Social interaction and rock art styles in the Atacama Desert (northern Chile).

Gallardo, Francisco

Introduction

The Atacama Desert is located in the north of Chile and extends 600km from the Loa River to the Copiapo River (Figure 1). It is the driest desert in the world and large portions of it are void of life. However, the supply of fresh water from the Altiplano rains and groundwater springs enable the formation of oases and forests of native algarrobo (Prosopis alba) and chanar (Geoffroea decorticans) trees (Figure 2). Up in the highlands, with an altitude of over 3000m, plant cover consists of seasonal grasses and extensive wetlands that sustain guanaco (Lama guanicoe), vicuna (Vicugna vicugna) and taruca (Hippocamelus antisensis), as well as vizcacha (Lagidium viscacia) and other rodents and a wide variety of bird life. The coastal desert is extraordinarily rich in marine resources, with vegetation limited to small saltwater marshes and the area around the mouth of the Loa River. This hostile environment, however, did not hinder the development of the human groups that settled along the coast and further inland.

Both sedentarism and the domestication of camelid livestock began in the Late Archaic and Early Formative periods, from 5000 to 2400 uncal BP. These herds were used mainly to transport trade items over long distances, an economic activity based around exotic goods that helped sustain large settlements in the Atacama foothills and fostered the emergence of a redistributive elite that resided in the Tulan Ravine south of the Atacama salt flat during the Early Formative period (Nunez 1992; Nunez et al. 2006a & b; Cartagena et al. 2007). In the Late Formative (2400-1600 uncal BP) the first permanent settlements appeared, the pastoral mode of production was entrenched and previously evident social complexity and inequality disappeared from the archaeological record (Aguero 2005; Nunez 2005).

It is within this context that some of the most complex repertoire of South American rock art is found. In the Atacama Desert the most fully contextualised rock art spans three cultural periods--the Late Archaic, the Early Formative and the Late Formative (Gallardo 200 Berenguer 2004; Nunez et al. 2006b). The oldest of these is the Kalina-Puripica style, which consists of engravings that have been linked to hunter-gatherer settlements dated from 5000-4000 BP (Nunez 1983; Berenguer et al. 1985). The Taira-Tulan and Confluencia styles of engravings and paintings, respectively, developed in a pastoral environment during the Early Formative period between 4000 BP and 2400 BP (Berenguer 1995; Gallardo et al. 1999). In the following Late Formative Period, which lasted until 1600 uncal BP the style of painting is known as Cueva Blanca. The compositional nature of this art form was influenced by the iconography and symmetrical structures of textile imagery (Sinclaire 1997).

[FIGURE 1 OMITTED]

In this paper I shall explore the social and symbolic relationships that link settlement and the production of surplus goods with the distribution of rock art and its formal content. The aim is to describe the rock art style as an expression of social and symbolic consensus that functioned ideologically to validate the different intercommunal hierarchies that were occurring at the regional level.

[FIGURE 2 OMITTED]

The Puripica-Kalina style (Late Archaic, 5000-4000 uncal BP)

The oldest rock art recorded in the Atacamena region consists mainly of engravings, with a distribution ranging from the Upper Loa River to the Tulan Ravine south of the Salar de Atacama (Berenguer et al. 1985; Berenguer 1995; Nunez et al. 1999, 2006c; Gallardo 2001). The main subject is camelids, forming aggregations rather than scenes. They are presented in profile, generally without hooves, with bodies and extremities displaying multiple anatomical attributes. They vary in size but do not exceed 0.3m (Figure 3). They are found on small boulders as well as on the rock faces of the ravines, always associated with residential sites not far from water and forage (Berenguer 1995; Nunez et al. 2006b).

The chronology assigned to this style is based on a settlement found in the Puripica Ravine, situated on a seasonal tributary of the Vilama River. There, under occupational waste dated at 4815 [+ or -] 70 uncal BP, engravings were found on a boulder that was part of a dwelling wall (Nunez et al. 1999), A temporal association that is similar, though spatially indirect, has been found at the Kalina site (Upper Loa) dated at 4370 [+ or -] 220 BP (Berenguer et al. 1985; Aldunate et al. 1986; Caceres & Berenguer 1996).

