INTRODUCTION: LAND, MINING AND POLITICS

Old Wapo: The causes of the most serious and most daily, immediate conflicts lie in the unlimited seizure of our lands. (Gope 2001:31)

Unquestionably, over the last five centuries and particularly in the past few decades, countless crimes

have been committed in the name of 'progress'. Nonetheless it is a mistake to assume -- as academic and popular literature has often done -- that small rural groups, faced with exogenous industrial development, are static, homogenous units that are being destroyed by 'modernity' (see Banks 1997:26-30 for a discussion of academic representations of communities as homogeneous, traditional, and/or disempowered). Rather, these communities are, as they have always been, composed of persons, each of whom is adapting to new circumstances, new possibilities, and new desires. Many villages are active, dynamic loci of intense internal conflicts provoked by differences in residents' expectations from the agents of landscape-altering economic development, with whom they engage in reciprocal relationships. The complexities of intracommunity conflicts and villagers' attempts to direct or prevent resource development acti vities may be elucidated by examining the discourses that people implement and especially by paying 'closer attention to the ways in which land, and the identities it confers, are deployed as terms in the debate' (Ballard 1997:61).

This paper engages with issues often addressed in the field known as 'political ecology', as an examination not of the macro-level politico-economic forces behind environmental change but of the engagements of local actors, in the context of huge economic and ecological stakes, with global players and with each other. Specifically, I analyse the relationships of the residents of a few New Caledonian villages with a multinational nickel mining venture known as the Koniambo Project. Perceptions of this project's potential costs and benefits have both created strains within the community and catalysed dormant intra-community tensions into factional conflicts.

Because of their different positions within local social hierarchies, which inform their relationships with the mining company, villagers stand to gain or lose more or less than their neighbours. Although nearly all the villagers with whom I spoke expressed a desire for the economic development the project would bring, particular sub-groups felt the need to make sure that certain components of the project would not result in a loss of their control of the land. To support their positions, some local people emphasised the importance of maintaining local ecosystems and cultural heritage while others highlighted the project's economic benefits. They also worried to varying degrees about the importance of appeasing the area's spirits. These concerns were genuine; however, in my analysis, what underlay and conditioned them were the villagers' desires for respect of their customary and/or legal rights. They especially aspired to determine what happened to their land, the source of their identity and dignity -- thei r primary 'symbolic capital' (Bourdieu 1994).

PARTICULARITIES OF NEW CALEDONIA AND THE KONIAMBO PROJECT

The archipelago of New Caledonia, an overseas possession of France, lies just north of the Tropic of Capricorn (Figure 1). Grande Terre, the main island which measures 400 by 50 km, possesses phenomenal mineral wealth, primarily nickel, making the island attractive to both local and multinational mining companies. New Caledonia has no legislation specifically requiring royalties or compensation to be paid to customary landowners, nor are such payments regularly demanded by local people (see Henningham 1992:74-76). This makes the situation there a sharp contrast to that of its neighbours, notably, Papua New Guinea which has abundant mineral resources and where 97% of land is owned by local people, and where compensation agreements form a crucial part of any mining negotiations.

New Caledonia has one of the lowest population densities in the South Pacific region, but this population is unevenly distributed with 60% of the population concentrated in greater Noumea where the population is only 22% Melanesian and 46% European. In contrast, the Northern Province's population is 79% Melanesian and 17% European (Ahmed-Michaux and Roos 1997). The majority of the residents of the Northern Province live in villages (ITSEE 1998), and nearly all these villagers are Kanak, New Caledonian Melanesians.

In Kanak societies, as Alban Bensa explains, '[I]and is not sacred as such but it is [...] charged with a social history where groups establish their identity' (1995:78). In other words, Kanak formulate their identities through their relationships to the landscape and its elements, both visible and invisible. In New Caledonia, as throughout Melanesia and Australia and indeed much of the South Pacific, landscape features are often believed to have been created by, and thus to mark the passage of, humans or spiritual beings. For the Kanak, a clan's name is often the name of the place customarily recognized as the origin of its founding ancestor (Leenhardt 1937; Bensa and Rivierre 1982). Since they demonstrate the itineraries of clans' ancestors, these sites and routes act as 'genealogies written in space' or 'living archives', as they were termed by the late independence leader, Jean-Marie Tjibaou (1996:69, 110-111).

The first clans to occupy an area, known as 'masters of the land', have a particularly high status; upon arrival, they received from the resident spirits (2) knowledge concerning hunting, fishing and gathering techniques as well as the first tuber to plant and the authorisation to settle in that spot and to welcome other clans (Bensa 1995:33; see also Bensa 1990 for a detailed description of different language groups' beliefs concerning these spirits). Furthermore, these clans' ancestral spirits also remain in the surroundings and thus clan members act as intermediaries between the living and the spirit world (Bensa 1995:58-60; see also Lambert 1980 [1900]).

In New Caledonia, conflicts that center around social status and land rights, and in which people draw upon multiple endogenous and/or exogenous sources of assistance, have occurred since the precolonial era. However, current tensions are exacerbated by the integration into the market economy of villagers and their natural resources, and by the Kanak people's attempts to recover their dignity and identity after 150 years of colonization.

Archaeological evidence indicates that the first humans arrived on New Caledonia's shores about 3,000 years ago (Sand 1995, 1996). They were first described by Europeans -- Captain Cook and his crew -- on 5 September 1774. From the first decades of the 19th century, European adventurers visited New Caledonia, with the first representatives of the London Missionary Society, as well as a few sandalwood traders, arriving in 1842 (Shineberg 1967). Kanak societies are known for their interest in new options and possibilities, and relationships with traders, missionaries and colonial officials soon became important sociopolitical stakes (Guiart 1962; Shineberg 1967; Douglas 1982, 1998; Naepels 1998; Bensa and Leblic 2000).

France took possession of the archipelago in 1853, and over the following century the Kanak saw their most fertile lands seized while they themselves were forcibly placed on reserves. Land claims, based on customary rights, began to emerge seriously around 1970, forcing public authorities to formulate land reform policies in order to address these demands. The first land reform plan was created in 1978. Since 1989, land is most often attributed to a Groupement de droit particulier local (GDPL), a corporate entity that allows clans to reclaim customary areas without having to use the property for commercial purposes. This group may consist of one or several clans or even all the clans of a village (ADRAF 2000:16).

In the 1970s, encouraged by the acquisition of independence by neighboring countries, Kanak began to form militant nationalist groups (Henningham 1992:66). The 1980s saw a series of violent uprisings, known as les Evenements (the Events), which ended in June 1988 with the Matignon Accords. This text promised restitution of customary lands, promotion of Kanak culture, preparation of a future New Caledonian elite, and rural development activities, and also made provisions for another referendum in 1998 (Freyss 1996:265). Ten years later, the promised referendum took place and gave an overwhelming (71.87%) approval to the Noumea Accord drawn up in April 1998. This accord also promised greater recognition of Kanak land rights and economic integration for Kanak.

Based largely on transfers from Metropolitan France, New Caledonia's GDP of approximately US$10,000 per person places it among the world's 20 wealthiest nations, despite its under-developed productive sector (Freyss 1996:261). However, this wealth has traditionally been concentrated in the capital, Noumea, and has thus been inaccessible to most Kanak. The political will to redress this imbalance is represented by the Koniambo Project, a joint venture nickel mining project involving Canada-based Falconbridge, the world's third largest producer of refined nickel, and SMSP, a Kanak-run mining company. If it goes ahead, this project will result in the exploitation of the Koniambo massif, located between the towns of Kone and Voh in the Northern Province (Figure 2), as well as the construction of a pyrometallurgic refinery (known to New Caledonians as the Northern Refinery) and a small dam.

STUDY SITES

Half a dozen villages surround the Koniambo Massif (Figure 3), which straddles the Hoot ma Whaap and Paici/Cemuki Customary Regions. My main study site, Oundjo (Figure 4), lies within the Hoot ma Whaap Region, mid-way between Kone and Voh. A village of approximately 300 residents (ITSEE and INSEE 1997), Oundjo is almost entirely Protestant. Since approximately 1990, some villagers have also formed a born-again Christian subgroup whose members meet regularly to pray, sing, and ask questions of God who responds by giving them visions. Other than cultivating yams and other garden vegetables, the villagers' main economic activities consist of harvesting marine products such as fish, crabs, trochus and sea cucumbers and selling them to itinerant merchants, as well as raising cattle at the nearby peninsula of Pinjen.

Oundjo is a focal point of many conflicts concerning the Koniambo Project. Pinjen (Figure 3), eight kilometres south-east of the village, had been reclaimed by Oundjo residents during the New Caledonia-wide, pro-independence 'Events' of the 1980s; at that time, the Land Bureau had purchased the land from a multinational enterprise and granted it to the villagers on the condition that they form a co-op to use the land for economic development purposes. In 1999, Falconbridge/SMSP identified this peninsula as the preferred site for construction of the Northern Refinery. The villagers subsequently divided themselves into two camps. The customary landowner clans pronounced themselves in favour of the refinery's construction on Pinjen while the cattle co-op operating on the peninsula was opposed. In March 2000, some of the co-op's members set up a road block to prevent the mining company from conducting feasibility studies for the refinery construction. In June 2001, approximately 20 members of customary landowner clans, led by 41-year-old Bernard Tchaounyane, moved to Pinjen which they began to occupy in order to ensure that the studies would proceed. Meanwhile, the cattle-raising co-op launched a lawsuit against Falconbridge/SMSP. Unable to resolve these conflicts, a few months later the mining company began research at another potential site, Vwavuto (Figure 3), three kilometres northwest of Oundjo, which it ultimately chose as the future site for the refinery in the event the Koniambo Project goes ahead.

In 1999, Falconbridge/SMSP also began to investigate potential sites for the dam that would provide a source of fresh water to cool the refinery's machinery. In early 2000, they determined that the most convenient site would be near Wapan (Figure 3), a bend in the Temala River ten kilometres north of the villages of Temala and Ouelisse. However, these villages' customary landowner clans refused the dam's construction, outlining several reasons. First, the river that would be dammed was an important source of fish and crayfish, and the lands that would be flooded were used as a hunting reserve for deer. Secondly, the area had been seized during the colonial period and had recently been successfully reclaimed by the villagers with the aim of using it for the village's economic development. Thirdly, as will be explained in greater detail below, the villagers needed to avoid the risk of severe punishment from their ancestors for disturbing their resting place. After much negotiation, the mining company finally ch ose another site for the dam, less convenient but far less controversial.

Meanwhile, in response to the advent of the Koniambo Project, the chieftainships from the southern end of the massif conferred with each other in order to grant their permission to Falconbridge/SMSP to conduct prospecting activities on the massif. Although there was some disagreement about who exactly had what rights, in the end they reached a consensus. In 1999, when Falconbridge/SMSP decided to commence prospecting activities on the massif, the mining company performed a highly publicised customary ceremony at the Voh stadium with the clans whom they had identified as customary landowners. Later that same year, the customary landowner clans from the southern end of the massif, concerned to protect the workers at the mine site, performed their own ceremony in which they apologized to their ancestors for the disturbance and explained to them what was happening.

METHODOLOGY

From January-June 2000, I worked as a consultant to Falconbridge and co-authored a 'Landscape Heritage' baseline study which included a map of the region's taboo sites. This paper is based on my own fieldwork, commenced during free moments while working for Falconbridge (as permitted by my contract) but primarily conducted during my subsequent return to the area in July-December 2001, when I was based at Oundjo with periodic trips to Ouelisse. Research methods consisted primarily of participant observation as well as semi-structured interviews with adults of both genders and of all ages, from each of the clans concerned.

POLITICO-ECONOMIC IMPORTANCE OF THE PROJECT

The Koniambo Project embodies a great many economic and political aspirations for people in the Voh-Kone' area. Differences in expectations of who exactly would receive the benefits to be generated created disputes within the community. In my understanding, what was ultimately at stake had less to do with financial gain than with social status. However, a pervasive discourse on the value of 'custom' led each side to accuse the other of greed for money or power at the expense of customary values.

'Greed, pride, selfishness'

At Oundjo, one 70-year-old man believed that the customary landowner clans of the Koniambo massif would receive payments from the mining company while other local residents, such as he, would suffer the economic consequences of the resultant pollution and destruction of their fisheries resources. A few months later, however, when the mining company began seriously to investigate the possibility of building the refinery on the peninsula of Vwavuto, to which this man had indirect customary rights, he joined with other customary landowners of the site in order to formulate demands for royalties and preference in employment. Such attitudes echo Colin Filer's observation, in Papua New Guinea, of

the tendency for all forms of compensation to be placed under the umbrella of 'custom' (a good thing) by those who actually claim them, but then to be repudiated as a form of 'politics' (a bad thing) when the claims are lodged by other people. (Filer 1997:179)

This ambivalence toward potential disparities in financial gains from the mining project must be examined in the context of Kanak people's relationships to money and its connection (or lack thereof) to social status, or 'symbolic capital' as termed by Pierre Bourdieu (1994:116, 161, 189). The Kanak became integrated into the market economy very rapidly at the beginning of the colonial period (Merle 1995; Naepels 1998). However, the economic sphere was, and remains, marked by discourse and practice emphasising equality (Bensa 1995:88), such that possessing greater wealth than one's neighbours is not a mark of high social standing. For example, it is not well regarded to consume conspicuously, as this evidences a lack of consideration for others with less spending power. (3) People especially disapprove of financial gain deemed to have been obtained at the expense of neighbours or relatives, who are expected to be given 'credit' at village stores (Faugere 1998:167) and a share, through gifts and 'borrowing', of any earnings. Meanwhile, relatively affluent individuals are constantly anxious that others will try, through sorcery, to prevent them from enjoying their wealth (Bensa 1995:89; Faugere 2000:42). (4)

Exchanges of goods and services, whether informally between relatives or friends or formally through customary ceremonies, still serve as frequent reminders of the interdependence of members of communities or kin groups. However, unlike many other areas of Melanesia, social status is not achieved by the accumulation and subsequent grandiose distribution of material goods (Bensa and Antheaume 1982). Instead, high social standing in Kanak societies is attained through proof of the ability to claim relative seniority and, especially, membership of a first-occupant clan. Historically, individuals constantly attempted to raise their families higher within the social hierarchy by manipulating myths and engineering marriages in order to associate themselves ever closer with the first-occupant clans, a position which, if less explicitly sought today, is still just as coveted. In the context of natural resource exploitation projects, the reception of royalties or priority in employment is high-profile proof of one's p osition as customary landowner. My findings thus differ, for example, from those of Strathern and Stewart (1998:213) who noted that a need for increased sums of cash was provoking internecine quarrels and accusations of jealousy among the Kawelka in Papua New Guinea. In the case of the Koniambo Project, I argue that while financial benefits in their own right undoubtedly interested people, their desire to receive visible, tangible forms of recognition from the mining company actually represented a wish for confirmation of a high social standing which was far more important, or at least could be enjoyed far more openly, than the money itself. Therefore, it is not surprising that villagers at Oundjo sought to affirm or improve their social status by attempting to convince the mining company that they had the rights to negotiate about, and receive financial benefits from, the resource exploitation activity. Meanwhile, they sought to belittle their opponents by accusing them of demonstrating excessive pride in fa lsely claiming to be customary landowner clans, thereby (according to these accusations) revealing a selfish, greedy desire for material wealth.

Who decides?

Decision-making authority is another resource that is a cause of conflict within the community and that, like the reception of mining money, is a sign of high social status. The decision about whether or not to accept the refinery's construction on local lands, and the determination of who has the right to make that choice, emerges at a time when people fear that their customary positions of authority are being challenged by various aspects of the villagers' integration into a market economy. For instance, the administrative position of Lesser Chief (petit chef, chief of a village, as opposed to grand chef, responsible for a district) is currently held by a member of the Tchaounyane family, whose members feel that their power base has been fundamentally shaken by the cattle-raising co-op. Similarly, people who hold a position of customary authority at Oundjo but who work in Noumea fear that others, back in the village, are taking advantage of their absence to invent land claims. In the context of a gradual er osion of their influence, it is not surprising that local customary authorities welcomed Falconbridge's recognition and implicit support of their positions, rights, and decision-making powers.

