible, adapted strategy to the morphology of the core. The objective was to get the greater advantage from the rock.

Positive Bases This is the structural category with most objects (Table 2.7.1) (Figure 2.7.5): 111 elements (34.9% of all the lithic industry). If we not take the Fragments into account neither the indeterminable the percentage of PB is 53.4%. The size of the Positive Bases is very small, only in the level 3 the average surpasses the 30mm of maximum length. In general there is uniformity in the dimensions of the PB of the three levels (Table 2.7.3).

In addition to the described pieces, there is another 1GNB that is very crusted, so that only it is possible to recognize a removal. This would be a 1GNB of exploitation whose knapping did not continue, perhaps because of the bad quality of the raw material.

The 36% of the PB have a length and a width equal or lower than 20 mm (n=40). In the level 1 this percentage is the 30.4%, in the level 2 the 37.9%, and in level 3 the 45.2%. Therefore an important part of the PB has very small size. A 11.7% of the PB of Vinyets are laminar (laminar index greater than 1.6). This percentage is of 8.5% if we counted only the PB of more than 20 mm.

1GNB of Exploitation and Configuration We have identified three 1GNB utilized first like cores (1GNBE), and next configured like instruments. These objects are bifacial, so much in their first stage as cores like later, when the retouches of configuration were accomplished. One of the objects has medium size (90x71x58,5mm); the rest are of small size, with very similar dimensions (about 55mm of length).

The PB have butts most of all no cortical (in a 86% of Average size of the PB (mm) Level Length Width 1 24,1 21,6 2 25,9 23,2 3 34,7 26,5 Without context 23,6 22,0 Global average 25,1 22,5

Two of these 1GNB have denticulated cutting edges, while the third has a dihedral straight convex cutting edge, at the lateral-transverse zone. The three objects were knapped with flint, and rests of

Thickness 11,5 10,5 16,3 10,2 11,5

Table 2.7.3.- Average size of the Positive Bases of Vinyets (for each level)

Figure 2.7.4.- First Generation Negative Bases of Configuration (1GNBC) found in Vinyets. 1: Distal trihedron configured on a pebble of flint, by means of bifacial retouches (level 2). 2: Denticulate of flint, with tendency to uniangular or convergent (level 2)

Figure 2.7.5.- Positive Bases of Vinyets. 1: Limestone with centripetal scars (level 3). 2 and 4: Positive Bases of flint (level 1). 3: Very small Positive Base of flint (level 2). 5: Positive Base of limestone (level 1)

86

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA the cases), unifaceted (in the 55%), and of platform type (88%). However, also there are many PB with bifaceted (20%), and multifaceted butts (9%). Adding the 2GNB with analyzable butts on this calculation, the proportions barely change. The dorsal faces are by majority no cortical (77%), and have two or three arrises frequently. The existence of an outstanding number of bifaceted and multifaceted butts agrees with the characteristics of 1GNBE, with orthogonal knapping and cubic morphologies. Second Generation Negative Bases The 2GNB have an important role in Vinyets’s lithic record, since they are the 25.2% of all the industry; this percentage is much more elevated (38.5%) taking like reference only determinable objects. The 2GNB of Vinyets are of small size, with an average of 34,1x31,0x15,4 mm (Table 2.7.5). However, the average of the 2GNB is slightly larger than the PB. Figure 2.7.5.- Second Generation Negative Bases of Configuration, from Vinyets. 1: Denticulate of flint (level 1). 2: Convex distal dihedron of flint (end scraper) (level 1). 3: Convex lateral dihedron of flint (level 1). 4: Lateral-transverse dihedron, with bifacial retouches, (flint, level 2)

2GNB of Configuration The 97.5% of the 2GNB are shaped tools (78 implements). The 2GNBC have a larger size than the PB, with an average of 31x28,5x13,5mm. The 2GNB not elaborated with flint are something larger that the 2GNB of flint: 38,5x32x17mm in front of 31x28x13mm. Average size of the 2GNB (mm) Level Length Width 1 37,9 35,1 2 32,0 29,9 3 45,8 46,0 Without context 39,2 29,6 Global average 34,1 31,0

it happens with the rest of objects, almost all Fragments are of flint, and were recovered in the level 2. Most of the Fragments are no cortical, and of very small format, with the exception of the Fragments of level 3, with larger size (Table 2.7.5).

Thickness 15,9 15,8 22,2 14,1 15,4

Average size of the Fragments (mm) Level Length Width Thickness 1 23,7 17,9 12,0 2 24,8 19,1 12,9 3 36,3 29,8 22,9 Without context 20,8 18,2 9,4 Global average 24,6 19,2 12,7

Table 2.7.4.- Average size of the Second Generation Negative Bases of Vinyets (for each level)

There is a preponderance of the unifacial configuration, accomplished with extractions that affect to half or less than half of the perimeter of the object. The angle of retouches is in half of the objects acute, nevertheless a third part of the implements has steep angle. Extractions are deep or very deep regarding the edge.

Table 2.7.5.- Average size of Fragments of Vinyets (for each level)

Strategies for producing Positive Bases We have found 9 cores in Vinyets (7 1GNBE and 2 2GNBE). We have identified two Indirect Technical Operational Themes (ITOT), with a very different proportion. On the one hand there is a ITOT with predetermination of the final morphology of the flakes. We have note the existence of this reduction strategy thanks to the finding of PB. Unfortunately we do not have more objects belonging to this Operational Theme.

Over a 60% of the retouches are direct, although the 22% of alternating retouches highlights. There are 14% of inverse retouches. 2GNB of Exploitation We have located two 2GNB of Exploitation. One of them, very crusted, show unipolar linear knapping, while it is not possible to identify in a clear way the orientation of the removals in the other. The dimensions of these objects (61x54x27, and 49x51x34 mm) surpass slightly the average of the 1GNB of Exploitation (41,7x32x26,2mm).

The bifacial ITOT with linear and orthogonal knapping is better represented. We have found in the lithic record of the various levels so much PB like 1GNBE, and 2GNBC that belong to this flaking technique. Furthermore, we have several 1GNB that were reconverted in retouched tools (1GNBC), and 2GNB that were utilized as cores (2GNBE). Without a doubt this is the Indirect Technical Operational Theme better represented in Vinyets’s lithic record.

Fragments A third part of Vinyets’s lithic material are Fragments. As 87

Xosé Pedro Rodríguez of maximum length (debris). In general, the shaped instruments are of small size.

In few cases we could have determined the direction of the knapping in the scars of removals of the dorsal faces of the PB. When it has been possible appears like more habitual the unipolar linear knapping, and the multipolar orthogonal knapping, coinciding with the 1GNB of exploitation. Also we have identified some PB that result from a reorganization of the flaking surfaces of the core. An object with the morphology of a preconfigured small point, although of very small size (24x26x6,5mm ) stands out. This PB would indicate the presence of a strategy of knapping focused to the preconfiguration of the final morphology of the products. Unfortunately we have not found 1GNB of Exploitation corresponding to this Indirect Technical Operational Theme.

The denticulated configuration is present at almost a third part of the instruments, although the majority have continuous retouches. The most habitual delineation of the cutting edges is convex or else straight. The shaping of dihedral cutting edges was the basic objective of the configuration of 2GNB (48 objects) (Figure 2.7.6.2., 2.7.6.3, 2.7.6.4). Concretely, in 18 objects lateral dihedrons were configured (Table 2.7.6) (Figure 2.7.6.3); in 13 concave dihedrons (notches); in 11 transverse dihedrons (Figure 2.7.6.2); and in 6 lateral-transverse dihedrons (Figure 2.7.6.4). The denticulated cutting edges are also frequent (13 objects) (Figure 2.7.6.1), as well as trihedrons (n=10).

The configuration of tools The processes of configuration of instruments have a great weight in this lithic record. These artifacts (1GNBC and 2GNBC) suppose almost the 40% of lithic identifiable material (once the Fragments were excluded).

There are 65 classifiable objects following Laplace’s typology. Denticulate are majority (n=19), followed by the pieces with steep angled retouch (n=17), the acute-angled points (n=11), and the side scrapers (n=10). Summing all points we obtain 15 pieces: 8 with continuous retouch (P1 and P2), and 7 denticulated (D4). Also there are 13 notches: 6 with acute angled retouch (D1), and 7 with steep angled retouch (A21). We have identified 9 end scrapers: 5 with acute angled retouch (G11, G12), and 4 with steep angled retouch (A15).

Pebble tools The most important Direct Technical Operational Theme consists in the configuration of a trihedron at the distal transverse sector of the artifact (Table 2.7.6). The three artifacts classified as 1GNBC fit in this model. It is pertinent to highlight the existence of three 1GNB utilized first like cores (1GNBE), and next configured like instruments. Two of these 1GNB have denticulated cutting edges, while the other has a dihedral straight convex cutting edge, at the lateral- transverse zone (Table 2.7.6).

To sum up, the objective was often to create dihedral cutting edges in a lateral or at the object’s transverse sector (Table 2.7.6). This type of configuration appears in 38 instruments (48.1%). Also the denticulates (19%), the concave dihedrons (6.5%), and trihedrons (16.5%) have a highlighted role.

It is possible that some of the cortical PB come from the configuration of these artifacts. However, the little amount of cortex in the 2GNB indicates that the flakes that come from this process for shaping pebbles almost never were retouched.

Conclusions Vinyets is an open air site with lithic industry in stratigraphic context. This site is placed close to the Gaià river, and close to Mediterranean sea. The lithic record studied in this work was recovered during two campaigns of excavation that affected three archeological levels. The lithic material is not much numerous, but there is a great homogeneity among the different levels, hence it has been studied jointly.

Flake tools The elevated presence of 2GNBC stands out specially. We have calculated that about half of all the blanks (initial flakes) derived from the knapping of cores were configured and transformed in 2GNBC. Concretely, of a total of 151 initial products (PB of more than 20 mm+2GNB), 78 were retouched (51.7%). The processes of configuration are represented also by numerous PB of less than 20 mm

Morphodynamic capacity of the retouched tools Transversal dihedral edge, with straight-convex morphology Lateral dihedral edge, with straight-convex morphology Transversal and Lateral dihedral edge, with straight-convex morphology Concave dihedral edge (“notch”) Transversal dihedron, and lateral denticulate Lateral dihedron, and lateral denticulate Distal Trihedral edge (“picks” and “points”) Straight-convex denticulate edge Uniangular (convergent) denticulate edge (“denticualte point”) Total identified

1GNBC 0 0,0% 0 0,0% 1 16,7% 0 0,0% 0 0,0% 0 0,0% 3 50% 2 33,3% 0 0,0% 6 7,6%

2GNBC 11 15,1% 18 24,7% 6 8,2% 13 17,8% 1 1,4% 1 1,4% 10 13,7% 6 8,2% 7 9,6% 73 92,4%

Table 2.7.6.- Morphodynamic capacity of the retouched tools of Vinyets

88

TOTAL 11 13,9% 18 22,8% 7 8,9% 13 16,5% 1 1,3% 1 1,3% 13 16,5% 8 10,1% 7 8,9% 79

MIDDLE PLEISTOCENE SITES IN THE NORTHEAST OF THE IBERIAN PENINSULA

Figure 2.7.7.- Morphogenetic matrix of the lithic industry of Vinyets

Taking all of the technical features into account, we propose the hypothesis that this lithic industry can be classified as initial Mode 3.

Unfortunately faunal remains were not recovered. The raw material utilized in Vinyets was mainly flint (nodules of the Gaià river). The source of raw material was very close to the site.

It is very difficult to propose a chronology for these occupations with the data that we have. Their stratigraphic position (eleven meters below present-day soil), and the characteristics of the infilling (that includes various fossilized soils), can indicate a final Middle Pleistocene age.

A bifacial strategy with linear and orthogonal knapping was utilized. This system of exploitation allows profiting from nodules of scarce quality. Also we verified the utilization of more complex strategies, that imply a predetermination of the final morphology of the flakes. The linear orthogonal strategy is well represented (1GNBE, PB, 2GNBC, 2GNBE), while the Indirect Technical Operational Theme with preconfiguration of the flake morphology only is present through PB. Probably the production of preconfigured objects was not accomplished in the site. In conclusion, the processes of production are concentrated in a reduction strategy (ITOT), that was accomplished on flint nodules of small size, picked up from the paleocourse of the river Gaià. However, the configuration has a notable variability, with search of various morphodynamic potentials. Most of all the objective was to create dihedral cutting edges in one lateral or at the transverse sector of the object. Nevertheless, also the denticulate, the concave dihedrons, and trihedrons stands out. The role of the processes of shaping flakes (2GNBC) is very significant. The scarce number of lithic implements, taking the excavated surface into account, and also the volume of evacuated sediment, seems to indicate the existence of not much intense occupations. These occupations would have like objective, most of all, the configuration of flake tools (2GNBC), previously obtained with unipolar and orthogonal, little complex knapping.

89

3 THE ATAPUERCA SITES

ATAPUERCA SITES

3.1.- INTRODUCTION

mation of a large karstic system. Overlying the cretaceous limestone are calcareous conglomerates, and red sandstone of Oligocene age (Olive et al., 1990; Pineda and Arce, 1993). Also there are Miocene deposits, made of (from old to young) marls, clays and gipsiferous sandstone overlaid by 20 m of limestone and marls with large silica nodules (Parés and Pérez-González, 1999).

Location The Sierra de Atapuerca is located 14 kilometers East of the city of Burgos (Figures 3.3.1 and 3.1.2). This small mountain range occupies an approximated surface of 25 square kilometers. The two more elevated peaks of Atapuerca are Matagrande, with 1,078 meters a.s.l. (at the north sector), and San Vicente with 1,082 meters (at the meridional sector) (Figure 3.3.1).

The landscape of Atapuerca (Figure 3.1.2) is dominated Atapuerca is nearly the limit between the hydrographic basins of the rivers Duero and Ebro. Concretely, Atapuerca is placed in the north-oriental border of the Duero basin, close to a natural corridor (the so-called Corredor de la Bureba), that connect with the Ebro Basin. This natural gateway allows crossing a mountainous zone characterized by two great mountain ranges: the Sierra de la Demanda (to the Southeast) and the Cordillera Cantábrica (to the Northwest). This strategic situation confers to Atapuerca some special features, that can explain, at least in part, the intensity of human occupation from the lower Pleistocene to historic times. The proximity of the Sierra de la Demanda influenced of important manner the human settlement of the Sierra de Atapuerca. The Sierra de la Demanda is an ecosystem with a notable variety of biotopes, that propitiates a great biodiversity. In this context, Atapuerca is a biological lure, attracting groups of settlers from the Sierra de la Demanda, specially in moments of crisis. This capability to attract biological communities would not exist if the Sierra de Atapuerca not be a potential source of exploitable resources. In this sense, this mountain range works like an ecotone: with hydric resources (rivers Arlanzón, Pico and Vena), and with vegetable resources of grassland and of low mountain. Another very important factor contributed to the human occupation of the Sierra de Atapuerca during Prehistory. During the Pliocene and beginnings of the Pleistocene a complex karstic system was generated. The excavation of the sediments accumulated at these cavities along the Quaternary has provided a magnificent archeological and paleontological record.

