New views on old hands: the context of stencils in El Castillo and La Garma caves (Cantabria, Spain).

Pettitt, Paul ; Castillejo, Alfredo Maximiano ; Arias, Pablo 等

[ILLUSTRATION OMITTED]

Introduction

Hand stencils form one of the most recognisable categories of European Palaeolithic 'cave art. Since their initial study by Cartailhac (1906-1909), Capitan (1911) and Breuil (Alcalde del Rio et al. 1911; Breuil 1952), research into their nature and meaning, particularly over the last half century, has essentially repeated questions as to what the age and gender of the stencilled individuals were and why some fingers or parts of them appear missing. We take here a contextual approach to the production and location of hand stencils in the caves of El Castillo and La Garma in Cantabria (henceforth 'stencils'--which are far more numerous than positive prints). These caves are in relatively close proximity and contain a relatively large number of stencils. Our rationale is to look past questions about the possible identity of those stencilled and reformulate our approach around their context in the caves. What follows is the first detailed examination of the physical context of stencils. Our resulting interpretations are preliminary, but reveal the potential of a new concept--palpation--for understanding cave art more widely.

Hand stencils in Palaeolithic cave art

In many societies the hand serves as a cultural icon. Although modern Homo sapiens societies tend to prioritise spoken language above hand-based communication, enough use is made of hand communication by small-scale societies to suggest that this has a considerable antiquity, and thus that it may have played an important role for Palaeolithic hominins capable of creating and understanding symbolic systems (e.g. Capitan 1911; Leroi-Gourhan 1967). Examples of hand stencils (outlines created by projecting fluid pigment over a hand placed against a cave surface) and prints (made by pressing a pigment-covered hand against a surface) are known in a variety of countries and areas, notably Argentina, Australia, Borneo, Mexico, Peru, the Sahara and the USA (Arizona, California and particularly Texas), all apparently of Late Pleistocene age or younger. European Upper Palaeolithic examples--the focus of this study--are known in France, Spain and Italy. No precise quantification of the number of stencils/prints and the number of caves in which they occur is known to us: we count them in 44 caves, of which 30 are in France, 13 in Spain and 1 in Italy (for details and references see online supplementary material). Among these the greater majority are in southern France (Dordogne, Lot, Ariege) and northern Spain (Cantabria), although isolated examples are known as far north as the Grande Grotte at Arcy-sur-Cure, Burgundy (Baffler & Girard 2007) and as far south as Ardales near Malaga (Espejo Herrerias & Cantalejo Duarte 2006).

Breuil (1952: 38) assigned prints and stencils to his Aurignacian-Perigordian cycle on stratigraphic grounds and their lack of association with anything other than dots or simple animal outlines, and although Leroi-Gourhan occasionally assigned them to a later phase (e.g. his Style III/Solutrean-Early Magdalenian for Tibiran; Leroi-Gourhan 1968: 321) most subsequent researchers, basing their conclusions on radiocarbon measurements and U-series measurements on stalactites overlying stencils support the view that they are Gravettian or earlier (e.g. Lorblanchet 1995: 245-46; Ripoll Lopez et al. 1999a: 13; Foucher et al. 2007: 83; Pike et al. 2012.). For the purposes of this paper, however, the only assumption one needs to make is that the stencils in each cave are broadly contemporary and thus can be interpreted as a thematic whole.

In only a few cases have stencils been meticulously studied and published, notably at Gargas (Sahly 1966; Barriere & Sueres 1993; Foucher & Rumeau 2007), Cosquer (Clottes et al. 1992; Clottes & Courtin 1996: 69-79) and Maltravieso (Ripoll Lopez et al. 1999a & b). Wider research has concentrated entirely on the identity of hand stencils rather than their context, i.e. the possible gender and age of the people whose hands were depicted, whether the left or right hand was shown (e.g. Faurie & Raymond 2004), and why in some caves fingers or parts of them appear attenuated, i.e. missing or bent back (in the last half century, for example, Janssens 1957; Sahly 1966; Hooper 1980; Wildgoose et al. 1982; Barriere & Sueres 1993; Ripoll Lopez et al. 1999a; Guthrie 2005: 114-32; Rouillon 2006). Most researchers agree that the left hand was overwhelmingly stencilled; that the ratio of lengths of ring and index fingers--which indicate gender from the age of two (Manning et al. 1998)--is often consistent with female hands, and that there is no reason to assume more than a small number of individuals are represented in each cave. Attenuated (usually described as 'missing') fingers in fact belong to stencils in only a small number of the caves, and the focus on these presumably resulted simply from their abundance in Gargas and neighbouring Tibiran. Given the relative rarity of attenuated fingers (a term we favour due to its neutrality) among this relatively rare category of parietal art one really should not be preoccupied with them.

New research in La Garma and El Castillo

We have undertaken an informed analysis of the context and positioning of stencils in La Garma and El Castillo, taking a number of observations and measurements of each visible stencil. By focusing on context we hope that we will widen our understanding of stencils and by so doing widen the appreciation of 'cave art' through exploring objectively processes of individual decision-making made in a very specific and constrained environment. Given that the stencils appear to be relatively early in the chronology of European Palaeolithic art we hope that this new perspective will contribute to current debates as to the origins and development of cognitively 'modern' behaviour and art in general.

