VICUÑAS AND DOMESTICITY

�

VICUÑAS AND DOMESTICITY

Alejandro F. Haber*

* CONICET, Escuela de Arqueología, (UNCa), Salas Martínez 464, K4703BKJ San Fernando del Valle de Catamarca, Argentina. E-mail:[email protected]

○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○

ABSTRACTThis paper is devoted to the analysis of three interrelated issues: a theoretical one, a methodological one and, finally, ahistorical one. The theoretical point deals with the question: can we build a theory of domesticity? Is domesticity, understoodas the recursive relationships between social relationships and the relationships between people and nature, a relevantcategory for the study of ancient human societies? The methodological issue seeks to disentangle the role of materialculture in the building of those networked relations. The historical matter of the role of the Puna de Atacama oases in thecircum-Puna region (South Central Andes) of the First Millennium A.D. is analised through data from Tebenquiche Chico.Finally, the importance of vicuñas in the study of domesticity is explored through the practices and representations relatedto animals.

RESUMENTres son los problemas que se analizan en este trabajo: uno teórico, otro metodológico y, finalmente, uno histórico. Lacuestión teórica plantea si es posible elaborar una teoría de la domesticidad. Dicho de otro modo, ¿es la domesticidad, comoconcepto que se refiere a las relaciones recursivas entre las relaciones sociales y las relaciones entre la gente y la naturaleza,una categoría relevante en el estudio de las sociedades del pasado? El análisis del problema metodológico busca desentrañarel rol que tuvo la cultura material en la construcción de dicha red de relaciones, y el que le cabe en la investigaciónarqueológica de la misma. El caso de Tebenquiche Chico aportará los elementos para discutir el problema histórico del rol delos oasis de puna en el contexto circumpuneño (Andes Centro-sur) del primer milenio D.C. Finalmente, las prácticas yrepresentaciones referidas a los animales permitirán evaluar qué lugar ocupan las ‘vicuña’s en el estudio de la domesticidad.

INTRODUCTIONTo begin this paper with a reference to the process ofdomestication of the so-called herbivore vicuña would beunsound. So, let me begin asking ‘why’? First, we knowthat the vicuña is a wild animal or, what is the same, a non-domesticated animal. Secondly, we know that ‘wild’ and‘domesticated’ are mutually exclusive historically successivecategories, mediated by a process of domestication. Andhere is where almost all the theoretical work on domesticationfocuses -the process of transformation from nature toculture. What transforms a natural being into a cultural beingis, of course, a debated matter among different theoreticalperspectives. Nevertheless, what the different theories dohave in common is that they regard domestication as afundamental step in human history. Along this line, plantand animal domestication led man to a new way of life withits diverse demographic, political, cultural and/or economicconsequences (Bender 1978; Binford 1968, 1984; Bökönyi1989; Braidwood 1960; Childe 1952; Cohen 1977; Dennell1983; Flannery 1969; Hassan 1977; Hayden 1990; Higgs 1975;Hodder 1990; Stark 1986). While an analysis of the best

known theories of domestication exceeds the scope of thiswork, I would like to go through some ideas grounded on ananti-essentialist philosophy (Boast 1997; Gosden 1994;Ingold 2000; Thomas 1996).

DOMESTICITYNothing in itself makes a being ‘natural’ as opposed to‘cultural’ or ‘domestic’. That is to say, the wild or domesticstate of an animal does not reside in any essential attributeof the animal itself, but in historically contingentclassificatory schemes. Domestication is not a process thattook place in a particular moment or during a given historicalperiod but, the result of the continuous working ofclassificatory schemes, and the related practices andrepresentations. Thus, domestication is not a fundamentalhistorical turning point. Moreover, it is not a historical turningpoint at all. The view of domestication as a fundamentalhistorical milestone in the history of mankind is more a partof the Western metaphysical self-narrating rhetoric than areal historical reference.

�

Within this theoretical framework, the relationship betweenpeople and animals is based upon an essential difference.The basis for this interpretation comes from the linguisticcontent of the word ‘domestication’, in which people andanimals are related to each other through domination. Thisconcept is significant because domestication is afundamental step for mankind in its progressive rising overnature (for instance, Childe 1952). Different versions of thisperspective identify natural forces with different phenomena:climatic change, demographic stress, or an evolutionaryprocess. Domestication means liberation from nature, sincematerialist theories hold that the fundamental interest ofhuman organizations is subsistence, and that theircontinuation -or social reproduction- is also dependent on it.

From a critical perspective of linear materialism, other diverseand recent dialectical perspectives have tended to regardsocial human organization as a major issue leading todomestication (Bender 1978). An alternative view deals withlinguistics. It is based on the idea of the house as the arenaof the household, as derived from the Indo-European *dom-(Hodder 1990). This perspective considers domesticationcannot be understood without accounting for the socialand symbolic relationships included in the domesticatingrelationship. The world has to be thought of in terms of awild nature and a domesticated culture; society has to bethought of in terms of corporate units capable of preventingpeers from having access to natural objects; animals haveto be thought of as objects of control and property. Fromthese perspectives, domestication is also seen as afundamental gap, even if it is in the social and/or symbolicrealms where change is to be more clearly realised.

