s t o n e a g e i n s t i t u t e
p u b l i c a t i o n s e r i e s
Series Editors Kathy Schick and Nicholas Toth
Number 1.
THE OLDOWAN: Case Studies into the Earliest Stone Age
Nicholas Toth and Kathy Schick, editors
Number 2.
BREATHING LIFE INTO FOSSILS:
Taphonomic Studies in Honor of C.K. (Bob) Brain
Travis Rayne Pickering, Kathy Schick, and Nicholas Toth, editors
Number 3.
THE CUTTING EDGE:
New Approaches to the Archaeology of Human Origins
Kathy Schick, and Nicholas Toth, editors
Number 4.
THE HUMAN BRAIN EVOLVING:
Paleoneurological Studies in Honor of Ralph L. Holloway
Douglas Broadfield, Michael Yuan, Kathy Schick and Nicholas Toth, editors
Stone Age Institute
Gosport, Indiana
and
Indiana University,
Bloomington, Indiana
the cutting
edge:
S T O N E A G E I N S T I T U T E P U B L I C AT I O N S E R I E S
N U M B E R 3
Kathy Schick
Stone Age Institute & Indiana University
Nicholas Toth
Stone Age Institute & Indiana University
New Approaches to the
Archaeology of Human Origins
Stone Age Institute Press · www.stoneageinstitute.org
1392 W. Dittemore Road · Gosport, IN 47433
Series Editors Kathy Schick and Nicholas Toth
Editors
COVER CAPTIONS AND CREDITS
Top: Homo habilis Utilizing Stone Tools. Painting by artist-naturalist Jay H. Matternes. Copyright 1995, Jay H.
Matternes. Inspired by a prehistoric scenario by K. Schick and N. Toth in Making Silent Stones Speak: Human Origins
and the Dawn of Technology (1993), Simon and Schuster, New York. Pp.147-149.
Lower right: Whole ﬂake of trachyte lava from the 2.6 million-year-old site of Gona EG-10, Ethiopia. Reported by S.
Semaw (2006), “The Oldest Stone Artifacts from Gona (2.6-2.5 Ma), Afar, Ethiopia: Implications for Understanding the
Earliest Stages of Knapping” in The Oldowan: Case Studies into the Earliest Stone Age, eds. N. Toth and K. Schick.
Stone Age Institute Press, Gosport, Indiana. Pp. 43-75. Photo courtesy of Tim White.
Lower left: Prehistoric cut-marks from a stone tool on Sterkfontein hominin partial cranium StW 53. Reported by T.
Pickering, T. White, and N. Toth (2000) in “Cutmarks on a Plio-Pleistocene hominid from Sterkfontein, South Africa”.
American Journal of Physical Anthropology 111, 579-584. Scanning electron micrograph by N. Toth.
Published by the Stone Age Institute.
ISBN-10: 0-9792-2762-3
ISBN-13: 978-0-9792-2762-2
Copyright © 2009, Stone Age Institute Press.
All right reserved under International and Pan-American Copyright Conventions. No part of this
book may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopying, without permission in writing from the publisher.
CHAPTER 4
ARCHAIC STONE INDUSTRIES FROM
EAST AFRICA AND SOUTHERN EUROPE
PRE-OLDOWAN AND OLDOWAN
HENRY DE LUMLEY, DEBORAH BARSKY AND DOMINIQUE CAUCHE
Over recent years, many discoveries have renewed
our knowledge about the oldest stone industries and also
about the behaviour and lifestyle of the hominids that
made them, not only in East Africa, but also in the Near
East, in Trans Caucasia and in southern Europe.
If the ﬁrst tool-making hominids appear in East Africa
as early as 2.55 million years ago, they are present
in the Levant a little over 2 million years ago, as early as
1.81 million years ago at the gates of Europe in TransCaucasia,
and a little over 1.4 million years ago on the
Mediterranean coasts of Europe (Figures 19 and 21).
PRE-OLDOWAN OR ARCHAIC OLDOWAN
The Gona and Ounda Gona sites in Ethiopia, dated
to a little over 2.5 million years old, have yielded lithics
characterized largely by knapping products (Figure 1):
ﬂakes and angular fragments (de Lumley, 2006; Semaw,
2000 and 2005; Semaw et al., 1997 and 2003). Manufactured
pebbles are rare. Bipolar knapping on an anvil was
often practiced while hand-held knapping methods employed
were sometimes unifacial, bifacial or even multifacial.
Core exploitation was relatively intense, with
numerous, generally small sized ﬂakes removed from
each piece. The assemblage does not include standardized
small retouched tools shaped by secondary retouch
modifying the edge of a ﬂake or angular fragment.
The Lokalalei 1 and 2C sites in Kenya, in the Nachukui
Formation dated to 2.340 million years old, discovered
and studied by Hélène Roche and Pierre-Jean
Texier, offer an industry similar to that from the Gona
sites (Roche et al., 1999; Delagnes et al., 2005). The assemblages
comprise mainly non-retouched ﬂakes, angular
fragments and cores (Figure 2).
Reﬁts of ﬂakes onto cores allowed deﬁne a systematic
reduction sequence resulting in the production of
numerous, generally small-sized ﬂakes. For some cores
as many as ﬁfty pieces were reﬁtted. As is the case for
the Gona region sites, the Lokalalei assemblages do not
comprise any standardized, small retouched tools on
ﬂakes or debris, with the possible exception of a few
pieces.
The archaic stone tools from the Omo Valley, such
as those from Omo 71 and Omo 84, dated to around 2,4
million years old, as well as those from Omo 57 and
Omo 123, dated to around 2.34 million years old, discovered
and studied by Jean Chevaillon, yielded very
numerous small quartz ﬂakes, with an average size of
2 to 3 cm, most often showing reduced or null striking
platforms and ﬂat inner surfaces, bearing witness to the
frequent use of bipolar knapping on an anvil. However,
pebble tools and percussion instruments are rare or absent.
There are no small retouched tools. Yet ﬂake edges
often show irregular use retouch and chips attesting to
their use.
Locally available raw materials, mainly quartz and
some basalt, were used in the fabrication of the stone tool
assemblage from the Fejej FJ-1 site in the southern Omo
area of Ethiopia, around ﬁve kilometres from the Kenyan
border (Barsky et al., 2006; de Lumley et al., 2004a
and 2004b; Chapon, 2007). Pebbles used were collected
from the environment immediately surrounding the site;
a few hundred meters at the most. Both bipolar knapping
on an anvil and hand held techniques were used during
knapping reduction sequences which were mainly unifacial,
sometimes bifacial and rarely multifacial.
The assemblage is mainly composed of small-sized
ﬂakes and angular fragments (Figure 3). Pebble tools are
56  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Figure 1. Gona EG 10, Hadar, Ethiopia. 2.55 million years. Pre-Oldowan or Archaic Oldowan.
1 to 3: ﬂakes; 4: core; 5: chopper
de Lumley et al.  57
Figure 2. Lokalalei 2C, West Turkana, Kenya. 2.2 million years. Pre-Oldowan or Archaic Oldowan.
1 to 4: ﬂakes; 5 and 6: cores; 7: percussion instrument. (from A. Delagnes et H. Roche, 2005).
58  The Cutting Edge: New Approaches to the Archaeology of Human Origins
rare: choppers (pebbles showing a bevelled cutting edge
shaped by one or several removals on a single face), a
few core-scrapers (rostro-carénés; massive choppers
with abrupt, narrow and convex cutting edges) and rare
chopping tools (pebbles showing a bevelled cutting edge
shaped by one or several removals on both faces). At this
site, once again, there are no intentionally shaped small
retouched tools on ﬂakes or angular fragments.
In East Africa, presently known lithic assemblages
dating to before around 1.9 million years ago show common
characteristics (de Lumley, 2006; de Lumley, 2004a
and 2004b). Such industries, dated to between 2.55 and
1.9 million years ago, have also been discovered in Ethiopia,
in the Hadar region (AL666) and in the Omo Valley
(Ftji1, 2 and 5, Omo 71, 84, 57 and 123) and at Kanjera
South in Kenya (Plummer et al., 1999). The tools were
essentially made from quartz or volcanic rock types collected
from the area immediately surrounding the site.
Knapping was generally executed using the hand held
technique most often with unifacial and unidirectional
removals, sometimes with multipolar or centripetal removals.
Bifacial knapping was rarely practiced. Multipolar
orthogonal knapping is also present. Bipolar knapping
on an anvil was quite often used. It is important to
underline that this type of percussion; bipolar simultaneously
in two opposite directions, does not necessarily
signify a poorly mastered knapping technique, but rather
a very efﬁcient way to extract ﬂakes or sharp fragments,
from small crystalline pebbles. Thus, knapping methods
and techniques for ﬂake production were chosen in acFigure
3. Fejej FJ-1, South Omo, Ethiopia. 1.96 million years. Pre-Oldowan or Archaic Oldowan.
1 to 3: ﬂakes; 4: core; 5: chopper.
de Lumley et al.  59
cordance with the quality of raw materials and/or initial
bloc or pebble morphology. It is thus that humans
chose essentially cubic blocks offering natural striking
platforms.
Generally cores were abandoned after the extraction
of only a few ﬂakes, especially when raw materials were
of poor quality. However, when rocks were of ﬁner quality,
for example ﬁne grained basalt, reduction sequences
were systematically more intensive, as Fejej FJ-1, and
even up to ﬁfty ﬂakes for a core at the Lokalalei 2C site
(Roche et al., 1999).
Relative abundance of raw materials near each site,
their nature, the morphology and size of available pebbles
or blocs all had a determinant inﬂuence on the technological
characteristics of each stone series.
These archaic lithic assemblages are essentially
composed of non-retouched ﬂakes, some cores or pebble
tools, notably choppers and a few chopping tools.
Flakes seem to have been the main objective sought out
by the knappers. Their non-modiﬁed edges sometimes
show chips attesting to their use. Flakes were often broken
during their production probably because of violent
blows with a hard hammer instrument or during bipolar
on an anvil knapping technique. Flakes obtained, generally
small sized, show numerous knapping accidents:
longitudinal fractures along or parallel to the knapping
axis, and fractures transversal or oblique to the knapping
axis.
In all of these assemblages, even the oldest, as at
Gona, hominids had acquired quite elaborate knapping
strategies attesting to their technical prowess, and their
relatively developed level of comprehension of raw
material bloc geometry and of the physical parameters
essential to knap hard rocks, as well as anticipatory
faculties. They had acquired and mastered knapping processes
according to certain constant technological principles.
The invention of the tool, 2.55 million years ago,
is a major cultural leap for humanity, with which begins
the real history of Mankind.
We have proposed to group together these archaic
industries under the name “Pre-Oldowan” or “Archaic
Oldowan” in order to underline the absence or extreme
scarcity of small retouched tools which characterizes
them (Barsky et al., 2006; de Lumley, 2006 and 2007;
de Lumley et al., 2004a and 2004b). In the case of retouched
tools, an edge is modiﬁed by tiny, continuous,
adjacent or overlapping removals, to obtain a predetermined
shape. The term “Pre-Oldowan” is not meant
to designate a single, homogenous and well deﬁned
cultural entity, but rather a cultural evolutionary stage
within which there are not yet stereotyped and standardized
small tools made by retouch that modiﬁed the initial
form of ﬂake, fragment or core clasts. It is a practical
denomination for designating a cultural milieu wherein
the technological (chaine opératoires) and typological
(resulting tool types) characteristics are more archaic
than in the Oldowan stricto sensu or Classical Oldowan.
It connotes a behavioural parallelism amongst diverse
human groups present in East Africa between 2.55 and
1.9 million years ago.
A drier milieu and the consecutive extension of a
savannah type environment constrained some hominids
to consume larger quantities of grains, thus favouring the
development of large teeth (Paranthropus), while others
were oriented towards ever increasing meat consumption.
They were scavengers who, with the help of their
primitive tools, collected meat left on large Herbivore
carcasses abandoned by large Carnivores (DominguezRodrigo
et al., 2005; Semaw et al., 2003). They also
gathered not only vegetables, but also small animals
such as rodents, reptiles and amphibians.
CLASSICAL OLDOWAN
Although a decisive step for human evolution, these
archaic tools were rudimentary. Growing sophistication
in tool making was soon to accompany the progressive
evolution of the human brain.
Around 1.9 million years ago in East Africa, at Olduvai
in Tanzania, but also in Kenya and Ethiopia, more
evolved stone industries appear. They too are characterized
by a predominance of ﬂakes, unifacial, bifacial or
multifacial cores and some pebble tools (choppers, chopping-tools)
(Figures 4 to 7). Small tools also make their
appearance, with intentionally retouched edges modifying
the initial form of the ﬂake or fragment support.
These are end scrapers, scrapers and denticulates, that is
to say, ﬂakes or fragments upon which adjacent notches
were intentionally made in order to form a saw-like edge
(Figures 4 to 7). Other types of tools also appear, such
as polyhedrons (globular shaped tools made by small
removals forming facets) (Figure 7, n° 11) and spheroids
(perfectly spherical stones intentionally made but
whose use remains unknown). It is this stone industry,
found at Olduvai Gorge at several sites dated to between
1.9 and 1.7 million years old that Louis, and then Mary
Leakey named “Oldowan” (Leakey, 1971; de la Torre et
al., 2005).