[FIGURE 1 OMITTED]

El Castillo, in the village of Puente Yiesgo, and La Garma, just north of the village of Omono, lie around 29km apart, to the south-west and east of Santander respectively (Figure 1). The c. 300m-long cave of El Castillo can be divided into two areas; a large entrance chamber (the Gran Sala), and a subsequent labyrinth of narrow galleries totalling almost 1km in length. Much of its art is figurative and can be attributed to the Solutrean or the Magdalenian, although its hand stencils belong to an earlier group which includes dots and lines, the antiquity of which was recently attested by the U-series dating noted above (Pike et al. 2012). Most of the stencils--at least 44 of which 38 are clear--cluster in a panel in a narrow gallery beyond the Gran Sala (the Gallery of the Hands) although isolated examples exist further into the cave's depth (Alcalde del Rio et al. 1911; Leroi-Gourhan 1968: 333-34; Figure 2) and as many surviving examples are faint one cannot rule out that their absence from the Gran Sala is due to their disappearance due to atmospheric erosion (Alcalde del Rio et al. 1911: 117). The c. 300m-long Lower Gallery of La Garma cave, 12km from Santander, can be divided into nine zones on the basis of the cave's topography. Contrasting patterns of spatial distribution distinguish the early (Pre-Magdalenian) and late (Middle Magdalenian) stages of Palaeolithic activity so far identified in the cave (Gonzalez Sainz 2003; Ontanon 2003; Arias et al. 2011). The former stage (including the stencils, series of red dots and other simple paintings, as well as animal representations also in red) occurs along the entire gallery; the later stage (comprising the remains of habitation structures and numerous paintings and engravings of Magdalenian style) is concentrated in the areas that are close to the cave's original entrance (zones I-IV, spanning roughly one third of the gallery's length). Recent pigment analyses of a number of examples of the cave's art has reinforced this division, demonstrating that the hand stencils are linked to non-figurative dots and lines rather than to the more diverse figurative images of the Magdalenian (Arias et al. 2011).

[FIGURE 2 OMITTED]

We examined as many hand stencils as were clearly visible in La Garma (n = 27) and El Castillo (n = 38) (Figure 3), taking a series of observations on each, notably: colour (in both caves always red); left or right hand (assuming a palm down position); the length (mm) of each observable digit; the width of palm (ram); the orientation of the hand (expressed as a point in a 360[degrees] circle); the angle between the thumb and the index finger; and the height above the cave floor. In addition to these we recorded contextual information about the location and associations of each hand stencil, specifically: the type of surface (e.g. flat, smooth, modified, bright, dull); the presence/absence of specific features (notably stalactite, concave or convex surface, cracks) and relation of these to the stencil; any artistic association (i.e. whether other stencils or other art is located close by); the wider positioning of the stencils (i.e. chamber type, proximity to shafts, stalactite columns and other features); and the specific position of stencils in relation to localised cave wall morphology, e.g. convex or concave surfaces, cracks and the like. In this paper we concentrate on these contextual associations.

[FIGURE 3 OMITTED]

Stencils: context and associations

Through the experimental replication of stencils we created a coded scheme reflecting the ease or difficulty of producing stencils. This confirmed that the most comfortable position in which to produce stencils was at 45[degrees] to the left (of vertical) when producing a stencil of the right hand, and 45[degrees] to the right (of vertical) when producing a stencil of the left hand. Table 1 compares coded ease/difficulty with the orientation of stencils in La Garma and El Castillo. Although most stencils in these caves were created within the most practical and comfortable range of 45[degrees] (left) to 135[degrees] (right), a small number of exceptions were identified (one in La Garma and five in El Castillo), which were created in uncomfortable and impractical conditions even though more comfortable and practical locations were available in their immediate vicinity (e.g. flat surfaces around head height). In four of these cases the stencils are associated with features of the cave walls, to which we return below. Choices about the placement of stencils were not simply about practicality, therefore; some were created at chosen locations irrespective of whether they were comfortable or practical or neither. Some examples are extreme: Hand 28 at the far right of the Main Panel of Hands in El Castillo--the only example of an uncomfortable or impractical stencil without any obvious association to a feature of the cave wall--was created at a low height on a steeply sloping flowstone surface close to where it meets the cave wall. This is a remarkably uncomfortable and impractical position to assume, strongly suggesting that a second individual created the stencil, i.e. that stenciller and stencilled were separate individuals (Figure 4).

