Artigo Revisado por pares

Burning the truck and holding the country

2016; HAU-N.E.T; Volume: 6; Issue: 1 Linguagem: Inglês

10.14318/hau6.1.035

ISSN

2575-1433

Autores

Fred Myers,

Tópico(s)

Land Rights and Reforms

Resumo

Previous article FreeBurning the truck and holding the country Pintupi forms of property and identityFred MyersFred MyersNew York University Search for more articles by this author New York UniversityPDFPDF PLUSFull TextEPUBMOBI Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreIf they are not given their shares, this denies their kinship. . . . A man's variegated relationships with others run through his chattels as well as his land; and the measure of how far he feels the correct sentiments in those relationships is the way he deals with his property and his produce (Gluckman 1965: 45).The social life of objectsI am concerned here with the indigenous meanings attributed to a variety of "objects" among Pintupi-speaking Aborigines of the Australian Western Desert.1 My argument is that "things" (objects, ritual, land, prerogatives, duties) have meaning—that is, significance or social value—for the Pintupi largely as an expression of autonomy and what I have elsewhere defined as "relatedness" or shared identity (Myers 1986a). In this regard, landownership is not a special kind of property, not a special set of rights defining relationships to an ecologically necessary "living space."2 Instead, it is one more form of objectifying social relationships of shared identity.Among the Pintupi, as among many hunter-gatherers, the use-value of rights to things is not at all obvious. Rather than, as Radcliffe-Brown (1930–31) once thought, corporate groups forming around some valued property or estate, Pintupi seem to constitute social aggregations (see also Sansom 1980) and give them identity through time by projecting them into shared relationships to objects.My essay begins, then, not so much in disagreement with Gluckman's views on property but in my discomfort with the notion of property as being too concrete and specific for the meanings that Pintupi give to objects. Legal language may be useful for characterizing certain similarities in the relationship between persons and things, but it does not make for entirely adequate translation. In exploration of this theoretical problem, and in search of deeper understanding of the social significance of things among hunter-gatherers, therefore, I want to focus on similarities and differences between land and other forms of property. The distinction on which I draw is not unlike the French contrast between propriété (personal property) and immeubles (real property), a contrast enshrined in Mauss' (1954) The gift and recently resurrected in Weiner's (1985) article, "Inalienable wealth."Two basic issues arise. First, is land tenure different from relationships between (and among) people and other objects? Second, the question of concerns over temporal continuity is, according to Woodburn (1980) and Meillassoux (1973), central to understanding hunter gatherers. Whereas Woodburn attempted to locate the source of Aboriginal concerns over enduring relationships in the "farming" of women (in bestowal), I see the basis of temporal continuity in the process of objectification. In articulating this process, some sorts of objects have different capacities than others. It is of critical importance, as I shall show, that a Pintupi can "give away" (or share) some rights to named places without losing his or her own intrinsic identity with the place. This inalienability of land, that it cannot really be lost, differs from the way most other objects enter into processes of exchange.While these issues cannot be fully resolved here, I would like to begin by exploring the different ways personal identification with objects occurs in Pintupi social life. How is identity extended in the negotiation of shared rights to objects and processes of exchange? And why, for example, are "personal effects"—but not country—destroyed, effaced, and given away at death? What, if anything, is inherited, or what does inheritance accomplish?Ownership and identificationA basic issue at all levels of Pintupi social organization is transaction in shared identity. What is most impressive to me about the Pintupi conception of objects is the continual negotiation about relationships to them and a willingness to include others as what, for the want of a better heuristic term, I will call "co-owners." Such ambiguity is deeply seated in the negotiated quality of much of Pintupi social life. Relationships among people are not totally given in the rules of a defining structure, not in landownership, kinship, or residence (Myers 1986a). Instead, the relationships must be worked out in a variety of social processes. The politics of Pintupi life, however, should not be confused with an aim for domination over others. Its roots lie in the emphasis placed on shared identity with others as a basis for social interaction.This framework suggests a simple and obvious conclusion, namely, that "property" be viewed as a sign. The immediate material use-value of most forms of propertied objects among the Pintupi (tools, clothing, food, etc.) is not great. While such objects are clearly useful, that which is necessary for simple production is rather easily obtained, constructed, or replaced. The rapidity and ease with which things move through a network of relatives and friends show that objects are important as opportunities to say something about oneself, to give to others, or to share. Like any signs, however obviously "useful" they might be, objects are tokens that represent an opportunity not so much to sustain exclusive use but to constitute other sorts of values defined by a larger system of exchange. That these values are ultimately convertible to labor or political support, far from subtracting from the significance of a semiotic analysis, suggests that such analysis should be based on a temporal