Common pool resource management among San communities in Ngamiland, Botswana

Abstract

The contestation for common pool resources in Botswana pits powerful institutions such as national governments and international organisations against powerless local communities who continue to rely heavily on these resources despite all attempts to dislodge them. This paper explores how common pool resource management has shifted from locally based and people-centred endogenous resource management to state-defined and controlled forms. This shift has marginalised San communities, who have historically relied on their commons for survival. The paper is informed by case studies of two San villages in Ngamiland District in northwestern Botswana, Mababe and Phuduhudu, both situated adjacent to national parks. The government and donor agencies introduced community-based natural resource management with the promise of reversing the loss of the commons for such groups as the San; however, the evidence on the ground suggests the promise has not been translated into practice.

Keywords:

• Common pool resources
• community-based natural resource management
• marginalisation
• San
• Ngamiland
• Botswana

1. INTRODUCTION

This paper examines how common pool resource management in Botswana has shifted from locally defined ‘endogenous’ systems to devolved, state-led systems in which the state defines the conditions for management, with different implications for different groups depending on their socio-economic and political status and ethnicity. It presents the case of the two villages of Mababe and Phuduhudu in the south-eastern part of the Okavango Delta. Both villages were established in the 1980s by the Botswana government as congregation centres for nomadic hunter-gatherer San groups where they could be provided with social services (water, food rations and medical supplies, among other things), a process referred to in the rest of this paper as ‘villagisation’. Previously, groups in both communities had used the southern Okavango Delta, the Mababe Depression (inside Chobe National Park; see Figure 1), all the way down to the Nxai Pan and Makgadikgadi Pans National Parks (Parry & Campbell, 1992).

Figure 1: Map showing Ngamiland District, WMAs, controlled hunting areas and the locations of Mababe and Phuduhudu

As with other San groups throughout southern Africa, the San of Mababe and Phuduhudu have suffered several injustices, chief among these being the non-recognition by government of their land rights and land use practices (Magole, 2003). The land on which they relied for their livelihood was considered unoccupied and was designated national parks by the British colonial administration (Hitchcock, 1985b). The post-independence government did the same. Most San villages are located in state land Wildlife Management Areas (WMAs) (see Figure 1). As various organisations wish to redress past socio-economic injustices, most San communities have become candidates for community-based natural resource management (CBNRM) projects. In theory, CBNRM aims to devolve the management of the commons back to the people. The respective communities that participate in CBNRM projects have established community trusts that operate CBNRM initiatives in the form of hunting safaris with joint venture private companies (Thakadu, 2005).

Communal systems were blamed, under the paradigm of the ‘tragedy of the commons’, as insecure and not providing the right incentives to invest in land (Berkes, 1989). Often it remained unrecognised that common property systems represented a specific kind of management that was far from being open to all but did, on the other hand, distribute access to vital livelihood resources to local people (Peters, 1994). The policies of the post-independence African governments, dominated by local elites and supported by global institutions such as the World Bank, however, were all based on the premise that ‘customary’ systems did not provide the security of tenure needed to promote investment in land and natural resources (Narayan et al., 2000). The result has been the non-recognition of indigenous land tenure systems such as that of the San.

The commons is a topic of interest to theorists interested in dilemmas of ‘collective action’ and ‘public goods’ (Peters, 1994). More empirical studies of the commons have resulted from newer concerns about environmental decline and the role of communal resources and about the increasingly aggressive role taken by states, aided financially and conceptually by external donors like the World Bank, in promoting policies designed to convert systems of tenure from communal to private and state owned.

This paper explores the implications of the shift in common pool resource management through CBNRM from an explicitly people-centred and traditionally based system for the San, to one defined by the state and its agents. The author used secondary data from previous studies on the San generally, and also specifically on the San of Mababe and Phuduhudu villages, and data obtained from 1-day workshops of the two villages conducted by the author in April 2008 under the Cross-Sectoral Commons Governance in Southern Africa project. The paper further explores how the shift has affected the livelihoods of the San people and the social relations that have now developed as a result of CBNRM; in particular, local conflicts over land use and tenure. The paper traces how the land tenure and livelihood dynamics of the San interface over time with the prevalent themes of knowledge, political economy and power in local, regional and global contexts. The paper concludes by focusing on the current era of CBNRM that had the intention (at least in theory) of redressing some of the injustices suffered by the San and solving their common pool resource problems.