Little attention has previously been paid to the context of stencils. What information has been published is typically informal, with formal associations generally restricted to those with other examples of cave art, which may or may not be contemporary. Leroi-Gourhan (1968: 148), for example, noted an association between hand stencils and rows or clusters of dots and parallel strokes in Gargas, Les Merveilles (Rocamadour), Pech Merle, El Castillo and Bernifal, and their proximity to or location in the midst of 'central compositions' in Bayol, El Castillo, Pech Merle and Bernifal. He interpreted them in terms of his gendered symbolism, although he was probably grouping art of different ages into this holistic view and thus they are not meaningful associations. Otherwise, published associations with the topography of cave surfaces are few. Lewis-Williams (2002: 219) noted a 'bulging' of the rock face between two hand stencils of the dappled horses panel in Pech Merle, and in the Grotte Cosquer eight stencils cluster on a stalagmite drapery and another was placed in a natural niche (Clottes et al. 1992: figs. 6 and 7, respectively). Most importantly Gonzalez Garcia (1987) noted that the stencils of El Castillo were created on concave surfaces in his wider analysis of the context of the cave's art. It is obvious from published photographs that stencils could be placed in or above niches (e.g. in Gargas (Foucher et al. 2007: 41, fig. 37; Foucher & Rumeau 2007: figs. 49, 52, 67 & 68) and Cosquer (Clottes & Courtin 1996: 71, fig. 35)), in proximity to cracks (e.g. in Gargas (Bourges 2007: fig. 31)) and in Cosquer (Clottes & Courtin 1996: figs. 33 & 171)), atop stalactite draperies (e.g. in Cosquer (Clottes & Courtin 1996: figs. 31 & 38)) and in apparent association with cracks (Les Fieux (Lorblanchet 2010: 325)), but no systematic study has to our knowledge been undertaken of stencils per se. Table 2 presents hand stencils in El Castillo and La Garma that are found in obvious association with natural features, notably fissures, convex bosses and concave depressions, observed examples of which are defined in the table. In total 80% of observable stencils at La Garma and 74% at El Castillo have some kind of association, either with fissures or undulations on the caves' surfaces. As areas of 'smooth' rock were easily accessible in each cave and within close proximity to stencils, such associations cannot be entirely fortuitous. Some stencils seem to have been 'fitted' to subtle topographic features in the wall, and some were positioned on bosses in the wall in such a manner that they appear to be 'gripping' the wall in a similar way that explorers use their hands to steady themselves when navigating the caves (Figure 5).

[FIGURE 4 OMITTED]

From Table 2 it can be seen that a concern with 'gripping' convex rock surfaces is observable on around 27% of stencils in La Garma and around 23% in El Castillo. The palms of a smaller number of stencils--17% and 5% respectively--were 'economically' fitted to slight concavities on the cave surfaces. Even using modern lighting these can only be seen very close up, and under the lighting conditions in which they would have been created, such fitting would have necessarily involved touch as well as very close-up scrutiny of the surface. Overall there seems to have been more of a concern with fitting stencils to surface morphology in La Garma (44% have either gripping or ergonomic characters) than El Castillo (28%). By contrast, an association with natural cracks or fissures typically accounts for 39% of stencils in La Garma and 60% in El Castillo. Details differ between the two: in La Garma the concern is more with placing the stencil to the right of fissures (15%) or in between two fissures (9%), whereas in contrast a concern with placing stencils directly on top of fissures (29%) or with two stencils spanning fissures (11%) is clear at El Castillo. It is of interest that the few examples of stencils placed in uncomfortable and impractical positions (with coded scores of 4 or 5 in each category) possess clear associations: in El Castillo these are Hand 7 (placed atop a stalactite), Hand 13 (associated with a crack), Hand 35 (placed in a chimney accessed only up a steep slope) and Hand 36 (ergonomically fitted to a palmar concavity, with fingers gripping a boss).

[FIGURE 5 OMITTED]

[FIGURE 6 OMITTED]

Specific examples illustrate our observations. Six stencils comprise El Castillo Area 1, among which one was created from a comfortable position on a relatively flat surface atop several fissures, despite the 'availability' of unfissured rock nearby (Figure 6). In the Main Panel of Hands (our Area 3) one stencil was created over a small concave depression, and thus was ergonomically placed (Figure 7). An isolated hand in the Gallery of Dots was similarly ergonomically fitted to an otherwise smooth surface, the palm placed over a slight concave depression and the fingers 'gripping' a boss in the wall next to a natural fissure (our 'ergonomic, 'gripping' and 'right of fissure' positions) (Figure 8). In the Main Panel two stencils of a left hand (probably the same hand) were created in very close proximity to each other to either side of a natural crack (Figure 9). A more obvious pairing, again 'spanning' a crack, was created by left and right hands placed on an otherwise smooth surface (Figure 10). We can be reasonably sure that the present floor level approximates fairly closely the level of the floor at the time these were created: the lack of excavation or modification and the survival of stalagmite flows over the floor suggest this is so. Assuming this is correct the height of these stencils--directly above the head of the stenciller/stencilled--would have rendered them difficult to create.

[FIGURE 7 OMITTED]

By comparing these associations with the specific morphologies of the relevant hand stencils we hoped to establish a pattern of individual choice in the placing of the stencils. In most cases stencils were too faint to establish finger lengths and other measurements confidently. Where stencils are clear, finger terminations may still be vague and thus decisions about where to begin and end measurements probably involve an error in the order of a few millimetres; thus we are cautious not to over-interpret these results and accept that the little data we do have may underestimate the actual number of individuals who left stencils of their hands in the caves. Table 3 documents finger and palm measurements for Stencils with the most common associations in the two caves. At La Garma, the morphology of stencils in the gripping position seems to reflect four separate individuals; the index/ring finger ratios (sensu Snow 2006) of two are consistent with females (Hands 16 and 24) and one with a male (Hand 23). At least two individuals placed stencils to the right of fissures based on the length of their middle fingers (Hands 14 and 16, the small length of the former possibly indicative of a juvenile). Hand 16 was placed with the fingers in the gripping position, the palm ergonomically fitted and to the right of a fissure. At least two individuals placed stencils in positions bordered by fissures (Hands 15 and 23 and 24 which may belong to the same individual and are consistent with a male). At least three individuals left stencils with no association visible today (Hands 2, 5 and 13).