Geological context From the structural point of view, the Sierra de Atapuerca belongs to the Iberian Range (Olive et al., 1990; Pineda and Arce, 1993), and is a NNW–SSE anticline verging on the NE (Parés and Pérez-González, 1999). Basically the structure of the mountain range is composed by Upper Cretaceous limestones, dolomites and calcarenites. The existence of dolomites and limestones has favored the for-

Figure 3.1.1.- Location of Atapuerca Sites, near de city of Burgos. 1: Elevation in meters, above the sea level. 2: Main river. 3: Small river. 4: Road. 5: Village. 6: Abandoned railway. 7: Gran Dolina site. 8: Trinchera Penal site. 9: Galería Complex. 10: Sima del Elefante site. 11: Sima de los Huesos site. 12: Portalón site. 13: Mirador site. 14: Valle de las Orquídeas site

93

Xosé Pedro Rodríguez

Figure 3.1.2.- The Sierra de Atapuerca, panomaric view from the South (©Atapuerca Research Team)

Figure 3.1.3.- Location of the sites of Cueva Mayor karstic system (shaded), and sites of the abandoned railway trench. TD: Gran Dolina. TP: Trinchera Penal. TZ: Cueva de los Zarpazos. TG: Galería. TN: Trinchera Norte (TZ, TG and TN form the Galería complex). TE: Sima del Elefante. P: Portalón. SH: Sima de los Huesos (adapted from the original topography by Grupo Espeleológico Edelweiss)

by some not much prominent tops. This relief is characterized for shallow valleys and structured slopes, and, in the Arlanzon Valley, for terraces with smoothed ledges. All these physiographic elements are dominated by the residual relief of the Sierra de Atapuerca, the highest point of which is 1082 m, and represents an Oligocene erosion-

al surface (Parés and Pérez-González, 1999; Zazo et al., 1983; 1987). At present the river Arlanzón flows to less than two kilometers of the southern border of Atapuerca range. This river

94

ATAPUERCA SITES has built in its right margin a system of terraces. After the first systematizations of Zazo, Goy and Hoyos (Zazo et al., 1983; 1987) , Pérez González et al. (Pérez-González et al., 1995; Pérez-González et al., 1999) have distinguished six old terraces: 0.50–1 m (actual alluvial plain), +3 m (T6, Ibeas de Juarros Terrace), +10 m (T5), +20 m (T4), +35 m (T3), +60 m (T2, La Laguna) and +75 m (T1, Escampa Colina). According to these authors this oldest terrace is the highest one in the transect at Ibeas de Juarros and is 1000 m high (a.s.l.), incised in lacustrine limestone.

Archaeological sites in the Sierra de Atapuerca There are many archeological sites in the Sierra de Atapuerca, from the Pleistocene to historic epoch (Tables 3.1.1 amd 3.1.2). Until now the archeological interventions in the Sierra de Atapuerca mainly have been accomplished at two zones: “Trinchera del Ferrocarril” (an abandoned railway trench), and “Cueva Mayor” (Figures 3.1.1 and 3.1.3). Gran Dolina is the most important site in the first zone, whereas the Sima de los Huesos is the most outstanding one in the second zone.

The fluvial deposits of the Arlanzón are composed mostly of slates and quartzites. In the present-day stream bars with some Tertiary limestone clasts can be observed which apparently are absent in the higher terraces because of the erosion (Parés and Pérez-González, 1999).

During the last years, archaeological surveys also were accomplished, as well as excavations in open air sites (Navazo, 2002), with archeological material in stratigraphic context (Valle de las Orquídeas, and El Hundidero).

Probably the limestones of Atapuerca experienced processes of karstification before the formation of the Oligocene erosional surface. Nevertheless, we can not demonstrate the existence of the karst during the Paleogene, because the exokarstic forms of this supposed karst were dismantled by the generation of the Oligocene erosional surface, and the internal forms are not recognizable nowadays. It is indubitable that during the Neogene an intense karstic activity had place, with development of exokarstic and endokarstic forms.

The sites of Gran Dolina (TD), Galería-Zarpazos (TGTZ), Penal (TP), and Sima del Elefante (TE) are located in the Trinchera del Ferrocarril (TF), in the meridional part of the Sierra de Atapuerca (Figures 3.1.1 and 3.1.3). The Trinchera forms an arch of some 500 meters of length (Figure 3.1.3). Its southern extreme is placed to 3° 31’ 05’’ W of Grenwich’s meridian, and to 42° 21’ 00’’ N , while its northernmost point is placed to 3° 31’ 18’’ W, and 42° 21’ 13’’N . The depth of the Trench attains at the most twenty meters.

The endokarst of Atapuerca has a phreatic origin, whose principal phase of development coincides, according to Zazo et al (Zazo et al., 1983; 1987), with the sedimentation of the Aragonian or Middle Vallesian limestones. Cueva Mayor and the caves of the abandoned railway trench, would be senile forms, along with the terrace T1 and T2, presumably of Lower Pleistocene age. During the Quaternary partial reactivations of the karst took place, relating to climatic changes and descents of the base level of the Arlanzón river.

Galería-Zarpazos and Dolina have been the most excavated sites (Figure 3.1.4). Later we will talk largely about these two sites, specially Gran Dolina. The excavations accomplished in Penal (TP) suggested, for the scarcity of findings, to concentrate efforts at another sites. The bottom part and the top of the Sima del Elefante were excavated simultaneously from 2001. At the bottom we have recovered fauna and lithic industry that would be from the end of the Lower Pleistocene (Rosas et al., 2001). On the southern slope of the Sierra de Atapuerca there is

Gran Dolina (TD) Galeria (TG)

Penal (TP) Sima del Elefante (TE) Sima de los Huesos (SH)

No excavaton

Cueva Zarpazos (TZ)

Cueva del Mirador (MIR) Portalón (P) Valle de las Orquídeas (OR) Hundidero (HUN) Table 3.1.1.- Campaings of excavation (shaded) in the sites of the Sierra de Atapuerca, from 1976 to 2004

95

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

1986

1985

1984

1983

1982

1981

1980

1979

1978

1977

1976

1987

excavation campaigns

SITES

Xosé Pedro Rodríguez SITES Lower Pleistoc.

Chronology Midde Upper Pleistoc. Pleistoc.

Archaeological remains Bronze Neolithic Age

Lithics

Fauna

Humans Ceramics

Gran Dolina (TD) Galeria (TG) Cueva Zarpazos (TZ) Penal (TP) Sima del Elefante (TE) Sima de los Huesos (SH) Cueva del Mirador (MIR) Portalón (P) Valle de las Orquídeas (OR) Hundidero (HUN) Table 3.1.2.- Main archaeological sites in the Sierra de Atapuerca, with their chronology (shaded), and the archaeological remains recovered (shaded)

another archaeological site, the Cueva del Mirador. Until now we have located at this site materials of the Bronze Age, and of the Neolithic (Vergès et al., 2002).

the possibility that this instrument have some symbolic significance (Carbonell et al., 2003). The totality of the skeletal elements are represented in the human sample of the Sima de los Huesos. The greater part of the researchers that have studied the fossils of the Sima de los Huesos think that it has to do with an intentional accumulation of corpses, made by hominids. However, some researchers thinks that the accumulation is consequence of a catastrophic phenomenon, that surprised a human group inside the cave (Aguirre, 2000).

Sima de los Huesos (SH) is the most important site of Cueva Mayor, but there is also a Neolithic site in the entrance of the cave (El Portalón), as well as Bronze Age funeral chambers in the Galería del Sílex (Figure 3.1.3).The Sima de los Huesos site is around 0,5 km from the Cueva Mayor present-day entrance (Arsuaga et al., 1997a). The sample of human fossils of the Sima de los Huesos is the more extensive and complete of the worldwide Middle Pleistocene. To this date about 4000 human fossils have been recovered. Three very complete skulls, and a pelvis stands out (Arsuaga et al., 1993; Arsuaga et al., 1997b; Arsuaga et al., 1999b). Taking into account the material recovered until now, the minimal number of individuals has been estimated in 28 (Bermúdez de Castro et al., 2003a; Bermúdez de Castro et al., 2003b; Bermúdez de Castro et al., 2004). The fossils correspond to a single biological population, classified as Homo heidelbergensis.

Discovery and first archaeological interventions The first systematic explorations in Cueva Mayor were accomplished at the beginning of the century XIX (Carbonell et al., 1999a; Rodríguez et al., 2001). Nevertheless, a long time before the cave was visited of more or less assiduous way. On the way in for the first time at the cave the great quantity of graffitis that exist in the walls attracts the attention. Some inscriptions have great antiquity, until now the most ancient belong to the century XVI. The first written documents about the cave date from the half of the XIX century, when Felipe Ariño requested the concession in property of the cave. Shortly after, in 1868, the first detailed description of Cueva Mayor was published by Sampayo and Zuáznavar (with drawings held by Isidro Gil). The pit that today we know as “Sima de los Huesos” was mentioned for the first time in this work. Second direct reference to the Sima de los Huesos appears in a request of 1890, to obtain authorization to accomplish a mining operation in Cueva Mayor.

The deposit of the Sima de los Huesos contains furthermore an important collection of carnivores, fundamentally Ursus deningeri, but also Panthera leo sp. fossilis, Panthera sp., Lynx pardina spelaea, Felis silvestris, Martes sp., Mustelidae indet., Canis lupus, Vulpes vulpes (García et al., 1997; García and Arsuaga, 2001; García, 2002). Among the micromammals Allocricetus bursae and Pliomys lenki stands out (Cuenca-Bescós et al., 1997). The faunal association, the characteristics of the human fossils, and the dates of the bones and speleothems indicate an age of less than 350.000 years, possibly between 400.000 and 500.000 years (Bischoff et al., 2003). We have not found herbivora’s remains at the Sima de los Huesos. Only a lithic instrument has been discovered: in 1998 a quartzite handaxe was found at a deposit that was containing human and carnivores remains. Some researchers have discussed

Parallel to these explorations, more or less scientific, the cave suffered frequently spoliations. There is an inscription in the walls of the Sima de los Huesos that dates back to the 1898. Precisely a determining fact for the future of the Sierra de Atapuerca took place to endings of the XIX century: the 96

ATAPUERCA SITES

Figure 3.1.4.- The sites of Gran Dolina and Galería Complex, located in the Trinchera del Ferrocarril (the abandoned railway trench) (©Atapuerca Research Team)

(Martínez Santaolalla, 1926) made reference to the Cueva Mayor in his study of the Neolithic in the Burgos area.

excavation, no far from Cueva Mayor entrance, of a trench for the circulation of a mining railway. In 1896 the British company “The Sierra Company Limited”, created by Richard Preece Williams, obtained the permission to initiate the construction of a railway between Monterrubio de la Demanda and Villafría. The railway should have crossed the Southwest sector of the Sierra de Atapuerca, for which was necessary to excavate a trench in Cretaceous limestones. The railroad was already in working in 1901.

Between 1926 and 1950 there are very few news of discoveries or investigations related with the Atapuerca sites, certainly due to the political and social contingencies of the moment. The interest for the fossiliferous fillings of the Trench (TF) started to grow from the fifties of the past century. In that epoch, the Trench was used as stone quarry for the extraction of limestone.

The railroad stopped working around 1910, and in 1917 the “Basque-Castilian society”, inheritor of “The Sierra Company Limited” went bankrupt and disappeared definitively. The Railway’s Trench (TF) had left in evidence in its walls some of the conduits of the karst, filled with sediments. Many years later it was proved that these sediments contained important archaeological and paleontological remains.

In 1962 members of the Edelweiss Speleological Team discovered the existence of fossils at the Trinchera del Ferrocarril (TF). In that year, Uribarri, a member of the Edelweiss Speleological Team (which had been discovering fossils since 1954), notified the Museum of Burgos of the presence of fossil bones in the locality presently known as Galería. Encouraged by this discovery, on 15 April 1963 B. Osaba, Director of the Museum of Burgos, visited the Trinchera, accompanied by members of the Edelweiss Group and the Association ‘‘Amigos de Burgos’’. During this visit Acheulian artifacts were also found. In 1964 a prospecting campaign was carried out. It demonstrated the importance of these karstic infillings because of the recovery of faunal remains and their associated industries (Palol, 1969a: 296). In 1966 an excavation campaign was carried out which led to the discovery of a lot of fauna, but scarcely lithic material (Palol, 1969b). On the other hand,

However, before recognizing the archaeological importance of the TF, the Cueva Mayor had become a site of interest. Carballo (Carballo, 1910) makes reference to the existence of a Bronze Age site in the Cueva Mayor entrance. Soon after, probably in 1912, Obermaier and Breuil visited the cave and analyzed the cave paintings of Cueva Mayor and Cueva del Silo (Breuil and Obermaier, 1913; Breuil, 1933; Obermaier, 1925). Probably the first researcher that highlighted the geological and paleontologic value of Atapuerca was Royo y Gómez, after a visit in 1923 (Royo y Gómez, 1926). At about that time, Martínez Santaolalla 97

Xosé Pedro Rodríguez Narciso Sánchez, led by the paleontologist Miquel Crusafont, director of the Paleontology Institute of Sabadell, recovered many fossils in the sites of the Trinchera (TF), especially in the Galería-Cueva de los Zarpazos (Carbonell et al., 1999a; Rodríguez et al., 2001).

gan in 1984, and numerous human fossils appeared. At this moment the sites of Gran Dolina, Galería and Sima de los Huesos were excavated simultaneously. In 1985, the vegetation in the area opposite Dolina (on the other side of the Trinchera del Ferrocarril) was cleared, yielding some stone tools. This area was named Trinchera Penal (TP) (Figure 3.1.3).

In 1972, a team of American archaeologists prospected in the Trinchera del Ferrocarril and drilled into the Cueva Mayor entrance. This work, directed by Geoffrey A. Clark and Lawrence G. Strauss, showed the existence of abundant fauna in two infillings in the Trinchera del Ferrocarril (Clark, 1979). Clark & Strauss also mentioned the presence of engravings in the Cueva del Silo entrance, which might be Calcolithic according to Breuil (Breuil, 1952). The materials recovered when drilling in July 1972 were studied by Apellániz (Apellániz, 1979), who excavated in the same place in 1973. Apellániz found Calcolithic, Bronze Age and late Roman (fourth and fifth centuries AD) occupation sites in the Portalón de Cueva Mayor (Apellániz and Domingo Mena, 1987).

In the meantime, the excavation of Gran Dolina, Galería and Sima de los Huesos continued, and samples were collected at the Sima del Elefante. There was a change of strategy in Gran Dolina’s excavation in 1990. No field work was carried out in the uppermost part of TD, only at its base (level TDW4, 25 m2 of excavated surface), which yielded important remains of macrofauna and few, nonetheless significant, remains of lithic industry. They are among the most ancient evidence of human presence in the Sierra de Atapuerca (Carbonell and Rodríguez, 1994).

In 1972 the Edelweiss Speleological Team discovered a new gallery in the karstic system of Cueva Mayor, known as Galería del Sílex (Martín Merino et al., 1981). This is an interesting site with Neolithic, Calcolithic and Bronze Age materials (Apellániz and Uribarri, 1976; Uribarri and Apellániz, 1975).

