This discussion of collective forest resources management in Latin America covers a wide and scattered literature. In "A conceptual framework", we discuss a number of theoretical and practical concerns or themes around the topics of common property, collective management and deforestation.
Local systems of forest management are examined next. In Latin America, indigenous and traditional management systems are still common, as exemplified by cases from Mexico and the Amazon. Traditional systems are well-adapted to ecological conditions but are under severe pressure by the rapid socio-economic changes that the countries of this region are experiencing. These systems offer potential sources of knowledge for the management of natural forests and of social organization for collective management. The insights they offer may be improved by research and experimentation.
Externally linked systems of forest management are taken up with examples from six countries. Commercial models of forest management involve the large-scale harvesting of natural forests and the establishment of a few industrial plantations. State agencies are often involved in leasing or managing commercial concessions for the harvest of forest resources.
In recent years, there has been interest in the region in introducing anew approach to collective forest resource management. This approach begins with the premise that it is very important to involve or link local communities in the direct management and control of forest resources. As will be seen, some examples express local initiative directly in the formation of innovative new systems. Several examples of externally linked systems are presented to illustrate both the development and difficulties of this approach. They include forest management programmes in Mexico, reforestation schemes in Guatemala, cooperative sawmills in Peru and Bolivia and extractive reserves in Brazil and Honduras. Most of these programmes are still in their early stages and have not been evaluated systematically. Research is needed to better understand the variables involved in them.
In the section "Further considerations", ten potential research questions are introduced which, we suggest, should guide future study. The chapter ends with annotated References.
Latin America is the most forested area of all developing regions, with 966 million ha of forests, covering 48 percent of the land area. More than half - 57 percent - of the earth's remaining tropical forests are found in this region .
Forests have important economic, ecological and social roles for national development. However, many Latin American countries have neglected forestry and forestbased activities in their development plans. In many cases, the economic contribution of the forest-based sector to national development has been minimal compared to its potential . At the other extreme, several countries have promoted permissive policies for forest exploitation and ambitious and unsound colonization programmes for the conversion of forests into agriculture . Forested lands often serve as an escape valve for a hungry population without access to agricultural land and are subject to aggressive colonization programmes with their highly visible negative effects.
One obvious consequence of these patterns is the rapid rate of deforestation in Latin America, the highest of the developing world. The existing forests disappear at a rate of about 1.3 percent per year, compared with 0.9 percent in Asia and 0.6 percent in Africa . This is equivalent to more than 12 million ha destroyed every year. In 1987, in Brazil alone, 9 million ha of tropical forest were destroyed.
Deforestation contributes to such problems as soil erosion, diminution of water sources and the elimination of wildlife habitats. There is also a growing concern for the effects of deforestation on global warming and the loss of genetic material. Such difficulties affect the ability of people who have traditionally depended on the forest for their livelihood to maintain their cultural and economic integrity.
One approach to resolving the problem of deforestation has been the management of forests by national governments. Another approach has been leaving the responsibility of forest management in the hands of those who own or use them, collectively or individually. Recently, certain programmes have recognized the importance of incorporating community groups in self-help programmes to be able to continue making use of forests and to contribute to correcting the environmental and social damage resulting from a massive tree-cover removal.
To distinguish the different types of collective forest management, certain terms need to be defined.
Community forestry is defined as:
any situation which immediately involves local people in a forestry activity. It embraces a spectrum of situations, ranging from woodlots in areas which are short of wood and other forest products for local needs, through the growing of trees at the farm level to provide cash crops and the processing of forest products at the household, artisan or small industry level to generate income, to the activities of forest dwelling communities ... [36, p. 42].
FAO [36, p. 44] identifies three approaches to community forestry:
Regarding the control or ownership of forest resources, these activities have the following sub-divisions and characteristics:
In this chapter, we deal primarily with collective forest management in which the custody and responsibility for creating, harvesting and improving the forest or the trees are conferred to the local organized community or user groups which cooperatively participate in project planning and implementation. This "includes projects based on the activities of community organizations such as schools or cooperatives which may establish nurseries or plant trees in small or fragmented private woodlots" [36, p. 46].
There are three basic strategies for forest management: plantation forestry, natural forest management, and agroforestry.
The latter two are addressed in this discussion. Plantation forestry is still limited in Latin America and generally not practised under collective management by small farmers and indigenous people.
Natural forest management is the controlled and regulated harvesting of different forest resources combined with protective and reproductive measures to sustain or increase subsequent stands and wildlife.
Agroforestry is a sustainable management system that combines agriculture and/ or animals with tree crops and/or forest plants on the same unit of land, either sequentially or simultaneously [33, p.1].
Forest management, technically speaking, includes many possible components, including silviculture which is indispensable for large plantation forests. Synnott [ 112, p. 741 distinguishes between the characteristic tools of silviculture and those of technical management. Among the former, he includes regulation of shade and canopy opening, treatments to promote valued individuals and species and to reduce unwanted trees, climber cutting, refining, poisoning, enrichment and selection. For Synnott, the term "management" involves the setting of management objectives, yield control, protection, working plans, felling cycles, concessions, roads, buildings, boundaries, sample plots, prediction, cost control, annual records and the organization of the silvicultural work.
In this discussion, we used a concept of forest management that allows for the many indigenous resource management strategies that do not necessarily adhere to the technical models described above. In our case, the term refers to how rural people harvest, use, take care of, reproduce and improve their forests (or trees) and associated resources, such as wildlife, water and plants, in order to attain sustainable long-term yields. This flexible use of the concept is necessary due to the diverse ways in which communities in Latin America utilize forests.
The definition of forest includes dense primary forests, land that is semi-densely covered with trees and brush and trees that are "dotted about the rural landscape, around houses, along field boundaries and roadsides, and in communal grazing areas ..." [36, p. 10].
The majority of the initiatives involving community management of forest resources is very recent and has not been the subject of rigorous evaluative research. Partial evaluations have been done which help identify the mistakes of certain approaches and enhance those projects or certain features of projects that have been successful.
Management is closely linked to the rules and conventions that determine access to a resource. One of these rules is known as property. Property can be defined as a social convention about the security of future benefit or income streams . Several researchers [e.g. 89,79] have explained property relations as divided into four categories: private property, state property, common property, and open access or no property.
