Content uploaded by Randall White
Author content
All content in this area was uploaded by Randall White on Jun 21, 2016
Content may be subject to copyright.
White, R. 2007. Systems of Personal Ornamentation in the Early Upper Palaeolithic:
Methodological Challenges and New Observations. In, Mellars, Paul, Boyle, Katie,
Bar-Yosef, Ofer and Stringer, Chris (eds.) Rethinking the Human Revolution: New
Behavioural and Biological Perspectives on the Origin and Dispersal of Modern
Humans. Cambridge, UK, McDonald Institute for Archaeological Research, 287-302.
(McDonald Institute Monographs).
287
Systems of Personal Ornamentation in the Early Upper Palaeolithic
Chapter 24
Systems of Personal Ornamentation
in the Early Upper Palaeolithic:
Methodological Challenges and New Observations
Inconvenient archaeological realities
As appealing as they are, such questions and
overarching propositions are situated far from the
complex reality of the archaeological record. The
promising results of my initial work on early Upper
Palaeolithic personal ornaments led me to undertake
an ambitious analysis of such objects from Aurigna-
cian-age sites across Europe. At the time, my rather
naïve goal was to create a detailed, pan-European
analysis of personal ornaments that would seek
geographic paerning in their form, raw material
and technique of manufacture. This research was
conducted against the backdrop of the ‘social ge-
ography’ and ‘boundary maintenance’ ideas then in
circulation (Conkey 1978).
To this end, I studied, in place, almost all of the
then-known Aurignacian-age ornaments from Europe
(White 1992; 1993b).3 While I recognized immediately
that there were some tantalizing geographic dier-
ences in bead form and technique, it became obvious
that this pursuit of regional paerning in Aurignacian-
age ornaments was ill founded, premature and could
only yield spurious results. Numerous methodologi-
cally-derived limitations to the known archaeological
record prevent, for the moment, robust, broad-scale
statements concerning regional and inter-regional
paerning. These include:
1. poor and/or inconsistent excavation quality;
2. lack of integration of ornament-recovery into ex-
cavation design;
3. secondary priority given to ornaments in publica-
tion and museum curation;
4. analytical and typological practices inappropriate
to the task.
These issues will be presented in more detail below
underscoring the unfortunate consequences of found-
ing broad interpretation on a weak foundation.
Randall White
Personal ornaments and the ‘human revolution’
When, in the course of the 1987 Human Revolution
conference, I presented my preliminary work on Early
Aurignacian personal ornaments (White 1989a,b),
I was something of a lone voice.1 Of the thirty-ve
contributors to the so-called ‘big red book’, I was
the only one to treat body ornaments as evolutionar-
ily important. It is gratifying to see that much has
changed over the past twenty years to the point that,
in the papers of the present conference, the presence
or absence of personal ornamentation has become
one of the key measures of modernity in the culture-
evolutionary record.
There is now broad consensus that, among
contemporary humans, material systems of personal
ornamentation enable the construction of a diversity
of social and personal identities (White 1999; 2000). In
large part, they do so by making reference to group
histories, values and beliefs via profound metaphori-
cal linkages between chosen materials and created
forms on the one hand, and various agreed-upon
social identities on the other. The initial invention of
such material systems2 of personal ornamentation in
the prehistoric past doubtless accompanied the rst
internally dierentiated human societies. The meta-
phorical references alone must have been a powerful
inuence on political and social relations.
My own work on Aurignacian personal orna-
mentation seeks to understand not merely the nal,
formal qualities of ornamental objects, but to tease out
the very processes (technological, aesthetic, symbolic)
by which socially meaningful forms are constructed. It
is my position that regional ornamental systems can-
not be dened or understood without addressing mat-
ters such as raw material choice, technical production,
display modes and archaeological/chronostratigraphic
context: a tall order indeed.
288
Chapter 24
A word on Aurignacian terminology
In the discussion that follows, I employ the following
terms proposed by François Bon with respect to Au-
rignacian cultural chronology and inter-assemblage
variability (Bon 2002):
• Aurignacien archaïque (the English term Archaic
Aurignacian is employed here). This term is syn-
onymous with what some specialists call Protoau-
rignacien;
• Aurignacien ancien (the English term Early Aurigna-
cian is employed here). This term is synonymous
with what some specialists call Aurignacien typ-
ique;
• Aurignacien récent (the English term Late Aurigna-
cian is used here). This term is synonymous with
what some specialists call Aurignacien évolué.
Caution is advised here. The degree to which there
is chronological overlap between these faciès remains
unclear. Some specialists see the rst two as dierent
independent traditions rather than stages in a single
chronological sequence. These terms are used here
to refer to assemblage types that most Aurignacian
researchers recognize. They are only given chrono-
logical meaning in the discussion below when they
are found ordered in the same stratigraphic section,
such as Isturitz and/or when they are radiometrically
dated in a specic context, such as at Castanet and
Brassempouy.
New questions require new data from new
One of the most fundamental issues restricting a
more thoroughgoing understanding of Palaeolithic
ornamental dynamics is the exceptionally poor quality
of older excavations and publications. Until recently,
virtually no sites had been excavated with a view to
the specic problems of personal ornaments. Into
the 1980s there were still Aurignacian excavations in
southwest France that used only dry-sieving through
5-mm mesh. According to my experience and experi-
ments, such methods result in recovery of less than
10 per cent of the Aurignacian ornaments present,
even worse when excavators are not sensitized to the
possible presence of such objects. Unfortunately, such
sensitization was seldom accomplished as post-World
War II Aurignacian researchers seem to have largely
forgoen the rich ornamental sites from the turn-of-
the-century.4 In sum, student excavators have rarely
been trained to recognize ornaments and ornament
fragments, not to mention production debris.
A more obvious problem with older excavated
sites is the absence of stratigraphic and spatial context,
both critical to an understanding of early systems of
personal ornamentation. In essence, there was (and is)
essentially no credible internal chronology for the c.
10,000-year-long Aurignacian techno-complex. I found
myself in the position, in comparing undated ornamen-
tal assemblages aributed to the Early Aurignacian, of
having to assume a single cultural time plane. My goals
were ‘palaeoethnographic’ but the record was spread
over a 10,000 year-long ‘ethnographic present’.
In France, even collections from ‘well-excavated’
sites were dispersed to numerous museums and pri-
vate collections (e.g. White & Breitborde 1992) before
they could be published or inventoried, making their
study through the published literature more than
complicated.
A historical lack of interest in personal orna-
ments oen results in their being mixed with faunal
portions of museum collections, making it necessary
for researchers to be sceptical of what museum cura-
tors place in front of them for study. Moreover, the
published record can be a mere shadow of what is in
museum drawers and is dominated by beaux objets.
When I undertook the analysis of the beads from Abri
Castanet in the 1980s, nearly half of the ornaments
were mixed in with the faunal collection at the Musée
national de Préhistoire in Les Eyzies. In sum, direct
analysis of existing collections is mandatory, rather
than the construction of databases founded upon the
published record.