[FIGURE 3 OMITTED]

The Late Archaic, which saw the production of this art, is a period characterised by an intensification of hunting and the domestication of the first llama herds (Nunez 1981, 1992; Dransart 1991; Yacobaccio 2001, 2004; Nunez et al. 2006c; Cartajena et al. 2007). Studies of archaeofauna indicate that in semi-permanent settlements such as Chiu Chiu (not far from Kalina), Puripica and Tulan, archaic populations would have kept herds of domesticated animals, especially llamas (Lama glama), which are morphologically interpreted as beasts of burden (Cartajena 1994; Cartajena et al. 2007). It appears to be clear that, along with the characteristic hunting and gathering of the time, long-distance trading and the circulation of exotic goods was also beginning, and seems to have involved the production and transport of beads of copper ore and Pacific shells. This activity appears to be represented by an overabundance of different forms of perforating tools in quantities that have not been recorded elsewhere for the Early or Middle Archaic periods, as well as by evidence of waste generated in producing those beads (Druss 1977; Nunez 1992; Jackson & Benavente 1994; De Souza 2004; Nunez et al. 2006c).

Caleta Huelen 42 is a coastal site dated at 4780 [+ or -] 100-3780 [+ or -] 90 uncal BP that consists of semi-subterranean dwellings built using large rounded boulders, a Late Archaic pattern that has been recorded from Puripica to Tulan (Figure 4). Excavation here produced a surprising number of shell beads--one area alone contained more than 1850 pieces--in direct association with materials that could have been obtained from inland populations, including obsidian, taruca deer antlers, thread and woven cloth from camelid yarn, woodworking and seeds from algarrobo and chanar fruit (Nunez et al. 1975; Zlatar 1983, 1989).

[FIGURE 4 OMITTED]

In this site there is multiple evidence of social interaction, but the most pertinent data in this regard came from the exterior of one dwelling, where a boulder was found to contain a camelid figure in the Kalina-Puripica style (Gonzalo Pimentel pers. comm.). Like the architectural style, the presence of this art on the coast suggests that these groups had integrated a cultural model from the highlands, contributing modified and unmodified shells that were used as ornaments, recipients and raw material. These products were circulated by hunter-gatherer groups that inhabited the Atacama Desert ravines on the outskirts of the region, such as those in northern Argentina, where Atacameno Pacific shell bead necklaces have been found as funerary offerings (Yacobaccio 2001, 2004).

The Taifa Tulan and Confluencia styles (Early Formative, 4000-2400 uncal BP)

This period features the Taira Tulan and Confluencia styles, whose distinctive forms have been found on the rock walls of the Tulan temple. This site is the only one from this period with complex architecture and is covered with litter from the numerous ceremonies held there; it has been dated at around 2840 uncal BP (Nunez et al. 2006c). The Taira Tulan style corresponds mainly to engravings of whole camelids as well as body parts including heads, ears and tails (Figure 5). They only occasionally appear alongside felines, birds and native rodents. The figures range from small to large (life-size) and are overlaid, forming large-scale works on rock walls that can be seen from a great distance away (Gallardo 2001).

[FIGURE 5 OMITTED]

The Confluencia style basically consists of small paintings (although there are also engravings) usually located in shelters (Gallardo et al. 1999). Here, the camelids are the principal subject, often far outnumbering humans. They are shown with anatomical detail, generally as animated or moving figures. Almost all are in profile and usually appear in group scenes. Notable among these scenes are moving herds, animals feeding their young, camelid fighting scenes, hunters with spears and darts and hunting by encirclement (see Montt 2004 and Gonzalez 2002) (Figure 6). A comparative and contextual study of the anatomical morphology of the images of wild and domesticated camelids has suggested that the former correspond to llamas and the latter to vicunas or guanacos (Gallardo & Yacobaccio 2005). The Taira Tulan and Confluencia styles are distributed throughout the Atacamena region, located in intermediate ravines in direct association with forage. This distribution coincides broadly with the majority of the residential sites from this period.

[FIGURE 6 OMITTED]

The Early Formative saw the consolidation of domestication processes begun in the previous period, and hunting and gathering were integrated into a more broad-based economic model. However, over time these activities became dominated by sedentism, copper ore bead production, animal husbandry for transport and interregional exchange (Nunez et al. 2006c). The husbandry of llamas used as beasts of burden intensified during this period (Nunez et al. 2006c; Cartajena et al. 2007), and this new activity brought with it the need to care for the animals, organise their feeding and grazing around the annual cycle, model the herd structure and control reproduction. Indeed, higher ranking and more complex settlements of this time are found in well-irrigated ravines, indicating the crucial role of livestock husbandry in determining the settlement pattern (Nunez et al. 2006c). While it is true that the first appearance of crops, such as maize, beans, quinoa and peppers, starts during the Early Formative, these occurrences have been limited to date. This fact, added to the absence of very large and/or complex settlements in oases with agricultural potential, suggests that agriculture was a less important technical and social process than those related to livestock husbandry, hunting and gathering (Holden 1991; Thomas et d 1995; Aguero 2005; Nunez 2005; Westfall & Gonzalez 2006).