In the case of dialectical and linear materialism theories,animals -or nature as a whole- are understood as objects ofsymbolic and/or social appropriation. Those who manageto domesticate nature, and constitute themselves assubjects, are the social units of different scales that succeedin preventing the appropriation of the objects of nature byothers (Ingold 1987). But nature is considered as equallypre-existent, as a non-cultural non-historical arena on whichsocial appropriation is constructed. We should askourselves whether there exists something essentially natural,as for instance an animal, on which domestication as afundamental cultural transformation could be carried out(Rindos 1984). Is not the concept of animal onlyunderstandable from inside contingent categorical schemes?(Douglas 1990; Ingold 1989). My intention is to shift from‘domestication’ -as the relationship between people andanimals-, to ‘domesticity’ -as the relations between two kindsof relationships-. Domesticity, I suggest, does not makereference to one or other pattern of relationships, but to apattern of the relations between both patterns ofrelationships, i.e. a meta-pattern (Herzfeld 1992). Thus, forexample, the relationship between a domestic unit and its

llamas will not be relevant. What will be relevant are therelations resulting from the relationships between socialrelationships and the objects of nature. My idea ofdomesticity is not based on the Indo-European meaningsderived from the root *dom-, but on those derived from theQueshuaymara root uyw- (Haber 1999, 2001a, 2001b). Thislinguistic Andean constellation is relatively similar to theIndo-European one, but it differs in: its emphasis onreciprocity and complementarity between terms (instead ofon control and domination); on inclusive relationships(instead of on oppositional ones); on the supernatural asimmanent to entities (instead of being transcendental to theworld); on objects as subjects (instead of on subjects asobjects).

At this point, it is worth considering a minor reflection onthis theoretical-methodological tool introduced above, inregards to clarify its differences with an ethnographicanalogy. I realise that -if it is true that to understand theworld through linguistic categories is inescapable- theadoption of a Queshuaymara constellation at least puts usat a distance from the Indo-European one already operatinginside our linguistic schemes. Thus, I argue for an optionthat sets up a methodological and theoretical self-criticismof what has always been assumed before any methodologyand theory1. In other words, if we ought to think the worldthrough given structures, i.e. relationships amongstconcepts, I prefer a less naïve conceptual standing, albeitequally contingent. This shift is suggested through my useof ‘domesticity’ instead of ‘domestication’. This means adouble challenge for archaeology, for not only must it proveto be able to trace the history of domesticity, but it shouldalso do it through material networks, and not merely throughlinguistic ones.

Now I will go over one particular case, that of TebenquicheChico, to illustrate how an archaeological study ofdomesticity may be carried on. My argumentation willfocalise on vicuñas and on the heterogeneous networksthey integrate, as my intention is to show the potential of anon-essentialist theory of domesticity.

TEBENQUICHE CHICOTebenquiche Chico is a small valley situated less than 10 kmaway from the Tebenquiche Chico Mountain (5,600 m abovesea level) to the Antofalla Salt Plain (3,350 m above sealevel) (Figure 1). A little permanent stream has eroded thebase of the valley forming two terrace levels. From 4,000 mabove sea level to the salt plain shore, the valley wastransformed into an agricultural oasis during the FirstMillennium A.D. Several irrigation canals carried the waterfrom the brook through the lower terrace and the slopes tothe higher terraces (Figure 2). Agricultural fields were

�

irrigated with independent parallel irrigationnetworks, each one disposed in successivealtitudes. These fields were built by slope terracing(in the slopes of the higher terraces and in thewestern slope of the valley), stone-lining the openspace of the higher terraces. The remaining spaceof the higher terraces was also irrigated, but thefields were not stone-lined (Figure 2).

In the middle sector of the valley (from 3,600 to3,800 m above sea level), 14 domestic compoundswere built, each one associated with a particularirrigation network (Figure 2). The houses and patioswere built with stone masonry made of roughboulders. Each house was built near the rigidityline (Barceló 1988) of its respective irrigationnetwork, that is, where the irrigation canal reachedthe higher terrace; in that rigidity line, the amountand quality of land to be irrigated and appropriatedby each associated domestic group was determined.The whole Tebenquiche Chico settlement systemof the First Millennium A.D. shows theoverwhelming importance of domestic socialorganization, determining the scale of both the livingplaces –houses- and the appropriation of productiveresources – agricultural irrigation networks-. In theupper area of the valley, near the water springs at4,100 m above sea level, some stone masonrystructures were built, probably associated withherding and/or hunting activities. Almost all of theFirst Millennium A.D. dwellings were abandonedduring the 12th. century A.D. Whereas almost all ofthe houses were reoccupied during the 16th. and 17th.centuries A.D. (the early period of Hispaniccolonization of the Peruvian viceroyalty, characterisedin this marginal area by a loose to non-existent colonialstate presence), this paper focuses on the FirstMillennium occupation (Figure 3). Several structureswere excavated in Tebenquiche Chico (TC) in TC1,TC27 and TC2 compounds. The data presented inthis work comes from TC1 (Figure 4).

