IMAGES

OF THE
ICE AGE
PauiG Bahn&
Jean Vertut

s
Facts On File
New York Oxford
 ~ I
r

'< d
~~
r r}
B ? r
I a 69

For Eric and Glyn

and
For Yvonne Vertut

Copyright Paul G Bahn 1988

Photographs copyright Jean Vertut 1988
All rights reserved. No part of this book may be reproduced
or utilized in any form or by any means, electronic or
mechanical, including photo copying, recording, or by any
information storage and retrieval systems, without
permission in writing from the publisher. For information
contact: Facts On File, Inc.
460 Park Avenue South
New York, New York 10016
Library of Congress Cataloging-in-Publication Data
Bahn, Paul G.
Images of the Ice Age.
Bibliography: p.
Includes index.
1. Paleolithic period - France 2. Paleolithic period -
Spain. 3. Cave-drawings - France. 4. Cave-drawings-
Spain 5. France - Antiquities. 6. Spain- Antiquities.
I. Vertut, Jean. II. Title.
GN772.22 .F7B32 1989 936.4 88-33405
ISBN 0-8160-2130-9
Facts On File books are available at special discounts when
purchased in bulk quantities for businesses, associations,
instirutions, or sales promotion. Please contact the Special
Sales Department at 212 I 683-2244.
(Dial1-800-322-8755, except in NY, AK, HI)
10 9 8 7 6 5 4 3 2 1
This book was designed and produced by
Bellew Publishing Company Limited
7 Southampton Place, London WC1 :\ 2DR
Printed and bound in Italy by New Interlitho, Milan

Frontispiece:
Bison drawn in the Reseau Clastres (Ariege)
with concretions on top of it. Probably Magdalenian.
Length: 1.14 m. (JV)
 Contents

Foreword by Count Robert Begouen 7

Preface and Acknowledgements 9
Introduction: Cave Life 13

1. The Discovery of Ice Age Art 17

2. A Worldwide Phenomenon 26
3. Making a Record 41
4. How Old is the Art? 53
5. Forms and Techniques 68
6. What was Depicted? 115
7. Reading the Messages 149
8. Conclusion 191
Appendix by Alexander Marshack 195
Notes 203
Bibliography 216
List of Illustrations 235
Index 237
Fig
the
len!
cu
use
bre
 Foreword
by Count Robert Begouen

It was inevitable that the destinies of Paul Bahn and Jean Vertut would lead
one day to their meeting on the path to a Palaeolithic sanctuary. Although
they had totally different academic backgrounds, both from their youth
had been attracted to and fascinated by the civilisations of the 'Reindeer
Age'.
Paul Bahn rapidly dedicated himself totally to their study, travelling
round Europe and, later, the world to obtain first-hand experience of
archaeological sites, materials and prehistorians. How can one ever forget
the sight of this young Englishman, thirsting for knowledge, arriving in
Ariege on an overloaded old velomoteur ! From museums to libraries he
took the time to see and read everything, a procedure that is quite
exceptional in this age of specialists. The result was a doctoral thesis whose
title sums up its approach: 'Pyrenean Prehistory: a palaeoeconomic survey
of the French sites' (1984). In it he poses a thousand new questions about
the various activities of prehistoric people, their movements, their way of
life, their mentality, all of which are avenues of research. He does not
hesitate to call into question some ideas which were considered solid , to
highlight certain contradictions, compare the observations of different
scholars, confront them with his own, and to produce a synthesis that is
admirable for its finesse and its good sense . This enormous work has
already become a precious tool for prehistorians.

After an education at the Ecole Centrale, from 1951 onward Jean Vertut
took part in the excavations of Arcy-sur-Cure , directed by Professor Andre
Leroi-Gourhan; his role was that of a specialist in colour photography
underground. When the decorated zone of the Grotte du Cheval was
Fig. 2 Jean Vertut photographing
the 'weasel' (probably Magdalenian ) discovered, Jean was naturally invited to take his first slides of parietal art.
letrgth 46 em) drawn in the R eseau Leroi-Gourhan very quickly noticed the originality of this talented and
Clastres (Ariege). He almost never likeable young man who had an intellectual and spiritual curiosity that was
wed a tripod, he just stopped always ready to marvel at le Reel. When the time came, he entrusted
breathing to take the shots. (Photo R. Vertut with the task of producing the illustrations for his master-work,
Simonnet) Prehistoire de J'Art Occidental.
8 IMAGES OF THE ICE AGE

Though he had an excellent 'eye', the principal quality of a

photographer, Jean Vertut was also an engineer in telemanipulation and
robotics; these disciplines also helped him to develop new, increasingly
sophisticated procedures enabling one to approach and to photograph
clearly the masterpieces of prehistoric art that were hidden away,
sometimes in inaccessible crannies . He was a pioneer in the problems of
focusing at a distance, computing light-measurements, and the use of
simultaneous multiple electronic flashes connected to the camera - all
techniques which have become run-of-the-mill today.
It was in 1956, during a visit to the cave of Trois-Freres, that the
Begouens first made the acquaintance of Jean Vertut and of Yvonne, his
wife, who was a precious assistant in his subterranean expeditions; but it
was only in 1964 that Jean came to take his first series of slides in our caves.
I was so enthusiastic about his results that I then asked him to undertake
the exhaustive photographic coverage of the Volp Caves with me. This
work, a long-term project , led us to spend hundreds of hours together, in
conditions that were sometimes unbelievable but which enabled Jean's
lens to capture the figures in their entirety, to eliminate as far as possible
the deformations caused by the irregularity of the rock, and to bring out
various little details that the artist had engraved carefully; all of this
improved notably the reading and understanding of the drawings.
Ceaselessly perfecting his technique, Jean Vertut became the
undisputed photographer of prehistoric art. Shortly before his death, he
had started important research on 3-D photography as applied to
prehistory, with possible extensions into photogrammetric recording.
Thanks to him, photography of parietal art passed the stage of simple -
albeit very beautiful - illustration, and became a separate technique of
research .
Readers will take great pleasure in discovering that technique in this
book. The magic of the images will give them wings, and everything will
then be transformed, transcended! Everyone will feel something human
and essential in these messages that were engraved, painted or sculptured
in the caves during the twenty millennia of the Upper Palaeolithic . These
images touch us, simply because they are beautiful. They remind us that
people never cease to improve what they have undertaken to do. Whatever
their motivations (and we know that the layouts are organised and
complex), it is nevertheless clear that at some point people loved their
works and, consciously or unconsciously, integrated Beauty into their
dialogue with Matter .

Count Robert Begouen

1988
 Preface and
Acknowledgements

Jean Vertut and I began planning this book in 1984. His sudden and
untimely death in May 1985 at the age of only 56 robbed the world of a kind
and remarkable man, and cave-art of its foremost photographer. It also
meant that the book could no longer be what we envisaged, since Jean
could neither choose the pictures himself, nor describe his methods and
techniques as he had wished to do, nor photograph the new sites and
objects that we had planned to include.
Nevertheless, with the help and encouragement of his widow, Yvonne ,
it has proved possible to produce an approximation of the original plan.
Naturally, it has been necessary to draw primarily on the range of sites
represented in Jean's existing collection, with only a few pictures
borrowed from other sources where required to illustrate precise points .
Inevitably, therefore, this book concentrates heavily on the Ice Age art of
France and Spain.
It was never our intention, however, to produce an encyclopaedic work
or a catalogue, or to attempt to replace the great tome by the late Andre
Leroi-Gourhan (which remains one of the glories ofJean's career as well as
of Leroi-Gourhan's); we wished to give a survey of past and current
theories about the art rather than promote a single favourite interpretation.
For reasons explained in the text , it will be some time before another
comprehensive volume of that type can appear - if indeed such a thing is
desirable. Sadly, our own book has now taken on an extra, unforeseen
role, that of serving as a tribute to Jean by displaying the best of his final
work. It goes without saying that, since the text was written in 1987, it is
entirely my own responsibility; Jean may not necessarily have agreed with
all the views expressed here, and the text is the poorer for the absence of his
influence and advice.
Since the book is meant for the interested layman as well as for the
academic, we decided not to get bogged down in arguments about the
definition of 'art', or in speculations as to how and why it arose and
disappeared. Much of that debate is redundant, since the earliest art to
appear archaeologically is at an already well-developed stage, and we have
lost its antecedents. In any case, these subjects have been treated
10 IMAGES OF THE ICE AGE

elsewhere, in great depth .' Art, or at least the ability to create images in
materials that have survived, appears relatively suddenly in the
archaeological record at roughly the same time as Homo sapiens sapiens;
clearly, something had developed in the mentality of our sub-species (and
perhaps in that of its predecessors) which enabled or stimulated it to
produce pictures, to adapt pre-existing shapes or to visualise new forms
before they exist.
The term 'art' is rapidly losing favour among those researching the
Palaeolithic period, since it presupposes an aesthetic function and lumps
together a wide range of objects and types of image-making spanning 25
millennia; other vaguer terms - 'pictures', 'iconography', 'images' ,
'pictograms/ideograms', 'symbolic graphisms' or 'decoration' - are now
preferred. In this book, therefore, the word 'art' is retained simply as a
convenient blanket-term for the decoration of objects, rocks and cave-
walls, whatever its function or motivation may have been.
Similarly, we chose not to devote the book to long, detailed and tedious
discussions of the style and dating of particular caves, and thus to avoid all
the entanglements and contradictions those subjective topics lead one into;
at least one recent book has dwelt at length on these issues; 2 nor did we
wish to devote much space to the equally impressionistic and subjective
topic of what it feels like to visit these sites - this too has been done quite
vividly in a recent popular work. 3 Moreover, it is superfluous to describe
the beauty of particular depictions - readers can make their own
judgements from the illustrations.
Very few books cover both the cave-art and the portable art of the last Ice
Age, although the two cannot effectively be studied without reference to
each other. We were determined to include both, but our book suffers
from the usual archaeological vice of displaying only the finest examples;
as such, they are not truly representative of the whole , but colour plates are
a limited and valuable resource, and it seems a pity to waste them on
illustrations which appear mediocre in execution, are in poor condition, or
would be unintelligible to the untrained eye.
Lack of space has precluded setting the art within the cultural, economic
and social context oflce Age life, although such aspects inevitably crop up
here and there. T o integrate these themes properly would require a
separate book; meanwhile, a number of texts about Ice Age life and culture
are readily available to the layman, while the specialists will have their own
views . Palaeolithic art includes so many subjects, and so many erroneous
facts and interpretations are already in print about it , that it seemed best to
cover it as fully and as factually as possible, in order to correct former
errors and to assess the contribution made by recent work and discoveries .
References to parallels in other cultures have been kept to an absolute
minimum .
.ioince the text is aimed at the non-specialist, radiocarbon dates have been
given as a straightforward figure - they should be taken as only a rough
guide to the true age of the object or layer in question; specialist readers
can find full details of the plus/minus range, the substance dated, and the
laboratory concerned, by following up the references provided.
More and more people are becoming interested in the decorated caves,
but fewer and fewer sites can be visited because of the sheer difficulties of
access, or the potential damage to the images from accidents, vandals or
bacteria- it is generally reckoned that a decorated cave open to the public
is probably doomed . T here is therefore a clear need for other ways of
PREFACE AND ACKNOWLEDGEMENTS 11

presenting the art to the public . Lascaux II, a copy almost

indistinguishable from the original, is the ideal answer to this problem,
albeit an extremely expensive one; for other sites , videos and television
documentaries are a partial solution, and so are books of this kind . Much
cave-art, however, can be fully appreciated only in three dimensions, so
that one can see how the artist was inspired by or utilised the shape of the
rock. Among Jean's last photographs are a series of stereoscopic studies
which he hoped to place in this volume, together with special glasses to
produce a 3-D effect. Such publications will inevitably appear within the
next decade, thanks in large measure to the expertise and vision of this
great pioneer.

My own acquaintance with Palaeolithic cave-art began in June 1963 when ,

at the age of nine, I heard the exciting story of the discovery of the art in the
Volp caves (by the Begouen brothers in 1912/1914) in a BBC radio
broadcast for schools (People, Places and Things); at about the same time I
acquired - as a free gift from some brand of washing powder - a waste-
paper bin, which I use still, decorated with Lascaux-type animals! A more
decisive influence was to come at Cambridge University where, during my
first undergraduate supervisions with Eric Higgs, I was held spellbound
by his tales of visits to the caves of France and Spain. Unlike some of his
disciples, and in contrast with his own image as a prehistorian
contemptuous of the dilettante art-historian approach to the past, Eric
loved Palaeolithic art. My interest was heightened by reading books on the
subject, and in lectures at Cambridge given by John Coles and Charles
McBurney.
Finally, it was Eric Higgs who suggested that I do the research for a
doctoral dissertation on the prehistory of the French Pyrenees- no doubt
because he knew I would be fascinated by the region's richness in
Palaeolithic art . During expeditions to southern France and Spain with
him in the mid-1970s, I was agreeably surprised to find him making
detours so that I could visit Gargas , Niaux, Le Mas d' Azil, Candamo, and
other sites. After his death in 1976, I pursued my research in France with
the active support and encouragement of the late Professor Glyn Daniel,
who helped me to obtain my first permission to visit Lascaux, took a keen
interest in my work and, I am proud to say, became a close friend. It is
therefore to these two men, entirely different yet both admirers of Ice Age
art, that I would like to dedicate this volume. I wish they had lived to see it.
During the year 1985/6 I was fortunate enough to be awarded one of the
first J. Paul Getty Postdoctoral Fellowships in the History of Art and the
Humanities, and used it to improve my knowledge ofthe cave-art sites and
literature in Europe (using Toulouse as a base, through the kindness of
Claude Barriere), as well as to travel abroad to see as much prehistoric art
as I could, especially in Australia. I am profoundly grateful to the Getty
T rust for such a wonderful opportunity, from which this book has
benefited enormously.
In the course of my 'Getty travels' as well as during visits to France and
Spain over the years, I have been helped by a great number of friends and
colleagues in my research on Ice Age art. In Paris, the late Leon Pales, the
late Andre Leroi-Gourhan, Arlette Leroi-Gourhan, Claude Couraud,
Henri Delporte, Denis Vialou, Michel Garcia, Georges Sauvet,
Dominique Buisson, Genevieve Pin<;on , Lucette Mons, Suzanne de St
Mathurin, Michel Orliac, Sophie de Beaune, Gilles Tosello, Jacques
12 IMAGES OF THE ICE AGE

Allain, Nicole Limondin , Marthe Chollot-Varagnac, Marina Rodna. In

the Pyrenees, Robert Begouen, Jean Clones, Dominique Sacchi, Claude
Barriere, Louis-Rene N ougier, Aleth Plenier, Carol Rivenq, Anne-
Catherine Welte, Jacques Omnes, Jean Vezian, Jacques Blot, Robert
Arambourou, Andre Clot , Rene Gailli, Luc Wahl, Romain Robert, Jean
Abelanet , Robert Simonnet, Georges Laplace, Roger Seronie-Vivien,
Claude Andrieux. In the Perigord, Brigitte and Gilles Delluc, Paulette
Daubisse, Jean Gaussen, Alain Roussot, Jean-Philippe Rigaud, Jean-
Marc Bouvier, Jean-Michel Mormone. In Spain, Jesus Altuna, Ignacio
Barandianin, Vicente Baldellou, Reyna ldo Gonzalez Garcia, Alfonso
Moure Romanillo, Francisco Jorda Cerda, Josep Maria Fullola Pericot,
Jose Manuel Gomez-Tabanera, Joaquin Gonzalez Echegaray, Eduardo
Ripoll Perell6, Jose Miguel de Barandiaran. Elsewhere in Europe, Lya
Dams, Marylise Lejeune , Yves Martin, Andre Thevenin, Martine Faure,
Jean Combier, Michel Lorblanchet, Gerhard and Hannelore Bosinski,
Joachim Hahn , Karl Narr , Hans Biedermann, Bohuslav Klima, Jan
Jelinek, Zo1a Abramova, Mike and Anne Eastham, Pat Winker, Warwick
Bray, Alex Hooper , Jonathan Speed; and outside Europe, David Lewis-
Williams and Anna Belfer-Cohen. In America, Alex Marshack, Meg
Conkey, Ray and Jean Auel, Tom and Alice Kehoe, Glen Cole, John
Pfeiffer, Noel Smith, Whitney Davis. In Australia, Robert and Elfriede
Bednarik, Geoff Aslin , John Clegg, Pat Vinnicombe, Sylvia Hallam,
Andree Rosenfeld , George Chaloupka, Jo Flood, Rhys Jones, Percy/
Steve/Matt Trezise , Paul T a<;on, Jo McDonald, Lesley Maynard, Charles
Dortch. In China , Zhu Qing Sheng, You Yuzhu; and in Japan, Yuriko
Fukasawa, Yoshiaki Kanayama, Gina Barnes and Hideji Harunari.
For supplying special pictures for this book I am particularly grateful to
Dominique Sacchi, Robert Begouen, Robert Simonnet, Carol Rivenq,
Claude Couraud, Gerhard and Hannelore Bosinski, Alfonso Moure
Romanillo, Glen Cole, Andree Rosenfeld, Brigitte and Gilles Delluc, Jean
Clones and Ulm Museum .
Special thanks go to Meg Conkey, Whitney Davis and John Pfeiffer for
their support and for detailed comments on the typescript.
The original idea for a volume co-authored with Jean Vertut came from
Ib Bellew; the project was initiated with the help of Brigitte and Gilles
Delluc, set up by my agent , Andrew Best, and brought to fruition thanks
to the unfailing help of Yvonne Vertut. Deirdre McDonald showed
remarkable skill in transforming text and pictures into a coherent book. I
am deeply grateful to my- and Jean's- good friends, Robert Begouen and
Alex Marshack, for writing the Foreword and Appendix.
Finally, I have a tremendous debt of gratitude to the Getty Grant Program
of the J. Paul Getty Trust for following up my postdoctoral Fellowship with a
publication grant towards the production of this book, which therefore owes its
fin~orm and, to a considerable extent, its very existence to the Trust's
generosity and support.

PAUL G. BAHN
1988
Introduction: Cave Life

Despite persisting popular belief, fuelled by the best efforts of movie-

makers and cartoonists, it is not true that people during the late Ice Age
habitually lived in caves. Over the last hundred years, archaeological
investigation has made it clear that they primarily occupied cave-mouths,
sunny rock-shelters, and huts or tents in the open air. This is hardly
surprising, since caves are not the most pleasant choice of habitat- they are
usually dark, wet, slippery, and full of rocks and jagged concretions.
Nevertheless there are cases , many of them in the French Pyrenees,
where hearths and other signs of occupation have been found far inside
very deep caverns, and even hundreds of metres from the entrance
(assuming that the present entrance was also the main one in prehistory).
What is the explanation for this? In some cases these are temporary
encampments linked to the creation of decoration on the cave-walls : for
example, in the cave ofTito Bustillo in northern Spain, occupation debris
and colouring materials have been found in excavations at the foot of the
principal painted frieze (p. 97).
lr1 '"~ther cases they may denote the seeking of refuge in particularly
harsh conditions: at present the caves of southern France and northern
Spain maintain a fairly constant temperature (usually c l4C ) which
makes them pleasantly cool in summer and mild in winter. However,
recent work by geophysicists has established that during the last Ice Age it
was not merely mild but quite warm inside the caves- a fact which helps to
explain why the Palaeolithic footprints found in their depths are of bare
feet and have no sign of frostbite 1 (but see p. lOS for the possibility of
frostbite on hands).
The footprints are quite often those of children: for example in the cave
of Aldene (Herault), or in Fontanet (Ariege) where a child- who also left
knee- and handprints - seems to have pursued a puppy or a fox into the
cave's depths. Children were clearly not afraid to explore the far depths,
narrow passages and tiny chambers of caverns, whether alone (as at
Aldene) or with adults; it is possible that the finger-holes and heel-marks
found around the clay bison of the Tuc d' Audoubert (Ariege) were made
by children playing while the adult(s) made the figures.
14 IMAGES OF THE ICE AGE

Fig. 3 Prehistoricfioot"'rints
r in F.I,
N iaux (A riege), carefully and th1
deliberately made by two young (A
children . Precise date unknown .
(JV)

It is a tragedy that we have irretrievably lost huge amounts of fascinating

information from cave-floors : in some cases, the caverns were frequented
throughout history, and thus any prehistoric traces on the floors were
destroyed centuries ago; in others, the prehistoric galleries were
discovered during our own century, but the discoverers inadvertently
obliterated the precious evidence because (no doubt with their attention
focused on the decorated cave-walls) they simply did not notice the
footprints, objects and even engravings at their feet . With one important
exception - the Tuc d' Audoubert - it is only in the most recently
discovered caves such as Fontanet and Erberua that all such traces have
been carefully preserved intact.
The Tuc d'Audoubert's inner depths were first explored in 1912; as the
cave is privately owned by the Begouen family, visits have been kept to a
strict minimum ever since, and nothing has been disturbed. Thus the cave
not only has artistic treasures (above all, the unique clay bison figures, as
well as engravings), it is also a treasure-house for all manner of evidence
about the activities of the Palaeolithic visitors: flint tools, teeth or pieces of
bone were carefully placed in rock-crevices, or stuck into the floor;
stalagmites were deliberately broken; cave-bear jaws were picked up, their
canines were removed (presumably for use in necklaces) and then the jaws
were thrown down again; the traces in clay around the bison have already
~en mentioned. In short, since no one visited these galleries between the
last Ice Age and 1912, one can actually follow the traces left by
Magdalenian people and reconstruct many oftheir actions (see Appendix) .
The same applies at Fontanet, a gallery blocked during the Magdalenian
period and only rediscovered in 1972. Here, the back part of the cave has
no traces of occupation but does have numerous prints, as mentioned
above. The front part near the blocked entrance has no prints, but it has
engravings and paintings on the walls, and, on the floor , a series of hearths
around which are the animal bones that the occupants probably tossed
over their shoulders. Here again, the vestiges and prints are so fresh that
 INTRODUCTION: CAVE LIFE 15

Fig. 4 Prehistoric footprints of

three children in the Reseau Clastres
(Ariege). (JV)

one would think they were made a few minutes ago rather than 13,000 or
14,000 years ago. 2
Apart from boulders which may have been used as seats, no cave
'furniture' has survived. Presumably it was all made of wood and has
therefore disintegrated through time- as yet , no one has been fortunate
enough to find a waterlogged Palaeolithic site with preserved wood: only a
few Palaeolithic wooden objects, including a couple of spears, have
survived in Europe and elsewhere .3 But we know that Palaeolithic people
were perfectly capable of working wood- as will be seen in a later chapter,
some of the decorated cave-walls definitely required ladders or
scaffolding, and the actual sockets for scaffolding-beams survive in
Lascaux.
In addition, other forms of evidence show us some improvements and
amenities which were brought to the caves: at Enlene (Ariege), for
16 IMAGES OF THE ICE AGE

example, thousands of small sandstone and limestone slabs were brought

in from local sources and laid down as a kind of pavement; and, as we shall
see, many hundreds of them were engraved. Pollen analysis of sediments
in caves such as Lascaux and Fontanet has revealed that great clumps of
grasses and summer flowers were brought in, presumably for bedding and
seating. 4 So even though few caves were actually inhabited by late Ice Age
people for any length of time , it seems they knew well how to adapt this
environment to their advantage and comfort.

Life in the Upper Palaeolithic could be harsh or pleasant, depending on

climatic and environmental fluctuations. At times there was plenty of food
even in the areas close to glaciers, since periglacial environments are rich
in flora and fauna, and there is no evidence of rickets in skeletons of the
period. On the other hand, some skeletons display evidence of phases of
slow growth caused by starvation or illness. Only a few cases of illness have
been detected, such as hydrocephaly in a child (Rochereil, Dordogne), or
a fungal infection in the old man of Cro-Magnon. Whereas the average
Neanderthaler died aged about 30, and very few of them passed their
mid-40s, Upper Palaeolithic people had an average age at death of 32, and
some were well over SO. 5
This may not seem a great age , but death between 30 and 40 was normal
until only a few centuries ago, and indeed remains so in some parts of the
world. On the whole, Upper Palaeolithic people were fit and healthy; little
evidence has been found for arm or leg fractures, and indeed the only
indication of any violence in this period in Europe (apart from the alleged
'pierced men' depicted in two caves; see p.l52) is a woman from Cro-
Magnon with a deep cut at the front of her skull (a wound to the brain
which she nevertheless survived for some time); seven skulls from the
upper cave of Zhoukoudian in China have enigmatic depressed fractures .6
On the whole, therefore, life was peaceful in the Upper Palaeolithic, as far -
as we can tell.

Since this book is primarily about the art of Franco-Cantabria, a brief

guide to the cultures and their dates is required. The Middle
Palaeolithic (period of Neanderthal people), also called Mousterian
after the site of Le Moustier (Dordogne), was succeeded by the Upper
Palaeolithic (period of Cro-Magnon people) which has been divided
into a series of phases named after different sites in France. The
following is a very rough guide to their dates :
Chatelperronian: starts c 35,000 BC
Aurignacian: c 30,000 BC
.,. Gravettian: c 25,000 BC
Solutrean: c 20,000 BC
Early Magdalenian: c 15,000 BC
Middle Magdalenian: c 13,500 BC
Late Magdalenian: c 10,000 BC
Azilian: c 8,000 BC Fig. 5
Palaeo,
The Magdalenian has also been divided into a sequence of phases (I to Veyrier
VI) which have to be mentioned occasionally in the text, although they pseudo-
are rapidly being abandoned by prehistorians. shown}
Pittartt
 1
The Discovery of
Ice Age Art

The saga of how we discovered and came to terms with the fact that our
remote ancestors could be brilliant artists is a fascinating one, filled with
missed chances, pioneering insights and, above all, the interplay of open
and closed minds which has always dogged archaeology, a subject in which
schools ofthought and personal antagonisms, even today, can play as great
a role as the actual evidence.

Portable objects
The first pieces of art from the Palaeolithic (Old Stone Age) were found in
about 1833, in the Magdalenian cave of Veyrier (Haute-Savoie), near
France's border with Switzerland, where a certain Dr Fran<;ois Mayor of
Geneva encountered a pseudo-harpoon of antler, engraved so as to
resemble a budding plant, and a perforated antler baton decorated with a
simple engraving which may perhaps be a bird (Fig.S); 1 another more
famous baton from the same site, with an engraving of a plant on one side
and an ibex on the other, was found only in 1868. A horse-head, engraved
on a reindeer antler, is said to have been found in 1842 at Neschers (Puy-
de-D6me);2 and a reindeer foot-bone, bearing a fine engraving of two
hinds (Fig.6), was discovered by Joly Leterme (an architect) and Andre
Brouillet (a notary) in the French cave ofChaffaud (Vienne) in 1852. 3

J At this time, however, the study of prehistory was in its infancy, and
there was as yet no concept of an Old Stone Age, let alone of Palaeolithic
art ; thus the Chaffaud engraving, for example, was declared to be Celtic in
date. It was not until some years later that, through comparison with
examples excavated in Palaeolithic occupation layers elsewhere, the
Chaffaud bone was properly identified.
The existence of Palaeolithic art was first established and accepted
5 The first decorated
through the discovery, in the early 1860s, of engraved and carved bones
'llliU!Olm!tcobjects ever found, from
(Haute-Savoie) . The antler and stones in a number of caves and rock-shelters in South-West France,
ISeiUU7-nairpoon, ofwhich half is particularly by Edouard Lartet, a brilliant French scholar funded by
here, is 11 .2 em long. (After Henry Christy, a London banker and industrialist. After finding his first
piece of portable art in 1860 (a bear's head engraved on antler, from the
18 IMAGES OF THE ICE AGE

cave of Massat in the French Pyrenees) Lartet published a drawing of it in Fig. 6 Detail of the reindeer foot-
1861 together with a sketch of the Chaffaud bone; he was thus the first bone from Chaffaud (Vienne),
person publicly to identify the latter as Palaeolithic. 4 showing engraved hinds. Probably
The depictions he encountered in his excavations came as a great Magdalenian . Total length: 13.5 em,
surprise: their quality was astounding, since it had been assumed that height 3. 7 em. (JV)
prehistoric people were primitive savages with no leisure time and no
aesthetic sense. Yet there could be no doubt that the objects were Fi!
authentically ancient - they were, after all, associated with Palaeolithic cez1
stone and bone tools, and the bones oflce Age animals; some of them, such cov
as the magnificent mammoth engraved on a fragment of mammoth ivory hin
from La Madeleine, 5 accurately depicted extinct species while others, such the
as the reindeer from Bruniquel (Fig.17), showed animals which had long
since deserted this part of the world.
The phenomenon was so new and unexpected that museums took some
time to make up their minds about it. For example, when the Vicomte de
Lastic-Saint Jal offered to the Louvre the art objects which he had dug up
at Bruniquel in 1863, the museum hesitated- in part, because of his high
price- and the British Museum snapped them up, subsequently buying
others from this site and elsewhere. By 1867, however, Palaeolithic art was
sufficiently well established for 51 pieces to be included in the 'History of
Work' section of the Universal Exhibition, in Paris.
These first discoveries triggered a kind of 'gold rush' which saw
numerous people excavating- or, more usually, plundering -likely caves
and shelters in search of ancient art treasures. Little attention was paid,
except in a few cases , to the exact stratigraphic position of these finds; they
were dug up like potatoes . At the same time, however, some diggers were
ignoring the art on the cave-walls. In some cases they must have seen it,
but.jbey did not observe it : in Archaeology, one tends to find what one
expects to find, see what one expects to see. In other words, the cave-art
was of no importance in their eyes because it could not be ancient: it was
inconceivable that it might have survived so long, and nothing of the sort
had yet been found . Nevertheless, a few pioneers had noticed cave-art and
wondered ...

Cave-art
Many caves in Europe, as elsewhere, have been frequented since the Ice
Age and throughout history by new occupants, by shepherds, by the
 THE DISCOVERY OF ICE AGE ART 19

curious or the adventurous. In some areas, such as the Basque country,

superstitions and religious traditions associated with caves are probably
extremely ancient, perhaps even extending back to the period of
Palaeolithic art . Certainly there is some evidence that rituals associated
with the art may have survived in some areas until quite recently: in 1458 ,
Pope Calixtus III (a Borgia from Valencia) prohibited religious ceremonies
in 'the cave with the horse pictures'. We do not know which cave he meant,
but from the description it is likely that its pictures dated to the Ice Age .6
Many decorated caves also contain graffiti of different periods- Gargas,
for instance, has engraved marks on its walls which are probably
protohistoric , Gallo-Roman and medieval in date . In some cases , it is
certain that people saw the art: in the Pyrenean cave of Niaux, a certain
Ruben de la Vialle wrote his name in 1660 on the only panel of the 'Salon
Noir' left undecorated by the Magdalenians; later graffiti there are placed
within or between animals, carefully avoiding the painted lines. 7
Seeing the art is one thing; reporting it is quite another, and no cave-art,
as far as we know, was ever mentioned in print before the nineteenth
century.8 This is quite understandable, since prehistory did not 'exist'
until then, and so the pictures had no significance. However, once
prehistoric studies got under way and portable art of the Ice Age had been
discovered and authenticated, a few scholars at last began to notice what
had been staring them in the face for so long.
7 Photomontage of the painted The first definite suspicions of this type can be attributed to Felix
atAltamira (Santander), Garrigou, a scholar who investigated scores of caves in the French
roughly 18m by 9 m. The Pyrenees in the 1860s and who found some pieces of portable art,
on the left is 2.25 m long, while including the famous bear engraved on a pebble, from the cave of Massat.
curled-up bison are 1.5 to 1.8 m During a visit to Niaux in 1864 he was taken to the Salon Noir, destined to
Magdalenian . (JV) be recognised as one of the glories of Ice Age cave-art (Fig.ll9). As
20 IMAGES OF THE ICE AGE

mentioned above , the Salon had often been visited in the past, and the local
guides even called this part of the cave the 'museum'! In his notebook
Garrigou wrote that 'there are paintings on the wall. What can they be?' 9
Alas, he did not make the crucial mental leap, and Niaux therefore
narrowly missed being the first true discovery oflce Age parietal (wall) art
(its figures were not rediscovered until 1906, although a local
schoolteacher called Claustre had mentioned the 'Salle des Betes' ofNiaux
in a note in 1885).
A few years later , in 1870, the archaeologist Jules Oilier de Marichard
wrote in his excavation notebook that he had seen 'signs of the zodiac'
engraved in a deep gallery of the cave ofEbbou (Ardeche), and in 1873 he
added that he had seen animal silhouettes sketched on the walls. 10 But, like
Niaux, Ebbou had to wait a considerable time (until 1946!) before being
rediscovered. In 1879 he wrote a letter to Emile Cartailhac stating that
some caves in Ardeche had red paintings of 'fantastic animals', 11 but
unfortunately he put nothing in print.
One scholar, however , did print something about wall-art: in 1878,
Leopold Chiron, a schoolteacher, noticed deep engravings in the cave of
Chabot (Gard); he not only took photographs and imprints of them, but
also published a note about them, although he could not know their date . 12
He mistakenly thought he could see birds and people among the lines;
unfortunately, the Chabot engravings are difficult to decipher, and the
figures are far from clear .
However, a more decisive event occurred in the same year, for it was at
the Paris Universal Exhibition of 1878 that Don Marcelino Sanz de
Sautuola saw Palaeolithic portable art on exhibit, an experience which led
eventually to his discoveries at Altamira. Unfortunately, whereas the
pictures in Chabot were too modest to make an impact, those of Altamira
were too splendid to be believed.

Altamira
The cave of Altamira , near the north coast of Spain, was found in 1868 by a
hunter, Modesto Cubillas, on freeing his dog which had become wedged
in some rocks. De Sautuola, a local gentleman and landowner, first visited
it in 1876; he noticed some black painted signs on a wall, but thought little
of them. His subsequent trip to the Paris exhibition left him profoundly
impressed and excited by the implements and art objects of the Ice Age on
exhibit there - we know that he was drawn back again and again to those
displays. He was also fortunate enough to meet Edouard Piette, the
greatest prehistorian in France, who was in the process of amassing a
remarkable and unrivalled collection of Palaeolithic portable art; the two
men discussed caves, artefacts and excavation methods. 13
~ck in Spain, de Sautuola began his own excavations in caves, and
eventually returned to Altamira in 1879, when he was 48 years old. As is
well known , during November that year he was digging in the cave-floor,
searching for prehistoric tools and portable art of the sort he had seen in
Paris, while his little daughter Maria (according to her own account) was
'running about in the cavern and playing about here and there ... Suddenly
I made out forms and figure s on the roof' . 14 She exclaimed 'Mira, Papa,
bueyes! [Look, Papa , oxen!]'. 15 She had seen the cluster of great
polychrome bison on the ceiling (Fig. 7), which had lain concealed in total
darkness for 15,000 years, awaiting her.
THE DISCOVERY OF ICE AGE ART 21

Her father was, to say the least, astonished. At first he laughed , then he
became more interested, particularly when he found that the figures were
done with what seemed to be a fatty paste . Finally 'he was so enthusiastic
that he could hardly speak' . 16 He must have noticed at once the close
similarity in style between these huge figures and the small portable
depictions with which he was familiar , but we do not know exactly when
he dared make the deduction that this wall-art was of equal age. He
certainly knew that it had not been done since the cave's discovery in 1868 ,
because no stranger could spend so much time and effort in there (and to
what purpose?) without everyone in the area knowing about it.
In any case, de Sautuola's greatness lies in the fact that he not only made
the crucial deduction, he dared to present his views to the sceptical
academic establishment. First, he wrote to Professor Juan Vilanova y
Piera, Spain's foremost palaeontologist, in Madrid. Vilanova visited him,
dug a little in the cave, and was convinced that de Sautuola was right about
the paintings. As a result of a lecture given about Altamira by Vilanova in
Santander, it made front-page news all over Spain, and tremendous
crowds of people came to see the cave. Even King Alfonso XII visited
Altamira, crawling into the smallest galleries, which were lit by the candles
of his entourage of servants, and leaving his name in candle-black on the
wall near the entrance!
In 1880 de Sautuola published a booklet on his discoveries at Altamira,
including the Ice Age paintings; 17 it was a cautious piece of work in which
he did not affirm the contemporaneity of the painted ceiling and the
Palaeolithic deposits, but merely posed the question . In the same year,
Vilanova presented the discovery at the International Congress of
Anthropology and Prehistoric Archaeology in Lisbon, a gathering of some
of Europe's greatest prehistorians - among them Montelius, Pigorini ,
Virchow, Lubbock, Evans and Cartailhac.
One would think that such a remarkable find, albeit unprecedented and
unexpected, would have met with interest - even excitement - at this
gathering; that there would have been a stampede of specialists to see the
site; that de Sautuola would have been feted and congratulated. It is to
archaeology's undying shame that the very opposite occurred, and that it
helped to bring about his premature death in 1888, a sad and disillusioned
man still under suspicion of fraud or naivety, with his discovery rejected
by most prehistorians. He had taken the hostility personally, as an attack
on his honour and his honesty. 18
There were a number of reasons for the rejection; for a start, he was a
complete unknown, rather than an established prehistorian. Secondly,
nothing remotely similar had been found before; after all, figures
scratched on bone were a very different phenomenon from sophisticated
polychrome paintings. Moreover, the discovery was made in Spain- at
this time, many of the leading prehistorians were French, and South-West
France was the region par excellence for Palaeolithic finds . So it must have
seemed illogical and suspect for a discovery such as Altamira to occur in a
different area.
At the Congress, Vilanova exhibited drawings of the Altamira figures in
the halls; but his presentation was met with incredulity and even an abrupt
and contemptuous dismissal of the very idea- Cartailhac, one of the most
influential French scholars, had been warned by his virulently anti-clerical
friend, Gabriel de Mortillet, that some anti-evolutionist Spanish Jesuits
were going to try to make prehistorians look silly. Where Altamira was
22 IMAGES OF THE ICE AGE

col'terned, therefore, Cartailhac smelt a rat, rather than a bison, and Fig. 8 Photomontage of the painted
walked out of the conference hall in disgust. When writing to de Sautuola panel at M arsoulas (Haute
in 1880 to acknowledge receipt of his booklet, he even informed him that Garonne), showing bison, horses and
his paintings of'aurochs' (wild cattle! ) were unlike the prehistoric animals, barbed signs. Probably
and had differently shaped horns! 19 Magdalenian. Total length: c 5 m.
Nate the bison composed of red dots
It was claimed that the paintings were far too good to be so ancient, and
(length: 87 em) . (JV)
that a modern artist (either duping de Sautuola or conspiring with him)
had faked them. Cartailhac referred to the affair as a 'vulgaire farce de
rapin' (a dauber's vulgar joke) . Yet none of the objectors had seen the
original pictures. Vilanova issued formal invitations to the leading
THE DISCOVERY OF ICE AGE ART 23

prehistorians to visit the cave- but, incredibly, all refused .

In 1881, one Frenchman, the engineer Edouard Harle, did turn up to
examine the art of Altamira; but, as an emissary of Cartailhac and de
Mortillet, he had clearly made up his mind beforehand that the paintings
had been done in the 1870s, between de Sautuola's two visits. His
judgement, published in the most influential journal of prehistory of the
time,Z 0 was based on a number of factors : the pictures were too good to
have been done by prehistoric savages (this despite the quality of the
authenticated portable art! ); they were anatomically inaccurate (here he
expanded Cartailhac's views about the anatomy of aurochs, totally
irrelevant since the Altamira animals are clearly bison! ); the paint looked
too fresh to be ancient, it came away on the finger [sic], some of it had been
applied with a modern paintbrush, and the cave was too humid and the
rock too friable to have preserved art for so long; some of the figures were
on top of stalagmite, while others were covered by only a thin layer of it,
which need not denote great age (this is perfectly true); in addition,
prehistoric artists would have spent long periods of time here , far from
daylight, and the soot from their lamps would have blackened the ceiling,
which was not the case. This last objection was more reasonable, given our
knowledge at the time (no Palaeolithic lamps had yet been authenticated,
although a few had already been found 21); but, as will be seen (p .109),
experiments now suggest that animal fat in lamps gives off no soot.
Cartailhac was pleased to publish Harle's account, since he was irritated
at the persistence of the two Spaniards at different meetings; 22 his own
report of the Lisbon meeting had a few disdainful words about Altamira,
while the official records of the Congress did not mention it at all! Other
anti-Altamira articles began to appear in scientific journals and in the
newspapers, both in France and in Spain. De Sautuola tried to present his
case again at a French Congress in Algiers in 1882, but in vain; he
submitted his booklet and a report to an international Congress in Berlin
the same year, but no discussion was held. Finally, Vilanova gave up the
fight , and de Sautuola stood alone. 23 Neither de Mortillet's book Le
Prehistorique (1883), nor Cartailhac's Les Ages Prehistoriques de
l'Espagne et du Portuga1(1886), nor the International Congress in Paris in
1889 mentioned Altamira at all.
However, it would be wrong to give the impression that all the leading
French prehistorians shared the opinions of Gabriel de Mortillet and
Emile Cartailhac, the two most virulent and powerful voices against
Altamira. For example, Henri Martin (grandfather of the famous
prehistorian of the same name) had to decline an invitation to visit the cave
in 1880, but said in his letter to Vilanova that he saw definite analogies
between the drawings and portable art; 24 and Edouard Piette, who had met
de Sautuola in 1878, was in no doubt about Altamira, since he had always
thought that portable art must have been accompanied by wall-arrZ5 -
indeed he wrote to Cartailhac in 1887 stating that he thought the paintings
were Magdalenian; and in the late 1880s, when he discovered the painted
pebbles of the Mas d' Azil which characterised the 'Azilian culture' at the
very end of the Ice Age, he presented them as the oldest known paintings
in France, matched or surpassed in age only by Altamira. 26 Needless to
say, this discovery too produced howls of rage from the establishment until
its validity was proved. Piette was courageous, imaginative and open-
minded throughout his career, and stands among the true pioneers of early
French prehistory.
24 IMAGES OF THE ICE AGE

Acceptance at last
How, then, did cave-art change from a source of ridicule to a subject
worthy of splendid monographs? It is no coincidence that the
metamorphosis took place in South-West France. After the Altamira
debacle, all was quiet on this academic front, but discoveries were still
being made: the stratigraphic position of Piette's pebbles, for example,
proved that ochre could adhere to rock for millennia.
The real breakthrough - literally - came at the cave of La Mouthe
(Dordogne), where in 1895 the owner decided to remove some of the fill
and exposed an unknown gallery; a group of four youngsters led by Gaston
Berthoumeyrou explored it on 8 April, returning on the 11th with Gaston's
father, Edouard; in the course of one of these visits, they spotted a bison
engraved on one wall, 100 metres inside the cave, as well as other figures;
because of Palaeolithic deposits in the blocking fill, it was clear that the
pictures must be ancient. 27
Emile Riviere - who had just visited Altamira - came to the cave and
carried out some excavations, in the course of which he found more
parietal pictures, and in 1899 unearthed a Palaeolithic lamp here, the first
to be accepted, 28 and still among the finest because of its carved red
sandstone and its engraving of an ibex. So La Mouthe provided proof of
age, and of a lighting system. Cave-art could no longer be denied.
Meanwhile, Chiron had again called attention to his discovery of
engravings in Chabot in 1889 and 1893; while in 1890 he also mentioned
engravings in the nearby Grotte du Figuier (Ardeche) . Moreover, in 1881,
a gentleman named Fran<;ois Daleau had begun digging in the cave of Pair-
non-Pair (Heads-or-Tails) near Bordeaux and found bones of Ice Age
species; in 1883 he first noticed engravings on the cave-walls exposed by
the removal of occupation layers, but he paid little attention to them, apart
from mentioning the fact in his notebook. It was not until 13 years later
that the discoveries at La Mouthe led him to clean the walls with a water-
spray from the vineyards and study the pictures- he drew sketches of the
animal figures in his notebook, and published an article about them. 29
Since some of them had been covered by Gravettian occupation layers, it
was indisputable that the pictures were at least as old as that Upper
Palaeolithic culture.
Daleau was visited by Piette, and also by Cartailhac who admired the
engravings and tried to photograph them. Cartailhac was growing
troubled: during a visit to La Mauthe, he himself had removed some
Palaeolithic sediment and exposed the legs of a painted animal on the wall.
Further crucial revelations awaited him in the Pyrenees. A clergyman,
the abbe David Cau-Durban, had been digging in the Pyrenean cave of
Marsoulas since 1883, where he had found cultural material and portable
art of the Upper Palaeolithic. The cave is small and narrow, and one wall is
fes,tj}Oned with engravings and red and black paintings. The paintings, at
least, are very visible; Cau-Durban had seen them but assumed they were
modern, like the names and dates on the wall, even though some were
partly covered by sediment.
After the discoveries at La Mauthe and Pair-non-Pair, Felix Regnault, a
local scholar, returned to Marsoulas in 1897 and discovered the animal
pictures. But his claims were greeted with hilarity: Cau-Durban mocked
him, saying that he was naive and gullible, and that the pictures had been
done by children; 30 although Regnault mentioned the paintings at a
session of the Societe Archeologique du Midi, stressing the analogy
THE DISCOVERY OF ICE AGE ART 25

between the portable and the wall-art in the cave, Cartailhac, cave-art's old
adversary, refused to publish a note about it in the Society's journal!
Regnault got two lines, with a possible misprint that described the
paintings as historic rather than prehistoric. Invitations were issued in
1898 to see the paintings, but Cartailhac did not come. Riviere, however,
did pay a visit, and found the colours rather fresh - consequently he had
doubts about the age of the paintings, though he accepted the engravings.
He may have been influenced by Cau-Durban and Cartailhac, both of
whom told him they had seen no animal paintings during the excavations.
Ironically, as we shall see, it was a visit to Marsoulas in 1902 which finally
opened Cartailhac's eyes .
In 1901, one of Riviere's diggers, Pomarel, had found engravings in the
cave of Les Combarelles, near Les Eyzies (Dordogne), and told his
teacher, Denis Peyrony, about them. The latter visited the cave with Louis
Capitan and the young Henri Breuil, and found many more pictures. A
few days later, Peyrony found the art in the nearby cave of Font de Gaume.
These discoveries, together with all that had gone before, finally
penetrated Cartailhac's stubbornness. In 1902 he visited Marsoulas and
from there he went to Altamira with Breuil and at last saw its art for himself
-after which he became one of the most enthusiastic scholars of cave-art,
even purchasing Marsoulas in order to ensure its protection!
As mentioned earlier, Piette had written to him about Altamira in 1887;
once cave-art began to be accepted at places such as La Mouthe, he had
again reminded Cartailhac about Altamira, comparing it with La Mouthe.
Yet in later years, Cartailhac claimed that he himself had re-opened the
debate; he boasted that he discovered many engravings at Marsoulas
which Regnault had missed; 31 above all, he published his famous article of
1902, '"Mea culpa" d'un sceptique'/ 2 which is still cited in France as the
epitome of an objective scientist admitting his mistakes. Certainly its title
and its intentions are laudable. However, on close inspection, the paper
contains only limited contrition: indeed, Cartailhac states that his actions
in 1880 were fully justified at the time, and that there is nothing to object to
in Harle's 1881 report. It is clear that Cartailhac changed his mind at the
last minute only when the evidence became overwhelming- it was thanks
to the new discoveries that 'we no longer have any reason to doubt the
antiquity of Altamira'.
Harle revisited Altamira in 1903, having seen Font de Gaume, and
realized his mistake, though somewhat grudgingly; he wrote to Cartailhac:
'it is the Font de Gaume that has made me change my mind . But for that I
would still say the Bison Ceiling is a forgery'. 33
Cartailhac later admitted that he should have expected parietal art to
exist, once portable art had been discovered, and he blamed early
opposition to Altamira on the fact that 'we were blinded by some
dangerous spirit of dogmatism'/ 4 his 'Mea Culpa' of 1902 was reported by
the whole Spanish press with satisfaction. 35 De Sautuola was vindicated .
On 14 August 1902, a number of prehistorians - including Emile
Cartailhac- attending the Montauban Congress of the French Association
for the Advancement of Sciences made an excursion to the Les Eyzies area
and visited the decorated caves there. This marked the official recognition
by science of the existence of cave-art. Had more minds been open and
willing to accept the unexpected, had Chabot been more splendid and
Altamira less so, cave-art studies could have begun two decades earlier.
However , they quickly made up for lost time.
 2
A Worldwide fro
Fi1
en~

Phenomenon Au
&

Mter these first discoveries, examples of portable art and new decorated
caves continued to be found in Europe (as will be seen, below); but for a
long time it was believed that, apart from the material in Siberia, Ice Age
art did not exist elsewhere, and was an exclusively European
phenomenon. Little by little, however, and especially in the last few years,
it has become apparent that, towards the end of the Pleistocene, artistic
activity was underway all over the world. The technical, naturalistic and
aesthetic qualities of European Palaeolithic images remain almost unique
for the moment, but it is still true that, at this period, in other parts of the
world, one can see traces of the same phenomenon.'

TheNew World
Amazingly, the first clue to Pleistocene art in America was found in 1870,
only a few years after Lartet's finds in France were authenticated.
Unfortunately, the object in question was badly published, and
disappeared from 1895 until its rediscovery in 1956! Consequently, very
few works on Palaeolithic art mention it. This mineralised sacrum (base of
spine) of an extinct fossil camelid was found at Tequixquiaq, in the
northern part of the central basin of Mexico. The bone is carved and
engraved- two nostrils have been cut into the end - so as to represent the
head of a pig-like or dog-like animal; the circumstances of its discovery are
unclear, but it is thought to be from a late Pleistocene bone bed, and to be
at l~t 11,000 or 12,000 years old/ it is on exhibit in Mexico's National 0
L_
Museum of Anthropology.
Other examples of portable art in the New World are less well F ig.
authenticated, although a bone with an engraving of a rhinoceros from frag~
Jacob's Cave, Missouri, is thought to be of Pleistocene date. 3 The Pau
mammoth engraved on a fossil shell pendant, found at Holly Oak, 25,0
Delaware, in 1864 (Fig. 9) is still the subject of some doubt regarding both aO
its Pleistocene date and its authenticity. 4
As for parietal art, this is far more difficult to date. The New World has
decorated caves and rock-shelters in many areas, such as southern Peru
 A WORLDWIDE PHENOMENON 27

Tracing of the mammoth

on a piece of a large whelk,
Holly Oak (Delaware).
\ucnemzcuy uncertain. (After Kraft
Thomas)

and Patagonia, 5 which some scholars believe to date back to the

Pleistocene; they may be right, but proof is lacking for the present.
Recently, however, a layer in the decorated rock-shelter of Pedra
Furada, Brazil, which is said to contain a painted fragment fallen from the
wall, has been dated to 17,000 years ago; 6 new methods of dating layers of
varnish formed on rocks are starting to suggest the possibility of very early
origins for rock-carvings in the western part of North America- in eastern
California, some layers formed on top of rock-engravings have been dated
to the eleventh or twelfth millennium before the present. 7

Africa
Once again, it is very difficult to date art in rock-shelters, but it is
extremely likely that some examples in various parts of the continent are of
late Pleistocene age- in Zimbabwe, fragments of painted stone are known
from layers at least 13,000 back to more than 40,000 years old, and
pigment has been in use for at least 125,000 years. 8 Portable Palaeolithic
art has been well authenticated in South West Africa: seven fragments of
stone found by Eric Wendt in the Apollo 11 cave , southern Namibia, have
paint on them, including four or five recognisable animal figures such as a
black rhino and two possible zebras; they display a use oftwo colours, and
were associated with charcoal which has provided a radiocarbon date of at
least 19,000 and perhaps even 26,000 years ago. 9 Engraved pieces of wood
and bone (including a baboon fibula with 29 parallel, incised notches) have
been recovered from Border Cave, Kwazulu, and dated to between 35,000
and 37,500 years ago, while a bored stone decorated with incisions has
been found in a layer of Matupi Cave, Zaire, dating to about 20,000 years
ago. 10

Arabia and India

10 Tracing of an engraved It has been claimed that the oldest rock-art in central Arabia dates back
of ostrich eggshell from some 14,000 years, though the only evidence for this is the apparently
India, dating to at least Pleistocene fauna depicted 11 - the same situation exists in India, where the
years ago. (After Kumar et
hundreds of caves and rock-shelters around Bhimbetka, near Bhopal,
contain parietal paintings spanning a long period: claims have been put
forward that the earliest are Upper Palaeolithic in age, especially since
engraved ostrich eggshells from excavated layers here are said to have been
dated to between 25,000 and 40,000 years ago. 12
28 IMAGES OF THE ICE AGE

The Far East

China's Palaeolithic 'art' is limited , for the moment , to beads and other
decorative objects from the upper cave at Zhoukoudian. However, You
Yuzhu is studying the animal bones from the important site of Shiyu, in
Shanxi Province (NW China), which dates to 28,135 years ago and has
yielded over 20,000 stone tools. He has found several hundred bones with
cut-marks and 'engravings' on them, but so far only one has 'figures' . He
believes that this specimen, a fragment of horse femur , shows an engraving
of an animal approached by two hunters, and a large bird pursued by other
hunters. 13 I have not seen the original bone, but no specialist to whom I
have shown a photograph of the object has been able to see anything
figurative in the marks, and it is not even clear that they were made by a
tool. No doubt the future will bring better examples from China; for the
moment, the problem remains open.
The same applies to Korea where a number of claims have been made
for portable art in the Middle Palaeolithic: these take the form of bones
supposedly modified to depict animals, 14 but serious doubts exist about all
of them. Similarly, primitive pecked rock-carvings in a number of sites are
claimed to be Palaeolithic, in part because they are thought to depict
extinct reindeer and grey deer, and partly because they display
'Palaeolithic mentality' , 15 but, once again, proof is lacking.
Fi
In Japan , on the other hand , there are some extremely interesting frc
engraved pebbles from the cave ofKamikuroiwa. Layer IX (Initial Jomon) ab
has been dated to 12 ,165 years ago, and contained several little pebbles Hi
with engravings on them , some of which seem to represent breasts and
'skirts' (Fig. 11).16 Fil
Mt
Australia Ro
It is in Australia, with its incredible wealth of rock-art, that one finds most
of the non-European examples of Pleistocene parietal art . The first site
where its existence was authenticated was Koonalda Cave, in southern
Australia, which was found to contain abundant 'digital flutings' (lines
made with fingers) on the ceiling and walls , in total darkness, hundreds of
metres inside; they seemed to be associated with the extraction of flint. 17
The site's archaeology showed that the mining activity took place at least
15,000 to 24,000 years ago, and so the marks are probably of similar age.
These finger flutings are identical to those known in several of the
European Palaeolithic caves.
Like those of Koonalda , finger flutings in the Snowy River Cave,
Victoria, can be dated only by assuming that they are contemporaneous
with archaeological deposits at the cave-entrance, which are about 20,000
years old.
~first direct proof of the antiquity of Australian art came from the
Early Man shelter in Queensland, where very weathered and patinated
engravings (circles, grids and intertwined lines) covered the back wall and
disappeared into the archaeological layer (Fig 12). As the layer yielded a
radiocarbon date of 13 ,000 years ago , 18 it is clear that the engravings must
be at least this old: it will be recalled that it was precisely this sort of proof,
at La Mouthe, which finally clinched the authenticity of parietal art in
Europe.
Engravings resembling those of Early Man shelter exist in Tasmania; as
that region became separated from the continent by a rise in sea-level
 A WORLDWIDE PHENOMENON 29

Fig.! I Tracing of'Venus' pebbles

Kamikuroiwa, Japan. Each
5 em long. (After Aikens &

12 Engravings in the Early

shelter, Queensland. (Photo A.

around 12,000 years ago, and since its art contains no engraved dingo
prints (the dingo having arrived in Australia after that date), some scholars
believe that the Tasmanian engravings are older than the separation;
others think they are much younger. In 1986, 16 red hand stencils were
found deep inside the Ballawine Cave in the Maxwell Valley, SW
Tasmania; they are undatable, but, from a comparison of the cave's
archaeological material with that of similar sites in the region, it has been
estimated that they may be at least 14,000 years old. 19 In 1987, hand
stencils and areas of wall smeared with red were found in another
Tasmanian cave, Judd's Cavern. 20
Arnhem Land, in northern Australia, is extremely rich in rock-art;
ochre 'pencils' with traces of wear have been found there in layers dating to
18 ,000 and 19,000 years ago, and perhaps even 30,000. Some ofthe oldest
paintings are covered by a thin siliceous film, which is deposited only in
very arid conditions - the last such period in this region occurring 18,000
years ago. Moreover, among these apparently ancient figures are animals
which have been interpreted as species extinct in Australia for at least that
length of time (eg the marsupial tapir Palorchestes), 21 although some
scholars disagree with these interpretations.
For the moment there is very little portable art of this period: in the cave
of Devil's Lair, Western Australia, three perforated bone beads have been
found in a layer that is 12,000- 15,000 years old, together with a possible
stone pendant, and some slabs of stone with engraved lines on them
(though it is not certain that the lines are artificial). 22
30 IMAGES OF THE ICE AGE

Figs. 13/14 Large, deeply carved

circle with internal lozenge lattice in
entrance of Paroong Cave,
S Australia, and deeply engraved
circles (diameter: between 15 and
45 em) on the opposite wall of the
entrance. (PG B)

Fi
as
no
s.
In NW Australia, in the Pilbara, there are thousands of engravings on
rocks in the open air, and many scholars are starting to think that the
patina on some of them proves that they are extremely ancient, probably
some 10,000-15 ,000 years old; and dating of 'desert varnish' (see New
World, above) covering petroglyphs in the Olary region of South Australia
recently produced a result of over 30,000 years!23
lJ;I.ere are not huge numbers of caves in Australia, and those which do
exist are shunned by modern Aborigines, who are wary of the dark and of
caverns. Consequently, any art found inside a cave is unlikely to be of
recent origin. The discovery of digital flutings at Koonalda was followed
by similar though lesser finds in the small Orchestra Shell Cave (Western
Australia) 24 and in two other caves in the south of the country. Recently,
however, a whole series of decorated caves has been discovered in South
Australia, near Mount Gambier. This region contains hundreds of caves in
its Tertiary limestone karst . Up to now, 140 have been examined; almost
all contain traces of animals (clawmarks on the walls, and so forth), and 22
 A WORLDWIDE PHENOMENON 31

Digital tracings in Karlie-

Cave, S Australia. (PGB)
of them also contain non-figurative marks made by humans on the walls
and ceilings. The discoverer, Geoff Aslin, and his collaborators, Robert
and Elfriede Bednarik, have undertaken an in-depth investigation of what
seems to be one of the most remarkable concentrations of non-figurative
parietal art in the world.
These caves are not like the well-known examples in France and Spain;
many of them have vertical entrances, at ground level in fields. Farmers
have always used them as garbage dumps, throwing in scrap metal,
machines and all kinds of refuse, to prevent livestock falling in;
consequently the descent into the depths can be dangerous, as one
negotiates a tottering mountain of rusting barbed wire and old
refrigerators. This state of affairs, however, has helped to protect the caves
and their art: very few potholers or explorers have entered them, especially
16 Deeply engraved circles
with extraction offlint
Karlie-ngoinpool Cave,
Australia. (PGB)
32 IMAGES OF THE ICE AGE

as they are on private land and often near the farmhouses . These caves are
to be cleared of rubbish and locked, one by one: until then, their exact
location must remain a secret.
It has been established 25 that they contain three traditions of
petroglyphs, stylistically distinct, which are sometimes superimposed on
the same wall, separated by thin layers of carbonate. It is hoped that
radiometric dating (the Uranium/Thorium method) of these deposits will
one day provide an approximate chronology for the traditions .
The earliest phase comprises the digital flutings; all22 caves have them ,
and they always lie beneath the engraved lines and seem to follow or
emphasise the topography of the walls. Like those of Koonalda, Orchestra
Shell Cave and others , they are probably at least 20,000 years old. At that
time, therefore, this tradition of finger-marking extended along the entire
southern part of the continent, a distance of 3,000 km!
The second phase, also very ancient, is called Karake (after the cave
where such traces were fir st seen); it is characterised by deeply engraved
and weathered circles- simple, concentric, or divided up by gouged lines
- which closely resemble the archaic petroglyphs of Early Man shelter
(13,000 years old) and the Tasmanian engravings. In other words, this
tradition seems to have extended down the eastern side of the continent.
Finally, the terminal phase, comprising shallow engravings, is probably
much more recent , dating to the Holocene period.
It is noteworthy that several of these caves, like Koonalda, served as
chert mines at the time of their decoration; the region around Mount
Gambier was already known to archaeologists for the great abundance of
Pleistocene stone tools and knapping-waste found there, around the good-
quality surface sources of stone. The decorated cave-mines are thus only
the nucleus of a considerable activity area.
Two of the caves in particular are outstanding. Paroong has an
important collection of Karake-type petroglyphs, especially at its
entrance: the vertical walls here are literally covered with a profusion of
deeply carved circles, one next to another, up to a height of 4 m above the
present floor (Figs . 13 ,14).
The cave of Karlie-ngoinpool (which means 'many' in the old local
indigenous language) contains one of the greatest known concentrations of
non-figurative parietal art. Its well-preserved finger flutings cover at least
75 square metres (Fig. 15). There are also quantities of deeply engraved
circles and other Karake motifs, and some of them (covering 10m width of
wall) remain hidden behind the mountain of rubbish in the entrance.
Many of these circles are closely associated with the extraction of nodules
of chert of very poor quality (Fig . 16). The cave ends in a little 'sanctuary' ,
a tiny chamber in total darkness and covered in engravings , especially
circles, which appears to have been of particular significance for the people
of Wlois remote period.

Clearly, therefore , the map of Pleistocene artistic activity is rapidly filling

up, with Australia providing a number of well-dated examples . It is
probable that both that country and many others will have innumerable
further surprises of this type in store for us . Nevertheless, Europe remains
supreme, for the present , in the quantity and quality of its surviving
Palaeolithic art.
A WORLDWIDE PHENOMENON 33

Europe
The Palaeolithic pictures of Europe are generally treated as two distinct
entities, the portable and the parietal- whereas in reality these are merely
the two ends of a continuous range; in other words, there is an overlap
between the two categories, comprising cases in which it is impossible to
decide whether detached fragments of wall were decorated before or after
falling, and blocks which could be moved but were too large to be carried
around (Breuil called them parois mobiles, or movable walls): at the
French rock-shelter of LaMarche, for example, the size of the engraved
slabs varies from specimens a few centimetres long to some over a square
metre in area and weighing tens of kilograms. However, for the sake of
convenience, the usual artificial division will be retained here.

Portable
The two categories have somewhat different distributions within Europe .
Portable art is found from the Iberian Peninsula and North Africa to
Siberia/6 and has notable concentrations in Central and Eastern Europe;
occasional specimens also turn up in countries around the fringes of
Europe, such as the crude engraving of a possible animal on a limestone
slab from the Aurignacian site ofHayonim Cave, Israel, 27 or the handful of
Upper Palaeolithic engraved bones from England. 28
It is very hard to quantify portable art objects, since many are broken,
and unknown numbers of them remain unpublished, whether in private or
clandestine collections, or lying forgotten and unstudied in museums
around the world. 29 One estimate in 198030 that there are well over 10,000
pieces of Palaeolithic portable art in Western Europe alone is certainly a
minimal figure. A 1965 study31 of the portable art from sites in the Perigord
region of France produced a total of 2,329 objects- once again a minimal
figure, since there are many more in foreign or private museums and
collections.
The scattering of these objects around the museums of the world poses
great problems for their study, which will be alleviated only when good
casts of the 'absentees' can be housed in some European research centres,
and when full details of every specimen are available from the
computerised data banks which are now being compiled. 32
On the other hand, the uneven distribution of the objects among
archaeological sites is a stimulating puzzle for the archaeologist. Many
Palaeolithic sites in Europe have no decorated objects at all, while others
have only one or a few. Some Magdalenian 'supersites', however, have
hundreds: the cave of Parpal16, in eastern Spain, produced over 5,000
plaques of engraved and painted stone, together with scores of decorated
bones, from its long Upper Palaeolithic occupation, and especially the
Magdalenian. 33 The site of La Marche, already mentioned above, has at
least 1,512 engraved slabs of stone (1 ,268 after some fragments were fitted
together), weighing a total of 4 tons! 34 The open-air site of Gonnersdorf in
West Germany has about 500 engraved slates .35 The cave ofEnlene, in the
French Pyrenees, has recently yielded over 1,000 engraved stones, with
more being found every year, in addition to its wealth of decorated bone
and antler 36 - in some areas of the cave floor , over 100 engraved slabs have
been found in less than a square metre!
Clearly, therefore, some sites are infinitely more important than others
in terms of decorated objects. This becomes particularly clear from the
34 IMAGES OF THE ICE AGE

above-mentioned survey of Perigord specimens: the 2,329 items were

sprMtd over 72 sites; but two rock-shelters alone, La Madeleine (582
objects) and Laugerie-Basse (560), account for 1,142, or 48% of the total; if
one adds Limeuil (176) and Rochereil (132), then only four sites account
for 1,450 objects, or 66% of the total. Similar figures would emerge from
studies of this kind in the Pyrenees or Spain and, as we shall now see,
parietal art's distribution is equally patchy.

Parietal
Cave-art is difficult to quantify, both in numbers of cav~<tnQ in numbers
of pictures. The most striking fact, however, is that its geographical
 A WORLDWIDE PHENOMENON 35

distribution in Europe is very different from that of portable art, although

it is most abundant in areas which are also rich in decorated objects: the
Perigord, the French Pyrenees and Cantabrian Spain.
Palaeolithic decorated caves are found from Portugal and the very south
of Spain up to the north of France: the most northerly known at present is
that of Gouy, just south-east of Rouen . 37 None has yet been found in
Belgium or Britain/8 though that may change in the future. They do not
extend to the east of France, except along the Mediterranean through
Sicily and Italy to western Yugoslavia and Romania .
The first problem in quantifying decorated caves is that they cannot
usually be dated directly (see below, p.54) and therefore many examples
remain doubtful, especially where they contain only non-figurative
designs. For example, claims appear sporadically for parietal art in
countries such as Hungary and Anatolia, but all are lacking in proof at
present . For a while, an engraving of a stag in the Bavarian cave of
Schulerloch was thought to be Palaeolithic/9 but , together with another
German candidate, it is now thought to be of more recent date. 40
The Soviet Union has a number ofthese doubtful cases from Georgia to
Fig. 17 Ivory carving depicting two Siberia , in caves, rock-shelters and on open-air cliff-faces; they comprise
, perhaps a male following a both engravings and paintings, and both realistic and stylised animals and
from Bruniquel (Tam et geometric motifs. 41 Their possible antiquity received a boost in 1959 when
. Probably Magdalenian . what seem to be genuinely Palaeolithic paintings were discovered in the
20.7 em. (JV) great cave ofKapova (or Kapovaya, or Shulgan-Tash) in the southern Ural
Mountains; the red figures include eight mammoths, two rhinos and two
18 Antler spear-thrower horses .42 The cave oflgnatiev, 100 km distant in the southern Urals, has
in the fonn ofa mammoth, also been accepted as having Palaeolithic parietal art, including an outline
Bruniquel (Tarn et Garonne). drawing of a horse. 43
trunk joins the legs while the The fact that these caves are 4,000 km away from the main clusters of
tm<T.nnrl1<. The ears are
decorated caves in western Europe serves to highlight the difference in
near the base of the tusks,
eyes are absent (the hole is distribution between portable and parietal art. No decorated caves have
to represent the eye since it is yet been found through all of Central Europe, between north-east France
correct position). Probably and the Urals, despite an abundance of portable art across this void. Of
MlltiUllei:!Ul.1n. Length: 12.4 em. course, cave-art can occur only in areas with caves (though, as we shall see
below, p.llO, examples of open-air rock-art are now being found); but .
Central Europe has plenty of suitable sites. Some theories about the
discrepancy will be examined later.
It should not be thought, however, that all decorated caves in western
Europe are well authenticated; apart from occasional suspected fakes (also
a problem with portable art), there are many cases in which reported
examples of cave art have never been rediscovered; have been proved to be
natural shapes, cracks or mineral colours in the rock (lusus naturae); have
been deemed post-Palaeolithic; or simply remain doubtful. 44 This means
that no two tallies of decorated caves are likely to be identical: in the most
recent authoritative survey of French cave-art, no fewer than 42 examples
were considered erroneous or very doubtful in the Aquitaine region
alone. 45 Another problem with quantification is that some caves which are
now separate may have formed a single system in the Upper Palaeolithic;
while other caves have been given several names which become confused
in the literature, so that the same site acquires two or three different entries
in the list!
According to the most recent survey (1984), there are 130 accepted and
authenticated examples of Palaeolithic parietal art in France, 46 although at
least seven more have been discovered since then; there are 106 examples
 mls

kms

PORTUGAL SPAIN
N
I Gouy Escoural 83 Ojo Guareiia 95 Malalmuerzo
2 Grotte du Renard 2 Mazouco 84 Penches 96 Ardales
3 La justice 85 Atapuerca 97 Pileta
4 Abri du Cheval 86 Domingo Garcia 98 Las Palomas
5 Le Croc-Marin 87 La Griega 99 El Toro
6 Mayenne-Sciences 88 El Reguerillo 100 La Cala
7 Grotte du Cheval (Arcy) 89 Los Casares 101 Navarro!V
8 Angles (Roc aux Sorciers) 90 La Hoz 102 Nerja
9 Blanchard 91 Moleta de Cartagena 103 Taverna
92 Cova Fosca 104 Fuente del Trucho
93 El Nino 105 Forcon
94 Morr6n I 06 Maltravieso

'
I e 0
: Oviedo
IL ____________ ___ -----
---'!..
83 84 Pamplona 0 II -......, ...._
~ """-.r ,,
BurgosO 85 L ______________ y_

105

104
86
87
Segovia 0

Caceres
0
106
s
 Map of Europe showing approximate locations of parietal art sites

Kapova & Ignatiev

\v -~
- ,.._,

I
/
(
)
(
l

I
I
I
/
~N
)

Cudubt'
~

YUGOSLAVIA

I Addaura
2 Armetta
3 Balzi Rossi Group
4 Buco della Sabia di Civate
5 Giglio
6 Giumente
7 lsolidda group
8 Levanzo
9 Miceli
10 Santa Rosalia group
II Niscemi
12 Paglicci
13 Pizzo Mulena
14 Puntali
IS Roca Rumeni
16 Romanelli
17 Romito
18 Sallinella
19 San Teodoro
20 Vaccari
21 Za Minica VII
22 Badanj
 Riade ~ "

I
3
6
5
7
4
..
Cantabrian Spain

SPAIN
I San Roman de Candamo 13 Los Azules (?) 25 Quintana! 37 Porquerizo 49 La Pila 61 San Carlos T
2 La Viiia 14 Les Pedroses 26 Las Herrerias 38 T raslacueva 50 Altamira 62 Peiia del Perro 7<
3 Los Murcielagos 15 La Lloseta 27 T ebellin 39 C hufin 51 El Castillo 63 El Otero 75
4 Entrefoces 16 Tito Bustillo 28 Covar6n 40 La Meaza 52 Las Monedas 64 Los Emboscados 76
5 Entrecueves 17 La Cuevona 29 Mazaculos 41 Micol6n 53 Las Chimeneas 65 El Patatal 77
6 La Lluera I 18 San Antonio 30 El Pindal 42 Las Aguas :54 La Pasiega 66 Co brantes
7 La Ll uera II 19 Sidr6n 31 T rauno 43 El Linar 55 La Flecha 67 Covalanas
8 Godulfo 20 Coverizas 32 Coimbre 44 El Perro 56 Homos de Ia Peiia 68 La Haza
9 LasMestas 21 Samoreli 33 Llonin 45 La C!otilde 57 E!Pendo 69 Cullalvera F
10 E!Conde 22 La Riera 34 Los Canes 46 Cud6n 58 Santian 70 Sotarriza 10
II Oscura d e Ania 23 C ueto de Ia Min a 35 La Loja 47 La Estaci6n 59 El Juyo 71 Covanegra 11
12 E!Buxu 24 Balmori 36 Fuente del Salin 48 Las Brujas 60 Salitre 72 El Cuco 12
13
14
,.. Dordogne
66 -----63 15
64
I 16
I 65
I
17
18 j
'
t"'
' I 19 !
20 I
I 21 f
rl OLimoges 39 22(
Angouleme 0 11 1
; J.-.. 54
23
24
F
l
,..,131.....,.. . - . . 55 25 R
10 1 29~ 40

S6es7 26 L
,-r
el30
e
\-,
l...,
38

58 27 L
( 14 131 IS ) 41 28 G
I 16 , 29 c
_,) !7 /1 OTulle /' 30 Lt
rJ Perig ueux { 83 31 Le
0
--,~I 134 ., 36 ( 84. 85
32M
33 Le
23 ~ 37 ,

r---- 1
24
',
>--- ,..- ~
<) 87 86
88
34 Ro
35 Sai
I 25 ) 30 I ,, I "\ IV
89 36 Pu
L--2 -.-'. (
6 135
e31 \ // -'\ I 37 Le
27 .. 33 I I 1 I 38 Bar
28J 32 34 I I I I 39 La .
1- 29 133. '"_.,__/ \( 40 Sai;
RhOne 92
," 128. 35 ', ' - - - - = - - - - - - - - + 41
Mu:
\. 0 Lau 42
\ r ., 43 Lau
Mende
I ~ 44 Orei
0 Rodez I -45 LeF

80
1...
.,_7 46 Lart
47 Croz
J 48 Patat

0
82: 81 """
96
.... 1
49 LaM
97 ; 50 Cour
Montauban ./
/
51 Font
NimesO 52 Com i
,..._J 53 Comt
I 54 La Ca
I
r""..J ,..._,., 55 Nanc1
I 0 56 Bemii
'-- ...\ Auch I
57 Le Bis
\

126 ,,
I I
-"""-v,_ ___ ,/
58 Sous-(
59 Comar
60 Lausse
I
0 Pau ,... 100 99 61 Cap Bl
123 ) 62 La Gre
II Carcassonne 53 Blanch
1'1.._ 124 1 22 121 I 120 117 liS 114 Ill S4 Castant
',____ // 112. \ S5 Labam
116 113 F . I S6 Reverd,
..._-- - " 119 118 108 0 OIX < i7 Le Roc
\ 110 107 132 \ i8 La Bigo
' <'----... 106 ..-- ,-' i9 Marcen.
'\ { ""'-.., 105 103 \.__
_J
, _ ...............\ ..i"'-.... "Y--, 104 '
1
'0 PechMt
'I Pergous
Regional maps of France and Spain showing \,-'- _....r-
102 0
t-j---.-'-.,---+--r-"
t2 Les Faw
1'TI-_.__3 Grotte 0
101
approximate locations ofparietal art sites 0 SO krns
 A WORLDWIDE PHENOMENON 39

in Spain, 47 up to 21 caves/groups in Sicily/ltaly, 48 2 sites in Portugal

(Escoural and Mazouco), 1 in Yugoslavia (Badanj), 49 and 1 in Romania
(Cuciulat). 50 As has already been mentioned, their distribution within
countries is very uneven, with important clusters occurring both in the
Perigord (48 sites), in the French Pyrenees (29 sites) and in Cantabrian
Spain (82 sites from Asturias to NavarreV '
Moreover, just as portable art varies markedly in quantity from site to
site, the same is true of parietal images. As mentioned above, it is difficult
to quantify the drawings : this is because of superimpositions,
deterioration and, above all , a lack of consensus about how to count 'signs'
78 Goikolau
-should a series of dots be considered separately or as a unit? And how can
79 Atxuri one quantify a mass of meandering fingermarks? Consequently, different
80 Altxerri
81 Ekain authors often have widely divergent totals of figures for the same cave.
82 Alkerdi
Nevertheless, it is clear that there are some sites with few figures and
others with hundreds . For example, the Grotte du Roc at Saint-Andre-
d' Alias (Dordogne), or the Cueva de los Murcielagos at Portazgo and the
74 Grone Carriot
75 LeMoulin Abrigo de Godulfo at Bercio (both in Asturias) have one figure apiece; 52
76 Le Cuzoul-de-Melanie
77 Cantal others, like the Grotte de Pradieres (Ariege) or the Cova Bastera (Pyrenees
78
79
Le Papetier
Cassegros
Orientales) have only a few red marks on their walls; 53 while the Grone du
80
81
La Magdeleine
Le Travers de Janoye
Cheval at Foix (Ariege) or the Abri du Poisson (Dordogne) have a mere
82 Mayriere Superieure handful of figures .54
IIII-\l'mnam-u<-ru1flere 83 Deroc
84 Mezelet On the other hand, Lascaux (Dordogne) has about 600 paintings and
85
86
Vacheresse
Bouchon nearly 1,500 engravings; 55 the Grotte des Trois Freres (Ariege) has over
87
88
Ebbou
Le Colombier I
1,100 parietal figures, 56 more than any other Pyrenean site- indeed, since
89 Le Colombier II that figure was calculated, whole new areas of engraving have been found
90 Huchard
91 Tete-du-Lion by careful removal of clay from the walls, and more remain covered for the
Saint-Front 92 C habot
93 Le Figuier moment; 57 some Spanish caves, such as Castillo and Altamira, are equally
94 Sombre
dominant in their region. It is very hard to see how caves with one figure
95 Grone des Deux Ouvertures
96
97
La Roque
La Baume-Latrone
and supersites like these could be equivalent in any way; similarly, some
98 Grone Bayol caves are huge while others are tiny, though size does not always equate
99 L'Alden e
I 00 Grone Gaze! with numbers of figures - there are large caves with few, and small caves
!01 Cove Bastera
102 Campiime (Fornols Haut) with many.
!03
104
Fontanet
Reseau Clastres
In an exercise similar to that on the Perigord's portable art, the 'graphic
105 Niaux units' of the parietal sites in Ariege (and part of Haute Garonne) have been
!06 Pradieres
I 07 Bedeilhac quantified: the total of over 2,600 'graphic units' is spread over 11 caves,
!08 Le Cheval
109 te Porte! but four of them (Trois Freres, Niaux, Fontanet, Marsoulas) account for
110
Ill
Massat
Le Mas d'Azil 2,216 units, or 84.9%, 58 with Trois Freres predominant of course (over
112
113
Les Trois Freres
Le Tuc d 'Audoubert
50%).
114 Marsoulas
II 5 Montconfort
116 Ganties Montespan Both types of art
117 Gargas
118 Tibiran Thus we have some sites with huge quantities of portable art, and some
119 Labastide
with similar concentrations of parietal images. Do the two phenomena
120 Bois du Cam et
121
122
Sainte-Colome
Etxeberri
ever coincide? Surprisingly, in those regions of Europe where the two
"flli;23 Sasiziloaga K o-Karbi a types of art are found, there are relatively few sites which have both (and
124 Sinhikole Ko-Karbia
125 Oxocelh aya-Hariztoya many which have neither form! ): a recent study ofthis topic, focusing on
126 Erberua
127 Isturitz figurative art, found only 27 sites in France, 9 in Spain and 3 in Itall9 -
128
129
Le Cuzoul-des-Brasconies
Fronsac
and some of these had only one portabl~ item, while others had over 100.
130 Font Bargeix Some local areas rich in one form are noticeably poor in the other.
131 La Croix
132 Les Eglises It seems reasonable to suppose that caves which have both art-types,
133 Les Escabasses
134 La Sudrie with one or both in large quantities, were very special places in the last Ice
135 Le Pigeonnier
136 Oulen
Age: one can certainly include the great river tunnel of Le Mas d' Azil
137 La Foret (Ariege) and Isturitz with its three decorated tunnel-caves - these are the
true 'supersites' of the Pyrenees, with their huge concentrations of tool-
40 IMAGES OF THE ICE AGE

production and artistic activity; they must have served as storehouses,

meeting-places, ritual foci and socio-economic centres not only for local
groups but also for a far wider area, as is confirmed by their artefactual
links with far-flung sites and with each other. 60 The same is true, though
on a lesser scale, of Altamira and Castillo. 61
However, perhaps the most dramatic example can be seen in the Volp
caves: Enlene (which, as we have seen, has huge amounts of portable art) is
connected by a passage to Trois Freres, the richest parietal site in the
whole region. This cave-system, therefore, was clearly of the greatest
importance in the Magdalenian period, an importance further underlined
by the immediate proximity of the remarkable Tuc d'Audoubert (see
p.94).
Enlene was inhabited; it has hearths , but no wall art. Trois Freres is
profusely decorated, but has almost no trace of habitation . These caves
thus suggested to early scholars of Palaeolithic art that living sites were
never decorated, and 'sanctuaries' were never inhabited. Their view
seemed to be further strengthened by the case ofNiaux and La Vache, two
caves facing each other across a narrow valley in Ariege. Niaux is rich in
wall-art but, as far as we know, was not inhabited; La Vache is a living site,
rich in portable art but with no Palaeolithic wall decoration. However, it is
now known that some decorated sites were inhabited (eg Lascaux,
Fontanet, Marsoulas), though in some cases it may still be true that a
decorated site was visited rarely, or only once, for whatever rites were
carried out there.
As will be discussed later (p.ll3), there are clear regional and
chronological differences in the techniques and content of both portable
and parietal art in Europe; taken as a whole, however, this phenomenon
now comprises about 275 decorated sites and many thousands of objects.
More are being found every year- the objects mostly by archaeologists,
and the caves (an average of about one per year) primarily by spelaeologists
and quarrymen. It may seem a lot but, seen against the 25,000 years of the
Upper Palaeolithic, or even just the few millennia of the Magdalenian, the
known quantity of Palaeolithic art actually remains quite small. One might
say that as a medium it is rare, but often well done ...
 3
Making a Record

Tracing and copying

Since the first discoveries and the acceptance of Palaeolithic art, a great
deal of effort has gone into copying and recording it in one way or another.
This is primarily to make reproductions available for scholars to work with
(many examples need to be studied for research or synthesis), and to
present the material to the public at large. Copying has the additional
advantage of reducing the number of occasions on which an original object
might need to be handled or a cave visited - and in a few cases even
specialists cannot visit the caves at present: the Reseau Clastres at Niaux is
closed off by an underground lake and kept sealed in this way so that
instruments inside it can send out data on the cave's undisturbed 'climate';
while Erberua (Pyrenees Atlantiques) lies at the end of a long 'siphon', and
can thus be visited only by experienced divers with modern equipment.
Fig. 19 The young Henri Breuil Finally, of course, the making of copies ensures that any example of
(1877-1961). (Photo PGB Palaeolithic art which might deteriorate or be lost, stolen or damaged will
at least survive in reproduction: for example, one of the Laussel 'Venus'
bas-reliefs is thought to have been destroyed in Berlin during the last war,
and survives only in the form of casts and pictures.' Photographs and
tracings made over the last 100 years also enable us to monitor changes to
the original images through time. 2

Breuil
The doyen of recorders of the art was the abbe Henri Breuil, who was a
young man at the turn of the century when the first decorated caves were
authenticated (Fig. 19): he was a talented artist, excelling in animal
figures, and had been employed by Edouard Piette to draw his remarkable
collection of Palaeolithic portable art. 3 He therefore had the great good
fortune to be present- young and vigorous, with time on his hands and
possessing the necessary skills- at the very time when the copying of cave-
art needed to begin. His first attempt entailed tracing a few figures in La
Mouthe in 1900;4 subsequently, he did Altamira in 1902, Marsoulas, Font
42 IMAGES OF THE ICE AGE

de Gaume, Combarelles, and all the other early finds- indeed he himself
found many caves or figures .
His workingmethods were very crude by modern standards, though he
was hampered by the limited materials then in existence. Sheets of florist's
or rice paper (the most transparent available, though merely translucent at
best) were held on to the cave-wall, often by assistants, and the Palaeolithic
lines beneath were traced with pencil or crayon. Carbide lamps had to be
held, often at arm's length for long periods, by helpers; Breuil was quick-
tempered, and might administer a sharp slap if a boy's fatigue or cramp
caused the light or the paper to move; 5 he was utterly absorbed in the work,
hardly speaking for hours on end, simply moving the arms of his 'human
candelabras' when necessary. 6
Later the tracings would be redrawn for publication, though not always
immediately - Breuil did so much copying, and had so many other
commitments, that publication sometimes occurred many years later (20
years in the case of Trois Fn3res, 24 for the Galerie Vidal at Bedeilhac, 26
for Pair-non-Pair), and the redrawing may have been done by a
collaborator (eg by Michaut in the case of the Galerie Vidal), and then
published unchecked. This inevitably led to mistakes: for example, his
published tracings of Pair-non-Pair contain numerous meanders and
circular images, but recent investigation has shown that these are ali
natural accidents in the rock-face : fissures, bumps and hollows; he
undoubtedly noted them as such on his original tracings, but when the
time came for publication the meaning of his graphic conventions had
been forgotten. 7
The 'direct tracing' method had its disadvantages, since contact with the
wall inevitably damaged the art very slightly in some places (as can be seen
in some modern macrophotographs), and Breuil was even known on rare
occasions to dust surfaces with a few flicks of his handkerchief to make
some lines clearer! 8 Even fragile engravings in the clay of cave-floors, such
as those at Niaux, were traced by the direct method, running a pencil
without pressure on to soft paper in contact with the clay. 9
At Altamira, his working conditions in 1902 were appalling: since the
ceiling figures were done in a very pasty paint, his paper could not be
placed on them, as this would detach pigment. Consequently, he had to
copy the animals while lying supine on sacks filled with ferns, using the
imperfect yellow light of candles which spattered his clothing with wax,
and undergoing constant fatigue and strain. His method in this cave was to
make rough sketches, then measure the originals, and finally make his
copies. He had brought a box of water-colour paints with him, but the
cave's atmosphere was too humid for the paper to dry, so he used pastels
instead; having no black (a colour of great importance in the cave), he had
to improvise with burnt wood and crushed charcoal mixed with water, but
th~esults were unsatisfactory; he returned to Altamira in 1932 to do the
job properly, redrawing a few figures and retouching the old copies of the
rest. 10 Not only was electric light now available; the floor had been
lowered, which meant that he could be comfortably seated and could see
the figures with less distortion and foreshortening than had been the case
in 1902. It is worth noting that a few figures had already shown marked
deterioration in the 30 intervening years.
Students of Palaeolithic art owe a huge debt of gratitude to Breuil for his
patient toil in the caves - by his own reckoning he spent over 700 days
deciphering and copying their art during the course of his long life. There
 MAKING A RECORD 43

20 'Spot the Breuil': his are a number of cases where paintings have faded since he copied them (eg
and a more recent tracing of the Galerie Vidal in the cave ofBedeilhac), so that only the Breuil versions
ibex engraving from Castillo give one any idea as to how the originals may have looked. 11 But a major
1(~12nta~rutt~): about 40 em long. problem - which has come to the fore only in the last few decades - is that
Almagro) all his copies are indeed a 'Breuil version' . His colossal output (including
his drawings of hundreds of portable items), his influence , 12 and his sheer
dominance (not only in Palaeolithic art but in all prehistory) until his death
in 1961, have ensured that Breuil copies are to be found in most textbooks
and in most works on cave-art; in many cases they are far more familiar to
us than the originals , and this is a very dangerous state of affairs because,
like every artist, Breuil had a style of his own . We are therefore seeing
Palaeolithic figures that have passed through a standard 'Breuil process':
they are subjective copies, not faithful facsimiles. It is probable that a
21 'Spot the Breuil': his scholar of Ice Age art, shown six versions of the same figure by different
and a more recent tracing of a prehistorians, could 'spot the Breuil' at a glance (Figs. 20,21 ).
engraving from Trois Freres There are three important consequences of this state of affairs. First, the
27.5 em long. (After Breuil style has inevitably made a wide range of Palaeolithic figures seem
& Clottes) somewhat similar to our eyes; it may therefore exaggerate our view of the
 44 IMAGES OF THE ICE AGE

art's unity through space and time (and, conversely, comparison of figures
produced by two different copiers may exaggerate the differences, since
their own styles enhance the contrast). 13
Second, Breuil's tracings are so well known and so firmly established
that they can hamper new readings of a figure, since they influence what
the researcher sees. 14
Third, Breuil's copies reflect his own preoccupations, his own notions
about the art. As will be seen (p.ISI), he believed in ideas of hunting
magic: this was a hunting art, and he therefore went all out to decipher
animal outlines, sometimes filling in missing elements, According to some
accounts, he was fairly objective in his copying: while working in Trois
Freres one day, he was asked the meaning of the particular line he was
tracing; 'I don't know,' he replied, 'I copy what there is, we'll explain it
later ... if it's possible'. 15
Nevertheless, it is clear from his books and articles that, despite this
attitude to his task, he was in fact extremely selective in the lines which he
chose to copy or at least 'extract' for publication. Where lines seemed to
have no relevance to the animal figures, he often ignored them, dismissing
them as 'traits parasites'. As a result, we have lots of nice clean tracings of
animals, copied from book to book and giving the impression that
Palaeolithic art contains nothing else; whereas in fact there is a vast amount
of non-figurative, abstract or geometric marking in the caves which may
have been just as important , if not more important, to prehistoric people.