In El Castillo it is possible that Hands 4 and 6 in the gripping position were left by the same individual, although two other individuals seems to have adopted this position too (Hands 32 and 38, the latter of which has finger ratios consistent with a female). Little can be said for stencils placed atop fissures except that three individuals seem to be represented by Hands 1, 6 and 18, the finger ratios of the latter consistent with a male. Three individuals placed stencils spanning fissures (Hands 12, 19 and 20, the latter of which is consistent with a female). This is of interest, as Hands 19 and 20 form a left and right pair that otherwise one would assume to represent the same individual, but the morphology suggests that one person placed their left hand to the left of the fissure and a second individual placed their right hand to the right of the fissure. Given the distinct differences in ring, index and little finger lengths of these stencils it is unlikely that these differences reflect measurement imprecision. Three or four individuals left stencils with no association (Hands 15, 27 and 28, which may be the same individual, and 31) and of these the finger ratios of Hand 28 are consistent with a female.

[FIGURE 8 OMITTED]

[FIGURE 9 OMITTED]

[FIGURE 10 OMITTED]

Given the indistinct nature of most stencils one should not make too much of these data. It is of interest that few examples exist of the same individual represented in different areas, but this may well be masked by the indistinctiveness of most examples. What it does reveal, at least, is that several individuals were making similar decisions about the meaningful placement of stencils; it is not the work of one individual, but represents similar decision-making among a group.

Palpation: a new understanding of hand stencils

Under the best of lighting conditions the exploration of caves is a tactile process; hands are often placed on floors, slopes and walls to steady the body. In the extremely low light conditions that pertained for the stencillers they would be brought into very close proximity with any surface they intended to mark. Hands would be placed on surfaces, fingers and palms traced across them, and the undulations of the cave walls and ceilings could be 'read' as much by touch as visual inspection. We will probably never know exactly why decisions were made to mark walls with pigments or engravings, although we have identified here a plausible link between touch and marking. In order to fix stencils, red pigments were presumably selected for their visual 'warmth', i.e. their visibility in low light conditions (Groenen 1997). The use of cave topographies and shadows to place and structure figurative art is well known (e.g. Lejeune 2004; Pigeaud 2004; Remade 2004; Vialou 2004), and our results suggest that similar processes were at work in the creation of stencils, irrespective of whether the process was easy or difficult. To an extent these are visual processes, but stencils in gripping positions and particularly with palms fitted to very slight concavities indicate that visual stimuli were not the only way that stencillers interacted with cave surfaces. These reveal that a close-up scrutiny of cave surfaces must have preceded the creation of stencils in the majority of cases in La Garma and El Castillo. For this, we suggest the term palpation, deriving from the surgical term for examination by touch. Not that palpation governed only the production of hand stencils; finger dots and lines connect the artist's hands with the cave wall, and finger flutings--that appear to have been created irrespective of discomfort (Sharpe & Van Gelder 2006)--could be interpreted as the visible record of the act of palpation. Furthermore, Lorblanchet (2009) has noted the gradation between bear claw marks and finger 'rubbings' and traces in the caves of the Quercy, which he interprets as a ritual interaction with the cave walls. We have, so to speak, barely scratched the surface here, but hope that we have demonstrated the heuristic potential of examining touch and context in Palaeolithic art. Far more formal examination of the context of parietal art is necessary, as is the investigation of levels of illumination and its effects on visibility, topography and placement of art. There is clearly much more to 'cave art' than meets the eye.

Supplementary material is published online at http://antiquity.ac.uk/projgall/pettitt339/

Acknowledgements

We are grateful to the Gobierno de Cantabria for authorisation to conduct this research and for their kind assistance. It was funded by a Leverhulme Trust grant to PP which employed RH as Research Associate. We are grateful to Marc Groenen for his support for the project. Paul Bahn and Joao Zilhao kindly commented on a previous draft. Beatrice Greenfield, Ellie Walton and the students of the Department of Archaeology at Sheffield University gave up much of their time reproducing hand stencils, for which PP is grateful. Luis Teira provided the photographs.

References

ALCALDE DEL RIO, H., H. BREUIL & L. SIERRA. 1911. Les cavernes de la region Cantabrique (Espagne). Monaco: Imprimerie A. Chene.

ARIAS, P., E. LAVAL, M. MENU, C. GONZALEZ SAINZ & R. ONTANON PEREDO. 2011. Les colourants dans l'art parietal et mobilier paleolithique de La Garma (Cantabrie, Espagne). L'Anthropologie 115: 425-45.

BAFFIER, D. & M. GIRARD. 2007. La Grand Grotte d'Arcy-sur-Cure. Les Dossiers d'Archeologie 324: 74-85.

BARRIERE, C. & M. SUERES. 1993. Les mains de Gargas. Dossiers d'Archeologie 178: 46-54.

