Remerciez-le!

Remerciez @Admin pour avoir partagé cet document gratuitement, de la manière la plus simple, en partageant sur les réseaux sociaux.

143 ANTHROPOZOOLOGICA • 2007 • 42 (2) © Publications Scientifiques du Muséum na

143 ANTHROPOZOOLOGICA • 2007 • 42 (2) © Publications Scientifiques du Muséum national d’Histoire naturelle, Paris. Andean camelid herding in the South Andes: ethnoarchaeological models for archaeozoological research Hugo D. YACOBACCIO Universidad de Buenos Aires-CONICET 25 de Mayo 217, AR-Buenos Aires (Argentina) hdyacobaccio@gmail.com Yacobaccio H. D. 2007. – Andean camelid herding in the South Andes: ethnoarchaeo- logical models for archaeozoological research. Anthropozoologica 42 (2): 143-154. ABSTRACT Culling strategies are a key issue in the analysis of pastoral economic systems. Mortality patterns have been used for reconstructing the strategies of exploitation of herds in different pastoral records, the main point being the distinction between primary and secondary productions as reflected in the age-classes of particular (or regional) bone records. Andean camelid pastoralism is no exception, and pioneering efforts have been made in constructing mortality profiles, but by using Old World models and assumptions for interpreting the information. Mixed exploitation patterns of camelid use (i.e. for meat and fiber production) is common in the Andes, and this paper presents mortality profiles resulting from this kind of economic activity, through the analysis of bone assemblages from contemporary pastoral sites in the Puna de Atacama. This case study will enable us to discuss these issues on new bases. RÉSUMÉ L’élevage des camélidés dans les Andes du sud : modèles ethnoarchéologiques appliqués à l’archéozoologie. Les stratégies d’abattage sont un élément clé dans l’analyse des systèmes économiques pastoraux. Les courbes de mortalité ont été utilisées pour la reconstruction des stratégies d’exploitation des troupeaux dans différents cadres pastoraux, la question essentielle étant la distinction entre productions primaire et secondaire, telles qu’elles sont reflétées dans les classes d’âges des enregistrements particuliers (ou régionaux) des ossements. Le pastoralisme des camélidés andin n’est pas une exception, et des efforts novateurs ont été accomplis dans la construction des profils d’abattage, qui utilisent toutefois des modèles et des hypothèses d’interprétation des données issus de traditions pastorales de l’Ancien Monde. Les modes d’exploitation mixte des camélidés (i.e. pour la production de la viande et des poils) sont communs dans les Andes, et cet article présente des profils de KEY WORDS Mortality patterns, age-classes, South American camelids, Puna de Atacama. Yacobaccio H.D. 144 ANTHROPOZOOLOGICA • 2007 • 42 (2) mortalité résultant de ce type d’activité économique, construits d’après l’analyse d’assemblage osseux de sites pastoraux actuels dans la Puna de Atacama. Cette étude de cas devrait permettre de discuter ces questions sur de nouvelles bases. RESUMEN Pastoreo de camélidos en los Andes Centro Sur : modelos etnoarqueológicos para la investigación arqueozoológica. Las estrategias de matanza selectivas son un punto clave en el análisis de los sitemas económicos pastoriles. Los patrones de mortalidad han sido empleados para la reconstrucción de las estrategias de explotación de los rebaños en diferentes registros pastoriles, cuyo objetivo principal ha sido la distinción entre producción primaria vs. secundaria y su reflejo en las clases de edad de registros óseos particulares (o regionales). El pastoreo de camélidos andino no es excepción y esfuerzos pioneros se realizaron en la construcción de perfiles de mortalidad, aunque se utilizaron modelos y presupuestos originados en el Viejo Mundo para interpretar la información. Patrones de explotación mixta en el uso de los camélidos (p. ej. producción de carne y fibra) es común en los Andes. Este trabajo muestra perfiles de mortalidad originados en esta clase de actividad económica a través del análisis de sitios pastoriles contemporáneos de la Puna de Atacama. Este caso de estudio permitirá ubicar estos estudios sobre nuevas bases. PALABRAS CLAVES Partones de mortalidad, clases de edad, camélidos sudamericanos, Puna de Atacama. INTRODUCTION The archaeozoological study of pastoralism has employed mortality profiles as an important tool for inferring past management practices. The model of kill-off patterns derived from the influential attempt by R. Payne (1973), were developed for caprine husbandry in a context of domestication, farming or nomadic pastoralism. These models have been used in discussing evidence for animal production, consumption and productive specialization. The distinction between production modes is based on the assumption that pastoral production tends to obtain particular sets of animal products or services (meat, milk, wool, transport) by diffe- rentially managing herd culling patterns. This situation has been emphasized in different instances for the Old World (i.e. Davis 1987, Legge 1996). The assumption of productive specialization has also dominated the study of Andean camelid pastoralism. In this paper I argue as