Modern methods of copying

Since Breuil was the first to do this kind of work, he had to learn as he went
along; and in view of this lack of precedent, and the materials available to
him, it is scarcely surprising that he made mistakes in his methods and his
results; indeed the wonder is that he did not make far more, and modern
scholars are unanimous in their admiration for the tremendous quality and
perception of his work.
Today, conditions have greatly improved, not only in terms of light-
1
sources, but also of the truly transparent, supple plastics and acetates
available to draw on, and new types of pens and markers. Nevertheless,
direct tracing is now (theoretically) taboo, since we are more aware of the
damage it can cause. Instead , 'tracing at a distance' is done, with the sheet
set up in front of the wall; and some new techniques are also available: in
the past, moulds made of parietal engravings (as at Pair-non-Pair) had to
be of plaster or clay, and the risk to the originals was enormous. Rubber
latex was a significant improvement, but today casts can safely be made of
engravings on hard surfaces using elastomer silicones and polyesters
which not only cause no damage but also are quick and easy to apply, and
p~duce far more precise and resistant results.
16

Indeed, the results are entirely faithful, exact replicas of the original in
size and volume . Moulds have the added advantage that they turn parietal
art into portable; they can be removed for study elsewhere under different
lighting systems, and can eventually be displayed to the public. Casts can
even be treated so that the engraved lines stand out far more clearly than on
the wall: water mixed with ink is spread over the cast; when it is wiped
away, some remains inside the engraved lines. This extra clarity has
enabled researchers to find no fewer than 25 new engraved figures on a cast
from Trois Freres cave. 17
 MAKING A RECORD 45

Fig. 22 Tracing of the plasticine

imprint of an engraved horsehead
from LaMarche (Vienne).
Magdalenian. Length from muzzle to
ear: 6.3 em. Note the probable
halter. (After Pales & de St Pereuse)

A similar technique has been applied to portable engravings such as

those of La Marche and La Colombiere. Plasticine or silicone imprints of
the engraved surface turn the incisions into raised lines , and in some cases
one can see in which order they were made. The imprints are also easier on
the eye, since they do not have the distracting nuances of colour and
texture of the originals. 18
The study of portable art has also made great strides through use of the
microscope, particularly in the pioneering work of Alexander Marshack; 19
this new methodology has revealed hitherto unnoticed details of content
and composition and, as we shall see (p. 75), can even suggest exactly how
the different marks were made. A recent approach, combining casts and
microscopes, involves making varnish replicas of the engraved surfaces of
pebbles; these replicas, unlike the stones themselves, can then be studied
under the Scanning Electron Microscope, at enormous magnifications;
this method not only provides data on the engraving technique (seep. 76),
but can also show in what order superimposed lines have been engraved .20

Photography
All these new techniques are used in close association with photography.
Thanks to the dominance of Breuil's tracings, little attention was paid to
photographs in the field of Palaeolithic art until comparatively recently,
despite some excellent early efforts: pictures were taken of the Altamira
figures in 1880 using electric light; from October 1912 onwards, Max
Begouen was achieving extraordinary results in the Volp caves. He took
the voluminous family camera to the farthest depths of the caverns; its
glass negatives required poses of an hour, or quantities of magnesium, and
were developed on the spot with water from the cave! He worked on them
to produce maximum contrast, and the results thus give the false
impression that the engravings have lost visibility since that time. 21
Nowadays, thanks to the major advances in cameras, films and lighting,
a wide range of techniques has been applied in the caves (see Appendix).
For example, photographs of panels or figures are enlarged to full size;
tracings are then made from these, though it is always necessary to check
46 IMAGES OF THE ICE AGE

Fig. 23 Pioneering photography in

Trois Freres (Ariege) in the late
1920s. The cramped conditions in
this gallery led to the use of the
mirror. Note the camera's reflection
and the horse engravings. (Photo
courtesy of the Field Museum of
Natural His tory, Chicago, negative
#62007)
the results against the original too- in fact, it is preferable to do the work in
the cave, in front of the original. This technique is superior to other tracing
methods, since the wall need never be touched, and one avoids the
difficulties of installing the paper parallel to the panel, and of reflections
and shadows caused by irregularities in the wall's surface.
Enlargements to greater than actual size, or macrophotographs of small
aw.as, allow one to see tiny details or superimpositions more clearly.
Photographs of paintings, of course, may be published as they are, but
engravings often require tracings to help the reader decipher what is there:
thus photography and drawing are complementary, not substitutes for one
another.
By and large, photographs are taken perpendicular to the wall; for large
figures or panels , or for those in narrow places where one cannot stand far
enough back, a number of overlapping pictures are taken and then
amalgamated into a photo-montage (Figs.30-2). The same technique is
also invaluable for parietal figures which cannot be seen all at once (such as
 MAKING A RECORD 47

Fig. 24 Ultraviolet photo of red

'claviform' sign in Trois Freres
(Ariege) painted on a surface
prepared by scraping. Probably
Magdalenian . The clavifonn is 76 em
in length. (JV, collection Begouin)
the horse 'falling' round a rock at Lascaux) or for portable objects such as
those where the composition is engraved around a cylindrical baton.
A single picture of a parietal figure is far from sufficient today: a whole
series is now taken, using different films (both monochrome and colour),
light-sources (lamps and electronic flash) from different angles, lenses
(normal, wide-angle, macro, etc), filters and degrees of contrast. Multiple
flashes can bring out the relief of a sculpture, and even the traces of
modelling clay by hand, as on the Tuc d' Audoubert bison.22 The next few
years will see the introduction of computers for the enhancement of photos
and the storage of copies.
Another new technique, pioneered in the caves by Alexander Marshack
and adopted by Jean Vertut and others, involves the use of infra-red and
ultraviolet lamps. Ultraviolet radiation makes certain materials fluoresce:
any calcite and living organisms on the cave-walls do so, but the ochres and
manganese used as pigments do not; consequently one can assess any
damage to the figures caused by growths or calcite-flows, while UN can

/
 48 IMAGES OF THE ICE AGE

also show up any paint beneath the calcite and thus 'restore' fading detail,
as shown by Marshack on some of the figures at Niaux. 23 Consequently,
UN lamps were used for both tracing and photography during the recent
systematic recording of this cave's art. 24
On the other hand, infra-red light or film makes red ochres appear
transparent, so that one can see other pigments beneath them; moreover,
any impurities in the ochres remain visible, and so different mixes of paint,
with different impurities, can be detected. By this method, Marshack
claims to be able to assess in what order the famous 'spotted horses' panel
at Pech Merle (Lot) was built up. 25
Infra-red may make the original composition clearer by 'removing' the
thin trickles where pigment ran: Jean Vertut's picture of a 'tectiform' sign
at Bernifal reveals that it was not painted in continuous lines , as some had
thought, but as a series of dots from which the colour had run (Fig.26). 26
The use of infra-red also counteracts the effects of changes in humidity and
wall-conditions, which can make certain painted lines visible on some days
and not on others - for parietal paintings 'live' in accordance with
atmospheric conditions. 27
In addition to the traditional tracings, therefore, it has become essential
for archives to include complete coverage in slide form (Jean Vertut took
about 1,200 slides in the Volp caves), and in some cases also on video and
film- as, for example, in the 'Corpus Lascaux', a cinematic archive of that
cave's decoration. 28
However, for art of this kind, the ideal is to have a record in three
dimensions; so the very expensive technique of photogrammetry (the
method by which contour maps are made from aerial photographs) is
increasingly being used ; in Palaeolithic art, by means of
stereophotographs and some key measurements, it is possible to produce a
detailed 'contour map' of an object or panel, which can then be used to
make an accurate 3-D copy- in other words, making a cast without even
touching the original! This has been done, for example, for the clay bison
of the Tuc d' Audoubert, which could not be copied by any casting
technique owing to their fragility 29 - Jean Vertut's expertise was invaluable
in making this project a success.
The best-known 3-D copy at present is the excellent Lascaux II,
completed in 1983, which reproduces part of that cave and its paintings; 30
but the technology which produced it is already almost obsolete, and even
more accurate copies of decorated caves can and will be made in the future.

The quest for objectivity

Every tracing or copy of a Palaeolithic figure is inevitably subjective to
some degree; it is also a distortion of reality since- except in the case of
~urate casts- three dimensions are being reduced to two. Every tracing
is a personal piece of work, and it is impossible to remove subjectivity
completely. A copy is only as good as the copier, and all copiers make
mistakes, the number and extent of which depend on the method they use
and on their experience and personality: good examples include a series of
18 published versions of an engraved human head from the Grotte du
Placard (Charente), none of which proved to be completely reliable; 31 five
different versions of an engraved reindeer from Les Combarelles; 32 nine of
a reindeer from Saut-du-Perron and seven of a rhino from La
Colombiere; 33 eight of the chamois on a disc from Laugerie-Basse; 34 and
 MAKING A RECORD 49

about 50 ofthe famous mammoth of La Madeleine, found in 1864. 35 Some

of them were clearly done from the original, others were simply copied
from book to book, with the distortions increasing each time . If the
original were destroyed or lost , which of the versions should one 'believe'?
There are even a few cases where a truly appalling copy has been
published then republished by others, who neglected to compare it with
the original object (or even with a cast or photograph), although a glance
would have sufficed to reveal the numerous errors.36
A great deal of emphasis is now placed on the 'depersonalisation' of
copying; the pendulum has swung away from Breuil's technique of
putting some spirit into tracings (the 'artistic approach'); we now have
teams producing 'collaborative copies', or individuals making accurate
but often lifeless versions of the Ice Age images (the 'cartographic
approach'). Both methods have their merits and their disadvantages.
Deciphering or copying images on a cave-wall is rather like an
excavation, except that the 'site' is not destroyed in the process; the
pictures are 'artefacts' as well as art and , if superimposed, they even have a
stratigraphy. 37
Moreover, instead of selecting and completing animal figures from the
Fig. 25 Exhaustive tracing of mass of marks, like early archaeologists seeking, keeping and publishing
engraved stone from Gannersdorf
(Germany), and a horse figure
only the' belles pieces' and ignoring the 'waste flakes', the aim for the last
'extracted' from the mass (see also 20 years has been to copy everything. This helps to reduce psychological
Fig. 93) . Magdalenian . The effects akin to identifying shapes in clouds or ink-blots: faced with a mass
plaquette is 20 em long, 18 em wide, of digital flutings or engraved lines, the mind tends to find what it wants to
and the horse 8.5 em high. (After find , in accordance with its preconceptions, and often detects figurative
Basinski & Fischer) images which are not really there .38 In addition, one needs to counteract
50 IMAGES OF THE ICE AGE

the psychological effect whereby the eye is drawn to the deeper lines Fig. 26 Infra-red photo ofpainted
(although these may have been of secondary importance) and to lines in tectifonn at Bernifal (Dordogne) .
concave areas which are generally better preserved than those on convex Probably Magdalenian . Width : c
areas which are more exposed to wear and rubbing. 39 30 em. (JV,)
To eliminate lines we do not understand is an insult to the artist, who did
not put them there for nothing; where there are so many lines that it is Fig. 27 Drawingoffullrockface
difficult to 'isolate' anything , however, it is still necessary to 'pull out' any showing engravings of horses.
M ontespan (Haute Garonne).
definite figures which exist hidden in the complex mass (this is also far less
Probably Magdalenian . (Drawing
of a strain on the eyes), though one should still try to publish the mass, byRivenq)
leaving the reader free to make a different choice (Fig.25). 40 In his
herculean 25-year study of the 1,512 slabs from LaMarche with their
terrible confusion of engraved lines , Leon Pales isolated and published
only those figures which his expert knowledge of human and animal
anft"omy revealed to his eye : but he estimated that only one line in 1,000
has been deciphered on these stones. 41 Unfortunately, there are very few
scholars with similar skills in deciphering and reproducing Palaeolithic
engravmgs.
Breuil's notion of 'traits parasites' has been completely abandoned:
Palaeolithic images are not restricted to the figurative but also include non-
figurative, abstract and geometric marks , some of them of enormous
antiquity, going by the new Australian evidence (see above, p .28).
Just as artefacts are no longer dug up for their own sake, but for what
 MAKING A RECORD
51

0 5
52 IMAGES OF THE ICE AGE

their manufacture, context and associations can tell us , so Palaeolithic

images are now studied with the same aims. We shall examine techniques
of execution in a later section (p. 91), while associations of figures with one
another and with differently shaped panels loom large in recent
interpretations of cave-art (p.l76).
As for context, it is obviously necessary for cave-plans to be as accurate
as possible; however, as with tracings of figures, versions have sometimes
been published which are so divergent that four plans of La Baume-
Latrone, for example , looked like different caves! 42 Mistakes have also
been made in many cases concerning the part of the cave or the particular
wall on which figures occur, and many figures are often missed out
altogether. 43 All of this requires checking and published corrections.
We have already seen that copying now includes as much as possible -
and it is no longer always restricted to artificial marks, figurative or non-
figurative, since it can often be difficult (and occasionally impossible) to
tell whether lines are incisions or cracks, 44 or to differentiate between faded
spots of paint and blobs of natural colour in the rock .
Photographs , of course, play a major role in studies of context, since a
normal lens sees roughly the same area as an eyeball, and a photo shows
everything at once: artificial lines, natural marks and fissures, the grain of
the rock, and the overall aspect ofthe panel. In recent publications of some
decorated panels, the whole rock-face has been drawn, though few
specialists have the ability or the time to do this 45 (Fig.27). A different
technique is 'morphometric cartography': ie the making of a detailed
contour map, such as that of the panel with the black painted frieze at Pech
Merle. 46
There is also a new emphasis on the physico-chemical interactions of
images and wall:47 close study of the changing conditions in the caves, and
of the natural phenomena affecting the images, helps to assess what has
happened to them through time, and provides crucial information for
conservation.
In short, since the time ofBreuil, the task of copying Palaeolithic art has
not only acquired many improved techniques, it has also adopted new
aims. It is no longer a question of simply accumulating collections of
animal figures for publication. All copies, whether tracings or
photographs, are seen as tools for further research- as starting points, not
as ends in themselves. Tracings are no longer seen as faithful
reproductions, but as explanations and interpretations of the images,
incorporating an inevitable degree of subjectivity, distortion and choice.
Consequently, no copy can ever be definitive, no cave-art can ever be
entirely known.
 4
How Old is the Art?

How do we know all this art was made in the Ice Age, and during which
part of the Upper Palaeolithic? Surprisingly, the methods have changed
little since the nineteeth century, and much remains uncertain.

Portable art
Where portable objects are concerned, their position in the stratigraphy of
a site, together with the associated bone and stone tools, give a pretty clear
idea of the cultural phase involved; of course, the development of
radiocarbon dating since the last war has led to fairly accurate
measurement of the age of some of these levels- though measurement of
the radiocarbon content of a piece of portable art is almost never done
because of the destruction involved (in any case, this would date the death
of the animal whose bone or antler was used, rather than the artistic
activity). Examples include the date of 14,280 years before the present for
the layer of LaMarche containing the engravings, or that of 11,900 BC for
the layer at Enlene which yielded an antler baton with a salmon in
'champleve', (see below, p.80), or dates of 10,430 and 10,710 BC for the
Gonnersdorf habitations containing the engravings. 1
There are a number of snags. First, many of the best-known pieces of
portable art were found in the last century, and they were dug up like
potatoes; only a few people like Piette noted the particular layer from
which the objects came: early excavators such as Lartet and Christy, or the
Marquis de Vibraye, paid comparatively little attention to exact
provenance, being carried away by the search for more Palaeolithic
'nuggets' .2 Consequently, lots of famous specimens can be given no more
definite attribution than 'probably Magdalenian'.
Second, a great many objects have no details of provenance at all- either
because they come from clandestine digs, or because their sites were
atrociously excavated or never published. As we shall see, very few of the
'Venus figurines' of western Europe have any stratigraphic context
whatsoever. At Lourdes (Hautes-Pyrenees), over 2,000 cubic metres of
deposits were removed from the cave ofLes Espelugues (so that it could be
54 IMAGES OF THE ICE AGE

turned into a chapel) and scattered over nearby land- the many items of
Palaeolithic portable art subsequently discovered in these discarded
sediments thus have no stratigraphic provenance. 3
Third, whilea layer may accurately date art-objects such as plaquettes
which seem to have been made and discarded quickly, there are many
portable objects, notably statuettes, that seem to have been much handled
(a point brought out clearly in analyses by Alexander Marshack, see
below, p .184) and which may have been carried around and used for
generations: consequently, their final resting place may be very far from
their place of manufacture, and the layer in which they are found is merely
an indication of when they were lost or discarded, not necessarily of when
they were made- they may in fact be considerably older. 4
Finally, there are inevitably occasional disagreements about the precise
attribution of the cultural material associated with art-objects: for
example, the industry with the engraved pebbles of La Colombiere was
thought to be an atypical Gravettian, but is now reckoned to be an atypical
Magdalenian, which agrees better with its radiocarbon dates around the
thirteenth millennium BC . 5
Much portable art without good stratigraphic context has been assigned
to different phases on the basis of style, but this involves subjective
judgements and relies on 'type fossils' which often have limited validity.
As will be seen (p.85), all the western 'Venus figurines' have unjustifiably
been lumped into two groups, the Gravettian and the Magdalenian. It is a
tautology to date an obj ect by its resemblance to standard examples, and
then to use that resemblance as proof of the date.
Nevertheless, in a few cases, the resemblance is so striking that there is a
strong probability that the same period is involved, and at times even the
same artist or group of artists; this is shown in examples where both objects
being compared are securely dated, such as the spear-throwers of Mas
d' Azil and Bedeilhac,6 or the portable engravings from the grotte des Deux
Avens (Ardeche) which are very similar in style and execution (especially
of fish) to those of La Vache (Ariege): both have been dated to the mid-
eleventh millennium BC . 7
Chronological schemes tracing the development of portable art have
been devised, but tend to be simplistic: one of the earliest was that of
Piette, who thought that sculpture in the round came first, to be followed
by relief sculptures, 'con to urs decoupes' (p. 81), and finally engraving. 8
Unfortunately, he ignored or forgot the fact that engravings , for example,
already existed in abundance in his pre-engraving phases! As we shall see,
a variety of techniques were already in use before the Upper Palaeolithic,
while some (such as the 'con tours decoupes') seem to characterise certain
regions and periods and cannot therefore be taken as fixed stages in the
development of portable art as a whole.

--
Parietal art
Wall-art is extremely difficult to date; as mentioned earlier, attempts are
underway to develop means of dating the varnish of patina which forms
over rocks and their engravings in desert areas of the world, or, in
Australian caves, the layers of carbonate between different phases of
engraving on a wall; and since some black drawings at Niaux, Las
Monedas and other caves were done with charcoal, rather than the usual
manganese dioxide, it may prove possible to date the pigment directly by
 HOW OLD IS THE ART? 55

the radiocarbon method- as with portable art, this would date the death of
the tree rather than the artistic activity, but the two events are unlikely to
be very dissimilar in age. In all other cases, however, different methods are
required.
Proof of Palaeolithic age can take a number of forms: for example, the
depiction of animals which are now extinct (such as mammoth), or which
were present only during the Ice Age (figures of reindeer in southern
France and northern Spain), is a solid argument which, it will be recalled,
first convinced the scientific world of the reality of Palaeolithic portable
art.
More recently, attempts have been made to date Palaeolithic parietal art
by the species depicted: it has been assumed that the pictures reflect the
faunal assemblage outside, and that therefore the cave of Las Monedas,
which has mostly horses (42%) with some reindeer (13%), represents a
cold phase, whereas the neighbouring cave of Las Chimeneas, which has
42% cervids, few horses and no reindeer represents an older, warmer
phase9 - yet both sets of figures seem to be of the same style.
However, whereas Monedas and Chimeneas seem to be simple,
homogeneous 'sanctuaries', matters become far more complicated for
heterogeneous collections of animal figures, such as the 155 of Castillo
which have been assigned to the Aurignacian (27), Gravettian (8),
Solutrean (25), Lower Magdalenian (88) and Upper Magdalenian (7); the
Fig. 28 Well-dated parietal art: ten species represented are not particularly indicative of a cold or warm
painted panel at Tece-du-Lion climate; and if the chronological attributions are correct, then there was a
(Ardeche), showing aurochs (70 em massive presence of red deer in the Solutrean, whereas only bison existed
long) . Solutrean. (]V) in the Upper Magdalenian! 10
56 IMAGES OF THE ICE AGE

Clearly, there are many other possible reasons for these differences in
depicted species; they should not be taken simply as a tally of what was
available outside. As will be seen in a later chapter, species percentages in
depictions rarely correspond to those in animal bones; both assemblages
are a conscious selection of what was available. While a picture of a
reindeer may prove that the animal was present, an absence of reindeer
pictures does not prove that the animal was no longer around; and in any
case, parietal and portable art of the same period, and even in the same site,
tend to depict different species - thus, at the Tuc d' Audoubert, there are
close analogies between the two art-forms in style and details of execution,
and the two are clearly contemporaneous. The portable material has been
dated to 12,400 BC. Yet the bison dominates on the walls, and felines and
reindeer are present , together with 100 claviform signs, but there are no
fish or anthropomorphs; the portable art, on the other hand, has no
claviforms, felines or reindeer, but it does have fish and anthropomorphs,
and it is dominated by horses! 11 Other examples of differences in the
overall content of the two art-forms will be given in the next chapter.
In the past, it was often assumed that a formation of stalagmite or calcite
on top of a painting or engraving was proof of great age, but nowadays we
are more cautious, since such layers can form very quickly under certain
conditions, and may merely prove that the figures are not recent fakes.
Very different rates of stalagmite accumulation can occur in the different
microclimates of a single cave . 12
Quite a few caves seem to have been blocked during or just after the Ice
Age - Fontanet's entrance, for example, collapsed during the
Magdalenian, sometime after 11,860 BC (a date obtained from some
charcoal in the cave); and it was occupation deposits of the Palaeolithic
which blocked and masked the decorated gallery at La Mouthe -
consequently any art located behind blockages of this type has to be of
Palaeolithic date .
Similar proof occurs in cases where all or part of the decorated walls
themselves are covered by Palaeolithic deposits (datable through their
bone and stone tools, and sometimes by the radiocarbon method). The
classic example of this type is Pair-non-Pair, whose engravings appeared
only when the Gravettian occupation layers were removed; in the same
way, the sculptured friezes of Cap Blanc, La Chaire-a-Calvin and Angles-
sur-}' Anglin were masked to some extent by Magdalenian deposits; and it
will be recalled that the engravings of the Early Man rock-shelter in
Queensland disappeared into a layer dating to 13,000 years ago.
There are also a few cases where a fragment of decorated wall has fallen
and become stratified in the archaeological layers; at Teyjat (Dordogne),
for example, a block of stalagmite with engravings on it was partially
covered by late Magdalenian deposits, and a detached portion bearing a
~on engraving was found in the lower layer ('Magdalenian V') . We shall
see that fallen decorated fragments may constitute a large part of the art
from Aurignacian and Gravettian layers.
All the above circumstances, however, merely provide a minimum age
for the parietal art. T he Pair-non-Pair engravings have usually been
attributed to the Gravettian, but they could just as easily be Aurignacian
(they were above the level of the Aurignacian occupation layers). The same
applies to the Laussel bas-reliefs, which were partly covered by Gravettian
layers, but could be older. Similarly, the Isturitz engravings were above
the Solutrean level and covered by Magdalenian layers - yet they have
HOW OLD IS THE ART? 57

often been assigned to the Magdalenian.

A piece of wall may fall to the ground years, centuries or even millennia
after it was decorated. Its stratigraphic position records its destruction, not
its execution. Only in exceptional cases does the opposite occur: at the abri
du Poisson, one of the excavated layers contained thin pieces of limestone
which had crumbled from the ceiling; since the sculptured fish itself bore
no signs of any such deterioration, it must have been carved after the
crumbling (dating to 'Perigordian IV', a Gravettian phase); 13 here,
therefore, we have a rare example of a maximum possible age, rather than a
mtmmum.
A different type of maximum age occurs in some high valleys, such as
that ofVicdessos (Ariege), where excavation has shown that caves were not
occupied before the mid-Magdalenian because of glacial activity; therefore
the art in caves in or near that valley, such as Niaux, Fontanet or Les
Eglises, cannot be older than that period. But how can one obtain a more
precise date?
One way is to find out if and when the decorated cave or shelter was
occupied. Some sites have no known Palaeolithic occupation (for example
Monedas , Chimeneas, Pindal, etc); in others the deposits were removed,
unexcavated, to make visits easier for tourists; and other caves were visited
on many occasions in the Upper Palaeolithic (Castillo, Font de Gaume,
etc). In some cases, however, there is only one brief period of occupation,
and it is therefore likely - though by no means certain - that the artistic
activity coincided with this occupation. For instance, the recently
discovered decoration of Font-Bargeix (Dordogne) was probably done
towards the end of the Magdalenian, since the only known Palaeolithic
occupation belongs to 'Magdalenian Vl'. 14 The cave of Gabillou had
atypical early Magdalenian material at its entrance, 15 and it is thought
extremely likely that its very homogeneous collection of parietal figures
can be attributed to that occupation, and perhaps even to one person.
Some caves- notably in Cantabria (La Pasiega, Chufin, El Buxu, etcY 6
and the RhOne valley (eg Chabot) - only have Solutrean deposits . At the
Tete-du-Lion , the charcoal fragments of the 'foyerd'eclairage' (see below,
p.109) were next to some spots of red ochre on the ground, analysis of
which proved them to be of exactly the same composition as the bovid
painting on the wall - the radiocarbon date of 19,700 BC therefore dates
this cave's decoration to the early Solutrean, like much of the art in this
region. 17 Since some caves were decorated but never inhabited, the two
phenomena were clearly sometimes separate, so that a single occupation
layer, while suggestive, may be much younger or older than the art .
The situation improves if evidence of artistic activity can be found in a
site's occupation layers- especially colouring materials in a painted site, as
at Tete-du-Lion. This is the case at Altamira, Lascaux and Tito Bustillo,
for example; however, one has to be careful how occupation layers are
dated: radiocarbon dates from charcoal in Lascaux indicate the early
Magdalenian, c 15,000 BC- but, as has been pointed out elsewhere, this is
rather like dating a church by analysing the residue from its candles! The
study of pollen from occupation layers may also help to assess the date. 18
A different type of artistic evidence from the occupation levels is
portable art; at some sites which have not only parietal art but also well-
stratified portable art, one can see clear analogies between the two in
technique and style; this method has been in use since the beginning of the
century - indeed the first monograph on Altamira devoted an entire
58 IMAGES OF THE ICE AGE

chapter to the shared engraving techniques of the two art-forms. 19 In some

cases it is quite probable that the same artist was responsible, and at any
rate the method can provide a fairly reliable date for the wall-art: at Gargas,
there are some resemblances between one or two Gravettian portable
engravings and those on the walls. Examples abound in the Magdalenian-
the Tuc d' Audoubert has already been mentioned, and in the same way
there are close analogies between the portable art ofEnlene and, beyond its
passage, the parietal figures ofTrois-Freres, so that many ofthe latter can
probably be dated to the twelfth millennium BC; engravings on plaquettes
at Labastide which are in the same style as its parietal figures 20 come from a
layer dated to 12 ,310 BC.
At Angles-sur-l' Anglin, a 'Magdalenian III' occupation was established
on bedrock, its upper surface eventually touching the base of the
sculptured frieze, while the 'Final Magdalenian' layer covered the
sculptures completely. The artists clearly belonged to the Magdalenian III
phase, especially as it featured massive picks, lumps of pigment, ochre
crayons, grinding stones and spatulas; in addition, a portrait of a man,
engraved on a plaquette from this layer, bears a close resemblance to the
big polychrome specimen detached from the site's parietal art, 21 and this
phase has been dated to 14,160 years before the present (as mentioned
above, the Magdalenian III engraved stones of La Marche, only 30 km
away, have been dated to 14,280 before the present, indicating that the two
sites were roughly contemporaneous, and there was clearly contact
between them. Angles, like La Marche, has limestone plaquettes with
animals engraved on them).
The best-known examples of identical portable and parietal figures are
the engraved hind heads found on deer shoulder-blades at Altamira and
Castillo, and on the walls of both caves. The respective excavators claimed
that the Altamira specimens came from a final Solutrean layer and the
Castillo ones from an initial Magdalenian, and since then there has been
debate about the reliability of these observations, especially as most
scholars were keen to include these figures in the Magdalenian. 22 The
culprit, as is often the case, was the artificial nomenclature created by
prehistorians: it has become clear from more recent excavations that there
was continuity between the two phases, and in fact they are probably
contemporaneous in Cantabria .23 It is more sensible to refer to this period
by its date (c 13,550 BC at Altamira) than to attach cultural labels to it.

Stratigraphy of figures
Although the Altamira ceiling has sometimes been taken as a single
accumulated composition , it actually comprises a series of superpositions:
Francisco Jorda has distinguished five separate phases of decoration,
lj@"ginning with some continuous-line engravings, followed by figures in
red flat-wash , then some multiple-trace engravings, some black figures,
and finally the famous polychrome paintings. 24 Since the multiple-trace
engravings (mostly heads of hinds, and a few other animals) are identical to
the portable specimens from the cave, it is clear that the two earlier phases
of ceiling-decoration predate 13,550 BC, while the black figures and
polychromes are younger (Fig. 7).
It was the cave of Marsoulas which first inspired in Cartailhac the idea
for dating by superposition - he noted the different styles present and
thought he could distinguish at least three layers: black animal figures,
 HOW OLD IS THE ART? 59

Fig. 29 Part of the painted panel

at Tito Bustillo (Asturias), showing a
horse and reindeer, both engraved
and painted. The reindeer's muzzle is
superimposed on the black horse. The
deer measures just over 1 m from
muzzle to antler-tip. Magdalenian.
(Photo A. Moure Romanillo)
polychromes, and finally red figures (Fig.8). 25 Subsequently Breuil
adopted and extended this approach.
Nevertheless, it is tricky to use superposition as a chronological guide-
quite apart from the problem of establishing the order in which layers were
applied, which (see below, p.l06) can be very difficult. Theoretically, of
course, all the layers could have been produced within a very short space of
time - a superposition could represent half an hour! - but it is perhaps
more likely that they span a few years, at least, and the timespan could be
decades, centuries or even millennia.
A problem similar to the Altamira ceiling is posed by the great painted
panel ofTito Bustillo, where nine superimposed phases of engraving and
painting have been differentiated, culminating in the famous polychrome
horses and reindeer. The last five phases are thought to date to the
thirteenth millennium BC, on the basis of comparisons with engraved
plaquettes from the cave and from the occupation at its entrance, and dates
obtained from the layers containing residues of artistic activity at the foot
of the panel (stone tools with traces of pigment, etc); 26 but the first four
phases are more difficult - the excavators believe that they are probably
not much older than the rest, but since phase IV comprises engravings
resembling the multiple-trace type, it is possible that it dates to the
fourteenth millennium as at Altamira, with the other three phases being
older still (Fig.29).
Even moreproblematical is Lascaux which, in recent years, has been
treated as a homogeneous collection of figures, all produced within about
500 years; but Breuil said he could discern 22 different episodes of
60 IMAGES OF THE ICE AGE

decoration in the cave, while the abbe Glory saw six in the Hall ofthe Bulls
alone. 27 There is certainly some heterogeneity of style in the cave. The
dates of around 15,000 BC for the cave come from charcoal (notably from a
layer down the 'well-shaft') and are supported by geometric signs carved
on some objects (a lamp , and spearpoints) which are similar to those
engraved or painted on the walls, and by the pigments and worn flints in
the thin archaeological layer. 28
There is no doubt that much of Lascaux's art can be attributed to this
period; but this does not prove that the whole thing is a coherent entity
spanning only a few centuries . It is worth bearing in mind that there is
another cluster of dates from the cave in the ninth millennium before the
present, and that the remarkable stylistic resemblances between some of
its figures (notably the bulls and some deer) and those of some Spanish
Levantine art-sites such as Minateda and Cogul, which Breuil noticed
immediately and mentioned in print/9 have yet to be satisfactorily
explained (Figs.30-2).

Dating by style
If a decorated site was unoccupied, or simply has no portable art of its own,
it becomes necessary to seek stylistic comparison with material from other
sites and even other regions. As with undated portable art, one inevitably
encounters all the problems of subjectivity, of 'type-fossils', and of over-
simplistic schemes of development. All stylistic arguments are based on an
assumption that figures which appear similar in style or technique were
roughly contemporaneous in their execution.
Suggested sequences of appearance of different forms of representation
tend to be highly subjective, and sometimes twist the facts in order to make
them fit- for example , Stoliar's scheme, based on only a handful of sites,
progresses from the exhibition of part of an actual animal body in the
Lower and Middle Palaeolithic, through life-size dummies of various
types to sculpture, bas-relief and finally engraving in clay. It requires the
Montespan clay statues to be early Aurignacian or even Chatelperronian in
date, though there is not the slightest evidence for this view, and the
figures are almost certainly Magdalenian like the rest of the cave's art. 30

Breuil' s cycles
Breuil based much of his chronological scheme on the presence or absence
of 'twisted perspective', a feature which he considered primitive, and
which means that an animal figure in profile still has its horns, antlers,
tusks or hoofs facing the front (rather like Mickey Mouse's ears, which
always face forward no matter what the position of his body, since this
11141kes animation easier.
It can be argued that this graphic convention is not necessarily
primitive, since it enhances the impact and beauty of features like horns
and antlers, 31 and can help to suggest depth; indeed, it occurs frequently in
the art of later cultures in various parts of the world, up to and including
western modern art .32 Moreover, in the Ice Age figures it is possible that
hoofs, at least, were drawn in twisted perspective so that they resembled
the animal tracks which must have been of fundamental importance to the
hunters and those to whom they had to teach their skills - a side-view of a
hoof shows nothing. 33 Indeed, it has been said that if one were to wipe out
HOW OLD IS THE ART? 61

the bison on the Altamira ceiling and leave only their feet , a professional
hunter would at once recognise them as a good representation of a bison's
spoor. 34
For Breuil, however, twisted perspective was a decisive indicator of
archaism: he had observed that, at Gargas, the supposedly Gravettian
engraved figures had horns seen from the front; the bison on the wall of La
Greze was engraved with horns in twisted perspective, and since this cave
had late Gravettian and Solutrean deposits, he thought the bison was
relatively early; but in Magdalenian figures , hoofs and horns were drawn
in proper perspective. He therefore developed an 'evolution of
perspective' which included a 'semi-twisted' stage at Lascaux. 35
Nevertheless, like Piette's, his scheme was inconsistent, since twisted
perspective is also known in the Magdalenian, as in the hoofs of the
Altamira bison mentioned above. It is also dominant in the cave of
Gabillou which , as we have seen , is probably attributable to the early
Magdalenian. The Magdalenian parietal male portrait from Angles-sur-
l'Anglin, mentioned above , has a 'full-frontal' eye, as in Egyptian and
Cretan art. 36 Breuil himself placed the portable engraving from Laugerie-
Basse known as the 'Femme au Renne' firmly in the mid-Magdalenian (its
stratigraphic position was uncertain), but the animal profile's hoofs face
the front! Yet he claimed that a bovid head with twisted perspective from
Isturitz, found in a final Magdalenian layer, had been brought up from a
Gravettian level! 37 On the other hand, true perspective is sometimes found
in early phases, as in some ibex at Gargas, or the deer of Labattut. 38
Twisted perspective is not , therefore, a reliable chronological marker.
Breuil began by proposing a four-stage scheme, but eventually he
conceived a development in two cycles, the 'Aurignaco-Perigordian' and
the 'Solutreo-Magdalenian', which were largely inspired by the art in the
cave of Castillo, and are now of only historical interest, especially as they
virtually ignore portable art. With the benefit of hindsight, it seems too
neat and symmetrical that Palaeolithic art should have developed as two
essentially similar but independent cycles of evolution: each of them
progressed from simple to complex forms in engraving, sculpture and
painting; each started with 'primitive' or archaic figures, and led on to
more complex and detailed images. The second cycle was derived from,
and built on the first; overall, he saw a progression from schematic to
naturalistic, and finally to degenerate forms .39
The 'Aurignaco-Perigordian' cycle featured hand stencils, macaroni
finger-markings, simple outline animals with legs omitted, and large red
animals in flat-wash (as at Altamira), eventually leading to bichrome and
giant figure s (as at Lascaux, which Breuil thought Gravettian) with an
improvement in perspective from twisted to semi-twisted. He thought that
finger markings led to engravings, and thence to bas-reliefs like those of
Laussel.
The 'Solutreo-Magdalenian' cycle again began with simple outline
animals, then black figures with flat-wash, followed by infill and hatching,
and finally the bichromes and polychromes, as at Altamira and Font de
Gaume . The sculpture of the Solutrean (such as the carved blocks of Roc
de Sers, found in a late Solutrean level) led to the delicate, detailed
engravings of the Magdalenian. The cycle ended with the abstract designs
of the Azilian . He believed that his system was applicable almost
unchanged to all regions with Palaeolithic decorated caves. 40
Clearly, this system posed problems; we have already seen that a rigid
62 IMAGES OF THE ICE AGE

application of the twisted perspective criterion led him to inconsistencies Figs . 30/31/32 Three
and to errors, such as the wrong dating of Lascaux. Another puzzle was photomontages ofparts of Lascaux
why the bas-reliefs of the Gravettian should be separated in this way from (Dordogne). The upper two are from
those of the Solutrean, when it was far more likely that they were a the 'Diverticule Axial', while the
continuum. third is ofpart of the 'Hall of the
One solution was proposed by Annette Laming-Emperaire; 41 her Bulls' . The first includes a falling or
jumping aurochs cow (1. 7 m long)
approach was somewhat different from Breuil's, since she believed that
with a hull's head above, as well as
superpositions were actually purposeful compositions representing a ibex, quadrangular signs, and small
group or a scene. She simplified the evolution of Palaeolithic art into three horses.
basic stages: her 'archaic phase' was roughly equivalent to Breuil's first
cycle, and included the simple outline animals, for example; her middle
phase covered the material where Breuil's two cycles overlapped- ie the
bas-relief sculptures, flat-wash paintings, and so forth; and her final phase
included the polychromes and other characteristic Magdalenian figures. A
division into three 'cycles' was proposed by Francisco Jorda, primarily for
Cantabria: they covered the Aurignaco-Gravettian, the Solutrean and
lower Magdalenian, and the Magdaleno-Azilian.42

Leroi-Gourhan's styles
All these schemes were soon superseded, however, by that of Leroi-
Gourhan,43 which was based primarily on the characteristics of what
seemed to be securely dated figures (both portable and parietal) and which
still reigns supreme. Like Laming-Emperaire, he believed that
superpositions were deliberate compositions, spanning a very short
period, and therefore he treated the Altamira ceiling, for example, as a
coherent entity- this is probably valid for its polychrome figures but, as
mentioned above, there are four earlier phases of decoration beneath
them, and it is very risky to assume that they are all contemporaneous,
especially as more than one occupation is represented in the archaeological
layers of the cave.
Leroi-Gourhan proposed a series of four 'styles', the most recent pair
being further subdivided into 'early' and 'late' phases. Unlike Breuil's
cycles, they were seen as an unbroken development, with a series of
'pushes' separated by long periods of transition. 44 Following a 'pre-
figurative' phase covering some of the Mousterian material mentioned in
Chapter 5, Style I comprises material from the Aurignacian and early
Gravettian, and notably the carved blocks from those periods, and the
fallen fragments of what may have been decorated walls, together with the
very few pieces of portable art which he attributed to this phase, such as a
few plaquettes from Isturitz. The motifs include deep incisions, figures
with stiff contours (the Belcayre herbivore being the only complete
specimen), an apparent obsession with vulvas (see below, Chapter 7), an
al.ence of decorated utilitarian objects, and parietal art only in daylight
areas of caves and shelters. 45
Style II corresponds to the rest of the Gravettian and part of the
Solutrean; it features good animal profiles, with a sinuous neck/back line,
often an elongated head , an oval eye, and twisted perspective, but with the
extremities rarely depicted - the bison of La Greze characterises this
phase. Style II includes far more portable art than parietal - partly
because so many 'Venus' figurines have been lumped into this phase, and
partly because there are so few portable animal figures of the period with
which parietal figure s could be compared - in the Franco-Cantabrian
 HOW OLD IS THE ART?
63

The second includes deer, aurochs,

dots and quadrangular signs, and the
three 'Chinese horses'. The second
horse is 1.5 m long, the large cow
2.8 m, and the deer on the right 1.5 m.
The third shows the 'unicorn' on
the left (1.65 m long), as well as the
great aurochs, horses, and some small
deer figures. Probably M agdalenian.
(JV)
64 IMAGES OF THE ICE AGE

region as a whole, it is reckoned that only about 20% of all portable art
predates the mid-Magdalenian! Indeed, in Cantabria, all but one or two
pieces are attributable to the late Solutrean or Magdalenian.46 What little
parietal art there is in this phase is still restricted to daylight zones .
Style III, covering the rest ofthe Solutrean and the early Magdalenian,
took as its prototypes the animal figures in caves such as Lascaux and Pech
Merle, with their undersized heads and limbs, and semi-twisted
perspective dominating: the recently discovered figures in the caves of
Domme have been attributed to Style III because of their striking
resemblance to depictions at Pech Merle - particularly the volume of the
front half of bison-figures, and their elongated bodies. 47 It is in this phase
that decoration of the dark parts of caves seems to have begun .
Finally, Style IV includes all the wonders of mid- and late-Magdalenian
art, and the decoration of really deep galleries , sometimes marked to their
furthest accessible points. The early part is closer in spirit to Style III , with
little movement in the figures, which are simply 'suspended in mid-air';
but the more recent phase features more supple animals. In the later part of
Style IV, Leroi-Gourhan believed that the decoration of caves gave way to
that of plaquettes and similar objects . Contrary to popular belief, the art of
the Azilian is not limited to spots on pebbles, but also has naturalistic
animal figures in some areas (for example, in portable engravings at Pont
d' Ambon, the abri Morin , and the abri Murat) .
Leroi-Gourhan's scheme, therefore, is fundamentally like Breuil's in
that, with the outlook of modern people, it sees an overall progression
from simple, archaic forms to complex, detailed, accurate figures of
animals, while signs develop from simple and naturalistic to abstract and
stylised forms. It treats Palaeolithic art as an essentially uniform
phenomenon. Diversity is played down in favour of standardisation, and
the development is greatly oversimplified. 48
It should be stressed , however, that Leroi-Gourhan, like Breuil before
him, was fully aware of the tentative nature of his scheme, and that his
Styles had very blurred boundaries. He was able to draw on more securely
dated examples than were available to Breuil; but since, as mentioned
above, little early parietal art can be dated, he still had to fall back on some
of the same criteria as Breuil, such as twisted perspective. Both found it
very hard to apply their schemes to painting.
The difference between their two schemes is therefore not so much one
of principle as of results- Breuil had the art spread throughout the Upper
Palaeolithic (he saw the parietal art of Altamira as dating from the
Aurignacian to the Late Magdalenian), but with a very low output in the
Solutrean, whereas Leroi-Gourhan compressed the majority of the art into

..
his last two Styles, and especially IV - in other words, their absolute
chronologies may differ, but their relative chronologies are similar .

The ladder and the bush

Palaeolithic art did not have a single beginning and a single climax; there
must have been many of both, varying from region to region and from
period to period. It is self-evident that within those 20,000- 25,000 years
there must have been periods of stagnation, improvement and even
regression, with different influences, innovations, experiments and
discoveries coming into play. The same is true ofthe cultures of the Upper
Palaeolithic, which come and go during this period, each with its own
 HOW OLD IS THE ART? 65

H BREUIL FJORDA A LEROI-GOURHAN

Approx. Approx.
daces BC cultures cycles cultures cycles cultures styles dates BC

Azilian AZILIAN Azilian 9

10 000 Upper Magdalenian Iarc 10 000
VI :L
<
z
' < fin al 8 MAGDALENO V _ VI U
v z @ z
o:: ;.u
-AZILIAN ~ style IV
UJ upper 7 Middle <
IV ~ ~ ~ Magdalenian d
Magdalenian a ...J a middle

v ~/~
5; ~
111 III - IV early
1 i LOWER
11 1--- M
- ag-d-al-en-ia_n___, 6 Early Magdalenian u
15 000 lower 5 SO LUTREO- late IS 000
II--- - - - - - 1 ' MAGDALENI AN
I- II :;;:
upper 4 ::r: style III
u
Solutrean Solutrean ~ early
middle
Solutrean 3
Inter-Gravetto-
20 000 Solu trean 20 000
style II

Upper AURIGNACO- Upper AURI GNAC0-

Perigord ian PERIGORDIAN Perigordian 2 GRA VETI.IAJ\

Gravettian -
t- r - - - - ----11

25 000 25 000

style I

Aurignacian Aurignacian Aurignacian

30 000 30 000

Fig. 33 Comparison of the dating specialities (such as Solutrean flintwork, or Magdalenian bonework),
systems (based on style) of Laming- constituting a complex interweaving of traditions , with continuities in
Emperaire,Jorda Cerda and Leroi- some aspects of culture and sharp breaks in others 49 0

Gourhan. (After Naber et al) The development of Palaeolithic art was probably akin to evolution
itself: not a straight line or a ladder, but a much more circuitous path- a
complex growth like a bush, with parallel shoots and a mass of offshoots;
not a slow, gradual change but a 'punctuated equilibrium' , with occasional
flashes of brilliance. One must never forget that art is produced by
individual artists, and the sporadic appearance of genius during this
times pan cannot really be fitted into a general scheme. Each period of the
Upper Palaeolithic almost certainly saw the coexistence and fluctuating
importance of a number of styles and techniques (both realistic and
schematic), 50 as well as a wide range oftalent and ability (not forgetting the
different styles and degrees of skill through which any Palaeolithic Picasso
will have passed in a lifetime). It is also naive to assume that all Palaeolithic
images are purposeful masterpieces- much of what we call stylisation may
simply be compensation for a lack of skill; there must be a certain
percentage of meaningless scribble , limited ability, or simply crude
attempts by children or beginners, and so we should not assume that
everything had complex symbolic meaning.51 Moreover, there must have
been different developments at different times in different regions , and
similar styles in two separate regions are not necessarily contemporaneous.
Consequently, not every apparently 'primitive' or 'archaic' figure is
66 IMAGES OF THE ICE AGE

necessarily old (Leroi-Gourhan fully admitted this point), and some ofthe
earliest art will probably look quite sophisticated: who, for example,
would assign the Vogelherd animals or the Brassempouy head to the
Aurignacian if they had not been found in a layer of that period? Leroi-
Gourhan preferred to put the Vogelherd animals into his Style II, thus
denying their actual provenance.
In some problems, both Breuil and Leroi-Gourhan seem to have been
wrong: at Altamira, for example, both of them attributed the polychrome
paintings of the ceiling to the late Magdalenian, but excavation revealed
only layers of the final Solutrean and the mid-Magdalenian, which were
rich in portable engravings and in colouring materials, as we have seen; so
it seems likely that all of Altamira's decoration can be attributed to those
periods, and that it was blocked during or after the mid-Magdalenian
(there is likewise no trace of Aurignacian occupation, contrary to Breuil's
theory) .52
There are many examples where Leroi-Gourhan's scheme proved more
satisfactory than Breuil's: for example, at Trois-Freres, Breuil thought
several parietal bison in the 'Sanctuary' were Gravettian, but Leroi-
Gourhan thought them Magdalenian, and showed they were identical to
other bison which Breuil considered Magdalenian! His view has been
confirmed by the discovery of a Magdalenian pebble from Enlene with a
bison and a horse engraved on it: a broken line next to one figure is very
similar to those on seven of the parietal bison. 53
Other problems remain open, until such time as excavation of the sites
in question may help to resolve them: for example, the parietal art of
Altxerri seems to be homogeneous, and has been attributed to the late
Magdalenian on the basis of Leroi-Gourhan's stylistic criteria, through
comparison with portable art from Isturitz, and because the reindeer is
depicted; but other scholars have thought it Solutrean and/or lower
Magdalenian, and point to supposed resemblances in technique and style
between the Altxerri engravings and the 'multiple-trace' engravings of the
late Solutrean/early Magdalenian of sites like Altamira and Castillo (and in
any case, reindeer bones have been found in Solutrean levels in northern
Spain). 54
It is simply impossible to fit everything into a rigid scheme which
minimises the variability of representations in any phase: in Palaeolithic
art every rule has exceptions - for example, where all polychrome
paintings were automatically thought contemporaneous, it is now clear
that those of Altamira are earlier than those ofTito Bustillo. There are no
real 'type fossils', and styles are very hard to define and separate. Rather
than attempt to define styles, and then fit the images to them, it seems
more sensible to take the archaeological and dating evidence at face value,
to make an inventory of the different works in each period and their
pr~ominant features, and then develop an overall 'evolutionary scheme'
(if such a thing is desirable) which incorporates that evidence without
distorting it. We must adopt a more flexible system of dating by style,
which avoids the problems of rigid systems like those above.
A classic example is provided by the cave of Parpall6, Spain, whose
numerous pieces of well-stratified portable art (see above, p.33), spanning
many phases of the Upper Palaeolithic, were somewhat neglected by
Leroi-Gourhan, perhaps because they display a number of features which
contradict his scheme. Work in recent years, particularly in Spain (at
Parpall6 and other sites), the Rhone valley and the Quercy region, has
HOW OLD IS THE ART? 67

established that , far from being the artistic desert of Breuil's view, the
Solutrean witnessed a tremendous amount of aesthetic activity. 55
The uncertainties in dating by style alone mean that all known parietal
sites are featured on the maps , pp.36-8 , with no attempt at dividing them
into chronological slices: it is certainly unsatisfactory to compress 25 ,000
years of wall decoration- two-thirds of art history!- into one map, but this
solution is more honest than a subjective division into stylistic periods .
Nevertheless, for the moment , style is the only means we have of dating
a great many sites. What , then, is the way forward? Rather than rely
blindly on a fixed scheme, or on any particular feature such as twisted
perspective, it is wiser to establish assemblages of features: in the same
way, archaeological layers are no longer given a cultural attribution
because of a single 'type-fossil', but rather on the basis of the whole stone
and bone industry present, on the associations, interrelations and
presence/absence of different features .
One example is provided by a recent study of a piece of Gravettian
portable art, in which the authors drew up a list of definite or extremely
probable specimens of portable and parietal art of the period (eg from
Parpall6, Isturitz, etc), and then looked for relevant criteria by which they
could be recognised .56 It was found difficult to produce firm criteria for
Leroi-Gourhan's styles - as already mentioned, t\Yisted perspective was
known in the Magdalenian, while an oval eye is not exclusive to Style II
either. However, the piece under study had an association of features
(elongated horse head, oval eye, hoof-style) which placed it firmly within
that phase.

Despite the variety of the forms which it encompasses , Palaeolithic art

does constitute a recognisable episode in art history. We have seen that, for
a number of reasons, almost all of it can safely be attributed to the 20 or 25
millennia of the Upper Palaeolithic, and most of it to the last ten. More
precise attributions are difficult at present and, where possible, one should
therefore try to restrict discussion to examples of portable and parietal art
which are securely dated, and thus hope to sidestep accusations of
subjectivity and tautology!
 5
Forms and Techniques

Before embarking on a survey of those forms and techniques of

Palaeolithic 'art' which have survived, it is worth remembering that they
probably represent only the tip of the iceberg: although it cannot be
proved (unless one day a waterlogged , desiccated or frozen site of the last
Ice Age is discovered, in which everything is preserved), it is virtually
certain that a great deal of artistic activity involved perishable materials
which are gone for ever: work in wood, bark, fibres, feathers, or hides (as
mentioned earlier, some decorated wooden rods are known from late
Middle Stone Age levels of Border Cave, South Africa, dating to c 37,000-
50,000 years ago 1) ; figures made in mud, sand or snow; and, of course,
body painting which, through finds of red ochre in living sites and in
graves, is thought to have very remote origins.

Music
In addition , dance and song leave no traces at all, and such things as reed-
pipes, wooden instruments , and stretched-skin drums will have
disintegrated; however, a few musical instruments have survived from the
Upper Palaeolithic- there are about 30 'flutes', spanning the Aurignacian
and Gravettian (18), the Solutrean (3) and the Magdalenian; a handful
come from Hungary, Yugoslavia, Austria and the USSR, but most are
from France, with 14 from the supersite oflsturitz alone. 2 The majority are
bl'fllken; the French ones are made of hollow bird-bones, while the eastern
specimens are of reindeer or bear-bone; they have between three and seven
finger-holes along their length, and are played like penny-whistles rather
than true flutes. Experiments with a replica by a modern musicologist have
revealed that, once a whistle-head is attached to direct the air-flow, one can
produce strong, clear notes of piccolo-type, on a five-tone scale. 3 It should
be noted that a possible flute , comprising a fragment of a very small
hollowed-out bone into which someone had begun to make two holes,
came from a pre-Aurignacian layer of the Haua Fteah cave, North Africa,
dating to c45,000 years ago. 4
FORMS AND TECHNIQUES 69

A few shaped, polished and engraved bird-bone tubes have been found
which have no holes, and have been interpreted as trumpet-like 'lures' for
imitating the call of a hind in the rutting season- one fine example from the
Magdalenian site of Saint-Marcel even has a series of what look like cervid
ears engraved on it! 5 Many perforated reindeer phalanges have been
interpreted as whistles in the past, though often the hole was made by
carnivore teeth or other natural breakage; those which were intentionally
made do produce a shrill, powerful note. A few definite whistles in bird
bone are also known, such as the Magdalenian specimens from le Roc de
Marcamps, Gironde. 6
A number of oval objects of bone or ivory, with a hole at one end, have
been interpreted as 'bull-roarers' ('rhombes', or 'bramaderas'), a type of
instrument which makes a loud humming noise when whirled round on a
string: a particularly fine example made of reindeer antler is that from the
cave of La Roche Lalinde (Dordogne) (Fig.34). The well-known parietal
engraving from Trois Freres of a 'sorcerer' with a bison-head (Fig .1 04) has
often been interpreted as playing a musical bow, but this seems an
extremely tenuous idea, and the enigmatic marks in front of its mouth
could be any one of a number of things (see later, p. l52).
As for percussion, a number of mammoth bones , painted with red
ochre, from the site of Mezin, near Kiev, dating to c 20,000 Be, are thought
to be musical instruments - a hip-bone xylophone (osteophone?), skull
and shoulder-blade drums, and jawbone rattles- and have been played by
Soviet archaeologists, who even cut a record of their jam-session. 7
Finally, there are possible lithophones in a number of caves: 'draperies'
of folded calcite formations often resound when struck with a hard object
(wooden sticks seem to produce the clearest and most resonant notes), and
this seems to have been noticed by Palaeolithic people, since some of the
lithophones are somewhat battered, and are decorated with painted lines
and dots. 8 Apart from Nerja in Spain and Escoural in Portugal, all known
examples are in the Lot region of France (Cougnac, Pech Merle, Les
Fieux, Roucadour) ; moreover, most of them are in or near large chambers
which could have held a sizeable audience.

Red skins or tanned hides?

Virtually all peoples around the world paint their bodies on certain
occasions, and we have no reason to doubt that the same was true during
the Palaeolithic- indeed, this was probably one of the very first forms of
aesthetic expression. Unfortunately, owing to the decomposition of
bodies, we have to infer it from other evidence. Lumps of natural pigments
are known from archaeological sites of very remote periods: for example,
one piece of red mineral, with vertical striations resulting from use, was
found in the Acheulian (c 250,000 Be) rock-shelter of Becov,
Czechoslovakia, which had been occupied by Homo erectus. The even
earlier site of Terra Amata (c 350,000 BC) at Nice, France, produced 75
bits of pigment ranging in colour from yellow to brown, red and purple;
most of them have traces of artificial abrasion and were clearly introduced
to the site by the occupants, since they do not occur naturally in the
vicinity. 9
During the period of Neanderthal people (c 100,000-35,000 Be), such
pigments become increasingly frequent, not only in occupation deposits
but also in burials, which now occur for the first time. In France, for
70 IMAGES OF THE ICE AGE

example, the cave of Pech de 1' Aze yielded 103 blocks of manganese
dioxide (black/blue), plus 3 of iron oxide (red); 67 of them were rounded or
polished into a 'crayon' shape, as if they had been used on some soft
surface. 10
A Neanderthal skeleton at Le Moustier was sprinkled with red powder;
red pigment was also found around the head of the famous skeleton of La
Chapelle aux Saints, near two skeletons at Qafze, Israel, and with many
others. In addition, there is evidence for actual mining of haematite (iron
oxide) in Southern Africa from c 45,000-50,000 BC onwards (it will be
recalled that pigment was in use in Zimbabwe at least 125,000 years ago),
and in Hungary from 30,000 BC. 11
It was in the Upper Palaeolithic period that pigments became really
abundant, being transported in tens of kilos; in some French occupation
sites, it is not rare to find habitation floors impregnated with red to a depth
of20 em. Well over 100 sites with pigment are known, as well as at least 25
burials. As we shall see, some of these pigments can be linked to the
decoration of cave-walls; but what about open-air sites or unpainted caves?
Can we assume that the presence of pigments here necessarily indicates
body-painting?
Unfortunately, matters are not so straightforward, for the simple reason
that mineral pigments of this type have a number of properties.
Ethnographic studies around the world show that ochre is often used in the
treatment of animal skins , because it preserves organic tissues, protecting
them from putrefaction and from vermin such as maggots . It is probably
this kind of function which explains the impregnated soil in some
habitation sites and the traces of red mineral on many stone tools such as
scrapers. Similarly, red pigment may have been applied to corpses not so
much out of pious beliefs about life-blood, as is commonly assumed, or in Fig. 34 Engraved <bull-roarer'
order to restore an illusion of health and life to dead cheeks, but rather to with geometric/linear motifs and
covered with red ochre ,from La
neutralise odours and help to preserve the body. 12
Roche at Lalinde (Dordogne).
Even if, as most prehistorians believe, the people of the Old Stone Age Magdalenian. Length: 18 em, width:
did indeed paint their bodies, the practice may have been purely 4 em . (JV)
functional in some cases, rather than aesthetic: ochre is very effective in
cauterising and cleaning injuries, and is still used in parts of Africa to dry
bleeding wounds- in fact , until the end of the last century, it was still used
by country doctors in parts of Europe as an antiseptic in the treatment of
purulent wounds. Another function which may have been important
during the last Ice Age is that of protection against the elements and
insects. Peoples such as Polynesians, Melanesians and Hottentots used red
pigments to maintain bodily warmth and ward off the effects of cold and
rain; among some North American Indian tribes, a mixture of red ochre
and fat was often applied to the cheeks of women and children as a hygienic
measure to protect their skins against the sun and dry winds; while in other
pllf'ts of the world red paint has been used as a protection against
mosquitoes, flies and other disease-carriers.
The more aesthetic uses of pigment on the body may well have arisen
from practices such as these: for instance, the medicinal properties of
ochre may have led to the painting of the dead or dying, as was commonly
done by certain tribes in the New World. In Palaeolithic cases, it is often
difficult to tell whether a body had its flesh painted or merely its bones. If
the whole body was painted at death, or just before, the lumps of pigment
placed with the corpse may represent supplies of body-paint for the
afterworld.
FORMS AND TECHNIQUES 71

But what of the living? As already mentioned, most authorities agree

that Upper Palaeolithic people must have painted their bodies , but the
evidence is very limited. The use-wear on lumps of pigment is often shiny,
indicating that they were applied to soft surfaces, which could be human
skin, but also animal hide; the traces of use on a rough surface may simply
indicate removal of powder to be used in a liquid paste. Some French caves
have yielded bone tubes or hollow bird-bones, often engraved, containing
powdered pigment, and not all of this material can be linked to cave-art; at
the Mas d' Azil, excavators found a flat cake of red ochre, pitted with holes,
and associated with sharp bone needles , which they interpreted as
evidence for tattooing in the Upper Palaeolithic. 13 Finally, as will be seen
below, some human figurines of the period were originally painted red -
but the same is true of bas-reliefs of horses and fish!

Techniques in portable art

These will be reviewed in order of apparent complexity, but this should
not be taken as a chronological progression: as we have seen in the
preceding chapter, the phenomenon of Palaeolithic art is a complex web of
forms and styles rather than a simple linear development.
Portable art is usually divided into what seem to be utilitarian and non-
utilitarian objects (ie decorated tools versus art objects or ornaments), and
their decoration is classed as figurative or non-figurative . A very wide
variety of materials and forms was employed.
The study of the techniques used in the period rests on two main types of
evidence: first, the traces left on the objects or images by the tools, together
with precise observation of their technological characteristics (in a very
few cases, production debris or what may be the original tools have
survived in close proximity to the images); second, experiments with
similar materials and tools have been carried out, followed by comparison
between the modern results and the originals.

Slightly modified natural objects

Much of what is called 'parure' (jewellery) belongs in this category- ie
fossils , teeth, shells or bones which have been incised, sawn or perforated.
Such techniques are by no means restricted to the Upper Palaeolithic: a
growing number of specimens are known from the preceding
(Mousterian) period, and can therefore be attributed to Neanderthalers:
two bones (a wolf foot-bone and a swan vertebra), with holes bored
through the top, from Bocksteinschmiede (Germany), dating to c 110,000
years ago; a carved and polished segment of mammoth molar, and a fossil
nummulite with a line engraved across it (making a cross with a natural
crack) , from Tata (Hungary), dating to c 100,000 years ago; a bone-
fragment from Pech de l'Aze (Dordogne), with a hole carved in it; a
reindeer phalange with a hole bored through its top, and a fox canine with
an abandoned attempt at perforation, from La Quina (Charente). 14 As will
be seen below, other forms of 'aesthetic expression' are also known from
the Mousterian.
The earliest phase of the French Upper Palaeolithic, known as the
Chiitelperronian (c 35,000 BC), has yielded a few more examples: the best
known are those from the cave of Arcy-sur-Cure (Yonne). These levels at
the site also contained a Neanderthal tooth; in view of the discovery of a
Neanderthal skeleton in a Chatelperronian layer at St Cesaire (Charente
72 IMAGES OF THE ICE AGE

~ R}fl tft aft~ ~F& ra

~ ~~ ~ ~~ ~ ~g \Yd
t&
, rav {).;1
.

..

ta ~ (Y
ii
Ull v ~ !)
~
. - .. Fig. 35 Perforated and engraved
deer canines from theM agdalenian
burial at StGermain-la-Riviere
(Gironde). (After Laurent)
Maritime)/ 5 it is more than likely that the earliest Arcy ornaments can be
attributed to Neanderthal craftsmanship. They include wolf and fox
canines made into pendants by incising a groove around the top, at least
one sawn reindeer incisor, a bone-fragment with a wide carved hole, a sea
fossil with a hole bored through its centre, and a fossil shell with a groove
cut around the top. 16
The next layer at Arcy, representing the Aurignacian (c 30,000 BC), has
material which features the same techniques, clearly drawing on what had
been developing for millennia: perforated fossils , a bone pendant with a
wide carved hole , and so forth. Even older Aurignacian sites, such as the
cave of Bacho Kiro in Bulgaria (over 41 ,000 Be), contain perforated
animal-tooth pendants. 17
The Aurignacian therefore has no sudden appearance of this kind of
material, but there seems to be a marked increase in its abundance,
perhaps linked to advances in lithic technology which improved or
facilitated the working of decorative objects. The three main classes of
'parure' are beads (of ivory, bone, stone, fossil wood, etc), animal teeth,
and shells.
The animal teeth, perforated through the root, are mostly bovine
incisors and the canines of fox, stag, wolf, bear or lion - fox teeth are often
the most abundant in the Aurignacian and Gravettian, especially in central
allft eastern Europe. 18 For example, the old man buried at Sungir, near
Moscow, about 24,000 years ago, had two dozen perforated fox canines
sewn on the back of his cap, 19 and more were found with the two children
buried at the site . Over 50 perforated fox canines were found in Kostenki
XVII (USSR) and are dated to 32,000 years ago / 0 while 150 covered the
head of the child buried beside Kostenki XV. In western Europe, on the
other hand, and particularly in the Magdalenian , stag canines were the
favoured decoration - the best known are the 70 perforated specimens
found round the neck or chest of the woman buried at St Germain-la-
Riviere (Gironde), 20 of which are engraved with crosses or parallel lines. 21
FORMS AND TECHNIQUES 73

The popularity of canine teeth continued to the end of the Upper

Palaeolithic, but in the Magdalenian there was an increase in the practice
of sawing reindeer incisors, which (as mentioned above) is already present
in the Chatelperronian period at Arcy: over 50 have been found at La
Madeleine and Gonnersdorf, and over 200 at Petersfels (Germany). Their
occasional discovery in rows shows that, as among some northern peoples
in historical times, their roots were sawn through and they were then cut
from the mouth as a group, still held inside a strip of gum which was handy
for hanging them as a string of eight 'pearls' .22 Occasionally one
encounters the teeth of other species used as pendants: eight perforated
human teeth are known from sites in France (one from the Aurignacian
site of La Combe, the rest from Magdalenian sites such as Bedeilhac), and
one from the Gravettian of Dolni Vestonice (Czechoslovakia) / 3 there are
pierced seal-teeth from the Magdalenian of Isturitz, and many sites have
imitations of canines (especially stag canines) made out of ivory, stone or
bone. 24
As for shells, only a few species were selected: primarily small, globular
gastropods (such as Littorina and Cypraea) which could easily be sewn to
clothing; long forms (such as Dentalia or Turri cella) which could be easily
strung; and a few scallops (Cardium, Pectunculus). 25 Many of these
species are inedible, and their function was clearly decorative rather than
nutritional: most were perforated with a pointed tool. They are often
found in considerable quantity, even in early sites- there were 300 in the
Aurignacian Cro-Magnon burial alone (and hundreds more in other
burials), while living-sites such as Isturitz (Pyrenees-Atlantiques) or the
abri Blanchard (Dordogne) contained hundreds of periwinkles . Fossil
shells were also utilised and sometimes came from great distances - for
example, those at Mezin (USSR), a site dating to 21,000 years ago, came
from a distance of at least 600 km. 26 The shells ofland molluscs were rarely
used, no doubt because they are thinner and more fragile.
Research shows that Dordogne sites (such as Laugerie-Basse, abri
Pataud, abri Castanet) generally yield a high proportion of species from
the Atlantic, particularly those which are common along the coast of
Charente, but shells from the Mediterranean are also clearly represented.
One finds the same ratio at the Atlantic end of the Pyrenees ; but sites in the
Central Pyrenees , such as Lespugue, at a distance of 200 km from either
coast, contain a more even ratio, while further to the east, in Ariege, the
proximity of the Mediterranean is reflected quantitatively in the shell
collections, 27 although even here Atlantic shells dominate slightly- this is
no doubt because all rivers in the French Pyrenees (apart from those at the
eastern extremity), like those of Dordogne, flow out to the Atlantic, and
this must have determined the movement of people and materials to a
considerable extent.
It is theoretically possible that all the shells came inland in an exchange
network involving 'maritime peoples' for whom we have no evidence
whatsoever thanks to the drowning of the coastlines of the last Ice Age
through the rise in sea-level since that time. Certainly a .great deal of
exchange went on: for example, Mediterranean shells have been found in
the German site of Gonnersdorf, 1,000 km away, and it is unlikely (though
possible) that the site's occupants travelled to that coast for them .
It has been argued 28 that the great number of shells in the sites of
Blanchard, Castanet and La Souquette means that this clutch of
Aurignacian sites represents a market centre for exotic materials . Equally,
74 IMAGES OF THE ICE AGE

however, it could simply mean that these sites were centres of production,
or the habitations of those who specialised in working these materials:
there is ample evidence throughout the Upper Palaeolithic for craft
specialisation, and for repeated contact with the coasts which involved not
merely vague 'exchange networks' but also probably the seasonal
movements of people, following herds, and dispersing or coming together
in certain places at different times of year. 29
In the past it was often assumed that the shells served as pendants and
ear-rings or, in groups , as necklaces or bracelets. However, as with beads
it is evident from finds such as the burials at Grimaldi on the
Mediterranean coast, or those of Sungir, that many were attached to caps
and clothing. At Sungir, for example, the three burials had only a handful
of perforated shells, but about 3,500 beads of mammoth ivory each,
arranged in rows across the forehead and temples, across the body, down
the arms and legs and around the ankles. Rather than being sewn on to
garments one at a time , it is far more likely that the beads were strung on
lengths of sinew, which were then attached to the clothing.
It has been estimated that a Sungir bead would have required about 45
minutes for its manufacture (cutting the tusk, drilling the hole, etc), which
means that each body had 2,625 hours of'beadwork' buried with it; 30 and
the very standardised and uniform appearance of these objects suggests
that they were produced by only a few people. In western Europe, far more
ivory beads are known from living sites than from burials; only the context
provided by a burial can indicate an ornament's true function, but it is
likely that most of the finds from occupation sites were also attached to
clothing. Some idea of the production sequence involved in beads can be
gained from the Aurignacian material at Blanchard, which includes pieces
at different stages of manufacture: small prepared rods of ivory were
divided into sections, separated into pairs, and then worked into a dumb-
bell form and perforated , before the final shaping (most are round or
basket-shaped). 31
Other types of bead include fish vertebrae, which were sometimes
strung together as necklaces, as in an example from a burial at Barma
Grande. 32 It is worth noting that ornamentation of this kind is by no means
restricted to Europe: for example, it will be recalled (see above, p.29) that
perforated bone beads are also known from the late Pleistocene of
Australia (Devil's Lair), while the Upper Cave at Zhoukoudian, China,
dating to 18,000 years ago, yielded over 120 decorative items such as
pierced vertebrae of carp and other fish, perforated seashells, animal teeth,
small pebbles, and engraved/polished bone tubes cut from the leg-bones of
birds. Many of the perforations are coloured red, suggesting that whatever
thread they were strung on was dyed. 33
As with any other category of portable art, there is a marked
dijferentiation in the distribution of ornaments: many sites in Europe
(including some burials) have none or a few, and others have hundreds. As
mentioned above, this probably reflects the presence of specialised
craftsmen, as well as the varying functions of different sites (including
clothing manufacture?), and perhaps even, where rich burials are
concerned, some form of incipient hierarchy.