BOURGES, F. 2007. Les peintures prehistoriques des Grottes, les raisons d'une conservation exceptionelle, in P. Foucher, C. San Juan-Foucher & Y. Rumeau (ed.) La Grotte de Gargas. Un siecle de decouvertes-, 53-57. Mazeres de Neste: Communaute de Communes du Canton de Saint-Laurant de Neste.

BREUIL, H. 1952. Four hundred centuries of cave art. Paris: Sappho.

CAPITAN, L. 1911. Les empreintes de mains sur les parois de las Grotte de Gargas. Bulletin Anthropologique 1911: 87-88.

CARTAILHAC, E. 1906-1909. Les mains rouges et noire et les dessins paleolithiques de la Grotte de Gargas. Bulletin de la Societe d'Archeologie du Midi de la France 37:140-41.

CLOTTES, J. & J. COURTIN. 1996. The cave beneath the sea. Palaeolithic images at Cosquer. New York: Harry N. Abrams.

CLOTTES, J., J. COURTIN, H. VALLADAS, H. CACHIER, N. MERCIER & M. ARNOLD. 1992. La Grotte Cosquer datee. Bulletin de la Societe Prehistorique Francaise 89: 230-34.

ESPEJO HERRERIAS, M. & P. CANTALEJO DUARTE. 2006. Cueva de Ardales (Malaga). Reproduction digital del arte rupestre paleolitico. Guadalteba: Comarca del Guadalteba.

FAURIE, C. & M. RAYMOND. 2004. Handedness frequency over more than ten thousand years. Proceedings of the Royal Society of London Series B 271: S43-S45.

FOUCHER, P. & Y. RUMEAU. 2007. Les galleries ornees de Gargas, in P. Foucher, C. San Juan-Foucher & Y. Rumeau (ed.) La Grotte de Gargas. Un siecle de decouvertes: 61-86. Mazeres de Neste: Communaute de Communes du Canton de Saint-Laurant de Neste.

FOUCHER, P, C. SAN JUAN-FOUCHER & Y. RUMEAU (ed.). 2007. La Grotte de Gargas. Un siecle de decouvertes. Mazeres de Neste: Communaute de Communes du Canton de Saint-Laurant de Neste.

GONZALEZ GARCIA, R. 1987. Organization, distribution and typology of the cave art of Monte del Castillo, Spain. Rock Art Research A: 127-36.

GONZALEZ Sainz, C. 2003. El conjuncto parietal paleolitico de la galena inferior de La Garma (Cantabria): avance de su organizacion interna, in R. de Balbln Behrmann & P. Bueno Ramirez (ed.) El arte prehistorico desde los inicios del siglo XXL: Primer Symposium Internacional de Arte Prehistorico di Ribadasella: 201-22. Ribadasella: Association Cultural Amigos de Ribadasella.

GROENEN, M. 1997. Ombres et lumieres dans l'art des grottes. Bruxelles: ULB.

GUTHRIE, R.D. 2005. The nature of Paleolithic art. Chicago (IL): University of Chicago Press.

HOOPER, A. 1980. Further information on the prehistoric representations of human hands in the cave of Gargas. Medical History 24: 214-16.

JANSSENS, P.A. 1957. Medical views on prehistoric representations of human hands. Medical History 1: 318-22.

LEJEUNE, M. 2004. Quelques reflexions sur le role de la paroi rocheuse dans l'art du Paleolithique Superieur, in M. Lej'eune (ed.) L'Art parietal paleolithique dans son contexte naturel (ERAUL 107): 15-19. Liege: Universite de Liege.

LEROI-GOURHAN, A. 1967. Les mains de Gargas. Essaie pour une etude d'ensemble. Bulletin de la Societe Prehistorique Francaise 64: 107--22.

--1968. The art of prehistoric man in Western Europe. London: Thames & Hudson.

LEWIS-WILLIAMS, J.D. 2002. The mind in the cave: consciousness and the origins of art. London: Thames & Hudson.

LORBLANCHET, M. 1995. Les grottes ornees de la prehistoire: nouveaux regards. Paris: Errance.

--2009. Claw marks and ritual traces in the Paleolithic sanctuaries of the Quercy, in P.G. Bahn (ed.) An enquiring mind: studies in honour of Alexander Marshack: 165-70. Oxford: Oxbow.

--2010. Art parietal. Grottes ornees du Quercy. Parc-Saint-Joseph: Rouergue.

MANNING, J.T., D. SCUTT, J. WILSON & D.I. LEWIS-JONES. 1998. The ratio of 2nd to 4th digit length: a predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen. Human Reproduction 13: 3000-3004.

ONTANON, R. 2003. Sols et structures d'habitat du Paleolithique superieur, nouvelles donnes depuis les Cantabres: la Gallerie Inferieure de La Garma (Cantabrie, Espagne). L'Anthropologie 107: 333-63.

PIGEAUD, R. 2004. Dialogue avec la paroi: cas des representations paleolithiques de la Grotte ornee Mayenne-Sciences (Thorigne-en-Charnie, Mayenne), in M. Lejeune (ed.) L'Art parietal paleolithique dans son contexte naturel (ERAUL 107): 21-43. Liege: Universite de Liege.

