Rural Community Agency in Cameroon: Interactions with Forest Policies and the REDD+ Climate Change Regime

Abstract

Community forestry around the world has demonstrated its potential for implementing the Reducing Emissions from Deforestation and Degradation (REDD+) climate change program. Secure tenure rights and access to rule-making are known as contributing to successful community forestry outcomes. Still, the effects of different aspects of rural ‘community agency’ are not well established. We investigate forest governance and conflicts and the relationships between aspects of rural community agency under the REDD+ climate change program in two forest communities—the villages of Fabe and Mosongiseli—near the southern portion of the Korup National Park in Cameroon. Using data from a survey instrument and interviews, we analyze, using “agency theory”, the concept of rural community agency according to dimensions of attitudes, understandings, and empowerment in the two communities in relation to forest governance and conflicts under REDD+. Our findings indicate a variety of power relations (e.g., on the communities’ use and management rights of their lands) and existential threats of conflicts within the communities (e.g., violation of the communities’ free, prior, and informed consent). The results also show that both communities share many of the patterns of diversity and integration to a similar extent. Although there is no definitive distinction between the two communities, the findings suggest that some differences exist in their degree of integration. Understanding and describing the nature of the power relations and threats of conflicts comprises an important component to begin an appreciation for the communities’ user group characteristics as these relate to the REDD+ program when implemented. The implication of this study is that threats of conflicts may increase when the villagers’ perception of the potential costs of losing their lands to REDD+ is formed by their experiences with current restrictions on the use and management rights of their lands.

1. Introduction

In December 2015, the Conference of the Parties (COP21) to the United Nations Framework Convention on Climate Change (UNFCCC) recognized and acknowledged in the Paris Agreement the vital role that forests play in the fight against climate change [1]. This recognition comes after decades of international cooperation and negotiations on the climate change program—the United Nations Reducing Emissions from Deforestation and forest Degradation and the role of conservation, sustainable management of forests, and enhancement of forest carbon stocks in developing countries (REDD+) [2,3].

Despite being hailed as a major step toward reducing global warming and climate change, REDD+ poses both challenges and opportunities to forest communities due to the dependence of the local people on the forest and its resources [4,5]. One challenge comes from similarities that REDD+ shares with integrated conservation and development projects (ICDPs), as well as top-down measures of forest management that have in the past alienated the local people from their customary lands and sources of livelihood in favor of other ecological, political, and economic interests [6,7,8].

Cameroon is currently developing its national REDD+ implementation policies [9,10]. However, implementing REDD+ in Cameroon poses risks to the livelihoods of forest communities [11,12,13]. These risks originate from two different, though complementary, frames. First, some observers, for example [14,15], argue that REDD+ will be implemented nationwide, but similar to ICDPs, many activities will occur at the sub-national project level, which may exacerbate ongoing land tenure disputes [11]. For example, the Korup project, an ICDP in the Korup National Park in the South West Region of Cameroon that began in 1988 [16], has been characterized by some as a “catastrophic failure” [17]. The Korup project was intended, in part, to resettle the inhabitants of villages living within the park, in addition to restricting hunting, gathering, and farming practices that are essential to the livelihoods of the local communities [6,7]. However, since the inception of the project, only one village has been relocated—the village of Ikondo-Kondo [18]. Resettlement was accomplished at a financial cost far higher than anticipated, while livelihood restrictions resulted in conflicts and considerable distrust between park authorities and those living in and around the park [7]. Second, REDD+ threatens to recentralize control of forest management because of the desire for states to obtain financial benefits from the program [19]. In studies of REDD+ implementation in the Philippines, Cambodia, and Papua New Guinea, Ref. [20] observed conflicts between national, regional, and local actors as governments tried to accrue benefits flowing from the control of the production of forest carbon.

Nevertheless, REDD+ also presents opportunities to forest communities in Cameroon [9,10], as community forestry around the world has demonstrated its potential as an effective, efficient, and equitable tenure program that may be appropriate for implementing REDD+ projects [19,21,22,23]. Ref. [24] asserts that REDD+ outcomes can be enhanced by involving communities in forest management but cautions that parameters for the success of community forestry need to be addressed in order to optimize the results of these initiatives. Widespread agreement exists that secure tenure rights in rural communities contribute to successful community forestry outcomes, but the effects of different aspects of rural community agency are not well established. Rural community agency is defined here as the intentionality and rationality of attitudes, understandings, and empowerment of individuals within a community and of the community as a whole in deciding what is best for them. It is important to understand how aspects of forest rural community agency and interactions may affect the REDD+ climate change program. Earlier publications have focused more generally on the issue of land tenure and carbon rights [25,26]. More recent publications look at benefit sharing and livelihoods [27,28]. However, there is little study on specific aspects of forest rural community agency and the potential impacts on the REDD+ climate change program.

This study investigates, with regards to forest governance and conflicts, relationships between aspects of rural ‘community agency’ under REDD+ among two forest communities—the villages of Fabe and Mosongiseli—near the southern portion of the Korup National Park (KNP) in Cameroon. The general objective is to understand and describe the range and qualitative nature of user group characteristics of the two communities—as it relates to institutional arrangements and the nature of forest rural community agency among various actors—and the potential impact on the REDD+ program when implemented. The specific objectives are threefold: (1) to understand the intricacies of integration, collective agency, and decision-making in the communities, (2) to understand the communities’ interactions with the KNP project and how the interactions may affect REDD+ when implemented, and (3) to understand power relations and implication for REDD+ and existential threats of conflicts. This will allow policymakers to make informed decisions about forest communities and socio-economic, cultural, and environmental impacts early on in the REDD+ program development process, thus, contributing to the improvement of REDD+ interactions with local communities. We examine prevailing norms, formal institutions, and interactions with external actors and the proposed REDD+ program in the study communities. The two communities were chosen for this study because of their experience with an internationally funded conservation program (i.e., the Korup project), which shares some similarities with the proposed REDD+ program in areas.

With the understanding that “agency” is the ability of social actors to make choices in a given institutional environment [29], we employ “agency theory”, which “encompasses ideas about the desirability of citizens actively engaging in the institutions, policies, and discourses that shape their access to resources” [30] (p. 223) to better understand community interactions in Fabe and Mosongiseli in relation to forest governance and conflicts. Rural community agency here is measured according to a diversity-based set of three dimensions: attitudes, understandings, and empowerment. This loosely follows from the identification by [31] of three bases of collective action: rationality, intentionality, and power. In this case, attitudes towards the REDD+ policies are seen as irrational on the part of individuals within demographic groups (education, age, gender, occupation, religion, and place of origin), and on the part of the community as a whole, in deciding what is best for them. Attitudes are understood as whether community members share favorable or unfavorable opinions regarding specific environmental policies and the intentions that community members have towards these policies, particularly towards interacting with the REDD+ program. The dimension of understandings informs the question of why community members have particular beliefs regarding the REDD+ program and current forest policies, based on their knowledge and experiences, thus informing the intentionality of individuals, demographic groups, and the community. Two key factors of understandings examined include knowledge of forest policies and local experiences with external actors. The dimension of empowerment is likely influenced by understandings, and informed by attitudes, but is largely based on two key factors, vis-à-vis capacity for decision-making and forest access.

2. Analytical Framework

This study uses two frameworks of agency theory for understanding the concept of “community” that may at first appear contradictory. Ref. [32] propose an institutional analysis model of communities and contrast this perspective with earlier notions advanced in the study of community-based conservation initiatives. The authors advocate a model in which diverse actors within the community interact in ways that establish institutions (seen as rules and norms that constrain individual actions) through “processes of decision-making and enforcement”. While institutions can direct actions, they are also in constant change as individuals and groups contest and negotiate their terms. They claim that this contrasts with earlier discourses in which communities were thought of as either spatially constrained territories, homogeneous social structures, or shared norms. While the authors acknowledge that these three attributes can be found to some extent in many communities, taking these characteristics as unexamined truths fosters a simplistic interpretation, which can lead to erroneous assumptions about how communities interact with their environment and conservation policies. A better understanding of conservation outcomes is obtained through a focus on institutions and the processes through which these rules and norms are shaped. Moreover, they argue that community institutions often cannot be clearly isolated from external institutions, so analyses should take into consideration interactions with external actors.

According to Ref. [33] (p. 632), Wilkinson’s field–interactional theory views the community as a “field of interaction” between individuals having diverse interests. In this conceptualization, the community is continually reconstructed through social interactions as individuals negotiate and define rules and norms. The perspective is primarily concerned with processes through which collective agency is produced. In the field–interactional model, collective agency is thought to come about through the empathic generalization of individual understandings to those of other individuals or groups within the community field. As a result of social interactions, “diverse lines of action coalesce around common interests” [33] (p. 629), leading to collective action. Despite similarities to “community as common interests and shared norms”, critiqued by [32] (p. 635), these authors do not deny that certain “community-level norms” do often exist and that, in some cases, they are amenable to conservation goals. They also note that shared understandings can encourage community members to work together in negotiating institutional arrangements.

In the field–interactional model, communities themselves are not necessarily composed of individuals having homogeneous understandings. Rather, the extent to which understandings are shared (due to social interactions between individuals from different groups) is theorized to result in a collective agency when individuals generalize their own interests as being synonymous with those of others in the community. This generalization then produces collective action, which is representative of a diversity of social groups rather than the interests of only a small subgroup. Thus, social interactions between members of the community having diverse interests would modulate the processes by which communities make rules, implement them, and resolve disputes in ways that give authority for these actions to broader sections of the community. The main concern is that researchers neglect the cases in which collective action, especially in heterogeneous communities—in interactions with community norms, formal institutions, external actors, and external policies—proves detrimental to ecological outcomes, individuals, subgroups, and even the broader community; thus, resulting to different degrees of internal and external conflicts.

This paper synthesizes these two community frameworks (Figure 1) by examining the demographics of the study communities; prevailing norms, formal institutions, and interactions with external actors; states and processes of integration between demographic groups; and an assessment of collective action in interactions with external policies as it relates to forest governance and conflicts under the REDD+ program.

3. Study Areas

Cameroon lies on the west coast of Africa. Fabe is a village situated to the northeast of Mundemba town in the Ndian Division in the Southwest Region of Cameroon. On the other hand, Mosongiseli is a village that lies to the southwest of Mundemba town, just outside the most southern end of the Korup National Park (KNP) (Figure 2).

The area currently designated as Korup National Park has a complex history. KNP began as a forest reserve in 1937, during which time enclaves were allotted to a number of villages, including, Fabe and Mosongiseli, within the reserve. The reserve became a national park in 1986 following a presidential decree. As its protected status evolved through the years, changes in tenure rights have accompanied changes in the interactions between local communities and park authorities. Understanding the characteristics of this place and the people who live there is important to understand what makes it an appropriate location to study the interrelationships between collective identities, collective agency, and collective action with respect to environmental policies, such as REDD+. The area has had a rich but troubled history stemming from both current and colonial-era policies of plantation agriculture. It has also been subject to attempts that seek to impose “Western” ideas of conservation on populations that are in many regards unwilling and unable to make the drastic livelihood changes required to satisfy these demands.

A roadside settlement with about 60 households, Fabe has a population of about 300 inhabitants, mainly people from the Bima clan, whose major occupation is farming (

Table 1. Summary of study villages.

Village	Population Size	Clan	Major Occupation	Settlement Classification
Fabe	300	Bima	Farming	Roadside
Mosongiseli	450	Balondo-Badiko	Farming/fishing	Creek/roadside

). A majority of the 450 inhabitants of Mosongiseli belong to the Balondo-Badiko clan. While Mosongiseli is considered a creek settlement, the categorization of the roadside settlement also applies to the village. Like Fabe, the major occupation of the people of Mosongiseli is farming. However, some of its inhabitants are involved in artisanal fishing. Fabe and Mosongiseli are two of five villages with a total population of about 1500 inhabitants located within the borders of KNP. The ecology of the KNP area consists of the dense, moist tropical forests of the Atlantic Biafran coastal region. The KNP itself, sharing a border with Nigeria to the west, occupies an area of 1250 km² (482 square miles) and is located between the coordinates of 4°54′ N to 5°28′ N and 8°42′ E to 9°16′ E.

4. Methodology

The primary research methods used for this study include a survey instrument consisting of closed- and open-ended questions (See supplementary material), interviews (See supplementary material), and other ethnographic methods such as “mapping” and participant observation. The survey was a combination of random and nonrandom sampling. Two non-random sampling techniques were used: purposive and convenience sampling (an approach described by [36], and snowball sampling. Snowball sampling was used to identify new respondents from the suggestions of previously sampled individuals. For the random sampling, a node-based technique was used—the administering of questionnaires to multiple respondents gathered at one location (the node). This technique was employed with two main variants: (1) at times, each member of the research team was assigned to record the responses of one respondent in a small group, for instance, inside a person’s home. One team member read each question, which was then translated into Pidgin-English (English-base creole language spoken as the language in the Anglophone regions of Cameroon) by another team member. If the respondent spoke only a local language, the translation was done by another team member fluent in that language, (2) on other occasions; it was assessed that the respondents were literate enough to write their own responses to the survey document. Each question was read aloud by a member of the research team, clarified, and translated into concepts more appropriate to the local context as needed. We used random sampling to ensure that every individual in the two communities had a nonzero probability of being selected. The non-random sampling technique was used to ensure that groups like women (gender distribution) and strangers that may not be properly represented in the sample are represented.

The survey instrument contained 29 questions, including both open- and closed-ended questions. A group of questions on demographics was used to gain information about the basic demographic makeup of the village (for instance, gender balance and age distribution), characteristics related to the respondents’ social identity (religion, place of origin, clan group), and characteristics more directly related to agency (such as occupation, level of education, land tenure status, and languages spoken). This was followed by questions about the respondent’s agency in terms of understandings of the forest (how forestland and resources are used culturally and for respondents’ livelihoods), and perceived empowerment (in terms of access to decision-making). Some questions sought information about the respondent’s current knowledge of REDD+ and their attitudes towards the program. These questions specifically sought to understand respondents’ opinions about how REDD+ might affect the community and the forest and their intentions towards proposed mechanisms of REDD+ benefit distribution. In all, 45 survey responses were received in Fabe, along with 4 in-depth interviews conducted (

Table 2. Summary of primary data collected.

Site	Surveys	In-Depth Standardized Interviews	Participant Observation	Field Notes and Unstandardized Interviews
Fabe	45	4 villagers	Korup project meeting	✓
Mosongiseli	36	3 villagers	Apostolic church service	✓
Yaoundé	N/A	4 NGO workers	N/A	N/A

). In Mosongiseli, we received 36 survey responses and conducted 3 in-depth interviews. In addition, 4 interviews were conducted in Yaoundé (

Table 3. Summary of NGOs interviewed.

Organization 1	Organization 2
The organization works as a facilitator for partnerships between smallholders and the agro-industry. Activities are mainly focused in the South and East regions but have had some recent involvement in the Southwest.	Technically not an NGO but a state development bank and bilateral donor. The organization is responsible for funding its technical branch and NGOs involved in community forestry, as well as conservation and development projects.
Organization 3	Organization 4
The organization works as a facilitator for partnerships between smallholders and the agro-industry. Activities are mainly focused in the South and East regions but have had some recent involvement in the Southwest.	A Cameroonian NGO working on a range of issues concerning forests of the Congo Basin.

) with employees of NGOs involved in environmental governance. They were knowledgeable about one or both issues of (1) forest policies in the Southwest Region and (2) the REDD+ process in Cameroon. The in-depth interviews were digitally recorded and transcribed using heterogeneity and collective action recognition software called “Dragon Naturally Speaking” (as described by [37]).

Other than routine informal qualitative observations noted during the research, in Fabe, there was an opportunity to observe a formal 2-h meeting with the villagers and the Korup Project staff. The meeting included a participatory rating of the park by the villagers and of the village by the park staff on various aspects of a Conservation and Development Agreement between the park and the community. In Mosongiseli, we observed a formal Sunday worship service of an Apostolic church held in a schoolhouse.

In analyzing the data, we employed the dual “technique of inductive and deductive thematic analysis”, as applied by [38], carefully reading through the data repeatedly in view of identifying peculiar themes for analysis. Deductive coding—reasoning broadly from the more general to the more specific—facilitated the organization and interpretation of the data, while inductive coding—from specific observations to the broader generalizations—allowed for the recognition of vital information before processing and interpreting the information.

We used secondary data from scholarly articles and numerous documents from different sources, including legislative texts, policy briefs, policy guidelines, policy papers, and expert reports to characterize forest governance and conflicts under REDD+ and to enrich the analysis in the two study communities.

5. Results

We asked survey respondents questions intended to determine how societal integration—socio-demographic characteristics of respondents—is related to collective agency towards current forest policies. We compared responses between respondents from the two study communities—the village of Fabe and the village of Mosongiseli. Different demographic groups within a community are expected to vary in their interests, and increased social interactions between different demographic groups are expected to contribute to the groups’ interests becoming generalized to other members of the community. Hence, we hypothesized that a greater degree of societal integration in Fabe and Mosongiseli would lead to collective action that embodies more diverse interests and involves the perspectives of multiple demographic groups in the two communities.

5.1. Patterns of Diversity and Integration

The results indicate that both communities share many of the patterns of diversity and integration to a similar extent (

Table 4. Proportions for dichotomous consolidation of demographic variables ¹.

Characteristic	Category	Fabe		Mosongiseli		Sample Total
		n	%	n	%	n	%
Village	Fabe	45	100	0	0	45	55.6
	Mosongiseli	0	0	36	100	36	44.4
Gender	Women	21	47.7	9	25.0	30	37.5
	Men	23	52.3	27	75.0	50	62.5
Age group	18–30	14	31.1	13	36.1	27	33.3
	>30	31	68.9	23	63.9	54	66.7
Origin	Indigene	24	53.3	23	63.9	47	58.0
	Strangers	21	46.7	13	36.1	34	42.0
Occupation	Farmer	34	79.1	18	50.0	52	65.8
	Non-farmer	9	20.9	18	50.0	27	34.2
Education	Below FSLC	13	30.2	14	38.9	27	37.2
	FSLC or above	30	69.8	22	61.1	52	65.8
Land tenure	No land	6	13.3	6	16.7	12	14.8
	Formal or customary	39	86.7	30	83.3	69	85.2
Location	Inner village	39	88.6	29	85.3	68	87.2
	Other quarter	5	11.4	5	14.7	10	12.8
Religion	Christian	38	86.4	35	97.2	73	91.3
	Non-Christian	6	13.6	1	2.8	7	8.8
Clan	Majority clan	31	68.9	24	66.7	55	67.9
	Minority clan	14	31.1	12	33.3	26	32.1

¹ The columns for n give the number of responses for each variable category, separately for each village, and combined in the Sample Total column. Percent is the proportion of responses in each category divided by the total number of responses for that question.

). For instance, each sample was similarly diverse in clan groups. Given the strong association between clan groups and place of origin, this likely indicates a similar diversity in a place of origin as well. Community celebrations create a space for processes of integration in both communities. Both communities show disparities between demographic groups in terms of education, with women having lower levels of educational attainment compared to men and youths having higher levels compared to non-youths. There do appear to be some differences in the degree of diversity and integration between the two communities, and in a majority of these cases, Fabe was identified as the more integrated community. Fabe is probably more integrated geographically according to the location of residence within the village, as strangers—those who trace their identity to another location—in the sample were found to live more in peripheral quarters in Mosongiseli. The study revealed processes of social integration in both communities. Fabe showed more potential for social integration according to occupation because a similar proportion of youths and the older population were farmers. Furthermore, a farmer’s field school operated by the Korup project likely provides a means of integration between indigenes—locally-born natives—and strangers. The Diyala—a society for women—in Fabe could also provide a setting for integration between women of different origins. On the other hand, Mosongiseli is probably more socially integrated into their monthly cleanup campaigns compared to Fabe, the latter of which apparently relies more on women to accomplish this task. Fabe appears to be more integrated hierarchically. In Fabe, women and men have separate secret elite societies, Ekpa and Motamo, respectively. Although genders are not integrated within each individual social group, the presence of a secret society for women, as well as for men, suggests that the community affords a certain hierarchical status to women. In addition, Fabe has a higher degree of religious diversity. Given that all religious minorities in the sample were indigenes, this probably enhances the hierarchical integration in terms of religion.

5.2. Agency-Based Factors

Regarding agency-based factors, the survey indicates that Fabe is likely more integrated in terms of access to decision-making because a higher percentage of indigenes and women in Fabe reported involvement in decision-making. Mosongiseli is probably more integrated with the agency-based factor of knowledge, as strangers in the sample had equivalent or even higher levels of education compared to indigenes, whereas those in Fabe had lower levels of education. Fabe may be less integrated in terms of attitudes, specifically towards industrial agriculture (plantations). Two oil palm companies operate in the communities—SG Sustainable Oils Cameroon (SGSOC), a division of the American-owned Herakles Farms (from now on referred to as Herakles) that operates in Fabe, and Pamol Plantations Plc (from now on referred to as Pamol), a parastatal company with majority government ownership that operates in Mosongiseli. Interview respondents in Fabe indicated that there had been disagreement between some village elders/elite and some members of the community on the issue of Herakles operating in the village. The elders/elites were in favor of Herakles, while a majority of the villagers did not want the company to operate in the village. In addition, different demographic groups did not have cohesive opinions on the Herakles matter; instead, opinions were held mainly as an individual choice. However, the disagreement could easily be an effect of the socially corrosive nature of Herakles’s actions more than any latent difference in opinions on the matter of allowing industrial operators access to communal lands.

5.3. Modes of Collective Action

While we cannot make a definitive distinction between the two communities, these findings suggest that some differences exist in their degree of integration. These differences are reflected in their modes of collective action toward current forest policies (

Table 5. Modes of interaction with forest policies.

Policy	Fabe	Mosongiseli
Forest conservation	Passive acceptance Community makes effort to act in accordance with the Community and Development Agreement (CDA); appreciation of modest benefits received	Active rejection Community wrote letter in opposition to peripheral zone
Industrial agriculture	Mixed: Passive acceptance followed by active rejection Elders/Elites accepted Herakles (previously signed agreement with Herakles); other members of community worked with NGOs in opposition	Mixed: Active rejection followed by passive acceptance Youth and women organized against Pamol; community then accepted the company
Community forestry	Passive acceptance Some community members worked with a local NGO to establish, but not widely known	Active acceptance 14-year attempt by management committee to establish, with eventual aid from international development agency, despite many difficulties and setbacks

). The modes of collective action for each community are categorized into qualities of acceptance or rejection and active or passive roles in the interactions.

6. Discussion

A close articulation between community agency and interactions with forest policies and REDD+ can help enhance our understanding of existential threats of conflicts as it relates to ownership and entitlement issues in forest-dependent communities. Here, we discuss the practical implications of integration and disparities among demographic groups for the implementation of REDD+ in Fabe and Mosongiseli. The focus is on two aspects of REDD+: one of conservation and one of the commodification of the forest, with analysis informed by perspectives on ownership and management of forest space.

6.1. Perspectives on Integration, Collective Agency, and Decision-Making

Addressing the hypothesis—that a greater degree of societal integration in the study communities would lead to collective action that embodies interests that are more diverse and involves the perspectives of multiple demographic groups—is limited by not knowing precisely what the interests of different segments of the communities are. A more detailed understanding of socio-economic circumstances would be needed [39]. For example, there are differences between the two villages in response to industrial agriculture, and there are also differences regarding where local farming activities take place relative to where plantations are established. If many community members in Fabe had farms where land was being set aside for Herakles, this would significantly disadvantage these families. Still, if the Pamol plantation in Mosongiseli was not where most farms were located, then this could explain the different responses more than societal integration and the recognition of different interests during processes of decision-making.

With regard to how community integration affects the communities’ response to forest policy, Mosongiseli has taken more active roles in policy interactions, especially in response to the Korup project and in pursuing community forestry. For example, segments of the community actively organized against Pamol when the company first came into the village. It is not clear, however, whether these responses are followed by social interactions between different demographic groups, creating common goals and motivating action. While integration was observed in more dimensions for Fabe, Mosongiseli pursued more active roles, which is in contrast to the hypothesis. On the other hand, there is uncertainty about whether the states and processes of integration found in the study actually constituted integration because, as [40] (p. 114) points out, “contemporary study of social integration treats it either as a multidimensional phenomenon or as a latent construct with multiple indicators”. Moreover, information on the processes by which collective actions were taken is not sufficient to make a definitive statement that integration or segregation of demographic groups led to the policy responses.

Our analysis of the survey and interviews show that in both communities, strangers were not involved at all levels of decision-making and were also disadvantaged in access to forest land and resources. Women and strangers report lower rates of participation in decision-making compared to youths. Despite their vital contributions to the cultural and economic life of their communities, women appear to have social barriers to decision-making. Women and strangers are stakeholders in the community and, therefore should be included in discussions about village matters [41,42] because they are particularly at risk of loss of livelihood and exclusion from benefits. It is worth noting that the KNP communities were not substantially involved in the processes that established the rules of the park and its surroundings, which they are now expected to obey [6]. A large number of forest management and governance studies have asserted the critical importance of involving all stakeholders in decision-making processes, for example [43,44,45]. The meeting observed in Fabe between the villagers and the Korup project staff is a step in the right direction by involving large parts of the community in the social processes of implementing park rules. However, the project should also draft new rules in collaboration with the communities in which it seeks to work. This would ensure that the rules are accepted by the current generation of villagers and community leaders and increase their legitimacy for standards of free, prior, and informed consent (FPIC).

6.2. Characterizing the KNP Project and the REDD+ Program

The characterization of the landscape concept and the issue of land rights (tenure entitlements) are essential to forest governance and the effective implementation of REDD+ projects. A KNP staff member noted that plans are underway to expand the Korup Project into a landscape-level project by expanding its range in the north to Takamanda and across the border into Nigeria. However, the question is, does expanding the range of the park make it a “landscape-level project as currently employed in forest governance under REDD+”? Discourses of landscape applications on the ground, for example [46,47], do not support this size-based, boundary object nature of categorizing the landscape concept. Rather, landscape discourses focus on the nature of objects and their interaction processes [48,49]. Therefore, to attain a true “landscape” status, the KNP project would have to aim for nested scales that include and integrates all actors and their livelihoods and spaces. However, the use of a nested scales landscape approach in forest governance and REDD+ that does not address ownership claims and tenure entitlements (see Section 6.3) can lead to conflicts around REDD+ implementation [50]. In this light, adopting a “landscape” approach that is focused not only on trees and on forests but encompasses social, economic, environmental, and policy systems would align the KNP project with the holistic, integrated approach of the REDD+ program [51].

Recent trends of decentralizing forest management and investing local communities with management responsibilities have been motivated in large part by the expense of centralized approaches to forest conservation [19]. To counter this approach, there has been a move toward community-based conservation measures. ICDPs were promoted as a means of encouraging both local participation in conservation measures and compensating these communities for costs they were being required to bear in order to comply with the rules established by the projects [14]. The Korup project has had mixed success in both compensating communities for livelihood restrictions and its efforts to relocate communities living within park boundaries, which have been largely unsuccessful in meeting conservation goals while also far outspending its estimated budget [6,7]. With REDD+, both the prospects for beneficial outcomes and risks of conflicts and other detrimental effects are increased as potentially much greater funding is introduced compared to current conservation spending [19].

Fabe and Mosongiseli exist within the Korup project support zone, so they currently interact with the park as the principal agent of forest conservation policy. Although the Korup project does not appear to actively restrict farming activities in Fabe, park authorities are threatening to enforce a 3 km peripheral zone in Mosongiseli. Faced with the peripheral zone restrictions, the Mosongiseli community acted collectively by expressing their concerns through the Village Forest Management Committee (VFMC) in a letter addressed to the conservator of KNP. Taking this active approach shows that the village has the channel to transmit their concerns and the organization to take coordinated action within the community. This suggests that this means of communication between the community and the park management through the VFMC could be a vital requirement for any similar REDD+ project as a way for the communities to express their disagreement. However, the community never received a response to the letter, and there has been no resolution of the matter, leaving the community in a position of uncertainty. In the meantime, the community has continued to farm in the peripheral zone; thus, the intended conservation goals are not being met.

Organization 2 (see Table 3) has been strongly involved with the Korup project. The respondent from this organization expressed the view that, because the communities have signed a conservation and development agreement (CDA) with the Korup project in exchange for receiving development assistance from the project, they must respect the rules and the boundaries of the park. While conservation along with socio-economic development may be the project’s intention, benefits to the communities do not appear to be very substantial. This is particularly the case in Mosongiseli, where the community is not allowed to carry out farming and other activities in their preferred locations. A cost-benefit analysis would indicate that complying with this demand requires considerably greater compensation than the community currently receives from the Korup project in development benefits.

6.3. Implications for REDD+ and Existential Threats of Conflicts

Even though REDD+ is primarily a carbon mitigation initiative, it has evolved to embrace broader approaches and methodologies for handling problems related to forest resource management [52]. Shortly after its inception in 2007, REDD+ began shifting its focus beyond just sequestering carbon in trees to including the social, economic, political, and environmental complexities of forest governance in developing countries [53]. The adoption of broader approaches and methodologies has redesigned the REDD+ program to operate more like an ICDP [14]. However, it should be recalled that the KNP project is an ICDP, and ICDPs have caused significant difficulties for forest communities because of top-down efforts to enforce conservation goals through the resettlement of local populations and livelihood restrictions [54]. Therefore, the question is, if implemented in Fabe and Mosongiseli, would the REDD+ program that operates like an ICDP exacerbate the threats of conflicts that the communities already face from the KNP project and from industrial agriculture ventures? Given that many authors, including [55,56,57], have documented accounts of conflicts that have arisen from forest people’s fears that they will be disenfranchised by REDD+ projects in their communities and that their land will be appropriated, there is a real possibility that the pending REDD+ projects in Fabe and Mosongiseli could exacerbate the threats of conflicts in the two communities.

6.3.1. The Monitoring, Reporting, and Verification Frame of Obstacles to REDD+

Some of the essential requirements of the REDD+ program are monitoring, reporting, and verification (MRV) requirements [58]. One frame of obstacles to REDD+ comes from MRV requirements. To have an effective MRV system, the land needs to be well defined and identified. In order to define and identify the land, it must be mapped for purposes of forest resource inventory, spatial zoning, and for a better understanding of people-environment interactions [59]. To the people of Fabe and Mosongiseli communities, mapping of their lands for the MRV requirements in REDD+ has the potential to raise conflicts. First, it would be a reminiscence of conflicts they experienced (and are still experiencing) with industrial agriculture from Herakles and Pamol. The establishment of oil palm plantations by Herakles and Pamol in Fabe and Mosongiseli, respectively, were alleged without a consensus of the communities. Respondents noted that the companies used oppressive measures to coerce the communities into accepting appropriation of their land with little to no compensation. They claimed that Herakles had made many false promises to the community. Many respondents considered Herakles to have “stolen our land” by claiming that the project was only to be temporary, when in fact, the company had intended to establish in the village on long-term bases. The respondents alleged that the community was harassed, intimidated, and pressed by Herakles to accept the company. In Mosongiseli, a respondent noted that Pamol has refused to pay royalties to the community on the grounds that the company is state-owned and therefore is not required to compensate communities. These allegations are tantamount to a violation of the free, prior, and informed consent (FPIC), which [60] described as “the central social safeguard” for protecting the rights and interests of forest communities. It is also a violation of the UN Declaration on the Rights of Indigenous Peoples, which is against REDD+ Decision 1/CP.16 [61], and contrary to the goals Cameroon has committed to in its REDD+ Readiness Preparation Proposal (R-PP) [62]. Therefore, on this backdrop of conflictual experiences and fears expressed by some respondents that the implementation of REDD+ projects in their community would mean further appropriation of their land are justified. The villagers could interpret any landmarking, zoning, and forest resource inventory activities on their land for REDD+ project purposes as initial steps to limiting their activities on the land or taking the land away from them altogether. Threats of conflicts may increase when the villagers’ perception of the potential costs of losing their lands to REDD+ is formed by their experiences from current restrictions on the use and management rights of their lands by Pamol and Herakles.

Second, mapping of their lands for the MRV requirements in REDD+ could exacerbate the generational territorial conflicts between local communities/indigenous peoples and their national governments over control of tangible geographic areas [63]. Territorial conflicts here are concerned with the issue of land tenure. Land tenure arrangements determine and accord user rights; the so-called bundles of rights in property—access, withdrawal, management, exclusion, and alienation [64,65,66]—privately to individuals and collectively to communities. Under Cameroon’s constitution (as is the case in many developing countries), all mineral rights belong to the national government. Given that carbon is designated as a mineral, ownership rights to biomass carbon under REDD+ could technically belong to the national government. These rights can be transferred to private project developers. According to [8], the establishment convention signed between Herakles and the government of Cameroon gives that company all carbon rights on the land allotted to it. Like in many customary and indigenous communities in Cameroon [26], tenure rights in the Fabe and Mosongiseli communities are not well established. Many respondents from both communities indicated that even though they have access to one or more parcels of land in the community, they do not have confidence in complete ownership of the parcel(s) of land. Lack of confidence in land ownership is also expressed at the community level, as illustrated by the draconian takeover of the communal land—invasion, restriction, seizure, displacement—for the KNP project and for industrial agriculture. This means that there is no security of tenure—no bundles of rights—because the communities do not feel safe with their individual and collective rights in the landed property. The spatial management modalities under REDD+ may therefore foster forestland grabbing [65], which further weakens the communities’ confidence in land ownership, and thus increase the threat of conflicts.

6.3.2. The Commodification (Finance) Frame of Obstacles to REDD+

Another frame of obstacles to REDD+ comes from the large amounts of money at stake, as the funding mechanism for REDD+ is tied in with the international market for carbon [66,67]. Given the contentious nature of disputes over commodified land in the two communities, it is important to look at concerns about REDD+ as it functions as a market for carbon credits. Large amounts of money are potentially at stake from the sale of credits, and control over this revenue, as well as the production of credits through ownership of land and carbon rights, could easily prove as problematic as control of the land for large-scale forest concessions and industrial agricultural. This thus raises the question: if REDD+ is implemented in these communities, what are some possibilities for handling funds, and what mechanisms are available for benefit distribution? Some REDD+ commentators, for example [68], have argued for the use of existing financial policies and financial distribution mechanisms rather than creating new mechanisms. In the case of Cameroon, one option proposed in the country’s R-PP is to use the model of the AFF tax (annual forestry fee) established by the 1994 Forest Law [69]. The law allocates 50% of annual timber concession fees to the national government, 40% to the village councils and FEICOM (the Special Equipment and Inter-municipality Intervention Fund), which is responsible for distributing the proceeds to village councils nationally [70], and a final 10% to the communities on whose communal land the concession is located. The problem is, as applied in the forestry sector, this system has been inadequate in delivering the 10% of revenues to the local communities that bear the greatest burden in hosting the timber companies [71]. Moreover, the respondent from Organization 4 (see Table 2) stated that a new finance law that supersedes the 1994 Forest Law has changed the distribution so that the 10% no longer goes to local communities but instead to tax officials.

Another option is the land rent model established by Decree 76–166 of 27 April 1976, used in the case of industrial agriculture concessions on national lands [72]. In similar proportions to the AFF, this system distributes 40% of land rents to the national government, 40% to village councils, and 20% to local communities. The analysis of [72] shows that typically, these rents are not paid by the industrial operators, or the funds are lost to corruption within the Ministry of State Property and Land Tenure. The law also has never established the minimum or maximum fees that companies must pay per hectare and lacks the record-keeping requirements of the AFF [72]. It seems clear that both of these mechanisms (the AFF and agricultural land fees) as they currently function will not meet the goals of Cameroon’s R-PP in respecting local rights, livelihoods, and the UN Declaration on the Rights of Indigenous Peoples [73]. Of course, these mechanisms only distribute the fees per area of land, not the revenue derived from the business operation itself. Even assuming that REDD+ funds received in either of these scenarios and not merely land fees are distributed in their respective proportions, they would not compensate for the opportunity costs calculated by [74] for communities in the KNP area, including Fabe and Mosongiseli, which found that 80% of the value of carbon credits would need to reach local communities.

In either option of REDD+ funds and benefits distribution mechanisms suggested above, it would be risky for REDD+ to operate through companies, whether government-owned, parastatal, or private enterprises, where the companies, and not the local communities, own the carbon rights themselves. This could be tragic for the Fabe and Mosongiseli communities because it would completely disenfranchise the communities from REDD+ benefits while still appropriating village land and imposing livelihood restrictions.

7. Conclusions

This study, in two forest communities—the villages of Fabe and Mosongiseli—near the southern portion of the Korup National Park in Cameroon, has broadened our understanding of relationships between aspects of rural community agency under REDD+. This would enhance the ability of policymakers to make informed decisions about the REDD+ program development process and thus, contribute to the improvement of REDD+ interactions with local communities. The study accounts for the communities’ understandings of the forest and forest policies by showing how understandings inform the opinions and intentions of members of the communities, explaining access to decision-making and access to forest land and resources as factors of empowerment. The study attempted to demonstrate the ability of communities to further their common interests, even when they are not advancing those interests in a collective way. In addition, the study has sought to demonstrate the existential threats of conflicts when REDD+ is implemented in communities without first addressing ownership claims and management of forest spaces.

Empirical findings of the study show that diversity in power relations characterizes community agency in the context of forest conservation, industrial agriculture, and community forestry. The communities’ experiences with industrial agriculture—Herakles and Pamol—show that there is the potential for conflicts to emerge, particularly between those with greater traditional authority in decision-making. That Herakles has claimed carbon rights in its establishment convention with the government of Cameroon is evidence that private entities could be involved in the communities’ carbon market when REDD+ is implemented to the exclusion of local communities.

In summary, these communities value the forest not only because of their strong dependence on its natural products and farmland, but they also appreciate the forest for a variety of social and cultural reasons. This suggests that the communities have reasons to implement effective conservation measures to preserve their heritage for themselves and future generations. Thus, REDD+ may offer the possibility of filling this livelihood gap if implemented in a way that respects the realities of forest societies and the necessities of life. Therefore, if these communities have the capacity to effectively assert their agency in Cameroon’s REDD+ development and implementation process, they will be in a better position to prevent the loss of livelihoods while gaining from REDD+ benefits.