You are on page 1of 25

Quaternary International 350 (2014) 180e204

Contents lists available at ScienceDirect

Quaternary International
journal homepage: www.elsevier.com/locate/quaint

Fragmented reduction processes: Middle Palaeolithic technical


behaviour in the Abri du Maras shelter, southeastern France
le
ne Moncel a, *, Mara Gema Chaco
 n a, b, c, Alice La Porta a, Paul Fernandes d,
Marie-He
Bruce Hardy e, Rosalia Gallotti f
D
epartement de Pr
ehistoire, UMR 7194, Les Hominid
es au Quaternaire, Mus
eum National d'Histoire Naturelle, Paris, France
 de Paleoecologia Humana i Evolucio
 Social, C/ Marcell Domingo s/n Campus Sescelades URV (Edici W3), 43007 Tarragona, Spain
IPHES, Institut Catala
c 
ria, Universitat Rovira i Virgili (URV), Avinguda de Catalunya 35, 43002 Tarragona, Spain
Area de Prehisto
d
Pal
eotime, Villars de Lans, France
e
Department of Anthropology, Kenyon College, Gambier, OH 43022, USA
f
Universit
e Bordeaux 1, UMR 5199 PACEA-PPP, Talence, France
a

a r t i c l e i n f o

a b s t r a c t

Article history:
Available online 5 June 2014

New excavations at the Abri du Maras, located in the southeast of France, have yielded Middle Palaeolithic assemblages with evidence of rock shelter occupations in a cold climatic context contemporaneous with MIS 4. Few MIS 4 sites are known in this part of France and especially in this state of
preservation. The paper is focused on one sedimentological layer divided into two archaeological levels
(sub-levels 4.1 and 4.2). Our goal was to examine the Middle Palaeolithic lithic assemblage of these two
levels by interdisciplinary approaches (technology, origin of int and functional analysis of stone tools) in
order to identify the technical strategies and the land-use patterns in a specic environmental context.
The two occupations do not show differences in behaviours. The technical strategies applied to int and
other stones indicate a fragmentation of the reduction processes in a local and semi-local perimeter
around the site. The main core technology is Levallois, generally on int cortical cores on akes. Flint
akes, blades and points are the main components of the series and the technological aims of the
debitage. Due to the small size of the akes used for aking, large int akes, blades (Levallois or cortical)
and Levallois points were produced elsewhere, to the north and south of the site (up to 20e30 km)
according to the geological study, and then brought to the shelter. Flakes in other lithic materials (quartz,
quartzite) were also knapped elsewhere before being transported to the shelter. Some of the large int
akes, but also nodules and fragments of slabs, were then used for onsite aking. Flake-tools are very
rare. Evidence of impact fractures and TCSA/TCSPs values of the corpus of unretouched Levallois points
suggest that some points, brought or prepared on the site, could have been used as projectile tips. The
lithics attest to management of local and semi-local stones in a perimeter of 30 km around the site
(possibly more due to some unidentied int) and an anticipation of domestic needs in relation to
reindeer hunting, the predominant activity. Imported artefacts and artefacts made on the site were used
for the same diversity of activities and materials (butchery, plant and woodworking). The technological
strategies and the type of management differ slightly from those from cave assemblages in the same area
^ ne corridor, possibly due to the type of the site, a vast
located in valleys and on low plateaux near the Rho
shelter. Data from the Abri du Maras are compared to data from the other Middle Palaeolithic sites of the
region and the role of the topographic aspect of the site on the type of occupations is discussed.
2014 Elsevier Ltd and INQUA. All rights reserved.

Keywords:
Upper Pleistocene
Technical behaviour
Neanderthals
Shelter
Southeastern France

1. Introduction

ontologie Humaine, MNHN,


* Corresponding author. Prehistory, Institut de Pale
 Panhard, 75013 Paris, France.
CNRS, 1 rue Rene
E-mail address: moncel@mnhn.fr (M.-H. Moncel).
http://dx.doi.org/10.1016/j.quaint.2014.05.013
1040-6182/ 2014 Elsevier Ltd and INQUA. All rights reserved.

The Abri du Maras site is located in a small valley less than 1 km


che River, a tributary of the Rho
^ ne River and close to
from the Arde
^ ne Valley in southeast France. This site is renowned for the
the Rho
Middle Palaeolithic (MP) deposits at the top of the sequence
bearing Levallois laminar debitage and seven distinct levels (levels

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

181

che River. Aspect of the present shelter.


Fig. 1. Location of the Abri du Maras in southeastern France. View of the topographical position of the site in a small valley near the Arde

182

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

8 to 2) with MP assemblages beneath this level (Baudet et al., 1955;


Combier, 1967; Moncel, 1994, 1996; Moncel et al., 1994). Little was
known of the bottom of the sequence due to limited excavations in
these levels during early eldwork. Since 2006, new excavations
have thus focused on this part of the sequence. The geological study
demonstrates that the roof of the shelter collapsed over time and
that the most recent occupations were installed underneath a small
shelter (Debard, 1988). The oldest occupations occurred below a
large shelter and are currently located in front of the present
shelter.
The aim of our research program is to provide as much data as
possible on Neanderthal occupations in their environmental
context in order to describe subsistence strategies and land use
through time and their possible associations with climatic changes.
This paper presents the rst results relating to the lithic series from
new excavations at the Abri du Maras, with particular emphasis on
layer 4, in which two archaeological levels have been distinguished.
We will focus on technical strategies, core technology, retting and
tool function.
^ne Valley: a
The geographic area under consideration is the Rho
vast corridor linking the north of Europe to the Mediterranean
zone, which is a key position between two mountainous regions;
the Massif Central to the west and the Alpine foothills to the east.
Rivers, valleys and limestone plateaux with many shelters and
caves provided a propitious environment for recurrent Neanderthal
occupations during the Middle and Upper Pleistocene, before the
arrival of the rst modern humans and Aurignacian cultures.
^ne Valley has yielded more
The right bank of the Middle Rho
than ten Middle Palaeolithic sites. They are located on low plateaux
^ ne corridor
and along rivers between the Massif Central and the Rho
(Moncel, 2003; Moncel et al. (Ed), 2008a; Moncel et al., 2008b,
che gorges
2010). Most of the sites are clustered around the Arde
and are cave sites or shelters, such as the Abri du Maras, but some,
such as the site of Payre, are further north (Moncel et al., 2008a).
Only one open-air site has been studied in detail (Saint-Bauzile;
Bernard-Guelle et al., 2011). Most of the sites are cave entrances
with vast, easily accessible entranceways but some are narrow
chambers in steep cliff faces. The context of most of these sites is
well-suited to hominin occupation: the cave openings are generally
south-facing and most of the sites have quasi-direct access to rivers
(most of the rivers were almost 40 m higher than present day levels
(Debard, 1988)). Moreover, the sites have a wide circulation zone
nearby, made up of the Rhone valley and its tributaries, and offer an

extensive range of local faunal and lithic resources. All of them are
located at the centre of varied biotopes, including plains, gentle and
steep-sloped forested plateaux and valleys. The faunal assemblages
indicate that abundant large and medium-sized herbivores were
present, representing the three main types of environment
(Daujeard and Moncel, 2010).
2. A vast rock shelter along the Rhone valley: Abri du Maras
The Abri du Maras is a shelter with an elevation of 170 m a.s.l.
che
and opens towards the southeast some 70 m above the Arde
River and would have been ~30 m above the river at the human
occupations (Fig. 1). Recent excavations revealed three sedimentary
layers. These layers are discontinuous and show a complex geometry which is the result of the multi-phased inlling in which
deposition, erosion and post-depositional phenomena occurred.
The oldest layer was designated layer 5 (levels 8 to 6 of earlier
excavations), and consists of an organic brown level with a sandyesilty matrix, overlying the limestone substratum. It is truncated
at the front of the site and covered by a sterile layer of rubble with a
reddish clayesilt matrix (upper layer 5). It probably completely
lled the thalweg before the Holocene incision. The next layer is
layer 4 (levels 5 and 4 of earlier excavations), made up of homogeneous aeolian silts (loess) with some large blocks and levels of
small limestone blocks. It is overlain by three levels of rubble with a
silty matrix and limestone slabs (levels 3, 2, and 1), similar to those
described during previous excavations and contains the most
recent occupations (Moncel et al., 2010).
Rich lithic and faunal remains and hearths characterize layer 4
which has been excavated over more than 40 m2 and contains two
phases of human occupations. Based on the vertical and horizontal
spatial patterning, two archaeological levels have been identied in
layer 4, named levels 4.1 and 4.2. Within these levels, accumulation
of limestone blocks attest to occupations of the shelter during more
humid phases of MIS 4. In level 4.1, the presence of ash lenses or
diffuse lenses, burnt int and bones provide evidence of the use of
re. Level 4.1 was excavated over a more extensive surface (38 m2
for an average depth of 30 cm) than level 4.2 (21 m2 with a depth of
10e20 cm).
In the past, U/Th dating applied to bones from layers m-n (in the
middle part of the sequence) yielded ages of 72 3 ka, 87 5 ka,
89 4 ka, and 91 4 ka. These ages correspond to the summit of
our layer 5/bottom of our layer 4 and date these levels to the end of

Fig. 2. Example of a large Levallois int ake transported to the site already knapped (Drawing: A. Theodoropoulou).