PIKE, A.W.G., D.L. HOFFMANN, M. GARC(A-DIEZ, P.B. PETTITT, J. ALCOLEA, R. DE BALBIN, C. GONZALEZ SAINZ, C. DE LAS HERAS, J.A. LASHERAS, R. MONTES & J. ZILHAO. 2012. U-Series dating of Paleolithic art in 11 caves in Spain. Science 336: 1409-13.

REMACLE, L. 2004. De la grotte au(x) mythe(s). Phenomenes d'interaction et lecture de l'image paleolithique, in M. Otte, L. Oosterbeek, D. Seglie, L. Remade & V. Bertollo (ed.) Art du Paleolithique Superieur et du Mesolithique (British Archaeological Reports international series 1311): 7-14. Oxford: Archaeopress.

RIPOLL LOPEZ, S., E. RIPOLL PERELLO & H. COLLADO Giraldo. 1999a. Maltravieso. El santuario extremeno de las manos. Caceres: Museo de Caceres.

RIPOLL LOPEZ, S., E. RIPOLL PERELLO, H. COLLADO GIRALDO, M. MAS CORNELLA & J.F. JORDA PARDO. 1999b. Maltravieso. El santuario extremeno de las manos. Trabajos de Prehistoria 56: 59-84.

ROUILLON, A. 2006. Au Gravettien, dans la Grotte Cosquer (Marseille, Bouches-du-Rhone), l'homme a-t-il compte sur ses doigts? L'Anthropologie 110: 500-509.

SAHLY, A. 1966. Les mains mutilees dans l'art prehistorique. Toulouse: Privat.

SHARPE, K. & L. VAN GELDER. 2006. The study of finger flutings. Cambridge Archaeological Journal 16: 281-95.

SNOW, D.R. 2006. Sexual dimorphism in Upper Palaeolithic hand stencils. Antiquity 80: 390-404.

VIALOU, D. 2004. Architecture de l'art parietal paleolithique, in M. Lejeune (ed.) L'Art parietal paleolithique dam son contexte naturel (ERAUL 107): 7-14. Liege: Universite de Liege.

WILDGOOSE, M., E. HADINGHAM & A. HOOPER. 1982. The prehistoric hand pictures at Gargas: attempts at simulation. Medical History 26: 205-207.

Received: 23 April 2013; Accepted: 20 June 2013: Revised: 3 July 2013

Paul Pettitt (1), Alfredo Maximiano Castillejo (2), Pablo Arias (2), Roberto Ontanon Peredo (2,3,4) & Rebecca Harrison (5)

(1) Department of Archaeology, Durham University, South Road, Durham DH1 3LE, UK (Author for correspondence, email: paul.pettitt@durham.ac.uk)

(2) The Cantabria International Institute for Prehistoric Research, University of Cantabria, Edificio Interfacultativo, Avda. los Castros sin, 39005 Santander, Spain

(3) Museo de Prehistoria y Arqueologia de Cantabria, Avda. los Castros 65-67, 39005 Santander, Spain

(4) Cuevas Prehistoricas de Cantabria, 39670 Puenta Viesgo, Spain

(5) Department of Archaeology, University of Sheffield, Northgate House, West Street, Sheffield S1 4ET, UK

Table 1. Orientation practicality and comfort levels of stencils at
La Garma and El Castillo

                   Number (% of total)
                     of stencils

Orientation of    La Garma   El Castillo                   Practicality
stencil (1)       (N = 21)    (N = 37)     Comfort (2)     (3)

10[degrees]                   1 (2.7%)          4             4
45-145[degrees]   20 (95%)   32 (86.5%)         1             1
150[degrees]                  1 (2.7%)          2             1
170[degrees]       1 (5%)                       3             2
190[degrees]                  1 (2.7%)          4             3
200[degrees]                  1 (2.7%)          4             4
280[degrees]                  1 (2.7%)          5             5

(1) from an angle of 0[degrees] which is horizontal and
pointing towards the left (9 o'clock on the clock face)

(2) levels defined from experimental work:
1 = very comfortable; 5 = very uncomfortable

(3) levels denned from experimental work:
1 = very practical; 5 = very impractical

Table 2. The context and association of stencils at
El Castillo and La Garma with natural features
of the cave surfaces.

                            La Garma hand
                            stencils (recorded
                            by our stencil        El Castillo hand
                            numbers). (N = 29,    stencils (recorded
                            includes two thumb    our stencil
Category of association     stencils)             numbers). (N = 38)

No observable               6 (Hands 2; 3; 5;     10 (Hands 15; 17;
associations                11; 13; 19)           21; 22; 23; 25; 27;
                                                  28; 31; 34)

Ergonomic: the stencil is   5 (Hands 1;9; 16;     2 (Hands 26; 37)
matched to undulations/     18; 26)
curves of wall (typically
the palm is placed over
a minor concave
depression)

Gripping: the fingers of    6 plus 2 thumbs       9 (Hands 3; 4; 5;
a stencil placed over       (Hands 6; 10; 16;     6; 8; 32; 35;
('grip') a convex boss      23; 24; 27; Thumb     36; 38)
or ridge                    1; Thumb 2)