As in the Late Archaic period, the production of beads for trade continued, with copper ore replacing shells as the primary raw material (Rees 1999; Rees & De Souza 2004; Nunez et al. 2006c). The most conspicuous evidence features objects of symbolic and social significance, including Pacific mollusc shells with little food value (Oliva peruviana, Turritela singulata) and marine mammal bones worked into cephalic omaments (Benavente 1982; Nunez et al. 2006a). From the jungles located east of the Andes came freshwater mollusc shells (Strophocheilus oblongus), tropical bird feathers, ceramic pipes and a hallucinogenic known as cebil (Anadenantbera colubrina) (Benavente 1982; Nunez 1992, 1994; Nunez et al. 2006c; Torres & Repke 2006).

The archaeological evidence indicates that established hunting practices were not affected by the appearance of domesticated camelids; on the contrary, they were assimilated and became a central part of the new economic and social scenario (Cartajena 1994; Yacobaccio et al. 1994; Nunez et al. 2006c; Cartajena et al. 2007; Gallardo & De Souza 2008). The development of these production processes must have impacted the social division of labour, promoting the participation of different interest groups in the economic and symbolic reproduction of these communities. While pastoral activity was essential for maintaining herds in the interregional exchange circuit, hunting contributed resources for subsistence and at the same time generated raw material for the production of prestige goods sucia as textiles that, like the products of mining, entered the exchange networks (Benavente 1982; Cartajena 1994). Out of this complex interweaving of production modes grew economic, political and social authorities and inequalities, which in turn led to the emergence of new kinds of social leadership associated with the administration of redistribution sites such as those described for Tulan. In this context, however, it is possible that the greatest privileges were concentrated among those social actors who controlled the means of exchange, as it was the exotic goods that played a leading role in the symbolic and ideological reproduction of social life during that time (Gallardo & De Souza 2008).

The Cueva Blanca style (Late Formative, 2400-1600 uncal BP)

The Tulan trading centre collapsed around 2720-2320 cal BP and soon after permanent and semi-permanent residential nuclei multiplied across the region (Sinclaire 2004; Aguero 2005; Nunez 2005). Small settlements of circular compounds grouped together occupied both the ravines and the oases, including the Loa River and its upper basin. Small-scale agriculture was adopted, but as in previous ages the dominant modes of subsistence were clearly the gathering of algarrobo and other fruit, the hunting of wild animals and the raising of domesticated camelids. Evidence of rock art during this time is limited, but there is a set of pictorial works called the Cueva Blanca style that introduced a new visual form. In this style camelids disappear almost entirely and human figures predominate (Figure 7). They are shown in frontal mode but lack movement, and geometric designs are much more prevalent, particularly wavy, zig-zag and criss-crossed lines. In terms of colour, red is predominant, though we also find combinations of two or more colours including green, black and/or yellow. From a compositional perspective, this art tends to be characterised by orthogonality, mirror symmetry and translation, and the images are frequently framed by a line drawn round them (Sinclaire 1997; Gonzalez 2005).

Such works are relatively numerous along the Salado River and a number of them have been recorded in other localities in the region (Nunez et al. 1997; Gallardo 2001; Berenguer 2004). This art represents a radical change over previous works and is directly related to the introduction of a new visual culture never before recorded: images from textile tapestry, a technique that could have derived from the Bolivian Altiplano in association with the Pukara culture (Sinclaire 1997). This is the style known as Alto Ramirez (in allusion to the type site of the Azapa Valley in Chile's far north), the iconography of which spread extensively northward of the Loa River beginning in the middle of the first millennium BC (Rivera 1991) (Figure 8).

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

The expansion of this style in the Atacama region was not limited to textile-related visual devices and rock art; the Alto Ramirez burial mounds--in some cases similar in structure and in others similar in form--have been recorded in different localities from the coast inland (Nunez 1971; Le Paige 1974; Moragas 1982; Aguero et al. 2006). Considering that in previous periods the hegemonic nucleus was located south of the Atacamena region, it is not unreasonable to suggest that there was a geographic inversion in the hub of interaction and its relative influence over internal social relations of the region. This gains special significance when we consider that during that time the monumental town of Guatacondo was built, some 70km from the locale of Mino at the source of the Loa and Quillagua rivers in the lower reaches. This settlement had a large central plaza with more than 180 circular compounds grouped together around it. In relative size it was six times larger than the largest contemporary Atacamena villages (Meighan 1980). Four other residential sites have been identified in the same valley, all with an archaeological design similar to that observed in the region under study. However, this community's prestige in social interactions is not so much the result of the circumstances enumerated above, but instead due to its role in an emerging technology: copper metallurgy (Graffam et al. 1990). This activity enabled the production of the metal sheets and tubular beads that have been found in Atacamena area sites and that also could have been used in the manufacture of tree cutting and woodworking instruments, which were used to build houses and make artefacts and accesories (Mayer 1980).

