ic
ro
polo
Keywords:
Archaeological demography
Upper Palaeolithic
Southwestern France
Hunter–gatherers
ncre
rec
sed
hunter–gatherers of Southwestern France as a case-study, this paper compares the evidence for changes
role o
a new
Cavalli-Sforza and Feldman, 1981), which maintains that in addi-
tion to a biological inheritance system, humans also possess a cul-
tural inheritance system which is subject to similar evolutionary
processes. Within this framework population dynamics (particu-
larly fluctuations in population size) are viewed as the most impor-
tant factor in understanding cultural change (Shennan, 2000: 821)
as ‘‘we cannot explain regional culture historical patterns without
and White, 2010;
ies draw o
minent in
uency of r
selection processes (Richerson et al., 2009: 211), affectin
rates of innovation and the maintenance of cultural traits (
lative culture’) (Derex et al., 2013; Ghirlanda and Enquist
Ghirlanda et al., 2010; Henrich, 2004, 2006; Kempe and Mesoudi,
2014; Kline and Boyd, 2010; Neiman, 1995; Riede and Bentley,
2008; Shennan, 2001; cf. Fitzhugh and Trusler, 2009; Read, 2006,
2012).
It has been proposed that the link between material culture and
demography espoused by dual inheritance theory is likely to be
more pronounced in the Palaeolithic, as the acquisition of theE-mail address: jcf35@cam.ac.uk
Journal of Anthropological Archaeology 39 (2015) 193–209
Contents lists availab
ol
w.Hill, 1993; Read and LeBlanc, 2003; Shennan, 2002, 2009).
Concomitant with this interest in the role of the individual in
demographic change is the study of material culture within a
framework of dual inheritance theory (Boyd and Richerson, 1985;
Hopkinson et al., 2013; Hosfield, 2005; Nowell
Premo, 2012; Premo and Kuhn, 2010). These stud
els and experiments which demonstrate the pro
of population size on social learning and the freqhttp://dx.doi.org/10.1016/j.jaa.2015.04.005
0278-4165/ 2015 The Authors. Published by Elsevier Inc.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).n mod-
fluence
andom
g both
‘cumu-
, 2007;trends documented in the archaeological record from the perspec-
tive of the decisions of individuals, designed to maximise their
reproductive success (Boone, 2002; Hammel and Howell, 1987;
2001) and the seeming lack of cultural diversity and innovation
across much of the global archaeological record of the Lower and
Middle Palaeolithic (Bocquet-Appel and Tuffreau, 2009;eral scholars have advocated studying the long-term populationthrough their investigation within human behavioural ecological
and evolutionary frameworks. Drawing on life history theory, sev-
record, most prominently the spatially and temporally
piece-meal appearance of ‘modern human behaviour’ (Culotta,
2010; Powell et al., 2009, 2010; Richerson et al., 2009; Shennan,1. Introduction
Archaeological theories about the
and cultural change have developedin relative population size as seen in three popular archaeological proxies for demographic change (site
counts, site sizes, and occupation intensity estimates). These proxies present conflicting results across the
sequence; a finding which is explored through the impact of taphonomic biases and past research agen-
das. Numbers of sheltered sites and quantities of retouched stone tools are suggested to be the most reli-
able demographic proxies. The problem of equifinality of interpretation in archaeological proxies for
demography is examined for the Aurignacian and Gravettian periods in the region, with changes in lithic
raw material, faunal acquisition strategies, and hunter–gatherer mobility all potentially contributing to
the patterns documented.
 2015 The Authors. Published by Elsevier Inc. This is an openaccess article under the CCBY license (http://
creativecommons.org/licenses/by/4.0/).
f demography in social
lease of life recently
first understanding regional demography’’ (Shennan and
Edinborough, 2007: 1344).
This link between population size and cultural change has been
used to explain several features of the Palaeolithic archaeologicalship relies on the ability to extract information about past population patterns from the archaeological
record. Using the extensive and well-studied record of the Upper Palaeolithic (39,500–11,500 cal BP)The demography of the Upper Palaeolith
Southwestern France: A multi-proxy app
Jennifer C. French
McDonald Institute for Archaeological Research, Department of Archaeology and Anthro
a r t i c l e i n f o
Article history:
Received 15 July 2014
Revision received 2 April 2015
a b s t r a c t
Demographic change is i
Palaeolithic archaeological
and material culture espou
Journal of Anthrop
journal homepage: wwhunter–gatherers of
ach using archaeological data
gy, University of Cambridge, Downing Street, Cambridge CB2 3ER, UK
asingly cited as an explanation for many of the patterns seen in the
ord, following the assumption of a relationship between population size
by dual inheritance theory. However, the empirical testing of this relation-
le at ScienceDirect
ogical Archaeology
elsevier .com/ locate/ jaa
techniques of artefact production would have focused heavily on
vertical transmission between parent and offspring (Shennan and
Steele, 1999). Variations in material culture would thus be strongly
related to the ebb and flow of the populations in question.
However, Collard et al. (2013a) have recently recommended cau-
tion in the use of population size as an explanation for patterns
in the Palaeolithic archaeological record, pointing out ambiguities
in previous studies of hunter–gatherer groups and the limited test-
ing of models against empirical archaeological data (see also
Collard et al., 2005, 2011, 2013b; Read, 2012). Palaeolithic archae-
ologists have clearly benefitted from the demographic data gener-
on population patterns from the archaeological record (Steele and
Shennan, 2009: 114).
emigration) (Hinde, 2002: 18). Changes in a population’s size, den-
sity, or growth rate are the result of variation in at least one of
these variables (Daugherty and Kammeyer, 1995: 11).
Unfortunately, data on these variables are unavailable for the
palaeodemographer. Palaeodemography is thus largely restricted
to the study of relative chronological and geographical changes in
population density, distribution and size, rather than the specific
demographic variations that caused these changes (although some
estimates of changes in fertility and mortality rates have been
inferred from osteological remains; e.g. Buikstra and Konigsberg,
1985; Buikstra et al., 1986; Greene et al., 1986; Konigsberg and
xam
K
B
1
(
f B
(
194 J.C. French / Journal of Anthropological Archaeology 39 (2015) 193–209To this end, this paper builds on earlier studies which examine
Palaeolithic demography using archaeological data (e.g. Ashton
and Hosfield, 2010; Ashton and Lewis, 2002; Bocquet-Appel and
Demars, 2000a, 2000b; Conard et al., 2012; Demars, 1996, 1998;
French and Collins, 2015; Gamble et al., 2004, 2005; Grayson and
Delpech, 2003; Grove, 2010; Hosfield, 1999, 2005; Meignen et al.,
2006; Mellars and French, 2011, 2013 (cf. Dogandzˇic´ and
McPherron, 2013); Morin, 2008; Petraglia et al., 2009; Schmidt
et al., 2012; Stiner, 2001, 2009; Stiner and Munro, 2002; Stiner
et al., 1999, 2000, 2008; Straus, 2011; Straus et al., 2000; see
French, 2015 for a review), focusing on the case-study of the hun-
ter–gatherers of the Upper Palaeolithic (39,500–11,500 cal BP) of
Southwestern France. A multi-proxy approach initially applied by
Mellars and French (2011) is used, in an attempt to overcome
the biases inherent with different types of data, and to; (1) explore
how the demographic patterns generated through each proxy
compare or converge, and; (2) assess which types of proxy data
can be obtained most reliably from the Palaeolithic archaeological
record.
2. Palaeodemography: how do we study demography from the
archaeological record?
Palaeodemography is the demography of past populations for
which no written source material is available. Three demographic
variables are the immediate causes of all population change; (1)
fertility (the process by which a population bears children); (2)
mortality (the process by which the members of a population are
reduced by death), and; (3) migration (both immigration and
Table 1
Summary of the three palaeodemographic methods employed in the study, and key e
Methodology Theoretical assumption
Site counts Number and distribution of sites reflects the relative
size and distribution of past populations
Site size Positive correlation between site area and number o
inhabitantsated from the modelling studies discussed above, as well as the
ever-increasing corpus of palaeo-genetic studies which estimate
past population sizes and trends (e.g. Atkinson et al., 2008;
Beaumont, 2004; Chikhi et al., 2010; Cox et al., 2009; Excoffier,
2002; Excoffier and Schneider, 1999; Fabre et al., 2009; Garrigan
et al., 2007; Harpending et al., 1998; Harris and Hey, 1999;
Scheinfeldt et al., 2010; Stajich and Hahn, 2005). Nonetheless,
the testing of the relationship between demography and cultural
change in the Palaeolithic is dependent on the availability of dataAccumulations research Positive correlation between the amount of cultural
material deposited at a site and the number of
inhabitants
A
G
aFrankenberg, 2005; cf. Bocquet-Appel and Masset, 1982;
Corruccini et al., 1989; Petersen, 1975, and these parameters can
be modelled; Bocquet-Appel and Degioanni, 2013; Sørensen,
2011; Surovell, 2000; Zubrow, 1989).
Three of the most common approaches to the study of demo-
graphic archaeology are summarised in Table 1 and form the basis
of this study. These are; (1) site counts; (2) settlement/site size
analysis, and; (3) accumulations research (see Varian and
Ortman, 2005). Within a palaeodemographic framework, the mag-
nitude of chronological variation in these proxies is indicative of
the magnitude of variation in population size or density
(Attenbrow, 2006: 13). The use of radiocarbon date summed prob-
ability distributions (‘dates as data’ (Rick, 1987)) is also a popular
palaeodemographic approach (e.g. Anderson et al., 2011; Armit
et al., 2013; Bocquet-Appel et al., 2005, 2009; Hinz et al., 2012;
Kelly et al., 2013; Martínez et al., 2013; Meeks and Anderson,
2012; Munoz et al., 2010; Shennan, 2009, 2013; Shennan and
Edinborough, 2007; Tallavaara and Seppä, 2011; Tallavaara et al.,
2010; Wicks and Mithen, 2014; Williams, 2012, 2013; Williams
et al., 2010) which has already been applied to the dataset under
discussion here (French and Collins, 2015).
Two main difficulties are common to these approaches
(French, 2015). Firstly, they rely on proxy data; the archaeological
material itself contains no direct demographic information
(Chapman, 1999). The methods listed in Table 1 provide the
required means to convert the archaeological proxy data into
statements about demographic change. However, all of these
approaches suffer from the problem of equifinality; multiple dif-
ferent explanations could be evoked validly to explain the pat-
terns seen in the data. For example, different rates of artefact
accumulation between sites could indicate changes in the num-
ber of individuals inhabiting them, but could also be related to
differing occupation lengths or alternative living arrangements
(e.g. Heizer, 1960: 93; Hiscock, 1986; Ross, 1985: 82–83). For
hunter–gatherers, such as those discussed here, the most fre-
quently cited alternative explanation for patterns in the data is
that of a change in mobility or land-use strategy (e.g.
Attenbrow, 2006; Schmidt et al., 2012).
The other difficulty is what is termed the ‘‘contemporaneity
problem’’ (Schacht, 1984). This refers to the practice of classifying
remains of the same period as contemporary when it is unlikely
ples of their application to the Palaeolithic archaeological record.
ey Palaeolithic examples
ocquet-Appel and Demars (2000b), Bocquet-Appel et al. (2005), Demars (1996,
998), Lahr and Foley (2003), Mellars and French (2011, 2013), Schmidt et al.
2012), Straus (2011), Straus et al. (2000), van Andel et al. (2003)
urke (2006), Grove (2010), Hayden (2012), Mellars (1973), Mellars and French
2011, 2013), White (1985)
shton and Hosfield (2010), Ashton and Lewis (2002), Conard et al. (2012),
rayson and Delpech (2003), Hosfield (1999, 2005), Meignen et al. (2006), Mellars
nd French (2011, 2013)
that they are strictly contemporary. For example, a sample of sites
in the landscape may date to the same period, but poor chronolog-
ical resolution and data palimpsests make it difficult to assess
whether they were occupied simultaneously or sequentially. This
distorts population estimates by combining centuries or millennia
into a single phase (Freter, 1997), ignoring change within periods,
and forcing a reliance on the equation of ‘populations’ with certain
classes of material culture (for attempts to overcome the contem-
poraneity problem see Ammerman et al., 1976; Grove, 2012; Hill,
1970; Schacht, 1981).
While archaeologists are increasingly citing demographic
change as an explanation for patterns in the Palaeolithic archaeo-
logical record, the dual problems of chronological control and equi-
finality of interpretations are more pertinent than for later
prehistoric periods. The range of cultural material available as
proxy data is more limited, and difficulties of data resolution and
palimpsests of occupation are more pronounced (Conard, 2001),
particularly as the applicability of 14C dating is restricted to only
the later stage (post 40 kya) of the period. Doubts have also been
raised about the suitability of palaeodemographic methods to the
study of archaeological records generated by archaic hominins
(e.g. Hassan, 1981: 84), although the focus of this study on Upper
Palaeolithic Homo sapiens populations bypasses this problem in
the current instance. Despite these reservations, results from ear-
lier studies (see Table 1) suggest that examining relative demo-
graphic trends in the Palaeolithic (particular the Upper
Palaeolithic), is within the limits of archaeological data, especially
when a multi-proxy approach is taken and the results are situated
within the wider framework of the demography of ethnographi-
cally documented hunter–gatherers.
3. Background: The Upper Palaeolithic geography and history of
Southwestern France
The Upper Palaeolithic of Southwestern France is an ideal case
study for exploring and testing a range of demographic approaches
to the Palaeolithic archaeological record. The archaeological record
of the region is exceptional, coupling continuous Pleistocene occu-
pation with a long history of research extending back to the
mid-19th century (see Sackett, 1981), the broadly uniform inten-
sity of which, both in chronological and geographical terms, goes
some way to eliminate the problems of sampling and taphonomic
bias which often hamper demographic studies. Despite this, previ-
ous palaeodemographic efforts are limited, and for the most part
superficial or based on only one type of archaeological data (e.g.
de Sonneville-Bordes, 1960, 1973; David, 1973; Mellars, 1985;
Smith, 1966 cf. Grayson and Delpech, 2003; Langlais et al., 2012;
Mellars and French, 2011).
The specific area of Southwestern France chosen for this study
centres on the modern administrative département of Dordogne
and incorporates the six surrounding départements of
Charente-Maritime, Charente, Corrèze, Lot, Lot-et-Garonne and
Gironde (Fig. 1). The area spans approximately 1.5 degrees in
latitude, from 44300N in Lot-et-Garonne, to 45700N in
Charente-Maritime, and covers 50,000 km2, although the area
available for occupation was undoubtedly larger in the late
Pleistocene and has since been reduced due to rises in global
sea-levels and altering coastlines (Lambeck et al., 2002).