With a few exceptions then, publications of older
excavations are incapable of responding to modern
research questions. Readers will perhaps understand
then, why I moved my research in a dierent direc-
tion, that of excavating and analyzing meticulously
recovered, well-dated assemblages of ornaments; and
why I remain intensely sceptical of recent broad-scale
syntheses of Aurignacian ornaments in Europe.
New research designs
Recent Aurignacian excavations at the sites of
Brassempouy, Castanet and Isturitz (Fig. 24.1), il-
lustrate the potential of personal ornaments for an
understanding of early Upper Palaeolithic society
and systems of meaning. Through my participation in
the long-term excavation of these previously known
sites, the meticulous recovery of personal ornaments
has been integrated into the overall research design.
Excavators have been trained in advance to recognize
personal ornaments and production debris and all
sediments have been wet-sieved through 1.5-mm
mesh.5 The booms of the sieves have then been dried
and hand-sorted to enable the recovery of small orna-
ments, ornament fragments and production debris.
289
Systems of Personal Ornamentation in the Early Upper Palaeolithic
At Castanet, systematic samples of raw sedi-
mentary matrix have been analyzed microscopically,
resulting in the recovery and identication of micro-
vestiges of ornament production. These include such
items as ivory shavings and talc dust from the nish-
ing of soapstone beads by abrasion and polishing.
The quantitative and qualitative analysis of these
painstakingly recovered assemblages is allowing our
research team to investigate:
• spatial and stratigraphic (as well as radiometric)
provenance within sites;
• the criteria underlying the choice of raw materials
for ornaments;
• microscopic and experimental evidence of perfora-
tion, suspension and aachment techniques, and
of stigmata of ornament use;
• on-site production versus o-site production;
• operational chains for the production of orna-
ments;
• geographic/geologic provenience of ornamental
raw materials;
• metric and formal standardization in ornament
manufacture;
• intra-regional similarities and dierences in all of
the above.
Some first results: Brassempouy, Castanet and
Isturitz
I will present here some preliminary results from re-
search currently in course from excavations at three
Aurignacian sites in southwest France (Fig. 24.1):
• Groe des Hyènes at Brassempouy (Landes);
• Abri Castanet (Dordogne);
• Groe d’Isturitz (Pyrénées-Atlantiques).
The Groe des Hyènes at Brassempouy
The well-dated Aurignacian levels of the Groe des
Hyènes at Brassempouy have produced an impor-
tant series of pierced teeth (fox, wolf, deer), shells,
facsimiles of cervid canines (one in ivory, the other in
stone). Four human teeth, prepared by perforation or
rainurage have also been found.6 Apart from the cervid
canine facsimiles, two tabular beads, two fragments of
possible diada, three conjoining fragments of a pos-
sible ‘bracelet’ and rare perforated mineral fragments,
nearly all the personal ornaments from the Groe des
Hyènes are basket-shaped beads.
A series of 14C dates has been obtained for the
Aurignacian sequence at the Groe des Hyènes (Hen-
ry-Gambier et al. 2004; Bon 2000; 2002; Fontugne 2007)
(Table 24.1). The dated archaeological levels indicated
in Table 24.1 extend over varying surface areas, with
the total area excavated being roughly 25 m2.
Abri Castanet (and probably by extension (?) identical
assemblages from the contiguous site of Abri Blanchard)
New excavations at Abri Castanet, rst begun in 1994
in collaboration between Jacques Pelegrin, and myself
continue today under my direction. The stratigraphy
is proving to be much more complicated than that
proposed by Denis Peyrony (1935), who recognized
two Aurignacian levels.
In our terms, a primary archaeological level situ-
ated on bedrock is aributed to a variant of the Early
Aurignacian very poor in burins and perforators. This
level (Niveau archéologique de base or NAB) is composed
of a number of sedimentary subdivisions that are
still being sorted out by our research.7 The total area
excavated for the moment is approximately 10 m2.
Figure 24.1. The Aquitaine basin with the three key sites
studied here: 1) Castanet and the other Castel-Merle sites;
2) Brassempouy; 3) Isturitz.
Table 24.1. Radiocarbon dates for the Aurignacian sequence at the
Groe des Hyènes.
Lab no. Stratigraphic layer Date cal.
GifA 9658 Ens.1 (end of Aurig. seq.) 30,600± 200
GifA 8174 Couche 2A-2C 32,190±620
GifA 8568 Couche 2A-2C 31,820±550
GifA 8569 Couche 2A-2C 31,690±780
GifA 9031 Couche 2D/2F 30,100±400
GifA 8570 Couche 2E 17,970± 150
GifA 9032 Couche 2E 26,870± 500
Gif/LSM 11035 Couche 2E 31,960±160
GifA 98105 Couche 2DE 32,410±370
Gif/LSM 11034 Couche 2DE 33,600±240
290
Chapter 24
A rst series of four AMS dates (Valladas et al. 2007)
yielded values as shown in Table 24.2. More dates are
currently being processed in collaboration with Paul
Mellars and the Oxford accelerator.
Groe d’Isturitz
Recent excavations by Christian Normand (2002;
2005; 2006) have revealed multiple Aurignacian strati-
graphic units, beginning with Archaic Aurignacian
(levels 4d and 4c) on the boom and ending with Early
Aurignacian (levels 4b through 3) on top. Level 4d is
bracketed top and boom by two dates (Normand
et al. 2007): 34,630±560 (Gif-98237) and 36,550±610
(Gif-98238). The total area excavated to date is less
than 10 m2.
The excavations at Brassempouy have now been
terminated and the massive publication will appear
within the next year or two (Henry-Gambier & Bon
2007). Long-term excavations at Castanet and Isturitz
are ongoing. As a result, I will use the more deni-
tive data from Brassempouy and, where appropriate,
make reference to the provisional results from the as
yet unnished excavations at Castanet and Isturitz.
In these laer cases, the small area excavated must be
taken into consideration in evaluating the signicance
of the provisional results presented here.
Since it is impossible in the space available to
provide detailed analyses, I focus here on a certain
number of provocative trends and results. For the
complete analyses of ornamental assemblages from
these sites, the reader is referred to the original pub-
lications now in course (White 2007a,b,c).
ornaments
Traditional morpho-typological approaches conven-
iently assumed that all objects prepared for suspen-
sion were interpretable as body ornaments. However,
a quick glance at any ethnographic record will show
scores of non-body-ornament objects adapted for
suspension/aachment on baskets, bags, blankets,
dwellings, etc. This fact requires rigorous experi-
mental, analytical and contextual data to distinguish
personal ornaments from others.... especially when
keeping in mind the absence of burial associations
for the European Aurignacian. While we have begun
the systematic experimental study of use-wear on
Aurignacian ornaments, we are far from reducing
the current ambiguity about modes of wearing and
suspension.