Engraved or painted stones

It will be recalled that the engraving of a bear on a pebble from the cave of
Massat was one of the first pieces of Palaeolithic portable art to be
 FORMS AND TECHNIQUES 75

36 Engraved plaquette from authenticated; and that a few sites have hundreds of incised slabs - mostly
Puy de Lacan (Correze), showing of sandstone, limestone, slate or stalagmite. Few examples of portable
duck-like bird, a fine bison head engraving are known from the early Upper Palaeolithic or from eastern
it, and, to the right, the Europe (one of the earliest is the geometric motif on a plaquette from the
.,.,n,111nt'tPr< of another bison. Total
Chatelperronian Grone du Loup, Correze, 34 and a late Mousterian
20 em. Magdalenian. (JV) limestone slab from the cave of Tsonskaia in the Caucasus has a cross
engraved on ie 5), and in fact this particular type of art characterises the
Magdalenian of western Europe.
The incisions on stone are sometimes deep and clear, but in many cases
they are so fine that they are almost invisible - this is why so many
engraved pieces are found discarded in the spoilheaps of earlier, less alert
excavators, as at Enlene. Only under a strong light coming in from the side
can one see the lines at all; indeed, this kind of fine engraving can almost be
classed as a drawing rather than an incision!
Alexander Marshack has pioneered the 'technological reading' of
Palaeolithic images; his studies of engraved stone, bone and antler under
the microscope have enabled him to follow the mechanics,
micromorphology and 'ballistic trace' of each incision- its point of impact
and subsequent path- and to identify marks made in different ways: from
left to right or right to left, as arcs, jabs, etc. He has also claimed that he can
detect changes of tool and of hand: for example, he believes that at least
seven different points were responsible for a fish with 'arrows' on a stone
from Labastide (Hautes Pyrenees), and that four or five different points
76 IMAGES OF THE ICE AGE

were responsible for renewmg the horns on the rhinoceros of La

Colombiere. 36
However, other scholars have argued that, at least where small stone
plaques are concerned, a single burin can produce a wide variety of traces
on them, depending on the part of the tool used, its position and angle, and
the strength of the hand - for example, the section of the incision changes
when a straight line is continued as a curve. 37 In any case, a tool may have
been resharpened in the course of use. Experiments with slates ofthe type
found at Gonnersdorf and Saut-du-Perron suggest that any changes in the
incised marks are due to differences in tool-pressure, not to different
tools .38
The new technique of placing varnish replicas of engraved surfaces
under the scanning electron microscope (see above, p.45) is now being
used in an attempt further to elucidate this problem; criteria are being
sought which would identify the use of the same tool in different kinds of
incisions- for example, it has been found that every tool leaves distinctive
secondary striations alongside the main incision whenever parts other than
the point have momentary contact with the stone. 39
A different question concerning 'plaquettes' (defined as slabs of stone
with parallel faces, under 20 em across and 4 em thick) is that of their
usage. Scholars such as Henri Begouen, who believed in hunting magic
(see p . ISI ), were inclined to see evidence of ritual in everything. Since
many of the engraved plaquettes from Enlene were broken, and fragments
of the same specimen might be found metres apart, he concluded that the
breakage and dispersal were purposeful; since most of them had been
burned, and appeared to lie with the engraved face downward, this too
formed part of the ritual. 40

------

0 5 10 IS em

Fig. 37 Tracing of an aurochs and

deer on a plaquette from Trou de
Chaleux, Belgium. Magdalenian.
Total width: 80 em . (After Lejeune)
FORMS AND TECHNIQUES 77

It is not always the case that most lie with the engraving face-down
(though this seems to be true for sites such as La Ferrassie and the abri
Durif at Enval 41 ); the fineness of many incisions makes it impossible to see
them within the cave, and if plaquettes are collected from a cave-floor, as at
Labastide, it is hard to remember which way up they were- in any case,
some have engravings on both sides: at Enlene, about 16% are engraved on
both sides; and, of 50 which are not but whose position in situ was noted,
27 had the engraving face-up and 23 face-down! 42
Moreover, in some caves, the upper face of a plaquette lying on the
surface becomes coated with a deposit of clay or calcite and needs to be
washed or treated with acid before the engravings underneath become
visible. 43 As for dispersal, it is true that pieces have often been well
scattered, and in some cases missing fragments have never been recovered;
this may sometimes denote purposeful breakage, but people and animals
trampling around on cave- or habitation-floors can also have drastic effects
on the material lying there.
The question of burning is more interesting. There is a definite link
between plaquettes and fire in some caves: many have marks of burning
and charcoal, some (as at Labastide) have been found in hearths, while
others have even been used in their construction, as at Mas d' Azil. 44 Breuil
and Begouen adopted the view that they had served as crude lamps, and
this may be true in some cases. However, a theory has also been put
forward that they represent a kind of heating device - sandstone has
thermal qualities and a resistance to tension which make it suitable for a
function of this kind. In France, old peasants in some areas still use heated
sandstone plaquettes, wrapped in cloth, as bedwarmers! 45 It may therefore
be thermal tension which shattered many stones, not some arcane ritual,
although a few (eg at Labastide and Enlene) do bear marks of blows
suggesting deliberate breakage.
It is difficult to assess how important these engraved stones were. In
sites such as Enlene, Gonnersdorf, Labastide and Enval there are many of
them (over 1,000, 500, 50 and 27 are known respectively), but there are
hundreds more without engravings in these sites. Enlene has tens of
thousands of plaquettes, brought in from a source some 200 m from the
cave; excavation has shown that an area of over five square metres of cave-
floor was paved with them, no doubt a measure against humidity, and this
paving includes engraved specimens; 46 a similar area of cave-floor at Tito
Bustillo (Asturias) had 83 plaquettes, 25 of which bore engravings ranging
from animal figures to simple incisions. 47 At the late Magdalenian open-
site ofRoc-la-Tour I (northernmost France), hundreds of schist plaquettes
were brought in as paving, but only about 10% have engravings. 48
Similarly, at Gonnersdorf, several tons of schist plaquettes were brought
to the open-air site as elements of construction and as foundations for
structures; only 5-10% of them were engraved, and these seem to be
distributed at random among the others. 49 This suggests that the
engravings lost all value once the ritual had been performed and they had
been broken and dispersed; or simply that the engravings never had any
ritual significance, and were done simply to pass the time, for practice, for
storytelling, or perhaps even to personalise one's private bedwarmer!
Future finds may help to clarify the situation further.
Occasionally, engraved specimens are stuck vertically in the floor- for
instance , around hearths at La Marche; and a few painted stones are
known from Enlene, Labastide, Parpall6, etc. Some are clear figures,
78 IMAGES OF THE ICE AGE

while on others simple staining by ochre in the soil may be involved: at

Tito Bustillo, it was thought at first that some plaquettes had paint on, but
careful study revealed that this was all contamination by pigments in the
soil; 50 the same phenomenon occurred on the engraved plaquettes of the
abri Durif at Enval, where red sand had coloured the side facing
downwards, but at LaMarche it is possible that some red traces on slabs
are paint, while others seem natural. 51
The best-known painted stones are those which characterise the very
end of the Palaeolithic in western Europe - the small 'Azilian pebbles',
first identified in the 1880s at the Mas d'Azil by Piette (see p.23). They
have been found in 28 sites in France, 5 in Spain, 3 in Italy and 1 in
Switzerland; but, of the nearly 2,000 known, over 1,400 are from the Mas
d' Azil. Their motifs, usually in reddish ochre, are simple (mostly dots and
lines), and seem to have been applied with the finger, less often with a fine
brush; but, as will be seen later (p .182), a recent study has produced
fascinating results from an analysis of the numbers and combinations of
these marks. 52
It will be recalled (see above, p.27) that painted animal figures are also
known on stones from the late Pleistocene Apollo 11 Cave in SW Africa,
while engraved pebbles have been found at Kamikuroiwa, Japan (p.28),
and non-figurative engraved specimens are quite common in the Azilian of
western Europe.53
Where engraved stones are concerned, it should be noted that, although
it is the fine figurative examples which tend to get published, there are far
more which are indecipherable, either because they are tiny fragments or
because they are non-figurative (at Roc-la-Tour I, about 600 small
engraved fragments have been recovered, but there are only 16 'readable'
figures so far 54). Some have a confused mass of superimposed lines (as on
the slabs of LaMarche, or the pebbles of La Colombiere), but experiments
with Gonnersdorf slabs show that a fresh engraving is very visible, due to
the presence of white powder in the incisions. When this is washed off, the
effect is like wiping chalk off a slate, and a new engraving can be made (the
incisions can be made quickly and easily, without effort55 ) - this suggests
that some of these engravings had significance for only a very brief time,
and the 'associations' of superimposed animals on a given surface are not
necessarily meaningful; on the other hand, there were plenty of stones
available, and each figure could easily have had one to itself if desired (as is
generally the case at Enlene, for example), so the superimpositions may
indeed have had some significance.

Painted bone, and engraving on bone and antler

Fig. 38 Reconstruction of the
Many ofthe above comments also apply to incisions on flat pieces of bone; mammoth skull painted with red
but experiments show that, unlike stone, fresh bone is hard to engrave: the ochre,fromMezhirich (USSR).
toaLtends to skid when it cuts bone fibres, and extremely sharp tools are Maximum width: 60 em . (After
required. Moreover, it is necessary to pass the tool backwards and Pidoplichko)
forwards , to widen the incision. One of the chief difficulties is that the
initial marks can barely be seen, though it has been found that covering the
bone with ochre beforehand makes incisions readily visible 56 - this may
explain why some Palaeolithic specimens, such as the bone-fragment from
Enlene with a grasshopper engraved on it, were covered in ochre when
found; 57 decoration of bones with pigment also survives occasionally- it
will be recalled that the mammoth-bone 'musical instruments' from Mezin
were painted with geometric motifs, chevrons, and undulations.
 FORMS AND TECHNIQUES 79

Fig. 39 Reindeer shoulder-blade

from the rock-shelter of Duruthy
(Landes) with a reindeer engraved on
it. Magdalenian . The animal
measures 9.5 em from muzzle to tail.
(PGB)

Fig. 40 Detail of a perforated

antler baton from Duruthy (Landes)
which bears a depiction in
'champ/eve' of two ibex, perhaps
fleeing a predator at left.
Magdalenian . The photo shows one
oftheibex (c 4 em long). (PGB)

Similarly, a mammoth skull from dwelling No 1 at Mezhirich, Ukraine (c

12,000 BC), is decorated with zigzags and dots of red ochre; engraved lines
on bone and ivory in eastern Europe are often highlighted with a filling of
black, manganese paste, whereas red ochre tended to be used for this
purpose in western Europe .58
Shoulder-blades are the bone equivalent of plaquettes, having a smooth,
large surface, and it is therefore surprising that they were not engraved in
great numbers ; nevertheless, decorated specimens are known from some
sites, and the Mas d' Azil and Castillo each have over 30, together with
undecorated ones; their engravings are both figurative and 'abstract'.
Those from the Mas d' Azil, at least, were clustered together in a small area
very poor in other finds. 59 At the abri Morin (Dordogne), 17 fragments
with figurative engravings have been found, and 29 with non-figurative
marks. 60
Although fibrous and relatively soft, such bones are by no means easy to
decorate: Jean Bouyssonie, the French prehistorian, found that it took
considerable muscular strength to engrave a fresh horse shoulder-blade,
and the flint point often broke.61 In some cases, the tool does not incise the
bone, but compresses its surface into a furrow.
Experiments also indicate that a burin is not the only tool which can
engrave bone, although it is the one which is always mentioned in this
connection. In fact, a wide range of stone tools are equally effective - awls,
Fig. 41 Engraved and carved pointed backed bladelets, and even the edge of a broken blade are just as
depiction of a bison calf, only a few good; it is the sharpness which counts, not the precise form. A copy of a
an in length, from Le Grand-Pastou small bison engraving on bone from La Madeleine, using different kinds of
(bmdes). Mfiltdalenian . (PGB) incisions and tools (which displayed no traces of wear afterwards), took
four hours; but a second attempt halved that time, showing that a practised
Palaeolithic craftsman would doubtless have taken little time to produce
this kind of image - similar results were obtained by Leguay, the pioneer of
this type of work, who, over a century ago, tried engraving on bone in the
prehistoric way using original stone tools from Palaeolithic sites, and
found that it could be done quickly with a little practice. 62 However, we
still have much to learn about the engraving of bone and antler, such as
whether specimens of different kinds and ages vary in their 'incisability' ,
or how special processes such as soaking may have affected the work.
80 IMAGES OF THE ICE AGE

Engraving on bone has a long history and early examples include: a

fragment ofbovine rib from an Acheulian layer at Pech de l'Aze, dating to Fig. 42 Tracing offigures (deer,
c 300,000 BC, bears an intentional engraving comprising a series of horse, isard, ibex, aurochs and
human) engraved around the gannet
connected double arcs, while several bones from Bilzingsleben (East bone from Torre (Guipuzcoa) .
Germany), of about the same date, have geometric designs engraved on Magdalenian . Length: c 16 em.
them which are clearly purposeful, and nothing to do with cutting or (After Barandiaran)
working; 63 a Mousterian fragment from the shelter of Riparo Tagliente,
Italy, also has a double arc incised on it; 64 La Quina, which, as mentioned
above, had two perforated pieces in its Mousterian, also yielded a bovid
shoulder-blade engraved with very fine, long parallel lines, while a
Neanderthal burial at La Ferrassie (Dordogne) contained a small bone
with a series of fine , intentionally incised marks which reminded the
excavator of the notched bones of the Aurignacian - similar fragments with
regular notches are known from the Mousterian of other European sites
such as Cueva Morin and several sites in Charente; 65 the Bulgarian cave of
Bacho Kirohas a Mousterian bone-fragment with a zigzag motif engraved
on it;66 while a Mousterian layer at Molodova-I, USSR, more than 40,000
years old, has a mammoth shoulder-blade decorated with little pits,
patches of colour, and notches that form complex patterns including
cruciform and rectangular figures in which some Soviet scholars have seen
the outline of an animal. 67 It will be recalled (p.27) that a baboon fibula
with 29 parallel engraved notches has been found at Border Cave in
southern Africa, and dates to 35,000-37,500 years ago. When faced with
examples of this kind, some scholars believe a genuinely continuous
tradition of marking is represented, while others see them simply as
sporadic recurrences of simple motifs.
In any case, only the Upper Palaeolithic has so far produced definite
figurative . designs, and these become particularly abundant and
im~ssive in the Magdalenian. The technique of 'champlew? was
invented, where bone around the figure is scraped away, making the
design stand out as in a cameo (see Fig.40); and the skill was developed of
engraving on bone shafts and on batons of antler, not only lengthwise
(there are numerous compositions involving lines of animals, or heads) but
also around the cylinder . Here, amazingly, perfect proportions were
maintained, even though the whole figure could not be seen at once: the
finest examples include that of Lortet, with its deer and salmon; that of
Montgaudier, showing seals and other figures ; and the gannet bone from
Torre, Spain, with its fine collection of human and animal heads (Fig.42)68
FORMS AND TECHNIQUES 81

- all decorated bird bones belong to the Magdalenian, as do most

depictions of birds.
A certain number of regional differences in these engravings have been
apparent since Piette's time 69 (although inevitably there are exceptions to
the rule): Dordogne specimens often have very deep incisions, almost bas-
reliefs, whereas in the Pyrenees and Cantabria engravings are generally
finer, with a mass of detail provided of the animals' coats. Scenes or
'tableaux' are also more common in the Dordogne than elsewhere.
It should also be noted that the art of engraving on teeth, already seen in
the simple motifs incised on pendants, was further developed in the
Magdalenian, as in the series of bear canines from Duruthy (Landes), with
their engravings of a seal, a fish, of 'harpoons', etc. 70

Carved bone and antler

As mentioned earlier (p.26), a carved and engraved bone is known from
the Pleistocene of Mexico; but once again it is Europe, and particularly the
Magdalenian of France, which can boast the finest and most numerous
specimens of the art. Apart from the perforated and sectioned bones
mentioned above, one of the earliest examples is the 'phallus' from abri
Blanchard, carved in a horn-core. In the Gravettian, tools and weapons
began to be decorated with both figurative and geometric motifs/' though
this practice really came into its own in the Magdalenian- for example,
almost all portable decoration in Cantabrian Spain comes from this period,
and increases through the period. However, it should be noted that much
of the simplest 'decoration' of tools and weapons- particularly transverse
incisions near the base - was probably intended to strengthen the
adherence to the haft, no doubt aided by gum or resin, and to improve the
grip of the user: it has been called 'technical aesthetics'. 72
Presumably, decorated objects were for long-term use, since there is
little point in investing time and effort in engraving an implement which
can be easily lost or broken- but this is not an absolute rule, since some
sites, such as abri Morin, have 'harpoons' and spear-points with carefully
made, figurative engravings. 73 It is worth noting that Garrigou interpreted
the small, perforated Azilian harpoons of La Vache (Ariege) as ear-
pendants! Semi-cylindrical rods of antler were carved with a variety of
motifs, such as the well-known 'spiral' decoration found in a cluster of
Pyrenean sites; perforated antler batons were also decorated, especially in
the Magdalenian. Antler, unlike bone, is relatively easy to engrave.
In the mid-Magdalenian, one encounters figures of animals and fish
('contours decoupes' or 'perfiles recortados'Y 4 and circular discs
('rondelles' or 'rodetes') 75 cut out of thin bone. The discs were often cut
from shoulder-blades, and several examples of the latter are known with
circles removed from them (see Fig.44); many are engraved, either with
animal or human figures or with abstract designs like sun-rays, and some
have tiny perforations round the circumference (Fig.43). Those with a
central perforation have occasionally been interpreted as buttons, which
seems an unlikely function, in view of their fragility. Similar discs are also
known in other materials: for example, the grave of Brno II
(Czechoslovakia), dating to c 23,000 BC, had specimens not only in bone
but also in stone, ivory, and cut/polished mammoth molar/6 while
Gonnersdorf has some in slate.
Animal figures are occasionally large, such as the 22 em bison from
Isturitz, found in two pieces 100m apart, and probably cut from a pelvic
82 IMAGES OF THE ICE AGE

bone; 77 but the majority of the (approximately) 150 'contours decoupes'

known are animal heads (about two-thirds of them horses) cut from a horse
hyoid (bone of the tongue), the natural shape of which already bears some
resemblance to a herbivore head .78 Many are perforated- some through
the nostril or eye, presumably for figurative effect, and others probably to
serve as pendants - and they have differing degrees of detail engraved on
them: eyes, muzzle, coat, and so forth (Fig.45).
Almost all of them have been found in the French Pyrenees, although
sites in Asturias and Cantabria have recently produced a few very fine
specimens of exactly the same type . No doubt the most outstanding find,
hidden in a corner of the cave of Labastide (Hautes Pyrenees), is what
seems to be a necklace of 19 identical perforated heads, apparently of the
isard (Pyrenean chamois) with its cold-season markings, together with one
perforated bison head, all cut from horse hyoids/ 9 this remarkable
ensemble was clearly made by a single artist.
It is worth noting that , like portable art as a whole, the distribution of
'contours decoupes' is extremely uneven: of those in France, about two-
thirds come from three Pyrenean sites (the 'supersites' Mas d'Azil and
Isturitz, plus Labastide because of its necklace); if those from Arudy are
added, it means that over 75% come from only four sites. 80 Bone discs are
rather more widespread , but over half of them come from the Pyrenees,
with Isturitz and the Mas d' Azil again the richest sites.
Figs . 43/44 Engraved bone disc,
The antler spear-throwers of the Magdalenian tend to have two kinds of
with a central perforation and 16
decoration: animal heads or forequarters carved in relief along the shaft (a others around the edge, from Le Mas
type found in both France and Switzerland); or figures carved in the round d'Azil (Ariege). The engraved motif
at the hook-end of the object, where the roughly triangular area of may be non-figurative.
available antler dictates the posture and size of the carving. 81 However, Magdalenian . Maximum diameter:
within these constraints the artists produced a wide variety of images - 5 em. (JV). Below: a shoulder-blade
fighting fawns, a pheasant, mammoths, a leaping horse and so forth from which a bone disc has been cut.
(Figs.18,103). 82 Le Mas d' Azil (Ariege).
Many of the finest of these carvings have been found in the Pyrenees, Magdalenian. Total length: 24 em.
and none finer than the intact spear-thrower from Mas d'Azil with its (JV)
image of a young ibex which stands, turning its head to the right and
looking back to where two birds are perched on what seems to be an
enormous turd emerging from its rear end (see Fig.46) ; this composition is
all the more startling because of the almost identical specimen found a few
years later at Bedeilhac, a few miles away- this one had lost its shaft, and
the ibex is kneeling and turns its head to the left, but otherwise is identical
in all respects . Broken specimens have also been tentatively identified
from other sites in and near the Pyrenees , with the result that up to ten
examples are known; if one allows for preservation, recovery, recognition
and publication it becomes obvious that these must represent a tiny
fraction ofthe dozens - perhaps hundreds- originally produced . One can
the.Jij[ore argue for a high output by an individual artist or a small group of
artisans with a favourite theme , since all the examples are attributed to the
Middle Magdalenian , a period which spans a few centuries in the
Pyrenees. 83 Even more remarkable is the virtually identical pose struck by Fig. 45 Bone 'contour decoupe' of a
Bambi and two birds (though without the turd! ) in the Disney cartoon horsehead, from Enlene (Ariege) .
made before the two intact spear-throwers were discovered .84 Magdalenian . Length: 5 em . (JV,
collection Begouin)
Statuettes and ivory carvings
The simplest free-standing figurines known from the Upper Palaeolithic
are terracotta models; their existence originally came as a surprise, since
 FORMS AND TECHNIQUES 83

Figs. 46/47 Antler spear-thrower one of the dogmas of archaeology was that fired clay was invented later, in
slwwing ayoung ibex with emerging the Neolithic period, and that the people ofthe Ice Age were incapable of
IUrd, on which two birds are perched. making it. This was clearly nonsense, as shown by the mastery of clay in
LeMas d'Azil (Ariege) . that period (see below, p. 93) and by the fact that any fire lit on a cave-floor
ma.~~aaumza'n. Total length: 29.6 em.
will have hardened the clay around it: indeed, lumps of fired clay around
aVJ. The detail shows the other side hearths bear a marked resemblance to crude potsherds. The recent
oftheibex. Length: c 7.5 em. (JV) .
discovery that pottery in Japan dates back at least 12,000 years merely
resembles a pose in Walt
s 'Bambi', featuring the fawn underlines the fact that, if Ice Age people did not make pottery, it was
two birds: this cartoon feature through lack of need rather than through ignorance.
released to movie theatres in Nevertheless, a number of terracotta figurines have survived and been
. TheM as d'Azil spear-thrower recovered in different areas: a few examples are known from the Pyrenees,
discovered between 1939 and North Africa, and Siberia; 85 but considerable quantities- 77 fairly intact,
, but the first illustration of it together with over 3,000 fragments (and others poorly fired, which have
published only in 1942 (by M. disintegrated) - have been found in Czechoslovakia, at the open-air sites of
S-J. Nquart in 'Ia Revue Dolni Vestonice, Pavlov and Predmost, where they are securely dated to
volJ,_O, Feb . 1942); the the Gravettian, c 22,400 BC. They comprise small figurines of animals and
not yet seen 'Bambi' but humans , and display some spatial differentiation (ie herbivores in one hut,
awmparison(pp. 9415) with
but human figurines in the centre of another, together with carnivores); a
in Disney's 'Snow White'
. In a subsequent publication hearth or 'oven' for their manufacture has also been found. 86 The best
p. 299) they also refer to known of these figurines is the 'Venus' of Dolni Vestonice, made of a
and reveal that illustrations mixture of clay and bone powder.
carving and ofBambi had been The Upper Palaeolithic also produced carvings in other materials such
side by side in the archives of as soft stone, as at Isturitz and Bedeilhac (figurines which seem to have
Louvre as a comparison! been deliberately and systematically broken), 87 and occasionally in
steatite, coal, jet or even amber; 88 there are also great numbers of 'pierres
84 IMAGES OF THE ICE AGE

figures', entirely natural objects which bear a fortuitous resemblance to Fig. 48 Drawing of a stone
something figurative - it is an open question as to whether this statuette of a kneeling horse>Duruthy
resemblance was noticed or considered significant by Palaeolithic (Landes) . Magdalenian. (After
people. 89 Laurent)
Sculpture is almost impossible in bone, which is hard and fibrous.
Consequently, the great majority of figurines are in limestone, sandstone
or ivory. The rock-shelter of Duruthy, Landes , has yielded fine horse
carvings in each of these materials (see Figs.48,51 ),90 while the cave of
Lourdes produced the best-known ivory horse. Human figures were
carved out of horse teeth at Mas d' Azil and Bedeilhac.
The famous 'Venus' ofWillendorf is made of a particular type of oolitic
limestone not found in Austria and was thus brought in from elsewhere. 91
Let us note that this figurine, like a few others, still bears traces of red
ochre. Stone figures were presumably carved with powerful flint tools, and
some traces of the process can occasionally be seen, as on the 'Venus' of
Tursac, made from a pebble.
Mammoth ivory was used quite extensively in the Ice Age, not only for
statuettes but also for beads, as we have seen, and for bracelets and armlets
such as those found on the Sungir skeletons, or the various objects of
Mezin (USSR) with their rich decoration of chevrons, zigzags and other
geMit'letric motifs.
Ivory is easy to engrave along its fibres , but not across them; a kind of
'champleve' could be achieved, as in the Aurignacian human figure with
raised arms from Geissenklosterle, Germany.92 However, three-
dimensional figurines were mastered in this material at a very early stage:
indeed, some of the earliest known pieces of Ice Age art are ivory
statuettes, including the human head and female figures from
Brassempouy (Landes), the animal and human figurines from Vogelherd
(Germany), and those of Geissenklosterle (fragments of two mammoths
and a feline).
 FORMS AND TECHNIQUES 85

Scholars have usually assigned the Brassempouy statuettes to the

Gravettian, simply because one or two 'Venuses', such as that of Tursac
and perhaps that of Lespugue, came from a layer of that period; 93 among
the very few 'Venuses' in western Europe with a stratigraphic context,
there are also specimens from the Magdalenian (eg Angles sur l'Anglin94),
and consequently all 'Venus' figurines have been unjustifiably lumped
into either the Gravettian (for the most part) or the Magdalenian. It is
extremely difficult to discover from Piette's excavation reports precisely
where the Brassempouy statuettes came from (assuming that those he
found came from the same layer as those dug out like potatoes by earlier
workers); but Breuil, who visited Piette's dig in 1897, saw the relevant
layer, and kept samples of the associated industry, stated quite clearly on
several occasions that the figurines came from the earliest Aurignacian,
and perhaps even the Chatelperronian. 95 In short, Palaeolithic figurative
art produced some of its greatest masterpieces- the Brassempouy head,
the Vogelherd animals, 96 and the male statuette from Hohlenstein-Stadel,
Germany (Fig.49),97 which dates to c 30,000 BC - in its initial phase, which
suggests strongly that they must have been preceded by a long tradition in
carving materials which have not survived.
Experiments have been carried out in carving 'Venuses' in ivory as well
as in stone and other materials; to make an ivory figure, a piece of tusk was
first prised out by cutting two deep grooves into the material; this was then

Fig. 49 Ivory statuette of a human

with a feline head, from H ohlenstein-
Stadel (Germany), together with
drllWings of the figure before the face
was fitted (after Hahn).
Aurignacian. Height: 28.1 em.
(Photo Ulmer Museum)

0
86 IMAGES OF THE ICE AGE

rubbed into a rough shape with sandstone, and a burin and broken blade
were used to carve the legs and other areas. It is thought that long handling
of the figures will have smoothed away any striations left by the work. 98 Fig. 52 Figure of a human drawn
Carving ivory is very hard, and takes many hours, even days, of effort. around a natural phallus-like
Ivory statuettes were also made in composite form, as in the case of the stalagmite protruding from the wall
Czech male figurine from Brno II, dating to c 23,000 BC, whose head, in Le Porte! (Ariege). Probably
trunk and arm were fitted together to make an articulated 'doll'. 99 Magdalenian. Height: 38 em. (JV)

Fig. SO Two ivory figurines still

joined at the head, from Gagarino
(USSR). Probably Graveuian.
(After Tarassov)

0 2 3cm

Some idea of how the human figurines of the Soviet Union were made is
provided by the finds at Kostenki and other sites which are so numerous
that there are examples of every stage of manufacture, from rough-outs to
final polishing; 100 one piece of ivory recovered from Gagarino, a site which
has yielded eight ivory 'Venuses' , has been carved to form two human
figures of different lengths, attached at the head like Siamese twins. 101
Although it is possible that this may have some special significance
(especially since the two children at Sungir were buried head to head), it
seems more likely that the piece is simply two separate figurines, made
together , which have not yet been separated (Fig.SO) .
It is worth noting that there are regional differences in the material used
for the 'Venuses' and in their locations: in the Pyrenees, eastern Europe
and Siberia they are predominantly of ivory, while those of Dordogne are
in stone, those of Grimaldi are steatite, and those of central Europe employ
ivory, stone and terracotta. 102 Most western specimens have been found in
caves or rock-shelters, while almost all those from central and eastern Fig. 51 H orsehead carved in
limestone, from Duruthy (Landes).
Europe come from open-air settlements, and seem to have had a special
Magdalenian. Length: 7.1 em,
role in the home (see below, p.l40). height: 6.2 em. (PGB)
As mentioned above, many of those from western Europe, including
most Italian specimens, have no context, but the dated examples are
scattered through the entire Upper Palaeolithic . It is therefore a somewhat
pointless exercise to treat them as a homogeneous or contemporaneous
gn~,p, as has often been done; 103 but, treated simply as a category of
object, like spear-throwers or jewellery, they have a strange distribution: a
virtually uninterrupted spread over 3,000 km from Brassempouy (Landes)
to the numerous specimens from Kostenki and other Russian sites; and
then a 5,000-km gap between Kostenki and the Siberian sites (such as
Mal'ta) around Lake Baikal.
There are many such enigmas in the regional variations and the
distribution of different types of portable art. In some cases, it is the
availability of materials which is the cause: central and eastern Europe
clearly had great quantities of mammoth bones, which were often used for
 FORMS AND TECHNIQUES 87

the construction of dwellings , 104 and this helps explain the dominant use of
ivory for carving; the raw material's properties , shape and volume will, in
turn, have determined the nature and extent of the decoration or images.
Some materials (ivory, antler, certain stones) were suitable for sculpture,
others (bone, stone, antler) for engraving. In some cases, the shape placed
clear limits on the composition- figures on spear-throwers have already
been mentioned, and the same is true of antler batons, long bones or ribs
which tend to have linear arrangements of figures . Similarly, hyoid bones
were merely made to look more like animal heads than they already do ,
which explains why some' contours decoupes' are slightly too long or short
to be truly naturalistic.
The basic question, of course, is whether the composition was chosen to
fit the support , or vice versa: this is particularly pertinent to plaquettes

Fig. 53 Vertical bison painted at

Castillo (SaniiiiUler), utilising the
shape ofa stalagmite for its back, tail
and hind leg (with explanatory
drawing, after Ripoll) . Probably
Magdalenian . Length: 80 em. CJV)
88 IMAGES OF THE ICE AGE

Fig. 54 Deeply engraved motifs on

limestone block (60 em high) from
abri Cellier (Dordogne). The
principal oval is c 10 em long.
Aurignacian. These motifs have often
been interpreted as vulvas . (JV)
which , unlike bones, come in a wide variety of sizes. Was a composition
planned, and then a stone of suitable size selected? Or was the stone chosen
first , and the figure s made to fit? No doubt, both scenarios occurred
frequently, and it is sometimes impossible for us to tell . In a few cases,
however, reconstruction has cast light on the problem - at Gonnersdorf,
for example, some big slates, more than a metre in length, have been
reconstituted from over 50 fragments; on one of them, a large horse figure
could be seen only after the reconstruction, whereas some figures were
engraved on fragments, their form and size respecting the borders of the
plaquettes. Clearly, these slabs were being engraved at every stage of the
breakage process, and the fi gures corresponded to the size and shape of the
slates available. 105
Almost all the Gonnersdorf figures are perfect, with only a few badly
proportioned 'failures'. Occasionally, however, we can see apparent
mistakes in the engraving process (as distinct from multiple versions of
some features; see below, p .l84): for example, a pebble from Mas d'Azil
bears an engraving of a horse whose body turned out to be too big for the
stone , so that the head's position had to be altered accordingly. 106 In other
cases, when a drawing proved too big, it was continued on the sides and
even on the other side of the stone. If nothing else, slip-ups of this kind
show that our ancestors were all too human, and that Palaeolithic art does
not consist exclusively of masterpieces!

lwtermediate forms- 'pseudoparietal art'

As mentioned earlier, the division between portable and parietal art is
simply one of convenience, since there is an area of overlap principally
comprising blocks of stone which, although movable, could not be carried
around and which may in some cases be fallen fragments of decorated wall.
All are from well-lit rock-shelters.
Most of the best-known blocks are thought to date to the Aurignacian
although, if they fell from the walls on to Aurignacian layers, they could be
considerably older. 107 Certainly the decoration of blocks is known at the
FORMS AND TECHNIQUES 89

Mousterian site of La Ferrassie (Dordogne), where a Neanderthal child

burial lay beneath a large limestone rock with a series of small cupmarks
(mostly in pairs) carved in it and apparently placed at random , 108 yet
another example of the wide variety of symbolic behaviour in this early
period.
Some Aurignacian blocks have deep engravings (almost bas-relief),
others have paint on them . The best-known painted specimen is that from
the abri Blanchard, where the rock was painted red, and then an animal
figure (the bottom part of which survives) was painted on top, with a black
outline and a reddish-brown infill.
Occasionally- as in the case of an Aurignacian block from Ferrassie,
where the 'broken side' has been painted, not the smooth side which
formed part of the shelter vault 109 - it is clear that the artwork was done
after the fragment fell from the wall; but by and large one cannot be
certain. However, there are so many painted chunks of rock in these early
collapsed shelters that it seems highly probable that some ofthe back-walls
and vaults must have been painted (and perhaps engraved) over wide areas
- the fact that nothing remains on the walls themselves is simply an
indication of how remote this art is : around 30,000 years old, more than
twice as ancient as most surviving Palaeolithic parietal art is estimated to
be .
Observation and experimentation show that some of the blocks were
prepared by grinding and rubbing to produce flatter and better surfaces
for engraving. 110 Fine engraving is rare- most of the incisions are deep and
wide , some of them clearly made by crudely joining together rows of
cupmarks or hammer-blows , and all very weathered.
Brigitte and Gilles Delluc have devised a system of representing the
different techniques used on these blocks - as well as in parietal deep
engravings and bas-reliefs- in cartographic form, with different symbols
and conventions to indicate the depth of the incisions, their shape in
section (curved, angular, etc) and so forth. 111 Such 'maps' avoid many of
the problems mentioned in an earlier chapter concerning copies of the art,
and at the same time convey a great deal of information about the figure's
three dimensions.

Parietal art
The principal advantage of parietal art over portable is that, whereas the
latter may have been made far away from its final resting place, and indeed
in a period earlier than the layer in which it is found, wall-art is still just
where the artist placed it. As will be seen in a later chapter, this simple fact
provides us with a great deal of information. Like portable objects, parietal
art encompasses an astonishing variety and mastery of techniques.

Slightly modified natural formations

In a few cases, it is hard to tell whether the use of natural features was
intentional or not, but this is usually pretty clear: 11 2 for example, in the
bosses of the Altamira ceiling used as bulging bison bodies (Fig .7); the
shape of cave-walls and stalagmitic formations incorporated into animal
backs and legs at Cougnac (Fig.89), a bison back and tail at Ekain, or a bird
at Altxerri; the little stalagmites turned into ithyphallic men at Le Porte!
(Fig.52); the antlers added to a hollow at Niaux which resembles a deer
head seen from the front (Fig. 55), or, in the same cave, the engraving of a
90 IMAGES OF THE ICE AGE

Fig. 55 This fissure in the S alan

Nair at Niaux (Ariege) looks
something like a deer head seen from
the front) and the artist added two
antlers. Probably Magdalenian.
Total length: c 40 em. (JV)
Fig. 56 Detail of one of the
engraved owls at Trois Freres
(Ariege) [see Fig. 92]) showing the
striations in the lines. The head is
about 10 em wide. (JV) collection
Begouiin)

bison on the clay floor which began with, and was composed around, an
eye formed by a cupmark left by water droplets There are countless other
examples.

Finger markings
The simplest form of marking cave walls was to run one or more fingers
over them, leaving traces in the soft layer of clay or 'Mondmilch' (a white,
clayey precipitate of calcium carbonate): as shown by the recent finds in
Australia (p.28), the technique is extremely ancient, and may have been
the first used: it is certainly a method which requires neither great effort
nor any kind of tool. Finger-markings in the European caves may well
span the entire Upper Palaeolithic and even part of the Mousterian; most
scholars followed Breuil in attributing them to a very early phase, but
those of caves such as the Tuc d' Audoubert and the Reseau Clastres are
iJJnost certainly Magdalenian.
It is possible that in some cases the Mondmilch was actually removed in
this fashion- either for body decoration or for medical purposes, since in
some Alpine regions in historic times it was used as an ophthalmic
analgesic 113 - but more often the fingermarks are splayed, and the
Mondmilch compressed rather than removed, suggesting that marking
was the aim.
In Australia , the lines made by fingers seem to be purely non-figurative;
but in some European caves, such as Gargas and Pech Merle, they also
include definite animal and anthropomorphic figures (Fig.S9). 114
 FORMS AND TECHNIQUES 91

Engraving
One possible source of inspiration for the finger technique is the
abundance of clawmarks of cave-bear and other animals on the cave-walls;
in a few cases these seem to have been incorporated into designs, such as a
circle at Aldene (Herault), or a hand at Bara-Bahau (Dordogne) (a very
similar, well-carved hand is known in Koongine Cave, South Australia); 115
and it is thought that some marks on cave-walls, at La Croze a Gontran
(Dordogne) and elsewhere, are engraved imitations of these clawmarks. 116
Engraving itself, which, as in portable art, is by far the most abundant
technique on cave-walls, encompasses a wide variety of forms, the choice
of which was largely dictated by the nature of the rock; incisions range
from the fine and barely visible to the broad, deep lines already seen on the
Aurignacian blocks; scratching and scraping were also used at times.
The deep engravings on the walls of Pair-non-Pair seem to have been
done with stone picks, whereas the finer incisions of later periods can be
attributed to sharp instruments producing a fine, V-shaped section. Jean
Vertut was developing new techniques to visualise the profile of fine
engraved lines , using microphotographic analysis and even computer
enhancement of contrast and equidensities. 11 7 1t will be recalled (p.42) that
his macrophotography reveals the very slight damage done by direct
tracing methods, which affects our ability to assess the tools used in the
lllCISlOnS .
A robust tool must have been used for the deep engravings and bas-
reliefs in the caves of Domme (Dordogne), and therefore the flint burin
found in a narrow fissure in the Grone du Pigeonnier here, in the
immediate proximity of the parietal figures, may simply have been used
for initial sketches, or for final details. 118 In any case, such finds are not
necessarily always of relevance to the art: for example, a burin-scraper

Fig. 57 Engraving of a vertical

horsehead in clay, M ontespan (Haute
Garonne) . Probably M agdalenian.
Length: 15 em. (Photo C. Rivenq)
Fig. 58 'Kneeling' reindeer
and scraped on a wall in
Freres (Ariege). Probably
Mas!tlalenlxa:n. Length: c 20 em. (JV,
Begouen)
92 IMAGES OF THE ICE AGE

found in a fissure between the legs of the engraved feline of Trois Freres
was traditionally assumed to be the artist's tool, left there when the job was
done, but in fact nine other objects (flints, bones, a tooth and a shell) have
also been found in fissures in this little 'sanctuary' . 11 9
As in portable art , the burin was not the only possible tool for the
engraver: almost any sharp flint could have been used, from carefully
made retouched pieces to simple waste flakes. This is seen clearly at
Lascaux, where use-wear was found on the sharp angles of 27 stone tools,
including burins, backed bladelets, and simple blades and flakes. 120
Experiments showed that a similar wear and polish could be produced by
passing a tool up to SO times across the rock, thus making an incision 1 mm Fig. 59 Digital tracings at
or 2 mm in depth. These tools were found only in those zones of Lascaux Altamira (Santander) with depiction
which have parietal engravings; they were absent in parts which do not. A of bovine head, and tracing by
few engraving tools were also found at Gouy, though their use-wear was Breuil. The head is just over 1m in
less pronounced than on the Lascaux tools, no doubt because the rock is length . (JV)
softer.
will be recalled that fresh incisions on bone are almost invisible unless
the surface is covered with ochre; a similar suggestion has been made for
parietal engravings: namely, that the wall may have been coated with a
substance such as clay or blood so that new lines stood out from those
beneath. 121 It is the great numbers of engraved figures one on top of
another at Trois Freres or Lascaux which have led to hypotheses of this
type, but in fact they are exceptions to the rule: by and large, parietal
figures, whether engraved or painted, carefully avoid superimposition:
even at Lascaux, juxtaposition is far more common than overlaps, while at
Niaux there seems to be a positive concern to avoid pre-existing figures.
FORMS AND TECHNIQUES 93

In most cases, fine engravings are almost invisible when lit from the
front but 'leap out' when lit from the side. This fact is of some importance,
for it provides an indication of whether the artist was right- or left-handed.
Right-handed artists tend to have their light-source on the left, to prevent
the shadow of their hand falling on the burin (or brush), and accordingly
the majority of Palaeolithic parietal engravings are best lit from the left.
Occasionally, however, one comes across the work of a southpaw - for
example , the Pyrenean cave of Gargas has many engravings, including a
fine, detailed pair of front legs of a horse; these had been known and
admired for decades; but recently, visiting scholars lit the figure from the
right instead of the left and were suddenly confronted with the rest of the
horse, which nobody had seen before! 122 It is possible, of course, that
Palaeolithic artists used the lighting of engravings to their advantage,
making them appear or disappear to great effect- alas, we shall never
know.
In places where walls were too rough for fine incisions, engraving
became almost a form of painting: it was either done more crudely, so that
it is best lit from the front, or it was done by scraping. In both cases, the
image came not from the relief and light/shadow of an incision, but from
the difference in colour between the white engraving and the darker
surrounding area of panel. This was the technique used on the Trois
Freres feline, mentioned above, and on many other figures. 123
There are also different regional/chronological styles of engraving. The
best example, one easily recognised and well dated (see p.S8), is 'striated
engraving', using multiple traces, found in a wide area of northern Spain,
both in portable art (Parpall6, Rascafio, El Cierro, Castillo, Altamira, etc)
and in parietal (Altxerri, Castillo, Altamira, Llonin, etc). 124 However, it is
dangerous to assume that every use of a technique such as this belongs to a
single chronological phase.

Work in clay
A different example of regional specialisation is work in clay, which (apart
from the terracotta figurines discussed above) is restricted to the Pyn:nees
on present evidence - presumably because the limestone of most of the
Pyrenean caves other than Isturitz was unsuitable for sculpture or bas-
relief.
The simplest forms of clay decoration were finger-holes, finger-
tracings, and engraving with tools in the cave-floor or in artificially set-up
banks of clay. Patterns of dots made by fingertip have been found in the
floor at Fontanet (Ariege), and apparently random holes were punched by
fingers into the frieze on clay at Montespan (Haute-Garonne), which itself
comprises horse figures and vertical lines drawn by finger. One of the
finest engravings known is the vertical horse-head which begins this frieze
(Fig.57). 125
Figures and signs were drawn in the floor at Bedeilhac and elsewhere,
but by far the best-known cave-floor engravings are those of Niaux, 126
mostly located around the Salon Noir; apart from the bison mentioned
above , they include figures such as two fish, horses, an ibex head, an
aurochs, and something vaguely resembling a human fist. Drawings in
clay at Massat include three heads of isard (the Pyrenean chamois).
Such drawings, for the most part, have survived only under rock
overhangs or in recesses where the feet of unwary visitors could not reach
them . It is therefore almost certain that originally there must have been
94 IMAGES OF THE ICE AGE

many more of them, and in far more sites, although at present they do seem
to be limited to the Pyrenees, like other work in clay. Untold numbers of
fine figures have probably been obliterated by casual visitors to caves in the
days before Palaeolithic art became known; only in caves such as Fontanet,
the Tuc d' Audoubert and Erberua, where the utmost care has been taken
since their discovery, and visits severely restricted, can one be certain that
all existing ephemeral traces of this kind have survived.
Since, as we have seen, Palaeolithic people could model clay into
figurines, it is hardly surprising that they were equally capable of parietal
modelling. Bas-relief exists in cave-floors, as in the series oflittle horses at
Montespan (some almost in haut-relief, since they rise to 9 and 10 em in
height), and also on banks of clay: the best-known example is in a side-
chamber at Bedeilhac, where there were four bison (two of them now
destroyed), with other marks and a carefully modelled and accurate vulva
nearby, with a small fragment of stalactite inserted at the position of the
clitoris (Fig.60). 127
Montespan has some broken haut-relief figures in clay, including at
least one possible feline (thought by some scholars to be a horse), but they
are so damaged that they pale into insignificance beside the classic
examples of the technique : namely the two bison of the Tuc d' Audoubert
(Figs . 61-2), which are about one-sixth normal size (63 and 61 em long)
and placed at the centre of a distant chamber, against some rocks. Around
them are marks including heel-prints, a crude engraving of a third bison in
the cave-floor, and a small clay statuette of a fourth.
The clay was brought from a neighbouring chamber; some 'sausages' on
the ground have traditionally been interpreted as phalluses, or

Fig. 61 The two clay bison of the

Tuc d' Audoubert (Ariege) in their
context, as the focal point of a small,
low-ceilinged chamber. Probably
M agdalenian. The bison are 63 and
61 em long respectively. (Wide-
angled photo by JV, collection
Begouen) . [See also Figs. 120-22]
Fig. 60 Clay bank in the cave of
Bedeilhac (Ariege), showing
modelled vulva on the left, and a bas-
relief bison about 30 em in li!ngth .
Probably Magdalenian. (]V)
Fig. 62 Detail of one of the Tuc
d'Audoubert bison, showing how the
clay was worked with fingers and
implements. (JV, collection Begouen)
FORMS AND TECHNIQUES 95

occasionally as horns, but a recent study by a sculptor 128 suggests, far more
plausibly, that they are simply the result of testing the clay's plasticity, and
the position of palm- and finger-prints on them supports this. Similarly,
marks on the haut-relief bison show that they were modelled with fingers
and further shaped with some sort of spatula (Fig. 62); a pointed object
was used to insert the eyes, nostrils, mouths, mane, etc. It is likely that the
cracks in the figures occurred within a few days of their execution, as they
dried out. Most of the artist's detritus around the figures seems to have
been carefully cleared away.
Finally, one fully three-dimensional large clay figure is known: the
famous bear at Montespan. However, although a fully rounded form, in a
Sphinx-like posture, it cannot really be classed as a statue- it is simply a
bear-shaped headless dummy, 1.1 m long and 60 em high, over which the
hide (perhaps with head still attached) of a real bear was probably
draped . 129 This would account for the polish on the clay. Nevertheless, a
great deal of work went into shaping this considerable mass of clay
(c 700 kg), and its haunches and stomach area are particularly fine
(Figs.63-64); its claws were also incised into the clay paws.
96 IMAGES OF THE ICE AGE

Bas- and haut-relief in stone

Just as work in clay is restricted to the Pyrenees, so parietal sculpture is
limited to other parts of France such as the Perigord, where the limestone
could be shaped in this way. One difference between the two techniques is
that clay figures are found (or have only survived) inside the dark depths of
caves; sculptures are always in rock-shelters or the front, illuminated parts
of caves. The reason for this discrepancy remains obscure, since the artists
were clearly capable of working for long periods far inside caves and, as we
shall see, had light-sources adequate to the task.
Among the earliest examples of parietal stone-carving are the bas-reliefs
ofLaussel (Dordogne), including the well-known woman holding a 'horn'
in her right hand ; this particular specimen was originally carved on a large
block of four cubic metres in front of the rock-shelter, which was not,
therefore, a 'movable wall'; 130 the figure includes a number of techniques -
her left side is merely deeply engraved, like the Aurignacian blocks
mentioned earlier, while her right is in 'champ]eve', since material has
been removed next to it which brings her body into relief. The figure
seems to have been regularised and perhaps even polished; finally, fine
engraving was used for her fingers and for the marks on the horn
(Fig.67). 131
The carvings ofFourneau du Diable (Dordogne), on the other hand, are
on a block of only half a cubic metre, weighing less than a tonne, and had
therefore been placed precisely where the artist wanted them, 132 while the
blocks of Roc de Sers (Charente) had originally been arranged in a semi-
circle at the back of the rock-shelter, but subsequently some had rolled
down the talus slope.
One of the best examples of haut-relief is the frieze ;it Cap Blanc
(Dordogne) (Fig.69), where figures reach a depth of 30 em in places. By
and large, we do not know precisely what implements were used for work
of this kind, although it is a safe assumption that they were quite robust
percussion tools (hammers , chisels, saws and abrasives), judging from the
traces of impact still visible on and around many sculptured figures. Few
of the sites have yielded plausible tools in their fill, but at Laussel some
Aurignacian 'picks' have battered ends; the picks of Angles-sur-1' Anglin
were mentioned earlier (see above, p.S8), and worn flint tools including
burins and scrapers were recovered from the Magdalenian of Cap Blanc. 133
The engraving on the Laussel female has echoes in incised lines on the
sculptured blocks at Roc de Sers; far more common, however, are traces of
pigment on these carved figures, similar to those on portable carvings of
various types. The Laussel woman has red ochre on her whole body,
especially visible on the breasts and abdomen, as well as in the groove
around her, and some of the other Laussel figures have traces too . In the
magnificent sculptured frieze of Angles-sur-l'Anglin (Vienne) many
f!i!ues (including the male portrait) have traces of paint, and vestiges of
red ochre have also been found on the fish in the abri du Poisson, and on
the friezes of La Chaire aCalvin (Charente) and Cap Blanc- indeed, at the
latter the ochre seems to cover large areas, both on the figures and around
them, but can now be seen only under very intense light or when the wall is
humid. 134
Pigment was also used in combination with engraving, as at Pair-non-
Pair, where there are many traces of ochre inside the incisions as well as on
the areas in relief; but in many sites it is difficult to decide whether the
 FORMS AND TECHNIQUES 97

engravings were 'redone' with colour or whether they were originally done
on painted surfaces, thus trapping some pigment in the grooves .
Pigments
Identification and preparation
As soon as Palaeolithic parietal art began to be accepted, at the end of the
last century , analyses were undertaken to identify the pigments used:
Riviere, the pioneer of La Mouthe , was the first to take samples , in 1898,
and had them analysed by Adolphe Carnot, who found that they were iron
oxide, with no trace oflead or mercury, and had been applied in the form
of a watercolour which had penetrated the rock; later , samples from
Gargas and Marsoulas, submitted by Regnault to C. Fabre, were also
identified as an iron-oxide paint. 135 H . Moissan did analyses for Font de
Gaume in 1902, with similar results.
Since that time, the red pigment on cave-walls has consistently proved
to be iron oxide (haematite, or red ochre), not only in western Europe but
also in remote caves like Kapova and Cuciulat. 136 Black, on the other hand,
is usually manganese dioxide, although recent analysis of pigment in the
Salon Noir, Niaux, has revealed that charcoal was used for figures here:
the plant cells are visible under the microscope , while the scanning
electron microscope has shown that it was a resinous wood similar to the
juniper; charcoal may also have been used instead of manganese at Las
Monedas . 137 Charcoal marks are known elsewhere, such as at Altamira,
and the Grone Bayol (Gard), but these may simply be torch-marks rather
than deliberate drawing.
White was used far more rarely; kaolinite was brought into the abri
Pataud (a non-painted site , as far as we know) and may have been used as a
white pigment or as an 'extending pigment' mixed with other colours; 138 at
Altamira, some white paste found in a shell proved to be a mix of mica and
illite, 139 and one nodule from Lascaux is also thought to be kaolin.
It has often been assumed that the Palaeolithic artists also used blues and
greens extracted from plants, and that these have not survived on the walls
-however, it requires complex chemical treatments developed in modern
times to extract these colours, and the plants concerned grow in warm or
tropical areas; so, apart from the charcoal mentioned above and the
possibility of a little woad, plants were probably not used for pigment. 140
Similarly, copper and lapis-lazuli were unknown as sources.
Consequently, Palaeolithic artists had only five basic colours to work with:
red, yellow, brown, black and white, but the latter is so rare that the choice
was really of four .
The main mineral colouring materials were usually readily available,
either casually collected as nodules or exploited from known sources (or
even occasionally mined, as we have seen in Africa and Hungary); lumps
of them have been found in abundance in some sites, and tens of kilos of
them , bearing traces of how they were used, lie unstudied in museums.
At Tito Bustillo, in front of the great Magdalenian painted frieze, there
were colouring materials, some of them still in the barnacle shells in which
they were mixed. In the far earlier cave-site of Arcy-sur-Cure (Yonne),
dating to the very beginning of the Upper Palaeolithic, there was a
limestone plate which had clearly been used for carrying ochre.
However, the greatest assemblage of evidence of this kind, and the best
studied so far, is that from Lascaux, where 158 mineral fragments were
98 IMAGES OF THE ICE AGE

Figs. 63/64 The front and right

flank of the clay bear of M ontespan
(H aute-Garonne), about 1 .1 m in
length. Probably Magdalenian.
(PGB). The view of the 'dark side'
(near the cave-wall) of the bear shows
the left forepaw and shoulder, and
the finely modelled belly and haunch.
Note the traces of red pigment
nearby. (PGB)
 FORMS AND TECHNIQUES
99

Fig. 67 'Venus with horn' from

Laussel (Dordogne). Probably
Gravettian. Height: 44 em. Although
now detached, it should be classed as
parietal art since originally it was
carved on a block of4 cubic metres.
aVJ

Figs. 65/66 The sculptured bas-

relief horsehead of Comarque
(Dordogne) , 70 em long. Probably
Magdalenian . (JVJ. Detail of the
....
horse's muzzle. (PGB)
Fig. 68 Red pigment in the cave of
Cougnac (Lot). (JVJ
100 IMAGES OF THE ICE AGE

found in various parts of the cave, together with crude 'mortars' and
'pestles', stained with pigment, and naturally hollowed stones still
containing small amounts of powdered pigment. 141 There are scratches
and traces of use-wear on 31 of the mineral lumps. Black dominates (105),
followed by yellows (26), reds (24) and white (3). It was found that there
were sources of ochre and of manganese dioxide within 500 m and 5 km of
the cave respectively.
The shades vary considerably: the colour of ochre is modified by heat,
and Palaeolithic people clearly knew this, since even in the
Chatelperronian of Arcy there were fragments at different stages of
oxidation still in the hearths. Yellow ochre, when heated beyond 250C,
passes through different shades of red as it oxidises into haematite. 142
A further stage in pigment preparation, in Lascaux at least, involved the
mixing of different powdered minerals, since unmixed pigments are rare
here . Chemical analysis of ten samples produced some surprising results:
for instance, one pigment contained calcium phosphate, a substance
obtained by heating animal bone to 400C. It was then mixed with calcite,
and heated again to 1000, thus transforming the mix into tetracalcite
phosphate. 143 A white pigment was found to comprise porcelain clay
(10%) , powdered quartz (20%) and powdered calcite (70%); while a black
pigment found in a mortar was charcoal (65%) mixed with iron-rich clay
(25%) and a few other minerals, including powdered quartz. Clearly,
therefore , the Palaeolithic artists were experimenting and combining their
raw materials in various ways. Analysis is now proceeding further, and the
scanning electron microscope is being used on tiny chips of some Lascaux
paintings in order to study the microstructure of the pigments and
compare them with natural samples, in an effort to assess the nature and
extent of the prehistoric processing. 144
A different problem is the binding medium used . In the past it was often
assumed that some form of fatty animal product was used for the purpose;
however, a series of 205 experiments in two caves has been carried out by
Claude Couraud , involving a variety of pigments and binding substances
(including fish glue, arabic gum, gelatin, egg white, bovine blood, and
urine), and a range of wall-types and degrees of humidity. Observation of
the results and deteriorations over three years led him to the conclusion
that fatty and organic substances were totally unsuitable binding agents,
and fail to adhere well to humid walls. In fact, the only substance which
seemed to be good at fixing and preserving the pigments on the rock-face
was water- especially cave-water, which is rich in calcium carbonate and
which was probably used at Lascaux. 145 It was also found that pigments
adhered better if they had been finely ground.

ApPJ!cation
The simplest way to apply pigment to walls was with fingers, and this was
certainly done in some caves : for example, the animal figures of La Baume-
Latrone (Gard) were clearly painted with fingers, as were the
'serpentiformes' of La Pileta (Malaga), versions in red clay or black
pigment of the simple finger-markings or 'macaronis' which also abound
in these caves. 146 La Pileta and other caves such as Cougnac also have series
of double finger-marks in pigment on or around animal figures. 147
Experiments suggest that painting by finger tends to produce poor
results. Nevertheless, when Lascaux II was being made, it became
FORMS AND TECHNIQUES 101

apparent to the modern artists that some painted lines had been done with
the finger in the original figures.
Normally, however, paint was applied with some sort of tool; again,
since none has survived, it is experimentation which suggests what was
used, although one must always remember that the original artists had
years of experience and familiarity with their materials, and acquired
'knacks' which cannot be revealed by modern short-term exercises. Some
lumps of pigment are in the form of'crayons' (there are 19 with use-wear at
Lascaux alone), and these may have been used to sketch outlines;
however, they do not mark the rock well, and really work only on humid
walls . Moreover, they wear down very fast - in tests, 1 em of length
produced a mark of only 15 em! 148 Consequently, lumps of pigment must
have been used mostly as sources of powder, which explains the traces of
scraping on many of them- and flints with ochre on their edge have been
found at Lascaux, Tito Bustillo and other sites, close to the minerals.
As mentioned earlier, mortars and other 'vessels' have been found at
Lascaux containing crushed pigment, and similar objects with traces of
colour inside are known from Altamira (including some vertebrae), Villars
(concave fragments of calcite), Tito Bustillo (barnacle shells) and
elsewhere. What we do not always know is whether the powder was made
into a paste or a liquid- no doubt it depended on the circumstances and the
desired effect, whether a dot, an outline or a flat-wash. In some cases, the
paint was probably in liquid form, since it is diffuse, or it has run - for
example on some figures of the Frise Noire at Pech Merle - although
trickling may also have been caused by wet walls at times .
Whether in paste or liquid form, how was the paint applied?
Experiments with a variety of materials produced their best results - ie
solid, precise and regular marks- with animal-hair brushes (especially
badger); brushes of crushed or chewed vegetable fibre were next best. On
the other hand, brushes of human hair were too supple and fragile, while
pads of bison fur transferred colour to the rock efficiently, but quickly
became flaccid and unusable. 149
Nevertheless, there are cases where some sort of pad must have been
used because other methods were unsuitable. Some surfaces at Lascaux
have a cauliflower-like covering (the crystalline calcite coating of the Hall
ofthe Bulls made it unsuitable for engravings), and once figures had been
outlined on them- presumably by crayon or brush- the infill was done
with hundreds of circular, diffuse spots that join up and give the
impression of an even wash. They appear to have been done with a pad and
dampened powder: for example, a red cow in the 'Diverticule Axial' has
been filled by about 200 such patches . Yet these figures have sharp edges,
which suggest that a hide or some such object was placed along the desired
line . 150 This pad-and-paste method would certainly have been easier to use
on ceilings and inclined walls than brushes and liquid paint. In some lines
of big dots in various parts of Lascaux, one can see a change after two or
three similar ones, which suggests that the pad or brush held enough
pigment for only a few at a time.
In some cases, such as the Covalanas and Arenaza deer (outline figures
made of dots) or the famous bison of Marsoulas, composed almost entirely
of red dots, each 2-3 em in diameter, 151 it is hard to decide whether fingers,
thumbs, pads or brushes were responsible . As mentioned earlier (p.48),
use of infra-red film has revealed that some apparently unbroken lines ,
such as in a tectiform at Bernifal, were originally made with dots whose
102 IMAGES OF THE ICE AGE
 FORMS AND TECHNIQUES 103

Fig. 69 Photomontage of the

sculptured frieze at Cap Blanc
(Dordogne), showing the series of
horses, some facing left and others
right. Magdalenian. Total length: c
8m. The central horse is 2.15 m long.
(TV)
Fig. 70 Three hinds drawn with
dots at Covalanas (Santander).
Total length: c 2.5 m. The central
hind is 60 em long. (JV)

Fig. 71 Some of the incomplete

hand stencils of Gargas (Hautes
Pyrenees) . (JV)
104 IMAGES OF THE ICE AGE

pigment ran together, because the earthy red pigment was soluble in
humid conditions and spread (unlike manganese dioxide, which is
insoluble and thus does not run and fade) . At Combe! (Pech Merle), infra-
red analysis showed that a group of red dots had been made with a pad
dipped in ochre, whereas an adjacent set had been applied by spraying. 152

Hand stencils
A technique of spraying paint was also clearly employed for the hand
stencils which are so numerous in certain caves (occasionally with forearm
included, as at Maltravieso and Fuente del Salin, Fig.72); 153 the
comparatively rare positive hand-prints (for example at Altamira, Santi{m,
La Pasiega and Fuente del Salin) were made simply by applying a paint-
covered palm to the wall. Very occasionally, hand stencils seem to have
been made with a pad , as in the case of two black specimens at Gargas
which have a regular area of paint around them; a white specimen in the
same cave apparently had white material crushed around the fingers; but
0 5 10 em
most have a 'diffused halo' which results from spraying. There are two L-....1-...J
possible methods: through a tube, or directly from the mouth; and was the
Fig. 72 Hand and forearm stencils
pigment in dry or liquid form? from Fuente del Salin (Santander).
Once again, observation and experiments have helped to clarify the (After Bohigas et al)
situation. When dry, the powder can have been applied only through a
tube; the site ofLes Cones (Vienne) contained a bone tube with powdered
red pigment inside it, which seems to indicate that this technique was
indeed used at times . If dry powder was projected, then a humid wall was
required or there would have been no adhesion; in any case, dry pigment
leaves 'fallout' beneath the hand, and this does not exist on the Palaeolithic
stencils. By contrast, if liquid paint is used, blowing it through a tube
concentrates it too much; experiments show that spraying liquid paint
from the mouth, about 7- 10 em from the wall, is not only the easiest
method but also the one which produces results which best resemble the
original stencils; 154 pursing the lips slightly projects a spray of fine droplets
which form the required halo with diffuse edges . On average, it takes
about 3 g of pigment and between 30 and 45 minutes to do each hand in this
way, whereas the 'dry method' can sometimes be quicker, but uses 9-10 g.
However, filling one's mouth with paint is hardly pleasant, and it is
possible that a combination of the two techniques was used : ie placing a
hollow tube upright in a paint-holder, then blowing across its top through
a second tube forces the paint up and out in a fine mist; or perhaps blowing
into a bent reed so that a spray emerges from a split at its base . 155
We have already seen that most engravings were probably done by
right-handed artists; and it has traditionally been assumed that because
most stencils are of left hands (at Gargas, 136 left hands have been
identified, and only 22 right), this too denotes predominantly right-
haalied people. Although this is likely to be correct, the hands are poor
evidence, partly because the painting may have been done by mouth, not
by the dominant hand, and partly because some hands may have been
stencilled palm-upward! However, one should also note that in the very
rare Ice Age depictions of people holding objects (particularly those of
Laussel, such as the 'Venus with horn') it is usually the right hand which is
doing the holding.
Unfortunately, the stencil experiments have not settled the old debate
about the apparently mutilated or deformed hands, most abundant at
Gargas but also known in other caves such as Tibiran, Fuente del Trucho
FORMS AND TECHNIQUES 105

(interestingly, located just across the Pyrenees from Gargas!),

Maltravieso, possibly Erberua and Fuente del Salin, etc. There are three
basic explanations for the missing phalanges (finger bones) on these
hands: either the fingers were bent (probably palm-upward) when the
stencils were done, or the phalanges were actually missing, through ritual
mutilation or through pathological conditions.
Many prehistorians have thought the fingers were deliberately bent, for
one reason or another, perhaps as a kind of language of gestures or
signals, 156 and some experimenters claim to have duplicated all the
'mutilations' at Gargas by bending their fingers, and point out that there
are stencils with clearly bent thumbs at Gargas and Pech Merle. 157 Another
experimenter, however, consistently found that placing his hand palm-
upward and bending the fingers caused pigment to infiltrate behind it, 158
which, as we have seen, was not the case in the originals. He therefore
supports the view that the phalanges were actually missing.
If they were, then ritual mutilation may have been responsible, as at
Maltravieso, where primarily the little finger was affected; but at Gargas
the cause is more likely to be pathological. Investigation by Ali Sahly
showed that the stencils here comprised adults, women and/or youths, and
children including infants; of the 124 best preserved, only 10 are free of
abnormality; the various combinations of missing phalanges are on only
the fingers, not the thumb. Almost 30% had the last two phalanges of all
four fingers gone. He claimed that repetitions of the same hands were
detectable, so that less than 20 people produced 231 prints between them.
His conclusion that conditions such as frostbite, gangrene and
Reynaud's disease (which attacks only the fingers and very rarely the
thumb) were responsible found support in his discoveries of actual hand-
imprints in clay, not only at Gargas but also at Lascaux, which seemed to
have phalanges missing; and of finger-holes in clay at Gargas, casts of
which seemed to end in stumps rather than finger-tips . 159

Some painting techniques

Strictly speaking, figures made by finger or in simple outline should
simply be seen as coloured drawings; the term 'painting' might better be
limited to those cases with infill of various kinds, and, as we have seen,
painted infill dates back to the Aurignacian block from abri Blanchard,
and perhaps beyond, as does the practice of colouring wide areas of rock-
wall. The sophisticated bichrome and polychrome figures of the
Magdalenian arrive relatively late, and are rare in Palaeolithic iconography
in comparison with engravings and outline drawings.
The latter were undoubtedly quick and easy things to do for an artist of
experience and talent. In an already famous experiment, Michel
Lorblanchet memorised every mark in the 'Frise Noire' of Pech Merle
(there are 25 animals on this panel measuring 7 m by 2.5 m)(Fig. 73), and
then reproduced the whole thing on an equally smooth panel of similar
dimensions in another cave, lit by a lamp in his left hand. Each figure took
an average of 1-4 minutes (compare this with the times for hand stencils,
above) , so that the whole frieze required about an hour, including initial
sketching with a stick. 160
Similarly, Claude Couraud has reproduced a Lascaux horse and a Niaux
bison, both full-size on paper and at reduced scale in a cave: the horse
required 30 minutes of preparation and another 30 to do on paper, but took
only 30 minutes in the cave; the bison took 20 minutes on paper, and only
106 IMAGES OF THE ICE AGE

10 in the cave. 161 Therefore, it seems that even quite detailed figures
needed little time for their execution.
Experiments have also been done to investigate problems of
superimposition of different paints . At Lascaux, it was long uncertain
whether the red cows were on top of the black bulls, or vice versa, because
the two pigments had mixed ; some scholars thought the cows were done
first , but infra-red pictures suggested the opposite. Using samples of
similar pigment, it was found that red on top of black did not mix, but
black on top of red did so - and so the bulls were clearly painted after the
cows. 162 As mentioned earlier (p.48), infra-red with special cut-off filters
 FORMS AND TECHNIQUES 107

has also been used by Alexander Marshack, Jean Vertut and others to
determine the use of different ochres and mixes of ochre within a single
composition or panel, where everything looks the same to the naked eye.
Similarly, ultraviolet light helps one to separate out overpainting and
overengraving more clearly. However, it must be stressed that these
techniques cannot be applied universally or at random, but need to be
developed for each particular problem or cave . 163
Combinations of painting and engraving are quite common and varied
in parietal art; at Trois Freres, clearly visible traces show that a panel was
Fig. 73 Photomontage of the 'black
frieze' in the cave of Pech Merle scraped extensively before a large red 'claviform' was painted on it (Fig.
(Lot), showing mammoths, bison, 24); 164 similarly, several panels at Altxerri were scraped extensively to
aurochs, horse and red dots. produce a light background for painted figures; 165 in some cases , an
Solutrean. Total size: c 7 m by engraved figure is filled in with ochre; elsewhere, as in the spotted bison of
2.5 m. The mammoth at top left is Marsoulas, a few incisions delimit certain parts of a painted figure, such as
1.4 m long. (JV) the head and leg; while at Tito Bustillo, some of the bichrome horses on the
108 IMAGES OF THE ICE AGE

main panel are silhouetted by wide areas of engraving which make them
stand out from the darker background and from earlier figures . 166 There
are even painted animals which are redone or 'highlighted' with
engraving, as at Trois Freres, Lascaux, Le Portel, or Santimamiiie, for
example.
As mentioned earlier , just as perfectly proportioned figures could be
engraved around a cylinder of antler, so the 'falling horse' of Lascaux is
painted around a rock: the artist could never see the whole animal, yet its
proportions remain sound. It is possible that an even more sophisticated
technique was occasionally used in the caves, namely anamorphosis, or
deliberate distortion . Claims have been made that four red cows at
Lascaux and a black horse at Tito Bustillo were purposely deformed in this
way so as to look normal from ground level or from the side; 167 given the
other accomplishments of Palaeolithic artists, there is no reason to suppose
that touches of this sort were beyond their capabilities.

Just as with portable art , there are clear regional differences in artistic
technique (quite apart from style and content): for instance, in the
Pyrenees there are roughly equal numbers of engraved and painted figures
in the Mas d' Azil, but engraving dominates in the decorated caves to the
west and painting in those to the east, although there are inevitably
exceptions (Tibiran has more painting than engraving, while Fontanet and
Massat are the reverse). 168 The regional difference in distribution between
clay and sculpture has already been mentioned; perhaps linked to this is
the fact that art on cave-floors is largely restricted to the Pyrenees but,
unlike Cantabria and the Perigord, this region has no decorated ceilings-
there are occasional engraved or painted figures on ceilings (engravings at
Montespan, for example), but there is nothing remotely like the Lascaux
passage, or the great painted ceiling of Altamira which Dechelette baptised
the 'Sistine Chapel of Rock Art' . 169
The Altamira ceiling was within easy reach - indeed, it was so low that
the artists could not have seen its whole surface at once; but how did they
manage to paint on high ceilings? Occasionally, as with a sign in Lascaux,
it might have been accomplished with a brush on the end of a long pole. At
La Griega (Segovia), it is thought that an engraving on a ceiling 3 m up
could have been done by a person straddling the passage with one foot on a
ledge on either side. 170 Elsewhere , as at Roucadour (Lot) where the
engravings are 6 m above ground-level, or at Baume-Latrone. where there
are broad sweeps of painting on a ceiling over 3 m high, 171 it is probable
that the floor has sunk down since Palaeolithic times. But for truly
monumental work , such as the great Labastide horse on a rock 4 m high,
the two black mammoths painted 5 m up on the vault ofBernifal, or much
of Lascaux's decoration , it is clear that ladders (perhaps tree-trunks with
bra~~eh-stumps as rungs) or scaffolding must have been used . Abundant
wood residues found in Lascaux may come from these constructions (some
are from big oaks) as well as from torches and so forth, but the clearest
evidence is the series of about 20 sockets cut into the rock on both sides of
the 'Diverticule Axial', about 2m above the floor; these were packed with
clay. Holes about 10 em deep in this clay suggest that branches long
enough to span the passage were fitted into the sockets and cemented into
place with the clay. This series of solid joists could then support a
platform, providing easy access to the upper walls and ceiling. 172
FORMS AND TECHNIQUES 109

Shedding light on the subject

Portable art may all have been done in daylight, like the art on blocks and
the decoration of rock-shelters. But the work inside caves required a
reliable source oflight. In a few cases, a hearth at the foot of the decorated
panel may have sufficed: at La Tete-du-Lion (Ardeche) a concentration of
pine charcoal only 1.2 m from the decorated wali may be a 'foyer
d'eclairage' of this type, though it could simply be the remains of
torches. 173 Certainly, portable light was necessary in most caves. What did
they have at their disposal?
Despite Riviere's discovery at La Mouthe, the very existence of
Palaeolithic lamps was not fully accepted until 1902, the same year that
cave-art was finally validated. This was no coincidence: once the pictures
deep inside caves, often hundreds of metres from daylight, were seen to be
authentic, it was obvious that the artists must have had some means of
illumination.
As often happens, the establishment of a previously rejected notion led
scholars to the opposite extreme, so that all hollow objects, and lots of flat
ones, were identified indiscriminately as lamps! A recent critical study of
the hundreds of objects claimed to be lamps has resulted in only 302 being
considered as possibilities, of which a mere 85 are definite and 31 others
probable. This is a very poor total when seen against the 25,000 years of
Upper Palaeolithic life and the hundreds of decorated caves, especially
when it is noted that 70% ofthem come from open-air sites, rock-shelters
or shallow caves. 174 It is certain that a different method was usually
employed in deep caves- probably burning torches, which have left little
or no trace other than a few fragments of charcoal or black marks on walls.
There are a few beautifully carved lamps; and some, such as those from
La Mouthe or Lascaux, even have engravings on them. Many ofthem are
of a red sandstone from the Correze region of France, which thus seems to
have been a centre of production.
Combustion residues in some specimens have been subjected to analysis
which indicated that they were fatty acids of animal origin; while remains
of resinous wood or of non-woody material clearly come from the wicks
(residues in the carved Lascaux specimen proved to be juniper).
Experiments have been carried out with replica lamps of different types,
different fuels (cow lard, horse grease, deer marrow, seal fat), and a variety
of wicks (lichen, moss, birch bark, juniper wood, pine needles, dried
mushrooms, and kindling). The results led to a number of interesting
insights, which were confirmed by study of the lamps used by Eskimos. 175
Firstly, a good fuel needs to be fluid, and easy to light. The initial
melting of the fat needs to absorb the wick material which, by burning,
continues to melt the fuel. This cycle can continue for hours, providing
both wick and fuel are replenished from time to time: one estimate is that
500 g of fat will keep a lamp going for 24 hours. 176 As mentioned earlier
(p.23), animal-fat fuel produces no soot.
There are two basic types of lamp: the open-circuit model, in which the
fuel is evacuated as it melts; and the closed-circuit, where the fuel is kept in
a cavity. The open type seems very rare in the Ice Age, although many
simple slabs of stone may have been used like this (about 130 limestone
slabs at Lascaux were interpreted in this way, although many are now lost,
and only about 36 seem likely lamps); most recognisable Palaeolithic
110 IMAGES OF THE ICE AGE

lamps are of closed-circuit type . In Eskimo communities, the open types

are for occasional use, while the closed types - in which far more work has
been invested- are used daily.
But how bright are these lamps? The answer, surprisingly, is that they
are pretty dim, even by comparison with a modern candle. The power of
the light given off depends on the quality and quantity of fuel; the flame is
usually unstable and trembling. Experiments with a stone lamp using
horse fat produced a flame of one-sixth the power of a candle, according to
measurements with a photometer. 177 With such limited radiance, it would
have been necessary for Ice Age people to use several at once, or to resort to
burning torches. In some deep caves, such as Labastide, there are also the
remains of hearths of various kinds, which would certainly have provided
strong light at times.
We, of course, are spoiled by artificial light, and are no longer
accustomed to dimmer sources. But in fact it is surprising how much can
be seen by the light of a single candle, and a large cave-chamber could be
adequately lit with two or three. It has been found that with only one lamp
one can move around a cave, read, and even sew if one is close enough to
the light- the eye cannot really tell that the flame is weaker than a candle.
One important factor is that flickering flames of this type have the effect
-no doubt observed and perhaps exploited in the Palaeolithic- of making
the depicted animals seem to move, an eerie experience. For the really big
figures, over 2m in length, more than one lamp would have been needed to
see the whole thing at once . The lamps would also have affected colour
perception- they give off a warm yellowish light which makes yellow look
orangey, and makes red pale or brown. The eye adapts well to red in this
light, and this may explain the frequent use of red dots and signs at various
points in the caves - by lamplight they were readily visible signals.

We still have much to learn about how Ice Age people - including many
children (p .13) - managed to make their occasional forays into the remote
depths of some caves. Even armed with bundles of torches, or lamps with a
plentiful supply of fat and wicks, it seems a very risky thing to do.
However, if archaeology has taught us anything, it is that we should never
underestimate the capabilities of our ancestors. We must not lose sight of
the fact that caves and darkness formed part of their environment, and
they clearly knew how to cope with them very successfully.

Open-air art
Equally, however, we should not let the terms 'cave-art' and 'cavemen'
blind us to the fact that they very rarely lived far inside caves but, rather,
occupied rock-shelters, cave-mouths, and a variety of open-air
hab*tions, from the tents of western Europe to the mammoth-bone huts
of central and eastern Europe.
Since they spent almost their whole lives in the great outdoors, it has
always been assumed that they must also have produced art outside the
caves , but that it has not survived the millennia of weathering and erosion.
As mentioned earlier , there have been occasional claims that open-air
figures were of Palaeolithic age - most notably at Chichkino, Siberia,
where hundreds of animal depictions over a distance of about 3 km include
a horse and a wild bovid considered characteristic of the end of the Ice Age
- but few scholars have been prepared to take them seriously. In recent
 FORMS AND TECHNIQUES 111

years, however, a series of important finds in western Europe has finally

brought the proof that Palaeolithic people did produce art in the open air,
that it can survive, and that the Soviet claims may therefore be valid after
all.
The first such finds occurred at a number of rock-shelters and caves
(including Murcielagos) in the Nal6n Valley, Asturias, which have some
eroded lines and animal figures deeply engraved in their exterior areas; 178
nevertheless, these engravings were still in or near shelters and cave-
mouths, rather like the sculptured friezes and carved blocks of France.
The next discoveries, however, were at truly open-air sites.
At Domingo Garcia, Segovia, the figure of a horse, almost a metre in
length, was found engraved (or, rather, hammered out) on a rocky
outcrop . 179 In style it resembles the engraved horses in the cave of La
Griega, in the same region; moreover, a schematic engraving of a different
style and period is superimposed on its outline (Fig. 75).
At around the same time, three animal figures , including a fine horse,
Fig. 74 Open-air engraved horse at 62 em long and 37 .5 em high (Fig.74), were discovered on a rock-face
Mazouco, Portugal. Length: 62 em, above the River Douro, at Mazouco in north-east Portugal; '80 they had
height: 37.5 em (Photo D. Sacchi) survived thanks to a position which protected them from the elements.
112 IMAGES OF THE ICE AGE

Fig. 75 The open-air engraved

horse of Domingo Garcia (Segovia).
(Photo and tracing A. Moure
Romanillo)

.
.
~
. ~
0
a
~

p a
.
<l

.
0

~ 0 e
0 II
t:>

0 50 em
 FORMS AND TECHNIQUES 113

This horse, too, resembles those of La Griega, and the three sites thus seem
to form a regional group.
Finally, a series of fine incisions has been found at Fornols-Haut,
Campome, in the eastern French Pyrenees , on a huge block of schist
located at an altitude of750 mona mountainside (Fig. 76). 181 The rock has
been greatly weathered by wind; but , because the eastern face is sheltered,
its engravings, although eroded, are clearly visible. This face is covered in
engravings, drawn in all directions and comprising about ten small
animals - none complete- as well as signs and zigzags. The finest figures
include the head of an isard, 7.5 em high , and that of an ibex (Fig .77).
As we have seen in the preceding chapter, all parietal art is notoriously
difficult to date, open-air examples especially so since they have no context
whatsoever. However, there is nothing remotely similar to these figures in
the Middle Stone Age and later art of eastern Spain , or in the schematic art
of more recent periods; moreover, had any of these figures been found in a
cave they would automatically have been accepted as fine specimens of
Palaeolithic art. These sites have therefore proved the existence and
survival of Ice Age open-air engraving (painting was doubtless also done
outside, but is far less likely to have survived weathering in this way); and
they have profound implications for our interpretation of the function s of
Palaeolithic art as a whole . Palaeolithic parietal art is by no means limited
to caves - they are merely the places where it has been best preserved.

Conclusion: portable versus parietal

In the past, portable art was sometimes seen merely as a 'sketchbook', a
repetition or rehearsal in miniature for parietal work . Today, however, we
see them as parallel and equally important art-forms, clearly part of the
same world, but perhaps entirely different in function .
As we have seen, one basic difference between the two is that portable
objects have been moved around, whereas parietal figures are precisely
where the artist wanted them, whether on the floor , the walls or the ceiling
of the cave. Another difference is that portable art, by definition , is
composed of small, light objects, and the figures and motifs drawn on
Fig. 76 The engraved rock on the
mountainside at F ornols-H aut
(Pyrenees-Orientales), altitude
750 m. The rock is 2.3 m high and
3.9m wide at the base. (Photo D .
Sacchi)
114 IMAGES OF THE ICE AGE

them are of a corresponding scale; as we have seen, the size and shape of Fig. 77 Ibex head engraved on the
the stones, bones and antlers either determined the form of the design rock at F ornols-Haut. Length: c
(lines of figures are common) or were chosen according to requirements. 7.5 em. (Photo D. Sacchi)
The same applies to parietal art (especially where rock-shape was
incorporated into a figure), but the height and width of the available
surfaces meant freedom from many of these constraints; hence, parietal
figures could be arranged in a variety of ways (individuals, clusters,
panels), and simple lines of animals are rare. Moreover, there was no
restriction on size: figures range from the tiny (such as those ofCampome)
to the enormous (such as the great Labastide horse, almost 2m long, for
which scaffolding was required 182 ) . The Hall of the Bulls at Lascaux has
relatively small deer and horses, together with the huge bulls, over 5 min
length; but of course not all the figures on a panel of this type are

-
necessarily of the same date, meaning or composition, and this is the
fundamental problem to which we must now turn our attention.
 6
What was Depicted?

For convenience, Palaeolithic images are normally grouped into three

categories although, as usual, there is a degree of overlap and uncertainty
in this division; the categories are animals, humans and non-figurative or
abstract (including 'signs'). Each presents its own problems, but the first
two share fundamental questions of zoology (identification of the species)
and ethology (what are they doing?). Neither question is straightforward.

