Year : 2004 | Volume
: 2 | Issue : 2 | Page : 411--432
Community Conservation, Inequality and Injustice: Myths of Power in Protected Area Management
School of Geography and the Environment, University of Oxford, Mansfield Road, Oxford OX1 4LT, United Kingdom
School of Geography and the Environment, University of Oxford, Mansfield Road, Oxford OX1 4LT
The principle of local support states that protected areas cannot survive without the support of their neighbours. It is the dominant motif of much writing about community conservation and the integration of conservation with development. However, we should be sceptical of it for several reasons. First, it implies that the weak can defeat the agendas of the strong. Second, the principle ignores the fact that inequality and injustice tend to be perpetrated about the globe. It is not existence of poverty or injustice that will cause problems for conservation, but their distribution within society. Third, a detailed case study from the Mkomazi Game Reserve in Tanzania shows how conservation can flourish despite local opposition. Advocates of community conservation need to pay more attention to fortress conservation«SQ»s strengths and especially its powerful myths and representations. Understanding how inequality and conservation are successfully perpetrated will make it easier to understand the politics of more participatory community conservation projects.
|How to cite this article:|
Brockington D. Community Conservation, Inequality and Injustice: Myths of Power in Protected Area Management.Conservat Soc 2004;2:411-432
|How to cite this URL:|
Brockington D. Community Conservation, Inequality and Injustice: Myths of Power in Protected Area Management. Conservat Soc [serial online] 2004 [cited 2021 Mar 5 ];2:411-432
Available from: http://www.conservationandsociety.org/text.asp?2004/2/2/411/49336
LOCAL SUPPORT IS not necessarily vital for the survival of protected areas. Conservation can be imposed despite local opposition and protected areas can flourish notwithstanding resistance to them. Rural poverty and injustice do not undermine the foundations of conservation. Indeed, they can underpin them.
These will be shocking ideas in many quarters. The new (and appealing) ideology is that the rural populations living around places of wilderness and high biodiversity must value these places as their own and desire their protection. Otherwise protected area policies will fail. By embracing these challenges conservation has acquired a human face (Bell 1987). In doing so it has become firmly established in the development options facing rural people and governments.
Denying its fundamental tenet may be unpopular. But doubts need to be voiced lest the ideology's unexamined repetition undermines the community conservation project. The thesis proposed here is that conservation can be imposed because the rural poor are weak, and resistance to conservation, however persistent, may be ineffective. We have to understand the forces that make protected areas powerful if we are to effectively meet the needs of protected areas' neighbours and win their support.
I should declare my allegiances. I believe that the goals of community conservation deserve support, but that its necessity for the success of conservation has been overrated. Protected areas do not all have to have the backing of their neighbours in order to survive. Local groups can be ignored by protected areas with impunity; passing over local groups may make no long-term threats to a protected area's security.
The purposes of this article are: first, to explain the reasoning behind this argument; second, to present data that have informed this thinking; third, to discuss limitations to generalising from this experience; and, fourth, to consider what circumstances facilitate sustained oppression in conservation. I will first examine the idea that local support is needed in more detail, and outline the circumstances in which local opposition may not matter. I will then discuss a case study, the Mkomazi Game Reserve in Tanzania, which illustrates the weakness of opposition. Finally, I will discuss the consequences of this argument for the practice of community conservation.
Power, Inequality, Resistance and the Principle of Local Support
The importance of local support for the survival of protected areas has been expressed in the strongest terms. Western (2001: 202) writes that 'a fallacy of protectionism is that we can ignore costs locally'. The president of the World Conservation Union (IUCN), opening the fourth World Parks Congress in Caracus, stated the importance in bald terms: 'quite simply, if local people do not support protected areas, then protected areas cannot last' (Ramphal 1993: 57). Barrow and Fabricius (2002: 77) state that 'ultimately, conservation and protected areas in contemporary Africa must either contribute to national and local livelihoods, or fail in their biodiversity goals'. Phillips, when asked to name one key lesson to be gleaned out of interactions between protected areas and their neighbours, found the answer 'very simple': it was 'the iron rule that no protected area can succeed for long in the teeth of local opposition' (Borrini-Feyerabend et al. 2002: 11).
In this article I will call this idea 'the principle of local support'. It is a principle to which there is little opposition. Sceptics of community conservation may vigorously deny the effectiveness of development programmes in achieving conservation's goals (Kramer et al. 1997; Oates 1995; Spinage 1998, 1999a, 1999b, 2002; Struhsaker 1998). They urge conservationists to continue to preserve nature as they did when protected areas first flourished, and not to get sidetracked into development. But, in their frustration over the challenges to protected areas, they do not challenge the principle of local support. Positive assessments of national parks' impact of nature sidestep the issue making little mention of their social impact, while also alluding to the importance of compensation schemes (Bruner et al. 2001). Indeed, many admit the importance of winning local people to the cause, preferring education programmes and perhaps some benefit sharing, to the more radical practices of community conservation.
Nor is the principle challenged in social scientists' criticism of community conservation. Observers of the CAMPFIRE project in Zimbabwe found flaws in the project's expectation that wildlife revenues will encourage local people to treasure wildlife. They found that people's expectations of development meant that they wanted wilderness to be tamed, wildlife to withdraw, more people to come in, and the tangible benefits of development and modernity to be made available. Denser settlements meant better shops and more services (Alexander and McGregor 2000). This is a possible consequence of participation, namely, that people who do not value wild fauna and flora will be empowered to make decisions over them. But the principle of 'local support' was not challenged by this critique. Rather, the difficulty of winning that support and changing values was emphasised.
Indeed, the principle of local support seems precisely the sort of idea that ought not to be challenged. If protected areas perpetrate injustices and mete out inequality, poverty, homelessness and rootlessness on their neighbours, should they 'get away with it'? Questioning its necessity may seem to deny the moral imperative of community conservation. On what grounds can it be challenged?
The nub of the argument is this. The local communities who oppose the existence and policies of their neighbouring protected areas tend to be politically weak rural groups. They can be opposed to powerful alliances of central and local governments, the police, park guards and paramilitary units, and national and international NGOs raising money and awareness for the cause of the protected areas. These are contests that the rural groups may be ill-equipped to win. Asserting the necessity of their cooperation ignores the realities of power. Some local groups can be ignored.