Below fissure: the          1 (Hand 9)            0
stencil is located below
a natural fissure/crack
when viewed upright (i.e.
fingers to top)

Bordered by fissures: the   3 (Hands              1 (Hand 24)
stencil is located          15; 23; 24)
between two natural
fissures/cracks, i.e. to
its left and right

Within multiple fissures:   2 (Hands 4; 17)       3 (Hands 9; 13; 14)
the stencil is in close
proximity to multiple
natural fissures/
cracks (>2)

Above fissure: the          1 (Hand 20)           1 (Hand 32)
stencil is located above
a natural fissure/crack
when viewed vertically

Atop fissure(s): the        1 (Hand 25)           11 (Hands 1; 2; 3;
stencil is placed                                 5; 6; 9; 10; 16;
directly on top of                                18; 29; 33)
fissure(s)

Atop stalactite: the        0                     1 (Hand 7)
stencil is placed on top
of a stalactite
(flowstone)

Spanning fissure: a pair    0                     4 (Hands 11 & 12;
of stencils spans a                               19 & 20)
natural fissure/crack,
i.e. one to either side

                            Percentage of
                            recorded stencils
                            by site (rounded
Category of association     up/down)

No observable               La Garma 20%
associations                El Castillo 26%

Ergonomic: the stencil is   La Garma 17%
matched to undulations/     El Castillo 5%
curves of wall (typically
the palm is placed over
a minor concave
depression)

Gripping: the fingers of    La Garma 27%
a stencil placed over       (includes
('grip') a convex boss      2 thumbs)
or ridge                    El Castillo 23%

Below fissure: the          La Garma 3%
stencil is located below
a natural fissure/crack
when viewed upright (i.e.
fingers to top)

Bordered by fissures: the   La Garma 9%
stencil is located          El Castillo 3%
between two natural
fissures/cracks, i.e. to
its left and right

Within multiple fissures:   La Garma 6%
the stencil is in close     El Castillo 8%
proximity to multiple
natural fissures/
cracks (>2)

Above fissure: the          La Garma 3%
stencil is located above    El Castillo 3%
a natural fissure/crack
when viewed vertically

Atop fissure(s): the        La Garma 3%
stencil is placed           El Castillo 29%
directly on top of
fissure(s)

Atop stalactite: the        El Castillo 3%
stencil is placed on top
of a stalactite
(flowstone)

Spanning fissure: a pair    El Castillo 11%
of stencils spans a
natural fissure/crack,
i.e. one to either side

Table 3. Finger lengths, palm widths and angles between thumb and first
finger for stencils in La Garma and El Castillo, ordered by apparent
association. It will be noticed that most are 'indistinct' i.e.
too faint to record measurements with confidence, but these are
included here for completeness. Shaded measurements of index fingers
indicate hands with finger ratios consistent with (but not necessarily
indicative of) females; shaded measurements of ring fingers are male.

             Thumb        Index        Middle        Ring
              (mm)         (mm)         (mm)         (mm)
La Garma

No association
Hand 2     Indistinct       70       Indistinct       70
Hand 3     Indistinct   Indistinct       73           68
Hand 5     50               78           65           77
Hand 11    Indistinct   Indistinct   Indistinct   Indistinct
Hand 13    Indistinct       58          >48       Indistinct
Hand 19    Indistinct   Indistinct   Indistinct   Indistinct

Gripping
Hand 6     Indistinct   Indistinct   Indistinct   Indistinct
Hand 10    Indistinct   Indistinct   Indistinct   Indistinct
Hand 16    Indistinct       75           68           59
Hand 23    Indistinct       58           75           72
Hand 24    Indistinct       74           76           69
Hand 27    Indistinct       58           61           56

Ergonomic
Hand 1     Indistinct   Indistinct   Indistinct   Indistinct
Hand 9     Indistinct   Indistinct   Indistinct   Indistinct
Hand 16    Indistinct       75           68           59
Hand 18    Indistinct   Indistinct   Indistinct   Indistinct
Hand 26    Indistinct   Indistinct   Indistinct   Indistinct

Right of fissure
Hand 8     Indistinct   Indistinct   Indistinct   Indistinct
Hand 12    Indistinct       71       Indistinct   Indistinct
Hand 14    Indistinct   Indistinct       44       Indistinct
Hand 16    Indistinct       75           68           59

Bordered by fissures
Hand 15    Indistinct   Indistinct      >72           74
Hand 23    Indistinct       58           75           72
Hand 24    Indistinct       60       Indistinct       65

Within multiple fissures
Hand 4     Indistinct   Indistinct   Indistinct   Indistinct
Hand 17    Indistinct   Indistinct   Indistinct   Indistinct

El Castille1

No association
Hand 15    Indistinct       65           85           66
Hand 17    Indistinct   Indistinct       76           80
Hand 21    Indistinct   Indistinct   Indistinct   Indistinct
Hand 22    Indistinct   Indistinct   Indistinct   Indistinct
Hand 23    Indistinct   Indistinct   Indistinct   Indistinct
Hand 25    Indistinct   Indistinct   Indistinct   Indistinct
Hand 27    Indistinct       70           80           70
Hand 28    Indistinct       75           80           70
Hand 31    Indistinct       68           76       Indistinct
Hand 34    Indistinct   Indistinct   Indistinct   Indistinct