Morpho-typological inventories of personal
ornaments are inadequate in the same way that ty-
pological approaches to the study of lithic artefacts
long prevented a dynamic and contextualized un-
derstanding of inter-assemblage variation. Whether
applied to stone tools or to personal ornaments,
such typological approaches confuse unfinished
production stages with nished objects. Moreover,
the form of the presumed ornament is predominant
in the denition of ornament types, to the exclusion
of other material aributes such as surface texture
(e.g. mae, lustrous), use-wear, breakage paerns
and raw material choice (e.g. ivory, stone, shell). As a
result, culturally-grounded ornamental systems and
related social distinctions lay hidden within a largely
typological structure of description (e.g. Vanhaeren
2002; Vanhaeren & d’Errico 2006).
In the same way that Bordesian types are now
seen to have been inappropriate as units of analysis for
Binford’s factor analysis of Mousterian assemblages
(Binford & Binford 1966; Binford 1973), ornament
types are inadequate for elaborate statistical treat-
ments that seek to extricate some original cultural
reality, such as regional ethnic groups. As emphasized
above, ornament typologies are ever more disconcert-
ing when created and applied to poorly excavated
collections known only from the literature.
Table 24.2. AMS dates from Abri Castanet.
Lab no. Stratigraphic layer Date cal.
GifA 99165 Stratigraphic Zone 114 31,430±390
GifA 99179 Stratigraphic Zone 122 32,310±520
GifA 99180 Stratigraphic Zone 122 32,950±520
GifA 99166 Stratigraphic Zone 131 34,320±520
Figure 24.2. Examples of the technique of basal
circumincizing or rainurage. Le: Aurignacian from
Vogelherd (courtesy of the Institut für Urgeschichte,
Tübingen); Centre: Châtelperronian, Groe du Renne,
Arcy; Right: Aurignacian, Abri Cellier.
291
Systems of Personal Ornamentation in the Early Upper Palaeolithic
Historically, when they were
even studied, techniques of ornament
preparation have tended to be viewed
as chrono-cultural markers. For ex-
ample, the so-called Arcy technique
of rainurage (basal circumincizing)
for the suspension of objects was
widely thought to be a marker for the
Châtelperronian culture, even though
modern excavations and analyses
show it to be a widespread part of the
Aurignacian repertoire of suspension
techniques (Fig. 24.2).
For mammal teeth, including
humans, perforation techniques are
highly variable, but this variability
does not paern stratigraphically.
Aurignacian artisans demonstrated
signicant technical exibility (Fig.
24.3). For example, four techniques
are present in the Groe des Hyènes,
ensemble 2 sample with respect to the
modication of objects for suspen-
sion: Perforation by bifacial goug-
ing; Perforation by demi-rotation;
Perforation by pressure or indirect
percussion; Perforation by basal cir-
cumincision (Rainurage).
Moreover, in the Early Aurig-
nacian of the Groe des Hyènes at
Brassempouy (White 2001), the ex-
istence of several objects perforated
by pressure or indirect percussion
(the technique that I have identied
at Quinçay and for some teeth at the
Groe du Renne at Arcy-sur-Cure),
contradicts the proposition that this
technique was exclusively Châtelper-
ronian (Granger & Levêque 1997).
Symbolic objects in the Archaic Aurignacian
Much aention has been focused recently on the exist-
ence of an Aurignacian industry that precedes the Early
Aurignacian. The alleged absence of symbolic objects
in this Archaic Aurignacian has been used by some
(d’Errico 2003; Zilhão & d’Errico 2003) to argue that
Châtelperronian personal ornaments were independ-
ently invented by Neanderthals rather than being the
product of acculturation with incoming Aurignacians
who already possessed them. Of course this argu-
ment relies on the as yet undemonstrated notion that
Châtelperronian ornaments are older than Aurignacian
ones in Western Europe. The severe dating problems
with Arcy-sur-Cure need to be kept in mind as well as
the fact that the pierced teeth from Quinçay come from
the upper half of a long Châtelperronian sequence.
Of course, the above argument would also
require the absence of ornaments from the Archaic
Aurignacian. A date of 37,200±1500 (GifA-97185) from
ornament-bearing level G at Caminade Est (Bordes
2000) does not comfort such an idea, nor do pre-40 kya
dates for symbol-bearing levels in eastern Europe and
Siberia (Kostienki 14, Kostienki 17, Kara-Bom).
Even less comforting to the Zilhão & d’Errico
position is the recovery of numerous ornaments from
an Archaic Aurignacian context in new excavations at
Figure 24.3. Groe des Hyènes, Brassempouy. Techniques used to transform
mammalian teeth for suspension, by stratigraphic unit.
Figure 24.4. Groe des Hyènes, Brassempouy. Two examples of fox canines
perforated by pressure or indirect percussion. The aack point in both cases is
clearly evident at the hole margin.
292
Chapter 24
Isturitz (Normand et al. 2006; White 2007d). The rich
and diverse ornament assemblage, from levels 4c6 and
4d1 (level 4d is associated with two dates: 34,630±560
(Gif-98237) and 36,550±610 (Gif-98238).
Included in the level 4d Archaic Aurignacian
assemblage are een perforated shells of Liorina
obtusata and a complex pendant of calcite (Fig. 24.5).
This constitutes well-documented proof of Archaic
Aurignacian ornament production and use prior to
any credible C14 dates for Châtelperronian ornament-
bearing assemblages. Level 4c, immediately overlying
4d and with many of the same archaic characteristics
of the level 4d assemblage, remains undated for the
moment. Its base has yielded the oldest known evi-
dence for amber jewellery in the world in the
form of amber pendants (Fig. 24.6),8 produc-
tion debris and raw chunks of amber.
Species choice in animal teeth
I wish to make three points concerning
species choice for the purpose of ornamen-
tation. Firstly, the number of teeth altered
for suspension is but an innitesimal pro-
portion of the mammalian teeth available.
Secondly, the species whose teeth were
chosen for the fabrication of ornaments are
not those that dominate the faunal assem-
blages (Fig. 24.7).
Finally, there is fascinating vari-
ability through time in species choice for
ornamentation. For several years now, I
have remarked signicant dierences in
the ornament-species frequencies between
Brassempouy and Isturitz, these sites being
only 60 km apart as the crow ies. I have en-
tertained two hypotheses to explain these dierences,
which are manifest in the overwhelming dominance
of bovid teeth at Isturitz and the absence of bovids in
the ornamental tooth assemblages from Brassempouy.
It seemed to me that either this variability was ran-
dom, that it signalled the boundary between dierent
regional groups, or that it was based in previously
unrecognized chronological changes in species choice
within the Aurignacian.
Figures 24.8 and 24.9 illustrate the result of not
controlling for chronology by studying bulk Aurigna-
cian samples that cover long periods of Aurignacian
time. When all of the Aurignacian ornaments from
Isturitz are grouped, they stand out in sharp contrast
Figure 24.5. Calcite pendant from the Archaic Aurignacian level 4d1
at Isturitz.
Figure 24.6. Amber pendant from the Archaic
Aurignacian level 4c6 at Isturitz.
Figure 24.7. Ensemble 2, Groe des Hyènes, Brassempouy, species
of teeth modied for suspension in relation to species of non-modied
teeth in the faunal assemblage (identications Letourneux 2003).
293
Systems of Personal Ornamentation in the Early Upper Palaeolithic
to those from the chronologically clustered
ensemble 2 levels from Brassempouy. How-
ever, when we separate out the ornaments
from levels 3 and 4a on one hand and levels
4b through 4d on the other we see an abrupt
chronological change in species choice. This
change from the overwhelming use of bovid
teeth to the predominance of other species
(fox, red deer, humans) has no relationship
to changes in available fauna, as bovids are
as abundant in the later levels at Brassem-
pouy and Isturitz as they are in the older,
Archaic Aurignacian levels at Isturitz. It is
quite astonishing to see the total absence
at sites like Brassempouy and Castanet of
tooth-ornaments of reindeer, horses and
bovids given that these three taxa dominate
in the consumed fauna, including teeth (Le-
tourneux 2003; Costamagno, in Normand
et al. 2006; Castel 2007). Dietary importance
and symbolic signicance seem to be nearly
mutually exclusive.
Not only do we see change through
time in the Aurignacian choice of ornament
species, but we are confronted with the dra-
matic consequences of methodologies that
use assemblages of ornaments recovered
with lile or no chronostratigraphic preci-
sion. This leads me to reject recent claims
for paerned regional variation within the
Aurignacian across Europe. We simply do
not know whether the variation claimed is
geographic or chronological in structure.
On a ner chronological scale, we can
now see that within a single Early Aurigna-
cian sequence such as the Groe des Hyènes
at Brassempouy, raw frequencies of tooth
ornaments as well as species proportions
vary through time (Fig. 24.10). Ignoring
such variation can only lead to spurious
conclusions.
While we are on the subject of changes
in species choice through time, it is worth
noting that Aurignacian human teeth modi-
ed for suspension are, to date, restricted
to Early Aurignacian contexts in the four
Aquitaine basin sites in which they occur:
La Combe; Groe des Hyènes, Brassem-
pouy; Isturitz (Fig. 24.11) and Tarté (White
et al. 2003; Henry-Gambier & White 2006;
Figure 24.8. Inter-site comparison of species of animal teeth
transformed into ornaments between the Groe des Hyènes at
Brassempouy and the Groe d’Isturitz (not distinguishing between
Early Aurignacian and Archaic Aurignacian at the laer)
Figure 24.9. Dierences in the species frequency of animal teeth
transformed into ornaments between Early Aurignacian and Archaic
Aurignacian at the Groe d’Isturitz.
Figure 24.10. Groe des Hyènes,
Brassempouy, teeth modied for suspension by
species and stratigraphic unit.
294
Chapter 24
Henry-Gambier et al. 2004). Some of these teeth show
evidence of deeshing implying purposeful extrac-
tion from esh-bearing bodies. The layers of Les Rois
(Charente) and that of Fontéchevade each delivered
a youthful mandible with cut-marks (Gambier 2000),
perhaps related to this process of tooth extraction for
ornamental purposes.
Insights from the application of a chaîne opératoire
based methodology to objects coming from
modern excavations
Raw materials
The beads and pendants from ensemble 2 at the Groe
des Hyènes, Brassempouy, were made from ivory,
chlorite, talc, calcite, bone, hematite and lignite. These
are the same materials as at Castanet and Isturitz with
some small but important dierences. For example,
proportions of ivory to so-stone beads (Figs. 24.12
& 24.13) are dramatically dierent between Brassem-
pouy and Castanet (as well as the other Castel-Merle
sites of La Souquee and Blanchard). It would seem
obvious to conclude that Brassempouy’s location some
250 km closer to the Pyrenean talc sources results in
much higher percentages of talc beads. This would be
a classic down-the-line distance drop-o paern, with
materials aenuating with distance from source
However, the laer conclusion is contradicted
by the fact that the beads at Isturitz, even closer to the
Pyrenees, are made almost exclusively of ivory (Fig.
24.13). Again, we nd complex variation that dees a
simple, regional model.
On-site versus o-site production
A small amount of fabrication debris and some un-
nished beads and pierced teeth (Figs. 24.14, 24.15 &
24.16) indicate on-site manufacture of at least some
of the basket-shaped beads and pierced teeth at both
Castanet and Brassempouy. However, in spite of 20
basket-shaped beads and fragments in levels 3 and
4a (Early Aurignacian) at Isturitz, only one unnished
bead and no production debris have yet been found
in the area excavated.
The systematic recovery and quantication of
fabrication debris and unnished beads results in a
rather interesting paern when we restrict our analy-
Figure 24.11. Groe d’Isturitz, level 4a (Early
Aurignacian). Human Lower le M2 or M3 perforated
by back-and-forth rotation. The hole, created by a rather
obtuse tool point, is heavily ‘worn’.
Figure 24.12. Proportions of dierent raw materials
used for basket-shaped beads in the Groe des Hyènes,
Brassempouy.
Figure 24.13. Global proportions of raw materials
of basket-shaped beads for three Vézère Valley sites
(Dordogne), the Groe des Hyènes, Brassempouy (Landes)
and the Groe d’Isturitz (Pyrénées-Atlantiques).
295
Systems of Personal Ornamentation in the Early Upper Palaeolithic
sis to Early Aurignacian assemblages where there
has been 100 per cent recovery (Fig. 24.15). The new
excavations at Castanet conrm the longstanding im-
pression of intense bead production, with the presence
of as many bi-products as beads. The Brassempouy
ensemble 2 assemblages show much lower produc-
tion intensity and may be interpreted as contexts in
which some ornament (clothing?) maintenance and
replacement was being undertaken. Traces of wear on
the inside of perforations also imply on-site use/loss/
discard of the beads (Fig. 24.17).
Finally, in the small area recently excavated at
Isturitz, evidence for ornament production is virtu-
ally absent, with one exception: strong traces of the
working of amber ornaments on site, a phenomenon
restricted in time to levels 4b and 4c.
A comparison of the intensity of activities related
to ornament production, among the Groe des Hyènes,
Figure 24.14. Groe des Hyènes, Brassempouy. Two
faces of the root of a wolf canine showing distinct
stigmata (hacking, scraping, gouging) of preparation for
perforation. The fact that this object was abandoned or
lost before perforation occurred implies on-site production
of at least some tooth-ornaments.
Figure 24.15. Ratios of basket-shaped beads to
production débris and unnished beads from modern
excavations at the Groe des Hyènes, Brassempouy, Abri
Castanet and the Groe d’Isturitz.
Figure 24.16. Vestigial canine of an old female red
deer showing a prehistoric break through the neck of
the ornament, implying on-site loss or discard. Surface
preparation by scraping preceded perforation by demi-
rotation or rotation.
Figure 24.17. Le, use keyholing from prolonged
wearing of a basket-shaped bead of chlorite (16x).
Right, polish on the neck of an ivory basket-shaped bead
(40x). Both objects are from the Groe des Hyènes,
Brassempouy.