To put it another way: there are countless examples throughout history of inequalities and injustices being perpetrated and perpetuated despite resistance to them, and despite the opposition and hatred they generated.  The Roman empire was not brought down by its slaves, enclosure in England and clearances in Scotland were not prevented by the people who lost their rights to the commons, nor were the iniquities of England's factory system overturned by a workers' revolt.  Indigenous peoples in Latin America, North America and Australia have been removed from their lands and violently treated for hundreds of years. Why should the injustices perpetrated by conservation be any different?
Against this, conservationists may protest that the ideologies justifying protection of nature are utterly different from those that justified the exploitation and maltreatment of people. Conservation's ideology is perhaps the most inclusive possible. It is concerned with protecting the planet not just for all people, but for all creatures and all time. Slavery, settlers taking indigenous lands, enclosures and the early Industrial Revolution were propped up by ideologies that justified treating some people as less than human. But these processes sustaining inequity do not have the same ideological justification in conservation.
There is some truth here. A large body of concerned conservationists will find this article disturbing and try to fight sustained inequity. Their ideologies do not tolerate oppression. But ideology rarely deliberately justifies callousness. Instead, it encourages forgetfulness of suffering or problems, or focusing attention on other, higher, more valued things. Ideologies set up principles or higher causes to which people strive, and in whose name sacrifices have to be made. These sacrifices will be a necessary short-term solution until present problems are overcome. In the meantime, they are fundamental for the higher cause. Thus, shoot-to-kill policies in Kenya and all the other violence that the practice of doing conservation can entail (evictions of people, elephant culls) become an unfortunate necessity in the service of conservation (Peluso 1993). Conservation ideologies, like other ideologies, also work by ignoring or overlooking problematic consequences of its policies. Because it is a good cause undertaken for posterity on behalf of nature, people do not expect any ill effects. Fund-raising or publicity campaigns do not need to have small print about the social impacts of the policies advocated. The good of conservation is such a powerful 'myth' that it dulls our expectations of ill effects (see later sections). As an illustration, consider the fact that the extent of protected areas on the planet has almost tripled in the last generation from about 3 per cent of the earth's surface in 1980 to more than 11 per cent now. Yet this expansion has taken place without any careful social impacts assessment of the consequences of gazettement. We have no idea how many people were evicted to create these protected areas and what the social consequences were (SchmidtSoltau 2003). We have not thought to look until now. Similarly, it is only recently that the social codes of conduct for the larger conservation NGOs have been advocated that would deter adverse social impacts from conservation. We had not thought this necessary until recently. Conservation ideology did not suggest that it might be necessary.
Other objections to my argument are posed in the reasoning of the principle of local support itself. Its adherents make two challenges. First, they object that the alliances that maintain coercive conservation-of governments, enforcement agencies and international support-are fickle. They are strong when they work together, but they have proven too unreliable too often to form the basis of a credible conservation policy. The history of national parks is replete with examples of paper parks for which there is little government support, where there is no enforcement of conservation law and which the international community has forgotten. Local groups have their way in these places, perhaps to the detriment of conservation. Oppression now will just store up trouble for the future.
This cannot be denied, but nor is it the complete story. Just as there are failures there are also successes-places where the alliances have sustained fortress conservation for a long time. Our understanding of power and local resistance must be able to incorporate both. Where things have broken down, it is wrong to assume that the damage to the protected area is entirely locally driven or locally endorsed. Much elephant and rhino poaching in Africa is driven by international gangs and government corruption not local people (Leader-Williams and Albon 1988; Smith et al. 2003). Locals may be disinclined to resist them, but this may be because the gangs are violent and heavily armed as much as because the locals dislike their neighbourhood park. Further, breakdown of the state alliance does not necessarily mean the irreversible destruction of the park. Nature's resilience can provide some cushion; as restored reserves saved from the brink of disaster testify.
In short, this response does not quite meet the point I am making. The argument is not that all neighbourhood resistance can be ignored with impunity. Rather, it is wrong to assume that doing so will necessarily threaten protected areas' longterm survival. We need to look at the circumstances and politics of oppression and resistance to understand how resistance arises to threaten protected areas, and how it might fail to mount a credible threat.
The second objection of the principle of local support is that, even where the alliance between the government, guards and the international community remains strong, it may not be equipped to resist the seething mass of resentment of numerous hostile neighbours. These conflicts will not be expressed so much in direct confrontations, but rather in the 'weapons of the weak', in the numerous opportunities to express their opposition on the landscape, flora or fauna of the protected areas through acts of vandalism and sabotage (Neumann 1998; Scott 1985). Few systems of oppression and surveillance can totally extinguish these weapons. And around protected areas their expression can mean the slaughter of valued animals, attacks on tourists, fires and mass disobedience, which are beyond the ability of the state and its international allies to control. The archetype here is the opposition to Amboseli National Park, described by David Western (1994), where local Maasai herdsmen expressed their opposition to conservation measures through the largescale slaughter of wildlife. In the face of such disturbing violence, it may seem downright perverse, to put it kindly, to suggest that local people can be ignored.
This is sound reasoning, but before we draw any conclusions on the basis of this experience, we must consider more carefully how local resistance to the park came to be so united and focused. What were the politics of this relationship, and what might make it unusual? What were the factors that allowed people near Amboseli to find common cause in their opposition to the park, and to overcome all the divisions and diversity found in such local 'communities'? We cannot assume that opposition will always be so united. Rather, I suggest the opposite- that it is in the divisions and diversity within protected areas' neighbours that the means to a political alliance in support of conservation may be found. Instead of assuming that local opposition makes conservation unsustainable, we should ask under what circumstances, and with what configuration of local alliances and external support, it might be sustained.
All societies distribute misfortune unequally, the least powerful people tending to experience it most. Misfortunes inflicted by protected areas can be concentrated upon a minority, who in their weakness and want of numbers are unable to do anything effective about it. In such situations, even if the protected area generates few benefits, the unaffected majority can ignore the harm it does. Fortress conservation will be strong not despite the misfortune it causes, but because of the way in which this is misfortune is distributed amongst its neighbours. Mkomazi Game Reserve in north-eastern Tanzania illustrates this mixture of power and injustice. Mkomazi is noteworthy because it demonstrates a two-fold exclusion of people, materially and symbolically, because of the patent injustices involved and because of the failure of local resistance to these policies and the contrasting success of fortress conservation policies. It also provides an interesting counterpoint to Amboseli because it is located quite close to it and involves some of the same ethnic groups.