At the Voh stadium on 27 April 1998, representatives of Falconbridge and SMSP exchanged ritual packages of cloth, money and tobacco (a gesture known as 'doing custom', faire la coutume) with representatives of the clans claiming customary ownership of the Koniambo massif. After seeing their mountains exploited against their will for over a century, the massif's customary landowner clans had finally been consulted by a mining company. Nearly everyone in the region with whom I spoke accepted these clans' right to decide whether or not to allow the mining company to operate on the mountains, explaining that only certain clans had the prerogative to negotiate with Falconbridge, and that each of these clans had the right to speak only about the area pertaining to it.

However, people were more unclear as to what this 'ownership' entailed. Throughout Melanesia, although rights to and authority over land have historically been restricted to certain groups, such rules were -- and still are -- flexible, nuanced and dynamic (for example, see Hviding 1996). Additionally, use rights are often distinguished from permanent relationships to geographical areas (see Jacka 2001:7), although temporary users may try to claim more definitive privileges. Especially, customary systems of land rights were obviously not designed to address the complexities of engagements with large-scale development, as 'none of the traditional uses of land offers a satisfactory analogue to mineral rights' (Jorgensen 1997:620).

Even putting aside questions of what exactly a landowner was, people did not agree on who exactly local landowners were. As explained above, in many parts of the South Pacific, land rights are determined by ancestors' itineraries (Malinowski 1922; Bonnemaison 1987; Parmentier 1987; Kahn 1990; Toren 1995; Hviding 1996) as is the case for New Caledonia (Leenhardt 1937; Bensa and Rivierre 1982; Guiart 1994; Bensa 1995). However, there almost always are differing versions of the history of the passage of these ancestors through each region (Naepels 1998:105).

In contrast to Knauft's findings in regard to Papua New Guinea (1998:145) and many non-Aboriginal Australians' complaints (see Merlan 1998:211), I do not suggest that villagers in the Voh-Kone area were creating new myths in order to claim land to which they had no customary rights. Rather, I argue that the oral histories and genealogies relative to local land rights are so multifaceted and ambiguous that individuals were able, as they have always done, to draw upon certain kin relations, ancestors' itineraries and remembered exchanges in order to assert their own perceived rights and to deny those of others. (5) In other words, conflicting claims did not necessarily involve dishonesty but rather divergent constructions of complex, overlapping legitimacies.

However, many of the villagers with whom I spoke expressed the opinion that their opponents were being audacious if not downright dishonest, corrupted by a desire for dominance. Several lamented the fact that there were no knowledgeable elders still alive to help them determine the correct social structure, and that each clan was therefore obliged to interpret the situation, leading to disagreement.

At the same time, both Pinjen and Vwavuto officially belonged to entities that fell under the jurisdiction of French, not customary, law: respectively, the cattle-raising co-op and private individuals. Despite the official legal status of these lands, however, the customary system of land rights was still recognized by the villagers. 'Competing cultural [and legal] constructions of the landscape' (Moore 1998a:377) here overlapped in complex ways. In New Caledonia, the co-existence of such contrasting discourses often leads to intra-community conflict, as individuals with various social positions experience different benefits and costs from each system (Teulieres-Preston 2000). Anxious to maintain if not improve their current economic and social status, people sought to assert their rights through whichever system of rights, Kanak or French, was most supportive of their interests of the moment. This does not indicate that Oundjo villagers were particularly wily negotiators, but rather that they did not hesitat e to use the intellectual tools that best served their cause.

Sometimes, co-op members based their arguments on customary law. One 38-year-old member of the Diela clan, a customary landowner clan at Pinjen that is also represented within the co-op, asserted that his clan had customary title to the largest portion of the peninsula. This could be proven by the the fact that his family members had the same names as particular landscape features (pers. comm. Andre Pwahmo Diela, July 2001, Oundjo). However, members of the co-op relied primarily on the argument that they had legal title to the peninsula under French law, as proven by the fact that they were the ones who paid the land tax every year. Members of the co-op filed suit against Falconbridge, claiming that the mining company had obtained no legal authorisation to drill, build roads, or do other work on their lands.

The clans in favour of the refinery's construction on Pinjen also made use of both discourses to justify their position. At times, they looked to French law as a means of attaining their goals. When the mining company showed signs of giving up on Pinjen, the Tchaounyane and their allies wrote a letter to the Agency for Rural and Land Development (ADRAF) to re-reclaim their customary lands. According to Dipiba Fouange, a 55-year-old man whose primary land rights were at Vwavuto but who also claimed a small area at Pinjen, the pro-construction clans had contacted SMSP's CEO. The latter reportedly promised that although the refinery would not be built at Pinjen, SMSP would reward them for their prior support by helping them with any other projects, such as a housing estate, that they might choose to initiate on the peninsula (pers. comm. November 2001, Noumea). Thus, the pro-construction clans made use of the French legal system and external capitalist institutions to support their claims and efforts. At the sam e time, they relied on customary law to insist that the customary authorities were the legitimate landowners.

ECOLOGICAL CONSEQUENCES

Like social status and decision-making authority, livelihoods are likely to change radically with the advent of the Koniambo Project. Currently, most local people's economic activities consist mainly of the exploitation of marine resources and land for gardening and raising livestock; understandably, they are concerned that mining activities will destroy these sources of subsistence and income. In my discussions with them, many of the people in the Voh-Kone area evidenced a genuine concern for what was to them the most important aspect of their environment: its role as both a source of natural resources and a seat of cultural heritage. However, most of the local people with whom I spoke described the mining project as involving a trade-off; on one hand the environment would undergo a certain degree of damage, and on the other hand employment and (they believed) royalties would be generated. (6) Whether or not the villagers deemed the trade-off worthwhile depended on, first, the benefits they expected from the project, in terms of money, social status, and control over land; and, secondly, the degree to which they trusted Falconbridge/SMSP. These expectations and this trust depended in turn upon the relationship they had with the mining company and whether or not they believed that this relationship, and the project, would increase their control over the land.

Concerns about pollution

When given the opportunity in casual conversations or more formal meetings, local people often expressed, to me as well as to Falconbridge representatives, the conviction that the prospecting work on the mountain had already begun to pollute local rivers, which would eventually result in siltation of the mangrove and reef ecosystems. In private discussions with me, Oundjo residents also lamented the fact that the area's marine resources were being exhausted because of an increase in the numbers of people fishing there, and feared that with the arrival of workers for the refinery project, this depletion would accelerate. One woman, who often went fishing for crabs and other shellfish along the coast of Pinjen, noted the irony of the fact that the villagers had long fiercely defended their fishing zone from outsiders but now had given the mining company leeway to destroy the area (Helene Phoea Goa, pers. comm. July 2001, Pinjen).

Those who supported the refinery's construction on Pinjen, however, were less concerned. While cautious, they believed that some ecological sacrifices would be necessary for the community's greater good, and that one must adapt to one's times. They were convinced that the mining project would provide benefits that would more than compensate for any decline in the fisheries resources. Bernard Tchaounyane, the leader of the pro-refinery group at Pinjen, admitted that mining projects always entailed negative ecological aspects; he claimed that his main worry was the environment and that it would be necessary to be very careful. Nonetheless, he insisted that 'we have to go that way' (pers. comm. July 2001, Pinjen). Meanwhile, supporters of the refinery's local construction expressed less anxiety about ecological consequences than did their opponents. Bernard was convinced that if the refinery were constructed at Pinjen, the situation would not be nearly as dire as the co-op's flimsy arguments would have suggested . Many supporters of the refinery's construction on local lands mentioned that the area had already been polluted by other economic activities practised by their fellow villagers, especially cattle fanning. Moreover, they believed that natural areas could easily regenerate.