The study area consists of a roughly triangular-shaped sedimen-
tary basin, lying between the Massif Central to the East, the
Pyrénees to the South and the Atlantic Ocean to the West. A wide
J.C. French / Journal of Anthropological Archaeology 39 (2015) 193–209 195Fig. 1. Map of France showing the location of the study region.
through Charente-Maritime, Gironde and Lot-et-Garonne as it
flows North from the Pyrénees before heading North-West, and;
be based on the results of multiple proxies, and that for each of these
proxies the study of multiple sites will average out or minimise
much inter-site variation (whether behavioural, taphonomic or
introduced by differing excavation/analytical techniques) to provide
an reasonable assessment of the wider trend.
(1) Numbers of archaeological sites
Data on the number of archaeological sites belonging to each
stage of the Upper Palaeolithic were collected following the
assumption that variations in the number of sites reflect fluctua-
tions in the relative size and distribution of past populations. A
‘site’ was defined as any location where at least one lithic artefact
chronologically diagnostic of any of the 5 periods of the Upper
Palaeolithic was present. Each period was divided further into
the sub-stages presented in Table 2. Sites from the study area
which have been radiometrically dated to the Upper Palaeolithic
but contain no artefactual evidence for human occupation were
not included. Painted/decorated caves were only included where
chronologically diagnostic artefactual material was also present.
The definition of a ‘site’ used explicitly does not assume that all
of those included in the database were permanent or long-term
habitation sites, nor that all of the sites documented for each per-
ical(2) The Dordogne, which runs through Corrèze, Lot, Dordogne,
and Gironde before joining the Gironde estuary.
The dominant geological features of the region are caves
(grottes) and rock-shelters (abris) from which the majority of
Upper Palaeolithic finds have been recovered. These occur
throughout the study region but are most common in the
Cretaceous limestone dominated landscapes of the Dordogne,
specifically the Dordogne and Vézère valleys in the southeast
(Laville et al., 1980: 4; Tixier, 2009: 12), creating a landscape of
sheltered river valleys and contrasting limestone plateaux, the
topography of which is suggested to have changed little since the
late Pleistocene (White, 1985: 51). The abundance of naturally
occurring shelters is considered to be one of the main factors that
favoured human occupation of the region throughout the span of
the Upper Palaeolithic (Mellars, 1985, 1996: 50).
The Upper Palaeolithic occupation of the region by hunter–
gatherer populations occurred in the late Pleistocene from
39,500 to 11,500 cal BP. Terrestrial and marine climatic proxy
records from the region including speleothem sequences (Genty
et al., 2003, 2010; Wainer et al., 2009), sediment profiles
(Bertran, 2005; Bertran et al., 2008, 2013), lacustrine pollen
sequences (Ampel et al., 2008, 2010; Wohlfarth et al., 2008), and
marine cores from the European Atlantic ocean margin of both
France and nearby Northern Iberia (Daniau et al., 2009; Genty
et al., 2010; Naughton et al., 2007, 2009; Sánchez-Goñi et al.,
2008) indicate that late Pleistocene global climatic changes
(Dansgaard–Oeschger cycles; Dansgaard et al., 1993; Grootes
et al., 1993; Svensson et al., 2008, and Heinrich events; Andrews,
1998; Hemming, 2004) had local climatic and environmental
effects in Southwestern France, although these may have been sub-
ject to time-lags (Blaauw et al., 2010).
Archaeologically, the Upper Palaeolithic of Southwestern France
is divided into five broad successive periods; Aurignacian
(39,500–34,000 cal BP), Gravettian (34,000–26,100 cal BP),
Solutrean (26,100–24,600 cal BP), Magdalenian (24,600–
15,500 cal BP) and Azilian (15,500–11,500 cal BP). These five
periods show a clear chronological and stratigraphic succession,
originally identified through the presence of diagnostic lithic
‘type-fossils’ (fossils directeurs), and later by radiometric dates.
These phases are divided further into sub-phases, although ambi-
guities are present throughout the chronological sequence as a
result of conflicts between absolute dates and stratigraphy, and
questions surrounding the chronologically diagnostic nature of
some type-fossils (e.g. Bon, 2002; Ducasse, 2012). In view of these
ambiguities simplified sub-divisions of each of the five periods
were adopted for this study. This simplified archaeological
sequence is presented in Table 2 (see Supplementary Material for
detailed discussion on both the absolute dating of these phases
and their sub-divisions).
4. Methods
The data for the study were collected through a comprehensive
literature review, compiled into a database which includes all the
recorded sites in the study region which can be dated either abso-range of landscapes, with associated underlying geology, are pre-
sent, including the Aquitaine plains, the plateaux of the Massif
Central, terraced alluvial valleys and the coastal region of the
Poitou in the North. Two great rivers flow through the region, both
of which have numerous tributaries; (1) The Garonne, which runs
196 J.C. French / Journal of Anthropologlutely or relatively (typologically) to any of the five main periods of
the Upper Palaeolithic of Southwestern France. A total of 542 sites
(865 occupations as many of the sites were occupied in multipleperiods) were included in the database (Table S3, Supplementary
Material). Specifically, data on the three different parameters listed
in Table 1 were collected and analysed as described below. These
proxies were studied with the aim of examining chronological vari-
ations between and, where possible, within periods, although the
lack of data restricted the comparative analysis of these
sub-stages to the analysis of numbers of archaeological sites. Due
to the varying sample sizes, and the resultant wide range of values,
medians were preferred over means as a measure of average
values. The availability of the required data for all three of the prox-
ies was heavily variable and dependent on such factors as location
within the study region, type of site (whether sheltered or
open-air), period(s) of occupation within the Upper Palaeolithic
sequence, and date of site excavation. As such, not all of the methods
described below were applied to all of the sites included in the data-
base. Nonetheless, the study adheres to the idea of ‘safety in
numbers’ within palaeodemographic analysis; that interpretations
Table 2
The Upper Palaeolithic succession in Southwestern France, showing start dates and
length of each phase for the simplified archaeological sequence used in this analysis.
See Supplementary Material for detailed discussion of how these phases were
determined.
Period Start date (kya cal BP,
IntCal 13)
Length of
phase (kyr)
Aurignacian 5.5
Early Aurignacian 39.5 3.5
Late Aurignacian 36.0 2.0
Gravettian 7.9
Early Gravettian 34.0 2.5
Middle Gravettian 31.5 2.0
Late Gravettian 29.5 3.4
Solutrean 1.5
Early/Middle Solutrean 26.1 0.6
Late Solutrean 25.5 0.9
Magdalenian 9.1
Badegoulian 24.6 2.8
Middle Magdalenian 21.8 3.6
Upper Magdalenian 18.2 1.2
Final Magdalenian 17.0 1.5
Azilian 15.5 4.0
Archaeology 39 (2015) 193–209iod were occupied simultaneously.
To account for the different lengths of each period and
sub-stage, site counts were standardised by converting these into
gicalestimates of the number of sites/1000 years, assuming the approx-
imate lengths given in Table 2 and rounded to the nearest whole
number. The figures generated represent averages over the
time-scale of the period in question, and do not imply that the
numbers of archaeological sites remained constant within periods.
To compensate for taphonomic bias relative to time-depth, and to
prevent their artificial over-representation (relative to the number
that existed in the past) later in the Upper Palaeolithic sequence,
the taphonomic correction curve of Surovell et al. (2009) was
applied to the open-air site counts, taking the chronological
mid-point of each period as the correction factor. While sheltered
sites are also likely subject to some taphonomic loss relative to
time-depth, no similar curve currently exists to accommodate for
this, and these deposits are generally better protected than those
at open-air sites (see discussion in Surovell et al., 2009).
(2) Site size
Data on the size (m2) of Upper Palaeolithic sites were collected
from the literature, following the assumption of a positive
(although not necessarily linear) correlation between site area
and the number of inhabitants. One difficulty is identifying how
many occupation episodes are represented by the archaeological
remains at a site, and the extent to which successive visits caused
the zones of occupation to shift laterally; large sites could be the
result of the single occupation of a large group, or successive
(re)occupations by smaller groups over an extended time period,
although small sites could only feasibly have been occupied by
small groups. While there is some ambiguity in correlating group
size and site size as far as larger sites are concerned, it is reasonable
to assume that small sites reflect occupation by a small group,
especially in instances where the site size is either spatially con-
strained or remains small despite the potential for expansion.
Data on site sizes were gathered from 3 main sources; (1)
White’s (1985) study of the Upper Palaeolithic sites of the
Périgord region; (2) written descriptions in site reports, and; (3)
excavation plans and sections. Very few authors offered a clear def-
inition of ‘site size’ in their reports which reduces the comparabil-
ity of the data. ‘Site size’ in this study refers to the overall extent as
delineated by cultural occupation remains. For cave and
rock-shelter sites this equates to the sheltered area, plus the asso-
ciated terrace (sensu White, 1985). Due to the lack of spatial con-
straints, the absence of clear sections from which to estimate site
size, and the difficulty of assessing whether outlying occupation
remains are associated with the site proper, or represent an addi-
tional find-spot on the landscape, the estimates are invariably
much cruder for open-air sites.
Site size data were used to calculate the number of small
(<500 m2) and large (P500 m2) sites present for each period. One
fundamental (but largely insurmountable) caveat of this is the
multi-period use of the majority of these sites, and the inability
to assess from the literature when the size quoted was attained.
However, this concern is limited to sites which lack clear spatial
restrictions and the use of only two broad size categories min-
imises the impact of this problem on the overall patterns
generated.
Where the data permit, specific areas of documented occupa-
tion for a period at a site were also collected in order to study
specific chronological fluctuations in the area occupied (Table S4,
Supplementary Material). For multi-period sites, the area over
which material diagnostic of each period was documented through
excavation was taken. For single-period sites, the total area of the
site was taken, following the assumption that all cultural material
belongs to the same period. Many of the sites in Southwestern
J.C. French / Journal of AnthropoloFrance have not been excavated or surveyed in their entirety and
these estimates remain therefore minimum estimates of both theoverall site area and the extent of occupation in each period as
known site size is clearly affected by the extent of excavation
(see, for example Roebroeks’ et al. (2011) recent work at the site
of Laugerie-Haute Est (Dordogne)). Sites where only small
sondage/trial excavations have occurred (defined here as <10 m2)
were excluded from the analysis.
(3) Occupation intensity (accumulations research)
Occupation intensity was calculated following the assumption
of a positive correlation between the amount of cultural material
at a site and the number of inhabitants. Four further assumptions
underpin the use of this proxy; (1) that the density of occupation
residue calculated is representative of that over the whole site;
(2) that differential selection and excavation strategies have not
distorted the estimates; (3) that the different potential functions
of the sites analysed (e.g. within part of a seasonal/annual round
of settlement) have only a minimal impact on the values produced,
once averages are taken across periods, and; (4) that occupation
intensity values are expressed in strictly relative terms and explic-
itly do not assume continuous human occupation of any of the
sites over the time-ranges in question.
Two different forms of occupation residue were analysed: (1)
quantities of retouched stone tools, and; (2) estimates of ungulate
meat weights (kg) (Tables S5 and S6, Supplementary Material).
(a) Retouched tool counts
Quantities of lithic artefacts were chosen as they are the most
common type of material culture found in the Upper
Palaeolithic archaeological record. The total quantity of lithic
material was rejected as a viable proxy as the collection and
curation of non-diagnostic flakes and perceived lithic ‘waste’
products was limited throughout a large part of the 20th cen-
tury, during which a significant number of the sites in the study
region were excavated (Clark and Lindly, 1991: 578; Rigaud and
Simek, 1987: 49; Sackett, 1981). To enhance comparability
across sites and assemblages, the analysis was restricted to
counts of retouched tools classified by the de
Sonneville-Bordes and Perrot (1953) ‘type-list’ of 97 formal
Upper Palaeolithic retouched tool types.
(b) Ungulate meat weight
Ungulate meat weight was chosen as a viable palaeodemo-
graphic proxy, following the assumption that faunal assem-
blages represent the food debris of the main source of dietary
protein for hunter–gatherers in the European Upper
Palaeolithic (Drucker and Henry-Gambier, 2005; Richards,
2009) and can provide an indication of the potential quantity
of food, and therefore number of consumers, at a site.
Counts of the Minimum Number of Individuals (MNI) repre-
sented by an assemblage were collected from the literature for
the 10 most prevalent ungulate taxa in the region (Table S6,
Supplementary Material). Only ungulate assemblages which
were deemed to have been accumulated by human groups were
included in the analysis (see Cruz-Uribe, 1991; Kuhn et al., 2010;
Pickering, 2002). The remains of carnivores and small vertebrates
were excluded, as they are likely to have accumulated largely
through non-human activity and to have made only a minimal
contribution to the diet of Upper Palaeolithic groups.
To calculate the total meat weight represented by an assem-
blage the MNI (per taxon) was multiplied by the amount of
meat one individual of the taxon would provide (see White,
1953), using the estimates given in Table S7 (Supplementary
Material). Estimates of available meat (defined as all parts of
the animal exclusive of bone and hide (Lyman, 1979: 536))
Archaeology 39 (2015) 193–209 197were used over estimates of consumable meat as the latter is
likely to be culturally defined (Reitz and Wing, 2008: 233).
In order to maximise the available database, where published
MNI counts were unavailable, estimates were used, extrapo-
periods, suggesting that the short duration of the phases in ques-
tion are driving this spike in the demographic signature (with
the exception of the longer Middle Gravettian and Badegoulian).
Dips in the number of archaeological sites are seen in the Early
Gravettian, the Middle Magdalenian and Azilian, with the lowest
number seen in the Final Gravettian (Fig. 2).
The impact of taphonomic loss with time-depth was assessed
for open-air sites by applying the correction curve of Surovell
et al. (2009). For each period the total number of sheltered sites,
the total number of open-air sites, and the corrected value for
open-air sites were scaled between 0 and 1 by dividing each value
by the maximum observed value in that category. The relative tem-
poral frequencies are displayed in Fig. 3. While similarities are seen
in places (most noticeably the decreases seen in both site types
33,000 cal BP), the distributions of open-air and sheltered sites
differ throughout the Upper Palaeolithic, even when open-air sites
are corrected for taphonomic bias. Mostly noticeably, a peak in the
frequency of open-air sites 23,000 cal BP (Badegoulian) contrasts
with a clear dip in the frequency of sheltered sites, and in the latter
part of the sequence (20,000–16,000 cal BP), the frequency of
open-air sites decreases, while that of sheltered sites increases.
These differences pose problems for the interpretation of relative
demographic trends from these data. One possible explanation
5.2. Site size
ical Archaeology 39 (2015) 193–209lated from published NISP (Number of Identified Specimens
Present) counts, following the method outlined in Mellars and
French (2011). Briefly, this involved a systematic comparison
between the reported NISP and MNI values for the 10 taxa stud-
ied from 12 sites included in the analysis for which quantitative
data on both values were given in the published faunal reports
(Table S8, Supplementary Material). A NISP:MNI ratio was
calculated for each (achieved by dividing the NISP value by
the MNI value) and the median value (rounded to the nearest
whole number) of the calculated ratios was taken to provide
an appropriate ‘conversion’ factor for each species. Meat weight
estimates based on these extrapolated MNI counts followed the
procedure given above.
All sites in the study region were included for which explicit
numerical counts of each/either of the two forms of occupation resi-
dues were available, along with a clear estimate of the area (m2)
from which these residues derived and reasonable confidence in
the integrity of the assemblage. Where possible, the results of the
most recent analysis of the site/assemblage were used, following
the assumption that these would adhere to stricter excavation and
analytical criteria. Clear cases of reworked sequences and mixed
assemblages were discarded from the analysis, although strati-
graphic integrity has only been studied for a handful of (mostly
Aurignacian) sites in the region (e.g. Le Piage, Roc-de-Combe
(Bordes, 2002, 2003), Le Flageolet, La Ferrassie (Michel, 2010)). As
with the site counts, to compensate for the differences in the dura-
tion of each period, densities were standardised for each site in
terms of the quantity of occupation residue recorded per square
metre of the excavated occupation levels, per 1000 years of the doc-
umented occupation sequence. Calculations based on areas of occu-
pationwere preferred to those based on volumeof sediments, due to
the influence of such factors as localised rock falls and other geolog-
ical and climatic factors on overall sedimentation rates during
occupation sequences. For each site, the overall duration of occupa-
tion in each period was based on the assumption that the sites were
potentially open and available for occupation during the whole of
the known time-span (see Dogandzˇic´ and McPherron, 2013: 312
for a critique) and the occupation density/m2 estimates were
divided by the figures given in Table 2. Where there is clear
evidence that only a certain sub-stage of a period is present at a site,
occupation density was only calculated for that stage. Where more
than one occupation level belonging to a sub-stage is present at a
site, the mean area of the levels in question was used in the
calculation.
5. Results
5.1. Site counts
Numbers of archaeological sites vary greatly across the
techno-complexes of the Upper Palaeolithic. Extreme peaks in
numbers of archaeological sites/1000 years are seen in theWhere data on age and sex accompany the published MNI
count the appropriate adjusted estimates given in Table S7
were used. Where these data were unavailable or difficult to
correlate with the published MNI counts, the mean adult
weight was used. Given the rarity of this information in the lit-
erature, the estimates given in this study should be considered
the maximum potential meat weight (food).
198 J.C. French / Journal of AnthropologSolutrean and Upper/Final Magdalenian, with smaller spikes in
the Middle Gravettian and Badegoulian. The gross numbers of sites
for these periods are not exceptionally higher than those for otherThe quantity of large (P500 m2) and small (<500 m2) sites show
remarkable consistency in their relative frequency throughout the
span of the Upper Palaeolithic. Small sites constitute between
68.2% (Azilian) and 77.1% (Gravettian) of the known sample for
any given period, and large sites between 22.9% (Gravettian) and
31.8% (Azilian) (Table 3).
EA LA EG MG LG E/M S LS B MM UM FM AZ
Open-air 7 5 3 4 2 2 8 20 4 8 7 7
Cave/rock-shelter 23 26 18 31 4 60 83 13 14 60 52 20
0
10
20
30
40
50
60
70
80
90
100
N
um
be
r o
f s
ite
s/
10
00
 y
ea
rs
Open-air
Cave/rock-shelter
Fig. 2. Fluctuations in the number of sites/1000 years across the main periods of
the Upper Palaeolithic in Southwestern France. EA = Early Aurignacian; LA = Late
Aurignacian; EG = Early Gravettian; MG = Middle Gravettian; LG = Late Gravettian;for this difference is that the study region is outside of the geo-
graphical region of applicability of Surovell et al.’s (2009) correc-
tion curve. Alternatively, periods of clear divergence between the
frequency distributions of the sheltered and open-air sites may
reflect the deliberate and/or preferential selection by the hunter–
gatherers of Upper Palaeolithic Southwestern France of either type
of site.E/M S = Early/Middle Solutrean; LS = Late Solutrean; B = Badegoulian; MM = Middle
Magdalenian; UM = Upper Magdalenian; FM = Final Magdalenian; AZ = Azilian.
Reproduced from French and Collins (2015).
gicalJ.C. French / Journal of AnthropoloData on documented areas of occupation were available for 139
occupations. Magdalenian sites show the largest documented area
of occupation (n = 57, Md = 120.0 m2) and Azilian the smallest
(n = 12, Md = 35.5 m2), although a Kruskal–Wallis test showed no
statistically significant differences between documented areas of
occupation based on techno-complex (n = 139, v2(4) = 5.186,
p = 0.269). Analysis of only the cave and rock-shelter sites (for
which the estimates are likely a truer reflection of past areas of
occupation due to their spatially restrictive nature) showed a sim-
ilar pattern with the largest documented areas of occupation seen
in the Magdalenian (n = 38, Md = 112.0 m2) and the smallest the
Azilian (n = 9, Md = 35.0 m2). While the differences between the
documented area of occupation was shown to be statistically
significantly across the techno-complexes (Kruskal–Wallis test
n = 104, v2(4) = 11.386, p = 0.023), pairwise comparisons
(corrected using the Bonferroni correction) revealed that this
in quantities of retouched tools between chronological periods
Fig. 3. Temporal frequency distribution of numbers of archaeological sites across
the Upper Palaeolithic of Southwestern France, showing the different distributions
of; (a) sheltered (cave/rock-shelter) sites; (b) open-air sites, and; (c) open-air sites
once corrected for taphonomic bias using the curve of Surovell et al. (2009). All
values have been standardised between zero and one, and show strictly relative
difference in the frequency of sites. Reproduced from French and Collins (2015).
Table 3
Site size data for the Upper Palaeolithic sequence in Southwestern France, docu-
menting number and frequencies of large (P500 m2) and small (<500 m2) sites, and
the median documented areas of occupation for cave and rock-shelter sites. See
Table S4, Supplementary Material for data on areas of occupation.
Period n Small sites Large sites
Number and frequency
of large and small sites
Aurignacian 100 76 (76%) 24 (24%)
Gravettian 83 64 (77.1%) 19 (22.9%)
Solutrean 74 54 (73%) 20 (27%)
Magdalenian 135 98 (72.6%) 37 (27.4%)
Azilian 44 30 (68.2%) 14 (31.8%)
Median documented
area of occupation (m2)
Documented areas
of occupation
Aurignacian 26 87.0
Gravettian 19 64.0
Solutrean 12 71.3
Magdalenian 38 112.0
Azilian 9 35.0(n = 74, v2(4) = 3.928, p = 0.416).
When only the more reliable data from the cave/rock-shelter
sites are considered, a similar pattern is found, with the greatest
densities of retouched tools still found in the Solutrean (n = 5,
Md = 40.0 tools/m2/1000 years) and the lowest in the Azilian
(n = 10, Md = 5.8 tools/m2/1000 years). Again, there were no statis-
tically significant differences in quantities of retouched tools
between chronological periods (Kruskal–Wallis test; n = 60,
v2(4) = 3.997, p = 0. 406) (Fig. 4).
5.3.2. Faunal meat weights
Data on faunal assemblage meat weights and the area (m2) from
which the assemblages derived were available for 24 sites (39
occupations) all of which were caves and rock-shelters, and the
majority of which were excavated and/or analysed post-1970. Of
these, 13 sites had published MNI counts and the rest were con-
verted from NISP counts using the conversion factors shown in
Table S8. Due to the small number of sites for which data were
available, sites with published MNI values and sites with converted
MNI values were combined in the analysis. The greatest faunal
meat weight densities was found in the Azilian (n = 5,
Md = 185.3 kg/m2/1000 years) and the smallest in the Gravettian
(n = 9, Md = 48.0 kg/m2/1000 years) (Fig. 4). A Kruskal–Wallis test
showed no statistically significant differences in faunal meat
weights between chronological periods (n = 39, v2(4) = 1.326,
p = 0.857).
6. Discussion
There are clear fluctuations in all the proxies suggesting that
relative population sizes were not uniform across the Upper
Palaeolithic in Southwestern France. This is not unexpected, as
ethnographic data demonstrate that regional hunter–gatherer pop-
ulations are subject to frequent fluctuations in relative population
growth and decline, often liked with environmental factors and
their associated impact on resource availability (Binford, 2001;
Pennington, 2001). This is likely to have been the case in the
Upper Palaeolithic, despite assumptions of relatively stable,
slow-growing, global Pleistocene populations (see Hassan, 1981).
Where these patterns have been tested for statistical significance,
in the majority of cases the results were deemed to not be statisti-
cally significant. Nonetheless, the samples are often small, and it is
incorrect to assume that because a relationship is not statistically
significant, it is not archaeologically significant (and vice versa).significance was restricted to the differences between the
Magdalenian and Azilian (p = 0.014) (Table 3).
5.3. Occupation intensity
5.3.1. Retouched tools
Data on quantities of retouched tools and the area (m2) from
which the tools derived were available for 60 cave/rock-shelter
and 14 open-air occupations, the majority of which were excavated
and/or analysed post-1970. An analysis of data from both open-air
and sheltered sites shows the greatest density of retouched tools
during the Solutrean (n = 5, Md = 40.0 tools/m2/1000 years) and
the lowest in the Magdalenian (n = 16, Md = 4.9 tools/m2/1000
years). However, the small relative sample size of the Solutrean
may be biasing the results, as well as the lack of data from
open-air sites for both the Solutrean and Azilian periods. A
Kruskal–Wallis test showed no statistically significant differences
Archaeology 39 (2015) 193–209 199The following discussion assumes that the patterns documented
in the data are archaeologically significant and interprets them as
such.
too
.
icalSite counts indicate a peak in population in the Solutrean, and a
clear dip in the Gravettian. In particular, the Late Gravettian is
identified by this proxy to be a period of low regional population
density. The peak in the Late Solutrean supports the long-held
hypothesis of Southwestern France as a population refugium dur-
ing the Last Glacial Maximum (Jochim, 1987). The results pre-
sented here also support the hypothesis of the Final Magdalenian
as a period of high population density in the region although this
is actually a decrease from the preceding Upper Magdalenian,
rather than the increase proposed by de Sonneville-Bordes
(1960: 93) and Mellars (1973: 271). A similar pattern of demo-
graphic fluctuations across the Upper Palaeolithic sequence of
Southwestern France was demonstrated by French and Collins
(2015) using summed probability distributions of radiocarbon
Fig. 4. Documented median occupation intensity values measured through retouched
in Southwestern France. See Tables S5 and S6, Supplementary Material for full data
200 J.C. French / Journal of Anthropologdates.
In terms of site size, the ratio of small:large sites is very consis-
tent across the Upper Palaeolithic, although due to the relationship
between site size and type of site discussed earlier, this fluctuates
slightly depending on the ratio of sheltered: open-air sites. This
pattern suggests similar group sizes amongst hunter–gatherers in
each period, and a similar regional population density as rates of
group population and dispersal, and the relative frequency of time
spent in groups at either end of this continuum, could reasonably
be interpreted as consistent across the Upper Palaeolithic
sequence. Nonetheless, chronological differences are apparent in
the more accurate proxy of documented area of occupation of shel-
tered sites, especially the contrast between the largest
Magdalenian sites and smallest Azilian sites, suggesting possible
differences in group size between techno-complexes.
In contrast, occupation intensity as measured through
retouched tools at sheltered sites suggests differing group sizes
between periods, with the highest values (and largest groups) seen
in the Solutrean and the lowest values (and smallest groups) in the
Azilian. Occupation intensity as measured through faunal meat
weights showed a different pattern, being highest in the Azilian
and lowest in the Gravettian.
As discussed earlier, a range of proxies have been used in this
study in an attempt to overcome the biases inherent in each, and
to provide a more holistic view of past demographic change
through the consideration of multiple strands of archaeological
evidence. However, the conflicting patterns presented in the datafor each period call into question the assumption that all the prox-
ies are measuring demographic change. These differences highlight
the range of possible patterns of relative demographic change that
can be inferred from the archaeological record depending on the
proxy used. It is worth considering the extent to which taphonomic
factors and data availability have affected both the different prox-
ies and/or introduced biases at specific chronological points of the
Upper Palaeolithic sequence.
6.1. The utility of old collections for demographic analysis
The aforementioned long and intensive history of Palaeolithic
research in Southwestern France, while providing a large corpus
l counts and faunal meat weight at archaeological sites across the Upper Palaeolithic
Archaeology 39 (2015) 193–209of data, has introduced its own set of biases and limitations which
have impacted the results of this study. As a high proportion of the
Upper Palaeolithic sites in the region were excavated and pub-
lished in the early 20th century, the collection and presentation
of the data reflect the contemporary research priorities (see
Sackett, 1981, 1991). Specifically, these refer to research strategies
which had the primary aim of data collection for building regional
chronologies. This predominant interest in chronology involved
greater attention to the vertical, as opposed to horizontal, exposure
of the site, and a tendency to concentrate on the richest archaeo-
logical sites (usually caves/rock-shelters) where the largest num-
ber of chronologically diagnostic lithic type fossils could be
found, often conflating several stratigraphic levels (Sackett, 1968:
68).
These research priorities and excavation strategies have
affected the availability of the data required for this analysis.
Many sites mentioned in the literature were unpublished, the
majority of which were excavated in the early 20th century prior
to the professionalization of archaeological research in the region
(White, 2002: 73). As the majority of the required data from these
sites were missing, these function as little more than ‘dots on a
distribution map’ of the Upper Palaeolithic settlement of
Southwestern France. Data on the spatial extent of occupations
was most clearly lacking; data on site size were only available
for 238 sites (44% of the total sample), and several of the larger
multi-level sites in the region, which would have permitted the
study of temporal inter-site occupation intensity fluctuations had
to be excluded due to a lack of data on the areas of the deposits
based on data from sheltered sites were deemed more reliable
due to factors similar to those listed above for the site size anal-
both within and between proxies and site types, and across the
chronological span of the Upper Palaeolithic.
7. Behavioural explanations for patterns in the proxy data:
focus on the Aurignacian and Gravettian
In addition to uncertainties introduced through the incomplete-
ness of the archaeological record, and the use of ‘old’ data, the
aforementioned difficulty of equifinality of interpretation requires
gical Archaeology 39 (2015) 193–209 201ysis, although it should be noted that rock-shelters in the valley
floors are susceptible to possible flooding and scouring of occupa-
tion deposits (Straus, 1990: 259), and those on slopes can suffer
intermittent episodes of slopewash processes and both scouring
of deposits and accelerated infill (Attenbrow, 2006: 106; see
Roebroeks et al., 2011 for the study region). However, detailedexcavated (e.g. La Madeleine, Laugerie-Haute Est/Ouest
(Dordogne)). However, while the exact impact of these different
research priorities and excavation strategies on the demographic
patterns presented here is difficult to quantify there is no a priori
reason to assume that they did not impact all of the sub-periods
of the Upper Palaeolithic equally.
6.2. Taphonomy
The application of the taphonomic correction curve of Surovell
et al. (2009) to the open-air sites including in the site count data-
set has, theoretically, prevented the over-representation of
younger sites in the demographic signature generated through
this proxy, although the open-air sites form only 1/3 of the total
site sample. The aforementioned similarities in the patterns gen-
erated through the use of site counts and the alternative demo-
graphic proxy of radiocarbon date summed probability
distributions (French and Collins, 2015) does, however, suggest
that time-transgressive taphonomic bias is not substantially
impacting the pattern seen in the site count analysis and
enhances the reliability of the trends documented (as the differ-
ent materials being studied – organic materials for radiocarbon
dating and stone tools as a proxy of site presence – are subject
to very different taphonomic processes). With regard to the other
proxies, the notion of a simple linear relationship between the
archaeological data and age (i.e. that older materials are less fre-
quent due to increased taphonomic loss with time-depth) is
refuted by the patterns presented above, which show no clear
chronological directionality.
Despite the application of the correction curve to numbers of
open-air sites, it is still suggested that the number of sheltered
sites documented in the site counts analysis is a better represen-
tation of the number of such sites occupied in the past, due to the
twin factors of enhanced visibility (resulting in a bias towards the
discovery of these sites; Laville et al., 1980; Wobst, 1974: 149)
and greater protection of associated archaeological deposits.
However, recent rescue excavations in the region conducted by
INRAP have demonstrated the presence of previously unknown
Palaeolithic open-air sites (Bourguignon et al., 2004). The ques-
tion thus remains for the other proxies as to whether data from
sheltered sites can be accepted as representative of all past
activity, or whether their use was seasonal, linked to wider
climatic/environmental conditions or reflected some other
behavioural preference. Some relationships are, however, demon-
strated between the type of site and the proxy data studied.
Estimates of site size from sheltered sites are suggested to be
more reliable, as open-air sites are more susceptible to problems
of shifting lateral occupations, horizontal displacement of materi-
als due to periglacial solifluction (Lenoble et al., 2008), and
limited excavation relative to the estimated overall extent of
the site. No data on faunal meat weight estimates were available
from open-air sites due to poor preservation of organic remains.
Data on retouched tool densities were, however, available from
both types of site. Despite the robusticity of lithics, estimates
J.C. French / Journal of Anthropolotaphonomic studies of many of the sheltered sites included in
this analysis are lacking, preventing a more detailed considera-
tion of the impact of differential destruction and preservationconsideration. Differences in the behaviour of hunter–gatherers
across the chronological span of the Upper Palaeolithic could feasi-
bly account for the variations seen both within and across the
proxy data. The Aurignacian–Gravettian (excluding the Late
Gravettian) sequence is the only phase of the Upper Palaeolithic
in the region with enough data (defined as P5 assemblages) to
examine changes in the proxies across sub-periods. The three most
important behavioural variables are discussed below with refer-
ence to this well-documented sequence.
7.1. Mobility
A change in mobility strategy is frequently cited as a possible
explanation for variations in the types of proxy data used in this
study, particularly numbers of archaeological sites (Attenbrow,
2006). Variations in numbers of sites could reflect differences in
the ways that people used the landscape and moved around the
region, with people possibly; (a) moving between sites more or less
frequently; (b) organising their movement differently, or; (c)
spending different amounts of time in various stages of group fis-
sion/fusion (aggregation and dispersal) as is characteristic of
ethnographic hunter–gatherers (Pedersen and Woehle, 1991;
Turnbull, 1968, 1972; Woodburn, 1968, 1972). In particular, differ-
ences in mobility strategy adopted by hunter–gatherers along
Binford’s (1980) forager/collector continuum affect both the num-
ber of potential archaeological sites generated and their subse-
quent visibility, likelihood of survival, and eventual incorporation
into archaeological datasets. However, the two explanations are
not mutually exclusive and the exact relationship between demog-
raphy and mobility is unclear, the two variables being both
inter-dependent (see Grove, 2009; Kelly, 2003) and affected by
external factors, including environmental setting (Kelly, 1983,
2013; Whallon, 2006). At a minimum there is a correlation
between group size and population density and mobility strategy,
with larger group sizes and higher population densities found
amongst groups practising logistical (collector) mobility (Binford,
2001).
As shown in Fig. 2, numbers of archaeological sites/1000 years
are highest in this period in the Middle Gravettian, and lowest in
the Early Gravettian and Late Gravettian with a considerable
decrease in the numbers of sites between the Middle Gravettian
and the Late Gravettian. The question remains as to what extent
changes in mobility across the Aurignacian and Gravettian, rather
than demographic changes, are affecting the patterns seen in the
Table 4
Numbers of new sites created and the relative frequency (%) of cave/
rock-shelter sites for the Aurignacian and Gravettian in the study region.
Period Number of new sites/1000 years
and % of total site sample
% of cave/
rock-shelter sites
Early Aurignacian 12 (41%) 77.5
Late Aurignacian 4.5 (15%) 85.2
Early Gravettian 6 (29%) 86.3
Middle Gravettian 7 (20%) 88.6
Late Gravettian 2.6 (39%) 65.2
number of archaeological sites. One way to address this is to con-
sider some secondary site characteristics (Table 4). For example,
the similar relative frequency of sheltered: open-air sites through-
out the Aurignacian and Gravettian sequence in the region (rang-
ing from 65.2% of the known sample in the Late Gravettian to
88.6% of sites in the Middle Gravettian) suggests that differences
in the settlement pattern in terms of relative use (and subsequent
preservation and archaeological discovery of) of open-air sites to
sheltered sites, cannot account for the large scale differences in
the documented number of archaeological sites. While data on
the number of sites established in each period (i.e. at which
the first traces of human occupation dated to the period under
consideration) support the notion of the Late Gravettian as a period
of low population density (with only 2.6 new sites estab-
lished/1000 years), these accounted for one of the highest propor-
tions of the total number of sites seen across the sequence (39% of
all sites). In contrast, in the Middle Gravettian only 20% of the doc-
umented sites were first occupied by humans during this period.
Differences in the relative visibility and importance to mobile
hunter–gatherer groups of sites that were either newly established
or well-known and had been occupied previously may affect the
correlation between regional population density and numbers of
archaeological sites. In particular, the creation of relatively more
new sites in the Late Gravettian period, compared with both the
Early and Middle Gravettian could be related to higher residential
mobility as small hunter–gatherer groups moved repeatedly
7.2. Lithic technology
As shown in Fig. 4, occupation intensity as measured through
retouched tool counts at sheltered sites was slightly higher in the
Gravettian than the Aurignacian, although this difference is not
statistically significant. Divided into sub-periods, the lowest rate
of occupation intensity was found in the Early Aurignacian
(n = 15, Md = 6.8 retouched tools/m2/1000 years) and the highest
in the Middle Gravettian (n = 13, Md = 17.7 retouched tools/m2/
1000 years) (Fig. 5). Again, there were no statistically significant
differences in quantities of retouched tools between the
sub-periods (Kruskal–Wallis test: n = 51, v2(3) = 2.906, p = 0.406).
There are several behavioural variables which could affect the
quantities of retouched tools independent of number of people,
including; (a) changes in the importance of retouched tools com-
pared to un-retouched flakes and blades; (b) changes in the impor-
tance of lithic tools in the cultural repertoire compared to organic
(bone, antler) tools; (c) changes in location of artefact discard; (d)
variations in lithic manufacturing technology, including differ-
ences in the rate of tool reduction/re-sharpening, and; (e) the
effects of variable raw material supplies on rates of tool manufac-
ture, use, and discard. Of these variables, the best available com-
parative data relates to raw material selection across the
Aurignacian and Gravettian and data on technological strategies
from the key site of Abri Pataud (Dordogne).
Variations in raw material and distances of procurement can
202 J.C. French / Journal of Anthropological Archaeology 39 (2015) 193–209around the landscape to collect food resources, although this is
somewhat at odds with the overall low number of sites docu-
mented for this period. The aforementioned lack of data on occupa-
tion intensity in the Late Gravettian makes this hypothesis difficult
to test. However, data from the Middle Gravettian show high num-
bers of sites, fewer new sites created, and high densities of
retouched tools (see below), possibly suggesting increased regional
population density, accompanied by an increase in logistical
mobility in which more sites acted as intensively occupied
‘home-bases’ where large groups camped, and fewer low-density
new sites were created.Fig. 5. Documented median occupation intensity values measured through retouched too
Aurignacian and Gravettian in Southwestern France. Data on the Late Gravettian was e
Supplementary Material for full data.affect the lithic assemblage, through the organisational strategy
that shows an inverse relationship between the amount of material
transported, which usually decreases with distance from sources,
and the extent to which the material is utilised, which increases
with distance (Blades, 1999a: 712). Due to the extra effort involved
in procuring the material, any artefacts made from non-local
sources are likely to be considered less expedient than those for
which the material is locally abundant, and as a result should be
subjected to a greater degree of re-working/re-sharpening.
Demars (1999) has demonstrated a link in the study region
between raw material type and tool types, with high-quality,l counts and faunal meat weight at archaeological sites across the sub-periods of the
xcluded due to an inadequate sample size (<5 assemblages). See Tables S5 and S6,
non-local flint being used preferentially for the making of ‘presti-
gious’ tool forms, and ‘essential’ tools, which form the bulk of the
tool–kit (primarily scrapers and burins) made from locally abun-
dant and easy to access sources. How could these factors have
impacted the quantities of retouched tools across the
Aurignacian and Gravettian sequence?
Studies of the raw material from numerous sites in the study
region demonstrate a positive correlation between the frequency
of high-quality, non-local Bergerac flint and Upper Palaeolithic
industries dominated by blade production (Chiotti, 2005;
Demars, 1984, 1999; Nespoulet, 2000). As such, the relative abun-
dance of this flint type is higher in Early Aurignacian and
Gravettian assemblages than Late Aurignacian assemblages.
Demars (1999: 7) also notes a high proportion of Gravette and
Font-Robert points manufactured in Bergerac flint. These assem-
blages largely adhere to the aforementioned link between mode
of raw material transport and distance from site, with non-local
flint introduced at a site in an already worked form, and more
stages of the production sequence evident at the site for local flint
of retouched tools relative to the total lithic assemblage. The mean
values of retouched tools percentages are similar across the Early
and Late Aurignacian (16.1% and 17.6% respectively), although
the values for each level fluctuate greatly (Table 5). The data from
Abri Pataud also permit the comparison of occupation intensity
values based on number of lithic pieces with those seen through
retouched tool counts (Fig. 6). The documented trends are similar
in both proxies, with the exception of an increase in lithic pieces
not seen in the retouched tool counts in the Early Gravettian.
This increase in lithic pieces is the opposite of what would be
expected for this period, as the increased quantity of high-quality
Bergerac flint found in this level (Nespoulet, 2008: 143) should,
as discussed above, be associated with a decrease in on-site tool
production and associated debris. While this pattern from the
Abri Pataud cannot be extended to all of the sites studied (partic-
ularly as the sequence from the Abri Pataud differs from the overall
ues b
008
ed t
sity
ls/m
0
10
20
30
40
50
60
Early 
Aurignacian
Late 
Aurignacian
Early 
Gravettian
Middle 
Gravettian
Late Gravettian
N
um
be
r o
f r
et
ou
ch
ed
 to
ol
s/
m
2 /1
00
0 
ye
ar
s
0
100
200
300
400
500
600
Early 
Aurignacian
Late 
Aurignacian
Early 
Gravettian
Middle 
Gravettian
Late 
Gravettian
N
um
be
r o
f l
ith
ic
 p
ie
ce
s/
m
2 /1
00
0 
ye
ar
s
Fig. 6. Comparison of occupation intensity values for the site of Abri Pataud
(Dordogne) through the proxies of retouched tool counts (above) and total number
of lithic pieces (below). See Table S5, Supplementary Material for data.
J.C. French / Journal of Anthropological Archaeology 39 (2015) 193–209 203(Demars, 1999: 8), although this does vary across the study region
(Morala, 1984; Morala and Turq, 1990). While the correlation
between an emphasis on blade production and raw material would
affect the total amount of lithic material brought to sites, this
would primarily affect differences in the quantity of all lithic deb-
itage, rather than retouched tool counts. The use of non-local flint
for Gravettian ‘Gravette’ and ‘Font-Robert’ points might partially
account for the decreased retouched tool count for the Early
Gravettian (following the assumption that these would have been
preferentially re-worked rather than discarded), although these
only form a small component of the total retouched tool repertoire.
The presence of increased retouched intensity in Early Aurignacian
assemblages (heavy ‘Aurignacian retouch’), indicative of increased
tool re-sharpening and re-use, in combination with the relative
abundance of flint from non-local sources, and the Late
Aurignacian technological strategy which produced more blanks
per core (Blades, 1999a, 1999b, 2001), could potentially accommo-
date the increase in numbers of retouched tools seen between the
Early and Late Aurignacian.
As one of the few sites in the region which covers the full
Aurignacian and Gravettian sequence (with the exception of the
Proto-Aurignacian) and for which seemingly reliable data on total
quantities of lithic material (including debitage and nuclei) are
present, the site of Abri Pataud (Dordogne) provides a rare oppor-
tunity to assess chronological differences in the relative frequency
Table 5
Total lithic counts, % of retouched tools in the assemblage and occupation intensity val
Abri Pataud (Dordogne). Data for the Aurignacian from Chiotti (2005) and Nespoulet (2
sub-period. See Table S5, Supplementary Material for full data on numbers of retouch
Level Total quantity of
lithic pieces
% of retouched
tools
Occupation inten
of retouched too
14 1466 15.8 25.5
13 1302 6.1
12 2819 18.1
11 6759 17.8
10/11 279 22.6
10 772 9.6 33.9
9 319 9.4
8 5676 11.8
7 upper 5308 15.8
7 lower 556 22.1
6/7 289 28.0
6 3802 26.3
5 5800 6.0 27.6
4 8833 8.8 48.5
3 1656 4.9 5.7
2 1845 7.3ased on the two proxies of retouched tool counts and total lithic counts for the site of
) for the Gravettian. Areas used in calculations are the mean of all levels dated to that
ools and areas used.
(number
2/1000 years)
Occupation intensity (number
of lithic pieces/m2/1000 years)
Cultural attribution
200.4 Early Aurignacian
261.3 Late Aurignacian
457.1 Early Gravettian
548.8 Middle Gravettian
95.7 Late Gravettian
reliance on converted MNI estimates for the meat weight analysis
makes taking this correlation further tenuous. However, a similar
processing strategy has been documented for the reindeer assem-
blage for the Middle Gravettian at the site of Le Flageolet I
(Dordogne) (Enloe, 1993).
ical Archaeology 39 (2015) 193–209trend of an increase in occupation intensity values between the
Early and Late Aurignacian), the near correlation of the retouched
tool and lithic piece values lends credence to the suggestion that
fluctuations in retouched tool intensity values may reflect a demo-
graphic signature and are not being obscured by behavioural and
technological variation.
7.3. Faunal acquisition and processing strategies
As seen in Fig. 4, occupation intensity measured through faunal
meat weights was much greater in the Aurignacian than the
Gravettian. Divided into sub-periods, the lowest rate of occupation
intensity was found in the Middle Gravettian (n = 5,
Md = 49.4 kg/m2/1000 years) and the highest in the Early
Gravettian (n = 5, Md = 178.5 kg/m2/1000 years) (Fig. 5). Again
none of the differences seen between periods were statistically sig-
nificant (Kruskal–Wallis test: n = 29, v2(3) = 2.604, p = 0.457).
Interestingly, the peaks and troughs in the data are the opposite
of those found with the retouched tools, although it should be
noted, that in contrast to the retouched tool analysis, the available
sample for the Gravettian sub-periods was up to 50% smaller
than those for the Aurignacian.
As with the retouched tools, there are several behavioural vari-
ables that could impact the quantity of faunal remains at archaeo-
logical sites independent of the number of consumers. These
include; (a) variations in hunting strategies and the effects of dif-
ferent age/sex profiles of the animals killed on the quantity of meat
provided; (b) variations in the storage of animal remains; (c)
potential destruction of the faunal remains through the actions
of various cultural practices (e.g. burning bones as a source of fuel);
(d) variable spatial patterns of discard of the faunal remains within
the sites, or changes in the location of discard; (e) variation in the
intensity and exploitation of the animal carcass for consumption,
and; (f) variation in transportation of animal carcasses due to fac-
tors such as distance from kill site. Data on these variables are
mostly related to faunal procurement and processing, although
the lack of comparative studies across the Aurignacian and
Gravettian sequences forces a reliance on ‘snap-shots’ of beha-
viour, and the impact of any differences on the patterns docu-
mented in the occupation intensity analysis are difficult to assess.
One of the defining differences between the Early and Late
Aurignacian in the study region is the decrease in the relative fre-
quency of reindeer in faunal assemblages between the two periods
(Boyle, 1990: 185; Grayson and Delpech, 2003, 2005; Grayson
et al., 2001; Mellars, 2004). Several studies of Early Aurignacian
processing strategies of both reindeer and other ungulates in the
study region suggest intensive on-site carcass exploitation, report-
ing cut-marks concurrent with on-site skinning of carcasses and
disarticulation of prime meat-yielding elements (Sollier and
Mallye, 2012), and evidence of long-bone fracturing for marrow
extraction (Morin, 2004), with reindeer remains indicating that
most of the consumption took place on site (Binford’s (1978)
‘inverse bulk curve’) (Chiotti et al., 2003 cf. Blades, 1999b: 107).
Unfortunately (for present purposes), these data are most often
compared to that from the earlier Mousterian and Châtelperronian
periods, rather than the Late Aurignacian.
Boyle’s (1990: 240) study of the butchery and carcass manage-
ment strategies used throughout the Gravettian at the site of Abri
Pataud (Dordogne) show some chronological differences across the
period. There is evidence of more intensive processing in the Late
Gravettian, corresponding with a ‘bulk curve’ and a ‘gourmet curve’
focusing on high-utility parts in the Early and Middle Gravettian.
Data on faunal meat weights from the site show a concomitant
2
204 J.C. French / Journal of Anthropologdecrease in the Late Gravettian (from 481.5 kg/m /1000 years in
the Middle Gravettian to 110.7 kg/m2/1000 years in the Late
Gravettian (Table S6, Supplementary Material), although theDifferences in processing strategies of different species further
complicates the assessment of the impact of this behavioural vari-
able on the occupation intensity estimates generated, with differ-
ences seen between reindeer and horse assemblages in both
Aurignacian (Chiotti et al., 2003: 199) and Gravettian (Boyle,
1990: 256) sites in the region. While the examples discussed above
provide insight into the range of treatment of faunal remains in the
Aurignacian and Gravettian, they show very little chronological
patterning, and as the assemblages probably reflect more than
one season of occupation, data on carcass management should be
treated with caution (Boyle, 1990: 240).
8. Conclusion
Palaeolithic archaeologists are increasingly citing demographic
causes for patterns seen in the archaeological record. While mod-
elling and genetic data provide useful estimates of past population
numbers and trends, there still remains a need for the empirical
assessment of archaeological data in the investigation of palaeode-
mography. This study has compared the patterns of relative
demographic change generated through the use of three different
proxies for the Upper Palaeolithic of Southwestern France.
This region was selected for study due to the presence of a
well-defined and dated chronological sequence which permitted
the study of diachronic change, the exceptional richness of the
archaeological record, and the concomitant long history of inten-
sive archaeological research, which minimised the effects of sam-
pling and taphonomic bias.
In contrast to previous studies which utilise the same archaeo-
logical proxies (Mellars and French, 2011), the proxies did not
show any clear chronological patterning,1 and frequently presented
conflicting demographic signatures. This raises some interesting
issues of interpretation. Taking the results shown by the majority
of proxies, should, at least in probability terms, be the best way to
proceed, although that does not preclude the possibility that the sig-
nature generated through the converging proxy is the ‘correct’ one
(see French, 2015). What is pattern and what is noise? One way of
assessing the relative value of the regional patterns generated is to
compare the results to wider-scale demographic studies (e.g.
Bocquet-Appel and Demars, 2000a, 2000b; Gamble et al., 2004,
2005), although these are often based on a single type of archaeolog-
ical data. While the focus of this study has been documenting rela-
tive demographic change rather than explaining it, another
possible line of enquiry to select between competing patterns could
be the comparison between the patterns generated and the patterns
expected. Both group size and population density of ethnographi-
cally documented hunter–gatherers are known to vary according
to environmental factors (Binford, 2001; Birdsell, 1953, 1958,
1968; Grove, 2009; Johnson, 2014; Layton and O’Hara, 2010;
Marlowe, 2005). The comparison between the results generated by
each proxy and the expected response of hunter–gatherers to the
prevailing climatic and environmental conditions during a given
1 It is unclear how the results of this current study impact on this previously
published study of demographic change during an earlier period in the region. In
Mellars and French (2011), a figure of a ten-fold population increase across the
Neanderthal-to-Modern Human transition was proposed, based on the convergence
in the results of the analysis between the same three proxies used in this study. Does
the lack of convergence seen in the data here suggest that this figure is likely an
overestimate, or does the lack of a similar pattern in this Upper Palaeolithic dataset
reinforce the strength and size of this difference in the size of populations in
Southwestern France across the Middle-to-Upper Palaeolithic transition?
Bocquet-Appel, J.-P., Demars, P.-Y., 2000a. Neanderthal contraction and modern
gicalperiod, particularly as it affects the availability of food resources,
provides a useful starting point for choosing between competing
demographic signatures.
For this dataset, I suggest that the number of sheltered sites and
quantities of retouched tools are the most reliable; the former
being well-preserved and the latter being robust, subject to limited
taphonomic degradation, and largely standardised across sites in
both collection methods and subsequent reporting. These suggest
relative population increases in the Solutrean, Middle Gravettian,
and Upper and Final Magdalenian with relative decreases in the
Late Gravettian and Middle Magdalenian. A similar pattern has
been documented for the study region using summed probability
distributions of radiocarbon dates as a demographic proxy
(French and Collins, 2015). Population increase at the regional level
is strongly supported for the Middle Gravettian by the high occu-
pation intensity as measured through quantities of retouched tools
seen at archaeological sites dated to this phase. The faunal meat
weight estimates are deemed to be a priori the least reliable proxy,
being more susceptible to taphonomic destruction, lacking a fully
standardised method of quantification (leading to reduced
inter-site comparability), and being based on a smaller sample.
These correlate with the proxy data which are obtained most easily
from the Palaeolithic archaeological record. Although attributable
largely to past regional research trends and trajectories, the diffi-
culties documented with the lack of availability of data from old
collections in the Southwestern French Upper Palaeolithic calls
into question whether similar palaeodemographic methods can
be applied to other, less intensively studied, regions.
Despite an imperfect archaeological record, it is possible to use
archaeological data to explore patterns of relative demographic
change in the Palaeolithic. Problems of equifinality of interpreta-
tion need to be considered, although as discussed above with the
case of the Aurignacian and Gravettian sequence, a lack of studies
of the long-term fluctuations in the behavioural variables which
provide alternative explanations for patterns in the data prevent
the empirical testing of competing interpretations. Difficulties of
contemporaneity of occupation between sites and limitations of
chronological resolution in Palaeolithic contexts also abound. The
question remains as to whether these relative estimates are useful
for the type of evolutionary and behavioural ecological approaches
that currently dominate research agendas. In particular, the forced
restriction in this study of the analysis of demographic change
across the broad chrono-typological periods, rather than the much
shorter sub-divisions, is unfortunately crude, conflating possibly
hundreds of human generations. This obscures, or at least confuses,
the role of the individual in past population processes, particularly
with regard to the role of external stimuli (i.e. local environmental
change). Nonetheless, the demographic trends seen in the archae-
ological record could serve as a starting point for further modelling
studies, being used to help select the most likely absolute popula-
tion estimates, and identify which parameter combinations pro-
vide the closest fit with the empirical archaeological data (e.g.
Porcˇicˇ, 2011).
Acknowledgments
This research was funded by a Research Fellowship at
Peterhouse, University of Cambridge and an Arts and Humanities
Research Council (AHRC) Doctoral Studentship. I am extremely
grateful to Paul Mellars for supervising the initial PhD research
upon which this paper is based, as well as for providing extensive
comments on earlier drafts of the manuscript. Thanks also to
Christina Collins, Robert Foley, Robert Kelly and Stephan Shennan
J.C. French / Journal of Anthropolofor discussion of many of the ideas covered in this paper, and
two anonymous reviewers for their constructive and useful
feedback.human colonization of Europe. Antiquity 74, 544–552.
Bocquet-Appel, J.-P., Demars, P.-Y., 2000b. Population kinetics in the Upper
Palaeolithic of Western Europe. J. Archaeol. Sci. 27 (7), 551–570.
Bocquet-Appel, J.-P., Masset, C., 1982. Farewell to palaeodemography. J. Hum. Evol.
11, 321–333.
Bocquet-Appel, J.-P., Tuffreau, A., 2009. Technological responses of Neanderthals to
macroclimatic variations (240,000–40,000 BP). Hum. Biol. 81 (2–3), 287–307.Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jaa.2015.04.005.
References
Ammerman, A.J., Cavalli-Sforza, L.L., Wagener, D.K., 1976. Towards the estimation of
population growth in old world prehistory. In: Zubrow, E.B.W. (Ed.),
Demographic Anthropology. University of New Mexico Press, Albuquerque
(NM), pp. 27–61.
Ampel, L., Wohlfarth, B., Risberg, J., Veres, D., 2008. Paleolimnological response to
millennial and centennial scale climate variability during MIS 3 and 2 as
suggested by the diatom record in Les Echets, France. Quatern. Sci. Rev. 27,
1493–1504.
Ampel, L., Bigler, C., Wohlfarth, B., Risberg, J., Lotter, A.F., Veres, D., 2010. Modest
summer temperature variability during DO cycles in Western Europe. Quatern.
Sci. Rev. 29, 1322–1327.
Anderson, D.G., Goodyear, A.C., Kennett, J., West, A., 2011. Multiple lines of evidence
for possible human population decline/settlement reorganization during the
early Younger Dryas. Quatern. Int. 242, 570–583.
Andrews, J.T., 1998. Abrupt changes (Heinrich events) in late Quaternary North
Atlantic marine environments: a history and review of data and concepts. J.
Quat. Sci. 13 (1), 3–16.
Armit, I., Swindles, G.T., Becker, K., 2013. From dates to demography in late
prehistoric Ireland? Experimental approaches to the meta-analysis of large 14C
data-sets. J. Archaeol. Sci. 40, 433–438.
Ashton, N., Hosfield, R., 2010. Mapping the human record in the British early
Palaeolithic: evidence from the Solent river system. J. Quat. Sci. 25 (5), 737–753.
Ashton, N., Lewis, S., 2002. Deserted Britain: declining populations in the British late
middle Pleistocene. Antiquity 76, 388–396.
Atkinson, Q.D., Gray, R.D., Drummond, A.J., 2008. MtDNA variation predicts
population size in humans and reveals a major Southern Asian chapter in
human prehistory. Mol. Biol. Evol. 25 (2), 468–474.
Attenbrow, V., 2006. What’s Changing: Population Size of Land-Use Patterns? The
Archaeology of Upper Mangrove Creek, Sydney Basin (Terra Australis 21).
Pandanus Books, Canberra.
Beaumont, M.A., 2004. Recent developments in genetic data analysis: what can they
tell us about human demographic history? Heredity 92, 365–379.
Bertran, P., 2005. Stratigraphie du site des Peyrugues (Lot), une coupe de reference
pour le dernier pléniglaciaire en Aquitaine. Quaternaire 16 (1), 25–44.
Bertran, P., Caner, L., Langohr, R., Lemée, L., d’Errico, F., 2008. Continental
palaeoenvironments during MIS 2 and 3 in southwestern France: the La
Ferrassie rockshelter record. Quatern. Sci. Rev. 27, 2048–2063.
Bertran, P., Sitzia, L., Banks, W.E., Bateman, M.D., Demars, P.-Y., Hernandez, M.,
Lenoir, M., Mercier, N., Prodeo, F., 2013. The Landes de Gascogne (southwest
France): periglacial desert and cultural frontier during the Palaeolithic. J.
Archaeol. Sci. 40, 2274–2285.
Binford, L.R., 1978. Nunamiut Ethnoarchaeology. Academic Press, New York (NY).
Binford, L.R., 1980. Willow smoke and dogs’ tails: hunter–Gatherer settlement
systems and archaeological site formation. Am. Antiq. 45 (1), 4–20.
Binford, L.R., 2001. Constructing Frames of Reference. University of California Press,
Berkeley (CA).
Birdsell, J., 1953. Some environmental and cultural factors influencing the
structuring of Australian aboriginal populations. Am. Nat. 87, 171–207.
Birdsell, J.B., 1958. On population structure in generalized hunting and collection
populations. Evolution 12 (2), 189–205.
Birdsell, J.B., 1968. Some predictions for the Pleistocene based on equilibrium
systems among recent hunter–gatherers. In: Lee, R.B., deVore, I. (Eds.), Man the
Hunter. Aldine, New York (NY), pp. 229–240.
Blaauw, M., Wohlfarth, B., Christen, J.A., Ampel, L., Veres, D., Hughen, K.A., Preusser,
F., Svensson, A., 2010. Were last glacial climate events simultaneous between
Greenland and France? A quantitative comparison using non-tuned
chronologies. J. Quat. Sci. 25 (3), 387–394.
Blades, B.S., 1999a. Aurignacian settlement patterns in the Vézère valley. Curr.
Anthropol. 40 (5), 712–719.
Blades, B.S., 1999b. Aurignacian lithic economy and early human mobility: new
perspectives from classic sites in the Vézère valley of France. J. Hum. Evol. 37,
91–120.
Blades, B.S., 2001. Aurignacian Lithic Economy. Ecological Perspectives from
Southwestern France. Plenum Press, New York (NY).
Bocquet-Appel, J.-P., Degioanni, A., 2013. Neanderthal demographic estimates. Curr.
Anthropol. 54 (S8), S202–S213.
Archaeology 39 (2015) 193–209 205Bocquet-Appel, J.-P., Demars, P.-Y., Noiret, L., Dobrowsky, D., 2005. Estimates of
Upper Palaeolithic meta-populations size in Europe from archaeological data. J.
Archaeol. Sci. 32, 1656–1668.
icalBocquet-Appel, J.-P., Naji, S., Linden, M.V., Kozlowski, J.K., 2009. Detection of
diffusion and contact zones of early farming in Europe from the space-time
distribution of 14C dates. J. Archaeol. Sci. 36, 807–820.
Bon, F., 2002. L’Aurignacian entre Mer et Océan. Société Préhistorique Française
Mémoire, Paris, p. 29.
Boone, J.L., 2002. Subsistence strategies and early human population history: an
evolutionary ecological perspective. World Archaeol. 34 (1), 6–25.
Bordes, J.-G., 2002. Les Interstratifications Châtelperronian/Aurignacien du Roc de
Combe et du Piage (Lot, France): Analyse Taphonomique des Industries
Lithiques, Implications Archéologiques. PhD Thesis, Université de Bordeaux I.
Bordes, J.-G., 2003. Lithic taphonomy of the Châtelperronian/Aurignacian
interstratifications in Roc de Combe and Le Piage (Lot, France). In: Zilhão, J.,
D’Ericco, F. (Eds.), The Chronology of the Aurignacian and of the Transitional
Technocomplexes: Dating, Stratigraphies, Cultural Implications. Trabalhos de
Arqueologia 33, Lisbon, pp. 223–244.
Bourguignon, L., Ortega, L., Sellami, F., Brenet, M., Grigoletto, F., Vigier, S., Daussy, A.,
Deschamps, J.-F., Casagrande, F., 2004. Les occupations paléolithiques
découvertes sur la section Nord de la déviation de Bergerac: résultats
préliminaires obtenus à l’issue des diagnostics. Préhist. Sud-Ouest 11, 155–172.
Boyd, R., Richerson, P.J., 1985. Culture and the Evolutionary Process. University of
Chicago Press, Chicago (IL).
Boyle, K.V., 1990. Upper Palaeolithic Faunas from South-West France. BAR
International Series 557, Oxford.
Buikstra, J.E., Konigsberg, L.W., 1985. Paleodemography: critiques and
controversies. Am. Anthropol. 87 (2), 316–333.
Buikstra, J.E., Konigsberg, L.W., Billington, J., 1986. Fertility and the development of
agriculture in the prehistoric Midwest. Am. Antiq. 51, 528–546.
Burke, A., 2006. Neanderthal settlement patterns in Crimea: a landscape approach. J.
Anthropol. Archaeol. 25, 510–523.
Cavalli-Sforza, L.L., Feldman, M.W., 1981. Cultural Transmission and Evolution: A
Quantitative Approach. Princeton University Press, Princeton (NJ).
Chapman, J., 1999. Archaeological proxy data for demographic reconstructions:
facts, factoids or fiction? In: Bintliff, J., Sbonais, K. (Eds.), Reconstructing Past
Population Trends in Mediterranean Europe (3000 BC–AD 1800). Oxbow Books,
Oxford, pp. 65–76.
Chikhi, L., Sousa, V.C., Luisi, P., Goossens, B., Beaumont, M.A., 2010. The confounding
effects of population structure, genetic diversity, and the sampling scheme on
the detection and quantification of population size changes. Genetics 186, 983–
995.
Chiotti, L., 2005. Les Industries Lithiques Aurignaciennes de l’Abri Pataud, Dordogne,
France. Les Fouilles de Hallam L. Movius Jr. BAR International Series 1392,
Oxford.
Chiotti, L., Patou-Mathis, M., Vercoutère, C., 2003. Comportements techniques et de
subsistence à l’Aurignacien ancien: la couche 11 de l’abri Pataud (Dordogne).
Gallia Préhist. 45, 157–203.
Clark, G.A., Lindly, J.M., 1991. On paradigmatic biases and Paleolithic research
traditions. Curr. Anthropol. 32 (5), 577–586.
Collard, M., Kemery, M., Banks, S., 2005. Causes of toolkit variation among hunter–
gatherers: a test of four competing hypotheses. Can. J. Archaeol. 29, 1–19.
Collard, M., Buchanan, B., Morin, J., Costopoulos, A., 2011. What drives the evolution
of hunter–gatherer subsistence technology? A reanalysis of the risk hypothesis
with data from the Pacific Northwest. Philos. Trans. R. Soc. B 366, 1129–1138.
Collard, M., Buchanan, B., O’Brien, M.J., 2013a. Population size as an explanation for
patterns in the Paleolithic archaeological record: more caution is needed. Curr.
Anthropol. 54 (S8), S388–S396.
Collard, M., Buchanan, B., O’Brien, M.J., Scholnick, J., 2013b. Risk, mobility or
population size? Drivers of technological richness among contact-period
western North American hunter–gatherers. Philos. Trans. R. Soc. B 368,
20120412.
Conard, N.J., 2001. Advances and problems in the study of Palaeolithic settlement
systems. In: Conard, N.J. (Ed.), Settlement Dynamics of the Middle Palaeolithic
and Middle Stone Age. Kerns Verlag, Tübingen, pp. vii–xx.
Conard, N.J., Bolus, M., Münzel, S.C., 2012. Middle Palaeolithic land use, spatial
organization and settlement intensity in the Swabian Jura, southwestern
Germany. Quatern. Int. 247, 236–245.
Corruccini, R.S., Brandon, E.M., Hander, J.S., 1989. Inferring fertility from relative
mortality in historically controlled remains from Barbados. Am. Antiq. 54, 609–
614.
Cox, M.P., Morales, D.A., Woerner, A.E., Sozanski, J., Wall, J.D., Hammer, M.F., 2009.
Autosomal resequence data reveal Late Stone Age signals of population
expansion in sub-saharan African foraging and farming populations. PLoS
ONE 4 (7), e6366. http://dx.doi.org/10.1371/journal.pone.0006366.
Cruz-Uribe, K., 1991. Distinguishing hyena from hominid bone accumulations. J.
Field Archaeol. 8, 467–488.
Culotta, E., 2010. Did modern humans get smart or just get together? Science 328,
164.
Daniau, A.-L., Sánchez-Goñi, M.F., Duprat, J., 2009. Last glacial fire regime variability
in western France inferred from microcharcoal preserved in core MD04-2845,
Bay of Biscay. Quatern. Res. 71, 385–396.
Dansgaard, W., Johensen, S.J., Clausen, H.B., Dahl-Jensen, D., Gundestrup, N.S.,
Hammer, C.U., Hvidberg, C.S., Steffensen, J.P., Sveinbjörnsdottir, A.E., Jouzel, J.,
Bond, G., 1993. Evidence for general instability of past climate from a 250-kyr
ice-core record. Nature 364, 218–220.
206 J.C. French / Journal of AnthropologDaugherty, H.H., Kammeyer, K.C.W., 1995. An Introduction to Population, second
ed. The Guildford Press, New York (NY).David, N., 1973. On Upper Palaeolithic society, ecology and technological change:
the Noaillian case. In: Renfrew, C. (Ed.), The Explanation of Culture Change:
Models in Prehistory. Duckworth, London, pp. 277–303.
de Sonneville-Bordes, D., 1960. Le Paléolithique Supérieur en Périgord. Imprimeries
Delmas, Bordeaux.
de Sonneville-Bordes, D., 1973. The Upper Paleolithic. In: Piggott, S., Daniel, G.,
McBurney, C. (Eds.), France before the Romans. Thames & Hudson, London, pp.
30–60.
de Sonneville-Bordes, D., Perrot, J., 1953. Essai d’adaptation des méthodes
statistiques au Paléolithique supérieur. Premiers résultats. Bull. Soc. Préhist.
Française 50 (5/6), 323–333.
Demars, P.-Y., 1984. Les matières premières lithiques du site du Paléolithique
supérieur de Puyjarige 2 (Brive, Corrèze). Rev. Archéol. Centre France 23 (1), 64–
66.
Demars, P.-Y., 1996. Demographie et occupation de l’espace au Paléolithique
supérieur et au Mesolithique en France. Préhist. Eur. 8, 3–26.
Demars, P.-Y., 1998. Les rapports de l’homme et du mileu dans le nord de
l’Aquitaine au Paléolithique supérieur l’implantations des habitats. Bull. Préhist.
Sud-Ouest 5, 13–30.
Demars, P.-Y., 1999. Circulation des silex dans le nord de l’Aquitaine au
Paléolithique supérieur: l’occupation de l’espace par les derniers chasseurs-
cueilleurs. Gallia Préhist. 40, 1–28.
Derex, M., Beugin, M.-P., Godelle, B., Raymond, M., 2013. Experimental evidence for
the influence of group size on cultural complexity. Nature 503, 389–391.
Dogandzˇic´, T., McPherron, S.P., 2013. Demography and the demise of the
Neanderthals: a comment on ‘Tenfold Population Increase at the
Neanderthal-to-Modern-Human Transition’. J. Hum. Evol. 64 (4), 311–313.
Drucker, D.G., Henry-Gambier, D., 2005. Determination of the dietary habits of a
Magdalenian woman from Saint-Germain-la-Rivière in Southwestern France
using stable isotopes. J. Hum. Evol. 49 (1), 19–35.
Ducasse, S., 2012. What is left of the Badegoulian ‘‘interlude’’? New data on cultural
evolution in southern France between 23,500 and 20,500 cal.BP. Quatern. Int.
272–3, 150–165.
Enloe, J.G., 1993. Subsistence organization in the Early Upper Palaeolithic: reindeer
hunters of the Abri du Flageolet, Couche V. In: Knecht, H., Pike-Tay, A., White, R.
(Eds.), Before Lascaux: The Complex Record of the Early Upper Palaeolithic. CRC
Press, Michigan, pp. 101–115.
Excoffier, L., 2002. Human demographic history: refining the recent African origin
model. Curr. Opin. Genet. Dev. 12, 675–682.
Excoffier, L., Schneider, S., 1999. Why hunter–gatherer populations do not show
signs of Pleistocene demographic expansions. Proc. Natl. Acad. Sci. USA 96,
10597–10602.
Fabre, V., Condemi, S., Degioanni, A., 2009. Genetic evidence of geographical groups
among Neanderthals. PLoS ONE 4 (4), e5151. http://dx.doi.org/10.1371/
journal.pone.0005151.
Fitzhugh, B., Trusler, A.K., 2009. Case study in technological evolution: innovation
and experimentation in and with the archaeological record. In: Shennan, S.
(Ed.), Pattern and Process in Cultural Evolution. University of California Press,
Berkeley (CA), pp. 203–223.
French, J.C., 2015. Demography and the Palaeolithic Archaeological record. J.
Archaeol. Method Theory. http://dx.doi.org/10.1007/S10816-014-9237-4.
French, J.C., Collins, C., 2015. Upper Palaeolithic population histories of
Southwestern France: a comparison of the demographic signatures of 14C
date distributions and archaeological site counts. J. Archaeol. Sci. 55, 122–134.
Freter, A.C., 1997. The question of time: the impact of chronology on Copán
prehistoric settlement demography. In: Paine, R.R. (Ed.), Integrating
Archaeological Demography: Multidisciplinary Approaches to Prehistoric
Population. Center for Archaeological Investigations Occasional Paper No. 24,
pp. 21–42.
Gamble, C., Davies, W., Pettitt, P., Richards, M., 2004. Climate change and evolving
human diversity in Europe during the last glacial. Philos. Trans. R. Soc. B 359,
243–254.
Gamble, C., Davies, W., Pettitt, P., Hazelwood, L., Richards, M., 2005. The
archaeological and genetic foundations of the European population during the
Late Glacial: implications for ‘agricultural thinking’. Camb. Archaeol. J. 15 (2),
193–223.
Garrigan, D., Kingan, S.B., Pilkington, M.M., Wilder, J.A., Cox, M.P., Soodyall, H.,
Strassmann, B., Destro-Bisol, G., de Knijff, P., Novelltto, A., Friedlaender, J.,
Hammer, M.F., 2007. Inferring human population sizes, divergence times and
rates of gene flow from mitochrondrial, X and Y chromosome resequencing
data. Genetics 177, 2195–2207.
Genty, D., Blamart, D., Ouahdi, R., Gilmour, M., Baker, A., Jouzel, J., Van-Exter, S.,
2003. Precise dating of Dansgaard-Oeschger climate oscillations in Western
Europe from stalagmite data. Nature 421, 833–837.
Genty, D., Combourieu-Nebout, N., Peyron, O., Blamart, D., Wainer, K., Mansuri, F.,
Ghaleb, B., Isabello, L., Dormoy, I., von Grafenstein, U., Bonelli, S., Landais, A.,
Brauer, A., 2010. Isotopic characterization of rapid climatic events during OIS 3
and OIS 4 in Villars cave stalagmites (SW-France) and correlation with Atlantic
and Mediterranean pollen records. Quatern. Sci. Rev. 29, 2799–2820.
Ghirlanda, S., Enquist, M., 2007. Cumulative culture and explosive demographic
transitions. Qual. Quant. 41, 591–600.
Ghirlanda, S., Enquist, M., Perc, M., 2010. Sustainability of culture-driven population
dynamics. Theor. Popul. Biol. 77, 181–188.
Archaeology 39 (2015) 193–209Grayson, D.K., Delpech, F., 2003. Ungulates and the Middle-to-Upper Paleolithic
transition at Grotte XVI (Dordogne, France). J. Archaeol. Sci. 30 (12), 1633–1648.
gicalGrayson, D.K., Delpech, F., 2005. Pleistocene reindeer and global warming. Conserv.
Biol. 19 (2), 557–562.
Grayson, D.K., Delpech, F., Rigaud, J.-P., Simek, J., 2001. Explaining the development
of dietary dominance by a single ungulate taxon at Grotte XVI, Dordogne,
France. J. Archaeol. Sci. 28, 115–125.