Until 1990 the director of the excavation was Emiliano Aguirre, whereas from 1991 the directors of the research project and of the excavations have been Juan Luis Arsuaga, José María Bermúdez de Castro and Eudald Carbonell. In 1991 the operation at the base of TD (TDW4) was completed, yielding new finds of macrofauna, but not of stone tools. New stratigraphic studies of this site were also carried out. An excavation was carried out in the Trinchera Penal in 1992, with two test trenches being made in order to evaluate the site’s archaeological potential. As the fauna and stone tool remains were scarce, it was decided not to continue the operation. This year TD was not excavated, but samples of microfauna were taken from the stratigraphic section, and the excavation areas were expanded, clearing the entrance to the upper part of the infilling.

Trinidad Torres excavated in the Trinchera del Ferrocarril in 1976 with the aim of finding cave bear (Ursus) remains for his Doctoral Thesis. Torres excavated in the base of Gran Dolina and Galería, where he recovered a fragment of human jaw, without stratigraphic context (Bermúdez de Castro and Rosas, 1992). Next, Torres decided to excavate in the Sima de los Huesos in which a great abundance of bones had been reported. During the excavation in the Sima de los Huesos he discovered a large amount of Ursus remains and human fossils, including a complete mandible. These fossils were studied by Aguirre and Lumley (Aguirre et al., 1976; Aguirre, 1977; Aguirre and Lumley, 1977), who classified them as Pre-Neanderthal, chronologically Middle Pleistocene. Fossils were shown for the first time in public in 1977 during a scientific Congress in Morella (Castelló, Spain).

During 1992 in Galería the level TG10 was excavated. This site was preserved this year by a structure scaffoldings with coverage. There were important findings at Sima de los Huesos, particularly three skulls, one of them complete (skull 5) (Arsuaga et al., 1993). This year an International Congress was held at the Castillo de la Mota (Valladolid), to show the new findings of the Atapuerca sites (Bermúdez de Castro et al., 1995).

The unquestionable interest of the Atapuerca sites encouraged Emiliano Aguirre to plan a scientific project, to undertake their systematic study. This project started in 1978, and Aguirre led the excavations himself (Aguirre, 1995a).

In 1993 it was decided to initiate a new strategy for the excavation of Gran Dolina. A biostratigraphic test pit was begun in view of the need to obtain systematic knowledge of the records of the whole series, and to determine the excavation method to be applied when extensive operations were to be started (Figures 3.1.5 and 3.1.6). The 6 m2 test pit had to cover Gran Dolina’s 18 m of stratigraphic thickness. The excavation of the test pit would provide a diachronic sample of the deposit, which could guide us in the future extensive excavation of the site. In this first year, the sounding test was in the level TD10 (which contained

In 1978, the sites were prepared for excavation; the vegetation at the uppermost part of Gran Dolina (TD) and Galería (TG) was cleared and paleontological, and sedimentological samples were collected. The excavation of Cueva de los Zarpazos began in 1980. In 1981, the unproductive levels of the uppermost part of Gran Dolina Dolina were removed, and then excavation by layers started. The systematic excavation of the Sima de los Huesos be98

ATAPUERCA SITES

Figure 3.1.5.- Schematic plan of Gran Dolina comprising the excavated zones: squares excavated between 1996 and 1997 in white, and squares excavated between 1981 and 1989, shaded. Both zones are excavated from 1998. The diagonal lines indicate the location of the biostratigraphic test pit, which was excavated between 1993 and 1997

abundant lithic artefacts and fauna), and TD9 (without archaeological remains). The 1994 fieldwork on Gran Dolina was thus started with a fresh outlook. Excavations began at TD8 (with fauna, but without lithic industry), and TD7 (with abundant fauna and one object of lithic industry). The TD6 level, where abundant fauna, lithic industry and human remains were found, was reached at the beginning of July (Figures 3.1.6 and 3.1.15) (Carbonell et al., 1995a). The microfauna indicate a minimum age of 500 ka (due to the presence of Mimomys savini), but the results of the paleomagnetic analyses, which had been undertaken at the beginning of the campaign, gave an age older than 780 ka (Carbonell et al., 1995a; Parés and Pérez-González, 1995; 1999). The site was covered and thus protected from natural deterioration. These findings confirmed the importance of Gran Dolina’s site, that had provided the more ancient human fossils of Europe. This year also a sounding at the Sima del Elefante was accomplished (Rosas et al., 2001).

Figure 3.1.6.- General view of the biostratigraphic test pit of Gran Dolina. In 1994 the excavation of the test pit reached level TD6 (©Atapuerca Research Team)

In 1997 the biostratigraphic test column reached level 5, while extensive excavation continued at level 10. The study of human fossils from TD6 culminated in the naming of a new species: Homo antecessor (Bermúdez de Castro et al., 1997). Between 1997 and 1999 interventions were carried out most of all in Gran Dolina and Sima de los Huesos. The excavation of a sounding of 9 m2 at the Cueva del Mirador began in 1999. This cave is located in the southern extreme of the Sierra de Atapuerca (figure 3.3.1). Lithic instruments, fauna and human remains correspondent to the Bronze Age were discovered during this excavation (Moral, 2002; Vergès et al., 2002). Neolithic materials appeared in later campaigns.

In 1995 the biostratigraphic test pit of Gran Dolina continued to yield material, including human fossils. Meanwhile, the upper part of Gran Dolina was prepared for extensive future excavation. In the meantime, at the Cueva de los Zarpazos a cranial human fragment was discovered. The human remain is similar to the Sima de los Huesos human sample (Arsuaga et al., 1999a). This fossil appeared associated to lithic instruments of Mode 2 technology. In 1996 the excavation of the test column in Gran Dolina was still concentrated on level 6, although it had gone beyond the layer with human remains. Simultaneously, extensive excavation of the upper part of Dolina was begun (with a surface of 80 m2), initiating operations at the upper part of level TD10 (Figure 3.1.5 and 3.1.7). The excavation of Cueva de los Zarpazos finalized that year. We began a biostratigraphic sounding at the bottom of Sima del Elefante. In the meantime, the excavation of the Sima de los Huesos continued.

Also archaeological surveys at various zones of the Sierra de Atapuerca began in 1999. These prospections have allowed discovering several open air sites (Navazo, 2002). The excavation of El Portalón site, located in the entrance of Cueva Mayor, began in 2000. This site also has provided Neolithic materials and of Bronze Age (Carretero, 2001). In 2000 and 2001, there were excavations at the open air 99

Xosé Pedro Rodríguez conduit that in the past was open to the exterior), and of TZ (where the sediment of the inside of the karst comes into Galería). Lithic industry and faunal remains have appeared at the three sectors. Besides, in 1995 one fragment of a human skull was found in TZ, with a dating about 300 Ka (Arsuaga et al., 1999a). Stratigraphy and chronology The stratigraphy of Galería was elaborated for the first time by Gil, Aguirre and Hoyos (Gil et al., 1987), and after it has been redefined by Pérez González (PérezGonzález et al., 1995; Pérez-González et al., 1999). According to Pérez González the stratigraphy of Galería has six Units (GI to GVI, numbered from bottom to top) (Pérez-González et al., 1995) (Figure 3.1.9). GI’s sedimentation corresponds to facies of the interior of the cave, without direct influence from the exterior. Archaeological material has not been recovered in this Unit. An speleothem placed in the top of this Unit has been dated, with 317±60 Ka (Grün and Aguirre, 1987) (Table 3.1.3). These conditions vary as from GII. Although the base of this new fill unit keeps some GI features, like the presence of guano horizons and discontinuous, thin speleothems in the most basal layers, the direct influence from outside is obvious, with the entry of external detritus of gravels, very fine angular/sub-angular gravels, and muds, organized into gravity flows. Free falls (debris falls) of non consolidated altedered material from the walls and top of Galería could accompany these mass movements. This latter process proves necessary to explain the opening of the cavity to the outside and the subsequent change in sedimentation. Various archeological levels, with fauna and lithic industry, have been located in GII.

Figure 3.1.7.- Excavation of the upper part of the level TD10 of Gran Dolina, during 2001 (©Atapuerca Research Team)

site called “Valle de las Orquídeas”, placed North of the Railway Trench (TF). This site corresponds to the end of the Upper Pleistocene. In 2001 the upper part of Sima del Elefante was excavated, discovering faunal remains and lithic industry of the end of the Middle Pleistocene. Simultaneously the excavation of the lower levels of this site continued. The excavation at Cueva de los Zarpazos (TZ) was reinitiated in 2002.

The GIII fill unit is installed in unconformity on GII. This new hiatus is also of an unknown time value. The deposits are organised into two main facies which interfinger. They both relate to two very different environments and their deposition time does not pertain to one and the same moment. Gravity deposits still accumulate on the Galería’s south wall in a much greater quantity than those which enter on the norht side, whilst in the central sectors, filling up starts by the accumulation of clastic elements arranged in a rhytmic series formed by a sequence of two lithological layers. One is the lower with very fine, homometric angular/subangular gravels with bed load water action. The other, the upper one, is mud and seems to represent deposits of vertical accretion or traction on some occasions, because of the possible presence of isolate ripples and very low angle cross planar bedding laminations. These muddy layers frequently display scours at their top, originated by the currents which transported the very fine gravels from the upper sequence. This unit has archeological material, distributed in various levels. In general, the material decreases

In 2003 and 2004 TD, TZ, TE, SH, CM and P were excavated. The most important discoveries were accomplished in Gran Dolina and Sima del Elefante. Two human fossils appeared in 2003 and six in 2004 when the stratigraphic section of the level 6 of Gran Dolina were cleaned. These fossils are of Homo antecessor. In 2003 lithic industry appeared in the lower part of Sima del Elefante, with an age that would be more than one million years ago.

The Galería Complex The complex of Galería is located at 3°31’07’’ East, and 42° 21’ 09’’ North. Galería’s archaeological complex includes three zones (Figure 3.1.3 and 3.1.8): The central zone, that it is Galería properly; Trinchera Norte (TN), that is a vertical pit placed East of Galería; and Cueva de los Zarpazos (TZ), a little cavity placed West of Galería. The three sectors are in contact, and Galería receives sedimentary contributions from TN ( through the vertical 100

ATAPUERCA SITES

Figure 3.1.8.- Galería Complex, to the beginnings of the decade of 1980, just after beginning the systematic excavations. TZ= Trinchera Zarpazos, TG= Trinchera Galería, TN= Trinchera Norte (©Atapuerca Research Team)

Figure 3.1.9.- Stratigraphic cut of Galería, with radiometric ages (adapted from Huguet 1997)

The GV Unit pertains to the fill phase of Galería,’s vertical chimney and, at the same time, represents the filling up of the cavity. There are no less than 6 or 7 mass clastic movements of supported subangular gravels, separated by texturally finer sandy silt or sandy silt-clayey horizons, which support very fine gravel clasts.

as we got close to the top of this Unit (Pérez-González et al., 1995). The GIV deposits are arranged in erosive discordance on GIII, and display a group of facies similar to those of the previous fill phase, but the alternating nature is not as evident, at least in the Galería’s southern half. Here, the current structures of the layers of very fine gravel, and filled scours enable aparent east-west and south inflow directions to be measured. This unit and the following have not provided archeological material. There are various dates of a stalagmitic crust (archaeological level TG12), the majority of which are between 180 and 200 ka.

GVI is the oldest edaphic formation preserved on the sides of Sierra de Atapuerca. This is a deep soil, the C horizon is 1,80 m down. The various dates carried on in Galería’s Complex indicate that the human occupations had place between 450 and 250 Ka, that is between the isotopic stages 7 and 11 (Table 3.1.3). 101

Xosé Pedro Rodríguez Stratigraphic Units GVI

Archaeological level

Dating (Ka)

Method

Reference

Top of TZ

GIV

TG12

GIII GII

TN8 TG8

GI

TG4

135 ±13 118 +71/-49 177,3 ±23 180 211 ±32 222 ±31 256 ±33 >350 >350 317,6 ±60

U/Th U/Th ESR ESR ESR ESR ESR U/Th U/Th ESR

(Pérez-González et al., 1999) (Grün and Aguirre, 1987) (Grün and Aguirre, 1987) (Rosas et al., 1998) (Falguères, 1986) (Falguères, 1986) (Falguères, 1986) (Carbonell et al., 1995b) (Grün and Aguirre, 1987) (Grün and Aguirre, 1987)

Taxon Ursus spelaeus Cuon alpinus europaeus Canis lupus Vulpes vulpes Lynx pardinus spelaeus Felis silvestris Mustela sp. (erminea/nivalis) Panthera leo Meles meles Martes sp. Megalocerso cf. giganteus Cervus elaphus Dama dama clactoniana Caprinii indet. Bovinae indet. Equus caballus cf stenheimensis Equus caballus cf germanicus Stephanorhinus hemiotecus Stephanorhinus hundsheimensis

GII

Table 3.1.3.- Radiometric dates for the stratigraphic untis of Galeria Complex

GIII

The information that we presented is the fruit of the work of all the team of the Area of Prehistory at the Universitat Rovira i Virgili. We are not going to delve now deeply into the specific characteristics of this record. We have preferred to hold our study in depth for Gran Dolina. In reality, Galería’s lithic industry belongs to more than 20 archeological levels, the majority with scarce lithic material. The levels with more objects are TN2 (with 248), TN5 (n=190), TG11 (n=176), TG10B (n=116), TN6DA (n=113), and TN7 (also with 113). The rest of the levels have less of one hundred artifacts. In spite of the fact that there are an elevated number of archeological levels, the lithic industry of TG and TN has been studied jointly because its homogeneity, as a general rule. An important part of the lithic record (32%) consists of indeterminable objects. For the most part they are objects that can not be examined because their bad preservation. This percentage is particularly high among the objects of flint (42%). The Positive Bases are the most numerous objects. Also the presence of 2GNBC (with a 23%), and of natural Bases (17%) is very outstanding. The 1GNBE are the category with less objects, with only a 2.1% (Table 3.1.5).

Table 3.1.4.- Macromammals found in Galería (Cáceres, 2002; Cervera et al., 1999; García, 2002; Huguet et al., 2001; Made, 1999; Made, 2001; Sánchez and Cerdeño, 1999)

Archaeological remains This site has provided an important archeological record. Concretely, we have recovered over 1600 lithic artifacts, about 7000 faunal remains, numerous micromammals remains, and two human fossils (Carbonell et al., 1999c). The majority of this archeological record has appeared in the sectors TG (Galería) and TN (Trinchera Norte, North Trench). In TZ (Cueva de los Zarpazos) around 200 lithic objects have been recovered, as well as thousand macrofauna’s remains, and the two human fossils above-mentioned (García-Medrano, 2003).

Raw materials Flint is the most commonly used raw material (66.8%), while quartzite reaches 19.9%, and sandstone 8.4% (Table 3.1.5). The rest of rocks has little significant percentages. All raw materials could be collected close to the sites, in a radium of less than three kilometers. The flint utilized at Atapuerca’s sites is of two types: flint of Cretaceous age, and flint of Neogene age (Carbonell et al., 1999c; García-Antón et al., 2002). The present-day identification of the type of flint of the archeological objects is not easy. For this reason we have not differentiated between cretaceous flint, and neogene flint in the general inventory that we presented here.

The most abundant herbivora’s species in Galería are the horse, the deer, the fallow deer and one bovine undetermined (Table 3.1.4). Among the carnivores, less frequent than the herbivora, Canidae stands out. The Ursidae acquire a more important weight in the upper levels.

At present, we can find neogene flint in the Sierra de Atapuerca, in the form of big blocks, that can surpass one meter of lenght. Cretaceous flint also appears in the proximities of the Trinchera del Ferrocarril in little nodules, that habitually have ten or fifteen centimeters of maximum length.