Southgate and Runge  note that, in Latin America, the state makes extensive claims on natural resources. These claims often outstrip the government's capacity to manage resources or even to ensure that its claims are honoured by the public at large. The state has not been able to manage forests well, which leaves a management vacuum and leads to situations of open access.
Oakerson  has suggested a model to analyse and explain the main factors involved in the management of common property resources (CPRs). In its simplest form, the Oakerson model is based on understanding the relationships between the physical characteristics of the resource, the decision making rules of the group or users involved, the patterns of interactions resulting from the appropriation and use of the resource and the outcomes of this process. Blaikie and Brookfield  have modified the Oakerson model to explain the dynamic interactions and adaptive changes when a resource is managed under a communal (or collective) regime. I
The concept of property comes from the liberal European tradition. It is usually associated with a piece of land or an object, although it may also include rights to factors that determine income streams such as patents or intellectual property. Thus, property refers to the triadic relationship between an individual, an object or a factor and the rest of the society; i.e. property needs to be socially sanctioned, usually through the state. This concept is deeply entrenched in Latin American societies as a result of the Iberian conquest and reinforced by the liberal reforms of the 19th century.
In a study of land tenure legislation of Mexico and Central America, Rendon-Cano  concludes that in all cases, the social function of property determines, in principle, certain limitations on private landownership. Alongside the fundamental concepts of private property, there is a strong tradition in most Latin American countries of an abiding social interest in private property. In addition to this penetration of social or collective interest in all property, there are numerous cases of collective or communal property of land in several countries of the region.
In analysing the Latin American cases, we have kept in mind the importance of the property relations and the access, use and management rules associated with forest resources . Issues related to land tenure, property rights and forest legislation are of paramount importance when assessing the viability of collective management of forests [161.
Many of the arguments about deforestation in Latin America and the policy options for dealing with this issue can be summarized in terms of the following ten themes. The first eight describe the underlying situation; the last two suggest alternative policy options.
First, since colonial times, there has been an expansion of export-oriented agriculture, due to land policies to stimulate crop export. This has resulted in the clearing of arable land and the displacement of indigenous peoples from control over the land.
Second, the development of forest resources for economic development were initially of secondary importance. Forests were traditionally considered as an agrarian frontier, i.e. places to clear in order to plant crops or raise cattle.
Third, by occupying land considered valuable for agriculture, cattle ranching or mining, indigenous peoples have been pushed into ecological zones considered of secondary importance to the agro-export sector, i.e. into the forests.
Fourth, this form of displacement, combined with population growth, has resulted in peasant occupation and clearing of forested lands, often in competition with indigenous agroforestry systems. Colonization programmes have pushed to reduce social conflicts due to highly skewed land-holding patterns in agriculture.
Fifth, after World War II, the increased economic value of some forest products encouraged efforts to exploit them commercially. The state played a major role in stimulating private enterprise development of forest resources mostly through tax and export incentives. In some cases, the state has directly controlled timber extraction and exportation of wood products. In general, the state has developed substantial regulations and involvements in forest resources management.
Sixth, industrial economic development strategies have also stimulated non-forest product development of forested areas, generating little inclination to sustain forest resources. Product development includes mining, oil extraction, road building and hydroelectric projects.
Seventh, people living in forested areas are typically limited to using small plots of land for subsistence crops and confined to the use and sale of forest products of no great value. In some cases, these forest people work as labourers for large-scale enterprises such as timber extraction, mining, oil, infrastructure installation, etc.
Eighth, commercially valuable resources found in forest areas are now being exploited by commercial enterprises; the forest people remain uninvolved in their management and are unable to enjoy their benefits. Commercial forest enterprises have multiple options for investment, tending to move their capital according to perceived risks and expected short-term returns. Forest resource extraction is one option, which in comparison to other investment options is often not very attractive, except for short run profits or for speculation. The forest people dependent on its resources for their livelihood, are essentially separated from its benefits, unable to enjoy the value of economic goods extracted from it. The result is little incentive to maintain the forest.
Ninth, one policy option is to stimulate large-scale private enterprises by removing state regulations and involvements in forest resource management, encouraging investment in forest resources and guaranteeing profits and the freedom to move capital into the most profitable and secure activities.
Tenth, a second policy option is to empower the forest people, themselves, to manage forest resources directly, thereby encouraging them to maintain the forests both for its ecological value and its economic benefits. The annotated bibliography covers experiences with this option in Latin America.
For indigenous and Mestizo groups in Latin America, the forests have traditionally provided areas in which to hunt animals and insects and to gather plants and inorganic materials. Forest plants are used as building materials, medicine, food, oils, perfumes, pigments, dyes, gums and resins. Insects are gathered as protein sources and are used for natural pest control. Animals are hunted for their meat, hides, feathers, furs and for ceremonial purposes. Management of garden sites in tropical forest regimes by indigenous peoples also provides additional food sources for forest wildlife species which, in turn, leads to an increase in the abundance of game animals for hunting . Access to forest resources permits the development of methods to allow for income generation, which decreases dependency on state-supported assistance programmes.
The cultural ecology of indigenous and Mestizo peoples demonstrates that they operate resource management systems that effectively influence the evolution of highly diverse forest environments. There are many examples in which local residents conceive of forests as much more than a source of timber and fuelwood. Their concepts of management are comprehensive, relating not only to the management of trees but to a wide variety of naturally associated forest resources in combination with annual crops and livestock. In this sense, agroforestry is not a new procedure but, rather, a new name for the old practice of mixing trees and shrubs with garden, annual and periodic crops and animal husbandry.
Hecht and Cockburn  effectively show that most of the Amazonian forest is the product of human activity that manipulates forest environments for human purposes. The forests are managed, and the mechanics of this management are understood through the cultural ecology of indigenous and Mestizo people.
There are many examples in the Americas of natural forest management and agroforestry carried out by local communities. Examples of this are: swidden-fallow in the Amazon by indigenous people, and Huastec agroforestry in Mexico by Mestizo people.