296
Chapter 24
the Castel-Merle (Dordogne) Aurignacian sites and the
Groe d’Isturitz (Pyrénées-Atlantiques), permits the
beginnings of the construction of a model regarding
the place of ornament fabrication within a regional
system of seasonal movement and site function. This
hypothesis remains extremely tenuous in light of the
fragmentary nature of available relevant data. It is
worth pointing out that available seasonal data indicate
warm season occupation at Brassempouy (Letourneux
2003) and Isturitz (Rendu, in Normand et al. 2006) and
winter occupation at Abri Castanet (Pike-Tay 2007).
in bead production
Fabrication stages for bead production in the Groe des
Hyènes are precisely those found at other southwest
French Aurignacian sites (White 1989a; 1993a,b). In
spite of signicant structural dierences between so
stone and ivory, the fabrication procedures were the
same for objects made from these two raw materials.
Operational chains for basket-shaped beads
So-called basket-shaped beads were rst recognized by
Marcel Castanet at Abri Blanchard in 1909 (Didon 1911;
1912; Delluc & Delluc 1981). This discovery prompted
Castanet, a farmer at Sergeac in the Dordogne, to
wet-sieve all of the sediments from Blanchard (more
than 200 such beads), as well as the proximate sites
of Castanet (Peyrony 1935) and La Souquee (White
1989a). Also recovered from the sieves were hundreds
of unnished beads and production debris, allowing
the reconstruction of a complex and ingenious chaîne
opératoire for bead production. I have described this
operational chain in several publications over the
years and will not repeat that discussion here except
to present a synthetic overview (Fig. 24.18).
In particular, I wish to focus on the fact that this
same operational chain characterizes Early Aurigna-
cian sites over a considerable area of southern France,
extending from Gatzarria (Sáenz de Buruaga 1989;
1991) on the Atlantic coast in the west to Abri Mochi
in Ligurian Italy to the east. This highly redundant
set of procedures seems not to extend southward
across the Spanish frontier and is unknown north of
the Aquitaine basin. Nonetheless, dierent versions
of this approach to Aurignacian bead production are
known from Belgium (Oe 1974) and Germany (Hahn
1972; White 1993a).
Figure 24.18. The ve main production stages for the
manufacture of Aurignacian basket-shaped beads. Those
pictured are from Abri Blanchard, the type-site for such
objects. Stage 1: creation of cylindrical, pencil-like rods
of talc/chlorite or of ivory (Figs. 24.18, 24.20 & 24.25).
Stage 2: circumincision of these rods into segments of 1
to 2 cm long, which are then snapped o the longer rod
(Figs. 24.18, 24.21 & 24.24). Stage 3: bifacial thinning
of one end of the detached cylinders to create a kind of
stem at one end and a bulb at the other Figs. 24.18, 24.22
& 24.26). Stage 4: perforation by bifacial gouging or
rotational drilling at the junction of the stem and the bulb
(Fig. 24.22). Stage 5: reduction of the stage 4 rough-
out by coarse abrasion and eventually by ne polishing
in order to obtain the characteristic basket shape (Figs.
24.18, 24.19 & 24.27).
Figure 24.19. Eight of the ivory basket-shaped beads from the Groe des Hyènes, Brassempouy (stage 5 of the chaîne
opératoire).
297
Systems of Personal Ornamentation in the Early Upper Palaeolithic
Figure 24.20. Groe des Hyènes, Brassempouy, stage 1 (ivory) of the production chain for basket-shaped beads.
Figure 24.21. Groe des Hyènes,
Brassempouy, stage 2 (ivory) of the production
chain for basket-shaped beads.
Figure 24.22. Groe des Hyènes, Brassempouy, stage 3/4 (ivory)
of the production chain for basket-shaped beads. This is one of the
rare objects that are intermediate between the dened stages. Visible
at le and right are stigmata of initial bifacial scraping to achieve
the perforation. Loss or abandonment of the object in the course of
production is implied.
Figure 24.23. Groe des Hyènes, Brassempouy, stage 4
(ivory) of the production chain for basket-shaped beads. Figure 24.24. Groe des Hyènes, Brassempouy, stage 2
(chlorite) of the production chain for basket-shaped beads.
Traces of scraping from the forming of the original pencil-
like cylinder are visible on the surface of this object.
298
Chapter 24
Such beads seem to be restricted to Early Aurig-
nacian contexts. They are known, for example, from
Saint Jean de Verges in the Ariège (Vézian & Vézian
1966) and from Groe XVI in the Dordogne, (Rigaud’s
excavations). It is certain that the recovery of these
minuscule objects (averaging 6 mm in length with
a range from 2 to 12 mm) is heavily conditioned by
recovery techniques, notably ne-mesh water sieving).
As a consequence, several excavations prior to the
modern period probably contained large quantities
of such beads, which were simply not recovered. In
the Passemard and Saint-Périer excavations at Isturitz
(Passemard 1944; de Saint-Périer 1930), a handful of
large examples were recovered from the same levels
that today yield many from the size range mentioned
above.
The high degree of inter-assemblage standardiza-
tion (Fig. 24.28) that emerges from this uniform and
broadly applied set of procedures implies detailed
knowledge shared across wide expanses of space.
Moreover, the uniform size of objects too small to be
visible individually implies that basket-shaped beads
were elements of larger arrangements, either strung
on human bodies or more probably sewn onto articles
of clothing and other decorated objects.
Discussion and conclusions
I have aempted here to provide a few illustrations
of ways in which meticulously conducted research,
aimed at the recovery and analysis of personal orna-
ments, allows us to rmly insert a few bricks in the
edice of archaeological knowledge of Aurignacian
culture. In turn, such knowledge informs contempo-
rary research into the processes involved in the rst
peopling of Europe by anatomically and culturally
modern humans.
The idea, recently re-presented by d’Errico and
Vanhaeren, of regionally distinct congurations of
personal ornaments corresponding to geographically
and linguistically distinct ethnic units, does injustice
to both ethnographic reality (see Wobst 1978) and to
archaeological reconstruction. Such geographic units
are not nearly as clear in the ethnographic record as
one would like. And the data used to argue their ex-
istence in the Aurignacian are inadequate for reasons
outlined at the beginning of this chapter.
Figure 24.25. Groe des Hyènes, Brassempouy, unused extremity of a stage 1 chlorite rod. At centre and at right are
visible the clear stigmata of circumincision.
Figure 24.26. Groe des Hyènes, Brassempouy, stage
3 (chlorite) of the production chain for basket-shaped
beads. At boom, stigmata of lithic tools used in bifacial
scraping to form the stem of the stage 3 rough-out (18x).
299
Systems of Personal Ornamentation in the Early Upper Palaeolithic
While we cannot help but be
inuenced and informed by ethno-
graphic knowledge of hunter-gather-
ers, we must be careful not to impose
that knowledge on the distant past.
Rather, we must evaluate with great
precision the degree to which the
archaeological record for the past
conforms to our expectations from
the ethnographic present. We must be
willing to depart from ethnographic
expectations if the archaeological ob-
servations contradict them. We must
also be willing to abandon badly ex-
cavated archaeological assemblages
whose ambiguity, incompleteness
and lack of precise chronostrati-
graphic context can be used to sustain
virtually anyone’s view of the past. In
particular, I am thinking of sites like
the Groe du Renne at Arcy-sur-Cure
(White 2001; 2002).