Lithic Industry The campaigns of excavation executed between 1982 and 1995 in Galería (TG) have provided 1428 lithic objects, framed in various levels (Table 3.1.5). Lithic industry of Cueva de los Zarpazos is not included in this recount, since his systematic study is under way. 102

ATAPUERCA SITES nB Flint Quartzite

1GNB

PB

2GNB

1GNBC

1GNBE

0,0%

1

0,2%

13 2,4%

128 47,6%

12

4,5%

4 1,5%

61 22,7%

45

2

31

31,3%

20

0

Sandstone

22

Limestone

17 47,2%

22,2%

14 14,1%

284 51,6% 154

FRAG

2GNBE

Total determinables

28,0% 20

3,6%

78

14,2% 550

16,7%

2

0,7%

17

6,3% 269

20,2%

0

0,0%

10

10,1%

INDET

TOTAL

57,7% 404

42,3%

954 66,8%

94,7%

15

5,3%

284 19,9%

99

82,5%

21

17,5%

120

8,4%

11 23,4%

2,8%

1 2,8%

13 36,1%

3

8,3%

0

0,0%

1

2,8%

36

76,6%

47

3,3%

Quartz

1

5,9%

2 11,8%

0 0,0%

5 29,4%

4

23,5%

0

0,0%

5

29,4%

17

94,4%

1

5,6%

18

1,3%

Other

0

0,0%

0

0,0%

0 0,0%

0

1

100%

0

0,0%

0

0,0%

1

20,0%

4 80,0%

5

0,4%

168 17,3%

30

3,1%

20 2,1%

Total

1

2,0%

2GNBC

0,0%

394 40,5% 227 23,4% 22 2,3% 111 11,4% 972

68,1% 456 31,9% 1428

Table 3.1.5.- Lithic industry found in Trinchera Galería (TG)

The principal source for the supply of the rest of materials (fundamentally quartzite and sandstone) were the terraces of the Arlanzón river, close to the Trinchera del Ferrocarril. The fluvial materials acquire important percentages of the lithic record (greater than the general average), in the archeological levels of the Unit GIII and GIIa. This circumstance is related to the abundance of natural Bases in these levels. General characteristics of the lithic record The high percentage of natural Bases (fundamentally of quartzite and of sandstone) is one of the most prominent features. The majority of the nB were very likely related to the processing of faunal remains. The 1GNB have scarce percentage relating to the total of the determinable industry. The instruments configured on quartzite and sandstone pebbles (1GNBC) are an important part of these 1GNB, while the NB1GE (cores) are not enough (Table 3.1.5). Only cretaceous flint has significant percentages of 1GNBE (eleven of the 20 objects of this sub-category are of this type of flint). The cores of neogene flint are 2GNBE, that is ancient PB that were not shaped for their utilization as instruments, but exploded as cores. The fact that this raw material appear in the form of big blocks favored the obtaining of PB of large size, that later were exploited as cores. For this reason we have only 2 cores of neogene flint that are 1GNBE, and 19 that are 2GNBE. The total number of cores (1GNBE + 2GNBE) is scarce (n=42), 4.3 % of the determinable industry.

Figure 3.1.10.- Lithic industry from Galería (Atapuerca). 1: 1GNB of Exploitation (core) of cretaceous flint, with bifacial centripetal removals, and with hierarchization of the faces. 2: Positive Base of quartzite. 3: Positive Base of quartzite, with cortex remains in his dorsal face, and multifaceted butt.

7) preconfiguration of large instruments, on neogene flint, quartzite and sandstone. In Galería, Positive Bases are the structural category which contains most objects (Figures 3.1.10 and 3.1.12). However, the PB of the initial sequences of the cores´ reduction practically do not appear. This circumstance, joint to the scarce number of cores, indicate that the production of blanks was developed basically outside the site. These blanks would be PB with dihedral cutting edges, straight and convex.

The reduction strategies (ITOT) identified are seven (Carbonell et al., 1999c): 1) bifacial centripetal. This is the most frequent Indirect Operational Theme, and it was put in practice on flint, quartzite and sandstone (Figure 3.1.10.1) 2) longitudinal or polarized, with flint, quartzite and sandstone. 3) longitudinal unipolar massive recurrent, put in practice on pebbles of quartzite, sandstone and quartz 4) bipolar orthogonal, with exploitation of cretaceous flint. 5) bipolar opposed, accomplished on neogene flint and sandstone 6) multipolar, on neogene and cretaceous flint.

The configuration of instruments was accomplished following two different strategies. On the one hand the objects of medium and large size were shaped directly on pebbles of sandstone and quartzite (1GNBC) (Figure 3.1.11.2). These artifacts have dihedral cutting edges, associates at times to trihedral and pyramids in their extremes. In addition, 2GNBC of small size were shaped 103

Xosé Pedro Rodríguez

Figure 3.1.12.- Lithic industry from Galería (Atapuerca).1, 2, 3 and 6: Positive Bases of quartzite. 4 and 5: Positive Bases of flint Figure 3.1.11.- Lithic industry from Galería (Atapuerca). 1: 2GNB of Configuration of quartzite, with lateral-transverse dihedral cutting edges (cleaver). 2: 1GNBC of quartzite, with bifacial retouches that configure a bilateral dihedral edge, and a distal trihedral edge (handaxe)

(Figure 3.1.13). These flake tools have dihedral cutting edges, and also denticulated cutting edges in similar proportions. Among the instruments of large size the presence of handaxes and cleavers stands out (Figure 3.1.11), while among the flake tools, side scrapers and denticulate are the most frequent artifacts (Figure 3.1.13). Everything it takes us to consider that Galería’s lithic record belongs to the Mode 2 or Acheulian. Conclusions In Galería we observed a fragmentation of the Operational Sequences, with absence of the initial phases of production. This circumstance has led to propose that the knapping was accomplished basically out of the site. The majority of the PB would be produced in the exterior, and after taken to the site. Part of these PB would be retouched and transformed in 2GNBC. Also the configured instruments of medium and large size (1GNBC) would be brought over the exterior. This phenomenon is characteristic of sporadic occupations, that last a short time, accomplished with a concrete objective. The data derived from the study of the fauna support the hypothesis of occasional occupations, that would have like objective the profiting of animal biomass. The vertical conduit that connects with Galería would work like a natural trap, that would favor (perhaps with the

Figure 3.1.13.- Lithic industry from Galería (Atapuerca). 1: 2GNBC of quartzite, uniangular (convergent) denticulate. 2: 2GNBC of quartzite, denticulate. 3: 2GNBC of quartzite, with a distal trihedron (point). 4: 2GNBC of sandstone, with a distal trihedron, and lateral dihedrons.

104

ATAPUERCA SITES humans’ participation) animals’ fall. These animals’ remains would be after profited by the hominids (and by other carnivores). The hominids came to the cavity provided with Positive Bases, with dihedral cutting edges, instruments of large size with dihedral cutting edges and with trihedral and pyramidal extremes, in order to tear up the fallen animals. Also the hominids came with natural Bases for the fracturation of bones. The instruments of second generation (flake tools, 2GNBC) were shaped into the cavity in some cases, certainly to revive PB that no longer they were usable. These knapping processes, and the eventual exploitation of cores, took place at TN’s zone, close to the vertical conduit, since the natural illumination would make easy the activity of knapping. These occasional occupations were more intermittent and marginal as from the level TG11 (Unit GIIIb), as the Galería was filling up, and less space remained to circulate in the interior of the cave.

Gran Dolina Stratigraphy and paleoclimatology Gran Dolina (3º 31’ 08’’ W, 42º 21’ 09’’ N; UTM coordinates: X=457279, Y=4689172) is located in the northern sector of the Trinchera del Ferrocarril. Gran Dolina (TD) is filled by interior and exterior deposits which are up to 18 m thick. The outcrop of the stratigraphic section has a N–S orientation; the bottom is at 987–988m a.s.l. and the top at 1005m. Gran Dolina has 11 lithostratigraphic levels, enumerated correlatively from bottom to top (from TD1 to TD11) (Gil et al., 1987; Parés and PérezGonzález, 1998; Parés and Pérez-González, 1999; PérezGonzález et al., 2001) (Figures 3.1.14 and 3.1.15).

Figure 3.1.14.- Gran Dolina stratigraphic section, with the lithostratigraphic levels (©Atapuerca Research Team)

accretion of reddish yellow colour (7.5YR6/6) or strong brown (7.5YR5/6), transported on suspension or with a low energy hydric regime. From the archaeological and paleontological point of view this level is sterile.

We can distinguish three major phases in the evolution of Gran Dolina’s karstic sequence (Hoyos and Aguirre, 1995): 1. Endokarstic phase; the cavity is not communicated with the exterior through the vertical conduit during this phase. The sediments of TD1 and TD2 correspond to this phase.

The stratigraphic unit TD2 is composed of angular boulders to pebbles of limestone and speleothem breakdowns from the ceiling and walls with large hollows between them. Yellow (10YR7/6) or reddish yellow (5YR6/6) clay fills these hollows between the clasts. The thickness of this deposit is very irregular, as it overlies TD1, the top of which is deformed by loading. The average thickness is about 1 m. In addition, there is a speleothem that increases its thickness by 50–60 cm northwards and overlies the authigenic facies of the bottom of Gran Dolina. Pollinic analyses indicate the existence of a cold and wet phase, in the one that the cupressaceae, pines, birches, Artemisia and Quercus dominate. This level is sterile from the archaeological and paleontological point of view.

2. Opening of the cavity to the exterior; the deposits receive the direct influence of the external conditions. The sediments of TD3 to TD9 correspond to this phase. 3. Exokarstic phase; the cavity’s roof disappears. The sediments of TD10 and TD11 correspond to this phase. The cavity becomes a depression at the slope of the hill, with prolongation toward the interior. The description of the lithostratigraphic levels of Gran Dolina begins with TD1 (Parés and Pérez-González, 1999) (Figures 3.1.14 and 3.1.15).

The TD3–4 unit is not less than 2 m thick and is made up of a subfacies of reddish yellow (7.5YR 6/6) or strong brown (7.5YR 5/6) sandy lutite that contains subangular and heterometric limestone clasts up to 15 cm long. Another subfacies is composed of an alternation of clastsupported (0.5– 3 cm) beds with clasts of up to 30 cm. The TD3–4 unit lies on top of the speleothem of unit TD2, previously eroded under phreatic conditions. Thus,

TD1 and TD2 are the lowermost stratigraphic levels, with interior facies typical of a closed cave. The lowermost of these units, TD1, is made up of 1,50 m of clay and mud, with very little sand (20mm) 1159 78 35 606 476 48 71 24 98 137 43 22 177 21

46,5% 51,7% 50,7% 56,7% 59,9% 44% 34,1% 24,7% 49,2% 39,9% 37,4% 31,9% 57,7% 8,3%

PB (350 kyr) and perhaps to 400-500 kyr: new radiometric dates. Journal of Archaeological Science 30: 275-280.

Cáceres, I. (2002). Tafonomía de los yacimientos antrópicos en karst. Complejo Galería (Sierra de Atapuerca, Burgos), Vanguard Cave (Gibraltar) y Abric Romaní (Capellades, Barcelona). Doctoral Thesis, Universitat Rovira i Virgili. Canal, J. & Carbonell, E. (1976). El Paleolític a les comarques gironines. Girona: Caixa d’Estalvis Provincial

Bisson, M. S. (2000). Nineteenth Century Tools for Twenty-First Century Archaeology? Why the Middle Paleolithic Typology of François Bordes Must Be Replaced. Journal of Archaeological Method and Theory 7: 1-48.

Canal, J. (1977). Identificación del Paleolítico inferior en las comarcas de Gerona. Eds) XIV Congreso Nacional de Arqueología, pp. 81-96. Vitoria.

Blackwell, B., García, J. F. & Straus, L. G. (1992). Uranium-series Isochron Dating at El Castillo Cave (Cantabria, Spain): The “Acheulean/ Mousterian” Question. Journal of Archaeological Science 19: 4962.

Canal, J. & Carbonell, E. (1978). Nova aportació per l’estudi del Paleolític Inferior i Mig al N.E. de Catalunya. Revista de Girona 84: 265278.

Boëda, E. (1986). Approche technologique et traces du concept Levallois et évaluation de son champ d’application: étude de trois gisements

Canal, J. & Carbonell, E. (1979). Les estacions prehistòriques del Puig d’en Roca. Una visió dels pobladors més antics de les nostres terres.

176

REFERENCES Girona: Associació Arqueològica de Girona

ment del Catllar/LAUT

Canal, J. & Carbonell, E. (1980). Los bifaces abbevillienses del Cau del Duc de Torroella de Montgrí. Butlletí de l’Associació Arqueològica de Girona III: 4-14.

Carbonell, E., Mosquera, M., Ollé, A., Rodríguez, X. P., Sala, R., Vaquero, M. & Vergès, J. M. (1992b). New elements of the logical analytic system. First International Meeting on Technical Systems to Configure Lithic Objects of scarce elaboration (Montblanc, 1992). Tarragona: Laboratori d’Arqueologia de la Universitat Rovira i Virgili/ Reial Societat Arqueològica Tarraconense (Cahier Noir, 6)

Canal, J. & Maroto, J. (1982). El Paleolític Inferior. In L’Arqueologia avui, pp. 21. Barcelona: Depart. de Cultura de la Generalitat. Canal, J. (1983). Una excursió al Cau del Duc de Torroella de Montgrí (Baix Empordà). Revista de Girona 103: 99-106.

Carbonell, E., Rodríguez, X. P., Costafreda, A., Llussà, A. & Serra, R. (1993). El jaciment plistocè mig del Clot del Ballester (Artesa de Lleida). Artsea de Lleida: Agrupació Cultural la Femosa (Quaderns d’Arqueologia del Grup de Recerques de “La Femosa”, 8).

Canal, J. & Carbonell, E. (1989). Catalunya Paleolítica. Girona: Patronat Eiximenis de la Diputació de Girona

Carbonell, E. & Rodríguez, X. P. (1994). Early Middle Pleistocene deposits and artefacts in the Gran Dolina site (TD4) of the ‘Sierra de Atapuerca’ (Burgos, Spain). Journal of Human Evolution 26: 291311.

Canals, A., Vallverdu, J. & Carbonell, E. (2003). New Archaeo-Stratigraphic Data for the TD6 Level in Relation to Homo antecessor (Lower Pleistocene) at the Site of Atapuerca, North-Central Spain. Geoarchaeology 18: 481-504.

Carbonell, E., Bermúdez de Castro, J. M., Arsuaga, J. L., Díez, J. C., Rosas, A., Cuenca-Bescós, G., Sala, R., Mosquera, M. & Rodríguez, X. P. (1995a). Lower Pleistocene Hominids and Artifacts from Atapuerca-TD6 (Spain). Science 269: 826-830.

Carballo, J. (1910). De espeleología. Boletín de la Real Sociedad Española de Historia Natural X. Carbonell, E. Achelense: término, concepto o falacia. Raña: VI-VIII.

Carbonell, E., Márquez, B., Mosquera, M., Ollé, A., Rodríguez, X. P., Sala, R., Vaquero, M. & Vergès, J. M. (1995b). Atapuerca Trinchera Galería (Spain): Strategies and operational models of lithic industry. Cahier Noir 7: 41-83.