Indigenous peoples in the Amazon manage forest resources in several ways. Many swiddeners abandon their swiddens and move on immediately after the land has lost productivity. Amuesha and Bora Indians of the Peruvian Amazon, however, continue certain secondary management practices for prolonged periods after the land has been abandoned for cultivation .
The Amuesha Indians inhabit the Palcazu Valley in the Selva Central of Peru. The Bora Indians inhabit the north eastern jungle near the city of Iquitos. The swidden-fallow management system of both groups is a form of agroforestry "involving a combination of annual crops, perennial tree crops, and natural forest re-growth" [34, p. 1]. In this system, the transition between swidden and fallow fields is not dramatic. Rather, management continues throughout the period of fallow; "... viewed properly, a swidden site is never completely abandoned as a resource zone" [34, p. 45]. This blend between primary swidden harvest and secondary fallow harvest has been described as "a continuum from a swidden dominated by cultivated plants to an old fallow with few or no cultivated plants" [33, p. 2].
In the Palcazu Valley, the Amuesha Indians have developed a complex and diverse agroforestry system based on adaptation to five distinct ecological zones where they grow and manage approximately 98 different species of forest plants and trees [95, p. 212]. In utilizing the diverse ecological zones of the valley, the Amuesha employ a complex land use scheme based on an intricate knowledge of soil type, fertility, natural vegetation and cropping potential . The limited availability of alluvial soils in the valley and the extensive distribution of acidic soils with a high aluminium toxicity makes it necessary to grow crops in five areas [95, p. 197]:
In the highly fertile lowland sites, there is a greater diversity of crops grown and a higher intensity of crop production overall. In contrast, the upland areas are less diverse and have lower crop variation. Although intercropping is practised in all of the five zones, the dominant crops are groundnuts and beans on the beaches, maize, cassava and plantains on the alluvial floodplains and rice and upland cassava in the upland areas. Fruit trees are also grown and are "left uncut in a swidden or are planted under ... maturing plantains forming an agroforestry system to an enriched and utilized fallow" [95, p. 189].
In a survey of 31 Amuesha yard gardens, Salick  found that approximately 50 different plant and tree species are grown. Planting cycles are carried out in stages in which different dominant crops are rotated on a given field. The crop rotation cycle ends when natural successional vegetation becomes predominant during a utilized fallow period in which left-over crops and forest plants are extracted. Due to the heavy rainfall, timing of cropping systems is extremely important in the Palcazu Valley. The Amuesha believe that successful crop production is often a matter of timing and luck. Heavy rainfall throughout the year makes field burning difficult and vegetation that does not burn is either left in the fields to be burned again later or is left as mulch. Annual crops (maize, beans, rice) are planted during the summer, requiring a spring burn when the vegetation is wet and there is less sunshine. In the fall, primary and secondary forest is often burned for a rotation of rice and cassava.
Management of tree species for timber production is not done on a consistent basis. Wood for fuel is produced in yard gardens along with a great variety of other items but there is no deliberate effort to manage specific tree species for timber production .
In their study of swidden-fallow management by the Bora Indians, who practise a similar system of agricultural and resource management in northeastern Peru, Denevan and Treacy note that "trees used for firewood include those specifically mentioned for that purpose; however any tree, especially if felled while preparing a swidden, is potentially a fuel source" [34, p. 44].
The Bora manage some tree species by selective cutting during swidden clearing. Valuable species of timber, such as tropical cedar, are often spared by the Bora along with various palms, while other trees are utilized. Trees that are spared from clearing are left in or at the edges of recently cleared fields . Like the Amuesha, the Bora plant fruit trees and other minor crops like manioc among swidden field crops.
Fields are managed in varying levels of intensity. It is noted that owners of swidden fields are recognized by neighbouring farmers as owners with harvest rights over the fallow that succeeds the swidden but that the ownership weakens over time. Transitional fields, for example, are still intensely managed in a stage when forest regeneration is just beginning. Fallow areas, described as orchard fallows (Fig. 4-1) which still retain productive tree species and some managed crop species, are retained and managed for periods of up to ten years with decreasing intensity of usage as the fallow increases in age [34, p. 12-15].
One 19-year-old forest fallow was identified that was originally cut from mature forest and planted with an estimated 11 tree species, including several species of fruit trees (Fig. 4-2). Over 22 useful tree species were identified in the plot, including trees used as construction materials, for medicinal purposes, as food, for artisan materials, salt extraction and for pitch to seal canoes .
The Amuesha and Bora both utilize an agricultural system that is in reality a conversion of a short-term cropping system into a long-term agroforestry system [34, p. 45]. The management of forest resources combines the production of agricultural crops with selective tree cutting and management for a variety of purposes, one of which is timber harvesting. Although much of the management is done by single families, "farms have no permanent boundaries; the fields operated by one family are each year scattered in different parts of the land under the jurisdiction of the community" [2, p. 65].
The Mestizo Haustecs live in the south eastern part of the state of San Luis Potosi, Mexico. Ancestral occupancy of this region dates back three thousand years. Alcorn [1, 2] describes an example of social forestry from here. The Haustec operate farms on communally owned lands, in communities and ejidos but most live on their own landholdings of from one to 15 ha. The Haustec-managed agro-ecosystem is sustainable enough to meet farm family needs and, at the same time, allows for forest regeneration and the protection of natural resources for future use. Their primary source of cash income is the sale of raw sugar or coffee. They also sell honey, fruits and a variety of other minor products (milk, eggs, poultry, wood). For self-consumption, they produce maize, a variety of domesticated and wild foods, construction materials, herbal medicines, craft materials and fuelwood.
The Haustec create patches of telom, or managed primary and secondary forests mixed with introduced species like coffee. Some are cycled into milpa swiddens, a type of agroforestry that integrates maize production with secondary successional forests in Middle America. Each telom is from 0.25 to 3 ha in size but, when viewed from a distance, the groves appear to be quite extensive. One farmer's telom borders another's, creating managed forest groves of irregular shape covering an area of 25 or more hectares. Some teloms have been established recently while others have existed for at least 80 years, indicating that the land use is sustainable.