mixed production sys- tems are characteristic of traditional herding in the Andes, new assumptions must be used in archaeofaunal analysis. In this line of thinking, I present an ethnoarchaeological case-study which provides bone-data for characterizing age-classes profiles indicative of mixed pastoral production. This model, in turn, will be useful for re-interpreting archaeozoological informa- tion, and for discussing the dynamics of herd utilization. OLD WORLD PRODUCTION MODES The assumption of the Old World model includes that herd size, demographic structure and sex ratios are more or less stable over time, and that herds are viable reproductive units. (Greenfield 1988). The models use age/sex classes to distinguish between different utilization in wool/meat/milk in sheep/goat production. MOTS CLÉS Courbes de mortalité, classe d’âges, camélidés sud-américains, Puna de Atacama Briefly, the age classes for the different modes of herd utilization are as follows: – 1. Meat production: males are slaughtered before maturity, while females are culled during adulthood. The expected percentages in each age-class for primary production, but with milk and wool component are: 30% infant-juveniles; 20% sub-adult; 50% adult. – 2. Milk production: males are culled as soon as the milk yield is assured or when lactation levels decrease; so they are slaughtered at very early ages (before a year). Females have higher rates of survival to adulthood. – 3. Wool production: emphasis is upon adults regardless of sex. The expectations derived from Payne’s model are 30% infant-juveniles, 10% sub-adults, 60% adults. Cribb’s simulation is somewhat different because he predicts only 4.9% of sub-adult, and 65.4% of adult, because he assumes that milk and meat is an important component in wool production (Cribb 1991). The difference between kill-off patterns is clearly seen only in the proportion of sub- adults, and in a lesser extent, in increasing adult percentages (Fig. 1). The production modes interpretation has ambiguities recognized by diverse authors, highlighting that the same pattern can be produced by different exploi- tation strategies. L. Martin’s (1999) study on pastoral production in the Jordan steppe showed the difficulty of identifying different production practices, espe- cially milk and meat from kill-off patterns, emphasizing that other lines of evidence must be included to obtain a clear interpretation. Besides, productive specialization may not be synchronic or spatially isomorphic. For example, Helmer, Gourichon and Vila (2006) showed for the nor- thern Near East that milk was utilized since the beginning of the middle PPNB, while a major change occurred in the exploitation of the meat of young animals. Also, Vigne & Helmer (2007), contrary to earlier statements, showed that secon- dary products were used as early as the beginning of the Neolithic in the Near East and Europe, implying that milk production may have played a role in caprine domestication in the former region. This situation adds another dimension of variation; that is, regional and/or local variability in herding practices. THE ANDEAN MODELS The same reasoning was employed in dealing with Andean camelid pastoralism, assuming the three production strategies. Although milk production was desemphasized for a new alternative use as pack animal, because camelids do not produce milk, the idea of specialized production herds was an important, if not the main, assumption. Thus, maintaining mature animals alive was considered an advantageous strategy if the llamas (Lama glama) were used as pack animals, and alpacas (Lama pacos) as sources of fibre (Wing 1972). E. Wing developed this idea: “Animals that are raised for work or products, for example, the llama burden bearing or alpacas for wool, would most advantageously be mantained for their full productive lives. Natural and cultu- ral events may mitigate this, and animals may be slaughtered or die while in their prime or as juve- niles” (Wing 1986:248, see also Hesse 1982). The same assumption of sex/age-classes diffe- rences relating to alternative uses is adopted based on contemporary management practices in Peru and, sometimes, using historical sources. Today camelid herding in Peru is specialized in the production of fibre for the international market, Andean camelid herding in the South Andes: ethnoarchaeological models for archaeozoological research 145 ANTHROPOZOOLOGICA • 2007 • 42 (2) 0 10 20 30 40 50 60 70 Meat Milk Wool Production Modes Percentages immature subadult adult FIG. 1. – Ideal Models of Old World Production Modes (data from Greenfield 1988). since the alpacas are the main producers of high quality fibre, while llamas are raised as beasts of burden. Only a minor proportion of camelid meat is consumed and can be either alpaca or llama (Flores Ochoa 1968: 115-118; Flores Ochoa & MacQuarry 1995). This is further complicate because of the difficulty for distinguish camelid sexes through bone analysis in fragmentary series. Moore (1989) based on ethnographic data from Peru (Sotomayor 1984) uploads/Industriel/ andean-camelid-herding-in-the-south-andes-ethnoarchaeological-models-for-archaeozoological-research.pdf