ZOOARCHAEOLOGY OF TC1The TC1 compound has two contiguous dwellingstructures. Room A1 has an entrance from the patioA5, and is communicated to room A2 through anopening in the adjacent wall. The zooarchaeologicalmaterial comes from the floors (palimpsests), pitsdug in the interior floors, fillings in the doublemasonry walls, and deposits from collapsed walls2.A total of 3,801 bone specimens were recovered fromTC1, 65.2 % (n=2,479) of which were identified, atleast, at familiar taxa level. The identified specimens

Figure 1. Circum-Puna region in the South Central Andes, showing theTebenquiche Chico settlement.

Figure 2. Central area of the Tebenquiche Chico valley, with the greatestconcentration of domestic dwellings. Irrigation networks, and the houses built

�

correspond to Camelidae (98.11%), Rodentia (1.53%), Felidae(0.18%), Canidae (0.11%), and Aves (0.07%).

Since there is neither zooarchaeological nor present dayexistence in the area of non-Camelidae Artiodactyla, severalspecimens that were not identified at familiar level but asorder Artiodactyla, were grouped with the Camelidaeaggregate. The Camelidae bones were analysed and classifiedaccording to genus and species through morphological clues(Benavente et al. 1993) on fragments of skull, mandible,teeth (Puig 1988), vertebrae, scapula, humerus, radius-ulna,pelvis, femur, and tibia, and metric analysis on phalanges

(Madero 1992). As the morphological analysisresulted in the species identification as Lamaglama (llama) and Vicugna vicugna (vicuña), andnot as Lama pacos (alpaca) or L. guanicoe(guanaco), several combined taxa aggregates werelumped together, whenever possible, as llama orvicuña aggregates. As a result of the comparisonof bone size and stage of epiphysial fusion, severalother specimens were classified as llama or vicuña.The remaining Camelidae specimens weregrouped in a llama/vicuña aggregate3. Only 9.7%(n=237) of the Camelidae aggregate wassuccessfully classified as llama (30%) or as vicuña(70%); the remaining 90.3% was grouped in thellama/vicuña aggregate. Of course, it could bereasonably assumed that the remaining specimensin that aggregate belong to one or the other group.As the Camelidae aggregate represents 99% ofthe mammal bone specimens from TC1, its internalclassification is of major relevance. Assuming that90.3% of the non specifically identified Camelidaespecimens are either of llama or vicuña, andassuming that the procedures followed for thespecies determination do not introduce significantbiases, the 9.7% classified as llama or vicuña is tobe considered as the key to the understanding ofthe whole zooarchaeological record of TC1.

Llama Cuisine

‘Llama’ bones deposited in the house at TC1 inthe highest proportion belong to the head (skull,mandible, and incisor), the sacrum-lumbar section(lumbar vertebrae and pelvis), the limb (scapula,humerus, and radius-ulna), the hindquarter (femurand tibia) and the feet (1st phalanx). Thosedeposited in the lowest proportions belong to theneck (cervical vertebrae), distal leg (metapodia)and foot (2nd phalanx). The bones from the thorax(thoracic vertebrae, ribs, and sternum) and carpals,tarsals, and 3rd phalanges could not be classifiedat species level through any of the techniques

mentioned above. The distribution of llama lumbar vertebraeand pelvis shows a peak that is not present in the case ofthe vicuña or the Camelidae aggregates distribution. Thisdifference may be associated with the overall higherfrequency of llama bones from the axial skeleton as a whole.In comparing the frequencies of vicuña and llama bones(Figure 5), the llama aggregate shows a higher proportion ofskull, incisor, lumbar vertebra, pelvis and scapula bones.

Considering the domestic context of deposition of animalbones at TC1, it seems to be related to consumptionmediated by processing activities that, for some reason,

Figure 3. Radiocarbon dates from TC1, calibrated with OxCal 3.5. All theradiocarbon dates are from charcoal samples.

Figure 4. Detail of domestic compound at TC1, showing rooms A1 and A2.

�

tended to be performed indoors. This may have been thecase as consumption needed complex and/or prolongedprocessing for nutrient extraction, including elaboratedculinary processes, and probably also boiling. If these werethe cases, the deposition expectation would not necessarilyinclude bones of high meat ranking, but those bonesassociated with food contents better extracted through theirprolonged boiling for the extraction of bone fat, marrow,endocraneal organs, etc. That would be the case for thehead (skull and mandible), and legs (humerus, radius-ulna,metacarpal, femur, tibia and metatarsal). All skull, mandibleand long bones from TC1 are fractured, and a good numberof the long bones shows traces of dynamic punctual load,as would be expected in fresh fractures in the diaphyses formarrow extraction. The head and long bones have a highfrequency in the Camelidae bone distribution. Thus, at leastpart of the bone deposition can be interpreted in relation tothe final consumption contexts, mediated by culinaryprocessing including prolonged boiling. Cooking andconsumption probably took place in the wind-protected areaof the house, around one of the hearths there located.Therefore, maybe for both functional (prolonged boiling)and social reasons (food transformation, service andconsumption) head and long bones tended to be finallyprocessed and consumed in the domestic indoors spaceand disposed in indoors refuse features (as pit fills).