2. HISTORY OF THE SAN IN BOTSWANA

The San (also called Khoisan or BaSarwa) are the earliest inhabitants of Botswana (Hitchcock, 1982; Tlou, 1985). The Sotho/Tswana groups moved into the Botswana region in several waves some time after 500 AD, and thereafter there was continual populating of the area, with gradual changes in culture, language, physical type and economy (Tlou, 1985). While it was previously thought that the presence of modern hunter-gatherer San groups emerged as a result of pressures from the more highly organised Sotho/Tswana groups, archaeological research combined with historical and ethnographic information now suggests that Botswana has been occupied by hunter-gatherers for a substantial period (Hitchcock, 1982).

The San, however, have the distinction of being perceived and depicted as the most marginalised of all groups of people in Botswana (Bolaane, 2004). Like the Hadzabe of Tanzania and the Pygmies of Cameroon, oppression, discrimination and dispossession have marked the San people's history (Bolaane, 2004). Over the past two centuries, pressure on land from other ethnic farmers, conservationists and mining companies has cut-off the San from their traditional lands, and new occupants have been more successful in obtaining legally recognised ownership, a concept unknown in San culture (Hitchcock, 1985a). Their political representation has been either non-existent or very limited, the consequence of which has been the lack of San visibility in the political map of Botswana and southern Africa.

Apart from their lack of military strength as a group, the worst enemy of the San has been misconceptions of other ethnic groups (especially Tswana and European settlers) about the San lifestyle, livelihoods and land tenure rights. The combination of these factors is probably the basis for the historical injustices that have bedevilled the San people even in the dawn of the post-colonial African democratisation and human rights era. The history of southern African conservation areas is replete with tales of evictions and other injustices meted out by governing authorities to local communities living in areas that were later declared parks and reserves (Carruthers, 1989, 1995; Brockington, 2002; Bolaane, 2004).

The tale of the people of Mababe and Phuduhudu is no different. When the parks adjacent to their villages were created and subsequently expanded, the residents of both villages lost access to land and resources within these new protected areas that they had regularly used in the past. Similarly, when the adjacent areas were designated with new land use titles such as WMAs or Controlled Hunting Areas, the terms and conditions for using these areas changed to those defined by the state. Although their history of marginalisation from authority over land and other natural resources should give the San a high claim in common pool resource management, many rural communities have been similarly disadvantaged. Their situation thus in a sense highlights the dynamics of the status of the commons for rural communities more generally throughout southern Africa.

3. THE POSITION OF THE SAN IN THE POLITICAL ECONOMY OF BOTSWANA

Plagued by poverty and unemployment, the San face socio-economic exclusion (Bolaane, 2004; Hitchcock & Biesele, 2004; Magole & Magole, 2007). More than one-half of them live below the poverty line, largely because of high unemployment and the restrictions on what they can hunt, and:

are often arrested and harassed as ‘poachers’ … the government has limited what they can get from the land, and it is not enough to sustain life … Alcoholism is rife, and sometimes the camps lack access to basic sanitation, health care and even water. (Minorities at Risk Project, 2003)

In order to respond to the impoverished situation of the San, the Government of Botswana has since 1974 developed programmes aimed at providing them with socio-economic support. In 1974 the government established the Bushmen Development Programme, renamed the Basarwa Development Programme in 1975. In 1976 its name was changed yet again to the Extra-Rural Development Programme (Ministry of Local Government and Lands, 1995). All along, there was a growing reluctance on the part of the Botswana government, on the pretext that ‘all Batswana [citizens of Botswana] are indigenous to the country’ (Saugestad, 2001:52), to single out a sector of preference for rural development on ethnic grounds. Finally, in 1978, the programme was renamed the Remote Area Development Programme−the name it retains to this day.

During the Fifth National Development Plan (1981–1986), the government introduced the National Settlement Policy, which categorised settlements according to their populations. The Settlement Policy recognised small remote San settlements developed through the Remote Area Development Programme. These San settlements would otherwise not qualify as villages by virtue of having populations of fewer than 500 people. It was hoped the process would reduce squatting, prevent bush fires and reduce stock theft, all of which were associated with the San. On the other hand, some San sympathisers−who feared that communal land was being privatised into the hands of commercial livestock farmers−hoped that villagisation would give some land rights to the San. However, the settlement policy was built on San citizenship, not their ethnicity, and was open to all groups. The official policy became one of integration and assimilation into the Tswana society through a villagisation policy.