The Mkomazi Game Reserve is in north-east Tanzania, bordering Kenya and the Tsavo National Park.  It is largely composed of semi-arid Accacia-Commiphora bushland with several hills supporting cloud forest. Its vegetation, birds and insects are diverse, its mammals largely dependent on the Tsavo National Park to which Mkomazi provides a wet season dispersal area (Coe et al. 1999). Mkomazi was created in 1951, but rose to international prominence after a large number of pastoralists were evicted in the late 1980s. It is a highly contested reserve, with large numbers of people (50,000 in 1988) living within a day's walk of its boundaries. The summary that follows is drawn from the research conducted by Homewood, Kiwasila and the author around the Reserve in the mid-1990s (Brockington 1998, 1999, 2001, 2002; Brockington and Homewood 1999, 2001; Homewood et al. 1997; Igoe and Brockington 1999; Kiwasila and Homewood 1999; Rogers et al. 1999).
The circumstances of the presence, and removal, of these people are complex. The plains in which Mkomazi are gazetted are largely dry and not good for farming. Accordingly, the reserve's main users have been hunters and herders. When the reserve was gazetted, a group of Parakuyo pastoralists (who are closely related to the Maasai) were living in the eastern half of the reserve, where they migrated between the Usambara mountains in the dry season and the Katamboi waterholes (in Kenya) in the wet.  A limited number of them, with their herds, were allowed to stay. This arrangement ignored the Pare and Sambaa people, who lived in and around the Pare and Usambara mountains, and formed the majority ethnic groups around Mkomazi. The reserve, however, was set up in turbulent times. The access arrangements made with the Parakuyo were not made with stable populations, rather at a particularly fractious moment in a dynamic history of migration and population change. The attempt to fix access rights to a few members of one ethnic group ignored the history and realities of pastoral ecologies and identities.
In the early nineteenth century these lands had been used by the 'Kwavi'- Maa-speaking herders from whom the present day Parakuyo are descended (Jennings 2003; Krapf 1860). These herders were expelled by the Maasai in internecine strife (known as the Iloikop wars), many seeking refuge in nearby agricultural communities (Beidelman 1960; Bernsten 1979; Galaty 1993; Waller 1979). Use of the vacated plains by Maasai herders grew gradually over the course of the century, but they were decimated when rinderpest struck in 1891 and again after German colonists attempted to confine them to lands west of the Pangani River in the late 1890s (Baumann 1891; Brockington 2002). It was only after the First World War that population growth and repopulation of the plains resumed.
The recovery of herds after rinderpest led to ever increasing use of the plains. Agriculturalists in the hills began to keep more stock in the lands below. The Parakuyo regained their herds, keeping cattle north of the Usambara mountains in the early years of the twentieth century. The Maasai began to move east across the Pangani river to the borderlands between Kenya and Tanganyika in which Mkomazi was to be gazetted.  Meanwhile, the British authorities on both sides of the Kenya-Tanganyika border were tightening their administrative boundaries, 'packaging of the land' into parcels of territory allotted to pastoralists, settlers, agriculturalists, national parks and the Crown, all of which were forbidden to immigrating pastoralists (Collet 1987). As a result, the state on both sides of the border found itself fighting the influx of herders into the borderland plains as they tried to move back into territory they had lost to war and disease. Things came to a head in 1950. Thousands of cattle and their herders were forced out of Toloha, on the border with Kenya, and moved to new grazing grounds normally used by Parakuyo at Katamboi. At this moment Mkomazi was gazetted.
The sudden arrival of the Toloha pastoralists at Katamboi was not initially welcomed by the Parakuyo. They complained to their district authorities. This local discontent and the establishment of Mkomazi were seen as powerful reasons by the government to justify the removal of the Toloha herders. But the government was reluctant to move the Parakuyo, whose right to stay rested on their long residence on the plains. The Toloha herders were therefore evicted, and the Parakuyo allowed to stay.
The expanding cattle populations across the Pangani river, however, meant there was ever-increasing pressure on the reserve's resources. Pastoral ecology (and ideology) in uncertain environments depend on reciprocity, and in particular being prepared to grant access to grazing and water resources (Behnke and Scoones 1993; Galaty and Bonte 1991; Johnson and Anderson 1988). Failure to do so contravened deep-rooted sentiments. The Toloha herders who were expelled shortly after their first, sudden arrival, returned, and others came with them, using gentler means of gaining access to Mkomazi's pastures. They exchanged stock and intermarried with the residents of the reserve. Regulations stipulated that only the Parakuyo could remain there, but ethnicity proved flexible (Spear and Waller 1993) so Maasai and other immigrants became Parakuyo, even buying the names of deceased herders listed as legal residents. In addition, permissible stock numbers and human populations within the reserve were also increasing.  Pare and Sambaa peoples also sought access to Mkomazi, pointing out that it was unjust to exclude one ethnic group but give access to another.
The early history of Mkomazi, therefore, was dominated by increasing cattle populations. Initially, residence and pasture were only allowed in the east of the reserve. Then, in the early 1970s, following drought, local pressure combined with weak management, and lack of government support, the western half of the reserve was also opened up to use by herders. The consequences of this local pressure on Mkomazi are clearly visible in the cattle counts for the reserve taken over this period [Figure 1].
At this point in the account you may well be wondering how this case study could possibly illustrate the power of fortress conservation when all local opposition, weak and failing administration, and unsuccessful conservation suggest the opposite. Surely this is where the principle of local support would have been most useful? If the reserve had had some relevance to people's livelihoods, if it had been valued in some way, then it would not have experienced the intense pressures it had been subject to, nor failed so dramatically.
But the story is only half told. In the 1980s the Department of Wildlife decided to reclaim the reserve. They had long been concerned about the environmental impact of so many cattle, and lobbied district officials to have the herders removed, with increasing success. In 1988 the final remaining residents were evicted. In 1989 the government agreed to invite the George Adamson Wildlife Preservation Trust to rehabilitate the reserve. This began with the provision of roads, airstrips, a plane, radios, cars and other equipment. In 1994 a wild dog breeding project was started, and in 1997 four black rhino were flown into a sanctuary from South Africa, joined by four more in 2001. Tourist facilities are basic, but plans are progressing. Mkomazi has joined the international arena and its conservation is thriving.
This conservation success, however, has come at considerable cost to the livelihoods and rights of the people who were moved. The failure to resist the evictions, or claim adequate compensation for their losses, demonstrates the power of fortress conservation and the weakness of local opposition. The evictions were forceful, the impact on livelihoods severe, the legal redress minimal, even farcical, and publicity about the reserve silent about the plight of evictees. They have been cast out physically and marginalised in its international images.