Animal figures
The vast majority of animal figures are adults drawn in profile; there are a
number of possible reasons for this - it is easier than drawing them full-
face, particularly if one wants to convey the important features of the
animals' anatomy, which are also far clearer on adults than on the young;
and it is certain that this was how the animals were observed and identified
by Palaeolithic people.
Allowing that we can never know what any prehistoric image was meant
to depict- only the artist can tell us- it is obvious that most of the animals
drawn seem easily 'recognisable' at genus level; but there are great
numbers of figures which are badly drawn (in our eyes), incomplete, or
(perhaps purposely) ambiguous. Numerous examples can be found of
figures which have received markedly different attributions - beaver or
feline, fish or porcupine, bear or reindeer, big cat or young woolly rhino,
and crocodile, reindeer or horse! 1 The same is true in Australian rock-art,
where there can be difficulties in deciding between possums and dingoes,
or tortoises and echidnas, since figures have few diagnostic features, and
Aboriginal identifications can differ from those reached by zoological
reasoning.2 However, in the absence of informants about Ice Age images,
zoological reasoning is all we have to go on- comparison with present -day
animals, or, in the case of mammoths or woolly rhinos, with their closest
living relatives or with frozen carcasses from Siberia - but such
comparisons have an inevitable degree of subjective assessment.
It is therefore no surprise that, just as scholars differ on the numbers of
figures in a cave, they also tend to produce very different totals and
116 IMAGES OF THE ICE AGE

percentages for each species: at Marsoulas, for example, the three most
recent studies have claimed 21/32/30 bison respectively, 11/32/26 horses,
and 16/18/12 humanoids; at Le Portel, three recent studies claim 33/26/31
horses, 27/23/30 bison , and 6/5/4 humans! 3 As with different tracings of a
single figure (see above , p.48), which version should one believe? The
most recent is not necessarily the most reliable.
Another problem is that many authors, including some of the foremost
authorities on Palaeolithic art, admit in the text their doubts about
particular interpretations; but in captions, statistics and tables these
uncertainties often become definite attributions, and thus distort the
results. 4
A hierarchy of identification should therefore be adopted - definite,
probable and uncertain - and must be maintained consistently throughout
a piece of work. Even so, differences of opinion will remain. 5 Identification
of figures in art can never fully become an exact science, although expert
knowledge of animal anatomy, as exemplified by the remarkable studies of
Leon Pales, can help to clarify many issues. 6 As shown in an earlier
chapter, it is also wisest to work from originals or from photographs,
rather than from someone else's tracings.
Another type of evidence to bear in mind is the range of species present
in the faunal remains of the site or region in question during the Upper
Palaeolithic. Inevitably, however, there are cases where bones are of no
help- for example, nobody denies that the ibex is depicted at Pair-non-
Pair, although no bones of the species have ever been found in Gironde;
similarly, pictures of seals do not coincide with the few remains of the
creature found in Palaeolithic sites; and despite the existence of a few
depictions of mammoth in northern Spain, remains of the animal have
been found at only one or two sites there.
Where complete or detailed pictures of animals are concerned, there is
generally a broad measure of agreement, providing they are well
preserved: no one could doubt the horses of Ekain or the bison ofNiaux.
Problems arise where there is a choice of more than one type of animal such
as species of deer, as will be seen below; or where there is an extreme
degree of stylisation in the figure. 7 A figure's size is of no help, since
Palaeolithic art has no scale or groundline, and, as we have seen, tiny
figures are found next to large ones, even of the same species, and
sometimes on the same panel.
Occasionally, complete animals may defy identification; they resemble
no known species, are presumably imaginary, and have been left
undetermined, though some have been given bizarre names: the spotted
'unicorn' of Lasniux with its two straight horns (which has been
interpreted as a Tibetan antelope, a lynx, and even two men in a skin- see
below, p.157); the 'antelopes' of Pech Merle, with their swollen bodies,
1~ necks and tiny goatlike heads (Fig. 78); or the two strange beasts ofthe
Tuc d' Audoubert, with their shapeless muzzles and short horns .8
There are also figures of beasts with extremely long necks at sites such as
Pergouset and Altxerri; and a few images seem to be composites, such as a
deer at Trois Freres which , according to Breuil's tracing, has a bison's
head; a fish at Pindal which has the fins and tail of a tuna on a salmonid
body; 9 and an animal at La Marche which seems to have the body of a horse
and the head of a bovid, while another has the body of a horse or bovid,
together with an antler. 10
The many incomplete figures offer a different set of problems. Lack of
WHAT WAS DEPICTED? 117

detail in apparently unfinished animals can lead to different views, as in

Niaux's Salon Noir, where a sketchy outline of a quadruped, 1.5 m long,
above a horse high on one panel has been interpreted as a feline by some
scholars, but as a definite horse by others. 11 Fragmentary animals are not
usually a problem when the head is present; where the head is missing, as
in a bison on the Altamira ceiling or the bears of Ekain (Fig. 90) and
Montespan which were purposely left headless, the characteristic shape of
the rest of the body leaves no room for doubt; but if only part of the body
survives, it can be difficult to decide: one such figure at Bedeilhac which
lacks both head and forelimbs is generally thought to be a horse. The back
legs on the Laugerie Basse engraving called the 'Femme au Renne' are
thought by most specialists to belong to a reindeer, though Leroi-Gourhan
would prefer it to be a bison in order to fit in with his theories (see below,
p.l69). 12
However, it is all too easy to make mistakes with incomplete figures,
particularly on fragmentary plaquettes, where much remains uncertain: a
piece from Labastide, for example, had the back half of a quadruped
whose shape was thought characteristic of a horse, but when later a further
piece was found, the figure proved to be a bison! 13 Interpretation can also
depend on which way up a plaquette is held- for example, an engraving on
a specimen from Lourdes was seen only as a horse head by some scholars,
but a fine ibex also appeared when it was turned upside down . 14
Schematisation, which involves reducing a figure to its essential traits
and leaving out the rest, is well known in Palaeolithic art, and is perhaps
clearest in the isolated neck- and back-lines of horses, bison or mammoths
(or, for horses, simply the mane) - for example, at Marsoulas, Cougnac
and Pech Merle . 15 Nevertheless, identification of the animal is still
possible: in the Reseau Clastres at Niaux, there are three painted bison;
the first is complete and 'naturalistic'; the other two are very incomplete,
but even without the proximity of the first they could confidently be
identified as bison through comparison with complete examples
elsewhere: the shape of the back and position of the mane are quite
distinctive (Figs.79 and Frontispiece). Such figures appear abbreviated,
perhaps a kind of Palaeolithic 'shorthand' in which a part stands for the
whole; however, this implies that we know the artist's intention- and it is
possible that the figures simply depict a part of the animal, since the shape
of the back and head would be fundamental in recognition at a distance. 16
As will be seen below, there have been many attempts to derive detailed
information about different species or 'races' of animals- especially horses
- from Palaeolithic figures; all were doomed to failure, precisely because
this is art. We cannot assume that the Palaeolithic view of'realism' was the
same as our own, 17 or that they were concerned in any way with presenting
accurately the external features of the fauna. There is no justification for
creating new species or subspecies on the basis of these images, as has
sometimes been proposed: the international rules of zoological
nomenclature require that species be defined on the basis of anatomical
remains ; it is unscientific to attempt it from pictures. For bison or horses,
each art-site could present at least one subspecies of its own, but the
diversity probably owes more to the artists than to nature.
Similarly, little reliable information can be obtained about the animals'
coats- which in any case vary with sex, age, season and climate- because
of omissions, distortions and conventions in the figures. 18 The magnificent
antlers and horns, and the optimal body growth suggested by the
118 IMAGES OF THE ICE AGE

prodigious amounts of fat on some animals , together with the lack of

emaciated animals in the art , have been interpreted as proof of the rich
environment and long growth season in these parts of Europe. 19
The age ofthe animals can almost never be estimated, except in the case
of the few juveniles such as the calf with two adult bison on a bone
fragment from abri Morin, the calf on a bone disc from Mas d' Azil (with an
adult bovid on the other side), or the fawns apparently suckling at does
drawn on the wall at La Bigourdane (Lot) and a plaquette from Parpall6. 20
In other cases, one can make only a subjective assessment of the size of the
antlers or horns, which may be an artistic convention rather than an
accurate record.
The animals' sex is sometimes displayed directly but almost always
discreetly- usually on male bison or deer, 21 but rarely on horses for some
reason (eg at La Sotarriza). Only a few animals are known which have
female genitalia depicted ;22 this may be partly due to the predominance of
profile views , in which the vulva is not visible, though this could easily be
overcome through twisted perspective if it was important to convey
gender. Even udders are rarely shown (eg on one of the curled-up bison
of Altamira), even where they could be, as in the ' jumping cow' of Lascaux.
One therefore has to look for secondary sexual characteristics, such as
antlers in red deer, differing horn shapes in ibex, or sheer size and
proportions in other species - though there is always a potential overlap
between females and juvenile males. 23 These secondary features were
sometimes exaggerated - particularly deer antlers and bison 'humps' - and
thus comparisons of size cannot be undertaken between sites, since the
massive mammoth-like humps of some Font de Gaume bison, when
placed beside some of the more streamlined, 'boar-like' bison of Niaux,
make the latter appear to be the females to Font de Gaume's males. 24

So much for identification ; but what are the animals doing? Many of them
are 'motionless', to the point where it has been claimed that they are dead,
or at least copied directly from carcasses25 - primarily because they seem to
be standing on tiptoe , and show no sign of weight-bearing in the limbs;
these features are not to be found in living animals . The hypothesis also Fig. 78 The so-called 'antelopes' of
accounts for some of the twisted perspective in hoofs (since the underside PechMerle (Lot): this panel is c 1m
of the foot is frequently turned up by a fallen animal), for the fact that long. (JV)
nearside feet are sometimes higher than the others , for the prominence of
the belly, and for projecting tongues and raised tails. It is , of course,
possible to attribute all of these features to muddled memories ofliving and
dead animals ,Z6 or simply to artistic conventions, but it must be admitted
that close observation of carcasses was certainly the artists' principal
source of anatomical detail.
One cannot, however , imagine them dragging such dead-weights far
int~aves to be copied directly! Perhaps some portable depictions are
outdoor sketches of carcasses, which were later carried into the caves and
copied on to the walls, but this is pure speculation . The fluent, effortlessly
drawn and well-proportioned animal figures in parietal art suggest that the
artists carried everything in their mind's eye. As Breuil, who used the
same technique, said of the Palaeolithic artists: they never took a
measurement - they projected on to the rock an inner vision of the
animal. 27 Even if the Palaeolithic figures were ultimately based on dead
specimens, it is noteworthy that their motionless figures are nevertheless
imbued with an impression of life and power.
 WHAT WAS DEPICTED? 119

Some hunting specialists have chosen to interpret many of the features

ofthese images in behavioural terms: for example , the raised tail is seen as
a signal of alarm, anger or threat, of submissiveness, or of a copulatory
invitation by females 28 - and thus as an illustration of autumn, when such
displays are common due to the interactions of rutting behaviour,
although if the threat were directed at a hunter it could occur in any season.
There have been two particularly famous examples of zoological
deliberation about posture in Palaeolithic images. The engraved reindeer
from Kesslerloch (Switzerland) was traditionally thought to be grazing,
until it was pointed out that its stance was exactly that of a male in rut -
either sniffing a female's track or, more likely, in a position for attacking a
rival. 29 The curled-up bison on the bosses of the Altamira ceiling have been
described as sleeping, wounded or dying (see below, p. l53), or as clear
pictures of females giving birth; 30 currently, the dominant view is that they
are males, rolling in dust impregnated with their urine, in order to rub
their scent on territorial markers - even though one of them has udders,
according to Breuil's copy! 31 In fact, they may simply be bison figures
drawn so as to fit the bosses- they have the same volume, form and dorsal
line as those standing around them, but their legs are bent and their heads
are down. 32
'Animated' figures are rare, and although they have occasionally been
Fig. 79 Three bison drawn in the
Reseau Clastres (Ariege) in different placed in the centre of a decorated 'panel', they are more usually to be
degrees ofcompleteness [see also found at the edges. 33 Such movement seems to appear in Solutrean and
Frontispiece]. The most complete, on early Magdalenian times (as at Roc de Sers), but is most common in the
the left, is 1.14 m long. Probably middle and later Magdalenian, although this kind of 'realism' never
Magdalenian. (JV) predominates. However, it is Lascaux which has the most abundant and
120 IMAGES OF THE ICE AGE

varied animated figures; the cave of Gabillou, in the same region and
thought to be of roughly the same period, also has lots of movement in its
figures.
'Scenes' are very hard to identify in Palaeolithic art, since without an
informant it is often impossible to prove 'association' of figures rather than
simple juxtaposition . In a few cases, however, this can be done: for
example , the Laugerie-Basse engraving, mentioned above, clearly shows a
supine woman beneath and beyond a deer. Different versions of one
possible scene, depicting what may be a wolf menacing a deer, have been
found on portable engravings from the Pyrenees (Lortet, Mas d' Azil),
Cantabria (Pendo) and Dordogne (Les Eyzies), which is an impressive
distribution. 34 A few other portable engravings such as the 'bear hunt' of
Pechialet (Dordogne), or the 'aurochs hunt' of La Vache have also been
interpreted as scenes, as has the parietal group of figures in the shaft at
Lascaux (see below, p .1 86), though one cannot be sure that a narrative is
involved rather than a symbolic mythology. 35
In short, therefore, one has to be very careful in making zoological and
ethological observations from Palaeolithic images. If one has photographs
of animals, one can accurately identify them and assess what they are
doing; but here we are dealing with drawings by artists with a message to
convey, and using stylistic conventions : nobody would assume, from the
spotted specimens at Pech Merle, that the horses of the period had big
bodies, small legs and tiny heads; so how reliable are the other features on
display? These are not 'Palaeolithic photographs'- we need to allow for
convention, technique, lack of skill, faulty memory, distortion, and
whatever symbolism and message were involved. Only the most complete
and 'naturalistic' figures are really informative , but even the finest
Palaeolithic pictures may prove a disappointment to the zoologist. Ice Age
images are exciting because they allow us to 'see' the fauna of the time, to
put flesh on the bones which we dig up; 36 but we are seeing the animals
through the artists' eyes.
Despite the different identifications and percentages of species produced
by each specialist, one fact remains clear: the overwhelming overall
dominance of the horse and bison among Palaeolithic depictions. In Leroi-
Gourhan's sample of 2,260 parietal figures, no fewer than 610 were of
horse, and 510 were bison ,37 which thus account for about half between
them. No other type of animal comes close, although, as we shall see, other
species do dominate at particular sites. The next most numerous include
deer, ibex and mammoth , followed by a cluster of rarer species.
The horse
Study ofthe most important animal in Palaeolithic iconography has, in the
past, wasted a great deal of effort on attempts to establish exactly which
'{,aCes' were depicted; Piette had a try at this in the nineteeth century/ 8 and
was later followed by other scholars who distinguished up to 37 varieties! 39
However, all such exercises were eventually abandoned; as mentioned
above, they were based on the shape and size of the figures, their manes,
and the colour and pattern of the coats, all of which were subject to artistic
whim - for example , Gravettian figures tend to have elongated heads,
while those of the Solutrean and lower Magdalenian tend to have tiny
heads! The only firm basis for subdividing horses into different species is
palaeontological differences, primarily in the skull and backbone, and it is
therefore a matter best left to those working with faunal remains.
WHAT WAS DEPICTED? 121

More recent studies have avoided this topic , and instead approached
important collections of horse figures from a more rigorous knowledge of
equine anatomy.40 At Gonnersdorf, 74 mostly fragmentary horse
depictions are known from 61 plaquettes, making this animal dominant
among the animals depicted at the site. Horses likewise dominate the
animal figures at LaMarche, where at least 91 whole or partial specimens
have been found on 64 stones; 62.4% of them face right, a percentage
similar to that (57-59%) at sites such as Ekain, Lascaux and Les
Combarelles.41 Not one of the LaMarche examples can be sexed; and it has
been found that, like artists throughout prehistory and history, those of La
Marche consistently made the horses' bodies too long42 - perhaps to give
an impression of speed and lightness. Certainly, in Palaeolithic art as a
whole, horses strike a number of different poses , which can be interpreted
from a behavioural point of view, such as the jumping horse on a spear-
thrower from Bruniquel, or the neighing horse head from Mas d' Azil. 43
One piece of zoological information which is probably reliable in these
horse depictions is that of the length of the coat, since present-day
Przewalski horses (thought to be among the closest surviving equivalents
to the Palaeolithic types) do display striking seasonal differences: hence
the shaggy ones recorded at sites like Niaux must be the long pelt of winter,
and, interestingly, no distinct 'winter horses' are known in Cantabrian
caves. 44 On the other hand, these differences might also denote different
climatic phases, as has been suggested for a similar range in bison
depictions (see below) .
Another frequent feature of Palaeolithic horse figures is the 'M' mark on
their side, which denotes a change in colour between the dark hide and the
lighter belly area - a shape and change still clearly visible on modern
animals such as the little semi-wild 'pottoks' of the Basque country
(Fig.ll4) . ~ 5 Stripes also sometimes occur on the horses' shoulders (for
example, at Ekain), a feature which Darwin noticed on living specimens,
but which the Palaeolithic artists may have exaggerated. 46
It is certain that the horse was of tremendous cultural importance to
people in western Europe during the Upper Palaeolithic- horse teeth and
bones have been found carefully placed in Magdalenian hearths in a
number of deep Pyrenean caves such as Labastide and Erberua, 47 as well as
near hearths at the Magdalenian open-air camp of Pincevent near Paris
(despite the fact that reindeer account for almost 100% of this site's faunal
remains). 48 At Duruthy (Landes), the carved horses (see above , pp. 84/6 )
were found in a kind of 'horse sanctuary' dated to the twelfth millennium
BC; the biggest, a kneeling sandstone figure, rested against two horse
skulls and on fragments of horse-jaw, while three horse-head pendants
were in the immediate vicinity. 49
Finally, it should be noted that the horse may well have been under close
control during the Upper Palaeolithic, as suggested by a variety of
evidence including some possible depictions of simple harnesses on
certain figures, especially one from LaMarche (Fig.22).50

The Bison
Like the horse, depiction of the bison varied through the Upper
Palaeolithic, in terms of shape, size and shagginess; some scholars have
argued that not only artistic convention was involved but also perhaps
climatic change, so that the heavy, shaggy bison would denote cold phases
and the lighter forms would be linked with milder phases 51 - it is a nice
122 IMAGES OF THE ICE AGE

Figs. 80/81 Perforated antler

baton with carving of horsehead with
ears raised and a stylised mane; the
shaft is also decorated. Le Mas
d'Azil (Ariege). Magdalenian. Total
length: 20 em. (JV). Detail of the
horsehead. From muzzle to ear-tip it
measures less than 4 em. (JV)

idea, but a lot of work and more accurate dating would be required to
prove that climate was indeed responsible, rather than style or different
species/subspecies. As mentioned above, different sites have their own
characteristic bison 'construction', seen in an extreme form at Font de
Gaume, and attempts to differentiate specific varieties have led no furth~r
than similar studies of horses.
The best study of bison figures, albeit of a small sample (13), concerns
those of La Marche; 51 an earlier attempt to establish the proportions and
rules of construction that characterise Palaeolithic depictions of bison
proved oflimited value because it was derived from the figures themselves;
it isJp. fact necessary, in studies of this kind, to proceed from precise
anatomical markers taken from living animals or at least from photographs
of them in profile. 53 Thus nature, rather than artistic convention, is the
reference point. Using this method, Leon Pales found that Palaeolithic
bison figures, even within a single site (such as Niaux or Fontanet),
showed differences in execution and style, whereas other groups (such as
at Altamira) were fairly homogeneous. Of all those examined, the La
Marche specimens were the 'closest to nature'.
It was also found that in this site, as at Trois Freres and Font de Gaume,
there was a tendency to have bovids facing left. Depiction of gender varies
 WHAT WAS DEPICTED? 123

Fig. 82 Bison engraved and

painted in Trois Freres (Ariege).
Magdalenian style. Length: 1.5 m.
(JV, collection Begouin)

from site to site: at Font de Gaume, 12 of the 39 bovids have a phallus,

whereas none ofNiaux's 47 bison has one (although many look male from
secondary sexual features such as the hump, the massive head and the
short, thick horns); only one at Trois Freres, standing on its hind legs, has
a phallus- one of the rare erect phalli in Palaeolithic art- and only one or
two of the La Marche bison may have the sex indicated. 54 Anatomical
details seem to be incorrect occasionally: for example, horns that point
forward, or which differ in volume from those found among faunal remains.
Most bison are 'immobile', but a few do show movement - some
hypotheses about the curled-up specimens at Altamira have been
mentioned above, and we shall return to them later (p.l53). One famous
bison,, a carving from La Madeleine, has turned to lick its flank; while that
in the 'shaft scene' of Lascaux appears to have lost its entrails and perhaps
to be charging a man. Other 'scenes' involving bison attacking people may
Fig. 83 Apparent scene of bison
pursuing a human, carved in bas- be depicted on the cave-wall at Villars (Dordogne), on a block from Roc de
relief around a block from Roc de Sers (Fig.83) and a portable engraving from Laugerie-Basse.55
Sers (Charente). Solutrean. The
block is c 35 em thick, the human is The aurochs
SO em high, and the bison 54 em. Wild cattle are the other large bovids represented in Palaeolithic art; the
(JV) most lifelike are perhaps those ofTeyjat, and the portable engraving from
124 IMAGES OF THE ICE AGE

Trou de Chaleux (Belgium) (Fig .37), but the biggest and best known are
those of Lascaux, including the 5-m bulls (Figs.30-2). Nobody has ever
questioned the attribution of the latter to Bos primigenius but, for some
reason, certain leading scholars thought the smaller cattle ofLascaux to be
Bos longifrons or brachyceros, a domestic type with long forehead and
short horns which is not dated earlier than the Neolithic period.
The simple explanation is that of the pronounced sexual dimorphism of
Bos primigenius: the big cattle of Lascaux are bulls, while the smaller are
probably cows of the same species. Studies of similar cattle in Morocco
with the coloration of their wild ancestors show that the great bulls are
black or reddish-black, with a light-coloured strip along the back and some
light hair between the horns; the cows are usually reddish-brown, with
darker legs, a darker and slimmer head, and thinner horns . All of these
features are on display at Lascaux , some to an exaggerated degree, together
with other specimens which are slightly abnormal in colour.56 The sexual
dimorphism can also be seen, to a lesser extent, in the bull and two cows
engraved at Teyjat (Figs.85-6).

Deer
Apart from a few depictions of the giant deer (Megaceros), such as those of
Cougnac which cleverly use natural rock shapes for the dorsal line and
hump (Fig.89), and sporadic claims for elk, and for fallow deer, as at
Tursac, 57 all Palaeolithic depictions are either of red deer or of reindeer.
In red deer, only the males have antlers , so that these can denote sex,
although some figures taken to be hinds may actually be males whose
antlers have been shed. In the reindeer, however, both sexeshave antlers .
Many depictions of reindeer show clearly the characteristic shape and
markings of the species, as in the series of engravings at Trois Freres;
ironically, some of the best figures are to be found in Cantabria (Altxerri
(Fig.87), Tito Bustillo (Fig.29), Las Monedas), a region where this animal
was never plentiful, according to faunal remains. 58
Occasionally, it is difficult to differentiate between the two types of deer;
on the famous baton from Lortet with a scene of deer and fish engraved
around it, some scholars see stags, but others see reindeer. (Indeed, Piette
first saw them as reindeer, and later as stags! ) In the portable engravings of
La Colombiere, most of the animals thought to be red deer have recently
been reinterpreted as reindeer. 59 Similarly, all but one of the 91 deer on the
walls of Lascaux have been assumed to be red deer by most specialists of
Palaeolithic art; 60 but a number of experts on caribou instead see some of
them as reindeer - particularly the series of heads, often called the
'swimming stags', since groups of swimming stags are very rare, whereas
swimming caribou are a common sight in the migration season (Fig.84).
This alternative identification may well be correct since 74 of the Lascaux
ce~ds (81 %) have antlers, and the wide variety of shapes includes some
which closely resemble those of tundra caribou and depictions of them by
Eskimo hunters during prehistoric and historic times. 6 1
As usual, some authors have chosen to see these Palaeolithic figures as
accurate depictions of individual animals ; thus, the remarkable examples
in caves such as Lascaux were detailed records of prize stags (mostly at
least 15 years old) with fine antlers / 2 whereas it is far more reasonable to
assume that the style and ability of individual artists played a major role,
and that the antlers are merely an artistic translation of reality.63 Attempts
 WHAT WAS DEPICTED? 125

Fig. 84 The so-called 'swimming'

deer of Lascaux (Dordogne).
Probably Magdalenian . This panel is
c5 m wide. (JV,)
to differentiate tundra and forest reindeer types from antler shape in the art
have also been made, but are unlikely to be reliable. 64
Pending publication of a detailed anatomical study of deer depictions, 65
one can still make a number of observations: many features, such as the
antlers, tail, hoofs, ears and eyes, are often incomplete or done badly - yet,
occasonally, even the gland below the eye of the rutting reindeer is
depicted, as on the engraving of Kesslerloch, and on seven or eight of the
Lascaux deer. 66
A number of different positions are known for cervids, such as the male
bending to lick a female's head at Font de Gaume, the rutting male of
Kesslerloch, and other variations at Limeuil and elsewhere - grazing,
bellowing, and so forth- and it has been pointed out that both reindeer and
red deer are most often depicted in what appears to be the rutting season
(autumn/early winter), characterised by features such as the white mane
on the reindeer bulls and the thick mane on mature stags, and bodies
rounded with fatY for example, a stag from abri Morin has been described
as having its antlers thrown back, its ears down, nostrils dilated, mouth
open and neck extended. 68

Ibex
One should really use the term 'ovicaprids' here, since it is quite possible
that some Palaeolithic depictions represent the mouflon or even the Tahr
Hemitragus, and a few have been attributed to these species; 69 but by and
large, animals of this type are automatically assumed to be ibex.
The question then arises whether they are the Alpine variety (Capra
ibex) or the Pyrenean (Capra pyrenaica); the former has curvilinear horns,
while the latter's horns are more lyre-shaped, and rise . at the tips.
However, they are often difficult to differentiate in art, for the simple
reason that in profile one cannot assess how sinuous or divergent the horns
are - indeed, this may be another role for twisted perspective - and in
126 IMAGES OF THE ICE AGE
 WHAT WAS DEPICTED? 127

Figs. 85/86 Engraved aurochs cow

(left) and bull (right) from Teyjat
(Dordogne), with tracing by Breuil.
The animals are 55 and 50 em long.
Magdalenian. (JV). The detail (left)
shows the cow's head. (JV)

Fig. 87 Engraved reindeer and fox

at Altxerri (GuipUzcoa).
Magdalenian style. The fox is 25 em
long. (JV)
128 IMAGES OF THE ICE AGE

many cases, particularly in portable art, the shape of the available space
caused distortion in the shape and curve of horns: one example from the
Mas d'Azil even has them sticking straight out in front! 70
Some depictions appear ambiguous- an ibex carved on an antler shaft
from the Mas d'Azil has an Alpine horn-shape when seen from the front,
but a Pyrenean shape in profile, while an engraving from Limeuil appears
to be a Pyrenean type which has been transformed into an Alpine ibex. 71
It is, of course, possible that the Limeuil example is really a young
animal which was turned into an adult through extension of its horns, or
that it represents the bone-stumps within the horns. However, there is
evidence that different varieties ofthis animal coexisted in western Europe
-at Isturitz, for example, there seem to be depictions of a number of types,
even if one allows for distortion, artistic licence, and so forth; and certain
figures seem to be accurate portrayals of the Alpine type (eg the figure on
the La Mauthe lamp), and of the Pyrenean (eg some of those in the Salon
Noir at Niaux). 72

Horn-size serves to distinguish males from females, and where two

animals with horns of different size are found together, as at Cougnac, it is
reasonable to suppose that they represent one of either sex; but, as
mentioned above, small horns may also denote a young male. An
occasional phallus was depicted (as on a La Marche animal with big
horns), and a beard is a useful secondary sexual characteristic.
Ibex figures were sometimes animated- the animals carved on spear-
throwers, with head turned back to see the emerging turd and birds, are
probably young ibex (Fig.46), as are the two headless animals fighting or
playing together on another famous spear-thrower fragment from Enlene.
A baton from Duruthy shows two ibex fleeing in apparent terror from
some large animal (Fig.40); elsewhere, ibex are often found in pairs, facing
each other, head to head. 73

The mammoth
Pending publication of the magnificent mammoths of La Marche/ 4 the
most thorough study of this species in Palaeolithic art concerns the 61
specimens on 46 plaquettes from Gonnersdorf. 75 About half are complete,
the rest having become fragmented along with the plaquettes. It proved
possible to divide them into adult and young animals; a few are grouped
together on the same plaquettes, but most are individual depictions.
Some details, such as the 'anal flap', coincide with observations on
frozen mammoths in Siberia; but others are troubling: a few of the
Gonnersdorf mammoths have small, short tusks, but most have none at
all, although the depictions seem very naturalistic; similarly, there are no
tusks on the far older mammoth figures from Vogelherd and
Geissenklosterle, on an engraving from Kostenki, and on a number of
paJ;i_etal depictions. 76 1t has therefore been suggested that some mammoths
had no tusks, perhaps through a depletion in natural resources; but other
scholars prefer to see it as artistic licence - or perhaps even as a sexual
dimorphism. One might compare it to the lack of hair on figures from
Pindal, Cougnac, La Baume-Latrone and other sites, which led to their
erroneous identification as warm-climate elephant species such as Elephas
antiquus.77
The Gonnersdorf mammoths show little sign of 'movement', and
mammoth depictions in general are not found in 'scenes'. They are,
however, sometimes found head to head, and other signs of animation
 WHAT WAS DEPICTED? 129

include a raised tail, and the trunk either raised or curled up; occasionally,
as in Pech Merle, one finds both detailed depictions and simple
abbreviations in the form of characteristic dorsal lines. 78
Finally, it is perhaps noteworthy that the few mammoth figures in Spain
all face left. 79

Big carnivores
Large carnivores in the art comprise bears and big cats. In the past, some
pretty uncritical inventories of supposed depictions of these animals were
drawn up, including many figures which even the authors found highly
dubious, and omitting others 80 - inevitably, they relied heavily on old
interpretations and unreliable tracings; they have since been superseded
by more objective studies based on anatomical expertise81 -once again,
primarily that of Leon Pales in his study of the LaMarche figures.
Bears, according to faunal remains, comprised two principal species
during the Upper Palaeolithic of Europe- the cave-bear ( Ursus spelaeus)
and the brown bear ( Ursus arctos). The former is characterised by the
domed shape of its skull, and traditionally it was assumed that any
depiction of a bear with a similarly domed forehead could safely be
assigned to that species. However, it has now been shown that a whole
spectrum of head-shapes is present in the art- only a few, such as those on
pebbles from Massat and La Colombiere, have really domed heads- and
this criterion is thus not sufficiently reliable to permit identification. 82
It is noteworthy that all the LaMarche bears face right; a number of
different postures can be found in Palaeolithic art, including bears
walking, or with nose raised, or with head down- the latter seen as the
threatening posture it would adopt towards humans. 83 The Pechialet
'scene', mentioned earlier, seems to show a bear on its hind legs and
menacing two 'humanoids'.
Bear and lion heads can appear very similar, and thus it has been hard to
differentiate between them in the fragmentary terracotta figurines ofDolni
Vestonice; this site, together with that of Pavlov, is important because its
art is dominated by carnivores: of 77 miniature figurines in good
condition, no fewer than 21 are thought to be bear, 9lions, 5 wolves and 3
foxes (=50%). 84 Only those of bear and lion are pierced and have little
'decorative' notches, but all the figures are caricatural, with few details,
Fig. 88 Engraved bone fragments and thus little can be deduced from them about the species depicted.
from La Vache (Ariege), showing a Indeed, in Palaeolithic art as a whole, carnivores, and especially felines,
line of three felines; the fragment on are consistently among the most inaccurate figures- perhaps because they
the right was discovered recently to be were far less easy to observe at close quarters than herbivores: the
part of this scene. (Drawing by D. commonest mistake is the position of the canine teeth in both felines and
Buisson & S. Rougane) bears- in nature, the lower canines are in front of the upper when the
130 IMAGES OF THE ICE AGE

Fig. 89 Photomontage of the

decorated panels at Cougnac (Lot):
note the use of natural rock
formations for the neck and chest of
the giant deer. The female ibex at the
top (facing left) is 50 em long; the
ibex on the right (and facing right) is
81 em in length. (J V)

Fig. 90 The two headless bears of

Ekain (Guipuzcoa). Probably
Magdalenian. Panel size: 1.2 m
(JV)

Fig. 91 Mammoth sculptured in

bas-relief in the Grotte du
Mammouth (or de Saint-Front))
Dordogne . It is 1.25 m high) 1.1 m
long) and located 4 m above the
present cave floor . (Photo B . & G.
Delluc)

mouth is closed, whereas in these figures, if the jaws were to be closed,

they would be in contact with the upper, or even stand behind them. 85
Caraj.vore canines clearly impressed Palaeolithic people, who, as we have
seen, often used them for jewellery; they are the only teeth ever drawn in
carnivore figures, and their size is often exaggerated, like deer antlers or
bison humps - in one case, however, the large teeth on a figure from
lsturitz have been taken at face value as a depiction of the scimitar cat. 86
Nevertheless, a few felines are accurately portrayed, including those of
La Colombiere and Labouiche; the IS of LaMarche are fairly realistic-
but only 43% of them face right (as opposed to 100% of the bears). We
cannot tell for certain whether or not the male big cats had manes , though
one or two depictions in Palaeolithic art seem to have one. 87
WHAT WAS DEPICTED? 131
132 IMAGES OF THE ICE AGE

The rare mammals

Finally, there is a whole range of animals which are depicted extremely
rarely, although often well .88 Many of them belong to the Magdalenian,
and their existence may thus simply be a result of the overall increase in
numbers of figures in this period: one can mention the rhino, musk-ox,
ass, sa:iga, chamois, wolf and fox, hare, glutton, otter and possibly the
hyena; there is also the remarkable weasel-like animal of the Reseau
Clastres (Fig.2). The 'boars' on the Altamira ceiling are now generally
thought to be 'streamlined' bison since at least one of them has horns! So
far, animals such as small rodents seem to be absent from Palaeolithic art.
Where sea-mammals are concerned, most are seals of various species, 89
some of the portable depictions found hundreds of kilometres inland.
Among the finest are the male and female grey seals (Halchoerus grypus)
on the famous baton of Montgaudier (Charente); they are characterised by
their elongated muzzle, their size difference and the male's neck folds.
Three enigmatic small shapes above them were recently claimed to be
whales seen in the distance. 90 Some parietal figures at Nerja (Malaga) have
been interpreted as dolphins, but others believe them to be seals. 91

Fish, reptiles, birds, insects and plants

Fish are quite plentiful, particularly in the form of portable depictions -
unfortunately, Breuil's corpus also includes a mass of decorative motifs
which he believed to have been derived from schematised fish; some of
them are quite plausible, many others are not. 92 Fish species are especially
difficult to identify in Palaeolithic art: most of them seem to be salmonids-
sometimes males with a kipe on their lower jaw in the characteristic state of
exhaustion after spawning, and thus perhaps representing the autumn. 93
Marine fish are also represented, both on portable objects (such as the
Lespugue flatfish figure, in the central Pyrenees) and on the walls of caves,
particularly those near the sea, most notably at Pileta. 94
One or two tortoises (St Cirq, Marsoulas) and perhaps a turtle (Mas
d'Azil) occur in portable art, and one possible salamander from Laugerie-
Basse; there are also a number of snakes, although once again Breuil's
attributions are excessive, with all manner of wavy lines being interpreted
as these creatures- one wonders why eels or worms were not envisaged:
the specimens next to the seals on the Montgaudier baton are clearly males.
Birds occur in both types of art; 95 the most recent survey of French
specimens found 121 depictions in 46 sites, and once the doubtful cases
had been weeded out there remained 81 birds in 31 sites96 - remarkably
few compared to numbers of animal figures. Portable depictions are more
abundant than parietal (the latter comprise IS figures, or 18 .5%). Most are
Magdalenian (86.4%; 69% are from the mid- and upper Magdalenian
alone), while the oldest are all parietal. Water-birds such as swans, geese,
d~s and herons seem to be the most numerous (about 37%). The species
and even the genus can be very difficult to identify, as the depictions are
often mediocre: indeed, in some cases it is not sure whether faces on cave-
walls belong to owl-like birds or to humanoids and 'phantoms', although
Leroi-Gourhan's beliefthat the owls of Trois Freres are 'anthropomorphs'
seems to rest on his theories about cave topography (see below, p.l72)
rather than on resemblance.
Another feature which birds can share with humans is a vertical, bipedal
posture; recently, portable engravings from Les Eyzies and from
Raymonden, traditionally seen as groups of tiny humanoids approaching a
WHAT WAS DEPICTED? 133

bison (in the former) apd standing around a bison carcass (in the latter) ,
have been reinterpreted by a specialist on birds as lines of swallows:97
certainly, their stance is identical, and microscopic study revealed tiny
incisions by the heads which seem to be beaks; another piece from
Raymonden seems to include two swooping birds.
Only about a dozen definite or probable depictions of insects are known,
all in portable art, and mostly Magdalenian.98 Apart from what seems to be
the larva of a reindeer-botfly from Kleine Scheuer, and the famous
grasshopper of Enlene, most are beetle-like bugs. However, even the
grasshopper, the clearest and most detailed depiction, cannot safely be
identified to the genus ievel.
Finally, plants are also rare in Palaeolithic art, 99 especially in parietal art,
and only a very few, such as those on a baton from Veyrier and a pebble
from Gourdan, can be classed as definite rather than possible. None can be
identified accurately.

To sum up, therefore: the 'animal' category in Palaeolithic art is oflimited

information value to zoology, and vice versa. The greatest caution and
rigorous objectivity must be applied to any attempt at identification of the
genera or species, and assessment of other external features and their
posture.
It is clear that large herbivores dominate the art, and that other types of
creature are either comparatively rare (carnivores, fish, birds, etc),
extremely rare (reptiles, insects, plants) or totally absent (rodents , and
many other species of mammal and birds). Since Palaeolithic people were
certainly familiar with all aspects of their environment, and since birds,
fish and plants were important resources for them , at least in the
Magdalenian, it follows that Palaeolithic art is not a simple bestiary, not a
random accumulation of artistic observations of nature. It has meaning
and structure.
This becomes even more obvious when one looks at the differing
frequencies of species in a variety of contexts . It has already been
mentioned (see above, p.56) that the Tuc d'Audoubert has different
species in its portable and its parietal art, and this observation has been
repeated elsewhere. Overall, reindeer, fish, birds and plants are primarily
found in portable art rather than parietal, 100 while insects are limited to
portable depictions.
Some themes are associated with particular forms of artefact: the horse is
rare on harpoons and 'baguettes demi-rondes' , while bison are not found
on spearpoints, are rare on perforated batons, but predominate on
plaques. 101 Horses and fish are common on perforated batons.
Different species also predominate in different periods and regions: for
example, the bison tends to dominate on cave-walls in Cantabria, although
it is rare in the region's portable art, where hinds are predominant on
shoulder-blades . The aurochs is more important than the bison in the art
of the early phases, but the situation is reversed in the middle and late
Magdalenian. The mammoth is of importance in the Rhone and Quercy
regions only during the early phases, but in Dordogne during the
Magdalenian, when the bison dominates in the Pyrenees. 102 The horse and
ibex seem to maintain roughly the same frequency in most places and
periods, though the horse appears particularly important in a cluster of
Dordogne caves in the Magdalenian (Cap Blanc, Comarque, St Cirq ,
etc) 103 like the ibex in the late Magdalenian ofCantabria. Such regional and
134 IMAGES OF THE ICE AGE

..
chronological differences may be at least partly ecological in origin, rather
than purely cultural or a matter of 'preference' .
Leroi-Gourhan's calculations concerning species frequency in different
Fig. 92 The owls engraved at Trois
Freres (Ariege) [see also Fig. 56} .
Panel width: 87 em. (JV, collection
regions have recently been presented in the form of graphs; and Begouen)
cumulative percentage frequency graphs have also been produced to
compare the depictions in different sites; exercises of this type obviously
have to assume that all the figures have been correctly identified, but
cumulative percentages are a particularly risky statistic. 104
Finally, it should be noted that different species are not scattered at
random through the sites; this is well known for parietal art (see below,
WHAT WAS DEPICTED? 135

p.l67), but the same phenomenon occurs with portable depictions. The
differential distribution of depicted species at Dolni Vestonice has already
been mentioned (p.83); at La Madeleine, the horse and bison dominate on
the objects of bone and antler, but the reindeer was predominant on a
cluster of engraved limestone slabs found in part of the shelter. 105 An
excellent example ofthis phenomenon occurs at Gonnersdorf, where most
of the mammoth engravings were found inside a habitation near the
hearth, with another concentration a few metres away; depictions of birds,
on the other hand, were concentrated precisely in those zones devoid of
mammoths, whereas horses and humans were found all over the excavated
area. 106 Since the mammoth engravings came from a winter habitation
(according to the faunal remains) and the birds from a summer one, it is
possible that there is a link between season and depicted species at this site.

Humans
The category of humans has traditionally been given the unwieldy title of
'anthropomorphs', comprising not only unmistakable human forms but
also images which overlap with the other two categories, and many which
seem to have been included simply because they looked figurative but did
not resemble any known animal. The fact that humans lack the specialised
anatomical features of other mammals (horns, antlers, etc) meant that all
kinds of vague, amorphous images were unjustifiably lumped together
with clear depictions of people. 107 Instead, a distinction should be made
between definite humans, 'humanoids' and composites, and the rest
should be left undetermined.

Definite humans
Perhaps the clearest images in this group are the painted stencils and prints
of hands which, as we have seen (p.l04), are quite numerous and can
dominate the art of whole caves (eg Fuente del Salin) or parts of caves
(Gargas, Castillo). Apart from a stencil on a block from abri Labattut
(which may have fallen from the wall), found between two Gravettian
layers, all hands are parietal; only the Labattut hand has any sort of date,
but studies of superposition, composition and association in different
caves suggest that this simple motif spans the entire Upper Palaeolithic.
As mentioned earlier, a few caves have incomplete hands, which may be
due to mutilation, pathologies, or some system of signals and gestures
using bent fingers. Ethnography provides a wide range of possible
explanations for hand stencils, both intact and incomplete: ega signature,
a property mark, a memorial, a wish to leave a mark in some sacred place, a
record of growth, 'I was here', or simply 'just put there'. In view of the
timespan involved, it is probable that all these roles and many more were
involved in the Palaeolithic stencils; and since the specific reasons for
making such marks seem to be remembered for only a couple of
generations 108 it is unlikely that we shall ever be certain why Palaeolithic
people painted them.
For decades, despite the steadily growing number of'Venus figurines',
it was believed that depictions of humans were rare in Palaeolithic art, and
that they were badly done (even though many animal figures were equally
incomplete or sketchy!). Now that at least 115 quite realistic human
figures are known among the engravings of La Marche/ 09 outnumbering
all other species depicted at the site, it is clear that they are not so rare; their
136 IMAGES OF THE ICE AGE

quality, together with that of the figures at nearby Angles-sur-1' Anglin,

and that ofthe ivory heads from Brassempouy and Dolni Vestonice, shows
that realistic images of people were by no means taboo, as had been
supposed.
Nevertheless, it remains true that few depictions of humans can match
the finest animal images in detail and beauty. More effort may have been
put into the animal figures for some ritual purpose, but the answer might
simply be lack of ability . Drawings of humans and animals require
different skills, and it is noteworthy that Breuil, who learned much of his
art from the 'Palaeolithic school', had a talent that was 'confined to
portraits of animals. When he ventured to reproduce human figures ... his
work was not on a higher level than that of any fairly efficient amateur. ' 110
Breuil drew amusing caricatures of his teachers at school, and it is quite
possible that some of the Palaeolithic portraits which look funny to our
eyes are actually caricatures rather than attempts at serious portraits.
Genitalia are rarely depicted on Palaeolithic humans, even on
figurines. 111 On bodies drawn in profile, the phallus can be seen, but the
vulva cannot. Where genitals are absent, or on isolated heads, males can
still be confidently recognised from beards and moustaches, whereas
breasts denote females (though it should be noted that the Brno male
statuette has small breasts or exaggerated nipples). At LaMarche, using
these criteria, there are 13 definite males (including 11 with beards). 112
Of 51 bodies drawn at the site, 4 are definitely male, but only 27 of the
Fig. 93 Tracing of engraved
rest can confidently be seen as female: 8 of these have breasts; the others plaquette from LaMarche (Vienne),
have been identified on the basis of the size of hips and buttocks . 113 and the face of a bearded man
Heads without facial hair have to be left 'neutral' -length of hair is not a extracted from the mass.
sure guide . Hence the Brassempouy head is unsexed, although it is usually Magdalenian. (After Airvaux &
called a 'Venus' or 'Lady' , and it has even been placed on a female body in Pradel)
WHAT WAS DEPICTED? 137

one reconstruction. 114 Other scholars have attempted to identify males on

the basis of nose-shape or faces that jut forward 11 5 - despite the fact that the
two females engraved at Isturitz have equally jutting faces (Fig.lOO).
There is a wide range of nose- and face-shapes at LaMarche alone.
The presence of adornments can denote females, though it is not an
infallible guide: the nine bodies at LaMarche that seem to have bracelets
and anklets are all female; analogous ornaments including 'necklaces' can
be seen on the above-mentioned females from Isturitz (identified by their
breasts), as well as on the similar' Femme au Renne' from Laugerie-Basse
(who has a vulva but no breasts), and on a number of female figurines from
the Soviet Union.
Clothing is rarely clear; the 'Venus ofLespugue' (Fig. 94) seems to have
a garment of some kind showing at the back, and belts are depicted
occasionally (eg at La Marche and Kostenki). The bearded male,
sculptured, engraved and painted at Angles-sur-l'Anglin, appears to be
wearing a garment with a fur collar, and perhaps also some form of
headgear. 116 However, the so-called 'Femme a l'anoral< at Gabillou is
unsexed, and the 'hood' may simply be a hair-line. Elaborate hairstyling is
extremely rare (eg the Brassempouy head, the 'Venus' of Willendorf).
Using the strict criteria outlined above, it has been established that La
Marche has 13 male, 27 female, and 69 undetermined humans; a study of
410 parietal and portable human figures from western Europe (including
those of La Marche but not those of Gonnersdorf) produced a very similar
picture: approximately 10% male, 25% female and 65% neutral. However,
70 figures from eastern Europe - all in portable art - proved to be
approximately 4% male, 60% female and only 36% neutral. 117 The
difference with the west is largely due to the fact that almost all the eastern
figures are statuettes, on which gender tends to be more easily recognisable.
In Palaeolithic depictions of humans as a whole, details such as
eyebrows, nostrils, navels and nipples are extremely rare. The legs are
often too short (as in the art of many later cultures); the legs and/or feet of
figurines are held together or slightly apart. Few figures have hands or
fingers drawn in any detail (examples include LaMarche, the 'Venuses' of
Laussel and Willendorf, the man in the Lascaux shaft-scene, etc). Arms on
statuettes are usually held close to the body' for technical reasons, and rest
on the stomach or, sometimes, the breasts. On other figures, the arms may
be at the sides, raised horizontally (as on the Sous-Grand-Lac man), or up
in the air- all these poses are represented at La Marche. 118
Except for the two ivory heads mentioned earlier, very few statuettes
(apart from those of Siberia) have any kind of facial detail; their heads are
held erect or tilt forward slightly (though one from Kostenki tilts
upwards). Many Palaeolithic humans are headless: in some cases, as in the
female bas-reliefs of Angles-sur-1' Anglin , it is clear that they never had
heads. The same is true of nine engraved humans at LaMarche, but others
at this site and elsewhere may have had their heads broken off, either
purposely or accidentally. 119
There are three heads seen full-face at La Marche; all the others,
together with 90% of the bodies, are in profile, and it is therefore worth
examining which way they face. Of 57 isolated heads, 35 face right and 22
left; of 51 bodies, 33 face right and 18left. In short, ofl08 people, 68 (63%)
face right and 40 (37%) left. Once again, as with sexing (see above), the
percentages are found to be consistent when the analysis is extended to 167
humans from other sites: 100 (60%) of them face right, and 67 (40%) left. 120
138 IMAGES OF THE ICE AGE

The depiction of women

Apart from a few definite males (eg Brno, Hohlenstein-Stadel) (Fig.49),
most carvings of humans are female. The 'Venus' figurines 121 have been
presented in so many art histories and popular works on prehistory that
they have come to characterise the period and its depiction of women; this
is unfortunate, partly because such statuettes are rare when seen against
the timescale of 25 millennia, and partly because the constant display of a
few specimens with extreme proportions presents a distorted view .
The term 'Venus' was first used by the Marquis de Vibraye in the 1860s
in connection with his 'Venus impudique' from Laugerie-Basse
(ironically, a very slim lady!), and was later adopted by Piette for the more
corpulent figures from Brassempouy and elsewhere; subsequently it
seems to have become attached primarily to the obese statuettes. 122
Despite the accepted view that Palaeolithic depictions of humans were
badly done, early scholars nevertheless tried to use them as evidence of
different races during the period. As with similar attempts involving
horses and other species (see above), they were doomed to failure. Piette
was among the first to have a try; apart from his erroneous belief that a
Grimaldi carved head was negroid, his major mistake was to see the obese
'Venuses' as steatopygous: ie having the special fatty deposits which
produce the massive, high and wide buttocks of some female Bushmen
and Hottentots. 123
In fact, he was confusing this phenomenon with the kind of buttock
development which is found in all races. Only one statuette (the
'Polichinelle' of Grimaldi) could conceivably be steatopygous; the rest
merely present proportions which one can see at any time and anywhere on
women who have produced lots of children. In short, 'Venus' figurines
have no value whatsoever as indicators of race. 124 Only four (Lespugue,
and one each from Willendorf, Grimaldi and Gagarino) have really
extreme proportions, and only Lespugue has truly monumental breasts
(Fig. 94). On the whole, the obese carvings are not anatomically abnormal
-they are simply bodies worn and altered by age and childbearing. They
seem well nourished, with their adipose tissue concentrated in discrete
areas rather than spread out as a continuous layer.
One recent study has looked at 132 Palaeolithic 'Venuses' and has tried
to estimate their age group, dividing them into the following categories:
young (pre-reproductive, with a firm body, high breasts and a flat
stomach); middle-aged (reproductive and potentially pregnant, with a
fleshy body, big breasts and a protruding stomach); and old (post-
reproductive, sagging all over). The result was 30 (22. 7%) young, 23
(17.4%) middle-aged and apparently pregnant, 50 (37.9%) middle-aged
but not pregnant, and 29 (22%) old. 125 Despite its tentative and subjective
nature, this exercise does at least suggest that the carvings probably
rep{,lloSent women throughout their adult life.
As with bison and other species, some scholars have attempted to
establish the measurements and conventions used in the construction of
Palaeolithic human figures, but made the same mistake of using the images
themselves, and not human anatomy, as the base of the study. Drawing
lozenges around the Venuses merely shows that they have the same basic
shape (as one would expect of human depictions), though with variations,
and exaggerates the degree of their deformation. Anatomical studies, on
the other hand, suggest that the figures' proportions are close to reality. 126
 WHAT WAS DEPICTED? 139

Fig. 94 The 'Venus' of Lespugue

(Haute Garonne) seen from the side.
Ivory, probably Gravettian. Height:
14.7 em. (JV). (Drawings show
front and back views after Pales and
Leroi-Gourhan respectively)
140 IMAGES OF THE ICE AGE

As was mentioned earlier (p. 85), 'Venuses' are not limited to the
Gravettian and Magdalenian; there is some reason to believe that those of
Brassempouy, at least , are Aurignacian; a possible 'rough' for one was
reported in the Solutrean of Roc de Sers, 127 but is a most unconvincing
Venus. So far, however, only Angles-sur-l'Anglin has produced both
parietal (four in bas-relief) and portable specimens (two statuettes),
together with over a dozen other human depictions, including the famous
large and small 'versions' of a male 'portrait' . 128 However, the 'Venus'
figurines were not found in any kind of privileged position. Very few such
statuettes in western Europe have any stratigraphic provenance, though
some (eg Lespugue, Courbet) seem to have been carefully hidden under
rocks in caves and shelters. In eastern Europe , however, the excavation of
open-air habitations has frequently encountered intact female figurines in
special pits in hut-floors: 129 for example, one from Kostenki I, found in
1983 and dating to about 23,000 years ago, was upright in a small pit,
leaning against the wall and facing the centre of the living area and the
hearths; the pit was filled with soil mixed with red ochre and was capped
by a mammoth shoulder-blade. 130 The eastern statuettes are often
interpreted by Soviet scholars as a mother- or ancestor-figure, a mistress of
the house.
These hidden figurines , in both western and eastern Europe, cast doubt
on the theory that Venuses were meant for public exhibition, and that
those with a protuberance at the base (Tursac , Sireuil) or with peg-like feet
were hafted on to stakes or stuck in the ground for display. 131
Finally, it is worth noting that no definite 'Venuses', either parietal or
portable , have yet been found in the Iberian Peninsula; this is partly
because bas-reliefs are limited to parts of France; but in view of the
remarkable distribution of female statuettes from the western Pyrenees to
Siberia, there seems no reason why Spain and Portugal should not contain
a few, and no doubt future discoveries will alter the situation.

'Humanoids'
This category should comprise all those figures interpreted, but not
positively identified, as being human. For example, there are many heads
which could belong to animals but also resemble some of the more
'bestialised' heads on definite human figures : one , engraved on the wall at
Comarque, represents a bear for some scholars, but others see it as
human. 132 It cannot safely be assigned to either group. A number of Fig. 95 Stylised humans engraved
rudimentary and sometimes grotesque heads on cave-walls, seen full-face , on plaquette from Gonnersdorf
are potentially human (eg Marsoulas, or Trois Freres), while others (such (Germany): two female figures are
as 'phantoms') are best left undetermined . As mentioned above depicted, 'face to face' .
(pp.132-3), the miniature figures grouped on some engraved bones have Magdalenian. The figure on the right
traditionally been interpreted as human (despite the lack of any is 11.3 em long. (Photo G. Basinski)
di~guishing features ) but may in fact be realistic birds .
This is by no means a new problem. From the start, Breuil was well
aware of the difficulties in separating 'elementary anthropomorphs' not
only from animals but also from 'signs'. It occurred to him that the
'claviform' (clublike) signs might be derived from stylised females, but he
dropped the idea because too many intermediate steps in the stylisation
were missing. 133 Today, however, we have far more 'stylised females', and
the idea has re-emerged.
Such stylisations are particularly abundant in the late Magdalenian,
though they are not exclusive to the period. They occur both on cave walls
 WHAT WAS DEPICTED? 141

Fig. 96 Engraved plaquette from (eg Fronsac 134) and on slabs and plaquettes as at Gare de Couze and La
Giinnersdorf (Germany), showing Roche-Lalinde: Gonnersdorf alone has 224 such figures on 87 engraved
four schematised females with what surfaces, as well as about a dozen figurines of similar type, in ivory, antler
may be a small child between them. and schist; some of the figurines are perforated. 135
Magdalenian. Width ofplaquette: The engraved figures are all stereotyped headless profiles (mostly facing
8em. (PhotoG. Basinski) right) with protruding buttocks; about a third can safely be identified as
female humans because they appear to have breasts, as well as arms and
other details. The other, more sketchy outlines from this site and
elsewhere have been interpreted as females because of their resemblance to
the more definite examples, 136 but they could be either young females or
males (none is definitely male), and are best left as 'neutral': some of them
would probably be classed as non-figurative if found out of context!
The Gonnersdorf females have a wide range of proportions, and occur
both singly and in groups of two , three, and more , which have been
interpreted as 'dances' for some reason. Some couples face each other, and
one woman has what may be a baby on her back (Figs.95-6). 137
142 IMAGES OF THE ICE AGE

In view of the position of Venus statuettes in open-air habitations (see

above), it is important to note that the intact figurines from Gonnersdorf
were all found in pits. 138 It will be recalled that such statuettes are thought
to have been made for long-term use while plaquettes had a short-term use
before being discarded and/or broken (see pp.54 and 77).
Various authors have proposed unilinear sequences of development
from realistic to stylised , and even to 'signs'; Breuil did this for a number
of species, and Leroi-Gourhan saw humans as being stylised into the
headless profiles and ultimately into 'claviform' signs which, like the
profiles, occur singly and in groups . 139 Alas, all such schemes are to some
degree subjective, and often ignore dating and other factors : for example,
true claviform signs are exclusively parietal, probably mid-Magdalenian
in date, and limited to the Pyrenees and Cantabria (and possibly also
earlier at Lascaux); but the headless female profiles are most abundant in
Germany (and to a lesser extent in Dordogne), and seem to be late
Magdalenian! The claviform comprises a vertical line with a bulge at one
side; but unlike the buttocks on the stylised females, the bulge is usually on
the middle or the upper half of the 'sign'! 140 It has been suggested that the
'upper bulge' on a claviform instead represents breasts, but this is based on
supposedly stylised statuettes from Dolni Vestonice and elsewhere,
comprising a rod with two little lumps near the top, and thus on analogies
with sites even further away in space and time from the claviforms. In any
case, stylisation of humans or of any other theme can and probably did take
place independently, in a number of phases and regions. 141
The claviform question demonstrates the dangers of subjective
impressions in these matters; a similar debate concerns some carvings
from Mezin (Soviet Union) which different scholars see as schematic
images of women, as phallic symbols or as birds. 142 They may, in fact, be
schematic female symbols (triangles, etc) engraved on phallic birds!
Clearly, in situations of this kind, the figures must be left undetermined,
and speculation about what they depict must never be claimed as a firm
identification.

Composites
We have just encountered some undiagnostic examples which could be
either animal or human ; but what of those figures which have clear and
detailed elements of both ? They been called 'anthropozoomorphs' .or
'therianthropes', but the term 'composites' is simpler, and does not give
priority to either the human or the animal features.
In the past, thanks to the dominance ofthe 'hunting magic' theories (see
below, p.l50), all such figures were automatically and unjustifiably called
'sorcerers', and were assumed to be a shaman or medicine man in a mask
or an animal costume.
Jlut how can we differentiate between people wearing masks, people
with jutting, bestialised faces (see above) and humans with animal heads?
Unfortunately we can't, and it is likely that all the above are represented in
Palaeolithic images . Inevitably, interpretations have been heavily
influenced by ethnographic parallels: for example, a bison-headed Fig. 97 Photomontage of the
humanoid at Trois Freres (with the supposed 'musical bow) (see Fig.104) 'sorcerer' of Trois Freres (Ariege),
bears some resemblance to North American Indians disguised as bison for together with Breuil' s version of this
a dance/ 43 while an equally famous figure from the same cave - the painted and engraved figure.
'Sorcerer', with his antlers and other animal parts (Fig. 97) - looks just like Probably Magdalenian. Length:
an eighteenth-century depiction of a Siberian shaman. 144 These 75 em. (JV, collection Begouen)
144 IMAGES OF THE ICE AGE

Palaeolithic images may indeed be disguised humans, although it cannot

be proved.
On the other hand, many scholars have noticed the strange resemblance
between the Hohlenstein-Stadel statuette (Fig.49) and Egyptian figures of
a deity with a human body topped by a lion's head; in this case, therefore,
the ethnographic analogy points towards an imaginary being - again
without proof. Likewise, Breuil called the Trois Freres 'sorcerer' the 'God
of the Cave' and considered him an imaginary figure, dominating the
'sanctuary'. Not all such 'sorcerer' figures, however, are in imposing
positions- indeed, some are hidden away among other images, and can be
very hard to see. 145
The Trois Freres figure is certainly strange: only the upright position
and the legs and hands are really human. The rest is a mixture of different
animals- the back and ears of a herbivore, the antlers of a reindeer, the tail
of a horse, and the phallus in the position a feline would have it. 146 Very few
such composite figures have horns or antlers (apart from the Trois Freres
examples, Gabillou has the clearest); overall, composites or 'monsters' are
fairly rare in parietal art- according to Leroi-Gourhan, only about 15 sites
have them, with no more than half a dozen in each, and some of those he
includes are probably not even humanoid . 147
The distribution of human figures is interesting: as with parietal and
portable art, there are many sites with none, some with one or two, while
others like LaMarche or Gonnersdorfhave many. Pales' sample of 410 in
western Europe (including La Marche but excluding Gonnersdorf)
comprised 187 realistic figures, 214 humanoids, and 9 composites 148 - the
vast majority (c 80%) were engravings, with carvings accounting for
almost all the others: painted humans are extremely rare .
Definite humans are scarce in parietal art, and there are few women
resembling the Venus figurines or the stylised profiles (examples include
Angles-sur-l'Anglin, Laussel, Fronsac). The distribution of sexed and
unsexed humans within caves will be studied later (p . l72).
Portable art accounts for over 75% of Palaeolithic human depictions; as
mentioned earlier, at Dolni Vestonice the human figures were not found in
the same hut as those of herbivores, but were with carnivore figures. At
Gonnersdorf they were scattered throughout the excavated area, unlike
engravings of mammoths and birds: animals predominated on some slates
at this site, while other slates were primarily devoted to stylised females;
no intentional grouping of human figures with animals or with symbols
could be discerned. 149
Non-figurative: 'signs'
The non-figurative category of Palaeolithic markings was neglected until
relatively recently, for the simple reason that it seemed uninteresting, or
i!JlPossible to explain or define. Many lines, as mentioned earlier (p.44),
were simply ignored as 'traits parasites'. Nowadays, however, thanks to
the attention paid to the abstract 'signs' by Leroi-Gourhan and to the
discovery of similar non-figurative motifs in Australia and elsewhere (see
above, p.28), we have to come to terms with the possibility that these
marks may have been of equal, if not greater, importance to Palaeolithic
people than the 'recognisable' figures to which we have devoted so much
attention. Certainly, it has been estimated that non-figurative marks are
two or three times more abundant than figurative, and in some areas far
more: for example, on 1,200 engraved pieces of bone and antler from 26
WHAT WAS DEPICTED? 145

Magdalenian sites in Cantabria, there are only 70 identifiable animals - all

other motifs seem non-figurative . 150
As already mentioned, it is hard to separate this category from the two
others; and it comprises a tremendously wide range of motifs , from a single
dot or line to complex constructions, and to extensive panels of apparently
unstructured linear marks. There are significant differences in content
between portable and parietal examples , which make it extremely difficult
to date parietal 'signs', particularly in cases where they constitute the only
decoration and figurative images are absent, as at Santian and other
sites. 151 Even in caves with both figurative and non-figurative images on
the walls, 'signs' can be either totally isolated, clustered on their own
panels or in their own chambers, or closely associated with the figurative -
and sometimes all of these .
In the past, certain shapes were assumed to be narrative or pictographic:
ie to represent schematised objects on the basis of ethnographic
comparisons (see Chapter 7, below) or, more often, subjective assessment
of what they 'looked like': hence, we have terms such as 'tectiforms' (huts),
'claviforms' (clubs), 'scutiforms' (shields), 'aviforms' (birds),
'penniforms' (feathers), etc. These are no longer taken literally but, as they
have entered the literature so decisively, are retained simply as rough
guides to particular shapes. Even if these motifs were ultimately derived
from real objects, they could now be purely abstract, 152 although it is
impossible for us to separate them fully from 'realistic' figures: some signs
may indeed be schematised or abbreviated (or unfinished!) versions of
some object. Some authors now regard 'signs' as 'ideomorphs' - ie
representations of ideas rather than objects, but we simply don't know
whether they are real or abstract or both.
We inevitably apply our understanding to what we see; our
background, culture and art history come into play, and hence the early
scholars tried to translate some signs as if they were hieroglyphics, in the
vain hope of finding a Palaeolithic Rosetta Stone. Even if these signs were
representational, their meaning would be far from straightforward without
an informant: for example, among the Walbiri of Australia, a simple circle
can mean a hill, tree, campsite, circular path, egg, breast, nipple, entrance
into or exit from the ground and a host of other things; similarly,
'translation' of simple signs and pictographs in North America can often
produce differing and unexpected results. 153 A mere resemblance of shape
does not necessarily mean that a motif is an image of an object.
What is clear is that the simpler motifs are more abundant and
widespread, as one would expect, since they could have been invented in
many places and periods independently. The more complex forms,
however, show extraordinary variability and are more restricted in space
and time, to the extent that they have been seen as 'ethnic markers',
perhaps delineating Palaeolithic groups of some sort: 154 hence,
'quadrilateral' signs seem concentrated in Dordogne, particularly at
Lascaux and Gabillou; a different kind of 'quadrilateral' is found in a
Cantabrian group of caves (Castillo, La Pasiega and Las Chimeneas, all in
the same hill, together with Altamira, some 20 km away); 'tectiforms' are
found in the Les Eyzies region (Dordogne), in a small group of caves only a
few kilometres apart, including Font de Gaume and Bernifal; 155 very
similar 'aviforms' are known at Cougnac and Pech Merle, 35 km apart;
triangles are predominantly found in the centre of Spain, but are rare in
Cantabria; perhaps most remarkable of all is the distribution of true
146 IMAGES OF THE ICE AGE

'claviforms', which are known in Ariege (Niaux, Trois Freres, Tuc

d'Audoubert, Le Portel, Fontanet, Mas d' Azil) and in Cantabria (Pindal,
La Cullalvera) some 500 km away (and perhaps also at Lascaux).
As will be seen below (p.169), Leroi-Gourhan divided 'signs' into two
basic groups, the 'thin' and the 'full', which he linked to a sexual
symbolism (phallic and vulvar); he later added a third group, that of the
dots. 156 In view of the wide range of shapes on display, such a division was
far too simplistic and schematic; using geometric criteria, other scholars
have proposed seven or even a dozen classes of parietal sign but still found
some very hard to fit in. 157
Similarly, the non-figurative motifs in portable art have been receiving
an increasing amount of attention lately, and likewise have been divided
into different groups. 158 Meg Conkey, for example, has sought out the
basic units of decorative systems in Cantabrian portable art, and their
structural interrelations. In a sample of 1,200 pieces, she found a set of
about 200 distinct 'design elements', which she organised into 57
'classes' . 159 Her analysis revealed that a core set of 15 elements was used
throughout the Magdalenian, and was widespread within and outside the
region. As with the parietal signs, therefore, it will be necessary to focus on
the more complex designs in order to identify regional variations in style
(and possibly distinct social groups of some kind).
In both parietal and portable art, a full survey of the presence and
interrelationships of different motifs, as well as of their association with
each other and with other figures, is required, but will entail a more
complete published corpus than we have, followed by computer analysis.
It is the presence and absence of particular combinations which is
revealing: in parietal signs, for example, very few binary combinations
occur out of the range of possibilities, and only signs found in binary
combinations also occur in 'triads' . 16 Clearly these marks were not set
down at random, but follow some set of rules and simple laws.
Can they therefore be seen as a primitive form of writing? Theories
about this go back to the discovery of portable art, when a variety of
enigmatic motifs associated with animal figures were seen as possible
artists' signatures by a number of scholars including Lartet, Garrigou and
Piette. 161 It is inevitable that, despite their wide variety of shapes, some of
the Palaeolithic signs will resemble some of the simpler characters in
certain early forms of writing; after all, the range of possible basic marks is
somewhat limited. It has recently been claimed that some Palaeolithic
signs have very close analogies with characters and letters in ancient
written languages in the Mediterranean, the Indus valley and China/ 62
but, as with resemblances to objects, this does not necessarily prove
anything. As far as parietal signs are concerned, it often appears to be their
presence which was important, rather than their layout or orientation; 163
inst~d of forming a script, they were sometimes joined through
superposition, juxtaposition or actual integration to form composites.
Where motifs on portable objects are concerned, however, it is still
possible that they include some sort of 'pre-writing'; we simply don't
know. What is almost certain is that the meaning of the signs and marks,
no matter how abstract they may appear to us, must have been clear to the
maker and to those who saw and/or used them. We can see this today with
our road signs and warnings: some have meanings obvious to everyone,
others have to be learnt, but all are known to those who operate within that
system. 164
 WHAT WAS DEPICTED) 147

Fig. 98 'Signs' in the cave of

Castillo (Santander) . Width: 58 em.
(JV)

Colour schemes
A final factor in the content of Palaeolithic art which we have yet to
consider is the use of different colours; we have already seen that the range
was limited and usually involved a straight choice between red and black.
At first sight, it might seem possible to present a theory that one of these
represented males and the other females (for example in hand stencils),
just as we have blue for boys and pink for girls; but the situation is far more
complex.
Where signs are concerned, their colour seems linked to different
regions and periods, but red usually dominates. As mentioned earlier
(p .llO), red is more visible in dark caves , so that whatever message (of
warning, topographic guidance, instruction, etc) the signs conveyed could
readily be seen and understood . In the Iberian Peninsula, 23% of the
148 IMAGES OF THE ICE AGE

parietal signs are engraved , 27% painted in black, and no less than SO%
painted in red; 165 in France too, about two-thirds of painted signs are red,
and they are often the only (or almost the only) red figures in the cave. 166
This is particularly noticeable on friezes of black outline animals, such as
Niaux's Salon Noir, or the frieze at Pech Merle (Fig.73).
The use of colour for both signs and animal figures alters significantly
through space and time: a survey of 38 caves in France and Spain found
that black accounted for 48% of figures and red for 43.7%, only 5.3% in
yellow and brown , and mixtures (bichromes and polychromes) a mere
3% 167 - a statistic which underlines the fact that the latter are exceptional
and unrepresentative of Palaeolithic art, despite their popularity in art-
history books.
However, when the paintings were dated according to Leroi-Gourhan's
styles (see above, p.62), a difference was found: in Style III (late Solutrean!
early Magdalenian), 56% of figures were red, and 34.7% black, while in
Style IV (mid- and late Magdalenian) these percentages were almost
exactly reversed, with black animals becoming particularly dominant at
the end of the period. Style IV appears to have a marked tendency for black
animals and red signs.
Black and red, whether in signs or figures, are sometimes found in
different parts ofthe same cave (eg Pech Merle, Castillo, Altamira), which
may denote a chronological difference, or some purposeful colour scheme.
Some signs are associated with a particular colour: about 94% of
claviforms are red, as are 100% of painted tectiforms in Dordogne; most
dots are red, and isolated dots are always red. 168
There are also a few such associations discernible in animal figures, but
these appear to be of regional rather than general significance: for
example, in Spain 95 % of hinds are red, whereas stags can be either colour;
Las Chimeneas has only stags, and they are all black, whereas all the 17
hinds in Covalanas are red . At Castillo and La Pasiega, however, red
dominates for deer, no matter what their sex. 169

All of the above should have made it clear that in Palaeolithic art there are
no absolute rules, no certainties, and a very broad range of complex
problems; but also that this is no random accumulation of pictures - on the
contrary, it is characterised throughout by careful selection of surface,
size, technique, colour, species, anatomical detail, degree of accuracy or
stylisation. It is now necessary to examine the sense that different people
over the last century have tried to make of all these facts, and in what
directions current research is taking us.
 7
Reading the Messages

So far, we have looked at Palaeolithic art and asked questions about where,
when, how and what; it is now time to grasp the nettle and tackle the most
difficult topic of all: why?

Early theories: art/or art's sake

The first and simplest theory put forward to explain the existence of art in
this period was that it had no meaning; it was simply idle doodlings,
graffiti, play activity: mindless decoration by hunters with time on their
hands . Such a view, formulated most clearly by Lartet and Piette in the
nineteenth century, reflected not only the anticlericalism of de Mortillet
(see p.21) who refused to believe that Palaeolithic people had any religion,
but also the spirit of the age: 1 art was still seen in terms of the recent
centuries, with their portraits, landscapes and narrative pictures. It was
simply 'art', its sole function was to please and to decorate. In addition,
Palaeolithic art indicated that the people of the period had plenty ofleisure
time, implying that hunting was easy, there was lots to eat, and life was
pleasant, calm and (in Piette's opinion) largely sedentary.
The 'art for art's sake' view arose from the first discoveries of
Palaeolithic portable art in the 1860s and 1870s, and it must be admitted
that there was little in the content and composition of these pieces to
suggest any deeper purpose; indeed, it remains quite possible that some of
the decoration of portable objects, and especially of utilitarian items, may
simply be decoration - although more may be involved even here, from
marks of ownership to the complex notations and seasonal images studied
by Marshack (see below, p.183). Similarly, the newly discovered
engravings in the open air do not seem as enigmatic as those inside caves;
but it should be remembered that those of Campome are all incomplete
figures, not simple pictures of animals which happened to be passing.
Of course, when parietal art began to be found, it rapidly became clear
that something more was involved: as will be seen later in this chapter, the
restricted range of species depicted, their frequent inaccessibility and their
associations in caves, the palimpsests and undecorated objects or panels,
150 IMAGES OF THE ICE AGE

the enigmatic signs, the many figures which are purposely incomplete,
ambiguous or broken , and the caves which were decorated but apparently
not inhabited , all combine to suggest that there is complex meaning
behind both the subject-matter and the location of Palaeolithic figures.
There are patterns that require explanation, repeated patterns which
suggest that individual artistic inspiration was subject to some more
widespread system of thought. It is worth noting that , even in 1878, when
only a few pieces of portable art were known, Cartailhac had already
sensed that there was a purpose and a meaning behind this art, having
probably been influenced by the writings of Tylor .2
Nevertheless , despite all the above evidence, the theory has recently
surfaced again ,3 with a claim that the art has no meaning, no religious,
mythic, magical or practical purpose; it is simply play, mere aesthetic
activity indicative of an inherent human desire to express ourselves
artistically. This view resembles the assumption that someone is innocent
until proven guilty: nothing in the periods before writing can be proved
absolutely to be religious; but if Palaeolithic art simply comprised 'objects
of contemplation' , one would expect to find a pretty broad spectrum of
subjects, accumulated more or less randomly on suitable surfaces . Yet the
opposite is the case. Besides, countless studies of modern 'primitive'
cultures make it clear that there is not necessarily a boundary between the
sacred and the profane - everything can have a meaning for them, whether
it be decoration , dances or mere gestures.4 To compare Palaeolithic art to
modern graffiti , or to attribute it to some inherent human desire to leave a
personal mark in places which are isolated or which stir the emotions, is a
projection of the present into the past.
It is undeniable that the artists must sometimes have derived great
personal satisfaction, and perhaps enhanced prestige, from the best of
their work, and certain 'realistic' figures are probably finer and more
detailed than was strictly necessary to make their message intelligible - just
as medieval craftsmen lovingly put minute details into stained glass which
was placed so high up that nobody could see them without binoculars!
However, simple aesthetics explains nothing and has no bearing on
meaning. An aesthetic interpretation 'reduces cultural phenomena to an
innate tendency and directs explanation inward to mental states about
which we can know nothing' - to infer a state of mind (pleasure) from the
art, and then use it as an explanation is a circular argument! 5 The theory
tells us more about its proponents ' reaction to Palaeolithic art than about
the artists.

Hunting magic, hunting and shamanism

After the death of Piette in 1906, utilitarian theories took over, and
~marily that of hunting magic. This was largely caused by the newly
published (1899) account by Spencer and Gillen of the life and beliefs of
the Arunta of central Australia, which inspired Reinach 6 to equate these
'primitive' users of stone tools with those of the Palaeolithic, and therefore
to assume that the same motivation lay behind the art of both cultures .
The Arunta were said to perform ceremonies in order to multiply the
numbers of the animals, and for this purpose painted likenesses of these
species on rocks. The analogy seemed perfect, and Palaeolithic art must
therefore have had a magico-religious cause. For a few decades, all ideas
and interpretations were subjectively chosen in this way from the ever-
 READING THE MESSAGES 151

growing mass of ethnographic material from around the world. This is

quite understandable - it appeared self-evident at the time that 'primitive'
societies provided lots of precious information about the making and
function of rock-art in the past. Almost a third of the first monograph on
cave-art was therefore devoted to a review of the available ethnographic
evidence: the present would teach us about the past , through the
parallelism of the artistic activity of all primitive hunting peoples. 7
'Sympathetic magic', including hunting magic, operates on the same
basis as pins in a wax doll: ie the effect resembles the cause. The depictions
of animals were produced in order to control or influence the real animals
in some way: any injury to the image would likewise be inflicted on its
subject. Desirable effects such as ensuring that the prey would be as fine
and well fed as the image were also possible. In short , this kind of art is
intended to exert influence at a distance, rather like modern
advertisements; it is purely functional, and directly related to the need for
food.
Breuil adopted the 'hunting magic' view because he felt that Palaeolithic
art sprang from the hunters' anxieties about the availability of game. Most
other scholars followed, particularly Begouen who saw evidence for ritual
and magic in almost every aspect of Palaeolithic art.8 One feature which he
constantly stressed was that some images seemed to have been 'killed'
ritually with images of missiles , or even physically attacked with a weapon
or by hand, presumably to improve the chances of the hunt, or (in the case
of carnivore figures) to vanquish a fearsome enemy. Marks at the mouths
or nostrils of animals were always interpreted as blood being vomited by a
dying beast; marks on their bodies were always wounds or missiles, and
dots were stones thrown at them: the famous bear of Trois Freres, covered
in 189 small circles, was 'thus an animal stoned to death , with blood
pouring out of its mouth.
The theory accounted for the broken and short-lived engravings on
plaquettes: once the animal had been 'killed' and the ritual was over, the
image had no further use or significance, and was discarded. It also had an
Fig. 99 'Tectifonn' signs painted explanation for why so many figures were drawn on the same pebble (as at
on a polychrome bison at Font de La Colombiere) or panel when so many equally good stones or surfaces
Gaume (Dordogne) . Probably were available but left blank: they were 'good hunting' pebbles or panels.
Magdalenian. Bison about 1.6 m If the first figure drawn there produced the desired result, they were used
long. (JV> again and again. 9
Such assumptions also influenced or dictated the intepretation of
'signs', which were thought to represent whatever object they looked like:
the 'claviforms' were clearly clubs or throwing-sticks, and 'penniforms'
were arrows, while the Dordogne 'tectiforms' were sometimes huts, but
mostly detailed diagrams of pit-traps and gravity-traps, especially in cases
where they were superimposed on a bison or mammoth (but why would
only this tiny area depict such devices?). Other motifs were assumed to be
lassos, nets and so forth. 10 Occasionally a more exotic ethnographic
analogy was used, as where some signs were compared to the traps found
in the East Indies that are intended for malignant spirits, for the souls of
the dead, or for the spirit of game.''
Needless to say, the subjectivity, presupposition and wishful thinking
which permeated this theory led to many errors: one of the most blatant
concerns the clay bear of Montespan which, from the start, was presented
as the epitome of the 'killed' image; the advocates of hunting magic
claimed that they could see holes in the bear which meant that it had been
152 IMAGES OF THE ICE AGE

stabbed with spears or struck with missiles in some sort of ritual attack.
Despite the fact that these 'holes' were simply the natural texture of the
clay, 12 that the bear is only 1 m from the cave wall , and that the ceiling
descends to a very low level immediately beyond it, one still finds dramatic
verbal and graphic reconstructions of the scene in popular books, with
warriors dancing round a bear head/hide draped over the dummy, and
throwing spears at it! These ridiculous flights of fancy are the legacy of the
hunting magic theory , whose proponents saw what they wanted to see.
Had they been a little more objective about the data , at Montespan and
elsewhere , they would have realised that they were constantly stretching
and adapting the theory to fit the evidence , or carefully selecting the facts
to fit the theory. For example, faced with the rarity of carnivores in parietal
art, they claimed that these were dangerous animals , difficult to hunt, and
in any case would not have been considered very tasty. Yet one would
think that hunting magic would be most useful for dangerous animals! In
any case , carnivores are not so dangerous (they tend to keep well away
from humans most of the time), and bulls or stags can be far more
aggressive; lions are caught and eaten in Africa, and bears and dogs are
considered delicacies in some parts of the world. 13 Besides, how can one
reconcile that explanation with the similar rarity of reindeer or birds in
parietal art, resources which were of great economic importance at times?
Here, the proponents claimed that it was unnecessary to depict these
species because they were in plentiful supply and easy to obtain! The
inconsistencies in logic are blatant.
Overall, there are very few Palaeolithic animal figures with 'missiles' on
or near them : bison seem to have the most, but less than 15% of bison are
marked in this way; many caves have no images of this type at all. 14 These
marks could be all manner of things (see other theories, below); even if
they were missiles, we cannot tell whether they were drawn at the same
time as the animal , or years or centuries afterwards: in some cases, such as
a pseudo-arrow under the neck of a reindeer from la Colombiere, the
'missile' was engraved before the animal! 15 One of the most famous
examples is a rhinoceros on a pebble from the same site which was thought
to be pierced by a number of feathered projectiles: yet such missiles are
weak and aerodynamically unstable, and certainly could not penetrate
rhino hide. 16
Similarly, the marks at the mouths and/or nostrils of animal figures are
open to many other interpretations , such as voice, dying breath, or breath
to show that the animals are very much alive (and see below, p.l58)- they
have a variety of forms , and are mostly drawn on bison, bears and horses. 17
Supposedly 'falling' animals are often simply drawn on vertical supports,
which dictated their orientation.
Besides, what is one to make of 'missiles' associated with human or
b.manoidfigures? The classic example is the engraving of two women
from lsturitz, one of whom has a harpoon-like motif on her thigh , which is
identical to a mark on the bison engraved on the other side of the same
bone (Fig.IOI)! A little ibex in Niaux appears to have lances sticking out of
it, but the same is true of the man at Sous-Grand-Lac, or the humanoids of
Cougnac and Pech-Merle (Fig.l02).
Even if the figures marked with 'missiles' were symbolic killings, there
are many possible reasons for them: they could be mythological, perhaps,
or some sort of sacrifice. Sympathetic magic is just one possibility.
However, most scholars now dismiss the labelling of these motifs by what
READING THE MESSAGES 153

they look like to our eyes , and avoid words such as 'arrow' in favour of
more objective terms, descriptive of shape (eg chevron with a median
longitudinal line).
Breuil and others believed that the carnivores were drawn so that the
artist might gain their strength, their skill in catching game , or perhaps be
protected from them- a mixture of envy and fear; but in that case, why are
carnivore images so rare? Clearly, hypotheses of this kind are untestable .
In the same way, opponents of hunting magic argued that many figures
were far too detailed and perfect for the job, since presumably a simple
outline would have done; but adherents of the theory countered that the
better the portrait, the more effective the magic! The arguments go around
in circles and lead nowhere .