Carbonell, E., Guilbaud, M. & Mora, R. (1983). Utilización de la lógica analítica para el estudio de tecno-complejos a cantos tallados. Cahier Noir 1: 3-64. Carbonell, E. & Mora, R. (1984). Diacronía y homogeneidad funcional entre dos yacimientos del paleolítico inferior del NE catalán: Pedra Dreta y Puig d’en Roca III. Eds) Arqueología Espacial 2. Coloquio sobre distribución y relaciones entre los asentamientos, pp. 147-158. Teruel: Seminario de Arqueología y Etnología Turolense, Colegio Universitario de Teruel.

Carbonell, E., Mosquera, M., Rodríguez, X. P. & Sala, R. (1996). The First Human Settlement of Europe. Journal of Anthropological Research 52: 107-114. Carbonell, E., Esteban, M., Nájera, A. M., Mosquera, M., Rodriguez, X. P., Ollé, A., Sala, R., Vergès, J. M., Bermúdez de Castro, J. M. & Ortega, A. I. (1999a). The Pleistocene site of Gran Dolina, Sierra de Atapuerca, Spain: a history of the archaeological investigations. Journal of Human Evolution 37: 313-324.

Carbonell, E. (1985). Méthode d’analyse appliquée aux industries lithiques des gisements du Pleistocène Moyen, du Massif de Montgrí (Catalogne, Espagne). Thèse de Doctorat de 3ème. cycle, Université de Paris VI (Pierre et Marie Curie) (Museum National d’Histoire Naturelle).

Carbonell, E., García-Antón, M. D., Mallol, C., Mosquera, M., Ollé, A., Rodríguez, X. P., Sahnouni, M., Sala, R. & Verges, J. M. (1999b). The TD6 level lithic industry from Gran Dolina, Atapuerca (Burgos, Spain): production and use. Journal of Human Evolution 37: 653693.

Carbonell, E. & Mora, R. (1985). “Cadena operativa “Achelense” en Catalunya,” in Actas de la Iª Reunión del Cuaternario Ibérico, vol. II, pp. 27-40. Lisboa.

Carbonell, E., Márquez, B., Mosquera, M., Ollé, A., Rodríguez, X. P., Sala, R. & Vergès, J. M. (1999c). El Modo 2 en Galería. Análisis de la industria lítica y sus procesos técnicos. In (E. Carbonell, A. Rosas & J. C. Díez, Eds) Atapuerca: Ocupaciones humanas y paleoecología del yacimiento de Galería, pp. 299-352. Zamora: Junta de Castilla y León (Consejería de Educación y Cultura).

Carbonell, E. & Mora, R. (1986). Anatomia morfotècnica del paleolític inferior a Catalunya. Fonaments 6: 35-100. Carbonell, E. (1987). Human Development in the framework of the Lithic Operational Chains. In (E. Carbonell, M. Guilbaud & R. Mora, Eds) Sistemes d’anàlisi en Prehistòria, pp. 68-82. Girona: Centre de Recerques Paleo-ecosocials (CRPES).

Carbonell, E., Mosquera, M., Ollé, A., Rodriguez, X. P., Sala, R., Vergès, J. M., Arsuaga, J. L. & Bermúdez de Castro, J. M. (2003). Les premiers comportements funeraires auraint-ils pris place à Atapuerca, il y a 350,000 ans? L’Anthropologie 107: 1-14.

Carbonell, E., Díez, C., Enamorado, J. & Ortega, A. I. (1987a). Análisis morfo técnico de la industria lítica de Torralba (Soria). Cuadernos de la Sección Antropología-Entografía de la Sociedad de Estudios Vascos 4: 201-216.

Carracedo, J. M., Heller, F., Soler, V. & Aguirre, E. (1987). Estratigrafía magnética del yacimiento de Atapuerca: determinación del límite Matuyama/Brunhes. In (E. Aguirre, E. Carbonell & J. M. Bermúdez de Castro, Eds) El hombre fósil de Ibeas y el Pleistoceno de la Sierra de Atapuerca, pp. 193-199. Valladolid: Junta de Castilla y León.

Carbonell, E., Mora, R. & Fullola, J. M. (1987b). Radiografia dels tecnocomplexos del Plistocè Superior de la Vall de la Femosa (Segrià). Cypsela VI: 201-210. Carbonell, E., Collina Girard, J., Guilbaud, M., Mora, R. & Sala, R. (1988a). Le gisement Pléistocene moyen de Puig d’en Roca (Espagne). B.S.P.F. 85: 204-209.

Carretero, J. M. (2001). El Portalón de Cueva Mayor. National Geographic España 9. Cerdeño, E. & Sánchez, B. (1988). Le Rinocéros du Pléistocène moyen d’Atapuerca (Burgos, Espagne). Geobios 21:81-99.

Carbonell, E., Guilbaud, M., Mora, R., Muro, I., Sala, R. & Miralles, J. (1988b). El compex del Plistocè mitjà del Puig d’en Roca. Girona: C.S.I.C.

Cervera, J., García, N. & Arsuaga, J. L. (1999). Los carnívoros del yacimiento mesopleistocénico de Galería (Sierra de Atapuerca, Burgos). In (E. Carbonell, A. Rosas, & J. C. Díez) Atapuerca: Ocupaciones humanas y paleoecología del yacimiento de Galería, pp. 175188. Zamora: Junta de Castilla y León (Consejería de Educación y

Carbonell, E., Márquez, B., Ollé, A., Rodríguez, X. P., Vallverdú, J., Vergès, J. M. & Zaragoza, J. (1992a). Els Vinyets. El Catllar. Els primers pobladors de la Catalunya meridional. Tarragona: Ajunta-

177

Xosé Pedro Rodríguez Esteban, M. (1996). Zooarqueologia del nivell 10 de la Trinchera Dolina, Sierra de Atapuerca, Burgos. Tesis de Licenciatura, Tarragona: Universitat Rovira i Virgili.

Cultura). Chang, K. C. (1967). Rethinking Archaeology: Random House

Falguères, C. (1986). Datations de sites acheuléens et mousteriéns du Midi Méditerranéen par la méthode de Résonance de Spin Electronique. Doctoral Thesis, Université de Paris VI.

Chavaillon, J. (1991). Les ensembles lithiques de Chilhac III (Haute Loire): typologie, situation stratigraphique et analyse critique et comparative. In (E. Bonifay & B. Vandermeersch, Eds) Les Premiers Européens, pp. 81-91. Paris: Editions du C.T.H.S.

Falguères, C., Bahain, J.-J., Yokoyama, Y., Arsuaga, J. L., Bermúdez de Castro, J. M., Carbonell, E., Bischoff, J. L. & Dolo, J.-M. (1999). Earliest humans in Europe: the age of TD6 Gran Dolina, Atapuerca, Spain. Journal of Human Evolution 37: 343-352.

Ciudad Serrano, A. (1986). Las industrias de cantos tallados en Ciudad Real. Aportación al Achelense inferior de la Submeseta meridional. Ciudad Real: Museo de Ciudad Real (Estudios y Monografías del Museo de Ciudad Real, 16).

Falguères, C., Bahain, J. J., Yokoyama, Y., Bischoff, J. L., Arsuaga, J. L., Bermúdez de Castro, J. M., Carbonell, E. & Dolo, J. M. (2001). Datation par RPE et U-Th des sites pléistocènes d’Atapuerca: Sima de los Huesos, Trinchera Dolina et Trinchera Galería. Bilan géochronologique. L’Anthropologie 105: 71-81.

Clark, G. A. Editor. (1979). The North Burgos Archeological Survey. Bronze and Iron Age Archaeology on the Meseta del Norte (Province of Burgos, North Central Spain). Tempe (Arizona): Arizona State University, Dept. of Anthropology. Anthropological Research Papers, Vol. 19).

Fernández-Jalvo, Y. & Andrews, P. (1992). Small mammal taphonomy of Gran Dolina, Atapuerca (Burgos), Spain. Journal of Archaeological Science 19: 407-428.

Clarke, D. L. (1968). Analytical Archaeology. London: Methuen Collina-Girard, J. (1975). Les industries archaïques sur galets des terrasses quaternaires de la plaine du Roussillon (Pyrénées-Orientales, France). Thèse de Doctorat, Université de Provence (Dept.

Fernández-Jalvo, Y., Diez, J. C., Bermúdez de Castro, J. M., Carbonell, E. & Arsuaga, J. L. (1996). Evidence of early cannibalism. Science 271: 277-278.

Crabtree, D. E. (1975). Comments on Lithic Technology and Experimental Archaeology. In (E. H. Swanson, Eds) Lithic Technology: Making and Using Stone Tools. Pocatello: Idaho State University Museum.

Fernández-Jalvo, Y., Diez, J. C., Cáceres, I. & Rosell, J. (1999). Human cannibalism in the Early Pleistocene of Europe (Gran Dolina, Sierra de Atapuerca, Burgos, Spain). Journal of Human Evolution 37: 591622.

Cuenca-Bescós, G., Laplana, C. & Canudo, J. I. (1994). “Precisiones sobre la edad de la Sima de los Huesos (Pleistoceno medio, Atapuerca),” in Comunicaciones de las X Jornadas de Paleontología, pp. 53-56. Madrid: Depto. de Paleontología-U.E.I.-S.E.P.

Fernández-Peris, J. (1993). El Paleolítico inferior en el País Valenciano. Una aproximación a su estudio. Recerques del Museu d’Alcoi 2: 721.

Cuenca-Bescós, G., Canudo, J. I. & Laplana, C. (1995). Los arvicólidos (Rodentia, Mammalia) de los niveles inferiores de Gran Dolina (Pleistoceno inferior, Atapuerca, Burgos, España). Revista Española de Paleontología 10: 202-218.

Fernández-Peris, J., Guillem, P., Fumanal, M. P. & Martínez, R. (1994). Cova de Bolomor (Tavernes de Valldigna, Valencia), primeros datos de una secuencia del Pleistoceno medio. Saguntum 27: 9-37. Fernández-Peris, J., Guillem, P. & Martínez, R. (2000). Cova de Bolomor (Tavernes de la Valldigna- Valencia). Datos cronoestratigráficos y culturales de una secuencia del Pleistoceno medio. In (R. Balbín, N. Bicho, E. Carbonell, B. Hackett, A. Moure, L. Raposo, M. Santonja & G. Vega Toscano, Eds) Actas do 3º Congresso de Arqueologia Peninsular (1999), vol. II: Paleolítico da Península Ibérica, pp. 81-100. Porto: ADECAP.

Cuenca-Bescós, G., Laplana, C., Canudo, J. I. & Arsuaga, J. L. (1997). Small mammal from Sima de los Huesos. Journal of Human Evolution 33: 175-190. Cuenca-Bescós, G., Laplana, C. & Canudo, J. I. (1999). Biochronological implications of the Arvicolidae (Rodentia, Mammalia) from the Lower Pleistocene hominid-bearing level of Trinchera Dolina 6 (TD6, Atapuerca, Spain). Journal of Human Evolution 37: 353-373.

Fernández-Peris, J. (2003). Cova del Bolomor (La Valldigna, Valencia). Un registro paleoclimático y arqueológico en un medio kárstico. SEDECK (Boletín de la Sociedad Española de Espeleología y Ciencias del Karst 4: 34-47.

Cuenca-Bescós, G., Canudo, J. I. & Laplana, C. (2001). La séquence des rongeurs (Mammalia) des sites du Pléistocène inférieur et moyen d’Atapuerca (Burgos, Espagne). L’Anthropologie 105: 115-130.

Flannery, K. V. (1967). Culture History v. Culture Process: A Debate in American Archaeology. Scientific American 217: 119-122.

Davidson, I. & Noble, W. (1993). Tools and language in human evolution. In (K. R. Gibson & T. Ingold, Eds) Tools, Language and Cognition in Human Evolution, pp. 363-388. Cambridge: Cambridge University Press.

Gabunia, L. & Vekua, A. (1995). A Plio-Pleistocene hominid from Dmanisi, East Georgia, Caucasus. Nature 373: 509-12.

Delson, E. (1989). Oldest Eurasian stone tools. Nature 340: 96. Gabunia, L., Vekua, A., Lordkipanidze, D., Swisher III, C. c., Ferring, R., Justus, A., Nioradze, M., Tvalchrelidze, M., Antón, S. C., Bosinski, G., Jöris, O., Lumley, M.-A. d., Majsuradze, G. & Mouskhelishcili, A. (2000). Earliest Pleistocene Hominid Cranial Remains from Dmanisi, Republic of Georgia: Taxonomy, Geological Setting, and Age. Science 288: 1019-1025.

d’Errico, F. & Nowell, A. (2000). A New Look at the Berekhat Ram Figurine: Implications for the Origins of Symbolism. Cambridge Archaeological Journal 10: 123-167. Dibble, H. L. & Rolland, N. (1992). On Assemblage Variability in the Middle Paleolithic of Western Europe. History, Perspectives, and a New Synthesis. In (H. L. Dibble & P. Mellars, Eds) The Middle Paleolithic: adaptation, behavior and variability, pp. 1-28. Philadelphia: University of Pennsylvania.

Gagnepain, J., Laurent, M., Bahain, J. J., Falguères, C., Hedley, I., Peretto, C., Wagner, J.-J. & Yokoyama, Y. (1998). Synthèse des donées paléomagnetiques et radiochronologiques du site de Ca’Belvedere di Monte Poggiolo (Romagna, Italie) et son environnement géologique. In (A. Antoniazzi et al. Eds.) XIII U.I.S.P.P. Congress ProceedingsForli 8-14 september 1996. Workshop, 6-II, pp. 877-888. Forli: A.B.A.C.O.

Díez, J. C. (1992). Zooarqueología de Atapuerca (Burgos) e implicaciones paleo-económicas del estudio tafonómico de yacimientos del Pleistoceno medio. Ph. D. Dissertation, Universidad Complutense (Departameto de Prehistoria).

178

REFERENCES Gamble, C. (1993). Timewalkers: The Prehistory of Global Colonization. Stroud: Alan Sutton

tion 14: 29-46. Gibson, K. (1996). “Technology, Language, and Cognitive Capacity,” in XIII International Congress of Prehistoric and Protohistoric Sciences. The First Humans and their Cultural Manifestations, vol. 4, Colloquia. Edited by F. Facchini, pp. 117-123. Forlí (Italia): A.B.A.C.O. Edizioni.

Gamble, C. (1994). Time for Boxgrove man. Nature 369: 275-276. Gamble, C. (1995). The earliest occupation of Europe: the environmental background. In (W. Roebroeks & T. Van Kolfschoten, Eds) The Earliest Occupation of Europe, pp. 279-295. Leiden: University of Leiden.

Gil, E., Aguirre, E. & Hoyos, M. (1987). Contexto estratigráfico. In (E. Aguirre, E. Carbonell & J. M. Bermúdez de Castro, Eds) El hombre fósil de Ibeas y el Pleistoceno de la Sierra de Atapuerca, pp. 47-54. Valladolid: Junta de Castilla y León.

García Antón, M. (1989). Estudio palinológico de los yacimientos mesopleistocenos de Atapuerca (Burgos): Reconstrucción paisajística y paleoclimática. Tesis Doctoral, Universidad Autónoma de Madrid (Dept. Departamento de Biología).