Oldowan culture was to spread throughout the entire
African continent: its traces are found not only in Olduvai’s
Bed I between 1.9 and 1.6 million years ago, but
also in Ethiopia, at the Gombore I site in Melka Kunturé,
at sites dated to 1.7 million years ago, in Algeria at Ain
Hanech, dated to around 1.8 million years old, and even
in south Africa, at Sterkfontein 5 and at Swartkrans.
“Classical Oldowan”, dated to between 1.9 and 1.6
million years old, is characterized by the predominance of
non-modiﬁed ﬂakes, the presence of unifacial and sometimes
bifacial or multifacial cores, pebble tools which
are present in low but signiﬁcant proportions polyhedrons
and spheroids and especially small retouched tools
(end scrapers, scrapers, denticulates) (de la Torre et al.,
2005). These elements denote type-specialization which
is progressively to become more and more standardized.
The Classical Oldowan marks the acquisition of a
new stage in human cognition, within which speciﬁc
60  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Figure 4. Olduvai, Tanzania. Bed 1, Site FLK North, Sandy Conglomerate. 1.75 million years. Oldowan sensu stricto or
Classical Oldowan. Small retouched tools on chert ﬂakes.
1 to 11: scrapers; 12: scraper + notch; 13 to 16: becs made from two continuous retouched notches.
de Lumley et al.  61
Figure 5. Olduvai, Tanzania. Bed I, site FLK NN 1. 1.75 million years. Oldowan sensu stricto or Classical Oldowan.
Small retouched tools on quartz ﬂakes.
1 and 7: end scrapers; 2 to 5 and 9 to 11: side scrapers; 6: notch; 8: denticulate scraper.
62  The Cutting Edge: New Approaches to the Archaeology of Human Origins
small-sized tools were fashioned using voluntary retouch,
and a certain standardization of manufactured
tools is achieved.
These early artisans lived at lake edges or alongside
small water sources or marshes, convenient for water
supply and scavenging. Though they continued to scavenge,
the desire for meatier alimentation led hominids
to practice their ﬁrst hunting activities. The growing diversity
of their stone tool-kit denotes a specialization in
their activities. Scrapers and end scrapers were probably
used to work skins, thus suggesting that they began to
dress themselves, denticulates and other tools with wide
deep notches may have been used for wood working. But
what were polydhedrons and spheroids used for? They
may have served as primitive bolas for capturing large
herbivores by immobilizing their legs.
THE PRE-OLDOWAN OUTSIDE OF AFRICA
According to our knowledge to date, australopithecines
appear to have been conﬁned toAfrica. However, it
was very early after their emergence that the ﬁrst human
tool making artisans left the African cradle of humanity
(Figures 19 and 21).
Figure 6. Olduvai, Tanzania. Bed I, site FLK North, levels 1 to 3. 1.75 million years. Oldowan sensu stricto or Classical
Oldowan. Small retouched tools on quartz ﬂakes (n°s 1, 2 and 4 to 6) and volcanic rock (n° 5).
1, 2 and 5: scrapers; 3, 4 and 6: denticulates.
de Lumley et al.  63
Figure 7. Olduvai, Tanzania. Bed I, site DK. 1.8 million years. Oldowan sensu stricto or Classical Oldowan. Small
retouched tools on ﬂakes, quartz (n°s 8 to 10) and volcanic rock (n°s 1 and 4 to 7). 1: scraper adjacent to an
end scraper; 2 to 9: scrapers; 10: end scraper; 11: polyhendron; 12: discoidal core; 13: chopping tool.
64  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Yiron, Israël
The Yiron site, located north-west of Haifa in Israel,
symbolizes this large dispersion. Reported by Avraham
Ronen as early as 1980, this site indicates to paleontologists
that stone tool manufacturers lived in the Levant
more than 2 million years ago (Ronen, 1991 and 2006;
Ronen et al., 1980).
The plateau is composed of a thick layer of basalt
ﬂow covering gravels trapped in red clays. The basalt
ﬂow has been dated by potassium-argon to 2.450 million
years old. Stone tools, a few ﬂint pieces, were discovered
in the gravel just beneath the basalt (Figure 8). Up
to date, too few objects have been found to allow for a
precise diagnostic of this industry. It is characterized by
the predominance of non-retouched ﬂakes, bifacial cores
and the absence of standardized small retouched tools
on ﬂakes. It appears therefore very similar to the PreOldowan
industries from East Africa.
Dmanissi
At the crossroads between Africa, Asia and Europe,
Georgia is situated between the Black and Caspian seas,
in the prolongation of the Rift Valley and the Palestinian
corridor, beyond the Zagros and Anatolian mountain
ranges and the Small Caucasus. It was the launching
point from which early humans migrated towards Europe
and Asia.
At the Dmanissi site, 85 kilometers south of Tbilissi
at an altitude of 100 meters, on the southern slopes of
the Small Caucasus, a Lower Pleistocene fauna, a PreOldowan
stone industry and numerous hominid remains
were discovered in a well dated stratigraphical context
(Lordkipanidzé et al., 2007; Celiberti et al., 2004;
Gabounia et al., 2000 and 2002; de Lumley, 2006; de
Lumley et al., 2002 and 2005). This site attests to Man’s
arrival at the gates of Europe around 1.810 million years
ago, thanks to the geochronological 40Ar/39Ar dating
method.
Today in ruins, the medieval city of Dmanissi, near
the village Patara Dmanissi, was built on a basaltic spur
at the conﬂuence of two rivers; the Mashavera, which
circles it to the north-west, and the Pinezaouri, to the
south-east. Constructed during the IXth and Xth centuries,
the city prospered between the XIIth and XIVth
centuries, before its destruction in the second half of
the XIVth century by Tamerlan, a Turkish Muslim conqueror,
at the head of the Golden Horde.
The prehistoric site was discovered beneath the medieval
city. Excavations brought to light trash pits in the
foundation basements. In 1991, a hominid mandible was
found in one of these, in very ancient deposits within
which it had been buried. Following this discovery,
other human remains were found at Dmanissi. Excavations
under the direction of David Lordkipanidzé have
revealed ﬁve skulls, four mandibles and around twenty
postcranial remains. They are the oldest known human
remains yet to have been found in Eurasia.
The Dmanissi spur is covered by a thick volcanic
lava ﬂow composed of basalt, covering Pliocene ﬂuvial
deposits. This lava, coming from the Dzavacheti
of Empeliki Mountains west of the site, ﬂowed through
the Mashavera paleo-valley. Heading towards the northeast,
it constituted an obstacle against which lakes were
formed. An age of 1.850 million years was obtained
from the lava ﬂow using the potassium-argon method.
A layer of volcanic ash (bed VI) covers the basalt
ﬂow. The hominid remains were found in this level, and
it has been dated using the 40
Ar/39
Ar method to 1.810
million years old. Subsequently, river and ﬂood silty
sands covered the layer of ash. The paleomagnetic limit
Olduvai-Middle Matuyama, around 1.770 million years,
was identiﬁed in the deposits, 80 centimetres above the
layer of ash. The archeological deposits at Dmanissi
which yielded the human remains, dated to between
1.810 and 1.750 million years, thus accumulated over a
period of less than 60,000 years (de Lumley et al., 2002).
Paleo-vegetation in the area may be reconstituted
through the study of the rare pollens, phytoliths and carporests.
It evokes a mosaic type landscape, with large,
mainly grassy open spaces and dispersed forest areas,
a gallery type forest along riversides and denser forest
covering the nearby mountains (Messager, 2006). It is
similar to that familiar to early African hominids.
The large mammalian fauna includes canids (Cansis
etruscus, Vulpes alopecoides), bears (Ursus etruscus,
Ursus sp.), hyenas (Pachycrocuta perrieri, Pachycrocuta
sp.), felines (Lynx issiodorensis, Panthera gombaszoegensis)
and notably sabre tooth tigres (Meganteron
meganteron, Homotherium crenatidens). Among
the herbivores present at the site: elephants (Mammuthus
meridionalis), horses (Equus stenonis, Equus sp. aff.
altidens), rhinoceros (Dicerorhinus etruscus), cervids
(Cervus perrieri, Cervus sp., Eucladoceros aff. senezensis,
Cervus (Dama) cf. nestii major), small bovids
(Paleotragus sp., Dmanisibos georgicus, Galagoral meniginii
sicenbergii, Capra sp., Sorgelia sp., Ovibovini),
antilopes (Gazella sp., Antilopini) and even, among the
birds, ostriches (Strutho dmanisensis).
This faunal assemblage may be dated according
to the evolutionary stage of the different species to the
transition period between the Tertiary and Quaternary
periods (Plio-Pleistocene boundary) dated to around 1.8
million years ago.
It comprises African species such as giraffe (Paleotragus
sp.) and ostrich, Eurasian species such as some
horses, elephants rhinoceros and deer, as well as sabre
tooth tigers which were widespread during this period
in Africa as in Eurasia, clearly indicating that Georgia is
situated at the crossroads between two major geographical
ensembles (Africa and Eurasia), with the Levant hyphenating
these two large spaces.
Sedimentological, paleobotanical and paleontological
data allow to reconstitute the environment within
which these early humans lived at the gates of Europe: a
savannah landscape with some trees and sectors of forest
de Lumley et al.  65
Figure 8. Yiron, Haifa Region of Israël. More than 2 million years. Pre-Oldowan or Archaic Oldowan. Flint lithic industry.
1 to 3: ﬂakes; 4 and 5: cores.
66  The Cutting Edge: New Approaches to the Archaeology of Human Origins
cover, under a hot and humid subtropical climate. The
hominids who lived on the southern slopes of the Small
Caucasus 1.8 million years ago rediscovered their ancestral
ecological niche. Fauna there was abundant, while in
East Africa the climate grew progressively drier.
The hominid remains include, ﬁve skulls, four mandibles
and around twenty post-cranial remains, belonged
to at least ﬁve individuals of both sexes and of varying
ages: one adolescent 13 or 14 years old, a pre-adult, two
adults and one toothless elder. They were quite small,
around 1.50 meters tall, with a cerebral volume of 600 to
770 cm3
(de Lumley et al., 2006).
Their anatomical characteristics, most notably metrical,
indicate that the Dmanissi hominid remains may be
an intermediary form between, on the one hand, Homo
habilis-rudolfensis and, on the other hand, Homo ergaster
which appeared in East Africa as early as 1.8 million
years ago. They do however appear closer to the former
group, particularly to rudolfensis ER 1470 found in Kenya.
Several characteristics differentiate them: their skull
has a greater transverse diameter (euroneuryon), less
however than that of Homo ergaster, a petro-tympanic
disposition at the base of the skull in posterior rotation,
as well as very strong canine pillars.
The relative gracility of their face, narrowness of
their occipital portion and the architecture of the base
of their skull clearly distinguishes them from Homo
erectus, whose predecessors appear to have been Homo
ergaster.
All of the anatomic characteristics of the Dmanissi
hominids combined led Marie-Antoinette de Lumley to
attribute the Dmanissi hominids to a new species: Homo
georgicus (de Lumley et al., 2006).
An abundant stone industry (Figure 9) was also
found at Dmanissi, among the fauna and human remains.
It is very archaic (Celiberti et al., 2004; de Lumley,
2006; de Lumley et al., 2005). The raw material is essentially
composed of volcanic rock pebbles, with some
metamorphic rocks as well, all collected from the nearby
Mashavera and Pinezaouri alluvials. Raw material supplying
was strictly local–there are no rocks brought from
any distance away.
Rocks selected are of ﬁne and of varying quality,
and hominids chose pebbles with angular morphology,
presenting natural planes, most favourable for knapping
and shaping.
The lithic assemblage is characterized by a large
number of non-modiﬁed whole or broken pebbles, perhaps
manuports, that is to say, objects brought to the site
in their natural state. Many show angular morphology,
with triangular or quadrangular sections. Their relatively
modest dimensions (755535 mm for the whole pebbles),
less than those of the pebble tools (837345
mm), may indicate a choice of larger clasts for the latter.
The assemblage is also characterized by numerous
non-modiﬁed knapping and shaping products–ﬂakes and
fragments–a relative frequency of cores, alongside the
larger unifacial or bifacial and more or less elaborate
pebble tools. Percussion tools include pebbles showing
marks from knapping blows or bone fracture, as well
as pebbles with isolated convex negative scars typical
of accidental removals. Small ﬂakes less than length
20 mm long are very scarce and there are no retouched
ﬂakes. There are no standardized small tools shaped by
intentional retouch.
All stages of knapping and shaping operational
chains are represented at the site, from whole pebbles
up to ﬁnished products (non-modiﬁed ﬂakes or pebble
tools) including knapping residues (cores and fragments)
and even percussion instruments. The ﬂakes themselves
carry traces of the different knapping stages; from entirely
cortical dorsal surfaces (test ﬂakes corresponding
to ﬁrst removals detached from a pebble and thus conserving
a cortical surface), to non-cortical ﬂakes (ﬂakes
detached within the volume of the blocks). Their presence,
as well as that of numerous cores, suggests that
knapping took place on the site, in situ. A reﬁt of a ﬂake
onto a chopper, both from the same layer, also shows that
shaping took place on-site. The edges of broken pebbles,
pebble tools, cores, fragments and ﬂakes often show irregular
use retouch.