Vicuñas for Export

If both Camelidae and vicuña aggregates (Figure 5) are takeninto account, phalanges have the highest frequency in bonedistribution. It is quite obvious that their pre-eminencecannot be explained on the basis of their meat, marrow or fatcontent (see De Nigris 2000). The phalanges from TC1 arenot fragmented nor do they display any dry or mineralizedfractures; neither do they present traces of punctual dynamicload for marrow extraction, or any kind of cutting markswhatsoever. Therefore, food consumption can be discardedas the determinant factor for the contextual deposition ofphalanges.

Following the interpretation outlined above forllama bones, the indoor bone deposition can bethought of in relation to indoor final processingand consumption. Even though the amount ofwool associated with the feet of the vicuñas is solittle that could be ignored, and considering therewas no real functional need for the feet to remainattached to the hides for the transportation of thevicuña hides, it is possible to explain the highfrequency of vicuña phalanges (the highestfrequency in TC1 Camelidae bone distribution) inrelation to the indoor location of the processingof vicuña hides (with phalanges attached) for wool

extraction. Vicuña wool is the most valuable vicuña resource,and it is of considerable higher quality than that of the llama.Not surprisingly, then, it was particularly required in pre-Columbian times for high quality textile production. Vicuñawool was one of the main local productions for exchangeduring the Colonial and Republican Periods (Puló de Ortiz1998).

The appropriation of vicuña wool, which was a primaryproduct, meant the death of the animal and it involvedhandling the hides. In the case of the extraction of llamawool, it may have involved live animal shearing, thus beinga secondary product. Several ethnographic observationshave shown that feet phalanges are not usually consumed-the hide being very difficult to separate from them- andthat, occasionally, they are taken from one place to anotherstill attached to the hide. These observations suggest thatthere might be a possible but not necessary relation betweenhides and phalanges. It is possible that phalanges were notconsidered as food, as they may have been transportedattached to the hides and separated from them at the momentof wool extraction. Once the shearing of the hides had takenplace, the phalanges could have been deposited –possiblytogether with the skin of the feet- as a by-product of thatactivity. If this was the case, there might have been otherby-products together with the phalanges, in the form oflithics with long cutting edges, wool, and hide discards.Moreover, the death age profile should be coherent with avicuña wool resource exploitation strategy.

Some of the most important stone-made tools found at TC1were black basalt cutting tools, varying in size from mediumto large, sometimes backed, with very acute natural edgesor acutely retouched ones. These tools may have been usedto separate the wool from the hides. There were 25 fragmentsof hair, hide, and wool found at TC1, 19 of which belong tovicuña, the remaining being plant, human or llama fibers(Haber 2001a). Vicuña fragments include: by-products ofthe whole wool and textile sequence of production -fromhide cuttings with wool and fleece fragments to disorderedbunches of fibers-; threads -both S and Z spun into threads-;twisted threads; a bobbin of threads and textile fragments.

Figure 5. Differences between llama and vicuña bone frequencies. Bones that werenot identified at species level have been omitted, as well as non-confidentidentifications.

�

Llama and vicuña bone specimens were classified accordingto the age of the animal at the moment of death. The resultantage categories were organised into two groups: ‘young’(specimens under 19 months) and ‘mature’ (specimens over18 months) (Figure 6). Of the 42 llama-bone specimens foundin the house at TC1 that were aged, 29 belong, mainly, tosexually immature young individuals killed before bodygrowth had been completed.

The 3:1 pattern of young to mature animals found in thellama assemblage might be an indication of the existence ofa selective culling pattern oriented, in general, to meatproduction. A total of 139 vicuña specimens were aged,only 10 of which were classified as young and the rest asmature. Nine out of ten specimens were from mature vicuñas,which, in most cases, were over two years old. The greatmajority of the vicuñas were already grown individuals, aculling pattern not compatible with meat-oriented predation.Instead, it could be proof of a selective hunting of matureanimals for their wool, which reaches its maximum productionat the age of two (Hofmann et al. 1983). In short, llamas andvicuñas can be said to have been clearly the object ofdifferent social practices.

Llamas and vicuñas as Categories of Practice

It is possible that llamas were socially appropriated andowned alive, and culled mainly for their food. The deadindividuals, sexually immature in most cases, were processedin different ways, except for the skull and long bones thatwere usually boiled indoors where the endocraneal organsand marrow were consumed. This might not have been thecase of vicuñas, which may not have been owned alive bypeople, but hunted by groups or individuals. The preys,generally mature individuals, were taken to the oasis,presumably already butchered and shared amongst thehunters. The hide, on the other hand, was sociallyappropriated by domestic groups, taken indoors (withphalanges attached) and processed for its wool. The wool,which underwent several processes of transformation, couldhave been exchanged in long distance trips to lower altitudevalleys, where vicuña wool must have been highly valued.Even though it would not be reasonable to expect to findevidence of long distance exchanges of a non-durable non-sourceable material as that, this remains a possibility giventhe huge amount of data on the demand of vicuña wool innon-producing areas and the interesting pattern ofallochthonous materials found in the same TC1 house thatvirtually maps the whole circum-Puna area (see Haber 2001afor a detailed account).