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

MIS 5/beginning of MIS 4 (Moncel et al., 1994; Moncel and Michel,


2000). These ages are conrmed by new preliminary ESR-U/Th
dating on Cervus elaphus teeth sampled at the summit of layer 5
(95 6 ka) (Richard, 2012). There does not appear to be a long gap
between the deposition of layers 5 and 4. These results corroborate
those from the large mammals which point to a gradual transition
between a temperate and a cold period between layer 5 and layer 4
(see below).
Humans inhabited the site during temperate conditions (layer 5)
and abandoned it when forests declined (upper layer 5). Then,
humans settled during two humid periods in a pleniglacial phase
(levels of blocks) and abandoned the site when conditions were
cold (layer 4). In layer 4, at least two human occupation phases, 4.1
and 4.2, were identied, among the partial collapse of the shelter
roof (levels of limestone blocks).
The large mammals remains provide abundant paleoenvironmental data whereas pollen analysis carried out on the whole
sequence of the new excavations showed that it was almost sterile in
sporopollinic material. Charcoal from the hearths led to the identication of four different taxa, due to hominid gathering of different
woods, including boxwood (Buxus sempervirens), Fabaceae, juniper
(Juniperus sp.) and Scots pine (Pinus type sylvestris) (Moncel et al., in
press). Very few micromammal remains were found, possibly due
to taphonomic processes or the fact that the shelter was not suitable
for micromammals and birds of prey. Only a fragment of a rst lower
molar may be related to Microtus cf. arvalis (Moncel et al., in press).
The large faunal spectrum is composed, in order of abundance, of
Rangifer tarandus, Equus caballus spp., C. elaphus, Bison priscus, Capra
ibex, Megaloceros giganteus. There are no carnivore remains and no
chewing or ingestion marks (Hardy et al., 2013). Layer 4 also yielded
three skeletal elements with cut marks, which have been assigned to
Oryctolagus cuniculus. Thus, faunal remains are mainly related to
Neanderthal activities, in a site close to an area where large herds of
migrating reindeer sought out favorable crossing points beside the

Fig. 3. Example of a large int Levallois ake brought to the site already knapped
(Photo).

183

che River or along the vast corridor offered by the Rho


^ne Valley.
Arde
Data suggest that this rock shelter was used for large seasonal gatherings by Neanderthals, especially during reindeer migrations in the
autumn. The varied faunal spectrum of layer 4 points to cold and open
environments and is consistent with sedimentary data and dating.

3. Material and methodology


The lithic assemblages from layer 4 of the Abri du Maras were
analyzed using technological and typological analytic methods in
order to dene the techno-economic processes and technological
da, 1982, 1993, 1994; Peresani,
sequences (chane op
eratoire) (Boe
2003). Because Levallois points form a part of the technological
aims of the reduction processes, tip cross-sectional area (TCSA) and
tip cross-sectional perimeter (TCSP) have been tested on the whole
corpus of Levallois points and compared to triangular akes,
following Hughes (1998), Shea (2006), Sisk and Shea (2008, 2011),
Hardy et al. (2013). We also include the preliminary results of the
analysis of the Raw Material Units (RMUs) and rets made on the
material to identify the technical events from a spatio-temporal
perspective. RMUs can be dened as the material outcome of a
knapping event or a series of knapping events carried out from a
specic nodule (Roebroeks, 1988; Vaquero, 2008). This methodology has been applied to other Middle Palaeolithic sites previously
 n et al., in press; Machado et al., 2011, 2013; Vaquero et al.,
(Chaco
2012a, 2012b).
Raw material gathering and the outcrops, especially int outcrops, were identied in order to discuss the chanes op
eratoires in
relation to stone procurement. Flint types have been determined
following the methodology developed by Paul Fernandes
(Fernandes et al., 2008; Fernandes, 2012). Results from functional
analysis of stone tools have already been published and we will
refer to Hardy et al. (2013) to compare data with the technological
categories (akes, blades and points).
The study was carried out according to the two archaeological
levels, grouped by the density of artefacts (vertical spatial distribution) (Table 1). Most of the material is in int (Table 2). The
density of lithic material is ~142 pieces/m3 for level 4.1, the
largest excavated level (38 m2). Level 4.2 was excavated over a
surface of 21 m2 and a depth of 10e20 cm. Some rets attest to
the lack of mixing between the two levels. The cutting edges of
the artefacts are fresh. The high density of artefacts in some areas
among broken blocks issued from roof collapse suggests recurrent human occupations for each level, which cannot however be
differentiated. This is conrmed by the location of the ash lenses
at different depths in both sub-levels. A double patina on some

Fig. 4. Length/width of elongated int products (bladelets and blades) from level 4.1.

184

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Fig. 5. Examples of int Levallois points.

pieces indicates the re-use of pieces abandoned on the site or


collected elsewhere.

Table 1
Total number of artefacts from levels 4.1 and 4.2 including all raw materials.
Level 4.1

Level 4.2

Total

Flakes
Pebbles (entire or broken)
Blades and bladelets (length > 2 width)
Cores
Levallois cores
Points and triangular akes

1296
30
185
28
13
67

450
17
59
22
6
30

1746
47
244
50
19
97

Total

1620

586

2206

Unplotted micro-akes

2793

Table 2
Total number of artefacts according to the type of raw material.
4.1
95
13
10

4.2
21
5
3
1
1

Total

Undetermined stones
Basalt
Limestone
Gneiss
Granite
Sandstone
Quartz
Quartzite
Schist
Flint

10
4
97
2
2
1387

25
2
1
527

116
17
18
1
4
4
122
4
3
1914

Total

1620

586

2206

4. Flint core technology


4.1. Products of the Levallois processing system: micro-akes, small
akes, akes, backed akes, Kombewa akes and Levallois akes
Both series are composed of a large majority of akes measuring
between 10 and 60 mm (length measured according to the

technical axis of the piece). However, some are 60e100 mm


(Table 3, Fig. 2). The ake assemblages comprise cortical akes,
akes with no cortex, elongated and backed (cortical back or core's
edge) akes. Most of the products appear to belong to the different
phases of a Levallois processing system, and secondarily to a
Kombewa debitage (Kombewa akes from the ventral face of a
ake) or a discoid-type core technology (pseudo-Levallois akes).
Several small akes could come from the rejuvenation of Quina
scrapers.
Table 3
Distribution of the lithic assemblage by technological categories and levels.

Flakes
Backed akes
Flakes with a cortical back (including
pseudo-Levallois akes)
Levallois akes with a back
Kombewa akes
Levallois akes
First cortical akes
Micro-akes (<5 mm)
Small akes (5e10 mm)
Blades and bladelets
Crest blades
Backed blades and bladelets
Blades with a cortical back
Levallois blades
Points
Backed points
Points with a cortical back
Levallois points
Bifacial tools
Fragments of
Fragments of
Fragments of
Fragments of

Levallois akes
akes and chunks
blades and bladelets
int pebbles

Level 4.1

Level 4.2

789
51
27

183
24
22

2
1
74
4.6%
12
160
251
84
5.2%
2
12
8
48
2.9%
30
1
3
35
2.1%
1
2
328
25
2

e
6
26
4.4%
7
50
113
32
5.9%
1
3
3
16
2.7%
14
e
e
16
2.7%
1
e
64
4
1

Half of the group of akes with a length of 10e60 mm bears


cortical patches. Few are completely cortical. The huge quantity of
micro-akes (<5 mm) attests to aking and retouch activities in
da,
situ. Levallois akes (less than 5%) are rare (cf. denition of Boe
1993, 1994), as are Kombewa akes issued from a aked surface.

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

185

Fig. 6. The 11 techno-morphological Levallois point types.

The largest akes, some of which are backed, are more diversied,
but most of them are issued from Levallois aking.
The scars on the akes indicate that aking was practiced by
unipolar and unipolar convergent removals, followed by centripetal removals, throughout the phases of the reduction

processes. Backed akes attest to the use of d


ebordement for
managing the core sides and maintaining convexities. The
platform is generally only rarely prepared (high ratio of at
butts) and facetted platforms tend to be related to nal Levallois
akes.

Fig. 7. (A) Comparative boxplots of the TCSA distribution of Levallois points (lp) and triangular akes (tf), (B) Distribution of Levallois points TCSA at Maras shelter.

186

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

4.2. Blades and bladelets


Initially, elongated products, blades and bladelets, were distinguished from akes by the ratio length > 2  width. However, their
size demonstrates that we cannot clearly separate the two groups.
Most of them are grouped between 20 and 60 mm and are more or
less elongated. Some measure up to 100 mm long (Fig. 3). The
length/width ratio indicates various degrees of elongation for these
products (Fig. 4).
Blades and bladelets cannot be distinguished from a technological point of view. Most, whether backed or not, can be related to
Levallois core technology. Removals are unipolar/unipolar convergent (one or two guide-axes) or centripetal. Flat butts are the most
common. Some crest blades suggest the voluntary or accidental
truncation of the core edges.
Two thick and long blades (100 and 110 mm long) probably
indicate that the longest products could be produced by Levallois
aking or another core technology. This blade could come from a
laminar core with management of the core volume.

4.3. Levallois points


The Levallois points comprise a low proportion of artefacts
(Fig. 5) of varying size, elongation and technology, indicating that a
wide variety of modalities were used to produce them. The largest
ones (>60 mm), as for the blades and akes, are isolated from the
main group of sizes. Levallois points (described by the presence of a
central triangle on the proximal part and two lateral convergent
da, 1982, 1993, 1994;
negative removals on the edges) (Boe
Bonilauri, 2010) have been compared to triangular akes (two
convergent edges) issued from Levallois or discoid technology
(corpus of 162 pointed products).
The Levallois points may be classied into three groups
depending on the direction of the removals: (1) unipolar (scars
have the same direction as the point axis), (2) bipolar (scars are
parallel and from two opposite striking platforms), and (3)
orthogonal-peripheral (removals from the edges of the cores or
perpendicular to each other). Considering the organization and
convergence of the removals, several subgroups were theorized
(unipolar convergent, unipolar longitudinal, bipolar convergent,
bipolar longitudinal, peripheral, orthogonal and indeterminate). In
relation to the direction, the convergence and the number of
negative removals present on the dorsal face of the points, 11
different techno-morphological sub-types of Levallois points were
described (Fig. 6). Levallois points produced by ve main recurrent
or preferential operational schemes were identied: unipolar
convergent operational scheme, unipolar longitudinal operational
scheme, bipolar operational scheme, centripetal operational
scheme and peripheral-orthogonal. The main method used to
produce Levallois points is unipolar (67%), followed by peripheralorthogonal (11%), and rarely centripetal and bipolar (respectively
4% and 2%).
The morphometric tip cross-sectional area (TCSA) for our
corpus of points indicates low values (Table 4a) compared to
ethnographic darts and arrow tips and experimental spear
thrusting data (Table 4b; Shea, 2006) (Fig. 7) (La Porta, 2013).
Shea (2006) suggests that thrusting spear points have TCSA
values ranging from 100 to 250 mm2. Tips that have TCSA values
less than 100 mm2 are considered too fragile for thrusting and
are more likely to have been thrown. Levallois point TCSA values
fall well within the spear tip and the dart tip range (Table 4b).
The Levallois points are signicantly smaller than the spear tip

sample (t 0.999686473, p < 0.05; Shea, 2006) but signicantly


larger than the dart tips (t 1.17095E-07, p < 0.05). However,
there is some overlap with dart tips if you include the entire
range of the sample. We also observe that triangular Levallois
akes present the lowest TCSA value (Mean 102 mm), followed
by triangular akes (Mean 114.64). If we compared the TCSA
values of our Levallois points with the TCSA of MSA points from
Porc Epic, that have already been identied as plausible projectile
points (Sisk and Shea, 2011), the data show that the TCSA values
of our points are not signicantly different (t 0.087309,
p < 0.05) to the Porc Epic bifacial point sample 1 and are slightly
higher (t 0.004996, p < 0.05) than the Porc Epic bifacial point
sample 2 (Shea, 2006). If we compare the TCSA values of European Middle Palaeolithic points from the site of Le Moustier
(Shea, 2006) with the TCSA of our Levallois points, we observe
that our TCSA value is signicantly lower (t 1, p < 0.05) than the
Le Moustier TCSA, but clearly higher than Chatelperronian points
(t 1.46313E-06, p < 0.05) and Gravette points (t 5.30471E-11,
p < 0.05). The Maras Levallois point TCSA values are not signicantly different to those of Solutrean bifacial points
(t 0.004472124, p < 0.05).