Pre-colonial land use management for Tswana groups in many areas of Botswana was based on a three-land settlement system (Peters, 1994). The system was based on the main economic activity being agriculture. Under the system, each household had access to land at three different levels that defined the settlement pattern consisting of a village, cultivation area and cattle posts (Magole, 2003). Although all members of the community had access to land and other land resources, the system was, as argued by Berkes (1989), about common pool resources, and far from being an open access system.

The San people, on the other hand, had a totally different concept of territoriality and land tenure. They lived in bands of extended families who moved around following wild game and collecting wild foods (Taylor, 2000; Hitchcock, 2002; Bolaane, 2004). Their roaming territories were large and there was little restriction on their mobility beyond what was imposed by terrain, seasons and disease pathogens (such as tsetse fly). They were therefore in constant movement and seasonal migration. The San of Mababe and Phuduhudu belong to the Ts'exa groups−that is, those associated with sandveld (Bolaane, 2004)−as opposed to River San. These San groups congregated at large settlements during summer, and dispersed as family units to various waterholes spread throughout the Mababe Depression and Nxai Pan areas during winter when water was scarce (see Figure 2). While this rotational movement by the San is understood today as having been ecologically in tune with the environment, the land use planners of yesteryear did not recognise the land use type and considered such areas as unoccupied, and as ideal candidates for designation as conservation areas (Magole, 2003).

Figure 2: Conservation areas of Botswana (national parks, game reserves and WMAs)

Throughout both the colonial and the post-independence eras in Botswana, the government has passed laws and policies to create and manage conservation areas. The implications of these legal instruments for local communities within and adjacent to the conservation areas has usually not been seriously considered, leading to the dynamics of marginalisation, misconceptions about land tenure rights and other factors mentioned earlier in this paper.

Table 1 summarises some of the legal instruments and policies for land tenure and management, and some of the major effects they had on San communities. The Tribal Grazing Land Policy has had the most impact on communal land resources in the recent past. The proposed system of land allocation under the Tribal Grazing Land Policy re-zoned the tribal grazing lands into three types: commercial, where groups of individuals were given exclusive rights (and leases) to specific areas and ranch development was encouraged; communal, where the traditional communal grazing system was not changed, but people were to be taught better management practices; and reserved areas, which would be unallocated land set aside for the future as safeguards for poorer members of the population (Peters, 1994). The third category of ‘unallocated land’ would later be converted into WMAs, through the Wildlife Conservation Policy of 1986 (see Figure 2). The primary land use in WMAs would be for commercial wildlife. The gazetting of these WMAs has been hotly contested by most stakeholders, including large and small cattle farmers (Magole, 2003). The rural communities whose settlements fell within WMAs also contested the WMAs, largely because of the restrictive regulations that came with their gazetting.

Table 1: Legal instruments and policies of land management and their effect on the San

Year	Legal instrument / policy	Proclamation	Main effect on the San
1933	Convention for the Protection of Fauna and Flora	Establishment of game reserves by the colonial administration throughout Africa	Evictions, denial of access to reserve areas
1961	Fauna Conservation Act	Establishment of game reserves in Botswana and anti-poaching laws	Evictions, denial of access, arrests, imprisonment for poaching
1967	National Parks Act	Establishment of National Parks	Evictions, denial of access, arrests, imprisonment for poaching
1968	Tribal Land Act	Establishment of Land Boards to manage tribal land and remove it from dikgosi	Exclusion as a tribe as Land Boards were based on tribal territories
1975	Tribal Grazing Land Policy	Establishment of privately owned ranches in communal land and establishment of WMAs	Reduction of access to some resources
1986	Wildlife Conservation Policy	Promotion of commercial wildlife uses in WMAs	Exclusion from commercial wildlife use that favoured large hunting companies
1988	Integrated Hunting Regulations	Hunting licences and hunting season	Issuing of special game licences
1992	Wildlife and National Parks Act	Expansion of parks and gazetting of WMAs	Further reduction of common pool resources as parks were expanded
2002	Draft regulations for WMAs	Centralisation of natural resource management in WMAs on government	Further reduction of access to livelihood resources
2007	CBNRM Policy	Formalisation of a de facto CBNRM	Abolition of special game licences

The San's hunter-gathering livelihood system is heavily dependent on access to wildlife (Taylor, 2000). However, over the years there has been a reduction in access to wildlife, due not to any decline in wildlife numbers but rather to a combination of factors, including the reduction of legally and physically accessible areas for the practice to continue. Another factor was the licence quota system introduced through the integrated hunting regulations of 1988 (see Table 1). Before this, subsistence hunting by the San had not been regulated. Now there were three restrictions. Firstly, although the new system did not prohibit the San from hunting, the regulations now required them to obtain a special game licence−to be caught hunting without a licence was a punishable offence. Secondly, the new regulations were based on a quota system that allowed licence holders to hunt only certain species and prescribed the number of animals to be hunted. Thirdly, only certain hunting methods were permitted; methods the state considered cruel were banned. The new system also opened up hunting grounds to all citizens throughout the country, completely disregarding the costs borne by those who lived closest to the hunted animals.

Remote areas dwellers – who were mainly San and mostly reliant on wildlife – complained bitterly that they had no firearms for hunting (Petje, 1988). Their traditional methods were outlawed and made illegal and the permissible methods were ineffective, while ‘outsiders’ were allowed to use vehicles and rifles to hunt in their area. Outlawing the most effective traditional methods of killing wildlife was equivalent to preventing legal access to wildlife resources. Opening up access to the greater citizenry did not translate into equitable access, as the less powerful local actors were typically excluded. The traditional methods, including wire snaring, were outlawed because of lobbying and pressure from the animal rights movements and the anti-hunting lobby groups, mainly from western countries.

Parry and Campbell (1992) record that the majority of the people of the Mababe Depression felt that the professional hunting companies and the Department of Wildlife and National Parks (DWNP) were now the major users of wildlife in their area. These authors argue that although the poor and less-educated people have a significantly higher need for wildlife meat than the more well-off, since their choices are more limited, at the same time they undervalue the potential benefits of wildlife, because they perceive only its utilitarian value and the dangers it poses to them, their livestock and their crops.

People from both Mababe and Phuduhudu said that their hunting in the past allowed for the selective removal of problem animals such as lion, hyena and elephant, so that their depredations were formerly fewer. They complained that the damage and danger from wildlife were getting worse, yet the control of problem animals provided by the DWNP was not increasing to meet demand. Parry and Campbell (1992) argue that the efforts put into control of problem animals by the DWNP did not relate to the amount of damage, but rather to the political power of the community and the DWNP perception of the land use status of the area. For instance, in the Mababe area, little problem animal control activity was carried out because the community is politically weak and the area has been zoned for wildlife development, as per the WMA regulations (Ministry of Trade, Industry, Wildlife and Tourism, 2002).

On the other hand, the San communities seem to have resigned themselves to being poachers liable to arrest by the Wildlife officers. Amongst the community members from Mababe and Phuduhudu, people who thought the DWNP was doing its job well said it did this by arresting poachers. Thakadu (1997) found that rural communities see wildlife authorities primarily as a law-enforcement agency serving the needs of the international tourists, safari companies and the global community that does not have to live day by day in close proximity to problem animals.

With the gazetting of WMAs in the Wildlife and National Parks Act of 1992 came the WMA regulations that tightened the state's grip on the San commons. Under the WMA regulations, the Director of the DWNP was given sweeping powers to grant or deny the use of resources within the WMAs. These regulations were hotly contested throughout the country, especially in the western region where cattle barons saw them as taking away their grazing. The CBNRM policy purported to redress some of the imbalances caused by the previous policies and laws (Taylor, 2007; Ministry of Environment, Wildlife and Tourism, 2007). In particular it promised to bring back to the local communities the benefits of wildlife to compensate for the costs they incurred by living adjacent to it, albeit on terms defined by the state. These conditions included communities organising themselves into community trusts, forming joint ventures with Safari Companies and accounting to government as to how they used the proceeds from the venture. The CBNRM policy goes further to demand that 65 per cent of the proceeds from the CBNRM projects be deposited into government coffers.

The above narrative is consistent with the five dimensions of power articulated by Boyce (2002). Inequalities in the dimensions of power play a role in the political economy of resource management. The identities of winners and losers generate the prediction that the distribution of negative impacts or costs will be correlated with other power-related variables such as income, race and ethnicity.

4. THE SAN TRADITIONAL KNOWLEDGE AND SKILLS

The San operate within traditional consensual decision-making processes (Taylor, 2000) that recognise the roles played by members of their community (men, women, boys, girls, the elderly, etc.). The traditional survival skills practised by the San include identifying and tracking and hunting wild animals, gathering and preparing bush foods, producing crafts, identifying and applying medicinal plants and healing individuals and communities through dances (Taylor, 2000; Hitchcock, 2002; Moses & Thoma, 2002; Magole & Magole, 2007).

Hunting is traditionally the domain of men and boys. It includes the performance of male rituals, and the transfer of skills from father to son, such as identifying animals, their tracks and mannerisms, and managing natural resources and interpreting seasons, weather and other conditions in the wild. The skills of a San hunter are so profoundly sensitised that he is able to tell from an animal's tracks whether it is male or female, young or old, weak or strong (Moses & Thoma, 2002).

During the colonial period, many hunting expeditions by European hunters used San men as guides and trackers in their hunting expeditions (Bolaane, 2004) within the ‘territories’ of the San. In those days there would be no negotiation over access to the ‘hunting grounds’ inhabited by the San, nor would there be monetary payments for the guiding services rendered, but instead the San guides and trackers would be paid with meat – especially those parts not of interest to the hunting party, such as the stomach and intestines (Bolaane, 2004). The dominant BaTawana in Ngamiland also engaged San hunters from Mababe and Phuduhudu in their hunting expeditions as their subjects (Tlou, 1985). In addition, San hunters were required by law to pay tribute to the BaTawana crown in the form of certain animals or parts thereof (Bolaane, 2004). The Mababe San hunters assisted hunting expeditions that used the areas known as the Chobe Triangle and the Mababe Depression, which have since become part of Chobe National Park. The Phuduhudu San hunters assisted hunting expeditions that used the area south of Nxai Pan National Park, called Bushman Pits, which has also become part of Nxai Pan National Park.

Many hunting safaris operating in Botswana have continued hiring the guiding and tracking services of San hunters even to this day (Magole & Magole, 2008). The South African Defence Force fighting South West African Peoples Organisation (SWAPO) guerrillas in then South West Africa (Namibia) also enlisted the services of San men for tracking the guerrillas (Ellis, 1994). Later, in post-independence Botswana, the Department of Wildlife and National Parks Anti-Poaching Unit would also enlist the services of San trackers, to combat poaching, especially cross-border poaching in the Chobe National Park. Whilst these institutions recognised the San hunters' invaluable skills and knowledge of the bush, their services either were not remunerated at all or were compensated meagrely or in kind.

San women and children are chiefly in charge of gathering bush foods (Moses & Thoma, 2002), and they often act as informants for the hunters about the game in the vicinity. The bush foods they gather have to be prepared to make them digestible, and care must be taken to ensure they are not poisonous. This knowledge could be invaluable to nutritionists and researchers if they were only aware of it.

In the treatment of disease, the San most often use their healing dances. These are performed under the guidance of a healer who has special powers, and are for curing not only individual illness but also community ills. Traditional San doctors have a vast knowledge of medicinal plants and their correct preparation and application in treating a range of diseases, and have remained popular amongst Tswana groups.

However, in the eyes of society at large the San knowledge and skills do not constitute professional knowledge (Moses & Thoma, 2002). Even in wildlife-based CBNRM projects, the ecological knowledge of the San is not prized. In fact, both the colonial administration and Botswana society, in particular the Tswana groups, have exploited San knowledge and labour for free or at minimal cost. In modern-day Botswana, the San are commonly engaged as cattle herders because they know the bush. The tourism sector, in particular the hunting fraternity, engage San men as guides, trackers and skinners. Again their labour is not compensated adequately. The transmission of knowledge through channels of unequal power relations leads to reinterpretation and renegotiation of the meanings to be ascribed to practices. While the rich ecological knowledge and survival skills of the San are acknowledged and indeed used for commercial gain, the unequal power relations between the San and other groups has limited the extent to which the San can negotiate and defend their livelihoods. Instead, the position and knowledge of other groups has strengthened, thanks to the San.

5. EVICTIONS – EVIDENCE OF THE POWERLESSNESS OF THE SAN

The earth is the home and common heritage of all humankind, but some people claim more of its bounty than others. Access to natural capital is filtered through our political and economic institutions (Twyman, 1998). Those people who are relatively wealthy and powerful generally reap more of the benefits from uses of the environment, and bear fewer of the costs from its abuse, than do those who are relatively poor and powerless (Boyce, 2002). This power dynamic creates winners who reap the benefits and losers who bear the costs.

The unequal power relations between San communities and other political and socio-economic groups have been demonstrated at various levels. Evictions of indigenous people from wildlife reserves or areas that were targeted to become parks were rife during the colonial administration, and have continued in the post-independence eras of many African states. Carruthers (1989:188–9, 1995:43), for example, describes how this happened during the 1930s and 1940s in the Kruger National Park, South Africa; and Brockington (2002) discusses the removal of indigenous residents from Tanzanian parks.

