The dilemma of contact: voluntary isolation and the impacts of gas exploitation on health and rights in the Kugapakori Nahua Reserve, Peruvian Amazon

Many small groups of indigenous peoples in the Amazon basin avoid and resist direct encounters with outsiders. As far as we know, they do so because of appalling experiences in earlier encounters with national society. When contacted today, they are extremely vulnerable to introduced diseases and exploitation. In this paper we draw on our experience in the Kugapakori Nahua Reserve for isolated peoples in SE Peru to discuss some of the current debates about whether isolated peoples should be contacted and how best to respect their right to life, health, autonomy and territory. The remote headwater regions where isolated peoples sought refuge during the last century are increasingly sought after for resource extraction. In particular, the extraction of oil and gas is increasing throughout the Peruvian Amazon. In the second part of the paper we give some examples of how oil/gas companies and the energy sector in Peru have affected the well-being of the peoples in this reserve in the 21st century. If this trend is not reversed the impacts for isolated peoples will be irreparable.


Introduction
There are indigenous peoples living in voluntary isolation in most of the countries that make up the Amazon basin (Brackelaire 2006). These are indigenous groups that avoid 1 Author to whom any correspondence should be addressed. or limit their relations with individuals outside their close kin network, a decision based to a large degree on appalling losses caused by disease and persecution during the rubber boom (late 1800s-1915), events which are still very much alive in their collective memories. Many indigenous peoples in the Peruvian Amazonian today have heard first hand of slave raids and widespread epidemics that their parents or grandparents survived; in some areas slave raids continued into the 1960s (Swierk 2004). In order to avoid encounters with outsiders, some groups retreated into ever more remote areas of the Amazon basin, particularly the headwater regions and they have remained there. Their territories are now being circumscribed further by new extractive activities: from small scale extractors, such as illegal loggers and artisanal miners, to much larger scale enterprises, such as international petrol/gas and timber companies.
Recent studies have shown that globally, indigenous people today have poorer indices of health and well-being than most other population groups in the same countries (Stephens et al 2006). Isolated peoples are particularly vulnerable with extremely high rates of morbidity and mortality related to the introduction of new diseases (Napolitano 2007, OGE 2003, Hurtado et al 2001. Both anthropologists and missionaries have described recent contact situations as genocide (e.g. Shepard 1999, SLOPA 1980-1989. Virgin soil epidemics-epidemics of 'novel' diseases introduced by outsiders-have accounted for the deaths of millions of indigenous Americans over the last five hundred years (Dobyns 1993, Myers 1988). First 'face-to-face' contacts are estimated to lead to the death of between a third and half of the population within the first five years (Hill and Hurtado 1996), sometimes more.
Section 1 of this paper presents some of the issues surrounding contact in the Amazon rainforest, and debates about whether and why it should or should not be encouraged. In section 3.1 we describe some examples of how oil and gas exploitation in the Urubamba Valley (SE Peru) is jeopardising the survival and rights of isolated peoples. We draw on our own experience gained working in the Kugapakori Nahua Reserve to inform the debate. The information available about isolated peoples is, by definition, either from retrospective testimonies or second hand.

Location
The Territorial Reserve for the Kugapakori, Nahua, Nanti and others 2 was the first Territorial Reserve for isolated peoples in Peru, created in 1990 in response to the tragedy of Nahua contact (see Napolitano (2007) among others) to protect the territories and resource base of its indigenous inhabitants. It includes the settled and migratory territories of at least four different groups: the Nahua, the Nanti 3 , the so-called Mashco Piro 4 and some Machiguenga families who have decided to 2 The Reserve's official title is the State Territorial Reserve for the ethnic groups in voluntary isolation and initial contact Kugapakori, Nahua, Nanti and others. Ministerial Resolution No. 46-90AG/DGRAAR, Supreme Decree No. 028-2003-AG. 3 The Nanti are often known as 'Kugapakori', hence the name of the reserve. Kugapakori is a Machiguenga term with pejorative connotations, which the Machiguenga use to refer to dangerous, bellicose or unpredictable groups. It was an expression of their generalised fear of the 'uncivilised' people in the headwaters, although it seems largely unfounded since there is no record of Nanti ever attacking the Machiguenga. 4 The 'Mashco Piro' (or 'Mashco') are a very little known group. The term is used most commonly to refer to several families, thought to be entirely nomadic, living between the headwaters of the Purús, Yurua, Los Amigos, limit their relations with outsiders 5 (Shinai 2004). All four groups rely on frequent hunting, fishing and gathering activities for their subsistence; in addition, the Nahua, Nanti, and Machiguenga plant a range of crops of which manioc and bananas are the staple 6 . Figure 1 shows the Reserve, the territories of its peoples and the titled native communities on its borders. Four oil/gas concessions affect the Urubamba basin: at present Blocks 88 and 56 are in production.

To contact or not to contact
Many institutions and organisations across the world (including government, civil society, indigenous organisations and missionary bodies) have at various times argued in favour of contacting indigenous people who have not sought to establish relations with outsiders.