But if hunting magic is no longer upheld as the primary motivation,

Palaeolithic art is still seen as a 'hunter's art' by some scholars who, even
today, take some of it at face value as a narrative. This is by no means a new
concept: as we have seen , many signs were interpreted as objects
associated with the hunt, and the scene of the wounded bison and the man
in the Lascaux shaft (Fig.ll8) has often been considered a depiction of a
real event, to the extent that a search was made for the unfortunate hunter's
grave nearby!
Proponents of 'hunting art' pick out the very few examples of what they
consider to be hunting or stalking scenes , dangerous situations, and
animals which are dead, wounded, alarmed or bolting, and present them
as if they were typical of Palaeolithic art as a whole; 18 however, not only are
these extremely rare, but their interpretation is by no means
straightforward, and we have already seen (pp . l32-3) that some of the
'groups of hunters' may really be birds, and that the 'dying' or 'wounded'
bison of the Altamira ceiling may actually be giving birth or rolling in the
dust. The art is considered to have been made by men , and to reflect their
obsession with game animals and the hunt, rather like the endless pictures
in modern American hunting magazines. 19
A similar view, which follows on from the idea mentioned earlier
(p.l24), that the deer figures at Lascaux and elsewhere are accurate
records of prize stags, is to see the whole thing as 'trophyism': ie the
hunting and display of trophies by men in order to impress and get the
girls. The demonstration and advertisement of their hunting prowess,
with the prey's attributes emphasised through twisted perspective,
improved their status and thus their reproductive success. 20 Palaeolithic
art was thus equivalent to a collection of scalps or medals. The idea is
appealing, but does not stand up to scrutiny; it does not explain the
enigmas in die art such as 'incomplete' or composite images, the large non-
figurative component, and the layout in caves .
The suggestion that the organisation of figures within the caves was
arranged according to trophy values makes little sense: in that case, one
would expect to find ferocious carnivores, bulls and prize stags
dominating the principal panels, which is not the case. However, one critic
ofLeroi-Gourhan's statistical approach to cave-layout (see below, p.l68)
did confirm that there seem to be three groups of figures: the large
herbivores (horse, bison, aurochs), the smaller herbivores (deer, ibex) and
the dangerous animals (rhino, bear, feline); rather than interpret them in
terms of some metaphysical plan, however, he suggested that their
distribution in the caves reflected their economic categorisation by a
154 IMAGES OF THE ICE AGE

Figs . 100/101 Engraved bone from

/sturitz (Pyrenees Atlantiques): on
one side there are two females, one of
whom has a barbed sign on her leg.
The other side depicts a bison with a
barbed sign on its side. Magdalenian.
Total length: 10 em. (JV)

Fig. 102 Humanoid apparently

'pierced by missiles', in the cave of
Pech Merle (Lot). Note the strange
'aviform' sign above . Probably
Solutrean. The humanoid is 50 em
high . (JV)
 READING THE MESSAGES 155

hunter, based on the amount of meat they provided and the risk they
involved. 21
Another interesting hypothesis, based on a literal reading of the art, is
that of corralling: once again, it is by no means a new idea, and the strange
circular images drawn at La Pileta, with double finger-marks dotted
around inside them, have often been interpreted somewhat tenuously as
corrals with animals (or ruminant tracks) inside them 22 (even though
identical marks are also found inside animal figures in the cave), while
animals with signs painted on or near them, at Marsoulas, Castillo, Niaux
and elsewhere, were thought by some authors to be in enclosures. 23
Recently, however, an expert on animal drives in North America has
proposed that much of the painted decoration of Lascaux constitutes a
diagram depicting an animal drive, with the rows of dots and the rectangles
representing the drive-lines (made of small piles of rocks or similar),
blinds and corrals. 24 The barbed signs near some animals are brush
barriers which help to funnel the animals into a corral. The little horses are
heading for one enclosure, the 'jumping cow' is skidding to a halt in front
of another, while the falling horse is going over the edge. This is not the
first time that the Lascaux signs have been interpreted in these terms: the
quadrilateral images have previously been seen as Texas-style gates for
shutting in cattle and horses/ 5 but the new hypothesis treats the cave itself
as a symbolic drive, forcing the animals down the narrow passages to the
'fall' at the end; the round Hall of the Bulls was perhaps conceptualised as a
pound, and the cave itself could be a memory-device to preserve
knowledge of how and where to place the cairns and corrals .
Similarly, the round chamber at Altamira is seen as a symbolic pound,
with a drive depicted on the ceiling; 26 the curled-up bison at the centre are
already dead, while those around them stand and face the hunters (the
male humans engraved at the edge of the group). The Altamira ceiling has
previously been interpreted as animals falling down a cliff, 27 but never
before as a symbolic pound as well.
The hypothesis has the merit that it concerns a technique which is
certainly the most efficient way to exploit large herbivores, and which was
surely used throughout the Upper and probably the Middle Palaeolithic; it
is also solidly based on ethnographic information, including the methods
of depicting such drives in other cultures. Indeed, the interpretation
seemed self-evident to the expert on animal drives as soon as he saw these
caves, so close were the analogies to ethnographic examples. It is not
claimed that all the 'signs' are corrals or drive-lines, but their close
association with the animals at Lascaux certainly make his hypothesis
thought -provoking.
In the last two decades we have become so accustomed to esoteric
interpretations of Palaeolithic art that we tend to forget that not all of it is
necessarily mysterious and metaphysical: much of it may well be mundane
and deal with everyday concerns. Certainly some aspects of animal
exploitation seem to be present: for example, there are one or two
Fig. 103 Carved antler (possibly a depictions of what may be stripped carcasses, or of animal hides spread
spear-thrO'Wer) from Le Mas d' Azil out; 28 and the earliest art on blocks includes a series of what appear to be
(Ariege) shOVJing three horseheads, animal prints (probably a bear's front paw - a large cupmark with an arc of
one of them apparently skinned. 3-5 little ones above); 29 other motifs which have traditionally been
Magdalenian . Length: 16.3 em. interpreted as vulvas may also be horse-hoof or bird-foot prints/ 0 while the
(JV) series of double incisions on the handle of the carved Lascaux lamp could
156 IMAGES OF THE ICE AGE

be the prints of an ungulate such as a boar. In any case, it is certain that the
observation and recognition of animal and bird tracks were of great
importance in Palaeolithic life, and it is no surprise that they were
occasionally reproduced .
However, even if hun ting played a role in the production of some
Palaeolithic art , for any or all of the reasons outlined above, the fact
remains that the art's content is far more complex. We have already seen
(Chapter 4) that the species depicted cannot be taken as an accurate record
of what was present in the environment; and it is equally true that they
often bear little relation to the use people made of what was available: ie to
the faunal remains in occupation sites .
It is often the case that we do not obtain an adequate picture of diet from
the bones, because of differential preservation, or because the sample is
small; and, as we have seen , it is often theoretically possible that the people
who ate the animals were not from the same period as the artists, though in
most cases they probably were.
However, the results from many sites suggest that depicted species
rarely correspond to eaten species in terms of quantity. This applies not
only to parietal but also to portable art, and was noticed by some of the
earliest specialists such as Cartailhac . For example, at Gonnersdorf the
mammoth is the second most frequently depicted animal (after the horse),
but is very rare in the fauna, while the reindeer, which abounds in the
fauna, is absent from the art. 31 At Dolni Vestonice, the mammoth
dominates the bones but is rare in the art, which is primarily devoted to
carnivores .32 At La Vache, the ibex is overwhelmingly predominant in the
fauna but only accounts for about 18% of the depictions, while the horse,
which predominates in the art (over 25%) represents only 0.01% of the
faunaP 3 The bones from the grotte des Eyzies and the abri Morin are
dominated by reindeer , but their portable art is primarily devoted to
horses. 34 On the other hand, both the slabs and the bones of Limeuil are
dominated by reindeer, and the horse dominates in both at Gonnersdorf.
In parietal art , the example always given is Lascaux, where reindeer
account for 90% of the bones but were drawn only once (0.16%); as
mentioned earlier (p.124), it is possible that many of the cave's stag figures
may actually be reindeer, which would alter this disparity a little; but
horses still account for 60% of the figures. Bison, aurochs and ibex are in
the cave's art but not in its bones; this situation is reversed for boar and roe
deer.
The same picture is repeated elsewhere: the ibex occurs in the art of
Pair-non-Pair but not in its fauna; at Comarque, over 75% of the bones are
reindeer, but the figures are mainly horses; 35 at Villars the people ate
reindeer, but did not draw any; in the Ardeche caves, the mammoth
predominates in the art (almost half of the figures) and the ibex and
r~deer are almost absent, whereas the bones show an abundance of the
last two species and an absence of mammoth; at Gargas there are no
reindeer figures although over a third of the bones are of that species,
whereas the mammoth, absent from the fauna, was depicted six times.
The situation recurs in Cantabria: at Altamira they drew bison and ate
red deer; at Ekain the 34 horses are 57.6% of the animal figures, but there
are almost no horse-bones in the fauna which is dominated by ibex and red
deer; while at Tito Bustillo, the fauna is heavily dominated by red deer (up
to 94%) with ibex second and very few horse bones, whereas the parietal
READING THE MESSAGES 157

art has 27 horses (37.5%) , 23 red deer (31.9%) and 9 ibex (12 .5%).36 On the
other hand, the red deer dominates both the art and the bones at Covalanas
and at La Pasiega.
In short , the relationship between depicted and eaten species ranges
from non-existent to quite good ; once again , there is no rule , no
homogeneity. Nevertheless, it remains true that the restricted range of
species depicted has no direct equivalent in any excavated level's sample of
bones, and it is clear that the motivations behind the art were different
from the environmental factors and economic choices which produced the
faunal remains. A similar situation occurs in Gallo-Roman sites , where the
animals depicted on pottery are over 75% wild, with dogs dominating the
other 25%, whereas around 90% of the bones come from domestic
species. 37
The disparities have been presented in the form of cumulative graphs
which demonstrate , for example, that art and bones vary far more widely
at Lascaux than at Gabillou .38 Another recent study , however, which
compared the combined percentages of depicted species in 90 French
caves with those of bones from 151 French sites spanning the entire Upper
Palaeolithic, and those of35 Spanish caves with the bones from 30 Spanish
sites, was a waste of effort: 39 it only makes sense to compare an individual
cave's figures with the bones from the same site, or at least from the same
locality and period, since climate and local topography played a major role
in economic choice- for example, ibex dominate the fauna in rugged ibex
country, the other herbivores in flatter areas.

Although we can accept that the practice of sympathetic magic and the
actual depiction of hunting may account for a small amount of Palaeolithic
art, the evidence for them is extremely scarce. However, there is a related
phenomenon which could lie behind rather more of the art: shamanism.
Many early scholars, as we have seen, interpreted the 'composite' figures
as sorcerers or shamans in masks, basing their argument simply on
similarities to ethnographic depictions and accounts . Today, the theory
survives in connection with particular hypotheses: for example, if the
Lascaux animals are indeed being driven, then the enigmatic 'unicorn'
figure, sometimes interpreted as men in a skin, may be a shaman driving
the little horses before him .
The shaman is a very important figure, being a person with spiritual
powers who combines the roles of healer, priest, magician and artist , as
well as poet, actor and even psychotherapist! His most important function
is to act as liaison between this world and the spirit world, a task usually
performed by means of trances and hallucinatory experiences - either self-
induced or using hallucinogenic fungi or other similar substances.
Cultures with shamans usually have a zoomorphic view of the world, and
things are seen and experienced in animal form. Hence, from this
perspective, Palaeolithic images are 'spirit animals', not copies ofthe real
thing, while the shaft-scene at Lascaux becomes a shamanistic fight, a
psychic conflict between two shamans (one in animal form) or between a
shaman and a malevolent spirit. 40
In a theory of this kind, based on Siberian ethnography, Glory
suggested that many of the figures in Palaeolithic art were 'ongones',
spirits which took the form of 'zoomorphs', 'anthropomorphs' and
'polymorphs', and which were asked to help in hunting, matters of health,
 158 IMAGES OF THE ICE AGE

Fig. 104 Macro-photo of the head

of a composite engraved figure in
Trois Freres (Ariege), together with
Breuil' s tracing of the full figure.
Probably Magdalenian. Total
length: 30 ern The head is about
3 em across. (JV, collection Begouiin)

and so on. Here, the lines emanating from animals' mouths would be an
evil 'illness spirit' being exorcised, while damaged images will have been
broken in anger when prayers were not answered. 41
A closely related view, based on concepts widely held among modern
hunting peoples, concerns the 'master-of-animals', usually a dead
shaman; this humanoid or composite figure represents a third force
mediating between living shamans and animals, one which constitutes the
life-force of an animal or which can impart life-force to it. The living
shamans derive their own life-force from the animals and then use it in the
service of their clients. 42 According to this scenario, the artists were
shamans maintaining links with the animals (with which they closely
identified) through the master-of-animals; the power derived from the art
came through the act of drawing, not from subsequent viewing. The lines
at the mouths and nostrils of animal depictions represented the entrance or
exit of life-forces.
A different possibility is that these lines represent the nose-bleeds often
experienced by shamans entering a trance - this is certainly a less fanciful
explanation of the marks in front of one Trois Freres composite than the
traditional one of playing a 'musical bow'! It is worth noting that what
appear to be lances or missiles hitting animals' bodies may likewise have an
alternative explanation : egan Apache myth concerning the creation ofthe
horse involves four whirlwinds penetrating the animal's flank, shoulders
aQji hips, enabling it to breathe and move. 43
Obviously, the shamanistic explanations, like the similar ones based on
totem!sm in other cultures, involve ethnographic parallels; but from what
we know about Upper Palaeolithic culture and life (particularly the
Magdalenian), it seems likely that beliefs of this kind existed and thus
played a role in the prodnction of Palaeolithic art. Certainly one of the
major directions of current research , involving hallucinations, trances and
'phosphenes' (see below , p .190), is derived largely from the study of the
shaman phenomenon in southern Africa and its role in the production and
' content of San rock-art. 44
 READING THE MESSAGES 159

Fertility magic and sex

One of the most popular and durable explanations for much Ice Age art has
been that it involves 'fertility magic': ie that the artists depicted animals in
the hope that they would reproduce and flourish to provide food in the
future. As mentioned earlier, such utilitarian theories about Palaeolithic
art were largely inspired by an early study of the Arunta of Australia who,
it was said, drew animals in order to make them multiply.
The proponents of this theory, like those of hunting magic (and they
were often the same people), selected examples which seemed to fit their
ideas, and they often saw what they wanted to see: animals mating, and an
emphasis on human sexuality too. Yet, as already mentioned, relatively
few animals have their gender shown, and any genitalia depicted are
almost always shown discreetly. There are no pictures of rampant
stallions, and the only known animal with a blatant erection is a bison at
Trois Freres which is probably a composite, and thus probably at least half
human (though the phallus is thought to be that of a bison).45
As for copulation , in the whole of Palaeolithic iconography there are
only a couple of possible examples , and they are dubious in the extreme.
T he best known is an alleged coupling of horses in the sculptured frieze at
La Chaire aCalvin (Charente): there may be two horse figures here, partly
superimposed, but the 'stallion' is not in the correct position for
copulation, and is smaller than the 'mare' (Fig.l 05).46 Some scholars
believe that the small horse is actually superimposed on a bison figure. The
other example, two bison at Altamira, is equally odd, as they are claimed to
be a female mounting a (larger) male in order to arouse him sexually!47 The
proponent of this view also sees the Altamira ceiling as the depiction of a
bison herd in the rutting season. 48

Fig. 105 Sculptured frieze of La

Chaire aCalvin (Charente) with
sketch showing the two superimposed
animals. Probably Solutrean. Total
length: c 3m. The superimposed
figures are 1.4 m long. (JV)
160 IMAGES OF THE ICE AGE

Faced with an absence of copulation scenes, proponents of fertility

magic have therefore interpreted many pairs of animal figures as being
engaged in 'pre-copulatory activity' : ie males advancing on females, or
sniffing them. 49 H ere , too , a great deal of wishful thinking is involved: for
example, at Teyjat we defi nitely have a male aurochs behind a female
(Fig.85), and his muzzle is close to her rump- but they are part of a line of
aurochs, and there is another cow behind the bull. Similarly, the two clay
bison in the Tuc d' Audoubert have often been interpreted as a male about
to mount a female: but neither has its sex shown , so that one can only go by
relative size; they are not close together, and the 'male' is actually behind
and 15 em beyond the 'female' (Fig.120); and, of course, the interpretation
ignores the other two bison here, a small clay model and an engraving in
the floor.
The last resort of those seeking evidence of an interest in animal fertility
was to see many figures , especially horses, as 'pregnant' . Yet even
Fig. 106 Two superimposed
humanoids on plaquette from E nlene
(Ariege), with a third at the other end
of the stone. Note also the legs and
belly of a bison . Magdalenian . Total
length: 21 em (Photo]. Clottes,
collection B egouiin)
 READING THE MESSAGES 161

veterinarians cannot tell if a horse is pregnant from its profile alone

(though they can do so from a back-view)/ 0 quite apart from possible
stylistic conventions involved in giving horses big round bellies (like short
legs, tiny heads, etc), inflated stomachs can also be the result of eating
quantities of wet grass; and some male horse figures with phallus shown
look equally pregnant! 51
In a couple of cases, a small figure is superimposed on a large animal's
body- most notably at Lascaux where a foal appears to be 'inside' a mare-
but they are unlikely to be pictures of pregnancy. The Lascaux case has the
foal in the wrong position, and one of its limbs protrudes below the bigger
figure's belly, so they are far more likely to be two superimposed images
or, according to another suggestion, a mare cantering alongside her foal. 52
Besides, as we have seen, there are very few young animals in Palaeolithic
art.

So much for animal fertility, therefore. Surprisingly, more attention has

been given to human sexuality, as if the artists were using fertility magic to
produce their own children. Here again, however, much is in the eye of the
beholder.
We have already seen that quite a few human figures can be sexed,
though relatively few have their genitalia marked: some are clearly female,
fewer are clearly male, and most are 'neutral' - ie asexual. Since some of
the definite males have erections, one might expect that intercourse would
have been drawn from time to time.
There have been one or two claims for depictions of human copulation,
but they are very sketchy, and it requires a great deal of goodwill and
wishful thinking to see them as any such thing. For example, some of the
'fat ladies' of LaMarche may actually be thinner ladies with men on top of
them, and the 'Femme au Renne' of Laugerie-Basse also has faint lines
engraved above her which may represent a person. 53 A large plaquette
from Enlene, with a bison as its principal figure, also has a couple of
superimposed human/humanoid figures which have been interpreted as a
male taking a female from behind 54 - yet, apart from possible whiskers on
the male, the only indication of gender is a difference in size (the smaller
figure's belt and long hair are insufficient evidence), and there is no
explicit trace whatsoever of sexual activity (Fig.106).
The only plausible scene of human copulation in Palaeolithic art is to be
found in the cave of Los Casares where, according to a tracing by Cabre, a
r---_
Fig. 107 Supposed copulation { 'J 0 5 10 em
I I I
scene, of dubious date,from Los )
Casares (Guadalajara). (After
Cabre)
162 IMAGES OF THE ICE AGE

~
''.I "" ""'- 0 2 3cm

Fig. 108 Breuil's version and a

more recent tracing of a plaquette
from Enlene (Ariege) showing the
differences in content. Magdalenian.
(After Begouen et al)

male human has a huge phallus which is superimposed on the stomach of

another human with prominent hips and abdomen, who is probably
female. 55 Even here, however, the scene is not a realistic depiction of
intetcourse, and in any case the date of the figures and the accuracy of the
copy are in some doubt (Fig.l07) .
As with animal figures, the proponents of fertility magic therefore
looked for scenes of 'pre-copulatory' activity: one example, on another
engraved stone from Enlene, has a male human with an erection, and
facing him is a possible second figure with arm outstretched. In Breuil's
original tracing , the second figure was 'completed' into a woman, with
breasts and prominent buttocks, who was thus reaching to grasp the
phallus in her hand; many of these features, however, are simply not
there .56,The 'woman' is extremely sketchy, and may not even be a figure,
 READING THE MESSAGES 163

while the 'arm/hand' could equally well be an animal's tail with a tuft of
hair at the end; and in any case it 'touches' the man's stomach , not his sex.
As the engraving is broken, the man's head is missing, but he may be a
composite figure , since he appears to have a furry tail hanging at the back
(Fig.l08).
Similar subjectivity influenced interpretation of the engraved rib from
Isturitz depicting two humans. They were seen as a male pursuing a
female, the latter identified by her breasts and the former by a 'gesture' or a
glint in the eye! Yet closer inspection reveals that the second figure, broken
off at the chest, clearly had breasts too; moreover, the position of the
breasts indicates that these are not two figures crawling along a horizontal
surface, with one pursuing the other, but two vertical figures , one placed
below the other (Fig .100).57
What about pregnancies, then? As with horses , any fat female figure was
interpreted as being pregnant; but a re-evaluation of'Venus' figurines (see
above, p.l38) has suggested that very few of them (17%) could plausibly
be interpreted in that way. 58 None is definitely shown in the act of
childbirth, although one stone figurine from Kostenki, different from all
the others, appears to be a pregnant woman about to give birth in the
Asiatic position (she is kneeling with legs apart, and her genitals are clearly
marked); and gynaecologists have interpreted the Tursac figurine as a
woman giving birth, with its 'peg' as the emerging child, and also one
Grimaldi figurine and the bizarre Monpazier statuette as being in the act of
Fig. 109 The 'playing card' figure giving birth. 59 Apart from these rare possibilities, 'Venus' figurines do not,
from Laussel (Dordogne) . Probably therefore, seem to glorify maternity or fertility; they appear instead to be
Gravettian . Size: 20 em. (JV) about womanhood . There are no known pictures of children, apart from
the possible baby at Gonnersdorf (see above, p.l41 ), and a few stencils of
infants' hands.
Finally, an enigmatic carving from Laussel which resembles a playing
card, with two half-figures joined at the waist, has often been claimed to be
either a copulation scene or a woman giving birth; but it is so vague that it
could be anything (Fig.109):60 indeed, some scholars see it as a full Venus
figure, while another view argues that it is a person standing waist-deep in
water!
The obsession with sexual interpretations, and particularly with female
genitalia, in the art can be traced back 75 years. In 1911 the abbe Breuil
was consulted about certain deeply engraved motifs which had been found
on some stone blocks in early Upper Palaeolithic sites in Dordogne. Breuil
declared these ovoid and subtriangular figures to be 'pudendum
muliebre', or vulvas. 61 This was a highly subjective interpretation but,
ever since then, most scholars have accepted it without question- even
though a man of Breuil's profession should not perhaps be considered an
expert on this particular motif! - and it was certainly a major factor in
Leroi-Gourhan's concept of the development and complementarity of
male and female signs (see below, p.l69).
T he 'identification' of so many examples of female genitalia led to ideas
about a Palaeolithic obsession with sex; there is tautology in the reasoning
-the figures are assumed to be vulvas, from which an obsession with sex is
inferred, the evidence for which is the vulvas! Independent data are
required to support the hypothesis; but when one searches, vulvas are
remarkably hard to find in Palaeolithic art . If one allows that the only
definite vulvas are those found in context, that is, in full female figures,
then they are very few in number; of the motifs without context, only the
164 IMAGES OF THE ICE AGE

vulva modelled in clay at Bedeilhac (see chapter 5) can be identified with

any confidence - all the rest are mere interpretations.
It is surprising to find that, even among the 'Venus' figurines, often seen
as proof of an intense interest in female sexuality, very few have the pubic
triangle marked, and even fewer have the median cleft. 62 In fact, these
figures accentuate the breasts, buttocks and hips; they may represent
fecundity but they do not draw attention to the vulva, unlike the headless
bas-reliefs of Angles-sur-1' Anglin which do seem to have this part of the
body as their focu s.
Where vulvas are depicted, nearly all are triangles, as one finds in female
figures from many other periods and cultures. Yet the early Upper
Palaeolithic motifs have a wide variety of shapes, and scholars have needed
a number of excuses in their desperate bid to fit all these shapes to the
chosen interpretation: eg there are 'incomplete vulvas', 'squared vulvas',
'broken, double vulvas', 'circular vulvas', 'relief vulvas', and even
'trousers vulvas' ! Others have claimed unfinished vulvas, atypical vulvas,
and even interpreted a single straight engraved line as an isolated vulvar
cleft! (Fig.54) 63
But in fact, as we know from rock-art in Australia and elsewhere, very
simple graphic designs can have a very wide range of meanings; and in any
case, if the Palaeolithic artists took pains to differentiate these shapes, it
makes little sense for us to lump them all together into a single,
subjectively chosen interpretation. Hence, more objective scholars have
divided them into descriptive rather than interpretative categories: horned
ovals, pear-shaped ovals, arched ovals, etc. 64
Wishful thinking , therefore, permeates the search for human sexuality
in Palaeolithic art. For example, the clay 'sausages' in the Tuc
d' Audoubert were automatically seen as phalluses by believers in fertility
magic , while others saw them as possible bison horns, but, as mentioned
earlier (pp. 94-5), they almost certainly represent the sculptor trying out
the clay. 65 Similarly, one of the portable statuettes of Angles-sur-l'Anglin
is a kneeling female, but Breuil originally thought it was a phallus. 66
It is possible, of course, that there is purposeful ambiguity here, a visual
pun; the same may apply to 'vulvas' which look more like horse-hoof or
bird-foot prints . We have seen that all boundaries and categories are
blurred in Palaeolithic art , and it would not be surprising if the artists
noticed and played with such ambiguities. 67 This kind of punning may also
explain why they drew a red human outline around each of two little
stalagmites, a few metres apart, sticking out of the wall at Le Portel, thus
turning them into ithyphallic men (Fig.52).
A different example of ambiguity probably owes more to the
preoccupations of the beholder than to the intentions of the artist: a series
of little ivory carvings from Dolni Vestonice have always been seen by
~st male scholars as abstract female figurines with long necks and huge
buttocks or breasts; the same applies to an ivory rod from the site, which
has a pair of'breasts' protruding from it, some way up. However, feminist
scholars suggest that these carvings should be inverted, and clearly
represent male genitalia!68
Just as the 'hunting magic' theory has developed into a view of the art as
reflecting a male preoccupation with hunting, so 'fertility magic' has given
way to a 'cheesecake' view of human depictions: ie this is early erotica
made by men for their own pleasure. This is why the female figures have
exaggerated breasts and buttocks, with no emphasis on the limbs or the
READING THE MESSAGES 165

face. They represent women's contours and were intended to arouse, to be

touched, carried and fondled. 69 Similarly, the act of engraving the alleged
'vulva' carvings of the early phases reflected the 'digital activity of
lovemaking around the labia'! 70
It is even suggested that the statuettes' heads are bowed in subservience,
submission or shyness, while the enlarged rumps on the schematised
females of Gonnersdorf and elsewhere are simply sexual lures. The rarity
of clothing and personal adornments in the depictions, as well as of birds
and other small resources, proves that it is an art which is limited to male
preoccupations, and not concerned with a woman's side of life. 71 In this it
resembles modern Eskimo art, comprising hunted animals and a few nude
females .
Finally, some Palaeolithic depictions of women have been compared
with pin-ups in Playboy, and it has been claimed that 'female figures often
[sic] appear in sexually inviting attitudes, which may be quite the same as
those in the most brazen pornographic magazines. There are also
anatomically detailed pictures of the vulva, showing the female sex organ
sometimes frontally, sometimes inverted and from the back, open to
penetration'! 72
This quotation reveals a lurid imagination. Apart from the bas-relief
reclining females of La Magdeleine, there are no such 'poses' in
Palaeolithic art. A few depictions do bear a superficial resemblance to the
pin-ups selected, but the range of positions in which the female body can
be depicted is fairly limited, as shown by the fact that all the comparative
material was found in a single issue of the magazine! Another basic point is
that we do not know the sex of the artists. The carvers of the 'vulvas' and
figurines could just as easily have been female , and one can extend this
argument to the whole of Palaeolithic art, invoking initiation ceremonies to
explain menstruation, with lunar notation (see below, p.l82) as
supporting evidence. 73

The theory that the art is about the male preoccupations of hunting,
fighting and girls comes from twentieth-century male scholars. Feminist
scholars have a different point of view about the supposedly sexual images
(for the same reason, it would be interesting to see the reaction of a
vegetarian culture to the animal figures). The macho view of Palaeolithic
art is both simplistic and anachronistic. Some of the female figurines may
have aroused men, but there is no reason to suppose that this was their
intended function: after all, many of them were carefully hidden in pits.
Moreover, images of female genitalia are far rarer than has been claimed in
the past, and those which were depicted may well have been intended
simply to indicate gender, rather than be erotic. The greater part of
Palaeolithic art is clearly not about sex, at least in an explicit sense. The
next major theoretical advance, however, introduced the notion of a
symbolic sexual element.

Laming-Emperaire and Leroi-Gourhan: the female and the

male
Despite his stubborn opposition to parietal art , Emile Cartailhac was a
very perceptive man; we have already mentioned (see above, p. lSO) his
early impression that 'art for art's sake' was an insufficient explanation. A
further example occurred in 1902, when he stated that the cave of Pair-
166 IMAGES OF THE ICE AGE

non-Pair appeared to contain a random accumulation of pictures , but that

of La Mouthe had been decorated systematically.74
Thanks m. the dominance of Breuil, who treated cave-art as a collection
of individuat~pictures, this pioneering insight was forgotten, and the
concept did not re-emerge for over half a century, when two scholars,
influenced by the unfinished work of Max Raphael, arrived more or less
independently at the same conclusion. Annette Laming-Emperaire and
Andre Leroi-Gourhan , in a series of publications between 1957 and 1965,
revolutionised the study of Palaeolithic art , which thus took its first leap
forward since 1902.
One of the basic features of their approach was to ask new questions,
while avoiding the intuition and the subjective use of ethnographic
parallels which had characterised previous work and led to an impasse.
Interpretation was to be based on the total corpus of figures in a cave, not
on one or two selected images. They treated parietal art as a carefully laid-
out composition within each cave; the animals were not portraits but
symbols; they did not reflect explicit practices but rather a complex
metaphysical system or 'mythogram' . Previous concentration on the
Fig. 110 The 'central composition'
aesthetic and magical aspects was inadequate, as the ideas behind
of black figures at S antimamiiie
Palaeolithic art were far more complex and sophisticated . In short, they (Vizcaya), showing a horse
moved from analogy to observation, from naive and particularist surrounded by a number of bison,
explanations to the structures reflected in the subject -matter and its including two drawn vertically, back
associations. Although there was no hope of grasping the art's meaning, to back. Probably Magdalenian .
one might gain some insight into the richness of thought behind this Panel size: 2.5 m. The horse is 45 em
symbol system. 75 long. (JV)
 READING THE MESSAGES 167

Although it was Laming-Emperaire who, following Max Raphael, first

treated parietal figures as a composition in her study of the sculptured
sites, as well as Lascaux and other caves, it was Leroi-Gourhan who
confirmed the notion, extended it to all major caves and continued to
develop the idea, while Laming-Emperaire eventually dropped it in favour
of an even more complex explanation (see below).
An investigation of how many figures of each species existed in each
cave, together with their associations and their location on the walls, led to
a number of insights: Leroi-Gourhan eventually divided animal figures
into four groups/ 6 based on the frequency of their depiction in parietal art
as a whole. Group A was the horse, which constitutes about 30% of all
parietal animals; B was the bison and aurochs, which also account for
about 30%; C was animals such as deer, ibex and mammoth, representing
another 30%; D, the finallO%, comprised the rarer animals such as bears ,
felines and rhinos.
He also divided caves into entrance zones, central zones, and side
chambers and dark ends. It appeared that about 90% of A and B were
concentrated on the main panels in the central areas, the majority of C
figures were near the entrance and on the peripheries of the central
compositions, while D animals clustered in the more remote zones. He
developed the concept of an ideal or 'standard' layout to which the
topography of each cave was adapted as far as possible (Fig.lll): these
were organised sanctuaries, with repeated compositions separated by
zones of transition marked with the appropriate animals or signs. 77
Different signs are found at the start and end of the sanctuary (mostly dots
and lines), while more elaborate signs occur before and within the big
animal compositions .

Fig.lll Leroi-Gourhan's

~
'blueprint' for the ideal cave layout.
(After Leroi-Gourhan)
40
~
65
~
50
121

0 ~ L~ t::;;:J I~
~22 ~ 91 50
Q
92
t::;;::J 47 71

~
29
~ 45
1/\ ~ 30
~ 1
35 86 84 58 37 _I 51
ENTREE POURTOUR CENTRE CENTRE POURTOUR PASSAGE ET FOND
Q
36

DIVERTICULE

However, as Leroi-Gourhan was well aware, many caves are extremely

complex topographically and may have had different entrances during the
Palaeolithic which are now blocked: one cannot simply assume that the
present entry has always been the only one. Our doors may have been their
windows! From the start, he sometimes had to adopt bizarre itineraries
within caves in order for his scheme to fit- even going to the wall furthest
from the entrance to find the 'first series' of big signs, and then coming
back again! 78 He knew that there is no 'average' cave but still believed that,
though topography may have dictated layout to some extent, it was still the
'mythogram' that conditioned the choice of topographic adaptation/9 ie he
168 IMAGES OF THE ICE AGE

imagined that the artist who had been 'commissioned 'to decorate the cave
checked its topography and reliefs, and noted any features which appeared
useful in accordance with the predetermined blueprint - hence the
Altamira ceiling-bosses could become only curled-up bison, not felines or
something else. 80
His divisions into central and peripheral areas are somewhat tenuous
(those in Gabillou, for example, are quite arbitrary- where do complex
panels start and end?) and decorated caves are far more heterogeneous
than the scheme suggests. It is not just a question of shallow caves and
rock-shelters versus simple corridors and complex systems of galleries,
but a problem of content: as we have seen, some caves have a single figure
or sign, others have animals and no signs or vice versa; 81 and how is one to
fit the newly discovered open-air engravings into this concept? Leroi-
Gourhan himself admitted that he omitted caves that were exceptions,
incomprehensible or lacking the appropriate animals; and he conceded
that there were different regional 'formulas', notably in Cantabria. 82 Far
from presenting his theories as impeccable, he constantly pointed out the
gaps and faults in them.
Even in caves with abundant animals and signs, the latter do not always
begin or end the decoration, and the former often do not conform to the
scheme: for example, bison are the last animals in the 'feline gallery' at the
end of Lascaux, and among the last in Les Combarelles, while the clay
bison of the Tuc d'Audoubert are at the far end of the cave. Leroi-Gourhan
sometimes assumed changes in cave topography in order to explain
'anomalies' of this kind (eg wrongly claiming that the Tuc d' Audoubert
was once joined to Trois Freres, so that the clay bison would not be at the
end) or divided the caves up on the basis of the themes represented rather
than topography, thus turning his scheme into a circular argument. 83 Even
so, his own figures revealed that 60% of bears and 66% of rhinos are in the
'central' areas; indeed, the rhino is rarer in cave-ends than are ibex and
deer! Ibex can be found in virtually any part of the cave. It is true,
however, that the D group does not usually occur in entrance zones, and
that felines tend to be in remote zones (a fact already noticed by Breuil).
A further problem with Leroi-Gourhan's scheme is his use of
quantitative data and percentages (which often contain small errors 84), and
especially the fact that the scheme works on a presence/absence basis, not
on abundance: hence a single horse figure is the equivalent of a mass of
bison, and vice versa. Few scholars have been able to accept that
abundance was irrelevant, quite apart from other variations such as
colour, size, orientation, technique and completeness. Leroi-Gourhan
pointed out that, in Christian churches, images of Christ are often far
outnumbered by the Madonna and saints, but admitted that the 17 hinds
of Covalanas probably had a more important role than the single (and
<W,bious) aurochs head in the cave. 85 A number of caves are similarly
dominated by a single species.
Yet in his final presentation of the scheme he allows NB/C!D animals to
be present in any proportions in a particular cave. Where the C group is
concerned, only one of the species need be present in the formula . Caves
with no A animals (Tete-du-Lion) or with no A orB (Cougnac) are seen as
exceptions that prove the rule! 86
In fact, there are so many exceptions, contradictions and variations that
one can seriously doubt the validity of the 'ideal layout'. Recent detailed
studies, both of individual caves and of regional groups, constantly stress
READING THE MESSAGES 169

that each site is unique and has its own 'symbolic construction' adapted to
its topography and concretions, and with multiple thematic linkages. 87
Few people would now uphold the universality of the layout, either in
space or time . Leroi-Gourhan lumped all Palaeolithic parietal art into his
scheme, and believed that its structure remained homogeneous and
continuous for 20,000 years- even in caves decorated in several phases,
the later artists respected the first sanctuary's layout. 88 There is certainly a
degree ofthematic uniformity over this timespan: caves are decorated with
the same fairly restricted range of animals in profile, and seem to represent
variations on a theme: as Leroi-Gourhan pointed out, nobody drew a frieze
of lions and storks surrounded by hyenas and eagles! 89 He therefore
thought that the animals depicted corresponded to a certain mythology.

The other foundation stone of the approach was the discovery of repeated
'associations' in the art, and the claim that there was a basic dualism. This,
too, was a concept first developed by Laming-Emperaire and
subsequently extended by Leroi-Gourhan. Having found bison
'associated' with both horses and women, Laming assumed that the horse
might in some way be equivalent to female, and thus the bovids with the
male. Leroi-Gourhan, on the other hand, saw the bison as equivalent to
females; his view was based almost entirely on a couple of crude figures in
Pech-Merle which he interpreted as human females derived from bison
with raised tails (Fig.ll2). 90 However, the animal figures are vague (they
could even be horses!) and any similarity is superficial, being limited to the
curve of the profile: in short, the link between females and bison is so
tenuous as to be non-existent. 91
Laming also found a frequent association of horses with either bison or
aurochs, a fact which Capitan and Bouyssonie had already noted in the
portable art of Limeuil in 1924, though in caves it is of little surprise in
view of their great abundance. In the new approach, the numerically
dominant horses and bovids, concentrated in the central panels, were
thought to represent a basic duality. One possiblity was that this was
sexual, though it was admitted that other explanations might be equally
plausible (eg sun/moon, heaven/earth, good/evil, or something more
complex). It was pointed out that horses live under the control of a stallion,
and cattle under that of cows. 92
Having tentatively claimed a sexual duality, it was natural for Leroi-
Gourhan to extend this to the abstract signs; his somewhat Freudian
viewpoint was that they could be divided into male (phallic) and female
(vulvar) groups, which he later dubbed 'thin' and 'full'- a third group was
later added to include the dots. This view was inspired by two bear-heads
in portable art, one of which (La Madeleine) seemed to be associated with a
phallus, and the other (Massat) with a 'penniform' sign. 93 Hence, where
proponents of hunting magic saw missiles and wounds, Leroi-Gourhan
saw phallic and vulvar symbols.
In the early years of his research, he felt certain that the two series of
signs were of a sexual nature; but it was never explained why such a wide
range of shapes should have only two basic meanings, even allowing for
artistic licence. Besides, there were inconsistencies such as the claviform
which is thin but, as we have seen (p.l42), was thought to be derived from
a female silhouette- in the same way as the Coca Cola bottle could be taken
either as a phallic symbol or as a stylised female form, according to one's
preference!
170 IMAGES OF THE ICE AGE

Many scholars rejected the sexual theory: Breuil, for example, referred Fig. 112 The 'bison-women' of
to it as a 'tentative deplorable' and a 'perspective sexomaniaque' / 4 which Pech Merle (Lot). Probably
was somewhat ironic coming from the man who had unleashed the Solutrean. Each about 10 em long.
obsession with vulvas in 1911! However, most critics were confusing (JV)
sexuality with genitalia; even if the division existed, it probably had
Fig. 113 The art in its setting: dots
nothing to do with eroticism , but simply reflected some sort of structure in and a hand stencil in P ech M erie
society. (Lot). Probably Solutrean. (JV)
From these horse/bison and thin/full oppositions and associations,
Leroi-Gourhan developed a view of the caves in which other animals were
allotted sexual roles (eg ibex were male, mammoths female). Even clearly
male bison were considered female symbols- in other words, animals were
sexed by species and location, not by gender (except for stags and hinds,
for some reason)! By 1982, however, he had dropped the emphasis on
sexual symbolism, although he retained the entirely valid point that the
cave itself probably constituted a female (womb) symbol, as it has in many
subsequent cultures. 95
Just as his scheme had fixed places for different species, with 'female'
animals and signs clustering in central areas, and male elements both
among them (horse) and around them, it also claimed that human males
were mostly at the back of the caves, while females were in the well-lit
entrance areas (as at La Magdeleine, etc) . Occasionally he allowed the
theory to dictate the evidence, to the extent that humanoids were identified
and gender attributed to unsexed humans/humanoids, on the basis of their
location in the caves! For example, the' Femme al'anorak' of Gabillou was
claimed to be a woman because of its position, but he ignored the fact that
the cave's clearest female is right at the very end.%
172 IMAGES OF THE ICE AGE

His claim that human males are always at cave-ends or on the periphery
of panels was untenable because he was including many unsexed heads,
and he himself admitted that males can be found all over the place- in
entrance zones, cave-ends or around central panels. 97 In fact, there is a
clear patterning of human/humanoid representations in caves, but it is
somewhat different from that proposed by Leroi-Gourhan. Applying strict
sexing criteria (see above, p.l36), and dividing caves into daylight,
obscurity and total darkness, it has been found that the percentages of
definite males, definite females and neutrals change dramatically: in
daylight zones, 56.2% of humans are female, 6.2% male and 37 .5%
neutral; in obscurity these figures have changed to 31.4%, 8.1% and
60.4%, while in total darkness only 3.6% of the humans are female, 16.3%
are male and a massive 80% are neutral. 98
Leroi-Gourhan felt that his formula comprised a variety of binary
oppositions, and that a symbol such as the horse or bison could certainly
have taken on or shed many associations in the course of the Upper
Palaeolithic- he never saw the art as a rigid, fossilised system. Some of the
associations which he and Laming discovered are certainly real, although
the terms 'juxtaposition' or 'coexistence' might be preferable since they do
not assume that the phenomenon is meaningful: for example, many hand
stencils are found in close proximity to painted dots (eg Pech Merle
(Fig.l13), Castillo), and often near horses too; over half of parietal signs
are with animal figures; in parietal art as a whole, bison and aurochs are
found with horses far more often than with any other species. In portable
art, however, the same rules do not apply: at LaMarche, horses are found
with practically every species depicted, while humans are found with
almost every species except felines .99
The overwhelming parietal coexistence of horses and bison suggests that
their duality is the basis of .the whole figurative system, as Leroi-Gourhan
believed; alternatively, these animals may each have a separate
importance, making their frequentcoexistence a matter of chance (the
limited range of species depicted also increases the chances of their being
found together). We simply do not know. However, a recent re-evaluation
of parietal figures and their associations, using factor analysis, not only
confirms the fundamental role and opposition of horses and bison, but also
suggests that the bison is dominant, since it has a greater 'effect' on a panel
centred on a horse than the presence of a horse does on a bison panel. 100
It was also found that some 'associations' were rare or non-existent -
such as bison and aurochs, bison and stag, or aurochs and reindeer;
however, some species are rarely associated because of regional
differences: as mentioned earlier (p.l33) the majority of hinds occur in the
early Magdalenian of Cantabria, while mammoths are predominantly
found in the mid-Magdalenian of Dordogne - this considerably lowers
their chances of being found together. 101
1">ther scholars have run statistical tests on Leroi-Gourhan's figures but
found little evidence to support the duality of signs or the male/female
dichotomy in animal species, especially the homogeneity of male signs and
animals, though the 'female animal' and 'dangerous animal' groups
remained plausible . 102 Others claimed that Leroi-Gourhan's use of
percentages of associations was 'idiosyncratic and fundamentally
unsound' because he ignored group size and relied simply on presence/
absence. A multivariate analysis of the animal figures in a number of sites
revealed two small groups of neighbouring caves containing depictions of
 READING THE MESSAGES 173

several species in similar proportions, but otherwise there was no overall

pattern. 103

The importance of Leroi-Gourhan's work is enormous, not so much

because of its results but through its impact on the field and on current and
future research. It constituted a leap forward and, like all first attempts, it
oversimplified, went too far too soon, and made many mistakes. As with
any new and powerful theory, there was an irresistible temptation to bend
ambiguous facts to fit, while ignoring others. There was an excessive
tendency to assume that Palaeolithic artists thought like twentieth-century
French structuralists. Leroi-Gourhan felt that he had established the
'blueprint' for cave-decoration, the framework on which a variety of
symbols and practices had been hung through the period. His results are
too erroneous to be fully accepted, but also too revelatory to be dismissed:
some of his results, as we have seen, cannot be faulted. He found order and
repeated 'associations', but not a universally applicable formula.
Inevitably, he modified his views over the years; Laming-Emperaire,
however, who initiated this approach, eventually ab:lndoned it in favour of
a theory inspired by the work of Raphael and Levi-Strauss, which virtually
reverted to the totemism so beloved of early ethnographers, and which
Reinach had presented (together with hunting magic) as an explanation of
Palaeolithic art. 104 Breuil and Begouen had opposed this view, since it
meant that animals were mythical ancestors of human groups, and there
would therefore have been taboos against killing or eating them- whereas
they saw the depictions as killed animals.
Laming criticised the 'topographic determinism' which had arisen
around the sexual dualism theory and which, as shown above, meant
figures and signs were 'sexed' by their location, in a perfectly circular
argument. 105 She maintained that the horse and bison were probably
fundamental, but did not necessarily constitute a sexual opposition.
Instead she wondered if they might be an expression of exogamy practices:
ie a decorated cave was a general model of a group's social organisation,
with animals of various sizes and ages representing either different
generations or the mythic ancestors of certain clans. For example, a big
bull and a horse at Lascaux would be clan ancestors. The male and female
aurochs at Lascaux should not be seen as a bull and cow, but as two
members of the bovid clan, such as a brother and sister. Two animals of
different species and sex would be a matrimonial couple, while two of
different species and the same sex would be an alliance. 106
She believed that the main panels were origin myths, the principal
animals were ancestors, and the secondary species were allied groups. The
artists were not displaying a system or formula, but telling a story within a
system. This approach was more flexible than that of Leroi-Gourhan
because it did not expect everything to remain stable through time and
space, but rather to be different in each cave. Unfortunately, Laming-
Emperaire died before she could develop the idea further, although it is
difficult to see how to do so without recourse to ethnographic analogies.
Whatever the faults and inadequacies of the work of these two scholars,
Fig. 114 Photomontage of the they did succeed in changing irrevocably the way in which we think about
horse panel at Ekain (Guipuzcoa), Palaeolithic art. The images could no longer be seen as simple
c4m wide. Note the red bison at top representations with an obvious and direct meaning; instead it was
right, and the 'M' mark on the horse realized that they were full of conceptual ideas - they were the means by
hides. Probably Magdalenian. (JV) which the artists expressed some more metaphysical concept. Their work
176 IMAGES OF THE ICE AGE

constitutes a very important transitional phase following the traditional

approach of compilation and intuitive and ethnographic explanation. All
current and future research will draw on their ideas and results, and
modify or discard them. Breuil, Laming-Emperaire and Leroi-Gourhan
would all have agreed that, for work to progress, we must not sit at their
feet but stand on their shoulders.
It now remains, therefore, to look at a few of the varied directions in
which current research is taking us.

Why is it where it is?

One of the most interesting new pieces of research has focused on the
distribution of figures, and the shape of the wall beneath each one, in four
neighbouring Cantabrian caves (Castillo, La Pasiega, Las Chimeneas, Las
Monedas). It was found that in each of them the figures had been
distributed according to the spatial characteristics of the cave, but there
emerged clear differences between the content of 'concave' and 'convex'
rock-surfaces: no less than 88 .8% of horses were on concave surfaces,
together with 93.1 % of hinds, 87 .7% of caprines, 100% of stags and 100%
of hand stencils; on the other hand, 78.7% of bison and 82 .8% of cattle
occurred on convex surfaces. 107
Moreover, the horse and bison were clearly the main representatives of
the 'concave' and 'convex' groups respectively, and were usually bigger,
more detailed and more complete than the rest of the animals. These
'hollow horses' and 'bulging bovines' confirm the existence of a horse/
bison dualism and the importance of these species in the art .
Of course there are exceptions to the concave/convex rule, even in these
caves - almost every statement made about Palaeolithic art can be
countered with exceptions - but a preliminary look at other caves suggests
that there may be a degree of consistency: for example, in Covalanas, the
only 'convex' figure is the bovid, while all the hinds and the horse are
'concave'; at Altamira , all the bison on the ceiling are 'convex' whereas the
deer and horse are concave; at Font de Gaume the bison and reindeer seem
to be on convex surfaces; the Tuc d'Audoubert clay bison are on convex,
the Cap Blanc sculptured horses and the Tito Bustillo painted horses on
concave surfaces; while almost all the hand stencils ofGargas, like those of
Castillo, are on concave surfaces. 108
Much more work will have to be done before we can assess how
widespread is this careful selection of surface-shape. Doubtless many
figures and caves will not conform, because no scheme can ever totally fit
an art-form spanning so many millennia. Nevertheless, this extension of
the approach ofLaming-Emperaire and Leroi-Gourhan may well prove to
be a breakthrough .
.,A different observation about the position of art in the caves concerns its
visibility and accessibility. It is not always easy to judge whether the
difficulties we experience in reaching certain galleries existed in
Palaeolithic times: for example, Montespan's art can be reached only by
wading upstream, with water up to one's waist and, at one point, having to
duck right under when the ceiling comes down to water-level; yet on very
rare occasions, as in 1986, the cave is dry. Similarly, graffiti in the Tuc
d' Audoubert (which now has to be entered by boat) show that it was visited
in 1685 and 1702 by clerics in periods of great dryness. 109 One therefore
cannot be sure whether the artists got their feet wet or not.
READING THE MESSAGES 177

Visibility and access can also be affected by stalagmites and stalactites,

and in some caves such as Cougnac these appear to have been broken by
Palaeolithic people, thus making the figures more visible from a distance.
One can divide cave-art into the clearly visible, the obscure, and the
hidden- or, in other terms, the public and the private. 110 The Hall of the
Bulls at Lascaux or the spotted-horse panel at Pech-Merle are good
examples of visible art in large chambers, whereas there are many
examples of art tucked away, either in inaccessible galleries (such as that of
Fronsac, 35 em wide) or in nooks and crannies of large chambers (eg the
signs hidden in the folds of the 'lithophone' at Nerja).
It has been noticed that the parietal art is frequently associated with
'bouches d'ombre': ie chasms or entrances to lower galleries, many of
which have water in them (eg at Tito Bustillo), and which may have been
linked with myths concerning the underworld, the cult of the earth
mother, and 'chthonic deities': 111 for example, at Ekain the farthest frieze
of horses is located above two holes which lead to the dark and mysterious
depths of the cave; in Travers de Janoye (Tarn) the parietal art is clearly
arranged around a large fissure. 11 2
In all regions and at all times, caves have been associated with placenta,
with ideas relating to maternity and birth, and with the entrance to the
underworld. Boundaries are areas of confusion to human beings and , like
the orifices of the body, are sacred places full of taboo; caves certainly
constitute a boundary between the outside world and the underworld, and
thus have a special significance in many mythologies . Water plays a major
role in virtually every known religion, and running water in particular is
frequently looked upon as potent, dangerous and linked with spirits. It is
therefore highly probable that where running water visibly crosses the
boundary of the underworld, in either direction , it may take on special
significance. This effect would be heightened where water enters the earth
in an impressive setting (Mas d'Azil, Labastide), or where it emerges as a
hot spring.
It has been suggested that these factors are likely to have played a role in
whatever beliefs and rituals lie behind Palaeolithic parietal art - what
might therefore be called the 'rites of springs': is it a coincidence , for
example, that large clay models have been found in only two caves (Tuc
d' Audoubert, Montespan) through which rivers flow, and that both
contained a headless snake skeleton? A large number of decorated caves
are located very close to springs, and there is a marked correspondence
between certain parietal sites and thermal/mineral springs: 113 for example,
the four caves in the Monte Castillo overlook the village of Puente Viesgo,
through which a geological fault runs, responsible for a number of hot
springs which have been found suitable for medicinal purposes; it would
be unsurprising if in the last Ice Age very special significance was accorded
to springs which did not freeze and/or which were thought to have
medicinal powers .
Recent work, therefore, has focused on the distribution of art within
caves, and on that of particular decorated caves within the regions where
they exist. Little attention has been paid to the problem of why the
distribution of parietal art is so restricted in Europe, unlike that of portable
art and despite the existence of perfectly good caves in Germany, for
example.
One theory is that southern France and northern Spain constituted
'refuges' into which people migrated in waves from other parts of Europe
178 IMAGES OF THE ICE AGE

during periods of climatic deterioration from about 25 ,000 years ago

onward: the 'cave-art regions' were more temperate, and had a greater
abundance and diversity of fauna available, and people were able to live
there by exploiting the plentiful reindeer and Atlantic salmon. 114 Salmon-
fishing would have led to increased sedentism, and thus a greater need for
elaborate procedures for resolving conflicts and marking territory: hence
the cave-art sites were established as permanent ceremonial locations;
while the portable art was perhaps linked to the more mobile life and
seasonal aggregations associated with reindeer exploitation. In short, the
art is seen as a manifestation of different social responses to processes of
climatic and economic change and population movement.
Unfortunately, the theory does not stand up to scrutiny, simply because
parietal art , as we have seen, is not limited to Franco-Cantabria, merely
concentrated there; in any case, reindeer were not the only staple resource
in the Upper Palaeolithic -many sites, depending on their location and
surrounding topography, had an economy based on horse , bison, red
deer , ibex, or a mixture of several species. In addition, fish began to be
exploited on a significant and systematic scale only at the very end of the
period, long after the main phases of art production; 115 and, of course,
some pockets of parietal art such as the Rhone valley were beyond the
range of the Atlantic salmon .
Consequently fish and reindeer exploitation can have had little or no
bearing on the production of Palaeolithic art. It is still possible that France,
Spain and the Mediterranean coast represented a refuge at times .
However, we are unlikely ever to discover precisely why this art flourished
where and when it did- it may constitute a kind of long-term Palaeolithic
equivalent of Classical Greece or Renaissance Italy.

How many artists were there?

But was Palaeolithic art produced in a steady stream over the millennia, or
does it represent a large but sporadic output by a relatively small number
of gifted artists during this long period? Some scholars still hold the former
view, but most now prefer the latter. 116
As mentioned earlier (p . IOS), experiments suggest that many parietal
and portable figures could each have been executed in a very short time by
an artist with a modicum of talent and experience: all the black-outline
animals ofPech-Merle could have been done in less than three hours, and
probably represent a single artistic episode: indeed, this cave is thought to
have had no more than three or four such episodes- for example, analysis
shows that its six black hand stencils were done with the same pigment and
used the same hand . 117 T ogether with the archaeological and pollen
evidence, these facts suggest that, far from being a great tribal sanctuary,
PecJl-Merle was visited and used only very rarely.
Similarities between Palaeolithic images - primarily in portable art -
have often led, in the past, to theories about 'schools' or workshops:
Limeuil , for example, was seen as a centre of artistic and magic education .
Images on stones and bones were thought to be artists' notebooks or
sketches, and the 'mediocre' specimens among the masterpieces were the
work of pupils. While such notions are somewhat fanciful and owe much
to nineteenth-century views of art , some modern scholars still attempt to
estimate and compare the percentages of 'masterpieces' and 'mediocre
READING THE MESSAGES 179

pictures' in different sites, 118 an exercise which is entirely subjective and

takes no account of the artist's intentions.
Some clusters of images are stylistically and technically so similar that it
is virtually certain that one artist or group is responsible. Hence , at La
Marche, a particular style, technique and set of conventions (and errors!)
have been recognised by those most familiar with its engravings . 119 Few
would doubt that the Mas d' Azil and Bedeilhac fawn/bird spear-throwers
(p. 82) were by the same artist (or, at least, two artists, one of whom had
studied the other's work); the same could be said of the portable and
engraved hind heads of Altamira and Castillo (p.S8), some of the human
profiles of LaMarche and neighbouring Angles-sur-l'Anglin (p.S8), the
Labastide isard heads (p.82) and certain portable bison-head engravings
from Isturitz. 120
On the other hand, other collections of portable art seem to be by a
number of different artists: for example, the Castillo engraved shoulder-
blades, or the slabs at Gonnersdorf where analysis of details suggests that
at least a dozen different artists were responsible for the mammoths. 121
The search for individual artists is by no means a new phenomenon in
Palaeolithic art: de Sautuola himself saw the Altamira ceiling as a unified
work, and in the Cartailhac/Breuil monograph on the cave , there was some
discussion of 'authors', and speculation that not all the cave's pictures were
by the same hand . Until the end of his life Breuil insisted that one artist of
genius could have done all the polychromes of Altamira. 122 Leroi-
Gourhan, on the other hand, refused to accept that one could attribute
even two adjacent images to a single artist. 123 He was obviously correct, in
so far as it cannot be proved that more than one picture was done by a
particular person; but there is a place here for intuitive reasoning: for
example, Pales was certain that at least three of the ibex in Niaux's Salon
Noir were by the same artist. 124
One scholar, Juan-Maria Apelhiniz, has for some years been trying to
establish firm criteria for recognising the work of individual artists, to
transform what have previously been intuitions into a reasoned method.
He looks for an original way of drawing, a particular 'tour de main' , the
repetition of idiosyncrasies, peculiarities and details of technique and
execution. There are variations, of course, since artists do not repeat
themselves exactly, and may well have changed style slightly through the
years: the same artist could certainly produce two very different figures ten
years apart, while, as we have seen, a change of tool can also affect the
hand. Only complete figures can be used .
It is advisable to work directly from the original images, rather than
from tracings and copies by other people which, as we have seen (chapter
3), can exaggerate the similarities or differences between figures .
Occasionally Apellaniz has attempted to work from other people's
tracings , 125 but more often uses direct observation, sometimes backed up
with photographs and copies, as in the case of the Altamira ceiling, where
he confirmed earlier scholars' intuitions that the ceiling was largely the
work of one 'master' . 126 It is relatively easy to compare the standing and
curled-up bison figures, since, as mentioned earlier (p.ll9), they share
many characteristics; however , it is much harder to extend the comparison
to animals of different species such as the horse and hind on the ceiling.
Studies of this type can be particularly easy on portable objects, where
similar figures often form friezes and were almost certainly the work of one
180 IMAGES OF THE ICE AGE

artist (eg on the Torre bird bone, Fig. 42, all the horns have their ends
drawn in the same way, probably by the same hand); the same applies to
objects such as bone discs with decoration on both sides: on that from the
Mas d'Azil with a calf on one side and a grown bovid on the other, or that
from Laugerie-Basse with a chamois on each side, the two figures are
identical in their use of space, their proportions, composition, silhouette
and technique. 127
In his work on parietal art, Apellaniz has produced mixed results . At
Altxerri and Ekain, he has 'identified' different artists or groups working
in different parts of the caves. 128 At Las Chimeneas he believes that all the
black stags are probably by the same hand, and at Lascaux the same
applies to the four huge black bulls, whereas the cows seem to have a
number of authors. 129 One of the clearest sets of similar figures is found in a
group of caves: in Covalanas and , 150m away, La Haza; in Arenaza, 45 km
to the east, and La Pasiega , 60 km to the west. The fidelity to the 'general
canon' of composition in hind and horse figures, the dotted outlines, the
similarity in details such as eyes and extremities of limbs are such that he
claims this group of caves can be seen as a 'school': ie the 'master' or
principal artist of Covalanas had disciples or companions who worked in
the other caves. 130 However, he does not believe an artist worked outside a
particular cave: each cave may have formed part of a 'school' ofthis type.
How valid is this approach? The criteria presented as objective and
reasoned are still rather subjective and intuitive; he is assessing the degree
of similarity between figures which may be by one hand, but which may
also simply display the accepted canons and conventions of a particular
period and culture. 131
In an effort to counteract these problems, Apellaniz is developing
statistical techniques, using a series of variables and measurements of
various parts of the animal outlines (like those made by other scholars on
bison figures, p.l22), and subjecting them to factor analysis in order to
assess their degree of similarity: so far this method has confirmed that the
Chimeneas stags, for example, are extremely similar in the treatment ofthe
neck, back and rump. 132 It will never be possible to prove that the same
hand was responsible, but this approach has certainly established a very
high degree of probability in some cases.
We can only speculate about how Palaeolithic art's 'rules' and 'canons'
were passed through space and time so faithfully. It is likely that most
decorated caves were visited only rarely in the Upper Palaeolithic, and in
any case many of their figures are hidden. Portable art, however, was
probably circulating over great distances, and people (including artists,
presumably) were often moving seasonally, and coming together annually
or periodically in aggregation sites when much could be taught and
learnt. 133 Finally, there must have been a great deal of art in perishable
materials aPr!, as we have seen, there were also 'permanent' images in the
open-"air. It seems certain , therefore, that Palaeolithic people were
surrounded by and familiar with many forms of art, and so there is really
no mystery about how its styles and techniques were passed down.

Art as information, art for survival

Indeed, it is almost certain that some ofthe art itself constituted a means of
storing and transmitting information of various kinds: this is particularly
likely for some non-figurative motifs. Once again, such notions are by no
 READING THE MESSAGES 181

means new: Lartet interpreted some sets of marks on bone and antler as
'hunting tallies' and, as mentioned earlier, other motifs were thought to be
artists' signatures or proto-writing.
A different kind of role for parietal non-figurative marks was first
suggested by Breuil in 1911 : he saw that some panels at Niaux , covered in
dots and lines, were located just where the main passage divided, and felt
that the marks might therefore be topographic guides. There do seem to be
occasional links between 'signs' and important places in a cave: at
Lascaux, for example , some lines of dots are located at points of
topographic transition, and similarly in other sites signs are positioned
where passages turn, become narrow or branch off. 134
In an extension of this view, and of the similar idea that certain caves
might have been decorated as symbolic animal drives/pounds (see p .155),
some scholars have even suggested that the artists took advantage of the
topography of cave-walls to provide or store information about the
surrounding landscape: for example, Leroi-Gourhan speculated that horse
and bison were primarily in the spacious, central parts of caves as these
represented the wide-open pastures , while figures of felines clustered in
the dark and narrow depths because these represented their lairs . 135 Others
believe that some cave-art might be a model of the local Palaeolithic
hunting territory: it has been pointed out that the axes of the main cave of
Ekain exactly parallel the line of flow of the streams in the area , while a
conical rock in the cave is like the hill housing the cave itself. 136 It is an
interesting idea, but very difficult to put forward in more than a handful of
cases , particularly as we do not know how certain features of the landscape
(such as stream-flow) may have changed. Nevertheless , it is quite possible
that some of the art acted as a mnemonic device in this way, to instruct the
younger members of the group (see below).
A few items of portable art have been interpreted as actual maps: a stone
from Limeuil with engraved meanders on it was thought to be a possible
river map; however, the best known is a fragment of mammoth tusk from
Mezhirich (Ukraine), dating to the twelfth millennium BC, which is
engraved with narrow bands, transverse lines and some geometric forms in

Fig. 115 Tracing of the so-called

engraved 'map' on the ivory plaque
from M ezhirich (USSR) . (After
Gladkih et al)
182 IMAGES OF THE ICE AGE

the middle. It has been claimed to represent a map of a river with four
dwellings along it and some fishing nets across it; 137 but in fact the object
was published upside down in terms of the direction in which the marks
were made, and the central geometric forms are totally unlike the
Palaeolithic huts of this region - in any case, the engraving appears to have
been accumulated rather than produced as a single composition. 138
The 'hunting tally' theory about sets of marks has been revised by a
number of scholars in recent years : some have used ethnographic analogy
to interpret groups of dots and lines on bones as being connected to games
of chance for telling fortunes; 139 others prefer to search for evidence of
Palaeolithic proto-mathematics. Absolon used to look for a system based
on 5 and 10, but since this is now considered to be imposing our modern
norms on the past, other scholars such as Boris Frolov claim to have found
repeated multiples of almost every number from 1 to 10 on portable objects
from the USSR and elsewhere. 140 There is said to be a particular frequency
of 5 and 7, with 3 and 4 as 'supplementary number sets' . Frolov sees this
numbering system as the basis of an incipient Palaeolithic science .
He has been vigorously opposed by Alexander Marshack, who points
out 141 that these numbers are quite random, and the discovery of any
repetitions in Palaeolithic engravings is more indicative of a researcher's
ability to count than of any number-system in the Palaeolithic, and that
claims about isolated sets of marks like these cannot be tested . There is no
reason to see them as numbers (let alone arithmetic) when, in Marshack's
view, they are more likely to be 'notations' - ie sets of marks accumulated
as a sequence, perhaps from observation of astronomical regularities . He
has tested linear sets of such marks (most notably a serpentine set on an
Aurignacian object from abri Blanchard) for 'internal periodicities and
regularities', particularly as possible examples of lunar notation, with
varying degrees of success, 142 though he believes that there are also non-
lunar notations.
Whatever the validity of their results, these studies have at least focused
attention on a type of marking which previously was ignored or dismissed
as random ; it is now clear that they are coherent and ordered , and were
carefully made over a period of time. Similarly, a recent study of the
painted pebbles from the Azilian (the very end of the Ice Age) revealed that
the 16 signs drawn on them were found in only 41 of the 246 possible
binary combinations (indicating that some sort of 'syntax' was employed)
and that there was a predominance of groups numbering from 21 to 29 or
their multiples , which may again have some connection with lunar phases
or lunations. 143
The phases of the moon would certainly have been the principal means
available to Palaeolithic people for measuring the passage of time, and
other scholars have likewise sought evidence of lunar observation, but of a
different kind: an ivory plaquette from Mal'ta (Siberia) has hundreds of
pit!"engraved on it in spirals; these have long been interpreted as symbols
of moon worship. Frolov claims that there are 7 spirals, the biggest of
which has 243 pits in 7 turns, and the others 122 (243 + 122 = 365). By
analogy with similar motifs found among modern Siberians, he interprets
it as calendrical ornamentation (24 3 days being the gestation period of the
reindeer, the staple food at Mal'ta, as well as the length of the winter, while
the summer lasts about 122 days) . 144
A different method of noting the passing of time, of course, is by means
of the seasons: those who uphold the 'hunting art' theory point out that
 READING THE MESSAGES 183

there are very few 'spring scenes' (emaciated animals, pregnancies, births)
and a preponderance of autumn images (fat animals, bouts, pre-
copulatory behaviour), while Marshack has interpreted many groups of
figures, especially in portable art, as representing particular seasons -
though some of his interpretations such as the 'moulting animals' of La
Colombiere or Lascaux are by no means definite. 145 We have already seen
(chapter 6) that little reliable zoological or ethological information can be
extracted from Palaeolithic images, though a few details such as the hook
on a salmon jaw are certainly indicative of a specific time of year.
Marshack has developed the concept of the 'time-factored' symbol to
explain the periodic use of images or accumulations of notation . They were
images of seasons, relevant to economic, ritual or social life, and made to
be used at the correct time and place. The distinctive features of a season
were depicted (eg on the Montgaudier baton) to provide a reference point
in the calendar. Hence, where previous scholars saw the 'penniforms'
around the 'Chinese horse' at Lascaux as missiles or phallic symbols
(depending on their pet theory), Marshack would see this as a horse in
summer coat running among plants or ferns .
Marshack's importance lies not only in devising new techniques for
investigating and documenting the art (see chapter 3 and Appendix), but
also in asking new questions about it, and in postulating that many images
Fig. 116 The enigmatic 'spade'
sign at Le Porte/ (Ariege) with a were built up and re-used over long periods. He believes that we are faced
painted horsehead nearby. The man with a very complex tradition of symbolic accumulations which were
painted around a stalagmite [see Fig. regionally specialised and differentiated. The notations are merely one
52] is below the sign. Probably type of sign system: many other 'marking strategies' were in use
Magdalenian . The sign is 45 em simultaneously. He has focused on what seem to be non-figurative,
across. (JVJ geometric markings found on non-utilitarian artefacts, always studying
184 IMAGES OF THE ICE AGE

the sequence and process of accumulation, rather than just the finished
Images.
In this way, he has found motifs of different types recurring throughout
the Upper Palaeolithic , such as zigzags, 'ladders', double-lines, and
(particularly in the Soviet Union) fish-scale or netlike patterns. One of the
most fundamental is the 'band motif on portable objects, which may be
equivalent to the finger-tracings on cave-walls and which, he has
speculated, may represent one aspect of a water-related symbolism 146 -
whereas , predictably, Breuil saw the parietal 'meanders' as serpents, and
Leroi-Gourhan saw them as perhaps phallic. Marshack does not claim that
they actually represent water, but that they were 'iconographic acts of
participation in which a water symbolism or mythology played a part', a
theory which fits well with the possible connection, mentioned above,
between certain caves and water entering or coming out of the earth.
These marks are no longer seen as random accumulations but instead as
a sophisticated, sequential, additive image system; no longer dismissed as
idle doodlings, but considered to be possibly the most complex element in
Palaeolithic iconography. T he act of image-making was the important
factor, and it was not always done in a public or participatory setting: as we
have seen, a great deal of parietal art is hidden away in inaccessible recesses
and tiny chambers, such as the horse and 84 'P' signs in a narrow passage
of the Tuc d' Audoubert (see Appendix), or the little chamber at La Pileta
with its animals and double fingermarks/ 47 or indeed the tiny engraved
'sanctuary' in Karlie-ngoinpool, Australia (see p.32). It is assumed that
different people squeezed into these crannies at different times to perform
some symbolic activity.
Some images, both portable and parietal, both hidden and public, were
clearly 'touched up' at times. Opinions still differ about what kind of
behaviour lies behind this phenomenon. Marshack firmly believes that
these images were used and re-used, and has published many examples
such as engraved animals on pebbles and bones from Polesini, Italy, which
had legs or muzzles added at some point after they were originally
drawn. 148 As we have seen (pp. 75-6), some of his claims that each of a
series of marks was made by a different point or hand are simply not
verifiable - a variety of lines or marks could have been done by the same
tool or hand , and one cannot take this as proof oflong-term retouching. 149
Many theories have previously been put forward to explain this
phenomenon which seems to be most frequent in the Magdalenian of
Franco-Cantabria. A proponent of hunting magic such as Begouen saw it
as the renewal of an image which had 'worked' before, the act of drawing
thus being more important than the finished picture (conversely, the
images would be broken if results did not comply with the prayers). Other
scJwlars believe that some figures, as at Lascaux, were simply improved or
rectified over a short period (they show a unity of style), the artist
experimenting until satisfied with the figure's position and quality; while
others suggest that the multiple hoofs, muzzles, etc (eg on reindeer at Ste
Eulalie, Lot) may be an attempt to depict movement. 150
The fact that we cannot assess the timespan involved prevents us from
establishing which explanation is correct; however, where carvings are
concerned, Marshack's view seems more plausible, since some of them -
such as those of Vogelherd, or even the Tata plaque - are worn and
polished from long handling, and the Vogelherd figurines seem to have
had marks added to them at times. 151
 READING THE MESSAGES 185

Fig. 117 Tracing of some painted

and engraved horses in the Panneau
de l'Empreinte at Lascaux
(Dordogne), showing extra eyes,
ears, muzzles and legs. Probably
Magdalenian . The <multiple' horse
C(Jl)ered in arrowlike marks is about
1m in length. (After Glory)

Even though some marking activity may have been private, lt IS

nevertheless true that much of it is on open view, both in caves and on
objects, and must therefore contain messages of various types.

