Evidence for cave marking by Palaeolithic children Kevin Sharpe^ & Leslie Van Gelder^ Amongst the numerous images found on the walls of Palaeolithic caves, fluted lines, made hy fingers dragged through a skin ofivet clay remain some ofthe most intriguing. In their study of images at Rouffignac, the authors undertook experiments with a range of modern suhjects who replicated the fiutings with their hands. Comparing the dimensions ofthe experimentalfiutings with the originals, they conclude that the patterns on the roofofChamherAI at Rouffignac were made hy the fingers of children aged hetween 2 and 5 years old. Given the current height ofthe chamher, such children would have needed to he hoisted aloft hy adults. Who knows what lessons in art or ritual were therehy imparted to the young persons... Keywords: Palaeolirhic, cave art, finger flutings, Rouffignac Cave
Introduction A wall in Gargas Cave, France, shows a baby's hand held by that of an adult while colour is blown over them. Footprints of youngsters have been immortalised into the floors of Pech Merle, Chauvet, Tuc d'Audoubert and Niaux caves. All these sites also contain prehistoric art. Children were present in the caves, but did they actually produce art or at ieast deliberately create any oi the markings? Whatever the minor impressions of Palaeolithic children in caves, this image is often forgotten in fevour of the popular image from the Charles R. Knight type of picture that shows the proverbial cave-man painting beautiful images of animals - with women and children only looking on. Some specialists of prehistoric parietal 'art' believe that children did participate in its creation. Bednarik argues that juveniles were responsible for some ofthe finger flutings (the lines that human fingers leave when drawn over a soft surface) made in caves in southern Australia at least 30 000 years ago (Bednarik 1986a; 1986b; 1987-88; 1990). Such Palaeolithic flutings occur in caves throughout southern Australia, New Guinea, and south-western Europe. As will be pointed out below, however, the case Bednarik makes is more suggestive than definitive, relying on a methodology that requires fiirther refinement with torensics. This report introduces a reliable methodology with which to ascertain children's authorship of flutings. Unlike Bednariks, and Sharpe and Van Celder's (2004), earlier publications on the subject, definitive evidence is presented that children did indeed create prehistoric 'art\ In particular we demon.strate that young children were responsible for '
Gradtiatc College. Union Institute & University, Cincinnati, Ohio, USA: Harris Manchester College, University of Oxford, UK: lOShirelakeClose, Oxford 0X1 ISN, UK (Frnail: ksharpe@khsarpe.com) Walden University, Minneapolis, Minnesota, USA
Received: 14June 2005; Accepted: 14 October 2005: Revised: 7 December 2005 ANTIQUITY 80 (2006): 937-947
937
^
2
"5
I
Evidence for cave marking by Palaeolithic children
Station 0
50m Porch
Figure I. I'Lin oj Roujfigttac Cave showing the various d>amben (developed from Harriere 1982: Figure 2). This paper especially concerns Chamber A! near its terminus.
fiutings in Rouffignac Cave in the Dordogne, France. This conclusion leads to further questions and insight into the activities carried out in the fluted chamber.
Chamber Al of Rouffignac Cave Despite previous controversy as to the authenticity of the art in Rouffignac Cave, it is now generally accepted as Palaeohthic and the date usually given for it, based on stylistic comparisons of the animal drawings in the cave, is 13-14000 years BP, in the Middle Magdalenian. Some scholars, however, suggest a much older date of around 27000 years BP and others a much younger date; but the stylistic means of dating is now questionable given the '''C dates from Chauvet Cave (Bahn 1994; Plassard 1999). The flutings that form the basis of this study are those near the terminus of Chamber Al, about 300m from the cave entrance (see Figure 1). The fluted sub-chamber here can be divided into natural alcoves or side chambers, numbered consecutively Alcoves I-IV from the top to lower left (facing the cave entrance), then V-VII from the lower to top right. The flutings were made into a thin red clay coating the white limestone, cutting through the red to expose the white underneath. They cover much of the 150m~ of the ceiling of this sub-chamber (Plassard 1999) (see Figure 2). The floor of the sub-chamber comprises red clay (smooth and compacted where frequented), which goes up the walls to varying 938
Kevin Sharpe & Leslie Van GeUer
-d
o
figure 2. A small portion of the flutings in Chamber A}
heights. No long open wall-spaces exist in the sub-chamber, few flint nodules appear on the floor, and large ceiling spaces generally have few flints protruding. The ceiling averages 1.6m above the floor (Plassard 1999).
The flutings in Chamber Al Five researchers have previously examined the flutings in Chamber Al and published their conclusions. Nougier & Robert (1958) introduced the world to the prehistoric artefacts of Rouffignac Cave, including these particular flutings. They title flutings in photographs of the ceiling of Chamber A1 (what they call the 'Serpents Dome') with such words as 'serpent' and 'anthropomorph' (Nougier & Robert 1958: Figures 16-18). Barriere (1982: 205; KS transl.) writes similarly o\ the ceiling, ''unique in all of prehistoric art, offering. . . interlaced macaroni, serpentines, and easily distinguishable individual snakei. Plassard (1999: 78; KS transl.) writes of the ' multitude of single, double, or triple lines that zigzag and become entangled in a swirling mass'. Marshack (1977:311) singles out Rouffignac as having 'the most numerous and complex [collection of flutings] in any cave in Europe'. Chamber Al, in particular, 'has thousands of [them] criss-crossing,.. . a random melange of interlacing lines running in every direction. There is neither structure, pattern, image, nor composition in the accumulation'. The collection contains both long and short units (a 'unit' is the set of flutings drawn with a sweep of a hand or finger), some appearing geometric while others not, and some overlying others. Many cover or are under the beginnings or endings of others. Sharpe and Van Gelder (in the press) classify these flutings as of the 'Mirian' form, a type of fluting characterized by lower-body movement and multi-fingered units. These 939
Evidence for cave marking by Palaeolithic children
differ from other forms found in the cave (with only upper-body movement or singleflngered units), some of which comprise flutings that appear more sequential than those in Chamber Al. Though some of the Chamber Al flutings are 60cm or less in length, many show a lack of constraint (not in the composition, which may employ a restricted range of shapes, but in the use of the space fluted). This involves some upper-body movement, but also twisting at the hips, locomotion (some lines are over 2m long, extending well beyond the arm range of a stationary fluter), or shifting weight. Circles and some of the zigzags on the ceiling, for example, required the twisting of the body rather than only of the wrist and arm. The above assumes that these lines are indeed flutings, made by humans with their fingers. Plassard (1999: 77) suggests that the lines may have been made 'with a bundle of sticks reminiscent of fingers'. This is refuted by the different starting points of the lines of many of the units; the differing line widths between some of the units; the lines of some units separating to avoid small obstacles; the finger-like cross-sectional shape of the lines; no more than five lines existing per unit; the fourth or fifth line of a unit, when it shows, looking like the line made by a little finger or a thumb; and the often uneven spacing between the lines in the units. The genuine finger-fluting origin of the lines therefore appears a reasonable assumption. Though they have not been dated directly, the flutings in Chamber A1 are considered Palaeolithic, for several reasons: art in the cave is considered to be so; there are line flutings, including zigzags, next to, inside, underneath, and on top of drawn mammoths in other passages of the cave; mammoths are drawn in Chamber A near to Al; and flutings are elsewhere dated to the Palaeolithic.
Attributing flutings to children Without proof, several observers have casually noted the young-person-like impression of the flutings in C^hamber Al. Did young people - or even children - in fact make them? Were young people the authors of at least some Palaeolithic fiutings (and therefore some Palaeolithic 'art') in Rouffignac or elsewhere? If it is possible to say that young people created the Chamber Al flutings, these questions are answered. If it is also possible to provide the age range(s) for these fluters, the question is answered with more precision. The most likely source of leads for this research would come from Bednarik because, as noted above, he suggested that juveniles fluted in caves in southern Australia. However, his comments and data (from measuring the interline spacing of flutings; see for example Bednarik 1987-88) fail to prove the young age of the fiuters, because: • he does not publish how many fingers he measured in each fluted unit; • he does not appear to have made allowance for the facts that o fingers held apart can produce wider inter-digit spacing, o fingers lightly stroked across a surface may produce narrower flutings than those fluted with more pressure, and o different fingers of one hand can be of different widths; • it is not clear what ages he means by the term 'juveniles' (e.g. babies, toddlers, prepubescents, adolescents?); and 940
Kevin Sharpe & Leslie Van Geider
• he does nor appear to have compared his results with measurements of flutings made by Hving people of different ages to gain a definitive association of widths with ages. His comments ought therefore only to he taken as a starting point for further research about the correlation between fluting width and fluter age.
General methodology This investigation is part of a research programme that bases its methodology on the flutings themselves (Sharpe 2004; Sharpe & Lacombe 1999; Sharpe et at. 1998; 2002; Sharpe & Van Geider 2004; in the press). In comparison with the methods of several previous researchers (for instance, Nougier, Robert, and Barriere), it does not first introduce ideas as to the meaning of the flutings (e.g. as depicting animals, humans, or symbols) and their significance, and then look at the lines through those ideas. Our programme seeks to establish an objective and experimental approach to the lines by seeing what data can be gleaned about the marks themselves as they were made and, thereby, what the marks might reveal about their makers. It also involves experimentation to ascertain how the markings may have been made and limitations on them given their means of manufacture. Marshack (1977), though he defers to his predecessors, pioneers strategies for this type of research. Bednarik (1 986a), d'Errico (1992), and Lorblanchet (1992) are some of the researchers who follow Marshack and on whose work the current methodology continues to build.
Method This study assumes that the people who made the flutings were anatomically a similar size to modern people, which we feel is justifiable, given the anatomical studies of Cro Magnon (Delporte 2004; Stringer 1992: 248-51). Common sense suggests that a fluting's width may relate to the age of the fluter; narrow fiutings perhaps suggest that young people created them. The study tested this hypothesis by measuring the finger widths of modern people of various ages, and those of the flutings in Chamber Al of Rouffignac Cave, and then comparing the two sets of data. The following procedures were followed. 1. The fiutings made by fingers F2--F4 (the 3 central digits) were studied. This is because if only one- or two-fingered flutings were studied, it cannot be told with certainty what finger or pair of fingers were used to create the flutings. The marks made by Fl (thumb) and F5 {litde finger) can usually be singled out because they are characteristic: Fl tends to make a scratch mark - if any mark, and ir seldom shows - because it is held at an angle to the plane of the other fingers, and F5 tends to trail the others forming a less significant mark. Further, F2-F4 were studied because the measured widths of different single-finger flutings are too similar given the smallness of the measurement and the potential measuring errors encountered. 2. Units of fluted fingers held apart may be wider than tmits of the same fingers held together. Therefore, measurements (rounded to the nearest mm) were made across the width of F2-F4 held close together, and beyond their beginning where they are narrower 941
^ 2 Q> Âť2
Evidence for cave marking by Palaeolithic children
with the rounding ot the finger tops. Many impressions of F2-F4 held close together are found in Chamber Al. 3. Measurements were restricted to the units of F2'F4 in Alcoves I and VII of Chamber Al. 4. For the experimental flutings, subjects (of various racial and demographic backgrounds) drew their fingers (held close together) over smoothed clay and the widths of the narrowest point of F2-F4 were measured. The subject's gender and age were also recorded. This phase of the experimental research followed a preliminary one, indicative but not as reliable as the current one, where the subjects drew around their hands placed flat on paper, fingers close together (Sharpe & Van Geider 2004).)
Errors Errors can occur in the measuring and recording processes for a number of reasons. 1 he width of the fiutings may depend on the firmness of the medium and the pressure applied; a fluting in a soft medium may be wider the more pressure applied. Rounding to the nearest mm may be too gross for accuracy. Shrinkage (or expansion) of the medium over time may alter the flutings width. The fingers may not be right against each other or may overlap each other. The width of a unit may vary over its length. Measuring accurately without touching the fiuted medium is difficult and it frequently involves adopting contorted positions. A series of measures were taken, and observations noted, to help understand and minimise the extent of these sources of error. A study was made of 10 units of clay flutings from the hands of two individuals (one male, one female) who applied different pressures (the results appear below). The results were looked at fot consistency in the width of fingers: ideally, a small number of fluters should lead to distinct narrow clusters of fluting widths. That the medium in Chamber A1 is now firm yet easily marked suggests its current state of hydration probably differs little from when it was fiuted. Measurements of fluting width were taken at the narrowest part of the unit away from the beginning of the unit, and where overlap was not obvious (overlap reduces the width of the overlapped finger relative to that of the overlapping finger). Lastly, the same person (LVG) made all the field measurements.
Results The following tables record the results of the clay fluting measurements of modern subjects, of the flutings on the ceiling of Chamber Al, of a comparison between the first two tables, and of the attempt to quantify the extent of the errors from different pressures. The study assumed that the impressions left by an individuals fingers from the left and right hands are symmetrically much the same size.
Discussion It was very hard and frequently impossible to have many children younger than 2-3 years old make fluted lines. They seemed to lack the ability to understand the command and to hold and control their hands in an appropriate manner. The best they could usually do 942
Kevin Sharpe & Leslie Van Gelder Table 1. For modern subjects, the narrowest widths (in mm) of F2-F4 close together, with the age (in years), number and gender of the subjects Individuals & Gender
Width
Age
Individuals & Gender
2
IF
5
4F
39 39 39
9 14 15 48 5 9
IF IM IF IM IM
Width
Age
22
23 26 27
2
lM
5
2F.2M
28 29
5 5
2M IF
30
5 5
2F, 2M
31 3t 31 32 32
33 33 34 34 34 34 34
35 35 35 35 35 35 35 35 36 36 ?>G 36 37 37 37 37 37 37 38 38 38 38 39
13 14 5 7
39 40 40 40 40 40 40 40 41 41 41
3M IF IF 2M
44
IF
2F IF IF
6 9
2F.2M
IF
42 42 42 42 42 42 42 42
10 11 12
IF IF
44 44
10
14
IF. IM
IF
44
14 15
IF, lM
45 45
5
IF, 2M
12 14 15 6 9 10 11 12 15
IF IF
15 9 13 14 46
2M 2M IF IM IM IF IM IM IM IM IM IM IF IM
6
IF
5 6
IF, 4M 2F, 2M
14 15 16
IF IF lM lM
5
6 9 14 15
5
IF
45 45 45
IF, lM IF, 2M IF, IM IF. IM
46 46 46 47 48 48 49 49 49 51
IF 3F IM IF, IM 3F 4F 2F IM
51
14 26 33
IF, 2M IF, IM
2M IF IM 2M IF IF IF IF IM IF IM 2M
5
lM IF, 2M
10 12 14 15
9 10 12 14
15 27 48 52
55 14 17
35 14 14 15 13 18
36 12 14
IM IF !M
was to smack the clay with the fingers of their open hand, even when assisted by an adult. Given that some children of this age could sometimes flute in the desired way, however, it is necessary to include this age bracket in the considerations.
943
13
I
Evidence for cave marking hy Palaeolithic children Table 2. For the narrowest point of flutings in Chamber A1 (Alcoves I and VII), the widths (in mm) of F2-F4 close together, with the number of flutings of such widths (widths not listed had no flutings) Width
Number
Width
Number
23 24 25 26 27
19 1 1 15 8
^ 31 33 34 36
2 1 1 1 1
Previous studies {Sharpe & Van Gelder 2004), as well a.s Table 1 above, suggest that by the time people have reached their teenage years, or even earlier, rhe finger widths have become adult-sized. In addition, for this study, the fluting width measurements for females and males can be combined (there being no significant difference between rhem). The study relating to possible errors from different pressures (see Table 4) shows the mean widths of the flutings of Fluter 1 and Fluter 2 are 40.5 Âą 0.5mm and 36.5 Âą 0.5mm respectively (disregarding the results marked*). This means that the error arising under different pressures and slight differences in finger closeness (subjectively unnoticed) may amount to about Âą 0.5mm. Further, the consistency ofthe Rouffignac results (significantly clustered around 23 and 26.5mm; see Table 2) suggests the methodology may overcome many ofthe potential errors. The age range for the fluters of Chamber Al suggested by Table 3 {working from Tables 1 and 2) is 2-5-years old, though one fluting may have been made by a person between that age range and l4-years old, and another between it and 16 years. With the error margin from Table 4 taken into account, these results remain the same. What do the flutings mean? This is the question of prime importance to most casual viewers and dedicated researchers. The flutings in Chamber Al have sometimes been seen as anthropomorphs (Nougier 1958), macaroni (Barriere 1982), meanders (Marshack 1977), serpentines {Barriere 1982), snakes (Nougier 1958), or related to water (Marshack 1977). More generally, flutings are also considered male symbols (Leroi-Gourhan 1958), related to initiation ceremonies (Bednarik 1987-88; Flood 1996), or to shamanisric ritual {LewisWilliams 2002). Mulvaney and Kamminga {1999: 365), commenting from Bednarik's work, consider that 'it is most likely to he play, children's "finger painting,". . . [or] done for decoration or identification, perhaps associated with rituaU. Flood {1996: 21) comments that they 'may simply mean that juveniles were more adventurous in exploring remote, hazardous places'. Other words that could be used are notation and symboling (following from d'Errico and Marshack), and iconography (an extension of Munn 1973). All these suggestions as ro the meaning of the flutings are speculative. They may and perhaps ought to lead to empirical research on the flutings, but to date none has been informed by such in-depth studies. The above results about young children and flutings in Chamber Al, however, help in answering the question of meaning. The age ofthe fluters rules out, for instance, initiation ceremonies at puberty. Similarly with the shamanistic interpretation: it may be unreasonable to consider 2-5-year olds shamans (though they still may have been acting within a ritual or otherwise shamanistic context). Besides being able
Kevin Sharpe & Leslie Van Gelder Table 3. For flutings in Chamber Al (Alcoves I and VII), the widths (in mm) of F2-F4 close together, with ihe number of flutings of sucb widtbs, and the probable age range of the fluter as indicated in Table 1 (from tbe youngest age at that or a larger widtb to the oldest age at tbat or a smaller width) Width
No.
Ages
Width
No.
Ages
23 24 25 26 27
19 1 1 15 8
2-5 2-5 2-5 2-5 5
28
2
31 33 34 36
1
5 5
1 1 1
5-14 5-16 5-16
Table 4. For the clay fluting study relating to possible errors from different pressures, for two subjects (male and female), the widths (in mm) of F2-F4 close together, listed by the fluting made (10 for each subject), against the narrowest (N't) part of the fluting (not the very top) and at 5cm intervals. "Indicates flutings where the fluter noted either a separation or overlap in the fingers. Fluter 1 (Male)
Fiuter 2 (Female)
Fl.#
N't
5cm
10cm
15cm
20cm
Fl.#
N't
5cm
10cm
15cm
20cm
1 2 3'
41 40 46 41 40 43 41 40 38 40
42 42 47
42 40 45
42 41 46 41
1 2* 3*
36 38 39
4
37 37
44 45 45
44 45 42
6
40 41
40
36 36 36 37 37
.37 41 39 37 37 36 36 36 37 37
56 39 40 37
40
41 41 46 43 41
36 39 40 37 38 37 37 36 37 38
36 40 39 38 38 37 37 37 37 37
4 5 6*
7 8 9' 10
43
41
43 41 44 41 40 40 39
40 43 42
7
40
8 9 10
39 39
37 37 37 36 37 37
to rule out some of the previously suggested connotations, what the fluters meant by their activities remains unknown; it will probably never be known and should probably not be expected to be known.
Conclusions Young children aged 2-5 niade many of the flutings in the fluted sub-chamber of Chamber Al in Rouffignac Cave. This is the first demonstrated case of young children creating Palaeolithic parietal 'art'. Given that this can be ascertained with a high degree of probability based on the physical evidence of the flutings, further matters present themselves for research and other information may be learned about the fluters. For instance, an aspect of Chamber Al to notice is the height oi the ceiling above the floor. The ceiling flutings are now in places just reachable by a man of 1.8m stretching up. It is unreasonable to think that young children marked unaided at such heights, yet the fluting size in some such places is small. Was the height of the ceiling above the floor at the time of fluting much the same as now? If so, or if tbe height were greater than now, the childreti would have had to have been held up to flute. In what direction did the children face when held aloft? Were the children acting as 945
O 'S
Evidence far cave marking hy Palaeolithic children
'paintbrushes' for those holding them up? Were the people holding up the children moving in some prescribed manner, such as in a dance? If so, could their feet and body movements bc reconstructed from the fiutings? Why did those holding up the children to fiute do this? The youngsters could have fluted where they could reach and the holders {if older people) could have marked, not only these sectiotis, but also sections where the youngsters could not reach. Here, however, they raised the children up to flute (and in some alcoves added their own fiutings). Further, the low sections ofthe ceilings that young children could comfortably flute by themselves usually show few or no flutings. While the archaeologist ought not to approach fiutings with dogmatic ideas as to what they mean, the fiutings' illusive meaning should not deter us studying them. They can offer a rich source of information about the behaviours ofthe fluters - Butings tell us about the fingers and hands that made them and these tell about people. Similar methodologies are now being applied to other fiutings in Rouffignac and elsewhere, relating information not only about the ages of the fluters, but also about data such as the fluters' genders and the number of individuals involved. At least three other forms of fiutings besides the Mirian Form exist in Rouffignac {Sharpe & Van Gelder, in the press) and work continues on them in Rouffignac and Gargas caves, to see if it is possible to elucidate further the behaviours and individuals behind their manufacture.
Acknowledgements Thanks are due to the many pt'ople who have helped support this rL'sesrch: Pat Brownley, Mary I^cotnbe, Sharleen Maddox, Ted Opderbeck, Nix Rowt- and Miriam .Sharpc; Jean and Marie-Odile Plassard, for discussions, their support and permission to work in Rouffignac Cave; Scverine [iesbordcs. Frederic (Jotirsolle and Frederic Plassard, for discussions and guiding and assistance in the cave; Union Institute & University, for financial support through its faculcy research grants; the pupils cif Clearview C'hristian Girls School. Pukalani, Maui, Hawaii, Franklin Sherman Elementary, McLean, Virginia, Preston Park Primary Schoo!, Wemhiey. Middlesex, anci Waldwick High School, Waldwick, New Jersey for hand width data; and Robert Bednarik, Jean Clottes, Francesco d'Errico, Sandor Callus (now deceased), Michel Lorblanchet, Alex Marshack (now deceased) and Hailam Movius Jr. (now deceased), for discussions and support over many years.
References
DEIPORTE, H.J. 2004. Human Evolution, in Encyclopedia Britannica, Encyclopardia Britannica Premium .Service, http://www.hritannica.com/ eh/article?eu= 127622 (accessed 18 June 2004). D'ERRICO, F. 1992. Technolo^, Motion, and the Meaning of EpiPalaeolithic Art. Current Antbropology 33 (1): 94-109.
BAHN, P.G. 1994. Some New Developments in Ice Age Art. Complutum 5: 197-202. BARRikRE, C. 1982. LArt Parietal de Rouffignac: La Grotte aux Cent Mammouths. Paris: Picard. BEDNARIK, R.t;. !986a. Parietal Finger Markings in Europe and Australia. Rock Art Research 3 (1): 30-61. —1986b. Cave Use by Au.stnilian Pleistocene Man. Proceedings ofthe University of Bristol Speleological Society \7{5): 117-A5. -1987-88. The Cave Art ofWestern Australia. The Artefact 12: 1-16. -1990. The Cave Petroglyphs of Australia. Australian Aboriginal Studies 2: 64-8.
946
FLOOD, J. 1996. Culture in Early Ahoriginal Australia. Cambridge Archaeological Journal 6 (1): 3-36. l.FROi-CotJRHAN, A. 1958. I,a Fonction des Signes dans les Sanctuaires Paleolithiques. Bulletin de la Societe Prehistoriques Fran^aise 55: 307-21. LEWIS-WILLIAMS, D . 2002. The Mind in the Cave: Consciousness and the Origins of Art. Lojidon; Thames & Hudson.
Kevin Sharpe & Leslie Van Gelder r, M, 1992. Finger Markings in Pech Merle and their Place in Prehlsroric Art, in M. Lorblanchec (cd,) Rock Art in the Old World: 451-90. New Deihir Indira Gandhi National Center for the Arts. MARSHACK, A. 1977. The Meander as a System: The Analysis and Reeognition of iconographic Units in Upper Palaeolithie Compositions, in P.J. Ucko (ed.) Form in Indigenous Art: Schematization in the Art of Aboriginal Australia and Prehistoric Europe, Prehistory and Material Ctilture Series, No. 13: 286-317. Canberra: Atistralian Instittiteof Aboriginal Studies. MuLVANEY.J. &J. KAMMINGA. 1')99. Prehistory of Australia. Washington (DC): Smithsonian Institiitioti Press. MuNN, N.D. 1973. Walbiri Iconography: Graphic Representation and Cultural Symbolism in a Central Australian Society. Ithac;i (NY): Cornell University Press. NOUGIER, L.-R. & R. ROBERT. 1958. The Cave of
Rouffignac. Transl. David Scott. London: George Newnes. PLASSARD, J. 1999. Rouffignac: Le Sanctuaire des Mammouths. Paris: Seuil.
t:, K. 2004. Line Markings: Htiman or Anitnal Rock Art Research 2\ (1): 57-84.
—, & M. LACOMBE. 1999. Line Markings as Systems of Notation? In News 95: International Rock Art Confess Proceedings: 46. Pineralo, Italy: IFRAO — International Federation oFRock Art Federations and NEWS 95 - International Rock Art Congress Proceed ings_files/sharp.htm. - , M. LACOMBE & H. FAWBERT. 1998. An Externalism
in Order to Communicate. The Artefact 21: 95-104. - , M. LACOMBE & H, FAWBERT. 2002. investigating
Finger Flutings. Rock Art Research 19 (2): 109-16. - & L. VAN GELDER. 2004. Children and Palaeolithic 'Art': Indications from Rouffignac Cave, France. International Newsletter on RaekArt 38: 9-17. - &: L. VAN GELDER, in the press. Trois Formes de Traces Digitaiix (ou Severines) en Grotte de Rouffignac, France. Prehistoire du Sud-Ouest. STRINGER, C^B. 1992. Evoltition of Harly Humans, in S. Jones, R. Martin & D. Pilbeam (ed.) The Cambridge Encyclopedia of Human Evolution: 241-51. Cambridge: Cambridge University Press.
947
-d
I