Pekka Virtanen

An increased role of local communities in natural resource management has recently been widely advocated as a solution to the problem of environmental degradation in the Third World. This conclusion is based on a broad debate on the role of endogenous institutions in which academics, politicians and practitioners working in southern Africa have participated. But, thus far, the debate has largely relied on dogma about the essential nature of rural communities: they are claimed to be clearly bounded, socially homogenous, and based on shared norms. In this article, the validity of these claims is studied in the context of the present administrative and legal reforms taking place in Mozambique. The analysis is based on field data from a case study from the Chôa highlands in Manica Province.

In Chôa, exclusive territoriality is not considered a valid strategy, as the local population remains part of a larger socio-economic network, which extends to neighbouring Zimbabwe. The linguistic and cultural continuity has created favourable conditions for a dynamic process whereby the border has become an asset instead of a barrier. The open character of the border made it possible to benefit from both the more developed economy of Zimbabwe and the abundant natural resources under common property regime in Mozambique. In the study area in Chôa customary authority remains practically the only functioning institution at the local level despite adverse government policies. However, it operates without a firm legal basis and may be losing its efficiency, because the source of consensus - traditional religion - is being challenged by the penetration of new religions and the market economy. The most acute changes concern customary rules of land tenure, which are presently being redefined locally. Such emerging rule-systems are dynamic and adaptive. But they can also be perceived as a risk to national unity, and their compatibility with the principles of modern state administration remains a contested issue.

1. INTRODUCTION

In recent years the poor outcomes of state-centred strategies of natural resources management and externally planned development interventions have forced both policy makers and scholars to reconsider the role of local-level institutions in resource management. In their contributions to this debate, various actors - ranging from international organisations to national NGOs - have advocated the increased role of local communities as the best way to overcome problems of the previous "top-down" approach. But, despite its popularity in current development jargon, the concept of "community" remains extremely vague and even confusing. In a recent critical review of the concept, and its application in the natural resource conservation discourse, Arun Agrawal and Clark Gibson[1] have identified three dogmas about "community" which are typically emphasised by those who advocate a positive role for communities: i) community is a small spatial unit, ii) it has a homogenous social structure, and iii) it is based on shared norms.

The authors also show that the uncritical emphasis given to these aspects leads to a simplified and undifferentiated understanding of community, as well as to a misconception of characteristics that may be beneficial for the development of community-based management systems. Instead, they highlight the dynamic side of community, which in reality always consists of multiple actors with multiple interests. These actors interrelate through different kinds of processes, which are structured by various institutional arrangements; that is, they operate in a specific institutional context.

In this paper I discuss the relationship between local communities and the state, starting from the different implications that the concept of territoriality has for them in Mozambique. The problem of defining a local community, and identifying or creating representative community institutions, has become a key issue in the debate about rural development in the present context of transition to market economy. Local community institutions in rural areas can be perceived either as a means to strengthen government control in the national periphery, or as a way to increase local autonomy and flexibility in the face of new economic opportunities. Both of the above approaches can be identified in the Mozambican discussion, but the emphasis of this paper leans towards the latter.[2]

2. COMMUNITY IN THE MOZAMBICAN POLITICAL CONTEXT

In the context of the present socio-economic transition process, universally popularised concepts like decentralisation and community-based resource management have gained popularity in the Mozambican politico-legal vocabulary. This has been concretised in the new laws, which refer to the principles of community participation and recognise some basic rights of local communities.[3] However, with respect to the concept of community, and especially the role of customary authorities, the new laws remain vague. This is largely due to strong party-political connotations of the customary authority issue, which has served to keep the discussion on a generalised and/or an ideological level. I shall first look into the national political context of the debate.

When Frelimo[4] took over the government of Mozambique from the Portuguese at independence in 1975, it sought to establish a totally new politico-administrative system in line with its radical modernisation policy. Customary authorities, who were labelled "obscurantist remnants of the feudal past" and "lackeys of the colonial regime", were excluded from participation by the new electoral law, along with others, who the new rulers took to be collaborators or exploiters.[5] Instead, the new government promoted production-based socio-economic units such as communal villages, agricultural co-operatives, and statefarms and factories as the building blocks of the new socialist state. This construction took place under the strict guidance of the Frelimo party-state, which was represented at the local level by party cells led by secretaries and activist groups (grupos dinamizadores). During the Chissano regime this policy has been subject to increasingly more open debate by politicians, members of state administration, and donor representatives. Some critics have maintained that destruction of indigenous local institutions and the imposition of new ones - such as the communal villages established by the Frelimo government - was actually the cause of the civil war between 1977 and 1992.[6]