The relatively low proportion of non-cortical ﬂakes
(less than a third) in association with numerous partially
exploited cores with only a few removal scars, underlines
non-intensive support exploitation. Most of the
non-cortical ﬂakes coming from within the volume of the
pebbles are in ﬁne grained tuff, a more exhaustive use of
this ﬁner quality rock type. Furthermore, a large number
of ﬂakes with cortical striking platforms also conserve,
at least partially, the cortical surface of the pebble from
which it was knapped on their outer side. This may be
explained by the summarily knapped clasts and pebble
tools most of which show a single removal negative
(primary choppers) or sometimes choppers with a few
removals. The orientation of removal scars conserved on
the ﬂakes’outer sides conﬁrms the frequency of unipolar
knapping. Flakes resulting from more complex knapping
strategies are rare. Likewise, ﬂakes show relatively few
removal scars (average of 2.1) also underlining minimal
exploitation of knapped or shaped clasts.
Knapping was done with a hard percussion instrument
using mainly hand held or occasionally bipolar
on anvil techniques. Cores are characterized by a
dominance of unifacial knapping, few removals and the
frequent use of unprepared cortical striking platforms.
However, presence of bifacial, multifacial or prismatic
cores indicates that the Dmanissi knappers were able to
apply more complex knapping strategies and thus possessed
well mastered knapping techniques. More numerous
removals are visible on some cores, notably those in
ﬁner quality raw materials.
Pebble tools and percussion instruments largely
dominate the assemblage. Many whole or broken pebbles
show traces of their use as percussion tools, either
for breaking bones or for knapping or shaping other
stone artefacts. Such traces consist of chipped areas or
de Lumley et al.  67
cupula, or, more often, accidental single convex edged
removal scars. Percussion instruments are most often
thick pebbles with traces typically located on the lateral
edges. Isolated convex removal scars have nearly rightangled
striking angles (average of 86 degrees), in other
words, the knapping platform is sub parallel to the main
pebble axis.
Pebble tools found in the different layers of the Dmanissi
stratigraphical complex attest that the clasts were
selected according to their size; favouring large pebbles;
either thick and angular or ﬂatter and angular. Nearly two
thirds of the pebble tools show a single concave removal
negative. These are primary choppers whose cutting
edge was shaped by only one short removal, with a more
acute angle (average of 79.5 degrees) than observed on
pebbles with convex edged removal scars. Choppers,
usually non-converging, are more elaborate; shaped by
few removals (an average of only 3.3), delineating an
angled cutting edge (average of 79 degrees). The few
core-scrapers present in the assemblage were made on
larger clasts by more numerous removals (average=4.3).
Chopping-tools are rare, made from thick, large pebbles.
On the average they show only ﬁve removal scars and
the angle of their cutting edge is the most acute among
Figure 9. Dmanissi, Georgia. 1.81 million years. Pre-Oldowan or Archaic Oldowan. Lithic industry in diverse rocks.
1 to 3: ﬂakes; 4: core; 5: chopper.
68  The Cutting Edge: New Approaches to the Archaeology of Human Origins
the pebble tools (average of 69.2 degrees).
On the other hand, small standardized tools shaped
by intentional retouch on ﬂakes or fragments were not
found in the stone tool assemblage from Dmanissi. Their
cutting edges do however often show irregular, marginal
micro-retouch or isolated irregular retouch, such as
dense notches, which may be continuous or sometimes
even overlapping, localized on parts of the edges and attesting
to the intensive use of these pieces.
Among pieces showing irregular retouch, isolated
notches are most frequent (56.2 %). Only about a quarter
of the pieces affected show continuous irregular retouch
located on an edge and evoking a scraper or a sort of
end scraper. Some ﬂakes (seven pieces) show a single
removal on one or the other of its faces.
In short, the Dmanissi stone assemblage (Figure 9)
is characterized by the following traits :
• a strong proportion of whole or broken pebbles
(manuports)
• numerous non-modiﬁed knapping products (ﬂakes
and fragments), most of which conserve a cortical
surface
• cores with a limited number of scars, made from selected
angular clasts (pebbles with natural planes)
• primary choppers (pebbles with single concave
ﬂake scars) dominate among the macro-tools and
there are some more elaborate choppers (notably
non-pointed), whereas percussion instruments are
most frequent
• many pieces, especially ﬂakes and fragments, with
traces of intense use wear such as notches, or dense,
continuous or overlapping irregular retouch, attesting
to their intensive use
• absence of standardized small tools shaped by intentional
retouch
The stone industry from Dmanissi has numerous
points in common with the oldest stone industries known
in Africa such as those from Kada Gona (EG 10 and EG
12, 2.550 million years old); Lokalalei 2C (2.340 million
years old) and Fejej FJ-1 (1.9 million years old):
• local raw material selection
• better quality raw materials were more exhaustively
knapped
• selectivity in choice of pebble morphology
• abundance of non-modiﬁed knapping products
(ﬂakes, debris, cores)
• frequent production of sharp-edged objects (essentially
ﬂakes, but also pebble tools) for cutting meat
and disarticulating animal carcasses
• mastery of rock fracturing methods
• use of hand held and bipolar knapping on an anvil
techniques
• parallel use of several knapping strategies
• predominance use of unidirectional, unifacial knapping
strategies
• presence of pebble tools, mainly with unifacial removal
scars (choppers) with little morphological
standardization
• absence or extreme rarity of small tools shaped by
intentional retouch
• high proportion of ﬂakes, fragments and broken
pebbles with tiny retouch or irregular retouch, notably
dense notches, resulting from intensive use
Pirro Nord, Foggia Province,
Southern Italy
The Pirro Nord site, also known as the Cava
dell’Erba, is located in southern Italy’s Foggia Province,
on the Apricena Commune, northeast of the Gargano
massif. The area has been known for a long time as a very
rich paleontological site; its karstic cavities were brought
to light during quarrying and have yielded remains of
mammals characteristic of the Lower Pleistocene.
In September 2006, at the occasion of the XVth
International
Congress of Prehistoric and Protohistoric
Sciences, Marta Arzarello and her collaborators reported
the discovery, among the faunal remains, of a few stone
pieces (Figure 10), attesting to the arrival of prehistoric
humans on the Mediterranean coast of southern Europe
as early as 1.4 million years ago.
The site is a vast karstic network of galleries and ﬁssures,
largely produced by erosion. The fauna discovered
in these galleries is very rich: more than one hundred
species of vertebrates have been found. Carnivores, notably
felines, are dominant (Homotherium crenatidens,
Megantereon whitei, Acinonix pardinensis, Lycaon lycaoanoides),
bear (Ursus etruscus) and giant hyena
(Pachycrocuta brevirostris). Among the herbivores there
are bison (Bison deguilii), horse (Equus altidens, Equus
cf. Equus stenonis), rhinoceros (Stephanorhinus hundsheimensis)
and cervids (Praemegaceros obscurus, Axis
sp.). A specimen of African monkey has also been found
(Theropithecus sp.).
This faunal assemblage is characteristic of the
Lower Pleistocene and dates the site to around 1.4 million
years old. It is later than the beginning of the Lower
Pleistocene, as indicated by the presence of Stephanorhinus
hundsheimensis, and earlier than the Vallonnet
site, which has been dated to 1.070 to 1 million years
old and where Ursus deningeri and Bison schoetensacki
have been found. Among the bird remains discovered,
the coexistence of such species as the great bustard (Otis
tarda), the bustard (Tetrax tetrax) and the sand grouse
(Pterocles orientalis) suggest an open environment, arid,
but with seasonally humid areas. This landscape corresponds
to that favoured by large herbivores.
A few pieces knapped from ﬂint were found in
three ﬁssures: three cores and some ﬂakes. They were
de Lumley et al.  69
knapped by violent percussion using the hand held technique.
No small ﬂakes with intentional retouch were
found. However, the edges of some of the ﬂakes show
micro retouch attesting to their use. The industry is like
other “Pre-Oldowan” industries older than Dmanissi,
Yiron and some East African assemblages (Fejej, Lokalalei
and Gona). The Pirro Nord assemblage is characterized
by the production of non-modiﬁed ﬂakes. Raw
material is local testifying to a limited area of transport.
This recent discovery of a prehistoric site 1.4 million
years old from southern Italy is a milestone for the
arrival of hominids in southern Europe, on the Mediterranean
shore of southern Europe.
The great Guadix-Baza paleo-lake,
Orce, Andalucia
Barranco León and Fuente Nueva 3
In 1976 the team from the Miguel-Crusafont Paleontological
Institute in Sabadell, near Barcelona, discovered
a rich stone industry among the large mammal fossils
in the Guadix-Baza Basin in Andalusia, northeast of
Granada. The Barranco León and Fuente Nueva 3 sites,
just 4 kilometres apart, are located to the east of the town
of Orce, around 115 kilometres to the north east of the
Provincial capital and 80 kilometres from the seashore
(de Lumley, 2006; Martinez-Navarro et al., 2003; Toro
et al., 2002, 2003a and 2003b). These two sites are of
exceptional interest as examples of typical Pre-Oldowan
butchery sites demonstrating, once again, that early humans
made tools because they had become meat-eaters.
The Barranco León site was ﬁrst made known by
Jordi Agusti and excavated in 1994 by Josep Gibert.
Excavations were later taken up by Isidro Toro Moyano.
Fuente Nueva 3 was discovered in 1991 by Alain
Bocquet and excavated from 1995 by Josep Gibert and
Alain Turq. Excavations were later undertaken by Isidro
Toro Moyano.
Two deeply embanked valleys cut into the GuadixBaza
basin Plio-Pleistocene formations that are over 100
meters thick. Important natural stratigraphical sections
have been left by erosion. They show the evolution of
an ancient lake, the Baza paleo-lake, which covered the
basin from the end of the Miocene up the Middle Pleistocene,
between 7 million years and 300,000 years ago,
before drying up, when waters feeding this endoreic basin
were captured by the Guadalquivir.
This basin, extending over around 3,000 km2
, was
formed during the Miocene, about 10 million years ago
(Middle Tortonian) following tectonic phenomena. Once
linked to the Mediterranean Sea and the Atlantic Ocean,
it was ﬁrst inundated by the sea which, at the end of the
Figure 10. Pirro Nord. Foggia Province, Italy. 1.4 million years. Pre-Oldowan or Archaic Oldowan. Flint core.
(from M. Arzarello et al., 2006).
70  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Miocene, leaving progressively thinner clayey deposits.
By the Messinian, around 6 million years ago, during
the Miocene-Pliocene transition, connections between
the Mediterranean and the Atlantic were permanently
closed and the basin became endoreic, that is to say, fed
only by continental waters, and so became a lake. It was
then progressively ﬁlled in by continental deposits during
the Pliocene, and throughout the Lower and Middle
Pleistocene.
During the middle of the Middle Pleistocene the
Guadalquivir captured drainage waters and the basin was
opened up once again, it became exoreic, as this Andalusian
river emptied into the Atlantic Ocean to the west of
Gibraltar. Its Plio-Pleistocene formations were then truncated
by a glacis, and then deeply cut into by erosion, the
base of the level then being lowered to 500 meters, after
which the Baza paleo-lake dried up permanently.
Numerous paleontological sites have been discovered
in the upper layers of this formation on the border
of the paleo-lake, most notably Venta Micena, dated to
1.5 million years ago, as well as the Barranco León and
Fuente Nueva 3 archeological sites, both dated to around
1.2 million years old. These last two sites have also revealed
stone industries and numerous bone remains.
These two Andalusian sites are very rich, not only for
information they furnish about fauna, but also for their
contribution to what we know about the early inhabitants
of Europe on the Iberian Peninsula.
The fauna is very abundant and these sites have
been dated according to the biochronology of large
mammals and microvertebrates. The sites have also been
dated by magnetostratigraphy. The sedimentary deposits
give a negative paleomagnetic reading which, taking
into account the fauna’s evolutionary stage, corresponds
to the Middle Matuyama inversion (1.780 to 1.070 million
years), preceding the direct Jaramillo episode (1.070
million years to 984,000 years). In fact, the large and
micro mammals are at the same evolutionary stage and
they are very close in age. They are older than those from
the Vallonnet Cave site, dated to between 1.070 million
years and 980,000 years, with the appearance of the deer
(Megaceroides cf. verticornis) and the archaic bear (Ursus
deningeri), species absent from both of these Spanish
sites, which may therefore be dated to around 1.2
million years old.
Fauna from Barranco León (Martinez-navarro et al.,
2003) includes saber-toothed tigers (Homotherium sp.),
giant hyenas (Pachycrocuta brevirostris), bears (Ursus
sp.), an archaic wolf (Canis mosbachensis), foxes
(Vulpes sp., Alopex cf. praeglacialis), badgers (Meles
sp.), archaic elephants (Mammuthus meridionalis), hippopotamus
(Hippopotamus antiquus), deer (Megaceroides
aff. obscurus, Pseudodama sp.), bison (Bison sp.),
small bovids (Hemitragus cf. albus), horses (Equus altidens,
Equus cf. bressanus), rhinoceros (Stephanorhinus
hundsheimensis), porcupines (Hystrix sp.), rabbits
(Oryctolagus cf. lacosti), and many rodents (Allophaiomys
aff. lavocati, Allophaiomys sp., Castillomys crustafonti
sp., Apodemus aff. mystacinus).