A typical telom contains over 300 species. Construction materials come from 33 species, utilitarian items such as tools, baskets, furniture, musical instruments, fuelwood are derived from 65 species, food from 81 species, medicine from 221 species and livestock feed from eight species. In total, 90 percent of the plant species occurring in Haustec telom, groves have use values [1, p. 188].
Alcorn affirms that "... the Haustec system could, with local modifications, be integrated as a farm component of smallholders using forested lands anywhere in the world" [1, p. 185].
In the past few decades, a variety of local collective forest resource management experiences have emerged in Latin America. The emphasis has been on the establishment of local community control over forest resources. Some of the new approaches to management have originated from outside, while others reflect local responses to new conditions combined with external support.
In this section, we examine systems of locally based, collective resource management systems in six countries:
We conclude with a discussion of systems and programmes derived from outside initiatives and of local initiatives with outside support.
There are approximately 37 million ha of land covered by forests in Mexico . This amounts to nearly 20 percent of the total land area, more than twice the amount of cultivated land. About one third of the rural population of 10 million people lives on these forested lands, and 70 percent of the forest resources are on lands designated as ejido and comunidad.3 In both ejidos and communities, users can benefit from the use of the land but cannot legally sell or lease it.
Despite its vast forest resources, Mexican forestry demonstrates poor performance in three ways: low forest productivity, high deforestation and widespread poverty.
The potential contribution of forestry to the national economy is high but the forestry sector represents only one percent of the gross national product. Every year, Mexico imports close to one million tons of cellulose and paper. In 1988, the import bill was nearly one billion United States dollars. Mexico's rate of deforestation is among the worst in Latin America, only after Brazil and Colombia. The forests are being destroyed at a rate in excess of 1 percent (somewhere between 400,000 and 615,000 ha), annually . And, in spite of the fact that most of the forest resources are one or community lands, the peasants, by and large, have not participated directly or benefited greatly from forest exploitation.
The exploitation of forests is a complicated and capital intensive operation. It requires roads and transport infrastructure, expensive equipment, skilled labour, ready access to markets and so forth. But without adequate financial and political support and proper organization, forestry activities cannot realize their productive potential or serve their protective and social functions.
Since the 1940s, Mexico has promoted the harvesting of natural forests in support of forest-based industry. Under this model, the government granted 25-year concessions to private firms and to state-owned enterprises, i.e. to Unidades Industriales de Explotacion Forestal. These private and public firms have carried out forest exploitation in large areas all over the country, many on ejido and community lands.
One of the consequences of the model is its negative impact on the peasants. It typically excludes them from control over their land, from participating in the productive process and from direct access to the economic benefits of forest exploitation. Therefore, there is no incentive for peasants to manage the resource with a long-term vision. The concessionaires, in many cases, have tried to obtain the maximum benefit in the shortest time, without concern for reinvestment of profits in reforestation or proper forest management as a renewable resource.
This model has also encouraged illegal exploitation and corruption, since government officials who control permits and the exploited areas also benefit from illegal extraction. Interest in short-term gains prevails over long-term sustainability, although there are exceptions.
Another effect of the concession model has been the inappropriate use of fragile lands. Since peasants realize no direct benefits from forest exploitation, they have come to value the land for more immediate gains by dedicating it to cultivation and other nonforest-related uses. With increasing population comes the need for increased food production.
The peasant response is to over-exploit scarce arable lands and to extend agriculture on to new and marginal land cleared from the forest. It takes only a short drive along any road through the forested areas of the country to see many peasant farm plots located on previously forested steep slopes quite inappropriate for cultivation.
By the end of the 1970s, it was clear that the situation was not sustainable and that, in the long run, the resource base would be destroyed. One alternative model of forest exploitation has recently emerged, based on the premise that it is crucial to find a way to incorporate the rural population into the management of the forests. Steinlin [ 108] and Janka  describe the fundamental elements required to address this "formidable challenge". Among the most important elements are a re-conceptualization of the management of forests and of the role and place for technical services offered by the government in order to incorporate local communities into the productive process. A shift in policy and practice of this magnitude requires the coordination of many actors, most especially the peasants themselves.
Peasant forestry organizations since 1975
One of the strong forces promoting peasant participation in forest management and benefits has been the emergence of various organizations in both the ejidos and communities since 1975. Since then, at least 23 organizations have been formed in 13 states to manage forests as an important natural resource and economic alternative. The most complete and current summary of the nature and status of these organizations is found in the Declaration of Tecpan Concerning Forests and Jungles of Mexico [31 ]. This declaration came out of the VIIIth National Meeting of Forestry Peasant Organizations held in the town of Tecpan, Guerrero. It was widely published in leading newspapers in Mexico on 23 June 1988.
At the Tecpan meeting, participants concluded that the main characteristic of the forestry peasant organizations is that they constitute a group committed to "developing a process of struggle and organization to recuperate the effective control of the forests and appropriate the productive processes and benefits derived from forest activity" [31 ]. These organizations have been holding meetings since 1984 in order to provide mutual support and maintain a network through which to exchange information and experience. They are concerned with the rapid destruction of the forests of Mexico and consider the key to forest protection and industrial development the "direct, profitable and autonomous participation of the peasants in the entire processes of production, protection and management of the forests" .
The peasants' demands go further than gaining simple access to the resource. They seek to assume full control of the productive processes. The declaration calls for a halt to forest concessions, which are considered a form of colonialism and a backward structure that work against long-term forest development. The peasants are asking for the transfer of state-owned enterprises and the technical services previously provided by the government to their direct control through the peasant organizations.
The peasant organizations have also called for a unified and consistent government policy in the forestry sector. They demand compliance with the new forest legislation, approved in 1986, which incorporates many of their demands.
But how feasible is this new approach to local management of forests? The review of a few case studies, which follow, will help analyse the elements that affect its implementation. Only the most relevant examples are given, although there are other reports of similar experience in various parts of Mexico [25, 45, 59].
Pilot Forestry Plan of Quintana Roo
One of the most successful examples of the application of the new approach to forest management is found in a peasant-generated Pilot Forestry Plan located in the southern part of the state of Quintana Roo on the Yucatan Peninsula [41, 42]. The plan has been in operation since 1983 and includes ten &s with a total land area of 300,000 ha, of which 40 percent is dedicated to permanent forest management. The plan area affects 1,989 d& members who, with their families, represent a total population of 14,000 people.
As background information, the area of the pilot plan is part of a larger forest industry concession of 500,000 ha, granted to Maderas Industrializadas de Quintana Roo (MIQRO) in 1954. The MIQRO concession included national lands and the lands of six existing ejidos. The concession was based on a technically sound silvicultural plan with a 25-year rotation cycle. It concentrated on the selective extraction of precious woods such as mahogany, and cedar. During the MIQRO concession period, most of the ejidatarios originally present in the area subsisted by swidden agriculture, supplemented by harvesting chicle from the forest and occasional employment with MIQRO.
The concession plan met with lack of interest and appreciation for the resource on the part of the peasants (as noted above) and with an inability of the MIQRO industrialists to react flexibly or constructively to change. In the 1960s, the government promoted a colonization programme in the region so that, by the end of the concession period, the number of ejidos had risen from six to 60, on the same 500,000 ha of concession land. Regardless of the demographic and economic change this implied, the industry did not adjust its management plans or style and, as a result, somewhere near half of the original forest was cleared by the end of the concession.
In 1983, when the concession ended, a team of individuals decided to apply some new ideas of forest management and implemented a new project based on a Pilot Forestry Plan in ten of the area's ejidos. The project required the coordinated action of numerous parties to address the necessary technical, organizational, political and financial issues for its implementation [ 110]. The government of the State of Quintana Roo and the undersecretary of forestry attended to the requests of the peasants and joined ideas and efforts in order to settle the ending of the forest concession to MIQRO, change the rules of the game and provide technical and organizational support to the ejidos.
Full political support of the then government of the state and the under-secretary for forestry was crucial in solving a series of political impasses. The project required a committed technical team willing to work with and not over the peasants. Flexibility was essential in order to try innovative technical solutions to increase the project's revenues and the efficiency of the operations.
The response of the peasants was also very important, especially during the period of trial and error. Only after three years of operation was the first forestry peasant organization in the state created. It is called the Society of Ejidal Forest Producers of Quintana Roo (SPFEQR) and is one of the most successful in the country. Each of the ten ejidos members is autonomous in its organization, operation and marketing. The Sociedad Civil (SPFEQR; a type of cooperative) is a center of services offering technical forestry assistance and support in financial, information, legal, marketing and political procedures.
The project team and the peasant organization point to a variety of indicators of their success. In the 1986-1987 season, the society's gross revenues were worth US$ 1.25 million. Most of these revenues were reinvested to capitalize on the enterprise. The potential benefit of the enterprise to local ejidatario was becoming clear. The technical innovation of increasing the number of species exploited and other processing operations had the effect of increasing the number of jobs available to members. The forest had acquired a tangible value in the perception of local residents who were now becoming more concerned with the conservation and judicious use of forest resources than ever before. In summary, once given control over forest resources, the local people were induced to do things themselves in new and more sustainable and beneficial ways that will not be easily co-opted or reversed.
Union of Forestry Communities and Ejidos of Oaxaca
Compared with the tropical region of Quintana Roo, the application of the peasant forestry organization model in the temperate regions of the country is more complex. In the temperate zones of Mexico, for example, where population pressures are greater, the forest industry is more deeply entrenched and has more well-established vested interests. Some promising attempts are, nonetheless, being made there.
In Oaxaca, a paper factory, the Fabrica de Papel Tuxtepec (FAPATUX), received a concession in the 1950s to exploit the high pine forests of the Sierra Juarez. Wood production in the region increased from 46,000 cu m in 1976 to 250,000 cu m in 1985. Without reforestation, the forested area in the region decreased from 43 to 28 percent during the same period. All indicators for social well-being show that area as the lowest in all of Mexico. About 46 percent of the Indians who live in the 69 counties (municipios) of the Sierra Juarez were forced to migrate out of the region, an indication of the lack of economic opportunity in the countryside.
In the early 1980s, the peasants and Indians of the region started to organize themselves to oppose a renewal of the concession to FAPATUX. At the same time, another similar concession ended in the Sierra Sur of Oaxaca to the company Forestal de Oaxaca. With the end of these two concessions, eight communities from the Sierra Juarez and the Sierra Sur joined forces in an attempt to gain autonomy and, after three years of trial and error, formed the UCEFO as a Sociedad Civil (a type of cooperative) to provide support services to their members [61 ].
During the brief period since UCEFO's creation, the peasant communities have accomplished the important goals of gaining direct control over local forests and achieving independence and autonomy. An outbreak of insect pests in the forest served as the catalyst to encourage peasants to take direct action in solving their problems. Currently, they are managing their own technical forest service, training their own technicians from member communities, contracting directly with industry and the federal government and capitalizing the organization. Members of the participating communities have found that the benefits of group forest exploitation are substantial and that organizing efforts themselves has enabled them to gain control over the benefits.
The members of UCEFO, while still learning to guide the process, now see that once peasant control is initiated it tends to develop momentum and strength. When allowed to flourish, the peasant organizational innovation provides a demonstrable and viable alternative to the previous concessions model that excluded peasants from forest resource control, management and benefits.
Commercial Forestry in Santa Cruz Tanaco, Michoacan
Not everything goes smoothly for the peasant communities and didos trying to gain control over their forest resources and destinies. In the state of Michoacan, for example, the community of Santa Cruz Tanaco was divided over the process. In 1963, a private entrepreneur established a sawmill and obtained rights to cut timber from community lands. The principal beneficiaries, however, were a few community authorities who enriched themselves by it. In 1972, a new era of peasant control started in Tanaco when local people rose up against the corruption and expelled the leaders from their positions of authority.
The new leaders of Tanaco, with some external support, formed a local enterprise to assure that the sawmill and the forest were managed directly by the people, with benefits distributed more equitably within the community. While not all of the conflict has been resolved, most community members have realized benefits from the new system of forest management. Many have raised their standard of living, and all have learned valuable lessons from their attempts at organizing to manage the forest . Further experimentation and experience will undoubtedly be necessary to solidify the potential gains.
A 1972 study conducted by the Organization of American States (OAS) drew attention to the alarming deterioration of Guatemala's soil and forest resources, particularly in the highlands or altiplano . As a result of this study, Guatemala's National Forestry Institute (INAFOR) signed an agreement with the American Peace Corps and with CARE's food-for-work programme to begin a grass roots reforestation and soil conservation programme aimed at helping Guatemala's subsistence farmers help themselves. The INAFOR/CARE/Peace Corps programme was part of a national effort to improve the use and condition of soil, crop, pasture, range, forest and watershed resources. The intent was to encourage self-sufficiency and self-respect among Guatemala's subsistence farmers and to help define a new future for the country's vital agricultural and forestry activities.
The programme was begun in 1974 with four pilot projects, and was soon expanded to 25 highland sites. As it gained momentum and attention, it was further expanded to 50 communities by 1978. In addition to the volunteers and food-for-work incentives, INAFOR provided salaries for administrators and promoters, as well as for the transport of materials and food rations. In some communities, the American volunteers and Guatemalan promoters helped create local groups to join the programme but in most they worked with already existing cooperatives and church groups. The farmers were trained in sound conservation practices .
In 1987, the donor agencies carried out several mini-case studies in the area. One examined the nurseries project at Santa Apolonia, providing a typical example of the soil conservation and forest management programme.
Santa Apolonia belongs to Region V of the Department of Chimaltenango, in the humid lower mountain life zone. This region is characterized by low rolling hills and fertile soils. The ICCP project was initiated in Santa Apolonia in 1975, and at the time of the study, employed one promoter on salary from INAFOR. Following a plan to switch from a focus on large central tree nurseries to smaller decentralized nurseries (viveros volante), the Santa Apolonia promoter was working to provide assistance to 13 outlying community nurseries.
One important aspect of the programme was to encourage participants in each of the thirteen communities involved to set their own goals. The Patzaj village committee, for example, set its 1987 production goal at 10,000 seedlings, using both the bare root method and plastic bags. According to the committee chairman, "There is no problem getting people to work in the nursery. Everyone participates" [74, p. 32]. This attitude is notable because no food had yet been provided them. The food-for-work incentive was scheduled to begin a few months later. And, according to a committee member, each family was receiving seedlings according to its needs. Some had set a goal of planting family forest plots with up to 25 trees.
From an evaluation conducted in 1987, it is clear that some leaders think the food-for-work programme was an important incentive for farmers. It helped them come together to work in conservation activities for their own benefit (1987:24). But one volunteer was sceptical of the food-for-work programme and commented his view of the food-for-work incentive structure saying that he would rather see low production numbers associated with good community participation than high production achieved by giving bribes. This particular volunteer used fruit trees and coffee trees as incentives for farmers working in the tree nurseries and adopting good soil conservation techniques.
Beyond this difference of opinion about incentives, there are other factors to consider. One CARE representative has said that Guatemala is becoming increasingly aware of the role of natural resources and small farmers in the country's social and economic structure. He stressed that "only by establishing natural resource policies that can survive changes in government and work for the benefit of the people will there be a chance for sustained development" [74, p. 25].
In both Bolivia and Peru, the development model for forest resource management focused on the organization of forestry cooperatives. In two projects in Bolivia  and Peru [49, 68, 99,109] sawmills were built for income generation, combined with management for sustainability. Both projects were designed, in part, to stem encroachment by colonists on common lands. Indigenous groups were helped to form cooperatives, receive aid in sawmill management and engage in selective timber extraction on common forest lands. Proceeds from timber sales were used for local community development.
In Bolivia, Davis  analysed a sawmill cooperative initiated by a private voluntary organization (PVO) to assist the Ayoreode Indians in the department of Santa Cruz. It illustrates the difficulty in applying market-oriented development strategies to indigenous groups that have little experience in marketing. The project was funded by the Inter-American Foundation and developed by a local Bolivian PVO called Ayuda Para el Campesino del Oriente Boliviano (APCOB). It formed part of a programme of indigenous development and resource protection promoted by the Central de Pueblos y Communidades Indigenas del Oriente Boliviano (CIDOP). Initiated in 1982, the project's purpose was to provide "ecologically sustainable development for ... one of the poorest, most culturally depressed indigenous groups in the eastern lowlands of Bolivia" [30, p. 3]. It was also designed to demonstrate that indigenous communities can protect their natural resources and develop sustainable economic activities.
To obtain timber for the sawmill, the Ayoreodes Indians managed and harvested approximately 30 species of trees in the forest. A self-management committee was organized by APCOB staff to plan the work and the production schedule, sign all legal documents and contracts, set wages and make reports. The entire community participated. Men were employed in the scheme while women made crafts for sale at a cooperative store in Santa Cruz.
Although the trees were managed collectively, it was necessary to obtain permission to harvest the timber from Bolivian government officials on an annual basis. The community reserved exotic and expensive tree species as "a savings account- to be used only when income was badly needed, and never all at once" [30, p. 9]. The project was heavily dependent on outside support and favourable market conditions. When outside support was reduced, sawmill production declined and the cooperative's organizational structure began to weaken.
In Peru, the Central Selva Resource Management Programme (CSRMP) was developed in I980 in reaction to plans by the Peruvian government to promote colonization of the Pachitea, Pozuzc and Palcazu valleys of the central Amazon [ I02]. The government had plans to construct roads and towns and permit over 150,000 colonists to move in. The plan was blocked by protests from a coalition of environmentalists, indigenous rights activities and, ultimately, by opposition from the U.S. government , which would have financed it through the Agency for International Development, or USAID. In its place, a limited development programme was planned through the CSRMP for the Palcazu Valley to promote the sustainable management of forest resources.
The programme was initiated in the Palcazu Valley in I983 for watershed and forest resource management. The goals of the CSRMP for this programme were to "provide a source of employment and cash income for community members ... to preserve the community's natural forests and manage them for the sustained yield of forest products and to preserve the cultural integrity of the Yanesha [Amuesha] people" [99, p. 5]. A total of US$ 22 million was allocated by USAID, on the condition that the 3,500 Amuesha Indians in the valley would become involved. USAID hoped to demonstrate that sustainable resource extraction was economically feasible. If it worked, perhaps the Peruvian government would adopt such a management model for the entire Peruvian Amazon region [I09].
The forestry component evolved as a main focus. Given the large amount of primary forest available and the potential for deforestation, project planners wanted to provide valley inhabitants with an incentive to protect forest reserves through "natural forest management for sustained yield" [ I09, p. I8]. Wood processing centres were to be established throughout the valley to allow residents to profit directly, rather than deal through middlemen. Due to concerns about implementing a programme that might subsidize private entrepreneurs, programme planners decided to start with one pilot forest management programme with the Amuesha. It was argued that due to Amuesha social and economic cooperative structures, benefits from the programme should be directed toward community improvement rather than individual capital accumulation.
The Amuesha formed a cooperative organization to harvest the timber in commonly managed forest areas and to operate a sawmill. The cooperative was composed of interested family members from each village. Each village was asked to donate a portion of its common forest land to the cooperative. They, themselves, mapped the lands prior to making allocations. Only members of the cooperative would have access to the allocated lands [ I09]. By March I989, the cooperative had approximately I00 members from five villages [ 117].
The Amuesha are traditionally organized at the village level through kinship and reciprocal forms of exchange. Decisions are made collectively, and surplus production is traditionally redistributed within the community [I02]. Several sub-committees responsible for administering different aspects of the work were organized, and cooperative leaders worked to gain consensus regarding project decisions. Informal weekly gatherings were held for the evaluation and coordination of activities . The overall cooperative structure and the organization of its tasks all conform to Amuesha social structure .
On the technical side, in order to harvest timber in a sustainable manner, the project borrowed a tropical forest management concept from Costa Rica, known as the "strip shelterbelt" system. This system requires that long, narrow strips be cut into the forest, surrounded by "intact forest, which is the source of seeds for natural regeneration of trees" [49, p. I34]. The length of a strip is determined by the quality of the forest, the topography and the ease of harvesting in a given area. Harvesting can only occur in areas that contain adequate soil type, slope and accessibility. The strips are clear-cut, and all of the trees, rather than selected species, are used for commercial purposes. The felled logs are taken away by ox cart, with minimum negative impacts to soil and vegetation . A rotation cycle of 30 to 40 years is projected between timber harvest activities .
An issue arose in the forestry component of the project over the use of toxic chemicals. After the logs are removed from the forest, they are injected with a compound of copper, chromium and arsenic to ensure preservation . Some manner of preservation is necessary in a tropical environment. The merits of using toxic chemicals were discussed by project planners and it was decided that, with proper training, the chemicals could be safely handled .
Although the forest management system adopted by the Amuesha cooperative is, in theory, a sustainable one, it nonetheless requires heavy outside financial and managerial support. Its success is dependent on marketing finished wood products.
In order to manage the sawmill correctly, training in accounting, management, cartography, land use planning and finance is necessary. And while the Amuesha are capable of learning these skills, Hartshorn notes that, if the "initial ... aid were cut-off or reduced before the project reaches full-development, then the Yanesha [Amuesha] Forestry Cooperative ... would be seriously jeopardized" [99, p. I37].
In 1988, when USAID personnel in Peru were targeted by the Shining Path guerrillas, aid was cut off and the agency withdrew its representatives and its funding. The World Wildlife Fund (WWF) then became the programme's source of outside support, and the programme continued operating.
Forests are managed by indigenous peoples for a variety of purposes, one of which is timber harvesting. But the harvesting of individual trees differs greatly from clearcutting large tracts of primary forest. There has been extensive research in the Amazon basin on indigenous resource management and how indigenous people manage and manipulate complex tree and forest systems [26, 34, 5I, 55, 57, 85,95]. There is, however, no precedent in the Peruvian Amazon for indigenous management of market-oriented production forests on a sustained basis . Overall, very little is known about how indigenous peoples can manage primary forests in the Amazon for large-scale timber extraction.
Schwartzman has described the case of 300,000 people in the Brazilian Amazon who depend on harvesting natural rubber in extractive reserves . The idea of "extractive reserves" emerged as a local solution to problems of deforestation caused by the advancement of cattle ranching. The rubber tappers need the forest to extract the product that represents their livelihood.
Extractive reserves are forested areas whose users are involved in harvesting forest products for example, rubber resin, and Brazil nuts in such a way as to assure sustainable supply of those products. The dynamics of extractive reserves provides for the preservation and even expansion of the tropical forests by having the users of the reserves defend the forests against others, such as cattle ranchers and agriculturalists who depend on removing the forest for their activities.
In the extractive reserve tenure system, reserve users have identifiable parcels of 200-500 ha in size within the boundaries of the reserve. Such a parcel is called a colocacao. Holders of these colocacoes have the right to extract forest products from the plants and animals of the area included.
Schwartzman  describes the colocacao of Seringal Cochoeira, located in the municipality of Xapuri, in Acre, Brazil. It occupies 24,898 ha and is inhabited by 67 rubber-tapper families, or about 420 people. Schwartzman indicates that the rubber tappers obtain an income that puts them above half of the economically active population of the region.
The World Bank and the Inter-American Development Bank (IDB) have shown interest in supporting this type of forest management. The World Bank is supporting the creation of five extractive reserves totalling I million ha in Rondonia, while the IDB will support the creation of four reserves totalling some 500,000 ha.
The price and market variations and the efforts to maintain rubber tappers in the economy mark the background of the rubber tappers' movement. Since the mid-I970s, the conflict for control of the resource base has created a struggle between cattle ranchers and rubber tappers, as the former try to evict the latter. The formation of unions has been the strategy of the rubber tappers to claim their right to the resource base and the option of surviving on rubber and Brazil nut extraction.
The fact that the rubber tappers need vast areas of forest to practise their activity has gained for them an alliance with international environmental organizations concerned with the preservation of the tropical forests. In I987, a leading spokesman for the rubber tappers who also practises the activity, Francisco Mendes, visited the United States and attended a meeting with the IDB to assure that the environmental concerns and rights of the rubber tappers were included when plans for road construction were implemented. More studies are needed to estimate the economic benefits of rubber tapping compared with other activities like agriculture and cattle ranching.
Another example of a community-organized and locally managed extractive activity is the collection of sap from pine trees in Honduras . There are about 6,000 resin collectors in Honduras, of which about half are organized in cooperatives affiliated with the Honduran Federation of Agroforestry Cooperatives (FECAFOR).
In considering success among the 46 cooperatives affiliated with FECAFOR, experience varies. Villa Santa is one of the most successful community forest management organizations in Honduras, with 196 members and large capital reserves. This cooperative contrasts with other groups which have little capital, declining membership and poor relations with the Honduran Government Forestry Corporation (COHDEFOR). The success of Villa Santa is attributed to the members' shared experience in protecting their forest resource rights from competing groups, the well-endowed natural resources of the region, good relations with the government and a strong leadership.
While there are a few successful cases, such as Villa Santa, resin tapper cooperatives in Honduras face such problems as insecure legal rights to forest resources, technological limitations for the tapping of resins and uncertain government action and policies concerning the role of the state forest management agency (COHDEFOR) and macro-economic resource conservation.
Programmes deriving from outside initiatives
Some of the relatively recent experiments in forest resource management, described above, are characterized by the support they receive either from the state or from local or international private voluntary organizations (PVOs) or non-governmental organizations (NGOs). These and similar initiatives have spread all over Latin America, growing out of a worldwide environmental preoccupation with the denudation of fragile lands on a continent supporting much of the earth's remaining natural forests.
Many of the initiatives are funded through, and depend on, bilateral or multilateral agreements or private international agencies. Organizations of the United Nations, the multilateral banks, national aid agencies in North America and Europe and various foundations, like the Ford Foundation, the Inter-American Foundation (IAF), the WWF and others, have participated in new and resourceful approaches, interpreting and addressing both the ecological and socio-economic needs of the Latin American people and, inevitably, introducing some of their own perspectives on resource management.
There are several features of these initiatives which condition their impact. Aside from the efforts of the IAF and some PVOs, they normally function as a spiral that starts with a general idea at the top, in most cases at the international level, descending through agreements or alliances with host governments, indigenous NGOs or PVOs to emerge, at long last, in the creation of community-based groups. Such projects are typically devoted to agroforestry or tree planting, often involving the creation of community nurseries.
Implicitly or explicitly, the ideologies and philosophies of the sponsoring international agencies, government ministries or PVOs and NGOs are transferred to projects and embodied in the way they are established - in their objectives, methodologies and expected outcomes. This influence in interaction with local conditions and experiences helps to explain why some projects have more economic and market-oriented objectives [28,7I] while others show more interest in poverty-solving approaches . For some, the final goal is mainly ecological with less emphasis on social or economic benefits. For others, tree planting is a means to organize communal groups ca became part of wider integrated rural development programmes [90, 9I].
Sometimes deforestation or reforestation issues are considered technical matters. Same authors argue that without the convergence of local people and their organizations, the technical approach is severely limited and insufficient [64, 90]. Same others strongly affirm that, as one of the first steps, reforestation requires the legal recognition of participants as private owners, which implies titling activities [ I09].
Donor agency and host country agendas do not always match. By one account, a donor agency undertook extensive negotiations with the host government aver the definition and goals of a project they wished ca fund but with which the host government did not agree [I09]. In another country, the international donor agency applied its regulations regarding social and environmental impacts aver government protests ca the contrary . In yet another case, a donor agency deliberately (and successfully) avoided the pitfalls of working with a corrupt and obstructionist government by channelling its programme through PVOs instead [28, 71 ].
Nat surprisingly, same agency ideologies and agendas are also ac odds with local social conditions and cultural traditions and experience. But there are projects in which local conditions and perspectives are quite consciously and firmly incorporated into programme and project design. The organizing methodologies of same programmes may vary between participatory and mare directed approaches . Ac times, the participants are compelled ca organize themselves under externally farmed group schemes  but same others emerge from or are consolidated into already existing groups .
Same projects encourage exotic tree plantations, promoting such fashionable, fast-growing "miracle" tree species as Leucaena leucocephala, Eucalyptus globulus, Cupressus lusitanica, Gmelina arborea and Tectona grandis. Others emphasize agroforestry initiatives, mixing a wide variety of craps with timber or fuelwood species. In only a very few cases have outside technicians started their work with an appraisal and recognition of indigenous knowledge in order ca enrich their own scientific knowledge. Very few have started their work by improving forest fallows [20, 52].
The majority of the projects encourage tree planting activities but do not embody long-range strategies ca recover or strengthen indigenous practices, nor do they strive ca recreate forest environments which existed prior ca their destruction or damage.
A common example of a community forest resource programme in the region is the community tree nursery. Sometimes the participants in such programmes establish nurseries or plant the trees an common property land and enjoy the benefits in common [19, 114]. Same others decide ca establish nurseries in common an private or public land, then, in a second stage, divide the seedlings equitably far transplantation ca private plots under individual responsibility [2I, 35]. Same nurseries are established in the care of larger town, others are developed as small nurseries (vireos valances), located in site communities, depending an the dispersion of the houses .
Local and national governments and the sponsoring private or international agencies have used diverse incentives ca encourage local people ca participate in these projects, including: food-for-work (or "food-for-peace" ), the donation of infrastructure (storerooms, instruments) and seedlings or the persuasion of people ca grow timber together with cash crops like cacao [67, 117]. Pilot programmes are sometimes started on the land of community leaders or other influential people to capitalize on the demonstration effect .
Local initiatives with outside support
A novel and abundant bibliography is now emerging in two fields that have direct application to developing successful common forest resource (CFR) management systems and projects. One set of literature deals with indigenous and traditional ways of farming and managing forests by indigenous, Mestizo and other peasant peoples. The other deals with project implementation strategies for incorporating and stimulating local participation and institutions. In contrast with top-down approaches which have ignored local knowledge and paid only lip service to participation, these new interests are more horizontally based [35, 67], in the sense that they advocate starting with thorough analyses of local farming systems and forest management practices, then proceed to ensure serious and beneficial participation.
Some examples of the application of these two approaches are: