93

Substantial Acts: From Materials (

to Meaning in Upper Paleolithic

(
Representation (

(
Randall White (
Department of Anthropology
(
New York University
25 Waverly Place (
New York, NY 10003
\
(
"What does it mean"? This is at one and the same time the most frequently asked
and the most naively conceived question asked ofUpper Paleolithic representations.
No a11 historian trained in the past two decades would ask this question of a Van Gogh
or a Picasso. Why then have generations of prehistorians treated entire corpora ofPa-
leolithic representations as if a single meaning and motivation lay behind them? Ex-
amples are numyrous: Female sculptures as fe11ility figures; painted and engraved
animals as instruments of hunting magic; geometric signs as markers of ethnic iden- (
tity and/or as the painted equivalents of gendered articles of speech.
(
The worst and most pervasive misconception is not even formulated as a question
but as an assumption embedded in the very te1m "Paleolithic art." Conkey and I have (
loudly and frequently decried the use of the concept "art," because of its status as an
historical artifact of the later stages of the so-called Western tradition. Indeed , 20th
century usage of the term "art" bears almost no relation to the Latin concept of "ars," (
which integrated the domains that we distinguish as " att'' and "savoir-faire."
Any thorough treatment of the anthropological literature on cultural esthetics (
makes cl ear the wide diversity of cosmologies, philosophies and social contexts that
underpin what we generalize as "art." I prefer the term "representation," which has a
wide and theoretically complex usage in anthropology. Thus representations can take
many forms , can have widely differing underlying logics, can be diversely motivated;
and, importantly, many representational media do not even operate in the visual/for-
mal channel.
This understood and accepted, we can forsake a focus on the origins of "a11" or
"the arts," which have enormous, but highly ethnocentric, cultural value. Rather we
can redefine what occurs at the beginning of the Upper Paleolithic as the invention of
material forms of representation (White 1992). Such an approach has the additional
advantage of allowing us to expand outward from graphic imagery to include, for ex-
Beyond Art: Pleis tocenl! Image am/ Symhol Memoirs of the Califomia Academy of Sciences
r .J: .. _ 'o r- - ' - A C' (T._ J"'\ C' -~ - P. X I r: l n l~\ ,... ,. ,.. f . ; 1'\!,,...-, h .,. ,..")l. r r. r.., .-i,..,ht fi'\ 1007
 )
) 94 WHITE
) ample, the representational domain of personal adornment, a critical area of study in
) anthropological analyses of meaning, value and social identity.
It is my position here that by asking, "What does it mean?'' or even the somewhat
) less anthropologically objectionable, "What purpose does it serve,?" and by conceiv-
in g Upper Paleo] ithic representations as "a11" in our sense, we have prevented a seri-
)
ous treatment of meaning(s). How then do we reconceptualize our subject matter, our
) notion of meaning and, more importantly~ how do we operationalize these new con-
ceptions in real archaeological research? My answer to thi s latter question is to focus,
) where, I have explored in detail raw matenal choice and fabncat10n techniques and
literally, on how meaningful representations are/were const111cted, what I have de-
scribed in the title of this paper as "substantial acts."
) Technological Orientati~n
) the Aurignacian ometime prior to 35,000 years ago. In light of what we know genei-
The tenn "technology"' has had a ve1y limited and theoretically uninteresting con-
) notation in American anthropology for decades (cf Dobres & Hoffman 1994 ). Mate-
rialist frameworks such as those ofWhite (1959) and Steward (1955) left relatively
) sentation of a diversity of individual and social Identities.
littl e room for considering tools and techniques as anything more than culturally pre-
) scribed means of production. Materialist th eo1y in American anthropology immedi-
ateiy before and after World Warii seems to have operated in ignorance of seminal
) th eo retical developments in what became European Sli11Cturalism (Mauss I 936;
) Leroi-Gourhan 1943 , 1945). My own orientation with respect to technology is de-
scended from that o f th e late Andre Leroi-G ourhan , who himsel fwas clearly ins pired
) by Marcel Mauss 's Techniques d11 C01ps, and whose major theoretical and methodo-
logical works on technology have only begun to be translated into English (Leroi-
) G c,urhan 1993). Critical to Leroi-Gourhan ' s research design was the notion of the
) chaines operatoire, which he saw as a conventionai ized, learned sequence of techn i-
cal ope rations implicated in all cultural production from the manufacture of stone
) tools to the painting of underground cavities to the modern assembly line. These
chains were constituents of culture rather than byproducts .
) At the level of individual material production episodes, these chaines operatoires
) are constituted of sequences of applied techniques, which Leroi-Gourhan viewed as
th manipulation of conventional tools by means of habitual, learned gestures.
) Vi ewed more broadly, chciines operatoires can be seen as organized into more en-
compass ing technical system , driven by a limited number of technical principl es.
)
Lero i-Gourhan clearly recognized that these underlying tpchnical systems were/are
) never the only ones possible and argued that they emerg~d from profound cultural
contex rs and historical/evolutionary trajectories.
) For him, regional variation in underlying technical systems man if, sted itself in
) what archaeologists recognize as style, which he saw as a new and revolutionary kind
of material behavior that replaced the need for regional biological diversification in
) hu mans . A new generation of scholars of technology and material culture (Lechtman
J 971; Lemonnier 1983 ; Schlanger 1994) have built on the foundation constructed by
) Lero i-Gourhan (cf Dobres & Hoffman 1994).
) My goa l here is to illustrate how detailed observation and experimentation
ai med at understanding the chaine opera to ire underlying the construction of material
) representations can lead to new insights into the social, economic and ideational con-
texts of th e representations themselves.
) age sites (White 1995b) were fabricated of1vory. Indeed , many of the fam ous Vogel-
) herd statuettes were pierced for suspensiOn. In the Western European Aungnacian,
)
MATERIALS AND MEANING IN PALEOLITHIC IMAGES 95
Case Study I: Aurignacian I Personal Ornaments
The Sample
European sites attributed to Aurignacian I (roughly 35-30,000 BP)_have yielded
an abundance of personal ornaments in the form o~beads, pendants, pterced ammal
teeth and pierced marine shells (White 1989; Tabonn 1993; Hahnl972, 1986). E~ se
the ways in which these varied across the European landsc~pe at the very begmnmg of
the Upper Paleolithic. It has been and it remamsmy position that thiS large corpus of
personal ornaments, following a near absence 111 the precedmg Moustenan ~White
1995a), is reflective of significant trans fonnat10ns lll human society With the 0.1set of
ally about the soc ial context of modern human bodily adornment, It seems reasonabl~
to hypothesize that the first appearance 111 thearchaeolog1ca I record oflai ge and var
ied assemblages of personal ornaments Implies the matenal constructiOn and 1ep1 e-
A Schematic Chaine Operatoire for Aurignacian I Personal Ornaments
A schematic outline of the constituents of a chaine operata ire for ~urignaci an (or
any other) system of personal ornamentation might look somethmg !Ike the follow-
ing:
I) Cultural assumptions about personhood. . . . . _
2) Be liefs about the relatiOnship between matenals, representatiOnal acts, 1epre
sentational constructs and socia~/supernatur~l efficacy. . .
3) Choice and acquisition (by direct ex tractiOn or by social mechamsms of ex-
change) of raw material based on above. .
4) Choice of forms , textures , colorsor subject matters. .
5) Organization of production, (social, temporal and spatial) . .
6) Combination of gestures and tools into technique~ for ornament production that
are coherent with the encompassmg, regional techmcal sys:ems and that enable_
7) Representation of desired signifiers of social Jdentity, age, reproductive
status,supernatural associations, ec.. . . . .
8) Use of the omamental representations m (socially, esthetically and cosmolog i-
ca lly) meaningful acts. . . ..
9) Purposeful (based on intentions for future retneval/use or Ideas about then re-
sidual power and efficacy) or acc idental (as a byproduct of human activity) di s-
posal of the ornaments. . , k .
It is important to emphas ize that, procedurally, archaeologists tend to NO . 111 re-
verse through these chains; beginning w1th basic pattern recogmt10n concern_mg the
distributi on of objects in the ground and movmg backward through multifaceted,
higher order analyses and inferences.
Formal Attributes of Beads
By far the majority of preserved personal ornaments in European Aurignacian-
 96 WHITE MATERIALS AND MEANING IN PALEOLITHIC IMAGES 97

facsimiles of red deer canines and marine shells (Figure 1) were executed in ivory (ex- talc beads so frequent in SW French Aurignacian I sites (see below) show a striking
amples are also known in talc and limestone). In addition, there is a variety of more- resemblance in form and size (size standardization is discussed below and has been
or-less idiosyncratic " pendants" manufactured of ivory and, less often, ofbone, antler quantified in White 1989) to a species of Mediterranean seashe ll found in so me
Aurignacian I sites in SE France: Cy clope neritea (Figure 3). In hght of th is hypothe- (
or talc. For reasons of space, I shall focu s here on operational sequences for the mass
production of ivory "beads," giving less detailed treatment to rarer ivory ornaments. sized tactile constituent of Aurignacian I representation, it is perhaps not surprising to
(
Two formal attributes of Aurignacian I beads and pendants are particularly note- observe that more than 95% of Aurignacian I personal ornaments are constructed of
worthy: surface luster and facsimile. First, texturally, a lmost all Aurignacian I beads ivory, talc, shell or animal tooth. (
and pendants exhibit remarkabl e surface luster th at is not a produ ct of postdeposi-
tiona l processes. The luster on ivory beads (Figure 2) was intentionally produced by
techniques (discussed later in this paper) of grinding and po lishing, es pecially the use
of an effective metallic abras ive, powdered ocher (hematite) .
The pote ntia l of ivo ry to take on lustrous polishes can be seen as having been ex-
ploited through the creation and application of appropriate tec hniques for replicating
(
the natura ll y occuning tactil e characteris tics of other omamenta l media, such as
mother of pearl , talc/s teatite and dental ename l. In other words, the poli shing of ivory (
was itse lf a representation of textures ex peri enced e lsewhere in the natural world .
Our own predominant cultural medium of represe ntation is visual; as a result we have (
tended to treat tactile charac teristi'cs of objects as nonrepresentational or minima lly
representational. H owever, a moment of refl ection will reveal that, even in the so- (
called " Wes tern Tradition," polished ivory's tac tile qualities have been so sought- (
after that elephants have been brought to the ve rge of ex tinction.
The hypothes is that the polishing of ivory was an attempt to represent the surface (
qua lities of other substances of tactile interes t is supported by the ex istence in A uri g-
nacian I s ites of ivory facs imil es of seas hell s (m other-of-pearl), and anima l teeth
(denta l ename l and natura ll y lustrous dentine). Indee.:l, the basket-s haped ivory and (

(
(
(
(
(
(
(

FIGURE 2. Ivory bas ket-shaped bead from Abri Blanchard, France, showing characteri stic luster. Scale in
(
mm.
(
(

FiGU RE I . Three of six facsim il es ofseashcllssculpted ii] ivory from the Aurignacian I s iteof La Souq uette,
(
France. ,.
:: : ,

(
 98 WHITE MATERIALS AND MEANING IN PALEO LITHIC IM AGES 99

Regional Perspectives

The fabrication sequence (a mere fragment of the much broader and inclusive
chaine operatoire outlined above) for Aurignacian ivo ry and stone beads varies both
intra- and interregi onally. In France, the most common form, represented by more
than 1000 specimens, is what has been called basket-shaped beads. Found in large
quantities early in this century atAbri Blanchard, Abri Castanet, Abri de Ia Souquette,
Istu ritz and Saint Jean de Verges, these have now been radi ocarbon dated by Del porte
at Brassempouy (De lporte & Bui sson 1990) to between 33 - and 32,000 years BP.
They were created (Figure 4) from pencil-like rods of ivmy or talc that were then cir-
cuminscribed and snapped in to cylindrical blanks one to two centim eters long. T hese
were then bilaterall y thinned at one end to form a sort of stem. A perforation was then
created at the junction of the stem and the unaltered end. Thi s was done by gouging
from each s ide, rather than by rotational drilling. These rough-outs were then ground
and polished into their fina l basket-shaped form using, first coarse abrasives and then
fine metallic abras ives (powdered hematite) as abras ive. My experiments indi cate
that a mean of one to two hours of labor per ivory bead and 30 minutes per talc bead
are req uired by thi s process. ..
Ivory beads in so uth Ge1man Aurignacian sites, also radi ocarbon dated to be-
tween 32- and 33,000 years BP, are substantially different, although the basic princi-
ple of reducing an ivozy baton was the same (Figure 5). In the case ofGeif3enklosterle
for example (Hahn 1986), a baton ellipti'cal in section was circumincised and
snapped . The blank was then thinned and perforated by go uging. In this case, how-
ever, two ho les separated by a bulge were du g into the blank. Thi s type of bead is as
unknown in France as the basket-shaped form is in Germany,

Raw Material Choice and Acquisition

The full range of ornamental raw materials preserved in the record includes va ri-
ous mineral and animal substances includin g limestone, schi st, talc-schist, talc, mam-
malian teeth, bone, antler and ivo ry, fossil and contemporary species of marine and
fresh-.vater shells, fossil coral, fossil bel emnite,jet, lignite, hematite, an d pyrite. How-
ever, this relatively extensive list shou ld not be taken to suggest a kind of rando m use
of materials encountered in the environment. A number of pronounced choices were
made.
If one exam ines the spec ies of animals whose teeth were chosen for Aurignacian
objects of suspension, there are clear choices that vary somewhat regionally. In.
France, Belgium, Germany and Russ ia fox canine teeth predominate fo llowed in
much small er quantities by the vestigial canines ofcervids, a[most always red deer. In
Spain and Italy, almost all pierced teeth are vestigial canines. of red deer, with fox ca-
nines being absent. At Ml adec in Czechoslovakia, beaver incisors dominate, closely
followed by moose and bovid incisors. In nearly all instances the species sample for
objects of s uspension is fund amentally differe nt from that found in the food debris,
suggesting choices based on ideas exclusive of dietary preferences .
Pierced marine, freshwater and foss il shells (Taborin 1993) constitute perhaps a
third of the objects of suspension found in French Aurignacian sites. Outside of
France, with the possible exception ofNorthern Italy (Bartolomei et al. 1992), they
are so rare as to be considered virtually absen t in Aurignac ian-age sites in Spain, Bel-
gium, Germany, Czechoslovakia, Hungary and Russia. Loca l marine fossils (bel em-
FIGURE 3. A specimen of Cyclope neritea, closely resembling many of the Aurignacian I basket-shaped
beads. Length 6.5 mm. (CoUI1esy ofY. Taborir.).
nite, co ral) served as raw materials for objects of suspens io n in so me
Aurignac ian-age sites in Russia (White 1993).
Occasional pierced objects in limestone are found throughout Europe. On the
other hand, rarer soft stones, such as talc, li gn ite and hemat ite were turned into
Aurignacian-age items of suspension most often in southern France, but a lso occa-
siona lly in Spain, Italy, and Gezmany.
Personal orn aments are freq uently manufactured of material s exotic to the
re-gions in which they are found. This is especiall y true of shells and rare mineral s,
but may a lso be true of ivory, since mammoth remains are virtually absent from
Aurignac ian I sites (Delpech 1983) in SW France (although the collection by Aurig-
nacian people of subfossil ivory from geological sources is a distinct possibility; see
below) . In general, rare minerals in France fall off with distance from point of natural
origin . For example, talc attenuates as one proceeds north from the Pyrenees. L ike-

)
 100 WHITE
MATERIALS AND MEANING IN PALEOLITHIC IMAGES 101
wise Atlantic and Mediterranean shell species attenuate as one proceeds into the
pulp cavity, leaving behind processes known as dentinal tubules (MacGregor 1985)
French interior.
or so-called Hunter-Schreger bands. These "radiate outward from the pulp cavity and
incline obliquely towards the tip" of the tusk. The result, in transverse section , is a
Contribution of Raw Materials to Representational Form: (
complex three-dimensional structure (Figure 7) that takes the form of arcs running
Mammoth-Ivory as a Raw Material counter to each other across the width of the tusk, and producing by their intersection (
what are frequently referred to as "engine-turnings" of the Hunter-Schregerpattern.
Proboscidean tusks are merely specialized teeth (upper incisors) , and are
composed predomimite!y of dentine. In contrast to many other ivories, mammoth and
elephant tusks have a cori1plex structure (Figure 6) fom1ed by specialized cells known
as odontoblasts. These odontoblasts produce new dentine along the lining of the pulp
cavity. As new dentine is produced, the odontoblasts migrate to the new surface of the
~-ae . , . . ,, :e=:.r. ;.;e,; --.s ... ,., . . ., .. ,.. -~ .. - .. > , ,: . ), .,. . . .~ .. ~ . : s"M ;' II . ' ;..i.!;C
I
(

( :. . . -=. .-~~ . . ., .. ..,.a~

(

II (

~- _;[
-
; >- - ~
-
I~ --cr. . a~.- _-.::; c
(

(
III
III
(a) (b) (c)
(

~l]Q ~8~ g&~ (

(
N .. '
' (

~ 88 ~ (
(
Fit1isbed Produd (

e
v ..

~
v
~.. ..
0 J
I I
I
2
I .I
= (
0Randal!Wiut~199)

I
FIGURE 4 . The production sequence for Aurignacian I basket-shaped beads.
. ~.
FIGURE 5. The production sequence for south German Aurignacian I two-holed beads (after Hahn 1986).
(
(
(
 MATERIALS AND MEANING IN PALEOLITHIC IMAGES 103
102 WHITE

In transverse section, this Hunter-Schreger pattern crosscuts broad concentric

rin gs often referred to as laminae (Figure 8). T hese are ac tually growth interfaces be-
tween superimposed cones of ivory. In living tusk, the boundaries between these
laminae are not areas of structural weakness , being bound together by the complex in-
tersecting dentinal tubules compos ing the Hunter-Schreger lines and by coll agen fi-
bers that fill the interstices between the dental tubules. However, upon desiccation
and accompanying deterio ration of dentinal collagen, tu sks tend to split or spa II a long
these concentric boundaries, which are fa r more developed in the outer zones of the
tusk's diameter. T he inner core of the tusk, in the area surroundi ng the centra l nerve
canal, is hi ghly compact, homogeneous and virtually imm une to such spa lling. As a
result, such inner ivo ry is exceedi ng ly diffi cul t to work w ith sto ne too ls. It is perhaps
no su rprise then that in ivo ry- ri ch Aurignac ian sites in SW france, this "core ivo1y" is
abundan tl y represented in the waste products of tusk red uction.
Ivory has properties (color, luster, softness o r warmth of to uch) that set it apart Nerve
from such media as bone an d antler. Howeve r, these qualities can on ly be reali zed by
Canal
means of polish ing with fine abrasives. Accord ing to Ritchie ( 1969), modern ivory
workers prefer fine meta llic abras ives iuclud ing j ewe ler's rouge; which is nothing
more than hematite/red ocher. Indeed, SEM analysis of Aurignacian I ivory beads
(W hite 1995b) revealed particles of red ocher embedded in the fine pol ishing striae on
their surfaces . Moreove r, large caches of red ocher have been recovered from two of
the ri chest iv01y-bearing Aurignacian sites in SW France: Abri Blanchard and Abri
Castanet (Didon 19 11; Peyrony 193 5).

Tools, Techniques and Gestu res: Exper imental Perspectives

My experimental research into the working of 1vory faisified several naive

misconceptions about probosc idea n tusks as raw materia l. My first mistake was to
presu me that modern African elephant tusk was a suitab le ex perimental analogue for
woolly mammoth tusk. A lthough, superfic ially, the d iffe rences do not appea r great,
the 1:\vo fo rms of ivory exhibit qHite differen t fracture patterns that are the product of
Enamel
!Tip Onh")
signi fi cant d iffere nces in the angle of intersection ofHunter-Schreger bands (known
as Hunter-Schreger ang les) .
Contrary to my ini tial pres umption, fresh tusks proved extreme ly difficu lt to break
into, and to reduce to usable parts . Fresh ivo1y does not fiacture along the large- sca le
concentric lam inae observable in prepared sections, as we had so naively imagined .
or can a fresh elephant tu sk be sectioned longitudi nally by beginn ing a spl it a1 the
thin, holl ow proximal end (pulp cavity) and propagating it toward the distal end of the

0
tusk.
Neither Alaskan permafrost iv01y coll ected in the 1920's, nor fresh Afiican
elephant ivory cou ld be worked effective ly by direct percussion. Attempts a t direc t '
.
percussion produced the same results that one mi ght expect ofh itting a stout piece of
hardwood with a hammers tone' Indeed, there are distinct s imila rities in the response
of hardwood and ivory to percussion and wedging, based upon s imi larities in struc-
ture and grain. Sizable flak es were removed by percuss ion onl y where the tusk had a l-
ready been partiall y spl it by desiccation. FtGURE 6. The morphology o f proboscidean tusks.
Neither fresh nor artificially desiccated (ki ln-dried) African elephant ivory could
be worked by piitting-and-wedging. Upon des iccation, elephant ivory deve loped in-
cipient concentric fractures that confonned to the broad internal laminae of the tusk.
However, these were too poorly developed _io serve as points of access for wedges .

104 WHITE
t. H nter Schreger bands seen here in transverse section in a late Pleistocene
FtGURE 7. Th e llltersec mg u - The use ofsubfos sil tusks as raw material has been hypothesized previously by Hahn
mammoth tusk. A single desiccation fracture is seen to crosscut the bands.
FtGURE 8. Concentric desiccation laminae, see~ in transverse section in a late Pleistocene mammoth tusk.
Note the radial fractures crosscutting the concentric desiccation fractures.
MATERIALS AND MEANING IN PALEOLITHIC IMAGES 105 (
Overheating in a kiln made modem elephant ivory so brittle that it was breakable by (
hand, and structurally weaker than many 35,000 year-old Aurignacian artifacts .
At the time I received it, the medial segment of Alaskan permafrost mammoth tusk
showed a quite developed concentric fracture at a lamellar boundary near its external (
surface. Attempts to exploit this desiccation fracture by percussion-driven wedging
from the distal extremity of this medial segment produced broad flakes that hinged (
outward part way down the segment's length. However, a split could not be propa-
gated through the entire length of the segment, presumably because the lines offorce
were directed outward when they encountered the interfaces between "superimposed
cones" of ivory. With enough labor however, it is certainly possible to scrape, grind
and polish such large flakes of fresh ivory into desired forms . Thus, their employment
by Aurignacian ivory-workers cannot be entirely excluded.
In addition, the slightly desiccated Alaskan permafrost tusk showed radial
fractures that crosscut the concentric laminar fractures (Figure 8). While these radial (
fractures did not provide an effective purchase for percussion-driven wedges to split
(
the tusk longitudinally, this is probably attributable to the extremely "fresh" state of
the permafrost tusk. Similar attempts with more highly desiccated lamellar fragments (
of mammoth tusk led to considerable success in creating long, workable splinters by
longitudinal splitting and wedging.
Splitting-and-wedging constituted a fundamental Aurignacian strategy (Knecht
(
1993) for working organic materials, but I conclude that they are/were not especially
appropriate to the structure of fresh or even slightly desiccated proboscidean tusks. (
(
( 1986) for the German Aurignacian and by Phillipov ( 1983) for th e Upper Paleolithic
of the Russian Plain. I think it nearly certain that, apart from some fotm of art ificial
drying or purposeful curing of fresh tusks, "subfossil" or at least mammoth tusks
(
ome years old were sought as raw material by Aurignacian ivmy workers. (
Even then, the production of long rods of ivmy by splitting and wedging of tusks
is not suggested by the structure of ivory, and is difficult to achieve. These rods seem
to have been a preexisting goal of Aurignacian ivory workers even if they required the
(
solution of significant mechanical problems. My best explanation for this determina-
tion by Aurignacian I ivory workers to obtain pencil-like rods of ivory is that it is a (
techn ique highly conducive to standardization. Cylindrical rods of roughly the same
(
diameter lead easily to the production of standardized bead-blanks and ultimately \
standardized beads. I have shown elsewhere (White 1989) the high degree of stan-
dardization in Aurignacian beads, a standardization that I believe to be motivated by
the anticipated arrangement of uniform , sewn-on beads on garments. In other words,
the visual impact of Aurignacian beads was not as individual objects, but as arrange-
ments constituted ofunifom1 elements.
Laboratoty experiments using Aurignacian blades of Bergerac flint, a fine
(
limestone grinding stone, powdered red ocher and liberal applications of water (es-
sential to softening and lubricating the surface of the ivory), produced stigmata and \
highly polished surfaces similar to those observed on Aurignacian ivory beads. Sig-
nificantly, while most Aurignacian beads show traces of hematite, few if any of them (
are profoundly stained. In my experience working pem1afrost mammoth tusk, indeli-
(
ble staining only occurred when powdered hematite was mixed with fat or oil. The su-
perficial nature of hematite deposits on Aurignacian beads supports the use of water
rather than fat as a softening/lubricating agent.
In my experience, there is no archaeological evidence before Sungir (dating to ca. (
28-25,000 years BP or earlier) for the preparation of whole tusks by softening (heat-
(
 )
) WHITE
106 MATERIALS AND MEANING IN PALEO LITHIC IMAGES 107
)
ing, boiling), which would have allowed the ivory to have been more easily worked.
) Even at Sungir however, the operational chain for bead production (White 1993)
seems not to have employed thoroughly softened ivory . Simple soaking of tusks in
)
water has only superficial effects. However, once the tusk is reduced to much thinner
) fTagments, such soaking can penetrate the entire thickness, making drilling, scraping
and gouging much easier. This soaking also works well on subfossi l tusk fragments.
) Comparison of our experimental sample with actual A urignacian production debris
) indicates clearly the use ofwater in the final stages of bead production.
In my experience with extant archaeological collections, there is no evidence in
) the Aurignacian for the oft-cited "groove-and-splinter" technique, so common in
late r periods. While grooving and splintering of fresh tusks is a feasible (if enor-
) mous ly tedious and time-consum ing) approach, the splinters that exist in Aurignacian
) assemblages show no traces ofhaving been incised out of the swface of the tusk. The
French sites that have produced the greatest quantiti es of iv01y ornaments and pro-
) duction debris have yie lded few significant tusk segments that might yie ld ins ight
into Aurignacian approaches to tusk reduction.
)
) Contexts of Production, Use and Disposal

) Lacking direct assoc iation between Aurignacian I beads and human skeletons, we
) are obliged to design researc h strategies to demonstrate first, that these were indeed
objects of suspension and second, how prec isely, they were suspended. A combi ned FIGURE 9. Wear facets on the inside of the holes of talc basket-shaped beads from Abri Blanchard, France.
) SEM/experimenta l replication program has already y ielded to us significant insight (Scale in mm).
into the attachment of Aurignacian basket-shaped beads. Sewn-on cloth ing beads are
) implied, but there is substantial variation in patterns of wear (Figure 9) on the surfaces Case Study II: Gravettian Anthropomorphic Figurines
) of bead-holes .
E lsewhere, I have proposed that sites w here beads were manufactured in quantity
) My goal in this case study is to illu strate how deta il ed observatio n and experi-
were special places on the landscape, perhaps loci of aggregations of othe1wise dis-
men-tation aimed at unde rstanding Uppe r Paleolithic fema le image production can
tant groups . This would ex plain the hypera bundance of exotic raw materia ls in these
) lead to new in sights into the soc ial, economic and ideat ional contexts of the objects
sites. I have also noted that adjacent sites show abundant but differing freq uencies of
themselves. In so doing, I seek to reemphasize (see White 1996b) and avo id some of
) bead production stages, perhaps indicating more complicated divisions of labor than
the dangers and misu nderstandings evident in recent writing on the subj ect.
previously imagined. There is substantial evidence that beads were worked down to a
) penultimate stage (e.g., perforated but not shaped by grinding/polish ing, stored as
such, and only finished at the mom ent at which they were to be sewn or strung). This The Sample
) may have allowed the ivory worker the fl ex ibility to create a fin al produc t ta il ored to
the size and form requirements of the moment. I am concerned here with anth ropomorph ic representation s, espec ially sculpted
) figurines , traditionally attributed to the G ravettian culture 28-22,000 yea rs BP. In
M ore than 25% of Aurignac ian I basket-shaped beads show preh istoric breaks,
) usually across the neck of the hole. This could eas ily be explained by beads broken off contrast to recent American writers on the subject (Rice 1981; Nelson 1990) , I do not
~garment during activity , or by the mending of garments at which time worn or bro- cons ider it acceptable to treat all of Upper Paleolithic female image ry as if it were a
) ken beads may have been removed and di scarded. However, if use and maintenance coherent whole. It is not 1 The Magdalenian and the Gravettian are worlds apart in
of beaded garments are suggested, so is intensive production of beads man ifest in term s of the diversity, form and context of fema le representations. However, Bisson
)
large quantities offabrication debris. Caching of large numbers of beads , or even of and others ( 1996) indicated that so-called Gravett ian fi gurines from Grimaldi cover a
) whole, decorated garments , cannot be excluded and might explain tpe high percent- much longer chronol ogical range. Dating problems notwithsta nding, the proportion
age of intact, unbroken beads. Unfortunately, excavation techniques ea rly in this cen- of dec idedly pregnant fema le images differs dra maticall y between these two cultures
) (Duhard 1993).
tury did not involve spatial provenienc ing of artifacts; thus, spatial ana lys is w ithin
) sites is imposs ible for the moment. My geographic scope over the past few years has been from the Atlantic sea-
board ofF ranee to the Don Valley ofEuropean Russ ia. I have examined, with varying
) degrees of precision, approximately 100 of these figurin es, including those from Gri-
maldi, Savignano, Sireuil, Tursac, Abri Pataud, Trou Magrite, Willendorf, Brassem-
) pouy, Lespugue, Dolni Vestonice, Predmosti , A vdeevo, Kostienki 1, and Gagarino.
)
)
 108 WHITE MATERIALS AND MEANING IN PALEOLITHIC IMAGES 109

Lying behind the previous lack of recognition of differences between the Gravet- 15 anthropomorphic figurines from Grimaldi (Liguria), Italy, seven of which had
tian and the Magdalenian are serious terminological problems that mask significant never been analyzed or published (Bisson & Bolduc 1994; Bisson & White 1996). In
differences in the technology of representation between the two cultures. In particu- concluding, I will attempt to place these in a broader European context with respect to
lar; the term "Venus," which is interpretive rather than descriptive, has been used in the technology of figurine production, use and disposition.
such an all encompassing way as to give the illusion that the repet1oire of female im- The 15 objects (Figure 10) in question were recovered from the sites ofBarma
ages in the Magdalenian is simply a continuation of that of the Gravettian. Nothing Grande and the Grotte du Prince between 1883 and 1895 by Louis Jullien. Presently,
could be further from the truth. seven of them are in the Piette Collection at the Musee des Antiquites Nationales at (
My position is that descriptive terminology should be based in representational Saint Germain-en-Laye outside Paris, one of them is in the Peabody Museum at Har-
techniques, rather than in presumed but undemonstrated function, meaning or view- vard , and the remaining seven are in private hands in Montreal, where Jullien emi- (
ing context. Even the terms statuette and figurine probably presume too much. Terms grated in 1898.
such as female bas-reliefin limestone, or sculptedfemale in ivory or woman molded (
and fired in loess are preferable, in that they provide an understanding of raw mate- Raw Materials
rial , technique. and subject, without embedding meaning in description.
The sample of 15 figurines breaks down as in Table 1: (
Female Representations as an Analytical Category
Tab le I The raw material breakdown for the 15 Grimaldi sculptures. (
A ques tion ne ver asked is whether female representations constitute a discrete, (
natural category of analysis. I find it troubling that associated imagery, and its rela- Raw Material Number of Figurines
Fossil ivory I (
tionship to female representations is seldom examined: It is noteworthy that in evety Bone or subfossil ivory 2
region in which female figurines exist in the Gravettian there are accompanying rep- Light green fibrous sellJentine 2
Dark green fibrous sellJentine 5 (
Teseritations of other subjects, from animal to geometric; and these vary greatly by re-
Dark green chlorite 4
gion. To my knowledge, the only scholar to have addressed this question with respect Yellow talc I (
to.female figurines is Henri Delporte, who notes an association in Siberia between im-
ages of women and birds; and in France between women and bison.
All of the soft stone materials are from relatively local alpine sources, that is,
Chaine Operatoire for Gravettian Female Sculptures within 100 km from the Grimaldi sites. Hardnesses vary from 1 to 4 on the Mohs
scale. Several other objects from the Grimaldi sites, especially beads and pendants, (
For our purposes, the general constituents of a chaine operatoire for female figu- are manufactured in similar soft stones.lv01y and soft stone are the same materials
rines might include the following, keeping in mind as previously stated that archae- that dominate the sample of Gravettian female images across Europe. Although the (
ologists usually address the components in reverse order: use of these silicates, such as talc, steatite and serpentine, can be explained by their (
I) Cultural assumptions about women. being relatively easy to work, they also have remarkable tactile qualities. Especially
2) Beliefs about the relationship between materials, representational acts, re- when polished, they are indistinguishable to the touch from the warm lustrous sensa- (
pre-sentational constructs and social/supernatural efficacy. tion of polished iv01y . I propose that, as in the case of Aurignacian beads discussed
3) Choice and acquisition of raw material. above, in our pursuit of the meaning ofGravettian female representations, we have (
4) Choice of subject matter. focused too little attention on carefully chosen and constructed tactile qualities that (
5) Organization of production, (social, temporal and spatial). had/have enormous evocative potential.
6) Combination of gestures and tool s into techniqu~s for figurine production that The ivory, constituted ofunmineralized to partially mineralized mammoth tusk, (
are coherent with the encompassing, regional technical system and that enable was probably obtained from known geological exposures (Giraudi 1981, 1983).
7) Representation of desired signifiers of social identity, age, reproduc"tive status, Mammoths were not present in Italy during the Late Pleistocene, although a small (
supernatural associations, etc. number of ivory objects, notably pendants, have been recovered from Upper Paleo-
~
8) Use ofthe female representations in (socially, esthetically and cosmologically) lithic sites in Northern Italy. Exchange with groups farther north is certainly a possi-
meaningful -acts. bility. <
9) Disposition of the figurines based on ideas about their residual power arid Certain raw materials are conspicuously absent, although they are key constitu-
efficacy. ' ents of organic and lithic tools such as spear points, awls, lamps etc. Notably, bone,
antler and limestone are absent. This severe choice may have nothing to do with
The Grimaldi Subsample workability, as some of the substances ignored are more plastic than those chosen for
figurine production. Raw material choice may have responded to the pursuit of visual
I wish now to illustrate the insight that can be gained by analyzing in detail a group and tactile qualities, and to the cosmology surrounding certain substances.
(
of figurines with the above operational scheme in mind. I will focus in particular on
(
(

(
 WHJTE MATERIALS AND MEANING IN PALEOLITHIC IMAGES Ill
110

a.

e.

b.

f.

Gl

d.

FIGURE I Qa-d. Fourteen of the anthropomorph ic figurines from Grimaldi, Italy. f iGURE I Oe-g.

)
112 WHITE MATERIALS AND MEANING IN PALEOLITHIC IMAGES 113
(

(
(
( '
(
(
(

1.

m.

J. I
\

(
(

n.
k. (
FIGURE I 0 1-n.
f'I GURE I Oh-k. <
(

(
(
 )
) 114 WHITE MATERIALS AND MEANING IN PALEOLITHIC IMAGES 1 15
)
Contribution of Raw Materials to Object Form However, the forming of the figurines is only part of the story. There is also clear
) evidence for burnishing, polishing and, seemingly, glazing (Figure 12). These latter
The form in which raw material was obtai1ied seems to have influenced in some techniques produce visual and, once again, tactile qualities not producable by stone
) tools alone. Both polished ivory and polished/glazed serpentine have very special tac-
cases at least the s1ze and form of the fini shed figurines . Octobon describes an un-
) worked se~entme pebble, oval in fonn and flattened in section like a skipping stone. tile qualities that may speak to one of the sensory means by which the figurines were
ThIS shapt u~:upport accounts for the flattened shape of the Femme au cou pe1fore , le experienced (White !996b). Moreover, the process ofheating talc dramatically deep-
) buste, Ia figu, me aplat1e: le figure and !a doublefigurine. For example, close exam i- ens its raw color throughout its thickness.
) natlon of Ia doublejigunne (F1gure lOa) reveals that the fissure separating the human
and the an!m~l figures was a natural flaw in the stone that conditioned the desi gn of Technological Perspectives on Hair and Clothing
) the_p1ec~. fhe mmuscule sc~le of the Gnmald1 spec1mens would be explained i fsup-
pOI ts fo1 the remammg figurmes were as small as the pebble described by Octobon . What technologies are codified in representation (e.g, hairstyles, clothing, etc.;
) cf. Dobres 1992) and how can careful technological analysis infonn such interpreta-
) Formal Attributes or the Grimaldi Fiaures tions (e.g., string aprons, hai r nets)? The Grimaldi figurines are essentially silent on
b
this point, in contrast to many of the Rus ian figurines , which make explicit reference
) All o: the l5 sculpted repres entations from Grimaldi appear anthropomorphic. to clothing and adornment. There is nothing in any of the Grim aldi objects to contra-
) The o~e1 all mventory w1th sexual/reproduct1ve attribution shows an obvious pre- dict the sense that nude women were intended.
dornmance ofp1 egnant women among complete figurines (Table 2). Three of the 15
) ha ve two faces shanng the same body. ln two cases these are human heads back to Other Subjects of Representation
back, whlle a third juxtaposes a human head with that of an animal. The three cases of
) two-faced figunnes IS, so far as I am aware, unique to "Grimaldi. It is also worth noti 1w Although there arc provenience problems, a small number of nonhuman repre-
that the several cases ofbiank visages are ve1y explicit; that is, faces are not simp!; sentati ons occur at the Grimaldi sites. These are essentially pierced talc objects with
II absent, empty faces are very carefully con tructed. patterned incisions similar to those found on at least two of the female figurines.
)! Tdbk 2. ex and reprod uct;vc status attributions tor the 15 Grimaldi sculptures.

)I - SQC2inc;;----
La 7eie m>groide (bmken)
Sex
- -.)- - - - _ _ _ _?St:lltl
___ ----
_-_-
_
Contexts of Use and Disposal

)I I.e Po!ichinelle F
Etght o f the 12 unbroken figurines from Grimaldi are perforated for suspension
pregnant (Figure 13) and others s uch asIa figurine en ivoire brun, have carved furrows sugges-
Lu Fig urine non decrite (bfoken) ?
) L "Hermaphrodite F
?
tive of suspens ion. Of course, suspension can take many forms in addition to the
pregnant
Le.Losange F
) La Femme au goitre pregnant wearing of the e objects on the body . For example, they may have been suspended in-
F pregnant
La Statuel/e en steatitejaune F side dwell ings attached to a11icles such as skin bags and baskets.
pregn ant
) La Femme au cou pe("(o re F pregnant
Like many ofthe Gravettian figurines recovered to date, the Grimaldi speci-mens
LC' busiC' (broken) F ')
were found (to the best of our knowledge) carefully pi aced in an area peripheral to in-
) LP Fig ure (pos, ibly animal) ? ?
!.a Figurine en il'oire bnm tense human occupation. They come from two sites, the Grotte du Prince and Banna
F pregnant
) Lo Fig uri11e oplarie F 0 Grande. While tho~e from th Barma Grande were recovered from occupational hori-
Ln Fig urine en ivoire Q I 'ocre rouge F zons, those from the Grotte du Prince were found in a smali niche adjacent to the main
pregnant
La Fig 11rine double
) a
La Femelie deux fetes
F pregnant cave . Mi chae l Bisson and I are working on comparisons of sediment on the figurines
F pregnant
with that still adhering to artifacts of known stratigraphic provenience. More interpre-
) tively, we have propo ed (B isson & White 1996) particular use contexts for the Gri-
) Tools, Techniques and Gestures maldi figurines, to which I now turn.
The Grimaldi sculptures, which are small and designed for suspension, fit the eth-
) nographic pattem of amulets or fetishes. However, the majority of human figurines
. Microscopic and experimental analysis reveal characteristic stigmata of tech-
) ntques of gougmg, scr~ptng, abrading and incising in the production of the Grimaldi made by living circumpolar peoples are significantly different from the Grimaldi
~gunnes (~tgure 11). fhe red undancy with which these techniques are applied con- figurines in gender ratio, a much higher frequency of facial and extremity detail, and a
) tJ tbutes to a remarkable consistency m breast and abdominal form, as well as hair much lower incidence of genital and abdominal prominence. Since the circumpolar
treatment. Techntcal dtfferences between ivoty and soft stone (chlorite/steatite/se r- ethnographic record is clear that recent human sculptures were used to promote fertil-
)I pentme) sculptmg at Grimaldi and more-or-less contemporaneous Russian sites ity, these contrasts stJengthen our doubts about a fertility magic explanation. Never-
theless, given the tendency of depictive amulets to represent unambiguously the
)I (where tvoty and marl predominate) could not be clearer. The Russian female ficrures
show a predominance of gouging and ~specially abrasion allowing for ~1Uch intended goal of the user (i.e., many hunting amulets were naturalistic sculptures of
)I smoother contours than the htghly incised and much more angular Grimaldi women.
)
)
 116 WHITE MATERIALS AND MEANING IN PALEOLITHIC IMAGES 117

!
(

{
FIGURE II. Dislinct stone tool marks on the surface of one of the Grimaldi figurines .

desired prey animals), it is most likely that the characteristics of the ?rimaldi figu-
-rines refer to a reproductive context, and that this context was childbirth 1tself
(
Use of the Grimaldi, and at least some other Gravettian female sculptures m the
context of childbirth is consistent with an archaeological context in which they are of-
ten found in clusters as ifcached away for future use; childbirth being an occasional
occurrence in small human groups. Moreover, the idea that the sculptures themselves
were perceived as having power is supported by recent finds from A vdeevo on the [
Russian Plain (Grigoriev 1996 and personal communication). There, in addition to
purposeful pit-burial of whole sculptures, sometimes more than one to a pit, Go-
vozdover and Grigoriev have found fragments of the same broken figure buned me-
<
FIGURE 12. Heal induced g laze on the abdominal surface of "Ia double.figurine. " (
ters apart in meticulously dug pits of a special, cone-like f01m . If the sculptures were
perceived as inherently powerful, it is easy to imagine that the disposal of broken ex- women, with no necessa1y inclusive or monolithic mea ning that derives from gender (
amples would have been attended by great care and ritual. . alone. Individual production probably accounts for the great variability of the figu-
Childbirth is both an emotionally charged and potentially dangerous event. It IS rines. Our interpretation also does not imply the subordination or commoditization of
predictable in its general timing (i.e., the average length of gestation), but unpredict- women as do the fertility goddess (Gimbutas 1989), paleopornography (Guthrie
able as to the timino0 of the onset oflabor, the sex of the offspring, and the survival of (
. 1979), and mating alliance (Gamble 1982) scenarios . Instead, we recognize the im-
the mother and/or child. We hypothesize that the Grimaldi figurines are best mter- portance of women in themselves, not just as sources of babies, since we suspect the (
pret~d as individually owned amulets meant to ensure the safe completion of preg- motivation behind these amulets was the survival of the mother rather than the baby.
nancy. Amulets employ the principle of similarity to influence the outcome of From this perspective, women are envisioned as taking active control of an important
uncertain events. They are often made by their owners, although they may also be ob- part of their lives using magical means that would have been entirely rational within
tained from shamans. Since the ethnographic record shows that in many societies (
their cultural context.
amulets are thought to gain power w ith age, the scu lptures may have been passed We conclude w ith the observation that the pregnancy symbolism of the Grimaldi (
from mother to daughter over a number of generations. figurines need not be their only symbolic meaning, although it would seem to be a pri-
This scenario also satisfies many of the legitimate demands of the feminist mary one. Clan or guardian spirits may also be invoked, particularly by parts of the {
critique. It does not require the figurines to represent a generalized concept of wom- body such as heads, hair, duplicate faces, associated animals, etc., that are not directly
anhood, but instead recognizes that they .may be produced by and for individual (
in volved in our notions of reproduction. In each case these additional referents can be

(
 118 WHITE MATERIALS AND MEANING IN PALEOLITHIC IMAGES 119

seen as statements by the makers of the figurines, perhaps with respect to the actual
spiritual source from which the amulet draws its power.

Conclusions

I believe that the above approach of close observation and analysis, conceived in
terms of a version ofLeroi-Gourhan 's chaine operatoireexpanded far beyond simple
production events to include broader social and cultural contexts in which those
events occur, serves as a substantial foundation for accessing domains of thought and
action in the distant past. In particular, a broad technological perspective, when com-
bined with indications of archaeological context and the prudent and restricted use of
relevant ethnographic analogues leads us to consider new interpretive frameworks.
This framework, I believe, has the merit of situating Aurignacian beads and Gravet-
tiilll!Epigravettian female sculptures, not in some broad, culturally mandated system
of bodily adornment, erotic or reproductive iconography, but in culturally appropri-
ate but vety personal and individual practices by which real people negotiated the
normal chailenges (physical, social and sp iritual) of everyday life.
To return to the original iss ue of meaning, I wish tor emphasize the multidimen-
sional nature of constructed and construed meanings, meaning viewed here not so
much as ideas but as highly contex tualized productions, performances, choices and
ski ll s. J hope that I have made obvious the futility of maintaining "art" and "tec hnol-
ogy" as distinct categories of analysis .

Acknowledgments

The author g ratefully thanks M. Pi erre Bolduc and Madame Laurence Jullien
Lavigne for perm ission to study the Grimaldi sculptures in their possession . He also
thanks Dani and Ofer Bar-Yosef ofHarvard University, and the directors and staff of
the Peabody Museum, Harvard University for permission and assistance in the study
of co llecti ons in the ir care. At various times assistance was provided by Michael Bis-
son, Henri Delporte, Marie-Helene Marino, Dominique Buisson, Martin Oliva,
Maria Govozdover, Natalia Leonova, Nicolai Praslov and Gennadi Grigoriev. He
w ishes to thank the Wattis Foundation, Jean and Ray Auel, and the editors, Meg Con-
key, Olga Soffer, Deborah Stratma nn , and Nina Jablonski for making this volume
possible .
.Fundi ng for this research was provided by New York University, the Institute for
Ice Age Studies, the American Council of Learned Societies, and th e National Sci-
ence Foundation. Any errors or omissions are the sole responsibility of the author.

Literature Cited

Bartolomei, G., A. Brogiio, P. Cassoli, M. Cremaschi, G. Giacobini, G. Malerba, A. Maspero,

M. Peresa ni & A. Tagiiacozzo. 1992. Risultati preliminari delle nuove ricerche al riparo di
Fumane. L 'Annuario storicco della Valpo!icella 1991-92.
FIGURE 13. The gouged perforation (top) and addi tional detail on the tiont of"la Figurine aplalie." Bisson, M. & P. Bolduc. !994. Previously Undescribed Figu1ines from the Grimaldi Caves.
Curr. Anthropol. 35:458-68.

j Bisson, M., Tisnerat, N. & R. White. 1996. Radiocarbon dates from the Upper Paleolithic of the
Sarma Grande. Curr. Anthropo!. 3 7: !56-62.

)
120 WHITE MATERIALS AND MEANING IN PALEOLITHIC IMAGES 121 (

Bisson, M. & R. White. 1996. Female imagery from the Paleolithic: The case ofGrimaldi. Cul- Peyrony, D. 1935 . Le gisement Castanet, vallon de Castel merle, commune de Sergeac (Dor- (
ture 16:5-64 . dogne). Aurignacien I et II. Bull. Soc. Prehist. Franr;. 32:418-43 .
Delpech, F. 1983 . Les Faunes du Paleolithique Superieur dans /e Sud-Ouest de Ia France. Ca-
(
Phillipov, A. 1983. Antropomorfniye statuetki is Mezina. Pages 34-38 in R.S. Vasil'evskii, ed.
hiers du Quaternaire, 6. CNRS, Bordeaux, France. Plastika i risunki drevnikh kul 'tur. Novosibirsk, Nauka, Russia. (
a
Delpo 1ie, H. & D. E3uisso n. 1991. Brassempouy: Les fouilles de 1988 1990. Bulletin de Ia So- Ritchie, C. 1969. fvo1y Carving. Barker, London, U.K.
ciete de Borda.! 16: 143-157. Rice, P. 1981 . Prehistoric venuses: symbols of motherhood or womanhood? J. Field A rchaeol. (
Didon, L. 1911. L' f.bri Blanchard des Roches (commune de Sergeac).: Gisement au rignacien 37:402-414.
moyen. Bull. Soc. His!. etA rchaeol. du Perigord 87:246-61 and 321-45 . Schlanger, N. 1994. Mindful technology: unl eash ing the chaine operatoire for an archaeology
Dobres, M.-A. ( 1992). Re-considering venus figurines : A feminist-inspired re-analysis. An- of mind. Pages 143-151 in C. Renfrew & E. Zubrow, eds., The Ancient Mind: Elements of (
cient Images, Ancient Thought: The Archae/ogy of Ideology. Calgary : Proceedings of the Cognitive Archaeology. Cambridge University Press, Cambridge, U.K.
1990 Chacmool Conference, pp. 245-262. Steward, J. 1955. Theory of Culture Change. University of lllinois Press, Urbana. (
Dobrcs, M. & C. Hoffman. 1994, Social agency and the dynamics of prehistoric technology. J. Taborin, Y. 1993. La Parure en Coquillage Paleolithique. XX!Xth supplement to Gallia
{
Arch. Me/hod and The01y I :211-58. Prehistoire. CNRS, Paris, France.
Duhard, J ~-P.I993. Rea/is me de / 'Image Feminine Pa/eolithique. Cahiers du Quatemaire, No. White, L. 1959. The Evolution of Culture. McGraw Hill, New York. (
19. CNRS, Bordeaux, France. White, R. 1989. Production complexity and standardization in Era ly Aurignacian bead and
Gamble, C.I982. Interaction and alliance in Palaeolithic society. Man I 7:92-107. pendant manufacture: evolutionary impli cat ion s. Pages 366-90 in P. Me liars & C. Stringer, (
Gimbutas, M.I989. The Language ofthe Goddess . Thames and Hudson, London; U.K. eds., Tlte Human Revolution Edinburgh University Press, Edinburgh, U.K. (
Giraudi, C. 1981. Presenza di depositi medio-pleistocenici intensamente deformati in Val __ _ . 1992. Beyond art: Toward an understanding of the origins of material representation in
Cerrina (Monferrato settentrionale). Geogr. Fis. Dinam. Quat. 4:69-74 . Europe. Ann. Rev. Anthropol. 21:537-64. (
. 1983. Provincia di Alessandria: I. Pontestura. Quadem{della Soprintendenza A rcheo- _ _. 1993. Technological and socia l dimensions of " Aurignacian-Age" body ornaments
logica del Piemonte 2: 143-144. across Europe. Pages 277-99 in H. Knecht, A. Pike-Tay & R. White, eds. , Before Loscaux: (
Grigoriev, G. 1996. Le travail de l'ivoire sur Ia plaine Russe: le casd' Avdeevo. in M. Menu, P. The Complex Record of the Early Upper Paleolithic. CRC Press, Boca Raton. (
Walter & F. Widemanri, eds. Le Travail etl 'Usage de I 'fvoire au Pa/eolithique Superieur, _ _. 1995a. Comment on symbol mediated marking in the Lower Paleolithic by R. Bed-
Centre Universitaire europeen pour les Biens Culturels, Ravello, Italy. narik. Curr. Anthropol. 37:623-625. {
Guthrie, D. 1979. Ethological observat ions from Palaeolithic art. Pages 35-74 in Bandi, H.-G. _ _. 1995b. Ivory perso nal ornaments of Aurignacian age: Technological , social and sym-
a
eta/., ed., La Contribution de Ia Zoologie et de I'Ethologie /'Interpretation de /'Art des bolic perspectives. Pages 264-267 in M. Menu, P. Walter& F. Widem ann, eds., Le Travail (
Ppeuples Chasseurs p,ehistoriques. Editions Universitaires Fribourg, Fribourg, France. elI 'Usage de I 'Ivoire au Pa/eolithique Superieur. Centre Univers itaire europeen pour les (
Hahn, J. 1972. Aurignacian signs, pendants and art objects in Central and Eastern Europe. Biens Culturels, Ravello, Italy.
World Archaeol. 3:252-56. . _ _ . 1996a. Les images feminines paleolithiques: Un coup d'oeil sur quelques perspectives (
. 1986. Kraft und Aggression: Die Botschaji der Eiszeitkunsl im Aurignacien Sud- americain es. in H. Del porte, ed. , Centenaire de Ia Decouverte de Ia Dame de Brassempouy:
----;/;utschlands. Archaeol. Venatoria, Tubingen, Germany . Ses Ancelres, ses Contemporaines, ses Herilieres. Sur Quelques Perspectives Americaines. {
Knecht, H. 1993. Splits and wedges: The techniques and technology of early Aurignacian ant - Etudes et Recherches Archeo/ogique de I'Universite de Liege. Liege, France. (
ler working. Pages 137-62 in H. Knecht, A. Pike-Tay & R. White, eds., Before Lascaux: _ _. 1996b. Comment on L. McDermott 's "Self-representation in Pavlovian, Kostenkian
The Complex Record of the Early Upper Paleolithic. CRC Press, Boca Raton . and Gravett ian female figurines during the European Upper Paleolithic." Curr. Anthropol.
Lechtman , H. 1977. Style in technology: Some early thoughts, Pages 3-20 in H. Lechtman & R. 37:264-267.
Merril, eds., Material culture: Styles, Organization and Dynamics ofTechnology Ameri- (
can Ethnological Society, St. Paul. (
Lemonnier, P. 1983. L'Etude des systemes techniques, une urgence en technologic culturelle.
Techniques et Culture I: 11-34. (
Leroi-Gourhan, A. 1943. Evolution et Techniques: /'Homme ella Matiere. Alb in Michel,
Paris, F ranee . {
. A. 1945. Evolution et Techniques: Milieu et Technique. Albin Michel, Paris, France.
(
- -.A . 1993 . Gesture and Speech. MIT Press, Cambridge.
MacGregor, A. 1985 . Bone, Antler, Jvo1y and Horn: Th e Technology ofSske/eta/ Materials (
. Since the Rotnan Period. Goon Helm, London, U.K.
Mauss, M. 1936. Les techniques d~ corps. J. d e Psych. Vol. 32. (
Nelson, S. 1990. Diversity of the Upper Palaeolithic "Venus" figurines and archaeological my-
thology. Pages 11-22 in S. Nelson & A. Kehoe, eds., Powers of Observation: Alternative
Views in Archaeology Archaeol~gical Papers of the American Anthropological Associa-
tion, Washington, DC. .!

r
(