2. SETTING THE SCENE

Mozambique became an independent state in 1975 after an armed struggle waged against the Portuguese colonial state that lasted over 10 years. In 1977, the ruling FRELIMO party declared itself a Marxist Leninist Party and embarked upon a modernization campaign that exhibited the classic characteristics of a Marxist model of development. The legacy left to Mozambique at the time was characterized by extreme underdevelopment of most rural areas and an externally focused economy that was dominated by the provision of transport routes, tourism services and migrant labour to neighbouring countries in the region. FRELIMO sought to sweep away this legacy by taking command of the economy, setting production quotas for all sectors and running the former colonial farms and plantations as large scale state enterprises.

The ensuing period, between 1977 and 1987, was an “intense but brief” phase of building a centrally-planned economy. As a result of the relative brevity of this period, the rural economy remained pretty much in its pre-independence state (Wuyts, 2001). Although drastic changes took place in the ownership of land (with state farms replacing colonial plantations and settler farms), the conditions under which most of the rural population operated remained the same, particularly in relation to the relative importance to their livelihoods of off-farm incomes and subsistence production.

The approach taken during the era of central planning, which treated the rural population as a homogenous group of subsistence producers operating outside of the cash economy, failed to capture the complexity of rural peoples’ livelihoods. Remittances from migrant, seasonal and casual labour occupied (and continue to do so) a significant part of the livelihoods of many of the rural population, albeit to different extents in different parts of the country.^[1] The model of a homogenous peasantry remains dominant, however, in the post central-planning phase of reform and reconstruction, though it now comprises a mass of “smallholder producers” rather than “subsistence producers”. The existence of this homogenous peasantry is largely a fiction; on the contrary, the picture is highly heterogeneous. Livelihoods are constructed from various survival strategies and diverse forms of “income-gathering” (Tanner, 1986). The difficulty of analysing rural livelihoods, understanding the characteristics of the rural poor and the mechanisms that may be needed to improve their conditions lie precisely in this complexity (Pontera, 1999).

This paper shows how most rural livelihoods in Mozambique, while highly dependent on access to natural resources, are not constructed on the “narrow” use of land for cultivation. It examines the extent to which new natural resource policies in Mozambique have taken cognisance of this diversity and looks at some of the initial evidence of the impact of these policies. It advances the proposition that some of the initial gains made from policy developments in the land sector, which were cognisant of livelihood diversity amongst the rural poor in Mozambique, may now be being rolled back by more narrowly constructed policy narratives in respect to the forms of access and rights frameworks of other natural resources and other sector legislation.

2.1 Rural livelihoods in context

Over the past 30 years Mozambique has suffered long periods of disruption as a result of the war, severe drought and extensive flooding. As a result it remains one of the poorest countries in the world. It has a Human Development Index of 170th of 173 countries globally in 2001, with a life expectancy at birth of 39 years on average and literacy rates of 44 percent. Those living on less than US$ 1 per day form 38 percent of the population and almost 80 percent survive on less than US$ 2 per day (UNDP, 2003). The spatial distribution of poverty is not uniform: on aggregate, poverty is more predominant in the centre of the country and less so in the north and south and there are significant differences between provinces.

Mozambique remains heavily dependent on the international donor community with the net disbursement of ODA in 2001 totalling US$934.8 million (US$51 per capita).

The vulnerability context of rural livelihoods

Major changes have occurred in the context in which rural livelihoods have been constructed in Mozambique over the last 20 years, many of which have enhanced the vulnerability of the majority of the rural population. These changes have impacted upon the natural, physical and social capital that is available to the rural poor. They include the widespread destruction of the war, the introduction of economic structural adjustment policies, the advent of market forces and a liberalized economy at the end of the war, the increasing onset of a severe HIV/AIDS epidemic, prolonged periods of drought and more recently, devastating floods and climatic events.

Smallholder farmers face a range of hazards that pose a threat to their productivity and farm-based livelihood strategies. These hazards include declining soil fertility through the practise of slash and burn agriculture. There are significant threats to crops and livestock from disease, insect infestations, weeds and storage pests.^[2] Cattle and goats were decimated by the war, and as populations recover, there are concerns about the capacity of veterinary support services. External shocks such as global commodity price volatility and the continuing shrinkage in migrant labour opportunities in South Africa have the potential to significantly reduce rural incomes.

Other threats include flood and drought. The most vulnerable areas to drought are those in the central and southern parts of the country (Maputo, Gaza, Inhambane, northern Manica and southern Tete). It is estimated that around 60 percent of the country has a probability higher than 30 percent of drought occurrence. Flood risk affects over 1.7 million hectares at an altitude of less than 20 meters above sea level and within 10 kilometres from the principal hydro-basins. The areas at most risk of flood are those around the Zambezi, Pungue and Buzi rivers, and, to a lesser extent, those surrounding the Limpopo and Incomati [GOM 1998a].

Mozambique’s agricultural sector has suffered from several changes in direction since independence, as well as the massive disruption from the war. At the end of the war, the severe lack of basic services and rural infrastructure was a binding constraint on agricultural growth. Many rural areas with large populations and considerable agricultural potential continue today to have extremely difficult access to the national road network. The IMF review of implementation of the PARPA^[3] in 2003 stated that the “nationwide coverage of transitable highways is still fragile, however, which has discouraged private-sector investment and slowed the development of rural markets for agricultural inputs and products” (IMF, 2003).

Rural trading is therefore beset with problems of transport availability, at costs that make Mozambican trading comparatively disadvantaged. A number of studies have identified market access and prices as the most important determinant for agricultural production. Physical capital in the form of the network of small stores (cantinas) that existed during the colonial period has been decimated - these used to offer the option of bartering agricultural produce for consumer goods and agricultural inputs, and provided an important bulking-up function. They may also have offered small-scale production or consumption credit to local people. Now, many farmers have to travel long-distances to local markets where their bargaining position is weak.

Since the mid-1990s state farms land was distributed to private enterprises and, to a much lesser extent, smallholders. Many of the state-operated farms near to the major population centres were occupied during the war by displaced people, often with the permission and encouragement of the authorities. These land holdings have been returned to the now privatised companies that formerly operated them or to new entrants to the scene, leading to the loss of land, trees and other resources that had formed a major part of rural peoples’ war-time livelihood strategies. Many of the companies have little capital and investment potential, however, and the land holdings in many cases are being held for speculative rather than productive reasons.

HIV prevalence rates were estimated at 12.2 percent in 2000, with the highest rates in the Central region (INE, 2002). The impact over the coming years is likely to enhance the vulnerability levels of many households, as families are affected by illness. Some families are likely to lose land, or to get much less benefit from the land than was assumed. The burden of care for family members who fall ill will largely be placed upon female members of the household, who are also predominantly those involved in agricultural production. Family savings will be consumed and assets sold to help pay for medical expenses. Farm-land utilization is likely to decline as the inputs become unaffordable and the household labour supply is reduced (Drimie, 2002). Land rights of women may also become increasingly insecure under patriarchal customary tenure systems (which predominate in Mozambique, despite their nominal unconstitutionality in this respect); widows may be required to return to their own families and lose the land that they had access to whilst married.

It is noticeable that the higher rates of infection also coincide with those areas where there is most competition for land; the major international transport corridors and the peri-urban areas (ibid). In addition, AIDS has already begun to affect the people working within land and natural resource administrations and related institutions as well as those involved in the supply of essential goods and services or those that provide markets. The implications for institutional capacity to carry out functions will be several: impacts in terms of productivity, on finances and on human resources.

Natural resource use as part of livelihood strategies

In some cases, like the 200 families of the Djabula village in Matutuine District, near Maputo, forest products represent 93 percent of their average yearly income of around 500 US dollars per family (63 percent from charcoal, 30 percent from traditional drinks made from the palm Hyphaene sp) (Pereira & Cossa, 2001).

Mozambique is essentially an agriculture-based economy^[4] and the contribution of agriculture, at almost 28 percent of GDP in 1999, is believed to be substantially underestimated in official statistics. More than 75 percent of the population is employed in the agricultural sector^[5]. It is estimated that of the total land area of 78.6 million ha, about 46 percent (36 million hectares) are considered suitable for arable use. However, in 1996, only some 3.4 million ha, or about 10 percent of the arable land, were estimated to be cultivated (ADB, 1996).

The war resulted in increased pressure on land near major towns of the coastal zones and safe rural areas, but out-migration to areas of origin has since occurred. With improvements to rural security and the tertiary road network has come increasing clearance of land for cultivation. Of the total area cultivated, it is estimated that about 90 percent is under production systems of the so-called “family sector”. The remainder is used by other agents: agribusiness firms (particularly sugar, tea and cotton) state/private joint ventures, cooperatives and private individual farmers. Most agriculture is rain-fed, and in the southern and central provinces of Mozambique rain falls erratically and most of it only during a short period of the year.

The use of natural resources in livelihood strategies is not limited to agriculture and the full-time cultivation of land, i.e. the low level of land that is cultivated paints a false picture of the actual area that is important to the rural poor. Other natural resources are collected, processed and/or marketed by many families, either as a predominant activity or as part of a diversified portfolio of livelihood strategies designed to spread and minimize specific risks. These include resources such as bush meat, honey, clay, roots and tubers, medicinal plants, building materials, thatching grass, firewood and the production of charcoal and salt. Forests, in particular, provide a range of resources central to peoples’ livelihoods. The majority of the population remains in settlements dispersed widely throughout the country’s forests. This is in marked contrast to other African countries, where colonial policy had been to relocate people from the forests to roadside communities. The effect in Mozambique is that a much higher percentage of the population lives in isolated forest communities that are directly dependent on access to surrounding forest resources and the health of forest ecosystems for survival. Mozambicans rely on woody biomass, for example, for nearly 85 percent of their total energy consumption.

Moreover, most of the population lives within a 40 kilometre wide coastal strip extending over 2,000 kms, making artisanal coastal fishing a major livelihood strategy for a huge part of the population. The main freshwater fishery in central and northern Mozambique is Lago Niassa (Lake Malawi). There are also a number of major rivers (e.g. Rovuma, Zambezi) which also host significant fisheries resources (although in some cases these are in decline due to destructive fishing techniques), and fishing in smaller rivers and water sources is also prevalent. The Sofala Bank (a wide extension of the continental shelf fed by the Zambezi delta) extends from Beira in Sofala along Zambézia to Nampula in the north. This fishery is rich in prawns. Inland fishing in rivers and lakes and the production of dried fish will very often be part of rural people’s subsistence or informal commercial activities (SLSA, 2003). It is estimated that direct employment in this sector provides for up to 80,000 jobs.

In contrast to other countries in the region, there are far fewer other opportunities for cash-earning activities and consequently a much higher reliance upon agriculture, alongside hunting, fishing and the gathering of other natural resources for sale (ibid). Although remittances from migrant, seasonal and casual labour are important elements of many livelihood portfolios^[6], many of these opportunities are in turn linked to the provision of agricultural labour (except in the case of migrants to the mines of South Africa, which is a phenomenon that is decreasing as the labour demands from this sector contract). In many rural areas the only opportunities for the provision of paid labour are through the system known as ganho-ganho, where services, usually for the heavy work of clearing previously uncultivated land, are provided to families that are able to provide remuneration in the form of consumer items unavailable on the local markets, food, beer and, less commonly, cash. It has been estimated that less than 20 percent of rural households use hired labour (Cramer & Pontara, 1997).

Even those livelihood strategies that include involvement in local crafts and trades, which in some areas have assumed an international dimension, are heavily based on access to natural resources. Charcoal and salt production, basket and mat making, beer and spirit production, carpentry, pottery and blacksmithing all rely upon local natural resource availability. Natural resources also play a major part in the coping strategies that people adopt during times of crisis or shocks.

2.2 Background to policy development processes and an overview of existing policy and institutions

Constitutional principles of importance to natural resource policies

A constitutional debate was initiated in 1990, largely as a result of decisions taken at the 5^th FRELIMO Party Congress. As a result of this debate, a new Constitution was approved in November 1990; several important changes to the Constitution related to natural resource ownership and related principles of use and management (see Box 2).

The first article of Chapter 4 (Economic and Social Organization) in the Mozambican Constitution is Article 35, which deals with the public domain of the state and entrenches the concept that the state is the paramount owner of the natural resources occurring within its territorial limits.^[7] The Constitution recognises the obligation of the state, in the national interest, to develop the natural resources of which it is the paramount owner and to determine the conditions under which citizens (and others) may access these resources for their use and enjoyment.

The constitution is unequivocal in its stipulation that land ownership is vested in the state and that no land may be sold, mortgaged, or otherwise encumbered or alienated^[8]. However, the same provision also stipulates that the use and enjoyment of land shall be the right of all the Mozambican people^[9]. The conditions under which citizens may enjoy such rights are the prerogative of the state, which is constitutionally obliged to develop specific laws governing these conditions.

The constitution also introduces qualifications and limitations on the eventual content and nature of these conditions (or mechanisms of access). First, it stipulates that rights of use and enjoyment may be granted to individual or collective persons, taking into account its social purpose. Second, there is a constitutional directive to the effect that direct users and producers must be afforded priority and that the laws developed by the state may not permit use and benefit rights of land to be used to favour situations of economic domination or privilege to the detriment of the majority of citizens^[10].

Most importantly, however, the 1990 constitution obliged the state, for the first time, to recognise rights acquired through inheritance or occupation^[11]. It was this amendment that heralded the subsequent revision of the land law and led to the legal recognition of customary and other rights to land. Mozambicans had, through this amendment, finally ceased to be squatters in their own country.

Policy formulation processes in natural resource sectors

Land

The process of developing the 1995 national land policy and the main elements that emerged in this and the subsequent legal instruments have been well described elsewhere (see J. Negrão, 2000, C Tanner, 2002, Norfolk et al, 2001a). It is important to note briefly, however, some of the elements of the process that have led to its being held up as a model in consultation and consensus-seeking amongst diverse groups within Mozambican society, and praised in particular for the space that was created during the process for the voice of the poor and the marginalized.

First, the policy development process was lead by an Inter-Ministerial Land Commission^[12] created by statute and involving 9 separate Ministries and coordinated by a Technical Secretariat. Meetings of the Technical Secretariat were attended by organizations representative of diverse interest groups: The Associação Rural de Ajuda Mútua (ORAM, Rural Association for Mutual Assistance), the União Nacional dos Camponeses (UNAC, National Peasants Union), church-based groups, national academic institutions, the private sector and land specialists. ORAM and UNAC enjoyed the status of “permanent invited members” to the Technical Secretariat. This structuring of the policy development process meant that non-governmental organizations were involved for the first time in legal reforms. Indeed the 1996 National Land Conference to discuss the draft law, involving 200 participants from government, private sector, aid agencies, academics and international and national NGOs has been called “...an exercise in democratic participation hard to equal...anywhere, north or south” (Tanner, 2001, quoted by Kanji et al, 2002).

Second, a broad-based campaign to raise public awareness and understanding was undertaken by some 200 different NGOs, churches, associations and cooperatives and appropriate and well-targeted materials were developed to put across the messages of the campaign (see also 4.2 below for further details).

Third, the important implementation elements of the new law, contained within the Technical Annex, were subjected to an iterative process of action-learning by the piloting of some of the procedures in the field. Further important elements that contributed to the development of the new law are identified in Box 3.

However, since the disbanding of the Inter-Ministerial Commission there has been a considerable reduction in the extent to which further important aspects of policy development and implementation are subjected to consultative and participatory processes. Often the high-level policy, laws and regulations require additional legal, statutory or bureaucratic instruments to be put in place and the content of these can have an impact upon the effectiveness of the policy and laws themselves. Often, however, the development of these instruments is not subjected to the same level of consultation with and examination by stakeholders and particularly not by those outside of the government agencies.

Thus, for example, the Norms and Instructions of the Direcção Nacional de Geografia e Cadastro^[13] (DINAGECA) is an internal operating manual used by this directorate to guide the work of its technicians and to lay down technical procedures. In referring to the legislated requirement to submit land applications to community consultation processes, the document contains this sentence:

“It is necessary to return the pending applications to consultations with communities, but certainly we ought to open an exception for the case of areas that are less and not greater than 10 hectares”^[14].

Other documents, formats, procedures and operating policies also have the potential to detract from the principles in the policy and law. These include the format and the nature of the community land certificates that are issued after a land delimitation is completed, the procedures adopted for the community consultations (and departmental guidelines on who should participate in these) and even the structure and nature of the official land register. The development of these “institutions” (or “rules of the game”, representing as they do the interface between policies and people) is rarely if ever subjected to the level of scrutiny given to policies and laws.

Forestry and Wildlife

The development of a Forestry and Wildlife policy framework and a corresponding sector investment programme began in 1991 with the drawing up of a provisional programme under a UNDP/FAO team. This was apparently the first attempt at a “programme approach” within the agriculture sector (Cuco, 2001). A “Forestry Pre-Programme”, based on the 1991 design phase, began in 1993 with finance provided from the UNDP. In 1995, a “National Programme of Forestry and Wildlife (1995-2000)” was prepared by the Direcção Nacional de Florestas e Fauna Bravia^[15] (DNFFB), followed in 1996 by the development of an “Investment Programme for the Forest and Wildlife Sector”, drawn up with technical assistance from FAO/UNDP. Also in 1996, a “Forestry and Wildlife Policy and Strategy” was developed (the investment programme was revised in 1997, following a Joint Donor Pre-Appraisal Mission and integrated into the PROAGRI programme^[16]). The Forestry and Wildlife Policy and Strategy document was adopted by the Council of Ministers in 1997^[17].

In these developments, the consultation process was seen as being limited to “foresters” and “technicians” and did not involve other important organizations with different perspectives and a greater awareness of the social aspects of the law. The Policy and Strategy document was followed by the passing of the new Forestry and Wildlife Law in 1999^[18]. Dealing as it did with high value resources there were powerful interest groups involved and consequently there was little consultation during its drafting. The law was greeted with dismay by some of those who worked on the land law and policy development processes and it was said that this new legislation created the “illusion of inclusion”, without clear and practical mechanisms for devolution (Tilley, pers. com., cited in Anstey, 2000).

From 2001 a shift appeared to take place and there was a much greater level of consultation and participation in the drafting of the (as yet unpromulgated) Regulations and of other legal instruments that were necessary in order to move towards implementation of the law. These issues included the following:

• the manner in which resource management powers could be delegated to communities and other entities (to be the subject of a Technical Annex to the law - Art. 109 Regulations);

• the manner in which communities were to receive financial benefits from the payment of local royalties (to be the subject of a Ministerial Diploma - Art. 112(3) Regulations);

• the methodologies and procedures for the establishment of local resource management councils, including their relationship to other planning and resource management bodies, both government and non-government.

To tackle these and other issues a broad-based Forest Forum (see Box 4) was established and a series of consultations undertaken throughout the country. Perhaps, in the same manner as the FAO technical assistance in land policy development had helped to encourage and ensure broad participation in the early phases of that process, it was the DFID-funded assistance to the forest sector that had an impact on the nature of forest policy formulation from 2001.

Although the new law was espoused as a Mozambican law made by Mozambicans, it is questioned whether the same would have happened with political pressure by donors and international technical assistance (see Pijnenburg, undated). The ability of civil society to interact and feed into policy formulation processes was increased by their vertical linkages (a form of social capital) with donors and others who were able to apply pressure so that the processes were transparent, open and consensual.

Despite considerable work the legal instruments have still not been finally agreed upon and promulgated and at this stage appear to be undergoing a process of “harmonization” with other sector legislation. The regulations, however, were published in 2002 and it was obvious that there had been some improvements as a result of the more inclusive approach^[19].

The policy consultation process resulted in the training of three regional civil society teams in order to gather information about the implementation of the new forest and land laws. Team leaders were then responsible for training and including members of the Serviços Provinciais de Florestas e Fauna Bravia^[20] (SPFFB) and local NGOs in the consultation process. Consultation missions were conducted initially in eight provinces involving interviews with more than 300 groupings of government, community, NGO and private sector stakeholders and was later extended to cover a further two provinces (IIED/DNFFB, 2002). Based on this field work, consultation reports prepared for the Northern, Central and Southern regions were presented to focus group meetings to refine and revise the final reports. A synthesis paper was drafted to crystallise the key findings from the consultations into policy relevant materials; this paper focused on the institutional structures and processes needed to ensure that rural communities had access to the rights envisaged in the law.

Consultations with stakeholders are continuing, most recently in regard to the development of operating manuals to provide guidance on the nature of community participation and representation on co-management bodies (COGEPs^[21]) and the formation of partnerships with private sector operators. While initial drafts of these manuals may put emphasis on issues that might be better left for community groups to decide, they are at least being circulated and discussed by various stakeholders.

The development of policy in respect to community rights within national reserve areas is not as clear. This is an issue which has a potentially massive impact upon the livelihoods of many thousands of people who are resident in areas that are, or are currently being declared as conservation areas^[22].

Water

The policy and regulatory environment for the water sector in Mozambique saw the promulgation of a Water Law in 1991 and the subsequent development of a National Water Policy (NWP) in August 1995. The latter therefore represents the first formulation of water sector policy by the government in the post conflict era. The NWP accords high priority to the basic needs of the poor in respect to water supply and sanitation and envisaged the participation of consumers in relevant planning and management of water supply systems, enshrining the principle of management at the most appropriate level. Development of the policy was the responsibility of the National Water Council, which had been formed as a result of the 1991 law. In rural areas there is no substantial difference between poor and non-poor in terms of type of water source and sanitation: most people depend on wells, rivers, lakes and latrines.

Fisheries

The Fisheries Law was approved in 1990^[23] and although it defines and prioritises three main areas of fisheries activity that can be classified as “micro level” (subsistence, artisanal and semi-industrial) there is no recognition in the law of community level rights to fishing resources nor of rights of participation in management decisions regarding the resource^[24]. Fishing resources are state property and the law is heavily weighted towards the establishment of conservation measures. These include the determination of resting periods, areas of forbidden or limited access, maximum quantities of exploration, prohibition or regulation of fishing as far as internationally protected species are concerned and the protection of rare and endemic species. There is also no evidence that the policy formulation for the fisheries sector was accompanied by any kind of broad consultation process. The lack of direct reference to communities’ rights and the weak capacity of the authorities responsible for enforcement are identified as major weaknesses (Chilundo & Cau, 2000). The law applies equally to marine and inland waters. For the latter there are as yet no specific regulations, as there are for other sectors such as aquaculture or marine fisheries. Access to the inland water fisheries is regulated by the General Regulations to the Law.

New policy tools and concepts

The 1995 Land Policy was built upon a set of principles that highlighted the need for greater protection of existing use rights to land and the establishment of an environment within which the rural poor could increase the benefits from the most common form of natural capital available to them: land. The policy was consciously designed to have a positive impact on the livelihoods of the rural poor^[25]. It resulted in the legal recognition of local community groups and of their land use rights, the incorporation of community representatives into formal institutional processes of land adjudication and the establishment of legislated participatory methodologies that permitted community members to register their rights in the national cadastre, either as individuals or as groups of co-title holders.

The Forestry and Wildlife Policy is also geared towards the greater involvement of local communities in the management of natural resources and ensuring that they get benefits from those resources. There are some important similarities and equally important differences between the actual legislation that resulted from the adoption of these new policies. The aim of this section is to briefly highlight some of these issues.

Local communities

The Land Law introduced a concept of “local communities” which related directly to a spatial area within which a group of people lived and made use of resources^[26]. The Technical Annex introduced a legally prescribed methodology for the identification of the community and the related areas, a process that was intended to be in the hands of the particular community and to be one largely of self-definition (with safeguards such as obligatory consensus with neighbours, etc.). Thus the communities could be anything from a traditional unit based on membership of a clan or chieftaincy to a simple group of neighbours (CTC, 2003). Section 4.3 examines the extent to which this flexibility has been utilised in the implementation of the Technical Annex.

The Forestry and Wildlife Law imports the Land Law’s definition of a “local community” but treats the community as a form of public body that has a legitimate interest in resource management, rather than a private body (as in the Land Law) that holds actual private use rights to a resource. It adds hunting as one of the areas considered as safeguarded by the local community^[27]. This is a positive change, since it recognises a further purpose for which customary user rights have existed. However, the Forestry and Wildlife Law only recognises these customary rights to forestry and wildlife resources for subsistence purposes. Instead of recognising more fully an inherent right to the resources (which could then not only be safeguarded by the community, but used by them as a natural capital asset with which they could negotiate), the law establishes a licensing framework for development and exploitation of such resources on a commercial basis. In doing so, it contains elements that put it beyond the reach of most of the existing national forest companies, let alone community groups that might wish to follow this route. In addition, the Forestry and Wildlife Law has introduced a level of confusion regarding the nature of the community as an entity, since it appears to be treating such groups as a form of public body that has a legitimate interest in resource management, rather than a private body (as in the Land Law) that holds actual private use rights to a resource.

The Regulations to the Environmental Framework Law^[28] adopt a different definition of local community. In this law the community is defined as a “group of people situated in the area of influence of a proposed activity and that is not confined necessarily to a village or to a distric^[29]t”.

Consultations and representation

Perhaps one of the most important aspects introduced by the new policies was that of mandatory consultation processes with local community groups, important largely because of the scale and breadth of its application. By requiring applicants (and government agencies) to consult with local groups that were potentially affected by the request for natural resource use rights, the balance of power was shifted away from the state, which had previously held the prerogative of deciding upon the occupation and use status of resources, and towards the direct users of those resources. Both the Land Law and the Forestry and Wildlife Law require these consultations, although they are termed a “renegotiation” in the latter law. One of the improvements to the Regulations for the Forestry and Wildlife Law was that the community were also to be involved in the decisions related to the adjudication of the “simple licences”, which had not been a requirement in the earlier drafts.

The inclusion of such a mandatory process meant that prima facie opportunities were created for negotiating agreements between the rights holders (local communities) and those who wished to gain access to resources. As such the consultations are “a critical ‘development moment’ for both the local population and the investor - one side gets benefits and an incentive to live peacefully with the newcomer; the other gets local cooperation and a secure environment to invest in” (Tanner, 2003).

Community benefits

The Land Law recognises customary rights and gives them the force of formal legal rights, whilst also encouraging the growth of private sector “take-up” of land use rights. It creates an enabling environment that allows local communities and investors to negotiate agreements around land use rights, with the state’s role limited to ensuring that minimum standards are applied in these negotiations, that registration complies with technical standards and that the taxation system functions effectively. The benefits to local communities are envisaged as coming in the form of payments or benefits to them as a result of negotiating the third party use of “their” natural capital.

Conversely, the Forestry and Wildlife Law creates an enabling environment that draws local communities and the private sector into decision-making forums that have management powers over resources. These resources are still owned by the state, however, which recognises no customary or inherent right to them, except in certain limited ways (e.g. the right to subsistence level use). Here, the benefits to local communities come from a royalty paid by the state from the revenue that it collects for use of the resources and a say in how the resources are managed.

It has been said of the two laws that they are essentially the same: rather than directly securing a role for local communities in the development and exploitation of natural resources, they create an enabling environment in which this can happen (Garvey, 2001). This is true, but hides an important difference in approach. When a resource has multiple stakeholders with conflicting objectives and differential power, it is common for governments to work out co-management arrangements. In this way they seek to strengthen local organization, and to provide technical assistance and to mediate the overlapping and conflicting claims on the resource. This allows governments to exercise a regulatory role and to retain control over components of the resource of direct value to the state (Singh and Gilman, 2000). In Mozambique, there are multiple stakeholders with conflicting objectives in land as much as in forest and wildlife resources and yet it is the forest and wildlife sector that has taken the classic co-management approach. The land policy has elements of co-management, but these are linked to the specific allocation of use rights in community areas and integrated into systems that permit community registration of rights.

Participation in resource management and conflict resolution

In addition to the mandatory consultations where user rights are to be allocated to private users, the new policies highlight the involvement of local communities in more general management decisions (through the COGEPs envisioned in the Forestry and Wildlife Law) and in the resolution of conflicts (specifically mentioned in the Land Law^[30]). Again, there is a difference in approach, since the decision-making powers that community groups may exercise over forest and wildlife resources operate within the co-management framework and have the nature of a public right of participation in state-mandated structures, whereas their rights to manage land resources, in terms of the Land Law, are recognised as private, “unadulterated rights” over the land within their jurisdiction. The Forestry and Wildlife Law and the Regulations contain provisions that permit the delegation of resource management powers over forestry resources to community groups; the joint Ministerial Diploma that would define the procedures for this has not yet been drafted.

The nature of new rights

Communities have no right to register their rights or to force the state to certify them; their registration and certification is subject to administrative discretion. Under the land law a community may only delimit their land and request its registration in the cadastral atlas if the district administrator approves this. Although the wording in the regulations is positive (“...at the request of local communities...the right of use and benefit of land acquired by occupation...can be identified and registered in the national cadastral atlas^[31]”) and although there is nothing in the law or the regulations that stipulates that the process is subject to state approval, the technical annex introduces this requirement by including within the pro forma documentation an obligatory process that must be approved by the district administrator^[32]. The Technical Annex, including the pro forma documentation, is a legal instrument and therefore represents a legal requirement in the registration of land use rights acquired through occupation^[33]. This has led to the blocking of registration of community rights, e.g. the Maputo Elephant Reserve authorities retrospectively reduced the area that had been delimited by a community (Nhantumbo, 2002).

In the Forestry and Wildlife law local communities are permitted to register a zone as a sacred forest and have it declared as such (thereby imposing limits on its use by third parties) but this is also subject to administrative discretion, this time at the level of the Provincial Governor.

The governance context

PROAGRI and the PARPA

The first phase of PROAGRI was set up as a five year agreement in 1998 between a major group of donors and the MADER, based on a Memorandum of Understanding. The Memorandum committed MADER to a transformation process that was to involve the identification of core functions at all levels and implementation of an institutional reform process and capacity building for MADER to implement its new functions in an efficient and effective way. Eight separate components were designed as a platform for this programme and a set of “Basic Principles” was established (see Box 5). In addition, PROAGRI was seen as a means to introduce a more coherent and effective way of channelling and managing donor assistance to the sector, replacing fragmented donor-driven projects with a coherent programme.

As far as the land component is concerned, over the period of the first phase of PROAGRI, the vast majority of government resources have been directed towards making provision for the private sector uptake of land rights in rural areas, rather than on implementation of the newly introduced concepts of land delimitation and formal registration of community rights.

The PARPA notes that the State has the obligation to promote and enforce the sustainable use of natural resources, with the participation of communities and local government, for the benefit of the country as a whole and in order to prevent irreversible exploitation. The PARPA further states that agriculture and rural development play a key role in any strategy for poverty reduction and economic growth and is also clear that success in this will depend on other sectors (such as infrastructure, telecommunications, markets, financial services, education and health and nutrition) that have an impact upon agricultural livelihoods in rural areas. The action plans in the PARPA are imported wholesale from the PROAGRI. Although the implication is that “agricultural development” is in fact cross-sectoral in nature, and that appropriate action in a number of sectors and cross-sectoral coordination is required, the PARPA does not elaborate on the mechanisms by which such coordination might take place (Whiteside, 2003).

Formal institutions

The two main role-players in respect to the regulation of natural resource use are the Directorates of Geography & Cadastre (DINAGECA) and of Forestry & Wildlife (DNFFB), both within the MADER. These national directorates are represented at provincial level by service units (the SPGC and the SPFFB) that also fall within the directorates of Agriculture and Rural Development at this level. This is an important constraint within the Mozambican administrative system: “dual subordination”. In theory, the provincial services are responsible to both their line ministry (i.e. MADER) and to the provincial governor. In cases of disagreement, it is not clear whose authority counts and outcomes appear to depend largely on the personalities and individual power bases of those involved (independent-minded provincial governors, especially those out of reach of Maputo have substantial autonomy). Coordination between the SPGC and the SPFFB is still lacking in many areas, although in some provinces there have been moves to implement the institutional reforms that posit the amalgamation of these institutions into a single natural resources oriented service unit (Durang, pers. com.).

As part of a ministerial restructuring process following the 1999 elections, Fisheries was separated from the former Ministry of Agriculture and Fisheries and became a Ministry in its own right. The Institute for Development of Small-Scale Fisheries (IDPPE), which provides support to small-scale fisheries, has always kept separate from the management of semi-industrial and industrial fisheries and from fisheries research. IDPPE’s focus is strongly oriented towards coastal fisheries and there is effectively no formal management of inland fisheries (except on Lago Niassa).

The Ministry of Tourism (MITUR), in respect to policy and regulation of conservation areas and the Ministry of the Environment (MICOA) are also both involved in the development of policy that has a direct bearing on natural resource access. Both ministries have relatively weak provincial structures.

At a district level the structure and composition of the agricultural directorates varies widely but is generally characterized by a very low level of human, physical and financial resources. Very few districts have specific representatives from the provincial land or forestry and wildlife services, which tend to be concentrated in the provincial capitals. For regulatory activities, therefore, the provincial offices will depend upon the participation of generalist technicians based in the districts.

Representatives of the district administrative authorities also play a role in land adjudication processes, but have been less prominent in decision-making in respect to forest and wild resources. Land consultations have to be accompanied by a representative of the district administrator, although in many cases this role will be allocated to the district directorate of agriculture. At sub-district level there is even less specialist capacity and this is usually restricted to extension workers.

Formal judicial structures extend only as far as district level and in many cases are absent even here. Those that do exist are beset with capacity problems; few judges have university level training, legal texts and other important material are lacking and corruption is recognised as a major problem (CTC, 2003). It is not easy for ordinary rural citizens, even when they are organised through a community structure, to get access to the judicial system. Factors that militate against them include:

• a lack of information on rights and entitlements, as a result of high levels of illiteracy and a lack of appropriate dissemination strategies and materials;

• the high costs involved, only partly alleviated by the potential to claim a “State of Poverty” and enjoy partial exemption from court costs;

• scepticism regarding the objectivity of official institutions;

• the imposition of administrative resolutions by local government authorities rather than encouraging the use of judicial recourse; and,

• the lack of an empowered ‘community voice’. (De Wit, 2001)

It is only since the initiation of a training and support programme for the judiciary that most district judges are aware of the land law. It will still require some time before these judges can respond more successfully to needs at the local level (ibid).

Community level institutions

During the colonial period, indirect rule occurred through a system of land estates (prazos), either granted by the crown or taken by conquest and governed essentially as fiefdoms. The colonial administration also co-opted traditional chiefs (régulos) as a further instrument of indirect rule; the chiefs became the main mechanism for levying taxes, recruiting labour and allocating land outside the prazos and company domains. At Independence, FRELIMO abolished the regulado and replaced it by a new cell-based system of centralized democracy, with grupos dinamizadoros, village councils and land commissions responsible for land allocation (O’Laughlin, 1995). In some areas, notably in the southern provinces and the liberation zones of the north, this system became established with local secretarios holding substantial power and influence. Geographical patterns in authority can still be discerned, although these have been disrupted by overlapping migration patterns as people fled from conflict.

In general, traditional authorities appear still to be the main influential actors with respect to natural resource allocation and control. In addition to their identification with spiritual roles and mystical values^[34] the traditional authorities are often used by NGOs as dispensers of aid and by companies as agents and generally have high stocks of social capital and influence. The extent to which traditional powers have been eroded by the official marginalization of the régulos after independence varies, but there is evidence that in some areas this process lead to the development of open-access systems for common pool resources that had previously been regulated through customary norms and practises (see Box 6).

In relation to land resources, however, the power and legitimacy of the traditional authorities seems to have been largely maintained. The end of the war and the consequent return of displaced populations in the early 1990s proved this continuing durability of traditional institutions of land allocation and adjudication: the re-establishment of legitimate and widely accepted land holding patterns (between groups and individuals that had remained in the countryside, those that had returned and those arriving to new areas) occurred within the framework of the customary rules of the rural populations. The process occurred largely without conflict and required little intervention from formal authorities.

In times of normality, the traditional authorities in an area may only be used by local people as a forum for resolving disputes. In many areas access to land can be through kinship networks or neighbours rather than through the chieftaincies. Outsiders who come to a new area in search of land would traditionally be expected to ask permission from the local traditional authorities, but in some cases (see Box 7) this may just consist of informing them after the fact in order for the boundaries to be confirmed.

Customary systems of resource tenure vary across the country and between different ethnic groups. The sophistication of these systems appears to be linked to population density and competition over resources. For example, in the central provinces local leaders can describe a complex set of rules and taboos, and the hunting and harvesting of a number of species is proscribed at particular times (SLSA, 2001).

Over large areas of the northern province of Niassa, on the other hand, where population density is extremely low, few taboos can be identified, even with respect to trees which are considered to facilitate communication with the ancestors. Where conflicts emerge, one section of the community simply breaks away and moves on and there is a reluctance to engage with either administrative or communal authority. On the other hand, in the same area, the paramount chiefs are able to describe and agree the boundaries of their territories across most of the province (Anstey, 2000).

The lack of a functioning judicial system in rural areas creates further reliance upon the traditional authorities, the Presidentes de Localidade and the Chefes de Posto to serve as higher authorities if a criminal or civil matter is irresolvable at the village level. As a result, rural communities govern themselves more often by traditional laws or remembered vestiges of colonial law than by current state law (Knight, 2002).

Trees, in particular, play an important evidentiary role in legitimizing local claims. Loans of land are often accompanied by a prohibition on the planting of new trees.

2.3 The implementation of natural resource policies

Information on rights and entitlements

Access to information is a critical factor in ensuring that services from a state to its citizens are in fact delivered and that standards are met. When it comes to rights and entitlements the information is a sine qua non since a right loses its force completely if you do not know that it exists; people have to know what their entitlements are in order to be able to claim them. Bureaucratic actors not only participate in constructing systems of rights and entitlements but can also impede the implementation of rights. One of the most effective mechanisms by which to ensure citizens do not claim rights is through the restriction, or lack, of information on entitlements.

The Land Campaign has been the main driver behind the dissemination of information, dealing not only with the new rights and entitlements contained within the Land Law but also some of those contained within the new Forestry and Wildlife Law. The campaign concentrated initially on getting across 6 main messages from the new legal framework. The means of communicating these messages included ”comic” strips, and audio-cassettes with recordings of the dramatised texts of these scripts. These were in Portuguese and 20 national languages (Negrão, 1999). In the second year of operation the campaign defined a further set of 6 messages. (See Box 8). Following on from the initial campaign and in addition to it, several NGO groups concluded separate funding arrangements with donors to finance the dissemination of information regarding the new law. This, for example, formed a major part of the DFID funding to ORAM in Zambézia which, in the period from 1999 to 2002 conducted literally hundreds of information sessions at various levels - with community groups generally, with community leadership, with local administrative structures and including district and provincial level seminars.

The Land Campaign later transformed itself to a Land Forum and became institutionalized in some provinces. The present level of energy and coordination of activities varies quite widely and there is evidence that some of the provincial fora are now more characterised by elements of competition rather than collaboration and coordination between the participating organizations (Ribeiro, pers. com.)

Dissemination of the Forestry and Wildlife Law has not been so extensive; initially, some organizations were reticent because they perceived the message to be one of control of community activities rather than the creation of space and opportunities for them^[35] (SLSA, 2001) and other organizations were awaiting the publication of the long-delayed regulations before they felt confident enough to become involved in dissemination. Some of the provincial Land Fora have, since publication, become involved in this and ORAM in most areas is now including a focus on the forestry laws and policies in its ongoing activities.

It appears that there have only been partial and fragmented attempts to monitor the effectiveness of the various information campaigns (an issue of concern to some donors) but there is considerable anecdotal evidence that suggests that knowledge of the laws and the associated rights and duties remains low, even in areas where NGOs have been operating quite intensively. In Zambézia, for example, even in areas where land delimitation had taken place, many community members still exhibited little knowledge of their rights (Wrangham, 2001). This is offset by evidence from other parts of the country that would seem to demonstrate that the information campaigns have in fact been very effective (see later for further discussion on this).

There is also anecdotal evidence that suggests that the intensive targeting of community groups by NGOs and civil society campaigns, coupled with the comparative lack of a state driven process of disseminating information, has created a situation in which the local populations’ awareness of the regulatory framework is in fact greater than that of local administrative structures, including at a district level. This is equally true within the forest sector. Conclusions from the consultation process carried out in the forestry sector also identified similar obstacles: (1) government agents did not have access to the laws nor the capacity to disseminate information to local community groups; (2) the law had not been translated into local languages; (3) NGOs involved in dissemination did not always understand the concepts and principles in the law; and (4) high levels of illiteracy were an obstacle to dissemination (Nhantumbo & Macqueen, 2003).

Consultations in land and forest resource allocation procedures

One of the most important aspects introduced by the new policies was that of mandatory consultation processes with local community groups. These are now necessary in every single application for natural resource rights in rural areas. The consultation process is an important opportunity for the establishment of a potential long-term partnership between a local community and private sector investors in rural areas and is of primary importance in reducing the potential for later conflict.

As a new institution these consultations are beset with myriad problems. In some cases they are not taking place at all or they may be performed in a perfunctory manner. Local elites may manipulate the process. Local administrative structures may not provide supportive guidance. Structural problems exist, such as the inclusion of a mandatory financial “incentive” for the community group and the lack of a system for capturing the terms of agreements and monitoring compliance. Box 9 shows the main findings of an analysis of over 100 consultation processes in Zambézia province and these are supported by much anecdotal evidence from other parts of the country. Further research conducted in Zambézia revealed a high awareness on the part of communities of the requirement that they be consulted on new applications for private land use rights within their areas. However, they registered complaints regarding the quality of consultations that have taken place to date and record that in some instances there are other ‘local structures’ that are consulted in their name^[36].

Even where local administrative structures may take a more neutral role in respect to the consultations, there is very little acceptance of the need to provide community groups with support and information that would assist them in making informed choices in the negotiations. Very rarely, if at all, do SPGC representatives who attend the consultations come equipped with maps and registers that show the extent of actual and pending private land holdings in a community area and it is in fact rare for even the district administrations to have access to this kind of information. Innovative and progressive solutions are not advanced by local state structures, which are often looking for quick fix solutions rather than long-term agreements that they will then be required to monitor. Thus, as a report from Cabo Delgado province states: “In reality, the new law has not turned out quite as well as planned. While it does defend community land rights, it has not produced the close relationships between investors and rural communities that its designers envisioned. Instead of contracts spelling out ongoing financial relationships between investors and communities, the practice of one-off (compensation) payments continues, leaving community members with a short-term flush of cash and long term loss of their lands” (Bechtel, 2001). There is also very minimal recording of the nature and elements of any longer-term agreements that may be being made, making the monitoring and enforcement of these an extremely unlikely scenario in the future^[37].

Bila & Nhantumbo (2002) point out that community consultations in the forestry sector are more complex in comparison with those regarding land resources, largely as a result of issues of scale. Forest concessions can cover hundreds of thousands of hectares distributed across district boundaries and covering various administrative posts and dozens of villages. They further note that given the apparent complexity and variety of stakeholders and interests involved in such large areas it would be expected that the consultations would raise a range of issues and expectations. However, “the consultation reports of those [concessions] already approved appear very simple and are not clear on the methodology used nor the principles used in the [consultations]” and it would seem that “the consultations were merely done to obtain a ‘rubber-stamp’ [from the community]” (Bila & Nhantumbo, 2002).

There is in fact little experience of consultations in forestry applications, given the fairly recent enactment of the regulations and the fact that most of the existing applications were in fact already authorized (despite their nominal illegality). However, concession agreements may proceed without operators taking on any meaningful obligations vis-à-vis the affected communities: “Other villagers lament that community leaders have negotiated away their forests in exchange for gifts such as a bicycle or food. Concessionaires admit that they often arrive in areas to begin harvesting and find villagers who have heard nothing of the concession. When problems arise after operations begin, such as the inadvertent destruction of crops as a result of tree felling, communities apparently find it impossible to speak with people at a level within the company who can make reparations (Reyes, 2003).

An often stated view is that the consultation processes act as a powerful disincentive to investment in rural areas, but it appears to be rarely the case that a community will reject an application during consultation. Most community groups in fact welcome the potential presence of a new local actor with resources and social capital that they do not possess, perceiving this to be a positive impact upon local development.

Land delimitation exercises

After the initial land delimitation exercises undertaken in 1999 as part of the piloting processes of the Technical Annex^[38], the level of government finance, resources and involvement in this area of implementing the new policies has been extremely limited. Most land delimitations since this time have in fact been undertaken through off-budget donor-supported exercises that have been implemented by various NGOs, with government participation occurring as a reimbursed service to these groups.

Indeed, although there was apparently considerable support for the resource and funding requirements that would be needed to implement and test the new poverty-focussed elements of the law^[39] (not least through PROAGRI) it soon became apparent that the government considered these to be of secondary importance and embarked instead upon a drive to attract outside investment and to facilitate the allocation of private land use rights. While pro-forma expenditure plans may include allocations for information dissemination and delimitation but actual expenditure has been minimal over the last few years. There is what has been characterized as “a tendency within the more conventional thinking of PROAGRI towards a dualist view of ‘family’ sector (community) land use and new, private sector land needs” (DfID, 2000).

Despite limited government funding for the pro-active delimitation of community land, there have been a considerable number of delimitations completed in various parts of the country, largely through the support of NGOs. Information on and monitoring of the implementation of these delimitations has been fragmented and partial, notwithstanding the recent DFID-funded appraisal of the land sector, which included a specific review of completed delimitations throughout the country (CTC, 2003). This review stated that the total of delimited areas now stands at 162; 59 of which have been issued with certificates and 24 of which have a full land title, implying that they were demarcated rather than delimited (ibid). One of the gaps left by the dissolution of the Inter-Ministerial Land Commission is a central collection point for the monitoring of the implementation of this aspect of the new law, a gap which has not been filled by the National Department of Geography and Cadastre which is still battling to provide the country with an efficient and transparent register on private land holdings.

The policy-makers in the mid 1990s were keen to highlight the flexibility of the land law and technical annex: the community (as an entity that can register land use rights) need not necessarily be a group that pays allegiance to a traditional paramount chief (régulo) and that the boundaries of delimited areas need not follow old “traditional” boundaries from the colonial period, nor present-day administrative divisions. The “community” can choose its own name, define its membership according to broad and flexible guidelines and appoint its own representatives, free of any stipulations in the law. The challenge of “how to (legally) recognise a group without converting it into something else” (Fingleton, 1998) seems to have been an important consideration in the framing of the law. In implementation, however, it appears that the vast majority of delimitations that have been completed are being defined along the lines of the old colonial-era regedorias or defining and refining the zones of influence of traditional chiefs and authorities at various levels.

It is the issue of scale that is perhaps one of the most important in relation to the effectiveness of delimitation exercises and their actual and potential impact upon livelihoods. Bruce (1999) suggests that common property management regimes are predisposed for success when the group of users is small and the resources not too extensive. Reasons for this include the low cost of the intra-group enforcement of rules, extra-group exclusion, easy detection of problems and infractions, better coordination and participation in decision making. The Mucombwe case had a small group, containing a population of over 600 people, and the area that was delimited covered 3,000 hectares. In contrast, some of the delimitations completed in Zambézia have covered very large areas of land with considerable populations^[40]. Again, this appears to be as a result of the enduring nature of the concept of regulados and regedorias^[41]. The “entry point” used by ORAM for the delimitation exercises in Zambézia, for example, has tended to be through the locality (state) structures and the “recognised” traditional authorities, many of which are identified through the use of old colonial-era registers or maps (Norfolk et al, 2003). Well-entrenched notions of administrative and traditional boundaries tend to be the dominant defining characteristics of the “community groups”, despite a policy framework that would allow for a much wider range of potential associations.

The extent to which the process of delimiting community land forms part of a “joined-up” implementation approach, that has as an objective the integration of the poor in the social and economic development of an area, has also been of concern. Many delimitation processes to date have been criticised for having been undertaken in isolation and without a clear vision of how the exercise would form part of further, obviously necessary, processes of local planning and development. To a certain extent this situation has come about as a result of two main drivers:

• An objective articulated by some NGOs that considered the defensive implementation of the law as necessary in order to secure and protect land rights from outside encroachment; and,

• A programmatic imperative on the part of some NGOs to deliver on targets set as a result of donor agreements.

Some commentators lay the blame for this more squarely with the NGO groups that have been implementing the delimitations, criticising them for not having stimulated further planning activities nor actively looking for potential alliances with private sector organizations that would be interested in negotiating access to some of the resources over which community groups had acquired legally-registered access (Nhantumbo, 2002). Others consider that long delays between activities and poor planning and coordination between state and NGO service providers is partly to blame. In the community of Canda, for example, land delimitation was undertaken as part of the FAO/DNFFB Community Based Management of Natural Resources Project, but the long delay between this and the initiation of a management component to the exercise was judged to have left people in the area feeling that the delimitation process had no real purpose (CTC, 2003).

The award of forest concessions to private sector entities

The procedures and principles for the concession management framework for forestry have been under discussion and development for the past two years and formed a further part of the DFID sponsored forest policy formulation process. There has been, however, a disjuncture between the principles and requirements that were eventually adopted and the actual application of these in the concession awarded to date.

The Regulations provided that concession contracts of up to 20,000 ha and all simple licenses can be authorized at the level of Provincial Governor without the involvement of the national government (concessions ranging in size from 20,000 to 100,000 ha must be approved by the Ministry of Agriculture). The basis for granting a concession begins with a direct request presented to the provincial director of forestry by those who seek it. This should be followed by several steps including an estimated timber inventory and the community consultation process. After the consultation a more detailed topographic representation of the area, its population and its timber should be forwarded along with plans for the operation of sawmills. Analysis of these materials is done at the provincial level, whilst the proposal is made public in national newspapers. Depending on the size of the possible concession, authorization is granted at the level of Provincial Governor or the Minister of Agriculture.

What actually takes place is less definitive. For example the detailed timber inventories call for a substantial expense for the would-be concessionaire prior to any guarantee of a return and have not been required by the DNFFB in the granting of the initial concessions. Management plans are weak, where they exist. The inclusion of sawmill plans may or may not happen, but a general shortage of functioning mills indicates a widespread lack of implementation. No mention is made of how monitoring might take place. There are indications that animosity and frustration are emerging among Mozambican communities as the ill-planned and intensifying timber harvest provides them with few benefits and little hope for the future (Reyes, 2003).

The methods and intensity of logging activities in a forest area can bolster some aspects of a local economy while undermining others. For example, wage earners from amongst the community can provide an increased market for locally produced goods, while on the other hand, those who depend on the harvesting of game or medicinal plants may find these things more difficult to come by.

Community forestry and other CBNRM initiatives

The Community Management of Forest and Wildlife Resources Assistance Unit within the DNFFB was established in 1998 under the direct control of the National Director. Funded by the Dutch Government and implemented by FAO^[42], the unit’s main objective was to coordinate, facilitate and promote the development of community-based management systems. The main aim of the project has been to develop methods, tools and arrangements for community forestry; despite the range of these that have been developed to date the models “are not yet ready for full replication.” (Mansur & Cuco, 2002).

A raft of other CBNRM initiatives exist. Many are driven by biodiversity conservation objectives or other objectives aimed at decreasing the degradation of natural resources in particular areas. Some of these projects have begun to use land delimitations as a tool for securing legal title to some areas but have been characterized by “narrow” delimitations of particular areas and resources, often attached to particular interest groups, rather than an identification of the variety and diversity of resources and land that may be of importance to various groups within a community. While these projects seek to provide the legal, institutional and economic frameworks for communities to become co-managers of communal area resources and advocate strong “ownership rights” for communities, they have been relying more upon state “permission to experiment” rather than a strong mandate for decentralization and tenurial security.

Decentralization initiatives

The devolution of powers of natural resource management to local groups has been complicated by a parallel process through which the government has been reinstituting the institution of “indirect rule” through ‘community representatives’. This legislation was not submitted to parliament but was issued as a decree from the Council of Ministers^[43]. The decree essentially re-appoints the traditional chiefs as legally-recognised representatives of community groups, although it mentions that these representatives can also be the Bairro or village secretaries or other leaders “legitimized as such by their respective communities”. In the decree and the associated regulations^[44] there are no clear procedures for how this choice of representative is to be made, except for provisions that state where a community has “legitimized” both a traditional authority and a “civic” leader (such as a Bairro secretary) it is the community that decides who has precedence. Research conducted in Zambézia in 2001 revealed that the implementation of the decree was understood rather differently by the local authorities who stated that they were “calling all the chiefs in the area to reinstate them” (SLSA, 2001).

Areas of confusion arise because one of the specific issues that is a subject of engagement between these representatives and the local administrative authorities is that of the “use and benefit of land”. However, the definition of a local community in the regulations to the decree varies from that in the Land Law and is strictly related to territorial administrative divisions; district, administrative post and locality. Community representatives of these groups are therefore state-appointed, state-remunerated and of a public character, whereas local community groups in terms of the land law are private land-holding entities. Difficulties arise when the state treats the “15/2000 representatives” as if they were the legitimate representatives of a local community as defined in the Land Law, which has in fact been the position of the DINAGECA (see later Box 11). The Land Law contains an article that states that the mechanisms for representation of community interests in respect to land use rights are fixed by law (Article 30); it therefore cannot be that the decree 15/2000 and the associated regulations serve this purpose, since it is not a law but merely a decree^[45]. What remains, however, is a confused situation understood in different ways by different actors.