perspective that focuses on value as being constituted in the process of reproducing social life.Analysis of the cultural relationships between persons and objects should begin appropriately with Pintupi ideas of "ownership"—a conception better translated as one of "identification." The Pintupi words most closely approximating "property" in English are walytja or yulytja. While the former may be translated as "relative(s)" or as "one's own personal effects" (see Hansen and Hansen 1977: 152) and the latter as "baggage and personal effects" (ibid.: 190), the significance of identification is clearer when one understands the entire semantic range of walytja. In addition to objects associated with a person, it can refer to "a relative," to the possessive notion of "one's own" (such as "my own camp" [ngayuku ngurra walytja] or "my own father" [ngayuku mama walytja]), or to reflexive conceptions like "oneself" (such as "I saw it myself" [ngayulu nyangu walytjalu] or "He sat there by himself" [nyinama walytja]).Examining this range of meanings offers some perspective on the almost natural reifying of legal conceptions of rights and duties in our common usage of "property." Indeed, a similar foundation of property in rights deriving from some concept of personal identification is suggested by the shared root of "proper" (my proper father) and "property" in the Latin term proprius, "private or peculiar to oneself" (Partridge 1983: 529).There is a certain ambivalence or ambiguity about Pintupi relationships to objects. There is clearly a sense that objects might "belong to" someone—the idea that X is the walytja of a person contrasts directly with the idea that an object X is yapunta, a word whose literal meaning is "orphan," without parents. To follow the linguistic usage further, one's parents are said to kanyininpa one, meaning "to have," "to hold," or, more loosely, "to look after one." Thus, an object that is yapunta does not belong to anyone, but it appears that this might mean also that it has no one who is holding or looking after it. The question remains of what it means for something to belong to—to be walytja of—someone. An object becomes yapunta when it is no longer "held," but has instead been wantingu, "lost" or "relinquished," released from an active association with a subject. For objects to belong to someone means as much that they are expressive of that person's identity as it does that they are simply identified with or related to that person. To say that something is "one's own" implies, for the Pintupi, that one does not have to ask (or defer to) anyone about its use.Leaving aside for a moment the issue of co-owners, the right to use an object without asking—even the claim that it is "my own"—expresses one's autonomy. By "autonomy," I mean self-direction, although we will have occasion to see that this is not necessarily self-created. In contrast, rights to objects that might be regarded, as personal effects still seem less exclusive among the Pintupi than Americans, for example, would tolerate. The very notion of ownership as identification provides also a sense that rights to objects can, and should, be more widely distributed, a willingness (not always ungrudging, of course) to include others with oneself as co-owners.Rights to objects enter into a system of exchange that constantly negotiates the relationships of shared identity. In the examples that follow, I want to show how negotiation of the meaning of ownership rights moves within a dialectic of autonomy (as in the right to be asked) and relatedness (exemplified in the tendency to include others and to share rights with others who one recognizes as identified). This sense of property as potentially providing a temporally extended objectification of shared identity has much in common with Sansom's (1980) treatment of fringe-dwelling Aborigines in Darwin as "people without property."To share, perchance to giveLet me begin with a simple and striking example of how Pintupi regard personal possessions. Cigarettes, purchased through the cash economy, are a popular item among Western Desert Aboriginal men. Men cadge each other's tobacco and cigarettes almost without thought. Although my Pintupi comrades were generous with me in sharing their cigarettes, I often found my supply depleted all too quickly. On just such an occasion, a young man named Jimmy came to my camp and asked if I had any cigarettes. Aggrieved, I replied with some anger that people had taken all of mine, more or less including him in the group of those who had taken advantage of me. Rather than taking offense, he was sympathetic and offered me some of his cigarettes. Further, he took it upon himself to explain how I should not give my things away so easily. Instead, he suggested that I should hide what I had. He showed me how he had hidden a packet of cigarettes in his socks under his trousers and suggested that if I did the same, I could simply tell people that, unfortunately, I had no cigarettes—although I would surely give them if I had. Giving me a whole packet, Jimmy told me he had several buried near his camp.This example illustrates some common themes in Pintupi action about sharing, owning, and asking. I interpreted Jimmy to mean I did not have to refuse anyone overtly. It is clear that it is very difficult for Pintupi to simply refuse or to accept from others. Sympathy and compassion are the appropriate and moral responses to co-residents or relatives. Outright refusal, conversely, constitutes an open rejection of the other's claim to having a relationship with one (see Myers 1979). To "say no to someone's face" is something very unusual and dangerous in Pintupi social life. Those so denied may respond with anger and violence.Were Jimmy's buried cigarettes to be taken by others, he would certainly be angry. In similar cases, people talk of "theft." That is, Jimmy might say, "Someone has taken my cigarettes, mulyartalu ("thieving"; ergative case)." Even though he might have been