Lesser Chief Francois Tchaounyane believed that people who cried 'pollution' were actually out to sabotage the mining project (pers. comm. October 2001, Pinjen). His brother's partner, a 32-year-old woman named Maguy Leack, did not feel that mining activities presented a genuine threat to local marine resources. She believed that people's apparent interest in the environment was spurious since it surfaced only when money was involved, and thus accused her opponents of being selfish and greedy.

We Kanak, we are too attracted by money. As soon as there is an economic benefit, when it's a question of money, we'll search for a little... We'll talk about the environment, we'll talk about the fish. (pers. comm. August 2001, Oundjo)

Unlike Maguy, I believe that some people were indeed nervous about losing their fisheries resources. However, because of their status as customary landowners and their resultant privileged relationship to the mining company, Pinjen's self-proclaimed customary landowner clans expected jobs and royalties that would more than compensate for the loss of natural resources. In contrast, members of the other clans feared that since they did not have a close relationship to the mining company or a position of decision-making authority in relation to the use of the peninsula, they would lose their livelihoods and receive nothing in return. Thus, the conflict largely centred on the socio-cultural significance of the peninsula itself: the dignity, social status, and control over land which the cattle co-op would lose, and the customary landowner clans would gain, if the refinery were built on Pinjen. Their expectations of the benefits and costs of the construction of the refinery influenced their discourses about and, i n my view -- contra Maguy -- their genuine perceptions of the project's environmental impacts.

Like Maguy, Falconbridge/SMSP representatives privately vented cynicism about local people's abilities to assess environmental risk and the real reasons behind residents' expressions of concern. However, the mining company's public response to villagers' anxieties was to try to reassure local people. At first, their official statements reflected a putative desire for an absence of any negative environmental impacts whatsoever from the project. By 2000, however, Falconbridge's representatives had nuanced their discourse, repeatedly insisting that in any instance of industrial development, 'zero risk is impossible.' Nonetheless, in November 2000, Falconbridge's Director of Environment assured the Mining Committee, composed of administrative officials, elected representatives, and local residents, that the company's goal was to design their antipollution dams so as to approach '100% efficiency'. In a similar vein, that same year the official answer to local residents' complaints that the rivers had become muddie r since the start of the prospecting activities was to insist that the real cause of this pollution was the record levels of rain witnessed in 1998 and 1999.

Whether or not villagers chose to believe such claims and promises depended on their expectations of other benefits and costs from the mining activity and the degree to which they felt that the project would increase, or decrease, their economic autonomy and control over land. Bernard Tchaounyane stated that the mining company's work on the massif compared very favourably to other, more polluting, mining activity that he had observed in the past. To him, this change vindicated Kanak people's desire to see mining projects on their lands; their attitudes had thus changed radically from the past when Kanak had been excluded from the mining sector:

Now, with the regulations that are in place, that has changed lots of things. So maybe that's why it's more the Kanak who are demanding it. We Kanak, we had talked about resource access; now it's the case, [...] so for us it's something positive. (pers. comm. July 2001, Pinjen)

This change in attitude may indeed have resulted, at least in part, from observations of improvements in environmental standards followed by mining companies. It is likely, however, that Bernard's last sentence reveals a more fundamental reason behind his interest in this project: the fact that, unlike past mining activity which had removed resources from their land against their will and generated wealth in which the Kanak did not share, the Koniambo Project's major goal was to benefit the people of the Northern Province, giving them 'resource access'. In other words, his sense that the project would increase his, and other Kanak people's, control over these resources made him inclined to place his trust in Falconbridge.

Bernard and other members of customary landowner clans who had received demonstrations of respect from Falconbridge through customary exchange ceremonies also tended to feel that, in being given a voice, they had been granted a degree of power over the mining activity. They were therefore more inclined to feel that the mining company would respect their decisions and concerns. Others, however, expressed a sense of powerlessness. They wished to be able to monitor and control the mining activities, but despaired of being granted such authority. For them, the fact that Falconbridge was a multinational company was a further reason not to trust it. Convinced that despite the company's promises, they would be no different from other, nearby projects which had caused substantial environmental damage, these people pointed out that Falconbridge had no long-term commitment to the area. Unlike the supporters of the refinery's construction at Pinjen, these villagers did not occupy a social position that caused them to be recognised by the mining company as figures of authority. Instead, they stood to lose control over their land and resources to a large, powerful, multinational mining company from whom they felt alienated and whom, therefore, they chose not to trust.

Cultural heritage

Customary relationships to the land constitute one of the defining characteristics of a contemporary pan-Kanak identity that is being formulated (see Horowitz 2001:244-247), partly as a rejection of Western cultural and economic values (Roux 1974:41), and in ways similar to trends found elsewhere around the world (see, for example, Conklin and Graham 1995; Morris-Suzuki 1998; Warren 1998). Throughout New Caledonia, many of the Kanak to whom I spoke lamented the fact that, in their view, much of their unique knowledge has been lost through colonisation and missionary activity and continues to disappear as younger generations turn their attention to Western forms of entertainment such as television. As they saw it, many details concerning clan histories and taboo places have been forgotten over the past few generations. Nonetheless, their notion of a special relationship to the land -- to both domesticated and uncultivated, uninhabited spaces -- gives them a sense of personal and Kanak identity.

Many of the nationalist identities that are being forged throughout Oceania are based on similar statements about the importance of the customary relationship to land (see, for example, Rumsey 2001:38). However, such claims in New Caledonia take on a particular valency in the context of the desire for sovereignty. As in Australia as described by Merlan, the discourse of connection to land is both 'cultural and profoundly political' (1998:164). The growing need to reaffirm relationships to land and to establish a sense of rootedness in the soil partially constitutes a response to the land spoliation of the colonial period and the desire of a dispossessed people to reassert their rights (Naepels 1998:237). This exemplifies what Moore, using 'a Gramscian metaphor for environmental resources' (1996:126) -- an emphasis on the interrelationships among power, history, culture, and ecology -- terms 'the simultaneity of symbolic and material struggles over territory' (1998b:347).

As explained above, certain sites, markers of clan history, serve important psychological and strategical purposes as both mnemonics and proof of clans' relationships to the land. The significance of such places was brought into relief by the advent of the Koniambo Project. The cultural meanings of particular natural features, key sources of identity, were threatened by the changes to the landscape that the mining project would inevitably entail. However, those who supported the construction of the refinery or dam on local lands were confident that the mining company would take care to avoid the important sites, which were small in area anyhow.

On the other hand, people who did not expect the Koniambo Project to increase their control over land expressed more concern about their loss of these cultural resources. At Ouelisse, the 50-year-old head of the Economic Development Committee of the Council of Elders recounted that he had initially been in favour of the dam's construction at Wapan, in view of the benefits that it would provide for the village in terms of increased water supplies and possible employment for his fellow villagers. However, he later realized that this reaction had been superficial, and that there were 'deeper' reasons to refuse the dam, based on Kanak people's desires to maintain their cultural heritage, of which they had lost much due to colonisation and acculturation. He explained that the late Lesser Chief, Mepone Ouedoy, had 'culturally defended' the area, which was the home of their ancestors and belonged to the villagers by birthright (pers. comm. Georges Bouigou Goa-Bealo, November 2001, Ouelisse).

Similarly, Oundjo villagers who opposed the refinery's construction on Pinjen sometimes spoke to me of the importance of the peninsula's cultural heritage value. Some evoked the importance to the Kanak of their lands of origin, the only place where they were sure of finding a place to live and garden, and where they could truly feel at home. Others spoke of the negative psychological effects they expected from seeing their ancestral residences bulldozed if the refinery were built there. Many recollected their efforts to reclaim the lands in the 1980s. A 45-year-old woman recalled the emotional impact of a moment during this period when, in digging the soil to build a hut on the land, people had discovered human bones. 'That's when I knew that our ancestors were there. [...] I truly saw that it belongs to our ancestors' (pers. comm. Henriette Cae Diela, July 2001, Oundjo).