In areas which fell under Renamo control during the civil war, the guerrillas sought to establish a combined military/civilian administration where customary authorities were responsible for the latter - including the management of natural resources and conflict resolution - much like the colonial period. Since the end of the war, Renamo has continued to promote the reinstatement of chiefs, claiming that they are not the "reactionary obscurantists" of the Frelimo doctrine, but an important source of social cohesion and cultural identity. As such, chiefs are seen by Renamo as a crucial factor in legitimating authority and regulating rural people's relations amongst themselves and with the environment. Meanwhile, Frelimo has softened its approach, but is still hesitant to acknowledge any formal position of customary authorities, even though their collaboration is sought locally.[7]

When administrative and legal reforms were made in Mozambique after the 1994 elections, two main institutional models to solve the problem of community representation were proposed by representatives of different governmental and non-governmental institutions. These have been backed up with various studies funded by different donor organisations, such as the FAO, USAID, Ford Foundation, and the German GTZ. While the two models have obvious points of convergence with the ongoing Frelimo-Renamo debate, they have so far managed to remain aloof from overt party-politics. According to the one model, appropriate resource management units must have discrete boundaries, which enclose identifiable groups of interdependent users. In practice, this is interpreted to mean de facto recognition of customary authorities as the representatives of rural communities, which shall be geographically delimited according to pre-independence chieftaincy borders. The alternative model is to vest power over resource management and the resolution of conflicts in a local council. This would be composed of representatives of the main local interest groups.[8] Both models accept the need for decentralisation. What is under dispute is the form of institutional representation based on the key characteristics of community: its territoriality, homogeneity, and its sharing of common norms.

3. PEOPLE AND BORDERS IN SERRA-CHOA

As a concept, "territoriality" refers to the existence of distinct geographical borders, which indicate who is included or excluded from a community. In Serra-Chôa there are two kinds of formal administrative borders, both of which are relatively recent. The international borderline along the British territories (to the west) only became effective during the short phase of an effective Portuguese administration after the final collapse of the historic Barwe kingdom in 1920. However, it should be noted that this border follows the northwestern border of the Barwe/Macombe kingdom closely, and thus has deep historical roots. The internal southeastern administrative border dates from the year 1957, when the present Administrative Post of Serra-Chôa was created, and which consists of two chiefdoms (regedorias).

Ever since the Barwe wars of the first two decades of this century, when the last Macombe warriors and their followers were given asylum on the British side, the international border has provided relative safety for political refugees. This function became exceedingly important during the recent civil war, which turned some four million Mozambicans into refugees or displaced persons.[9] According to official statistics, about 100,000 people from Manica Province lived abroad in the early 1990s[10], most of whom found refuge in the neighbouring Manicaland Province of Zimbabwe. Nearly 80,000 Mozambicans were registered in this area during the 1992 Census.[11] Along with refugees, migrants to the more industrialised neighbouring countries have traditionally included a large number of labour migrants.[12] But there is also a less well known counter-current: Zimbabweans moving into Mozambique in search of land because of its increasing scarcity in their own country.

In the study area the immigrants have in recent years been mainly individual peasants, as described in the following case:

Case 1.

Mr. Maruva, a young man from a nearby Zimbabwean village in Nyanga District, moved to Adhabi area in Serra-Chôa three years ago. As an orphan he had been sent to school by an Irish missionary, and had completed sixth class: a good basic education by the standards of his new home-country. After working in various agricultural jobs, which allowed him to accumulate some savings, he decided to move to the Mozambican side in search of better pastures for the cattle he had acquired. He contacted the sabhuku (sub-chief) of the area and was granted permission to settle. However, he has been able to keep on cultivating some land in his home area in Zimbabwe. Two years later he decided to move again, this time within the area under the same sabhuku, to a site with good pastures left vacant near the administrative post.

Now he has a prospering herd of some 70 head of grade cattle and imported 'boer' sheep, and he is actively developing his farm. As he is culturally from the same Shona sub-area and obviously skilled and hard working, he is on good terms with the traditional and modern administration, both of which seek to encourage market oriented agricultural production in the area. He plans to settle permanently in the present site, provided that the road network will be improved, as planned, and cattle rustling does not worsen.[13]

One reason why the immigrant (in the above example) was well received is because of the direct link which is perceived between population, political power, and development. More people means increased agricultural production and more power to the respective sabhuku. In the Chôa highlands the homesteads are dispersed over a large area, and the population density remains low, as there were only some 12,000 people living in an area of 1,226 square kilometres in 1997.[14]

Another interesting point is the evident porosity of the administrative borders: the local communities are surrounded by soft boundaries - irrespective of their formal status as local or international borders. In addition to individual and group migration, for both economic and political reasons throughout the years, the international border is crossed daily by Mozambicans seeking medical aid or markets for their agricultural products, as well as shops to buy foreign industrial products. Even most of the children living in the Caeredzi and Jora valleys go to Zimbabwean schools.[15] This process is caused by lack of social infrastructure and marketing channels on the Mozambican side, and physical proximity of the Zimbabwean services especially for those living in the Caeredzi valley.