Gripping
Hand 3     Indistinct   Indistinct   Indistinct   Indistinct
Hand 4     Indistinct       68           72           65
Hand 5     Indistinct   Indistinct   Indistinct   Indistinct
Hand 6     Indistinct       65           70           66
Hand 8     Indistinct   Indistinct   Indistinct   Indistinct
Hand 32    Indistinct   Indistinct       75           66
Hand 35    Indistinct   Indistinct   Indistinct   Indistinct
Hand 36    Indistinct   Indistinct   Indistinct   Indistinct
Hand 38    Indistinct       60           65           55

Atop fissure
Hand 1     Indistinct       75           71       Indistinct
Hand 2     Indistinct   Indistinct   Indistinct   Indistinct
Hand 3     Indistinct   Indistinct   Indistinct   Indistinct
Hand 5     Indistinct   Indistinct   Indistinct   Indistinct
Hand 6     Indistinct       65           70           66
Hand 9     Indistinct   Indistinct   Indistinct   Indistinct
Hand 10    Indistinct   Indistinct   Indistinct   Indistinct
Hand 16    Indistinct   Indistinct   Indistinct       60
Hand 18    Indistinct       70           80           75
Hand 29    Indistinct   Indistinct   Indistinct   Indistinct
Hand 33    Indistinct   Indistinct   Indistinct   Indistinct

Spanning fissure
Hand 11    Indistinct   Indistinct   Indistinct   Indistinct
Hand 12    Indistinct       70           66       Indistinct
Hand 19    Indistinct       66       Indistinct       67
Hand 20    55               90           90           80

Within multiple fissures
Hand 9     Indistinct   Indistinct   Indistinct   Indistinct
Hand 13    Indistinct   Indistinct   Indistinct   Indistinct
Hand 14    Indistinct   Indistinct   Indistinct   Indistinct

             Little     Angle between      Palm
              (mm)      thumb & index   width (mm)
La Garma

No association
Hand 2         60            60             95
Hand 3     Indistinct        47             81
Hand 5         69            62             82
Hand 11    Indistinct    Indistinct     Indistinct
Hand 13        51        Indistinct         94
Hand 19    Indistinct    Indistinct         90

Gripping
Hand 6     Indistinct    Indistinct         82
Hand 10    Indistinct    Indistinct     Indistinct
Hand 16    Indistinct    Indistinct         90
Hand 23        50            45             90
Hand 24        62            45             90
Hand 27    Indistinct        35             85

Ergonomic
Hand 1     Indistinct    Indistinct     Indistinct
Hand 9     Indistinct    Indistinct     Indistinct
Hand 16    Indistinct    Indistinct         90
Hand 18    Indistinct    Indistinct     Indistinct
Hand 26        52        Indistinct     Indistinct

Right of fissure
Hand 8     Indistinct    Indistinct     Indistinct
Hand 12    Indistinct    Indistinct     Indistinct
Hand 14    Indistinct    Indistinct     Indistinct
Hand 16    Indistinct    Indistinct         90

Bordered by fissures
Hand 15     V v-n 00     Indistinct         76
Hand 23        50            45             90
Hand 24        50            55             89

Within multiple fissures
Hand 4     Indistinct    Indistinct     Indistinct
Hand 17    Indistinct    Indistinct     Indistinct

El Castille

No association
Hand 15        60            35             94
Hand 17        70        Indistinct         84
Hand 21    Indistinct    Indistinct         85
Hand 22    Indistinct    Indistinct     Indistinct
Hand 23    Indistinct    Indistinct     Indistinct
Hand 25    Indistinct    Indistinct     Indistinct
Hand 27        60            40             80
Hand 28    Indistinct        45             85
Hand 31        63            35             79
Hand 34    Indistinct    Indistinct     Indistinct

Gripping
Hand 3     Indistinct    Indistinct     Indistinct
Hand 4     Indistinct        36             84
Hand 5     Indistinct    Indistinct     Indistinct
Hand 6         60            35             84
Hand 8     Indistinct    Indistinct     Indistinct
Hand 32        50            40             87
Hand 35    Indistinct    Indistinct     Indistinct
Hand 36    Indistinct    Indistinct     Indistinct
Hand 38        60        Indistinct         85

Atop fissure
Hand 1         53            25             95
Hand 2     Indistinct    Indistinct     Indistinct
Hand 3     Indistinct    Indistinct     Indistinct
Hand 5     Indistinct    Indistinct     Indistinct
Hand 6         60            35             84
Hand 9     Indistinct    Indistinct     Indistinct
Hand 10    Indistinct    Indistinct     Indistinct
Hand 16    Indistinct        35             83
Hand 18        70            30             85
Hand 29    Indistinct    Indistinct         80
Hand 33    Indistinct    Indistinct     Indistinct

Spanning fissure
Hand 11    Indistinct        27         Indistinct
Hand 12    Indistinct    Indistinct         90
Hand 19        58            30         Indistinct
Hand 20        65        Indistinct         90

Within multiple fissures
Hand 9     Indistinct    Indistinct     Indistinct
Hand 13    Indistinct    Indistinct     Indistinct
Hand 14    Indistinct    Indistinct     Indistinct