In the Late Formative Period the mining of copper ore and the production of copper ore beads for trade increased proportionally to the number of residential sites of all hierarchies (Rees 1999; Rees & De Souza 2004; Nunez 2000). Exchange with regions like the Argentine north-west increased (though without intermediaries), and the flow of goods from the Altiplano (Qeya/Tiwanaku I ceramics) is established in a site on the Salado River. Turi 2 is probable the only multi-component site of this epoch, and though it is a settlement of modest size, it is situated beside an extensive plain that could have provided forage for the llama caravans coming from present day Bolivia (Castro et al. 1992).

The communities in the Atacama region during this period intensified interregional relations and came under the influence of a cultural model centred mainly in southern Peru and the Bolivian Altiplano.

Discussion

In general, rock art distributions have been described as the outcome of social interactions, processes related to maintaining social boundaries or strategic alliances between different communities. To date, these social practices are most frequently explained as a response to environmental stress, population increase, migration and cultural contact (Conkey 1978, 1980; Gamble 1982; Jochim 1983; Jolly 1996; Wilson 1998). The functionalist perspectives and their limitations have been explored sufficiently in the post-processual literature, but they can still be productive avenues of analysis if taken somewhat beyond the simple 'rock art styles equal social solidarity' formula (e.g. Lewis-Wiliams 1982; Faris 1983; Bradley et al. 1994; Franklin 1994; Whitley 1994; David & Lourandos 1998).

The rock art style of the Puripica-Kalina in the Late Archaic period coincides with processes of sedentarisation, the production of exchange goods and the appearance of the first beasts of burden. If the reduced mobility of these hunter-gatherer groups is clearly indicated in the proliferation of semi-permanent camps, then evidence of surplus production and the appearance of llamas being used for transport offer independent proof of the intensification of social relations beyond the region. The rock art distribution is evidence of a certain consensus among the different Atacamena communities, particularly with regard to the images of camelids, a resource central to subsistence (in wild form) and especially social interaction (in domesticated form). These mechanisms fostered the flow of economically and symbolically valuable goods, a form of capital whose introduction into neighbouring regions (and indeed within the same region) would have contributed to the ideological and social reproduction of these communities, especially among those that controlled circulation and redistribution.

The rock art of the next period, the Early Formative, expresses the presence of two social identities: while the Taira Tulan style alludes to domestic livestock, the Confluencia style makes reference to hunting and to wild animals (Gallardo & Yacobaccio 2005; Gallardo & De Souza 2008). The tension between these two activities resulting from their different roles in social reproduction appears to be reflected in the active character of the former works and the passive nature of the latter. While Confluencia favours small figures with multiple visual attributes that are organised into scenes, i.e. designed to be 'read', the Taira Tulan style favours 'monumental' works organised randomly and densely overlapping, with later figures carved over earlier engravings from which they took their form. These expressions also include anatomical parts and re-carving of lines--formal attributes suggesting their construction in successive stages--and render visual results that hinder rather than help the recognition of their content.

Although the Taira Tulan sites outnumber the Confluencia ones, both are found throughout the Atacamena region, bearing tacit witness to how the different communitios shared a diverse but collective imagery. However, this apparent homogeneity is contradicted by the importance and scale of certain settlements of the period, most notably those of the Tulan Ravine (Pollard 1970; Benavente 1982; Nunez 1992; Sinclaire 2004). Indeed, the semi-underground structures found at this site would have required a large workforce to construct. Judging from the waste and offerings found, this place must have functioned as a ceremonial gathering place (Nunez et al. 200oa). The abundance of exotic goods found at these sites and their evident ritual nature lead one to believe that the Tulan pastoral communities were able to control and manage the access to, circulation and distribution of trade goods both into and out of the region. The social and economic inequality between those who raised livestock and those who hunted is masked by the formal content of rock art, which endows the activities of each with the same level of prestige (Gallardo & De Souza 2008).