Table 4a
TCSA values for the different classes of pointed tools from layer 4 at Abri du Maras.
The whole corpus of pointed artefacts (Corpus), Levallois points (LP), Levallois
triangular akes (LTF) and triangular akes (TF).
1a Samples

Mean

SD

Min

Max

Corpus
LP
LTF
TF

133.92
124.20
102.3
114.64

85.22
82.54
55.17
49.95

11
13
26
54

414
370
210
192

72
42
13
7

Table 4b
TCSA values from ethnographic and recent archaeological stone points given by Shea
(2006).
1b Samples

Mean

SD

Min

Max

Arrowheads
Dart tips
Spear tips

33
58
168

20
18
89

8
20
50

146
94
392

118
40
28

A small subset of artefacts showed evidence of impact fractures. As reported by Hardy et al. (2013), tip cross-sectional area
(TCSA) and tip cross-sectional perimeter (TCSP, calculated for
triangular cross-section for unifacial points or rhomboid crosssection for bifacial points) have been calculated for these artefacts (4 Levallois points, a ake and a triangular ake, Table 5).
Within this small sample, some of the artefacts have TCSA values
suggesting they were too thin and fragile to function effectively
as thrusting spear tips (Shea, 2006). Recently, TCSP has been
suggested as a more reliable measure of penetration for stone
points (Sisk and Shea, 2011). TCSPs of these artefacts fall well
within the range of various MSA assemblages analysed by Sisk
and Shea (2011). Using the methods they outline, these artefacts are good candidates for thrown projectile tips. Furthermore,
four of the six have TCSPs which fall within the range of ethnographic dart points. On balance, these artefacts from Maras match
the criteria proposed for thrown projectile points and some t
the criteria proposed for dart points used in a complex projectile
system (darts and spear throwers) suggesting the possible use of
multiple modes of projectile technology. For a more detailed
discussion, see Hardy et al. (2013).

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

187

Table 5
Possible projectile points with impact fractures (from Hardy et al., 2013).
Square

No.

Type

Width

Thickness

TCSA (mm2)

TCSPRhom (mm)

TCSP Tri (mm)

Impact type

Scar length (mm)

F6
I8
K6
K6
K7
L6

39
5
168
429
21
148

Triangular ake
Levallois Point
Levallois Point
Flake
Levallois Point
Levallois Point

37
26
32
37
26
30

10
6
5
8
9
7

185
78
80
148
117
105

76.7
53.4
64.8
75.7
55.0
61.6

79
54.6
65.5
77.3
57.6
63.1

Crushing and spin-off


Snap
Step termination
Burin-like
Burin-like
Burin-like

3.0
n/a
4.5
13.9
3.2
6.3

4.4. Cores
Different core technologies co-exist at the site in both levels. The
da, 1993, 1994) (Fig. 8). Some other cores
main one is Levallois (Boe
are cores on akes (some removals on the inferior face of the ake
without a specic preparation), discoid-type cores and other rare
types (cortical cores with few removals on one or two surfaces and
orthogonal cores) (Table 6, Fig. 9). Most of the cores, including the
Levallois cores, measure between 20 and 60 mm long and are made
on akes, except for some large nodules and a few worked slab
fragments.
Table 6
Number of cores according to core technology for each level.
Level 4.1
Fragments of cores
Fragments of Levallois cores
Fragments of orthogonal cores
Irregular cores
Bifacial discoid cores
Unifacial discoid cores
Crude cores
Kombewa cores
Levallois cores with a preferential ake
Levallois unipolar convergent cores (points)
Levallois centripetal cores
Orthogonal cores
Cores on akes
Trifacial cores

12
2
1
3
1
1
1
1
5
3
1
1
8
1

Total

41

Level 4.2

one is the production of points by unipolar convergent removals.


The at nal surface and the location of the axis do not suggest a
long recurrent production of pointed products. For cores with one
large ake, it is not possible to determine whether or not we are at
the end of the reduction process. Some show a large triangular scar.
Among the other cores, the discoid-type cores seem to belong to a
parallel reduction process to the Levallois core technology, made by
bifacial or unifacial management on nodules or thick akes and
angular/slab fragments.
The dominant use of akes for aking explains the number of
cores that we have grouped under the name of ake-cores by centripetal removals and the lack of prepared convexities. Some Kombewa cores are also present. The lower face of these is managed by a
small ake, by one invasive removal or by several centripetal removals. According to the type of blank, core technology differs:

- Cortical and small akes: Levallois technology, Kombewa technology, ake-cores. The preparation of the platform is limited.
- Fragments of at and thin slabs or nodules: (1) a short centripetal sequence of cortical removals on one or two faces
(discoid-type), (2) for a trifacial debitage using the faces of the
blank, (3) occasionally for Levallois debitage.

4
5
1

2
4
2
4

27

The distinction between Levallois cores and ake-cores was based


on the organization of removals and the management of the distal
and lateral core convexities. Most of the Levallois cores are made on a
large and cortical ake. Few core platforms are managed and suggest
low productivity of the aking surface (no traces of re-working the
core convexities and platforms). Some deep and small akes are
sometimes produced at the end using the remaining convexities (for
instance, distal convexity on the Levallois unipolar convergent).
Two groups dominate the Levallois cores. The rst group indicates the nal production of a large preferential ake. The second

Three cores are distinguished from the others by the following


technological features: (1) a fragment of nodule with two orthogonal surfaces, one Levallois unipolar convergent, the second one
with unipolar parallel scars; (2) a core with several orthogonal
surfaces with orthogonal aking surfaces. One surface is covered by
unipolar convergent removals, the others by short removals; and
(3) a very small core with one deep and short removal on the underside, a second deep and short removal on the ake edge.
4.5. A low ratio of int tools
Flake-tools are rare (Table 7) and located all over the excavated
area, probably due to the palimpsest of occupations for each level
(Fig. 13). This ratio is perhaps the only parameter which differentiates the two levels (n 38, 2.3% for sub-level 4.1 and n 30, 5.1%
for sub level 4.2).

Table 7
Example of the sub-level 4.2. Number of tools according to the type of blank.
Invasive and
thin removals
Fragments
Flakes
Flakes with a cortical back
Levallois akes
Blades
Blades with a cortical back
Bifacial tool on a slab
Points
Levallois points

Total

Becs

Denticulates

2
2

End-scrapers

Convergent
points
1
1

Peroirs

Scrapers
2
20
2
1
3

1
1
2
2
2

30

188

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Fig. 8. Examples of int cores: (A) Levallois core for points (B) Levallois core with an invasive pointed ake (C) Centripetal core on ake (D) Core on a fragment of slab with three
knapped faces.

Several tool groups have been identied: (1) lateral and


transversal scrapers with marginal retouch (Fig. 10), only two
bear Quina retouch (level 4.2, Fig. 11); (2) points with marginal
retouch; (3) micro and macro-denticulates; and (4) some becs
and borers.
The retouch does not modify the shape of the blank apart from
the two Quina scrapers made on large cortical akes and a ake
with no cortex (with partial bifacial retouch). Some artefacts
display thin and invasive retouch on the inferior face (for hafting?).
Most of the Levallois akes and blades are not retouched, regardless
of size.
Two small bifacial tools exist, one for each level (Fig. 12). One is
on a thin slab (level 4.2) with a partly worked periphery. The section is plano-convex and the retouch forms a rounded tip and a
lateral cutting edge opposite a back. The second one (level 4.1) is a
small and thick pointed tool (40 mm) with some invasive bifacial
removals.

4.6. Flint Raw Material Units (RMUs) and rets


The results presented here are preliminary, particularly for int,
as the study is still in progress. Badly preserved pieces (with patina
or thermal alteration) and pieces less than 1 cm long (n 921) have
not been taken into account. In some cases, the cortical aspect
(fossils, inclusions ) allowed us to indirectly associate pieces
belonging to the same nodule. Raw Material Units incorporates the
artefacts produced during one or several technical events on the
same nodule (Roebroeks, 1988; Vaquero, 2008). In some cases,
products can be retted into this technical event.
A total of 16 RMUs (n 356 pieces) have been identied for the
moment and correspond to 16 different int nodules. All of them
are concerned by knapping activities: akes of different types, ake
fragments and some cores (Fig. 14). Some RMUs indicate several
complete or almost complete knapping sequences conducted
onsite. All the technical categories are observed in the same RMU,

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

189

Fig. 9. Examples of int cores: (A) Discoid core (B) Levallois core on a ake (C) Bifacial core with centripetal removals on a nodule (D) Pyramidal core (discoid-type?) (Drawing: A.
Theodoropoulou).