The Central Kgalagadi Game Reserve (CKGR) was an exception. Here the resident San group was allowed to continue living their traditional hunter-gathering lifestyles in the reserve (Silberbauer, 1965; Good, 1992). However, the post-independence government of Botswana later overturned this decision and also evicted the San group from the reserve. The establishment of all the other parks in Botswana (Chobe, Moremi, Nxai Pan, Makgadikgadi Pans, and Gemsbok, now Kalahari Transfrontier) involved the eviction of local people. The San of Khwai village were evicted from Moremi in 1963 (Bolaane, 2004), the San of Mababe from Chobe (Magole, 2007), and the San of Phuduhudu from Nxai Pan. These evictions meant that San activities such as hunting and gathering would not be allowed in these new parks. Subsequently the parks were expanded, infringing further on the San groups' livelihood areas.

While both the Mababe and Phuduhudu communities have been roped into CBNRM by the government, there has not been a visible upward transformation of the socio-economic and political status of the people. Unemployment levels remain high in these villages, coupled with low levels of education and poverty (Magole & Magole, 2008). Even more worrisome is that the San, more than any other ethnic group, generally remain likely candidates for any future eviction (Hitchcock, 1985a). The real powerlessness of the San in the face of evictions was epitomised in the Botswana High Court case of Roy Sesana & 254 Basarwa vs. the Government of Botswana (Nkala & Piet, 2006). In this milestone case, which attracted international attention, the San of the CKGR, with the support of Survival International, took the Botswana government to court on three counts: forced removal from the CKGR, the termination of special game licences that allowed them to hunt game animals using traditional methods and equipment, and the termination of social services for those who defied the eviction orders. The judgement delivered by three judges favoured the complainants in two of the three counts, saying that the San were entitled to special game licences and that their removal from the CKGR was unlawful (Nkala & Piet, 2006). As a result of the judgement, some of the concerned San people relocated back into the CKGR but the government refused to reinstate the social services for them, tactically ensuring that their return into the reserve remained untenable without the support of the social services that are afforded to all other citizens.

6. CONCLUSION

The San believe that their cultural practices form the backbone of a healthy and socially intact community (Taylor, 2000). This is evident from their resilience in managing to remain coherent communities despite a long history of marginalisation. Injustices such as land and resource dispossession have been so disruptive that the affected communities have often been unable to uphold their traditional consensual decision-making processes. Although the San people's traditional hunting and associated practices have been banned, they still depend heavily on the commons for their livelihood.

CBNRM has allowed group boundaries for remote local communities to become more exclusively defined. At the same time, economically and politically empowered outsiders who seek to broaden their resource use base are permitted by national laws to access resources without bearing the costs of living adjacent to them. This they achieve by claiming to be citizens and thus entitled to use the country's resources because they come under a widely inclusive definition of users. While this argument holds in theory, in practice it serves only those with the means and power to access resources throughout the country and limit the poor to their localities. This logic has been to the fore when national elites have resisted CBNRM in Botswana (Magole, 2003), especially in the Okavango Delta. They define the Delta ecosystem as a national resource to which they as citizens have rights, and deny that the local people have stronger rights to manage or profit from it. This effectively legitimises the marginalisation of poor and weak groups by the wealthy and powerful.

There is a structural conflict of interest here between the rural poor wanting land rights and the other citizens (elites) with their own interests in the common pool resources, which cannot be solved by a public policy of free settlement for all citizens (Swatuk, 2004). Efforts to provide a sustainable livelihood and increase secure access to land for the San have largely failed. The Botswana government has to acknowledge that poverty in Botswana has ethnic dimensions that cannot be ignored. It needs to invest resources in inter-ethnic relationship-building, promote increased access to natural resources for the San communities, and invest in community rights education. CBNRM has the potential to offer San people an opportunity to regain some lost rights to manage resources on their land. However, this may only be possible if the power imbalances resulting from a long history of marginalisation are acknowledged, addressed and resolved. In its current format, CBNRM stands to benefit the government much more than the people it claims to support.

This special issue was produced with the support of the European Union's Sixth Framework programme through the Cross-Sectoral Commons Governance in Southern Africa Project No. 043982. This work does not reflect the Commission's views and in no way anticipates its future policy in this area.

Notes