Giles, F. & Santiago, A. (1987). El poblamiento del sur de la Península Ibérica en el Pleistoceno inferior a través de Gibraltar. Eds) Actas del Congreso Internacional «El estrecho de Gibraltar», Ceuta 1987. Tomo I, Prehistoria e Historia de la Antigüedad, pp. 97-109. Ceuta.

García Antón, M. (1995). Pollen analysis of Middle Pleistocene palaeovegetation at Atapuerca. In (J. M. Bermúdez de Castro, J. L. Arsuaga & E. Carbonell, Eds) Human Evolution in Europe and the Atapuerca Evidence (Workshop, Castillo de la Mota, Medina del Campo, Valladolid, 1992), pp. 147-165. Valladolid: Junta de Castilla y León.

Giles, F., Gutiérrez, J. M., Mata, E. & Santiago, A. (1996). Laguna de Medina, Bassin du fleuve Guadalete (Cadiz, Espagne): Un gisement acheuléen ancien dans le cadre des premières occupations humaines de la Péninsule Ibérique. L’Anthropologie 100: 507-528.

García Antón, M. (1998). Reconstrucciones de paleovegetación en Atapuerca según análisis polínico. In (E. Aguirre, Eds) Atapuerca y la Evolución humana, pp. 61-71. Madrid: Fundación Ramón Areces.

Giles, F., Gutiérrez, J.M., Santiago, A., Mata, E. & Aguilera, L. (1992). Secuencia paleolítica del valle del Guadalete. Primeros resultados. Revista de Arqueología 135: 16-26.

García, N., Arsuaga, J. L. & Torres, T. (1997). The carnivore remains from the Sima de los Huesos Middle Pleistocene site (Sierra de Atapuerca, Spain). Journal of Human Evolution 33: 155-174.

Giralt, S., Vallverdú, J., Sala, R. & Rodríguez, X. P. (1995). Cronoestratigrafia i paleoclimatologia de l’ocupació humana a la vall mitjana del Ter al Plistocè mitjà i superior inicial. In (B. Agustí, J. Burch & J. Merino, Eds) Excavacions d’urgència a Sant Julià de Ramis (Anys 1991-1993), pp. 23-36. Girona: Centre d’Investigacions arqueològiques de Girona.

García, N. & Arsuaga, J. L. (2001). Les carnivores (Mammalia) des sites du Pléistocène ancien et moyen d’Atapuerca (Espagne). L’Anthropologie 105: 83-93. García, N. (2002). Los carnívoros del Pleistoceno de la Sierra de Atapuerca. Doctoral Thesis, Universidad Complutense (Depto. de Paleontología).

Gladiline, V. & Sitlivy, V. (1991). Les premières industries en Subcarpatie. In (E. Bonifay & B. Vandermeersch, Eds) Les premiers européens. Actes du 114e. Congrès National des Sociétés Savantes (Paris, 3-9 avril 1989), pp. 217-231. Paris: Editions du CTHS.

García-Antón, M. D., Morant Sabater, N. & Mallol, C. (2002). L’approvisionnement en matières premières lithiques au Pléistocène inférieur et moyen dans la Sierra de Atapuerca, Burgos (Espagne). L’Anthropologie 106: 41-55.

Goguitchaichvili, A. & Parés, J. M. (2000). A recognition palaeomagnetic study of volcanic and sedimentary rocks from Dmanissi (Caucasus): implications for the oldest human occupation in Europe. Comptes Rendus de L’Academie des Sciences Serie II, Fascicule A Sciences de la Terre et des Planetes 331: 183-186.

García-Medrano, P. (2003). Estudio tecnológico de Covacha de los Zarpazos (Complejo Galería, Atapuerca, Burgos) en el contexto del Pleistoceno medio de la submeseta norte de la Península Ibérica. Tesis de Llicenciatura, Universitat Rovira i Virgili (Dept. d’Història i Geografia).

González García, M. I., López Cerezo, J. A., Luján López, J. L. & De Melo Martín, M. I. (1994). Las concepciones de la tecnología. Arbor: 125-145.

Geneste, J.-M. (1985). Analyse lithique d’industries moustériennes du Périgord: une approche technologique du comportement des groupes humains au Paléolithique moyen. Doctoral Thesis, Université de Bordeaux.

Grün, R. & Aguirre, E. (1987). Datación por ESR y por la serie del U, en los depósitos cársticos de Atapuerca. In (E. Aguirre, E. Carbonell & J. M. Bermúdez de Castro, Eds) El hombre fósil de Ibeas y el Pleistoceno de la Sierra de Atapuerca, pp. 201-204. Valladolid: Junta de Castilla y León.

Geneste, J.-M. (1991). Systèmes techniques de production lithique: variatons techno-écnomiques dans les processus de réalisation des outillages paléolithiques. Techniques et culture 17-18: 1-35.

Grup de Recerques Arqueològiques de La Femosa. (1976). El Paleolític de la Vall de la Femosa. Artesa de Lleida: Grup de Recerques Arqueològiques de l’Agrupació Cultural de La Femosa

Gibert, J., Agustí, J. & Moyà-Solà, S. (1983). Presencia de Homo sp. en el yacimiento de Pleistoceno inferior de Venta Micena (Orce, Granada). Paleontologia i Evolució publicación especial: 1-12. Gibert, J., Campillo, D., Arqués, J. M., García-Olivares, E., Borja, C. & Lowenstein, J. (1998a). Hominid status of the Orce cranial fragment reasserted. Journal of Human Evolution 34: 203-217.

Guilbaud, M. (1985). Elaboration d’une méthode d’analyse pour les produits de débitage en typologie analytique, et son application à quelques industries des gisements de Saint-Césaire (Charente-Maritime) et de Quinçay (Vienne). Doctoral Thesis, Université de Paris VI.

Gibert, J., Gibert, L., Iglesias, A. & Maestro, E. (1998b). Two “Oldowan” assemblages in the Plio-Pleistocene deposits of the Orce region, southeast Spain. Antiquity 72: 17-25.

Guilbaud, M. (1987). Le debitage comme expression d’une realité psychique. In (E. Carbonell, M. Guilbaud & R. Mora, Eds) Sistemes d’anàlisi en prehistòria, pp. 149-179. Girona: C.R.P.E.S.

Gibert, J., Campillo, D., Eisenmann, V., García-Olivares, E., Malgosa, A., Roe, D., Walker, M. J., Borja, C., Ribot, F., Gibert, L., Albadalejo, S., Iglesias, A., Ferrandez, C. & Maestro, E. (1999). Spanish late Pliocene and early Pleistocene hominid, palaeolithic and faunal finds from Orce (Granada) and Cueva Victoria (Murcia). Human Evolu-

Guilbaud, M. (1996). Psycotechnic analysis and culture change: origins of the upper Paleolithic as seen through the example of Saint-Césaire. In (E. Carbonell & M. Vaquero, Eds) The Last Neandertals, The First Anatomically Modern Humans: A Tale about the Human Diversity, pp. 337-354. Tarragona: Universitat Rovira i Virgili.

179

Xosé Pedro Rodríguez Leroi-Gourhan, A. (1952). Sur la position scientifique de l’ethnologie. Revue philosophique oct-des. 1952: 506-518.

Hoyos, M. & Aguirre, E. (1995). El registro paleoclimático pleistoceno en la evolución del karst de Atapuerca (Burgos): el corte de Gran Dolina. Trabajos de Prehistoria 52: 31-45.

Leroi-Gourhan, A. (1964). Le geste et la parole I- technique et langage. Paris: Albin Michel

Huguet, R. (1997). Estudi zooarqueològic de la Unitat GII del complex Galería (Sierra de Atapuerca, Burgos). Tesis de Licenciatura, Tarragona: Unversitat Rovira i Virgili.

Lewis, C., Waldridge, S. & Asmeron, Y. (1998). Neogene astenosphere derived volcanism in northeast Spain: constraints on the geodinamic evolution of the western Mediterranean sea.

Huguet, R., Díez, C., Rosell, J., Cáceres, I., Moreno, V., Ibáñez, N. & Saladié, P. (2001). Le gisement de Galería (Sierra de Atapuerca, Burgos, Espagne): un modèle archéozoologique de gestion du territoire durant le Pléistocène. L’Anthropologie 105:237-257.

López-Antonanzas, R. & Cuenca-Bescós, G. (2002). The Gran Dolina site (Lower to Middle Pleistocene, Atapuerca, Burgos, Spain): new palaeoenvironmental data based on the distribution of small mammals. Palaeogeography, Palaeoclimatology, Palaeoecology 186: 311-334.

Inizan, M.-L., Roche, H. & Tixier, J. (1992). Technology of Knapped Stone. Meudon: Centre de Recherches et d’Etudes Préhistoriques, CNRS

Lumley, H. d. (1971). Le Paleolithique Inferieur et Moyen du Midi Mediterranéen dans son cadre géologique. Paris: CNRS

Isaac, G. L. (1978). The food-sharing behaviour of proto-human hominids. Scientific American 238: 90-108.

Lumley, H. d., Fournier, A., Krzepkowska, J. & Echasoux, A. (1988). L’industrie du Pleistocene inférieur de la grotte du Vallonet, Roquebrune-Cap-Martin, Alpes Maritimes. L’Anthropologie 92: 501-614.

Isaac, G. L., Harris, J. W. K. & Marshall, F. (1981). “Small is informative: the application of the study of mini-sites and least-effort criteria in the interpretation of the early Pleistocene archaeological record at Koobi Fora, Kenya,” in Las Industrias más Antiguas. X Congreso de la Unión Internacional de Ciencias Prehistóricas y Protohistóricas. Edited by J. D. Clark & G. L. Isaac, pp. 101-119. México.

Made, Jan van der (1999). Ungulates from Atapuerca TD6. Journal of Human Evolution 37:389-413. Made, Jan van der (2001). Les Ongulés d’Atapuerca. Stratigraphie et biogéographie. L’Anthropologie 105:95-113.

Isaac, G. L. (1984). The archaeology of human origins: studies of the Lower Pleistocene in East Africa 1971-1981. In (F. Wendorf & A. Close, Eds) Advances in Old World Archaeology, pp. 1-87. New York: Academic Press.

Mallol, C. (1999). The selection of lithic raw materials in the Lower and Middle Pleistocene levels TD6 and TD10A of Gran Dolina (Sierra de Atapuerca, Burgos, Spain). Journal of Anthropological Research 55: 385-407.

Jaubert, J. (1995). Schémas opératoires et outillages peu élaborés: le cas du Paléolithique inférieur et moyen de Coudoulous I (Lot). Cahier Noir 7: 85-100. Johnson, L. L. (1978). A History of Flint-Knapping Experimentation, 1838-1976. Current Anthropology 19: 337-372.

Mallol, C. (2004). Micromorphological oservations from the archaeological sediments of ‘Ubeidiya (Israel), Dmanisi (Georgia) and Gran Dolina-TD10 (Spain) for the reconstruction of hominid occupation contexts. Doctoral Thesis, Harvard University (Department of Anthropology).

Karlin, C. & Julien, M. (1994). Prehistoric technology: a cognitive science? In (C. Renfrew & E. B. W. Zubrow, Eds) The ancient mind. Elements of cognitive archaeology, pp. 152-164. Cambridge: Cambridge University Press.

Manzi, G., Mallegni, F. & Ascenzi, A. (2001). A cranium for the earliest Europeans: Phylogenetic position of the hominid from Ceprano, Italy. Proceedings of the National Academy of Sciences of The United States of America 98: 10011-10016.

Kozlowski, J. K. (1992). Les permiers habitants de l’Europe Centrale et Orientale. In (C. Peretto, Eds) Il piu antico popolamento della valle Padana nel quadro delle conoscenze europee. Monte Poggiolo, pp. 69-91. Milano: Jaca Book.

Márquez, B., Ollé, A., Sala, R. & Vergés, J. M. (2001). Perspectives méthodologiques de l’analyse fonctionnelle des ensembles lithiques du Pléistocène inférieur et moyen d’Atapuerca (Burgos, Espagne). L’Anthropologie 105: 281-299.

Kuhn, S. (1993). Mousterian Technology as Adaptative Response: A Case Study. In (G. Larsen Peterkin, H. M. Bricker & P. Mellars, Eds) Hunting and Animal Exploitation in the Later Palaeolihtic and Mesolithic of Eurasia, pp. 25-31. Washington: American Anthropological Association.

Marshack, A. (1997). The Berekhat Ram figurine: a late Acheulian carving from the Middle East. Antiquity 71: 327-337. Martín Merino, M. A., Domingo, S. & Antón, T. (1981). Estudio de las cavidades de la zona BU-IV-A (Sierra de Atapuerca). Kaite Estudios de Espeleología Burgalesa 2: 41-76.

Laplace, G. (1972). La typologie analytique et structurale: Base rationnelle d’étude des industries lithiques et osseuses. Banques des données archéologiques. Colloques nationaux du CNRS 932: 91-143.

Martín Nájera, A. (1986). Aplicación del Sistema Lógico Analítico al estudio del Complejo Industrial del Pleistoceno medio de la Trinchera de la Sierra de Atapuerca (Burgos). Memoria de Licenciatura, Universidad de Extremadura.

Laplace, G. (1974). De la dynamique de l’analyse structurale ou la typologie analytique. Rivista di scienze preistoriche XXIX: 3-71.

Martinet, G. & Searight, S. (1994). Le Maghreb préhistorique et la navigation. Bulletin de la Société d’Études et de Recherches Préhistoriques des Eyzies 43: 85-111.

Laplace, G. (1987). Un exemple de nouvelle ecriture de la grille typologique. Dialektikê. Cahiers de typologie analytique 1985-1987: 16-21.

Martínez Santaolalla, J. (1926). Prehistoria burgalesa. Neolítico y Eneolítico. B.A.C.A.E.B. IV.

Leakey, M. D. (1971). Olduvai Gorge, 3. Excavations in Beds I and II, 1960-1963. Cambridge: Cambridge University Press

Martínez-Navarro, B., Turq, A., Agustí Ballester, J. & Oms, O. (1997). Fuente Nueva 3 (Orce, Granada, Spain) and the first human occupation of Europe. Journal of Human Evolution 33: 611-620.

Lemonnier, P. (1976). La description des chaînes opératoires: contribution a l’analyse des systèmes techniques. Techniques et Culture 1: 100-151.

Mauss, M. (1947). Manuel d’ethnographie. Paris: Payot

180

REFERENCES gia), Áridos (Arganda, Madrid) i Galería-TN (Sierra de Atapuerca, Burgos). Doctoral Thesis, Universitat Rovira i Virgili (Dept. Història i Geografia).

Mir, A. (1979). La fauna de la Cueva Mollet I en Serinyà. Campañas de excavación 1947-1972. Eds) IV Reunión del Grupo de Trabajo del Cuaternario. Vigo.

Ollé, A., Vergès, J. M. & Rodríguez, X. P. (2004). El jaciment de La Cansaladeta (La Riba, Alt Camp): primers resultats. Quaderns de Vilaniu. Miscel.lània de l’Alt Camp 45: 127-144.

Mithen, S. (1994). From domain specific to generalized intelligence: a congnitive interpretation of the Middle/Upper Palaeolithic transition. In (C. Renfrew & E. B. W. Zubrow, Eds) The ancient mind. Elements of cognitive archaeology, pp. 29-39. Cambrdige: Cambridge University Press.