Interpretations of Palaeolithic art have been governed largely by the

knowledge and preoccupations of academics in different decades. Just as
'art for art's sake' was the theory which best suited the end of the last
century, and 'sympathetic magic' was taken from the ethnography of the
early years of this century, so it was no coincidence that the structural and
sexual interpretation arose during the 1950s and 1960s (the era of the
sexual revolution, and of Levi-Strauss) or that ideas about astronomical
observation were developed during the Space Age.
As with Stonehenge, each generation re-invents the evidence in its own
image. Just as our computer age produced grossly exaggerated views of
that monument as an accurate eclipse predictor, the new technology has
inevitably led to some novel ideas about the purpose and meaning of the
earliest art. A great deal of research has been done in recent years into
information and its processing, and into the nature of memory and how we
can use our memory centres more efficiently. One of the major aspects of
this work concerns how art can be employed to transmit information and
imprint it on the mind .
One such approach has been an attempt to apply 'Information Theory'
to cave-art; it was thought a useful technique since it ignores the potential
meaning of the figures, and concentrates solely on their 'quantity of
information'. First applied to language in 1945, the theory defines a
message's quantity of information as a function of the number of letters of
which it is formed and the language in which it is written. The unit of
information is called a 'bit' (an abbreviation of 'binary digit').
As we have seen, it is a fair assumption that in the last Ice Age most of the
cave-art contained a 'message' which was not aimed at us and which we
cannot understand clearly. This should not matter to Information Theory,
but there is a more crucial obstacle in the fact that we do not know what
kind of classificatory system should be employed in our analysis. This is
important , because the frequency of each 'symbol' directly affects its
quantity of information; the more examples of a particular motif in a
sample, the fewer 'bits' are allotted to each of them . But how should we
divide up the figures? Should bison be separated from aurochs, or stags
from hinds?
In short, the applicability of Information Theory to cave-art is severely
limited, and the few results achieved so far could have been produced
equally well by common sense: 152 it is obvious that in a panel of similar
186 IMAGES OF THE ICE AGE

figures, such as the horses of Ekain (Fig.ll4) or the bison-ceiling of

Altamira (Fig.7), each individual figure is ofless information value than a
single, unique figure, such as the bizarre 'sign' in the cave of Le Porte!
(Fig.116), which resembles a 'spade' from a playing card and which has
also been seen as a hut, a vulva or a bird in flight. But it does not mean that
we are any closer to understanding what that information might

Exercises of this type are akin to a Martian coming to Earth and trying to Fig. 118 Photomontage of the
calculate the information contained in an incomprehensible English entire 'shaft scene' in Lascaux
message by lumping it together with some Polish, Basque and Latin in (Dordogne), shmving rhinoceros,
order to establish the probabilities of the different symbols used. They dots, apparently wounded bison,
involve an attempt to measure the degree of complexity of a structure or a bird-headed ithyphallic man and
'bird-on-stick'. Probably
message, but do not take into account the context; and, as we have seen, Magdalenian. Total width:
wit&.cave-art, the context may be even more important than the content. c 2. 75 m. (JV)
In this computer age, it was perhaps inevitable that a cave with figures
on its walls should be interpreted as a great 'floppy disk', a storehouse of
information for long-term retention and recall. Such an explanation does
have the merit of taking into account the position and distribution of
figures within a cave, and sees the blank spaces between the groups as part
of the message. Indeed , the art itself is only part of the experience: the task
of reaching it, and the strange sensations and occasional dangers involved,
are all part of the process of imprinting the message indelibly. It has long
been assumed that cave-art, at least, has much to do with the initiation of .
READING THE MESSAGES 187

young members of the group, and the 'information' and 'survival' theories
have again brought this to the fore .
We have seen (p.llO) that rock-art was done out of doors in the same
period, but caves are where parietal art has survived most successfully,
and the secrecy and darkness involved in a visit to them must have been a
memorable experience for any initiate. Quite apart from the physical
difficulties encountered in some caves, whether rock-formations, flowing
water or holes to crawl through, there is the utter blackness, the total
silence, the loss of sense of direction in the often labyrinthine passages, the
change of temperature, and the frequent sense of claustrophobia . 153
The fear of being abandoned , lost and alone in the dark, would have
concentrated the mind wonderfully and prepared the apprehensive initiate
for anything. Similar techniques are still in use today for more sinister
purposes, such as brainwashing or debriefing: victims are taken to alien,
unpleasant or unfamiliar surroundings, subjected to physical and mental
discomfort, made confused and uncertain about what is to happen; their
188 IMAGES OF THE ICE AGE

entire frame of reference is removed, the everyday world is completely

undermined- a phenomenon which Henry Miller called 'moving day for
the soul'. The technique of sensory deprivation also concentrates the
mind, and a cave would certainly affect all the senses of scared initiates
quite radically and leave them vulnerable to indoctrination.
The advocates of this idea do not assume that Ice Age people were aware
of all these factors- no doubt they used instinct as much as planning- but
the experience of centuries will have taught them which techniques were
most successful; and success was crucial since the tribe's very survival
would depend on its accumulated learning and rules being transmitted and
imprinted effectively and indelibly. Hence the careful positioning of
certain figures in niches, or in places where they could be effectively and
suddenly revealed by cunning lighting - these were the rudimentary
'special effects' of the time which must have had considerable impact on
impressionable and expectant minds. One can speculate that the mass of
unsexed human figures in the dark cave-depths represent these
adolescents and their ambiguous, transitional status.
But how exactly could these images transmit information? Clearly, we
cannot know precisely what the different animals or signs were
'metaphors' or mnemonics for, which rules or events they were connected
with. But they must have been vivid, overwhelming visual impressions,
especially by the flickering flames of small lamps which, as mentioned
earlier (p.llO) make the animals seem to move; and pictures of this sort are
extremely powerful conveyors of information: the cliche that 'a picture is
worth a thousand words' is all too true here, and cave pictures still shout
down the millennia, though now their language is alien to us. 154
We know from modern studies that the human mind can 'take in' up to
half a dozen items at a glance, and these convey far more information to the
mind when arranged in familiar patterns: for example, if one glanced at the
above cliche, and at seven unconnected words, the cliche would be
perceived more clearly and remembered. The same is probably true of
Palaeolithic art, which certainly comprises a 'vocabulary' of symbols,
some of which must have had considerable information value, and certain
combinations of which may have had special significance.
One further idea concerning mnemonics involves not the thinking of the
computer age, but methods developed by the Ancient Greeks and
Romans, and maintained in medieval Europe. These entailed memorising
vast quantities of information through a technique of impressing 'places'
and 'images' on the memory. Before the invention of printing, a trained
memory was often of vital importance; people would therefore imagine, in
their 'mind's eye', a large deserted place such as a great cathedral or a
whole series of rooms. By walking around it mentally, and memorising
thousands of places, each associated with something they wished to
re~ber, they could later repeat the 'journey' and recall everything
perfectly. The Greek poet Simonides, who wrote the first text about this
method, called it 'inner writing'; and some Roman scholars used it to
commit long texts to memory - Cicero's speeches, legal points, or Virgil's
entire Aeneid (9,700 lines , which could even be recited in reverse
order!). 155
Committing information to memory in this way is second nature to
Australian Aborigines who use vast stretches of desert, containing
thousands of special places (water-holes, hollows in trees, etc) which they
can associate with the Dreamtime wanderings and adventures of their
READING THE MESSAGES 189

ancestors. In this way, a huge amount of tribal lore, vital to survival, is

preserved and imparted little by little to the next generation during
initiations and festivals. Art plays a crucial role here; and it is quite
possible that similar methods were employed during the Upper
Palaeolithic period, both in the open air and in the caves. The latter are
suitably quiet, deserted settings for a 'mental walk', containing numerous
special places such as niches and concretions by means of which
information could readily be committed to memory.
In short, the concept of computerised information has combined with
the ancient art of mnemonics to produce a new view of the earliest art,
worthy of the 1980s. Naturally, little evidence survives to support the idea,
apart from the obvious fact that some of the art is in caves, and carefully set
out within them. We have no proof of Palaeolithic initiation ceremonies,
nor of how information was transmitted or imprinted.
Some scholars believe that parietal art constitutes a system of
communicating information, though we need to learn much more before
we can feel confident that it is a semiological system capable of
transmitting coded messages. 156 Others claim that portable objects are
more likely than parietal art to be involved in the storage and rapid
transmission of information, for the simple reason that they are portable
and not all of them are decorated. Certainly their different motifs do
display interesting distribution patterns, as, for example, in the
Magdalenian ofCantabria, where there is distinct regional variation: some
motifs are widespread, while others are unique to major centres
('aggregation sites') such as Altamira, or to particular areas, and may thus
mark group boundaries. 157
It is believed that having a set of motifs or a common art -style, different
from those of one's neighbours, helps to maintain a group's identity and
cohesion (an adherence to norms is the basis of a culture's stability), while
motifs shared with neighbours would foster intergroup solidariti 58 - a
'portable art version' of Leroi-Gourhan's view of complex signs as ethnic
markers; it has even been claimed that some objects such as 'Venus
figurines' were display items indicative of a system of communication and
information-exchange, designed to maintain contact and alliances between
the participants in mating networks which had become strained through
climatic deterioration and expansion of hunting territories! 159
Unfortunately, there is no evidence whatsoever to support these fanciful
ideas- we have seen, for example, that 'Venus figurines' are by no means a
homogeneous group limited to two short periods, and thus there is nothing
to link them with 'alliance networks', while the environmental
deteriorations thought to have triggered this crisis (and the above-
mentioned migrations of people into Franco-Cantabria) have been
assumed, not demonstrated. There were many climatic fluctuations irt the
course of the Upper Palaeolithic.
Like all the other views outlined above, these theories merely reflect one
of archaeology's current preoccupations- prehistoric social systems, the
delineation of group boundaries, and the links between style and
information content; and, like the preceding theories, they may well
contain some truth. Time will tell, but meanwhile at least one of the
proponents has decided that the 'art as adaptive information' idea is really
a truism and based on some very weak assumptions 160 - and in any case
should not be assumed to apply to all of Palaeolithic art. Other recent ideas
are simply too vague to be useful: for example, the view has been expressed
190 IMAGES OF THE ICE AGE

that group cooperation was required for large-scale communal hunts

(another truism), and that cave-art may represent the rituals involved in
reinforcing the authority of the hunt coordinators; alas, there was no
suggestion as to how this was done. 161

All in the mind

Finally, the most recent approach to Palaeolithic art has focused on the
origin of the oldest motifs, both in Europe and Australia, and drawn on
current research in neuropsychology. 'Phosphenes' are subconscious
images, geometric shapes that seem to be present in the neural system and
visual cortex of all human beings: some of the earliest art is thus thought to
be an externalisation of these images. This would explain the similarity
between the finger tracings in Europe and Australia - they sprang from a
common neural circuitry. 162
These geometric forms are entoptic (ie can be seen by the eyes when the
eyelids are shut) and are experienced, for example , during migraine
attacks. One way of inducing them is to alter the state of one's
consciousness by trance or hallucination: psychologists have shown that
the early stages of trance can produce such hallucinations in geometric
forms. Hence, it has been argued that many simple Palaeolithic 'signs' are
shapes of this type , 163 although any collection of non-figurative art is likely
to include lots of marks which resemble entoptic phenomena, simply
because there are very few basic shapes that one can draw, whatever the
motivation or source: dots, lines, squiggles and basic geometric forms. 164
As we have seen , there is some likelihood that shamanism played a
major role in the production of Palaeolithic art, and it has recently been
claimed, largely through analogy with well-documented shamanistic art in
southern Africa, that, in addition to the 'signs' , many animal figures and,
especially, the composites in Palaeolithic art may be hallucinatory images
- ie they represent the insights and experiences of shamans in trance
performances, and should thus be 'read' metaphorically. 165
This view may well account for some of the more disturbing cross-
cultural parallels in images that are far more complex than meanders or
simple geometric shapes . For example, in southern African rock-art, there
are images of a medicine man in a trance next to a dying eland (an analogy
between the animal's state and the shaman who 'died' in trance 166) which
resemble the shaft scene of Lascaux with its bird-headed man and
wounded bison (Fig.11 8); similarly, the Lascaux man , with his erection
and his 'bird on a stick', is remarkably like an Arizona rock-engraving
depicting an ithyphallic man with a bird on a stick. 167
And what are we to make of the fawn/bird spear-thrower of Mas d' Azil
(p.83) which is identical (apart from the turd!) to a pose struck by Disney's
BiiJllbi in a film made before the object was discovered? There is also a
striking analogy between one Palaeolithic scene and a Greek amphora of
the eighth century BC: the Magdalenian baton from Lortet shows a fish that
seems to be leaping up towards the genitals of a deer , and above are two
lozenges with dots inside; the amphora has a very similar composition,
except that the deer is replaced by a horse, and there are three lozenges. 168
Are these simply remarkable coincidences? Or are they indicative of
some widespread phenomenon such as 'universal myths' of very remote
antiquity, or perhaps something akin to the phosphenes - images seen
under trance conditions? Future research may help us decide.
 8
Conclusion

Readers may by now be wondering what exactly we do know about

Palaeolithic art- and the answer is: both a great deal and virtually nothing.
For a start, although we have about 275 decorated sites in Europe alone,
and thousands of pieces of portable art, they probably represent a tiny
fraction of what existed originally - most outdoor art has gone, so has
anything in perishable materials, and we do not know how many decorated
caves remain undiscovered. Even in the caves we have, untold quantities
of figures may have been weathered or washed away, trampled into the
floor or covered by stalagmite and clay - in rich sites such as Castillo or
Trois Freres new figures are still being discovered today.
There is a vast literature on the subject, and hundreds of tracings of
figures; but, as we have seen, many tracings, cave-plans and studies are
inaccurate, incomplete or biased, and everything needs to be checked,
improved and amplified. Much portable art lies unknown and unstudied
in museums and private collections, while some major parietal sites still
await adequate publication.
It is only with full and accurate facts at their disposal that prehistorians
can hope to draw valid conclusions- but , as shown in chapter 6, it can be
difficult enough to identify the figures, let alone assess what they are doing
or what they might mean; indeed, some would argue that we should
abandon the attempt to extract zoological or ethological information from
the art, whereas others see these topics as crucial to our understanding of
the art's message .
Thanks in large measure to the work of Leon Pales, rigorous and
objective anatomical criteria have been applied to a large number of
figures, and many doubtful cases have been weeded out. Currently much
effort is being devoted to distinguishing between secure identification and
mere interpretation, between what is definite and what is only probable: in
future, prehistorians working on Palaeolithic depictions will have to be far
more careful about presenting probabilities as certainties, or mere
possibilities as probabilities.
As we have seen, the current emphasis is on total recording of the art:
not just all the marks present but also their context. In a cave this entails
192 IMAGES OF THE ICE AGE

study of its topography, sediments, climate, archaeological contents, and

its use between the Palaeolithic and the art's discovery. Moreover, not only
the art's location, date and subject-matter are investigated but also the
technique of execution, and the time taken to produce it. Such topics may
seem less fascinating than the art's meaning, but they are fundamental to
any theorising, and it is no coincidence that Breuil devoted more attention
to the question of 'when', and Leroi-Gourhan to 'how', in the latter part of
their lives.
We need these solid facts . As Sherlock Holmes said, in A Scandal in
Bohemia, 'it is a capital mistake to theorise before one has data. Insensibly,
one begins to wish facts to suit theories, instead of theories to suit facts.' On
the other hand, facts do not speak for themselves: at some point you have
to choose a theory and test it against the data. Unfortunately, this has been
done badly in the past: as we have seen, all-embracing theories were
presented, and facts bent or ignored in order to make them fit. It is
unlikely that anyone well acquainted with Palaeolithic art will ever try this
agam.
The reason is simple: we are faced here with two-thirds of known art-
history, covering 25 millennia and a vast area of the world (it is ironic that
so many books on art begin with a token picture of Lascaux, bearing in
mind that it stands at the half-way mark in the history of art! -indeed ,
probably closer to us than half-way, since Palaeolithic art makes sense only
if it was preceded by a long period of development on perishable
materials). What we are trying to investigate ranges from simple scratches
on bones and stones to sophisticated statuettes, and from figures on blocks
and rocks in the open air to complex signs hidden in the inaccessible
crannies of deep caverns. Almost every basic artistic technique is
represented, together with an often simultaneous use of different
conventions such as realism , schematisation , stylisation and abstraction.
Most scholars would now agree that it is futile to try and encompass all of
this within one theory.
Not all of it is necessarily mysterious or religious, though some parietal
art is almost certainly tied to ritual and ceremony. Some is private, some
public; some figures are discreet while others draw attention to
themselves.
There is a certain thematic unity - profile views of the same restricted
range of animals are present throughout in some areas, together with a
mass of apparently non-figurative motifs - but, beyond this, any
homogeneity is more assumed than real. As we have seen, the most recent
studies emphasise that every cave has its own unique symbolic
construction, and it is becoming ever harder to establish regularities in the
midst of such wide variation .
Similarly, portable art is marked by great regional variation, both in
~ms and in motifs; it is a parallel, perhaps complementary art-form,
reflecting different preoccupations, and it is extremely rare (as with the
Altamira/Castillo engravings) that one artist seems to have tackled both.
One can therefore study the complex phenomenon of Palaeolithic art at
three principal levels: the individual sites (either a cave, an object, or a
group of objects); regional groupings (of motifs, complex signs, dominant
animal species, styles, etc); and as a whole, seeking homogeneity and
regularity. As the last option loses favour, efforts are being focused on the
other two levels of study, as well as on how things changed through the
period. As our data base becomes more complete and objective, it will
CONCLUSION 193

become easier to identify regional and chronological groupings and

characteristics, and hence theorise more securely about social systems and
information exchange. In the past , we have tried to 'run' before we can
'walk', trying to extract more information than was available, simply
because we dislike unintelligible things. But Palaeolithic art is perverse-
every new 'piece of the jigsaw' raises fresh questions and casts doubt on the
conclusions already reached. There are no absolute rules, there are always
exceptions.
We have seen that the currently fashionable theories involve a
cosmological structure, the marking of group boundaries, the fostering of
cooperation, the recording of seasonal phenomena, the maintenance of
mating networks, the initiation of the young, and shamanistic trance
images. The more traditional theories include art-as-pleasure, hunting
and fertility magic, totemism, and a macho preoccupation with hunting
and girls. In view of the timespan and vast area through which Palaeolithic
art is spread, it is probable that all these theories contain some truth, and
that there were many other motivations which may never be known.
It is generally agreed that Palaeolithic art contains messages, no doubt of
many kinds (signatures, ownership, warnings, exhortations,
demarcations, commemorations, narratives , myths and metaphors); their
basic function was probably to affect the knowledge or the behaviour of
those who could read them. We, alas, do not know how to read them . We
can analyse their content, their execution, their location and their
'associations'; but the way everything was combined into an experience
has gone for ever. Without informants, we can use the art only as a source
of hypotheses, never as a confirmation of them; without the artists, our
chances of correctly interpreting the content and meaning of a decorated
cave are very slight- and how would we know if we were right or not?
Nevertheless, we are constantly learning more about the art and the
people who produced it; every new discovery, every new idea add to the
picture we are building up. Despite its frustrations, the study of
Palaeolithic art is extremely rewarding, not only through the beauty of the
figures and the exhilarating experience of visiting the caves - the very
places where the artists worked - but also because it still represents our
most direct contact with the beliefs and preoccupations of our ancestors,
and therefore constitutes one of the most fascinating episodes of
prehistory.
Fig. 119 Jean VertutintheSalon Noir,Niaux(Ariege). (]V)
 Appendix
by Alexander Marshack

Jean Vertur had intended co write an appendix co this On these visits we recognised that these areas of
book, giving an accoun r of the methods used in his work Jean's work and interest both equally represented the
in the caves. As he is unable to do so, Alex Marshack capacity, mind and work of man . We would talk
has stepped into the breach with the following text without cease for hours about the theoretical and
concerning jean's methodology. He presents a practical problems involved m robotics and
transcript of a discussion recorded in the USA, and engineering. These discussions included the
published here by kind permission of David Abrams. extraordinary complexity of dealing with the real
physical world with instruments and sensors, and the
I even more complex problems of dealing with the real
For almost a dozen years, Jean Vertut, one of the physical and cultural world of man through concepts
leading engineers and theoreticians in the field of and symbols. For Jean and me, these problems, if not
robotics, would fly to engineering and nuclear equivalent, were comparable. We went imperceptibly
laboratories in the USA to discuss developments in the from one field of discourse to the other. Both theory and
subject and research projects being conducted between technology formed the base of our discussions within
France and the United States. On each such visit, he these areas. The constraints and limitations present in
would stop for some days at our apartment in New these human efforts were discussed, as well as the
York, arriving with a valise of documents, a loaf of techniques and potentials for increasing analysis,
French farm bread, fresh camembert cheese, a sealed understanding and production. It is doubtful whether
packet of strong French coffee, and hundreds of the Upper Palaeolithic symbol systems and cultures
transparencies documenting his recent photographic were ever discussed in quite this philosophical and
work in the Upper Palaeolithic caves . In his last years technological manner or within such a context.
this included his work in the Volp Caverns, Trois Jean brought to these highly technical and
Freres and the Tuc d' Audoubert. The engineering and philosophical discussions a deep interest in religion,
robotics projects represented his theoretical and that is, in the affective, emotional contents of symbolic
practical PJj2fessional work, part of the sophisticated imagery and ritual. He once brought with pride a
contemporary effort to replace man in certain recording in which he, his wife and children, as well as
specialised areas of production and mechanical motion. others, had participated in the reconstruction and
The study of Upper Palaeolithic art represented Jean's production of early Christian songs and hymns. In our
equally professional anq technical, but private, discussions about image and symbol, Jean referred
personal involvement in the effort to understand the often to the work of Marcel Jousse, the French Catholic
work of the hand and mind of early man, particularly as philosopher and semanticist, whose earlier theories on
found 'at home' in the Franco-Cantabrian region of imagery, symbolism, information exchange and
France and Spain. education, Jean suggested, had apparently influenced
196 IMAGES OF THE ICE AGE

Andre Leroi-Gourhan, whether consciously or not. We working with Robert Begouen to document the caves of
had long discussions about the levels of information in Tuc d'Audoubert and Trois Freres, Jean photographed
image and symbol, and how these changed during the clay bison in a still more complex manner. He used
production, during repeated viewing, and during six lamps to enhance the three-dimensional rendering,
analysis. I added to these discussions an interest in photographed the bison in stereoscopic photography,
neuropsychology and cognition, in the way in which the made a wide-angle documentary photograph of the clay
human hand and mind work to solve problems in the bison in their context within the low-ceilinged
real and cultural worlds , and the many levels of chamber, for the first time giving us a sense of their size
information that are contained in any cultural set of within the chamber (they had seemed over-large and
images and symbols. life-sized in the earlier photographs), and finally he
In the evening, when the sky and room had documented them in photogrammetry, and then, by
darkened, we would project our hundreds of computer use of the data, helped reconstruct a model of
transparencies and photographs, discussing the data or the bison for exhibition at the Musee de !'Homme in
information apparent and missing in each photograph Paris.
and image. We discussed the techniques that might be And still Jean was dissatisfied. He made endless
used to secure different levels of information from the close-up, detailed photographs of the bison,
same image. In the 1970s I had pioneered the use of documenting the way in which the statues were carved,
special analytic techniques for study of the cave images, using tools and hand-modelling, documenting the
as I had earlier pioneered the use of the microscope to remnants of unfinished clay left on and around the
study the problems of manufacture and use among the carvings, the evidence of bat clawmarks around the
mobiliary symbolic materials. With grants from the bison, etc. In this sense, no single photograph was a
National Science Foundation and the National final or adequate document. Jean and I had for years
Geographic, I had developed the use of infra-red, insisted that the usual flat photographic document of
ultraviolet, and fluorescence techniques, as well as the the cave image and the even more schematic tracing or
technique of microscopic photography for the study of a 'releve' were inadequate for documenting the analytic
host of specialised problems in the caves. Each image, complexity to be found in the images and tableaux. It
panel and cave represented different sets of problems was, therefore, with a sense of profound appreciation
and required different sets of analytic techniques. I had that we welcomed and discussed the analytical work of
deposited a set of results with the Minister for Cultural Brigitte and Gilles Delluc in deepening methods of
Affairs in Paris, but also lent Jean my filters, lamps, analysis and documentation in their studies of
charts and results, and he tested these in his own cave limestone carvings in the homesites and caves of the
work. We compared results in these evening Aurignacian period in the Dordogne. We welcomed
discussions, suggesting to each other still newer modes this work as another step in the direction of the
of analysis and documentation . necessary methodological analysis that went beyond the
Jean had pioneered in the effort to recreate the simple visual surface observations and analyses of the
ambience or atmosphere of the cave environment, the earlier researchers.
overall impression of three-dimensional space and Jean had done the photographs for Leroi-Gourhan in
complexity that could not be captured in the flat the 1960s and had slowly come to realise that these
photographs of animals or details of tableaux. His photographs, while dramatic and beautiful, were also
photographic reconstructions of the ceiling of bulls in 'falsifications' and subjective 'abstractions' , and that
Altamira and the panel of horses and cattle in Lascaux, one must, in the next stage, go beyond these simple
prepared for Leroi-Gourhan's classic volume surface images or 'documents'. He attempted to deepen
Prehistoire de l'Art Occidental, remain the most widely photography as a documentary means, and therefore
known images of these two famous tableaux from the welcomed the cartographic and illustrative method,
two giants of Franco-Cantabrian cave-art. In the 1960s developed by the Dellucs, of providing information
he pJl..otographed the famou s clay bison in the Tuc that supplemented the photograph . I had experimented
d' Audoubert, using two lamps. Later he photographed with comparable but simpler techniques of
the bison using four lamps, a dramatic three- schematically rendering the analytical data that is
dimensional photograph that has become the classic documented in the photograph but is not apparent to
representation of these famous clay-carvings. The the untrained observer. For these reasons, Jean had
Lascaux panel and the Tuc d' Audoubert photographs arranged a meeting with the Dellucs in Perigueux in
were key images in the first exhibition of European 1983, where we discussed technology, theoretical,
Upper Palaeolithic art in the United States, which I analytical and interpretative problems; and a need for a
prepared, with Jean's help, in 1978. In the 1980s, while joint effort by researchers in the field to discuss the
 APPENDIX 197

problems and possibilities being opened up in the found in the clay' in other parts of the cave, they added
contemporary period. to the sense of variable and complex symbolic use to
Leroi-Gourhan had instituted an earlier stage in the create at times images of 'art' and, at other times,
investigation of the cave-art by his visual and structural images of participatory ritual marking .
studies of the images; the relation of the images to each In the same cave, we also squeezed into a tiny,
other; and their position, form and style. While Jean difficult-to-reach 'chamber' whose upper portion was
had provided most of the photographic documentation large enough only for our two heads and our arms which
for Leroi-Gourhan's volume, Leroi-Gourhan's work were used to point out the images and the problems we
was not analytical in the sense of the work that Jean, would like to document by photography and
among others, then attempted to develop; no single microphotography. The chamber, with its rounded
image or composition was given intensive internal ceiling, had at its centre a tiny, crudely engraved horse
study. The young researchers who followed Leroi- encircled by a large number of 'P' signs, referred to as a
Gourhan (including his students the Dellucs, Michel feminine sign in Leroi-Gourhan's classification, a
Lorblanchet and Denis Vialou , and an older generation schematic figure apparently derived from the female
including Jean and myself) had begun experimenting form. Jean and I examined each sign, trying to recreate
with new levels of analysis and interpretation. Jean, as its mode of manufacture . Some were clearly made by
an advocate and discussant, was attempting to circulate stone tools, but others by a finger; some were over-
information on these many efforts among the engraved or renewed; most were made by right-handed
researchers and to bring together the separate persons persons, some by the left hand; most were in a sequence
and technologies. Many of the symposia and colloquia or row of 'P' signs, though some, apparently added
that began to appear were in some measure instituted by later, were isolated. We stood there and discussed
his persistent advocacy and efforts. which images needed microscopic documentation, how
There was no better opportunity to appreciate Jean's the entire composition and sequence of manufacture
perception of the caves as an area of analytic problems could be reconstructed, and what levels of information
than a visit to one with him . I remember a decade ago in required documentation.
one deep chamber in the Tuc d' Audoubert that we Jean began the documentation we had discussed, and
came upon an unpublished 'Aurignacian' panel of the following year we spent one long evening going over
intertwined incised macaronis or meanders. He and I his preliminary photographs and sketches of the
lay down on the damp clay floor and, like students at composition . This was not the Jean Vertut of the
work on a puzzle, went line-by-line through the entire beautiful photographs but Jean the scientist, analyst
composition, attempting to determine the point of and investigator, the technician and theoretician. The
beginning (apparently within a crack in the wall) and to excitement of these developing analytic efforts and his
reconstruct the sequence and logic or 'strategy' of its increasing steps to document the dynamics of cave use
subsequent development or construction . The and symbolism made him speak often of retiring early
macaroni, exiting at the left from the crack in the wall, from engineering to take a degree in Palaeolithic
then wandered, twisting in every direction, intertwined archaeology, and devote all his time to the study of the
and interlaced, and crossed over to the other side. We art. He died in mid-career, both in the field of robotics
lay on our backs for a long time, discussing how to and in the study and documentation of the caves.
unscramble and document the composition, and what
photographs would be required at points of intersecting
and overcrossing lines to determine priority in
II
engraving. In traditional documentation by tracing or Jean Vertut had begun taking photographs of the art in
photography, such a panel would have been published the caves in mid-century, before the introduction of
as a meaningless melange of wandering lines (see the modern flash and strobe or of high-quality colour films.
abbe Glory's tracings of the Magdalenian macaronis in With each introduction of technology, he drew on his
the_volum~ascaux Inconnu, edited by A. Leroi- knowledge of engineering to update his cameras,
Gourhan and J. Allain, 1979). meters, lights, films, and eventually began to use
Jean later began the photographic documentation special filters and films. A large part of our discussions
and we discussed his first photographs and the need for in the last few years concerned his use of new
closer analysis, but he died before its completion. The techniques and equipment for the analysis or
macaronis in the Tuc d'Audoubert were perhaps the documentation of particular types of problems. At
least interesting evidence of human activity in the cave times he developed his own photographic apparatus
in the traditional frame of archaeological evidence but, and technical devices when these were not
like sets of intentional fingerprints and spear marks commercially available. He had begun to use infra-red
198 IMAGES OF THE ICE AGE

and ultraviolet in conjunction with special filters, in practical and theoretical concerns, both m
part following on and enlarging on my earlier efforts. documenting and in attempting to understand the
He also began using high contrast colour films, dynamics of the caves and their uses. His answers to
developed for biological microscopy, to enhance the Abrams almost always attempted to indicate the
cave images. He did macroscopic studies of the inadequacy of traditional modes of documentation and
engravings to study faint and difficult images. He inquiry, and they reveal his own developing interest in
experimented with stereoscopic photography and going beyond the flat , if magnificent, photographs of
three-dimensional representations of carved and single images for which he had become known,
sculptured forms. He conducted photogrammetric particularly as the major photographer of Leroi-
studies of the bison of the Tuc d' Audoubert to create a Gourhan's volume. The videotape is a record of Jean's
computerised reconstruction of the sculptures . He many levels of interest in the caves, and the diverse
developed a method of piecing together major cave approaches he was using to document them .
tableaux (Lascaux, Cougnac, etc) from separate The tape has been edited somewhat because the
photographs by using arrays of projectors to create informal mealtime conversation during which it was
overlapping seamless images . He had begun to use made, and Jean's own occasional search for an
Munsell colour charts in recent years to verify the expression in English to convey an idea that had come
accuracy of the colours in his reconstructed panels and to him in French, tended to break the flow of his
cave tableaux, but he also intentionally altered the thoughts. I have here and there added the unexpressed
colour values of certain images to increase contrast or information needed to clarify a thought.
heighten faded images. We had both begun an
investigation of the uses of computer enhancement to To a question from David Abrams: 'The question
recover images and to unscramble particularly complex reminds me of Marcel Jousse ... Leroi-Gourhan,
accumulations, and were preparing a program to incidentally, denied that he had read Marcel Jousse. It
initiate use of the method for specific analytical may be that he forgot. Marcel Jousse was a Jesuit, a
problems in some of the major caves. priest, who had attempted to take a direction taken by
Jean never developed techniques merely for their use anthropologists today [that is , a study of the way in
but always as a means of attacking and solving which information is both remembered and exchanged
particular problems. He not only pioneered in the use -A.M .]. Jousse learned the ancient Latin and, by going
of new techniques but encouraged their testing by to the early Biblical texts, he tried to indicate the
colleagues. There was never a sense of proprietorship or manner in which Christ, for instance, taught his
secrecy in his development of new approaches. This followers. Jousse attempted to "play back" Christ's
combination of technical advocacy and an ongoing words as they are presented in the Bible (within their
effort to document and present the cave images at original cultural context) and to give them a meaning
different levels of 'information', as he put it, made him greater than they at first appear to have. In 1965 Jousse
one of the pioneers of the present new stage of research presented what may have been the first anthropological
and documentation, a research which he had begun theory of communication. He wrote four volumes
when assisting Leroi-Gourhan in mid-century and called L 'Anthropologie du Geste, the "Anthropology of
continued to develop until the time of his death. Gesture" ['gesture', however, not in its English
Without offering a hypothesis as to the meaning of meaning but in the French sense of providing meanings
Upper Palaeolithic cave-art, he contributed beyond the overt significance of the words - A.M.].
enormously to the documentation and to our Jousse used the term "anthropology" in the way it is
understanding and appreciation of its complexity and used in English, not in the way in which it is used in
beauty. French. He was interested in the fact that information
and memory proceed by way of continuous "playback"

... III
Some months before he died in 1985 , a videotape was
-by the "playback" of emotion, by the "playback" of
the word, by the "playback" of the image, and so on,
and that it is by such continuous and repeated playback
made in California of Jean's informal comments and that we reach a higher level of knowledge [that is, a
thoughts on Upper Palaeolithic art and his developing synthesis of meanings at a higher level - A.M.] . This
analytical studies. The tape was made at the home of occurs not by a straightforward intellectual approach
David Abrams, an American who for a number of years but by an incremental method which is, incidentally,
has led some of the American tourists who pass through the way that one actually learns ... One learns by way of
the Franco-Cantabrian caves. Jean's responses to the image, the gesture, and the word. There are many
Abrams's questions , while informal and sketchily levels of information in these modes . . . and also,
conversational, reveal the wide range and depth of his therefore, many levels of analysis. In the study of Upper
 APPENDIX 199

Figs. 120/1211122 Different views

of the clay bison of the T uc
d'Audoubert (Ariege), showing how
they were made and set upright
against the rocks. (JV). [See also
Figs. 61 and 62}
200 IMAGES OF THE ICE AGE

Palaeolithic art it was Alex who put me on to this track historians of art attempt to explain these icons makes no
. . . The Dellucs have recently added to this type of sense to someone who knows their origins and contents.
analysis of the images , and Robert Begouen has also Religion helps ten times more in giving them meaning.'
been a powerful influence in developing new levels of
analysis. There are many prehistorians of cave art who ABRAMS: 'You don't know how far one has to go to find
still think that the "releve" (the tracing) is the real (or anyone in the United States who thinks this way, much
valid) level of information for the cave images.' less talks this way about cave-art.'
QUESTIONER: 'You talked about the importance of
"gesture" in symbolism . Do you mean by "gesture" the VERTUT: 'Alex does this ... I think that you, too, are
actual act of painting?' probably moving in this direction ... in publishing and
in trying not to criticise but to advise most
VERTUT: 'This is the first level only of what I mean. archaeologists who are interested in cave-art, please do
There are, necessarily, other levels. I am not sure that not go to the drawings of Breuil. The printed
"gesture" in English has the same meaning as "geste" in [archaeological] material often contains too little
French. "Geste" in French has a greater power. It has at information to discuss this art intelligently. One must
the same time the meaning of emotion [effect and go to the material and do better research .. . everything
emphasis- A.M.] . It may refer to the roots of strong that has been published by researchers in cave-art ...
emotion that were always present in the ancient oral may be, in some measure, true, but the truth is, in fact,
tradition. When you hear the mere recording of a text, far more complex.'
for instance, you lose a significant part of the original
meaning. Such a recording, of course, is better than ABRAMS: 'I have some pictures here of Altamira made
losing the whole ... The fact that one draws on a wall by Breuil. I now wonder how much the pictures can be
does not provide the same meanings as would the trusted ... They are so ... '
gestures that would accompany your words or the
meaning of the words themselves. These are difficult VERTUT: 'You have to understand that he was
and profound questions . . . They concern the interested in one image at a time. His effort was to
differences in teaching by a use of words, images, or remove the one image from its context, separating the
gestures and emotions ... images [documenting each image- A.M.]. He was, at
'After the discovery of Lascaux, Marcel Jousse that time, thinking of the "art for art's sake"
attempted to go beyond Breuil's limited approach to the explanation, which was part of the problem. It took
images by giving additional life and depth to their Breuil a long time to see and document everything
possible meaning, stating that they were "mimograms" separately. If you use these separate images merely for
[images containing complex levels of information - guidance, and to keep a record and not as the
A.M.]. Of course they were mimograms. But of what fundamental information but as guides to memory
type? That question could not be answered in the which can be useful in reconstruction ... '
Forties because archaeologists at that period believed
that a particular image, for instance, was a "shaman" or
a "hut" , applying ethnographic analogy and ABRAMS: 'Jean Noel, a guide at Niaux, took us to the
ethnographic information to the images in the caves. Salon Noir. He didn't say anything. He started to sweep
The problem is far more complicated .. . with his lamp the whole side of the cave .. . He brought
'With Leroi-Gourhan's work it was finally realised me to the concept of not the individual painting but the
that Upper Palaeolithic imagery was, in fact, more panel ... It was the most dramatic ... '
complicated. Actually, it is as complicated as religion
itself and the most complicated religious imagery and VERTUT: 'I had a marvellous visit last summer in Niaux
syw.bolism. It represents, in fact, another universe. We with young friends I met in Niaux with Clastres when
do not today live in that universe .. . Leroi-Gourhan was he was alive. I always work in Niaux in the night time. I
the last in the generation of what we call in French spent many nights, sometimes six or seven nights, in
"litteraire!l' , who thought of Upper Palaeolithic art as 1961 and 1962 and '64, and then later again in '6 7. I had
the beginning or introduction of art, as part of the the privilege to visit the Clastres Gallery with Leroi
history of art. You know , of course, that the history of Gourhan and to photograph the hall with footprints in
art is a "game" that provides us only with certain types the sand . . . That was very impressive . . . I have
of analytic information, but it is not the basic or original documented about 95% and reconstructed some of the
information contained in the image itself. My wife, for panels ... I did it again in 1972 for Le Thot [the Centre
instance, has been painting icons. The way in which the for Prehistoric Art, in the Dordogne].'
 APPENDIX 201

ABRAMS: 'The bison in Niaux in the book . That picture This is exactly the same tradition of "graphisme" that
is extremely powerful. It has a power as big as the book Alex is studying.
itself ... ' 'Concerning the head of the Niaux bison which opens
Leroi-Gourhan's volume and which you [Abrams] find
VERTUT: 'I had a Graflex at the time ... This is the so powerful, this is precisely the point at which trouble
simplest technique of just taking the picture. The starts. To me your reaction to the head of the bison is
subject is powerful. The picture is just a good picture . not excessive, but the full bison is far more interesting
This is the easiest gesture to make. In these days there is than the face .. . Showing the face at full scale [as it
no difficulty at all in taking a good picture ... We have appears in the book] favours the emotional reaction to
high resolution and great enlargement ... That profile it. This is, of course, accepted, but if you read too much
of the bison is about full scale. There is great difficulty into the emotion you can disadvantage your approach
in photography. When you take a picture, you keep
certain things about the "graphisme" [or image itself] 'So to substitute a cave visit is one of my aims ... to
but you lose others. I did for Robert Begouen the first show the art properly to the public. The only way to do
microphotographic enlargements, three to four times this is to show a full-scale print ... This was attempted
natural size, of engravings [in T rois Freres and the Tuc approximately at Le Thot, but it is also necessary to
d'Audoubert]. He was amazed at what could be seen ... properly prepare a person before he sees the picture.
If you want to reproduce the Lascaux paintings, for That is, of course, impossible ... In fact, it is only
instance, natural size is sufficient, but if you want to do during recent years in the preparation for Le Thot that I
a replica of the surface , it must be done in the small began to try to counteract the problem of growing
tunnel where you descend to the red deer ... I think you difficulty in visiting the caves.'
must produce a 3 X target replica ... but this again is
merely a technique.' ABRAMS: 'Some years ago there were no pictures of
people in the caves to give a sense of scale to the images.'

ABRAMS: 'I think that some of the power of the Niaux VERTUT: 'This is a problem of the photography ...
bison is their full size ... ' [Later] 'Back to the scale problem. You don't need
people in the picture for scaling as far as I'm concerned.
VERTUT: 'But that is only one of the factors .. . Those who have once been in a cave already have a
[Later] 'This is one of the particular ways of drawing sense of scale .. . We have to reveal the scale carefully in
the bison which is not an exact bison but is instead very another way ... Those who have never been in a cave on
human-like. The eyes are human eyes ... You have two seeing the pictures lose their sense of scale and .. . time
types of bison in cave-art. There are those that have [the spatiaVtemporal sense of distance, size, depth, etc.
nearly human features .. . In Altamira in the deep -A.M.]. Those who live in a structured environment
gallery you have reliefs where the added eyes and horns [have their own normal sense of time and space from
... [make it] impossible to tell whether the head is a their environment].
human or bison ... but here in the Breuil drawing of an [Showing a photograph taken in the Tuc
Altamira bison the individual is complete .. . This is one d' Audoubert] 'These are finger marks [in the clay floor]
of those interesting ambiguities . When everything is ... We can see them in detail now, made with my zoom
clear and simple, it is not interesting ... And this is one so the field is very limited ... It's very difficult to see
of the interesting things about the cave-art ... The with flash ... This is the closest I could go without a
"woman-bison" of Pech-Merle just underneath the red tripod. I never use a tripod because it is too time-
hand and dots, as seen by Leroi-Gourhan, and the consuming ... This was thought to be a bear footprint
woman on the ceiling made by finger-marks, ... When you get close up, you can see the traces of the
apparently represent a form of transformation ... But bear hair in the clay and there are hundreds of them in
thffl is fullo@f ambiguity ... You know, Leroi-Gourhan the cave ... Here is a bear skull on the ground. Next to it
found the curious relationship of female figures and a [Palaeolithic] person came, and here are the traces of
bison, but we still do not know if the horse is related to knees on the ground. The person broke off the bear
the male ... He found many examples where the female canine which is the "nice" tooth. This is the act of a few
and bison were associated , and this was one reason he seconds which you have to trace [and document by
suggested that the bison was "sexual" and female. But photography- A.M.] . Just two metres away is a small
the horse [which Leroi-Gourhan suggested was male] is "gour', a small dam [or clay mound or cone] . Here at
ambiguous. The female and bison association is the right you see the slippage of the right footprint ...
frequent but not the male with the horse ... There are a other footprints here ... and you have the skeleton of a
few examples: in Lascaux, Villars and Trois Freres ... snake that was left. One of the vertebrae was recently
202 IMAGES OF THE ICE AGE

removed and studied , and it was found to be a viper. earlier was made with four flashes. Here is a picture of
You have many traces of this kind which are traces of the rock upon which the bison were placed. The two
life ... Further on you have footprints, heel prints ... bison were, at first. badly supported, so support clay
This is a rib that was removed by prehistoric man from a was added below ... Here is a photo of the female bison .
very old skeleton of a cave bear .. . A little further on ... We can see the horns and ears, and the photo reveals the
footprint slippage ... with holes from water dripping complete process of finishing it. These marks were
from the ceiling . made with tools , but these others by fingers (Fig .62).
'The visit to this cave, and one or two others, was to a 'This is a long-distance shot of the chamber with the
very secret place ... where maybe once three or four two bison . It was made with a wide-angle lens. This is a
people came and left ... Closer to the clay bison is a deep unique picture never made before. It gives you a
depression in a small chamber called the "heel detailed view of the ceiling including much new
chamber". The clay has been covered by calcite, some information. You can see the red of the ceiling which
of the calcite has been removed [in modern times] and may have been one of the reasons for placing the bison
underneath you see absolutely fresh clay still soft where here (Fig .6l.)
finger marks are still fresh . The problem is how to study 'Here is the picture of another bison, a very small
this area of human activity without destroying all of the third one, behind the other two. This bison is a
covering ... duplicate made from clay of the cave, a duplicate that
'Here are heelprints in a very low-ceilinged room. has been officially returned to the cave in a casting made
Here again you see fingerprints, as typical in the Tuc at StGermain [from the original] by a special process.
[d'Audoubert], double prints of the fingers .. . Here is a 'Here is a close-up photo of a piece of clay that had
sign like the branch of a tree that Alex would call been left on the rock with all the fingerprints still on it.
"vegetal" ... The full study of the area has still to be The carvers left it there after the bison were finished.
done ... 'Here is a picture of the supporting clay under the
'This is a big, big vulva that is in the book .. . I am bison. Here is a close-up of the broken tail. Here on the
now preparing microphotographs by which it will be rock and the bison are small, intriguing marks. They
possible to identify the types of tool used ... On the right seem to be marks made by bats that landed on the sharp
is a small mound, apparently made by humans since it is edges.
impossible otherwise to account for the mound in an 'Here is the photo of an extra piece of clay that was
area where everything else is flat. On this mound there left at the spot [by the original makers] that should have
are [spear] marks. One can't walk in this area because it been removed at that time.
is covered by calcite so the pictures were taken at some 'Here is the photo of a very dramatic recent
distance. discovery. On the supporting rock opposite the bison
[Referring to another photograph] 'In this picture is a we found the only engraving. That engraving, a "sign",
stalagmite that was used to cut the clay and remove a big is like a signature.'
plate ... approximately in the shape of a bison ... Here
the finger marks again, so this was attended with a lot of Almost every picture that Jean took in the caves of the
ceremony. On the other side there are pieces of Tuc d' Audoubert and Trois Freres in the last few years,
macaronis which were covered with clay. These were while working with Robert Begouen, was taken to make
tests in my campaign to indicate the way in which we a point, or to present data on the complexity of the
could work ... to get more information or data on the human activity associated with use of the caves and the
footprints in a precisely measured , good, metered way creation of the. diverse images and symbol systems
for photogrammetry construction. Here are the spear found in these major caves. When he showed his slides,
marks and scratches in the mound. It will, however, be there was a lesson in documentation, analysis and
possible to make a mould and get a print of the spear. complexity in each. There is a profound sense ofloss in
'The original photo of the two bison was made with his passing and the silence that now accompanies his
twa..flashes ... but the magnificent picture you saw massive documentation.
 Notes

Preface authors, there is not a single mention by anyone of

drawings in that cave before the mid-twentieth
1. See discussions and references in Davis 1986, 1987; century (see de Saint Mathurin 1958 ; Delluc &
Bednarik 1986; Pfeiffer 1982; Stoliar 1977/8, 1981; Delluc 1981).
Ucko 1987; Zhurov 1977/8, 198 1/2. 9. Molard 1908 .
2. Sieveking 1979. 10. Chabredier 1975 , p. 12; Oilier de Marichard 1973,
3. Pfeiffer 1982. p . 28.
Introduction: Cave Life 11. Cartailhac 1908, p. 522; 0 llier de Marichard 197 3,
p. 28.
1. Andrieux 1974, and pers. comm. See also Pales & de 12. Decouvertes 1984.
St Pereuse 1976, pp. 129, 133 ; 1979, p. 140; Pales et 13. Kuhn 1955, pp. 45/6.
al1976. 14. Maria gave this account in conversation with
2. L'Art des Cavernes 1984, pp . 433-7; Vialou 1986. Herbert Kuhn in 1923 (see note 13). She claimed
3. Bahn 1987b; Tyldesley & Bahn 1983. she was five years old at the time of the discovery;
4. Arl. Leroi-Gourhan 1983. other informed sources place her age at eight
5. Vallois 1961; Dastugue & de Lumley 1976, pp. (Garcia Guinea 1979, p. VIII) or nine (ibid . p. 19).
616/ 17. 15. Contrary to popular accounts, Maria would not
6. Roper 1969, p . 448. have used the word 'taros' for the animal figures
she saw (see Garcia Guinea 1979, pp . VIII and 20).
1. The Discovery of Ice Age Art 16. Kuhn 1955, p. 46.
1. Pittard 1929. 17. de Sautuola 1880.
2. Faure 1978, p . 75. 18. Kuhn 1955, p. 48.
3. Contrary to most textbook accounts, the Chaffaud 19. Garcia Guinea 1979, pp. 42-4. On de Mortillet's
bone was not found in 1834 (see de Saint Mathurin anticlericalism, see Reinach 1899 and Bahn 1988.
- 1971; 'Worsaae 1869). 20. Harle 1881.
4. Lartet 1861. For his later finds, see Lartet & 21. The earliest known find of a Palaeolithic lamp
Christy 1864, 1875. occurred in 1854 at La Chaire aCalvin (Charente),
5. Bouvier 1977, pp. 54-7. and several others were also discovered before
6. Kuhn 1971, p. 14. Altamira, but they were either contested or
7. Simonnet 1980, p. 13. ignored, while some remained unpublished for
8. Contrary to the often repeated claims (eg Nougier years because of the controversy over the existence
& Robert 1957) that the 'paintings' of Rouffignac of a Palaeolithic lighting system (de Beaune 1987,
were first mentioned by Fran<;: ois de Belleforest in a p. 12).
publication of 1575, and occasionally by later 22. Begouen 1947, pp. 494/5.
204 IMAGES OF THE ICE AGE

23. Kuhn 1955 , p. 49. Anati (1985, p. 54; 1986, pp . 789/791 ) believes
24. Garcia Guinea 1979, p. 42; Madariaga de Ia Campa that some rock-art in Tanzania may be 40,000 years
1981, p. 302. old, partly because levels of that date have been
25. Breuil 1964, p . 11. excavated in decorated sites, but also because some
26. Piette 1889, p. 206. parietal figures seem to resemble those on the
27. Roussot 1965; Delluc & Delluc 1973, p. 205. Apollo 11 Cave's stones.
28. As mentioned in note 21, early finds of lamps had 9. Wendt 1974, 1976.
met with resistance or ridicule; Riviere himself had 10. For Border Cave , see Butzer et a11979, p. 1212,
originally failed to take seriously the claims of aM. note 44; for Matupi Cave, see Van Noten 1977, p.
Detrieux who informed him in 1896 that he had 38.
found a lamp near Le Moustier (de Beaune 1987, p. 11. Anati 1986, pp. 789/792.
12). In short, the history of the discovery and 12. Kumar et al 1988; Wakankar 1984, 1985; Anati
acceptance of Palaeolithic lamps closely mirrors 1986, pp. 789/793.
that of cave-art itself. 13. You Yuzhu 1984.
29. Roussot 1972/3. 14. Sohn Pow-Key 1981.
30. Begouen 1932, p. 8; 1952, p. 289. 15. Sohn Pow-Key 1974, p . 11.
31. Cartailhac & Breuil I 90 5 . 16. Aikens & Higuchi 1982, p. 107.
32. Cartailhac 1902 . 17. Wright 1971.
33. Brodrick 1963 , p. 63. 18. Rosenfeld et a11981.
34. Cartailhac 1908, p. 520. 19. Brown 1987, p. 59; Brewster 1986.
35. Cartailhac 1908, p. 512 . 20. Rhys Jones, pers. comm .
21. Chaloupka 1984; Murray & Chaloupka 1983/4.
2. A Worldwide Phenomenon For arguments against, and a reply by Chaloupka
1. Bahn 1987. & Murray, see Archaeology in Oceania 21, 1986,
2. Aveleyra 1965; Messmacher 1981, p. 94. 140-47. On the problems of identifying species in
3. Messmacher 1981 , p. 84. Australian rock-art, see Chapter 6, p.115).
4. Kraft & Thomas 1976. 22. Dortch 1979, 1984.
5. Anati (1986, pp. 789/796) claims that some rock art 23. McNickle 1985; Nobbs & Dorn 1988.
at Rio Pinturas, Chubut (southern Patagonia, 24. Hallam 1971.
Argentina) is 'associated' with archaeological levels 25. Aslin & Bednarik 1984a/b, 1985; Bednarik 1986.
dating to 12,000 years ago, but this remains 26. For regional studies, see Barandiaran 1973, Moure
unsubstantiated. The Peruvian cave of Toquepala Romanillo 1985 (N Spain); Challot 1964, 1980, de
has red figures of camelids, deer and armed Saint Perier 1965, Piette 1907 (France); Lejeune
hunters on the wall , and two small 'brushes' of 1987, Twiesselmann 1951 (Belgium); Adam &
wood with tips of wool impregnated with red ochre Kurz 1980, Basinski 1982 (Germany and
were found stratified here in levels dated to almost Switzerland); Eppel 1972 (Austria); Graziosi 1973
10,000 years ago (Julien & Lavallee 1987, pp. 49- (Italy); Freund 1957, Hennig 1960, Valoch 1970,
50). See also Linares 1988. Muller-Beck & Albrecht 1987 (Central Europe);
6. Guidon & Delibrias 1986. Bednarik (pers. comm. Abramova 1967, Praslov 1985 (USSR).
27/8/87), who recently visited the site, does not 27. Belfer -Cohen & Bar-Yosef 1981 , p . 35. There is
accept this date for the paintings; some of the art is also a faint possibility that some linear rock-
certainly about 9,500 years old, and there is also engravings in a number of caves at Mount Carmel
'much older art' which remains undated. are Palaeolithic (see Ronen & Barton 1981); and an
7. Whitney & Dorn 1987. enigmatic engraved figurine was recently found in
8. On Zimbabwe, see Walker 1987, p. 142. Mori (eg the Acheulian (Lower Palaeolithic) site ofBerekhat
..,974), like Frobenius before him, has proposed Ram in the Golan Heights (Goren-Inbar 1986) .
that the earliest rock-engravings of the Sahara may 28. Campbelll977, vol. 2, Figs. 102, 105, 143. A brief
date back to the late Pleistocene. However, other account, without illustrations, can be found in
scholars such as Muzzolini (1986, pp. 312-14) Sieveking 1972 . There is some doubt about the
point out that this reasoning is based on a few authenticity of a horse-head engraved on a
isolated, and possibly anomalous, radiocarbon fossilised bone from Sherborne (see Antiquity 53 ,
dates from occupations at the foot of decorated 1979, 211-16; Nature283, 1980, 719-20; Antiquity
rocks, and which may have no association at all 55, 1981,44-6 and 219-20).
with the engravings. The question therefore 29. See, for example, Bahn & Cole 1986.
remams open. 30. Chollot-Varagnac 1980, p. 457.
 NOTES 205

31. de Saint-Perier 1965. 55. Leroi-Gourhan & Allain 1979.

32. Bouet et al 1986/7; Delporte et al 1986; a 56. Vialou 1986, p. 187 . It is noteworthy that Vialou
computerised data bank has also been set up for the (p. 166) counts clusters of the same motif as a single
French parietal sites (Djindjian & Pin<;on 1986). unit: thus the 189 circles engraved on a bear figure
33. Pericot 1942; Llongueras 1972 ; Fortea 1978. at Trois Freres are counted as one sign. If a
34. Pales 1969; Pales & de St Pereuse 1965. different system were adopted, the resulting
35. Bosinski & Fischer 1974, 1980. figures and percentages would be totally different.
36. Begouen & Clones 1987; Bahn 1983. 57. Begouen & Clones 1987.
37. Graindor & Martin 1972; Martin 1973. 58. Vialou 1986, p. 351.
38. Breuil thought some red streaks in the Welsh 59. Tosello 1983.
coastal cave of Bacon Hole might be Palaeolithic 60. Bahn 1982, 1984.
(1952, p. 25), but these have now faded away, and 61. Conkey 1980.
are thought to have been either a natural
phenomenon or the traces of someone having 3. Making A Record
recently cleaned a paintbrush on the wall! The 1. Delporte 1979, p. 62.
recent claims for art on the walls of a cave in the 2. See, for example, Rivenq 1976, which compares
Wye Valley (see Illustrated London News Jan. modern photographs of a frieze in Montespan with
1981, 31-4; Current Anthropology 22(5), Oct. those taken in 1926; Begouen 1980, p. 689, which
1981, p. 501-2) proved to be based on a mixture of does the same for Trois Freres; or Omnes 1982, pl.
natural features and wishful thinking (see XIV, which does the same in Labastide.
Antiquity July 1981 , pp. 81-2, 123-5; I.L.N. May 3. Brodrick 1963, p. 53.
1981 p. 24; Current Anthropology 23(5), Oct. 4. Breuil, letter to Glyn Daniel, 20/8/55 , in author's
1982, p . 567-9). possession; see also Aujoulat 1987, p. 22.
39. Maringer & Bandi 1953, p. 23. 5. Begouen & Clones 1987.
40. Bosinski 1982, p. 6; Breuil 1952, p. 24; Freund 6. Boyle et a11963, p . 15 .
1957, p. 55 . 7. Delluc & Delluc 1984, p. 57. Non-existent circles
41. Bader 1965 , p. 25; Maringer & Bandi 1953, p. 27; also appear on some ofBreuil's versions of portable
Anati 1986, p. 789/93 ; Shimkin 1978, p. 280. engravings from Enlene- see Begouen et all98415,
42 . Bader 1965, 1967; Shimkin 1978, pp. 280-82. pp. 29, 40; this article (pp. 7112) shows that
43. Praslov 1985, p. 181; Abramova, pers. comm. Bouyssonie's copies of portable art are often more
44. For example, the lone engraved bison of Segries reliable than those of Breuil which feature
(Alpes de Haute Provence) is thought by most omissions, additions and a certain amount of
(though not all) specialists to be a fake, and was wishful thinking.
therefore not included in L 'Art des Cavernes: see 8. Begouen 1980, p. 684; Vertut 1980, p. 674.
de Lumley 1968. For the doubts, see Pales & de St 9. Breuill952a, p. 11.
Pereuse 1981, pp. 75-86: the figure is quite 10. Cartailhac & Breuil 1906; Breuil & 0 bermaier
different in its proportions, technique and style 1935 .
from unquestionably authentic bison, and IS 11. Breuil 1949.
isolated in a region empty of Palaeolithic art. 12. For unquestioning acceptance ofBreuil's tracings,
45 . Aujoulat et al1984. see, for example, Leason 1939, p. 51: 'In verifying
46. L'Art des Cavernes 1984. his theory the Author has had recourse to
47. Arte Rupestre en Espana 1987 is the most up-to- published sources, but, coming from the
date synthesis of the Spanish caves, pending the distinguished hand ... of Professor Henri Breuil,
announced publication of an equivalent of the their accuracy can hardly be called into question.'
French Atlas (note 46). See also Straus 1987. 13. Pales 1969, pp. 105/6 .
48 . Grazi-1973; Naber et all976. 14. This was a problem, for example, in a recent re-
49. Basler 1979. examination of the heads of the Trois Freres felines
SO. Carciumaru & Bitiri 1983. - see Begouen & Clones 1986/7; and comment by
Sl. For detailed regional distribution maps, see notes Clones in Marshack 1985, pp. 175-7 .
46 and 47. 15. Begouen 1942, p. 6.
52. Delluc & Delluc 1973; L'Artdes Cavernes 1984, p. 16. Garcia 1979; Martin 1974.
208/9; Fortea 1981. For other examples of caves 17. Begouen & Clones 1987.
with a single figure, see Jorda 1979, p. 332; 1985. 18. Pales 1969, p. 40; Airvaux & Pradell984, p. 213;
53. L'Artdes Cavernes 1984, pp . 376/7,347-9. Faure 1978.
54. L'Art des Cavernes 1984, pp . 378-80. 19. Marshack 1972.
206 IMAGES OF THE ICE AGE

20. d'Errico 1987. 16. Straus 1982, p. 78.

21. Begouen 1980, p . 685; 1984, p . 79; Vertut 1980. 17. Combier et a11958; Combier 1984.
22. Vertut 1980, p . 667 . 18 . Arl. Leroi-Gourhan 1983, and in Leroi-Gourhan &
23. Marshack 1975 . Allain 1979.
24. Vialou 1982; 1981, p. 81. 19. Cartailhac & Breuil 1906; see also Barandianin
25. Marshack 1975; see also Marshack 1985, pp . 68, 1973, pp. 302-7 .
77-9 (and pp. 63/4 for criticism by Lorblanchet). 20. For Labastide, see Omnes 1982, pp . 182, 187; for
26. Vertut 1980, p . 672. Gargas, Barriere 1976, pp . 406-9; Ucko 1987, pp.
27. Lorblanchet 1984; see also Aujoulat 1987. 40-1.
28. Ruspoli 1987, pp. 177-86. 21. de Saint Mathurin 1973, 1975, 1978.
29. Begouen 1984, p. 79 . 22. Almagro 1976.
30. Delluc & Delluc 1984a. 23 . Utrilla 1979; see also Almagro 1976, p . 99; Straus
31. Laurent 1963, 1971. 1982.
32. Tosello 1983, p . 285. 24. Jorda Cerda 1972, 1980.
33. Faure 1978, pp. 68/9. 25. Cartailhac & Breuil 1905.
34. Roussot 1984c. 26. de Balbin Behrmann & Moure Romanillo 1982;
35. Bouvier 1977, pp. 54-7. Moure Romanillo 1986.
36. See Pales 1970 which puts an end to an error of this 27. Glory 1964.
kind. 28. Leroi-Gourhan & Allain 1979.
37. Lorblanchet 1984. 29. For example, Breuil1952, p. 149.
38. Begouen 1942, p . 9. 30. Stoliar 1977/8, 1981, 1985 ; similarly, Hornblower
39. Pales & de St Pereuse 1965, p. 221. (1951) assigned these models to the Gravettian, but
40. Pales 1964, p . 12; Pales & de St Pereuse 1965, p. there is no evidence for this either. For a reply, see
230. Zhurov 1977/8, 198112; Bahn 1984, p. 343; 1987a.
41. Pales 1969, p. 26; Pales & de St Pereuse 1976, 31. See Altuna 1975, pp. 104/5, or Altuna & Apellaniz
1981, & forthcoming . 1978, p. 22, for an illustration of the difference
42. Bednarik 1986, p . 35; Meylan 1986, p. 47 . between a side- and a front view of a living red
43. Meylan 1986, pp . 46-50; Lhote 1972, p . 323 . deer's antlers.
44. Ucko 1977, p. 8. 32. Rousseau 1984; Pales & de St Pereuse 1981, pp.
45. Rivenq 1984. 94-7 .
46. Lorblanchet 1981. 33. Guthrie 1984, p. 58; Mithen (1988) has also
47. Vialou 1982. adopted this view.
34. Riddell1940, pp. 158/9.
4. How Old is the Art? 35. Breuil 1977.
1. Airvaux & Pradel 1984; Begouen & Clones 1979, 36. de Saint Mathurin 1973, 1975.
p. 19. 37. Breuil1977, pp. 57/8.
2. Piette 1894, p. 129. 38. Ripoll1964.
3. Bahn 1984, pp . 164/5. 39. Breuil 1912; 1952.
4. Berenguer 1986, p. 669. 40. Breuil & Lantier 1959, p. 225.
5. Faure 1978. 41. Laming-Emperaire 1962, 1959 .
6. Bahn 1982. 42. Jorda Cerda 1964, 1964a; see also Altuna &
7. Combier 1984. Apellaniz 1976, pp . 158-63; Barandiaran 1973, p.
8. The development of Piette's views is explained in 310.
Breuil1909; and see Breuil1959, p. 15; Delporte 43. Leroi-Gourhan 1971.
1987. 44. Leroi-Gourhan 1962.
9.~onzalez Echegaray 1968 ; 1974, pp. 39-42. 45. Delluc & Delluc 1978, p. 386.
10. Gonzalez Echegaray 1972. A similar attempt to 46. Moure Romanillo 1985, pp. 124/5; Barandiaran
date caves from the dominant species depicted was 1973.
made by Jorda, quoted in Altuna 1983, p. 236. 47. Delluc & Delluc 1983, pp. 32/3.
11. Begouen & Clottes 1985 , p. 43 . 48. See Ucko & Rosenfeld 1967, p . 78, for a discussion
12. Rouzaud 1978, p . 32. of these points.
13. Laming-Emperaire 1959, pp . 32/34; Roussot 49. See, for example, Bahn 1984 for Pyrenean
1984a. cultures, or Soffer 1985 for those of the Central
14. Carcauzon 1986,1988. Russian Plain .
15. Gaussen 1964. 50. Lorblanchet 1977.
 NOTES 207

51. Guthrie 1984, p. 72. 31. Pond 1925; for a more detailed and illustrated
52. Almagro et al1972 , p. 469 . account of Aurignacian production of ivory
53. Begouen & Clones 1985 , p. 49; 1979a, p. 60. beads, see Otte 1974.
54. Altuna & Apellaniz 1976, pp . 157, 165; Apellaniz 32. Taborin 1982.
1982, p. 92; Almagro 1976. 33. Jia Lanpo 1980, p. 52; Atlas of Primitive Man in
55. Pericot 1942; Jorda 1964; Lorblanchet 1974, pp. China 1980, pp. 114/15.
82/3; see also Straus 1982. 34. Leroy-Prost 1984, p. 45.
56. Airvaux et al1983. 35. Frolov 1981, pp. 63, 77; 1977/9, pp . 153/4,85. It
will be recalled (Chapter 2, Note 27) that an
5. Forms and Techniques engraved stone figurine has even been found in an
Acheulian site in Israel.
1. Marshack 1987. 36. Marshack 1972, pp. 255-8; Faure 1978, p. 65.
2. Fages&Mourer-Chauvire 1983;Absolon 1937. It 37. Omnes 1982, p. 185; Bosinski 1973, p. 39.
should be noted that the so-called 'Aurignacian Marshack has recently admitted (1985, pp. 72/3)
flute from the abri Blanchard, located in the that his own experiments showed that a line's
Musee des Antiquites Nationales' mentioned by cross-section altered as the line/tool changed
White (1986 , p. 109) is actually a specimen from direction; it is the changes of direction and of
La Roque (Dordogne), probably Gravettian, and rhythm which suggest to him the presence of
housed in the British Museum! different tools or hands.
3. Pfeiffer 1982, pp . 18112. 38 . Bosinski & Fischer 1974, pp. 5/6; Faure 1978, p.
4. McBurney 1967; Marshack 1987. 65; Marshack 1985, pp. 72/3.
5. Allain 1950. 39. d'Errico 1987.
6. Roussot 1970, pp. 9/10. On perforated reindeer 40. Begouen & Begouen 1934; L. Begouen 1939, p.
phalanges and experiments on their breakage and 298; Begouen et al198415. See also de Saint Perier
use, see Harrison 1978. 1930, p . 81; Capitan & Bouyssonie 1924, p. 40;
7. Bibikov 1975. For a cautionary note, see Soffer and Pales & de St Pereuse 1976, p. 17.
1985,pp.420,468-9. 41. For Ferrassie & abri Durif, see Pales & de St
8. Dams 1984, 1985, 1987a, pp. 60, 193-5; Glory Pereuse 1979, pp . 137/8.
1968, pp. 55/6. 42. Begouen et al 1984, p. 145; many of the La
9. SeeMarshack 1981 and Wreschner 1975, 1980for Marche slabs are also engraved on both sides- see
references. Pales & de St Pereuse 1981, p. 22.
10. Bordes 1952. 43. Omnes 1982, p. 185; Bahn 1983.
II. See note 9; Meszaros & Vertes 195 5. 44. Pequart 1960/3, p. 222; see also Pales & de St
12. Audouin & Plisson 1982. Pereuse 1981, p. 22, for the situation at La
13. Pequart 1960/3, pp. 211-14. Marche.
14. Bordes 1969; Marshack 1987, 1988. 45. Omnes 1982, p. 184.
15 . Leveque & Vandermeersch 1980. 46. Begouen & Clones 1981, p. 42; 1983; 1985.
16. Marshack 1987, 1988; Poplin 1983, p. 62. 47. Moure Romanillo 1985, p. 103.
17. Kozlowski 1982. 48. Rozoy 1985.
18. Hahn 1972; Soffer 1985 . 49. Bosinski 1973; 1984, p. 318.
19. Bader 1967 (1976). 50. Moure Romanillo 1985, p. 103.
20. Praslov 1985. 51. Pales & de St Pereuse 1979, p. 141; 1976, p. 17.
21. Taborin 1982, pp. 50/51. 52. Couraud 1985, 1982; Bahn & Couraud 1984;
22. Poplin 1983. Bahn 1984a.
23. Le Mort 1985; see also Begouen, Begouen & 53. Thevenin 1983; Couraud 1985.
Vall~ 1937. 54. Rozoy 1985.
24. Taborin 1977. 55 . Bosinski 1973, p. 41. See also Russell 1987, pp.
25. Taborin 1985; Desbrosse et al 1976. 206-9, for similar experiments.
26. Soffer 1985, pp. 373, 440; 1985a. 56. Delporte & Mons 1980.
27. Bahn 1977, pp. 252/3. 57. Begouen & Clones 1985, p. 45; Pales & de St
28. Taborin 1985 . Pereuse 1979, pp. 14112.
29. Bahn 1982, 1977. 58. For painted bones, see Gladkih et al1984, p. 142;
30. Bader 1967 (1976); Soffer (1985, p. 456) estimates Soffer 1985, pp. 78, 84; and Shimkin 1978, pp.
only 15 minutes per bead, and hence 2,500 hours 275, 283; for paste inside engraved lines, see
for the 3 burials together . Marshack 1981; 1979, p. 288.
208 IMAGES OF THE ICE AGE

59. Pequart 1939/40 , p . 451 ; Almagro 1976. For Le 87. Forlsturitz,seeStPerier l930,pp.109-16; 1936,
Mas d'Azil, see also Pales 1970 ; Delporte & Mons pp. 126-33; Mons 1986. For Bedei1hac, see Bahn
1975 . For a brief review of the topic , without & Cole 1986.
illustrations, see Sieveking 1983 . For plaquettes 88. St Perier 1936, pp. 123/5.
as a whole, see Sieveking 1987. 89. For a fine example, see Pales & de St Pereuse
60. de Sonneville-Bordes 1986, p . 633. 1979, pp. 118/19; on the topic as a whole, see
61. Almagro 1976, p. 68 note. Chollot-Varagnac 1980, p. 452.
62. Delporte & Mons 1980, p. 43 - this paper also 90. Arambourou 1978, pp. 116-24.
gives references to some of the earlier 91. Eppel1972, p. 78.
experimental work by these authors . For the 92. Hahn 1979, 1982.
nineteenth century experiments, see Leguay 93 . Delporte 1979.
1877, 1882. 94. de Saint Mathurin 1978.
63. For Pech de l'Aze, see Bordes 1969; Marshack 95. Breuil 1959, pp . 15-17 & 109; Breuil & Lantier
1977, pp. 289-92 ; 1976, p. 140; 1976a, pp . 278/9. 1959, pp. 193/4. In Breuil1909, p. 399, he states
For Bilzingsleben, see Marshack 1981, p. 189; that Piette's 'Vallinfernalien' layer ( = Gravettian)
Muller-Beck & Albrecht 1987, p. 68; and Mania was above the 'Eburneen ancien' layer containing
& Vlcek 1987, pp . 41, 43; and Mania in Rock Art the ivory sculptures, which must therefore be pre-
Research 5(2), 1988 (in press). Feustel (in Gravettian. Yet Del porte (1979, p. 221) claims
Muller-Beck & Albrecht 1987, p. 60) even claims that there is no evidence of any French statuette
that marks on one Bilzingsleben bone may be the being Aurignacian! Other Breuil references to
depiction of a large animal. this topic are cited by Pales 1972, p. 250.
64. Leonardi 1983. However, the recent excavations by Del porte at
65. Marshack 1976, p. 139; 1976a; Freeman & Brassempouy have found ivory fragments only in
Gonzalez Echegaray 1983. the Gravettian .
66. Marshack 1976, pp. 139-41; 1976a, p. 277. 96. Marshack 1984; 1976; 1976a; Basinski 1982;
67. Frolov 1981, pp. 63 , 77. Hahn 1984.
68 . Barandiaran 1971. .97 . Hahn 1971; Basinski 1982; Marshack 1987, 1988.
69. Breuil 1909, p. 380. 98. Dauvois 1977; for other observations on how
70. Bahn 1984, p. 116 . female statuettes of ivory and stone were carved,
71. Baulois 1980. incised and polished, see Delporte 1979, pp. 249-
72. Allain & Rigaud 1986; Praslov 1985 , p. 187. 51.
73. de Sonneville-Bordes 1986, pp . 634-6. 99. Jelinek 1975, p. 411.
74. Hellier 1981/2 , 1984. 100. Praslov 1985; Abramova 1967.
75. Barandiaran 1968 ; Sieveking 1971. 101. Tarassov 1971; Delporte 1979, p. 178.
76. Jelinek 1975, pp. 458/9. 102. Del porte 1979.
77. Passemard & Breui11928 . 103. Gamble 1982 .
78. St Perier 1930, p . 97; 1936, pp . 116-20; Bellier 104. Soffer 1985 ; G1adkih et a11984; Jelinek 1975.
1984, p. 26. 105 . Basinski 1984, p. 315; Basinski & Fischer 1980,
79. For the new Spanish specimens , see, for example, pp. 47, 99, 119 & pl. 101.
Fortea 1981; for Labastide , see Simonnet 1952; 106. Leroi-Gourhan 1978; for other examples of
Omnes 1982, p. 192. distortion to fit available space, see Pales 1976/7,
80. Bellier 1984, p . 30 . pp. 90/91 , and Pales & de St Pereuse 1976, p. 53,
81. For the French and Swiss type , see Bandi & and 1981, pp. 128-31.
Delporte 1984; for the relationship with antler 107. Berenguer 1986, p. 665.
shape, see Leroi-Gourhan 1971, Figs 34/35. Not 108 . Marshack 1987, 1988.
"111!1ll compositions of this type, however, were 109. Delluc & Delluc 1978, p. 393.
determined by the antler's shape- see Pales & de 110. Delluc & Delluc 1978, pp. 428, 391.
St Pereuse 1965 , pp . 227/8. 111. Delluc & Delluc 1984b; 1978, pp. 215-21.
82. Cattelain 1977/8. 112. Lejeune 1981 & in press; Eastham 1979, pp. 367-
83. Bahn 1982, pp. 257/8; 1984; Robert et a11953. 70; see Vezian 1956 for a variety of examples in
84. Pequan 1960/3, p. 299. one cave.
85. See Bahn 1978; Bahn & Otte 1985; Pe.quart 113. Bednarik 1986, pp. 44/5; see also Breuil1926, p.
1960/3, p. 219; Vasil'ev 1985. 366.
86. Klima 1984, 1983 , 1982.
 NOTES 209

114. Marshack 1977. For Gargas, see Barriere 1976; 146. For Baume-Latrone see L'Art des Cavernes 1984,
for Pech Merle, see L 'Art des Cavernes pp. pp. 333-9, and Bednarik 1986, p. 36; for Pileta,
470/7 1. see Dams 1978, pp. 86/7.
115. Bednarik 1986, pp . 43/4. 147. Marshack 1985, pp . 78, 81.
116. Delluc & Delluc 1983a; 1985, p. 60; Bednarik 148. Perpere 1984, p. 43.
1986. 149. Couraud 1982, p. 4; and in Leroi-Gourhan &
117. Vertut 1980. Allain 1979, p. 165 .
118. Delluc & Delluc 1983, pp . 58/9. 150. Leroi-Gourhan 1981, p. 30; 1982, p. 13.
119. Begouen & Clottes 1987, 1980. 151. Plenier 1971, pp. 69/70.
120. Allain & Rigaud , in Leroi-Gourhan & Allain 152. Marshack 1985, pp. 77-9.
1979, pp . 103, 106 . 153. Bohigas et all986; Moure Romanillo et all984/5;
121. Begouen 1925/6, p. 508 ; Bouillon, quoted in Almagro 1969.
Ucko & Rosenfeld 1967, p . 241. The idea was 154. Couraud 1982, p. 4; Barriere 1976, p. 75;
first put forward by Max Begouen, and used in his Lorblanchet 1980, p. 34.
prehistoric novel Les Bisons d'Argile (Fayard: ISS. Pfeiffer 1982, pp. 24112; Wildgoose et al1982.
Paris, 1925). 156. Leroi-Gourhan 1967. See also Moore 1977; Walsh
122. Eastham & Eastham 1979, pp . 375/6. On the 1979; Wright 1985 .
subject of light sources and visibility of 157. Lorblanchet 1980, pp. 35, 37; Pfeiffer 1982, pp.
engraving, see also Pales & de St Pereuse 1965, 24113 .
pp. 227/8. 158. Couraud 1982, p. 4. The experiments reported by
123. Begouen 1931. Wildgoose et al produced somewhat equivocal
124. A1magro 1981 , 1980, 1976 . results on this point and others.
125 . For Fontanet, see L 'Art des Cavernes 1984, p. 159. Sahly 1963; Barriere & Sahly 1964. The medical
435 ; for Montespan , see Rivenq 1976, 1984. view of the Gargas hands was presented earlier by
126. Breuil1952a, p. 12; Beltran et al1973. Janssens 1957. For a review of incomplete hand
127. Beltran et all967 . stencils, see Pradel 197 5.
128. Beasley 1986; Begouen , Clottes & Del porte 1977. 160. Lorblanchet 1980, pp. 37/8.
129. Begouen , Casteret & Capitan 1923 . 161. Couraud 1982, p. 5.
130. Roussot 1984. 162. Glory 1964; Couraud 1982, p. 5; and in Leroi-
131. Del porte 197 3. Gourhan & Allain 1979, p. 166.
132. Roussot 1984. 163 . Marshack 1985, p. 68.
133. Delluc & Delluc 1978, p. 389; a picture of the Cap 164. J. Begouen 1939, p. 284.
Blanc tools can be seen, though with no indication 165. Altuna & Apellaniz 1976.
of size, in White 1986, p. 135 . 166. Plenier 1971, pp. 69/70; de Balbin Behrmann &
134. Roussot 1981 ; Laming-Emperaire 1962, p. 186. Moure Romanillo 1982, pp. 62/3.
See also Simek 1986, p . 406 . 167. Aujoulat 1985; Pfeiffer 1982, p . 142 .
135. Regnault 1906, p. 332. 168. Rouzaud 1978, pp. 126/7.
136. Bader 1965, p . 28 ; Carciumaru & Bitiri 1983. For 169. Breuil & Lantier 1959, p . 223 .
a list of pigment analyses since 1950 (mostly at 170 . Sauvet 1983, p. 52.
Lascaux & Altamira) see Couraud 1983, p. 108. 171. For Roucadour, see L 'Art des Cavernes 1984, pp.
137. For Niaux, see Brunet 1981, 1982; for Las 511 -13; for Baume-Latrone, see Bednarik 1986,
Monedas, see Ripoll1972, p. 53. p . 34.
138. Judson 1959. 172. Delluc & Delluc, in Leroi-Gourhan & Allain
139. Couraud 1982, p. 4. 1979, pp. 175-84; Barriere & Sahly 1964, p. 179.
140. Couraud 1983, p. 107; 1983a, p. 6. 173. L'Art des Cavernes 1984, pp. 596-9.
14 . Couratlld & Laming-Emperaire, m Leroi- 174. de Beaune 1987, 1987a.
Gourhan & Allain 1979, pp . 15 3-70. 175 . de Beaune 1987, 1987a, 1984; Delluc & Delluc, in
142. Couraud 1982. Leroi-Gourhan & Allain 1979, pp. 121 -42;
143. Ballet et al in Leroi-Gourhan & Allain 1979, pp. Ruspoli 1987, pp. 28-30.
171-4. 176. Roussot 1984, p. 76.
144. Vandiver (in press). 177. de Beaune 1987, 1987a, 1984.
145. Couraud 1982, p. 4; and in Leroi-Gourhan & 178. Fortea 1981.
Allain 1979, pp. 162-4. Couraud's experiments 179. Martin Santamaria & Moure Romanillo 1981. For
thus confirmed the opinions ofRottlander (1965). La Griega, see Sauvet 1983.
210 IMAGES OF THE ICE AGE

180. Jorge et al1981, 1982 . 25. Leason 1939, 1956; Souriau 1971 likewise
181. Sacchi, Abelanet & Brule 1987; Bahn 1985. believes that the deer on the Lortet baton are
182. Omnes 1982, pp. 105-7. stretched dead on the ground, one with head bent
back, whereas most scholars have seen them as
galloping and bellowing.
6. What was Depicted? 26. Riddell 1940.
1. Examples cited in Bahn 1985a, from pp. 166, 27. Pales & de St Pereuse 1965, p. 229; 1981, p. 96;
283/284 and 328 of Bandi et a11984. Pales 1969, p. 115.
2. See Rosenfeld 1984, and Macintosh 1977; 28. Guthrie 1984, p. 46; Riddell1940, p. 160.
Macintosh compared his identifications with the 29. Vojkffy, quoted in Bandi 1968, p. 16; Bandi et al
answers later given by an Aboriginal informant, 1984, p . 29.
and found that out of 22 items he had been wrong 30. Vojkffy, quoted in Giedion 1965, p. 193.
about 15, and only superficially right about the 31. Bandi 1968, p. 17; Breuil & Obermaier 1935, p.
other 7! 31 and pl. 27; Leroi-Gourhan 1984, p. 77; see also
3. Cited by Clones 1986, p. 24. Jorda 1980, p. 282.
4. Examples given by Clones 1986, p. 10, and by 32. Baffier 1984, p. 153.
Pales 1969, p. 52. 33. Leroi-Gourhan 1984, p. 76.
5. Rousseau adopted 'valable' and 'probable' 34. Graziosi 1960, pp. 9112 and pl. 90; Barandiaran
categories in his 1967 study of felines, but Pales 1971a; 1973, pp. 187/8 and pl. 45, No . 1; Breuil
(1969, p. 52) disagreed with some of his 1936/7, p. 6; Bahn 1982, pp. 257/8. Three of the
assessments, finding him too ready to accept 'wolf/cervid' pieces were also studied in Sieveking
examples into both groups. 1978.
6. Pales 1969; Pales & de St Pereuse 1976, 1981, & 35. Delporte 1984, pp. 112-14; see also Caralp et al
forthcoming. 1973.
7. Lorblanchet 1977. 36. For a general survey of the fauna and its
8. Leroi-Gourhan 1983, p. 258. depictions, see Powers & Stringer 197 5.
9. Leroi-Gourhan 1982 , p . 49; 1983, p. 258. 37. Leroi-Gourhan 1971, p. 463.
10. Pales & de St Pereuse 1981, Nos 66/7, pls 84-7; 38. For example, Piette 1887.
1976, pl. 52. 39. Capitan, Breuil & Peyrony 1924; Bourdelle 1956;
11. Rousseau 1967, p. 76 ; 1974; Pales 1969, p. 64. Blanchard 1964; for a summary of the problems,
12. Lhote 1968; Leroi-Gourhan 1958b, pp. 520/1; for see Barandiaran 1972, p. 346; Pales & de St
examples of headless animals, see Leroi-Gourhan Pereuse 1981, p. 58; Prat 1986; and de
1971, p. 471; and Ripoll1972, pp. 56-60. Sonneville-Bordes & Laurent 1986/7, p. 72.
13. G., L. & R. Simonnet 1984, p. 29; also illustrated 40. Lion 1971; Altuna & Apellaniz 1978, pp. 107-9,
in Clones 1986. 113-19; Basinski & Fischer 1980; Pales & de St
14. Bahn & Cole 1986, p . 144 & Fig. 28. Pereuse 1981.
15. Lorblanchet 1974, p. 70; 1977, p. 46. 41. Pales & de St Pereuse 1981, pp. 26/7.
16. Bahn 1986, p. 117, & p. 107 (comment by 42 . Pales & de St Pereuse 1981, p. 38.
Clones); Davis 1986, p . 516. 43. Leroi-Gourhan 1984, pp. 76-8.
17. Ucko & Layton 1984. 44. Schmid 1984, p. 157.
18. Rousseau 1984a; Barandiaran 1972, pp. 351-8. 45. Altuna 1975,pp.112/3;Altuna&Apellaniz 1978;
19. Guthrie 1984, p. 38; Mithen (1988) prefers to see see also Barandiaran 1972, pp. 365-8; and Bahn
these features as teaching aids for hunters. 1982a, p. 23.
20. For Morin, see de Sonneville-Bordes 1986, p. 46. Rousseau 1973; also 1984a, pp. 187-9; Altuna &
632; for the fawns, see Lorblanchet in Marshack Apellaniz 1978; and Barandiaran 1972, p. 373 .
...,1985, pp. 62/3; for the Mas d'Azil bone disc, see 47 . Bahn 1982a, p. 21.
Chollot-Varagnac 1980, p. 408 (many scholars, 48 . Bahn 1983a.
including Graziosi 1960, Barandiaran 1968, 49. Arambourou 1978, p. 49; Bahn 1982a, pp. 22/3.
Leroi-Gourhan 1971 and Apellaniz 1982, 1986, 50. Piette 1906; Pales & de St Pereuse 1966, & 1981,
have wrongly attributed this piece to Laugerie- pl. 1; Bahn 1978a; Jorda 1987.
Basse). 51. Ziichner 1975, p. 23; for an earlier study of bison
21. See examples in Lhote 1968. figures, see Capitan, Breuil & Peyrony 1910.
22. Baffier 1984, pp . 143 , 150. 52. Pales & de St Pereuse 1981, pp. 63-106; 1965.
23. Baffier 1984. 53. de Lumley 1968a; for objections, see Pales & de St
24. Pales & de St Pereuse 1981, p. 87. Pereuse 1981, pp. 75-8.
 NOTES 211

54. Pales & de St Pereuse 1981, p. 72. 83. Leroi-Gourhan 1984, p. 84; Guthrie 1984, p. 44.
55 . Leroi-Gourhan 1984, pp . 77-80. 84. Klima 1984, p. 326.
56. It was primarily Breuil & Laming-Emperaire who 85. Pales 1969, pp. 53, 84, 102 .
said the small bovids were Bas longifrons; the 86. Mazak 1970; Rousseau 1974.
sexual dimorphism solution was put forward by 87. Pales 1969, pp. 58-60; Rousseau 1984a, p. 170.
Zeuner (1953) and Koby (1954) . 88. For general studies of rare species, see Capitan et
57. Koby 1968. all910, 1924; Novell986; for examples of studies
58. Barandiaran 1969; Ripoll 1972, & 1984, p. 272; (both critical and uncritical) of individual species,
see also Olivie 1984/S for cervids in Cantabrian see Barandiaran 1974 (glutton); Nougier &
art. Robert 1957a (rhino), 1958 (saiga), and 1960
59. Faure 1978, pp. 47 , 50, 76 . (canids).
60. Vialou 1984, and in Leroi-Gourhan & Allain 89. Capitan et al 1924; de Sonneville-Bordes &
1979. Laurent 1983; de Sonneville-Bordes 1986, pp .
61. Kehoe 1987 & in press; and letter to T.F . Kehoe 639-41; Bahn 1977, pp. 253/4; 1982 , p. 255;
from Danish zoologist M. Meldgaard , 30/3/87 . 1984.
Other possible examples of 'swimming' animals 90. Marshack 1970, 1972, 1975. For the whale
may include the deer on the Lortet baton, with theory, see Robineau 1984.
fish between their legs and 'above' their backs; 91. Dams 1987, p . 17; 1987a, pp . 219-23 .
and some horses at Lascaux are said to hold their 92. Breuil & de St Perier 1927.
heads just as if they were crossing a river 93 . Bahn 1984, pp. 153, 157,243-5, 276-8; Marshack
(according to Gunn, in Marshack 1979, p. 299). 1970, 1972 . For other features offish depictions ,
62. Blanchard 1964a. see Barandiaran 1972, pp. 348-51.
63. Vialou 1984, p. 214. 94. Dams 1987; 1978, p. 85.
64. Bouchud 1966. 95. Alcalde, Breuil & Sierra 1912, ch. XVI; Vayson
65 . Pales & de St Pereuse, forthcoming. de Pradenne 1934; Breuil & Begouen 1937;
66. Schmid, in Bandi et al1984, p. 90. Lorblanchet 1974, pp . 85-96.
67. Leroi-Gourhan 1984, pp . 81 , 84; Guthrie 1984, p. 96. Buisson & Pin<;on 1986/7, p. 79.
45 . 97. Eastham 1986.
68. de Sonneville-Bordes 1986, p. 633. 98. Bahn & Butlin 1987.
69. Pales 1976/7; Pales & de St Pereuse 1981, pp . 99 . Tyldesley & Bahn 1983; Delcourt-Vlaeminck
107-40. 1975; Marshack 1972, 1975 . .
70. Pales 1976/7, pp . 90/91; Pales & de St Pereuse 100 . Leroi-Gourhan 1971.
1981 , pp . 128-31. 101. Conkey 1981.
71. Pales 1976/7, pp. 91 , 10112; Pales & de St Pereuse 102. Sauvet 1979, pp . 342/3; Leroi-Gourhan 1971 ,
1981, pp . 132/3. Fig. 764; Vialou 1986.
72 . Pales & de St Pereuse 1981, pp. 137-9. For an 103. Nougier 1972, p. 265.
uncritical review of ibex figures in Pyrenean art, 104. Roussot 1984b. For the statistical problems
see Brielle 1968. involved in making cumulative percentage
73. Welte 1975/6. frequency graphs to compare assemblages
74. Pales & de St Pereuse, forthcoming . ranging from hundreds of figures to fewer than
75 . Bosinski 1984; Bosinski & Fischer 1980. For an ten, see J.E. Kerrich & D .L. Clarke, 'Notes on
earlier, less critical review of mammoths in art, the possible misuse and errors of cumulative
see Berdin 1970. percentage frequency graphs for the comparison
76. Bosinski 1984, p. 301; Bosinski & Fischer 1980, of prehistoric artefact assemblages', in Proc.
pp. 38-40; Hahn 1984, p. 288; Praslov 1985, p. Prehist. Soc. 33, 1967, pp. 91-116.
_ 186. . . 105. Delporte 1984, p . 132; Capitan & Peyrony 1928,
77. Ripoll1984, p. 278; Bandi 1968, p. 15. pp. 107-15 .
78. Welte 1975/6; Leroi-Gourhan 1984, p. 83; for 106. Bosinski 1984, p. 320; Bosinski & Fischer 1980,
Pech Merle, see Lorblanchet 1977, pp. 54/5. pp. 56-60, 124-6.
79. Jorda 1983. 107 . Ucko & Rosenfeld 1972, p. 156. For an early
80. Capitan et al 1924; Breuil , Nougier & Robert study of humans in the art, see Capitan et al1924,
1956; Nougier & Robert 1965, 1966. pp . 91-116.
81. Rousseau 1967, 1974; Pales 1969; Altuna 1972, 108. Macintosh 1977, p. 191. See also Moore 1977;
pp. 308-15 . Walsh 1979; Wright 1985.
82. Pales 1969, p. 102.
21 2 IMAGES OF THE ICE AGE