There is little doubt that the evictions themselves were violent and traumatic.  Oral histories spoke of houses and possessions being burnt, families being dumped outside the reserve, without rehabilitation. In the longer term, the evictions were disastrous for the livestock economy. The over-concentration of cattle around the reserve meant that many died of disease and want of food. Fines for illegal grazing have been heavy. The reserve was particularly valuable for its nutritious wet season grasses, which enabled stock to recover from the rigours of the dry season. Grazing outside the reserve is cramped, with pastures limited by the proximity of farms and mountains.
The facts of change are stark. Average herd size plummeted [Table 1] and many people left the region (Igoe and Brockington 1999). [Table 1] also shows that over half of the households do not have the lowest estimated number of cattle (four per capita) needed for subsistence requirements (Fratkin and Roth 1990; Roth 1996). With the decline in herds almost all families now cultivate (Dahl 1979; Dietz 1987). Indices of herd fertility and milk yields at Mkomazi compare poorly to pastoralists in comparable situations (Brockington and Homewood 1999; Homewood 1992).
Families with inadequate milk for subsistence sell it, on a daily basis, to buy staples (Fratkin and Smith 1995). The returns from milk sales are normally controlled by women. But at Mkomazi a common grievance voiced by the women was that the income they formerly used for their own purposes was now required to support the whole family. Responsibility for household expenditure is a contested issue within the household. Men, in charge of the herds, try to defray some household expenses they might otherwise meet by selling animals, to women, who have to meet them by selling milk or other goods. But where there is no herd, or it is greatly reduced, women have less bargaining power. At Mkomazi, more of the responsibility of buying food now falls on the women (Brockington 2001).
The consequences at the household level are mirrored in changes in the district livestock economy. Livestock sales have declined [Figure 2]. The decrease reflects the collapse of the Kisiwani market, which was closest to the reserve. Its collapse has considerably decreased the circulation of cash in the local economy and is a major cause of complaint amongst local businessmen (Brockington and Homewood 1999). In addition, the proportion of female animals sold has dramatically increased [Figure 3]. Female stock are the productive nucleus of the herds. Their sale is a sign of stress amongst pastoralists.
Agricultural people living around the reserve have also suffered from reduced access to firewood, wild foods and beehive sites. Kiwasila has documented a number of costly limitations imposed by the reserve (Kiwasila and Homewood 1999). However, it must be emphasised that the brunt of the costs of the eviction have been visited upon a pastoral minority. Their livelihoods have been most affected by banishment. The evictions moved about 5,000 people who were not, primarily, from the majority ethnic groups in the area. Most were Parakuyo and Maasai herders.
The local social and economic costs have not been offset by new revenues. There are no tourist hotels at Mkomazi. Those envisaged have tens of beds rather than hundreds. Even if there was a mechanism for distributing revenues to local villages, there are too many people involved, and their losses are too great for revenue from tourism ever to compensate them. It is difficult to see on what material grounds an alliance between Mkomazi and its neighbours could be built (Brockington 2002; Homewood et al. 1997).
Yet, despite this impoverishment, herders have not been able effectively to resist their eviction. This is not for want of trying. As the evictions took place, herders sent representatives to plead with the district and regional commissioners, the minister of internal affairs and the party head offices in Dodoma (at a time when Tanzania was a one-party state). They sought help from the local Catholic church. This was to no avail (Brockington 2002; Mustaffa 1997). With the failure of these petitions, they sought help from 'indigenous peoples' non-governmental organisations (NGOs) set up to contest land rights for pastoralists elsewhere in Tanzania. With the help of these organisations, the Legal Aid Committee of Dar es Salaam University and Western donors, a court case was launched seeking compensation and recognition that customary rights had been lost and should be returned. Ninetyeight plaintiffs were named, including herders, and their descendants, who had originally been allowed to live in Mkomazi when it was first set up, by virtue of their long residence on its plains.
But the outcome of the court cases was disappointing. The court recognised that the evictees had suffered as a result of the violent and poorly-planned evictions, but did not award compensation for damages incurred because the case had been made after the statutory three-year limit. It recognised that the plaintiffs had customary rights to the eastern half of the reserve, and that the evictions were unlawful because they were carried out before these rights had been properly extinguished and alternative lands found for the residents. But the judge decreed that, since the evictions had taken place over ten years ago, these rights had been 'overtaken by events'. Compensation of 300,000 Tz shillings (about US$ 350) each was awarded to thirty-eight plaintiffs for the hardships suffered due to this unlawful eviction. Herders contested the decision in the appeal courts arguing that their customary rights could not be 'overtaken' thus. But the appeal courts reduced the number awarded compensation to twenty-seven. Most extraordinarily, they concluded that the plaintiffs had no customary rights to the eastern portion of the reserve because they were members of a Maasai community, which only began to reach the area in the second half of the 1940s.
But it is the absence of this story from so much public literature about the reserve, and especially the fund-raising literature on which recent conservation successes have been based, that best illustrates the marginality of Mkomazi's evictees. At one point the literature of the George Adamson Wildlife Preservation Trust, and its sister trust in the USA, the Tony Fitzjohn/George Adamson African Wildlife Preservation Trust, claimed that the people evicted from Mkomazi were not indigenous to the area. In other literature and web sites the needs of the reserve's neighbours and the vital importance of projects such as primary and secondary schools, are given great weight.  But the costs of the evictions are not mentioned. There is no sense of whether the gains offered by school support match the losses to livelihoods and cattle markets.
Indeed, to some extent the success of the international programme and the marginalisation of the herders are interdependent. Mkomazi's restoration was in part facilitated by the herders' marginalisation. When the rhino sanctuary was first planned, two South African rhino specialists concluded:
There appears to be limited resentment towards the Mkomazi Game Reserve by the Msaai [sic], as they were well aware that their permission to graze within the reserve was only a temporary one . . . . The more numerous Wapare and Wasambar [sic] tribe members within the Kisiwane [sic] and Uzambaras [sic] areas were never historically associated with the reserve and thus have no negative feelings towards it . . . . [I]t would appear that the introduction of black rhino into the MGR would be . . . little affected by the limited to dwindling negative feelings towards the surrounding communities (Knight and Morkel 1994: 6-7). 