The fauna from Fuente Nueva 3 (Martinez-Navarro
et al., 2003) includes saber-toothed tigers (Machairodontinae
indet.), giant hyenas (Pachycrocuta brevirostris),
bears (Ursus sp.), an archaic wolf (Canis mosbachensis),
foxes (Vulpes sp., Alopex cf. praeglacialis), badgers (Meles
sp.), archaic elephants (Mammuthus meridionalis),
hippopotamus (Hippopotamus antiquus), deer (Megaceroides
aff. obscurus, Pseudodama sp.), bison (Bison
sp.), small bovids (Ammotragus europaeus, Hemitragus
cf. albus), horses (Equus altidens), rhinoceros (Stephanorhinus
hundsheimensis), porcupines (Hystrix sp.), and
rodents (Allophaiomys aff. lavocati, Allophaiomys sp.,
Mymomys savani).
The stone industries from both sites (Figure 11) are
very similar, concerning the raw materials used, and the
types of pieces realized and the knapping techniques
used (Toro et al., 2003a and 2003b; de Lumley, 2007).
They attest to a human presence in southern Europe as
early as the middle of the Lower Pleistocene.
The rocks used as clasts for these tools are mainly
ﬂint and limestone, collected in the environment immediately
surrounding the sites where they are abundantly
available as blocks, pebbles, nodules or plates. Different
types of ﬂint of varying quality, as well as marly limestone
or limestone siliciﬁed to different degrees have
been identiﬁed.
Rock exploitation apparently began at the source
where the materials were collected by some initiating
blows. An under-representation of ﬂakes from the early
stages of the operative schemas is seen: entirely cortical
or with few removal scars on their dorsal face, suggests
that this activity took place away from the sites. The
scarcity of possible reﬁts of considerable size conﬁrms
that most of the knapping was not performed in the same
area where the meat was consumed.
Different types of more or less massive percussion
tools–whole or broken pebbles, stones and pebbles with
isolated convex removal scars–are numerous. Traces of
their use are conserved on their surfaces as crush-marks,
cupules, or isolated removal scars. The surfaces of the
limestones are chemically altered, often impeding observation
of any such traces. These implements apparently
served early humans frequenting the swampy area on the
edge of the Baza paleo-lake for breaking large herbivore
bones and as percussion instruments for shaping and
knapping stones.
Large tools on pebbles or stones are very rare in both
sites among numerous knapped ﬂakes. They are mainly
pebbles with single concave removal scars (“primary
choppers”), while in both sites there are few choppers
with edges shaped using a series of continuous removals.
Other types of choppers, irregular and of poor quality are
also rare.
Flake knapping is thus the main characteristic of
the Barranco León and Fuente Nueva 3 assemblages.
It took place using hand held or bipolar on anvil techniques.
Small series of unidirectional removal scars are
de Lumley et al.  71
often observed on the cores. Core scars are sometimes
also centripetal or crossed, on one or two sides of a
clast, or multifacial in orthogonal series, leading to the
production of cores with globular or polyhedron morphology.
Some of the ﬂakes obtained were themselves
subsequently knapped. The latter, as well as overall core
reduction strategy, reveals a desire to optimize raw material
exploitation, especially for ﬁner quality rocks such
as some of the ﬂint.
Angular fragments are also numerous, resulting
from the percussion of relatively poor quality rocks, with
mineral inclusions or lithoclases transformed into planes
during knapping.
Most of the ﬂakes conserve little of the original cortical
surfaces from pebbles or nodules while others were
knapped from the interior of the clast and conserve no
cortical surface at all. The latter were obtained from the
central mass of the raw material and are often small with
a low extension (length) index. The model sought after
was a small, more or less square ﬂake, on the average
2 to 3 centimetres long. The frequency of small ﬂakes
may be explained by frequent use of the bipolar on an
anvil technique which produces chipping on the transversal
ﬂake edges, posed directly onto the anvil, and the
detaching of small fragments.
Cores are remarkably rare compared to ﬂakes. At
Barranco León, a total of 252 ﬂakes larger than 2 cm
were counted in comparison to only 18 cores. Such a
ratio (14 ﬂakes for one core) seems impossible given the
poor quality of some of the rocks used. At Fuente Nueva
3, a total of 300 ﬂakes larger than 2 cm in comparison to
21 cores giving a ratio of around 16 ﬂakes for each core,
even less probable for poor quality rocks. It therefore appears
likely that a large proportion of the non-modiﬁed
Figure 11. Barranco León and Fuente Nueva 3, Orce, Guadix-Baza basin, Andalucia, Spain. 1.2 million years.
Pre-Oldowan or Archaic Oldowan.
1 to 4: ﬂakes; 5: core; 6: pebble with an isolated concave removal negative (primary chopper).
72  The Cutting Edge: New Approaches to the Archaeology of Human Origins
ﬂakes, especially those larger than 2 cm, and also the
cores, were transported to the site at least partially ﬂaked.
However, numerous exhausted polyhedron, globular
or cubic cores from which no further ﬂakes could have
been produced, knapped using either hand held or bipolar
on an anvil techniques, as well as ﬂake-cores from
which small ﬂakes were produced, seem to suggest that
much of the knapping did take place on-site, according
to needs, to carry out certain activities. There are mainly
very small ﬂakes knapped on-site from small cores (very
reduced) or from other ﬂakes. In fact, frequent irregular
retouch on these small ﬂakes show that their edges were
rapidly used and they probably had to be replaced often.
However, numerous cores showing intense exploitation
and from which no further ﬂakes may be obtained
are seen; polyhedrons, globular or cubic cores knapped
using either hand held or bipolar ﬂaking on an anvil,
seem to indicate that a large number of pieces were obtained
on the site, according to needs, for some activity.
Many pieces, be they broken or shaped pebbles,
angular stones, cores, fragments or (especially) ﬂakes,
show traces on their more-or-less sharp cutting edges of
irregular use retouch and micro retouch. However, intentional
retouch for the elaboration of standardized small
tools was not practiced by Orce’s early humans.
The quasi-exclusive use at Barranco León and
Fuente Nueva 3 of local raw materials can be seen; all
coming from sources accessible within a range of 5 kilometres,
suggest that these two sites were not habitation
sites. They seem to have been areas for a speciﬁc activity
for which hominids used nearby rocks.
This system of exploitation of rock resources is
particularly interesting. The behaviour of these early humans
is, in this respect, very opportunistic.
Quaternary conglomerates near the sites, accessible
500 meters to 5 kilometres away, furnished nodules of
limestone and ﬂint. Collection of these materials could
have taken place during daily foraging in search of food.
Although this kind of exploitation of stone resources
may appear simplistic or even instinctive, selective behaviour
did preside in the choice of raw materials. Pebbles
with summarily rolled surfaces and, consequently
few ﬁssures, seem to have been selected. Among the latter,
limestone was carefully selected for the production
of large tools while nodules of silicious rocks (ﬂint and
radiolarite) were reserved for ﬂake production. This selectivity
shows that hominids practiced a reasoned use
of raw materials, taking into account their physical properties,
in spite of their simple acquisition approach. It
proves that these early humans were perfectly adapted
to their environment and that they had already adopted
a behavior that was efﬁcient in its industrial enterprises.
The presence of stone artefacts among the large
mammal bones, sometimes preserved in anatomical position,
such as the hippopotamus at Barranco León or
the elephant at Fuente Nueva 3, as well as apparently
human-caused breakage and some traces of butchery on
bones, leads to believe that the stone tools were destined
to be used for processing the carcasses of these animals.
The relative scarcity of pebble tools and their small
size precludes the idea of a site reserved for disarticulating
and cutting up of large mammal carcasses killed by
humans or which were bogged down in the swamps.
The predominance of non-modiﬁed ﬂakes, most often
small in size, on the average 10 to 25 mm long at
Barranco León and 20 to 40 mm long at Fuente Nueva 3,
seems to suggest that they were used for cutting up and
scraping meat remaining on animal carcasses abandoned
by large carnivores in the swampy areas around the Baza
paleo-lake (Figure 12). Humans competed with hyenas
for the carcasses. They practiced secondary scavenging
: they followed satiated carnivores and obtained the
pieces of meat still remaining on the bones using their
small ﬂakes.
The technological (operative schemas) and typological
(resulting tool types) characteristics of the stone
industry, as well as the behaviour of the early humans
from Barranco León and Fuente Nueva 3, ﬁt these sites
within the Pre-Oldowan cultural horizon.
La Sima del Elefante,
Sierra d’Atapuerca, Spain
Around 500 kilometres to the northwest of the
Guadix-baza basin, the Atapuerca Sierra, near Burgos,
in the Castille-León Province, is an imposing karstic
massif with numerous cavities which sheltered remarkable
prehistoric sites covering all periods of human
history in Europe (Carbonell I Roura et al., 1995 and
2001; Carbonell I Roura et al., 2008; Cuenca-Bescos
et al., 2004; Garcia et al., 2008; Huguet et al., 2007;
Pares et al., 2006; Rofes et al., 2006; Rosas et al., 2001
and 2006).
Amongst these sites, La Sima de l’Elefante is a
cavity opened up by the trench of an old railway track
(la Trinchera del Ferrocaril). It is a vast cave, rich in
large mammal faunal remains–such as elephants–from
which comes the name of the site. It was in-ﬁlled by
sandy clays rich in stones over 15 meters thick. The
lower levels are very old; magneto-stratigraphical studies
have established that they were deposited prior to the
geomagnetic polarity change Matuyama-Brunhes, dated
to 770,000 years ago. Results from the cosmogenic nuclide
method gave a date of 1.13 ± 0.16 and 1.22 ± 0.14
million years (Carbonell et al., 2008).
Fauna from these levels includes archaic cervids,
primitive bison, hippopotamus, rhinoceros, macaques,
turtles, beavers and other rodents and insectivores for
which numerous remains have been found. The evolutionary
stage was determined by paleontological studies,
particularly by the rodents, including an archaic form
of Iberomys huescarensis, Allophaiomys lavocati and
Catillomys rivas, and the insectivores, especially Beremendia
ﬁssidens and Asoriculus gibberodon, all suggest
a date of Lower Pleistocene, earlier than 1 million
years old.
Presence of hippopotamus, beavers and turtles sug-
de Lumley et al.  73
Figure 12. The Pre-Oldowan site Barranco León and Fuente Nueva 3, Orce, Guadix-Baza basin, Andalucia, Spain,
were located in a swampy area between the southern edge of the large Baza paleo-lake and a Jurassic
limestone massive, rich in ﬂint.
In the swampy area, on the Baza lake-edge, prehistoric humans indulged in scavenging around 1.2 million
years ago.
74  The Cutting Edge: New Approaches to the Archaeology of Human Origins
gest a humid landscape with rivers and swamps. During
excavations in 2000, Eudald Carbonell’s team brought
to light the ﬁrst ﬂake knapped from ﬂint: it is proof that
hominids stayed at the Atapuerca Sierra more than a million
years ago.
Since then, other non-modiﬁed knapped ﬂint ﬂakes
have been recovered. None of these ﬂakes had been
transformed by intentional, regular retouch. They are
non-modiﬁed ﬂakes. These tools appear to correspond
with those from other sites with Pre-Oldowan industries.
In July 2007 an isolated premolar and the lower portion
of a hominid mandible, belonging to the same individual,
was attributed to Homo antecessor (Carbonell et
al., 2008), were discovered.
Thus, a little over 1 million years ago in the GuadixBaza
basin at Barranco León and Fuente Nueva 3 as well
as at La Sima del Elefante in the Atapuerca Sierra, the
earliest inhabitants of Europe lived in a humid, wooded
environment, at a lake's edge in the Andalusian basin or
near the rivers and swamps of the Castilian Sierra.
Who were the artisans of these primitive stone tools?
Were they similar to Dmanissi’s Homo georgicus, at the
gates of Europe, or were they closer to the European
forms of Homo erectus, Homo heidelbergensis that we
ﬁnd later at the Ceprano site in Italy (Homo cepranensis)
or at Gran Dolina (Homo antecessor), another cave
site in the Atapuerca Sierra, around 880,000 years ago?
Ongoing excavations in the Guadix-Baza basin and the
Sima d’Elefante site may soon provide answers to this
question so fundamental to understanding the morphological
evolution of the earliest inhabitants of Europe.
The hominid remains found in June and July in the lower
levels of the Sima del Elefante yield the ﬁrst answers to
this question.
The Vallonnet Cave site, RoquebruneCap-Martin,
Alpes-Maritimes
The Vallonnet Cave at Roquebrune-Cap-Martin in
the Alpes-Maritimes is located on the western slopes of
Cap Martin, around 800 meters from the Mediterranean
seashore (de Lumley et al., 1988). It opens up onto the
right-hand side of a small ravine, the Vallonnet, which
descends towards Menton Bay. It is a small cavity carved
into a Jurassic dolomitic-limestone massif, pitched, enveloped
in a Miocene conglomerate made up of pebbles
and sandy concretions. The low and narrow porch joins
a 5 meter long corridor which opens onto a 4 meter wide
chamber. Stratigraphical and sedimentological studies of
the deposits have deﬁned ﬁve ensembles.
The base of stratigraphical ensemble I is made up of
a stalagmitic ﬂoor dated to between 1.4 and 1.370 million
years old. Pollen analysis in the deposits evokes a
forest landscape with Mediterranean essences dominated
by plane trees.
Above this level, stratigraphical ensemble II is made
up of marine sands rich in foraminiferous, marine mollusc
shells and ﬁsh bones amassed by a transgressing sea
which dismantled the existing continental in-ﬁll. This
marine beach is slightly older than 1.070 million years.
Among the ﬁsh, the presence of diodon (globe ﬁsh) and,
among the molluscs, the presence of species typical of
warm seas, indicate a tropical or subtropical sea. The latter
data, in association with that from the fossil pollen
analysis, suggests a relatively warm, dry climate with
mild winters.
With a thickness of 1.5 meters, stratigraphical ensemble
III is the most important of the series deposited
in the cave. It is composed of continental deposits made
up of clayey-silty sands rich in stones and pebbles from
the conglomerate overhanging the cave. This relatively
homogenous continental in-ﬁll has been sub-divided into
three main layers which were subsequently sub-divided
into several human or carnivore occupation levels. Many
bones brought into the cave by humans and/or by carnivores
have been found in these deposits. Pollen analysis
in this stratigraphical ensemble suggests an open landscape
made up of composites, mainly chicories, clusters
of trees, and a sort of scrubland passing gradually into
deciduous-leafed oak forests dominated by white oak.
The relatively dry climate observed at the beginning of
this ensemble becomes more humid later on. Magnetostratigraphical
studies show that this ensemble corresponds
with the Jaramillo geomagnetic polarity episode
within the larger reverse polarity period of Matuyama.
This episode, situated between the two reverse geomagnetic
polarity events that are the Middle and Upper Matuyama,
is dated to between 1.070 and 1 million years
ago.
A little over 1 million years ago, the entranceway
corridor’s existing inﬁll was emptied by erosion and
only the deposits located deep in the chamber of the cave
were preserved.
Stratigraphical ensemble IV is a thick stalagmite
ﬂoor that was formed and that sealed what remained of
the geomagnetically reverse deposits thus preserving
them from further erosion. It was formed later than 1
million years ago (Upper Matuyama) and has been dated
using electron spin resonance to between 900,000 and
890,000 years old.
Palynological analysis suggests a deciduous forest
comprising a variety of species suggesting a colder and
especially more humid climate than at present.
The deposits making up this upper ensemble V consist
of silty-sandy plastic clays, extracted from earlier
deposits and accumulated during the different humid
phases of the Quaternary period.
Nearly one hundred knapped stone tools were found
in the sandy-clayey-silty sediment of ensemble III (Figure
13) among Quaternary faunal remains. They are
dispersed over sixteen carnivore and human occupation
levels identiﬁed as different archeostratigraphical units
by Anna Echassoux (2004) on computerized vertical
projections allowing a view of the archeological material
on proﬁles 25 centimetres wide.
Six units were identiﬁed in the lowest layer C, ﬁve
in the middle layer B2 and ﬁve in the upper layer B1.
de Lumley et al.  75
These archeostratigraphical units were not only human
occupations. Research carried out has not revealed
any particular pattern in the site. The Vallonnet Cave
served as a den for large carnivores, mainly bear, and
felines such as panther and saber-toothed tigers; large
hyenas broke many of the bones (Essachoux, 2004).
Large carnivores brought numerous herbivore carcasses
into the cave: deer, bison, small bovid, rhinoceros, horse,
wild boar. Apart from the carnivores, humans could have
also frequented the cave, leaving their tools behind.
Large mammal fauna is very abundant and includes
more than twenty-ﬁve species. Carnivores account for a
third of the material which is characteristic of the end of
the Lower Pleistocene and which includes some archaic
species: macaque (Macaca sylvanus ﬂorentina), felines
like the Eurasian jaguar (Pantera gomaszoegensis),
cheetah (Acinonyx pardensis) or the giant hyena (Pachycrocuta
brevirostris), southern elephant (Mammuthus
meridionalis), rhinoceros (Stephanorhinus hundsheimensis,
horse (Equus stenonis), pig (Sus), deer (Cervus
nesti vallonnetensis), musk ox (Praeovibos).
Other, more evolved species, announce the Middle
Paleolithic: an archaic wolf (Canis mosbachensis), an
archaic fox (Alopex praeglacialis), the cave lynx (Lynx
spelaea), a bovid (Bison shoetensacki) and thar (Hemitragus
bonali).
The evolutionary stage of each species of this well
dated fauna (from 1.070 to 1 million years old), delimits
the Vallonnet stratigraphical horizon which has become
a referential milestone in Europe for biochronology.
Paleoecological afﬁnities evoke different landscapes:
the numerous deer remains suggest forest areas,
while bison and small bovids evoke more open spaces.
Stone tools (Figure 13) left by humans during their
short stays in the cave–a hundred odd pieces–were
knapped from pebbles collected in the Roquebrune
Miocene conglomerate (de Lumley et al., 1988). Three
pieces were knapped from translucent beige ﬂint pebbles
which come from 700 meters to the north of the cave at
Ciotti, near Menton. Limestone was most often used for
the manufacture of these tools, as well as a few sandstone
and some rare ﬁne quartzite and ﬂint pebbles. The
assemblage is mainly composed of pebbles used as percussion
instruments (showing isolated convex removals),
pebble tools, cores and non-modiﬁed ﬂakes.
Percussion tools are by far the most numerous. Pebbles
showing single concave removals (primary choppers)
are frequent, although are poorly manufactured.
Pebble tools showing multiple removals (choppers,
chopping-tools) are signiﬁcant but most of them are of
poor quality and without standardization. Many cortical
ﬂakes present in the collection were produced by accidental
breakage of percussion instruments with isolated
convex removals. Those produced from intentional production
of pebble tools are also well represented. Finally,
ﬂakes from cores are rare. A core showing multidirectional
orthogonally oriented removal scars and another
with bipolar unifacial scars have been unearthed. All
ﬂakes are non-modiﬁed; none were transformed by intentional
retouch.
The middle portion of a bison femur (Bison shoetensacki)
showing a series of removals and apparently
having served as a percussion instrument (Figure 14),
and around ten deciduous deer antlers were brought into
the cave and used as tools. While tooth marks present on
most of the bones attest to their transport and breakage
by carnivores, characteristic spiral-type breaks resulting
from intentional human percussion on fresh bone are visible
on others.
Some bones show parallel, ﬁne, short, obliquely
oriented striation marks from meat cutting with a stone
implement. The high proportion of percussion tools,
whole or broken pebbles and pebbles with convex edged
removals, gives this industry a particular character and
bears witness to the dominant human activity practiced in
the Vallonnet Cave. These tools may be associated with
numerous intentionally broken large herbivore bones
found in the cave, while a few knapped ﬂakes, especially
those in ﬂint, bear witness to meat cutting activity.
Around one million years ago humans occasionally
came to Vallonnet Cave to scavenge carcasses abandoned
there by large carnivores, breaking the bones and
consuming the marrow.
At sites where the presence of early humans is attested
to by archaic stone industries, large herbivore
bones are often associated with large feline remains such
as saber-toothed tigers (Homotherium and Megantereon
megantereon), or Eurasian jaguars (Panthera gombaszoegensis).
Did early humans who had become meat
eaters follow large carnivores who, once satiated, left
behind carcasses of large herbivores, just like the giant
hyenas did (Pachycrocuta brevirostris) ? There apparently
was real competition between hyenas and humans
for scavenging.
These humans, more scavengers than hunters, left
no traces of domestic arrangement in the Vallonnet Cave.
No evidence has yet been discovered to suggest that they
had domesticated ﬁre. With Pirro Nord, Barranco León
and Fuente Nueva 3 and the Sima del Elefante, the Vallonnet
Cave is among the oldest known evidence of a
human presence in Europe in a well-dated stratigraphical
context.
Ca’Belvedere di Monte Poggiolo,
Emilei-Romagne, Italy
In Emilia-Romagna, on the eastern slopes of the
Apennines, between Rimini and Bologna, numerous
Lower Paleolithic sites were found rich in non-modiﬁed
ﬂakes, cores and pebble tools: Ca’Belvedere di
Monte Poggiolo, Ca’Romanina and Rio Sanguinario.
Ca’Belvedere, discovered in 1984 and studied by Carlo
Peretto, is the principal site.
Stratigraphical studies of the site identify, at the
base, a thick layer of greyish-blue clayey sediment deposited
by a coastline sea and overlain by a level of pebbles.
Nonetheless, magnetostratigraphical studies have
76  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Figure 13. Vallonnet Cave, Roquebrune-Cap Martin, Alpes-maritimes, France. Between 1.7 and 1 million years.
Pre-Oldowan or Archaic Oldowan. Limestone industry.
1: Pebble with an isolated convex edged removal negative (percussion instrument); 2, 3 and 5: ﬂakes; 4:
core; 6: chopper.
de Lumley et al.  77
Figure 14. Vallonnet Cave, Roquebrune-Cap Martin, Alpes-maritimes, France. Between 1.7 and 1 million years.
Pre-Oldowan or Archaic Oldowan.
Bison schoetensacki diaphysis fragment that served as a percussion instrument, showing a series of
invasive removals.
78  The Cutting Edge: New Approaches to the Archaeology of Human Origins
revealed a reverse geomagnetic polarity allowing to date
these deposits to the Upper Matuyama epoch, between
one million and 780,000 years ago. Unfortunately, fauna
was not preserved: it is therefore difﬁcult to precisely
date the site.
Avery rich stone tool industry comprising thousands
of pieces (Figure 15) was discovered on this shoreline.
It is characterized by abundant ﬂake production, almost
all obtained from ﬂint pebbles. Cores are numerous,
showing use of mainly unifacial, sometimes bifacial and
rarely multifacial knapping methods.
Large tools are represented by choppers, most of
which show single removal scars, and there are some
chopping-tools. Contrary to what one generally observes
in archaic Pre-Oldowan industries, ﬂakes from this site
sometimes show retouch, at the limit of regular retouch,
dominated by notched tools, bringing to mind non-standardized
small tools which may have served for cutting
meat.
Numerous ﬂakes have been reﬁtted onto one another
or onto the cores from which they were extracted, attesting
to complete on-site exploitation. These reﬁts help in
reconstituting knapping technology used by the artisans
at this site. There is a very simple technique using mostly
single removals; ﬂakes obtained may sometimes show
irregular use retouch. In addition, analysis of spatial distribution
of the ﬂakes shows that certain areas of the site
were preferred.
The Latium Rift
In central Italy, between Rome to the north and
Monte Cassino to the south, the Apennines to the east
and the Lepini mountains to the west, the Latium rift
was covered with large lakes during the Lower or Middle
Pleistocene. Ancient volcanoes were identiﬁed along
this rift by ashes deposited over time in the Quarternary
formations and that allow for radiometric dating. The
chemical composition of these ashes makes them characteristic
chronological milestones interstratiﬁed in the
Quaternary deposits.
Archaic stone tools were discovered at several locations
by Italian Paleontologists, notably Italo Biddittu,
Aldo Segre and Eugenia Segre-Naldini, at Colle Marino,
Arce, Fontana-Liri and Castro dei Volsci (Biddittu, 1971,
1972, 1983 and 1984; Biddittu et al., 1992; Cauche et
al., 2004). The levels comprising these stone tools are
sandwiched between deposits containing Lower Pleistocene
faunal remains below and more recent levels above
which seem to correspond to eruptive activity known in
the Latium. The oldest of these levels has been dated by
potassium-argon to 0.7 million years.
It is difﬁcult to date these sites. Because of the acidity
of the sediment, fauna has mostly disappeared. However,
at Colle Marino, a humerus fragment from a large
hyena, Pachycrocuta brevirostris, was found, as at Vallonnnet,
Barranco León and Fuente Nueva 3.
The prehistoric stone tools of the Latium are characterized
by frequent percussion instruments–pebbles or
blocks showing traces of percussion activity and pebble
tools–as well as numerous ﬂakes and debris. As at
Ca’Belvedere, some ﬂakes and debris show retouch that
may have been intentional, especially notches.
THE OLDOWAN IN EUROPE
In East Africa, especially at Olduvai, humans voluntarily
retouched ﬂakes and fragments to shape them into
small tools as early as 1.9 million years ago. The retouch
modiﬁed the natural edges into notches, denticulates,
scrapers and end scrapers. Along with polyhedrons and
spheroids, these small tools characterize the Oldowan
culture.
Fabrication of such small tools appears much later
in Europe (Figure 21), where they appear only around
900,000 years ago. Although present at Olduvai at
DK1 (Figure 7) and FLK NN1 (Figures 4 to 6) 1.9 to
1.8 million years ago, they have been identiﬁed at some
southern European sites, such as at Terrassa in Catalonia
around 0.9 million years ago, at Ceprano in Italy (Figure
16) and at Gran Dolina in Spain (Figure 17) only around
0.8 million years ago and again in Italy at La Pineta in
Isernia (Figure 18) only around 0.62 million years ago
(Figure 20).
The temporal discrepancy (Figure 21) between the
emergence of this large cultural horizon in Africa and
its appearance in Europe is considerable: one million
years. This temporal gap leads one to believe that the
spatial diffusion of these new acquisitions was extremely
slow. The passage between Pre-Oldowan and Oldowan
industries thus marks a new stage in the development of
human cognitive capacities and represents an important
cultural advancement.
Terrassa, Catalonia, Spain
The Terrassa site in Catalonia, particularly rich in
stone industry and in fauna, is located 30 kilometres to
the northeast of Barcelona at the edge of a small river,
the Val Paradis. It was unearthed in geomagnetically reverse
deposits corresponding to the Upper Matuyama.
The fauna is slightly more recent than that from Vallonnet
and dates the site to around 900,000 years old. It includes
a Eurasian jaguar (Panthera gombaszoegensis),
a gracile rhinoceros (Stephanorhinus hundsheimensis), a
stenonian horse, a suid close to the one from Vallonnet,
a cervid (Pseudodama nestii vallonnetensis) and antique
elephant (Elephas antiquus).
The stone tools include a few poorly worked pebbles,
numerous ﬂakes and a few intentionally retouched
small tools on ﬂakes or debris.
Ceprano, Latium, Italy
The Campo Grande di Ceprano site is located in
central Italy, in the Latium, around 100 kilometres southeast
of Rome (Ascenzi et al., 1996 and 1997; Mallegni
et al., 2003). The site is of particular interest because a
human skull was discovered there in 1994 by Italo Bid-
de Lumley et al.  79
Figure 15. Ca’Belvedere di Monte Poggiolo, Emilie-Romagne, Italy. Around 1 million years. Pre-Oldowan or Archaic
Oldowan. Flint industry.
1 and 2: ﬂakes; 3: core; 4: chopper; 5 to 8: chopping tools.
80  The Cutting Edge: New Approaches to the Archaeology of Human Origins
dittu. The skullcap was found isolated in clay ﬂows during
roadwork, and is the oldest human skull presently
known from Mediterranean Europe.
The skullcap was discovered during roadwork, isolated
in clay ﬂows. These clayey deposits cover ﬂuvial
sand formations, rich in fresh water molluscs. They show
reverse geomagnetic polarity and appear to be slightly
earlier than the Matuyama-Brunhes dated to 780,000
years ago.
The massive, low skullcap bears relief marks above
the orbits and temporal crests. The latter end in a torus
showing a parietal postero-inferior angle similar to that
observed on most Homo erectus. Its morphology is very
different from Homo georgicus who lived 1.8 million
years ago. In spite of a large variability, it appears close
to that of the African and Eurasian Homo erectus. It is
close to the European Anteneandertals (Homo heidelbergensis),
and thus has been attributed to a new species
(Homo cepranensis) (Mallegni et al., 2003).
One million years separate Homo georgicus from
Ceprano Man (Homo cepranensis). Is there an evolutionary
link between them or did a new population come
to Europe from Africa or Asia? The question remains to
be answered.
Archaic stone tools were found in association with
large mammal bones including elephant (Elephas antiquus),
rhinoceros (Stephanorhinus hundsheimensis),
hippopotamus (Hippopotamus sp.), cervis (Pseudodama
nestii vallonnensis).
The lithics (Figure 16) were knapped from a large
variety of raw materials–ﬂint, jasper, siliciﬁed breccia,
quartz, quartzite, limestone–collected as pebbles
or plates, and are characterized by a large number of
knapped ﬂakes, cores and pebble tools (Cauche et al.,
2004). Core reduction was intense with frequent directional
changes. Most ﬂakes show little or no cortical
residue.
Some ﬂakes or fragments were intentionally retouched
into scrapers, denticulates and side scrapers or
shaped into notches, becs and short thick points shaped
by adjacent notches. These small tools denote a diversity
of specialized activities corresponding perhaps to a
higher level of cognition for Oldowan humans.
Gran Dolina, TD 6, Atapuerca Siera, Spain
The Gran Dolina site is located in the Atapuerca Sierra,
Castille-León Province, near the Sima del Elefante.
This huge cavity, like that of the Sima del Elefante, was
opened up during railway construction work to build
the Trinchera del Ferrocaril (Carbonell et al., 1995 and
2001).
The lower levels (from the base to the top) TD 4, TD
5 and TD 6, immediately underlying the polarity change
limit Matuyama-Brunhes, are earlier than 780,000 years
old. Level TD 6, familiarly called the “Aurora Strata”
is rich in fauna including rhinoceros (Stephanorhinus
etruscus), elephants (Mammuthus sp.), horse (Equus
altidens), cervids (Cervus nestii vallonnetensis, Cervus
elaphus, Eucladoceros giulii), bison (Bison cf. voigtstedtensis)
and large carnivores such as saber-toothed tigers
(Homotherium latidens), Eurasian jaguar (Panthera
gomaszoegensis), lynx (Lynx sp.), as well as very numerous
rodents (Allophaiomys chalinei, Pliomys espiscopalis,
Mimomys savani). This faunal ensemble is characteristic
of the end of the Lower Pleistocene (800,000 years).
Nearly a hundred human remains were discovered
in this level, including a frontal, a maxillary, a mandible,
a parietal and numerous teeth and post-cranial elements:
ribs and vertebra. All of these bones come from six individuals;
two children , two adolescents and two young
adults.
Some traits allow to characterize them: they are
relatively slender, there is a depression on the maxillary,
their tooth relief is relatively complex with numerous
cusps and deep grooves, and the molars have multiple
roots. These characteristic traits led Jose Bermudez de
Castro, Juan LuisArsuage and Eudald Carbonell to name
them Homo antecessor (Carbonell et al., 1995).
These fragmentary human remains show striation
marks from meat removal and spiral-type fractures typical
of fresh bone breakage. It would appear that the site
was a lair for hominids who scavenged large herbivore
and human remains, more speciﬁcally young individuals.
The stone industry (Figure 17) discovered among
these human and faunal remains was knapped from diverse
rocks: limestone, ﬁne and coarse grained quartzite,
quartz, ﬂint and sandstone. They must have been collected
as pebbles or blocks.
Flakes dominate, indicating an important knapping
activity. Cores are also numerous, conﬁrming the latter
hypothesis. Knapping was almost exclusively done using
hand held percussion in several directions: centripetal,
orthogonal or simply unipolar. The few ﬂakes showing
intentional retouch (Figure 17, n°s
4 to 6) to make
scrapers and denticulates or to shape notches, link this
assemblage to the Oldowan.
La Pineta, Isernia, Italy
The La Pineta site, south of the town of Isernia in
the Molise Province of central Italy, is located 150 kilometres
to the northeast of Naples (Peretto, 1994). The
natural Isernia basin constitutes a vast depression once
occupied by a large lake at the beginning of the Quaternary
period. A thick layer of white lacustrine clays were
left by the lake and then covered by sand and gravel.
Travertine is interstratiﬁed at the top of this formation.
Several layers containing volcanic ash allow to precisely
determine the age of this site.
In the sand and gravel layers with travertine, two archeological
levels rich in fauna and stone industry were
uncovered over a large area by Carlo Peretto. Radiometric
dates using two argon isotopes (40
Ar /39
Ar) enable to
attribute an age to these two archeological levels of between
620,000 and 604,000 years. The clayey upper levels,
situated above the archeological layers, have been
dated to 504,000 and 474,000 years. The very abundant
de Lumley et al.  81
Figure 16. Campo Grande di Ceprano, Latium, Italy. Around 800,000 years. Oldowan sensu stricto or Classical
Oldowan. Industry in diverse rock types.
1: pebble with an isolated concave removal negative (primary chopper); 2 and 3: choppers; 4: chopping tool;
5 and 6: denticulated scrapers; 8: end scraper; 7, 9 and 12: cores.
82  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Figure 17. Atapuerca, Gran Dolina, level TD 6, Castille León, Spain. Around 800,000 years. Oldowan sensu stricto or
Classical Oldowan.
1 to 3: ﬂakes; 4 notched tool; 5: scraper; 6: denticulate; 7: core; 8: chopper.
de Lumley et al.  83
Figure 18. La Pineta, Isernia, Molise Province, Italy. 620,000 years. Flint industry.
1 and 6: denticulated scrapers; 2: double convex edged dejeté scraper; 3: ﬂake produced from another
ﬂake; 4: core; 5: scraper; 7: denticulate; 8: chopper.
84  The Cutting Edge: New Approaches to the Archaeology of Human Origins
fauna includes bison (Bison schoetensacki), rhinoceros
(Stephanorhinus hundsheimensis), elephants (Elephas
(Palaeoloxodon) namadicus), bear (Ursus deningeri),
hippopotamus (Hippopotamus cf. antiquus), wild boar
(Sus scrofa), small bovids (Hemitragus cf. bonali), cervids
(Megaloceroides solilhacus, Cervus elaphus acoronatus,
Dama dama cf. clactoniana, Capreolus sp.) and a
large feline, the lion (Panthera leo fossilis). This fauna is
characteristic of the morphological species evolution at
the beginning of the Middle Pleistocene.
Within this faunal assemblage, numerous cervid
bones and the presence of hippopotamus remains suggest
a humid forest landscape.
The stone tools from this site (Figure 18) are very
numerous, with several thousand pieces. The products
were essentially knapped from ﬂint and some limestone.
Two main operative schemas may be distinguished. For
the shaping of pebble tools, humans chose limestone, but
they preferred ﬂint for knapping ﬂakes. Pebble tools are
relatively rare but well made. Flakes are most numerous
in the assemblage. Most of the ﬂintknapping took
place using bipolar on an anvil technique but hand held
knapping was also practiced. Unidirectional percussion
predominates, without excluding the use of bipolar percussion.
Many ﬂakes were further reduced as cores.
This industry is characterized by intentionally retouched
small tools, mainly on ﬂint ﬂakes. The collection
includes mainly notched tools, scrapers and end
scrapers. This making of small tools on ﬂakes (Figure
18, n°s
1, 2, 5 and 7) by successive notches or continuous
raised retouch, that is to say, wide, concave and sub
vertical or more or less abrupt micro removals, is one of
the principal traits of the la Pineta assemblage. This type
of shaping is practically unknown in more archaic industries
(Pre-Oldowan) but present in Classical Oldowan.
Most of the bones, dispersed over a large surface
area, are broken and rarely found in anatomical connection.
They often show intentional fractures (on fresh
bone) for marrow extraction. Striation marks on the
bones indicate that meat was cut off of them. Some bison,
elephant and rhinoceros skulls were opened, suggesting
that the brains of these large herbivores were consumed.
It appears that this was a scavenging site where humans
were attracted by the carcasses of drowned animals. The
pebble tools may have served for butchery activities
such as disarticulating large herbivore carcasses, while
the ﬂakes and small tools may have served to remove
and cut up ﬂesh.
No particular spatial pattern has been detected at the
site. An accumulation of travertine blocks surrounded by
ﬂint ﬂakes may suggest a natural area where hominins
could have withdrawn to a dry place after having scavenged
in the surrounding muddy swamps.
CONCLUSIONS
In East Africa, stone industries earlier than 1.9 million
years present common characteristics. They are essentially
knapped from quartz or volcanic rocks, raw
materials collected close to the sites. Most of the knapping
was realized by hand held technique, using unifacial,
unipolar knapping, or sometimes multipolar or centripetal
removals. Bifacial knapping was rare. Multipolar
orthogonal knapping was also used. Bipolar knapping
on an anvil was quite often used (Omo 57, Omo 123,
FtJi1, FtJi2, FtJi5, Fejej FJ-1) (de Lumley et al., 2004;
de Lumley, 2006 and 2007). It is important to underline
that bipolar on an anvil knapping was not necessarily
a poorly mastered technique but rather the best way to
extract ﬂakes or fragments from small pebbles in crystalline
rocks. Methods and techniques of ﬂake production
are often tied to the quality of raw materials and
to initial block or pebble morphology. Generally, cores
were abandoned after the extraction of a few ﬂakes, especially
when raw materials were of poor quality. Nonetheless,
when rocks were of ﬁne quality, for example ﬁne
grained basalt, reduction sequences were longer, up to
ﬁfty ﬂakes for a core at Lokalalei 2C. The apparent technological
differences between these assemblages reﬂect,
for a large part, differences in available raw materials.
Relative availability of raw materials in the immediate
environment of each site, their nature, morphology and
the size of available pebbles or blocks, conditioned the
technological characteristics of each lithic series.
The lithic assemblages are mainly composed of nonmodiﬁed
ﬂakes, some cores and pebble tools, mainly
choppers and rare chopping-tools. For these sites ﬂake
production seems to have been the main objective of the
knappers. The non-modiﬁed ﬂake edges are sometimes
chipped, indicating their use. Flakes are often broken
as a result of the violent blows using a hard percussion
instrument, either by hand held or bipolar on anvil
techniques. Flakes obtained were small and often present
knapping accidents: longitudinal breaks along (Siret
type accidents) or parallel to the knapping axis, or breaks
that are transversal or oblique to that axis.
Industry from these sites whose age is between 2.55
to 1.90 million years old present the following common
characteristics :
• local rock selection
• choice of raw materials for which the best ones were
more exhaustively knapped
• selectivity in choosing pebble or block morphology
• presence of numerous percussion instruments
• well-mastered rock fracture
• use of both hand held and bipolar knapping on an
anvil techniques
• presence of pebble tools, generally unifacial (choppers)
and pebbles with single removals (primary
de Lumley et al.  85
choppers) probably used for disarticulating animal
carcasses
• parallel use of several well mastered knapping
strategies and, mainly unifacial knapping using
unidirectional removals, more rarely multipolar or
centripetal and sometimes bifacial or multifacial
orthogonal knapping. The methods used generally
follow a least effort strategy
• dominance of unifacial cores
• abundance of non-modiﬁed knapping products
(ﬂakes, fragments and cores)
• dominance of non-modiﬁed ﬂakes, generally small
sized and without pre-determined form, probably
used for cutting meat
• abundance of ﬂakes broken from knapping with violent
blows
• frequency of remnant cortex on ﬂakes whose striking
platforms are also most often cortical or non-
prepared
• absence or extreme rarity of small intentionally retouched
tools on ﬂakes or fragments
• a high proportion of irregular marginal micro-retouch
on ﬂake, debris, pebble tool and broken pebble
edges, especially dense notches.
We have proposed (de Lumley et al., 2004; de Lumley
et al. 2005; de Lumley, 2006 and 2007) to group
together these archaic industries under the name PreOldowan
(or Archaic Oldowan) to designate lithic assemblages
which are characterized by the absence or
extreme scarcity of small retouched tools. The term
Pre-Oldowan does not designate here a single homogenous,
autonomous and well deﬁned culture, but simply
a cultural evolutionary stage where standardized, stereotypical
small tools on ﬂakes and fragments with retouch
that modiﬁes the original form of their edges are not yet
made while such tools are present in the Oldowan levels
DK 1 (1.9 myrs.) and FLK NN (1.8 myrs.) at Olduvai,
where there are a relatively signiﬁcant amount of true
end scrapers, scrapers and denticulates made from quartz
(Figures 4 to 7), associated with polyhedrons and spheroids
(Leakey, 1971; de la Torre and Mora, 2005).
In a large sense it is possible to distinguish two
stages of Oldowan that are probably tied to the level of
cognition attained by early humans.
1. the Pre-Oldowan or Archaic Oldowan, like at Gona
EG 10 and EG 12, Ounda Gona OGS-6 and OGS-7,
Lokalalei or Fejej FJ-1, sites with an age between
2.5 and 1.9 million years, where stone industries are
characterized by the dominance of non-modiﬁed
ﬂakes, mostly unifacial cores, some pebble tools,
numerous percussion instruments and the absence
of standardized small retouched tools on ﬂakes or
debris.
2. The Oldowan s. s. (sensu stricto) or Classical Oldowan,
like at Olduvai DK 1 and FLK NN 1, with an
age of between 1.9 and 1.6 million years, also characterized
by the dominance of non-modiﬁed ﬂakes,
presence of unifacial and sometimes bifacial cores,
pebble tools, percussion instruments, and where
standardized small retouched tools appear (end
scrapers, scrapers, denticulates), as well as polyhedrons
and spheroids.
Pre-Oldowan industries are characterized by a very
strong dominance of non-modiﬁed ﬂakes, a large proportion
of cores, the presence of pebble tools, a relative
abundance of percussion instruments and the absence of
standardized intentionally retouched small tools. They
correspond with an early stage of hominid cultural evolution
whose degree of cognition did not yet allow them
to realize speciﬁc small sized tools or for relative standardization
in their tool manufacture.
The term Pre-Oldowan is a practical denomination
to designate a cultural horizon whose technological and
typological characteristics are more archaic than those
of the Oldowan s. s. or Classical Oldowan and correspond
to comparative behaviour between diverse groups
of hominids living in East Africa between 2.55 and 1.9
million years ago.
Outside of Africa, archaic Pre-Oldowan industries
are present at Yiron in the Levant, as early as 2 million
years ago and at Dmanissi in Georgia 1.81 million years
ago.
They are present in Mediterranean Europe as early
as 1.2 million years ago at Barranco León and at Fuente
Nueva 3 in Andalucia, as early as 1.1 million years ago
at La Sima del Elefante in the Atapuerca Sierra, between
1.07 and 0.984 million years ago at the Vallonnet Cave
site in the Alpes-Maritimes and at Ca’Belvedere di Montepoggiolo
in Italy around 1 million years ago.
An important time-lag thus separates these assemblages
from Classical Oldowan assemblages in East Africa
and Mediterranean Europe where standardized small
retouched tools on ﬂakes or fragments were sometimes
made, as well as spheroids: DK 1 at Olduvai around 1.9
million years ago, FLK NN 1 at Olduvai around 1.8 million
years ago, Terrassa in Catalonia around 0.9 million
years ago, Ceprano in Itay and Gran Dolina TD4 TD 5
and TD 6 in Spain around 0.8 million years ago and at la
Pineta at Isernia around 0.65 million years ago.
The large chronological gape separating the emergence
in Africa of these major cultural horizons tied to
the development of cognitive capacities of hominids and
their presence in Europe bring to light the very slow spatial
diffusion of these new cultural acquisitions.
If all these archaic stone industries, called here PreOldowan
or archaic Oldowan, whose technological and
typological characteristics correspond to the earliest
stage of human cultural evolution tied to the attainment
of a certain cognitive level, are very similar, they may
nonetheless present a certain variability conditioned by
the environment and human adaptation to milieu.
86  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Thus the industry from Gona EG 10 and EG 12 was
essentially knapped from trachyte, rhyolite and basalt
pebbles of excellent quality abundant in the Gona alluvials,
those from Fejej FJ-1 in quartz pebbles relatively numerous
in the alluvials of water courses from the Hamar
mountain range, those from Dmanissi in various rock
types from the Mashavera and Pinezaouri rivers, while
those from Barranco León and Fuente Nueva 3 were
made from ﬂint nodules from nearby Jurassic limestone
bordering the Guadix-Baza basin (Celiberti et al., 2004;
de Lumley et al., 2005).
Well made pebble tools are quite well represented at
Gona EG 10 and EG 12, at Fejej FJ-1 and at Dmanissi,
but are very rare in the Barranco León and Fuente Nueva
3 assemblages because of the absence of alluvial layers
with pebbles in the environment near the sites. Pebbles
are replaced by Jurassic limestone blocks whose edges
show irregular use wear attesting to their use as cutting
instruments.
A common structure between lithic assemblages is
tied to a speciﬁc cognitive level, where non-modiﬁed
small ﬂakes dominate, most probably used for cutting or
scraping meat off bones abandoned by scavengers, and
by the absence or extreme scarcity of small retouched
tools on ﬂakes or fragments, but a relative variability due
to environmental constraints; thus may be viewed the arFigure
19. Sites with Pre-Oldowan or Archaic Oldowan type industries, without small, intentionally retouched tools,
corresponding to a behavioural and cognitive parallelism.
de Lumley et al.  87
chaic stone industries from East Africa up to the shores
of Mediterranean Europe.
With the emergence of conceptual thought as early
as 2.55 million years ago, hominids were capable of
conceiving and manufacturing a model. That model was
small non-modiﬁed ﬂakes for cutting up meat or pebble
tools for slicing. With the slow conquest of the world by
early humans, this cognitive capacity was to be diffused
up to the gateways of Europe, at Dmanissi 1.81 million
years ago and in Mediterranean Europe, in the GuadixBaza
basin at Barranco León and Fuente Nueva 3 and
the Sima del Elefante at Atapuerca 1.2 million years ago.
Pre-Oldowan lithic assemblages, as well as the Oldowan
lithic assemblages which followed them in time,
each correspond to behavioural and cognitive parallelisms
of hominids groups present in diverse regions of
the world and that may be largely separated in time
(Figure 21).
REFERENCES
Antoniazzi, A., Cattani, L., Cremaschi, M., Fontana, L.,
Peretto, C., Posaneto, R., Proli, F., Ungaro, S., 1988. Le
gisement du Paléolithique inférieur de Ca’Belvedere
di Monte Poggiolo (Forli, Italie). Preliminary results.
L’Anthropologie 92, 629-642.
Figure 20. Sites with Oldowan sensu stricto or Classical Oldowan type industries, with small, intentionally retouched
tools, corresponding to an identical behavioural and cognitive stage.
88  The Cutting Edge: New Approaches to the Archaeology of Human Origins
Figure21.MainmajorculturalsequencesinAfrica,theNearEast,TransCaucasiaandsouthernEuropeintheirgeochronologicalandpaleoclimaticQuaternaryframework.
Importanttemporalgapsseparatingtheemergenceofdifferentcultures,correspondingtoidenticalbehaviouralandcognitivestagestiedtoacertaincognitivelevel,
maybeobservedindifferentlargeregionsoftheworld.
de Lumley et al.  89
Arzarello, M., Marcollni, F., Pavia, G., Pavia, M., Petronio,
C., Petrucci, M., Rook, L., Sardella, R., 2006. Evidence
of earliest human occurrence in Europe: the site of Pirro
Nord (Southern Italy). Naturwissenschaften 94, 107-112.
Ascenzi, A., Biddittu, I., Cassoli, P.F., Segre, A.G., SegreNaldini,
E., 1996. A calvarium of late Homo erectus from
Ceprano, Italy. Journal of Human Evolution 31, 409-423.
Ascenzi, A., Segre, A.G., 1997. Discovery of a Homo erectus
calvarium at Ceprano, central Italy. Anthropologie 35,
241-246.
Barsky, D., Cauche, D., Celiberti, V., Pleurdeau, D., Beyenne,
Y., de Lumley, H., 2006. les industries lithiques
archaïques du site de Fejej, Ethiopie. XIVèrne
Congrès
de I’Union Internationale des Sciences Préhistoriques et
Protohistoriques, Liege (Septembre 2001), pp.1-9.
Biddittu, I., 1971. Paleolitico inferiore di Arce e Fontana Liri
(Frosinone). Archivio per I’Antropologia e la Etnologia,
vol. CI, Firenze: Stamperia Editoria1e Parenti, pp.
251-254.
Biddittu, I., 1972. Pleistocene e industrie 1itiche pre-acheuleane
ad Arce e Fontana Liri (Frosinone). Quaternaria 16,
35-52.
Biddittu, I., 1983. II paleolitico inferiore del Lazio, industrie
archaiche. In: Atti XXIII riunione scientiﬁca dell’ Istituto
italiano di Preistoria e Protostoria, Firenze.
Biddittu, I., 1984. Le piu antiche industrie del Paleolitico
inferiore del Lazio. Atti della XXIV riunione scientiﬁca
dell’ Istituto Italiano di Preistoria e Protostoria. 1982. pp.
31-38. Roma.
Biddittu, I., Segre, A.G., 1982. Pleistocene medio-inferiore
con industria arcaica su ciottolo nel bacino di Anagni
(Lazio). Atti della XXIII riunione scientiﬁca dell’Istituto
Italiano di Preistoria e Protostoria, 7-9 maggio 1980,
Firenze, pp. 567-576.
Carbonell, E., Bermúdez de Castro, J.M., Arsuaga, J.L., Diez,
C.J., Rosas, A., Cuenca, G., Sala, R., Mosquera, M.,
Rodriguez, X.P., 1995. Lower Pleistocene Hominids
and artifacts from Atapuerca TD-6 (Spain). Science 269,
826-830.
Carbonell, E., Bermúdez de Castro, J.M., Parés, J.M., PérezGonzalez,
A., Cuenca-Bescós, G., Ollé, A., Mosquera,
M., Huguet, R., van der Made, J., Rosas, A., Sala, R.,
Vallverdú, J., Garcia, N., Granger, D.E., Martinón-Torres,
M., Rodriguez, X.P., Stock, G., Vergès, J.M., Allué,
E., Burjachs, F., Cáceres, I., Canals, A., Benito, A., Diez,
C., Lozano, M., Mateos, A., Navazo, M., Rodriguez,
J., Rosell, J., Arsuaga, J.L., 2008. The ﬁrst hominin of
Europe. Nature, Letters, vol. 452/27, March 2008, doi:
10.1038/nature 06815, pp. 465 à 469.
Carbonell, E., Mosquera, M., Olle, A., Rodriguez, X.P.,
Sahnouni, M., Sala, R., Verges, J.M., 2001. Structure
morphotechnique de l’industrie lithique du Pléistocène
inférieur et moyen d’Atapuerca (Burgos Espagne).
L’Anrhropologie 105, 259-280.
Cauche, D., Celiberti, V., Barsky, D., Notter, O., Biddittu,
I., de Lumley, H., 2004. les plus anciennes industries
lithiques du Latium, Italie. Actes du XIVème
Congrès de
l’UI.S.P.P., Université de Liège, Belgique, 2-8 septembre
2001, section 4, Premiers hommes et Pa1éolithique
inférieur. BAR International Series 1272, 49-57.
Celiberti, V., Barsky, D., Cauche, D., Notter, O., Nioradze,
M., Lordkipanidze, D., Gabunia, L., de Lumley, H.,
2004. les industries lithiques archaïques du site de Dmanissi,
Géorgie. Actes du XlVème
Congrès de l’UI.S.P.P.,
Université de Liège, Belgique, 2-8 septembre 2001, section
4, Premiers hommes et Paléolithique inférieur. BAR
International Series 1272, 29-36.
Cuenca-Bescós, G., Rofes, J., 2004. Insectívoros (Mammalia),
clima y paisaje de los niveles inferiores de Trinchera
Elefante (Pleistoceno inferior, Atapuerca). Zona Arqueológica
4, 150-156.