Fig. 10. Example of a int scraper.

from the rst cortical akes to the products coming from the
different phases of the reduction sequences and cores. Others show
the fragmentation of the reduction sequences. The smallest int
RMU is formed by a single object (ake or retouched artefact).
These pieces were introduced to the site already knapped. Other
pieces (even small debris) cannot be retted to the same RMU. The
biggest RMU is in int (RMU int-1) and is formed by 105 pieces.
Results obtained by the RMU analysis were conrmed by the
rets. To date, a total of 13 int rets were identied (n 4 for level
4.1 and n 9 for level 4.2). They are formed by 31 pieces and
provide a ret rate of 1.4%. The technical categories include all the
phases of the chane op
eratoire: cores (n 2), akes (n 14), ake
fragments (n 8), fragmented akes (n 5) and retouched pieces
(n 2). The ret with the most elements is formed by 6 pieces,
followed by a ret of 3 pieces. The others (n 11) only join 2 pieces.
The total number of connection lines is 18 and they are issued
from knapping sequences (n 11) or fractures (n 7). A high
number of fracture connection lines are produced by post

190

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Fig. 11. Example of a int Quina scraper (Drawing: A. Theodoropoulou).

Fig. 12. Small int bifacial tools.

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

191

Fig. 13. Spatial distribution of the int tools by type for levels 4.1 and 4.2. Limits of the excavated area for each level indicated by black lines. The largest limestone blocks are in grey.

depositional factors (except two technical fractures including a Siret


breakage). The distances of the connection lines range from several
centimetres to two meters (the longest one) and are within the
normal dispersion range observed by experimental archaeology for
onsite knapping sequences (Cziesla, 1990).
The spatial patterning of the int rets does not show any
specic dispersion. In level 4.1, they are scattered over the whole

excavated surface, whereas in level 4.2 they are concentrated in the


middle part of the excavated area (line 6) (Fig. 15).
4.7. Fragmented int knapping system
Comparing core and product sizes, it is clear that part of the
series could not have been produced in situ, namely the pieces

192

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

ranging from 60 mm to 100 mm. Most of the cores are on akes and
are smaller than 60 mm. The rare nodules or fragments of slabs
measure around 50e70 mm. Depending on the reduction processes, these pieces could not have produced the longest products.
Thus, the akes used for knapping had to be brought onsite as well.
The largest cores on nodules are crude cores and could not have
supplied the largest akes for the core technology.
Consequently, the composition of our two series composed of
ake-cores, large akes, Levallois akes, points and blades, but also
cortical akes, small akes and micro-akes indicates (Fig. 16):
- Several complete and in situ chane op
eratoires, made on akes
for most of the cores. The main one was Levallois and produced
points and akes. Bladelets and small elongated products could
be accidentally produced from these cores. Some Kombewa
akes and the high number of 15 mm akes come from ake
knapping. Cores on ake and some discoid-type cores could be
related to a parallel core technology or a nal phase on Levallois
cores (Lenoir and Turq, 1995). No core may be clearly related to
the Quina system (Bourguignon, 1997).
- One or several partial external chanes op
eratoires which provided large akes for knapping or for direct use, large blades
and points (more than 60 mm long). Most of these artefacts are
not retouched. Moreover, as the numerous ake-cores could
not have produced the ratio of small cortical akes counted in
the series (25%), we deduce that some of the small cortical
products, whatever their size, could also have been brought to
the site already knapped, or else come from another part of the
site (eroded in front of the shelter or from a non-excavated
area). These products are mainly of Levallois-type (from the
different phases of the reduction process including the noncortical backed akes for maintaining the convexities) but
some thick blades could represent a form of laminar core
technology.
The in situ reduction processes indicate a real intent to produce
small pieces. However, long blades, large akes and points, such
as the large Quina scraper, conrm that large products were also
sought after (long cutting edges) even though they were not

produced onsite. The range of sizes of useful products is consequently wide.


The ratio of Levallois products and cores attests to the preferential use (in and outside the site) of this technology for producing
akes, elongated products and points. It is mainly recurrent, unipolar convergent or centripetal, followed by preferential akes
(some of which are pointed as suggested by the scars). Large akes
imported as matrixes for knapping or for direct use (Bourguignon
et al., 2004; Turq et al., 2013) conrm the application of fragmented
reduction processes characteristic of the Middle Palaeolithic. The
detection of several pieces with a double patina suggests a complex
management of artefacts, the re-use of artefacts left on the site and
the recurrence of occupations as demonstrated by the spatial
patterning of artefacts (Fig. 17).
The fragmentation of the reduction processes takes place in
parallel to the gathering of various int types in a local and semilocal perimeter (from the site up to a distance of 30 km). According to the rst geological analysis applied to a sample of more than
200 pieces, several types are still unknown, and may have been
^ne Valley coming from the Alps Mountains.
collected from the Rho
The main int types known attest to: (1) int procurement on the
plateau to the north of the site at a distance of 3e30 km (F14 and
F34, local to semi-local Barremian formations) from secondary int
outcrops and conglomerates, (2) secondary collecting to the south,
on a plateau located between 2 and 20 km from the site, involving
che River (F124, F165, Ludian and Lutecian forcrossing the Arde
mations). The diversity of types is due to frequent gathering from
conglomerates and indicates the indiscriminate use of all kinds of
existing int in a given outcrop. Neanderthals demonstrated procurement behaviour suggesting int gathering during other activities and a strategy of collector.
This int diversity is observed for Levallois akes, elongated
products and points and suggests a large mobility of products in
both a semi-local and local perimeter, whether they were introduced as large akes for knapping, for direct use or as raw material
stocks for future use. The sample of the largest products observed
suggests that the largest akes, blades and points come from the
south of the site. Flakes for knapping were brought in from both the
north and the south (Tables 8 and 9).

Table 8
Types of int and procurement perimeter, sub-level 4.2.
Location of outcrops

South of the site

North of the site

Perimeter of gathering

Semi-local

Local

Local to semi-local

4.2

F124
F124b
Ludian

F165
Lutecian

F14
Barremian

Fragments
Flakes
Backed akes
Kombewa akes
Levallois akes
Blades
Levallois blades
Levallois cores
Small akes
Points
Levallois points
Other cores
Total

2
1

2
3
1

Unknown

F34
Barremian

F122
F121

F190

3
5

F199b
Cenozoic
5
7

F199c

F200b to
F216
1
2

F210

1
2

1
1
2
1

1
29
1

13

26

39

25

1
1

23

1
2

22

15

18

12

1
2
31

28

40

17

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

193

Table 9
Types of int and procurement perimeter, sub-level 4.1.
4.1

South of the site

North of the site


Local

Local to semi-local

F124
F124a
F124b
Ludian

F165
F165b
Lutecian

F14
Barremian

F34
Barremian

F122

F199b
Jurassic

F210

11
2
3
1
2
1
5

3
1
1

Flakes
Backed akes
Levallois akes
fragments
Blades
Backed blades
Levallois blades
Micro-akes
Small akes (<10 mm)
Levallois points

10

6
1

1
3
1

2
1
1
2

Total

14

13

Table 10
Types of tools and use-material (from Hardy et al., 2013) in sub-level 4.1.
Thin and
invasive removals

Denticulates

1
2

Flint types from a local and semi-local perimeter cannot be


correlated with specic types of artefacts or with the specic
use of pieces (Hardy et al., 2013). Traces of plant, wood and
butchery activities predominate on all types of products,
whether they are retouched or not (Tables 10 and 11), regardless
of their origin and their technological status when they were
introduced to the site (nodules, cores, pre-knapped akes or
nished products). Bird and sh residues are present on elongated int products from the north of the site. Retted pieces
were used for butchery, woodworking and plant-related activities. Impact fractures were found on both points and elongated
akes, suggesting that shape was not a strong factor in selecting
of projectile tips (cf. Table 5).

Animal
Animal/plant
Fibrous plant
Fish
Hard/high silicate
Hide
Insect?
Plant
Wood

Unknown

Semi-local

Convergent
points

Scrapers

1
1
1

1
1
1

1
1
1

1
2

30

pebbles and were collected further up the plateau, along the


che or the Rho
^ne rivers.
Arde
The series are composed mainly of whole or broken round
pebbles (5e10 cm long). Percussion marks are not clearly visible
and therefore all of these pieces cannot be attributed to hammerstones. One small and at pebble appears to be a retoucher despite
the lack of percussion marks. One large pebble-tool (granite)
(Fig. 18) and a thick ake (basalt) with deep removals on one edge
are the only large tools on pebbles.
The series also includes some cortical akes, points, blades and
elongated akes, in spite of the poor quality of some of these stones,
especially quartz. They attest to knapping activities on these stones.
Some are Levallois (especially in limestone) and produced by
methods similar to those applied to int management. The lack of
cores suggests that most of the products were brought to the site, as
is the case for the largest int artefacts. However, some microakes and chunks suggest that these stones were used and
reworked at the site.
Of the 16 RMUs that have been identied, 1 is limestone, 1 is
basalt, 3 are quartzite and 11 are quartz. As is the case for int, all
of them illustrate knapping activities in and outside the site. The
smallest RMU is formed by a single object (1 quartz RMU is a
hammerstone, 3 quartzite RMUs are isolated akes or retouched
artefacts). The biggest RMU is in quartz and comprises 32 pieces.
The available data show that the reduction sequences of local
raw materials (especially quartz) took place onsite. Even when
pebbles were introduced as fragments, the reduction processes
took place in the shelter. The only RMU in basalt shows the fracture

Table 11
Types of tools and use-material (from Hardy et al., 2013) in sub-level 4.2.
Thin and invasive removals
Plant
Wood
Wood/sh?

Becs

Denticulates

Endscrapers

Convergent points

Peroirs

Scrapers

5
1

16
1

1
2

5. The management of other lithic materials


The int is associated with a small corpus of limestone (black,
silicied or marly), quartz, quartzite, granite, sandstone, schist and
basalt (Table 12). These raw materials are widely available as

of a hammerstone during use. After the breakage, some akes were


produced. The RMUs identied on limestone attest to the use of a
high quality stone. It is made up of 14 pieces related to a Levallois
reduction sequence which may have occurred onsite, but there is

194

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Table 12
Types of artefacts in stones other than int for level 4.1.
Basalt
Chopper
Chunks
Half-pebbles
Whole pebbles
Pebble fragments
Flakes
Backed akes
Flakes with a cortical back
Levallois akes
Points and triangular akes
First cortical akes
Fragments of akes
Blades
Backed blades
Levallois blades
Micro-akes (<5 mm)
Micro-chunks
Small akes (5e10 mm)
Total

Limestone

Silicied limestone

Granite

1
2
3
3

1
4
2
1

Sandstone

Quartz

1
23

1
1
1

6
5
26
3
1

Quartzite

Schist

1
3

2
2
2

1
13
7
1
1

1
1

1
2
1
4

1
12

no abandoned core. This could suggest that the core was transported as part of the tool kit after aking.
The results from the rets analysis corroborate the RMU results.
A total of 4 rets have been identied on quartz (n 3 for level 4.1
and n 1 for level 4.2) and one for basalt in level 4.2. They are
formed by 10 pieces and provide a ret rate of 0.1%. The technical
categories include pieces from all the phases of the chane
op
eratoire: cores (n 3), akes (n 2), ake fragments (n 2),
fragmented akes (n 1) and fragments (n 2). All the rets group
2 pieces. There are 5 connection lines, which belong to knapping
sequences for the pieces in quartz (n 4) and to a fracture for those
in basalt (n 1). As for int, the distances for the connection lines
vary from several centimetres to two meters (the longest one is in
quartz).
As explained above (Fig. 19), the spatial patterning of the rets in
quartz and basalt does not show any clear-cut distribution. However, the number of rets is very low for the moment and does not
enable us to identify specic knapping areas.

6. Discussion: Neanderthal occupations in a shelter in southeastern France


6.1. A witness of Neanderthal occupations
The Abri du Maras shelter provides a very well-preserved
archaeological dataset allowing us to interpret Neanderthal occupations during humid and cold phases during MIS 4. Ash lenses in
situ and rets attest to a lack of signicant disturbance to the site.
Thus, the site provides an excellent opportunity to see how Neanderthals occupied this area and managed their territory in a small
^ne River, in a position of shelter
valley along a tributary of the Rho
away from the thick and large glaciers located in the Alps. The lithic
corpus from both levels of layer 4 indicates analogous technical
behaviour, geared towards the Levallois production of akes,
elongated products and blades. The mobility of artefacts, especially
the longest ones (Levallois blades, akes, points, other types of
large akes and elongated products), suggests management of the
reduction processes in a local and semi-local perimeter, as indicated by the geological study of int. The site is located among a
well-known network of outcrops, at a crossroads of different
geological formations, which explains the diversity of stones used
at the site. The Neanderthal raw material collection behaviour
observed for other sites is conrmed at the Abri du Maras. They

10

97

Total
1
25
1
15
10
33
3
2
1
1
19
10
4
1
1
2
1
5
235

collected everything they found at a given outcrop. This is an


opportunistic strategy of raw material procurement from multiples
sources (collector behaviour).
Cores and large akes were brought to the Abri du Maras for
knapping, associated with large artefacts with long and/or pointed
cutting edges for direct or for future use. They were possibly used
without being retouched. RMUs and retting analysis conrm that
some products were imported to the site and others were exported
from the site. These data correspond well to patterns obtained in
other areas in Europe (Vaquero et al., 2012a; Turq et al., 2013) and
conrm the fragmentation of reduction processes, and the use and
the mobility of cores on akes during the Middle Palaeolithic. This
behaviour is not related to the quantity or quality of raw materials
or to the distance of the procurement zones. We observed the same
pattern for local and non-local stones. The Abri du Maras is located
in an area with abundant, easily available int outcrops regardless
of the environmental context, especially during cold climate with
open landscapes. Perhaps this opportunistic strategy can be
described as a exible strategy in which Neanderthals are able to
take advantage of the material that is available without the need to
seek out only certain types of high quality material. Another possibility is that something besides raw material is driving their
mobility strategies (animal or plant resources?).
This behaviour is related to multiple seasonal occupations. Many
ash lenses are dispersed throughout the site and located at various
depths (Fig. 19). Generally speaking, needs were anticipated. Part of
the materials from both the south and the north of the site were
already worked before being transported to the Abri du Maras.
Some points show impact fractures and may be related to projectiles. Some of them were brought to the site and left after use.
The lithics were used for a variety of activities performed on the site
or outside (projectile points related to hunting) on products made
outside or on the site. Cores were only exploited for brief sequences
and tools were rarely resharpened. The duration of the occupations
appears to be short. This organization of territory management at
the Abri du Maras can be associated with a circulating type,
considering the exploitation of local int (Binford, 1981).

6.2. The site in its regional technological framework


The assemblages from layer 4 at the Abri du Maras both
corroborate data from other sites in the area (Fig. 20) and bring new
detailed patterns due to its state of preservation. The regional sites

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

195

Fig. 14. Examples of lithic rets. A. Quartz rets, B, C. rets on int cores, D. rets of int akes.

can be differentiated by core technologies and the assemblages


generally point to variable technical behaviour by humans between
MIS 9 (including the top of the sequence of Orgnac 3) (Moncel et al.,
2011, 2012a) and the end of MIS 4/beginning of MIS 3.
The Levallois method was used in Ranc Pointu 2, Baume d'Oullins, Balazuc and Baume Flandin, dated from MIS 5 to the beginning
of MIS 3 (Moncel, 2003; Moncel and Daujeard, 2012). While most of
the Levallois cores are centripetal in the upper levels of Orgnac 3
(MIS 8), unipolar and bipolar methods are the main methods used

during MIS 4 to 3, even if there is some evidence of the centripetal


method (Baume Flandin) (Moncel, 1996, 2005). In each level at
Payre (MIS 8 to 6), a single small core indicates the debitage of small
elongated akes or bladelets around the core from a at platform
(Moncel et al. (Ed), 2008a; Moncel et al., 2008b).
The Abri du Maras layer 4 assemblages t in well with the
youngest Levallois corpus of the area. Among the Levallois assemblages, some present original characteristics, such as a laminar
debitage, which is either Levallois or based on the aking of the

196

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Fig. 15. Lithic rets for levels 4.1 and 4.2. Limits of the excavated area for each level indicated by black lines. The largest limestone blocks are in grey.

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

197

Fig. 16. Schema of the knapping strategies by raw materials.

entire periphery of the block without preparation. This is the case of


the upper levels at Abri du Maras and at Baume Flandin (Moncel,
2005). The two upper levels from previous excavations at the
Abri du Maras (levels 1 and 2, MIS 4 and 3?) yielded a laminar
debitage (5e10% of blades) mainly produced by a Levallois method
(uni-bipolar and convergent unipolar) on int cores (Combier,
1967; Moncel, 1996, 2003; Moncel and Michel, 2000). Blades are
more or less elongated and pointed. The Levallois cores in these
assemblages are not blade cores. Only some cores could be associated with laminar aking, with elongated removals made directly
on the cortical surfaces of small round pebbles (direct aking). The
Levallois core technology in our layer 4 differs somewhat from
these levels in that some of the products are not very elongated.
There are no laminar cores of any type at the site, whether Levallois
or not. The largest and longest blades were introduced to the site
and not produced in situ.
In Baume Flandin (MIS 5), laminar debitage was also associated
with Levallois debitage (centripetal and unipolar for ake debitage)
(Moncel, 2005). While the blades are thin in Abri du Maras, elongated products (10% of blades) are mostly thick in Baume Flandin
and some have a partial crest related to the debitage of angles of
cubic int fragments with no preliminary preparation. The usual
Upper Palaeolithic features have however not been observed.
On the other hand, discoid aking (unifacial or bifacial) was
mainly used at Payre (MIS 8/7 and MIS 6/5), Saint Marcel (MIS 5 to
^cheurs (MIS 4) (Moncel
3), Le Figuier (MIS 4) and the Abri des Pe
et al., 2004; Moncel et al. (Ed), 2008a; Moncel et al., 2008a,
2012b) while it is a minority in layer 4 at Abri du Maras. Many cores
present one or two aking surfaces with crossed and sometimes
orthogonal removals.
This diversity of technological choices bears no relation to the
age of the site, subsistence strategies or the climatic context.
Levallois and discoidal knapping were used throughout the Middle
Palaeolithic in this geographical area. Although the discoidal
method is considered to represent the aking of unspecialized
products, production variability is high in all the series. The unipolar Levallois method is more frequent in the numerous MIS 4 and
3 assemblages although it was already used in MIS 9/8 in Orgnac 3.

The frequency of elongated blanks (less elongated than blades) is a


trend in many assemblages in the area before MIS 5 (Moncel, 2001,
2005), produced both by Levallois, discoid, orthogonal and laminar
methods. However, the debitage of blades and points did not
emerge in the area before the beginning of MIS 5 (Moncel, 2001)
and was occasionally employed at the end of MIS 4 or the beginning
of MIS 3 under various climates whereas it existed as early as MIS 9
in southern Europe (Fontana et al., 2013). Different methodologies
were applied to Levallois cores and semi-tournant cores, such as
villion and Tuffreau, 1994). For the Levallois
in northern France (Re
method, the main ake tools are often thin scrapers with marginal
retouch. However most of the akes remain unretouched. Artefacts
with thin inverse retouch (lateral or proximal) or bifacial tools are
present in small quantities (possibly related to hafting?). In the
assemblages rich in blades, such as the Abri du Maras, retouched
products are mainly on akes (side-scrapers). Blades remain unretouched, while in the Baume Flandin the thicker blades are more
retouched.
The variability observed in the area corresponds well to most of
the observations in the southeast of France, for example recent
studies on the site of Loton (Bourguignon and Meignen, 2010) or
gue et al., 2010). On the other side of the
the Hortus cave (Lebe
^ ne, the upper deposits of La Combette (Vaucluse) are characRho
terized by Levallois, Quina and Kombewa core technologies and the
presence of exogenous rocks from distant deposits (tens of kilo^me) yielded
metres away) (Texier et al., 1998). Grotte Mandrin (Dro
MIS 3 lithic series which were aked onsite, producing mostly
elongated and pointed Levallois products, on mainly local int
(Giraud et al., 1998; Slimak, 2004).
The Levallois core technology differs from that observed in
southwest France, which generally involves centripetal methods
during MIS 4 (Delagnes, 1995; Delagnes et al., 2007). On the other
hand, we observe similar behaviour in the north of France with
Levallois point aking at the end of MIS 5 and beginning of MIS 4
(Goval, 2012). Moreover, as in layer 4 of the Abri du Maras, a
mobility of points between sites is observed for these northern
open-air occupations. The late development of point manufacture
in the South of France (MIS 5) while earlier in the North (from MIS

198

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Fig. 17. Spatial patterning of the int lithic assemblage by technical categories and levels. Limits of the excavated area for each level indicated by black lines. The largest limestone
blocks are in grey.

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

199

Collection of plants suitable for cordage production may have


also factored into decisions about seasonal movement
(Hurcombe, 2000; Hardy et al., 2013).
6.4. A technological behaviour related to short-term seasonal
occupation at the Abri du Maras?

Fig. 18. Large chopper on a granite pebble.

7; Rots, 2013) opens questions on the status of this tool in Neanderthal groups according to areas: hand tools and/or projectiles,
mobile tools (Goval, 2012)?
6.3. Land-use and mobility: raw material gathering and transport
of artefacts
Raw material provenance and diversity display a link between
the type and length of the occupation but does not vary over time
or according to technical choices. Flint was always collected in large
quantities from the geological formations around the sites. For brief
stopovers, the int reduction sequence is largely partial as at Abri
^cheurs (a specic case compared to the other sites of the area
des Pe
due to the predominant use of quartz; Fernandes et al., 2010;
Raynal et al., 2012), as in layer 4 at Abri du Maras. Some int
types are diverse and others not. In areas where int is rare, for
^cheurs in a brief stopover context, other
example at the Abri des Pe
stones were used on an occasional basis (e.g. quartz, good quality
limestone). In this case, the purpose of these stones is similar to
those in int. Some int artefacts are also brought for short or longterm seasonal occupations from long distance outcrops as akes or
retouched akes, for instance at Payre, or from a semi-local area as
nodules and akes (Fernandes et al., 2008, 2010). The quantity of
exogenous stones is not related to the complexity or type of aking
techniques, or to the duration of occupation. These artefacts
demonstrate mobility of a tool-kit between sites, artefacts not always being retouched. They could be considered as a reserve of
tools travelling from site to site for unspecialized activities (Hardy
and Moncel, 2011).
In the context of the discussion of transportation of int artefacts and raw material over long distances, it is important to
discuss the mode of transport. Although raw material sourcing
clearly indicates long distance movement of stones, archaeologists rarely discuss how this transport occurs. Recently, Hardy
et al. (2013) observed twisted bres on artefacts from Maras.
These bers were interpreted as evidence of the manufacture of
string or cordage. Although it is not possible to predict what
Neanderthals at Abri du Maras were doing with cordage, one
possibility is the construction of string bags for the transport of
objects (Hurcombe, 1994). With this in mind, it is important to
remember that use of resources across the landscape was not
limited to collecting of raw material and subsistence items.

In this south-eastern area, three types of occupation durations


have been established on the basis of faunal remains and subsistence strategies: long-term residential camps, shorter-term regular
hunting camps and brief stopping places. Each type of occupation
attests to a different strategy of animal resource acquisition
(spectrum, prey mortality, carcass transport and seasonality)
(Daujeard and Moncel, 2010; Daujeard et al., 2012). Considering the
relation the subsistence strategies and data coming from functional
lithic analyses, the different activities did not tend to play a role in
the choice of specic knapping techniques, the intensity of the
retouch of the products, or in the composition of the tool kits in the
area and their mobility. This variability is possibly more related to
topographical inuences than environmental inuences and
chronological attributions. Topography, biotope variety, accessibility, site size, water availability and exposure clearly inuence the
subsistence activities carried out at the different sites.
The last occupation (upper level, 1 of the old excavations) at the
Abri du Maras can be ascribed to the rst occupation category. On
the other hand, through the faunal remains, we observe evidence of
short-term occupations for layer 4 (Moncel et al., in press), dominated by reindeer hunting, in particular in sub-levels 4.1. Evidence
from the lithic series conrms this hypothesis (imported cores on
akes, short reduction sequences, pieces brought to the site in int
and other stones, abandon of points with impact fractures) (Fig. 21).
These characteristics differ from those of the bivouac sites of Abri
^cheurs (Moncel et al., 2008b) and Balazuc, where the whole
des Pe
corpus was imported during these bivouacs. The lower levels of
^cheurs have yielded sparse lithic assemblages on quartz
Abri des Pe
(Moncel and Lhomme, 2007; Moncel et al., 2008a). The artefacts are
whole pebbles in metamorphic stone and unretouched akes from
discoid debitage (unifacial and bifacial) on local quartz pebbles. The
few int artefacts introduced from various long distance zones to
^cheurs could have been produced either by Levallois
the Abri des Pe
or discoid aking.
6.5. Occupation type in a rock shelter?
A large variety of technical behaviour and management is
recorded in caves of this area but we have scant information concerning occupation types in rock shelters. The Saint-Marcel cave
contains Middle Palaeolithic occupations under a vast porch and
data indicate a common strategy over time revolving around cervid
hunting and discoid debitage with few ake-tools (Moncel et al.,
2004). Data from open-air sites is still too sporadic to appraise
differential behaviour according to the type of site, as in the
southwest of France where open-air butchery sites predominated
by herbivore hunting are observed (Costamagno et al., 2006). The
site of Saint-Bauzile, dated to MIS 5, attests to discoid-type core
technology, similar to that observed at the nearby site of Payre
(Moncel et al. (Ed), 2008a; Bernard-Guelle et al., 2011). Only the
int gathering tends to display the exploitation of less diverse
outcrops in open-air sites.
7. Conclusion
New excavations at the Abri du Maras provide evidence of occupations under a shelter during a cold context contemporaneous
with MIS 4. The technical strategies indicate the fragmentation of

200

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Fig. 19. Spatial patterning of the charcoal and other remains associated with hearths by level. Limits of the excavated area for each level indicated by black lines. The largest
limestone blocks are in grey.

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

201

Fig. 20. Positioning of the different archaeological Middle Palaeolithic records from southeast France on a NGRIP d18O (permil) curve (from Andersen et al., 2006; Rasmussen et al.,
2006; Vinther et al., 2006; Svensson et al., 2007; Wolff et al., 2010). Discontinuities of sedimentation and occupations over time are not taken into account in the schema.

202

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

Fig. 21. Schematic map indicating the types of mobile artefacts (largest int akes, blades, points) in a local and semi-local perimeter around the site. Large akes are devoted to
che River which is possible at a ford
direct use of aking. Large blades and points are devoted to direct use without retouches. Southern procurement assumes a crossing of the Arde
near the site, downstream of the gorges.

the reduction processes in a local and semi-local perimeter around


the cave. Large akes, blades and points were produced elsewhere,
north and south of the site, and brought to the occupations with
int and other stone akes, nodules and fragments of slabs for in
situ aking. The main core technology is Levallois, associated with
some other aking methods. Flake-tools are rare. The series attest
to lithic management and anticipation of needs in relation to
prevalent reindeer hunting (at least for level 4.1). Although longdistance transport of raw material has been cited by some as evidence of more advanced behaviour (Bar-Yosef, 2002; Mellars, 2005)
or the lack of modern cognitive way nding skills (Burke, 2012), we
see the lithic procurement at the Abri du Maras as evidence of a
exible strategy that shows forethought and planning by Neanderthals. Because int sources are abundant in the local region,
Neanderthal mobility was likely organized around the procurement of other important resources, including not only fauna but
plants for food or tools as well (Hardy et al., 2013).
These results differ somewhat from those of cave assemblages in
the area, and we have little data pertaining to occupation types
under shelters and at open-air sites. Future work in this area rich in
^ ne corridor will shed light on regional subsistence
int near the Rho
strategies and the role of the topographic aspect of the site in
resource acquisition and scheduling.
Acknowledgements
re de la Culture
Fieldwork was nanced by the French Ministe
through several scientic programs and by the administrative area

^ne-Alpes (South-East France). M.G. Chaco


n's research was
of Rho
funded by a postdoctoral grant from the Juan de la Cierva Subprogram (JCI-2010-07863, MINECO, Spain). The English manuscript
was edited by L. Byrne, an ofcial translator and native English
speaker. We would like to thank Huw Groucutt and the three
anonymous reviewers for their advice which help us to enrich the
paper.

References
Andersen, K.K., Svensson, A., Johnsen, S.J., Rasmusse, N.S.O., Bigler, M.,
thlisberger, R., Ruth, U., Siggaard-Andersen, M.-L., Steffensen, J.P., DahlRo
Jensen, D., Vinther, B.M., Clausen, H.B., 2006. The Greenland Ice Core Chronology 2005, 15e42 ka. Part 1: constructing the time scale. Quaternary Science
Reviews 25 (23e24), 3258e3267.
Bar-Yosef, O., 2002. The upper Paleolithic revolution. Annual Review of Anthropology 31, 363e393.
s, M., Bouchud, J., Gilles, R., Grenier, P., Nicolas, J., 1955. L'Abri du
Baudet, J.L., Barthe
che). Revue Arche
ologique 45 (1), 1e16.
Maras (Saint-Martin-d'Arde
, M., Fernandes, P., 2011. Le site mouste
rien d'Andance (SaintBernard-Guelle, S., Rue
che): un habitat de hauteur en contexte basaltique dans la moyBauzile, Arde
e du Rho
^ ne. Bulletin de la Socie
 te
 Pre
historique Franaise 108 (4),
enne valle
671e697.
Binford, L.R., 1981. Bones: Ancient Men and Modern Myths. Academic Press, NewYork.
 1982. Etude expe
da, E.,
rimentale de la technologie des pointes Levallois. In:
Boe
histoire et technologie lithique II,
Cahen (Ed.), Tailler ! Pour quoi faire: Pre
Recent Progress in Microwear Studies, Studia Praehistorica Belgica, vol. 2,
pp. 23e56. Leuven.
 1993. Le de
da, E.,
bitage Discode et le de
bitage Levallois re
current centripe
te.
Boe
 te
 Pre
historique Franaise tome 90 (6), 392e404.
Bulletin de la Socie
 1994. Le concept Levallois: variabilite

da, E.,
 des me
thodes. CNRS Editions,
Boe
Paris.

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204


olithique moyen : une me
moire technique
Bonilauri, S., 2010. Les outils du Pale
e? Approche technofonctionnelle appliquee a un assemblage lithique de
oublie
conception Levallois provenant du site d'Umm el Tlel (Syrie centrale) (Ph.D.
 de Paris ouest, Nanterre, Paris.
thesis). Universite
rien de Type Quina: Nouvelle de
nition d'une
Bourguignon, L., 1997. Le mouste
 technique (Ph.D. thesis). Universite
 Pars X, Nanterre, p. 672.
entite
ratoires:
Bourguignon, L., Faivre, J.-P., Turq, A., 2004. Ramication des chanes ope
cicite
 du mouste
rien. Paleo 16, 37e48.
une spe
s: nouvelles conBourguignon, L., Meignen, L., 2010. Ioton (Gard) 30 ans apre
rations technologiques et techno-e
conomiques sur l'industrie mouste
riside
te
 pre
historique Franaise 107, 433e451.
enne. Bulletin de la Socie
Burke, A., 2012. Spatial abilities, cognition and the pattern of Neanderthal and
modern human dispersals. Quaternary International 247, 230e235.
n, M.G., Bargallo
, A., Gabucio, M.J., Rivals, F., Vaquero, M., 2014. The NeanChaco
derthal behaviors from a spatio-temporal perspective: an interdisciplinary
approach to interpret the archaeological assemblages. In: Conard, N.J.,
Delagnes, A. (Eds.), Settlement Patterns Dynamics IV. Kerns Verlag, Tubingen (in
press).
Cziesla, E., 1990. On Retting of Stone Artefacts. In: Cziesla, E., Eickhoff, S., Arts, N.,
Winter, D. (Eds.), The Big Puzzle. Studies in Modern Archaeology, Bonn,
pp. 9e44.
olithique de l'Arde
che dans son cadre pale
oclimatique.
Combier, J., 1967. Le Pale
moire de l'Institut de Pre
histoire, 4. Delmas e
ditions, Bordeaux.
Me
Costamagno, S., Meignen, L., Beauval, C., Vandermeersch, B., Maureille, B., 2006. Les
Pradelles (Marillac-le-Franc, France): a mousterian reindeer hunting camp?
Journal of Anthropological Archaeology 25, 466e484.
Daujeard, C., Moncel, M.-H., 2010. On Neanderthal subsistence strategies and land
use: a regional focus on the Rhone Valley area in southeastern France. Journal of
Anthropological Archaeology 29, 369e391.
Daujeard, C., Fernandes, P., Guadelli, J.-L., Moncel, M.-H., Santagata, C., Raynal, J.-P.,
2012. Neanderthal subsistence strategies in Southeastern France between the
plains of the Rhone Valley and the mid-mountains of the Massif Central (MIS 7
to MIS 3). Quaternary International 252, 32e47.
 1988. Le Quaternaire du Bas-Vivarais d'apre
s l'e
tude des remplissages
Debard, E.,
dimentaire,
d'avens, de porches de grottes et d'abris sous roche. Dynamique se
oclimatologie et chronologie. Documents des Laboratoires de Ge
ologie, 103.
pale
partement des Sciences de la Terre, Universite
 Lyon 1, Lyon.
De
Delagnes, A., 1995. Variability within uniformity: three levels of variability within
Levallois system. In: Dibble, H.L., Bar-Yosef, O. (Eds.), The Denition and Interpretation of Levallois Technology, Monographs in World Archaeology. Prehistory Press, Madison, Wisconsin, pp. 201e211.
olithique
Delagnes, A., Jaubert, J., Meignen, L., 2007. Les technocomplexes du Pale
ographique.
moyen en Europe Occidentale dans leur cadre diachronique et ge
andertaliens. Biologie et
In: Vandermeersch, B., Maureille, B. (Eds.), Les Ne
historiques (CTHS), Paris, pp. 213e229.
cultures. Collection Documents Pre
raires et Transformations du Silex: Une Pe
troarche
ologie
Fernandes, P., 2012. Itine
e, Application au Pale
olithique Moyen (Ph.D. thesis). University of
Refonde
Bordeaux I, Bordeaux.
Fernandes, P., Raynal, J.-P., Moncel, M.-H., 2008. Middle Palaeolithic raw material
gathering territories and human mobility in the southern Massif Central,
France: rst results from a petro-archaeological study on int. Journal of
Archaeological Science 35, 2357e2370.
rales et comFernandes, P., Moncel, M.-H., Lhomme, G., 2010. Ressources mine
olithique moyen: Payre et l'Abri des Pe
^cheurs (Arde
che,
portements au Pale
France). Rivista di Scienze Preistoriche LVII, 31e42.
Fontana, F., Moncel, M.-H., Nenzioni, G., Onorevoli, G., Peretto, C., Combier, J., 2013.
Widespread diffusion of technical innovations around 300,000 years ago in
Europe as a reection of anthropological and social transformations? New
comparative data from the western Mediterranean sites of Orgnac (France) and
Cave dall'Olio (Italy). Journal of Anthropological Archaeology 32, 478e498.
rien en
Giraud, Y., Brugal, J.-P., Jeannet, M., 1998. Un nouveau gisement mouste
e du Rho
^ ne: la grotte Mandrin a
 Malataverne (Dro
^me). Bulletin
moyenne valle
te
 Pre
historique Franaise 95 (1), 7e15.
de la Socie
andertaliens dans le nord de la France. INRAP-CNRS
Goval, E., 2012. Peuplements ne
Editions, Paris.
Hardy, B.L., Moncel, M.-H., 2011. Neanderthal use of sh, mammals, birds, starchy
plants and wood 125e250,000 year ago. PLos One 6 (8), e23768.
arez, P., Desclaux, E.,
Hardy, B.L., Moncel, M.-H., Daujeard, C., Fernandes, P., Be
Chacon Navarro, M.G., Puaud, S., Gallotti, R., 2013. Impossible Neanderthals?
Making string, throwing projectiles and catching small game during Marine
Isotope Stage 4 (Abri du Maras, France). Quaternary Science Reviews 82, 23e40.
Hughes, S.S., 1998. Getting to the point: evolutionary change in prehistoric weaponry. Journal of Archaeological Method and Theory 5, 345e408.
Hurcombe, L.M., 1994. From functional interpretations to cultural choices in tool
use. In: Ashton, N., David, A. (Eds.), Stories in Stone. British Museum, London,
pp. 144e155. Lithic Studies Society, Occasional Paper 4.
Hurcombe, L., 2000. Time, skill and craft specialization as gender relations. In:
Donald, M., Hurcombe, L. (Eds.), Gender and Material Culture in Archaeological
Perspective. St. Martin's Press, New York, pp. 88e109.
La Porta, A.O., 2013. Points and Convergent Tools in the European Middle Palaee (Unpublished Master). In:
olithic Sites: the Shelters of Maras and Baume-Valle
The Regional Context of the South-East of Massif Central (France). National
Museum of Natural History, Paris, France.
gue, F., Boulbes, N., Gre
goire, S., Moigne, A.-M., 2010. Syste
mes d'occupation,
Lebe
 des Ne
andertaliens de l'Hortus (He
rault,
exploitation des resources et mobilite

203

France). In: Conard, N., Delagnes, A. (Eds.), Settlement Dynamics III. Kerns
Verlag, Tbingen, pp. 455e483.
Lenoir, M., Turq, A., 1995. Recurrent centripetal debitage (Levallois and Discoidal):
continuity or discontinuity? In: Dibble, H.L., Bar-Yosef, O. (Eds.), The Denition
and Interpretation of Levallois Technology Monographs in World Archaeology.
Prehistory Press, Madison, Wisconsin, pp. 249e256.
n, B., 2011. Contribucio
 n teo
 rico-metodolo
gica
Machado, J., Hern
andez, C.M., Galva
lisis histo
 rico de palimpsestos arqueolo
gicos a partir de la produccio
n
al ana
n para el Paleoltico medio en el yacimiento de El
ltica. Un ejemplo de aplicacio
Salt (Alcoy, Alicante), vol. 20. Recerques del Museu D'Alcoi, pp. 33e46.
ndez, C.M., Mallol, C., Galv
Machado, J., Herna
an, B., 2013. Lithic production, site
formation and Middle Palaeolithic palimpsest analysis: in search of human
occupation episodes at Abric del Pastor Stratigraphic Unit IV (Alicante, Spain).
Journal of Archaeological Science 40, 2254e2273.
Mellars, P., 2005. The impossible coincidence. A single-species model for the origins
of modern human behavior in Europe. Evolutionary Anthropology 12, 12e27.
rieurs de l'Abri du
Moncel, M.H., 1994. L'industrie lithique des trois niveaux supe
che). In: Re
villion, S. (Ed.), Les industries laminaires au
Maras (Arde
olithique moyen. CNRS, Paris, pp. 118e123. Dossiers de documentation
Pale

ologique 18. Editions
arche
du.
olithique moyen de l'Abri du Maras
Moncel, M.H., 1996. L'industrie lithique du Pale
che). Gallia Pre
histoire 38, 1e41.
(Arde
Moncel, M.H., 2001. Middle Palaeolithic Assemblages with blades in south-east
France. The question of the technical variability during Middle Palaeolithic
and its meaning. Archaeology, Ethnology and Anthropology of Eurasia 2 (6),
37e47.
 des groupes humains
Moncel, M.-H., 2003. L'exploitation de l'espace et la mobilite
 la n du Ple
istoce
ne moyen et au de
but
au travers des assemblages lithiques a
istoce
ne supe
rieur. La moyenne valle
e du Rho
^ne entre Dro
^me et Arde
che
du Ple
(France) British Archaeological Reports S1184, Oxford.
bitage
Moncel, M.-H., 2005. Baume Flandin et Abri du Maras : deux exemples de de
but du Ple
istoce
ne supe
rieur dans la Valle
e du Rho
^ ne (sud-est,
laminaire du de
France). L'Anthropologie 109, 451e480.
Moncel, M.-H., et al. (Eds.), 2008a. Payre. Des occupations humaines de la moyenne
e du Rho
^ ne de la n du Ple
istoce
ne moyen et du de
but du Ple
istoce
ne
valle
rieur, vol. XLVI. Me
moire de la Socie
te
 Pre
historique Franaise.
supe
re tentative de datation par UeTh du site
Moncel, M.H., Michel, V., 2000. Premie
olithique moyen de l'abri du Maras. Bulletin de la Socie
te
 pre
historique
pale
franaise 93 (3), 371e375.
Moncel, M.H., Lhomme, G., 2007. Les assemblages lithiques des niveaux du
olithique moyen de l'Abri des Pe
^cheurs (Arde
che, Sud-Est de la France). Des
Pale
andertaliennes re
currentes dans un fosse
 . L'Anthropologie
occupations ne
111 (3), 211e253.
Moncel, M.-H., Daujeard, C., 2012. The variability of the Middle Palaeolithic on the
^ne Valley (southeast France): technical traditions
right bank of the Middle Rho
or functional choices? Quaternary International 247, 103e124.
che): Une
Moncel, M.-H., Gaillard, C., Patou-Mathis, M., 1994. L'Abri du Maras (Arde
olithique moyen (1993).
nouvelle campagne de fouilles dans un site Pale
 Pre
historique Francaise tome 91, 363e368.
Bulletin Societe
gut-Bonnoure, E., Fernandez, P., Faure, M., Gue
rin, C.,
Moncel, M.-H., Daujeard, C., Cre
che, France) au
2004. L'occupation de la grotte de Saint-Marcel (Arde
olithique moyen: strate
gie d'exploitation de l'environnement et type
Pale
te

d'occupation de la grotte. L'exemple des couches i, j et j'. Bulletin Socie
historique Franaise tome 101 (2), 257e304.
Pre
Moncel, M.H., Brugal, J.P., Prucca, A., Lhomme, G., 2008b. Mixed Occupation during
^cheurs
the Middle Palaeolithic: case study of a small pit-cave site of Les Pe
che, south-eastern France). Journal of Anthropological Archaeology 27 (3),
(Arde
382e398.
 Boulbes, N., Puaud, S., Debard, E.,

gut-Bonnoure, E.,
Moncel, M.H., Daujeard, C., Cre
 Roger, T., Dubar, M., 2010. Nouvelles donne
, E.,
es
Bailon, S., Desclaux, E., Escude
but du Ple
istoce
ne supe
rieur de la moyenne
sur les occupations humaines du de
e du Rho
^ne (France). Les sites de lAbri des Pe
^cheurs, de la Baume
valle
che). Quaternaire
Flandin, de lAbri du Maras et de la Grotte du Figuier (Arde
21 (4), 389e415.
Moncel, M.-H., Moigne, A.-M., Sam, Y., Combier, J., 2011. The Emergence of Neanderthal Technical Behaviour: new evidence from Orgnac 3 (Level 1, MIS 8),
Southeastern France. Current Anthropology 52 (1), 37e75.
Moncel, M.-H., Moigne, A.-M., Combier, J., 2012a. Towards the Middle Paleolithic in
Western Europe: the case of Orgnac 3 (South-Eastern France). Journal of Human
Evolution 63, 653e666.
Moncel, M.H., Puaud, S., Daujeard, C., Lartigot-Campin, A.S., Millet, J.J.,
 Ge
gut-Bonnoure, E.,
ly, B., Vercoute
re, C., Desclaux, E.,
Theodoropoulou, A., Cre
che): bilan
Roger, T., Bourges, F., 2012b. La grotte du Figuier (Saint-Martin-d'Arde
cents sur un site du Pale
olithique moyen et supe
rieur de la
des travaux re
e du Rho
^ne (Sud-Est de la France). Bulletin de la Socie
te

moyenne valle
historique franaise 109 (1), 35e67.
pre
, E., Bailon, S., Barshay-Szmidt, C., Bearez, P., Cre
gut, E.,
Moncel, M.-H., Allue
Daujeard, C., Desclaux, E., Debard, E., Lartigot-Campin, A.-S., Puaud, S., Roger, T.,
2014. Evaluating the integrity of palaeoenvironmental and archaeological records in MIS 5 to 3 karst sequences from southeastern France. Quaternary International (in press).
Peresani, M., 2003. Discoid Lithic Technology. Advances and Implications. British
Archaeological Reports International Series S1120, Oxford.
Rasmussen, S.O., Andersen, K.K., Svensson, A.M., Steffensen, J.P., Vinther, B.M.,
Clausen, H.B., Siggaard-Andersen, M.-L., Johnsen, S.J., Larsen, L.B., Dahl-

204

M.-H. Moncel et al. / Quaternary International 350 (2014) 180e204

thlisberger, R., Fischer, H., Goto-Azuma, K.,


Jensen, D., Bigler, M., Ro
Hansson, M.E., Ruth, U.A., 2006. New Greenland ice core chronology for the last
glacial termination. Journal of Geophysical Research 111, D06102.
Raynal, J.-P., Moncel, M.-H., Daujeard, C., Fernandes, P., Fiore, I., Tagliacozzo, A.,
n, M.G., Theodoropoulou, A., 2012. Neanderthal landLecorre-Le Beux, M., Chaco
use and related tool-kits at the MIS 5/4 boundary in the South-East portion of
the French Massif Central. In: Ruebens, K., Romanowska, I., Byone, R. (Eds.),
Unravelling the Palaeolithic. Ten years of research at the Centre for the
Archaeology of Human Origins (CAHO, University of Southampton). BAR International Series, Southampton, pp. 53e72.
villion, S., Tuffreau, A., 1994. Les industries laminaires au Pale
olithique moyen.
Re
DAF, CNRS, Paris.
le
othe
mes de sites du
Richard, M., 2012. Datation de restes dentaires et de spe
olithique moyen d'Arde
che. Application de la re
sonance de spin
Pale
lectronique (ESR) et des se
ries de l'uranium (UeTh). Master in Prehistory.
e
um national d'Histoire naturelle, Paris.
Muse
Roebroeks, W., 1988. From Find Scatters to Early Hominid Behaviour. A Study of
de
re (The
Middle Palaeolithic Riverside Settlements at Maastricht-Belve
Netherlands). In: Analecta Praehistorica Leidensia 21. Leiden.
Rots, V., 2013. Insights into early Middle Palaeolithic tool use and hafting in
Western Europe. The functional analysis of level IIa of the early Middle Palaeolithic site of Biache-Saint-Vaast (France). Journal of Archaeological Science 40
(1), 497e506.
Shea, J.J., 2006. The origins of lithic projectile point technology: evidence from
Africa, the Levant, and Europe. Journal of Archaeological Science 33, 823e846.
Sisk, M.L., Shea, J.J., 2008. Intrasite spatial variation of the Omo Kibish Middle Stone
Age assemblages: artifact retting and distribution patterns. Journal of Human
Evolution 55, 486e500.
Sisk, M.L., Shea, J.J., 2011. The African origin of complex projectile technology: an
analysis using tip cross-sectional area and perimeter. International Journal of
Evolutionary Biology 2011, 1e8.
res expressions du Mouste
rien entre Loire et Rho
^ ne
Slimak, L., 2004. Les dernie
 de Provence-Centre d'Aix, Aix en Provence.
(Ph.D. thesis). Universite

Svensson, A., Andersen, K.K., Bigler, M., Clausen, H.B., Dahl-Jensen, D., Davies, S.M.,
thlisberger, R., Seierstad, I.,
Johnsen, S.J., Muscheler, R., Rasmussen, S.O., Ro
Steffensen, J.P., Vinther, B.M., 2007. A 60,000 year Greenland stratigraphic ice
core chronology. Cimate of the Past, 47e57.
Texier, J.-P., Brugal, J.-P., Lemorini, C., Wilson, L., 1998. Fonction d'un site du
olithique moyen en marge d'un territoire. L'Abri de La Combette (Bonnieux,
Pale

Vaucluse). In: Brugal, J.-P., Meignen, L., Patou-Mathis, M. (Eds.), Economie
historique: Les comportements de subsistence au Pale
olithique. XVIII
Pre
ologie et d'Histoire d'Antibes, Antibes.
Rencontre Internationales d'Arche

Editions
APDCA, Sophia Antipolis, pp. 325e348.
Turq, A., Roebroeks, W., Bourguignon, L., Faivre, J.-P., 2013. The fragmented character of Middle Palaeolithic stone tool technology. Journal of Human Evolution
65 (5), 641e655.
Vaquero, M., 2008. The history of stones: behavioural inferences and temporal
resolution of an archaeological assemblage from the Middle Palaeolithic. Journal of Archaeological Science 35 (12), 3178e3185.
n, M.G., Cuartero, F., Garca-Anto
n, M.D., Go
mez de Soler, B.,
Vaquero, M., Chaco
Martnez, K., 2012a. The lithic assemblage of level J. In: Carbonell, E. (Ed.), High
Resolution Archaeology and Neanderthal Behaviour: Time and Space in Level J
of Abric Roman (Capellades, Spain), Vertebrate Paleobiology and Paleoanthropology Book Series. Springer, Dordrecht, pp. 189e311.
n, M.G., Garca-Anto
n, M.D., Go
 mez de Soler, B., Martnez, K.,
Vaquero, M., Chaco
Cuartero, F., 2012b. Time and space in the formation of lithic assemblages: the
example of Abric Roman Level J. Quaternary International 247, 162e181.
Vinther, B.M., Clausen, H.B., Johnsen, S.J., Rasmussen, S.O., Andersen, K.K.,
Buchardt, S.L., Dahl-Jensen, D., Seierstad, I.K., Siggaard-Andersen, M.-L.,
Steffensen, J.P., Svensson, A.M., Olsen, J., Heinemeier, J., 2006. A synchronized
dating of three Greenland ice cores throughout the Holocene. Journal of
Geophysical Research 111, 2967e3013.
Wolff, E.W., Chappellaz, J., Blunier, T., Rasmussen, S.O., Svensson, A., 2010. Millennial-scale variability during the last glacial: the ice core record. Quaternary
Science Reviews 29, 2828e2838.

You might also like