The nonsense of this assessment is apparent from the fact that it was written just three months after aggrieved pastoralists met with lawyers to prepare their court cases, claiming rights to Mkomazi and compensation. Yet the erasure of pastoralists' presence from Mkomazi has been a repeated theme in representations of the reserve. When concern over the move of South African rhino to Mkomazi resurfaced in 1998, following lobbying in South Africa by NGOs supporting the pastoralists' case, the Tanzanian government was able to allay South African fears by assuring worried conservationists that the court cases brought against the government were from 'a group of Maasai who originally came from Kenya' (Koch 1997: 109). A subsequent draft report by the Department of Wildlife stated: 'When the Mkomazi/Umba Game Reserves were established, six pastoral families were living inside and they were compensated in order to move to areas outside the Reserve' (Mbano 1997: iii).
The same web sites that emphasise the need to care and provide for Mkomazi's neighbours are silent about its former residents, except to stress that they were causing environmental degradation and had to be removed for the reserve to survive.  The international representation of Mkomazi ends up being an Orwellian rewriting of the reserve's, and its people's, histories.
Yet the rhino are thriving. The wild dog were hit by canine distemper, but their populations are recovering. These interventions have been successful, at least on their own terms. They have not been impeded by local opposition. Indeed, the success of the conservationists, and the existence of these reports, has confirmed the marginalisation of the herders. With its rhino and wild dog, Mkomazi is now established on a more secure footing, and the pastoralists' case looks even less attractive to potential allies in government.
It is important to point out that, while the marginalisation of herders may be the result of the conservationists' campaigns, it was not their intention. Nor was the labelling of evictees as 'not indigenous' deceitful (the history of the reserve was not well known at the time, and the label has not been repeated). Supporters of the trusts are experienced professionals, respected pillars of the political and conservation establishment. They support these actions because they believe that these policies are necessary to save Mkomazi. They believe that human costs have to be paid to prevent the destruction of the environment.
There are good reasons, however, to question this urgency, and the ecological arguments on which it is based. This issue revolves around the dynamics of vegetation change in semi-arid areas in response to grazing pressure, of which there are conflicting interpretations (Behnke and Scoones 1993; Illius and O'Connor 1999; Sullivan and Rohde 2002). Conservationists have portrayed the reserve as fragile and destroyed by cattle. The alternative view is that nature here is resilient: it recovered remarkably quickly after the evictions, and was able to sustain far larger numbers of cattle than experts thought, and continued to do so even while it was being degraded. I cannot rehearse the detail of these arguments here, but refer readers to the relevant publications if they wish to explore it further (Brockington 2002; Brockington and Homewood 2001; Canney 2001; Homewood and Brockington 1999). Conservationists may have thought they had a good cause to expel pastoralists, but these reasons, and their justification of eviction, are controversial.
What lessons can be learned from this situation? There are a number of elements that may make this case study unrepresentative. The environment of the reserve has proven resilient to the impact of large numbers of cattle. Sustained occupation of the reserve for several decades has still left levels of biodiversity exciting to scientists and conservationists six years after the evictions (Coe et al. 1999; Homewood and Brockington 1999). Pastoralists, despite their numbers, did not exploit the resources intensively; stocking rates at their maximum were between 3 and 4 ha per cow. Mkomazi is not an isolated reserve, but an extension of the Tsavo ecosystem. It is not surrounded by people. In such circumstances temporary failures of fortress conservation are possible, because the potential still remains for its recovery.
More important are the divisions within the local 'communities' that surround the reserve. The majority of people who live close to Mkomazi (50,000 at the time of the evictions) were not intensively involved in cattle keeping. The evictions have not harmed them significantly. In the west of the reserve, which was occupied most recently, Maasai immigrants were unpopular. There have been long histories of conflict and cattle theft between them and the Pare people, whereas in the east of the reserve relations between the Parakuyo and Sambaa people were generally better. In the west, however, the prospect of large populations of herders being returned to the reserve was not popular. Here local lobbying has argued for the boundaries to be redrawn further away from the mountains rather than allowing residence. The evicted pastoralists are thus politically weak, without a broad base of popular support. Indeed, many have now dispersed and live far from Mkomazi. They are unable effectively to press their claims. Harsh policies work because the misfortune has been concentrated on to relatively few people.
This is a relatively recent case study. Only fifteen years have passed since the pastoralists were evicted, and much less time since the rhino sanctuary was established. But the alliances that sustain Mkomazi are strong and show little signs of weakening. When hotel facilities are developed and the reserve starts to generate income for the government, it will become still more valuable. This reserve now has international importance.
Mkomazi has four lessons for community conservation. First, writings by conservation progressives can appear to suggest that injustice is inherently unsustainable. In doing so they attribute too much power to weak rural groups. They overlook some of the ways in which fortress conservation can be successful, and the way in which oppression works. For example Stuart Marks (1984: 130) wrote:
The romantic vision of keeping Africa as an unchanged paradise teeming with wildlife is a foreign nonsense, for to ask East Africa to perpetuate such an image is to ask it to stay poor and undeveloped.
But does this mean that this poverty causes the vision to fail? If the poverty is unequally experienced, if the benefits are experienced by the elite in Africa and abroad, then the vision has a good chance of success. East Africa is not one unit with the sentience to accept or reject protected areas. It is profoundly divided. The distribution of misfortune and benefit is therefore vital to conservation's prospects in the region.
Adams and McShane (1992: 241) argued that a more realistic understanding of Africa's history, and of the role of people in its landscapes is necessary because they believe that 'conservation based on myth is bound to fail'. But why should they fail because they are historically wrong? If they can generate money, gain the support of the foreign and political elite, and widespread public sympathy ('global' opinion) these are enough for success.
The double meaning to the word 'myth' makes it even more unlikely that policies based on myth will fail. Myths can refer to ideas which are simply, and totally, wrong. But the word also refers to ideas which are just immensely powerful.  These myths guide and structure our world views. In Bourdieu's (1998: 34) words, they 'obtain belief'. Myths shape our thoughts and interventions, they determine our interpretations about what is wrong with the world and its solutions (Horne 1993; Lewis 1961,1971). Myths are, in short, immensely influential. Policy underpinned by myths are almost bound to succeed. 
The idea of wilderness and wild Africa is an example of this power (Anderson and Grove 1987; Cronon 1995; Grove 1995). The wilderness myth is powerful because it evokes ideas of pristine purity, unspoilt origins, a world not marred by people. When fund-raising literature for Mkomazi invokes an image of a wilderness restored to its former glory (and implicitly or explicitly writes people out of the reserve's history), it does not matter that this may be a poor reflection of people's place there, or its recent history. It works because it invokes a 'great idea', it responds to a powerful need in the West. It raises the funds to make that idea a reality. This literature, and the mass of novels, coffee-table picture books and films that Adams and McShane were criticising, live in symbiosis with the protected areas they describe. Each shelters the other. Fortress conservation policies preserve not just wilderness, but also a dream of Africa, and in the process they reproduce and sustain its supporters (Brockington 2002).