For others, memories of the land claim struggle imbued the peninsula with a particular political significance. One 38-year-old man, actively involved with the cattle co-op, noted the irony of the fact that if the villagers rented the peninsula to Falconbridge/SMSP they would no longer be at home on the land they had striven so hard to reclaim: 'We will have done just the opposite of what we fought to obtain.' He pointed out that without the physical evidence of the sites of former habitations, others could deny the validity of Kanak land claims: 'I often hear people say that "the Kanak claim all that, but it doesn't show that they are the ones who did all that'" (pers. comm. Andre Pwahmo Diela, August 2001, Oundjo). (7)

Several people feared that a loss of the particular features of their clan's lands, which commemorated their history, would entail a sense of disorientation. One man worried that 'if we accept any old thing, I am no longer me. I won't have my reference points any more. [...] Where will I have my identity?' He also spoke of the strength that he acquired from visiting sites that commemorated his clan's history. For him, this connection was a defining characteristic of the pan-Kanak culture: 'Our own identity, that cultural identity, it's what we have with nature' (pers. comm. August 2001, Oundjo). Decrying the upheaval caused by the inevitable presence of large-scale industrial projects, which were radically altering both nature and local lifestyles, he insisted that people needed to maintain important sites as reference points in order to be able to assimilate all the new elements in their surroundings.

In contrast, although some of the supporters of the refinery's construction on Pinjen agreed that it would be painful to see the bulldozers destroy the landscape and its features such as the resident animals and the marks left by their ancestors, such as yam terraces, they insisted that such sacrifices were counterbalanced by the economic gain that the project represented. One 32-year-old woman declared,

We're going to lose somehow a little bit of the heritage of our elders before, but it will be for our children later on. If we have to make sacrifices for our children, I think we have to go for it. (pers. comm. Maguy Troyko Leack, August 2001, Oundjo)

The secretary-general of the Voh Town Hall, a 30-year-old man from Ouelisse, expressed a similar sentiment but qualified people's attachment to certain places as stemming from ignorance: 'I think that people haven't quite grasped economic development' (pers. comm. Joel Boatate, June 2001, Voh).

SPIRITS AND TABOO PLACES

In addition to their cultural importance as evidence of clan histories and identities, taboo places are respected for the dangers they represent. Disrespect of a taboo can result in illness, accidents, or natural disasters which often affect not the individual concerned but another family member or even his entire group of relatives (see Salomon 2000:98-102). Mountain tops are also considered to be the domain of ancestral spirits. In the Voh-Kone area, many of the Kanak whom I interviewed, both young and old, unequivocally demonstrated anxiety regarding taboo places and associated ancestors. Yet again, however, whether or not people expressed concern about the potential impact of the mining project on taboo sites and spiritual beings depended on their beliefs as to whether the mining project would increase or decrease their control over the land.

Dangerous?

Several people explained that taboo sites were gradually becoming less dangerous, due to an increase in the numbers of automobiles, airplanes, and other noisy machinery that disturbed the ancestors' previously peaceful surroundings, causing them to seek refuge in places ever farther from human habitation. People also often mentioned the loss of forest cover from bush fires and cattle ranching as another factor contributing to the spirits' departure. Christianity was another force that was gradually eliminating dangers from spirits based at taboo sites. For the time being, however, these areas still possessed a mysterious yet perceptible force. (8)

Even if such places were no longer to be feared per se, they still had to be respected out of consideration for the clans that owned them. Moreover, in many cases, respect for taboo places reflected less a fear of punishment from spirits than a desire to maintain a sense of cultural identity. One man from Temala in his mid-40s, who had become the head of a small enterprise, explained his interpretation of the meaning of taboo sites.

If we stop believing in all that there, well it's all over, eh? No use saying 'I'm a Kanak.' That's sort of the characteristic of the Kanak. It was believing in all that (pers. comm. Auguste Fouagne, June 2000, Noumea)

In their conversations with me, people rarely mentioned taboo sites as a reason to oppose the entire Koniambo Project. Rather, the existence of such places most often constituted a reason to be very careful to avoid certain precise areas, small in size, where spirits resided or where ancestors had forbidden entry. It was also a factor evoked retrospectively to explain mysterious misfortune.

At the same time, knowledge concerning these sites became a political stake during Falcon-bridge's baseline studies. Part of my research for the 'Landscape Heritage' study included locating, photographing, and mapping as many taboo sites as possible. Although the precise location of such places is often kept secret by a select few, and I was later told that I had been the first person of European origin to visit the sites, I met with almost no resistance from those whom I approached. On the contrary, they expressed approval of Falconbridge's intentions to avoid the dangers involved in destroying such places. However, the late CEO of SMSP, himself a Kanak, believed that another reason was involved in people's willingness to reveal their taboo sites. According to him, self-proclaimed customary authorities hoped to manoeuvre themselves into a position of 'privileged representatives' in order to obtain a 'parcel of power' by presenting themselves as the possessors of these sites. Other community members' recognit ion of the sites as taboo would allow the knowledgeable individuals to claim the position of 'landowners' and thus to rise in status (pers. comm. Raphael Pidjot, March 2000, Noumea). Indeed, various people, some of whom had chosen at the time not to provide me with information, later told me that the map was 'all wrong'. They maintained that their rivals had invented some of the taboo sites indicated on the map and neglected to mention others. Thus, in the specific social and political contexts of various communities around the massif, the placements of taboo sites and their associated spirits were evoked or denied by different individuals to support their particular arguments.

Pinjen

For example, opponents and supporters of the refinery's construction on Pinjen spoke differently of the significance of taboo places. These sites were a source of worry for those who opposed the refinery's construction on the peninsula. However, the supporters of construction there, while agreeing that ancestral spirits were still powerful at certain places, were convinced that few such places existed on the peninsula and that the necessary steps had been taken to ensure that those sites would not be destroyed.

Christian faith did not exclude recognition of the power of ancestral spirits. Even born-again Christians evoked the power of spirits as well as that of God, in order to support their arguments for or against the refinery's construction. Roda Kahmene, a member of the group known as Children of God, was certain that the refinery would never be built on Pinjen. She and other members of her religious group had experienced visions that demonstrated that the customary landowner clans' spirits were creating conflicts amongst the clan members and thus preventing the project from going ahead. Seeing this, God had decided that the clans would first have to be reconciled before the project could proceed. In contrast, at Vwavuto, where her clan had customary land rights, the families concerned had quickly come to an agreement; this was a sign that God wanted the refinery to be built there (pers. comm. October 2001, Oundjo).

Vwavuto

Few of the Oundjo residents with whom I spoke were concerned about the dangers from spirits at Vwavuto. The adjacent area of Traa was known to be the residence of mwaxheny, mischievous dwarves who were the totem of the Gwa-Cidopwan. However, the president of the Council of Elders, who preferred to see the refinery constructed there, assured me that Vwavuto itself was free from these beings and that to his knowledge there were no taboo places there (pers. comm. Henri Cip Kahmene, November 2001, Oundjo). Dipiba Fouange, strongly in favour of the refinery's construction at Vwavuto, laughed when I asked whether such dangers were important (October 2001, Noumea). In contrast, another Oundjo elder, who claimed land rights at Vwavuto but had been excluded by rival villagers from negotiations and consequently opposed construction at this site, affirmed that it could be hazardous to build the refinery there because of the mwaxheny, and because knowledge of the location of taboo areas had not been passed on by the elde rs (pers. comm. September 2001, Oundjo). Thus, people's evocations of dangers from non-human creatures or spirits on the peninsula correlated with their opposition to the refinery's construction at the site, which in turn reflected concerns about losing their rights over the land.

Wapan

Although many people at Oundjo worried about possible punishments from spirits if the proper customary procedures were not followed, I rarely heard anyone mention this concern as a main reason to oppose either work on the massif or the refinery's construction. At Ouelisse, in contrast, the danger of disturbing the ancestors was one of the primary arguments used in refusing a dam at Wapan. The man who represented the villagers in discussions with Falconbridge/SMSP explained that the area contained ancestral burial grounds as well as taboo sites marked by trees or stones. Members of the landowner clans were the only people who would have been able to cure illnesses contracted from accidentally behaving disrespectfully in a taboo place, but they had lost the relevant knowledge. Moreover, the water that would flood the ancient cemeteries would unearth the ancestors' spirits, who would then head directly for their descendants to punish them for having disturbed their rest. My informant, as well as several members of the chiefly Ouedoy family, was certain that such an event would entail the disappearance of the landowner clans. Thus in September 2000, after a series of debates in which the members of other families were allowed to express their opinions, the Council of Elders accepted the customary landowner clans' decision and sent a letter of refusal to the mining company.