Greene, D.L., Van Gerven, D.P., Armelagos, G.J., 1986. Life and death in ancient
populations: bones of contention in Paleodemography. Hum. Evol. 1 (3), 193–
207.
Grootes, P.M., Stuiver, M., White, J.W.C., Johnsen, S., Jouzel, J., 1993. Comparison of
the oxygen isotope records from the GISP 2 and GRIP Greenland ice cores.
Nature 366, 552–554.
Grove, M., 2009. Hunter–gatherer movement patterns: causes and constraints. J.
Anthropol. Archaeol. 28, 222–233.
Grove, M., 2010. The archaeology of group size. In: Dunbar, R., Gamble, C., Gowlett, J.
(Eds.), Social Brain, Distributed Mind. Proceedings of the British Academy 158,
pp. 391–411.
Grove, M., 2012. Scatters, patches and palimpsests: solving the contemporaneity
problem. In: Reubens, K., Romanowska, I., Bynoe, R. (Eds.), Unravelling the
Palaeolithic. Ten Years of Research at the Centre for the Archaeology of Human
Origins (CAHO, University of Southampton). University of Southampton Series
in Archaeology 8. Archeopress, Oxford, pp. 153–164.
Hammel, E.A., Howell, N., 1987. Research into population and culture: an
evolutionary framework (and comments and replies). Curr. Anthropol. 28 (2),
141–160.
Harpending, H.C., Batzer, M.A., Gurven, M., Jorde, L.B., Roger, A.R., Sherry, S.T.,
1998. Genetic traces of ancient demography. Proc. Natl. Acad. Sci. USA 95,
1961–1967.
Harris, E.G., Hey, J., 1999. Human demography in the Pleistocene: do mitochondrial
and nuclear genes tell the same story? Evol. Anthropol. 8 (3), 81–86.
Hassan, F.A., 1981. Demographic Archaeology. Academic Press, New York (NY).
Hayden, B., 2012. Neanderthal social structure? Oxford J. Archaeol. 31 (1), 1–26.
Heizer, R.F., 1960. Physical analysis of habitation residues. In: Heizer, R.F., Cook, S.F.
(Eds.), The Application of Quantitative Methods in Archaeology. Viking Fund
Publications in Anthropology 28, pp. 93–157.
Hemming, S.R., 2004. Heinrich events: massive late Pleistocene detritus layers of
the North Atlantic and their global climatic imprint. Rev. Geophys. 42, 1–43.
Henrich, J., 2004. Demography and cultural evolution: how adaptive cultural
processes can produce maladaptive losses: the Tasmanian case. Am. Antiq. 69
(2), 197–214.
Henrich, J., 2006. Understanding cultural evolutionary models: a reply to Read’s
critique. Am. Antiq. 71 (4), 771–782.
Hill, J.N., 1970. Broken K Pueblo: Prehistoric Social Organisation in the American
Southwest. Anthropological Papers of the University of Arizona 18. The
University of Arizona Press, Tucson (AZ).
Hill, K., 1993. Life history theory and evolutionary anthropology. Evol. Anthropol. 2
(3), 78–88.
Hinde, A., 2002. Demographic perspectives on human population dynamics. In:
MacBeth, H., Collinson, P. (Eds.), Human Population Dynamics: Cross-
disciplinary Perspectives. Cambridge University Press, Cambridge, pp. 17–
40.
Hinz, M., Feeser, I., Sjögren, K.-G., Müller, J., 2012. Demography and the intensity of
cultural activities: an evaluation of Funnel Beaker societies (4200–2800 cal BC).
J. Archaeol. Sci. 39, 331–340.
Hiscock, P., 1986. Technological change in the Hunter River valley and the
interpretation of late Holocene in Australia. Archaeol. Ocean. 21, 40–50.
Hopkinson, T., Nowell, A., White, M., 2013. Life history, metapopulation ecology,
and innovation in the Acheulian. PaleoAnthropology 2013, 61–76.
Hosfield, R., 1999. The Palaeolithic of the Hampshire Basin. A Regional Model of
Hominid Behaviour during the Middle Pleistocene. BAR British Series 286.
Archeopress, Oxford.
Hosfield, R., 2005. Individuals among palimpsest data. Fluvial landscapes in
Southern England. In: Gamble, C., Porr, M. (Eds.), The Hominid Individual in
Context. Archaeological Investigations of Lower and Middle Palaeolithic
Landscapes, Locales and Artefacts. Routledge, London, pp. 220–243.
Jochim, M., 1987. Late Pleistocene refugia in Europe. In: Soffer, O. (Ed.), The
Pleistocene Old World: Regional Perspectives. Plenum Press, New York (NY), pp.
317–331.
Johnson, A.L., 2014. Exploring adaptive variation among hunter–gatherers with
Binford’s frames of reference. J. Archaeol. Res. 22, 1–42.
Kelly, R.L., 1983. Hunter–gatherer mobility strategies. J. Anthropol. Res. 39 (3), 277–
306.
Kelly, R.L., 2003. Colonization of new land by hunter–gatherers. Expectations and
implications based on ethnographic data. In: Rockman, M., Steele, J. (Eds.),
Colonization of Unfamiliar Landscapes: The Archaeology of Adaptation.
Routledge, London, pp. 44–58.
Kelly, R.L., 2013. The Lifeways of Hunter–Gatherers. Cambridge University Press,
Cambridge.
Kelly, R.L., Surovell, T.A., Shuman, B.N., Smith, G.M., 2013. A continuous climatic
impact on Holocene human population in the Rocky Mountains. Proc. Natl.
Acad. Sci. USA 110 (2), 443–447.
Kempe, M., Mesoudi, A., 2014. An experimental demonstration of the effect of group
size on cultural accumulation. Evol. Hum. Behav. 35 (4), 285–290.
Kline, M.A., Boyd, R., 2010. Population size predicts technological complexity in
J.C. French / Journal of AnthropoloOceania. Proc. R. Soc. B 277, 2559–2564.
Konigsberg, L.W., Frankenberg, S.R., 2005. Paleodemography: ‘‘Not quite dead’’.
Evol. Anthropol. 3 (3), 92–105.Kuhn, B.F., Berger, L.R., Skinner, J.D., 2010. Examining criteria for identifying and
differentiating fossil faunal assemblages accumulated by hyenas and hominins
using extant hyenid accumulations. Int. J. Osteoarchaeol. 20 (1), 15–35.
Lahr, M.M., Foley, R., 2003. Demography, dispersal and human evolution in the last
glacial period. In: van Andel, T.H., Davies, W. (Eds.), Neanderthals and Modern
Humans in the European Landscape during the Last Glaciation: Archaeological
Results of the Stage 3 Project. McDonald Institute Monographs, Cambridge, pp.
241–256.
Lambeck, K., Yokoyang, Y., Purcell, T., 2002. Into and out of the Last Glacial
Maximum: sea-level change during oxygen isotope stage 3 and 2. Quatern. Sci.
Rev. 21, 343–360.
Langlais, M., Costamagno, S., Laroulandie, V., Pétillon, J.-M., Discamps, E., Mallye, J.-
B., Cochard, D., Kuntz, D., 2012. The evolution of Magdalenian societies in
South-West France between 18,000 and 14,000 calBP: changing environments,
changing tool kits. Quatern. Int. 272–3, 138–149.
Laville, H., Rigaud, J.-P., Sackett, J., 1980. Rock Shelters of the Perigord. Geological
Stratigraphy and Archaeological Succession. Academic Press, New York (NY).
Layton, R., O’Hara, S., 2010. Human social evolution: a comparison of hunter–
gatherer and chimpanzee social organisation. In: Dunbar, R., Gamble, C.,
Gowlett, J. (Eds.), Social Brain, Distributed Mind. Proceedings of the British
Academy 158, pp. 83–113.
Lenoble, A., Bertran, P., Lacrampe, F., 2008. Solifluction-induced modifications of
archaeological levels: simulation based on experimental data from a modern
periglacial slope and application to French Palaeolithic sites. J. Archaeol. Sci. 35,
99–110.
Lyman, R.L., 1979. Available meat from faunal remains: a consideration of
techniques. Am. Antiq. 44 (3), 536–546.
Marlowe, F.W., 2005. Hunter–gatherers and human evolution. Evol. Anthropol. 14,
54–67.
Martínez, G., Flensborg, G., Bayala, P.D., 2013. Chronology and human settlement in
northeastern Patagonia (Argentina): patterns of site destruction, intensity of
archaeological signal, and population dynamics. Quatern. Int. 301, 123–134.
Meeks, S.C., Anderson, D.G., 2012. Evaluating the effect of the Younger Dryas on
human population histories in the Southeastern United States. In: Eren, M.I.
(Ed.), Hunter–Gatherer Behaviour. Human Response during the Younger Dryas.
Left Coast Press, Walnut Creek (CA), pp. 111–138.
Meignen, L., Bar-Yosef, O., Speth, J.D., Stiner, M.C., 2006. Middle Palaeolithic
settlement patterns in the Levant. In: Hovers, E., Kuhn, S.L. (Eds.), Transitions
before the Transition: Evolution and Stability in the Middle Palaeolithic and
Middle Stone Age. Springer, New York (NY), pp. 149–169.
Mellars, P.A., 1973. The character of the Middle-Upper Palaeolithic transition in
Southwestern France. In: Renfrew, C. (Ed.), The Explanation of Culture Change:
Models in Prehistory. Duckworth, London, pp. 255–276.
Mellars, P.A., 1985. The ecological basis of social complexity in the Upper
Palaeolithic of Southwestern France. In: Price, T.D., Brown, J.A. (Eds.),
Prehistoric Hunter–Gatherers: The Emergence of Cultural Complexity.
Academic Press, London, pp. 271–297.
Mellars, P.A., 1996. The Neanderthal Legacy. An Archaeological Perspective from
Western Europe. Princeton University Press, Princeton (NJ).
Mellars, P.A., 2004. Reindeer specialization in the early Upper Palaeolithic: the
evidence from south west France. J. Archaeol. Sci. 31 (5), 613–617.
Mellars, P., French, J.C., 2011. Tenfold population increase in Western Europe at the
Neanderthal-to-modern human transition. Science 333, 623–627.
Mellars, P., French, J.C., 2013. Population changes across the Neanderthal-to-
modern-human transition in western France: a reply to Dogandzˇic´ and
McPherron (2013). J. Hum. Evol. 65 (3), 330–333.
Michel, A., 2010. L’Aurignacien Récent (Post-Ancien) dans le Sud-Ouest de la France:
Variabilité des Productions Lithiques. Révision Taphonomique et Techno-
Économique des sites de Caminade-Est, Abri Pataud, Roc-de-Combe, Le
Flageolet I, La Ferrassie et Combemenue. Unpublished PhD Thesis, University
of Bordeaux 1.
Morala, A., 1984. Périgordien et Aurignacien en Haut-Agenais. Etude d’Ensembles
Lithiques. Ecole des Hautes Etudes en Sciences Sociales, Toulouse.
Morala, A., Turq, A., 1990. Les strategies d’exploitation du milieu mineral, du Riss à
l’Holocene, en Haut-Agenais (Sud-Ouest de la France). In: Séronie-Vivien, M.R.,
Lenoir, M. (Eds.), Le Silex de sa Genèse à l’Outil (Tome II). Cahiers du
Quaternaire 17. CNRS, Paris, pp. 385–390.
Morin, E., 2004. Late Pleistocene Population Interaction in Western Europe and
Modern Human Origins: New Insights Based on the Faunal Remains from Saint-
Césaire, Southwestern France. Unpublished PhD Thesis, University of Michigan.
Morin, E., 2008. Evidence for declines in human population densities during the
early Upper Palaeolithic in Western Europe. Proc. Natl. Acad. Sci. USA 105 (1),
48–53.
Munoz, S.E., Gajewski, K., Peros, M.C., 2010. Synchronous environmental and
cultural change in the prehistory of the northeastern United States. Proc. Natl.
Acad. Sci. USA 107 (51), 22008–22013.
Naughton, F., Sánchez-Goñi, M.F., Desprat, S., Turon, J.-L., Duprat, J., Malaiz´e, B., Joly,
C., Cortijo, E., Drago, T., Freitas, M.C., 2007. Present-day and past (last 25 000
years) marine pollen signal in western Iberia. Mar. Micropaleontol. 62, 91–114.
Naughton, F., Sánchez-Goñi, M.F., Kageyama, M., Bard, E., Duprat, J., Cortijo, E.,
Desprat, S., Malaiz´e, B., Joly, C., Rostek, F., Turon, J.-L., 2009. Wet to dry climatic
trend in north-western Iberia within Heinrich events. Earth Planet. Sci. Lett.
284, 329–342.
Archaeology 39 (2015) 193–209 207Neiman, F.D., 1995. Stylistic variation in evolutionary perspective: inferences from
decorative diversity and interassemblage distance in Illinois Woodland ceramic
assemblages. Am. Antiq. 60 (1), 7–36.
icalNespoulet, R., 2000. Le Gravettien final de l’abri Pataud, (Les Eyzies de Tayac,
Dordogne). Nouvelles données technologiques et typologiques sur l’industrie
lithique provenant du niveau 3. L’Anthropologie 104 (1), 63–120.
Nespoulet, R., 2008. L’ Gravettien de l’abri Pataud. Bilan et perspectives. Paléo 20,
373–380.
Nowell, A., White, M., 2010. Growing up in the Middle Pleistocene: life history
strategies and their relationship to Acheulian industries. In: Nowell, A.,
Davidson, I. (Eds.), Stone Tools and the Evolution of Human Cognition.
University of Colorado Press, Colorado, pp. 67–82.
Pedersen, J., Woehle, E., 1991. The complexities of residential organization among
the Efe (Mbuti) and the Bamgombi (Baka): a critical view of the notion of flux in
hunter–gatherer societies. In: Ingold, T., Riches, D., Woodburn, J. (Eds.), Hunters
and Gatherers: History, Evolution and Social Change. Berg, Oxford, pp. 75–91.
Pennington, R., 2001. Hunter–gatherer demography. In: Panter-Brick, C., Layton,
R.H., Rowley-Conwy, P. (Eds.), Hunter–Gatherers: An Interdisciplinary
Perspective. Cambridge University Press, Cambridge, pp. 170–204.
Petersen, W., 1975. A demographer’s view of prehistoric demography (and
comments and replies). Curr. Anthropol. 16 (2), 227–245.
Petraglia, M., Clarkson, C., Boivin, N., Haslam, M., Korisettar, R., Chaubey, G.,
Ditchfield, P., Fuller, D., James, H., Jones, S., Kivisild, T., Koshy, J., Lahr, M.M.,
Metspalu, M., Roberts, R., Arnold, L., 2009. Population increase and
environmental deterioration correspond with microlithic innovations in South
Asia ca. 35,000 years ago. Proc. Natl. Acad. Sci. USA 106 (30), 12261–12266.