In the Late Formative period the appearance of the Cueva Blanca rock art, and the related Alto Ramirez style in textiles, coincides with the appearance of copper alloy metallurgy. The Cueva Blanca style not only adopted the Alto Ramirez structure and iconography but recreated it through painting, in its own particular mode of expression that enabled its creators to move beyond simple reproduction. This close iconographic relationship could be ascribed to a visual culture, that not only enabled sharing of the prestige associated with the populations north of the Loa River, but also was a way of neutralising in the collective imaginary the unequal access to new technologies that were adopted in those times. The Late Formative was a time when new avenues of trade produced symbolic and technological clientelism in these communities in relation to others situated in the north of the region. A case in point is the village of Guatacondo, the design and scope of which suggests that it could have served as a centre for ceremonies and redistribution (Nunez 1974).

Conclusion

This paper offers a provisional model relating the rock art styles to different forms of social interaction in the Archaic and Formative periods. In all of the periods mentioned the social primacy of some communities over others was reflected in the settlements, whose complexity, size and design suggest the existence of special places for social gatherings where social surpluses could be redistributed. The rock art styles not only expressed social identities at the intercommunal level, but were an active symbolic medium in the reproduction of the different social structures alluded to.

The early rock art styles' regional character, formal content and effects on social integration laid the foundation for a set of symbolic practices that in each period deposited an ideal closely linked to prevailing economic and social imperatives. While the prestige of the hunt was incorporated into the Kalina-Puripica and Confluencia styles, the new expectations introduced by pastoralism were expressed in the Taira-Tulan style. The incorporation of the Cueva Blanca style represents a profound rupture in Atacamena social prehistory, as for the first time rock art imagery was reduced to an object, a textile product whose economic and symbolic importance reached its pinnacle during the Inka Empire (Murra 1975). This is highly important because the textiles alluding to Atacameno rock art were produced using an embroidery technique introduced from the far north of Chile around 500 BC (Rivera 1970; Mujica 1985; Sinclaire 1997). This would have lent prestige to the communities possessing this textile that, along with their use of new technologies such as metallurgy, would have allowed them to exercise influence over the Atacameno groups of the Late Formative period.

The needs associated with producing surpluses for exchange as prestige goods fostered alliances among communities across the region, in a framework of relations that seems to have contributed to an asymmetry in the redistribution system, given the relative position of different communities in regard to transportation, trade or prestige goods. However, social inequality was manipulated through rock art styles, the role of which was to suggest a consensus through a common imagery.

Acknowledgements

Funding for this research was provided by FONDECYT (1070083). I would also like to acknowledge the contribution of the referees and Martin Carver, whose comments improved the presentation of the ideas contained in this article.

Received: 3 April 2008; Accepted: 20 June 2008; Revised: 21 July 2008

References

AGUERO, C. 2005. Aproximacion al asentamiento temprano en los oasis de de San Pedro de Atacama. Estudios Atacamenos 30: 29-60.

AGUERO, C., P. AYALA, M. URIBE, C. CARRASCO & B. CASES. 2006. El periodo formativo desde Quillagua, Loa inferior (norte de Chile), in H. Lechman (ed.) Esferas de interaccion prehistoricas y fronteras nacionales modernas: Los Andes sur centrales: 73-118. Lima: Instituto de Estudios Peruanos.

ALDUNATE, C., J. BERENGUER, V. CASTRO, L. CORNEJO, J. L. MARTINEZ & C. SINCLAIRE. 1980. Cronologia y asentamiento en la region del Loa Superior. Santiago: Direccion de Investigaciones y Biblioteca, Universidad de Chile.

BENAVENTE, M. 1982. Chiu-Chiu 200. Una comunidad pastora temprana en la provincia del Loa (II Region), in Actas del IX Congreso Nacionalde Arqueologia Chilena: 75-94. La Serena: Museo Arqueologico de La Serena.

BERENGUER, J. 1995. El arte rupestre de Taira dentro de los problemas de la arqueologia atacamena. Chungara 27: 7-43.

--2004. Cinco milenios de arte rupestre en los Andes atacamenos: imagenes para lo humano, imagenes para lo divino. Boletin del Museo Chileno de Arte Precolombino 9: 75-108.

BERENGUER, J., V. CASTRO, C. ALDUNATE, C. SINCLAIRE & L. CORNEJO. 1985. Secuencia del arte rupestre en el alto Loa: una hipotesis de trabajo, in C. Aldunate, J. Berenguer & V. Castro (ed.) Estudios en arte rupestre. 87-108. Santiago: Museo Chileno de Arte Precolombino.

BRADLEY, R., F.C. BOADO & ILE VALCARCE. 1994. Rock art research as landscape archaeology: a pilot study in Galicia, north-west Spain. World Archaeology 25: 374-90.