Missionary groups commonly have an overt integrationist strategy predicated on the desire to evangelize 'primitive tribes', who they believe live according to 'their ancestral customs, isolated from one another, without aspirations, without science, art or industry, without organised religion, at the mercy of their brutal instincts, without law or king to govern them' (Sarasola 1938in Monnier (1984, our italics). Although missionaries today might not write in quite the same tone, many still present their efforts to search for new converts as a heroic struggle in a hostile rainforest to save 'abandoned' peoples, even drawing an analogy between isolated indigenous groups and disabled children (Echevarría 2003 p 68). In spite of associated humanitarian efforts, missionary attempts at contact since the 16th century have been responsible for the decimation of many indigenous populations of the Americas (Dobyns 1993, Myers 1988. Some government bodies have also sought contact at various times: the State Department for Indigenous Affairs in Brazil, FUNAI 7 , had a policy from the late 60s to the early 90s to contact, 'pacify' and relocate many groups in the Brazilian Amazon. This led to the deaths of thousands of isolated peoples, some at the hands of the military in brutal repressions, Las Piedras, Manu and Mishagua rivers in Eastern Peru. Some Mashco groups have, in recent years, moved further and further West, towards Nahua territory (upper Mishagua) during dry season migrations. This is probably a direct result of massive illegal logging operations in their traditional lands and resulting violent encounters (Huertas Castillo 2002 p 94) although a large part of their territory is ostensibly protected, since April 2002, by a Territorial Reserve (Ministerial Resolution 0427-2002-AG). Nothing is known about their mortality rates due to infectious disease, although many are thought to have died in violent clashes with loggers and other indigenous groups who consider them dangerous and are very afraid of them (e.g. Feather 2005). 5 The Machiguenga are a much larger indigenous people than either the Nahua or the Nanti peoples with an estimated population of 10 000 (Rosengren 2004). The majority of Machiguenga live in titled communities along the Urubamba river, although there are small groups who maintain their distance from national society like those living along the Paquiría river inside the Kugapakori Nahua Reserve. 6 Very little is known about the Mashco Piro but no permanent settlement is known and no gardens have been found in the vicinity of their temporary camps. 7 Fundação Nacional doÍndio, the department for indigenous affairs of the Ministry of Justice in Brazil. FUNAI elected in the 1990s to adopt an approach that was more respectful of indigenous peoples' autonomy: it stopped organising forced contact and refused evangelical groups access to isolated peoples' territories 9 (an estimated 50 isolated groups had managed to maintain their independence in spite of the earlier policy, Huertas Castillo 2002). This approach is now broadly espoused by most indigenous rights organisations and support groups across Latin America (Declaración de Belem 2005). It avoids making judgements about the way isolated peoples live and privileges the premise that no one can decide on their behalf what is best for them; their position-expressed through action or inaction-not to seek contact with outsiders should be respected by protecting their land from potential threats to that decision (e.g. extraction of resources by third parties).
From the testimonies of some recently contacted individuals, it is clear that whether or not to seek contact with outsiders is the subject of frequent and intense debates for many peoples. They are at the crux of a dilemma, caught between fear of capture, disease and ill treatment (which they associate with outsiders), and a desire for metal tools and other manufactured goods (which they know outsiders possess) and social interaction with outsiders (see for example Townsley 1989, Montag 1998, Shepard 1999, Beier and Michael 2004. It has been said that an aversion to outsiders based on fear is not a voluntary choice (Hurtado et al 2001, Ossio et al 2004, but Zarzar suggests that it must be seen in context: isolated peoples make a choice 'neither isolated by the free exercise of their will, as some would have it, nor absolutely forced, since in the final analysis the process that leads to isolation is preceded by an option, an act of will' (Zarzar 2000).
In 1953 the Dominican missionary PadreÁlvarez Lobo (O.P.) went up the Mishagua River into Nahua territory. He 'saw no one, but found clear signs of the presence of indigenous peoples who wanted contact with outsiders, like bamboo on the beaches and along the banks of the river' (SLOPA May 1984).Álvarez Lobo's inference is a non sequitur since the evidence of habitation that he describes is hardly sufficient to be interpreted as evidence of peoples who want contact with outsiders. If the Nahua had wanted to establish a direct relationship withÁlvarez Lobo they had plenty of opportunities; they were certainly aware of his presence and apparently collected the pots and machetes he left out for them on the trips he made in the 1960s, during which they never revealed themselves to him (SLOPA 1980(SLOPA -1989. Even after one group of Nahua had made their first trip to Sepahua in 1984 (a visit more or less coerced, depending on your narrator) and received many gifts, most Nahua, by their own accounts, did not want to maintain contact. But a few did decide to go back to receive more 'gifts'. However questionable the basis of their gift giving (the population of Sepahua had every intention of 'buying' the Nahua's friendship in order to extract timber from their territory, according to Wahl 1990), at least some Nahua at this point made a conscious decision to seek sustained relations with Sepahua. In contrast, the Nanti living to the South on the upper river Camisea made no efforts to contact the downriver Machiguenga communities after their first encounters with a Machiguenga man in the late 1980s who offered them machetes in exchange for labour, they simply waited for him to return the following year. This is despite their desire for metal tools having been a motivating factor in their recent move from the head waters of the Timpía (Beier and Michael 2004). In this sense, the Nanti and Nahua approaches to contact have been different. Nanti relations with outsiders have increased in intensity due to outsiders' political or social impulses, tempered in varying measure by the Nanti themselves. In 2003 Migzero, peresetente 10 of Montetoni, was clear: People should not come here for no reason, they should come for good reason . . . The doctors should come when we are sick, but it is not good when they come if we are well. . . . Before (recent past) when there was illness, the health promoter went down to Malanksiari and gave the alarm (by radio). He explained what the illness was like (symptoms) and the doctors said that they would come and they came in a helicopter. We do not call just because, only when things are serious. The radio is not a toy (cited in Napolitano and Stephens 2003).