The Mozambicans are reciprocated by Zimbabwean buyers and peddlers, and people from neighbouring areas looking for better pastures for cattle, wild-game for meat, or firewood. The latter are typically common property resources, which are becoming increasingly scarce on the Zimbabwean side. The same kind of openness characterises the situation between neighbouring wards and chiefdoms, but at present, the opportunities created by the different economic prospects in the two countries make international exchange more lucrative. As a large proportion of resources must be acquired from outside of one's own village, the principle of reciprocity is tacitly accepted, and strict exclusion rules are not supported by the local population. This was also stated quite openly by the population.[16] Similar findings about soft borders between culturally related local communities have also been reported from studies conducted in Zimbabwe.[17]

4. HOMOGENEITY OF POPULATION

In Serra-Chôa, the limits of ethno-linguistic and modern administrative territory are - apart from the international frontier - practically identical, as the border effectively separates the ChiManica speaking enclave inhabiting the Chôa highlands (and some intervening valleys, shown on Map 1) on the Mozambican side, from the ChiBarwe speakers of the predominantly lowland areas to the east.[18] Even though migration from Zimbabwe still occurs, it is generally limited to individuals or families that are from the same Shona sub-culture and geographical region, and many of the newcomers already have relatives in the area.[19] Migration has thus caused very few changes in the ethno-linguistic composition of local population. However, in terms of economic differentiation the impact has been more notable. Even though access to markets remains a problem in peripheral areas like Chôa even today, some seven years after the war, signs of increasing production are prominent. But the development is rather uneven: extensive production of cash crops and large-scale breeding of cattle are the domains of a few wealthy individuals. Many of those who now have big herds or manage to cultivate extensive areas have worked for quite a few years in Zimbabwe, where they have accumulated both technical skills and the capital to put them into effective use. A number are also members of the ruling lineages.[20]

Economic differentiation is not a new phenomenon in Serra-Chôa, as some local African farmers have managed to benefit from the vacuum left by the Portuguese settlers at independence, as exemplified by the case of Mr. Gumbo:

Case 2.

Like most polygamist peasants in Nhacapanga, Mr. Gumbo has spread his agricultural activities between two homesteads, one at the highland area, where most of his cattle are kept, and another at the Caeredzi valley, where he used to have large fields planted with cash crops (cotton). Soon after independence he managed to acquire a tractor and a mechanical grinding mill from an emigrating Portuguese farmer, which allowed him to expand production. And, even though the tractor is by now defunct and Mr. Gumbo lost a lot of assets during the civil war, in the 1998/9 growing season he was still doing relatively well with a large area under cultivation.

More active re-entry to commercial production is presently prevented by unfavourable prices in Zimbabwe, which is the main market for agricultural products from the valley, and difficult access to markets on the Mozambican side, due to a lack of transport facilities. This situation also discourages investment in machinery, irrigation material and fertilisers, which would be necessary for increased commercial production. On the other hand, both Mr. Gumbo and other wealthy individuals are watching the situation closely, ready to intensify their production when conditions improve.[21]

Mr. Gumbo's case, although somewhat special, brings up a number of important issues with regard to economic differentiation in Serra-Chôa. In the highland area land is not the limiting factor in agricultural production, even though it is becoming such in the valley. The opening up of new fields is limited by labour and/or money to invest in cattle, which is the main source of draught power and manure in the highlands. There are already a number of peasant households that have an embryonic capacity for limited commercial production (cattle and irrigated agriculture), which is presently limited by the lack of access to markets.

While those who have concentrated their efforts in the highlands are usually monogamists, there is a group of relatively wealthy polygamist families who have diversified their labour force between cattle husbandry in the highlands and cash-crop cultivation in the valley. This makes it possible to combine the positive sides of the valley (fertile soils and warmer climate) with those of the highlands (good pastures with few cattle diseases). Up to 1997 the valley also provided easier access to the main markets in Zimbabwe. In many cases the economic power of such families is reinforced by access to capital and modern technology from Zimbabwe through migrant labour or even salaried jobs. If these families manage to secure access to more agricultural land and pastures through the communal tenure system based on customary norms, they have quite good economic prospects.[22]

5. COMMON NORMS AND CONSENSUS

In Barué District the present administration has recognised explicitly the predominant role of customary authorities at a local level, where they are expected to represent the state authority in collaboration with the Chief of Administrative Post and the President of Locality.[23] This situation reflects a change in the official position, even though this is more in terms of giving increasing leeway for individual approaches than a consistent national policy.