CACERES, I. & J. BERENGUER. 1996. Arte rupestre mobiliar de estilo Kalina en el alto Loa. Boletin de la Sociedad Chilena de Arqueologia 23:14-15.

CARTAJENA, I. 1994. Determinacion de restos oseos de camelidos en dos yacimientos del Loa Medio (II Region). Estudios Atacamenos 11: 25-52.

CARTAJENA, L, L. NUNEZ M. GROSJEAN. 2007. Camelid domestication on the western slope of the Puna de Atacama, northern Chile. Anthropozoologica 42:155-73.

CASTRO, V., C. ALDUNATE, J. BERENGUER, L. CORNEJO, C. SINCLAIRE & V. VARELA. 1992. Relaciones entre el noroeste argentino y el norte de Chile: el sitio 02-Tu-002, Vegas de Turi, in M. Albeck (ed.) Taller: De Costa a Selva: Produccion e Intercambio entre los Pueblos Agroalfareros de los Andes Centro-Sur: 215-39. Tilcara: Instituto Interdisciplinario de Tilcara, Universidad de Buenos Aires.

CONKEY, M. 1978. Style and information in cultural evolution: toward a predictive model for the Paleolithic, in C. Redman, M. Berman, E. Curtin, W. Langhorne, N. Versaggi & J. Wanser (ed.) Social archaeology: 61-85. New York (NY): Academic Press.

--1980. The identification of prehistoric hunter-gatherer aggregation sites: the case of Altamira. Current Anthropology 5: 609-30.

DAVID, B. & H. LOURANDOS. 1998. Rock art and socio-demography in north-eastern Australian Prehistory. World Archaeology 2: 193-219.

DE SOUZA, P. 2004. Cazadores recolectores del Arcaico Temprano y Medio en la cuenca superior del rio Loa: Sitios, conjuntos liticos y sistemas de asentamientos. Estudios Atacamenos 27: 7-42.

DRANSART, P. 1991. Llamas, herders and the exploitation of raw materials in the Atacama Desert. World Archaeology 22: 304-19.

DRUSS, M. 1977. Environment, subsistence economy and settlement pattern of the Chiu Chiu Complex (ca. 2700 to 1600 BC) of the Atacama Desert, Northern Chile. Unpublished PhD dissertation, Columbia University.

FARIS, J. 1983. From form to content in the structural study of aesthetic systems, in D. Washburn (ed.) Structure and cognition in art: 90-112. Cambridge: Cambridge University Press.

FRANKLIN, N. 1994. Research with style: a case study from Australian rock art, in S. Shennan (ed.) Archaeological approaches to cultural identity: 278-90. New York (NY): Routledge.

GALLARDO, F. 2001. Arte rupestre y emplazamiento durante el Formativo Temprano en la cuenca del rio salado (Desierto de Atacama, norte de Chile). Boletin del Museo Chileno de Arte Precolombino 8: 81-95.

GALLARDO, F. & H. YACOBACCIO. 2005. Wild or domesticated? Camelids in Early Formative rock art of the Atacama Desert (Northern Chile). Latin American Antiquity 16:115-30.

GALLARDO, F. & P. DE SOUZA. 2008. Rock art, modes of production and social identities during the Early Formative Period in the Atacama Desert (Northern Chile) in I. Domingo, D. Fiore & S. May (ed.) Archaeologies of art: time, place, and identity: 79-97. Los Angeles (CA): Left Coast Press.

GALLARDO, F., C. SINCLAIRE & C. SILVA. 1999. Arte rupestre, emplazamiento y paisaje en la cordillera del desierto de Atacama, in J. Berenguer & F. Gallardo (ed.) Arte rupestre en los Andes de Capricornio: 57-96. Santiago: Museo Chileno de Arte Precolombino.

GAMBLE, C. 1982. Interaction and alliance in Palaeolithic society. Man 1: 92-107.

GONZALEZ, J. 2002. Etologia de camelidos y arte rupestre de la Subregion no Salado (norte de Chile, II region). Estudios Atacamenos 23: 23-32.

GONZALEZ, P. 2005. Codigos visuales de las pinturas rupestres Cueva Blanca: Formas, simetria y contexto. Boletin del Museo Chileno de Arte Precolombino 10: 55-72.

GRAFFAM, G., A. CAREVIC & M. RIVERA. 1996. Ancient metallurgy in Atacama: evidence for copper smelting during Chiles early ceramic period. Latin American Antiquity 7: 101-13.