Fifteen years after establishing their first relationship with Machiguenga who lived downstream, the Nanti were still committed to holding the outside world at a distance.
These two examples illustrate the contrasting approaches of two Amazonian peoples after their first direct encounters with outsiders. They show that there is a purposeful decision behind the actions these people take, and whilst they may almost always be reacting to external initiatives or aggressions, they are reacting according to their own outlooks and on the basis of perceived positive and negative experiences. The action of avoiding contact, practiced by peoples still living in isolation, should therefore be seen as a conscious and careful decision.
Some anthropologists, like Hill and Hurtado, on the basis of their own extensive experience, have argued for 'controlled contact' as the only means of minimising health impacts. Hill and Hurtado favour forcing and managing contact, at least for some groups, due to their otherwise 'probable extinction in the immediate future' (Hurtado et al 2001). Their experience with the Paraguayan Ache 'suggests that almost all these deaths could have been avoided if antibiotics against pneumonia had been administered, and food provided to the sick population during the epidemics' (Hill and Hurtado 1996 p 167). Hill observed and participated in 'carefully executed' contacts with some Ache groups which resulted in very few deaths (Hill and Hurtado 1996 p 55).
Their argument is compelling, especially backed as it is by successful examples, but would it be replicable and is 10 Nanti pronunciation of the Spanish 'Presidente'. See Beier and Michael (2004) for more on the curious evolution of this novel authority figure in the Nanti communities of the upper Camisea. it ethically acceptable for us to decide who 'gets contacted' when it goes against the carefully considered decisions of these peoples? Hill and Hurtado describe a complex situation: for decades the Ache's territory had been increasingly reduced and access to resources restricted by the presence of settlers and the activities of 'enemy' Ache groups that had been contacted earlier. There was virtually nowhere left for them to go and settlers were terrorising them; in this context the forced controlled contact of the 'last' two groups (of 24 and 28 individuals, respectively), one by other Ache and the other by a Protestant missionary family, seems to have been a decision which the two groups accepted with relief (Hill and Hurtado 1996 p 55).
However, our experience in the Peruvian Amazon suggests that this is not always the case. While isolated peoples have sufficient territories, resources and health, and relatively large population numbers, they are far from 'disabled' 11 . As long as their autonomy is respected and access to their natural resources is secure they insist their quality of life is high and they 'live well' (Shinai 2004). Although susceptible to introduced diseases, it is only after contact with outsiders that isolated groups become vulnerable to their effects and they are reduced to less viable numbers (Napolitano 2007). The sudden and high intensity contact of the Nahua (Zarzar 1987, Wahl 1990, Dagget 1991 and the lower intensity, intermittent establishment of relations of the Camisea Nanti (Beier and Michael 2004) in the context of oil/gas and timber exploitation in the Urubamba basin both resulted in extremely high mortality, estimated between 40-60% (see Shinai 2004, Napolitano 2007.
There is thus good reason to believe that deliberate decisions and choices are behind isolated peoples' control of their relations with outsiders and it seems hard to justify a paternalistic approach of 'pre-emptive contact' that overrides their deliberate decisions, however flawed or limited an outsider may imagine their perception of threats to be.
Hurtado et al (2001) are concerned for the best interests of isolated peoples and argue that 'All Indians we have ever spoken to gladly accept improvements in their physical conditions and health situation if offered by true friends'. Our experience has also shown that people want improvements in physical and health conditions, but in practice the post contact years do not lead to such improvements for most peoples, with morbidity and mortality remaining extremely high, sometimes for decades (Strongin 1982, Montag 1998, Shepard 1999, OGE 2003. In addition 'true friends' are rarely the instigators of contact. In the Peruvian rainforest, indigenous people will most often be living side by side with settlers, loggers, miners and/or oil/gas workers: most of them poor and trying to make a living under difficult conditions, a very few extremely rich and ruthless. Nahua and Nanti experiences of contact were both managed by other indigenous intermediaries to significant degrees. These mediators facilitated access to their territories for loggers and benefited personally from the Nahua and Nanti's naiveté about the mechanisms and interests of the outside world (Shinai 2004, Beier andMichael 2004) 12 .
High mortality and morbidity are not the only consequence of direct relations with outsiders. A whole series of other changes accompany the process of establishing new relations. Both Nahua and Nanti have been exploited as cheap labour, for their natural resources and sexually by their nearest indigenous neighbours and mestizos 13 . Depression and disorientation followed the epidemics, and relations with outsiders and involvement on the margins of a market economy introduced them to the lowest rungs of Peruvian social stratification. Their experience of these changes affects aspects of their psychosocial experience, factors which are not usually included in a biomedical concept of 'health'.
Our experience suggests that it would be difficult to imagine how these changes could be 'controlled', even by the best-intentioned and best-equipped scientists, medics, anthropologists or missionaries. Control in this context could apply to two things: careful medical assistance (undoubtedly a duty for health services and a right of the newly 'contacted') and strategies to prevent exploitation. Good medical care can make an enormous difference to mortality, as Hill and Hurtado (1996) describe. That we have not observed a situation where medical efforts managed to substantially reduce mortality is in large measure the result of poorly coordinated responses and no shared government policy on how to respond. For example, two mission organisations responded to the Nahua epidemics (1984-86) from the Urubamba: the Dominican Mission, with a hospital in Sepahua, and the Summer Institute of Linguistics (SIL), which set up a field hospital in Putaya, upper Mishagua, however the two did not coordinate. The response in the Manu river basin, managed by the authorities of Manu National Park, also appears to have been slow and uncoordinated. Response to recent Nanti outbreaks tends to be very variable with medication, equipment, fuel and funds often lacking.
The logistics of responding to epidemics are also a challenge: providing intensive round the clock care for 28 individuals for several years is one thing; providing the same level of care for more than 300 (some estimates of Nahua population in 1984 are as high as 600) in dispersed settlements in the rainforest is much less feasible, especially given the usual availability of resources in most Latin American countries. Medical teams were exhausted attending to the Nahua population of up to 200 that gathered in the vicinity of the SIL field hospital in 1984-5 (Dagget 1991) and although these people came to receive treatment, many others probably continued to avoid outsiders in spite of their illness and died in the forest (Huertas Castillo 2002 p 102).
The second aspect to consider in controlling the impacts of contact is exploitation. Minimising the exploitation of 12 These indigenous intermediaries can play both positive and negative roles. In practice they usually establish kinship ties and some traditional exchange and support relations, but at the same time often take on exploitative roles. Their actions are often key determinants of the isolated group's contact experience. 13 Both Nahua and Nanti have accused indigenous and mestizo teachers of sexual assault and abuse of women and children (Shinai 2004). newly contacted peoples (which can come from powerful and unpredictable quarters) is difficult. In the Kugapakori Nahua Reserve, few efforts were made to prevent exploitation and the further opportunities for disease transmission that this contact offered. To mention just a couple of examples: following the Nahua's contact, the rich timber resources in their territory led to a free for all, as the Nahua had previously attacked all loggers (Wahl 1990). Following reports of the Nanti's presence in the Upper Camisea, the regional education authority opened a post for a Machiguenga teacher there, but instead of being a positive influence he became the principal instigator of commercial and sexual exploitation of the Nanti (see Beier andMichael 2004, Shinai 2004).
We have referred often to the work of Magdalena Hurtado and Kim Hill in this section. We find many points of agreement in their work, our differences stem from a difference at the level of action: their focus is on treating the sick; we would advocate an approach that also explicitly recognises isolated peoples' rights and decisions to remain in isolation, protects their territories and instigates a comprehensive emergency health response plan in case of contacts. Hill and Hurtado's position may seem appropriate in hindsight in certain contexts, like the Paraguayan rainforest, where abuses and widespread colonization are already a fait acompli; in many parts of the Peruvian rainforest this was not the case until recently. In Peru, it is only in the last few years that new extractive policies are actively promoting activities which will lead to tragic confrontations and probably epidemics of introduced diseases for isolated peoples. Not forcing contact implies, in addition to respecting isolated peoples' right to autonomy, respecting their right to health and life. By extension, promoting policies that are likely to lead to situations of forced contact is a violation of their right to health and life. By ensuring territorial security, governments can allow isolated peoples time and space to live in peace, and perhaps lose their fear of outsiders which may lead them to reconsider their position regarding contact 14 . The current policies of resource extraction only perpetuate isolated peoples' perception of a violent, destructive outside world, worthy of fear and unworthy of sustained relations.

Threats from oil and gas activities
The rest of this paper describes recent events related to the Kugapakori Nahua Reserve and the Peruvian Amazon. We illustrate the role that extractive industries, in this case oil/gas exploration and exploitation, have had on the health of isolated peoples today and how industry lobby groups attempt to delay and obstruct the development of polices to defend the rights and well-being of indigenous peoples living in voluntary isolation.
A number of extractive industries affect the territories of voluntarily isolated peoples, including agriculture (predominantly by Andean settlers), logging and small scale gold mining. However, of all of these, the oil/gas frontier has advanced the most in the last seven years. The total area available for oil/gas exploitation in the Peruvian Amazon has increased dramatically since 2000. In the 13 month period from May 2006 to 14 Eventual contact should not be considered as inevitable.
June 2007 the area under hydrocarbon concessions almost doubled to 81% of the land area of the Peruvian Amazon (MEM 2006a(MEM , 2007 15 .
Oil and gas exploration, exploitation and associated transport installations have been reported to have a variety of environmental, social and health impacts for indigenous communities across the world. Among the negative impacts are deforestation for the laying of seismic lines, roads and drilling installations, causing ecosystem destruction and spills leading to the loss of fish and game and contamination of vital water and land resources (Epstein andSelber 2002, O'Rourke andConnolly 2003). Reported social impacts for local communities include loss of territory and colonization (O'Rourke and Connolly 2003), alcohol and substance abuse, domestic violence and suicide (Wernham 2007). Although there are few epidemiological studies on the impact of the oil/gas industry on local communities, there is evidence of increases in infectious diseases such as HIV/AIDS and malaria (Jobin 2003), as well as pollutant-related increases in rates of dermatologic and pulmonary conditions (Epstein and Selber 2002), cancer, spontaneous abortion and other health indicators (San Sebastián et al 2001a, 2001bSan Sebastián 2002, 2004;Ahtuangaruak 2003in Wernham 2007. Shell's exploratory activities in the early 80s in the Mishagua watershed contributed to inter-ethnic conflict between 'isolated' Nahua and settled Machiguenga in Manu National Park (Rummenholler 1997in Huertas Castillo 2002, Clousdey 1988) and increasing logging pressure in the area, which set the scene for the Nahua's first direct contact and the ensuing epidemics. Shell's explorations throughout the Urubamba basin led to the discovery of the Camisea gas fields, with estimated reserves of 8.7 trillion cubic feet of gas and 411 million barrels of associated gas liquids (Camisea Project website 2007) 16 . The concession to exploit the fields (known as Block 88) is operated by Pluspetrol 17 and 75% of it overlaps the territory of the Kugapakori Nahua Reserve, with three proposed platforms inside the reserve 18 (see figure 1). The project has been the subject of some controversy since 2002 for social and environmental violations along the pipeline and in titled indigenous communities, as well as for impacts on the peoples of the Kugapakori Nahua Reserve (e.g. Caffrey 2002, Goodland 2003, Shinai 2004. Six spills have taken place along the pipeline route to date, four in the rainforest, and there have been fifteen leaks (Giraldo Gómez 2007).

Forced contact
In 2002, as part of their seismic exploration, a Veritas (subcontracting company) work group spent some time 15 Percentage calculated using the estimated area of the Peruvian Amazon to be 65 100 000 hectares (Samanez-Mercado 1990 working near the homes of three Machiguenga families who avoid encounters with outsiders. They were members of a group of loosely connected families living along the upper Paquiría. Seismic work involves clearing tracks in the forest, clearing many heliports and campsites for the workers and digging holes for detonation; it is usually labour intensive. A few weeks later these Machiguenga families abandoned their homes and went to live on the Paquiría River near other relatives (Swierk 2002). They told an anthropologist working on the Paquiría at the time that Machiguenga guides with the seismic teams had told them they should leave or risk dying or being taken away (Swierk 2002).
When Pluspetrol was accused of causing forced relocation (Shinai Serjali 2002), the company alleged that these families were not in 'absolute isolation' because they knew some members of the local communities who had worked timber in that area (Pluspetrol 2002). This kind of manipulation of the concept of 'isolation', facilitated by the confusion of terms applied to these peoples (see Huertas Castillo 2002, Shinai 2004, 2007a, enabled Pluspetrol to evade its critics. The fact that these families had had occasional encounters with Machiguenga loggers did not reduce their susceptibility to disease or exploitation at the hands of outsiders (all commercial logging in this Reserve is illegal and there are reports that the families had 'not been treated well' by the loggers, Swierk 2004). Playing on the implied difference between 'absolute isolation' and some alternative kind of isolation ('partial'?) is to make a false distinction. Almost all indigenous peoples living in voluntary isolation today have been 'contacted' at some point and then made or maintained a decision to keep outsiders at a distance; companies, missionaries and others should respect the choice they express in the present. In addition, the vulnerability resulting from indigenous peoples' susceptibility and the introduction by newcomers of new diseases does not change quickly: the fact that a Machiguenga family has had a few encounters with a group of loggers does not reduce the potential impact of introduced diseases or exploitation by outsiders.
This case also touches on the issue of prior consultation: Pluspetrol claimed that it was talking to all the families around their proposed well platforms 'so they would have information about company activities', with the implication that they were addressing one of the reiterated concerns of the indigenous movement, consultation. But can and should isolated peoples be consulted? If the central tenet of policy is to respect their autonomous decision to avoid contact, seeking them out for 'consultation' is impossible without violating their rights 19 . This position has since been accepted by the Inter American Development Bank's Operational Policy on Indigenous Peoples (IADB 2006). 19 Not to mention that even with very able translators it is hard to convey 'gas extraction megaproject' in the abstract to titled Machiguenga communities with 50 years of relations with national society. See Caffrey (2002) for more on inappropriate negotiations with the titled native communities around the Kugapakori Nahua Reserve. Any 'consent' given as a result of these consultations is unlikely to be 'free' or 'informed', and much more likely to be contingent on misunderstandings, fear, gifts or promises of future work.

Effects of IADB involvement
In 2002 the Inter-American Development Bank 20 (IADB) and Export Import Bank 21 (ExIm) took an interest in funding the Camisea Project. A civil society campaign was initiated to fully inform the banks about the existing and potential impacts of the project (AIDESEP et al 2003). ExIm eventually withdrew citing its environmental guidelines (Environmental Defense 2003). But the IADB went ahead with the loan ($75 million from IADB and $423 million in associated loans, IADB 2004), after appending additional conditions to its original environmental and social impact report (IADB 2003). Some of the conditions referred specifically to the Kugapakori Nahua Reserve (see Shinai 2004 p 85).
The IADB signed the loan approval in September 2003 in spite of cited civil society opposition. However, some IADB conditions required fulfilment before disbursement and an intense period of campaigning followed as indigenous and non governmental organisations tried to monitor the fulfilment of loan conditions by the Camisea consortia. Several attempts to force disbursement through were stalled, but the IADB eventually went ahead during the Christmas period of 2004while civil society organisations were on holiday-granting full disbursement in only one instalment just days after the first major spill from the Camisea pipeline. This spill left many Machiguenga communities located outside the reserve without fish for months.
A parallel $5 million loan to the Peruvian government for institutional strengthening for supervision of the Camisea Project (disbursal began in August 2003) has also been highly questioned. Among other things, it contributed to the preparation of an anthropological report about the Reserve which recommended the continuation of construction work inside the Reserve and dismissed evidence of impacts upon the inhabitants (Ossio et al 2004). AIDESEP, the national Amazonian indigenous federation, disseminated its critique of the report and the 'anthropological experts' who prepared it, declaring the lead author persona non grata (AIDESEP 2003). Nonetheless, the report led to the preparation of a Government 'Programme of protection and defense of isolated peoples'. More like company guidelines on how to handle encounters than how to defend isolated peoples rights, the measures proposed were considered bizarre and insulting 22 ; AIDESEP identified flaws which amounted to indigenous rights violations, including the failure to recognise indigenous control of their lands, incoherent use of terminology and lack of useful criteria to evaluate how to manage access to the reserve (AIDESEP 2004). In sum the 'Programme of protection' seemed to validate a company presence inside the reserve, although the project's impacts on health were already widely accepted. 20 The IADB is a multilateral bank, established in 1959, to finance the development of Latin America and the Caribbean. Its headquarters are in Washington, with country offices throughout the region. 21 The Export Import Bank is the official export credit agency of the United States. Its mission is to help fund the purchase of US goods and services in international markets. 22 For example, AIDESEP wrote in its critique that it 'encourages company workers and others to use violent means of repulsion (if they encounter isolated peoples) such as smoke guns, lights, alarms and whistles' (AIDESEP 2004).
One positive outcome of IADB pressure was the accelerated preparation and acceptance of a Supreme Decree for the Kugapakori Nahua Reserve, raising its status from 'Ministerial Resolution', which explicitly prohibited granting licenses for resource exploitation inside the reserve 'except that which already existed' (DS 028-2003-AG), i.e. the Camisea Project. The Supreme Decree was rushed through with little consultation (Shinai Serjali 2003) at a time when the government of Peru had to be seen to be fulfilling at least some of the IADB conditions in order for the loan to be approved. At the time some sectors of the government seemed to believe that similar decrees would be forthcoming for the other four territorial reserves for isolated peoples. To date none of these reserves, whose inhabitants are just as vulnerable but which lie in less politically and strategically visible locations, have been strengthened to this extent.
Over all, the experience with the IADB seems to suggest that there is little interest in the bank's private sector in protecting people or their rights per se. The bank is concerned about its public image and needs to be seen to be contributing to an improvement to justify its loan, but it does not carry out its threats (e.g. not to disburse until all conditions were met). The situation also demonstrated the government's lack of concern for rights violations-it only took action, and then minimally, when under pressure from the bank.

Pressure on health personnel
In May 2003, the epidemiology division of the Peruvian Ministry of Health, in coordination with AIDESEP, sent a team to investigate reports of child deaths in the Nanti communities in the headwaters of the Camisea. The team arrived in time to witness the end of another outbreak, this time of respiratory disease. Among other specific findings, the report made a link between an increase in reported outbreaks in 2002-3 and Camisea Project activity in the area and considered the project's presence key to understanding the development of outbreaks in headwater communities (OGE 2003).
These findings took over a year to be published 23 . The publication should have come in late 2003 and thus coincided with the period when the Camisea consortia, Peruvian government and IADB were trying to demonstrate 'good behaviour' to get the loan disbursed. Pressure was applied inside the government: the Minister of Energy and Mines himself presided over several meetings with the authors, sometimes accompanied by the General Manager of Pluspetrol, demanding that the Ministry of Energy and Mines (MEM) be given final approval of the text, in short an opportunity to censor it (Barclay 2004). At some of these meetings the CONAPA 24 representative was as heavy handed as the company and MEM, an instance of the flagrant loyalty to industrial and economic interests in many government sectors over the well-being of the Nanti (and other isolated peoples by extension). In an attempt to undermine the legitimacy and validity of the report, government and company representatives focused their criticisms on the Cabeceras Aid Project (CAP), a two-man humanitarian NGO that has supported the Nanti since 1997, and whose regularly updated census was the basis of the demographic analysis in the OGE report. MEM claimed, among other things, that CAP had an 'ulterior motive' for supporting the Nanti (Barclay 2004). However, Dr Pilar Mazetti, the Minister of Health, stood her ground, claiming publicly that the 'health of our citizens is more important than economic interests' (introductory remarks at presentation of report in July 2004). When the report was finally published in mid-2004, it had not been approved by MEM and appeared in its original form, but it had been held up for 6 key months for the establishment of the Camisea Project.

Law for isolated indigenous peoples
During 2004 a group of Peruvian NGOs collaborated with AIDESEP to develop a draft law to protect the rights, lives and territories of peoples living in voluntary isolation. This law was presented to a Congressional Commission for Andean, Amazonian and Afroperuvian Peoples in 2005. A small group of NGOs closely followed discussions within this Commission and it was widely known that the National Society for Mining and Petrol sent a lawyer to take notes at all the Commission's meetings. On one occasion an advisor to the Commission-responsible for the final text of the lawinformed NGO participants, that a group of energy sector and oil company representatives had requested a meeting with the Commission to express their point of view (Matos 2006). This was presumably done privately because it did not take place during the open meetings of the Commission which indigenous and NGO representatives attended. The same advisor later admitted that the Commission had been under considerable pressure from the energy sector and petrol companies (Matos 2006). Despite months of campaigning from the NGOs by the time the law came to be discussed in Congress significant changes had been made. The law was approved by Congress in May 2006 25 , but it does not recognise the intangibility of the Reserves for isolated peoples, allowing, under certain conditions, for the realisation of extractive industries 'in the interests of the nation' (Gamboa Balbín and Santillán Bartra 2006). The law also removed the term 'territory' from the title of Territorial Reserves (referring to them as Indigenous Reserves), an omission which indicates the Peruvian government's lack of respect for ILO 169 and its failure to meet obligations to protect indigenous territory. The law also creates additional obstacles that hinder the adequate protection of the current Territorial Reserves and the creation of any new ones (Gamboa Balbín and Santillán Bartra 2006).

Steps forwards, steps backwards: standing still?
Some government agencies in Peru do now profess a strong rights based stance in favour of isolated peoples, which interprets encouraging contact as a violation of indigenous and human rights to health, well-being and life as well as the rights to self determination, autonomy and territory (e.g. Defensoría del Pueblo 2006, CENSI 2006, OGE 2003.
However, less change is in evidence in the extractive sectors. Between 1990 and 2002 five 'Territorial Reserves' were created to protect the territories of isolated peoples in the Peruvian Amazon. All five are currently invaded by illegal loggers and/or overlapped by 'legal' oil and gas concessions. Five more reserves have since been proposed by AIDESEP but none have been approved by the government and all are affected by oil/gas exploitation and logging activities. Figure 2 shows all ten existing and proposed territorial reserves, indicating existing threats to each.
In January 2007 Perupetro (the state company responsible for managing oil and gas concessions) opened the bidding for 18 new concessions (Perupetro 2007) many of which affect existing and proposed Territorial Reserves (AIDESEP 2007). After international outcry during Perupetro's promotional visit to Houston, and a strong internal campaign in Peru led by AIDESEP, some of these lots were modified to exclude areas overlapped with existing Territorial Reserves (El Comercio 2007). However the guarantees for isolated people's rights are hollow given that Law 28736 does not make these areas intangible and those living in Proposed Reserves receive even less protection. For example, Blocks 67, 39, 104, 117 and 121 entirely cover the proposed Napo Tigre Territorial Reserve 26 , which would protect the territories of three, or perhaps four, peoples living in voluntary isolation on the border with Ecuador 27 (Shinai 2007b), and Petrolífera's Lot 107 overlaps the Proposed Cacataibo Territorial Reserve (see figure 2).
The examples described for the Kugapakori Nahua Reserve, lead us to conclude that in order to give the peoples who have chosen to avoid interactions with outsiders the opportunity to make their own choices about if, when and how they might like to establish relations with national society in the future, the territories of isolated peoples have to be protected from intrusions and the areas around them should be given priority health care to minimise potential disease transmission, while effective response plans must be put in place in case of emergency. Both Ecuador and Bolivia have, in the last year, created protected areas for isolated people that are intangible and prohibit all extractive industries (Servindi 2006, ENS 2007. In Peru, not even the Kugapakori Nahua Reserve has similar status, the activities of the Camisea Gas Project being allowed to continue within its borders. The 26 Also known as the Pucacuro Territorial Reserve or the Arabela Territorial Reserve. 27 In December 2006 the Peruvian President, Alán García, visited Block 67 in the northern Peruvian Amazon to make a statement about the importance of the oil discovery there (an estimated 250 million barrels) for the future of Peru. Barrett Resources plans to drill 117 wells, lay miles of road and pipelines to connect the platforms, and build a refinery (MEM 2006b). No mention was made of the evidence of isolated peoples living in the area. In 2007 AIDESEP made a demand to the Inter-American Commission for Human Rights for cautionary measures to be taken to protect the isolated peoples in the proposed Napo Tigre Territorial Reserve. The demand was accepted and the Peruvian Government has been asked to present evidence of how it is ensuring the protection of their rights (Servindi 2007a).