obligated on the basis of his relationship to a thief to give some cigarettes, to take "without asking" (tjapintja wiya) would be regarded as a violation of his personal rights to them. In other cases like this, people often argued that the owner got them with his own money. This is to say that the cigarettes were not a product of cooperative or joint activity through which others could claim identification with the product.No one else would be very concerned about such an event, as long as his or her ox was not gored. Indeed, it might be claimed, in counter, that Jimmy should not have hidden (yarkatjunu) his cigarettes, or, as I have heard in "thefts" of radios and cassette players, that he should have hidden them better (so the thief would not be tempted). Jimmy's claim would be that he should have been asked. Under such circumstances, in which refusal is impossible, the only way to maintain one's possessions is to place them out of immediate reach. To give one's cigarettes in response to a request, however, is to build a right to ask others because reciprocity in these matters is expected. By not asking, it would appear, one is doing more than simply taking an object; one is denying to the "proprietor" the opportunity to give it, that is, to be generous and to thereby build a debt. "Theft," defined as taking without asking, is a serious interruption to one's ability to express the self, albeit through a gift, and to build through exchange an expanded, shared identity with the other.It is illuminating, finally, to reverse the conception offered here. Consider that one might claim a right to be given some object, claiming co-ownership or a relationship with the proprietor that obligated him or her to give. While the proprietor might choose to give, for propriety's sake, or to be diplomatic, he or she might still claim that the other was really "nothing to do" (mungutja)—having no basis for shared identity either with the object or with the proprietor.The exchange of foodIt is obviously not possible here to circumscribe the entirety of Pintupi relationships to things. Nonetheless, consideration of rights in food that is gathered or hunted can inform us more deeply about what is at issue in establishing one's primary identity with things.While women frequently forage for vegetable foods and small animals in a group, each woman has exclusive rights to what she produces when the actual productive activity is not cooperative. Much of her production is brought back to the residential camp for final preparation and consumption. Characteristically, women prepare what they collect themselves. In camp, the product is (1) shared with the immediate family, (2) given in exchange for child care services rendered, for example, while a mother was foraging (B. Clark, personal communication), and/or (3) distributed to co-residents who had not fared well. While such people may have claims on the producer's services, they have no special claim to the product itself. What she gives is conceptualized, therefore, as exchange. Although co-residents in this situation are expected to share with each other, the sharing often takes place only on request, giving to the distribution of production a character of "mutual taking." Based on their technology and resources, cooperation in production among Pintupi men is unnecessary, though beneficial, in certain circumstances. When men engage in cooperative drives, the kill is distributed among all who participate. Most hunting, however, takes place alone or in small groups, and when a man is successful in hunting large game, it is distributed interdomestically, to those others in the residential group who have shared their production with the hunter. Gifts of meat could satisfy other exchange obligations as well, namely, to one's in-laws or parents.Anthropologists have frequently pointed out the practical economic benefits of such interdomestic distribution, so I need not emphasize that point here. In any case, the preparation of large game treats it as a social product. A hunter is supposed to give the kangaroo he kills to others for preparation, although his activity in the hunt provides for him the right and responsibility to direct disposition of the cooked animal in exchange (if he uses another's spear or rifle, the owner of the hunting implement has this privilege). Success in the hunt provides food for the hunter, of course, but this does not exhaust its significance. His success also secures a particular set of rights to the animal, providing him with an opportunity to give—to engage in interdomestic exchange that establishes or promotes a kind of moral identity with recipients.The social value of this exchange of meat is not simply that of caloric satisfaction. Rather, these exchanges among individuals provide a moral basis for continued and ongoing co-residence and cooperation among members of a group. They constitute, that is, a moment in the reproduction of shared identity (people who "help each other") that is the foundation of band organization. Failure to share or, as I would prefer to describe the activity, to exchange within such groups, has predictable consequences. The idiom of shared identity sustained through exchange provides the very basis of possible criticism: conflict ensues as those neglected make accusations of being rejected or neglected as "relatives" (walytja). Such neglect is understood as "not loving," not regarding people as related.Distribution of foraging production in a large group does present a problem. Frequently, the conflicts that develop in such aggregations concern sharing and are brought on by the difficulty of allocating products and services among a large group of co-resident "kin." Since all such co-residents have claims on each other, at least to some extent, large groups place a considerable strain on individuals, producing conflicts of loyalty as well as continuous imposition on one's generosity (not to say diminishing people's incentive to