Oms, O., Agustí, J., Gabàs, M. & Anadón, P. (2000a). Lithostratigraphical correlation of micromammal sites and biostratigraphy of the Upper Pliocene to Lower Pleistocene in the Northeast Guadix-Baza Basin (southern Spain). Journal of Quaternary Science 15: 43-50.

Mithen, S. (1996). The Prehistory of the Mind. A search for the origins of art, religion and science. London: Thames and Hudson

Oms, O., Parés, J. M., Martínez-Navarro, B., Agustí, J., Toro, I., Martínez-Fernández, G. & Turq, A. (2000b). Early human occupation of Western Europe: Paleomagnetic dates for two paleolithic sites in Spain. Proc Natl Acad Sci U S A 97: 10666-10670.

Monnier, J.-L., Hallégouët, B., Hinguant, S., Laurent, M., Auguste, P. , Bahain, J.-J., Falguères, C., Gebhardt, A., Marguerie, D., Molines, N., Morzadec, H. & Yokoyama, Y. (1994). A new regional group of the Lower Palaeolithic in Brittany (France), recently dated by Electron Spin Resonance. C.R. Acad. Sci. Paris 319:155-60.

Ortega, A. I. (1994). La industria lítica de Torralba del Moral (Soria). Valladolid: Universidad de Valladolid

Montes Barquín, R. & Morlote, J. M. (1994). Aportación al estudio de los materiales líticos del Paleolítico Inferior de los alrededores de Altamira. Eds) Homenaje a González Echegaray, pp. 17-15. Santander: Museo y Centro de Investigación de Altamira.

Pallarès, M. & Pericot, L. (1931). Els jaciments asturians del Montgrí. Anuari de l’Institut d’Estudis Catalans (Secció Històrico-arqueològica) VII: 27-39. Pallí, L. (1976). Morfolitología de las terrazas del Ter en Girona. Anales de la Sección de Ciencias del Colegio Universitario de Gerona 1.

Montes Bernárdez, R. (1992). Los primeros grupos humanos depredadores en el sur de la Península (Andalucía, Murcia, Albacete). Munibe 44: 3-12.

Pallí, L. (1982). “Mapa geològic de Girona,”. Girona: Ajuntament de Girona/ Departament de Geologia del Col.legi Universitari de Girona.

Mora, R., Carbonell, E. & Martínez, J. (1987). Can Garriga: un tecnocomplejo en contexto estratigráfico (Sant Julià de Ramis, Girona). Cuaternario y Geomorfología 1: 195-218.

Palmqvist, P. (1997). A critical re-evaluation of the evidence for the presence of hominids in Lower Pleistocene times at Venta Micena, Southern Spain. Journal of Human Evolution 33: 83-89.

Moral, S. (2002). La Cueva del Mirador. La Edad del Bronce en la Sierra de Atapuerca. Burgos: Editorial Monte Carmelo (Ediciones Sierra de Atapuerca (Textos, 2).

Palol, P. d. (1969a). Informe que presenta el delegado de excavaciones de la zona universitaria de Valladolid, profesor Pedro de Palol, sobre los trabajos realizados durante el año 1964. Noticiatio Arqueológico Hispano X-XII (1966-1968): 294-300.

Morales, J., Soria, D. & Soto, E. (1987). Los carnívoros del Pleistoceno medio de Atapuerca. In (E. Aguirre, E. Carbonell, and J. M. Bermúdez de Castro, Eds), El hombre fósil de Ibeas y el Pleistoceno de la Sierra de Atapuerca, pp. 135-151. Valladolid: Junta de Castilla y León.

Palol, P. d. (1969b). Informe que presenta el delegado de excavaciones del distrito universitario de Valladolid, profesor Pedro de Palol, sobre los trabajos y hallazgos realizados durante el año 1966. Noticiario Arqueológico Hispano X-XII (1966-1968): 301-306.

Mosquera, M. (1995). Procesos técnicos y variabilidad en la industria lítica del Pleistoceno medio de la Meseta: Sierra de Atapuerca, Torralba, Ambrona y Áridos. Doctoral Thesis, Universidad Complutense (Departamento de Prehistoria).

Parés, J. M. & Pérez-González, A. (1995). Paleomagnetic age for hominid fossils at Atapuerca archaeological site, Spain. Science 269: 830-832.

Mosquera, M. (1998). Differential raw material use in the Middle Pleistocene of Spain: evidence from Sierra de Atapuerca, Torralba and Aridos. Cambridge Archaeological Journal 8: 15-28.

Parés, J. M. & Pérez-González, A. (1998). Contexto estratigráfico y cronológico de Gran Dolina (Yacimiento de Atapuerca). In (E. Aguirre, Eds) Atapuerca y la Evolución humana, pp. 49-60. Madrid: Fundación Ramón Areces.

Moyà-Solà, S. & Köhler, M. (1997). The Orce skull: anatomy of a mistake. Journal of Human Evolution 33: 91-97.

Parés, J. M. & Pérez-González, A. (1999). Magnetochronology and stratigraphy at Gran Dolina section, Atapuerca (Burgos, Spain). Journal of Human Evolution 37: 325-342.

Navazo, M. (2002). Poblamiento y uso del espacio prehistórico: prospección de las terrazas del río Arlanzón y estudio de sus yacimientos a través de la industrialítica. Burgos: Editorial Monte Carmelo (Colección Sierra de Atapuerca, 1,

Pelegrin, J. (1986). Technologie lithique: une méthode appliquée à l’étude de deux séries du Périgordien ancien, Roc de Combe couche 8, La Côte niveau III. Thèse de Doctorat, Université de Paris X (Dept.

Obermaier, H. (1925). El hombre fósil. Madrid: Comisión de Investigaciones Paleontológicas y Prehistóricas (Memorias, 9).

Peretto, C. & Ferrari, M. (1995). Techno-typological considerations on the industry from Ca’Belvedere di Monte Poggiolo (Forli, Italy). Cahier Noir 7: 3-16.

O’Connell, J. F. & Allen, J. (2004). Dating the colonization of Sahul (Pleistocene Australia–New Guinea): a review of recent research. Journal of Archaelogical Science 31: 835–853.

Peretto, C., Amore, F. O., Antoniazzi, A., Antoniazzi, A., Bahain, J. J., Cattani, L., Cavallini, E., Esposito, P., Falguères, C., Gagnepain, J., Hedley, I., Laurent, M., Lebreton, V., Longo, L., Milliken, S., Monegatti, P., Ollé, A., Pugliese, A., Renault-Miskovsky, J., Sozzi, M., Ungaro, S., Vannuci, S., Vergès, J. M., Wagner, J. J. & Yokoyama, Y. (1998). Industrie lithique de Ca’Belvedere di Monte Poggiolo: Stratigraphie, matière première, typologie, remontages et traces d’utilisation. L’Anthropologie 102: 343-465.

Olive, A., Ramírez Merino, J. L. & Ortega, L. I. (1990). Mapa Geológico de España, E.1:50000 (Belorado, 201). Madrid: Inst. Tecnológico Geominero de España Ollé, A. (2002). Variabilitat i patrons funcionals en els Sistemes Tècnics de Mode 2. Anàlisi de les deformacions d’ús en els conjunts lítics del RIparo Esterno de Grotta Paglicci (Rignano Garganico, Fog-

181

Xosé Pedro Rodríguez Pérez-González, A., Aleixandre, T., Pinilla, A., Gallardo, J., Benayas, J., Martínez, M. J. & Ortega, A. I. (1995). An approach to the Galeria stratigraphy in the Sierra de Atapuerca trench (Burgos). In (J. M. Bermúdez de Castro, J. L. Arsuaga & E. Carbonell, Eds) Human Evolution in Europe and the Atapuerca Evidence (Workshop, Castillo de la Mota, Medina del Campo, Valladolid, 1992), pp. 99-122. Valladolid: Junta de Castilla y León.

Roche, H. (1980). Premiers outils taillés d’Afrique. Paris: Société d’Ethnographie Roche, H. & Texier, P.-J. (1991). “La notion de complexité dans un ensemble lithique. Application aux séries acheuléennes d’Isenya (Kenya),” in 25 Ans d’Études technologiques en Préhistoire. XI Rencontres Internationales d’Archéologie et Histoire d’Antibes, pp. 99-108. Juan-les-Pines, France: APDCA.

Pérez-González, A. & Santonja, M. (1995). Los yacimientos de Ambrona y Torralba. Eds) IX Reunión sobre Cuaternario. Excursión postreunión, pp. 1-17. Madrid: AEQUA/CSIC.

Rodríguez, J., Díez, J. C., Laplana, C. & Nicolás, E. (1996). Estudio Paleontológico de la asociación de mamíferos del nivel TD6 (Pleistoceno inferior, Sierra de Atapuerca, Burgos, España). Revista Española de Paleontología 11: 199-206.

Pérez-González, A., Parés, J. M., Gallardo, J., Aleixandre, T., Ortega, A. I. & Pinilla, A. (1999). Geología y estratigrafía del relleno de Galería de la Sierra de Atapuerca (Burgos). In (E. Carbonell, A. Rosas & J. C. Díez, Eds) Atapuerca: Ocupaciones humanas y paleoecología del yacimiento de Galería, pp. 31-42. Valladolid: Junta de Castilla y León (Consejería de Educación y Cultura).

Rodríguez, J. (2001). Structure de la communauté de mammifères pléistocènes de Gran Dolina (Sierra de Atapuerca, Burgos, Espagne). L’Anthropologie 105: 131-157. Rodríguez, X. P. (1991). El complejo mesopleistocénico de la Sierra de Atapuerca (Burgos): variabilidad técnica de la industria lítica de Trinchera Dolina. Tesis de Licenciatura, Universitat de Barcelona (Dept. Departament de Geografia i Història).

Pérez-González, A., Parés, J. M., Carbonell, E., Aleixandre, T., Ortega, A. I., Benito, A. & Merino, M. A. M. (2001). Géologie de la Sierra de Atapuerca et stratigraphie des remplissages karstiques de Galería et Dolina (Burgos, Espagne). L’Anthropologie 105: 27-43.

Rodríguez, X. P. & Rosell, J. (1993). Contribución al conocimiento del Paleolítico inferior del Noreste de la Península Ibérica: el yacimiento de Nerets (Conca de Tremp, Catalunya). Cuaternario y Geomorfología 7: 15-22.

Pericot, L. (1923). L’Asturià del Montgrí. Butlletí de l’Associació Catalana d’Antropologia, Etnologia i Prehistòria: 206. Perles, C. (1985). Les industries lithiques de Franchthi (Argolide). Thèse de Doctorat, Université de Paris X (Dept.

Rodríguez, X. P., Sala, R., Casellas, S. & Vallverdú, J. (1995). Ocupació antròpica de la vall mitjana del Ter en l’inici del Plistocè superior. In (B. Agustí, J. Burch & J. Merino, Eds) Excavacions d’urgència a Sant Julià de Ramis (Anys 1991-1993), pp. 37-65. Girona: Centre d’Investigacions arqueològiques de Girona.

Perlès, C. (1991). Économie dees matières premières et économie du débitage: deux conceptions opposées? Eds) 25 Ans d’Études technologiques en Préhistoire. Bilan et perspectives. XI Rencontres Internationales d’Archéologie et Histoire d’Antibes, pp. 35-45. Antibes: APDCA.

Rodríguez, X. P. (1997). Los Sistemas Técnicos de Producción Lítica del Pleistoceno Inferior y Medio en la Península Ibérica: Variabilidad Tecnológica entre Yacimientos del Noreste y de la Sierra de Atapuerca. Ph. D. Thesis, Universitat Rovira i Virgili (Departament d’Història i Geografia).

Piaget, J. (1971). Biology and Knowledge. Edinburgh: Edinburgh University Press Pigeot, N. (1987). Magdaléniens d’Étiolles: économie de débitage et organisation sociale. Paris: CNRS (Gallia Préhistoire, XXV).

Rodríguez, X. P. & Lozano, M. (1999). El Pleistoceno medio y superior inicial del Noreste de la Península Ibérica. Pyrenae 30: 39-68.

Pineda, A. & Arce, J. M. (1993). Mapa Geológico de España a Escala 1:50.000, Burgos, 200. Madrid: Instituto Tecnológico GeoMinero de España

Rodríguez, X. P. & Lozano, M. (2000). Situación actual y problemática de la investigación del Paleolítico inferior en el noreste de la Península Ibérica. Eds) Actas do 3º Congresso de Arqueologia Peninsular, vol. II: Paleolítico da Península Ibérica, pp. 71-80. Porto: ADECAP.

Ploux, S. (1991). Technologie, technicité, techniciens: méthode de détermination d’auteurs et comportements techniques individuels. Eds) 25 Ans d’Études technologiques en Préhistoire. Bilan et perspectives. XI Rencontres Internationales d’Archéologie et Histoire d’Antibes, pp. 201-214. Juan-les-Pines, France: APDCA.

Rodríguez, X. P., Carbonell, E. & Ortega, A. I. (2001). Historique des découvertes préhistoriques de la sierra de Atapuerca (Burgos, Espagne), et perspectives du futur. L’Anthropologie 105: 3-12.

Prats, J. M. (1991). Can Garriga, N-1. Un registre de 87.000 a 103.000 anys B.P. del Nord-Est de la Península Ibèrica. Tesis de Licenciatura, Universitat de Barcelona (Facultat de Lletres de Tarragona).

Rodriguez, X.P. (2004). Atapuerca y el inicio del Paleolítico medio en Europa. In Baquedano E. y Rubio, S. (Eds.): Homenaje a Emiliano Aguirre, vol IV. Madrid: Museo Arqueológico Regional de Madrid, pp. 416-431.

Querol, M. A. & Santonja, M. Eds. (1979). El yacimiento achelense de Pinedo (Toledo). Madrid: Ministerio de Cultura. Excavaciones Arqueológicas en España, Vol. 106).

Rodríguez, X. P., Maroto, J., Vaquero, M., Ortega, D., García, J. & Lozano, M. (2004). El Paleolític inferior i mitjà a Catalunya. Fonaments 10/11: 23-66.

Querol, M. A. & Santonja, M. (1983). El yacimiento de cantos trabajados de El Aculadero (Puerto de Santa Maria, Cádiz). Madrid: Ministerio de Cultura (Excavaciones Arqueológicas en España, 130).

Rodríguez-Asensio, J. A. (1978). The Early Paleolithic site of Bañugues (Gozón, Asturias, Spain). Current Anthropology 19: 615-616.

Quintanilla, M. Á. (1988). Bases para la filosofía de la técnica. (La estructura de los sistemas técnicos). Arbor: 11-28.

Rodríguez-Asensio, J. A. (2001). Yacimiento de Cabo Busto. Los orígenes prehistóricos de Asturias. Luarca: GEA (Gran Enciclopedia Asturiana)

Raynal, J.-P., Sbihi-Alaoui, F. Z., Geraads, D., Magoga, L. & Mohi, A. (2001). The earliest occupation of North-Africa: the Moroccan perspective. Quaternary International 75: 65-75.

Roebroeks, W., Conard, N. J. & van Kolfschoten, T. (1992). Dense Forest, Cold Steppes, and the Palaeolithic Settlement of Northern Europe. Current Anthropology 33: 551-586.

Ripoll, E. & Lumley, H. d. (1965). El Paleolítico Medio en Cataluña. Ampurias XXVI-XXVII: 1-67.