However, the change has been gradual and is quite recent on a more formal level. After independence, representatives of customary authority were removed from the office in Serra-Chôa, even though some individuals retained an influential position informally. When those holding the positions during the colonial period died they were not replaced, and the ritual functions - rain ceremonies and rituals for the ancestral spirits - were discontinued. It was only during the early 1990s that new sabhukus and chiefs were elected, and the ritual functions are only restarting now, after most of the refugees have returned.[24] But we should not confuse the lack of formal recognition of customary authority by the state with the practical relevance of customary norms for the local population.

In general discussion, the continuing validity of customary norms is emphasised by the population. There are various area-specific rules at the local (dunhu) level, which still regulate the use of land and natural resources with some effect. In Nhacapanga, for example, the area is divided into separate areas for grazing, on the one hand, and for homesteads and cultivated areas, on the other, and owners are held responsible if cattle destroy property inside the latter. The cutting of trees is also regulated, and some places, such as sacred forests, are totally protected according to the norms of traditional religion. However, in everyday life these kinds of idealised rules are not made explicit, and their enforcement relies mostly on internalisation and voluntary observation. Conflict resolution generally follows the consensus-principle, while punishment is limited to fines, ceremonies to appease ancestral spirits or, in serious cases, expulsion from the community.

The norms become explicit when an outsider wants to use resources or move into the area. In that case he must first ask permission from the sabhuku, and in this process he must explicitly commit himself to respect local rules before he is granted permission.[25] On the other hand, the basic rules are almost identical throughout the region, and all users are subject to them. This facilitates a system of reciprocal use rights in natural resources between neighbouring communities.[26]

At the same time there are signs that respect for customary norms are weakening with regard to sacred forests, which are threatened by uncontrolled bush-fires and desecration. The most sacred areas have a ritual basis for their sacredness: typically they are burial places or propitiation sites for ancestors.[27] But, even though the area is replete with traditions about powerful ancestral spirits that protect the natural resources, and especially the sacred places in the form of mhondoro lions or snakes, non-respect of cult places was constantly presented as one of the main environmental problems during PRA meetings.[28] While this may seem a rather marginal issue in terms of natural resources management, it actually has more profound implications. This is because transcendental beliefs, which are grounded in traditional religion, are an integral part of an holistic world view that gives legitimacy to a broader system of common norms. The flagrant disregard by some youth and outsiders can be interpreted as a sign of decreasing recognition of the whole institution.[29] With increasing integration of the present peripheries with national and international markets, the common cultural basis may gradually disappear.

The problem with a norm-system based on traditional religion becomes increasingly evident with the ongoing penetration of new religious orders. In addition to the persistence of a few catholics, new religions have also entered the area from Zimbabwe, notably the African Apostolic Church of Johane Maranke (Vapostori), which is influential in the Nhacapanga area. So far, this has not created major conflicts because, in contrast to universal (European) Christian churches, such syncretist movements encourage African customs, such as polygamy, and accept that the spirits of the dead can influence the physical world. Even participation in traditional religious rituals seems to be tolerated in the Mozambican context.[30] For example, the present sabhuku Nhacapanga is a member of the Vapostori, while his father (who was the previous sabhuku) was a follower of traditional religion. This did not prevent the son's election for sabhuku in 1995, although in this position he is responsible for the observation of traditional ceremonies in his area. In fact, through relatives, he is linked to both spheres: one of his sons-in-law is a Vapostori preacher, while another relative performs the traditional rain ceremonies.[31] Overall, at present the impact of Christian churches seems to be less penetrating in Serra-Chôa than on the Zimbabwean side.[32]

In the present socio-economic context, customary rights to land seem to emerge as the most problematic issue in Serra-Chôa. According to customary norms, every male member of the community has an inalienable right to land within the chiefdom (nyika), even if he has moved away. With the shortage of land prevalent in Zimbabwe, combined with its more developed infrastructure, there are a number of cases where people who fled across the border during the civil war are now reclaiming their rights to land in Mozambique, even though they prefer to keep their residence on the other side.[33] With increasing demand for fertile valley land for cash-crop production in Mozambique, and new generations of ex-Mozambicans growing up in Zimbabwe, the customary norms have become subject to dispute, and consensus is increasingly hard to establish.

New norms to resolve the problem have been sought from joint decisions by customary and modern institutions from both sides of the border, as the following case from Nhaterere demonstrates:

Case 3.