Pickering, T.R., 2002. Reconsideration of criteria for differentiating faunal
assemblages accumulated by hyenas and hominids. Int. J. Osteoarchaeol. 12,
127–241.
Porcˇicˇ, M., 2011. An exercise in archaeological demography: estimating the
population size of Late Neolithic settlements in the Central Balkans. Doc.
Praehist. XXXVIII, 323–332.
Powell, A., Shennan, S., Thomas, M.G., 2009. Late Pleistocene demography and the
appearance of modern human behaviour. Science 324, 1298–1301.
Powell, A., Shennan, S.J., Thomas, M.G., 2010. Demography and variation in the
accumulation of culturally inherited skills. In: O’Brien, M.J., Shennan, S.J. (Eds.),
Innovation in Cultural Systems: Contributions from Evolutionary Anthropology.
MIT Press, Cambridge (MA), pp. 137–160.
Premo, L.S., 2012. Local extinctions, connectedness, and cultural evolution in
structured populations. Adv. Complex Syst. 15 (1 & 2), 1150002. http://
dx.doi.org/10.1142/S0219525911003268.
Premo, L.S., Kuhn, S.L., 2010. Modeling effects of local extinctions on culture change
and diversity in the Paleolithic. PLoS ONE 5 (12), e15582. http://dx.doi.org/
10.1371/journal.pone.0015582.
Read, D., 2006. Tasmanian knowledge and skill: maladaptive imitation or adequate
technology? Am. Antiq. 71 (1), 164–184.
Read, D., 2012. Population Size Does Not Predict Artifact Complexity: Analysis of
Data from Tasmania, Arctic Hunter–Gatherers, and Oceania Fishing Groups.
Human Complex Systems, UCLA, <http://escholarship.org/uc/item/61n4303q>.
Read, D.W., LeBlanc, S.A., 2003. Population growth, carrying capacity and conflict.
Curr. Anthropol. 44 (1), 59–85.
Reitz, E.J., Wing, E.S., 2008. Zooarchaeology, second ed. Cambridge University Press,
Cambridge.
Richards, M.P., 2009. Stable isotope evidence from European Upper Palaeolithic
diets. In: Hublin, J.-J., Richards, M.P. (Eds.), The Evolution of Hominin Diets:
Integrating Approaches to the Study of Palaeolithic Subsistence. Springer,
Dordrecht, pp. 251–257.
Richerson, P.J., Boyd, R., Bettinger, R.L., 2009. Cultural innovation and demographic
change. Hum. Biol. 81 (2–3), 211–235.
Rick, J.W., 1987. Dates as data: an examination of the Peruvian preceramic
radiocarbon record. Am. Antiq. 52 (1), 55–73.
Riede, F., Bentley, R.A., 2008. Increasing the relevance of mathematical model
approaches to demographic history. Qual. Quant. 42, 275–281.
Rigaud, J.-P., Simek, J.F., 1987. ‘‘Arms too short to box with God’’. Problems and
prospects for Paleolithic prehistory in Dordogne, France. In: Soffer, O. (Ed.), The
Pleistocene Old World: Regional Perspectives. Plenum Press, New York (NY), pp.
47–61.
Roebroeks, W., Kamermans, H., Mol, J., Turq, A., van Kolfschoten, T., 2011. Watching
the river flow: a small-scale survey of the floodplain deposits in the Vézère
valley, between Le Moustier and Les Eyzies (Dordogne, France). Analecta
Praehist. Leidensia 41, 1–40.
Ross, A., 1985. Archaeological evidence for population change in the middle to late
Holocene in southeastern Australia. Archaeol. Ocean. 20, 81–89.
Sackett, J.R., 1968. Method and theory of Upper Paleolithic archaeology in
Southwestern France. In: Binford, L.R., Binford, S.R. (Eds.), New Perspectives in
Archaeology. Aldine, Chicago, pp. 61–83.
Sackett, J.R., 1981. From de Mortillet to Bordes: a century of French Palaeolithic
research. In: Daniel, G. (Ed.), Towards a History of Archaeology. Thames and
Hudson, London, pp. 85–99.
Sackett, J.R., 1991. Straight Archaeology French style: the phylogenetic paradigm in
historic perspective. In: Clark, G.A. (Ed.), Perspectives on the Past: Theoretical
Biases in Mediterranean Hunter–Gatherer Research. University of Pennsylvania
Press, Philadelphia, pp. 109–139.
Sánchez-Goñi, M.F., Landais, A., Fletcher, W.J., Naughton, F., Desprat, S., Duprat, J.,
2008. Contrasting impacts of Dansgaard-Oeschger events over a western
European latitudinal transect modulated by orbital parameters. Quatern. Sci.
208 J.C. French / Journal of AnthropologRev. 27, 1136–1151.
Schacht, R.M., 1981. Estimating past population trends. Annu. Rev. Anthropol. 10,
119–140.Schacht, R.M., 1984. The contemporaneity problem. Am. Antiq. 49 (4), 678–695.
Scheinfeldt, L.B., Soi, S., Tishkoff, S.A., 2010. Working toward a synthesis of
archaeological, linguistic and genetic data for inferring African population
history. Proc. Natl. Acad. Sci. USA 107 (2), 8931–8938.
Schmidt, I., Bradtmöler, M., Kehl, M., Pastoors, A., Yafelmaier, Y., Weninger, B.,
Weniger, G.-C., 2012. Rapid climate change and variability of settlement
patterns in Iberia during the Late Pleistocene. Quatern. Int. 274, 179–204.
Shennan, S., 2000. Population, culture history and the dynamics of culture change.
Curr. Anthropol. 41 (5), 811–835.
Shennan, S., 2001. Demography and cultural innovation: a model and its
implications for the emergence of modern human culture. Camb. Archaeol. J.
11 (1), 5–16.
Shennan, S., 2002. Genes, Memes and Human History. Darwinian Archaeology and
Cultural Evolution. Thames and Hudson, London.
Shennan, S., 2009. Evolutionary demography and the population history of the
European early Neolithic. Hum. Biol. 81 (2–3), 339–355.
Shennan, S., 2013. Demographic continuities and discontinuities in Neolithic
Europe: evidence, methods and implications. J. Archaeol. Method Theory 20
(2), 300–311.
Shennan, S., Edinborough, K., 2007. Prehistoric population history: from the late
Glacial to the late Neolithic in Central and Northern Europe. J. Archaeol. Sci. 34,
1339–1345.
Shennan, S., Steele, J., 1999. Cultural learning in hominids: a behavioural ecological
approach. In: Box, H., Gibson, K. (Eds.), Mammalian Social Learning. Symposia of
the Zoological Society of London 70. Cambridge University Press, Cambridge,
pp. 367–388.
Smith, P.E.L., 1966. Le Solutréen en France. Delmas, Bordeaux.
Sollier, M.-C., Mallye, J.-B., 2012. Hominid subsistence strategies in the South-West
of France: a new look at the early Upper Palaeolithic faunal material from Roc-
de-Combe (Lot, France). Quatern. Int. 252, 99–108.
Sørensen, B., 2011. Demography and the extinction of European Neanderthals. J.
Anthropol. Archaeol. 30, 17–29.
Stajich, J.E., Hahn, M.W., 2005. Disentangling the effects of demography and
selection in human history. Mol. Biol. Hum. Evol. 22 (1), 63–73.
Steele, J., Shennan, S., 2009. Introduction: demography and cultural macro
evolution. Hum. Biol. 81 (2–3), 105–119.
Stiner, M.C., 2001. Thirty years on the ‘‘Broad Spectrum Revolution’’ and Palaeolithic
demography. Proc. Natl. Acad. Sci. USA 98 (13), 6993–6996.
Stiner, M.C., 2009. Prey choice, site occupation intensity and economic diversity in
the Middle-early Upper Palaeolithic at the Üçag˘izli Caves, Turkey. Before Farm.
2009 (3), 1–20.
Stiner, M.C., Munro, N.D., 2002. Approaches to prehistoric diet breadth,
demography, and prey ranking systems in time and space. J. Archaeol.
Method Theory 9 (2), 181–214.
Stiner, M.C., Munro, N.D., Surovell, T.A., Tchernov, E., Bar-Yosef, O., 1999.
Palaeolithic population growth pulses evidenced by small animal
exploitation. Science 283, 190–194.
Stiner, M.C., Munro, N.D., Surovell, T.A., 2000. The tortoise and the hare: small game
use, the broad spectrum revolution and palaeolithic demography. Curr.
Anthropol. 41 (1), 39–73.
Stiner, M.C., Beaver, J.E., Munro, N.D., Surovell, T.A., 2008. Modeling palaeolithic
predator-prey dynamics and the effects of hunting pressure on prey ‘choice’. In:
Bocquet-Appel, J.-P. (Ed.), Recent Advances in Palaeodemography: Data,
Techniques, Patterns. Springer, Dordrecht, pp. 143–178.
Straus, L.G., 1990. Underground Archaeology: perspectives on caves and
rockshelters. Archaeol. Method Theory 2, 255–304.
Straus, L.G., 2011. Were there human responses to Younger Dryas in Cantabrian
Spain? Quatern. Int. 242, 328–335.
Straus, L.G., Bicho, N., Winegardner, A.C., 2000. The Upper Palaeolithic settlement of
Iberia: first-generation maps. Antiquity 57 (1), 99–115.
Surovell, T.A., 2000. Early Paleoindian women, children, mobility and fertility. Am.
Antiq. 65 (3), 493–508.
Surovell, T.A., Finley, J.B., Smith, G.M., Brantingham, P.J., Kelly, R., 2009. Correcting
temporal frequency distributions for taphonomic bias. J. Archaeol. Sci. 36 (8),
1715–1724.
Svensson, A., Andersen, K.K., Bigler, M., Clausen, H.B., Dahl-Jensen, D., Davies, S.M.,
Johnsen, S.J., Muschleler, R., Parrenin, F., Rasmussen, S.O., Röthlisberger, R.,
Seierstad, I., Steffensen, J.P., Vinther, B.M., 2008. A 60 000 year Greenland
stratigraphic ice core chronology. Clim. Past 4, 47–57.
Tallavaara, M., Seppä, H., 2011. Did the mid-Holocene environmental changes cause
the boom and bust of hunter–gatherer population size in eastern Fennoscandia?
The Holocene 22 (2), 215–225.
Tallavaara, M., Pesonen, P., Oinonen, M., 2010. Prehistoric population history in
eastern Fennoscandia. J. Archaeol. Sci. 37, 251–260.
Tixier, J.-P., 2009. Histoire Géologique de Vue Préhistoriques Classiques du Périgord:
Une Vision Actualisée. CNRS, Paris.
Turnbull, C.M., 1968. The importance of flux in two hunting societies. In: Lee, R.B.,
deVore, I. (Eds.), Man the Hunter. Aldine, New York (NY), pp. 132–137.
Turnbull, C.M., 1972. Demography of small scale societies. In: Harrison, G.A., Boyce,
A.J. (Eds.), The Structure of Human Populations. Clarendon Press, Oxford, pp.
283–312.
van Andel, T.H., Davies, W., Weniger, B., 2003. The human presence in Europe during
the last glacial period 1: human migrations and the changing climate. In: van
Archaeology 39 (2015) 193–209Andel, T.H., Davies,W. (Eds.), Neanderthals andModernHumans in the European
Landscape during the Last Glaciation: Archaeological Results of the Stage 3
Project. McDonald Institute Monographs, Cambridge, pp. 31–57.
Varian, M.D., Ortman, S.G., 2005. Accumulations research in the Southwest United
States: middle-range theory for big-picture problems. World Archaeol. 37 (1),
132–155.
Wainer, K., Genty, D., Blamart, D., Hoffman, D., Couchoud, I., 2009. A new stage 3
millennial climatic variability record from a SW France speleothem.
Palaeogeogr. Palaeoclimatol. Palaeoecol. 271, 130–139.
Whallon, R., 2006. Social networks and information: non-‘‘utilitarian’’ mobility
among hunter–gatherers. J. Anthropol. Archaeol. 25, 259–270.
White, T.E., 1953. A method of calculating the dietary percentage of various food
animals utilised by Aboriginal peoples. Am. Antiq. 18, 393–399.
White, R., 1985. Upper Palaeolithic Land-use in the Périgord. A Topographic
Approach to Subsistence and Settlement. Archeopress, Oxford.
White, R.W., 2002. The historic and legal context of foreign acquisitions of
Paleolithic artifacts from the Périgord: 1900–41. In: Straus, L. (Ed.), The Role
of American Archaeologists in the Study of the European Upper Paleolithic. BAR
International Series 1048. Archaeopress, Oxford, pp. 71–83.
Wicks, K., Mithen, S., 2014. The impact of the abrupt 8.2 ka cold event on the
Mesolithic population of western Scotland: a Bayesian chronological analysis
using ‘activity events’ as a population proxy. J. Archaeol. Sci. 45, 250–269.
Williams, A.N., 2012. The use of summed radiocarbon probability distributions in
archaeology: a review of methods. J. Archaeol. Sci. 39, 578–589.
Williams, A.N., 2013. A new population curve for prehistoric Australia. Proc. R. Soc.
B 280, 1–9.
Williams, A.N., Ulm, S., Goodwin, I.D., Smith, M., 2010. Hunter–gatherer response to
late Holocene climatic variability in northern and central Australia. J. Quat. Sci.
25 (6), 831–838.
Wobst, H.M., 1974. Boundary conditions for Palaeolithic social systems: a
simulation approach. Am. Antiq. 39, 147–178.
Wohlfarth, B., Veres, D., Ampel, L., Lacourse, T., Blaauw, M., Preusser, F., Andrieu-
Ponel, V., Kéravis, D., Lallier-Vergès, E., Björk, S., Davies, S.M., de Beaulieu, J.-L.,
Risberg, J., Hormes, A., Kaspar, H.U., Possnart, G., Reille, M., Thouvney, N.,
Zander, A., 2008. Rapid ecosystem response to abrupt climate changes during
the last glacial period in Western Europe, 40–16 ka. Geology 36 (5), 407–410.
Woodburn, J., 1968. Stability and flexibility in Hadza residential groups. In: Lee, R.B.,
deVore, I. (Eds.), Man the Hunter. Aldine, New York (NY), pp. 103–110.
Woodburn, J., 1972. Ecology, nomadic movement and the composition of the local
group among hunters and gatherers: an East African example and its
implications. In: Ucko, P.J., Tringham, R., Dimbleby, G.W. (Eds.), Man,
Settlement and Urbanism. Duckworth, London, pp. 193–206.
Zubrow, E.B.W., 1989. The demographic modelling of Neanderthal extinction. In:
Mellars, P., Stringer, C. (Eds.), The Human Revolution. Edinburgh University
Press, Edinburgh, pp. 212–231.
J.C. French / Journal of Anthropological Archaeology 39 (2015) 193–209 209