Adams and McShane direct attention to the realities of the world in which conservation works, exposing facts, histories and processes that conservation has to deal with in order to succeed. I would endorse their call, but add that one of these realities is the power of myth. The inequalities that accompany conservation policies may not threaten conservation areas if there are myths to sustain them. The representation of conservation, its images and publicity, and its blindness to the harm it causes, plays a significant role in sustaining its activities. These can prevail against local opposition, especially if this opposition can be written out of conservation's representations. The lesson from Mkomazi is that injustices are easier to sustain if there is strong international support for the conservation policies being practised.  Strong external support makes it easier to override local opposition.
Second, the presence of social injustice perpetrated by conservation does not mean that, automatically, these injustices will be rectified and conservation values suffer as a result. People may absorb abuse as much as challenge it. There are many injustices that remain unsolved. This is not the sort of place that many would choose, as Rawls (1971) suggested, randomly to be born into. The daily abuses of structural power and inequality wreak unanswered havoc on the lives of millions of people every day (Bourgois 1995, 2001).
The belief that oppression and injustice are inherently unsustainable is ahistorical. There are many examples when justice has won-such as the ending of slavery, the successes of the US civil rights movement, the ending of apartheid. But we cannot conclude that these causes were successful because they were right. This would not only be ahistorical, but also apolitical and, indeed, naive. If change occurs, it is not always the struggles of the oppressed that drive it. The abolition of slavery was achieved by the moral decisions of English politicians, not the battles of slaves (Iliffe 1995). Nor can we assume that progress is permanent, as the reoccurrence of slavery shows. The battle for civil rights in the United States had been pursued for a long time before concessions were won, and even then these were vulnerable to a renewed white backlash (Tuck 2001).
The point is not to accept that these causes were successful, but to ask how and why, to establish what configuration of circumstances lead to these changes. If conservationists are to challenge inequality and injustice then this will require a proactive engagement with the political, social and economic forces that cause and reproduce it.
The iron rule that Adrian Phillips invoked (see p. 412) exists, but it is a moral necessity rather than a practical problem. This is the third point. The statements endorsing the principle of local support quoted at the start of the paper were all pragmatic. They argued that we have to take the principle of local support seriously or lose that which we value as conservationists. But this stance means that the imperative to reform is lost should these values not be threatened by injustice. It would be possible for these statements to be interpreted to mean that we can worry less about injustice if it does not impede our goals. I doubt this is the authors' intention. The social injustices of conservation do not become tolerable if they cease to hinder conservation (Brown and Kothari 2002). We will have, therefore, to insist that these social injustices are addressed because they are unjust, not because they are inconvenient.
Fourth is the practical lesson from this discussion. For the politics of inclusion, participation and distributing benefits from conservation will be similar to the politics of distributing its costs and misfortunes. Both will consist, in Lonsdale's words, of attempts 'to deflect the costs . . . onto their fellows and to appropriate its benefits as their own' (Berman and Lonsdale 1992: 71). Community conservation is likely to consist of a myriad of marginalisations and inequalities enforced on smaller and smaller scales. Men and large animals might be included, but women and the products they value are left out (Sullivan 2000). Local government is empowered-but only to the district level and not to the ward level (Murombedzi 2001). Inequality, injustice and exclusion will be inherent in the solutions to largerscale injustices that protected areas have imposed. There is no way out of them (Murphree 2000). This is what participation entails (Mosse 1994). The task, therefore, is to understand who will win and who will lose from whatever solutions community conservation offers.
Acknowledgements: Thanks to two anonymous reviewers, Grazia Borrini-Feyerabend, Christo Fabricius, Adrian Phillips, Ed Barrow, Ashish Kothari and Sejal Worah for an interesting exchange that lead to this article, and particularly to Stephen Tuck for his advice, comments and suggestions; to the Royal African Society, the Biodiversity and Inequality Groups of the School of Geography and the Environment in Oxford, the Geography Department of Canterbury University, Christchurch, and the Governance Stream of the World Parks Congress in Durban for the opportunity to air these ideas.
|1||Adams, J.S. and T.O. McShane (1992), The Myth of Wild Africa: Conservation without Illusion. Berkeley: University of California Press.|
|2||Alexander, J. and J. McGregor (2000), 'Wildlife and Politics: CAMPFIRE in Zimbabwe', Development and Change, 31: 605-27.|
|3||Anderson, D.M. and R. Grove. (1987), 'Introduction: The Scramble for Eden: Past, Present and Future in African Conservation', in D.M. Anderson and R.Grove (eds), Conservation in Africa: People, Policies and Practice, pp. 1-12. Cambridge: Cambridge University Press.|
|4||Barrow, E. and C. Fabricius (2002), 'Do Rural People Really Benefit from Protected Areas: Rhetoric or Reality?' Parks, 12: 67-79.|
|5||Baumann, O. (1891), Usambara Und Seine Nachbargebiete. Berlin: Dietrich Reimer.|
|6||Behnke, R.H. and I. Scoones (1993), 'Rethinking Range Ecology: Implications for Rangeland Management in Africa', in R.H. Bhenke, I. Scoones and C. Kerven (eds), Range Ecology at Disequilibrium: New Models of Natural Variability and Pastoral Adaptation in African Savannas, pp. 1-30. London: ODI.|
|7||Beidelman, T. O. (1960), 'The Baraguyu', Tanganyika Notes and Records, 54: 245-78.|
|8||Bell, R. (1987), 'Conservation with a Human Face: Conflict and Reconciliation in African Land Use Planning', in D.M. Anderson and R. Grove (eds), Conservation in Africa, People, Policies and Practice, pp. 79-101. Cambridge: Cambridge University Press.|
|9||Berman, B. and L. Lonsdale (1992), Unhappy Valley: Conflict in Kenya and Africa. London: James Currey.|
|10||Bernsten, J. (1979), 'Pastoralism, Raiding and Prophets: Maasailand in the Nineteenth Century', Ph.D. thesis. Masidon: University of Wisconsin.|
|11||Borrini-Feyerabend, G., T. Banuri, T. Farvar, K. Miller and A. Phillips (2002), 'Indigenous and Local Communities and Protected Areas: Rethinking the Relationship', Parks, 12: 5-15. |
|12||Bourdieu, P. (1998), Acts of Resistance: Against the New Myths of Our Time. Cambridge: Polity Press.|
|13||Bourgois, P. (1995), In Search of Respect: Selling Crack in El Barrio. Cambridge: Cambridge University Press.|
|14||(2001), 'The Power of Violence in War and Peace. Post-Cold War Lessons from El Salvador', Ethnography, 2: 5-34.|
|15||Brockington, D. (1998), 'Land Loss and Livelihoods. The Effects of Eviction on Pastoralists Moved from the Mkomazi Game Reserve, Tanzania'. unpublished Ph.D. thesis. London: University College.|
|16||(1999), 'Conservation, Displacement and Livelihoods. The Consequences of the Eviction for Pastoralists Moved from the Mkomazi Game Reserve, Tanzania', Nomadic Peoples, 3: 74-96. |
|17||(2001), 'Women's Income and Livelihood Strategies of Dispossessed Pastoralists. The Case of Mkomazi Game Reserve', Human Ecology, 29: 307-38.|
|18||(2002), Fortress Conservation: The Preservation of the Mkomazi Game Reserve, Tanzania. Oxford: James Currey.|
|19||Brockington, D. and K. Homewood (2001), 'Degradation Debates and Data Deficiencies: The Case of the Mkomazi Game Reserve, Tanzania', Africa, 71: 449-80.|
|20||Brockington, D. and K.M. Homewood (1999), 'Pastoralism around Mkomazi: The Interaction of Conservation and Development', in M. Coe, N. McWilliam, G. Stone and M. Packer (eds), Mkomazi: The Ecology, Biodiversity and Conservation of a Tanzanian Savanna, pp. 513-30.|
|21||London: Royal Geographical Society (with the Institute of British Geographers). Brown, J. and A. Kothari (2002), 'Editorial', Parks, 12: 1-4.|
|22||Bruner, A.G., R.E. Gullison, R.E. Rice and G.A.B da Fonseca (2001), 'Effectiveness of Parks in Protecting Tropical Biodiversity', Science, 291: 125-28.|
|23||Canney, S. (2001), 'Satellite Mapping of Vegetation Change: Human Impact in an East African Semi-Arid Savanna', Ph.D. thesis. Oxford: University of Oxford.|
|24||Coe, M., N. McWilliam, G. Stone and M. Packer (1999), Mkomazi: The Ecology, Biodiversity and Conservation of a Tanzanian Savanna. London: Royal Geographical Society (with the Institute of British Geographers).|
|25||Collet, D. (1987), 'Pastoralists and Wildlife: Image and Reality in Kenya Maasailand', in D. Anderson and R. Grove (eds), Conservation in Africa: People, Policies and Practice, pp. 129-48. Cambridge: Cambridge University Press.|
|26||Cronon, W. (1995), 'The Trouble with Wilderness; or Getting Back to the Wrong Nature', in W. Cronon (ed.), Uncommon Ground: Rethinking the Human Place in Nature, pp. 69-90. New York: W.W. Norton.|
|27||Dahl, G. (1979), Suffering Grass: Subsistence and Society of Waso Borana. Stockholm: University of Stockholm, Department of Social Anthropology.|
|28||Dietz, T. (1987), Pastoralists in Dire Straits: Survival Strategies and External Interventions in a Semi-Arid Region at the Kenya/Uganda Border-Western Pokot, 1900-1986. Amsterdam: Instituut Voor Sociale Geografie, Universiteit van Amsterdam.|
|29||Eltringham, S.K. (1994), 'Can Wildlife Pay its Way?' Oryx, 28: 163-68.|
|30||(1997), 'Letter to the Editor', Tanzanian Affairs, Journal of the Britain-Tanzania Society, 58: 30.|
|31||Fratkin, E. and E.A. Roth (1990), 'Drought and Differentiation among Ariaal Pastoralists of Kenya', Human Ecology, 18: 385-401.|
|32||Fratkin, E. and K. Smith (1995), 'Women's Changing Economic Roles with Pastoral Sedentarisation: Varying Strategies in Alternate Rendille Communities', Human Ecology, 23: 433-54.|
|33||Galaty, J.G. (1993), 'Maasai Expansion and the New East African Pastoralism', in T. Spear and R.Waller (eds), Being Maasai. Ethnicity and Identity in East Africa, pp. 61-86. Oxford: James Currey.|
|34||Galaty, J.G. and P. Bonte (1991), Herders, Warriors and Traders: Pastoralism in Africa. Boulder, CO: Westview Press.|
|35||Grove, R. (1995), Green Imperialism: Colonial Expansion, Tropical Island Edens and the Origins of Environmentalism. Cambridge: Cambridge University Press.|
|36||Homewood, K.M. (1992), 'Development and the Ecology of Maasai Pastoralist Food and Nutrition', Ecology of Food and Nutrition, 29: 61-80.|
|37||Homewood, K.M. and D. Brockington (1999), 'Biodiversity, Conservation and Development', Global Ecology and Biogeography Letters, 8: 301-13.|
|38||Homewood, K.M., H. Kiwasila and D. Brockington (1997), 'Conservation with Development? The Case of Mkomazi, Tanzania', report submitted to the Department for International Development. DHD.|
|39||Horne, G. (1993), '"Myth" and the Making of "Malcolm X"', The American Historical Review, 98: 440-50.|
|40||Igoe, J. and D. Brockington (1999), 'Pastoral Land Tenure and Community Conservation: A Case Study from North-East Tanzania', IIED Pastoral Land Tenure Series, 11.|
|41|| Iliffe, J. (1995), Africans: The History of a Continent. Cambridge: Cambridge University Press. |
|42||Illius, A. and T. O'Connor (1999), 'On the Relevance of Non-Equilibrium Concepts to Arid and Semi-Arid Grazing Systems', Ecological Applications, 9: 798-813.|