109. Pales & de St Pereuse 1976; 1981, p. 28; Airvaux 142. Abramova 1967; Praslov 1985, p. 190; Stoliar
& Pradel1984 . 1977/8, pp. 11112; 1977/8a, pp . 51-9, 73; Shimkin
110. Brodrick 1963 , pp. 126/7. 1978, p. 279. See also Rosenfeld 1977, p. 93 .
111. Bahn 1986. 143. Rousseau 1967, p . 158/9. On masks, see also
112. Pales & de St Pereuse 1976, p. 30; Airvaux & chapter in Cartailhac & Breuil 1906.
Pradel 1984. 144. Giedion 1965, p. 374.
113. Pales & de St Pereuse 1976, pp. 31 -4. 145. Ucko & Rosenfeld 1967; 1972, p. 205.
114. Delporte 1979, p. 27. 146. Begouen & Breuil 1958; Rousseau 1967, p. 102.
115 . Leroi-Gourhan 1971, Fig. 777; see Pales & de St 147. Leroi-Gourhan 1983 .
Pereuse 1976, p. 49 ; Ucko & Rosenfeld 1972, pp. 148. Pales & de St Pereuse 1976, p. 144.
194. 149. Bosinski 1984, p. 318; Bosinski & Fischer 1974,
116. de Saint Mathurin 1973 ; 1975, pp. 25/6. 1980.
117. Pales & de St Pereuse 1976, p. 156. 150. Conkey 1981, p. 24.
118. Pales & de St Pereuse 1976, pp . 83-5; Ucko & 151. For examples, see Jorda 1979, 1985 .
Rosenfeld 1972, pp. 18112. The abundance of 152. Lorblanchet 1977; Bahn 1986, p. 104.
humans with arms raised was noticed by 153 . See quotations from Forge, Munn and Warner, in
Cartailhac & Breuil1906, p. 52. Bahn 1986, pp. 104, 109. See also Speck &
119. Pales & de St Pereuse 1976, p. 52. Schaeffer 1950.
120. Pales & de St Pereuse 1976, pp. 28/9. 154 . And. Leroi-Gourhan 1980; Sauvet & Wlodarczyk
121. Out of a vast literature, see Delporte 1979; 1977, pp. 548-50; Casado 1977.
Passemard 1938; Narr 1960; Abramova 1967 . 155. Capdeville 1986; Vialou 1987; there is also a
122 . Pales 1972, p . 21 8. possible tectiform engraved on a bone tool from
123. Piette 1895, 1902; Ucko & Rosenfeld 1972, p. Altamira.
161. 156. Leroi-Gourhan 1958, 1958a, 1968.
124. Pales 1972, p. 220; Pales & de St Pereuse 1976, 157. Casado 1977; Sauvet & Wlodarczyk 1977.
pp. 9112, 32. 158 . Mons 1980/1; Chollot-Varagnac (1980) estab-
125. Rice 1981; see also Guthrie 1984, p. 62; the Rice lished 25 basic motifs.
paper is less useful when it compares its results to 159. Conkey 1980, 1981.
the supposed age structure of Upper Palaeolithic 160. Sauvet & Wlodarczyk 1977, p. 553.
adult females , since it has lumped together 161. eg see Piette 1905.
figurines spanning the entire period. 162. Forbes & Crowder 1979.
126. Leroi-Gourhan 1971, Fig. 52 bis; Pales 1972 ; 163. Sauvet & Wlodarczyk 1977, p. 552.
Pales & de St Pereuse 1976, pp. 70-74. See also 164. See, for example, Clegg 1985, 1986.
Delporte 1971; 1979, pp . 264-8. 165. Casado 1977, p. 251.
127. Delporte 1962, p. 58. 166. Leroi-Gourhan 1958, p. 318; 1971; Sauvet &
128. de Saint Mathurin 1978,1973. Wlodarczyk 1977, p. 557.
129. Abramova 1967; 1984, p . 333; Shimkin 1978, p. 167. Geoffroy 1974, p. 47.
278; Berenguer 1986, p . 668. 168. Geoffroy 1974, p. 57.
130. Praslov 1985 , pp . 182/3. 169. Gonzalez Garcia 1985, 1987; Geoffroy 1974, p.
131. Delporte 1979, p . 76; Gamble 1982, p . 98. 55.
132. Delluc 1981a, p . 72.
133. Breuil & Lantier 1959, p. 237; Pales & de St
Pereuse 1976, p . 101.
7. R eading the Messages
134. Carcauzon 1984, 1988. 1. For an account of Piette's view, see Pales 1969, p.
135 . Bosinski & Fischer 1974; Bosinski 1973. 111. On de Mortillet, see Reinach 1899; Bahn
136 ...j.osenfeld 1977; Lorblanchet & Welte 1987. 1988; Laming-Emperaire 1962, p. 70, and Ucko
137 . Bosinski & Fischer 1974, p. 45 & pl. 59. & Rosenfeld 1967, p. 118.
138. Bosinski & Fischer 1974, p. 94. 2. Quoted in Laming-Emperaire 1962, p. 66; and
139. Breuil 1905; Breuil & de St Perier 1927; Graziosi Pales 1969, p. 112.
1960, p. 196 and pl. 291; Leroi-Gourhan 1958a, 3. Halverson 1987; this paper also provides part of
p. 388; 1971 , p. 480, Fig. 792. the history of the theory . See also Souriau 1971.
140. Rosenfeld 1977, p. 101. 4. Chollot-Varagnac 1980, p . 448.
141. Ucko & Rosenfeld 1972, p. 168; Lorblanchet 5. Lewis-Williams 1982 , p. 429; Bahn 1987a.
1977. On claviforms, see also Stoliar 1977/8a, p. 6. Reinach 1903; see also Laming-Emperaire 1962,
64. pp. 72-5.
 NOTES 213

7. Cartailhac & Breuil 1906 (see especially pp. 143 46. Laming-Emperaire 1962, p. 219/20 & pl. 13;
and 225). Leroi-Gourhan 1958a, p. 396; 1966, p. 41 ; 1971 ,
8. Breuill952; Begouen 1929. p. 387, Figs . 491-3; Bouvier 1976.
9. Giedion 1965 , p. 57. 47. Freeman 1978; 1984, p. 214; but see Breuil &
10. Lips 1949, pp. 84/5; Lindner 1950, pp. 53-67. Obermaier 1935, pp. 85/6, who wisely leave the
11. Obermaier 1918. second animal undetermined.
12. Graziosi 1960, p. 152. On the Montespan case, 48. Freeman 1978; 1984, p. 222,
and other fallacious evidence for hunting and 49. Nougier & Robert 1974.
magic, see Bahn 1988a. 50. ].R.B. Speed, Vet.M.B., M.R .C. V.S., pers.
13. Pales 1969, p. 47. comm.
14. Leroi-Gourhan 1958a, p . 390; 1971, p. 30. 51. Leroi-Gourhan 1966, p. 39; 1971, pp . 89, 98;
15 . Faure 1978, p. 61. Bandi 1968, p. 16; Baffier 1984, pp. 148/9.
16. Faure 1978. 52. Glory 1968, p. 57; Rousseau 1984a, p. 195.
17. Barandianin 19 84. 53. Pales & de St Pereuse 1976, pp . 114-23.
18. Guthrie 1984, p. 51 , states wrongly that 54. Begouen et all982, 1984, 1984/5 pp. 66-70.
'portrayals of hunting scenes constitute the bulk 55. Cabre 1934.
of Palaeolithic large animal art'. Similarly, 56. L. Begouen 1939, pp. 293(4; Begouen & Clones
Mithen (1988) has interpreted all manner of 1984; Begouen et al1984!5, pp . 29-33 . A further
figures and shapes as examples of tracks, example of such wishful thinking is provided by a
droppings, and other visual 'cues' in an attempt to small limestone sculpture from Laussel which
see the art as a means of communicating seems to be the gland of a phallus; Lalanne
information about hunting. claimed in 1946 that it came from a carved
19. Guthrie 1984, p. 50. erection, but in fact this cannot be deduced from a
20 . Eaton 1978. gland alone (see Duhard & Roussot 1988 , p. 43).
21. Parkington 1969, p. 12. See Rice & Paterson 57 . St Perier 1936, p . 115.
1986, p. 665, for a similar suggestion . . 58. Rice 1981.
22. eg Breuil & Lantier 1959, pp . 237/8; Lindner 59. On Kostenki, Praslov 1985 , p. 185; on Tursac,
1950, p. 69; Marshack 1976b, pp. 7112. Delporte 1979, p. 76; on Grimaldi, Duhard 1987;
23. Lindner 1950, pp. 56/7 . on Monpazier, Duhard 1987a.
24. Kehoe 1987, in press . 60. Leroi-Gourhan 1971, p. 480 , Fig. 794; Bahn
25. Weissen-Szumlanska 1951 , pp. 457/8. 1986, pp . 109, 119.
26. Kehoe, in press . 61. Delluc 1978, p. 239; Stoliar 1977/8a, p. 42 ; Bahn
27 . Pales & de St Pereuse 1976, p. 78. 1986, p. 99.
28. eg Chollot-Legoux 1961. 62. Bahn 1986.
29. Delluc 1985, 198lb; Breuill925 . 63. Collins & Onians 1978; Delluc 1978, p. 353; and
30. Bahn 1986; Delluc 1985, see Bahn 1986, p. 102.
31. Basinski 1973, p. 45 ; 1984, p. 315. 64. Stoliar 1977/8a, p . 39.
32. Klima 1984, pp. 324-6. 65. For the phallus theory, see Begouen & Breuil
33. Delporte 1984, p. 132. 1958, p. 98, and Vialou 1986, p. 198; for the horn
34. de Sonneville-Bordes & Laurent 1986/7, pp. theory, Ucko & Rosenfeld 1967, pp. 57 & 178; for
69/70. the sculptor's view, Beasley 1986.
35. Delluc 1984c; 198la. 66. de Saint Mathurin 1978, p. 17 .
36. Altuna 1983; Altuna & Apellaniz 1978 , pp. 106-7. 67 . Bahn 1986, pp. 107, 117; Delluc 1985 , pp. 56/7,
37. Delporte 1984, p. 125. 61.
38. Roussot 1984b, pp. 495/6. 68. For illustrations, see Jelinek 1975 , pp. 406-10.
39._Rice ~aterson 1985 , 1986. 69. Guthrie 1984, pp. 62-6.
40. See for example Lommel 1967; Davenport & 70. Collins & Onians 1978.
Jochim 1988. 71. Guthrie 1984, pp. 70/1.
41. Glory 1968, pp . 37/8, 57. 72. Guthrie 1984 made the comparison; it was
42 . Smith, in press. adopted by Kurten 1986, from whom the
43 . Clark 1966, pp . 12/ 13. quotation is taken.
44. Lewis-Williams 1982; Lewis-Williams & Dowson 73 . Bahn 1986, 1985a; Gimbutas 1981.
1988. 74. Cartailhac 1902, p. 349.
45. Vialou 1986, p. 139.
214 IMAGES OF THE ICE AGE

75. Raphael 1986; Laming-Emperaire 1962; for 108 . Gonzalez Garda 1985, pp. 517-19.
Leroi-Gourhan, see especially 1971, 1982. Two 109. Rivenq, pers. comm.; Begouen 1984, p . 77.
collections of his papers have been published in 110. Gonzalez Garda 1985, pp. 457-63; Eastham 1979,
Spanish: 1984a has his major articles and pp . 370/71.
bibliography from 1935 to 1983; 1984b has his 111. Nougier 1975; 197Sa, p . 115.
lectures at the College de France from 1969 to 112. L'Art des Cavernes 1984, pp. 544-8.
1983 . A critical review of the development of his 113. Bahn 1978b, 1980.
views on Palaeolithic art can be found in Meylan 114. Jochim 1983.
1986, and a critique of his early work in Ucko & 115. Bahn 1984; Mellars 1985, p. 283.
Rosenfeld 1967, pp. 195-221. 116. Sieveking 1979a, p. 106, supports the former
76. Leroi-Gourhan 1982 , pp. 45-50. view; against it, see Bahn 1977, p. 251; 1978b, p.
77. Leroi-Gourhan 1958a, p. 395; 1971. 125; and Conkey 1983, pp. 206, 221.
78. Leroi-Gourhan 1958 , p . 321. 117. Lorblanchet 1980; and comment in Marshack
79. Leroi-Gourhan 1982 ; 1972, p. 285. 1985, p. 64 .
80. Leroi-Gourhan 1972, p. 307. 118. Nougier, eg 1972.
81. Jorda 1979, 1985. 119. Pales & de St Pereuse 1965, p . 230; Pales 1969,
82. Leroi-Gourhan 1971 , pp . 86, 322; 1972, p. 308 . pp. 110/11.
83. Parkington 1969, pp. 5/6; Ucko & Rosenfeld 120. Apellaniz 1982, pp. 38/9; Mons 1986/7, p. 91.
1967; Bandi 1972, p. 315 ; Lhote 1972, p. 322. 121. Almagro 1976, p. 71; Bosinski 1973, p. 43; 1984,
84. Meylan 1986, pp . 45/6. p. 274.
85. Leroi-Gourhan 1972, p. 291. 122. Breuill962, p. 356; for early views on Altamira,
86. Leroi-Gourhan 1982, p . SO . see Apellaniz 1982, p. 48; 1983, p. 274;
87. eg Altuna & Apellaniz 1976, p. 163; 1978; Jorda Cartailhac & Breuil 1906.
1980; Vialou 1986. 123. Leroi-Gourhan 1971, p. 29.
88. Leroi-Gourhan 1972 , p. 306; 1968. 124. Pales & de St Pereuse 1981, p. 139; see also
89. Leroi-Gourhan 1972 , p. 289. Baffier 1984, p. 153.
90. Leroi-Gourhan 1971 , p. 481. 125. Apellaniz 1980.
91. Lhote 1968a; 1972, pp . 324-8 . 126. Apellaniz 1982, pp. 45-63; 1983.
92. Leroi-Gourhan 1966, p . 44. 127. Apellaniz 1982, pp. 34-8; 1986, pp. S0-52 . It will
93. Leroi-Gourhan 1958a; 1968. On the La be recalled that, like other authors before him, he
Madeleine 'phallus', see Bahn 1986 pp. 103, 114, attributes the Mas d' Azil disc wrongly to
118. Laugerie-Basse, and that of Laugerie-Basse to
94. Breuil & Lantier 1959, p. 237. abri du Souci.
95 . Leroi-Gourhan 1982, p. 58; see also Bahn 1978b, 128 . Altuna & Apellaniz 1976, pp. 148-53 ; 1978, p.
1980. 141; Apellaniz 1982 .
96. Leroi-Gourhan 1971, p. 259; but see Ucko & 129. For Chimeneas, see Apellaniz 1982, pp. 63-8;
Rosenfeld 1967, pp. 195-221; 1972, pp . 177, 1984. For Lascaux, see 1984a .
193-7. 130. Altuna & Apellaniz 1978, pp. 125-33; Apellaniz
97. Leroi-Gourhan 1971, pp . 95/6; Pales & de St 1982, pp. 71-92; Arte Rupestre 1987, p. 40.
Pereuse 1976, pp. 49 , 154. 131. Sauvet 1983, pp. 56/7; but see reply by Apellaniz
98. Pales & de St Pereuse 1976, pp. 153-5. in Ars Praehistorica 314, 1984/5, pp. 259-60.
99. Pales & de St Pereuse 1976, p. 27; 1981, p. 28; for 132 . Apellaniz 1984, pp. 534-7; 1987.
other examples of associations in Palaeolithic art, 133. Bahn 1982, 1984; Conkey 1980; Soffer 1985.
see Nougier & Robert 1968, Caralp et all973. 134. Breuil 1952, pp. 194/5; Leroi-Gourhan 1968, p.
100. Sauvet 1979, p . 347. 69; Casado 1977, p. 16.
101~auvet 1979, p. 348. 135. Leroi-Gourhan 1966, p. 47.
102. Parkington 1969 . For a different statistical test, 136. Eastham 1979, pp. 378-84.
see U cko & Rosenfeld 1967, p. 208 . 137. For Limeuil see Capitan & Bouyssonie 1924; for
103. Stevens 1975, 197Sa . Mezhirich, see Gladkih et al 1984, p. 141;
104. Laming-Emperaire 1962 , pp. 115-23 . On Shimkin 1978, pp. 274, 283 ; Soffer 1985 , p. 79.
Raphael, see pp. 118/9; see also Raphael1986. 138. Marshack 1979, pp. 287-92.
105. Laming-Emperaire 1972, pp. 66/7 . 139. Dewez 1974.
106. Laming-Emperaire 1970; 1971; 1972, p. 70. 140. Frolov 1977/9; 1981; and in Marshack 1979, pp.
107. Gonzalez Garda 1987; 1985, p. 490. 605-7.
 NOTES 215

141. Marshack 1979, pp. 271 , 309, 607; 1972a, p. 329. 154. Pfeiffer 1982 .
142. Marshack 1970a, 1972, 1972a , 1975. See also 155. Yates 1966; Pfeiffer 1982, pp . 210-25 .
review by Rosenfeld in An tiq uity 45, 1971 , pp . 156. Sauvet 1979.
317-19, and reply in 46, 1972 , pp . 63-5 . 157. Conkey 1978, 1980, 1985 .
143. Couraud 1985; Bahn & Couraud 1984. 158. Conkey 1980.
144. Frolov 1977/9; 1979; 1981. 159. Gamble 1982, p. 104.
145. Guthrie 1984, p. 47; Marshack 1970, 1972, 1975; 160. Conkey 1983, pp. 218-19; 1984, p . 264 .
but see Faure 1978 , Bahn 1986. 161. Hammond 1974.
146. Marshack 1977; 1979, p. 305. 162. Bednarik 1986; Davis 1986.
147. Marshack 1985, pp . 70, 78-81. 163. Lewis-Williams & Dowson 1988.
148. Marshack 1969, 1972, 1985. 164. Comment by Bahn, in Lewis-Williams & Dowson
149. Lorblanchet 1980a, p . 476; and in Marshack 1988, pp. 217/8.
1985, pp. 62/3. 165. Lewis-Williams 1982; Lewis-Williams & Dawson
ISO. Leroi-Gourhan 1981 , pp. 25, 32; Lorblanchet 1988.
1980a, pp. 474, 476. 166. Lewis-Williams 1982, p. 434.
151. Marshack 1984, 1985. 167. Pericot 1962; Bahn 1986, pp. 110, 119.
152. Collot et al1982. 168. Forbes & Crowder 1979, p. 362 & plate.
153. Pfeiffer 1982.
 Bibliography

This is by no means intended to be an exhaustive list of AIRY A UX, J. ec al 1983 . La plaquette gravee du
references on the subject of Palaeolithic art: instead it Perigordien superieur de l'abri Laraux, commune de
concentrates on recent studies. Extensive Lussac-les-Chateaux (Vienne). Nouvelle lecture et
bibliographies covering the older literature, and also comparaisons. Bull. Soc. Prehisc. fram;aise 80, 235-
concerning individual sites, can be found in: Breuil 46.
1952; Zervos 1959; Graziosi 1960; Laming-Emperaire ALCALDE DEL RIO, H., BREUIL, H. & SIERRA,
1962; Leroi-Gourhan 1971 ; Naber er a11976; and L'Art L. 1912. Les Cavernes de fa Region Canrabrique.
des Cavernes 1984. Monaco.
The most up-to-date guide to the few French ALLAIN, J. 1950. Un appeau magdalenien. Bull. Soc.
decorated caves open to the public is Vialou 1976 Prehisc. fram;aise 47, 181-92.
(though it predates Lascaux II); L'Art des Cavernes ALLAIN, J. & RIGAUD, A. 1986. Decor et fonction.
also has some information on opening hours; the only Quelques exemples tires du Magdalenien.
equivalent for northern Spain, though now very L'Anrhropologie 90, 713-38.
outdated, is Sieveking & Sieveking 1962. ALMAGRO BASCH, M. 1969. Las Pinruras
Rupescres deJa cueva de Malcravieso en Oiceres.
ABRAMOVA, Z.A. 1967 . Palaeolithic art in the Min. de Educaci6n y Ciencia: Madrid.
USSR. Arctic Anthropology4 (2), 1-179. Also 'L'art ALMAGRO BASCH, M. 1976. Los omoplatos
mobilier paleolithique en URSS', in Quartar 18, decorados de Ia cueva de 'El Castillo', Puente Viesgo
1967, 99-125, 9 pl. (Santander). Trabajos de Prehiscoria 33, 9-99, 12 pl.
ABRAMOVA, Z.A . 1984. Les correlations entre l'art ALMAGRO BASCH, M. 1980. Los grabados de trazo
et la faune dans le Paleolithique delaPlaine russe (La multiple en el arte cuaternario esp:u1ol, in Alramira
femme et le mammouth), in Bandi er aJ1984, 333-42. Symposium, 27-71. Min. de Cultura: Madrid.
ABSOLON, C. 1937. Les fl utes paleolithiques de ALMAGRO BASCH, M . 1981 . La tecnica del grabado
1' Aurignacien et du Magdalenien de Mora vie de trazos multiples en el arte cuaternario espaii.ol, in
(analyse musicale et ethnologique comparative, avec Arce Paleolicico, Comisi6n XI, Xth Congress
demonstrations). 12e Congres Prehist. de France, UISPP, Mexico City, 1-52.
~ulouse/Foix 1936, 770-84. ALMAGRO BASCH, M., GARCIA GUINEA, M.A.
ADAM, K.D. & KURZ , R . 1980. Eiszeickunst im & BERENGUER, M. 1972. La epoca de las pinturas
siiddeutschen Raum. Konrad Theiss Verlag: y esculturas policromas cuaternarias en relaci6n con
Stuttgart. los yacimientos de las cuevas: revalorizaci6n del
AIKENS, C.M. & HIGUCHI , T. 1982. Prehiscoryof Magdaleniense III. in Sanrander Symposium 467-
japan. Academic Press : London/New York. 73.
AIRVAUX, J. & PRADEL, L. 1984. Gravure d'une ALTUNA, J. 1972. Fauna de mamiferos de los
tete humaine de face dans le Magdalenien III de La yacimientos prehist6ricos de Guipuzcoa . Munibe
Marche, commune de Lussac-les-Chateaux 24, 464pp.
(Vienne). Bull. Soc. Prehist. fram;:aise81, 212-15.
 BIBLIOGRAPHY 217

ALTUNA, J. 1975. Lehen euskal herria - Guide ARTE RUPESTRE EN ESPANA 1987. Revista de
illustre de prehistoire basque. Mensajero: Bilbao. Arqueologia, numero extra.
ALTUNA, J. 1983 . On the relationship between ASLIN, G.D. & BEDNARIK, R.G. 1984a. Karlie-
archaeofaunas and parietal art in the caves of the ngoinpool Cave: A preliminary report. Rock Art
Cantabrian region, in Animals and Archaeology, I: Research 1 (1), 36-45.
Hunters and their Prey (J. Clutton-Brock & C. ASLIN, G.D. & BEDNARIK, R.G. 1984b. Koorine
Grigson , eds), 227-38. British Arch. Reports, Int. Cave, South Australia. Rock Art Research 1 (2), 142-
series 163, Oxford . 44.
ALTUNA, J. & APELLANIZ, J.M. 1976. Las figuras ASLIN, G.D., BEDNARIK, E.K. & BEDNARIK,
paleoliticas de la cueva de Altxerri (Guipuzcoa). R.G. 1985. The 'Parietal Markings Project' - a
Munibe 28, 3-242. progress report. Rock Art Research 2 (1), 71-74.
ALTUNA, }. & APELLANIZ, J .M. 1978 . Las figuras ASLIN, G.D. & BEDNARIK, R.G. 1985. Mooraa
rupestres de Ia cueva de Ekain (Deva). Munibe 30, Cave - a preliminary report. Rock Art Research 2
1-151. (2), 160-65.
ANA TI, E. 1985. The rock art of Tanzania and the East ATLAS OF PRIMITIVE MAN IN CHINA 1980.
African sequence. Bollettino del Centro Camuno di Science Press: Beijing.
Studi Preistorici 23, 15-68 . AUDOUIN, F. & PLISSON, H. 1982. Les ocres et
ANATI, E. 1986. Etat de la recherche sur !'art leurs temoins au Paleolithique en France: enquete et
rupestre: rapport mondial. L 'Anthropologie 90, experiences sur leur validite archeologique . Cahiers
783-800. du Centre de Recherches Prehistoriques 8, 33-80.
ANDRIEUX, C. 1974. Premiers resultats sur !'etude AUJOULAT, N. 1985. Analyse d'une oeuvre parietale
du climat de la salle des peintures de Ia Galerie anamorphosee. Bull. Soc. Prehist. Ariege 40, 185-
Clastres (Niaux, Ariege) . Annales de Speleologie 29, 93.
3-25 . AUJOULAT, N. 1987. Le Releve des Oeuvres
APELLANIZ, J .M. 1980. El metodo de determinacion Parietales Paleolithiques. Enregistrement et
de autor en el Cantabrica. Los grabadores de Llonin, Traitement des Donnees. Documents d' Archeologie
in Altamira Symposium, 73-84. Min. de Cultura: fran<;aise No. 9. Maison des Sciences de !'Homme:
Madrid. Paris.
APELLANIZ, J .M. 1982. El Arte Prehist6rico del Pais AUJOULAT, N., ROUSSOT, A. & RIGAUD, J-P.
Vasco y sus Vecinos. Desclee de Brouwer: Bilbao. 1984. Traces peu explicites et attributions douteuses
APELLANIZ, }.M. 1983. El autor de los bisontes ou erronees, in L'Art des Cavernes, 72-85. Min. de
tumbados del techo de los policromos de Altamira, Culture: Paris.
in Homenajeal Prof. M. AlmagroBasch, vol. 1, 273- AVELEYRA ARROYO DE ANDA, L. 1965. The
80. Min. de Cultura: Madrid. Pleistocene carved bone from Tequixquiac, Mexico:
APELLANIZ , }.M . 1984. La methode de A reappraisal. American Antiquity 30, 261-77.
determination d'auteur appliquee a !'art parietal
paleolithique. L'auteur des cervides a silhouette BADER, O.N. 1965. La Caverne Kapovaia, Peinture
noire de Las Chimeneas (Santander, Espagne). Paleolithique. Nauka: Moscow.
L'Anthropologie 88, 531-37 . BADER, O.N. 1967. Die palaolithischen
APELLANIZ , }.M. 1984a. L'auteur des grands Hohlenmalereien in Osteuropa. Quartiir 18, 127-38.
taureaux de Lascaux et ses successeurs. Ibid. 539-61. BADER, O.N. 1976. Upper Palaeolithic Burials and
APELLANIZ, }.M. 1986. Analisis de Ia variaci6n the Grave at the Sungir Site. Sovetskaia Arkheologiia
formal y Ia autoria en la iconografia mueble del 3, 1967, 142-59, translated for the U.S. Geological
Magdaleniense Antigua de Bolinkoba (Vizcaya). Survey.
Munibe 38, 39-59 . BAFFlER, D. 1984. Les caracteres sexuels secondaires
APELLANJ.l, }.M. 1987. Aplicaci6n de tecnicas des mammiferes dans !'art parietal paleolithique
estadisticas al analisis iconografico y al metodo de franco-cantabrique, in Bandi et a/1984, 143-54.
determinacion de autor. Munibe 39, 39-60. BAHN, P.G. 1977 . Seasonal migration in S.W. France
ARAMBOUROU, R. 1978. Le Gisement during the late glacial period. journal of Arch.
Prehistorique de Duruthy a Sorde-l'Abbaye Science 4, 245-57.
(Landes). Bilan des Recherches de 1958 a 1975. BAHN, P. G. 1978. Palaeolithic pottery, the history of
Memoire 13 de la Soc. Prehist. fr., 158pp. an anomaly. Anthropos (Athens), 5, 98-110.
L'ART DES CAVERNES 1984. Atlas des Grottes BAHN, P.G. 1978a. The 'unacceptable face' of the
Ornees Paleolithiques fram;aises. Min. de Ia West European Upper Palaeolithic. Antiquity 52,
Culture: Paris . 183-92' 1 pl.
218 IMAGES OF THE ICE AGE

BAHN, P.G. 1978b. Water mythology and the de BALBIN BEHRMANN, R. & MOURE
distribution of Palaeolithic parietal art. Proc. ROMANILLO, }.A. 1982. El panel principal de Ia
Prehist. Soc. 44, 125-34. cueva de Tito Bustillo (Ribadesella, Asturias). Ars
BAHN, P.G. 1980. 'Histoire d'Eau' : L'art parietal Praehistorica 1, 47-97.
prehistorique des Pyrenees. Travaux de l'Inst. d 'Art BANDI, H-G. 1968. Art quaternaire et zoologie, in
Prehist. Toulouse 22 , 129-35 . Simposio de Arte Rupestre, Barcelona 1966, 13-19.
BAHN, P.G. 1982. Inter-site and inter-regional links BANDI, H-G. 1972. Quelques reflexions sur Ia
during the Upper Palaeolithic: the Pyrenean nouvelle hypothese de A. Leroi-Gourhan concernant
evidence. The Oxford journal of Arch. 1, 247-68. Ia signification de !'art quaternaire, in Santander
BAHN, P.G. 1982a. Homme et cheval dans le Symposium, 309-19.
Quaternaire des Pays de 1' Adour, in Les Pays de BANDI, H-G. & DELPORTE, H. 1984. Propulseurs
l'Adour, Royaume du Cheval, Guide-Catalogue, magdaleniens decores en France et en Suisse. in
Musee Pyreneen, Lourdes, pp. 21-26. Elements de Pre et Protohistoire europeenne,
BAHN, P.G. 1983. A Palaeolithic treasure-house in the Hommages a]-P. Millotte, Annales Litt. de l'Univ.
Pyrenees. Nature 302, 571-72. de Besan<;on, 203-17.
BAHN, P.G. 1983a. New finds at Pincevent. Nature BANDI, H -G. et al (eds) 1984. La Contribution deJa
304, 682-83 . Zoologie et de l'Ethologie aJ'interpretation de J'art
BAHN, P.G. 1984. Pyrenean Prehistory. Aris & des peuples chasseurs prehistoriques. 3e colloque de
Phillips: Warminster. Ia Soc. suisse des Sciences Humaines, Sigriswil
BAHN, P.G. 1984a. How to spot a fake azilian pebble. 1979. Editions Universitaires: Fribourg.
Nature 308, p. 229. BARANDIARAN, I. 1968. Rodetes paleoliticos de
BAHN, P.G. 1985. Ice Age drawings on open rock hueso. Ampurias 30, 1-37 .
faces in the Pyrenees. Nature 313, 530-31. BARAND IARAN, I. 1969. Representaciones de reno
BAHN, P.G. 1985a. Review of Bandi et al 1984. en el arte paleolitico espaiiol. Pyrenae 5, 1-33 .
Antiquity 59, 57-58. BARANDIARAN, I. 1971. Hueso con grabados
BAHN, P.G. 1986. No sex, please, we're paleoliticos en Torre (Oyarzun, Guipuzcoa).
Aurignacians. Rock Art Research 3, 99-120. Munibe 23, 37-69 .
BAHN, P.G. 1987. A Ia recherche de l'iconographie BARANDIARAN, I. 1971a. 'Bramaderas' en el
paleolithique hors de !'Europe. Travaux de J'Institut Paleolitico superior peninsular. Pyrenae7 , 7-18.
d'Art Prehist. de Toulouse 29, 7-18. BARANDIARAN, I. 1972. Algunas convenciones de
BAHN, P.G . 1987a. Comment on article by J. representaci6n en las figuras animales del arte
Halverson. Current Anthropology 28 , 72-3. paleolitico, in Santander Symposium 345-84.
BAHN, P.G. 1987b. Excavation of a Palaeolithic plank BARANDIARAN, I. 1973. Arte Mueble del
from Japan. Nature 329, p. 110. Paleolitico Cantabrica. Monog. Arq. 14: Zaragoza.
BAHN, P.G . 1988. Expecting the Spanish Inquisition: BARANDIARAN, I. 1974. El Glot6n (Gulo gulo L. )
the rejection of Altamira in its 19th century context. en el arte paleolitico. Zephyrus 25, 177-96.
Paper for the 5th Int. Conference on Hunting and BARANDIARAN, I. 1984. Signos asociadas a hocicos
Gathering Societies, Darwin, September 1988. de animales en el arte paleolitico. Veleia 1, 7-24.
BAHN, P.G. 1988a. Where's the beef? The myth of BARRIERE, C. 1976. L'Art Parietal de la Grotte de
hunting magic in Palaeolithic art . Paper for the 1st Gargas. Memoire III de l'Inst. Art Prehist.
AURA (Australian Rock Art Research Association) Toulouse; British Arch . Reports (Oxford), Int.
Congress, Darwin, September 1988. series No. 14, 2 vols.
BAHN, P.G. & BUTLIN, R.K. 1987. Les insectes BARRIERE, C. & SAHLY, A. 1964. Les empreintes
dans l'art paleolithique: quelques observations humaines de Lascaux, in MisceJanea en Homenajeal
nouvelles sur Ia sauterelle d'Enlene (Ariege). Actes Abate H. Breuil, vol. 1, pp. 173-80. Barcelona.
~Colloque Int. d'Art Mobilier Paleolithique, Foix! BASLER, D. 1979. Le Paleolithique final en
Le Mas d'Azil (in press) . Herzegovine . in La Fin des Temps Glaciaires en
BAHN, P.G. & COLE, G. 1986. La prehistoire Europe (D . de Sonneville-Bordes, ed.), vol. 1, 345-
pyreneenne aux Etats-Unis. Bull. Soc. Preh. Ariege- 55. C.N.R.S.: Paris .
Pyrenees 41, 95-149. BAULOIS, A. 1980. Les sagaies decorees du
BAHN, P.G. & COURAUD , C. 1984. Azilian pebbles Paleolithique superieur dans Ia zone franco-
-an unsolved mystery . Endeavour 8, 156-58. cantabrique. Bull. Soc. Prehist. Ariege 35, 125-28.
BAHN, P.G. & OTTE, M . 1985 . La poterie BEASLEY, B. 1986. Les bisons d'argile de Ia grotte du
'paleolithique' de Belgique: analyses recentes. Tuc d' Audoubert. Bull. Soc. Prehist. Ariege-
Helinium 25, 238-41. Pyrenees 41, 23-30.
 BIBLIOGRAPHY 219

de BEAUNE , S.A. 1984. Comment s'eclairaient les BEGOUEN, R. 1984. Les bisons d'argile du Tuc
hommes prehistoriques? La R echerche 15, No. 152, d' Audoubert. Dossiers de l'Archeologie 87, octobre,
247-49. 'Les Premiers Artistes', 77-79.
de BEAUNE, S.A. 1987. Lampes et Godets au BEGOUEN, R. & CLOTTES, }. 1979. Le baton au
Paleolithique. XXIIIe Supplement a Gallia saumon d'Enlene. Bull. Soc. Prehist. Ariege 34, 17-
Prehistoire. 25.
de BEAUNE, S.A. 1987a. Palaeolithic lamps and their BEGOUEN, R. & CLOTTES, J. 1979a. Galet grave de
specialization: a hypothesis . Current Anth. 28, 569- la caverne d'Enlene a Montesquieu-Avantes
77. (Ariege). Caesaraugusta 49/50, 57-64.
BEDNARIK, R.G. 1986. Parietal finger markings in BEGOUEN, R. & CLOTTES, J. 1980. Apports
Europe and Australia . Rock Art Research 3, 30-61 & mobiliers dans les cavernes du Volp (Enlene, Les
159-70. Trois-Freres, Le Tuc d'Audoubert), in Altamira
BEGOUEN, H. 1925/6. Observations nouvelles dans Symposium, 157-88. Min. de Cultura: Madrid.
les grottes des Pyrenees, in M elanges Gorgianovitch- BEGOUEN, R. & CLOTTES, }. 1981. Nouvelles
Kramberger, 501-9. Zagreb. fouilles dans la Salle des Morts de la Caverne
BEGOUEN, H. 1929. The magic origin of prehistoric d'Enlene a Montesquieu-Avantes (Ariege), in 21e
art. Antiquity 3, 5-19. Congres Prchist. de J-<rance, Montauban!Cahors
BEGOUEN, H. 1931. La technique des gravures 1979, vol. 1, 33-56.
parieta1es de quelques grottes pyreneennes. XVe BEGOUEN, R. & CLOTTES, J. 1983. El arte
Congres Int. d'Anth. & d 'Arch. Prehistorique, mobiliar de las cavernas del Volp (en Montesquieu-
Portugal1930, 8pp, 13 pl. Avantes/Ariege). Revista de Arqueologia Afio IV,
BEGOUEN, H . 1932. La Prehistoire a la Societe No. 27, 6-17.
Archeologique du Midi de la France. Memoires deJa BEGOUEN, R. & CLOTTES, }. 1984. Un cas
Soc. Arch. du Midi deJa France, 1-11. d'erotisme prehistorique. La Recherche 15, 992-95.
BEGOUEN, H. 1942. De la Lecture des Gravures BEGOUEN, R. & CLOTTES, ]. 1985. L'art mobilier
Prehistoriques. Conseils am es Etudiants. Editions des Magdah~niens. Archeologia 207, novembre, 40-
du Museum: Toulouse. 10pp. 49.
BEGOUEN, H. 1947. E1oge de M . Emile Cartailhac. BEGOUEN, R. & CLOTTES, }. 1986/7. Legrand
Bull. Soc. Arch. du Midi de Ja France, 3e serie, 5 felin des Trois-Freres. Antiquites Nationales 18/19,
(1942-5), 438-51. 109-13.
BEGOUEN, H . 1952. Seance du 24 juillet. Bull. Soc. BEGOUEN, R. & CLOTTES, J. 1987. Les Trois-
Preh. fr. 49, 289-91. Freres after Breuil. Antiquity 61, 180-87.
BEGOUEN, H. & L. 1934. Quelques plaquettes de BEGOUEN, R., CLOTTES, J. & DELPORTE, H.
pierre gravees ou peintes des cavernes pyreneennes. 1977. Le retour du petit bison au Tuc d'Audoubert.
lle Congres Prehist. de France, Perigueux, 3pp. Bull. Soc. Prehist. fr. 74, 112-20.
BEGOUEN, H., BEGOUEN, L. & VALLOIS, H. BEGOUEN, R. et a11982. Plaquette gravee d'Enlene,
1937. Une pendeloque faite d'un fragment de Montesquieu-Avantes (Ariege). Ibid. 79, 103-12.
mandibu1e humaine (Epoque magdalenienne). 12e BEGOUEN, R. et al1984. Complements ala grande
Congres Prehist. de France, Toulouse-Foix 1936, plaquette gravee d'En1ene. Ibid. 81, 142-48.
559-64. BEGOUEN, R. et al1984/5. Art mobilier sur support
BEGOUEN, H. & BREUIL, H. 1958. Les Cavernes lithique d'Enlene (Montesquieu-Avantes, Ariege),
du Volp: Trois-Freres- Tuc d 'Audoubert. Arts et Collection Begouen du Musee de l'Homme. Ars
Metiers Graphiques: Paris. Praehistorica III/IV, 25-80.
BEGOUEN, H., CASTERET, N . & CAPITAN, L. BELFER-COHEN, A. & BAR-YOSEF, 0 . 1981. The
1923. La caverne de Montespan (Haute-Garonne). Aurignacian at Hayonim Cave. Paleorient 7 (2), 19-
Revue 4,;Jth. 33, 1-18, 1 pl. 42.
BEGOUEN, ]. 1939. De quelques signes graves et BELLIER, C. 198112. Contribution aJ'etude de l'art
peints des grottes de Montesquieu-Avantes, in paJeolithique en Europe occidentale, le contour
Melanges Begouen, 281-87. Toulouse . decoupe en os. Memoire de licence en histoire de
BEGOUEN, L. 1939. Pierres gravees et peintes de l'art et arch., Univ. libre de Bruxelles, 2 vols.
l'epoque magdalenienne, in Melanges Begouen, BELLIER, C. 1984. Contribution a l'etude de
289-305. Toulouse. l'industrie osseuse prehistorique: les contours
BEGOUEN, R . 1980. La conservation des cavernes du decoupes du type 'tetes d'herbivores'. Bull. Soc.
Volp. Son histoire, son bi1an. in Altamira royale beige Anth. Prehist. 95, 21-34.
Symposium, 681-93. Min. de Cultura: Madrid.
220 IMAGES OF THE ICE AGE

BELTRAN, A., GAILLI , R. & ROBERT, R. 1973. BOURDELLE, E. 1956. Les parentes morphologiques
La Cueva de Niaux . Monograf. Arq. No. 16, des Equides caballins d'apres les gravures rupestres
Zaragoza. du Sud-Ouest de la France. Mammalia 20, 22-23.
BELTRAN, A., ROB ERT, R. & GAILLI, R. 1967. BOUVIER, J-M. 1976. La Chaire aCalvin, Mouthiers
La Cueva de Bedeilhac. Monograf. Arq. No. 2, (Charente) . Donnees et problemes, in Sud-Ouesc
Zaragoza. (Aquitaine et Charente), Livret-Guide de
BERDIN, M.O. 1970. La repartition des mammouths !'Excursion A4, U.I.S.P.P. Nice, 133-36.
dans l'art parietal quaternaire. Travaux Inst. Art BOUVIER, J-M. 1977. Un Gisement Prehistorique:
Prehist. Toulouse 12, 181-36 7. La Madeleine. Fanlac: Perigueux.
BERENGUER ALONSO, M. 1986. Art parietal BOYLE, M. et al 1963. Recollections of the Abbe
paleolithique occidental. Techniques d'expression et Breuil. Antiquity 37, 12-18.
identification chronologique. L'Anthropologie 90, BREUIL, H. 1905. La degenerescence des figures
665-77. d'animaux en motifs ornementaux a l'epoque du
BIBIKOV, S. 1975 . A Stone Age orchestra. UNESCO Renne. C.r. Acad. Inscr. Belles-Lettres, 105-20.
Courier, June, 8-15 . BREUIL, H. 1909. L'evolution de !'art quaternaire et
BLANCHARD, ]. 1964 . Informations recherchees les travaux d'Edouard Piette. Revue Arch. 4e serie,
d'apres les equides europeens figures, in Prehistoric 13' 378-411.
Art ofthe Western M editerranean and the Sahara(L. BREUIL, H. 1912. L'age des cavernes et roches ornees
Pericot & E. Ripoll , eds) 3-34. Viking Fund de France et d'Espagne. Revue arch. 19, 193-234.
Publications in Anth . No. 39, New York. BREUIL, H. 1925. Les origines de !'art. journal de
BLANCHARD,] . 1964a. Selection intentionnelle des Psychologie 22, 289-96.
belles tetes de cerfs gravees et peintes, in Miscel:inea BREUIL. H. 1926. Les origines de !'art decoratif. Ibid.
en Homenaje al abate H. Breuil, vol. I, 249-58. 23, 364-75.
Barcelona. BREUIL, H. 1936/7. De quelques oeuvres d'art
BOHIGAS ROLDAN, R . et al1986 . Informe sabre el magdaleniennes inedites ou peu connues. I.P.E.K.
santuario rupestre paleolitico de Ia Fuente del Salin 11, 1-16.
(Mufiorrodero , Val de San Vicente, Cantabria). Bal. BREUIL, H. 1949. Les fresques de Ia Gal erie Vidal aIa
Cantabro de Espeleologia No.7, diciembre, 81-98, I Caverne de Bedeilhac (Ariege). Bull. Soc. Prehist.
pl. Ariege4, 11-16, 11 pl.
BORDES, F . 1952. Sur !' usage probable de la peinture BREUIL, H. 1952. Four Hundred Centuries of Cave
corporelle dans certaines tribus mousteriennes. Art. Centre d'Etudes et de Documentation
Bull. Soc. Prehist. fran9aise 49, 169-71. Prehistoriques: Montignac.
BORDES, F. 1969. Os perce mousterien et os grave BREUIL, H. 1952a. La Caverne de Niaux.
acheuleen du Pech de l'Aze II. Quaternaria 11, 1-6. Complements inedits sur sa decoration. Bull. Soc.
BOSINSKI, G. 1973. Le site magdalenien de Prehist. Ariege 7, 11-35.
Gonnersdorf (Commune de Neuwied, Vallee du BREUIL, H. 1959. Notre art de l'Epoque du Renne: sa
Rhin Moyen, RFA). Bull. Soc. Prehist. Ariege 28, repartition, sa succession, ses musees. in Zervos
25-48. 1959, pp. 13-20 & 109.
BOSINSKI , G. 1982. Die Kunst der Eiszeit in BREUIL, H. 1962. Theories et faits cantabriques
Deutschland und in der Schweiz. Habelt: Bonn. relatifs au Paleolithique superieur et a son art des
BOSINSKI , G. 1984. The mammoth engravings of the cavernes. Munibe 14, 353-58.
magdalenian site Gonnersdorf (Rhineland, BREUIL, H. 1964. Preface, in Musee des Antiquites
Germany), in Bandi et al1984, 295-322. Nationales. Collection Piette., by M. Chollot.
BOSINSKI , G. & FISCHER, G. 1974. Die Musees Nationaux, Paris, pp. 11-13.
Menschendarstellungen von Gonnersdorf der BREUIL, H. 1977. La perspective dans les dessins
J);usgrabungen von 1968. Steiner: Wiesbaden. paleolithiques anterieurs au Solutreen, in Cougnac,
BOSINSKI, G. & FISCHER, G. 1980. Mammut- und grotte peinte, by L. Meroc & ]. Mazet, 53-61.
Pferdedarstellungen von Gonnersdorf. Steiner: Editions des grottes de Cougnac: Gourdon.
Wiesbaden. BREUIL, H. & BEGOUEN, H. 1937. Quelques
BOUCHUD, ]. 1966. Essai sur le Renne et Ja oiseaux inedits ou meconnus de !'art prehistorique.
Climatologie du Paleolithique Moyen et Superieur. 12e Congres Prehist. de France, Toulouse-Foix
Magne: Perigueux . 1936, 475-88.
BOUET, B. et al198617 . Le Centre d'Information et de BREUIL, H. & LANTIER, R. 1959. Les Hommes de
Documentation (C.I. D .) H. Breuil. Antiquites la Pierre Ancienne. Payot: Paris. 2nd edition.
Nationales 18/19, 9-15.
 BIBLIOGRAPHY 221

BREUIL, H ., NOUGIER , L-R & ROBERT, R. 1956. CARALP, E., NOUGIER, L-R. & ROBERT, R.
Le 'lissoir aux ours' de la grotte de la Vache, aAlliat , 1973. Le theme du 'mammifere aux poissons' dans
et l'ours dans l'art franco-cantabrique occidental. l'art magdalenien. Bull. Soc. Prehist. Ariege 28, 11 -
Bull. Soc. Prehist. Ariege 11, 15-78, 7 pl. (& 12, 23.
1957, 53-54). CARCAUZON, C. 1984. Une nouvelle decouverte en
BREUIL, H. & OBERMAIER, H . 1935 . The Cave of Dordogne: la grotte prehistorique de Fronsac.
Altamira at Santillana del M ar, Spain. Tipografia de Revue Archeologique Sites 22, aoO.t, 7-15.
Archivos: Madrid . CARCAUZON , C. 1986. La grotte de Font-Bargeix .
BREUIL, H. & de SAINT PERIER, R. 1927. Les Bull. Soc. Hist. et Arch . du Perigord 113, 191-98.
Poissons, les Batraciens et les Reptiles dans l'Art CARCAUZON,C. 1988. Decouvertedequatregrottes
Quaternaire. Archives de l'Inst. Paleont. Humaine ornees en Perigord. Archeologia 235, 16-24.
2, Paris. CARCIUMARU, M. & BITIRI, M. 1983. Peintures
BREWSTER, R. 1986. Aboriginal cave paintings rupestres de la grotte Cuciulat (Roumanie). Bull.
among oldest in world. Popular Archaeology 7 (6), Soc. Prehist. franc;aise 80, 94-96.
p. 44. CARTAILHAC, E. 1902. Les cavernes ornees de
BRIELLE, G. 1968. Les bouquetins dans l'art parietal dessins. La grotte d'Altamira, Espagne. 'Mea culpa
quaternaire pyreneen. Tra vaux Inst. Art Prehist. d'un sceptique' . L'Anthropologie 13, 348-54.
Toulouse 10, 31-96. CARTAILHAC, E. 1908. Les plus anciens artistes de
BRODRICK, A.H. 1963. The Abbe Breuil, l'Humanite. Revue des Pyrenees 20, 509-27.
Prehistorian; a Biography. Hutchinson: London. CARTAILHAC, E. & BREUIL, H. 1905. Les
BROWN, S. 1987. The 1986 AURA Franco- peintures et gravures murales des cavernes
Cantabrian field trip: summary report. Rock Art pyreneennes: Marsoulas . L'Anth. 16, 431-43.
Research 4 (1 ), 56-60. CARTAILHAC, E. & BREUIL, H. 1906. La Caverne
BRUNET, J. 1981. Niaux: Du charbon de bois dans les d'Altamira a Santillane, pres Santander (Espagne).
peintures. Archeologia 156, juillet, p. 72. Imp. de Monaco.
BRUNET, ] . 1982. La Grotte de Niaux. Archeologia CASADO LOPEZ, M.P. 1977. Los Signos en el Arte
162, janvier, p. 74. Paleolitico de la Peninsula Iberica. Monografias
BUISSON, D . & PIN(:ON , G. 1986/7. Nouvelle Arq. 20, Zaragoza.
lecture d'un galet grave de Gourdan et essai CATTELAIN, P. 1977/8. Les propulseurs au
d'analyse des figurations d'oiseaux dans l'art Paleolithique Superieur. Essai d'un inventaire
paleolithique franc;ais . Antiquites Nationales 18/19, descriptif et critique. Memoire de licence en histoire
75-90. de l'art et arch., Univ. libre de Bruxelles, 2 vols.
BUTZER, K.W . et al1979. Dating and context of rock CHABREDIER, L. 1975. Les gravures paleolithiques
engravings in southern Africa. Science 203, 1201-14. de la grotte d'Ebbou (Ardeche). Archeocivilisation,
nile serie, numero special, 14/15.
CABRE AGUILO, J. 1934. Las Cuevas de Los Casares CHALOUPKA, G. 1984. From Palaeoart to Casual
y delaHoz. Archivo Esp. deArtey Arq. 30,Madrid. Paintings. Monograph 1, Northern Territory
CAMPBELL, ].B. 1977. The Upper Palaeolithic of Museum of Arts and Sciences: Darwin.
Britain. 2 vols, Clarendon Press : Oxford. CHOLLOT-LEGOUX, M. 1961. Une figuration
CAPDEVILLE, E. 1986. Aperc;us sur le probleme des magdalenienne de depouille animale. Antiquites
signes tectiformes dans l'art parietal paleolithique Nationales et Internationales lie annee, mars-j uin,
superieur d'Europe. Travaux de l'Institut d'Art p. 13 & 1 pl.
Prehistorique de Toulouse 28, 59-104. CHOLLOT-LEGOUX, M. 1964. Musee des
CAPITAN, L. & BOUYSSONIE , J. 1924. Un atelier Antiquites Nationales: Collection Piette (Art
d 'art prehistorique. Limeuil. Son gisement a Mobilier Prehistorique). Musees Nationaux: Paris.
!f!..avures.,j.ur pierres de l'age du Renne. Nourry: CHOLLOT-VARAGNAC, M. 1980. Les Origines du
Paris. Graphisme Symbolique. Singer-Polignac: Paris.
CAPITAN, L. , BREUIL, H. & PEYRONY, D . 1910. CLARK, L.H . 1966. They Sang For Horses: The
La Caverne de Font-de-Gaume aux Eyzies Impact of the Horse on Navajo and Apache Folklore.
(Dordogne). Monaco. Univ. of Arizona Press .
CAPITAN, L. , BREUIL, H. & PEYRONY, D . 1924. CLEGG,]. 1985 . Ethnography, pictures and theory.
Les Combarelles aux Eyzies (Dordogne). Masson: Paper for A.A.A . Conference, Valla, 20pp.
Paris. CLEGG, J. 1986. Berkeley Addendum to the above,
CAPITAN, L. & PEYRONY, D . 1928. La Madeleine. 7pp. See also 'Prehistoric messages in art and the
Son gisement, son industrie, ses oeu vres d'art. Pink Panther',in Univ. ofSydneyNews 29/4/86,p.89 .
Nourry: Paris.
222 IMAGES OF THE ICE AGE

CLOTIES, J. 1986. La determination des figurations DAMS, L. 1984. Preliminary findings at the 'organ'
humaines et animales dans !'art paleolithique sanctuary in the cave of Nerja, Malaga, Spain.
europeen, in Cultural Attitudes co Animals Oxford journal of Arch. 3, 1-14.
including Birds, Fish and Invertebrates, vol. 3, DAMS, L. 1985. Palaeolithic lithophones: descriptions
39pp. World Arch . Congress, Allen & Unwin : and comparisons. Oxford journal ofArch. 4, 31 -46.
London. DAMS, L. 1987. Poissons et contours de type
COLLINS, D. &ONIANS, }.1978. Theoriginsofart. pisciforme dans !'art parietal paleolithique. Bull.
Art History 1, 1-25 . Soc. Royale beige Anth. Prehist. 98, 81-132 . (Also
COLLOT, F. et al La theorie de !'information en 'Fish images in palaeolithic cave art' in Archaeology
Prehistoire. Bio-Mathematique 77, 32pp. Today 8 (2), 16-20).
COMBIER, }. 1984. Grottes ornees de l'Ardeche, in DAMS, L. 1987a. L'Art Paleolithique deJa Grotte de
Les Premiers Artistes, Dossier de I' Arch. 87, 80-86. Nerja (Malaga, Espagne). British Arch. Reports,
"COMBIER, J., DROUOT, E. & HUCHARD, P. Int. Series No. 385, Oxford.
195 8. Les grottes sol utreennes a gravures parietales DASTUGUE, ]. & DE LUMLEY, M.A. 1976. Les
du canyon inferieur de l'Ardeche. Mem. Soc. maladies des hommes prehistoriques du
Prehist. fr. 5, 61-117 . Paleolithique superieur et du Mesolithique, in La
CONKEY, M.W. 1978. Style and information in Prehistoire franfaise, vol. 1:1, 612-22 (H. de
cultural evolution: toward a predictive model for the Lumley, ed.). C.N.R.S .: Paris.
Paleolithic, in Social Archaeology, Beyond DAUVOIS, M. 1977. Travail experimental de l'ivoire:
Subsistence and Dating (C.L. Redman et al, eds), sculpture d'une statuette feminine, in Methodologie
61-85. Academic Press: New York/London . appliquee a l'industrie de l'os prehistorique (H.
CONKEY, M.W. 1980. The identification of Camps-Fabrer, ed. ), 270-73 . C.N.R.S.: Paris.
prehistoric hunter-gatherer aggregation sites; the DAVENPORT, D. &JOCHIM, M.A . 1988 . Anoteon
case of Altarnira. Current Anth. 21, 609-30. the scene in the shaft at Lascaux. Antiquity 62 (in
CONKEY, M. W. 1981. A century of Palaeolithic cave press) .
art. Archaeology 34, 20-28 DAVIS, W. 1986. The origins of image making.
CONKEY, M.W. 1983. On the origins of Paleolithic Current Anthropology 27, 193-215, 371, 515-16.
art: a review and some critical thoughts, in The DAVIS, W. 1987. Replication and depiction in
Mousterian Legacy (E. Trinkaus, ed .), 201-27. Paleolithic art. Representations 19, 111-47.
British Arch . Reports, Int series 164: Oxford. DECOUVERTES DE L'ART DES GROTTES ET
CONKEY, M .W . 1984. To find ourselves: art and DES ABRIS. 1984. Catalogue de !'Exposition,
social geography of prehistoric hunter-gatherers, in 59pp. Rejou: Perigueux.
Past and Present in Hunter-Gatherer Studies (C. DELCOURT-VLAEMINCK, M. 1975. Les
Schrire, ed.), 253-76 . Academic Press: New York! representations vegetales dans !'art du Paleolithique
London. superieur. Paleontologie et Prehistoire: Rev. Soc.
CONKEY, M.W. 1985 . Ritual communication, social Tournaisienne de Geol., Preh. et Arch. 31, 70-96.
elaboration and the variable trajectories of DELLUC, B. & G. 1973. Quelques figurations
Paleolithic material culture, in Prehistoric Hunter- paleolithiques inedites des environs des Eyzies
Gatherers: the Emergence of Social and Cultural (Dordogne): Grottes Archambeau, du Roc et de Ia
Complexity (T.D. Price & ].A. Brown, eds), 299- Mouthe. Gallia Prehistoire 16, 201-9.
323. Academic Press: New York/London . DELLUC, B. & G. 1978. Les manifestations
COURAUD, C. 1982. Techniques de peintures graphiques aurignaciens sur support rocheux des
prehistoriques: Experiences . Information Couleur environs des Eyzies (Dordogne). Gallia Prehisroire
19, 3-6. 21, 213-438.
COURAUD, C. 1983. Pour une etude methodologique DELLUC, B. & G. 1981. Une visite a Ia grotte de
~s colorants prehistoriques. Bull. Soc. Prehist. fr. Rouffignac en 1759. Bull. Soc. Hi st. et Arch. du
80, 104-10. Perigord 108, 3-11.
COURAUD, C. 1983a. La couleur dans !'art DELLUC, B. & G. 1981a. La grotte ornee de
paleolithique. Information Couleur 22, 5-9. Comarque a Sireuil (Dordogne). Gallia Prehisroire
COURAUD, C. 1985 . L'Art Azilien. Origine - 24, 1-97.
Survivance. XXe Supplement a Gallia Prehistoire. DELLUC. B. &G . l981b. Lessignesen 'empreinte'du
debut du Paleolithique superieur, in 21e Congres
DAMS, L. 1978. L'Art Paleolithique de la Caverne de Prehist. de France, vol. 2, 111-16.
Ja Pileta. Akademische Druck- u. Verlagsanstalt:
Graz.
 BIBLIOGRAPHY 223

DELLUC,B . &G. 1983. Les grottesorneesdeDomme DJINDJIAN, F. & PIN(:ON, G. 1986. Unexemplede
(Dordogne): La Martine, Le Mammouth et le banque de donnees sur microserveur Videotex:
Pigeonnier. Gallia Prehistoire 26, 7-80. L'Art Parietal Paleolithique. Bull. Soc. Prehist.
DELLUC, B. & G. 1983a. La Croze a Gontran, grotte fran~aise 83, 332-34.
ornee aux Eyzies-de-Tayac (Dordogne) . Ars DORN, R.I., NOBBS, M. & CAHILL, T.A. (in
Praehistorica 2, 13-48. press). Cation-ratio dating of rock engravings from
DELLUC, B. & G. 1984. L'art parietal avant Lascaux. the Olary Province of arid South Australia.
Dossiers de l'Archeologie 87 , octobre, 'Les Premiers Antiquity (forthcoming).
Artistes', 52-60. DORTCH, C. 1979. Australia's oldest known
DELLUC, B. & G. 1984a. Lascaux II , a faithful copy. ornaments. Antiquity 53, 39-43.
Antiquity 58, 194-96,2 pl. DORTCH, C. 1984. Devil's Lair: A study in
DELLUC, B. & G. 1984b . Lecture analytique des prehistory. Western Australia Museum: Perth .
supports rocheux graves et releve synthetique. DUHARD, J-P. 1987. Le soi-disant 'hermaphrodite'
L'Anthropologie 88, 519-29. de Grimaldi: une nouvelle interpretation
DELLUC, B. & G. 1984c. Faune figuree et faune obstetricale. Bull. Soc. Anth. Sud-Ouest 22, 139-44.
consommee: une magie de la chasse? Dossiers de DUHARD, J-P. 1987a. La Statuette de Monpazier
l'Arch. 87, oct. , 'Les Premiers Artistes', 28-29. represente-elle une parturiente? Bull. Soc. Prehist.
DELLUC, B. & G. 1985 . De l'empreinte au signe, in Ariege-Pyrenees42, 155-63.
Traces et Messages de la Prehistoire, Dossiers de DUHARD, J-P. & ROUSSOT, A. 1988. Le gland
l'Arch. 90, 56-62 . penien sculpte de Laussel (Dordogne). Bull. Soc.
DELPORTE, H. 1962. Le probleme des statuettes Prehist. fran~aise 85, 41 -44.
feminines dans le leptolithique occidental.
Mitteilungen der Anth. Gesellschaft in Wien 92, 53- EASTHAM, A. 1986. The season or the symbol: the
60. evidence of swallows in the Paleolithic of Western
DELPORTE, H. 1971. A propos du style des Europe. Paper for International Council for
figurations feminines gravettiennes. Antiquites Archaeozoology Conference, Bordeaux 1986, 14pp.
Nationales 3, 5-20. EASTHAM, A. & M. 1979 . The wall art of the Franco-
DELPORTE, H. 1973. Les techniques de la gravure Cantabrian deep caves. Art History 2, 365-87.
paleolithique, in Estudios dedicados al Prof Dr. EATON, R.L. 1978. The evolution of trophy hunting.
Luis Pericot, 119-29. Inst. de Arq . yPreh.., Univ. de Carnivore 1, 110-21.
Barcelona. EPPEL, F. 1972. Les objets d'art paleolithique en
DELPORTE, H. 1979. L'Image de la Femf!l.e dans Autriche. Bull. Soc. Prehist. Ariege 27, 73-81.
l'Art Prehistorique. Picard: Paris. d'ERRICO, F . 1987. Nouveaux indices et nouvelles
DELPORTE, H. 1984. L'art mobilier et ses rapports techniques microscopiques pour la lecture de l'art
avec Ia faune paleolithique, in Bandi et al1984, 111- grave mobilier. C.R. Acad. Sc. Paris 304,-Serie II,
42. No. 13, 761-64, 1 pl.
DELPORTE, H . 1987. Piette, pionnier de la
Prehistoire. Picard: Paris. FAGES, G. & MOURER-CHAUVIRE, C. 1983. La
DELPORTE, H. , KANDEL, D . & PIN(:ON, G. flute enos d' oiseau de la grotte sepulcrale de Veyreau
1986. Le C.I.D. Breuil, domaine A.P.M.: Un (Aveyron) et inventaire des flutes prehistoriques
systeme documentaire sur I' Art Paleolithique d'Europe. in La Faune et l'Homme Prehistoriques,
Mobilier. Bull. Soc. Prehist. fran ~aise 83, 299303. Mem. 16 de la Soc. Prehist. franc;aise, 95-103.
DELPORTE, H . & MONS , L. 1975. Omoplate FAURE, M. 1978. Revision critique d'une collection
decoree du Mas d'Azil (Ariege). Antiquites de gravures mobilieres paleolithiques: les galets et
Nationales 7, 14-23. les os graves de la Colombiere (Neuville-sur-Ain,
DELPOR41E, H. & MONS, L. 1980. Gravure sur os et Ain, France). Nouv. Arch. Mus. Hist. nat. Lyon
argile, in Revivre la Prehistoire, Dossier de fasc. 16, 41-99, 6 pl.
I'Archeologie 46, 40-4 5. FORBES, A. &CROWDER, T.R.1979. The problem
DESBROSSE, R., FERRIER, J . & TABORIN, Y. of Franco-Cantabrian abstract signs: agenda for a
1976. La parure. in La Prehistoire fran~aise, I: 1 (H. new approach. World Arch. 10, 350-66, 2 pl.
de Lumley, ed.), 710-13. C.N .R. S.: Paris. FORTEA, F.j. 1978. Arte paleolitico delMediterraneo
DEWEZ, M. 1974. New hypotheses concerning two espafiol. Trabajos de Prehistoria 35, 99-149.
engraved bones from La Gron e de Remouchamps, FORTEA, J. 1981. Investigaciones en la cuenca media
Belgium. World Arch. 5, 337-45, 1 pl. del Nal6n, Asturias (Espana). Noticia y primeros
resultados. Zephyrus 32/33, 5-16.
224 IMAGES OF THE ICE AGE

FREEMAN, L.G . 1978. Mamut , jabali y bisonte en GLORY, A. 1968. L'enigme de l'art quaternaire peut-
A1tamira: reinterpretaciones sugeridas por la historia e11e etre resolue par la theorie du culte des ongones?
natural, in Curso de Arce Rupescre Paleolicico (A. in Simposio de Arce Rupestre, Barcelona 1966, 25-
Beltran, ed.), 157-79. Universidad Int. 'Menendez 60.
Pelayo': Santander. GONZALEZ ECHEGARAY, ]. 1968. Sobre Ia
FREEMAN, L.G. 1984. Techniques of figure dataci6n de los santuarios paleoliticos, in Simposio
enhancement in Paleolithic cave art, in Scripta de Arte Rupestre, Barcelona 1966, 61-65.
Praehiscorica. Francisco ]orda Obla ca, 209-32. GONZALEZ ECHEGARAY, ]. 1972 . Notas para el
Univ. de Salamanca. estudio cronol6gico del arte rupestre de la cueva del
FREEMAN, L.G . & GONZALEZ ECHEGARAY, ]. Castillo, in Santander Symposium 409-22.
1983. Tally-marked bone from Mousterian levels at GONZALEZ ECHEGARAY, ]. 1974. Pincuras y
Cueva Morin (Santander, Spain), in Homenaje al grabados de la cueva de las Chimeneas (Puente
Prof M. Almagro Basch, I, 143-47. Min. de Viesgo, Santander). Monografias de Arte Rupestre,
Cultura: Madrid. Arte Paleolitico No. 2, Barcelona.
FREUND, G. 1957. L'art aurignacien en Europe GONZALEZ GARCIA, R. 1985. Aproximaci6 al
centrale. Bull. Soc. Prehisc. Ariege 12, 55-78, 2 pl. Desenvolupament i Situaci6 de Jes Manifestacions
FROLOV, B.A. 1977/9. Numbers in Paleolithic Arcisciques Quaternaries: a Jes cavicacs del Monee del
graphic art and the initial stages in the development Castillo. Tesis de Llicenciatura , Univ . de Barcelona,
of mathematics . Soviet Anch . and Arch. 16, 1977/8, Fac. de Geografia i Historia , Dept. d'Historia de
142-66; and 17, 1978/9, 73-93,41-74 & 61-113 . l'Art. 547pp.
FROLOV, B.A. 1979 . Les bases cognitives de l'art GONZALEZ GARCIA , R. 1987. Organisation,
pa1eo1ithique, in Valcamonica Symposium III. The distribution and typology of the cave art of Monte del
Intellectual Expressions of Prehistoric Man: Arc and Castillo, Spain. Rock Arc Research 4, 127-36.
Religion, 295-98. GOREN-INBAR, .N. 1986. A figurine from the
FROLOV, B.A. 1981. L'art paleolithique: Prehistoire Acheulian site of Berekhat Ram. Mi'Tekufat
de 1a science? in Arce Paleolicico, Comisi6n XI, Xth Ha'Even 19, 7-12 .
Congress UISPP, Mexico City, 60-81. GRAINDOR, M .]. & MARTIN, Y. 1972. L 'Art
Prehiscorique de Gouy. Presses de la Cite: Paris.
GAMBLE, C. 1982. Interaction and alliance in GRAZIOSI, P . 1960. Palaeolithic Arc. Faber: London.
Palaeolithic society. Man 17, 92-107. GRAZIOSI, P. 1973. L'Arce Preistorica in Icalia.
GARCIA, M. 1979. Les silicones elastomeres R .T.V. Sanzoni: Florence.
appliques aux releves de vestiges prehistoriques GUIDON, N. & DELIBRIAS , G. 1986. Carbon-14
(Arts, Empreintes Humaines et Animales). dates point to man in the Americas 32,000 years ago.
L'Anchropologie 83, 5-42 & 189-222. Nature 321, 769-71.
GARCIA GUINEA, M .A. 1979. Alcamira y otras GUTHRIE, R.D . 1984. Ethological observations from
cuevas de Cancabria. Silex: Madrid. Palaeo1ithic,art, in Bandi et a11984, 35-74.
GAUSSEN, ]. 1964. La Grocce Orneede Gabillou (pres
Mussidan, Dordogne). Memoire 4, Inst. de Prehist. HAHN, J. 1971. La statuette masculine de la grotte du
de l'Univ. de Bordeaux. Hohlenstein-Stadel (Wurtemberg).
GEOFFROY, C. 1974. La couleur dans l'art parietal L'Anthropologie 75, 233-44.
pa1eo1ithique. Cahiers du Centre de Recherches HAHN, J. 1972. Aurignacian signs , pendants and art
Prehist. 3, 45-64. objects in central and eastern Europe. World Arch.
GIEDION, S. 1965. L 'Ecernel Present. La Naissance 3, 252-66, 2 pl.
de l'Art. Editions de la Connaissance: Bruxelles. HAHN,] . 1979. Elfenbeinplastiken des Aurignacien
GIMBUTAS, M. 1981. Vulvas, breasts and buttocks of aus dem Geissenklosterle. Archiiologisches
.-.he Goddess Creatress: Commentary on the origins Korrespondenzblatt 9, 135-42 , 2 pl.
of art, in The Shape of the Past: Studies in Honor of HAHN, J. 1982 . Demi-relief aurignacien en ivoire de Ia
Franklin D. Murphy (G. Buccellati & C. Speroni, grotte Geissenklosterle, pres d'Ulm (Allemagne
eds), 15-42. Inst. of Arch. : Los Angeles. federale). Bull. Soc. prehist. fr. 79 , 73-77.
GLADKIH, M.I. , KORNIETZ, N.L. & SOFFER, HAHN, ]. 1984. L'art mobilier aurignacien en
0. 1984. Mammoth-bone dwellings on the Russian Allemagne du Sud-Ouest: Essai d'analyse
Plain. Scientific American 251, 136-42 . zoologique et ethologique, in Bandi et al1984, 283-
GLORY, A. 1964. La stratigraphie des peintures a 93.
Lascaux, in MisceJan ea en Homenaje al Abate H.
Breuil, vol. 1, 449-55. Barcelona.
 BIBLIOGRAPHY 225

HALLAM , S.] . 1971. Roof marking in the 'Orchestra JORDA CERDA, F . 1983. El mamut en el arte
Shell' Cave, Wanneroo, near Perth, Western paleolitico peninsular y la hierogamia de Los
Australia. Mankind 8, 90-103. Casares, in Homenaje al Prof. M. Almagro Basch,
HALVERSON, J . 1987 . Art for art's sake in the vol. 1, 265-72. Min . de Cultura: Madrid.
Paleolithic. Current Anth. 28, 63-89, 203-05. JORDA CERDA, F. 1985. Los grabados de Mazouco,
HAMMOND, N . 1974. Paleolithic mammalian faunas los santuarios monotematicos y los animales
and parietal art in Cantabria: a comment on dominantes en el arte paleolitico peninsular. Revista
Freeman. American Antiquity 39, 618-19. de Guimaraes 94, 23pp.
HARLE, E. 1881. La grotte d' Altamira, pres de JORDA CERDA, F. 1987. Sobre figuras rupestres
Santander (Espagne). Mau?riaux pour l'Histoire paleoliticas de posibles caballos domesticados.
Primitive et Naturelle de J'Homme 2e serie, XII , Archivo de Prehistoria Levan tina 17, 49-58.
275-83, 1 pl. JORGE, S.O. et al 1981. Gravuras rupestres de
HARRISON, R.A. 1978. A pierced reindeer phalanx Mazouco (Freixo de Espada a Cinta). Arqueologia
from Banwell Bone Cave and some experimental (Porto) 3, 3-12.
work on phalangeal whistles . Proc. Univ. Bristol JORGE, S.O . et al 1982. Descoberta de gravuras
Spel. Soc. 15 , 7-22 , 1 pl. rupestres em Mazouco, Freixo de Espada-a-Cinta
HENNIG, H. 1960. L'art magdalenien en Europe (Portugal). Zephyrus 34135, 65-70 .
centrale. Bull. Soc. Preh . Ariege 15, 24-45. JUDSON, S. 1959. Paleolithic paint. Science 130, p.
HORNBLOWER, G.D . 1951. The origin of pictorial 708 .
art . Man 51, 2-3 . JULIEN, M . & LAVALLEE, D. 1987. Les chasseurs
de Ia prehistoire, in Ancien Perou: Vie, Pouvoir et
JANSSENS, P.A . 1957. Medical views on prehistoric Mort. Exhibition catalogue, Musee de l'Homme,
representations of human hands. Medical History 1, pp. 45-60. Nathan: Paris.
318-22, 2 pl.
JELINEK, J. 1975. Encyclopedie illustree de KEHOE, T.F. 1987. Corralling life. Wisconsin
l'Homme Prehistorique. Grund : Paris. Academy Review, March, 45-48.
JIA LANPO 1980. Early Man in China. Foreign KEHOE, T.F. (in press) Corralling life. Plains
Languages Press: Beijing. Anthropologist
JOCHIM, M . 1983. Palaeolithic cave art in ecological KLIMA, B. 1982. Neue Forschungsergebnisse in
perspective, in Hunter-Gatherer Economy in Dolni Vestonice, in Aurignacien et Gravettien en
Prehistory (G.N. Bailey, ed .), 212-19. Cambridge Europe, Fascicule II, Cracovie/Nitra 1980. Etudes et
Univ. Press. Recherches Arch . de l'Univ. de Liege, 13, 171-80.
JORDA CERDA, F. 1964. El arte rupestre paleolitico KLIMA, B. 1983. Une nouvelle statuette paleolithique
de Ia regwn cantabrica: nueva secuencia a Dolni Vestonice . Bull. Soc. Prehist. fr . 80, 176-78.
cronol6gico-cultural, in Prehistoric Art of the KLIMA, B. 1984. Les representations animales du
Western Mediterranean and the Sahara (L. Pericot & Paleolithique Superieur de Dolni Vestonice
E. Ripoll, eds), 47-81. Viking Fund Publications in (Tchecoslovaquie), in Bandi et al 1984, 323-32.
Anth., No . 39, New York. KOBY, F.E. 1954. Y a-t-il eu, a Lascaux, 'un Bos
JORDA CERDA, F. 1964a. Sobre tecnicas, temas y longifrons'? Bull. Soc. Prehist. fr. 51, 434-41.
etapas del Arte Paleolitico de la Region Cantabrica. KOBY, F .E. 1968. Les 'rennes' de Tursac paraissent
Zephyrus 15 , 5-25. etre plut6t des daims. Bull. Soc. Prehist. Ariege 23,
JORDA CERDA, F. 1972 . Las superposiciones en el 123-30.
gran techo de Altamira , in Santander Symposium, KOZLOWSKI, J.K. (ed.) 1982. Excavation in the
423-56. Bacha Kiro Cave (Bulgaria): Final Report.
JORDA CERDA, F. 1979. Sur des sanctuaires Panstwowe Wydawnictwo Naukowe: Warsaw .
..monotru.natiques dans l'art rupestre cantabrique, in KRAFT, J.C. & THOMAS, R.A. 1976. Early man at
Valcamonica Symposium III. The Intellectual Holly Oak, Delaware. Science 192, 756-61.
Expressions of Prehistoric Man: Art and Religion, KUHN, H. 1955. On the Track of Prehistoric Man.
331-48 . Hutchinson: London.
JORDA CERDA, F. 1980. El gran techo de Altamira y KUHN, H. 1971. Die Felsbilder Europas.
sus santuarios superpuestos, in Altamira Kohlhammer: Stuttgart.
Symposium, 277-87. Min . de Cultura: Madrid. KUMAR, G., NARVARE , G. & PANCHOLI , R.
1988.Engraved ostrich eggshell objects: new
evidence of Upper Palaeolithic art in India . Rock Art
Research 5 (1 ), 43-53.
226 IMAGES OF THE ICE AGE

KURTEN, B. 1986. How to Deep-Freeze a Mammoth. LEJEUNE, M. (in press) L'Art Parietal dans son
Columbia Univ . Press: New York. Contexte Nature]. Etudes et Recherches Arch. de
l'Univ. de Liege 14.
LAMING-EMPERAIRE,A . 1959. Lascaux. Paintings LE MORT, F. 1985. Un exemple de modification
and Engravings. Pelican: Harmondsworth. intentionnelle: la dent humaine perforee de Saint-
LAMING-EMPERAIRE , A. 1962. La Signification de Germain-la-Riviere (Pal. Sup.). Bull. Soc. Prehist.
l'Art Rupestre Paleolithique. Picard: Paris. fr. 82, 190-92.
LAMING-EMPERAIRE, A. 1970. Systeme de penser LEONARDI, P. 1983. Incisioni musteriane del Riparo
et organisation sociale dans l'art rupestre Tagliente in Valpantena nei Monti Lessini presso
paleolithique, in L 'H omme de Cro-Magnon, 197- Verona (ltalia), in Homenaje al Prof M. Almagro
211. Arts et Metiers Graphiques: Paris. Basch, vol. 1, 149-54. Min. de Cultura: Madrid.
LAMING-EMPERAIRE , A. 1971 . Une hypothese de LEROI-GOURHAN, And. 1958. La fonction des
travail pour une nouvelle approche des sooetes signes dans les sanctuaires paleolithiques. Bull. Soc.
prehistoriques, in Melanges Varagnac, 541-51. Prehist. fr. 55, 307-21.
Ecole Pratique des Hautes Etudes: Paris. LEROI-GOURHAN, And. 1958a. Le symbolisme des
LAMING-EMPERAIRE , A. 1972. Art rupestre et grands signes dans l'art parietal paleolithique. Ibid.
organisation sociale, in Santander Symposium, 65- 55, 384-98.
82. LEROI-GOURHAN, And. 1958b. Repartition et
LARTET, E. 1861. Nouvelles recherches sur la groupement des animaux dans l'art parietal
coexistence de !'Homme et des grands mammiferes paleolithique. Ibid. 55, 515 -28.
fossiles. Annales des Sciences Naturelles (Zoologie), LEROI-GOURHAN, And. 1962. Chronologie de l'art
4e serie, XV, 177-253, 3 pl. paleolithique, in Atti del VI Congresso UISPP,
LARTET, E. & CHRISTY, H. 1864. Sur des figures Rome, vol. 3, 341-45.
d'animaux gravees ou sculptees et autres produits LEROI-GOURHAN, And. 1966. Reflexions de
d'art et d'industrie rapportables aux temps methode sur l'art paleolithique. Bull. Soc. Prehist.
primordiaux de la periode humaine. Revue Arch. fr. 63, 35-49.
IX, 233-67, 2 pl. LEROI-GOURHAN, And. 1967. Les mains de
LARTET, E. & CHRISTY, H. 1875. Reliquiae Gargas. Essai pour une etude d'ensemble. Ibid. 64,
Aquitanicae. Jones: London . (1865-1875). 107-22.
LAURENT, P. 1963 . La tete humaine gravee sur bois LEROI-GOURHAN, And. 1968. Les signes parietaux
de renne de la grotte du Placard (Charente). du Paleolithique superieur franco-cantabrique, in
L'Anthropologie 67, 563-69. Simposio de Arte Rupestre, Barcelona 1966, 67-77.
LAURENT, P. 1971. Iconographie et copies LEROI-GOURHAN, And. 1971. Prehistoire de l'Art
successives: la gravure anthropomorphe du Placard. Occidental. Mazenod: Paris. 2e Edition.
Mem. Soc. Arch. Hist. Charente, 215-28. LEROI-GOURHAN, And. 1972. Considerations sur
LEASON, P.A. 1939. A new view of the Western I' organisation spatiale des figures animales dans l'art
European group of Quaternary cave art. Proc. parietal paleolithique, in Santander Symposium
Prehist. Society 5, 51-60. 281-308.
LEASON, P.A . 1956. Obvious facts of Quaternary LEROI-GOURHAN, And. 1978. Le cheval sur galet
cave art. Medical and Biological Illustration 6 (4), de la galerie Breuil au Mas d' Azil (Ariege). Gallia
209-14. Preh. 21, 439-45.
LEGUAY, L. 1877. Les procedes employes pour la LEROI-GOURHAN, And. 1980. Les signes parietaux
gravure et la sculpture des os avec le silex. Bull. Soc. comme 'marqueurs' ethniques, in Altamira
. Anth. Paris 2e serie, 12 , 280-96. Symposium 289-94. Min. de Cultura: Madrid.
LEGUAY, L. 1882. Sur la gravure des os au moyen du LEROI-GOURHAN, And. 1981. Le cas de Lascaux.
.ilex. C.r. Assoc. fran ~aise pour J'A vancement des Monuments Historiques 118, 24-32.
Sciences 11, La Rochelle, 677-79. LEROI-GOURHAN, And. 1982. The Dawn of
LEJEUNE; M . 1981. L 'Utilisation des Accidents European Art. Cambridge Univ. Press.
Naturels dans Je trace des figurations parietales du LEROI-GOURHAN, And. 1983. Les entites
Paleolithique Superieur franco-cantabrique. imaginaires. Esquisse d'une recherche sur les
Memoire de fin d'etudes, Univ. de Liege, 274pp. monstres parietaux paleolithiques, in Homenaje al
LEJEUNE , M . 1987. L 'Art Mobilier Paleolithique et Prof M. Almagro Basch, vol. 1, 251-63. Min. de
Mesolithique en Belgique. Artefacts 4, Editions du Cultura: Madrid.
Centre d'Etudes et de Documentation
Archeologiques: Treignes-Viroinval, Belgium.
 BIBLIOGRAPHY 227