All the villagers felt obliged to respect the landowner clans' decision, but not all agreed that a categorical refusal was necessary. Emile Tein-Boanou, the president of Ouelisse's Council of Elders, stated that during the intravillage debates, he had argued that there were rituals that could be performed in order to displace the taboo sites. Moreover, he pointed out that the villagers were all Christians, and their religion taught that such beliefs and practices concerning ancestral spirits were 'the works of Satan.' In addition, modem science had proven that humans themselves were the ones who created strange phenomena, through their own beliefs. Compromises would be necessary, both on the part of the villagers and the mining company; however, he felt that the dam project had much to offer in terms of economic development, and that it would be 'more beneficial than those things [i.e. the preservation of taboo sites]' (pers. comm. November 2001, Ouelisse).

It is no coincidence that Emile's parents were from the nearby village of Ouengo, where he was raised, and where -- rather than at Ouelisse or Wapan -- he had land rights. As another community member maintained, the 'further' that families were from the Lesser Chief (that is, the more recent their arrival at the village), the greater their tendency to support the dam's construction (pers. comm. Georges Bouigou Goa-Bealo, November 2001, Ouelisse). Thus, as at Oundjo, people's actions and discourses relating to ancestral punishments linked to the mining activity depended directly on their expectations of the project's implications for their own control over the land. In contrast to the customary landowner clans, who would literally and permanently lose their lands if they were flooded, those with no land rights to relinquish at Wapan stood to gain from the increased water supplies and employment possibilities offered by the dam's construction.

The massif

Not only taboo places but also mountain tops were known to be frequented by spirits. Many people from the villages surrounding the massif reported that mysterious beings, often in the form of silent old men, had been seen by workers on the mountains. These beings resembled relatives who had passed away, and thus were assumed to be ancestors. Reactions to these apparitions were varied; people who trusted Falconbridge to provide employment opportunities and to cause as little ecological damage as possible -- those who also supported the refinery's construction on Pinjen -- were prone not to worry about the spirits' emergence. Others, more mistrustful of Falconbridge's promises, believed that the sightings of the spirits demonstrated that people had entered places they should have avoided. The ancestors were thought to express themselves in other ways as well. Two years in a row after the beginning of exploratory activities on the Koniambo massif, the Voh-Kone area experienced record levels of precipitation. One senior member of a customary landowner clan wondered whether the excessive rains were due to ancestors' annoyance (pers. comm. Kowi Poangone, February 2000, Tieta). Others were convinced that, indeed, the weather was a warning. The head of a customary landowner clan near Voh pointed Out that his clan's totems, mwaxheny (mischievous dwarves) and xhwaala (thunder), could control the weather. Angry that the mining company had not consulted the customary landowner clans at every step, they had been demonstrating their discontent (pers. comm. Felix Yeiwene Foawy, October 2001, Tieta).

These spirits were dangerous because, if angered or bewildered by unfamiliar activity, they could cause workers on the massif to become ill or have accidents. (9) As described by Taussig (1980) and Nash (1979) in their respective studies of Bolivian miners, workers involved in dangerous resource-exploitation often try to prevent disaster by communicating with spiritual beings. As the only people who could transmit information directly to their ancestors, customary landowners of the massif were able to claim a position as the only group that could ensure the workers' safety. These clans justified their prerogative to approve or reject the mining project by explaining that the resident ancestral spirits needed to be properly informed of the activities that would occur on the mountain tops. The mayor of Kone, a member of a clan from the Paici/Cemuki Customary Region that claimed rights to the massif, explained the situation. 'As long as it's the clan itself that gives permission, the spirits follow automatically . And that doesn't follow simply from the spoken word, it's with the ceremony that must be done' (pers. comm. Joseph Goromido, June 2001, Kone).

The ceremony to which he referred had occurred in 1999, when the customary landowner clans realized that the mining company's prospecting activities were disturbing the spirits and causing accidents. They needed to apologize to their ancestors and to explain what was happening on the massif, in order to ensure the workers' safety. The ceremony involved placing packages of cloth and tobacco in a hole in the ground, pronouncing speeches to the ancestors, and then planting columnary pines or kauri trees in the hole. This was performed at several sites.

However, no similar ceremony was performed by the customary landowners of the northern end of the massif: the Xuti, Xutapet and Gwa-Cidopwan clans. This was an important oversight because each customary area had its own set of ancestral spirits who needed to be addressed separately. As one of the participants in the ceremony explained, 'The customary ceremony that we did has no value for their place' (pers. comm. Raymond Hauli Diela, February 2000, Oundjo). Several members of customary landowner clans from the Kone region were concerned that this negligence might lead to misfortune. One member of a clan from the southern end of the massif suggested that the fact that the ceremony of addressing the ancestors had not yet been performed at Voh might explain the fact that, according to him, there had been accidents in that area as well as river siltation from soil erosion.

Jacob Couthy, the person from the Voh region to whom the demands for a similar ceremony had reportedly been addressed, agreed that the current mineral exploration was upsetting to the ancestors who expressed themselves through signs such as small footprints. However, he denied that anyone had approached him with a request to speak to the ancestors; nor, according to him, would it be necessary as a ceremony had already been performed with the mining company at the Voh stadium two years previously. Jacob was strongly in favour of the mining project, for whom he worked on a contractual basis and which would provide employment to the members of his association for local unemployed people. His lack of anxiety concerning the need to communicate with the ancestors may have been due to his trust in Falconbridge/SMSP and his desire for the mining project's economic benefits. In contrast, one elderly Oundjo resident who was concerned about the mining activity's potential impacts claimed that after a series of minor ind ustrial accidents, some workers had presented the customary landowner clans of the northern end of the massif with a ritual gesture in an attempt to appease the ancestral spirits. All three clans accepted the gesture, although my informant expressed his own reluctance to do so (pers. comm. September 2001, Oundjo). Concerned that the mining project would result in a loss of the area's marine resources, he resented the fact that his gesture would protect the miners while providing no benefits for his clan or the other members of his village.

In some ways, like the Papua New Guineans at Kurumbukare described by Zimmer-Tamakoshi, the villagers in the Voh-Kone area were '[c]aught between their desires for wealth and their fear of the ancestors' and thus strove 'to bring the ancestors into agreement with their mundane goals' (1997:650). However, the ceremonies that took place also made clear that the two objectives, economic development and respect of the ancestors, were perfectly compatible -- as long as the correct protocol was followed by both the appropriate local customary authorities and the mining company. This action was thus not only an attempt to prevent potential accidents but also a means for the local customary landowner clans to affirm their authority over the massif (see Jacka 2001 for a similar example from the Porgera mine in Papua New Guinea). The ceremony was carried out in private, in the presence of a restricted set of individuals, and without informing the mining company. Thus, it was performed as a statement to other community members, and quite probably also as a means of restoring or bolstering their own confidence through a reaffirmation of their influence with the spirits. One participant, the head of a customary landowner clan at Tieta, described what he had said to the ancestors. While reassuring them that Falconbridge had ritually asked permission to operate on the mountains, he also reminded the spirits to punish any future mining activity that might occur without authorisation from the customary landowner clans: 'I said that, "I'm going to do 'custom' to protect our young people and to let the others work who have already... but those who go above our heads, who will go without telling us, you will supervise all that"' (pers. comm. Felix Yeiwene Foawy, May 2000, Tieta).