Thus, llamas and vicuñas belong to two different categories,not because they diverge in their biological taxonomy or intheir functional status, but also because they were object ofdifferent social practices.

GRAFFITTI OF TC1There is an additional series of features that are relevant inthe discussion of animal categories at TC1. In room A2 –theinnermost room of the house-, seven pictographies werefound, four on wall boulders of the western wall (Figure 7),and the remaining three on boulders that had presumablyfallen from the walls. Red and yellow ochre was utilised indepicting llamas/vicuñas and other less clear Figures.Despite the fact that the pictographies do not seem tointegrate a scene, they are stylistically and thematicallyrelated (Figure 8). Llama/vicuña figures in yellow, depictedin profile and in static attitude, are a recurrent theme; thesefigures appear associated with red vertical elongated figures,believably humans (though this cannot been confirmed).There are also a couple of black and white figures, but theyare hardly interpretable.

The pictographies at TC1 are difficult to interpret; aninteresting starting point for their analysis could be theircontextual situation inside room A2. As mentioned above,this is the innermost domestic space whose access –bothphysical and visual- was more restricted. Whether the red

���

�����

����

��

�����

�����������

�

��

��

��

��

���

���

���

�����������

Figure 6. Frequencies of ‘young’ and ‘mature’ aged-bone specimensof llama and vicuña.

�

vertical elongated figures accompanying the llama/vicuñayellow figures were meant to be humans or not, that cannotbe confidently said. What can be said is that the llama/vicuña yellow figures were not abstract and isolated figures,but were represented in relation to other beings. Theserelated with llamas/vicuñas images were depicted in thespace where social domestic relationships were producedand reproduced. There is nothing in the pictographiesunambiguously saying that the llama/vicuña figuresrepresent animals appropriated by particular domesticgroups. But those related-with-llamas/vicuñas images wererepresented in the innermost restricted domestic space. Thedomestic space, as an arena in which the relationships ofdomestication of nature were reproduced by, and reproducethe relationships of being domesticated, included therepresentations of the relationships with llamas/vicuñas in

the innermost domestic inclusion andextra-domestic exclusion.

ANIMAL RE-CATEGORIZATION:PRACTICES ANDREPRESENTATIONSOne way to analyse the relationshipsbetween practices and representationsis comparing bone deposition indoorsat the TC1 dwelling and thepictographies on the walls of the same

house. The practices of disposal of bones in the innermostpart of the house, linked with the last phases of foodprocessing, boiling, service, and consumption, point atdiverse interesting issues. The category distinction betweenvicuñas and llamas seems to be clear.

Llamas were culled, in general, during their first year of life.After a long process of butchering and consumption –which,for the moment, we know only by intuition- some of theparts of the animals were processed and consumed indoors,probably involving complex culinary elaboration, such asprolonged boiling. The existence of remains of hearths andceramic containers in the same domestic context may favorthis interpretation. At the same time, culinary andconsumption practices may have symbolically reinforcedthe social meaning of the indoor space as domesticallyrestricted. Vicuña remains point at different patterns of

socially restricted consumption notso heavily linked to cooking andconsumption -as in the case of llamas-, but to wool extraction from hidesand its eventual appropriation as agood, directly or indirectly orientedtowards exchange.

The spatial context of this practicecould not be more sociallysignificant: it is the same sociallyrestricted space that formed andreinforced the production andreproduction of the peasant domesticgroup. The preponderance of overtwo-year-old animals in the vicuñaage-profile suggests a carefulselection of culling practices,probably linked to wool as aproductive objective. We can assumethat the llamas were herded animalswhile the vicuñas were hunting preys.That distinction seems to bestrengthened, as both categories

Figure 7. Western wall of Room A2 at TC1, showing the location of the pictographicrepresentations.

Figure 8. Details of pictographic representations in Room A2 at TC1, on both standing (70, 85,87, 88) and fallen (89 and 93) wall boulders. Upper row: 89, 70, 86. Lower row: 85, 93, 88, 87.

�

were demarcated by differences in the social access to theresources: socially restricted access to llamas and sociallygeneralised access to vicuñas.

But, even if such was the case, a major issue should beraised: how was it that the most important resources of bothproductions ended up being appropriated by the same socialsubject –that is, the domestic group. It is at this point thatthe ambiguity of the pictographic representations couldshed some light on the ways in which the poetics ofdomesticity complemented the politics of domesticity. Itcould be expected that two so clear-cut categories of practicebe equally clear-cut represented. However, as I have pointedout, two of the most important features of the pictographiesat TC1 are: first, that they are ambiguous as regards categoryanimal demarcation, and secondly, that the figures are relatedto other equally ambiguous beings, nevertheless neitherllamas nor vicuñas. Thus, the pictographies could be readas the representations of ambiguously defined relationshipswith llamas or vicuñas.