LEROI-GOURHAN, And. 1984. Le realisme du LORBLANCHET, M. 1974. L'Art Prehistorique en

comportement dans l'art paleolithique d'Europe de Quercy. La Grotte des Escabasses (Themines- Lot).
l'Ouest, in Bandi et al 1984, 75-90. PGP: Morlaas.
LEROI-GOURHAN; And . 1984a. Simbolos, Artes y LORBLANCHET, M. 1977. From naturalism to
Creencias de la Prehistoria. lstmo: Madrid. abstraction in European prehistoric rock art, in
LEROI-GOURHAN, And . 1984b. Arte y Grafismo en Form in Indigenous Art (P.J. Ucko, ed.), 44-56 .
la Europa Prehist6rica. lstmo : Madrid. Duckworth: London.
LEROI-GOURHAN, Arl. 1983. Du fond des grottes LORBLANCHET, M. 1980. Peindre sur les parois des
aux terrasses ensoleillees , in Hom enaje al Prof M . grottes, in Revivre la Prehistoire, Dossier de 1' Arch .
Almagro Basch, vol. 1, 239-49 . Min. de Cultura: 46, 33-39.
Madrid. LORBLANCHET, M. 1980a. Les gravures de !'ouest
LEROI-GOURHAN, Arl . & ALLAIN, J . (eds) . 1979. australien . Leur renovation au cours des ages . Bull.
Lascaux Inconnu . Xlle Suppl. aGallia Prehistoire. Soc. Prehist. fr. 77 , 463-77.
LEROY-PROST, C. 1984. L'art mobilier, in Les LORBLANCHET, M. 1981. Les dessins noirs du
Premiers Artistes, Dossier de 1' Archeologie 87, 45- Pech-Merle. in XXIe Congres Prehist. de France,
51. Montauban/Cahors 1979, vol. 1, 178-207.
LEVEQUE, F. & VANDERMEERSCH, B. 1980. LORBLANCHET, M. 1984. Les releves d'art
Decouverte de restes humains dans un niveau prehistorique, in L'Art des Cavernes. Atlas des
caste1perronien a Saint-Cesaire (Charente- grottes ornees paleolithiques fram;aises, 41-51. Min.
Maritime). C.r. Academie des Sciences (Paris) D, de Ia Culture: Paris.
291, 187-89. LORBLANCHET, M . & WELTE , A-C. 1987. Les
LEWIS-WILLIAMS, J.D. 1982. The economic and figurations feminines stylisees du Magdalenien
social context of Southern San rock art. Current Superieur du Quercy. Bull. Soc. Etudes du Lot 108,
Anth. 23, 429-49. f.3, 3-58.
LEWIS-WILLIAMS, J.D . & DOWSON, T.A. 1988. de LUMLEY, H. 1968. Le bison grave de Segries,
The signs of all times: entoptic phenomena in Upper Moustiers-Ste-Marie, Bassin du Verdon (Basses-
Palaeolithic art. Current Anth. 29, 201 -45. Alpes), in Simposio de Arte Rupestre, Barcelona
LHOTE, H . 1968. La plaquette dite de 'la Femme au 1966, 109-21.
Renne', de Laugerie-Basse, et son interpretation de LUMLEY, H. 1968a. Proportions et constructions
zoologique, in Simposio de Arte Rupestre, dans l'art paleolithique: le bison, in Ibid. , 123-45.
Barcelona 1966, 79-97.
LHOTE, H. 1968a. A propos de l'identite de la femme MACINTOSH, N.W.G. 1977. Beswick Creek cave
et du bison selon les theories recentes de l'art parietal two decades later, in Form in Indigenous Art (P.J.
prehistorique, in Ibid. , 99-108 . Ucko, ed .), 191-97. Duckworth: London .
LHOTE, H. 1972. Observations sur la technique et la MADARIAGA DE LA CAMPA, B. 1981. Historia del
lecture des gravures et peintures quaternaires du descubrimiento y valoraci6n del arte rupestre
Sud-ouest de la France, in Santander Symposium, espafi.ol. in Altamira Symposium, 299-310. Min . de
321-30. Cultura: Madrid.
LINARES MALAGA, E. 1988. Arte mobiliar con MANIA , D . & VLCEK, E. 1987. Homo erectus from
tradici6n rupestre en el Sur del Peru. Rock Art Bilzingsleben (GDR) - his culture and his
Research 5 (1 ), 54-66. environment. Anthropologie (Brno) 25 (1), 1-45 .
LINDNER, K. 1950. La Chasse Prehistorique. Payot: MARINGER, J. & BANDI, H-G. 1953 . Art in the Ice
Paris. Age. Allen & Unwin: London .
LION VALDERRABANO , R. 1971. El Caballo en el MARSHACK, A. 1969. Polesini: a reexamination of
Arte Cantabro-Aquitano. Public. del Patronato de the engraved Upper Palaeolithic mobiliary materials
las Clljflas prehist. de la prov . de Santander VIII: of Italy by a new methodology. Rivista di Scienze
Santander. Preistoriche 24, 219-81 .
LIPS, J.E. 1949. The Origin of Things . Harrap: MARSHACK, A. 1970. La baton de commandement
London. de Montgaudier (Charente): Reexamen au
LLONGUERAS, M. 1972. Gnificos estadisticos sobre microscope et interpretation nouvelle .
las placas de la cueva del Parpall6 (Gandia, L'Anthropologie 74, 321-52 .
Valencia), in Santander Symposium , 393-405. MARSHACK, A. 1970a. Notation dans les Gravures
LOMMEL, A. 1967. The World of the Early Hunters. du Paleolithique Superieur. Memoire 8, lnst.
Medicine-men, Shamans and Artists. Evelyn, Prehist. Univ. Bordeaux; Delmas: Bordeaux.
Adams & Mackay: London.
228 IMAGES OF THE ICE AGE

MARSHACK, A. 1972. The Roots of Civilization . MAZAK, V. 1970. On a supposed prehistoric

Weidenfeld & Nicolson : London . representation of the Pleistocene scimitar cat,
MARSHACK, A. 1972a. Cognitive aspects of Upper Homotherium FABRINI 1890 (Mammalia.
Paleolithic engraving . Current Anth. 13, 445-77; Machairodontidae). Zeitschrift f. Siiugetierkunde
also 15, 1974, 327-32, & 16, 1975, 297-98. 35, 359-62 .
MARSHACK, A. 1975. Exploring the mind of lee Age McBURNEY, C.B.M. 1967. The Haua Fteah
man. National Geographic 147 (1), 64-89 . (Cyrenaica) and the Stone Age of the South-east
MARS HACK, A . 1976. Implications of the Paleolithic Mediterranean. Cambridge Univ. Press.
symbolic evidence for the origin of language. McNICKLE, H.P. 1985. An introduction to the Spear
American Scientist 64, 136-45 . Hill rock art complex, Northwestern Australia.
MARSHACK, A. 1976a. Some implications of the Rock Art Research 2 (1), 48-64.
Paleolithic symbolic evidence for the origin of MELLARS, P.A. 1985. The ecological basis of social
language . Current Anth. 17, 274-81. complexity in the Upper Palaeolithic of
MARSHACK, A. 1976b . The message in the Southwestern France, in Prehistoric Hunter-
markings. Horizon 18, 64-73 . Gatherers: the Emergence of Social and Cultural
MARSHACK, A. 1977. The meander as a system: the Complexity (T.D. Price & J.A. Brown, eds), 271-
analysis and recognition of iconographic units in 97 . Academic Press: New York/London.
Upper Paleolithic compositions, in Form in MESSMACHER, M. 1981. El arte paleolitico en
Indigenous Art (P.J. Ucko, ed.), 286-317 . Mexico, in Arte Paleolitico, Comisi6n XI, Xth
Duckworth: London . Congress UISPP, Mexico City, 82-110.
MARSHACK, A. 1979. Upper Paleolithic symbol MESZAROS, G . & VERTES, L. 1955. A paint mine
systems of the Russian Plain: cognitive and from the early Upper Palaeolithic age near Lovas
comparative analysis. Current Anth. 20, 271-311, (Hungary, County Veszprem). Acta Arch.
604-08 . Hungarica 5, 1-34.
MARSHACK, A. 1981. On Paleolithic ochre and the MEYLAN, C. 1986. Leroi-Gourhan au bestiaire de la
early uses of color and symbol. Current Anth. 22, prehistoire. Bull. Soc. Prehist. Ariege-Pyrenees4l,
188-91. 31-62.
MARSHACK, A. 1984. Concepts theoriques MITHEN, S.J. 1988. Looking and learning: Upper
conduisant a de nouvelles methodes analytiques, de Palaeolithic art and information gathering. World
nouveaux procedes de recherche et categories de Arch. 19, 297-327.
donnees. L'Anthropologie 88, 575-86. MOLARD, Cdt. 1908. Les grottes de Sabart (Ariege):
MARSHACK, A. 1985. Theoretical concepts that lead Niaux et les dessins prehistoriques. Spelunca VII,
to new analytic methods, modes of inquiry and No . 53, 177-91.
classes of data. Rock Art Research 2, 95-111; and 3 MONS, L. 1980/ 1. Les baguettes demi-rondes du
(1986), 62-82 & 175-77. Pa1eolithique superieur occidental: analyse et
MARSHACK, A. 1987. Early hominid symbol and reflexions. Antiquites Nationales 12/13, 7-19.
evolution of the human capacity. Paper for The MONS, L. 1986. Les statuettes animalieres en gres de
Origin and Dispersal of Modern Humans la grotte d'Isturitz (Pyr-Atl). Observations et
conference, Cambridge, March . hypotheses de fragmentation volontaire.
MARSHACK, A. 1988. La pensee symbolique et l'art, L'Anthropologie 90, 701-11.
in L'Homme de N eandertal, Dossiers de MONS, L. 1986/7. Les figurations de bisons dans l'art
1'Archeologie 124, 80-90. mobilier de la grotte d'Isturitz (Pyrenees-
MARTIN, Y. 1973. L'Art Paleolithique de Gouy. Atlantiques). Antiquites Nationales 18/ 19, 91-99.
Martin: Gouy. MOORE, D.R. 1977. The hand stencil as symbol, in
MARTIN, Y. 1974. Technique de moulage de Form in Indigenous Art (P.J. Ucko, ed.), 318-24.
gJ;il{Ures rupestres. Bull. Soc. Prehist. franraise 71, Duckworth: London.
146-48. MORI, F . 1974. The earliest Saharan rock-engravings.
MARTIN SANTAMARIA, E. & MOURE Antiquity48, 87-92.
ROMANILLO, J .A. 1981. El grabado de estilo MOURE ROMANILLO, J.A. 1985. Nouveautes dans
paleolitico de Domingo Garcia (Segovia). Trabajos l'art mobilier figuratif du Paleolithique cantabrique.
de Prehistoria 38, 97-105 , 2 pl. Bull. Soc. Prehist. Ariege 40, 99-129.
MOURE ROMANILLO, J.A. 1986. New data on the
chronology and context of Cantabrian Paleolithic
cave art. Current Anth. 27, p. 65 .
 BIBLIOGRAPHY 229

MOURE ROMANILLO, }.A. et al 1984/5. Las NOUGIER, L-R. & ROBERT, R . 1974. De
pinturas paleoliticas de la cueva de la Fuente del l'accouplement dans !'art prehistorique . Ibid. 29 ,
Salin (Mufi.orrodero, Cantabria). Ars Praehistorica 15-63.
III/IV, 13-23. NOVEL, P. 1986. Les animaux rares dans !'art parietal
MULLER-BECK, H. & ALBRECHT, G. (eds). 1987 . aquitain. Bull. Soc. Prehist. Ariege-Pyrenees41, 63-
Die Anfiinge der Kunst vor 30,000 ]ahren. Theiss: 93 (&in 42, 1987, pp . 83-118).
Stuttgart.
MURRAY, P. & CHALOUPKA, G. 1983/4. The OBERMAIER, H. 1918. Trampas cuaternarias para
Dreamtime animals: extinct megafauna in Arnhem espiritus malignos . Bol. Real Soc. esp. de Hist. Nat.
Land rock art. Archaeology in Oceania 18/19, 105- 18 , 162-69.
16. OLIVIE MARTINEZ-PENALVER, A. 1984/5.
MUZZOLINI, A. 1986. L'Art Rupestre Pn3historique Representaciones de cervidos en el arte parietal
des Massifs Centraux Sahariens. British Arch. paleolitico de Cantabria. Ars Praehistorica 314, 95-
Reports, Int. Series 318. Oxford. 110. '
OLLIERDEMARICHARD, P. 1973. Unpionnierde
NABER, F .B. et all976. L'Art Parietal Paleolithique la prehistoire ardechoise: Jules Oilier de Marichard
en Europe Romane. 3 vols. Bonner Hefte fi.ir (1824-1901 ). Etudes Prehistoriques 4, mars, 25-29 .
Vorgeschichte Nos. 14-16. OMNES, ]. 1982. La Grotte Ornee de Labastide
NARR, K.J. 1960. Weibliche symbol-plastik der (Hautes-Pyrenees). Omnes: Lourdes .
alteren Steinzeit. Antaios II (2), 132-57. OTTE , M. 1974. Observations sur le debitage et le
NOBBS, M . & DORN, R.I. 1988 . Pleistocene age fa<;onnage de l'ivoire dans 1' Aurignacien en
determinations for rock varnish within petroglyphs: Belgique. in ler Colloque Int. sur l'Industrie de l'Os
Calibrated dates from the Olary region of South dans la Prehistoire (H. Camps-Fabrer, ed .), 93-96 .
Australia. Rock Art Research 5 (2), in press. Univ. de Provence.
NOUGIER, L-R. 1972. Nouvelles approches de l'art
pn!historique animalier, in Santander Symposium, PALES, L. 1964. Preface, in Gaussen 1964, 11-1 6.
263-78 . PALES, L. 1969. Les Gravures de LaMarche: I, Felins
NOUGIER, L-R. 1975. L'importance du choix dans et Ours. Memoire 7, Publications de l'Inst. de
!'explication religieuse de !'art quaternaire, in Prehistoire de Bordeaux; Delmas: Bordeaux.
Valcamonica Symposium 1972: Les Religions de la PALES, L. 1970. Le 'Coco des Roseaux' ou Ia fin d'une
Prehistoire, 57-64. erreur. Bull. Soc. Prehist. franraise 67, 85-88.
NOUGIER, L-R. 1975a. Art prehistorique et PALES , L. 1972. Les ci-devant Venus steatopyges
topographie. Bull. Soc. Prehist. Ariege 30, 115-17. aurignaciennes, in Santander Symposium, 217-61.
NOUGIER, L-R. & ROBERT, R . 1957. Rouffignac, PALES, L. 1976/7. Les ovicaprides prehistoriques
ou la Guerre des Mammouths. La Table Ronde: franco-iberiques au nature! et figures . Sautuola II,
Paris. 67-105, 1 pl. Publicaciones del patronato de las
NOUGIER, L-R. & ROBERT, R. 1957a. Le cuevas preh. de Ia Prov. de Santander, XV.
rhinoceros dans l'art franco-cantabrique occidental. PALES, L. & de ST PEREUSE, M .T . 1965. En
Bull. Soc. Prehist. Ariege 12 , 15-52, 2 pl. compagnie de !'abbe Breuil devant les bisons graves
NOUGIER, L-R. & ROBERT, R . 1958. Le 'lissoir aux magdaleniens de Ia grotte de Ia Marche, in
Sai:gas' de 1a grotte de 1a Vache a Alliat et l'antilope MisceJanea en Homenaje al Abate H. Breuil, vol. II,
Sai:ga dans l'art franco-cantabrique . Ibid. 13, 13-28, 217-50 . Barcelona.
2 pl. PALES, L. & de ST PEREUSE, M.T. 1966. Un
NOUGIER, L-R. & ROBERT, R. 1960. Les '1oups cheval-pretexte: retour au chevetre. Objets et
affrontes' de 1a grotte de 1a Vache (Ariege) et 1es Mondes 6, 187-206.
canides ~s l'art franco-cantabrique, in Festschrift PALES, L. & de ST PEREUSE, M.T. 1976. Les
fiir Lothar Zotz, 399-420, 1 pl. Rohrscheid Verlag: Gravures de LaMarche: II, Les Humains. Ophrys:
Bonn. Paris .
NOUGIER, L-R. & ROBERT, R. 1965 . Les Felins PALES, L. & de ST PEREUSE, M .T. 1979. L'abri
dans l'art quaternaire. Bull. Soc. Prehist. Ariege 20, Durif a Enval (Vic-le-Comte , Puy-de-Dome).
17-84; and 21, 1966, 35-46, 2 pl. Gravures et sculptures sur pierre. Gallia Prehistoire
NOUGIER, L-R. & ROBERT, R. 1968. Les 22, 113-42.
processions, les associations et les juxtapositions PALES, L. & de ST PEREUSE, M.T. 1981. Les
d'anthropomorphes et d'animaux dans !'art Gravures de La Marche: Ill, Equides et Bovides.
quaternaire. Ibid. 23, 33-98. Ophrys: Paris.
230 IMAGES OF THE ICE AGE

PALES, L. & de ST PEREUSE, M.T. (forthcoming) PITTARD, E. 1929. La premiere decouverte d'art
Les Gravures de La Marche: IV, Cervides, prehistorique (gravure et sculpture) a ete faite dans Ia
Elephants et divers. station de Veyrier (Hte-Savoie) par le Genevois
PALES, L. et al 1976. Les Empreintes de Pieds Franc;ois Mayor. Revue Anth ., 39e annee, no. 7-9,
Humains dans Jes Cavernes. Archives de l'Inst. de 296-304.
Paleontologic Humaine 36, Paris. PLENIER, A. 1971. L'Art deJa Grotte de Marsoulas.
PARKINGTON, J. 1969. Symbolism in cave art. Memoire 1, In st. Art Prehist. Toulouse.
South African Arch. Bull. 24, 3-13. POND, A.W. 1925. The oldest jewelry in the world.
PASSEMARD, E. & BREUIL, H . 1928. La plus Art and Archaeology 19, 131-34, 1 pl.
grande gravure magdalenienne acontours decoupes. POPLIN, F. 1983. Incisives de renne sciees du
Rev. Arch. 27, 4pp. Magdalenien d'Europe occidentale. in La Faune et
PASSEMARD, L. 1938. Les Statuettes Feminines ]'Homme Prehistoriques, Mem. 16, Soc. Prehist.
Paleolithiques dites venus steatopyges. Teissier: franc;aise, 55-67.
Nimes. POWERS, R . & STRINGER, C.B. 1975. Palaeolithic
PEQUART, M. & S-J. 1939/40. Fouilles cave art fauna. Studies in Speleology2 (7/8), 265-98.
archeologiques et nouvelles decouvertes au Mas PRADEL, L. 1975. Les mains incompletes de Gargas,
d'Azil. L'Anthropologie 49, 450-53. Tibiran et Maltravieso. Quartar 26, 159-66.
PEQUART, M. & S-J. 1960/3 . Grone duMas d'Azil PRASLOV, N.D. 1985. L'art du Paleolithique
(Ariege) - Une nouvelle galerie magdalenienne. Superieur a I'est de !'Europe. L'Anthropologie 89,
Annales de Paleontologie, collected papers, 351pp. 181-92.
PERICOT, L. 1942. La Cueva del Parpall6 (Gandfa). PRAT, F. 1986. Le cheval dans I'art paleolithique et les
Consejo Sup. de Investigaciones Cientificas, Inst. donnees de la paleontologic. Arqueologia (Porto) 14,
Diego Velazquez: Madrid. dec., 27-33.
PERICOT, L. 1962. Un curioso paralelo. Munibe 14,
456-58. RAPHAEL, M. 1986. L'Art Parietal Paleolithique.
PERPERE, M. 1984. Les instruments de !'artiste, in Kronos: Paris.
Les Premiers Artistes, Dossier de !'Arch. 87,41-44. REGNAULT, F. 1906. Empreintes de mains
PFEIFFER, }.E. 1982. The Creative Explosion. An humaines dans la grotte de Gargas (Hautes
inquiry into the origins ofart and religion. Harper & Pyrenees). Bull. Soc. Anth. 5e serie, 7, 331-32.
Row: New York. REINACH, S. 1899. Gabriel de Mortillet. Revue
PIETTE. E. 1887. Equides de la periode quaternaire Historique 24e annee, 69, 67-95.
d'apres les gravures de ce temps. Materiaux 3e serie, REINACH, S. 1903. L'art et la magic. A propos des
4, 359-66. peintures et des gravures de 1' Age du Renne.
PIETTE, E. 1889. L'epoque de transnwn L'Anth. 14, 257-66.
intermediaire entre l'age du Renne et l'epoque de la RICE, P.C. 1981. Prehistoric venuses: symbols of
pierre polie. Comptes rendus du JOe Congr. Int. motherhood or womanhood? journal of Anth.
d'Anth. et d'Arch. Prehist., Paris (publ. 1891), 203- Research 37, 402-14.
13. RICE, P.C. & PATERSON, A.L. 1985. Cave art and
PIETTE, E. 1894. Notes pour servir al'histoire de I' art bones: exploring the interrelationships . American
primitif. L'Anthropologie 5, 129-46. Anthropologist 87, 94-100.
PIETTE, E. 1895. La station de Brassempouy et les RICE, P.C. & PATERSON, A.L. 1986. Validating the
statuettes humaines de la periode glyptique. L'Anth. cave art - archeofaunal relationship in Cantabrian
6,129-51, 7pl. Spain. Ibid. 88, 658-67.
PIETTE, E. 1902. Gravure duMas d'Azil et statuettes RIDDELL, W.H. 1940. Dead or alive? Antiquity 14,
de Menton. Bull. Soc. Anth. Paris 5e serie, 3, 771- 154-62.
-iJ.J. RIPOLL PERELLO, E. 1964. Problemas
PIETTE, E. 1905. Les ecritures de l'Age glyptique . cronol6gicos del arte paleolitico, in Prehistoric Art of
L'Anth. 16, 1-11. the Western Mediterranean and the Sahara (L.
PIETTE, E. 1906. Le chevetre et la semi-domestication Pericot & E. Ripoll, eds), 83-100. Viking Fund
des animaux aux temps pleistocenes. Publications in Anth. 39, New York.
L'Anthropologie 17, 27-53. RIPOLL PERELLO, E. 1972. La Cueva de las
PIETTE, E. 1907. L'Art pendant l'Age du Renne. Monedas en Puente Viesgo (Santander).
Masson: Paris. Monografias de Arte Rupestre, Arte Paleolitico No.
1, Barcelona.
 BIBLIOGRAPHY 231

RIPOLL PERELLO , E. 1984. Notes sur certaines ROUSSOT, A. 1970. Flutes et sifflets paleolithiques en
representations d'animaux dans l'art paleolithique Gironde. Rev. Historique de Bordeaux et du Dept.
de la peninsule iberique, in Bandi et ali984 , 263-82. de la Gironde, 5-12 .
RIVENQ, C. 1976. La 'Scene de Chasse' de Ganties- ROUSSOT, A. 1972/3. La decouverte des gravures de
Montespan. Rivenq: Toulouse. Pair-non-Pair d'apres les notes de Fran<;ois Daleau.
RIVENQ, C. 1984. Grotte de Ganties-Montespan, in Cahiers du Vitrezais I , july, 5-7, & oct, 15-17; 2,
L 'Art des Cavernes. A tlas des grottes ornees april, 22-24 .
paleolithiques fram;aises, 438-45. Min . de la ROUSSOT, A. 1981. Observations sur le coloriage de
Culture: Paris . sculptures paleolithiques. Bull. Soc. Prehist. fr. 78,
ROBERT, R. et ali953. Sur !'existence possible d'une p. 200.
ecole d'art dans le Magdalenien pyreneen. Bull. ROUSSOT, A. 1984. Les premiers bas-reliefs de
Arch. 187-93. l'humanite, in Les Premiers Artistes, Dossier de
ROBINEAU , D. 1984. Surles mammiferes marins du !'Arch. 87,73-76.
baton grave prehistorique de Montgaudier. L 'Anth . ROUSSOT, A. 1984a. Peintures, gravures et
. 88, 661-64. sculptures de l'abri du Poisson aux Eyzies. Quelques
RONEN, . A. & BARTON, G.M. 1981. Rock nouvelles observations. Bull. Soc. Prehist. Ariege
engravings on western Mount Carmel, Israel. 39, 11-26.
Quartiir 31/32, 121-37. ROUSSOT, A. 1984b. Approche statistique du
ROPER, M.K. 1969. A survey of the evidence for bestiaire figure dans 1'art parietal. L 'Anthropologie
intrahuman killing in the Pleistocene. Current Anth. 88, 485-98.
10, 427-59. ROUSSOT, A. 1984c. La rondelle 'aux chamois' de
ROSENFELD , A. 1977. Profile figures: Laugerie-Basse, in Elements de Pre et Protohistoire
schematisation of the human figure in the europeenne, Hommages a ]-P. Millotte, Annales
Magdalenian culture of Europe, in Form in Litt. Univ. de Besan<;on, 219-31.
Indigenous Art (P .]. Ucko, ed.), 90-109. ROUZAUD, F. 1978. La Paleospeleologie. L'Homme
Duckworth: London. et le Milieu Souterrain pyreneen au Paleolithique
ROSENFELD , A. 1984. The identification of animal superieur. Archives d'Ecologie Prehist. 3,
representations in the art of the Laura region, North Toulouse .
Queensland (Australia), in Bandi et al 1984, 399- ROZOY, J.G. 1985. Peut-on identifier les animaux de
422. Roc-la-Tour I? Anthropozoologica 2, 5-7.
ROSENFELD , A., HORTON , D. & WINTER, J. RUSPOLI, M . 1987. The Cave of Lascaux: The Final
1981. Early Man in N orth Queensland. Terra Photographic Record. Thames and Hudson:
Australis 6, Australian National University: London.
Canberra. RUSSELL, P .M. 1987. Women in Upper Palaeolithic
ROTTLANDER, R. 1965. Zur Frage des Europe. M.A. Dissertation, University of
Pigmentbinders der Franko-Kantabrischen Auckland.
Hohlenmalereien. Fundamen ta Reihe A, Band 2,
340-44. SACCHI, D ., ABELANET, ]. & BRULE, J-L. 1987.
ROUSSEAU, M. 1967. L es Gran ds Felins dans l'Art Le rocher grave de Fornols-Haut. Archeologia 225,
de notre Prehistoire. Picard : Paris. juin, 52-57.
ROUSSEAU, M . 1973. Darwin et les chevaux peints SAHLY, A. 1963. Nouvelles decouvertes dans la grotte
paleolithiques d'Ekain . Bull. Soc. Prehist. Ariege de Gargas. Bull. Soc. Prehist. Ariege 18,65-74,2 pl.
28 , 49-55 . de SAINT MATHURIN , S. 1958. Rouffignac, ses
ROUSSEAU, M. 1974. Recentes determinations et textes, ses plans. Bull. Soc. Preh. fr. 55, 588-92 .
refutations de 'Felin' dans l'art paleolithique. de SAINT MATHURIN, S. 1971. Les biches du
SiiugetittKkundliche Mitteilungen 2, 97-103. Chaffaud (Vienne). Vicissitudes d'une decouverte .
ROUSSEAU , M. 1984. Torsions conventionnelles et Antiquites Nationales 3, 22-28 .
flexions naturelles dans l'art animalier paleolithique de SAINT MATHURIN, S. 1973. Bas-relief et
et au dela, in Bandi et ali984, 243-49. plaquette d'homme magdalenien d'Angles-sur-
ROUSSEAU, M. 1984a. Les pelages dans l'Anglin. Antiquites Nationales 5, 12-19.
l'iconographie paleolithique, in Ibid. 161-97. de SAINT MATHURIN, S. 1975. Reliefs
RQUSSOT, A. 1965 . Les decouvertes d'art parietal en magdaleniens d' Angles-sur-1' Anglin (Vienne).
Perigord. in Centenaire de la Prehistoire en Perigord Antiquites Nationales 7, 24-31.
(1864-1964), 99-125. Fanlac: Perigueux.
232 IMAGES OF THE ICE AGE

de SAINT MATHURIN , S. 1978. Les 'Venus' SIMEK, J. 1986. A Paleolithic sculpture from the abri
parietales et mobilieres du Magdalenien d'Angles- Labattut in the American Museum of Natural
sur-1' Anglin. Antiquites Nationales 10, 15-22. History collection . Current Anth. 27, 402-07.
de SAINT PERIER, R. I930 . La Grone d'Isturitz, I: SIMONNET, G. 1952. Une belle parure
Le Magdalenien de la Salle de St-Martin. Archives magdalenienne, in 12e Congd:s Prehist. de France,
de l'Inst. Pal. Humaine 7, Paris . Paris 1950, 564-68.
de SAINT PERIER, R . I936 . La Grone d'Isturitz, II: SIMONNET, G., L. & R. 1984. Quelques beaux
Le Magdalenien de la Grande Salle. Archives de objets d'art venant de nos recherches dans la grotte
l'Inst. Pal. Humaine I7, Paris. ornee de Labastide (H -P). Approche naturaliste.
de SAINT PERIER , R-S . I965 . Inventaire de l'art Bull. Soc. Meridionale de Speleol. et Prehist. 24, 25-
mobilier paleolithique du Perigord. in Centenaire de 36.
la Prehistoire en Perigord (1864-1964), 139-59. SIMONNET, R. 1980. Emergence de la prehistoire en
Fanlac: Perigueux. pays ariegeois. Apen;u critique d'un siecle de
de SAUTUOLA, M.S. 1880. Breves apuntes sobre recherches. Bull. Soc. ariegeoise Sciences, Lettres et
algunos objetos prehist6ricos de la provincia de Arts 35, 5-88.
Santander. Santander, 27pp, 4 pl. SMITH, N.W. (in press) A Psychologyoflce Age Art:
SAUVET, G. I983. Les representations d'equides An Analysis. Stanford Univ. Press.
paleolithiques de la grotte de la Griega (Pedraza, SOFFER, 0. 1985. The Upper Paleolithic of the
Segovia). A propos d'une nouvelle decouverte. Ars Central Russian Plain. Academic Press: Orlando.
Praehistorica 2, 49-59 . SOFFER, 0. 1985a. Patterns of intensification as seen
SAUVET, G. & S. I979 . Fonction semiologique de from the Upper Paleolithic of the Central Russian
l'art parietal animalier franco-cantabrique . Bull. Plain, in Prehistoric Hunter-Gatherers. The
Soc. Prehist. fr. 76 , 340-54. Emergence of Cultural Complexity (T.D. Price &
SAUVET, G. & S., & WLODARCZYK, A. I977. ].A. Brown, eds), 235-70. Academic Press: New
Essai de semiologie prehistorique (pour une theorie York/London.
des premiers signes graphiques de l'homme). Ibid. SOHN POW-KEY 1974. Palaeolithic culture of Korea.
74, 545-58. Korea Journal, april, 4-11.
SCHMID, E. I984. Some anatomical observations on SOHN POW-KEY 1981. Inception of art mobilier in
Palaeolithic depictions of horses , in Bandi et ali984, the Middle Palaeolithic period at Chommal Cave,
155-60. Korea. in Resumenes de Comunicaciones,
SHIMKIN, E.M. I978. The Upper Paleolithic in Paleolftico Medio, Xth Congress UISPP, Mexico
North-Central Eurasia: evidence and problems, in City, 31-32.
Views of the Past (L.G . Freeman, ed.), I93-3I5. de SONNEVILLE-BORDES, D. 1986. Le bestiaire
Mouton: The Hague. paleolithique en Perigord. Chronologie et
SIEVEKING, A. I971. Palaeolithic decorated bone signification. L'Anthropologie 90, 613-56.
discs. British Museum Quarterly 35, 206-29, II pl. de SONNEVILLE-BORDES, D. & LAURENT, P.
SIEVEKING, A. 1978. La significaci6n de las 1983. Le phoque ala fin des temps glaciaires, in La
distribuci6nes en el arte paleolitico. Trabajos de Faune et l'Homme Pd:historiques, Mem. Soc.
Prehistoria 35, 61-80 . Prehist. fr. 16, 69-80.
SIEVEKING, A. I979 . Th e Cave Artists. Thames & de SONNEVILLE-BORDES, D. & LAURENT, P.
Hudson: London. 1986/7. Figurations de chevaux a l'abri Morin.
SIEVE KING, A. 1979a. Style and regional grouping in Observations complementaires. Antiquites
Magdalenian cave art. Bull. Inst. of Arch. 16, 95- Nationales 18/ 19, 69-74.
109. SOURIAU, E. 1971. Art prehistorique et esthetique du
SIEVEKING, A. 1983 . Decorated scapulae from the mouvement, in Melanges A. Varagnac, 697-705.
~per Palaeolithic of France and Cantabrian Spain, Sevpen: Paris.
in Homenajeal Prof M . AlmagroBasch, vol. 1, 313- SPECK, F.G. & SCHAEFFER, C.E. I950. The deer
17. Min. de Cultura : Madrid. and rabbit hunting drive in Virginia. Southern
SIEVEKING, A. 1987. Engraved Magdalenian Indian Studies 2, 3-20.
Plaquettes. British Arch . Reports Int. series No. STEVENS, A. 1975. Animals in Palaeolithic cave art:
369: Oxford . Leroi-Gourhan's hypothesis. Antiquity49, 54-57.
SIEVE KING, A. & G. 1962 . The Caves of France and STEVENS, A. 1975a. Association of animals in
Northern Spain. Vista Books: London. Palaeolithic cave art: The second hypothesis.
SIEVEKING, G. 1972. Art mobilier in Britain, m Science and Archaeology I6, 3-9.
Santander Symposium, 385-88. '
 BIBLIOGRAPHY 233

STOLlAR, A.D. 1977/8. On the genesis of depictive UCKO, P.J. & LAYTON, R. 1984. La subjectivite et
activity and its role in the formation of consciousness le recensement de l'art paleolithique. Typescript for
(Toward a formulation of the problem). Soviet Anth. Colloque Int. d'Art Parietal Paleolithique,
& Arch . 16, 3-42; & 17 ( 1978/9), 3-33. Perigueux. 39pp.
STOLlAR, A.D. 1977/8a . On the sociohistorical UCKO, P.J. & ROSENFELD, A. 1967. Palaeolithic
decoding of Upper Paleolithic female signs. Ibid. 16, Cave Art. World Univ. Library: London.
36-77 . UCKO, P.J. 1987. Debuts illusoires dans l'etude de la
STOLlAR, A.D. 1981. On the archeological aspect of tradition artistique. Bull. Soc. Prehist. Ariege-
the problem of the genesis of animalistic art in the Pyrenees 42, 15-81.
Eurasian Paleolithic. Ibid . 20, 72-108. UCKO, P.J. & ROSENFELD, A. 1972.
STOLlAR, A.D. 1985 . Proisxozdenie izobrazitel'nogo Anthropomorphic representations in Palaeolithic
iskusstva (The origin of figurative art). Iskusstvo: art, in Santander Symposium 149-215.
Moscow. UTRILLA, P. 1979. Acerca de la postcton
STRAUS, L.G. 1982 . Observations on Upper estratignifica de los cervidos y otros animales de
Paleolithic art: old problems and new directions. trazo multiple en el Paleolitico Superior espaii.ol.
Zephyrus 3415, 71-80. Caesaraugusta 49150, 65-72.
STRAUS , L.G. 1987. The Paleolithic cave-art of
Vasco-Cantabrian Spain. Oxford journal ofArch. 6, VALLO IS, H.V. 1961. The social life of early man: the
149-63. evidence of skeletons, in Social Life of Early Man
(S.L. Washburn, ed.), 214-35. Viking Fund
TABORIN, Y. 1977 . Quelquesobjetsdeparure. Etude Publics. No. 31, New York.
technologique: les percements des incisives de VALOCH, K. 1970~ Oeuvres d'art et objets en os du
bovines et des canines de renards . Methodologie Magdalenien de Moravie (Tchecoslovaquie). Bull.
Appliquee a l'Industrie de J'Os Prehistorique (H. Soc. Prehist. Ariege 25, 79-93.
Camps-Fabrer, ed), 303-10. C. N.R.S.: Paris. VAN NOTEN, F. 1977. Excavations at Matupi Cave.
TABORIN, Y. 1982. La parure des morts. La Mort Antiquity 51, 35-40.
dans Ja Prehistoire, Dossier del' Archeologie 66, 42- VANDIVER, P. (in press) Paleolithic Pigment
51. Processing: a soft stone technology. Univ. of
TABORIN, Y. 1985. Les origines des coquillages Chicago Press.
paleolithiques en France . in La Signification VASIL'EV, S.A. 1985. Une statuette d'argile
culturelle des Industries Lithiques (M. Otte, ed), paleolithique de Siberie du sud. L 'Anthropologie
278-301. British Arch. Reports Int. series No. 239: 89, 193-96.
Oxford. VAYSON de PRADENNE, A. 1934. Les figurations
TARASSOV, L.M. 1971. La double statuette d'oiseaux dans l'art quaternaire. I.P.E.K. 3-17.
paleolithique de Gagarino. Quartiir22, 157-63, 1 pl. VERTUT, J. 1980. Contribution des techniques
THEVENIN, A. 1983 . Les ga1ets graves et peints de photographiques a l'etude eta la conservation de I' art
l'abri de Rochedane (Doubs) et le probleme de l'art prehistorique, in Altamira Symposium, 661 -75.
azi1ien. Gallia Prehistoire 26, 139-88. Min. de Cultura: Madrid.
TOSELLO, G. 1983. In ven taire des Sites VEZIAN, J. 1956. Les utilisations de contours de la
Paleolithiques d 'Art Parietal et Mobilier Figuratif. roche dans la grotte du Portel. Bull. Soc. Prehist.
Maitrise de Prehistoire, 2 vols. Univ. de Paris I, Ariege 11, 79-87, 4 pl.
Pantheon-Sorbonne. VIALOU, D. 1976. Guide des Grottes Ornees
TWIESSELMANN, F. 1951. Les Representations de Paleolithiques Ouvertes au Public. Masson: Paris.
l'Homme et des Animaux Quaternaires decouvertes VIA LOU, D. 1981. L'art prehistorique: questions
en Belgique. Memo ire 113 , Institut Royal des d'interpretations. Monuments Historiques 118,
..SCience~aturelles de Belgique: Brussels . Nov/Dec., 75-83.
TYLDESLEY, }.A. & BAHN , P.G. 1983. Use of VIALOU, D. 1982. Une lecture scientifique de l'art
plants in the European Palaeolithic: a review of the prehistorique. La Recherche 13, 1484-87.
evidence. Quaternary Science R eviews 2, 53-81. VIALOU, D. 1982a. Niaux, une construction
symbolique magdalenienne exemplaire. Ars
UCKO , P.J . 1977 . Opening remarks, in Form in Praehistorica 1, 19-45.
Indigenous Art, (P.J . Ucko, ed.), 7-10. Duckworth: VIALOU, D. 1984. Les cervides de Lascaux, in Band!
London. et all984, 199-216.
234 IMAGES OF THE ICE AGE

VIALOU, D. 1986. L 'Art des Grottes en Ariege WILDGOOSE, M., HAD INGHAM , E. & HOOPER,
Magdalenienne. XXIIe Supplement a Gallia A. 1982 . The prehistoric hand pictures at Gargas:
Prehistoire, 28 pl. attempts at simulation. Medical History26, 205-7 .
VIALOU, D. 1987. D'un tectiforme al'autre, in Sarlat WORSAAE, J.J.A. 1869. Communication ... relative a
en Perigord, Actes du 39e Congr . d'Etudes l'authenticite des os de renne presentant des dessins
Regionales, Suppl. to Bull. Soc. Hist. et Arch. du et trouves en France dans les cavernes du Perigord.
Perigord 114, 307-17. Congres Int. d'Anth. et d'Arch. Prehistoriques, 4e
session, Copenhagen , 127-34. (Publ. 1875).
WAKANKAR, V.S . 1984. Bhimbetka and dating of WRESCHNER, E. 1975. Ochre in prehistoric
Indian rock paintings, in Rock Art of India (K.K. contexts. Remarks on its implications to the
Chakravarti, ed. ), 44-56. Arnold-Heinemann: New understanding of human behavior. Mi'Tekufat
Delhi. Ha'Even 13, 5-11.
WAKANKAR, V.S . 1985. Bhimbetka: the stone tool WRESCHNER, E. 1980. Red ochre and human
industries and rock paintings, in Recent Advances in evolution: a case for discussion. Current Anth. 21,
Indo-Pacific Prehistory (V .N. Misra & P. Bellwood, 631-44.
eds .), 175-76. Oxford & IBH Publishing Co: New WRIGHT, B. 1985. The significance of hand motif
Delhi. variations in the stencilled art of the Australian
WALKER, N.J. 1987 . The dating of Zimbabwean Aborigines. Rock Art Research 2, 3-19.
rock art . Rock Art R esearch 4, 137-49. WRIGHT, R.V.S. (ed.) 1971. Archaeology of the
WALSH, G.L. 1979. Mutilated hands or signal Gallus Site, Koonalda Cave. Australian Institute of
stencils. Australian Archaeology 9, 33-41 . Aboriginal Studies: Canberra.
WEISSEN-SZUMLANSKA, M. 1951. A propos des
gravures et peintures rupestres. Bull. Soc. Prehist. YATES, F.A. 1966. The Art of Memory. Univ. of
fr. 48, 457-59. Chicago Press.
WELTE, A-C. 1975/6. L' Affrontement dans I' art YOU YUZHU 1984. Preliminary study of a
prehistorique. Travaux de l'Inst. d'Art Prehist. Palaeolithic bone engraving. Kexue Tongbao 29 (1),
Toulouse 17,207-305 & 18, 187-330. 80-82.
WENDT, W.E. 1974. 'Art mobilier' aus der Apollo 11 -
Grotte in Si.idwest-Afrika. Acta Praehistorica et ZERVOS, C. 1959. L'Art de l'Epoque du Renne en
Archaeologica 5, 1-42. France. Cahiers d'Art: Paris.
WENDT , W.E. 1976. 'Art mobilier' from the Apollo ZEUNER, F.E . 1954. The colour of the wild cattle of
11 cave , South West Africa: Africa's oldest dated Lascaux. Man 53, 68-69.
works of art. South African Archaeological Bulletin ZHUROV, R.I. 1977/8. On one hypothesis relating to
31 ' 5-11. the origin of art. Soviet Anth. & Arch. 16, 43-63 .
WHITE , R. 1986. Dark Caves, Bright Visions, Life in ZHUROV, R.I. 198112. On the question of the origin
Ice Age Europe. Exhibition Catalogue, American of art (an answer to opponents). Soviet Anth. &
Museum of Natural History: New York. Arch. 20 (3), 59-82.
WHITNEY, D.S. & DORN, R .I. 1987. Rock art ZUCHNER, C. 1975. Der Bison in der eiszeitlichen
chronology in eastern California. World Arch. 19 Kunst Westeuropas . Madrider Mitteilungen 16,
(2), 150-64. 9-24.
 Illustrations

I Bison in the Reseau Clastres (Ariege) frontispiece 26 Infra-red picture of painted tectiform at Bernifal
2 Jean Vertut photographing the 'weasel' in the (Dordogne) so
Reseau Clastres (Ariege) 6 27 Drawing offull rock face and horse engravings at
3 Prehistoric footprints in Niaux (Ariege) I4 Montespan (Haute Garonne) 51
4 Prehistoric footprints in the Reseau Clastres 28 Painted panel at Tete-du-Lion (Ardeche) 55
(Ariege) IS 29 Part of painted panel at Tito Bustillo (Asturias) 59
5 Decorated objects from Veyrier (Haute Savoie): 30 Photomontage of paintings at Lascaux (Dordogne) 63
Musee d'Art et d'Histoire, Geneva I7 31 Photomontage of paintings at Lascaux (Dordogne) 63
6 Reindeer bone from Chaffaud (Vienne): Musee des 32 Photomontage of paintings at Lascaux (Dordogne) 63
Antiquites Nationales, St Germain-en-Laye I8 33 Comparison of three dating systems 65
7 Photomontage of the Altamira ceiling (Santander) I9 34 Engraved 'bull-roarer' from La Roche at Lalinde
8 Photomontage of the painted panel at Marsoulas (Dordogne): Musee des Antiquites Nationales,
(Haute Garonne) 22 St Germain-en-Laye 70
9 Mammoth engraving from Holly Oak (Delaware): 35 Decorated teeth from StGermain-la-Riviere
National Museum of Natural History, Smithsonian (Gironde): Musee des Eyzies 72
Institution, Washington DC 27 36 Engraved plaquette from Le Puy de Lacan
IO Engraved eggshell from Patne (India) 27 (Correze): Musee de Brive 75
II Engraved pebbles from Kamikuroiwa (Japan): 37 Engraved plaquette from Trou de Chaleux
National Museum of Japanese History, Chiba 28 (Belgium): Institut Royal des Sciences Naturelles,
I2 Engravings in Early Man shelter, Queensland 29 Brussels 76
13 Large engraved circle in Paroong Cave, S. Australia 30 38 Decorated mammoth skull from Mezhirich
I4 Deeply engraved circles in Paroong Cave, (USSR): Kiev 78
S. Australia 30 39 Engraved shoulder-blade from Duruthy (Landes):
IS Digital tracings, Karlie-ngoinpool Cave, S. Australia 3I Musee d' Arthous 79
I6 Engraved circles and flint extraction , Karlie- 40 Decorated antler baton from Duruthy (Landes):
ngoinpool Cave, S. Australia 3I Musee d' Arthous 79
I7 Carving of two reindeer from Bruniquel (Tarn et 4I Carving of bison calf, from Le Grand-Pastou
Garonne): British Museum 34 (Landes): Musee d'Arthous 79
I8 Antler spear-thrower carved in the shape of a 42 Engraved bird bone from Torre (Guipuzcoa):
mammoth, from Bruniquel (Tarn et Garonne): Museo San Telmo, San Sebastian 80
British Museum 34 43 Engraved bone disc from Le Mas d'Azil (Ariege) :
I9 The young Henri Breuil 4I Musee duMas d'Azil 82
20 'Spot the Breuil', engraved ibex from Castillo 44 Shoulder-blade with disc extracted, from Le Mas
(Sant:wll.er) 43 d'Azil (Ariege): Musee duMas d'Azil 82
2I 'Spot the Breuil', engraved horse from Trois 45 'Contour decoupe' ofhorsehead from Enlene
Freres (Ariege) 43 (Ariege): Musee de Pujol 82
22 Tracing of engraved horse from La Marche 46 Antler spear-thrower carved in the shape of a fawn
(Vienne): Musee de !'Homme, Paris 45 with emerging turd, from Le Mas d'Azil (Ariege):
23 Pioneering photography in Trois Freres (Ariege) 46 Musee duMas d'Azil 83
24 Ultraviolet picture of claviform in Trois Freres 47 Detail of antler spear-thrower from Le Mas d' Azil
(Ariege) 47 (Ariege): Musee duMas d'Azil 83
25 Tracing of plaquette from Gonnersdorf (Germany), 48 Stone statuette of kneeling horse from Duruthy
with horse figure extracted: Landesamt fur (Landes): Musee d'Arthous 84
Denkmaipflege von Rheinland-Pfalz , Aussenstelle 49 Ivory statuette of human with feline head, from
Koblenz 49 Hohlenstein-Stadel (Germany): Ulm Museum 85
236 IMAGES OF THE ICE AGE

SO Ivory figurines joined at the head , from Gagarino 89 Photomontage of decorated panels at
(USSR) 86 Cougnac (Lot) 130-1
51 Horsehead carved in limestone, from Duruthy 90 The headless bears of Ekain (Guipuzcoa) 130
(Landes): Musee d'Arthous 86 91 Bas-relief mammoth in the Grone du Mammouth,
52 Human drawn around a stalagmite, at Le Porte! Domme (Dordogne) 131
(Ariege) 87 92 Engraved owls at Trois Freres (Ariege) 134
53 Vertical bison painted at Castillo (Santander) 87 93 Bearded man engraved on plaquette from
54 Engraved motifs ('vulvas') on block from abri LaMarche (Vienne): Collection Pradel,
Cellier (Dordogne): Musee des Eyzies 88 Musee de Lussac 136
55 Deer head in the Salon Nair , Niaux (Ariege) 90 94 The 'Venus' ofLespugue (Haute Garonne):
56 Detail of engraved owl in Trois Freres (Ariege) 90 Musee de !'Homme, Paris 139
57 Engraved vertical horsehead in clay at Montespan 95 Stylised humans engraved on plaquette from
(Haute Garonne) 91 Gonnersdorf (Germany): Landesamt fiir
58 'Kneeling' reindeer engraved in Trois Freres Denkmalpflege von Rheinland-Pfalz, Aussenstelle
(Ariege) 91 Koblenz 140
59 Digital tracings at Altamira (Santander) 92 96 Schematised females and possible child on
60 Clay bas-reliefs at Bedeilhac (Ariege) 94 plaquette from Gonnersdorf (Germany):
61 Wide-angle view of the clay bison of the Landesamt fur Denkmalpflege von Rheinland-
Tuc d'Audoubert (Ariege) 95 Pfalz, Aussenstelle Koblenz 141
62 Detail of one of the clay bison of the 97 Photomontage of the 'sorcerer' from Trois Freres
Tuc d'Audoubert (Ariege) 95 (Ariege) 143
63 View of the 'dark side' of the clay bear of 98 'Signs' in Castillo (Santander) 147
Montespan (Haute Garonne) 98 99 'Tectiform' signs on bison at Font de Gaume
64 View of the front of the clay bear of Montespan (Dordogne) 151
(Haute Garonne) 98 100 Females depicted on one side of the engraved bone
65 The sculptured horsehead ofComarque (Dordogne) 98 from Isturitz (Pyrenees Atlantiques): Musee des
66 Detail of the horsehead of Comarq ue (Dordogne) 99 Antiquites Nationales, St Germain-en-Laye 154
67 'Venus with horn' from Laussel (Dordogne): 101 Bison depicted on the other side of the engraved
Musee d'Aquitaine, Bordeaux 99 bone from Isturitz (Pyrenees Atlantiques): Musee
68 Red pigment in the cave of Cougnac (Lot) 99 des Antiquites Nationales, St Germain-en-Laye 154
69 Photomontage of the sculptured frieze of Cap 102 Humanoid and 'missiles' in Pech Merle (Lot) 154
Blanc (Dordogne) 102 103 Carved antler showing three horseheads,
70 Three hinds in Covalanas (Santander) 102 from Le Mas d-Azil (Ariege): Musee des
71 Incomplete hand stencils at Gargas Antiquites Nationales, St Germain-en-Laye 155
(Hautes Pyrenees) 103 104 Head of composite figure in Trois Freres (Ariege) 158
72 Hand and forearm stencils at Fuente del Salin 105 Sculptured frieze of La Chaire aCalvin (Charente) 159
(Santander) 104 106 Humanoids on plaquette from Enlene (Ariege):
73 Photomontage of the 'black frieze' at Pech Musee de Pujol, and Musee de !'Homme, Paris 160
Merle (Lot) 106-7 107 'Copulation' scene at Los Casares (Guadalajara) 161
74 Open-air engraved horse at Mazouco (Portugal) 111 108 Two versions of a plaquette from Enlene (Ariege):
75 Open-air engraved horse at Domingo Garcia Musee de !'Homme, Paris 162
(Segovia) 112 109 'Playing card' figure from Laussel (Dordogne):
76 Engraved rock at Fornols-Haut (Pyrenees- Musee d'Aquitaine, Bordeaux 163
Orientales) 113 110 Central composition at Santimamine (Vizcaya) 166
77 Ibex head engraved at Fornols-Haut (Pyrenees- 111 Leroi-Gourhan's 'blueprint' for cave layout 167
Orientales) 114 112 The 'bison-women' of Pech Merle (Lot) 170
78 'Antelope' figures at Pech Merle (Lot) 118 113 Hand stencil and dots at Pech Merle (Lot) 17i
79 Three bison in the Reseau Clastres (Ariege) 119 114 Photomontage of the horse panel at Ekain
80 Antler baton carving of horsehead , from Le Mas (Guipuzcoa) 174-5
d'Azil (Ariege): Musee du Mas d'Azil 122 115 Tracing of the 'map' from Mezhirich (USSR): Kiev 181
81 Detail of the antler baton from Le Mas d'Azil 116 The 'spade' sign at Le Porte! (Ariege) 183
(Ariege): Musee duMas d'Azil 122 117 Tracing of horses at Lascaux (Dordogne) 185
82 Engraved and painted bison in Trois Freres 118 Photomontage of the 'shaft scene' at Lascaux
(Ariege) 123 (Dordogne) 186-7
83 Glifo.ving of a bison pursuing a human , Roc de Sers 119 Jean Vertut in the Salon Nair, Niaux (Ariege) 194
(Charente): Musee des Antiquites Nationales, 120 General view of the clay bison of Le Tuc
St Germain-en-Laye 123 d'Audoubert (Ariege) 199
84 'Swimming' deer in Lascaux (Dordogne) 125 121 Rear view of one of the clay bison of Le Tuc
85 Detail of engraved aurochs cow's head from d'Audoubert 199
Teyjat (Dordogne) 126 122 View looking down on the clay bison of Le Tuc
86 Engraved aurochs from Teyjat (Dordogne) 127 d' Audoubert 199
87 Engraved reindeer and fox in Altxerri (Guipuzcoa) 127
88 Engraving of felines from La Vache (Ariege):
Musee des Antiquites Nationales, MAPS
St Germain-en-Laye 129 Maps of Palaeolithic parietal art sites in Europe 36-8
 Index
Numbers in italics refer to pages on which illustrations appear

Aborigines, Australian, 30, 115, 88 , 210 Aurignacian, 16, 33, 55, 56, 60, 62, 64-6, Berthoumeyrou, Gaston & Edouard, 24
see also Aruma, Walbiri 68, 72-4,80,84,85,88,89,91,96, Bhimbetka (India), 27
Abrams, David, 195, 198 105, 140, 182, 196, 197,207,208 Bigourdane, La (Lot), 118
Absolon, Karel, 182 Aurochs, 22, 23, 76, 123, 124, 126, 127, Bilzingsleben (E. Germany), 80, 208
Acheulian, 69, 80, 204, 207 133 , 160, 167, 169, 172, 173, 176, 180 Birds, 17, 81, 82, 90,132,133,134, 135,
Africa, 27, 33, 70, 78, 83, 97, 152, 158 , Australia, 28-32,30,31, SO, 54, 74, 90, 140, 144, 152, 153, 186
190 91, 115, 144, 145, 150, 159, 164, 184, Bison, 2, 20,22-4,61, 64,89 94, 94, 95,
Aldene (Herault), 13, 91 190,204 117,119, 119-23123, 133, 135, 152,
Alfonso XII, King, 21 Austria , 68, 84 153,154,155, 159-60, 167-70, 172,
Altamira (Santander), 19, 20-5, 39-42, Azilian, 16, 23, 61, 64, 78, 81, 182 173, 176, 181, 198,199,201,202,205,
45, 57-9, 61, 62, 64, 66, 89, 92, 97, 210
101, 104, 108, 117-19, 122, 123, 132 , Bacho Kiro (Bulgaria), 72, 80 Blanchard, abri (Dordogne), 73, 74,
145, 148, 153, 155, 156, 159, 168, 176, Bacon Hole (Wales), 205 81, 89, 105, 182,207
186,189,192,196,200,201,203,209, Badanj Cave (Yugoslavia) , 39 Bocksteinschmiede (Germany), 71
213,214 Baikal, Lake (Siberia), 86 Border Cave (S. Africa), 27, 68, 80, 204
Altxerri (Guipuzcoa), 66, 89, 93, 107, Ballawine Cave (Tasmania), 29 Bouyssonie, Jean, 79, 169, 205
116, 124, 127, 180 Bara-Bahau (Dordogne), 91 Brassempouy (Landes), 65, 84-6, 136-8,
America,26,27, 145,155 Barma Grande (Liguria, Italy), 74 140,208
Anamorphosis, 108 Barriere, Claude, 11 Brazil, 27
Anati, Emmanuel, 204 Bas-relief, 56,60-2,71, 81, 89, 91, 93, Breuil, Henri, 25, 33, 41,41-5 , 49-50,
Anatolia, 35 94,96, 137,140,164,165 52,58-62,64,66, 77,85,90,93, 116,
Angles-sur-!' Anglin (Vienne), 56, 58 , 61, Basque Country (France/Spain), 19, 121 118, 119, 132, 136, 140, 144, 151, 153,
85, 96, 136, 137, 140, 144, 164, 179 Bastera, Cov_a_(Pyrenees Orientales), 39 162-4, 166, 170, 173, 176, 179, 181,
Anthropomorphs, see Humanoids Baume-Latrone, La (Gard), 52, 100, 184, 192,200,201,205,208,210
Antlers, 17, 82, 87, 89, 90, 108, 118, 108, 128, 209 Britain, 35
124, 125, 130, 144, 206, 208 Bayol, Grone (Gard), 97 see also England, Wye Valley
Apelhiniz, Juan-Maria, 179-80,214 Bears, 14, 17, 19, 81, 91, 95, 98, 117, Brno II (Czechoslovakia), 81, 86 , 136,
Apollo 11 Cave (Namibia), 27, 78, 204 129-30,130,151,152,167-9,201,205 138
Arabia, 27 Becov (Czechoslovakia), 69 Brouillet, Andre, 17
Arc_- ur-Cur-Yonne), 7, 71-3,97, 100 Bedeilhac (Ariege), 42, 43, 54, 73, 82-4, Bruniquel (Tarn-et-Garonne), 18, 34,
Ardeche(France),20, 156 93, 94, 94, 117, 164, 179,208 121
Arenaza (Vizcaya), 101, 180 Bednarik, Robert & Elfriede, 31,204 Bull-roarer, 69, 70
Argentina, 204 Begouen, Henri, 76, 77, 151, 173, 184 Buxu, El (Asturias), 57
Ariege (France), 39-40, 146 Begouen,Max,45,209
Arizona (USA), 190 Begouen, Robert, 196, 198,201,202 Cabre Aguilo, Juan, 161
Arnhem Land (N. Australia), 29 Belcayre (Dordogne), 62 California, 27
Arudy (Pyrenees Atlantiques), 82 Belgium, 35 Calixtus III, Pope, 19
Aruma, The, 150, 159 Belleforest, Fran~ois de, 203 Camp6me (Pyrenees Orientales),
Aslin, Geoffrey, 31 Berekhat Ram (Israel), 204 see Fornols-Haut
Asturias (Spain), 39, 82, 111 Bernifal (Dordogne), 48, 50, 101, 108,145 Candamo (Asturias), 11
238 IMAGES OF THE ICE AGE

Cantabria (Spain), 35 , 39, 57, 62 , 81, 82, Cubillas, Modesto, 20 Fronsac (Dordogne), 140, 144, 177
108 , 120, 121, 124, 133 , 145 , 146, Cuciulat (Romania), 39, 97 Fuente del Salin (Santander), 104, 104,
156,168, 172, 176, 178, 184, 189,211 Cullalvera, La (Santander), 146 105, 135
Cap Blanc (Dordogne), 56 , 96, Czechoslovakia, 83 Fuente del Trucho (Huesca), 104
102,103, 133, 176,209
Capitan, Louis, 25, 169 Daleau, Fran<;ois, 24 Gabillou (Dordogne), 57, 61, 119 137,
Carnot, Adolphe, 97 Daniel, Glyn, II, 205 144, 145, 157' 168, 170
Cartailhac, Emile, 20-5, 58, ISO , 156, Darwin, Charles, 121 Gagarino (USSR), 86, 86, 138
165, 179 Dechelette, Joseph, 108 Gare de Couze (Dordogne), 140
Casares, Los (Guadalajara), 161 , 161 Deer, see Red deer, Reindeer Gargas (Hautes-Pyrenees), 11, 19, 58, 61,
Castanet, abri (Dordogne), 73 Delluc, Brigitte & Gilles, 89, 196-8 90, 93, 97,103, 104, 105, 135, 156,
Castillo (Santander), 39-40, 43 , 55, 57, Delporte, Henri, 208 176,209
58, 61,66,79, 87, 93,135 145, 147, Detrieux, M., 204 Garrigou, Felix., 19, 20, 81, 146
148, 155, 172, 176, 177, 179, 191, 192 Deux Avens, Grotte des (Ardeche), 54 Geissenklosterle (Germany), 84, 128
Cau-Durban, David, 24, 25 Devil's Lair CW. Australia), 29, 74 Germany, 35, 84, 85, 177
Cellier, abri (Dordogne), 88 Discs, 81, 82, 82, 180,210,214 Gillen, F.J., ISO
Chabot (Gard), 20 , 24, 25, 57 Dolni Vestonice (Czechoslovakia), 73, Gironde (France), 116
Chaffaud (Vienne), 17, 18, 203 83, 129, 135, 136, 144, 156, 164 Glory, Andre., 60, 157, 197
Chaire-a-Calvin, La (Charente), 56 96, Domingo Garcia (Segovia), Ill, 112 Godulfo, abrigo de (Asturias), 39
159, 159, 203 Domme, Grottes de (Dordogne), 64, 91, Gonnersdorf (Germany), 33, 49, 53, 73,
Chamois, see Isard 131 76-8,81,88, 121, 128, 135, 137,140,
Champleve, 53, 80, 84, 96 see also Mammouth 141,141, 144, 156, 163, 165, 179
Chapelle aux Saints, La (Correze), 70 Dordogne, see Perigord Gourdan (Haute Garonne), 133
Charente (France), 73, 80 Durif, abri, see Enval Gouy (Seine-Maritime), 35, 92
Chatelperronian, 16, 60, 71 , 73 , 75, 85, Duruthy (Landes), 79, 81, 84, 84, 86, Grand-Pastou, Le (Landes), 79
100 121' 128 Gravettian, 16, 24, 54-8, 61, 62, 66-8, 72,
Cheval, Grotte du (Ariege), 39 73, 81,83, 85,120,135,140,206-8
Chichkino (Siberia), 10 Early Man shelter (Queensland), 28 29, Greze, La (Dordogne), 61, 62
Chimeneas, Las (Santander), 55 , 57, 32,56 Griega, La (Segovia), 108, Ill
145, 148, 176, 180,214 Ebbou (Ardeche), 20 Grimaldi (Liguria, Italy), 74, 86, 138, 163
China, 28, 74, 146 Eglises, Les (Ariege), 57
Chiron, Leopold, 20, 24 Ekain (Guipuzcoa), 89, 116, 117, 121, Hand stencils/prints, 29, 103, 104, 104,
Christy, Henry, 17, 53 130, 156,174,175, 177, 180, 181, 186 105, 135, 147, 163,171, 172, 176,
Chufin (Santander), 57 England, 33 178,201,209
Cierro, El (Asturias), 93 Enlene (Ariege), 15, 33, 40, 53, 58, 66, Harle, Edouard, 23, 25
Clastres, Reseau (Ariege), 2, 6, 15, 41, 75-8,82, 128, 133,160, 161, 162, Haua Fteah (Libya), 68
90, 117,119, 132,200 162,205 Hayonim cave (Israel), 33
Claviforms, 47, 56, 107, 140, 145, 146, En val (abri Durif), 77, 78, 207 Haza, La (Santander), 180
151' 169, 212 Erberua (Pyrenees Atlantiques), 14, 41, Higgs, Eric, 11
Clay, 13, 14, 60, 83,93-5,95, 98, 105 , 94, 105, 121 Hohlenstein-Stadel (Germany), 85, 85,
108, 160, 164, 168, 176, 177, 196,199, Escoural (Portugal), 39, 69 138, 144
201,202 Eskimos, 109, 124, 165 Holly Oak (Delaware), 26, 27
Cogul (Lerida), 60 Espelugues, Les, see Lourdes Hornblower, G.D., 206
Coles, John, 11 Eyzies, Grotte des (Dordogne), 120, 132, Horse, 17, 35, 82, 84, 84, 86, 93,98, 108
Colombiere, La (Ain), 45 , 48 , 54, 76, 156 111,111,112, 120, 121,122, 133, 135,
78, 124, 129, 130, 151, 152 , 183 Eyzies, Les (Dordogne), 25, 145 155, 156, 159, 167, 169, 172, 173,
Comarque (Dordogne) , 98-9 , 133, 140, 174,175, 176, 177, 181,201
156 Fabre, C., 97 Humanoids, 90, 129, 132, 135, 140-2 ,
Combarelles, Les (Dordogne), 25, 41, Felines, 84, 91, 93, 94, 129, 130,129, 140,141, 144, 152,154, 158,160, 161,
48 , 121' 168 167, 168, 172, 181,205,210 170, 172
Combe!, Le, see Pech Merle F errassie, La (Dordogne), 77, 80, 89, 207 Hungary, 35, 68, 70,97
Composites, 142-4,143, 157-9 , 158, 163, Fieux, Les (Lot) 69 Hunting, 60, 61, 142, 149, 150-3, 155-9,
180 Figuier, Grotte du (Ardeche), 24 164, 165, 182, 183, 193,210,213
Conkey, Meg, 146 Fish, 74,116,132,133,178,211
Co~rs decoupes, 54, 81, 82, 82, 87 Flutes, 68, 69 Ibex, 17,24,43,61,82,83, 113,114,
Correze (France) , 109 Font-Bargeix (Dordogne), 57 116, 125, 128,130, 131, 133, 156, 157,
Cottes, Les (Vienne), 104 Font de Gaume (Dordogne) 25, 41, 57, 167, 168, 170, 176,211
Cougnac (Lot), 69, 89, 99, 100, 117, 124, 61, 97, 118, 122, 125, 145,151, 176 Ignatiev (USSR), 35
128,130,131, 145, 152, 168, 177, 198 Fontanet (Ariege), 13, 14, 16, 39, 40, 56, India, 27, 146
Couraud, Claude, 100, 105, 209 57,93,94, 108,122,146,209 Infra-red, 47, 48, 50, 101, 104, 106,
Courbet (Tarn-et-Garonne), 140 Footprints, 13, 14, 15 107' 196' 197
Covalanas (Santander) , 101, 102, F ornols-Haut (Pyrenees-Orientales), Insects, 78, 133
148, 157' 168, 176, 180 113,114,113,114, 149 Isard, 48, 82, 93, 179-80
Cro-Magnon, burial of (Dordogne) , 16, Fourneau du Diable, Le (Dordogne), 96 Israel, 33, 204, 207
73 Frobenius, L., 204 see also Hayonim, Mount Carmel,
Croze a Gontran, La (Dordogne), 91 Frolov, Boris., 182 Berekhat Ram

- v j a...V t t.,
 INDEX 239

lsturitz (Pyrenees Atlantiques), 39, 56, Levi-Strauss, Claude, 173, 185 Mortillet, Gabriel de, 2I, 23, 149, 203,
61, 62,66-8, 73, 81-3, 93, 128, 130, Lighting, see Lamps 212
137, 152, 154, 163, 179,208 Limeuil (Dordogne), 34, 125, 128 156, Mount Carmel (Israel), 204
Italy, 35, 39, 78, 86, 184 169, 178, 181,214 Mount Gambier (S. Australia), 30, 32
Ivory, 18, 74, 79, 81,84-7, 136, 137, Lithophones, 69, 177 Mousterian, 62, 71, 75, 80,89-90
141,164,182,207,208 Llonin (Asturias), 93 Moustier, Le (Dordogne), I6, 70,204
Lorblanchet, Michel, 105, 197,206 Mauthe, La (Dordogne), 24, 25, 28, 4I,
Lortet (Hautes-Pyrenees), 80, 120, 56,97, 109,128, I66
Jacob's Cave (Missouri), 26 124, 190, 210, 211 Murat, abri (Lot), 64
Japan, 28, 78, 83 Loup, Grotte du (Correze), 75 Murcielagos, cueva de los (Asturias),
Joly Leterme, 17 Lourdes (Hautes-Pyrenees), 53, 84, II7 39, 111
Jorda Cerda, Francisco, 58, 62, 65, 206 Music, 68, 69
Jousse, Marcel, 195, 198, 200 McBurney, Charles, II Muzzolini, Alfred, 204
Judd's Cavern (Tasmania), 29 Madeleine, La (Dordogne), I8, 34, 48,
73, 79, 123, 135, I69 Nal6n valley (Asturias), Ill
Magdalenian, 14, I6, I7, I9, 23, 33, 40, Neanderthalers, 16, 69-7I, 80,89
Kamikuroiwa (Japan), 28, 28, 78
53-8,60-2,64,66-9,72,73,75, 77, Nerja (Malaga), 69, 132, I77
Kapova/Kapovaya (USSR), 35 , 97
80-2,85,90, 96, 97 , 105, II9-2I, I32, Neschers (Puy-de-Dome), I7
Karake (S. Australia), 32
133, I40 , 145, I46, I48, 158, I72, I84, Niaux (Ariege), 11, 14, I9-20 , 39-42,48,
Karlie-ngoinpool (S. Australia), 31, 32 ,
I89-90, I97 54,57, 89, 90, 92,93,97, 105,116-18,
184
Magdeleine,La(Tarn), I65, I70 121, 122, I28, I46, 148, I52 , ISS,
Kehoe, Tom, 211
Mal'ta (Siberia), 86, I82 179, I81,194, 200,201 , 209
Kesslerloch (Switzerland), 119, 125
Maltravieso (Caceres), 104, 105
Kleine Scheuer (Germany), 133
Mammoth, 18, 26, 27, 35, 48, 55, 69, 71, Ochre,24,29,48,57,58, 70, 78 , 79,84,
Koonalda cave (S. Australia), 28 , 30, 32
78, 80, 8I, 84, 86,108 , 110,115, II6, 96, 97, 100, 101, I07 , 140,204,
Koongine cave (S . Australia), 91
I28, I29, 131, 133, 135, 140, 144, 156, see also Pigments
Korea, 28
167, 170, I72, 179, 181,211 Olary (S . Australia) , 30
Kostenki (USSR), 72 , 86, 128, 137, 140,
Mammouth, Grotte du (Dordogne), 131 Oilier de Marichard, Jules, 20
163
Marche, La (Vienne), 33, 45, 50, 53, 58 , Orchestra Shell cave (W. Australia), 30,
Kiihn, Herbert, 203
77, 78, II6, I21 -3, 128-30, 135-7 , 136, 32
I44, 161, 172, I79 , 207 Ostrich eggshell, 27
Labastide (Hautes-Pyrenees), 58, 75, 77 , Marshack, Alexander, 45, 47, 48, 54, 75,
82, 108, 110, 114, 117, 121, 177 , 179, 107, 149, 182-4, I95, I98, 200-2,207 Pair-non-Pair (Gironde), 24, 42, 44, 56,
205,206,208 Marsoulas (Haute Garonne), 22,24, 25, 9I' 96, Il6, 156, I66
Labattut, abri (Dordogne) , 61, 135 39-4I, 58, 97, 10I, 107, 116, 117, I32, Pales, Leon, 50, 116, 122, 129 I44, 179,
Labouiche (Ariege), 130 I40, ISS l9I,210
Ladders, see Scaffolding Martin, Henri, 23 Paroong (S . Australia) , 30, 32
Lalinde, La Roche (Dordogne), 69, 70, Mas d'Azil, Le (Ariege), 11, 23, 39, 54, Parpall6 (Valencia), 33, 66, 67, 77, 93,
141 71 , 77-9, 8I-4, 82, 83, 88, 108, 118, 118
Laming-Emperaire, Annette, 62, 65 , 120, 121,122, 128, 132, 146,155, I77, Pasiega, La (Santander), 57, I04, 145,
165-9,172, 173,176,210 179-80, I90,210,2I4 I48 , IS7, 176, I80
Lamps, 23, 24, 42, 60, 77, 93, 109, 110, Massat (Ariege), 18, I9, 74, 93, 108, Pataud, abri (Dordogne), 73, 97
155,203,204 129, 169 Patne (India), 27
Lartet, Edouard, 17, 18, 26, 53, 146, Matupi cave (Zaire), 27, 204 Pavlov (Czechoslovakia), 83, I29
149, 181 Mayor, Dr Franc;:ois, 17 Pech de l'Aze (Dordogne), 69, 71, 80,
Lascaux (Dordogne), 11, 15, 16, 39-40, Mazouco (Portugal) , 39, Ill, Ill 208
47,48,57,59-61,63,64,92,97, 100, Mexico, 26, 8I Pech Merle (Lot), 48 , 52, 64, 69, 90, 101,
101,105, 106, 108, 109, 114, 116, Mezhirich (USSR), 79, 181,181, 2I4 104, 105,106, 107, 116, 117,118, 120,
118-21, 123, 124,125, 137, 145 , 146, Mezin (USSR), 69, 73, 78, 84 129, 145, 148, 152,154, I69, 170, 171,
153, 155-7' 161' 167' 168, 173, 177' Michaut, Lucien, 42 I72, 177, 178,20I,209
180, 181,183, 184,185-7, 190, 192 , Mickey Mouse, 60 Pechialet (Dordogne), 120, I29
196-8,200-1,209,211 Miller, Henry, 188 Pedra Furada (Brazil), 27
Lascaux II (Dordogne), 11, 48, 100, 216 Minateda (Albacete), 60, Pendo , El (Santander), I20
Lastic-Saint Jal, Vicomte de, 18 Moissan, H., 97 Pergouset (Lot), 116
Lauget.ie-Bass~ordogne), 34, 48, 61, Molodova (USSR), 80 Perigord, 33-5, 39, 96, 108, I45, 148,
73, 117' 120, 123, 132, 137' 138, 161, Monedas, Las (Santander), 54,55, 57, lSI, 163, 172, 196
180, 210, 214 97' I24, 176, 209 Peru, 26, 204
Laussel (Dordogne), 41, 56, 61, 96, 99, Monpazier (Dordogne), 163 Petersfels (Germany), 73
104,137,144,163,163 Montespan (Haute Garonne), 51, 60, 91, Peyrony, Denis, 25
Leguay, L. , 79 93 -5,98, 108, 117, lSI, I52, 176, I77, Picasso, Pablo, 65
Leroi-Gourhan, Andre, 7, 9, 62,64-7 , 205,209 Piette, Edouard, 20,23-5, 4I, 53, 54, 61,
117,120, 132, 134, 144, 146, 148, 153 , Montgaudier (Charente), 80, 132, I83 78, 8I, 85, 120, I24, 138 , 146, 149,
163,165-70, 172, 173, 176, 179, 181 , Mori, Fabrizio, 204 l50,206,208,2I2
184, 189,192, 196-8,200,201,213 Morin, abri (Dordogne), 64, 79, 81, II8, Pigeonnier, Grotte du, see Domme
Lespugue (Haute Garonne), 85, 132, 125, 156, 210 Pigments , 27, 42, 54, 58-60, 69-70, 78,
137' 138, 139, 140 Morin, cueva (Santander), 80 96, 97, 99, 100, lOI, 104-6, 178,209
 240 IMAGES OF THE ICE AGE

Pilbara (Australia), 30 Saint Andre d'Allas, Tibiran (Hautes-Pyrenees), 104, 108

Pileta, La (Malaga), 100, 132, 155, 184, see Roc , Grone du Tito Bustillo (Asturias), 13 , 57, 59, 59,
209 Saint Cesaire (Charente Maritime), 71 66 , 77, 78 , 97 , 101 , 108 , 124, 156,
Pincevent (Seine-et-Marne), 121 Saint Cirq (Dordogne), 132, 133 176 , 177
Pindal(Asturias) , 57, 116, 128,146 Saint-Front, Grone de (Dordogne), Toquepala (Peru), 204
Placard, Grone du (Charente), 48 see Mammouth Torre (Guipuzcoa), 80, 81, 180
Plants, 16, 17,133, 183 Saint Germain-la-Riviere (Gironde), Travers de Janoye (Tarn) , 177
Plaquettes, 16, 33, 49, 54, 58, 59, 62, 64, 72,72 Trois Freres, Grone des (Ariege), 8,
75-8 , 87, 117, 118, 121 , 128, 140, 151 , Saint Marcel (lndre), 69 39-40, 42 , 43, 44 , 46, 47, 58, 66, 69 ,
160, 161, 182, 208 Sainte Eulalie (Lot), 184 90, 91 , 91-3 , 107, 108, 116, 122-4,123,
Playbay, 165 San rock-art, 158 132 , 134, 140, 142 , 143, 144, 146, lSI ,
Poisson , abri du (Dordogne), 39, 57, 96 Santian (Santander), 104, 145 158 , 158, 168, 191 , 195 , 196,201 ,
Polesini (Italy), 184 Santimamiiie (Vizcaya), 108, 166 202 , 205
Pomarel, 25 Saut-du-Perron (Loire), 48 , 76 Trou de Chaleux (Belgium), 76, 123
Pont d'Ambon (Dordogne), 64 Sautuola, Marcelino Sanz de, 20-3, 25 , Tsonskala (USSR), 75
Porte!, Le (Ariege), 87, 89, 108, 11 6, 179 Tuc d'Audoubert, Le (Ariege), 13, 14,
146, 164, 183, 186 Sautuola, Maria de, 20, 203 40, 47, 48 , 56, 58, 90, 94, 95, 95, 116,
Portugal, 35, 39, 111, 140 Scaffolding, 15 , 108, 114 133 , 146, 160, 165, 168, 176, 177, 184,
Pradieres (Ariege), 39 Schulerloch (Bavaria), 35 195-8,199,201,202
Predmost (Czechoslovakia), 83 Seals, 73, 80, 81, 116, 132 Tursac (Dordogne), 84, 85, 124, 140,
Puente Viesgo (Santander), 177 Segries (Alpes de Haute Provence), 205 163
Puy de Lacan, Le (Correze), 75 Shamanism, 144, 157, 158 , 190, 193 Twisted perspective , 60-2 ,64,67, 118
Pyrenees , 7, II , 13, 18 , 19, 33, 34, 39, Sherborne (England), 204 Tylor, Edward B., 150
73,81-3,86, 93 , 94 , 96 , 104, 108, 113, Shiyu (China), 28
120, 121 , 128, 132, 133 , 140,21 1 Shulgan-Tash , see Kapova Ultraviolet , 47,47, 48 , 107, 196, 197
Siberia, 26, 33, 35 , 83, 86, 110, 115 , 128 , USSR, 35 , 68 , 182 , 184, 206
Qafze (Israel), 70 137, 140, 144, 157, 182
Quercy, 66 , 133 Sicily, 35 , 39
Quina, La (Charente), 71 ,80 Signs, 39, 115, 140, 144-8,147, 150, 151, Vache, La (Ariege), 40, 54, 81, 120 , 129,
153, 155 , 163, 167, 169-70, 172 , 173, 156
Raphael, Max, 166, 167
181 , 183,183, 186, 189-90, 192 'Venus' figurines, 53, 54, 62 , 83-6, 104,
Rascafto (Santander), 93
Sireuil (Dordogne), 140 135-8,139, 140,144, 163 , 164, 189
Raymonden (Dordogne), 132, 133
Snowy River cave (Australia), 28 Vertut, Jean, 6, 7-9,47, 48 , 91 , 107 , 194,
Red deer, 124, 125 , 133 , 148, 153, 156,
Solutrean, 16, 55 -8, 61 , 62 , 64, 66, 68 , 195-202
157, 167, 170, 172 , 176, 180, 185 , 206
119-20, 140, 148 Veyrier (Haute Savoie), 17, 17, 133
Regnault , Felix, 24, 25 ,97
Sotarriza, La (Santander), 118 Vialle, Ruben de Ia , 19
Reinach , Salomon, ISO, 173
Souci , abri du (Dordogne), 214 Vialou, Denis, 197, 205
Reindeer, 17, 18, 28, 48, 55, 56, 66 ,91,
Souquette, La (Dordogne), 73 Vibraye , Marquis de, 53, 138
119, 121 , 124, 125 , 127, 133, 135 , 152 ,
Sous-Grand-Lac (Dordogne), 137 , 152 Vicdessos valley (Ariege), 57
156, 167, 176, 182, 184
Spear-throwers, 34, 82, 83, 86, 87, 121, Vilanova y Piera, Juan , 21-3
Rhinoceros, 26 , 27 , 35 , 48, 76, 115,132,
128,179 Villars (Dordogne), 101, 123 156, 201
152, 167, 168, 172, 178 ,207
Spencer, Baldwin, 150 Vogelherd (Germany), 65, 66, 84, 85 ,
Rhone valley, 57, 66 , 133 , 178
Statuettes, 82-6, 136-8, 140, 147 , 164, 128, 184
Rio Pinturas (Argentina), 204
165, 184, 192,208 Volp caves, 8, 11, 40 , 45, 48 , 195
Riviere , Emile, 24, 25, 97, 109,204
Stoliar, A.D., 60 see also Enlene , Trois Freres ,
Roc, Grone du (Dordogne), 39
Sungir (USSR), 72 , 74, 84, 86 Tuc d'Audoubert
Roc-de-Marcamps , Le (Gironde), 69
Switzerland, 17 , 78 , 82 Vulvas,62 ,88, 94, 118 , 136, 137, 146,
Roc de Sers (Charente), 61 , 96 , 11 9, 123 ,
155, 163-5 , 169-70, 186, 202
123, 140
Roc-la-Tour I (Ardennes), 77, 78 Tagliente, Riparo (Italy) , 80
Roche-Lalinde , La (Dordogne), Tanzania , 204 Walbiri , The , 145
see Lalinde Tasmania, 28, 29, 32 Wendt , Eric , 27
Rochereil (Dordogne), 16, 34 Tata (Hungary), 71, 184 Willendorf (Austria), 84, 137 , 138
Romania , 35, 39 Tectiforms, 48 ,50, 145 , 148, 151 , 151, Wye valley (Britain), 205
Rottliinder, Rolf, 209 213
R~adour (Lot), 69 , 108 , 209 Tequixquiaq (Mexico), 26
You Yuzhu, 28
Rouffignac (Dordogne), 203 Terra Amata (Alpes-Maritimes), 69
Yugoslavia , 35, 39, 68
Rousseau, Michel, 210 Terracotta , 82, 83, 86, 93, 129
Tete-du-Lion (Ardeche), 55, 57,109, 168
Sahara, 204 Teyjat (Dordogne), 56, 123 , 124, 126, Zhoukoudian (China) 16, 28, 74
Sahly, Ali , 105 127, 160 Zimbabwe, 27 , 70, 204

Lh.JU htOUE 0
,9 ov 19