CONCLUSIONS: (DE)CONSTRUCTING DISCOURSES

As Merlan (1991) points out in the context of Aboriginal people's engagements with mining development at Coronation Hill, Australia, it is misleading to divide people into mutually exclusive categories of 'traditionalists' and 'modernists'. In the contemporary context, there is a multiplicity of ideas in which to believe and with which to formulate an identity, and many Kanak interweave various elements of the belief systems they discover around them into their discourses on topics of concern to them. As far as I could determine, nearly all the villagers with whom I spoke worried about their natural resources and the potential anger of their ancestral spirits, at the same time that they accepted, or even embraced, what they saw as the inevitable advance of economic development.

In the Voh-Kone area, the degree to which people expressed concern about the potential dangers of industrial expansion, or interest in its benefits, depended on how much they believed they would gain from or be harmed by the mining project. Their concerns were not only or primarily directly economic but, rather, centred around 'recognition capital' (Bourdieu 1994:189), which in the Kanak case takes the form of authority over land. Clearly, different people in the Voh-Kone area had very different expectations from and concerns about the Koniambo Project, depending on their particular socio-political positions within the local community. These hopes and fears were translated into beliefs and statements about the potential impact of the project on local ecosystems and ancestral spirits. Those, such as the customary landowners at Pinjen, who expected that their relationship with the mining company would increase recognition of their social status, were little inclined to worry about marine resources and taboo pla ces. In contrast, members of the cattle-raising co-op at Pinjen as well as the customary landowners at Wapan anticipated a loss of control over land from the construction of the refinery or dam. As a result, they used arguments about ecological impacts or spiritual dangers in order to oppose these components of the project.

As described by Ernst, the Onabasulu of Papua New Guinea changed their discourses -- and even adapted certain myths -- in order to stake land claims and to reify social categories 'in the context of an emerging ecopolitics' (1999:91). Thus, in the Papua New Guinea highlands as well as in New Caledonia, '"cognized models" as deployed are not always merely mimetic but may constitute highly politicized discursive practices' (Ernst 1999:96, original emphasis). In other words, the ways in which people react to and even think about the situations they find themselves in -- such as their anticipations of the benefits and/or threats a mining project may pose -- reflect their socio-political objectives.

As Pierre Bourdieu explains, the quest for this 'symbolic capital' need not be viewed as cynical, 'conscious calculation of utility' (1994:158); rather, 'social agents have "strategies" that are only rarely based on a genuine strategic intention' (1994:156). Thus, people may internalise culturally appropriate goals and act accordingly, without explicitly formulating or even conceptualising their intentions. In the context of the Koniambo Project and the daily, immediate conflicts it created, as I have tried to demonstrate, Voh-Kone area villagers selected -- whether consciously or unconsciously -- from among the range of beliefs available to them, the arguments that would best support their need to ensure respect of their social positions and control of their own destinies.

NOTES

(1.) My sincere gratitude goes to all those at Oundjo, Ouelisse and the other villages of the Voh-Kone area who gave so generously of their time and warmth, especially my hosts Henri and Antoinette Kahmene. I would also like to thank Bronwen Douglas, Alban Bensa, Chris Ballard, Jimmy Weiner, Richard Baker, Jean Guiart, Glenn Banks, and Adrian Muckle for their stimulating comments on fieldwork results and theoretical analysis. Special thanks also to Clive Hilliker for his patient technical assistance and artistic advice. Finally, I am grateful to two anonymous reviewers and to Yuri Saalmann for their comments on an earlier version of this article. Any errors of fact or interpretation are, of course, exclusively my own responsibility.

(2.) These beings are given various labels by different authors, all of which inevitably contain inappropriate resonances. I choose to call them spirits, while Bensa (1990) refers to them as "genies" (genies). In Bwatoo and Haveke, the dialects spoken at Oundjo (my primary study site), they are called mwaxheny.

(3.) As an example of this, my host family, upon returning from a fund-raising fair late one evening, declared that they were very hungry because they had not eaten anything there. Although they could have afforded the food that was for sale, they had been in the company of relatives who could not; unlike people of European origin, they explained, they could not eat something in front of a friend or relative who did not have the ability to do the same.

(4.) Fundamentalist Christianity provides the only escape from the social constraints surrounding material wealth, first by providing an excuse not to participate in customary exchange ceremonies and secondly by relieving people of their fear of sorcery. At Oundjo, the largest houses and newest ears were possessed by members of the highly religious 'Children of God' group, who lived on the outskirts of the village and only rarely participated in customary ceremonies. Faugere (2000:59-60) describes the 'weapon against [the financial obligations of] custom' provided by conversion to Pentecostalism on Mare.

(5.) See Jorgensen (1997) for a similar example, from Papua New Guinea, of the invocation of myth to assert land rights and to deny those of others, in the context of a mining project.

(6.) In contrast to villagers' beliefs, Falconbridge maintained that it would not pay royalties, or give preference in employment, to customary landowner clans.

(7.) See Hviding and Bayliss-Smith (2000:242) for a discussion of similar concerns in Marovo, Solomon Islands.

(8.) See Toren (1995:171) for a discussion of similar beliefs in Fiji.

(9.) In Melanesia, most events are believed to be caused by the agency of individual actors, whether human or spiritual, and thus are rarely considered to be true accidents. See Strathern and Stewart (2000) for a discussion of this belief in the Highlands of New Guinea.

REFERENCES

ADRAF 2000. La Reforme fonciere en Nouvelle-Caledonie 1978-1998. Noumea: Agence de developpement rural et d'amenagement foncier and Grain de sable.

AHMED-MICHAUX, P. and W. ROOS. 1997. Images de la population de la Nouvelle-Caledonie, Principaux resultats du recensement 1996. Noumea: Institut territorial des statistiques et des etudes economiques.

BALLARD, C. 1997. It's the land, stupid! The moral economy of resource ownership in Papua New Guinea. In P. Larmour (ed.) The Governance of Common Property in the Pacific Region, pp.47-66. Canberra: RSPAS, ANU.

BANKS, G. 1997. Mountain of Desire: Mining Companies and Indigenous Community at the Porgera Gold Mine, Papua New Guinea. Unpublished Ph.D. thesis, Australian National University, Canberra.

BENSA, A. 1990. Des ancetres et des hommes: Introduction aux theories kanak de la nature, de l'action et de l'histoire. In R. Boulay (ed.) De jade et de nacre: patrimoine artistique kanak, pp.130-159. Paris: Editions de la Reunion des musees nationaux.

1995. Chroniques kanak: L'Ethnologie en marche. Paris: Peuples autochtones et developpement.

BENSA, A. and I. LEBLIC 2000. Introduction. In A. Bensa and I. Leblic (eds.) En pays kanak, Pp.1-6. Paris: Editions de la Maison des sciences de l'homme.

BENSA, A. and B. ANTHEAUME. 1982. Le terroir et l'organisation sociale en zone cemuhi (region de Touho, Nouvelle-Caledonie): Quelques questions d'anthropologie economique. Journal d'agriculture traditionnelle et de botanie appliquee 29(3-4):275-284.

BENSA, A. and J.-C. RIVIERRE. 1982. Les Chemins de l'alliance: L'Organisation saciale et ses representations en Nouvelle-Caledonie (Region de Touho-Aire linguistique cemuhi). Paris: SELAF.

BONNEMAISON, J. 1987. Tanna: Las Hommes lieux. Paris: ORSTOM.

BOURDIEU, P. 1994. Raisons pratiques. Sur la theorie de l'action. Paris: Seuil.

CONKLIN, B. A. and L. R. GRAHAM. 1995. The shifting middle ground: Amazonian Indians and eco-politics. American Anthropologist 97(4):695-710.

DOUGLAS, B. 1982. "Written on the ground": Spatial symbolism, cultural categories and historical process in New Caledonia. The Journal of the Polynesian Society 91(3):383-416.

1998. Across the Great Divide: Journeys in History and Anthropology. Amsterdam: Harwood Academic Publishers.

ERNST, T. M. 1999. Land, stories and resources: Discourse and entification in Onabasulu modernity. American Anthropologist 10l(l):88-97.

FAUGERE, E. 1998. L'Argent et la coutume, Mare (Nouvelle-Caledonie). Unpublished Ph.D. thesis, Ecole des hautes etudes en sciences sociales, Paris and Marseille.

2000. Transactions monetaires en pays kanak. Genses 41:41-62.

FILER, C. 1997. Compensation, rent and power in Papua New Guinea. In S. Toft (ed.) Compensation for Resource Development in Papua New Guinea, pp.156-189. Canberra: Law Reform Commission of Papua New Guinea; RSPAS, ANU; and NCDS, ANU.

FREYSS, J. 1996. Nouvelle-Caledonie: le <<reequilibrage>> et ses contraintes. In A. Saussol and J. Zitomersky (eds.) Colonies, territoires, societes: L'Enjeu francais, pp.253-272. Paris: L'Harmattan.

GOPE, P. 2001. Le Dernier Crepuscule. Noumea: Grain de sable.

GUIART, J. 1962. The millenarian aspect of conversion to Christianity in the South Pacific. In S. L. Thrupp (ed.) Millenial Dreams in Action: Essays in Comparative Study, pp.122-138. The Hague: Mouton & Co.

1994. Fonction du mythe et realite empirique en Oceanie. In C. Vielle, P. Swiggers and G. Jucquois (eds.) Comparatisme, mythologies, langages, pp.73-142. Louvain-la-Neuve: Peeters.

HENNINGHAM, S. 1992. France and the South Pacific: A Contemporary History. Sydney: Allen & Unwin.

HOROWITZ, L. S. 2001. Perceptions of nature and responses to environmental degradation in New Caledonia. Ethnology 40(3):237-250.

HVIDING, E. 1996. Guardians of Marovo Lagoon: Practice, Place, and Politics in Maritime Melanesia. Honolulu: University of Hawai'i Press.

HVIDING, E. and T. BAYLISS-SMITH. 2000. Islands of Rainforest: Agroforestry, Logging and Eco-tourism in Solomon Islands. Ashgate: Aldershot and Burlington.

ITSEE 1998. Panorama des tribus. Noumea: Institut territorial de la statistique et des etudes economiques.

ITSEE and INSEE 1997. Recensement de la population de la Nouvelle-Caledonie: Principaux tableaux 1996. Notes et documents no. 77. Paris: Institut territorial de la statistique et des etudes economiques, Noumea; Institut national de la statistique et des etudes economiques.

JACKA, J. 2001. Coca-Cola and kolo: Land, ancestors and development. Anthropology Today 17(4):3-8.

JORGENSEN, D. 1997. Who and what is a landowner? Mythology and marking the ground in a Papua New Guinea mining project. Anthropological Forum 7(4):599-627.

KAHN, M. 1990. Stone-faced ancestors: The spatial anchoring of myth in Wamira, Papua New Guinea. Ethnology 29:51-66.

1996. Your place and mine: Sharing emotional landscapes in Wamira, Papua New Guinea. In S. Feld and K. H. Basso (eds.) Senses of Place, pp.167-196. Santa Fe: School of American Research Press.

KNAUFT, B. M. 1998. How the world turns upside down: Changing geographies of power and spiritual influence among the Gebusi. In L. R. Goldman and C. Ballard (eds.) Fluid Ontologies: Myth, Ritual and Philosophy in the Highlands of Papua New Guinea, pp.143-161. Westport, Connecticut and London: Bergin & Garvey.

LAMBERT, P. 1980 (1900). Maurs et superstitions des Neo-Caledoniens. Noumea: Societe d'etudes historiques de la Nouvelle-Caledonie.

LEENHARDT, M. 1937. Gens de la Grande Terre. Paris: Gallimard.

LEWIS, G. 1995. Revealed by illness: Aspects of the Gnau people's world and their perceptions of it. In D. de Coppet and A. Iteanu (eds.) Cosmos and Society in Oceania, pp.165-188. Oxford and Washington, D.C.: Berg.

MALINOWSKI, B. 1922. Argonauts of the Western Pacific: An Account of Native Enterprise and Adventure in the Archipelagoes of Melanesia New Guinea. London: Routledge & Kegan Paul.

MERLAN, F. 1991. The limits of cultural constructionism: The case of Coronation Hill. Oceania 61:341-352.

1998. Caging the Rainbow: Places, Politics, and Aborigines in a North Australian Town. Honolulu: University of Hawai'i Press.

MERLE, I. 1995. Experiences coloniales: La Nouvelle-Caledonie (1835-1920). Paris: Belin.

MOORE, D. S. 1996. Marxism, culture and political ecology. In R. Peet and M. Watts (eds.) Liberation Ecologies: Environment, Development, Social Movements, pp.125-147. London and New York: Routledge.

1998a. Clear waters and muddied histories: Environmental history and the politics of community in Zimbabwe's Eastern Highlands. Journal of Southern African Studies 24(2):377-403.

1998b. Subaltern struggles and the politics of place: Remapping resistance in Zimbabwe's Eastern Highlands. Cultural Anthropology 13(3):344-381.

MORRIS-SUZUKI, T. 1998. Re-inventing Japan: Time, Space, Nation. Armonk, New York and London: M. E. Sharpe.

NAEPELS, M. 1998. Histoires de terres kanakes. Paris: Belin.

NASH, J. 1979. We Eat the Mines and the Mines Eat Us: Dependency and Exploitation in Bolivian Tin Mines. New York: Columbia University Press.

ORTNER, S. B. 1995. Resistance and the problem of ethnographic refusal. Comparative Studies in Society and History 37(1): 173-193.

PARMENTIER, R. J. 1987. The Sacred Remains: Myth, History, and Polity in Belau. Chicago and London: The University of Chicago Press.

ROUX, J. C. 1974. Positions melanesiennes sur les caracteres juridiques et coutumiers de la reserve autochtone. ORSTOM, Unpublished document.

RUMSEY, A. 2001. Tracks, traces, and links to land in Aboriginal Australia, New Guinea, and beyond. In A. Rumsey and J. F. Weiner (eds.) Emplaced Myth: Space, Narrative, and Knowledge in Aboriginal Australia and Papua New Guinea, pp.19-42. Honolulu: University of Hawai'i Press.

SALOMON, C. 2000. Savoirs et pouvoirs therapeutiques kanaks. Paris: PUF.

SAND, C. 1995. 'Le Temps d'avant' -- La Prehistoire de la Nouvelle-Caledonie. Paris: Editions l'Harmattan.

SHINEBERG, D. 1967. They Came for Sandalwood: A Study of the Sandalwood Trade in the South-West Pacific 1830-1865. London and New York: Cambridge University Press.

1996. Recent developments in the study of New Caledonia's prehistory. Archaeology in Oceania 31:45-71.

STRATHERN, A. J. and P. J. STEWART. 1998. Shifting places, contested spaces: Land and identity politics in the Pacific. The Australian Journal of Anthropology 9(2):209-224.

2000. Accident, agency, and liability in New Guinea Highlands compensation practices. Bijdragen tot de Toal-, Land- en Volkenkunde 156(2):275-295.

TAUSSIG, M. T. 1980. The Devil amid Commodity Fetishism in South America. Chapel Hill: The University of North Carolina Press.

TEULIERES-PRESTON, M.-H. 2000. Le droit de la mer kanak et ses transformations. In A. Bensa and I. Leblic (eds.) En pays kanak, pp.129-146. Paris: Editions de la Maison des sciences de l'homme.

TJIBAOU, J.-M. 1996. La Presence kanak. Paris: Editions Odile Jacob.

TOREN, C. 1995. Seeing the ancestral sites: Transformations in Fijian notions of the land. In E. Hirsch and M. O'Hanlon (eds.) The Anthropology of Landscape: Perspectives on Place and Space, pp.163-183. Oxford: Clarendon Press.

WARREN, C. 1998. Tanah Lot: The cultural and environmental politics of resort development in Bali. In P. Hirsch and C. Warren (eds.) The Politics of Environment in Sourheast Asia: Resources and Resistance, pp.229-261. London and New York: Routledge.

ZIMMER-TAMAKOSHI, L. 1997. When land has a price: Ancestral gerrymandering and the resolution of land conflicts at Kurumbukare. Anthropological Forum 7(4):649-666.