HOLDEN, T. 1991. Evidence of prehistoric diet from northern Chile: coprolites, gut contents and flotation samples from the Tulan Quebrada. World Archaeology 22:320-31.

JACKSON, D. & A. BENAVENTE. 1994. Secuencia y cambios y adaptacion de los cazadores recolectores de la microcuenca de Chiu-Chiu, Provincia El Loa. Chungara 26: 49-64.

JOCHIM, M. 1983. Paleolithic cave art: some ecological speculations, in G. Bailey (ed.) Hunter-gatherer economy in prehistory: 212-19. Cambridge: Cambridge University Press.

JOLLY, P. 1996. Symbiotic interaction between black farmers and south-eastern San: implications for Southern African rock art studies, ethnographic analogy, and Hunter-Gatherer cultural identity. Current Anthropology 37: 277-306.

LE PAIGE, G. 1974. El yacimiento de Tchaputchayna, Estudios Atacamenos 2: 59-74.

LEWIS-WILIAMS, J.D. 1982. The economic and social context of southern San rock are Current Anthropology 23: 429-49.

MAYER, E. 1986. Armas y herramientas de metal prehispanicas en Argentina y Chile. Munchen: C.H. Beck.

MEIGHAN, C.W. 1980. Archaeology of Guatacondo, in C.W. Meighan & A.L. True (ed.) Prehistoric trails of Atacama: archaeology of northern Chile (Monumenta Archaeologica 7): 99-126. Los Angeles (CA): Institute of Archaeology, University of California.

MONTT, I. 2004. Elementos de atuendo e imagen rupestre en la Subregion del rio Salado (Region del Loa Superior). Chungara, Volumen especial 36(2): 651-62.

MORAGAS, C. 1982. Tumulos funerarios de la costa sur de Tocopilla (Cobija), II Region. Chungara 9: 152-73.

MUJICA, E. 1985. Altiplano-coast relationships in the south-central Andes: from indirect to direct complementarity, in S. Masuda, I. Shimada & C. Morris (ed.) Andean ecology and civilization: 103-40. Tokyo: University of Tokyo Press.

MURRA, J. 1975. Formaciones economicas y politicas del mundo andino. Lima: Instituto de Estudios Peruanos.

NUNEZ, L. 1971. Secuencia y cambio en los asentamientos humanos de la desembocadura del rio Loa en el norte de Chile. Boletin de la Universidad de Chile 112: 3-25.

--1974. La agricultura prehitorica en los andes meridionales. Santiago: Editorial Orbe.

--1981. Asentamiento de cazadores recolectores tardios en la Puna de Atacama: Hacia el sedentarismo. Chungara 8: 137-68.

--1983. Paleoindio y Arcaico en Chile: diversidad, secuencia y procesos. Mexico: Editorial Cuicuilco.

--1992. Emergencia de complejidad y arquitectura jerarquizada en la puna de Atacama: las evidencias del sitio Tulan-54, in M. Albeck (ed.) Taller "De Costa a Selva': produccion e intercambio entre los Pueblos Agroalfareros de los Andes Centro-Sur: 85-115. Buenos Aires: Instituto Interdisciplinario Tilcara, Universidad de Buenos Aires.

--1994. Cruzando la cordillera por el norte: senorios, caravanas y alianzas, in F. Mena (ed.) La cordillera de los Andes: ruta de encuentros: 9-21. Santiago: Museo Chileno de Arte Precolombino.

--2005. La naturaleza de la expansion aldeana durante el formanvo tardio en la cuenca de Atacama. Chungara 37: 165-93.

--2006. La orientacion minero metalurgica de la produccion Atacamena y sus relaciones fronterizas, in H. Lechman (ed.) Esferas de interaccion prehistoricas y fronteras nacionales modernas: Los Andes sur centrales: 205-51. Lima: Instituto de Estudios Peruanos.

NUNEZ, L., I. CARTAJENA, J. LOO, S. RAMOS, T. CRUZ & H. RAMIREZ. 1997. Registro e investigacion del arte rupestre en la Cuenca de Atacama (Informe Preliminar). Estudios Atacamenos, 14: 307-25.

NUNEZ, L., I. CARTAJENA & M. GROSJEAN. 1999. Un ecorrefugio oportunistico en la Puna de Atacama durante eventos aridos del Holoceno Medio. Estudios Atacamenos 17: 125-74.

NUNEZ, L., I. CARTAJENA, C. CARRASCO & P. DE SOUZA. 2006a. Templete Tulan de la Puna de Atacama: emergencia de complejidad ritual durante el Formativo Temprano (Norte de Chile). Latin American Antiquity 17: 445-73.