Peruvian State currently grants more protection to National Parks 28 than to peoples living in voluntary isolation, some of its most vulnerable citizens.
Other steps forward are that the National Centre for Indigenous Health (CENSI) has prepared a guideline on how to manage and respond to contact situations. This is an important first step towards an effective contingency plan, but it depends greatly upon the capacity of local medical teams which will need extensive additional training and does not include details of how to maintain long-term vigilance (CENSI 2006). INDEPA 29 has also prepared a Protection Plan for the Kugapakori Nahua Reserve 30 but its efficacy, and that of INDEPA, have been called into question. In December 2006 there were reports of new deaths in Montetoni. INDEPA had funding, as part of the Protection Plan, to send an emergency team there immediately. The speed of the 'emergency' response is questioned by some observers: the team arrived in Montetoni four days after the first death, but 20 days after the first documented cluster of cases of severe diarrhoea (at least 7 people) (Beier 2007a). On their return to Lima, INDEPA put out a press release indicating that two old men had died and that 'death was due to excessive consumption of masato' (INDEPA 26/12/2006). Masato is manioc beer; most Amazonian indigenous peoples consume it daily and rarely drink anything else. The suggestion that individuals would have died from drinking it in excessive quantities is insulting to indigenous people (see Inoach 2007). More up-todate information indicates that four people died-two young and able-bodied men and an elderly couple-and perhaps a child, during an epidemic of diarrhoea that affected at least half the population 31 (Beier 2007b). That the official report should attribute the deaths to excessive masato consumption shows a disturbing want of understanding from the very government body that is supposed to defend indigenous peoples, even after four years of work on a Protection Plan for the reserve. It appears that their visit to Montetoni was cursory and uninformative and that attention by the local health post was also insufficient: 'the level of ignorance demonstrated by current Camisea health centre personnel (setting aside the pejorative and racist attitude embedded in their analysis of the situation in Montetoni) greatly contributed to the severity of the recent epidemic' (Beier 2007b). That INDEPA stated publicly 28 Oil and gas exploitation is prohibited within National Parks yet even they are not necessarily safe from the petrol frontier. In September 2007 a new law was proposed to the Peruvian Congress to cut a 200 000 hectare section out of the Bahuaja Sonene National Park, in the Southern Amazon, to allow gas exploitation to occur (see: http://salvemoscandamo.com/). 29   (INDEPA 2006, this press release has now been removed from their website) that masato was the immediate cause of death is evidence of the negligence and racism which at once blames 'their' customs and draws attention away from the proximal causes of death and sickness and the underlying causes that need to be investigated, such as forms of transmission, sources of infection and the social context that enables them.

Conclusion
In the first half of this paper we presented the debate on the ethical and practical implications of encouraging contact with indigenous peoples who are actively avoiding it. Regardless of the health impacts, the human rights arguments should be sufficient to prevent contact, but this requires national governments to take responsibility for effectively protecting the territories of voluntarily isolated peoples from invasions and environmental impacts that could affect their livelihood strategies, quality of life and health. Yet, even as the UN Declaration on the Rights of Indigenous Peoples is accepted, underscoring the rights of indigenous peoples to selfdetermination, autonomy, life, freedom, integrity and health (UN General Assembly 2007), national policies are being promoted which will undermine all of these for the most vulnerable peoples.
In the second half of this paper, we have presented four detailed examples of how the oil/gas industry in the Urubamba Valley has directly threatened the isolated peoples of the Kugapakori Nahua Reserve or affected attempts to strengthen national legislation in their favour. These examples illustrate the role the oil/gas industry plays at all levels-on the ground, in the preparation of national legislation and in the international arena-to undermine the promotion of safeguards for indigenous peoples in voluntary isolation, based on the principles debated in the Amazon over the last few decades.
If this trend in the quest for oil and gas (and other extractive industries, particularly timber) is not reversed soon, the violation of rights and eventual death of hundreds of indigenous peoples seems inevitable. This kind of genocide has occurred before but it would appear that the Peruvian government and the international oil sector have learnt little from the 500 year history of indigenous decimation in the Americas. In the words of Marcus Colchester: If we deplore the horrors of death and destruction that ineluctably accompanied previous penetrations of the Amazon, can we now show that modern industrial society is more civilised? Can we respect the choice of other societies to avoid contact and leave them in their homelands undisturbed until, perhaps, some future time when they themselves decide on the risky venture of contacting a world that they have learned by bitter experience is not safe to interact with? If we cannot, then it is almost certain that future generations will condemn us for the same avarice, indifference, selfishness and greed, for which we today condemn the conquistadores and the rubber barons.' (Colchester 2004) References AIDESEP