Mr. Mano, a son of a Mozambican family in Nhaterere decided to marry a girl from the Zimbabwean side. After marriage the new family settled on the Zimbabwean side of Caeredzi where the husband had also found a job, but continued to cultivate fields and graze cattle on the Mozambican side. However, their right to cultivate was disputed by other families residing on the Mozambican side of the border, who claimed the same land. Eventually the case was taken to a representative of the customary authority, who failed to settle the problem. It was subsequently taken to the formal state administration, which decided to create an ad-hoc committee to study the issue. This group, which consisted of representatives of both state and customary authority, as well as local elders, visited the area and found the fields in question under cultivation. After careful deliberation their verdict was that the family could not keep on using land on the Mozambican side without having residence there, but that they could harvest the present crop before giving up the fields.[34]

These kinds of new institutional bodies, even if temporary, can play a crucial role in bringing together community and state representatives to discuss solutions for situations in which the customary norms need revision in the face of increasing integration into the market economy. But, while important during an early phase of the process, they cannot be a substitute for permanent management institutions in the long run.

6. CHIEFS OR COUNCILS?

In most of rural Mozambique there are presently no functioning management institutions at the community level that have a formal legal status, because the local assemblies (locality and communal villages created in some areas by the 1977 and 1980 elections) have practically stopped functioning. Despite the prevalent community rhetoric adopted by the government, the new laws have not clarified the situation,[35] and the heavily boycotted municipal elections of 1998 covered only selected urban or peri-urban areas (33 in all). On the other hand, the role of customary institutions has been tacitly recognised in many districts, including Barué. However, even there the status granted to these institutions vis-a-vis the state and local population remains problematic.

In the list of duties for a chief compiled during an official meeting in Catandica in January 1999, the role of chief (régulo) - as a direct representative of state administration - was clearly emphasised: his task was to act as the link between the government and the community. He is not primarily a representative of his community towards the state, but the other way around. The objectives are given from above, and the chief's role is to carry them out.[36] In this context, the source of power for customary authority, and consequently its accountability towards its constituency, has been effectively disregarded. If carried out without due regard to local interests, these kinds of proposals threaten to revive the old problem of chiefs having a dual loyalty - towards both the community and the state - which plagued the colonial system of indirect rule.

The formal state laws and norms concerning natural resources have relatively little practical meaning for the local population.[37] For example, forest and wildlife legislation are little known locally, and hunting is a common practise throughout the year without any kind of licence. This is linked to the shallowness of the state's presence in the area, as the war destroyed a large part of the social infrastructure (schools, shops, administrative buildings). And, since the separation of the party from the state in the early 1990s, the administrative role of party functionaries has gradually diminished. Nowadays they are increasingly left with party matters only, while resource management is returned to the customary authorities.[38]

On the other hand, co-option of the customary authorities to serve the state administration has turned out to be a rather complicated process. Even the status and legitimacy of chieftaincies (nyika) within the Chôa area is disputed. While the official system recognises only one chief (régulo Macufa), people of the north claim that the head of Sanhamahwe lineage is also a chief (régulo/mambo), and not a sabhuku under Macufa.[39] There is also some discrepancy between the borders of political and spiritual communities. The political border of sabhuku Muswipa's area coincides with the international border following Caeredzi river, but the spiritual domain of a rainmaker called Sabadza, based in Nyamaropa (in Zimbabwe), includes the Caeredzi valley north of Nhaterere.[40] At the same time, influential local sources contest the right of the Macufa lineage to the post of paramount chief. They claim that the lineage displaced the ruling Sahatziro lineage only in the 1940s. The dismissal of the then ruling chief, Janeiro Sahatziro, by the colonial authorities is said to have resulted from a conflict of interests, which the other lineage manipulated to get their candidate, Macufa Bango, selected instead of the legitimate candidates from the ruling lineage.[41]

Some representatives of the traditional authority insist that the legitimate ruling lineages are those nominated by the Macombe kings, and not the others, which they claim arose from colonial era intrigues. On the other hand, representatives of the latter, who were not acknowledged by the Barwe rulers, tend to claim pre-Macombe heritage, and some claim both.[42]

Another complicating factor, in the present situation, is the Renamo policy to recruit local representatives of customary authority to serve their cause. While there was clearly some resentment towards the Frelimo policy to do away with the customary authority after independence, it seems that Renamo never enjoyed widespread support in the area. Its activities were limited to the destruction of government structures and the occasional raids on villages, which were accompanied by looting of moveable property, including cattle. Those of the local population who were caught by the rebels were forced to carry the loot to bases, but were usually allowed to return to their home areas afterwards.[43] The rebels were more interested in recruiting or, if necessary, capturing members of chiefly lineages. In 1986, Renamo kidnapped chief Sanhamahwe who was forced to stay in their base for one year. It also tried to kidnap the present chief Macufa, but he managed to escape. The sabhuku of Doro was also taken by Renamo and never returned.[44]

Despite the confusion created by early Frelimo policy and the civil war, the legitimacy of customary institutions seems to be quite high in Serra-Chôa. In every instance, the sabhuku was mentioned as the person to whom one should first turn in the case of a dispute. This practice is also encouraged by the local representatives of state administration. They rely heavily on sabhukus to resolve local conflicts, even though party secretaries still play a minor role. On the other hand, some kind of accountability mechanism towards the population seems to be functioning. For example, during this study it was reported that there were two recent cases where an unpopular or incompetent representative of the authority (one chief and one sabhuku) were dismissed by the people.[45]

The increasing influence of customary authorities is also manifested in the development of a local council created in 1998 by the chief of the administrative post. It originally comprised of local representatives of line ministries, including the chief of administration and his assistant/clerk, the head teacher, a medical orderly, the president of the locality of Nhauroa (a wealthy peasant from the area who is also the Frelimo party secretary), and two other relatively prosperous local peasants. The latter were nominated by the state authorities, but subsequently presented to the people for approval in a mass meeting. Customary authorities were not included, but were invited to participate when specific issues were discussed.[46] But, during the following year, the composition of the commission changed as the local members managed to strengthen their position. Now the chief of administration is the only representative of state administration, while the traditional authority is represented by Chief Macufa and one sabhuku. In some ways the new commission substitutes for a traditional chief's council (Makurukotas), which the present chief never established. Thus far, it has dealt with a number of minor disputes over marriage payments and damage caused by stray cattle, as well as a few land tenure conflicts. It has also been involved in the apprehension of cattle thieves. Aside from the latter, who were taken to the police in Catandica, many of the cases remain unsolved because the institution lacks formal status and the means of legal enforcement.[47]

While the technical efficiency of the above commission remains to be seen, it brings up another, and more interesting problem from the community point of view. The original Chôa commission, in particular, was dominated by state officials who were outsiders in the area, many staying for only two or three years, and who thus had no permanent links with the local population. For example, both the chief of administration and the medical orderly were transferred in 1999. The selection of community representatives - which followed an established Frelimo practice of nomination by state/party officials and subsequent approval by population - is another example of the same state-bias. In both commissions, those nominated from amongst the local community were all relatively wealthy, which raises the question of whether they are representative of the people. In Chôa, women, youth and the poor sections seem to be excluded. In addition, the enforcement dilemma has not been solved, because the new institutions lack a formal legal basis in municipal law and, along with other similar uncoordinated initiatives, they seem to face an insecure future.

7. CONCLUSIONS

The administrative borders established by the various historical state formations do not function as exclusive territorial boundaries between local communities in Serra-Chôa. In terms of social infrastructure, market services and natural resources, the customary socio-political units or even the formal administrative units are not self-sufficient. This interdependence is not limited to the national level, but extends to neighbouring Zimbabwe, which has a crucial role to play in local livelihood strategies. Geographical proximity, new economic opportunities and common socio-cultural roots turn out to be more meaningful than national borders for the local population. In this sense the emerging political community of "Mozambican people", which Frelimo sought to create through its nation-building programme, has failed to absorb or subdue the other factors, at least in the peripheral regions.

As noted by Yuri Lotman,[48] borders (both physical and mental) can turn peripheral areas into transition zones, where something that is external is transformed into something that is internal. Ethno-linguistically, the Serra-Chôa forms a highly homogeneous socio-cultural unit with strong historical ties to Zimbabawe. This linkage has also facilitated the creation of an incipient market-oriented production capacity - despite adverse official policies and virtual neglect of peripheral areas by the Frelimo government. Although the disturbance of the recent civil war is still strongly felt, and access to Mozambican markets remains difficult while Zimbabwean markets have declined rapidly, the number of commercially oriented producers has increased in the study area. At present, the economic strategy of this group of producers is based on a combination of the customary system of resource management, which gives them open access to land, and accumulation of financial and technical resources through a prolonged stay in Zimbabwe - either as migrant workers or war refugees. This emerging economic elite has recently been reinforced by a steady stream of people from neighbouring Zimbabwe in search of more land. Even though this kind of in-migration is not new as such, the commercial orientation of the newcomers has now become a typical feature.

In general, customary norms over natural resources are still prevalent in Serra-Chôa. However, reliance on internalisation of cultural norms as enforcement mechanisms risks losing their efficacy because the basis of consensus - traditional religion - is being challenged by the penetration of new religions and other "modern" world-views. The changes are most visible with respect to the status of traditionally protected areas like sacred forests, some of which are increasingly threatened by forest fires and tree felling. But the effects of the penetration of new people and ideas are not always antithetical to traditional values. For example, the growth of Vapostori religion in the area may actually reinforce traditional customs like polygamy, even though it is critical of some other aspects of traditional religion. However, the most acute changes seem to concern customary rules of land tenure, which are presently being redefined locally. The emerging rule-systems tend to be strongly rooted in local people's own experience and perceived needs, and as such they are capable of reacting to external factors of change. But they can also be perceived as a risk to national unity, and their compatibility with the principles and accountability structures of modern state administration remains an intensely debated issue.

In Mozambique those local institutions that are functional at the community level - that is, customary ones - are now acknowledged at the district level, but continue to operate without a firm legal basis. Creation of such a basis is actually rather problematic, for customary institutions are not uniform, and some do not even enjoy undisputed support within the respective communities, because members of the different ruling lineages are engaged in a continuing power struggle. This is manifested for example in the different and contradictory lineage histories that are used to legitimise access to traditional positions of authority. On the other hand, the representativeness of the new "ad-hoc" structures introduced in some areas by NGOs or local state authorities can also be questioned. Important sub-groups like women and the poor seem to be left out. As the new councils also lack formal status, their power of implementation is very limited, especially in cases that involve outsiders to the local community. The same constraint applies to customary institutions, which have at their disposal very few effective means to control and sanction outsiders. To a large extent this "administrative vacuum" results from lack of political consensus between the ruling party and the opposition concerning the status of local community and customary authority in local administration. In this case a political stalemate at the centre threatens to thwart development of innovative new mechanisms for community based resource management at the periphery. In the context of the rapid socio-economic transition, which is taking place in Mozambique, such failures in institutional development can lead to drastic social and environmental losses for the local population.

NOTES

1. The field data consists of two PRA exercises (covering Nhacapanga and Barauro villages) carried out in collaboration with the Mozambican Ministry of Environment and the District Authorities of Barué, and a number of individual interviews by the author with representatives of customary and modern state authorities and the local population. Altogether, over one hundred people were contacted. The interviews were carried out by the author in Portuguese or English, or with the help of a local interpreter in ChiManica. Reference to field data is made by indicating the location and date of the interview. The fieldnotes are in the author's personal archives. The names of individuals in the cases presented have been changed.

2. The study is based on fieldwork carried out during 1998-1999 in the Chôa highlands in the frontier area of Manica Province facing Zimbabwe (Map 1).

3. The most important ones are the Land Reform Law (Lei 19/97, Art 1, 10 12, 15, 16, 24), the Environmental Framework Law (Lei 20/97, Art 4, 7, 8), and the revised Law on Municipal Authorities (Lei 2/97, Art. 28).

4. Frelimo (Frente de Libertação de Moçambique) was the main national movement, and after gaining independence it created a Marxist one-party state in Mozambique. Only a few years later an armed opposition movement called Renamo (Resistencia Nacional Moambicana) was started with the support of the minority regimes in the neighbouring Zimbabwe (then Rhodesia) and South Africa. By 1994 Frelimo had officially renounced Marxism, and Renamo had turned into a legal political party.

5. Assembleia de Republica, 1977 Lei no. 1/77 Art. 14

6. Hall and Young 1997

7. Kloeck-Jenson 1998

8. The communitarian model was adopted an develop by technical advisors to the official Land Commision (De wit et al. 1995: 18-21; Tanner et al. 1998: 12-15, 19-21), While the modernists solution has been promoted by (among others) some leading activist from the leading NGO-based Land Campaign (Negrao 1996, 62; Negrao et al. 1999, 16, 28-29).

9. Newitt 1995

10. Instituto Nacional de Estatistica 1999

11. Central Statistical Office 1992, 19

12. Newitt 1995

13. Choa Sede 19/05/1999; Nhacapanga 22/05/1999, 16/11/1999

14. Instituto Nacional de Estatistica 1999, 5.

15. Nhacapanga 19/11/1998, 22/05/1999; Adhabi 13/12/1999

16. Chôa Sede 18/11/1998; Nhacapanga 19/11/1998; Barauro 18/03/1999

17. Mandondo 1998

18. Nhaterere 12/05/1999; Tewetewe and Chowzo 19/04/1999; Inyazonia 18/11/1999

19. Nhacapanga 19/11/1998, 22/05/1999

20. Nhacapanga 19/11/1998; Tewetewe 30/04/1999; Nhaterere 12/05/1999; Barauro 20/05/1999

21. Nhacapanga 22/05/1999, Adhabi 13/12/1999

22. Nhacapanga 19/11/1998, 22/05/1999, 03/11/1999

23. Catandica 29-30/01/1999

24. Barauro 18/03/1999; Tewetewe 30/04/1999; Nhaterere 12/05/1999

25. Chôa Sede 18/11/1998; Nhacapanga 19/11/1998; Barauro 18/03/1999; Nhaterere 12/05/1999

26. Cf. Mandondo 1998:11-14

27. Nhacapanga 12/11/1999, Chinda 14/11/1999

28. Chôa Sede 18-19/11/1998; Barauro 17-18/03/1999

29. cf. Matowanyika 1991

30. Nhacapanga 12/12/1999

31. Nhacapanga 13/05/1999, 03/11/1999

32. Cf. Matowanyika 1991, 162, 243

33. Nhacapanga 22/05/1999

34. Chôa Sede 25/05/1999

35. A similar relationship between formally recognised district level authority and local communities which have no legal status has caused problems in Zimbabwe, where it has hampered devolution of natural resources management powers down to local level (Matowanyika 1991, 245-247, 252).

36. Catandica 30/01/1999

37. Again the situation seems to be alike in neighbouring Zimbabwe, where enforced district level by-laws in communal areas are largely ineffectual as local communities often quietly ignore them.. Mandondo 1998, 17; Matowanyika 1991, 223, 239

38. Chôa Sede 18/11/1998; Tewetewe 30/04/1999; Barauro 20/05/1999

39. Nhaterere 12/05/1999; Barauro 17-18/03/1999; Tewetewe 30/04/1999

40. Nyamaropa 12/12/1999, 13/12/1999

41. Barauro 21/05/1999; Inyazonia 18/11/1999

42. Chigodo and Chowzo 18/11/1999; Nhacapanga 12/12/1999

43. Catandica 29-30/01/1999; Barauro 18/03/1999

44. Tewetewe 30/04/1999; Nhaterere 12/05/1999

45. Nhaterere 12/05/1999; Chôa Sede 13/05/1999

46. Chôa Sede 25/05/1999

47. Nhacapanga 04/11/1999

48. Lotman 1990

REFERENCES

Agrawal, A. and Gibson, C.C. "Enchantment and Disenchantment: The Role of Community in Natural Resources Conservation" World Development 27 (1999): 629-649.

Assembleia De República
Lei No. 1/77 De 1 De Setembro, Boletim De República, I Série - No. 101, Suplemento, 1977.
Lei No. 2/97 De 18 De Fevereiro, Boletim De República, I Série - No. 7, 2o Suplemento, 1997a.
Lei No. 19/97 De 1 De Outubro, Boletim De República, I Série - No. 40, 3o Suplemento, 1997b.
Lei No. 20/97 De 1 De Outubro, Boletim De República, I Série - No. 40, 3o Suplemento, 1997c

Central Statistical Office Census, Manicaland Provincial Profile, Government Printer, Harare, 1992

De Wit, P.V., Nyamuno, C.,- Shumba, M., Mufandaedza, L. "Land Use Planning Proposals, Part 1. Land Use Issues and Land Use Needs at Different Operational Levels." National Family Sector Agricultural Development Programme, Pre-Programme, 1995.

Hall, M. and Young, T. Confronting Leviathan: Mozambique Since Independence. London: Hurst &Co., 1997.

Instituto Nacional De Estatística II Recenseamento Geral Da População E Habitação 1997, Resultados Definitivos, Província De Manica. Maputo, 1999.

Kloeck-Jenson, S. "Locating the Community: Local Communities and the Administration of Land and Other Natural Resources in Mozambique." Paper Presented At The International Conference on Land Tenure in The Developing World with a Focus on Southern Africa. Cape Town, South Africa, January 27-29, 1998.

Lotman, Y. Universe of The Mind: A Semiotic Theory of Culture. I.B. London: Tauris &Co, 1990.

Mandondo, A. "The Concept of Territoriality in Local Natural Resource Management, and its Implications on Livelihood in Nyamaropa Communal Land." A Paper Prepared for the: WWF Programme Office-Zimbabwe, 1998.

Matowanyika, J. "Indigenous Resource Management and Sustainability in Rural Zimbabwe: An Exploration of Practices and Concepts in Common Lands." Ph.D. Thesis, Ontario: University of Waterloo, 1991.

Negrão, J. Alternativas Para O 'Sector Familiar': Notas À Política Nacional De Terras E Ao Anteprojecto Da Lei De Terras, Extra (1996) 17: 58-63.

Negrão, J., Brito, L., Da Silva, P. Parecer Técnico Ao Projecto Do Lei De Florestas E Fauna Bravia. Maputo, 1999.

Newitt, M. A History of Mozambique. London: Hurst & Company, 1995.

Tanner, C., De Wit, P. Propostas Para Um Programa De Delineação Das Comunidades Locais. Document Presented to The National Seminar on Community Land Delimitation and Management. Beira, Mozambique, August 12-14 1998.

Virtanen, P. Research Field Notes are in Author's "Personal Archives." Reference to Field Data is Made By Indicating the Location and Date of the Interview. Villages Covered: Adhabi, Barauro, Tewetewe and Chowzo, Nhaterere, Inyazonia, Catandica, Chinda, Nyamaropa, Chigodo and Chowzo, and Choa Sede and Nhacapanga. 1998