In the time since the last Human Revolution
conference in 1987, I, among other researchers (e.g.
Kuhn et al. 2001; Ambrose 1998), have worked hard
to develop an emerging archaeological record for Au-
rignacian-age personal ornaments; a record capable of
responding to higher level questions about Aurigna-
cian social symbolism. Like a gradually developing
photographic print, small patches are distinct while
others remain without detail. As far as the southwest
French Aurignacian is concerned, several trends begin
to emerge with some clarity.
The animals whose teeth are worn are not those
whose meat is consumed. Phrased another way, the
consumed fauna and the displayed fauna are almost
mutually exclusive. This implies that the animals
behind the parts transformed into ornaments are
construed in terms that are largely of the collective
symbolic imagination.
In like fashion, raw materials to be transformed
into ornaments were not chosen for their ready avail-
ability; au contraire. Materials used to communicate
social value and identity, such as ivory, amber, lignite,
mother of pearl, dental enamel and soapstone, share
a common character: visual and tactile lustre. It is
hard to escape the notion that this evocative, qualita-
tive characteristic must have been construed in some
meaningful way within Aurignacian society at a given
place and time. My own experimental work shows the
remarkable time (three hours per ivory basket-shaped
bead) and technology (polished surfaces created by
the use of metallic abrasives) necessary to produce
evocative surfaces.
Dierences in nished bead to production debris
ratios among Early Aurignacian levels lead us to
identify a massive production signature (Castanet),
an ornament maintenance signature (Brassempouy)
Figure 24.27. Groe des Hyènes, Brassempouy, stage 5 (calcite) of the production chain for basket-shaped beads. This is
a nished bead with a decoration created by sub-parallel incisions on the bulbar end.
Figure 24.28. Basket-shaped bead length distributions for Brassempouy and
Vézère Valley sites.
300
Chapter 24
and a ‘no-production-whatsoever’ signature (Isturitz);
permiing the construction of a provisional model of
ornament fabrication within a regional system of sea-
sonal movement and site function Such results allow
us to begin to focus on the way in which ornament
production, use and discard were organized across
Aurignacian-age cultural landscapes; and how that
organisation changed over the course of the Aurigna-
cian in southwest France.
The results presented above also force us to deal
with the reality of assemblage similarities across dis-
tance and, conversely, signicant dierences between
proximate Aurignacian sites. As complex as this reality
is, addressing it head-on moves us beyond largely dis-
credited notions of one region/one ornamentation/one
ethnic group/one language that require the much too
convenient assumption that the available ornamental
record for the Aurignacian is adequate to the task.
Acknowledgements
I am extremely indebted to the members of the Castanet,
Brassempouy and Isturitz research teams for their care and
patience in the recovery of ornaments and related materials.
I especially wish to recognize the support and condence
of Henri Delporte, Dominique Buisson, Dominique Gam-
bier, François Bon and Christian Normand in entrusting
to me the study of the remarkable ornamental materials
from Brassempouy and Isturitz. My own research on or-
naments, as well as the Castanet excavations themselves,
have received generous support from the National Science
Foundation (project number SBR-9806531), the LSB Leakey
Foundation, New York University, the Service Régional de
l’Archéologie pour l’Aquitaine and the Direction Régionale
des Aaires Culturelles d’Aquitaine.
Notes
1. Other researchers concerned with the evolutionary and
cultural signicance of early Upper Paleolithic personal
ornaments, such as Yvee Taborin (e.g. 1988; 1990; 1993)
and Joachim Hahn (1972), were not present at the origi-
nal Human Revolution conference.
2. I insist here on this notion of systems of personal or-
namentation. As a result, rare instances of pre-40,000
year-old pierced objects, isolated in space and in time,
may indicate the technological and aesthetic capacity
of these early humans but their ornamental activities
appear sporadic at best, never seeming to coalesce into
organized regional entities. I leave aside in this paper
the recently claimed outlier of Blombos in South Africa,
noting only that I have serious doubts about the arti-
cially altered nature of the mollusk shells aributed to
the Middle Stone Age. I will have much more to say on
this subject once the dust seles on this highly media-
tized case.
3. It is surprising that several methodological and theo-
retical statements (in both English and French) over the
past two decades, have gone largely uncited by some
researchers currently dealing with Aurignacian orna-
ments (Vanhaeren 2002).
4. It is likely that the very small ornamental assemblages
from relatively recent ‘landmark’ Aurignacian excava-
tions such as those at Abri Pataud, Le Flageolet I, La Fer-
rassie recovered only a tiny proportion of the ornaments
and production debris originally present and preserved;
this largely due to the absence of appropriate recovery
techniques.
5. In my opinion, claims for the absence of personal
ornaments (in Europe or elsewhere) cannot be taken
seriously for sites where ne-mesh wet-sieving and
manual sorting of sieve booms were not undertaken.
6. Given the total absence of Aurignacian burials, we
(White et al. 2003; 2006; Henry-Gambier et al. 2004) have
argued that these human teeth, prepared for suspen-
sion, may indicate Aurignacian mortuary practices that
did not involve burial.
7. The geological control of the new excavations has been
undertaken by Jean-Pierre Texier, Thierry Gé, Romain
Mensan, Philippe Gardère and Paul Goldberg.
8. Almost certainly made from amber obtained from
Cretaceous sources in the Pyrenean forelands (Beck et
al. 1987).
References
Ambrose, S., 1998. Chronology of the later Stone Age and
food production in East Africa. Journal of Archaeological
Science 25, 377–92.
Beck, C.W., F. Chantret & D. Sacchi, 1987. L’Ambre paléo-
lithique de la groe d’Aurensan (Hautes-Pyrénées).
L’Anthropologie 91, 259–62.
Binford, L., 1973. Interassemblage variability: the Moust-
erian and the functional argument, in The Explanation
of Culture Change, ed. C. Renfrew. London: Duckworth,
227–54.
Binford, L.R. & S.R. Binford, 1966. A preliminary analysis of
functional variability in the Mousterian of Levallois
facies. American Anthropologist 68, 238–95.
Bon, F., 2000. La question de l’unité technique et économique
de l’Aurignacien: reexions sur la variabilité des
industries lithiques à partir de l’analyse comparée de
trois sites des Pyrénées françaises. La Tuto de Camal-
hot, Régismont-le-Haut et Brassempouy. Unpublished
PhD dissertation, Université de Paris I.
Bon, F., 2002. L’Aurignacien entre Mer et Océan. Réexion sur
l’unité des phases anciennes de l’Aurignacien dans le sud
de la France. (Mémoire XXIX.) Paris: Société Préhis-
torique Française.
Bon, F., 2002. Les termes de l’Aurignacien, in Autour des
concepts de Protoaurignacien, d’Aurignacien initial et
ancien : Unité et variabilité des comportements techniques
des premiers groupes d’Hommes modernes dans le Sud de
la France et le Nord de l’Espagne, eds. F. Bon, J.M. Maillo
Fernandez & D. Ortega Cobos. (Actes de la Table-
ronde de Toulouse, 2003.) Special issue of Espacio,
Tiempo y Forma 15, 39–65.
301
Systems of Personal Ornamentation in the Early Upper Palaeolithic
Bordes, J.-G. 2000. La séquence aurignacienne de Cami-
nade revisitée: l’apport des raccords d’intérêt strati-
graphique. Paléo 12, 387–407.
Castel, J.-C., 2007 (in press). Résultats de l’étude paléon-
tologique. Les fouilles du nouveau secteur de l’abri
Castanet: Contexte géographique, historique et prob-
lématique, in Abri Castanet: Fouilles du nouveau secteur
1994–1998 (dedicated issue of the journal Paléo).
Conkey, M., 1978. Style and information in cultural evolu-
tion: Toward a predictive model for the Paleolithic, in
Social Archaeology: Beyond Subsistence and Dating, eds.
C. Redman, W. Langhorne, M. Berman, N. Versaggi,
E. Curtin & J. Wanswer. New York (NY): Academic
Press, 61–85.
d’Errico, F. 2003. The invisible frontier: a multiple species
model for the origin of behavioral modernity. Evolu-
tionary Anthropology 12, 188–202.
Delluc, B. & G. Delluc., 1981. Les manifestations graphiques
aurignaciennes sur support rocheux des environs des
Eyzies (Dordogne). Gallia Préhistoire 21, 213–438.
Didon, L., 1911. L’Abri Blanchard des Roches (commune
de Sergeac). Gisement aurignacien moyen. Bulletin
de la Société Historique et Archéologique du Périgord 87,
246–61 & 321–45.
Didon, L. ,1912. Faits nouveaux constatés dans une station
aurignacienne, L’Abri Blanchard des Roches prés de
Sergeac. L’Anthropologie 23, 603.
Fontugne, M., 2007 (in press). Datations des couches auri-
gnaciennes de la groe des Hyènes, in L’Aurignacien
ancien de la groe des Hyènes (Brassempouy, Landes), eds.
D. Henry-Gambier & F. Bon. Supplement to Gallia
Préhistoire.
Gambier, D., 2000. Aurignacian children and mortuary
practice in western Europe. Anthropologie (Brno) 38(1),
5–21.
Granger, J.-M. & F. Levêque, 1997. Parure castelperronienne
et aurignacienne: étude de trois séries inédites de
dents percées et comparaisons. (Sciences de la terre
et des planètes 325.) Paris: C.R. Académie des Sci-
ences, 537–43.
Henry-Gambier, D. & F. Bon (eds.), 2007 (in press).
L’Aurignacien et le Châtelperronien de la groe des
Hyènes à Brassempouy (Landes). Supplement to Gallia
Préhistoire.
Henry-Gambier, D. & R. White, 2006. Modications articiel-
les des vestiges humains aurignaciens de la groe des
Hyènes et de la galerie Dubalen. Quelle signication.,
in El centenario de la cueva de El Castillo: el ocaso de
los Neandertales, eds. V. Cabrera & F. Bernaldo de
Quiros. Madrid: Ministerio de Educacion y Ciencia,
71–88.
Henry-Gambier, D., B. Maureille & R. White, 2004. Vestiges
humains des niveaux de l’Aurignacien ancien du site
de Brassempouy (Landes). Bulletins et Mémoires de la
Société d’Anthropologie de Paris, n.s. 16, 49-87.
Hahn, J., 1972. Aurignacian signs, pendants and art objects
in central and eastern Europe. World Archaeology 3,
252–6.
Hahn, J., 1988. Das Geißenklösterle I. Stugart: Konrad Theiss
Verlag.
Kuhn, S., M. Stiner, D. Reese & E. Güleç, 2001. Ornaments of
the earliest Upper Paleolithic: new insights from the
Levant. Proceedings of the National Academy of Sciences
of the USA 98(13), 7641–6.
Letourneux, C., 2003. Devine qui est venu diner a
Brassempouy? Approche taphonomique pour une
interprétation archéozoologique des vestiges osseux
de l’Aurignacien ancien de la groe des Hyènes
(Brassempouy, Landes). Unpublished PhD thesis,
Université de Paris I – Panthéon Sorbonne.
Normand, C., 2002. Nouvelles données sur l’Aurignacien
de la groe d’Isturitz (communes d’Isturitz et de
Saint-Martin-d’Arberoue ; Pyrénées-Atlantiques):
l’industrie lithique de la Salle de Saint-Martin (re-
cherches 2000–2002), in Autour des concepts de Pro-
toaurignacien, d’Aurignacien initial et ancien. Unité et
variabilité des comportements techniques des premiers
groupes d’Hommes modernes dans le Sud de la France
et le Nord de l’Espagne, eds. F. Bon, J.M. Maillo Fer-
nandez & D.Ortega Cobos. (Actes de la Table-ronde
de Toulouse, 2003.) Special issue of Espacio, Tiempo y
Forma 15, 145–74.
Normand, C., 2005. Les occupations aurignaciennes de la
groe d’Isturitz (communes d’Isturitz et de Saint-
Martin-d’Arberoue; Pyrénées-Atlantiques; France):
synthèse des données actuelles. Saint-Sébastien,
Munibe (numéro spécial en hommage au Dr J. Altuna)
57, 119–29.
Normand, C., 2007 (in press). Les recherches archéologiques
dans les groes d’Isturitz et d’Oxocelhaya de 1912
à nos jours : synthèse des résultats. Actes de la table-
ronde d’Hasparren, 14 et 15 novembre 2003. Paléo,
hors-série.
Normand, C. & A. Turq, 2006. Bilan des recherches 1995–
1998 dans la Groe d’Isturitz (communes d’Isturitz et
de Saint-Martin-d’Arberoue, Pyrénées-Atlantiques),
in Préhistoire du Bassin de l’Adour: bilans et perspectives,
ed. C. Chauchat. Navarre: Izpegi, 69–98.
Normand, C., S.A. de Beaune, S. Costamagno, et al., 2007 (in
press). Nouvelles données sur la séquence aurignaci-
enne de la groe d’Isturitz (Saint-Martin-d’Arberoue ;
Pyrenees-Atlantiques). Congrès Préhistorique de France,
Avignon.
Oe, M., 1974. Observations sur le débitage et le façonnage
de l’ivoire dans l’Aurignacien en Belgique, in Premier
Colloque International sur l’Industrie de l’Os dans la
Préhistoire, ed. H. Camps-Fabrer. Aix: Editions de
l’Université de Provence, 93–9.
Passemard, E., 1944. La caverne d’Isturitz en Pays Basque.
Préhistoire 9, 1–84.
Peyrony, D., 1935. Le gisement Castanet, Vallon de Castelmer-
le, Commune de Sergeac (Dordogne). Aurignacien I et
II. Bulletin de la Société Préhistorique Française 32, 418–43.
Pike-Tay, A., 2007 (in press). Etude de la saisonnalité sur
dents de renne, in Abri Castanet: Fouilles du nouveau
secteur 1994–1998. (dedicated issue of the journal
Paléo, .
Sáenz de Buruaga, A., 1989. Colgantes y otras manifesta-
ciones artisticas en los niveles del Paleolitico superior
inicial de la cueva de Gatzarria. Veleia 6, 21–48.
302
Chapter 24
Sáenz de Buruaga, A., 1991. Paleolítico superior de la cueva de
Gatzarria (Zuberoa, País Vasco). (Dedicated issue of the
journal Veleia.) Vitoria: Santander.
Saint-Périer, R. de, 1930. La Groe d’lsturitz. Paris: Archives
de 1’Institut de Paléontologie Humaine, no. 7.
Taborin, Y., 1988. Les prémices de l’expression symbolique.
in De Néandertal à Cro-Magnon, ed. C. Farizy. Nemours:
Musée de Préhistoire d’Isle de France, 73–5.
Taborin, Y., 1990. Les prémices de la parure, in Paléolithique
moyen récent et Paléolithique supérieur ancien en Europe,
ed. C. Farizy. Nemours: Mémoires du Musée de
Préhistoire de l’Isle de France 3, 335–44.
Taborin, Y. 1993. La parure en coquillage paléolithique. (Gal-
lia Préhistoire supplement 29.) Paris: CNRS.
Valladas, H., J. Pelegrin, R. White, H. Knecht & M. O’Farrell,
2007. Présentation des datations C14 et TL in Abri Cas-
tanet: Fouilles du nouveau secteur 1994–1998. (dedicated
issue of the journal Paléo).
Vanhaeren, M., 2002. Les fonctions de la parure au Paléo-
lithique supérieur: de l’individu à l’unité culturelle.
Unpublished PhD dissertation, Université de Bor-
deaux I.
Vanhaeren, M. & F. d’Errico, 2006. Aurignacian ethno-lin-
guistic geography of Europe revealed by personal
ornaments. Journal of Archaeological Science 20, 1–24.
Vézian, J. & J. Vézian, 1966. Les gisements de la groe de
Saint-Jean-de-Verges (Ariège). Bulletin de la Société
Préhistorique de l’Ariège 25, 29–65.
White, R., 1989a. Production complexity and standardisation
in early Aurignacian bead and pendant manufacture:
Evolutionary implications, in The Human Revolution:
Behavioural and Biological Perspectives on the Origins of
Modern Humans, eds. C. Stringer & P. Mellars. Edin-
burgh: Edinburgh University Press, 366–90.
White, R., 1989b. Toward a contextual understanding of the
earliest body ornaments. in The Emergence of Modern
Humans: Biocultural Adaptations in the Later Pleistocene,
ed. E. Trinkaus. Cambridge: Cambridge University
Press, 211–31.
White, R., 1992. Beyond art: toward an understanding of the
origins of material representation in Europe. Annual
Review of Anthropology, 21, 537–64.
White, R., 1993a. A social and technological view of Aurig-
nacian and Castelperronian personal ornaments in
France, in El Origen del Hombre Moderno en el Suroeste
de Europa, ed. V. Cabrera Valdès. Madrid: Ministerio
de Educacion y Ciencia, 327–57.
White, R., 1993b. Technological and social dimensions of
‘Aurignacian-age’ body ornaments across Europe, in
Before Lascaux: the Complex Record of the Upper Paleo-
lithic, eds. H. Knecht, A. Pike-Tay & R. White. Boca
Raton (FL): CRC, 277–99.
White, R., 1997. Substantial acts: from materials to meaning
in Upper Paleolithic representation, in Beyond Art:
Upper Paleolithic Symbolism, eds. D. Stratmann, M.
Conkey & O. Soer. San Francisco (CA): California
Academy of Sciences, 93–121.
White, R., 1999. Integrating social and operational com-
plexity: the material construction of social identity at
Sungir, in L’Os : Festschri for Henriee Camps-Fabrer,
eds. A. Averbouh, P. Caelain & M. Jullien. Aix-en-
Provence: Université de Marseille, 120–37.
White, R., 2000. Un big bang socioculturel: Une nouvelle
façon d’apprehender le monde est née voici 40,000
ans La Recherche 4, 10–16.
White, R., 2001. Personal ornaments from the groe du
Renne at Arcy-sur-Cure. Athena Review 2(4), 41–6.
White, R., 2002. Observations technologiques sur les objets
de parure, in L’Aurignacien de la Groe du Renne. Les
fouilles d’André Leroi-Gourhan à Arcy-sur-Cure (Yonne),
ed. B. Schmider. (Supplément to Gallia Préhistoire,
XXXIV.) Paris: CNRS, 257–66.
White, R., 2007a (in press). La parure aurignacienne de la
groe des Hyènes, Brassempouy, in L’Aurignacien et
le Châtelperronien de la groe des Hyènes à Brassempouy
(Landes), eds. D. Henry-Gambier & F. Bon. (Supple-
ment to Gallia Préhistoire.)
White, R., 2007b (in press). Les fouilles du nouveau secteur
de l’abri Castanet: Contexte géographique, historique
et problématique, in Abri Castanet: Fouilles du nouveau
secteur 1994–1998. (dedicated issue of the journal
Paléo).
White, R., 2007c (in press). Objets de parure et témoins
esthétiques à l’Abri Castanet, nouveau secteur, in
Abri Castanet: Fouilles du nouveau secteur 1994–1998.
(dedicated issue of the journal Paléo).
White R., 2007d (in press). Les parures aurignaciennes
d’Isturitz: perspectives technologiques et régionales,
in Les recherches archéologiques dans les groes d’Isturitz
et d’Oxocelhaya de 1912 à nos jours: synthèse des résultats,
ed. C. Normand. (Actes de la table-ronde d’Hasparren,
14 et 15 novembre 2003.) Paleo, hors-série.
White, R. & L. Breitborde (eds.), 1992. French Paleolithic
Collections in the Logan Museum of Anthropology. (Bul-
letin, New Series, 2.) Beloit (WI): Logan Museum of
Anthropology.
White, R., D. Henry-Gambier & C. Normand, 2003. Human-
tooth ornaments from the French early Aurignacian:
implications for early Upper Paleolithic treatment of
the dead. Annual Meetings of the Paleoanthropology
Society, Tempe, Arizona. Published abstract, hp://
www.paleoanthro.org/abst2003.htm.
Wobst, M.H., 1978. The archaeo-ethnology of hunter-gath-
erers or the tyranny or the ethnographic record in
archaeology. American Antiquity 43(2), 303–9.
Zilhão, J. & F. d’Errico, 2003. The chronology of the Aurig-
nacian and Transitional technocomplexes. Where do
we stand?, in The Chronology of the Aurignacian and of
the Transitional Technocomplexes. Dating, Stratigraphies,
Cultural Implications, eds. J. Zilhão & F. d’Errico. (Tra-
balhos de Arqueologia 33.) Lisbon: Instituto Português
de Arqueologia, 313–48.