Delagnes, A., Roche, H., 2005. Late Pliocene hominid knapping
skills: The case of Lokalalei 2c, West Turkana,
Kenya (2005). Journal of Human Evolution 48, 435-472.
de Lumley, H., 2006. Il y a 2.5 million d’années… Un seuil
majeur de l’hominisation. L’émergence de la pensée conceptionnelle
et des stratégies maîtrisées du débitage de la
pierre. Palevol. Comptes Rendus de l’Académie des Sciences,
Paris, Paléontologie Humaine et Préhistoire, tome
5, fascicule 1-2, janvier-février 2006, n° 1-2, p. 119-126.
de Lumley, H., 2007. La grande histoire des premiers Hommes
européens. Édition Odile Jacob, 269 pages.
de Lumley, H., Beyene, Yonas (Ed.), 2004a. Les sites préhistoriques
de la région de Fejej, Sud-Omo, Ethiopie, dans
leur contexte stratigraphique et paléontologique. Préfaces
de Dominique de Villepin et de Teshome Toga. ADPF,
Association pour la diffusion de la pensée française. Editions
Recherche sur les civilisations. Ministère des Affaires
Etrangères. Direction générale de la Coopération
internationale et du développement, sous-direction des
Sciences sociales et de l’Archéologie, 18 articles, 635 p.
de Lumley, H., Beyene, Y., Barsky, D., Byrne, L., Camara,
A., Cauche, D., Celiberti, V., Fournier, A., Pleurdeau,
D., 2004b. L’industrie lithique préoldowayenne du
site de Fejej FJ-I, in «les sites préhistoriques de la
région de Fejej, Sud-Omo, Ethiopie, dans leur contexte
stratigraphique et paléontologique», édité sous la
direction de Henry de Lumley et Yonas Beyene. Preface
by Dominique de Villepin and Teshome Toga. ADPF,
Association pour la diffusion de la pensee française.
Editions Recherche sur les civilisations. Ministère des
Affaires Etrangères. Direction générale de la Coopération
intemationale et du développement, sous direction
des Sciences sociales et de l’Archéologie, pp. 391-564.
de Lumley, H., Fournier, A., Krzepkowska, J., Echassoux, A.,
1988. L’industrie du Pléistocène inférieur de la grotte du
Vallonnet, Roquebrune-Cap-Martin, Alpes-Maritimes.
L’Anthropologie 92, 501-614.
de Lumley, H., Lordkipanidze, D., Feraud, G., Garcia, T., Perrenoud,
C., Gagnepain, J., Saos, T., Voinchet, P., 2002.
Datation par la méthode 40Ar/39Ar de la couche de
cendres volcaniques (couche VI de Dmanissi, Géorgie)
qui a livré des restes d’hominidés fossiles de 1.81 Ma.
Comptes Rendus Palevol 1, 181-189.
de Lumley, H., Nioradze, M., Barsky, D., Cauche, D., Celiberti,
V., Nioradze, G., Notter, O., Zvania, D., Lordkipanidze,
D., 2005. les industries lithiques préoldowayennes
du début du Pléistocène inférieur du site de Drmanissi en
Géorgie. L’Anthropologie 109, l-182.
90  The Cutting Edge: New Approaches to the Archaeology of Human Origins
de Lumley, M.A., Gabounia, L., Vekua, A., Lordkipanidze,
D., 2006. Les restes humains du Pliocène ﬁnal et du
début du Pléistocène inférieur de Dmanissi, Géorgie
(1991-2000). I–Les crânes, D 2280, D 2282, D 2700.
L’Anthropologie, 110, 1-110.
de la Torre, I., Mora, R., 2005. Technological strategies in the
Lower Pleistocene at Olduvai Beds I & II, Erau1 112,
Etudes et Recherches Archéologiques de l’Université de
Liège, 2005.
Dominguez-Rodrigo, M., Pickering, T.R., Semaw, S., Rogers,
M.J., 2005. Cutmarked bones from Pliocene archeological
sites at Gona, Afar, Ethiopia: implications for the
function of the world’s oldest stone tools. Journal of
Human Evolution 48, 109-121.
Echassoux, A., 2004. Étude taphonomique, paleoecologique
et archeozoologique des faunes de grands mammifères
de la seconde moitié du Pléistocène inférieur de la grotte
du Vallonnet (Roquebrune-Cap-Martin, Alpes-Maritimes,
France). L’Anthropologie 108, 11-53.
Gabounia, L., de Lumley, M.A., Vekua, A., Lordkipanidze,
D., de Lumley, H., 2002. Découverte d’un nouvel hominidé
à Dmanissi (Transcaucasie, Géorgie). Comptes
Rendus Palevol I, 243-253.
Gabounia, L., Vekua, A., Lordkipanidze, D., Swischer
III, C.C., Ferring, R., Justus, A., Nioradze, M., Tvalchrelidze,
M., Anton, S.C., Bosinski, G., Jöris, O., de
Lumley, M.A., Majsuradze, G., Mouskhelishvili, A.,
2000. Earliest Pleistocene hominid cranial remains from
Dmanisi, Republic of Georgia: taxonomy, geological
setting and age. Science 288, 1019-1025.
Garcia, N., Howell, F.C., 2008. New discovery of a large mustelid
(Carnivora: Mammalia) from the early Pleistocene
locality of Sima del Elefante (Sierra de Atapuerca,
Spain). Palaeontographica Abteilung A 283.
Huguet, R., 2007. Primeras ocupaciones humanas en la Península
Ibérica: paleoeconomia de la Sierra de Atapuerca
(Burgos) y la Cuenca de Guadiz-Baza (Granada) durante
el Pleistoceno Inferior. Ph.D. Dissertation, Dept. of
History and History of Art, University Rovira i Virgili,
Taragona, Spain.
Leakey, M.D., 1971. Olduvai Gorge. volume 3. Excavations
in Beds I and II 1960-1963 Cambridge University Press.
Lordkipanidze, D., Jashashvili, T., Vekua, A., Ponce de Leon,
M.S., Zollikofer, C.P.E., Rightmire, G.P., Pontzer, H.,
Ferring, R., Oms, O., Tappen, M., Bukhsianidze, M.,
Agusti, J., Kahlke, R., Kiladze, G., Martinez-Navarro,
B., Mouskhelishvili, A., Nioradze, M., Rook, L., 2007.
Postcranial evidence from early Homo from Dmanisi,
Georgia. Nature 449, 305-310.
Mallegni, F., Carnieri, E., Bisconti, M., Tartarelli, G., Ricci,
S., Biddittu, I., Segre, A., 2003. Homo cepranensis sp.
novo and the evolution of African-European middle
Pleistocene hominids. CR. Palevol 2, 153-159.
Martinez-Navarro, B., Espigares, M.P., Ros, S., 2003. Estudio
preliminar de las asociaciones de grandes mamiferos de
Fuente Nueva-3 y Barranco Leon-5 (Orce, Granada, Espana)
(informe de las campanas de 1999-2002). In: Toro,
I., Agusti, J., Martinez-Navarro, B. (Eds.), Excavaciones
arqueologicas en los yacimientos del Pleistoceno inferior
de Barranco Leon y Fuente Nueva 3, Orce (Granada).
Memoria Cientiﬁca Campanas 1999-2002. Junta de
Andalucia. Consejeria de Cultura. E.P.G. Arqueologia
Monograﬁco, pp. 115-136.
Messager, E., 2006. Apports des analyses paléobotaniques à la
reconstitution paléoenvironnementale du site de Dmanisi
et de sa région (Géorgie). Thèse de Doctorat. du Muséum
National d’Histoire Naturelle. Quaternaire: Géologie,
Paléontologle humaine, Préhistoire, Institut de Paléontologie
Humaine, Paris, samedi 2 décembre 2006, 243 p.
Parés, J.M., Pérez-González, A., Rosas, A., Benito, A., Bermúdez
de Castro, J.M., Carbonell, E., Huguet, R., 2006.
Matuyama-age lithic tools from the Sima del Elefante
site, Atapuerca (northern Spain). Journal of Human Evolution,
50, 163-169.
Peretto, C. (Ed.), 1994. Le industrie litiche del giacimento
paleolitico de Isernia la Pineta. Isernia Cosmo Iannone
ed. Istituto Regionale per gli studi storici del Molise «V.
Cuoco», 433 pages.
Peretto, C., 2000. les outillages sur galet du site de Ca’Belvedere
di Monte Poggiolo (Forli, Italie). Résumé des
communications. Colloque de Tautavel sur «les Premiers
habitants de l ‘Europe», p. 99.
Peretto, C., Amore, F.O., Antoniazzi, A., Bahain, J.J., Cattani,
L., Cavallini, E., Esposito, P., Falgueres, C., Gagnepain,
J., Hedley, I., Laurent, M., Lebreton, V., Longo,
L., Milliken, S., Monegatti, P., Olle, A., Pugliese, N.,
Renault-Miskovsky, J., Sozzi, M., Ungaro, S., Vannucci,
S., Verges, J.M., Wagner, J.J., Yokoyama, Y., 1998.
L’industrie lithique de Ca’Belvedere di Monte Poggiolo:
stratigraphie, matière première, typologie, remontages et
traces d’utilisation. L’Anthropologie 102, 343-465.
Plummer, T., Bishop, L.C., Ditchﬁeld, P., Hicks, J., 1999. Research
on late Pliocene Oldowan sites at Kanjera South,
Kenya. Journal of Human Evolution 36, 151-170.
Roche, H., Delagnes, A., Brugal, J.P., Feibel, C., Kibunjia,
M., Mourre, V., Texier, P.J., 1999. Early hominid stone
tool production and technical skill 2.34 Myr ago in West
Turkana, Kenya. Nature 399, 57-60.
Rofes, J., Cuenca-Bescos, G., 2006. First evidence of the
Soricidae (Mammalia). Asoriculus gibberodon (Petényi,
1864) in the Pleistocene of North Iberia. Rivista Italiana
di Paleontologia e Stratigraﬁa 112, 301-315.
Ronen, A., 1991. The Yiron Gravel lithic assemblage. In:
Artifacts older than 2.4 My in Israel Archeologisches
Korrespondenzblatt Verlag des Romisch-germanischen
Zentralmuseum, Mainz. pp. 159-164, 21.2.
Ronen, A., 2006. The oldest human groups in the Levant.
Compte Rendu Palevol 5, 343-351.
Ronen, A., Inbar, M., Klein, M., Brunnacker, K., 1980. Artifacts-bearing
gravels beneath the Yiron Basalt. Israelian
Journal of Earth Sciences 29, 221-226.
Rosas, A., Pérez-Gonzàlez, A., Carbonell, E., van der Made,
J., Sánchez, A., Laplana, C., Cuenca-Bescós, G., María
Parés, J., Huguet, R., 2001. Le gisement pléistocène de
la “Sima del Elefante” (Sierra de Atapuerca, Espagne).
L’Anthropologie 105, 301-312.
Rosas, A., Parés, J.M., Sánchez, A., Bastir, M., MartínezMaza,
C., Benito, A., Saladie, P., Sala, R., Rodríguez, J.,
Martínez-Pérez, R., García, N., Allué, E., van der Made,
J., Vallverdú, J., Bermúdez de Castro, J.M., Carbonell,
E., Pérez-González, A., Huguet, R., 2006. The Sima del
Elefante cave site at Atapuerca (Spain). Estudios Geologicos
62, 327-348.
Semaw, S., 2000. The world’s oldest stone artefacts from
Gona, Ethiopia. Their implications for understanding
stone technology and patterns of human evolution
between 2.6-1.5 million years ago. Journal of Archaeological
Science 27, 1197-1214.
Semaw, S., 2005. Les plus anciens artefacts lithiques (2.6-2.5
millions d’années) des sites archéologiques du pliocène
ﬁnal de EG-10 et EG-12 in Gona Est, Afar, Ethiopie. In:
Sahnouni, M. (Dir.), Le Paléolithique en Afrique. L’histoire
la plus longue. Editions Artcom. pp. 13-52.
Semaw, S., Renne, P.R., Harris, J.W.K., Feibel, C.S., Bernor,
R.L., Fasseha, N., Mowbra, Y.K., 1997. 2.5 millionyear-old
stone tools from Gona, Ethiopia. Nature 385,
333-336.
Semaw, S., Rogers, M.J., Quade, J., Renne, P.R., Butler, R.F.,
Dominguez-Rodrigo, M., Stout, D., Hart, W.S., Pickering,
T., Simpson, S.W., 2003. 2.6 million-year-old stone
tools and associated bones from OGS-6 and OGS-7,
Gona, Afar, Ethiopia. Journal of Human Evolution, 45,
169-177.
Toro, I., Agusti, J., Martinez-Navarro, B. (Ed.), 2003a. El
pleistoceno inferior de-Barranco Leon y Fuente Nueva 3,
Orce (Granada). Memoria cientiﬁca campañias 1999-
2002. Arqueologia Monograﬁas. Junta de Andalucia,
Consejeria de Cultura, Direccion Général de Bienes
culturales, 206 pages.
Toro, I., de Lumley, H., Barsky, D., Celiberti, V., Cauche, D.,
Moncel, M.H., Fajardo, B., Toro Cano, M., 2003b. Las
industrias liticas de Barranco Leon y Fuente Nueva 3 de
Orce. Estudio tecnico y tipologico. Las cadenas operativas.
Amilisis traceologico. Resultados preliminares. In:
Toro, I., Agusti, J., Martinez-Navarro, B. (Eds.), Memoria
cientiﬁca campañas 1999-2002. Junta de Andalucia.
Consejeria de Cultura. E.P.G. Arqueologia Monograﬁco.
pp. 183-206.
Toro, I., Turq, A., Agusti, J., Martinez-Navarro, B., Oms,
O., 2002. Los yacimientos de Barranco Leon y Fuente
Nueva 3 de Orce (Granada). Contribucion al conocimiento
del primer poblamiento humano de Europa.
SPAL 9 (2000), 179-188.
de Lumley et al.  91
92  The Cutting Edge: New Approaches to the Archaeology of Human Origins