Local people and conservation officials’ perceptions on relationships and conflicts in South African protected areas

ABSTRACT Protected areas (PAs) are often conflict-ridden, but conflict resolution mechanisms are often constrained by little appreciation of the perceptions of the principal agents (PA managers and local communities) about such conflicts. Getting local people’s support in PA management efforts is considered important for achieving conservation and livelihood goals. Using data from 13 nature reserves in South Africa, this study explores the perceptions of reserve managers and local communities about their relationships and the existence and underlying causes of conflicts. The findings showed sharp contrasts in perceptions between reserve managers and local communities. Reserve managers generally perceived that there were no conflicts with local communities and that their relationship with them was positive while local communities thought otherwise, claiming conflicts were centred around restricted access to PAs, lack of benefits from PAs and communication problems. These findings have profound implications for conservation, especially considering the importance of getting local people’s support in PA management. EDITED BY Isabelle Durance


Introduction
For a greater part of the twentieth century, biodiversity conservation in protected areas (PAs) was characterised by socially exclusive 'fortress' strategies aimed at creating 'pristine environments' (Hulme & Murphree 1999;Adams & Hutton 2007). More often than not, the proclamation of PAs informed by fortress approaches resulted in economic and social costs to local communities such as the loss of land ownership rights and access to natural resources, which in turn strained relationships and generated conflicts between local communities and conservation officials (Brockington 2002;Brockington et al. 2006). A conflict is herein defined as a disagreement and dispute over access to use, control and manage natural resources. Conservation conflicts emerge out of competition between two or more actors over some aspects of biodiversity management and when there are perceptions that some actors assert their interests over others' interests (White et al. 2009;Chuenpagdee et al. 2013).
Against a backdrop of the shift towards peoplecentred conservation, literature on local people's perceptions towards PAs has been growing exponentially, as a basis for understanding and evaluating the impacts of conservation interventions (Kideghesho et al. 2007;Mutanga et al. 2015Mutanga et al. , 2016Bennett 2016). Perceptions are considered as the way in which a referent object is regarded, understood or interpreted by individuals to produce meaning (Lindsay & Norman 1977;Bennett 2016). Actors can perceive reality differently depending on their respective interests, experiences and knowledge among other factors. In other words, interpretations of reality are socially constructed, as influenced by people's history and surroundings (Bennett 2016). Though perceptions may differ from reality, they shape actions of individuals, and this means that differences in perceptions may form the basis of conflicts. Mounting evidence shows that perceptions towards conservation are shaped differently by a diverse range of demographic factors including education levels, gender, household size and age (Kideghesho et al. 2007;Mutanga et al. 2015) and socio-economic factors such as past experiences with PAs, wealth status, benefit accrual from PAs and the perceived state of relationships with PAs and PA officials (Allendorf et al. 2007(Allendorf et al. , 2012Kideghesho et al. 2007;Tessema et al. 2010). Socioeconomic factors like past experiences, benefit-sharing in PAs and conflicts with PAs are particularly considered key to understanding perceptions on conservation (Holmes-Watts & Watts 2008;Bennett 2016), given the goal of ensuring conservation with a 'human face'.
Proponents of people-centred approaches to conservation argue that if local communities have secured access to PAs, can benefit from PAs and participate in PA managementthey are likely to support conservation efforts (Balint 2006;Kideghesho et al. 2007;Cundill et al. 2013;Thondhlana et al. 2016). In fact, the shift to people-centred approaches in PA management is premised on ensuring social equity in conservation as a basis for minimising conflicts between local communities and conservation officials (IUCN 2000;Brockington 2003;Holmes-Watts & Watts 2008;Bennett & Dearden 2014). From a practical perspective, it is argued that promoting local people's participation in defining conservation goals including benefit-sharing arrangements may increase local support for conservation efforts since it builds trust and improve relationships with PA officials (Adams et al. 2004;Ezebilo & Mattsoon 2010;Tessema et al. 2010;Redpath et al. 2013;Mutanga et al. 2016). From an ethical position, while conservation can succeed even without people's participation (Brockington 2003), it should reflect moral norms as it is the right thing to do particularly in areas with a history of physical and economic displacement of local communities (Holmes-Watts & Watts 2008; Thondhlana & Muchapondwa 2014).
Central to local people's participation in conservation is decentralisation of decision-making powers from state to local authorities and committees that represent local people. Decentralisation in principle facilitates local communities' participation in the design and implementation of conservation strategies, which in turn ensures a shared understanding of problems and ways of solving them (Ribot et al. 2010). Without this, local communities are often against conservation efforts and can be set on a collision course with conservation officials (Tessema et al. 2010;Vedeld et al. 2012;Cundill et al. 2013;Thondhlana et al. 2016). Therefore, examination of perceptions of actors with an interest in conservation is increasingly becoming an important focus for conservation work aimed at addressing conservation conflicts. Adams and Sandbrook (2013) argue that an understanding of qualitative aspects of conservation like perceptions can provide better insights into the often power-laden, complex and messy policymaking processes.
However, while conservation-related conflicts have been increasingly studied, relatively few studies focus on the perceptions of protagonists (Adams et al. 2003;Mutanga et al. 2016). Many studies arguably focus more on local people's perceptions towards PAs and relationship with PA officials (Tessema et al. 2010;Allendorf et al. 2012) than on conservation officials' perceptions of their relationship with local communities with a few notable exceptions (McClanahan et al. 2005;Allendorf et al. 2012;Mutanga et al. 2016). Further, most studies are based on individual case studies such that efforts for conflict resolution are often fragmented (White et al. 2009). This also makes it challenging to generalise findings given varied contextual realities. Therefore, an understanding of the perceptions of both the PA officials and the local people on their relationships may provide the first step towards identifying points of conflict and mapping pathways for achieving positive relationships (Weladji et al. 2003;Allendorf et al. 2012;Bennett 2016;Mutanga et al. 2016;Thondhlana et al. 2016). According to Bennett (2016), an examination of perceptions can offer insights into how the socio-economic and ecological impacts of conservation are evaluated, understood and interpreted by people.

PA management in South Africa
PAs in South Africa consist of national parks and provincial reserves including marine parks and reserves managed by the national (South African National Parks) and provincial conservation agencies, respectively. The main mandate of the conservation agencies is to implement programmes to meet conservation targets such as expanding areas under protection consistent with the Convention of Biodiversity targets, which South Africa is a signatory to. Most PAs in South Africa have a historical legacy of forced displacement of local communities, which denied these communities access to land for various livelihood activities and involvement in PA management (Ramutsindela 2003;Kepe et al. 2005). In response, the South African government has, since the transition to democratic rule in 1994, attempted to redress this via granting local communities land tenure and use rights in PAs and promoting their participation in decision-making in PA management (Kepe et al. 2005;Holmes-Watts & Watts 2008;Cundill et al. 2013;Thondhlana et al. 2016). In postapartheid South Africa, the involvement of local communities in PA management is part of the national conservation discourse and policy, related to local community demands for historical redress for land alienation and more equitable access to natural resources as part of the country's broader socio-economic transformation (Ramutsindela 2003;Holmes-Watts & Watts 2008;Kepe 2008). Despite this attempt, many PAs are often marred by tensions epitomised by cases of periodic skirmishes and violent conflicts between local communities and conservation officials, even in cases where land settlement agreements have been finalised (Kepe et al. 2005;Fay 2007;Thondhlana et al. 2011Thondhlana et al. , 2016. Nonetheless, South Africa is pursuing the goal of integrating biodiversity conservation with local development in line with global trends. That there have been conflicts in South African PAs is an undisputed fact. However, of interest to conservation practitioners and researchers is the emergence of conflicts in an era where the government has increasingly made efforts to redress past land injustices (Ramutsindela 2003;Kepe et al. 2005).
In South Africa, despite the growing literature on PA management, there is a limited understanding of the perceptions of conservation officials and local communities on their relationships with each other and existence and sources of conflicts. A lack of understanding of the different actors' positions can make it difficult to embark on negotiated settlement agreements aimed at achieving conservation and livelihood goals. Thus, this study aims to reach a better understanding of the perceived relationships between conservation officials and local communities from both stakeholders' views using evidence from nature reserves in the Eastern Cape province of South Africa. An understanding of both actors' perspectives may allow deeper probing of the factors behind different perceptions as a basis for suggesting interventions to curb growing tensions in PAs. On a national level, this study may have profound implications for conservation, since many PAs in South Africa are under land claims by local communities who were previously displaced when these PAs were established. Beyond South Africa, the contribution of this study lies in informing a growing audience of actors with an interest in fostering positive relations between local people and PAs and meaningful people-centred conservation.

Study sites
This study was conducted at and around 13 nature reserves in the Eastern Cape province of South Africa ( Figure 1). The Eastern Cape province is the second largest province in South Africa, with an area covering about 169,580 km 2 . Its physical characteristics are quite variedarid areas of the Great Karoo are found in the northern and north-western parts of the province while the Drakensburg Mountains cover most of the north-eastern areas. The province is bordered by the Indian Ocean in the southern and eastern parts with a relatively long coastline. All the reserves included in this study, except for Oviston, fall within the boundaries of the Maputaland-Pondoland-Albany biodiversity hotspots, which are characterised by high levels of endemic and threatened species (CEPF 2010). The reserves are managed by the Eastern Cape Parks and Tourism Agency (ECPTA), a provincial conservation authority primarily responsible for conservation issues in the province. All the reserves (Table 1), ranging in size from about 160 ha to 15 000 ha, have a history of forced land displacement of local communities. Only one (Silaka) out of the five reserves with cases of land claims lodged by adjacent local communities (Table 1) has a settled land claim (Thondhlana et al. 2016). Resource use arrangements for local communities in the reserves varysome reserves allow resource use on a permit-based system, while others have 'no take zones', where resource use is strictly prohibited.
Each reserve in principle has a Reserve Foruma platform designed to communicate key reserve issues to stakeholders, including those issues affecting and relating to natural and cultural heritage conservation goals and the livelihoods of adjacent communities. Consistent with national parks, provincial reserves are required to engage in stakeholder participation, including meetings with local communities to foster good relationships. It is expected that community members via their representatives participate in all the processes of reserve management plans. It is required that at least four meetings per year with Reserve Forums take place in each reserve to ensure continuous information sharing. However, Reserve Forums have no decision-making powers.
The Eastern Cape province has the highest poverty levels in South Africa, with the majority of local people living below USD2 per day (Statistics South Africa 2012). Most PAs in the Eastern Cape province are located in rural areas inhabited by poor people (DEAT 1997;Pollard et al. 2003). Consistent with patterns in the Eastern Cape province, all the reserves, except for Kwelera and Commando Drift, are generally surrounded by poor local communities. Local people are involved in different livelihood strategies including arable agriculture, livestock production and natural resource harvesting. Most poor households are beneficiaries of government social welfare programmes and depend on state grants such as child-care, old-age and disability grants.

Data collection
Data collection took place between July and August 2014. We approached managers of 25 nature reserves under ECPTA's management and local community leaders in communities neighbouring these reserves. This represented almost all provincially managed reserves. We purposefully excluded two reserves, Dwesa-Cwebe and Mkambati, where long-term conflicts have already been well researched and documented. Fifteen officials from 13 reserves responded positively. We conducted face-to-face semi-structured interviews with reserve officials (reserve manager or the senior game ranger in instances where a reserve did not have a resident manager or both). We also interviewed 18 most senior community leaders (chiefs and ward committee members) of individual communities living adjacent to all the sampled reserves. Four out of 15 reserve and 8 out of 18 community respondents were females.
We employed an inductive approach to enable exploration and building of in-depth understanding of a complex phenomenon. We adopted a semi-structured approach with open-and closed-ended questions to encourage open and free-flowing discussions. Both the reserve officials and community leaders were asked about (i) their perspectives on the nature of their constituency's relationships with each other, with responses captured as negative, positive or neutral; and (ii) the existence, nature and sources of conflicts. Other themes that emerged during the course of the discussions were noted and recorded. To capture all the details of the discussion, we taperecorded the discussions with the full consent of the interviewees. Each interview, which took between 30 min and 1 h, was conducted in either IsiXhosa (the local language) or English depending on the interviewee's preference. Interviews were conducted with the help of an experienced translator, who has more than 15 years of experience of working with local communities and researchers. Interviews were supported by document analysis, which allowed us to examine arrangements around resource use by local communities and land claims in the reserves. The research was conducted with permission from ECPTA (Research Agreement No. RA0142) and local traditional leadership following a comprehensive elucidation of the research objectives and purpose. The anonymity and confidentiality of responses were explained to the informants before interviews, and the interviews were conducted with the respondents' informed consent.

Data analyses
After qualitative data were transcribed, content analysis was used to identify, summarise and synthesise the perceived relationships and sources of conflicts. Descriptive statistics were used, where relevant, to quantify the number of informants in agreement or disagreement with existence of conflicts and citing a particular cause of conflicts. Direct quotes from informants and quantitative data were used as evidence to support claims and express meanings (Newing 2010). The potential limitations of this study are that the respondents' perceptions may not be representative of community views and it may not be possible to generalise the implications of the findings. Thus, the results should be interpreted cautiously. Despite these limitations, the results provide valuable insights into the local communities' and reserve management's perceptions of their relationships and points of conflicts with each other, thereby providing the basis for potential interventions.

Perceptions on relationships
Only 2 of the 18 (11%) community informants interviewed felt the relationship with the reserve management was positive. Eleven (61%) of the community informants interviewed perceived their constituency's relationship with the reserve management to be negative while the remaining five said it was neutral. In contrast, nearly all reserve officials interviewed (12 of the 15) perceived their relationship with the surrounding communities to be positive, while the remaining three felt it was either negative or neutral. In general then, our results point to a disjunction in perceptions between local community and reserve informants about the nature of their relationships with each other.

Perceptions on the existence and sources of conflicts
Fourteen (78%) of the community informants interviewed believed that there were conflicts between their respective communities and reserve management. During discussions on the topic, several issues perceived to be sources of conflicts were raised by the community informants including restricted access to the reserve for resource harvesting and livestock grazing, lack of preferential employment, limited community developmental opportunities and lack of community consultation on reserve management issues (Table 2). It was also reported by two of the community informants interviewed that extension of reserve boundaries into community land without local people's approval was behind conflictual relationships. Analyses of responses seem to suggest that existence of the conflicts was mostly reported in communities who had settled or ongoing land claims in the reserves, located close to the reserves, and directly dependent on land-based activities (e.g. fuelwood harvesting and livestock grazing) for their livelihoods. However, all reserve officials interviewed except one attributed the positive relationship to granting of local communities access to reserves, preferential employment of local communities and consultation with local communities in the running of the reserves (Table 3).

Restricted access to reserves
Twelve community informants (67%) interviewed said that access to the reserve for livestock grazing, fuelwood collection and visiting culturally important sites were restricted (Table 2). Other community informants reported that fires often broke during dry periods, expressing concerns that this negatively affected the availability of fodder for their livestock. This source of conflict was also supported by one reserve official who said that: 'Elders have different attitude towards us, and I think the reason is that the elders have livestock and sometimes during drought their livestock face big challenges in getting grazing areas. They then ask us to allow their livestock to come and graze in the reserve but we can`t allow that, so we say no'. Some of the reserve informants blamed local communities for fire outbreaks. They also said that while resource use was granted in principle, access to resources in the reserves remained restricted due to fear of illegal harvesting of threatened plant species, and poaching of wildlife by local communities, clearly indicating a lack of trust between local communities and PA officials. Some community informants (5) claimed that they were denied access to reserves for the purposes of performing ritual activities. In one of the reserves, conflicts were reported between an indigenous Khoisan community and reserve management over access rights to ancestral land. The community informants claimed that the land occupied by the reserve belonged to them, citing that the graves of their fathers and forefathers and visible ancient drawings in the reserve were evidence of their historical occupation and ownership of the reserve land. The informants also said that the culturally important medicinal plants that they relied on for their health were found inside the reserve. Though there were reports of suggestions by reserve officials to grow the desired medicinal plants outside of the reserve, the local community rejected the offer arguing the healing properties would be lost if medical plants were cultivated. The following quote by a community leader captures the value given to performing cultural activities in the reserve and the concern arising from denied access: Table 2. Summary of community leaders' perceptions on causes of conflicts.

Perception
No of informants (n = 18) Sample statement by key informant Restricted access to reserves for livestock grazing, fuelwood collection and cultural activities 12 (67%) A few years back, we experienced a drought here, and our livestock suffered, we spoke to reserve management to help by allocating a portion of land for our livestock to graze. They promised us, saying that they will speak to senior people and we waited for a response and eventually the report arrived and the request was rejected. We used to be given loads of fuelwood when a household had a funeral or traditional ceremony, but that is now limited to staff members only and community leaders, but ordinary community members are not allowed to harvest. These are the issues that make our relationship with the reserve sour. Lack of preferential employment opportunities 12 (67%) It is not a good relationship because when there are job opportunities we are not told as the community. They have taken our land and gave us nothing. They don't even tell us how many vacancies are there or who is to be employed and as we are closest to the reserve we find it strange that when they employ people they take from outside this community. Lack of community development and support services 9 (50%) The reserve promised this community that they will support our children's education through funding generated from game sales. That promise was never fulfilled, and up to this day not a single child was supported by this reserve. We used to get a community share from reserve animal sales for community developmental programmes, but we are no longer receiving it. Lack of consultation with local communities 9 (50%) This reserve does not consult with the community when implementing their activities or plans. One example is when employment opportunities arise, they employ people without consulting with local leadership. This reserve is in our neighbourhood and would like to know what is happening inside there. Most people are not aware of what is going on inside the reserve; it would be nice as our neighbours to let us know. Expansion of reserves in communal land 3 (17%) We have demarcated the boundaries of our land and reserve land, but we have observed that they went beyond their boundaries into our own land when they were fencing off their land. We realized that they expanded the reserve and took much of our land leaving no space to graze for our livestock. In the spirit of working together with locals, we have opened our dam which is situated within the reserve for fishing at affordable rates.
Communities are allowed to harvest some resources they use during certain times of the year. Employment opportunities to local communities 12 Whenever there are employment opportunities, community members are the first to be offered those opportunities. 99% of the staff is from the community and also stays in the community; they have brothers, sisters, parents and uncles. We are one family here. Community consultation 4 We manage this reserve in partnership with the community, we work very closely, and our activities and plans are discussed and approved by them. For any activity or development that is to happen here, I have to consult with community. In other words, they are part of decision-making in terms of how the reserve is being run. Wealth status of community 2 Communities surrounding the reserve are farms not villages, and the relationship is very goodwe work together in many ways. Our relationship is very good. Our reserve is surrounded by small white residential areas.
'This community needs to perform traditional ceremonies inside the reserve because there are some rituals that are performed only on sacred pools of the river and those pools are inside the reserve. They (local people) used to do those rituals inside the reserve but now the reserve staff is not allowing them to perform those rituals'.
This suggests that local people's negative perceptions towards reserve officials are attributed to restricted access to nature reserves for both economic and cultural reasons. In South Africa, livestock production and collection of wild natural resources are key livelihood sources for many poor households living adjacent to PAs. However, most of the decisions are made by the ECPTA, and the reserve officials do not have the authority to make decisions with regard to livestock grazing. Thus, the inflexibility of the reserve officials in terms of local requests that are perceived to be legitimate by local communities, for example, livestock grazing in the reserves following fire or reserve access for cultural activities, is contributing towards conflictual relationships and negative attitudes towards reserve officials.

Lack of preferential employment opportunities
About 67% (n = 12) of the community leaders believed that a lack of job opportunities and preferential employment were behind perceived negative relationships and conflicts with the reserve management (Table 3). They expressed that despite the fact that employment-related benefits were often touted as key returns from reserves, not enough and meaningful jobs had been realised. The respondents further added that the limited job opportunities were often offered to 'outsiders' rather than locals. However, contrary to community informants' views, 12 out of the 13 reserve officials interviewed said that all the job opportunities were offered to local communities first, saying that jobs were only offered to outsiders if there were no qualified local residents, which yielded a good working relationship with local communities. Only one reserve official acknowledged that there were conflicts over jobs with local communities but attributed this to local peoples' lack of understanding of the skills needed for working in the reserves, saying: 'Some people are complaining about the system used in hiring people here. They don't understand that some positions need certain special skills which might not be available around here'. These results show clearly contrasting views between reserve officials and local community leaders regarding employment opportunities in reserves, and these differences in understanding could be the basis of negative relationships and conflicts.

Lack of community development and support services
Nine community leaders (50%) interviewed expressed a growing dissatisfaction with the reserve management, citing failure of community development projects to materialise. It was mentioned that some reserves had promised to cover the costs of their children's education through the provision of schools and bursaries, but nothing had materialised in their view. In comparison, only one reserve official appeared to support the communities' perspectives regarding lack of community development as a source of conflicts, saying that: Things are not the same, for example, people used to buy meat from the reserve and get sponsorship for local soccer clubs and soccer fields used to be maintained by the reserve staff and all that is no longer happening". This reserve official perceived that things had changed since the arrival of a new reserve managerwhom the surveyed official said seemed unresponsive to community social development projects. An official at one of the reserves explained that the communities were no longer benefiting from development projects because the ECPTA introduced a centralised tender system and communities had to register as companies for them to bid for tenders.
Other reserve informants said that development projects for local communities were desirable but not their primary mandatearguing that their mandate was predominantly conservation. Reserve informants viewed PAs as an important tool for biodiversity conservation that fulfils certain national and global conservation targets. For example, two reserve managers mentioned that their mandate, as directed from head office, was to implement programmes to meet conservation targets such as expanding areas under protection in line with the Convention of Biological Diversity targets, which South Africa is a signatory to. This demonstrates that though the idea of integrated conservation and development is embraced at policy levels, varied interpretations of what this means in practical terms could be a source of conflict between actors. It was also said by reserve officials that community development initiatives were hampered by budget constraints beyond their control.

Lack of consultation with local communities
Lack of consultation with local communities on reserve management issues was perceived as another source of conflict by nine (50%) of the surveyed community leaders. The informants believed a negative relationship was due to a lack of consultation with local communities, with no information provided to them when new job opportunities came up, when there were managerial changes in the reserve and when there were plans to extend the physical boundaries of the reserves. Some community leaders' typical remarks towards lack of consultation are provided in Table 2. However, community leaders' perceived lack of consultations was in stark contrast with reserve management perceptions (Table 3). All the interviewed reserve informants believed that they consulted all community members via Reserve Forums, before making decisions on yearly budgets, employment opportunities and other reserve management issues. The managers said that Reserve Forums provided a platform for communities to air their views and concerns regarding reserve management issues, as illustrated by typical statements in Table 3. The Reserve official's position on Reserve Forums was disputed by local community leaders, as illustrated by the following statements: We are not part of the Reserve Forum and have never heard of it yet the reserve management says it closely works with the community.
We don't have any such forums, and we really need them to improve working relationships. I can't even remember the last time we had a meeting between the game reserve and community members, probably 3 years ago.
Some community informants claimed that the level at which local communities were involved in decision-making for reserve management was no more than being informed about pre-identified and set goals. Two community leaders were also concerned about reserve expansion into communal land without consultation with or notifying local communities, which had obvious livelihood implications such as decreased livestock grazing areas ( Table 2). The contrasting perceptions presented here are perhaps symptomatic of a mismatch between community expectations of benefit accrual and involvement in reserve management and reserve officials' views on what this means in practice. Overall, the findings illustrated conflicting accounts between local community leaders and reserve officials on their relationships related to reserve management.
The results also showed that there were no reported conflicts in areas where (a) local communities lived far away from the reserves and hence there was neither a direct interaction with the reserve nor need for reserve resources, (b) reserves were surrounded by relatively well-off households (commercial farmers) who had no direct dependence on the reserve resources and (c) communities were made up of people without a history of displacement from the reserve. Typical comments regarding positive relationships and reasons for no conflicts included the following: The relationship is very good. Our reserve is surrounded by small white residential areas (Reserve official).
The relationship with surrounding community is good. Most if not all communities surrounding the reserve are white commercial farmers who work close with us. They tell us when the fence is broken and when our animals get out of the reserve and they even help us fix the fence. There is no poaching here at all (Reserve official).
Members of this community come from all over and are quite a new establishment and maybe people are focusing on building relationships with one another and strengthening their community spirit (Community informant).
Overall, our analysis of responses by reserve showed that positive relationships were reported at reserves located close to commercial farms but farther away from poor local communities. Commercial farming communities reported here are often sparsely populated and consist of well-off farmers, supporting the hypothesis that people who are less dependent on the direct use of natural resources for livelihoods often have positive attitudes towards PAs. Contrastingly, conflictual relationships were reported at reserves surrounded by densely populated and poor rural communities with a history of land dispossession, supporting the hypothesis that the need to access resources in the reserves for livelihood needs (firewood and livestock grazing) resulted in conflicts between local communities and PAs.

Discussion
This study aimed to better understand the perceived relationships and sources of conflicts between local communities and reserve officials and its potential implications with regard to leveraging support for conservation goals using evidence from nature reserves in the Eastern Cape province of South Africa. Although disparities in perceptions existed within actors, the findings generally showed sharp contrasts in perceptions between reserve officials and local communities. A majority of reserve officials generally perceived that there were no conflicts with local communities and that their relationship with them was positive while the later thought otherwise. These differences in perceptions between local communities and conservation managers are consistent with recent findings by Mutanga et al. (2016) in Zimbabwe. According to local community leaders' perspectives, conflicts were centred around (i) restricted access to reserves, (ii) lack of preferential job and other community development opportunities, (iii) a perceived lack of consultation with local communities in decision-making and (iv) reserve expansion into communal land. With respect to reserve access, the results illustrate that despite the increasing recognition of the importance of factoring the economic and cultural needs of local communities in setting conservation goals, the inability of reserve management to respond to community requests makes them appear socially and culturally unresponsive, which sets local communities and reserve management on a collision course. Taken together, the results are indicative of the fact that without addressing community needs and concerns, it may be difficult to leverage community support for conservation goals.
The local community leaders' concern about restricted access to reserves is not baseless. These findings corroborate previous findings that highlight resistance to conservation in cases where PAs impact local people's livelihoods. For example, evidence suggests that restricted access to resources for economic and cultural purposes yielded conflicts at Kgalagadi Transfrontier Park (Thondhlana et al. 2011(Thondhlana et al. , 2015 and Silaka Nature Reserve in South Africa (Thondhlana et al. 2016). Similar sources of conflicts have been reported at Mikumi National Park in Tanzania (Vedeld et al. 2012) and various parks in Zimbabwe (Mutanga et al. 2016). In western Serengeti, Tanzania, Kideghesho et al. (2007) found that inadequate access to pasture negatively affected attitudes towards conservation. Thus, the results support the view that conservation policies that do not factor in the livelihoods of poor communities may lead to the emergence of adverse social outcomes such as conflicts. However, Cundill et al. (2013) caution that promises of benefits from reserves ought to be informed by detailed quantitative assessments of what the various reserves can offer to avoid creating grandiose but difficult-to-fulfil expectations that could lead to conflicts in PAs.
Related to the benefits issue, the perceived failure of reserves to satisfy community development projects can be tracked to the integrated conservation and development literature, where local welfare improvement is widely considered critical for the success of conservation (Adams et al. 2004;Adams & Hutton 2007;Ezebilo & Mattsoon 2010). Often the support of local communities for conservation is leveraged through promises of community development projects such as development of infrastructure (e.g. roads, schools and clinics), job provision, provision of bursaries for school-going children and other support services. Therefore, without these, local communities are less likely to view conservation positively (Cundill et al. 2013). Thus going forward, increasing the visibility of reserves in contributing towards community infrastructure development projects while embarking on skills support for local communities to be employable in PAs may yield positive attitudes towards the PAs.
Our findings also show that many local community informants felt that there was no good communication between them and reserve officials. Though some of the reserves have Reserve Forums, established to allow participation of local people in PA management, local community informants perceived the forums to be either non-existent or merely platforms for rubber stamping and legitimising reserve projects without their meaningful involvement. A case in point relates to reports of reserve expansion into communal land without community's knowledge or approval, implying a lack of local community involvement in PA management. It is reasonable to suggest that this could explain their negative perceptions towards and the reported conflicts with reserve officials, in support of findings elsewhere (e.g. Bruyere et al. 2009;Thondhlana et al. 2015Thondhlana et al. , 2016. Further, in our view, expansion of PAs in community territories may invoke emotive experiences of past displacements, which could trigger negative perceptions towards PAs. It is important to note that conservation agencies have conservation interests as their primary motive, and in most collaborative arrangements, they initiate and facilitate the process of involving local communities (Cundill et al. 2013;Thondhlana et al. 2015). This means that conservation agencies are likely to assume a stronger position than that of local communities and assert and protect their interests at all stages of negotiations for collaborative management of PAs which can result in conflicts (White et al. 2009;Chuenpagdee et al. 2013;Thondhlana et al. 2016)a scenario parallel to 'fortress' conservation approaches. It is therefore conceivable to suggest that local people's negative perceptions towards PA officials may be shaped by a historical legacy of exclusionary approaches to conservation. According to Bennett (2016), people's perceptions are mediated by past experiences including tenure insecurity. Mombeshora and Le Bel (2009) argue that the history of PA creation may affect perceptions of local people towards PAsciting that there is often animosity towards PAs with a history of displacement and dispossession of local resident communities. Addressing this would require both local communities and reserve officials to build trust to ensure good relationships and pave way for mutually binding and beneficial agreements. Establishing open channels of communication may also allow difficult questions to be asked and addressed including the conflict between local communities' expectations (livelihood benefits) and conservation targets (reserve expansion). Given that most countries like South Africa set to expand their networks of PAs in line with global agreements, the contradictory perceptions on what these PAs can offer may exacerbate conflicts. Therefore, in support of Booth et al. (2009), we suggest that support for conservation goals in PAs may only be achievable if the nature and goals of conservation are understood by, known and acceptable to, all actors. This work has also revealed that although local participation is considered a key aspect of PA management, in reality, decision-making power is largely centralised and held by conservation agencies. A key source of tension appeared to be a perceived inflexibility on the part of reserve officials. It could be argued that while reserve officials may have an interest in responding positively to community concerns and needs, they actually do not have the power to make decisions regarding issues raised by communities. Rather, decision-making power is centralised in ECPTA executivewhich means that important decisions are taken by management officials who are possibly not in sync with local realities, which subsequently results in conflicts. It also undermines any gains that might have been made through the creation of Reserve Forums. Further, it is our view that in a centralised system, local community representatives may find it difficult to have a say in decision-making and their involvement would be merely nothing more than rubber stamping. Thus, local people's negative perceptions of relationship with reserves may be symptomatic of the unequal power dynamics in decision-making forums. West et al. (2006) in a review of the social impacts of PAs argue that top-down approaches by the state institutions that fail to consider local interests and practices often yield conflicts with local communities.
A growing number of critics argue that despite the local participation and benefit-sharing rhetoric of such PA programmes, local communities still do not participate in defining 'conservation and development' agendas (e.g. Adams et al. 2004;Holmes-Watts & Watts 2008). Particularly in South Africa, though the importance of community participation is embraced in the management of previously restrictive PAs, there is little progress in practice (e.g. Cundill et al. 2013;Thondhlana et al. 2015Thondhlana et al. , 2016. Instead agendas are pre-defined and set by conservation agencies who then try to get local community by-in, which subsequently yields conflictual relationships between local communities and conservation officials (Cundill et al. 2013). Thus going forward, consultations with local communities may help in co-defining goals for and expectations from conservation that reflect local realities. McClanahan et al. (2005) and Ciocănea et al. (2016) similarly suggest that opening up more communication channels between local people and reserve management may ensure that actors have a shared understanding of each constituency's goals for and expectations from conservation. However, we suggest as others have (e.g. Cundill et al. 2013) that it may be more beneficial to consider the use of 'neutral' actors as negotiators for local communities to trust the process. This is because conservation agencies are subservient to the statehence it is impossible for the state to represent the best interests of local communities while at the same time attempting to fulfil local, national and international conservation imperatives (Cundill et al. 2013).
Our findings showing positive relationships in reserves surrounded by sparsely populated and welloff communities corroborate findings elsewhere that show that wealthier households tend to be supportive of PAs and PA officials because they are not directly dependent on natural resources for survival (Kideghesho et al. 2007;Snyman 2014). Commercial farming communities reported in this study do not compete with PAs for land and access to natural resources for survival, as they earn their living through agri-business (livestock and crop farming). The opposite is true for poor households, whose livelihoods are directly dependent on natural resources. Thus, the results also support the findings by Mombeshora and Le Bel (2009) and Mutanga et al. (2016) who showed that strained relationships are often a result of a history of forced displacement that compromised livelihood activities including resource access for fuelwood and livestock grazing. Often local communities responded to restricted conservation approaches via illegal resource harvesting, poaching of wildlife and destruction of PAs through fire, which only saved to perpetuate conflicts and negative relationships between local communities and PA officials. Addressing these conflicts would require promoting interactions between local communities and reserve officials, for example, through regular and consultative Reserve Forum meetings where the interests, goals and responsibilities of each actor are sincerely shared and discussed. Further, communication between reserve officials and ECPTA should be improved as a potential solution to some of the issues in the reserves.

Conclusion
This study has provided insights into the differences in perceived nature of relationships and existence and sources of conflict between local communities and conservation officials, as a basis for designing interventions aimed at eliminating or repairing conflictual relationships. We have highlighted that examining both actors' perspectives is profoundly important because local communities are often directly affected by conservation efforts while conservation officials have to implement any policy changes. In other words, achieving the goal of integrating biodiversity conservation with local development can translate from principle to reality if the views of local communities and conservation officials are considered. From the communities' side perceptions towards PAs are mediated by an array of factors including restricted access to PAs, lack of benefits accrual from PAs, deeply held cultural values and beliefs and communication problems. On the other hand, reserve officials believe their mandate is primarily conservation. Such conflicting understanding of conservation goals and expectations points to the fact that conflict resolution will not be any easy process.
Ameliorating conflicts would require making mutual concessions, but this can only be achieved if communication channels and consultations are believed to be sincere. For example, in cases where reserves may be unable to produce sufficient benefits for local communities due to their location, size and large number of people expecting benefits, these issues should be openly discussed. To achieve this however, it is critical that PA management first develop good working relationships with local communities via comprehensive dialogue, listening actively to local communities and undertaking a needs-based analysis as a basis for developing conservation strategies in line with community expectations. Without establishing positive relationships between local people and conservation officials, simmering discontent may escalate into bigger conflicts that could jeopardise both conservation goals and livelihood needs. Going forward, lessons from this study point to the fact that interventions for conflict resolutions should address local livelihood needs and cultural values. Further, we believe that the issue of centralisation of power and therefore the dehumanisation of conservation is a major aspect that should be addressed in conservation to improve relations between local people and PAs.