|43||Jennings, C. (2003), 'They Called Themselves Iloikop: Rethinking Pastoralist History in Nineteenth Century East Africa', in T. Falola and C. Jennings (eds), Sources and Methods in African History: Spoken, Written and Unearthed, pp. 173-94. Rochester, NY: University of Rochester Press. |
|44||Johnson, D. and D.M. Anderson (1988), The Ecology of Survival. Boulder, CO: Westview Press. |
|45||Kiwasila, H. and K. Homewood (1999), 'Natural Resource Use by Reserve-Adjacent Farming Communities', in M. Coe, N. McWilliam, G. Stone and M. Packer (eds), Mkomazi: The Ecology, Biodiversity and Conservation of a Tanzanian Savanna, pp. 531-55. London: Royal Geogra phical Society (with the Institute of British Geographers).|
|46||Knight, M.H. and P. Morkel (1994), 'Assessment of the Proposed Mkomazi Rhino Sanctuary. Mkomazi Game Reserve, Tanzania'. IUCN African Rhino Specialist Group.|
|47||Koch, E. (1997), 'On the Bicornis of a Dilemma', Out There, November: 107-9.|
|48||Kramer, R., C. von Schaik and J. Johnson (1997), Last Stand: Protected Areas and the Defense of Tropical Biodiversity. Oxford: Oxford University Press.|
|49||Krapf, J.L. (1860), Travels, Researches and Missionary Labours during Eighteen Years of Residence in Eastern Africa. London: Trubner and Co.|
|50||Leader-Williams, N. and S.D. Albon (1988), 'Allocation of Resources for Conservation', Nature, 336: 533-35.|
|51||Lewis, C.S. (1961), An Experiment in Criticism. Cambridge, Canto: Cambridge University Press. |
|52||(1971), God in the Dock: Essays on Theology. London: Fount.|
|53||Marks, S. (1984), The Imperial Lion: Human Dimensions of Wildlife Management in Central Africa. Epping: Bowker.|
|54||Mbano, B.N.N. (1997), 'Foreward' in The Mkomaz/Umba Game Reserves General Management Plan. Dar es Salaam: Wildlife Division, The Ministry of Natural Resources and Tourism. |
|55||Mosse, D. (1994), 'Authority, Gender and Knowledge: Theoretical Reflections on the Practice of Participatory Rural Appraisal', Development and Change, 25: 497-526.|
|56||Murombedzi, J. (2001), 'Committees, Rights, Costs & Benefits: Natural Resource Stewardship & Community Benefits in Zimbabwe's CAMPFIRE Programme', in D. Hulme and M. Murphree (eds), African Wildlife and Livelihoods: The Promise and Performance of Community Conservation, pp. 244-55. Oxford: James Currey.|
|57||Murphree, M. (2000), 'Community-based Conservation: The New Myth?' paper presented at conference on 'Community-based Conservation: Old Ways, New Myths and Enduring Challenges'. Mweka, Tanzania: African Wildlife Management in the New Millennium, College of African Wildlife Management.|
|58||Mustaffa, K. (1997), Eviction of Pastoralists from the Mkomazi Game Reserve in Tanzania: An Historical Review', IIED Pastoral Land Tenure Series, 8.|
|59|| Neumann, R. (1998), Imposing Wilderness: Struggles Over Livelihood and Nature Preservation in Africa. Berkeley: University of California Press.|
|60||Oates, J.F. (1995), 'The Dangers of Conservation by Rural Development: A Case Study from the Forests of Nigeria', Oryx, 29: 115-22.|
|61||Peluso, N.L. (1993), 'Coercing Conservation? The Politics of State Resource Control', Global Environmental Change, June: 199-217.|
|62||Polanyi, K. (2001 ), The Great Transformation: The Political and Economic Origins of Our Time. Boston: Beacon Press Books.|
|63||Ramphal, S. (1993), 'Para Nosotros la Patria es el Planeta Tierra', in J. McNeely (ed.), Parks for Life: Report of the IVth World Congress on National Parks and Protected Areas, pp. 56-58. Gland: IUCN.|
|64||Rawls, J. (1971), A Theory of Justice. Oxford: Oxford University Press.|
|65||Rogers, P.J., D. Brockington, H. Kiwasila and K. Homewood (1999), 'Environmental Awareness and Conflict Genesis: People versus Parks in Mkomazi Game Reserve, Tanzania', in T. Granfelt (ed.), Managing the Globalized Environment: Local Strategies to Secure Livelihoods, pp. 26-51. London: IT Publications.|
|66||Roth, E.A. (1996), Traditional Pastoral Strategies in a Modern World: An Example from Northern Kenya', Human Organisation, 55: 219-24.|
|67||Schmidt-Soltau, K. (2003), 'Conservation-related Resettlement in Central Africa: Environmental and Social Risks', Development and Change, 34: 525-51.|
|68||Scott, J.C. (1985), Weapons of the Weak: Everyday forms of Peasant Resistance. New Haven: Yale University Press.|
|69||Smith, R.J., R.D.J. Muir, M.J. Walpole, A. Balmford and N. Leader-Williams (2003), 'Governance and the Loss of Biodiversity', Nature, 426: 67-70.|
|70||Spear, T. and R. Waller (1993), Being Maasai: Ethnicity and Identity in East Africa. Oxford: James Currey.|
|71||Spinage, C. (1998), 'Social Change and Conservation Misrepresentation in Africa', Oryx, 32: 265-76.|
|72||(1999a), 'A Reply to Colchster', Oryx, 33: 5-8.|
|73||(1999b), 'A Reply to Martin', Oryx, 33: 282-84.|
|74||(2002), 'A Reply to Adams and Hulme', Oryx, 36: 16-17.|
|75||Struhsaker, T. (1998), 'A Biologist's Perspective on the Role of Sustainable Harvest in Conservation', Conservation Biology, 12: 930-32.|
|76||Sullivan, S. (2000), 'Gender, Ethnographic Myths and Community-based Conservation in a Former Namibian "homeland"', in D. Hodgson (ed.), Rethinking Pastoralism in Africa: Gender, Culture and the Myth of the Patriarchal Pastoralist, pp. 142-64. Oxford: James Currey.|
|77||Sullivan, S. and R. Rohde (2002), 'On Non-equilibrium in Arid and Semi-Arid Grazing Systems', Journal of Biogeography, 29: 1595-1618.|
|78||Tuck, S.G.N. (2001), Beyond Atlanta: The Struggle for Racial Equality in Georgia 1940-1980. Atlanta: Georgia University Press.|
|79||Waller, R. (1979), 'The Lords of East Africa. The Maasai in the Mid-Nineteenth Century. 1840-85', Ph.D. thesis. Cambridge: Cambridge University.|
|80||Western, D. (1994), 'Ecosystem Conservation and Rural Development: The Case of Amboseli', in D. Western, R.M. Wright and S.C. Strum (eds), Natural Connections: Perspectives in Communitybased Conservation, pp. 15-52. Washington, DC: Island Press.|
|81||(2001), 'Taking the Broad View of Conservation: A Response to Adams and Hulme', Oryx, 35: 201-3.|