NUNEZ, L., I. CARTAJENA, C. CARRASCO, P. DE SOUZA & M. GROSJEAN. 2006b. Patrones, cronologia y distribucion del arte rupestre arcaico tardio y formativo temprano en la cuenca de Atacama, in D. Fiore & M. Podesta (ed.) Tramas en lapiedra: 191-204. Buenos Aires: Asociacion Amigos del Instituto Nacional de Antropologia.

NUNEZ, L., I. CARTAJENA, C. CARRASCO, P. DE SOUZA & M. GROSJEAN. 2006c. Emergencia de comunidades pastoralistas formativas en el sureste de la puna de Atacama. Estudios Atacamenos 32: 93-117.

NUNEZ, L., V. ZLATAR & P. NUNEZ, 1975. Caleta Huelen 42: una Aldea Temprana en el Norte de Chile (Nota Preliminar). Hombrey Cultura 2: 67-103.

POLLARD, G. 1970. The cultural ecology of ceramic-stage settlement in the Atacama Desert. Ann Arbor (MI): University Microfilms, Inc.

REES, C.H. 1999. Elaboracion, distribucion y consumo de cuentas de malaquita durante d periodo Formativo en la vega de Turi y sus inmediaciones, subregion del rio Salado, Norte de Chile, in C.A. Aschero, M.A. Korstanje & EM. Vuoto (ed.) En los tres reinos: practicas de recoleccion en d cono sur de America: 83-94. Tucuman: Universidad Nacional de Tucuman.

REES, C. & P. DE SOUZA. 2004. Produccion litica durante d Periodo Formativo en la subregion del rio Salado (Norte de Chile). Chungara 36 supl. esp: 453-65.

RIVERA, M. 1976. Nuevos aportes sobre el desarrollo cultural altiplanico en los valles bajos del extremo Norte de Chile durante d Periodo Intermedio Temprano, in H. Niemeyer (ed.) Homenaje al R.P. Gustavo Le Paige: 71-81. Antofagasta: Universidad del Norte.

--1991. The prehistory of northern Chile: a synthesis. Journal o fWorld Prehistory 5: 1-47.

SINCLAIRE, C. 1997. Pinturas tupestres y textiles formativos en la region atacamena. Estudios Atacamenos 14: 327-38.

--2004. Prehistoria del periodo formativo en la cuenca alta del rio Salado (Region del Loa superior) Chungara 36 supl. esp: 619-39.

THOMAS, C., A. BENAVENTE, I. CARTAGENA & G. SERRACINO. 1995. Topater, un cementerio temprano. Hombre y Desierto 9:159-70.

TORRES, C, & D. REPKE. 2006. Anadenanthera. Visionary plant of ancient South America. New York (NY) & London: Haworth Press.

WESTFALL, C. & C. GONZALEZ, 2006. Cementerio regimiento Chorrillos de Calama: Testimonios funerarios formativos en el Loa medio, Region de Antofagasta, in Actas del XVII Congreso Nacional de Arqueologia Chilena. Valdivia: Universidad Austral.

WHITLEY, D. 1994. By the hunter, for the gatherer: art, social relations and subsistence change in the prehistoric Great Basin. World Archaeology 25: 356-72.

WILSON, M. 1998. Pacific rock art and cultural genesis: an multivariate exploration, in C. Chippindale & P.S. Tacon (ed.) The archaeology of rock art: 163-84. Cambridge: Cambridge University Press.

YACOBACCIO, H. 2001. Cazadores complejos y domesticacion de camelidos, in G. Mengoni, D. Olivera & H. Yacobaccio. (ed.) El uso de los camelidos a traves del tiempo: 261-82. Buenos Aires: Ediciones del Tridente.

--2004. Social dimensions of camelid domestication in the southern Andes. Anthropozoologica 39: 237-47.

YACOBACCIO, H., D. ELKIN & D. OLIVERA. 1994. El fin de las sociedades cazadoras? El proceso de domesticacion animal en los Andes Centro-Sur. Arqueologia Contemporanea 5: 23-32.

ZLATAR, V. 1983. Replanteamiento sobre el problema Caleta Huelen 42. Chungara 10: 21-8.

--1989. Un yacimiento preceramico y su problematica desde la perspectiva de sus recintos habitacionales. Hombre y Desierto 1: 1-36.

Francisco Gallardo Museo Chileno de Arte Precolombino, Bandera 361, Santiago. Chile (Email: fgallardo@museoprecolombino.cl) Translated by Joan Donaghey