The locus where these representations where found shouldbe included in a discussion about their meaning. They weredepicted in the innermost room of the domestic compound,where the architecture can be seen as a device of socialexclusion. The ambiguity of the representation (Tilley 1991)of the related-with llamas or vicuñas may be read as anambiguity in the politics of access to resources, whetherrestricted and domestic or generalized and non-domestic.Moreover, this pictographic discourse about therelationships with ambiguously defined llamas or vicuñaswas socially restricted to the domestic group.

Domesticity at TC1

When bone data from the same dwelling are brought intothe discussion, several questions about domesticity arise.As it can be seen from the integration of pictographic andbone data, domesticity cannot be defined merely as therelationships with animals -whether llamas or vicuñas- or associal relationships, but as a network of practices andrepresentations elaborated on the relationships betweenrelationships with animals and relationships about animals.Though vicuñas were probably hunted for the domesticappropriation of their wool, this activity may have involvedsomething else than just the technical processes. Maybe,the socially accepted definitions of categories of objectswere being challenged, perhaps making place for newrepresentations of reality leading to alternative practices asregards the social subjects of the appropriation.

Reflection on Animal Categories

The ideas that have been discussed in this text may shockthe reader, in a way, for their strangeness. This could be the

case since we are used to thinking about vicuñas as naturalobjects and not as objects of domestic appropriation. Whatis the origin of our own representations of the vicuña as anatural object? My answer is: the present rhetoric aboutvicuñas as a natural resource, closely linked to the statepolicies for the preservation of the fauna, particularly thepolicy for the preservation of vicuñas. One of the effects ofthat policy is the display of state power for the exclusionfrom that resource of both peasants and traders. If we relatethe representation of the natural naturalness of vicuñas withthe present projects for natural resource management andvicuña-breeding in captivity conditions, sponsored by thenational and/or provincial states, there we have a secondexample of the ways in which domesticity links recursivelythe relationships with animals and the social relationshipsabout animals. Only in this case, the representations do notdraw on ambiguity, but on the categorization of vicuñas asnatural objects; and the subjects of appropriation are notthe peasant domestic groups, but the national and provincialstate and its associated agencies.

FINALE: POETICS AND POLITICSOF DOMESTICITYThe vicuña as a resource was much exploited by theinhabitants of the First Millennium A.D. of the Puna deAtacama oases. Vicuña meat was presumably sharedamongst the hunters, while the wool was appropriated byeach domestic social group, probably as an exchange good.Given that the vicuña wool was particularly apt for theelaboration of high quality textiles, it was a valuable good inthe lower valleys of the South Central Andes. Pottery fromall the regions surrounding the Puna de Atacama was foundin the excavations at TC1, suggesting long distanceexchange relations with the broader circum-Puna region(Haber 2001). While wool remains are hardly found in theregional archaeological inventories, locally availableobsidian has been found to have had a broad regionaldistribution during the First Millennium A.D., an indicatorthat the circum-Puna region was articulated through sporadicor periodic long distance trips (Scattolin and Lazzari 1997).The domestic appropriation of the vicuña wool also pointsat the social scale of these long distance ventures.

Domestic appropriation of vicuñas as a resource implied aseries of practices leading to the technical and socialmanagement of the wool, and the political exclusion of similardomestic groups from particularly appropriated resources.The political differences regarding the appropriation of llamaand vicuña resources were grounded on both being differentcategories, but the domestic appropriation of vicuñas couldintroduce an incongruence between practices andrepresentations. The animal representations in sociallyrestricted spaces could open the possibility for performing

discourses about resources, and the ambiguousrepresentation of practices belonging to different categoriescould be the means for introducing a change in politicallydemarcated resource categories. The ambiguity in meaningof the figures represented can be regarded as a metaphor, apoetical means of introducing a plus of signification usingsignificants at hand. Examining the mutual implications ofthe politics and the poetics of domesticity, as they arerevealed in the pictographic, architectural, zooarchaeologicaland landscape data from Tebenquiche Chico, the meta-pattern of the human-animal and social relationships can behistorically traced.

Reproduction and change in human-animal relationshipscannot be adequately understood under the limited scopeof the theory of domestication. Uywaña-derived theoreticalelements can be refreshingly introduced in the discussion.The recursive relations between relationships amongstpeople and the relationships between people and animalsare necessary for the understanding of both patterns. Onepattern of relations is not fully understood without the other,and vice versa. Thus defined, the concept of domesticity isfundamental to understand the reproduction and change ofpast social, economic and symbolic practices and structures.

As a way of conclusion, I would like to go back to thebeginning. As I said, I am not writing about the domesticationof vicuñas. I hope readers do not think that I have adopteda mere rhetorical strategy. There is a theoretical backgroundthat must be understood also as a methodological one. Imight say that the issue is neither about domestication, norabout vicuñas or llamas. The relevant question is: what arewe talking about when we speak of domestication? In thefirst place, if it is considered the present representations ofvicuñas as a protected natural resource, our own theoreticaland discursive categories about these animals are not merelylinguistic concepts without operation on reality. Hence, ourpoetics of domesticity involves politics of domesticity. But,on the other hand, domestication –exemplified with the caseof TC1-, or the social and cultural practices of appropriationof nature, cannot be simply understood as technical devicesof control and domination of the external world. Thus, thepolitics of domesticity involves a poetics of domesticity.

AcknowledgementsField research was made possible through a research grantfrom the Universidad Nacional de Catamarca. Severalstudents from the Universidad Nacional de Catamarca andthe Universidad Nacional de La Plata took part in theexcavations at TC1, from 1990 to 1997. Juanito Ferreyracollaborated with the zooarchaeological analysis of theassemblage from TC1, Marcos Quesada helped with theediting of the Figures, and Claudio Revuelta contributedwith some bibliographical requirements. Both Francisco

Gallardo and Laura Miotti contributed greatly to theimprovement of this text. A preliminary version of this paperwas presented in Rosario in September, 2001. Errors andmisunderstandings are solely mine.

REFERENCES CITED

Barceló, M.

1988 La arqueología extensiva y el estudio de la creacióndel espacio rural. In Arqueología medieval. En lasafueras del ‘medievalismo’, edited by M. Barceló, H.Kirchner, J. M. Lluró, R. Martí and J. M. Torres, pp.195-274. Crítica, Barcelona.

Benavente A., M. A., L. Adaro A., P. Gecele C. and C.Cunazza P.

1993 Contribución a la determinación de especies animalesen arqueología: familia Camelidae y taruca delnorte. Universidad de Chile, Vicerrectorría Académicay Estudiantil, Departamento Técnico de Investigación,Santiago.

Bender, B.

1978 Hunter-Gatherer to farmer: a social perspective. WorldArchaeology 10: 204-222.

Binford, L. R.

1968 Post-pleistocene adaptations. In New perspectives inArchaeology, edited by S. Binford and L. Binford, pp.313-341. Aldine, Chicago.

1984 In pursuit of the past. Thames and Hudson, London.

Boast, R.

1997 A Small Company of Actors. A Critique of Style.Journal of Material Culture 2(2):173-198.

Bökönyi, S.

1989 Definitions of Animal Domestication. In The WalkingLarder. Patterns of Domestication, Pastoralism, andPredation, edited by J. Clutton-Brock, pp. 22-27. OneWorld Archaeology, vol. 2, edited by P. Ucko. UnwinHyman, London.

Braidwood, R.

1960 The agricultural revolution. Scientific American203:130-148.

Childe, V.G.

1952 The new light on Most Ancient East. Routledge andPaul, London.

Cohen, M. N.1977 The Food crisis in Prehistory. Yale University, NewHaven.

�

De Nigris, Mariana E.

2000 Procesando para el consumo: dos casos de Patagoniameridional. In Desde el país de los gigantes.Perspectivas arqueológicas en Patagonia, vol. 2, pp.401-414. Universidad de la Patagonia Austral, RíoGallegos.

Dennell, R.W.

1983 European Economic Prehistory. Academic Press,London.

Douglas, M.

1990 The Pangolin Revisited: a New Approach to AnimalSymbolism. In Signifying Animals. Human Meaningin the Natural World, edited by R. Willis, pp. 25-36.One World Archaeology, vol. 16, edited by P. Ucko.Unwin Hyman, London.

Flannery, K. V.

1969 Origins and ecological effects of early domesticationin Iran and the Near East. In The domestication andexploitation of plants and animals, edited by P. Uckoand G. W. Dimbleby, pp. 73-100. Aldine, Chicago.

Gosden, Ch.

1994 Social Being and Time. Blackwell, Oxford.

Haber, A. F.

1999 Uywaña, the House and its Indoor Landscape: ObliqueApproaches to, and Beyond, Domestication. In ThePrehistory of Food. Appetites for Change, edited byC. Gosden and J. Hather, pp. 59-82. One WorldArchaeology, vol. 32, edited by P. J. Ucko. Routledge,London and New York.

2000 Una arqueología de los oasis puneños. Domesticidad,interacción e identidad en Antofalla. I y II milenios d.C. Unpublished Doctoral Thesis. Facultad de Filosofíay Letras, Universidad de Buenos Aires, Buenos Aires.

2001a El oasis en la articulación del espacio circumpuneño.Actas del XII Congreso Nacional de ArqueologíaArgentina 13(1):251-267. Córdoba.

2001b Observations, Definitions and Pre-understandingsin the Ethnoarchaeology of Pastoralism. InEthnoarchaeology of Andean South America.Contributions to Archaeological Method and Theory,edited by L. Kuznar, pp. 31-37. EthnoarchaeologicalSeries, vol. 4. International Monographs in Prehistory.Ann Arbour.

Hassan, F.

1977 The dynamics of agricultural origins in Palestine: atheoretical model. In The origins of agriculture, editedby C. Reed, pp. 685-609. Mouton, The Hague.

Hayden, B.

1990 Nimrods, Piscariots, Pluckers and Planters: theemergence of food production. Journal ofAnthropological Archaeology 9:31-69

Herzfeld, M.

1992 Metapatterns: Archaeology and the Uses of EvidentialScarcity. In Representations in Archaeology, editedby J.-C. Gardin and C. S. Peebles, pp. 66-86. IndianaUniversity Press, Bloomington and Indianapolis.

Higgs, E. S.

1975 Palaeoeconomy. Cambridge University Press,Cambridge.

Hodder, I.

1990 The Domestication of Europe. Structure andContingency in Neolithic Societies. Basil Blackwell,Oxford.

Hofmann, R. K., K.-Ch. Otte, C. F. Ponce y M. A. Ríos

1983 El manejo de la ‘vicuña’ silvestre. Tomo II. DeutscheGesellchaft für Zusammenarbeit, Eschborn.

Ingold, T.

1987 The Appropriation of Nature. Essays on HumanEcology and Social Relations. University of IowaPress, Iowa City.

Ingold, T. (editor)

1989 What is an animal?. Unwin Hyman, London.

Ingold, T.

2000 The perception of the environment. Essays inlivelihood, dwelling and skill. Routledge, London.

Madero, C. M.

1992 Análisis faunístico de Huachichocana III (Jujuy):identificación arqueológica de la caza y el pastoreo decamélidos. Palimpsesto 2: 107-122.

Martínez, G.

1989 Espacio y pensamiento. I. Andes meridionales.Hisbol, La Paz.

Puig, S.

1988 Craneología y craneometría de camélidos:diferenciación interespecífica y determinación de laedad. Xama (Publicación de la Unidad deAntropología) 1:43-56.

Puló de Ortiz, M.

1998 La ‘vicuña’: el oro que camina por los Andes. Andes.Antropología e Historia 9:243-280.

��

Rindos, D.

1984 The Origins of Agriculture. An EvolutionaryPerspective. Academic Press, Orlando.

Scattolin, M. C. and M. Lazzari

1997 Tramando redes: obsidianas al oeste del Aconquija.Estudios Atacameños 14: 189-209.

Stark, B.

1986 Origins of food production in the New World. InAmerican Archaeology, past and future, edited by D.Meltzer, D. Fowler and J. Sabloff, pp. 277-321.Smithsonian, Washington.

Thomas, J.

1996 Time, Culture and Indentity. An InterpretiveArchaeology. Routledge, London and New York.

Tilley, C.

1991 Material Culture and Text. The Art of Ambiguity.Routledge, London.

Van Kessel, J. & D. Condori Cruz

1992 Criar la vida: trabajo y tecnología en el mundoandino. Vivarium, Santiago.

NOTES1 Two reviewers of this paper criticised this contention. Oneof them argued that it is inadequate to adopt an Andeancategory instead of an Indo-European one. The other basedher criticism on the difference between ‘emic’ and ‘etic’categories. I must say that both of them miss my point. I amnot trying to introduce a term that represents reality morefaithfully. There is, from my perspective, no essentialdifference between reality (to be represented) and linguisticcategories (to represent). I agree with the idea that both theIndo-European and Andean categories are historicallydetermined. In fact, I am not trying to ignore the

determinations of history, nor am I trying to contribute toany universal language accounting for objective reality. Thatso-called ‘universal language’ bears no relation whatsoeverto my analysis. Hence, domestication does not belong to anetic perspective (scientific) as opposed to an emic one (as‘uywaña’). Even for those who consider that there is a realdifference between etic and emic perspectives, the fact that‘domestication’, as a linguistic category, comes from thenatural language and then is applied by science isunquestionable. Its Indo-European root, that relates it tothe exercise of power over other beings, is historicallyprevious to any scientific idea on it. In this sense, theconcept is emic before being etic. It is our language, and notscience, that provides our minds with categories to describethe world. I am not suggesting that because I use an Andeancategory, language will resemble reality more accurately. Itsimply provides me with a clearer notion of the dis-adjustment between my language and the world. Only whenthat displacement has been operated, one begins tounderstand that the concept of ‘uywaña’ can contributeimmensely to the understanding of Andean society,economy and symbolism (Martínez 1989; van Kessel andCondori Cruz 1992).

2 For a detailed zooarchaeological analysis -includingtaphonomic issues of the bones at TC1- and for a descriptionof the architecture and stratigraphy, see Haber 2000. Eventhough all the contexts excavated at TC1 have been includedin this study, not all of them can be considered ‘domestic’.Several contexts are related to the construction of the houseand not to its dwelling. In the Figures included in this paperit can be observed that all the contexts are grouped together,but the NISP counts for the construction contexts representonly 4% of the total NISP, thus not introducing a strongbias to the Figures.

3 The llama/vicuña aggregate was so named on theassumption that the bones that could not be successfullyidentified -but were assigned to the Camelidae family-,belong to one or the other species effectively identified.

��

��

��

��

��

��

��

