Knowledge, power, livelihoods and commons practices in Dwesa-Cwebe, South Africa

Abstract

This paper explores two aspects of the management of the commons in the Transkei area of Dwesa-Cwebe and traces them from the 1870s to today. The first is the intersection of power, politics and knowledge. National interests are privileged over local interests in the management of forests and grasslands that were controlled by local people before the onset of colonialism. This leads to the marginalisation of certain groups. The second is the economics of common resources. The paper contrasts communities' multiple livelihood strategies with the state's single strategy system. It looks at how state policies have changed people's status and their dependence on resources, and reflects on Dwesa-Cwebe's prospects of managing its commons now that local institutions have been undermined and livelihood patterns changed.

Keywords:

• Knowledge
• power
• livelihoods
• local people
• grasslands
• forests
• commons

1. INTRODUCTION

This paper is an exploratory review of the governance of the commons in the Dwesa-Cwebe area of the Transkei region of the Eastern Cape Province of South Africa, using secondary research materials. The case study traces the continuity and discontinuities in commons practices in villages surrounding the Dwesa-Cwebe Nature Reserve from the beginning of settler conquest in the region in the 1870s to today and explores how issues of knowledge, power and livelihoods have shaped the management of the commons here. The reserve contains a number of marine, grassland and forest resources that neighbouring people use for their livelihoods. The status of these, access to them, and their contributions have been shaped by different political regimes over a long period of time.

The paper examines the power relations involved in the management of these resources and how they marginalise the local communities. This is especially pertinent for exploring how national interests are pursued at the expense of local interests: national interests are abstract issues such as forest conservation, whereas the people themselves focus on the trees for construction materials or medicines. These divergent interests illustrate the different economic perspectives at play. The pursuit of a conservation strategy in the form of a protected nature reserve contrasts with the local people's multiple livelihood strategies for using a variety of natural resources. Within such divergent interests and economic perspectives, the power relations become evident in the way political relations combine with knowledge forms to structure natural resource management practices. The deployment of technical knowledge in contrast and opposition to local knowledge represents the use of power through the privileging of one form of knowledge over others in the management of resources. Further, this use of knowledge has to be situated within the broader colonial conquest project, which has nonetheless continued in post-apartheid South Africa through the pursuit of nature conservation at the expense of local livelihood systems around the Dwesa-Cwebe Nature Reserve.

The Transkei region was selected for this review because of its long history of communities struggling against outside knowledge and colonialism (Beinart, 2003; Tropp, 2003; Ntsebeza, 2006). It is a living example of how people's lives were reorganised under the colonial and apartheid political systems to stop what was seen as environmental degradation by local people, thereby subjugating their management systems. The paper is based on literature about the Dwesa-Cwebe Nature Reserve, which provides a historical profile of resource management since the nineteenth century, and the author's research in 2000 in the village of Hobeni (see Grundy et al., 2004). The following questions guided the review. How do power and political relations intersect in influencing knowledge about natural resource management (commons practices) and livelihood patterns? How have these relations evolved over time to their current status? What are the emerging issues? What is the future of commons management in the Dwesa-Cwebe Nature Reserve?

The rest of the paper is organised as follows. Section 2 discusses the conceptual issues that guided the review and Section 3 gives an overview of Dwesa-Cwebe and its governance. Section 4 discusses the way political relations and power have evolved in determining knowledge and livelihoods here since the nineteenth century. Section 5 describes local livelihoods in the nature reserve, and Section 6 concludes by considering the future of commons management in Dwesa-Cwebe in the light of past political interventions and changing livelihood patterns.

2. CONCEPTUAL ISSUES

The main concern of this paper is to show how power is deployed through political relations, which in turn determines how certain technical practices (knowledge) affect the livelihood systems of people in Dwesa-Cwebe. Foucault (1982:217) states that power is an ‘all-embracing and reifying term’ such that in order to understand it more meaningfully, what becomes pertinent is to study the multiple power relations context. Struggles over power relations are universal: they are found in multiple localities, transcending socio-economic and political boundaries. Such struggles are over the effects of knowledge (science and state control) as power, as is the case of people living around Dwesa-Cwebe Nature Reserve. In this discussion the main emphasis is placed on power in terms of knowledge and allied practices around natural resource management as well as the economics of all this (especially in relation to livelihoods), which are traced through the way forestry and grazing policies have been framed in the Dwesa-Cwebe area since the nineteenth century.

Scholarship on the effects of knowledge in furthering social and political control by the state−for example, Peet and Watts (1996) and Singh and van Houtum (2002)−has largely drawn from Foucault's ‘governmentalisation of the state’ and subject formation, with the main emphasis being on understanding how science, scientific disciplines and knowledge influence governance through the exercise of power (Foucault, 1982:221). A constant interplay between power and knowledge results in the expansion of both power and new disciplines of knowledge. The outcome is disciplinary knowledge (Foucault, 1982). Hill (1996), to illustrate this argument, traces the development of conservation policy in Zimbabwe and how it led to the consolidation of state power, through the expropriation of community-owned lands to make them protected areas (national parks and forests) and, in the post-colonial period, through decentralised natural resources management programmes. Hill (1996:106) contends that ‘the state uses conservation policies in much the same way it uses taxation, investment, interest rate or land resettlement policies: to establish and extend its own interests’. Along the same lines, the Dwesa-Cwebe case is selected to trace how the state since the 1870s has transformed the management of natural resources in the area and taken it away from communities through the introduction of various technical practices.

Another aspect of the Foucauldian perspective that is relevant here is the concept of subject formation, which is achieved through ‘technologies of power, which determine the conduct of individuals and submit them to certain ends or domination … and technologies of the self, which permit individuals to effect … a certain number of operations on their bodies and souls’ (Foucault, in Agrawal, 2005:165). As a result, certain ideas, knowledge and discourses are called upon and used in environmental contestations (see Robbins, 2004). Such contestations are characterised by relationships of domination and resistance. The focus of the state does not necessarily result in the desired outcomes (Agrawal, 2005), as those the state seeks to dominate also exercise their power in the form of resistance and opposition to state interventions.

The creation of Dwesa-Cwebe Reserve has always faced enormous resistance from the surrounding communities (Palmer et al., 2002). The nature of resistance often reflects the coercive nature of the state, in colonial and post-colonial settings. Hence the duality of the state political entity, which can be perceived ‘both as a modern power regulating the lives of citizens and as a despotic power that governed peasants as subjects’ (Mamdani, 1996:136). While Mamdani focuses on subjugation from a legalistic perspective through the creation of tribal authorities within the colonial state, a continuum exists between his ideas and subject formation in terms of disciplinary knowledge as argued by Foucault (1982). Singh and Van Houtum (2002) also explore the way expert or scientific knowledge is linked to the advancement of both state power and tribal elites (along the lines of Mamdani's decentralised despots). From a historical perspective, the authors provide insights into the intersection of forms of knowledge such as conservation and forestry, on the one hand, and political relations (State control) during colonial conquest and continued in the post-colonial era, on the other.

Scott (1998) sheds further light on this Foucauldian notion of power by identifying two main attributes of state policies that advance the state's control and reach. The first is the administrative ordering of nature and society through the production of landscapes through which states move and organise society. The second is ‘high modernism’, by which a state, through ideology, organises and justifies policies on the basis of scientific and technological legitimacy (Scott, 1998:4). In this second sense, the state pursues some preferred form of modernity that is based on ‘epistemic knowledge’ (Scott, 1998:340) or some rational principles that often lead to the destruction of local systems of knowledge. In some cases that Scott cites, such rationalisation resulted in social engineering experiments of the kind that South Africa witnessed with the ‘betterment programmes’ of the early to mid-twentieth century (Tropp, 2003). Both of these attributes are pertinent to the discussion presented in this paper about the way the colonial and post-colonial state operated in relation to grazing and forestry controls in the Dwesa-Cwebe Nature Reserve.

3. DWESA-CWEBE AND ITS MANAGEMENT

The Wild Coast region of the northeastern Eastern Cape is so called because of its largely undisturbed coastline and relatively undeveloped interior (Eastern Cape Parks Board [ECPB], 2007). Within this region is the Dwesa-Cwebe Nature Reserve, managed jointly by the ECPB and communities living around it. Dwesa-Cwebe has a complex environment comprising ecosystems that are considered rare and have high conservation relevance (ECPB, 2006). The Reserve is a 5700-hectare conservation area of unique biodiversity. It has coastal forests, the southern sour grassland and a Marine Protected Area. It also provides a habitat for the survival of a number of fish species. Dwesa-Cwebe is bordered on the one side by the Indian Ocean and on the other by the rugged grasslands of the former Transkei. It has a number of introduced game species including Red Hartebeest, Cape Buffalo, Eland, Blue Wildebeest, Burchell's Zebra, White Rhinoceros and Blesbok that roam the grasslands and forests, and crocodiles have been reintroduced into the rivers (ECPB, 2006).

Fay (2007:90) describes the people of the area as having two culturally distinct Xhosa-speaking groups, whose total population comes to around 15 000 people and who live in villages around the nature reserve, and makes the point that ‘Dwesa-Cwebe is a community formed through struggle’, as officially there is no such designated group or administrative entity within the area. In terms of institutional outlay for resource management, first there is the ECPB (on behalf of the state) that manages the reserve, including the Dwesa bungalows and campsite, through a resident manager. Second, there is the Amatole District Council (the local government authority) that manages all the land claim settlement funds and is the implementation authority for settlement. Third, there is the authority of traditional leaders (headmen and sub-headmen), which has not been easy to incorporate into democratic structures. They are also responsible for land allocation and they have a paramount role of controlling general resource use and access (Grundy et al., 2004). Fourth, there is a Dwesa-Cwebe Land Trust, which was set up in the 1990s as the legal entity that would own the land (from the claim over the nature reserve) on behalf of the communities, with a mandate for development as well as land ownership. As a result of the land claim settlement, communities around Dwesa-Cwebe also transformed their communal land into seven Community Property Associations (Palmer, 2003; Fay, 2007). Each Community Property Association has two representatives on the Trust, and these constitute the Trust. Community Property Associations are encouraged under the new democratic dispensation as a way to be more democratic and representative than traditional authority (Grundy et al., 2004; Ntsebeza, 2006).

4. KNOWLEDGE, POWER, RESOURCE CONTROL AND RIGHTS IN DWESA-CWEBE SINCE THE 1870s

In the Dwesa-Cwebe area the colonial government was driven by technocratic considerations that led to the marginalisation of local resource management practices, and the same is true of the post-colonial government. The state's science-based knowledge tends to be privileged over local knowledge systems for managing grazing, forest and marine resources. In the following, each of the historical periods is examined in the light of the paper's main focus on power and political relations and how they influence knowledge and livelihoods for the Dwesa-Cwebe area.

4.1 1870s–1930s

Before the advent of colonial settlers in the 1870s, there were three Xhosa-speaking groups in the Dwesa-Cwebe area who were engaged in various conflicts to assert their authority. The first Xhosa paramount chief who has been dated was Ngconde, who ruled from near the Mbashe River (the river bisects the nature reserve) from the mid-seventeenth century onwards (Palmer, 1998:3). Over the eighteenth century, under a different chief, the Xhosa moved westwards, thereby leaving the area to the east of the Mbashe River, which included the Cwebe forest, not belonging to any group. The second group, the Bomvana, who originated from Natal, needed a place of refuge, so their chief, Gambushe, purchased the unclaimed territory from Hintsa, the chief of the Gcalecka, the heirs of Chief Ngconde, who had the rights to the land but no use for it as they inhabited the area to the west of the Mbashe River (Palmer, 1998:3). Over the course of time, territory to the west of the Mbashe River, including the Dwesa part of the nature reserve, came to be regarded as Gcaleckaland, while the area to the east, including the Cwebe part of the reserve as it is today, came to be associated with the Bomvana (Palmer, 1998:3). The nature reserve's history of occupation became more complex with the arrival of a third group of people, the so-called Fingo (Mfengu). Historians say that the Fingo were refugees from Natal chiefdoms that had been defeated by Shaka, who fled southwards through the Transkei only to be dominated by the Gcalecka Xhosa (Beinart, 2003). The Fingo are said to have sided with the British in frontier wars with other Xhosa groups. They settled on the west bank of the Mbashe River in the Dwesa part of the nature reserve (Palmer, 1998).

During this time, resource control was under the chiefs and other tribal authorities. However, after the 1878 frontier war the whole of the Transkei region was eventually annexed by white settlers under the colonial administration, with the power of the chiefs being systematically broken by the direct rule of resident magistrates for each of the districts into which the whole region was divided (Palmer et al., 2002; cf. Mamdani, 1996). Local people still retained residency inside the future forest reserve, which at that time had not yet been demarcated for the state.

There was pressure on the colonial administrators to make colonies financially solvent and forests were prized for their valuable hardwoods, so the state intervened in the forest sector to raise revenue (Palmer et al., 2002). One of the earliest foresters in the then Cape Colony was Dr J S Henkel, who played a major role in state forestry in South Africa and Rhodesia. It was with his visit to the area in 1891 that an end to residency in the reserve was initiated in earnest. Forest residents were perceived to be a threat to the preservation of the forest and ‘a constant danger to the forest on account of grass fires and also very inconvenient’ (Henkel, 1891, cited in Palmer et al., 2002:52). Subsequently Henkel reported that by 1894 communities had been removed from the reserve after demarcation, but some families continued to live in it, with permits, until 1924 (Palmer et al., 2002). By 1904 the first commercial exploitation of sneezewood and yellow-wood had been made for export, with the first roads being opened in the 1920s.

During this early colonial period, management visions differed. On the one hand there was the conservationist Forest Department, which was inclined to set forests aside for the state and deny communities access to them, and on the other was the Native Affairs Department, which supported communities being allowed access to the reserve for their livelihood needs (Palmer et al., 2002). Even among the early foresters there was lack of clarity about whether the local people's use of trees for building would not encourage the growth of commercial hardwoods (Palmer et al., 2002). The denying of community access to timber in the nature reserve bears out Scott's (1998) point, mentioned above, about the state's administrative ordering of nature and society.

It was during this period that the Haven Hotel and cottages were built in the centre of the reserve to mark the beginning of tourism around Dwesa-Cwebe. While the owners of the tourist ventures were allowed to graze their livestock in the reserve, local people were officially not allowed to do so, although they carried on illegally. Outside the nature reserve, headmen's forests provided most of the timber needs of the local communities, but their governance too was constantly contested among local people, the Native Affairs Department and the Forest Department. On the agricultural side, local communities produced for the market in competition with a burgeoning white commercial agriculture. Because of this competition, measures were put in place to make it difficult for black agricultural producers to succeed (Palmer et al., 2002).

4.2 1930s–1994

Policies that were initiated in the previous period were put into practice more ruthlessly during this period. Between the late 1930s and 1981, communities around Dwesa-Cwebe lost residency inside the reserve, albeit with continued rights of access to forest products and grazing. After 1981 the Transkei region became a homeland area under apartheid, and communities' rights to resources in the reserve were completely denied. These two periods are discussed below.

With the nationalist government takeover in 1948 and the subsequent granting of homeland status to Transkei in 1976, following a period of partial self-rule from 1963, conservationists finally gained their objective of having the reserve's resources closed to the surrounding communities (Palmer et al., 2002). Following the establishment of ‘self-government’ in the Transkei, a Department of Agriculture and Forestry was established, which transformed the former demarcated forests into ‘forest reserves’ (Palmer et al., 2002). Overnight, people were barred from access to the reserves and they were forbidden to own tools for cutting and harvesting wood. Outside the reserves, restrictions on listed trees were enforced. Following the Nature Conservation Act of 1971, the Dwesa-Cwebe Nature Reserve was established in 1975. The reserve was fenced and stocked with a variety of wildlife (including alien species such as buffalo, hartebeest, wildebeest, rhinoceros and crocodile), which led to the total cessation of access rights to forests, grasslands and the seashore for local people (Palmer et al., 2002). This was then supported by a study from Dr E J Moll under the Wildlife Society, who recommended the ‘examination of the practice of forest resource harvesting, the implementation of controlled burning of the grasslands, the controlling of cattle grazing, and the discontinuation of shellfish harvesting to allow for recovery, after which limits should be placed on harvesting’ (Palmer et al., 2002:91). At the time, the surrounding communities were increasingly seen as a threat and the Transkei administration's preservationist attitude to the reserve made them impose an entrance fee of R1, to be levied on everyone entering the reserve, so as to deter locals.

Another state practice that illustrates the deployment of knowledge and power in resource management is the ‘betterment’ planning or rehabilitation, which, although it was launched in 1929, only came to be experienced in the Dwesa-Cwebe area in the 1980s (Palmer et al., 2002). The focus of the programme was to divide land into ‘discrete residential, arable and residential areas; the relocation of people from their dispersed homestead sites into villages; the fencing of residential areas and grazing camps …’ (Palmer et al., 2002:100). Under ‘betterment’, local people were forcibly removed from communities around the reserve following the closure of access to the resources in it. The ultimate result was the creation of a buffer zone around the nature reserve, and communities were moved even further away (Palmer et al., 2002). Betterment was premised on the scientific argument that if people were located in one part of the area and natural resources in another, this would enhance natural resource management and conservation. Headmen would be approached by staff from the Department of Agriculture and Forestry before areas were surveyed for conversion to different land uses. In the Hobeni village, entire sub-localities were converted into arable land after residents were relocated further away from the reserve. In many parts of the Transkei, the betterment scheme led to the destruction of the role of sub-headmen as an authority structure that controlled resources, as they were perceived to be part of a system that local people resisted and despised under apartheid (Fay, 2007). In effect, the local communities' capabilities for resource management were being severely undermined by political relations that not only weakened local institutions but also ensured that people would now resort to gaining access to resources by covert means.

In the early 1990s there was a drought period that adversely affected crops in non-irrigated fields and grazing for most communal areas. Livestock owners thus turned to grasslands inside the reserve that they had traditionally turned to in times of natural disasters such as drought (Palmer et al., 2002). While community leaders appealed to conservation authorities and traditional authorities to no avail, illegal grazing was carried on in the reserve. Subsequently, buffer zone communities sought to have their rights restored by the conservation authorities even after the drought period, for access to the reserve for cultural purposes and for access to natural resources.

With the democratic dispensation being established in April 1994, protests by the surrounding communities led to invasions of the nature reserve, resulting in the self-fulfilment of rights of access that had been denied for over a century (Fay, 2007). A land claim between the Dwesa-Cwebe Land Trust and the Eastern Cape Nature Conservation Department (now the ECPB) was eventually formally settled in June 2001 through the Department of Land Affairs. Land ownership was restored to the communities and a compensation package of R14 million was paid by the state through the Department of Land Affairs but administered through the local municipality (Transkei Land Service Organisation, n.d.). Of this, R9 million was invested in the development of local tourism infrastructure (Palmer et al., 2002). Since the opening up of the reserve, harvesting of forest resources continues to be regulated by the ECPB and Marine and Coastal Management, initially under conditions that will allow for the regeneration of trees and marine resources. However, harvesting of grass for thatching is permitted without restriction. Overall, since the formal handover of the land in 2001, very little appears to have changed in relation to resource access to the nature reserve. The main areas of frustration for the surrounding communities are primarily the severely limited access to various resources, the lack of implementation of the development plan (tourism was to drive development in the area), the lack of support for the land trust (to develop capacity for management) and the very limited progress on the Haven Hotel (Palmer, 2003; Fay, 2007; Transkei Land Service Organisation, n.d.).

This section has provided illustrations of Scott's (1998) thesis about the way the state imposes order on nature and society, and the way it justifies policies by appealing to scientific legitimacy. Since the late nineteenth century, the Dwesa-Cwebe nature reserve has witnessed various state policies through control of forestry, social engineering in the form of ‘betterment’ (Palmer et al., 2002) and, more recently, conservation strategies (including the setting aside of the reserve in the 1970s for conservation), all of which have systematically undermined local resource management practices or, at the very least, access to resources.

5. LIVELIHOODS AROUND DWESA-CWEBE NATURE RESERVE SINCE 1994

A number of studies have been conducted in the area focusing on the contribution of various commons resources to local residents' livelihoods in more recent times (Palmer et al., 2002; Timmermans, 2002; Andrew & Fox, 2004; Grundy et al., 2004; Shackleton et al., 2007). All of the studies point to the importance of natural resources, both inside and outside the nature reserve, to the livelihoods of the people in the villages.

In situating resource use around the reserve it helps to understand the contribution of agriculture to rural livelihoods in Transkei. In a study of a nearby Nompa Village on the west bank of the Shixini River, adjacent to the nature reserve, Andrew and Fox (2004) observe that people have not given up cultivation entirely, but instead they have changed their cultivation practices−adapting to garden cultivation that is ‘more productive, less risky, and more viable’ (Andrew & Fox, 2004), in the sense that they have been able to invest more labour, time and resources in gardens than in the main fields. Given the shortage of men among rural households and the decline in livestock and financial resources, this has been a more viable option. Another constraint identified by Andrew and Fox is loss of soil fertility in the main arable lands. In a study conducted between 2000 and 2002, speaking of the decline in the importance of agriculture to livelihoods in the Cwebe area, Timmermans also observes that the greater part of ‘household income originates from social grants and remittances’ and that more than 30 per cent of residents were earning less than R1500 per annum and 34 per cent between R5500 and R10 250, which meant that natural resources were very important in ‘maintaining local agricultural and residential infrastructure’ (n.d.:334). Forest resources provide some limited opportunities for income generation in the form of crafts, herbal medicines and the loan of labour or oxen for harvesting wood.

Focusing in more detail on the two villages of Ntubeni (south of the reserve along the coastline) and Cwebe (northern part of the reserve and inland), Shackleton et al. observe that:

there were a number of resources for which either there was well-developed trade in terms of a large proportion of households buying and selling, or that … only a few people sold a specific resource, but cash incomes from trade were high for those that pursued it as a primary livelihood strategy. (2007:151)

They point out that trade in resources has therefore allowed some households to ‘escape poverty’, with trade and use being more important for poorer households. Reeds and woven products contributed the majority of traded products across both villages (28.2 per cent of all households), followed by thatch grass (14.7 per cent). Table 1 presents an overview of the direct-use values to user households per year of a select group of resources, with fish added for comparative purposes.

In an earlier study, Timmermans (2002) had observed that the nature reserve, ‘pocket’ forests under the control of headmen, communal woodlands and other grazing areas outside the reserve were important for a variety of the natural resource products that communities used or traded, or both. With regard to restricted grazing inside the reserve, there are mixed views, with some people arguing that the former grasslands were experiencing bush encroachment due to long periods without livestock grazing (Fay, 2007; Transkei Land Service Organisation, n.d.).

Table 1: Summary of gross direct-use annual values per household

Resource	Ntubeni	Cwebe
Woven products	R774 (13.2%)	R2864 (59.5%)
Weaving reeds	R51.74 (84.2%)	R139.19 (89.5%)
Thatch grass	R70.15 (96%)	R105.59 (96%)
Grass brooms	R8.68 (N/A)	R13.03 (N/A)
Fish (largely marine)	R1641 (86.8%)	R3282 (10.8%)
Wild spinaches (imifino)	R348 (92.1%)	R508 (97.1%)
Source: Shackleton et al. (2007).

6. CONCLUSION

This paper has brought to the fore two issues of commons practices in the Transkei. The first is the impact of colonialism and state intervention on resource management and governance systems in the Transkei region. Since the advent of a settler regime in the late 1870s there has been systematic undermining of traditional authority, weakening it to such an extent that there are only a few pockets where it is effective in resource management. Perhaps with the new formations emerging under the Common Property Associations that the post-apartheid state is promoting, effective governance structures for natural resource management may develop.

The second issue is the interplay of knowledge and power in resource management. Since the early colonial period, scientific and outside knowledge have taken precedence over local forms of knowledge in the management of the forests and grasslands of the Dwesa-Cwebe Nature Reserve. There has been continuity in the subjugation of local people's way of managing resources since the area was made a reserve (Dwesa in 1891 and Cwebe in 1893, but since then managed as one unit – Dwesa-Cwebe). Local people's management practices have always been treated as inferior and inappropriate, with the result that areas such as the Dwesa-Cwebe reserve have been taken from them for preservation. Local people depend on a variety of resources at any given time and manage them more holistically than the conservation scientists who tend to segment them into forests or wildlife or grasslands or biodiversity. Local resource utilisation practices that were premised on tribal authority systems have been undermined by the state-imposed management of the reserve. Even with the supposed co-management set-up that has been in place since 2000, community participation is hardly factored into the management plans and decisions−as witnessed in the unilateral prohibitions of resource harvesting from time to time (Fay, 2007).

This paper illustrated the continued importance of selected resources for certain groups of people in order to situate livelihoods within the intersection of power, politics and knowledge. Overall, natural resources are of diminishing importance to rural people as more and more men get absorbed into the labour market elsewhere (Andrew & Fox, 2004; Shackleton et al., 2007; Timmermans, n.d.). The formal economy is changing livelihood patterns in the Transkei, leaving fewer people dependent on natural resources (cf. Du Toit & Neves, 2007). However, for the few, especially the poorer people, such resources are even more important for their livelihoods and in some cases in terms of income-generating opportunities.

Overall, it is hard to see a future for the commons in the former homelands of South Africa, such as Dwesa-Cwebe, as local institutions for resource governance have been undermined over time. But poorer households still depend significantly on the commons, including those of the nature reserve – reeds and grasses for mats and crafts, firewood, pasture for small stock – and policies need to be developed that will strengthen the management systems by taking into cognisance local practices.

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 (CROSCOG) Project No. 043982. This work does not reflect the Commission's views and in no way anticipates its future policy in this area. The author wishes to thank Professor Charlie Shackleton and the staff of the Department of Environmental Science, Rhodes University, for help with data collection; Professor Robin Palmer for help with locating materials; Dr Derrick Fay for insights about the research area; Dr Stephen Turner for helpful comments on earlier drafts of this article; colleagues on the CROSCOG project for their support; Dr Doug Wilson for his patience with the team; the Institute for Poverty, Land and Agrarian Studies, University of the Western Cape, for its support and encouragement; and the anonymous reviewers for their help in shaping this article.