Roebroeks, W. (1994). Updating the Earliest Occupation of Europe. Cur-

182

REFERENCES Sala, R. (1997). Formes d’ús i criteris d’efectivitat en conjunts de Mode 1 i Mode 2: Anàlisi de les deformacions per ús dels instruments lítics del Plistocè inferior (TD6) i mitjà (TG11) de la Sierra de Atapuerca. Ph. D. Thesis, Universitat Rovira i Virgili (Dept. Història i Geografia).

rent Anthropology 35: 301-305. Roebroeks, W. & van Kolfschoten, T. (1994). The Earliest Occupation of Europe: A Short Chronology. Antiquity 68: 489-503. Roebroeks, W. & van Kolfschoten, T. (1995). The earliest occupation of Europe: a reappraisal of artefactual and chronological evidence. In (W. Roebroeks & T. van Kolfschoten, Eds) The earliest occupation of Europe, pp. 297-315. Leiden: University of Leiden.

Sala, R., Aulines, A., Garcia, J., Martínez, K., Matarrodona, M., Bargalló, A., Sánchez, P., Martínez, I., García, P., Pujadas, R., Gómez, B., Campeny, G. & Barceló, P. (2002). El jaciment del pleistocè mitjà i superior inicial del volcà del Puig Marí (Maçanet de la Selva, Girona). Eds) Sisenes Jornades d’Arqueologia de les comarques gironines, Sant Joan de les Abadesses, pp. 23-27.

Roebroeks, W. & van Kolfschoten, T. (1998). The earliest occupation of Europe: a view from the North. In (E. Aguirre, Eds) Atapuerca y la evolución humana, pp. 153-168. Madrid: Fundación Ramón Areces.

Sánchez, B., & Cerdeño, E. (1999). Perissodactyla del yacimiento del Pleistoceno medio de Galería (Sierra de Atapuerca). In (E. Carbonell, A. Rosas, and J. C. Díez, Eds), Atapuerca: Ocupaciones humanas y paleoecología del yacimiento de Galería, pp. 169-174. Zamora: Junta de Castilla y León (Consejería de Educación y Cultura).

Rosas, A., Carbonell, E., Cuenca, G., García, N., Fernández Jalvo, Y., Made, J. van der, Ollé, A., Parés, J. M., Pérez González, A., Sánchez Marco, A., Sánchez Chillón, B. & Vallverdú, J. (1998). Cronología, bioestratigrafía y paleoecología del Pleistoceno medio de Galería (Sierra de Atapuerca, España). Revista Española de Paleontología 13:71-80.

Santonja, M. & Querol, M. A. (1980). Estudio técnico y tipológico de la industria lítica del sitio de ocupación achelense de Áridos-1. In (M. Santonja, N. López Martínez & A. Pérez González, Eds) Ocupaciones achelenses en el valle del Jarama, pp. 253-278. Madrid: Publicaciones de la Diputación Provincial de Madrid.

Rosas, A., Pérez-González, A., Carbonell, E., van der Made, J., Sánchez, A., Laplana, C., Cuenca-Bescós, G., Parés, J. M. & Huguet, R. (2001). Le gisement pléistocène de la “Sima del Elefante” (Sierra de Atapuerca, Espagne). L’Anthropologie 105: 301-312.

Santonja, M. & Pérez González, A. (1984). Las industrias paleolíticas de La Maya I en su ámbito regional. Madrid: Ministerio de Cultura (Excavaciones Arqueológicas en España, 135).

Rosell, J. & Rodríguez, X. P. (1991). Paleolític inferior a la conca de Tremp: la localització arqueològica dels Nerets. Collegats 5: 133139.

Santonja, M. (1994). Los últimos diez años en la investigación del Paleolítico inferior de la Cuenca del Duero. Veleia 8-9: 7-41.

Rosell, J. (1993). Impacte biològic a la base de «Gran Dolina» (Sierra de Atapuerca, Burgos). Tesis de Licenciatura, Universitat Rovira i Virgili (Departament d’Història i Geografia).

Santonja, M. (1995). El Paleolítico inferior en Europa: Apuntes en un momento de revisión. Boletín de la Asociación Española de Amigos de la Arqueología 35: 53-62.

Rosell, J. (1998). Les premières occupations humaines à la Sierra de Atapuerca (Burgos, Espagne) les niveaux TDW4 et TDW-4B. Eds) Économie Préhistorique: les comportements de subsitance au Paléolithique. XVIIIe Rencontres Internationales d’Archéologie et d’Histoire d’Antibes, pp. 154-161. Sophia Antipolis: Éditions APDCA.

Serra, S., Gutiérrez, R., Carbonell, E. & Canal, J. (1981). Puig d’en Roca III. Un nuevo lugar de ocupación del Paleolítico Inferior en el Valle Medio del Ter (Girona). Butlletí de l’Associació Arqueològica de Girona 4: 4-15.

Rosell, J., Cáceres, I. & Huguet, R. (1998). Systèmes d’occupation anthropique pendant le Pléistocene inférieur et moyen à la Sierra de Atapuerca (Burgos, Espagne). Quaternaire 9: 355-360.

Solé Sabarís, L. & Ribera, M. (1949). Explicación de la Hoja n. 334 del Mapa Geológico de España, E. 1:50.000. Madrid: Instituto Geológico y Minero de España

Roux, V. (1991). Peut-on interpréter les activités lithiques préhistoriques en termes de durée d’apprentissage? Apport de l’ethnologie et de la psychologie aux études technologiques. Eds) 25 Ans d’Études technologiques en Préhistoire. Bilan et perspectives. XI Rencontres Internationales d’Archéologie et Histoire d’Antibes, pp. 47-56. Antibes: APDCA.

Soler, N. (1982). El Cau del Duc de Torroella de Montgrí. Eds) Excavacions Arqueològiques a Catalunya en els darrers anys, pp. 31-32. Barcelona: Generalitat de Catalunya. Soto, E. (1987). Grandes herbívoros del Pleistoceno medio de la Trinchera del ferrocarril de Atapuerca (Burgos, España). In (E. Aguirre, E. Carbonell, and J. M. Bermúdez de Castro, Eds) El hombre fósil de Ibeas y el Pleistoceno de la Sierra de Atapuerca, pp. 93-111. Valladolid: Junta de Castilla y León.

Royo y Gómez, J. (1926). Tectónica del Terciario continental ibérico. Boletín del Instit. Geológico y Minero de España 47: 129-158.

Straus, L. G. (2001). Africa and Iberia in the Pleistocene. Quaternary International 75: 91-102.

Sahnouni, M. (1985). Industries sur Galets du gisement Villafranchien supérieur d’Ain Hanech (Sétif, Algérie Orientale). Approche morphotechnologique. Ph. D., Université de Paris VI.

Tavoso, A. (1984). Réflexion sur l’économie des matières premières au Moustérien. Bull. Soc. Préhist. Fr. 81: 79-82.

Sahnouni, M. (1998). The Lower Palaeolithic of the Maghreb: Excavations and Analyses at Ain Hanech, Algeria. Oxford: BAR International Series 689 (Cambridge Monographies in African Archaeology 42),

Texier, P.-J. (1985). Chilhac III: un gisement paléntologique villafranchien soliflué ? Bull. Soc. Préhist. Fr. 82: 68-70.

Sahnouni, M. & Heinzelin, J. D. (1998). The site of Ain Hanech revisited: New investigations at this Lower Pleistocene site in northern Algeria. Journal of Archaeological Science 25: 1083-1101.

Texier, P.-J. (1995). Polyèdre, sub-sphéroïde, sphéroïde et bola: des segments plus ou moins longs d’une même chaîne opératoire. Cahier Noir 7: 31-40.

Sahnouni, M., Derradji, A., Hadjouis, D., Canals, A., Medig, M., Belahrech, H., Abdesselam, S., Harichane, Z. & Rabhi, M. (2001). Continuing investigations in the Early Pleistocene locality of Ain Hanech, northeastern Algeria. Journal of Human Evolution (Abstracts for the Paleoanthropology Society Meetings) 40: A18-A19.

Thieme, H. (1997). Lower Palaeolithic hunting spears from Germany. Nature 385: 807-810. Thompson, E. P. (1981). Miseria de la Filosofía. Barcelona: Crítica Thorne, A., Grün, R., Mortimer, G., Spooner, N., Simpson, J., McCul-

183

Xosé Pedro Rodríguez loch, M., Taylor, L. & Curnoe, D. (1999). Australia’s oldest human remains: age of the Lake Mungo 3 skeleton. Journal of Human Evolution 36: 591-612.

Virgili. Vega Toscano, L. G. (1989). Ocupaciones humanas en el Pleistoceno de la Depresión de Guadix-Baza: elementos de discusión. In (M. T. A. F. Bonadonna, Eds) Geología y Paleontología de la Cuenca de Guadix-Baza, pp. 327-345. Madrid: CSIC.

Tissoux, H. (1999). Géochronologie de sites paléolithiques de Catalogne. Résultats préliminaires obtenues par les méthodes U-Th et ESR sur les sites de l’Arbreda, de Cau del Duc d’Ullà et de Cau del Duc de Torroella de Montgrí. Mémoire de DEA, Musem National d’Histoire Naturelle (Institut de Paléontologie Humaine).

Vergès, J. M. (1996). Impacte antròpic i pautes tecnofuncionals al Plistocè mitjà: La indústria lítica del nivell TD10 de Gran Dolina (Sierra de Atapuerca, Burgos). Tesis de Licenciatura, Unviersitat Rovira i Virgili (Departament d’Història i Geografia).

Tixier, J. (1991). Cogitations non conclusives. In 25 Ans d’Études technologiques en Préhistoire. Bilan et perspectives. XI Rencontres Internationales d’Archéologie et Histoire d’Antibes, pp. 391-394. Juan-Les-Pines, France: APDCA.

Vergès, J. M. (2002). Caracterització dels models d’instrumental lític del Mode 1 a partir de les dades de l’anàlisi funcional dels conjunts litotècnics d’Aïn Hanech i el-Kherba (Algèria), Monte Poggiolo i Isernia la Pineta (Itàlia). Doctoral Thesis, Universitat Rovira i Virgili (Dept. Història i Geografia).

Toro, I. (1999). El proyecto de investigación sobre el Plio-pleistoceno en la cuenca de Guadix-Baza: De la investigación al desarrollo. Hacia un modelo de gestión del patrimonio arqueológico y paleontológico. Boletín del Instituto Andaluz del Patrimonio Histórico 29: 156-167.

Vergès, J. M., Allué, E., Angelucci, D. E., A., C., Díez, C., Fontanals, M., Manyanós, A., Montero, S., Moral, S., Vaquero, M. & Zaragoza, J. (2002). La sierra de Atapuerca durante el Holoceno: datos preliminares sobre las ocupaciones de la Edad del Bronce en la Cueva de El Mirador (Ibeas de Juarros, Burgos). Trabajos de Prehistoria 59: 107-126.

Torres, T. (1984). Úrsidos del Pleistoceno-Holoceno de la Península Ibérica. Doctoral Thesis, Universidad Politécnica de Madrid. Toth, N. (1982). The Stone Technologies of Early Hominids from Koobi Fora, Kenya: an Experimental Approach. Ph. Doctoral Thesis, University of California.

Vert, J. (1976). El Montgrí dintre el Paleolític inferior. In El Paleolític a les comarques gironines, pp. 68-76. Girona: Ed. Curbert.

Toth, N. (1987). Behavioral inferences from Early Stone artifact assemblages: an experimental model. Journal of Human Evolution 16: 763-787.

Vert, J., Puig, X., Canal, J. & Carbonell, E. (1977). El poblament del Montgrí en el Paleolític Inferior. Revista de Girona 80: 249-262.

Turq, A. (1992). Le Paleolithique inférieur et moyen entre les vallées de la Dordogne et du Lot. Doctoral Thesis, Université de Bordeaux I.

Vidal Encinas, J. M. (1982). Las Gándaras de Budiño: balance preliminar de dos campañas de excavaciones (1980 1981). El Museo de Pontevedra XXXVI: 91-113.

Turq, A., Martínez-Navarro, B., Palmqvist, P., Arribas, A., Agustí, J. & Rodríguez Vidal, J. (1996). Le Plio-Pleistocene de la région d’Orce, province de Grenade, Espagne: bilan et perspectives de recherche. Paleo 8: 161-204.

Villa, P. (1996). Book review: The First Italians. Le industrie litiche del giacimento paleolitico di Isernia La Pineta. Lithic Technology 21: 71-79.

Uribarri, J. & Apellániz, J. M. (1975). “Problemas prehistóricos de la Galería del Sílex de la Cueva de Atapuerca,” in XII Congreso Nacional de Arqueología, pp. 167-172.

Wynn, T. (1979). The Intelligence of Later Acheulean Hominids. Man 14: 371-391. Wynn, T. (1993). Layers of thinking in tool behavior. In (K. Gibson & T. Ingold, Eds) Tools, Language and Cognition in Human Evolution, pp. 389-406. Cambridge: Cambridge University Press.

Valet, J.-P. & Meynadier, L. (1993). Geomagnetic field intensity and reversals during the past four million years. Nature 366: 234-238. Vallespí Pérez, E. (1994). El Bajo Guadalquivir en el Paleolítico Inferior y Medio peninsular. Eds) Homenaje al Dr. Joaquín González-Echegaray, pp. 13-16. Altamira: Museo y Centro de Investigación de Altamira (Monografías, nº 17).

Wynn, T. (1995). Handaxe enigmas. World Archaeology 27: 10-24. Zazo, C., Goy, J. L. & Hoyos, M. (1983). Estudio geomorfológico de los alrededores de la Sierra de Atapuerca (Burgos). Estudios geológicos 39: 179-185.

Vallespí Pérez, E. & Díaz del Olmo, F. (1996). Industries in quartzite and the beggining of the use of flint in the Lower and Middle Palaeolithic sequence of the Bajo Guadalquivir. In (N. Moloney, L. Raposo & M. Santonja, Eds) Non-Flint Stone Tools and the Palaeolithic Occupation of the Iberian Peninsula, pp. 135-140. Oxford: Tempus Reparatum.

Zazo, C., Goy, J. L. & Hoyos, M. (1987). Contexto geológico y geomorfológico. In (E. Aguirre, E. Carbonell & J. M. Bermúdez de Castro, Eds) El hombre fósil de Ibeas y el Pleistoceno de la Sierra de Atapuerca, pp. 41-46. Valladolid: Junta de Castilla y León.

Vallespí Pérez, E. J., Ciudad Serrano, A. & García Serrano, R. (1979). Achelense y Musteriense en Porzuna (Ciudad Real). Materiales de superfície I. (Colección E. Oliver). Ciudad Real: Museo de Ciudad Real (Estudios y Monografías I, Vallespí Pérez, E. J., Ciudad Serrano, A. & García Serrano, R. (1985). Achelense y Musteriense de Porzuna (Ciudad Real). Materiales de superfície II (Muestras de las colecciones de A. Retamosa y M. Expósito). Ciudad Real: Universidad de Castilla La Mancha Vallespí Pérez, E. J., Díaz del Olmo, F., Alvarez, G. & Vallespí García, E. (1988). Secuencia Paleolítica del Bajo Guadalquivir. Revista de Arqueología 82: 8-17. Vallverdú, J. (1993). Dades per a l’estratigrafia quaternària del Camp de Tarragona: el paleosòls. Tesis de Licenciatura, Universitat Rovira i

184