overproduction). Thus, the rights to the kangaroo as property are involved immediately with an exchange used to maintain one's relatedness to others. It is possible, of course, for a person to assert his (or her) autonomy, the right to decide who gets a kangaroo. Rights exist, but what do they mean? Insofar as exercise of choice may defy other people's claims about their relationship or obligations, these choices are likely to create a threat. Herein lies the tension between a valued autonomy and the claims and necessity of shared identity.How Pintupi people manage this tension is clear in a case of "hidden meat." In 1979, I lived at the small outstation community of Yayayi, then, population 15. Having been successful the previous day in hunting kangaroos and bush bustards, we were enjoying the cooked fruits of our labor on one cold midmorning when we heard the approaching sounds of the tractor from the nearby community of Yinyilingki. Longtime and frequent co-residents with the Yayayi mob, the Yinyilingki people were close and often generous relatives only temporarily separated. To my surprise, the male leader of Yayayi decided we should hide our cooked meat, which we did, inside the many flour drums around the camp. When the Yinyilingki people arrived, they asked, of course, whether we had any meat. One of the Yayayi, Ronnie, sitting on top of some drums, replied that we were, unfortunately, empty-handed. His good friend from Yinyilingki, Ginger, was not fooled, since he could plainly see the evidence of recent cooking as well as the feathers we had plucked from the birds. Laughing and without any rancor, he opened a few flour drums until he found what he was looking for. Inescapably, for better or worse in his case, Ronnie's identification with the cooked meat truly spoke to the world about him—communicated, that is, about his identity. Nonetheless, the case is instructive about the cultural meaning of "property" as an expression of identity, a node of personal identity caught at once in webs of shared identity.Materially, Ronnie's strategy of the polite rejection through hiding was a failure, although it did not lead to the conflict or antagonism that shatters the sense of shared identity. I believe this is so because often enough he was generous. Indeed, the strategy of hiding one's property to avoid either having to give it away or having to overtly deny the other is a common enough practice. Ginger recognized Ronnie's rights to the meat, but these rights did not really sustain exclusive use. What Ronnie did with the meat was necessarily and unavoidably meaningful, a sign of their relationship, just as Ginger's jocular but insistent demand highlighted his sense of a closeness that allowed him to intrude. The potential dangers of such selfishness, however, are clearly outlined in Pintupi myths, where long cycles of vengeance follow on a constant failure to share. And conflict over food has altered the relationships in many Pintupi camps, so that the continued identity of a community, what the Pintupi call "the people with one camp" (ngurra kutjungurrara), is essentially a temporary objectification of these relationships of exchange.Motor vehicles as media of identificationAmong the most valuable objects in contemporary Pintupi life are motor vehicles, especially the trucks and four-wheel drive Toyotas that are able to carry large loads (and people) in difficult terrain. Vehicles are necessary and valuable for getting supplies from the store, for hunting expedition, and for visits near and far. The problem of ownership is compounded in the case of such objects, of which there are few, in that Pintupi recognize two categories of "property"—what they call (in English) "private" as opposed to "community" or "company."Vehicles that are private are those purchased with money that belongs to an individual or, occasionally, to some individuals who jointly buy a car. The purchaser is understood to be the proprietor of the vehicle, that is, to have the right to decide on its use and nonuse. It is unlikely that the proprietor will be the sole user of a vehicle, but as in the case of other personal effects, relatives and friends must ask permission to gain its use. The very autonomy of ownership (the possibility of giving) is the basis for the establishment of extending shared identity with others. To have a car, one might say, is to find out how many relatives one has.Those without their own vehicles draw especially on their kin relations for help, placing those in possession of a car under almost constant pressure of such demands. Where the moral rubric of shared identity guides the relationships of those who live in the same camp, such requests are difficult to refuse, and open rejection is impossible. Those who refuse are said to be "hard" or "jealous" for the motorcar. Anyone who has lived in a Pintupi community will recognize just how much conflict, how many fights, are occasioned by relationships to motor vehicles because of misuse and requests for and refusals of use. Indeed, sometimes owners are relieved when their cars break down, only then seeming to be free of demands. Disputes among kin about the use of a vehicle have, I understand, led proprietors to set fire to and destroy their own cars as a desperate and angry resolution.In contrast, "ownership" provides an opportunity for a person to "give," and one who helps his relatives is not only understood to be generous but also gains a degree of respect and authority for having "looked after" them. The demands on a proprietor are not entirely a bad thing, then, insofar as they provide one with an opportunity to "be someone." Proprietorship can make a person central to the planning of activities that require transportation. More important, however, is the general attitude toward personal possessions. No matter what they cost, it would appear, Pintupi regard vehicles as replaceable, like spears or digging sticks. Thus, even when a $4,000 car is destroyed after only a few weeks of use, they say, "There are plenty more motorcars; no worries." To be sure, this reflects an expectation that someone else will have a car, that another relative will help one to obtain the bare necessities—that is, an expectation of long-term reciprocity prevails.I believe this attitude to property underlies much of Pintupi social life. Put in familiar terms, if faced with a choice between caring for their property or for their relatives, they prefer to invest in people rather than things. Without granting this any special moral status, let me say that under the conditions of settlement in the traditional life of hunter-gatherers, such "investment priorities" may be a realistic appraisal of resources. Nonetheless, the conception of property as replaceable guides our attention to a conception of things as relatively transparent signs of social relationships, vehicles for another sort of value. Many Pintupi recognize a difference between their conception of the relationship between persons and things and that of whites. A long discussion I had with a middle-aged man resulted in his contemplating the difference thus: "You white people are always worrying for money. You don't think about who will cry for you when you die." This is a salutary comment on commodity fetishism as Marx comprehended it. The accumulation of private objects, while not entirely ignored (Pintupi do work and save to get cars), is not the means through which one's identity is transmitted through time. That is secured in part through giving such things away.If it is difficult to refuse help, private cars are the basis on which a sort of shared identity is constructed, those with whom one shares the use reflecting an ongoing exchange. Pintupi themselves regard particular vehicles as representing a cluster of associates who, often for the life of the car, travel together: a certain blue Holden is "that Yinyilingki motorcar," identified with a group of young men. The car is the occasion for the temporary realization of their relationship and obligations to each other. When driving past old wrecks on the road, Pintupi habitually identify them with the persons, communities, and events they were involved with. Cars become, in other words, objectifications of a set of social relationships.The other category of motor vehicle, the "community" one, has been the occasion for conflict and confusion in many settlements. Such objects are usually the product of government or foundation grants, not the result of a community's joint, voluntary contribution to a collective enterprise. The problem with community vehicles is, who actually can be said to own something that belongs to a community?Ordinarily, Pintupi have assimilated this problem of property to their ambiguous notion of kanyininpa ("to have, to hold, to look after"), and they recognize that Pintupi Village Councillors "look after" such vehicles. Indeed, in all the cases I have seen, a particular councillor assumes responsibility for a vehicle. "Shorty," they might say, "is looking after that Bedford truck." They do not mean, of course, that he is its "owner." Their conception, as I understand it, is that Shorty must be asked, but conversely, as a Village Councilor he must help them. Although he might try to explain that there are other uses for the truck and try to relate these to everyone's benefit, he cannot really refuse.These community vehicles are interesting from another point of view. To the Pintupi who are granted them, such vehicles seem to embody a recognized "community" identity. When Pintupi refer to the Yayayi Toyota or the New Bore Toyota, they are saying more than that the vehicle belongs to that place. In a particular Pintupi sense, such objects represent the community's collective identity as an autonomous social entity. They do not do so, however, legally; if it crashes, the community is not held liable collectively, and individual members would undoubtedly claim that they were "nothing to do" (mungutja) with the action.For historical reasons, Pintupi associate the founding of past outstation communities with the granting of four-wheel drive Toyotas by the Department of Aboriginal Affairs or the Aboriginal Benefits Trust Fund. As a result of this association, it appears, they believe that a community's autonomy will be recognized in the granting of such a vehicle. Men often say that they are the "boss" of an outstation at such and such a place, but they are waiting to go there because the government has not yet given them their Toyota. Possibly, the attraction of gaining control over such a vehicle is the very reason that people have been eager to establish outstations. Much of the politics of autonomy that matter so much to men gets worked out around control of the motor vehicles of the community. At the first Yayayi community where l worked in 1973, there were two community vehicles, each associated with historically and geographically distinctive segments of the camp, one for the people "from the east" and the other controlled by those whose traditional country had been farther "from the west." The controllers of these resources became the nodes of community organization.The ambivalence about proprietorship of community vehicles is made clear in the case of the death of a person who controlled a vehicle. When the "boss" at Alumbra Bore died in 1981, the community was faced with the problem of what to do with the orange Toyota that was theirs from a community grant. Personal property of a deceased is always destroyed or given away to that person's "mother's brothers" from far away, because a person's effects are identified with him or her and make close relatives sad. The Alumbra community planned initially to swap vehicles with another community in order to remove their truck from sight because it reminded them of the dead man. They eventually thought better of this, realizing they would still have to see this truck often since it would be in the area. They decided

Referência(s)