CONCEPTUAL DEVELOPMENT

Like democracy, CBNRM is both an idealistic theoretical goal and an imperfect practical struggle towards this goal (Dahl Reference Dahl1998). The conditions for successful CBNRM are onerous, requiring principles that incorporate (1) the economic allocation of resources to higher valued uses, (2) the political organization of society to do so, (3) creating technical, environmental and social capabilities far beyond environmental technocrats, and (4) building in the capability to learn from and manage complex social ecological systems adaptively (Murphree Reference Murphree1991). The conditions for democracy are similarly onerous, and the history of mankind shows how rarely these conditions have been met (Tilley Reference Tilley2007): the light of democracy flickering briefly around the times of Greek and Roman civilization, but democracy only emerging more generally in the last two centuries.

CBNRM is conceptually far richer than community conservation seen as park outreach or environmental education, or the passive receipt of benefits. CBNRM requires a fundamental shift in the political economy of natural resources. It transforms the role of the bureau-technical elite from managerial control to research, facilitation, institutional design and conformance monitoring. It shifts the benefits, power and responsibility for natural resources from a few highly qualified scientists and bureaucrats into the hands of thousands of rural people living with, and depending upon, biodiversity resources (Murphree Reference Murphree1999; McShane & Wells Reference McShane and Wells2004).

The new institutions introduced by CBNRM convert low-value uncontrolled uses into controlled high-value uses that include both monetary values (for example income from hunting) and non-monetary values (for example proprietorship or ecosystem services). Yet CBNRM's benefits extend beyond this economic and environmental realm. Since development and poverty reduction are multi-dimensional, perhaps the greatest value of CBNRM is political transformation through democratization, equity, transparency and accountability.

It is this vision that has captivated and motivated CBNRM scholar-practitioners in southern Africa for the past three decades (Metcalfe Reference Metcalfe, Western, Wright and Strum1994). However, operationalizing this vision requires considerable tenacity and a well-crafted implementation policy which in turn depends on a deep and interdisciplinary technical and scholarly understanding of issues like the economic allocation of bio-resources, the political organization of society, capacity-building and collaborative adaptive management of complex social ecological systems.

ECONOMICS, PROPERTY RIGHTS AND RESOURCE ALLOCATION

The London Convention of 1933, a response to the depletion of Africa's wildlife associated with white exploration and emerging technologies and markets, radically altered the political economy of wildlife (Child Reference Child, Suich and Child2009a). On the positive side, it set off the movement to establish protected areas that, today, are so important for the environment and economy. However, it also centralized the control of wildlife (and most bio-resources like forestry) in the state, and severely restricted the commercial use of wildlife. Coupled with the expansion of human population and agriculture, these policies were devastating to wildlife by the 1950s because they encouraged resources that were owned and marketable (such as agriculture and livestock) to replace resources like wildlife, which were neither. Policy makers gradually recognized that the primary threat to wildlife was no longer the free-for-all hunting and trading of a frontier economy, but the replacement of wildlife habitats by the cow and plough: if wildlife could not compete for space economically, it would continue to disappear at an alarming rate (B. Child Reference Child1995). Forward thinking conservationists, including the eminent zoologist Reay Smithers, intuitively recognized that the solution lay in devolving property rights and allowing wildlife to pay its way. Starting in the 1960s, southern Africa began to challenge the assumptions of conventional conservation, iteratively developing working examples and conceptual foundations for a new ‘sustainable use’ approach (SASUSG [Southern Africa Sustainable Use Specialist Group] 1996; Suich & Child Reference Suich and Child2009). These changes were driven by a core of motivated professionals in wildlife agencies and ministries with a long term commitment to the wildlife sector who, through lasting interpersonal connections (for example annual week-long professional meetings), simultaneously modified legislation across the region to devolve the authority for wildlife management to private landholders (Child Reference Child1971; G. Child, personal communication 2008). The evolution of CBNRM, similarly, owes much to these regional ‘communities of practice’.

The rapid recovery of wildlife on private land in southern Africa quickly validated the radically new conservation paradigm of devolved incentive-led conservation, albeit in the institutionally simple circumstances of private property and commercial farming. This was an important precursor to CBNRM, and provided a reliable commercial model (Bond et al. Reference Bond, Child, Harpe, Jones, Barnes, Anderson and Child2004) based on the competitive marketing of hunting and tourism by private outfitters. Thus wildlife provided the economic foundation for the CBNRM experiment, with the challenge being to develop appropriate institutions whereby thousands of communal inhabitants could benefit from and manage wildlife. Although the CBNRM model was developed using wildlife, its principles apply generally to collective management of wild resources.

Wildlife officials recognized that the financial value of wildlife did not reflect its comparative economic advantage (namely undistorted costs and benefits seen from a societal perspective) because landholders bore the costs of wildlife on their land (such as use of grazing and water, or conflicts) but were excluded from legal benefits. Market failures were partly corrected by encouraging sustainable commercial use and devolving wildlife use rights to landholders. The theoretical objective of these new policies was to internalize both the costs and benefits of wildlife so that improved pricing signals allocated land to higher valued uses (Stroup & Baden Reference Stroup and Baden1983, Martin Reference Martin2008). It was also recognized that property rights and pricing signals extended well beyond financial benefit to intangible values like the sense of security that encourages landholders to value, notice and husband their environment.

The proponents of the ‘sustainable use approach’ believed that economic institutions that reflect the true value of natural resources in allocation decisions would enable wild resources to outcompete economically and environmentally questionable agriculture and livestock practices in semi-arid savannahs where rain-fed crop production was not viable. They deliberately removed many bureaucratic practices that undervalued wildlife, such as licence fees for hunting and tourism, outdated restrictions on commercial use, bureaucratic hurdles on landholders deciding off-take levels that would never be countenanced by livestock producers, and subsidized access to hunting and tourism on state land that undercut the market.

The strongest argument against privatizing wildlife is that it is a fugitive resource associated with too many externalities to trust market price solutions. The conventional solution has been to scale rights at a sufficiently high level (to the state) so that costs and benefits are internalized; unfortunately this ignores the problem that feedback loops (such as mechanisms of accountability) are too long to be meaningful, especially to landholders who determine de facto land use.

Zimbabwe (and its neighbours) experimented with an alternative approach to the problem of fugitivity, and stumbled across the issues of scale and sequencing that Murphree (Reference Murphree2000) discussed so eloquently. Following the principle of subsidiarity (Handy Reference Handy1994), the rights to wildlife were scaled out to individual landholders. While the activities of many species like smaller animals with smaller ranges could be internalized at this level, there was need for a mechanism to scale up this process to manage the spatial or temporal externalities associated with larger animals (or processes like soil erosion). Zimbabwe recognized that ‘one-size-fits-all’ national regulation was insensitive to the specific needs of different wildlife species and geographic areas, reduced the relative value of wildlife and, furthermore, was unenforceable. Instead, they experimented with an intriguing example of devolved regulation (G. Child Reference Child1995). Through the Natural Resources Act of 1951, farming areas (usually catchments) voluntarily created self-regulating communities (called Intensive Conservation Areas or ICAs). These were effective at managing externalities like soil erosion, deforestation and overgrazing through mechanisms of peer pressure, education and, occasionally, their legislative powers to order and enforce destocking or anti-erosion measures. In 1975, ICAs were co-opted to manage the externalities associated with wildlife. This allowed the community, rather than the state, to establish their own regulations, such as setting off-take quotas for valuable and mobile species (such as sable). This experience demonstrated the efficacy of building a regulatory framework from the bottom up, including specificity of design (to maximize benefits and minimize costs), jurisdictional parsimony (Murphree Reference Murphree2000), social legitimacy and the lower costs of peer-based implementation. It also improved wildlife's competitiveness by reducing the regulatory costs and increasing regulatory conformance by managing the externalities associated with fugitive wildlife resources.

The economic model developed for wildlife ranching, and bureaucratic comfort with bottom-up collective action, laid an important foundation for the emergence of CBNRM.

Even before black majority rule, Zimbabwe's wildlife authorities introduced small but critical changes through parliament at a time when empowering black people in communal lands was controversial (G. Child Reference Child1995). In the late 1970s, for example, the wildlife agency forged a parliamentary agreement that income from wildlife in communal lands (and from game reserves established in communal areas) would circumvent Treasury and be paid directly to district councils for projects in communities that lived close to wildlife. Hence, administratively, Zimbabwe's Operation WINDFALL [Wildlife Industries New Development For All] programme was born (Murundigomo Reference Murundigomo1992; Matzke & Nabane Reference Matzke and Nabane1996), the most positive outcome of which was to show that building community projects through councils was a far cry from achieving the economic principles described above for private land. Operation WINDFALL coincided with a failed attempt at top-down land use planning to address land and wildlife degradation caused by immigration in Zimbabwe's Sebungwe region (Martin Reference Martin1981), and evolved into the CAMPFIRE [Communal Areas Management Programme For Indigenous Resources] programme (Martin Reference Martin1984a, Reference Martin, Bell and McShane-Caluzib, Reference Martin1986). Martin conceptualized the transfer of the economic principles described for private land into the more complex administrative conditions of communal land by devolving wildlife proprietorship and benefits to self-defining ‘village companies’, and removing as many central fees and controls as possible. In naming this the Communal Areas Management Programme for Indigenous Resources, Martin (personal communication 2008) argued that there would be many surprises along the way, but provided communities were legislatively empowered and performance was tracked, an adaptive management process (as opposed to blue-print planning) could address these.

POLITICS, DEMOCRATIZATION, ORGANIZATIONAL DEVELOPMENT AND COGNITIVE EMPOWERMENT

Martin's conceptualization established an idealistic theoretical goal for CBNRM, a vision of natural-resource-based rural development that has provided powerful motivation in the imperfect practical struggle towards this goal. This vision is built around the principles that free discretionary economic choice allocates land and resources to high value uses, is good for both the environment, and individual and societal well-being. However, discretionary choice also provides individuals and communities with political and economic freedoms, so that CBNRM can also be a powerful democratizing process.

CBNRM advocates a shift from top-down planning by the bureau-technocratic guardians of global biodiversity, to a more organic and complex approach that places more rights, responsibilities and opportunities in the hands and service of ordinary people. Working out the most effective configurations of rights, responsibilities, organizations, institutions and collective action is a daunting challenge, the complexity of which can only be addressed adaptively. The devolution of rights and responsibilities from the centre to communities through CBNRM is pioneering this process. However, devolution requires a well crafted and disciplined process (Peters & Waterman Reference Peters and Waterman1982). Simply ‘letting go’ is a recipe for anarchy and chaos, a symptom of which is local elite capture at lower levels.

The importance and nature of local democratization in CBNRM is hotly debated in southern Africa. The main arguments for representational democracy are logistical or related to managerial effectiveness; larger communities can afford to employ specialists to manage, for example, their natural resources and, in any case, it is not possible to service large numbers of small communities. Furthermore, they argue that members are too busy to engage fully in the many processes, and that representational or indirect democracy is managerially efficient (Overdevest Reference Overdevest2000).

The counter-argument is sustainability requires that discretionary power and governance is rooted in individuals through a democratic process, so that CBNRM must evolve from and encapsulate direct participatory democracy, even if functions are delegated upwards later.

The essence of participatory democracy is that all significant decisions are made face-to-face by the whole community, which limits communities, ideally, to about 150 members (Dunbar Reference Dunbar2003) but pragmatically to a few hundred households. However, most CBNRM communities are larger than this, so that decision-making power is vested in elected representatives linked to the constituency through Annual General Meetings and elections. Proponents of participatory democracy argue that face-to-face participation is essential for building a sense of democratic citizenship, proprietorship and accountability (all highly valued by members) and that, ultimately, interpersonal accountability is essential for managerial effectiveness. As Murombedzi (Reference Murombedzi1992) warned, CBNRM can promote recentralization at the local level, which may improve natural resource management, but, without face-to-face accountability, risks the elite capture and alienation that CBNRM is designed to overcome.

In our preliminary analysis of CBRNM communities in Zambia, Zimbabwe, Namibia and Botswana, the results from small single villages differ clearly from large multi-village communities (Child et al. Reference Child, Mupeta, Lubilo and Diane2007). In multi-village communities, ordinary people are invariably less informed about natural resources and finances, report low levels of participation, empowerment and personal involvement, are concerned about financial mismanagement, and receive few direct benefits in the form of cash or social projects. Single-village community-based organizations [CBOs] are rare, limiting our sample to three villages in Botswana, two in Zimbabwe and 45 Village Action Groups in Zambia that were closed down by the Zambia Wildlife Authority in 2002. In Botswana, the only country where a direct comparison exists, levels of knowledge, participation, buy-in and benefit were far higher in single villages (B. Child, unpublished data). In Zambia, the single-village CBOs in Lupande performed an order of magnitude better than their multi-village counterparts, constructing large numbers (>250) of social projects, paying dividends to c. 20,500 people annually and demonstrating high levels of participation (70,000 + person days annually) and accountability (<1% of annual village income of US$ 180,000 unaccounted) (Child & Dalal-Clayton Reference Child, Dalal-Clayton, McShane and Wells2004). Similarly, Mahenye and Masoka communities are the poster-children of Zimbabwe's CAMPFIRE programme (Taylor & Murphree Reference Taylor and Murphree2007; Murphree Reference Murphree2009). Despite the turmoil of local leadership changes and Zimbabwe's political and economic travails, both programmes in these communities apparently ‘collapsed’ at least once, yet have demonstrated the capacity to recreate themselves. High levels of democratic participation in the formative years of these programmes (Child & Peterson Reference Child and Peterson1991; Child et al. Reference Child, Ward and Tavengwa1997) have apparently altered people's paradigms of governance, empowering them with a ‘capacity to aspire’ (Rihoy et al. Reference Rihoy, Chirozva and Anstey2007) to something better than governance by a local, predatory and unaccountable elite. When circumstances were negative, people have bided their time, but have recreated participatory governance when favourable opportunities have presented themselves (M. Murphree, personal communication 2008; C. Jonga, personal communication 2008). While the same desires are prevalent in Lupande in Zambia internal community dynamics have been unable to overcome a (dis)enabling macro-environment.

Financial data emphasizes the dichotomy between single- and multi-village CBNRM communities. In single villages, some 60–80% of CBNRM income is allocated to projects or household benefits, and most people understand both their income and expenditure. In multi-villages, ordinary people have little clue about their finances, and it is rare for even 20% of income to be used for benefits through social projects or cash dividends; overhead and management costs invariably absorb over 80% of the income. Where oversight is good, overhead expenditure is used for salaries of employees like game guards, sitting/meeting allowances and vehicles (R. Diggle, personal communication 2009). Some argue that overheads are the legitimate cost of natural resource management, and that employment is a benefit rather than a cost. On the contrary, committees with a vested interest in boosting overheads seldom make good budgetary trade-offs between those costs and broader community benefits like cash and projects, and employment is often skewed towards the favoured few and is not always productive. Ultimately, a representative committee can manage natural resources well, but if ordinary people are excluded from participation and benefits, this will compromise sustainability. With strong oversight (top-down accountability) often funded by donors, representational CBNRM works for wildlife even where it does little for ordinary people. However, where both top-down and bottom-up accountability (Ribot Reference Ribot2008) are missing, as happens with Community Resource Boards in Zambia, money is consumed by sitting allowances, unpaid loans or is misappropriated with few benefits to people or the resource base.

Cases where management by a committee or even a self-appointed elite is unencumbered by upward and downward accountability misrepresent the concept (often within the context of short-term projects; Hulme & Murphree Reference Hulme and Murphree2001), are all too common, and give CBNRM a reputation of being in crisis (Adams et al. Reference Adams, Aveling, Brockington, Dickson, Elliott, Hutton, Roe, Vira and Wolmer2004; Brockington Reference Brockington2007). However, these imposters strengthen the case that CBNRM is not achieved by wishful thinking but through carefully crafted institutions that get both prices and politics right, and are administered rigorously if sensitively. Successful cases of CBNRM tend to (1) internalize costs and benefits locally and (2) incorporate participatory democracy. Rigorous procedures start with well-designed constitutions that place the origin of decisions and government in ordinary people through carefully articulated roles, responsibilities and procedures such as:

• • annual (or at most bi-annual) elections,

• • face-to-face revenue allocation (annually) and through participatory quarterly assessment of expenditure compared to budget,

• • genuine participation in natural resource management through processes like quota-setting (Rigava et al. Reference Rigava, Taylor and Goredema2006) and open competitive marketing, and

• • external auditing of finances and conformance to democratic procedure.

Top-down accountability (and capacity-creation) is provided by:

• • long-term light-touch facilitation and technical support, and

• • protection of the policy and procedural environment that allows participatory democracy to occur.

Both Murphree (Reference Murphree2000, Reference Murphree2004) and Ostrom (Reference Ostrom1990) recognize that the central dilemma for governing wild resources is to balance the optimal scales for human (small) and ecosystem (large) management through nested hierarchies. Thus ‘governance’ concerns the sequencing and configuration of power and accountability across micro-, meso- and macro- levels (such as individuals, villages, districts, provinces and states), and includes government, the private sector and an emerging civil society that operates locally, nationally and internationally.

Because of its origins in natural resource management, many CBNRM initiatives are, in the name of efficiency, scaled to rely on representational governance. This necessitates considerable top-down oversight to counter elite capture and corruption, but is ineffective for achieving the goals of equitable participation and benefit sharing. So how can both constituency accountability and efficiency be developed? Through careful sequencing, CBNRM must first be scaled down to the level of participatory democracy to create accountability and social legitimacy. Then, and only once people are practiced in their democratic rights and responsibilities, appropriate functions are delegated upwards through an organic process to take advantage of scale-efficiencies. Thus participatory democracy (including the devolution of property and use rights to this level), becomes the fundamental building blocks of the new form of natural resource governance, and once these building blocks have solidified, appropriate functions are delegated upwards to take advantage of efficiencies or to manage externalities. This approach (1) greatly reduces the risks of elite capture associated with representational governance, (2) creates stronger proprietary links between people and wildlife resources, and (3) roots conservation and development goals within a democratizing process.

At the micro-level, CBNRM principles are conceptually and operationally generic. However, establishing the cross-scale enabling policy environment for CBNRM to emerge is historically, administratively and politically specific. This is illustrated by the linked, but also divergent, history of CBNRM in southern Africa, which we will briefly review to identify common success factors, factors that worked well in at least one country, and factors about which there is still considerable debate.

CAMPFIRE IN ZIMBABWE

With its innovative wildlife department, effective national natural resource movement and strong university, Zimbabwe's well-documented CAMPFIRE programme (Child Reference Child1993, Reference Child1996; Metcalfe Reference Metcalfe, Western, Wright and Strum1994; Child et al. Reference Child, Jones, Mazambani, Mlalazi and Moinuddin2003; Murphree Reference Murphree, Brosius, Tsing and Zerner2005, Reference Murphree2009; Taylor Reference Taylor, Suich and Child2009) made the early running in the evolution of CBNRM. CAMPFIRE was initiated by a cohort of professional civil servants in the wildlife agency who amended the Parks and Wild Life Act of 1975 to devolve the ‘appropriate authority’ for managing and benefiting wildlife to district councils. In what Murphree (Reference Murphree, Brosius, Tsing and Zerner2005) termed a ‘pragmatic compromise’, the conceptually pure intention to devolve authority directly to ‘producer communities’ was thwarted by the government's insistence that rights be allocated to district councils, a situation partially corrected by a verbal agreement to ensure that at least 50% of revenues reached communities. CAMPFIRE acquired considerable conceptual rigour and operational coordination through the CAMPFIRE Collaborative Group, which brought together an eclectic group of wildlife managers, social advocates, political scientists and local politicians. Through a well facilitated process, the roles and goals of the partners were clearly defined and lead responsibility was passed to the nascent CAMPFIRE Association, a political producer organization that represented CAMPFIRE districts. Key features of CAMPFIRE were:

• • clear conceptual principles, operationalized through the ‘CAMPFIRE Guidelines’;

• • implementation partnerships in which goals negotiated by government and the producer association were supported operationally by effective non-governmental organizations (NGOs);

• • regular monitoring and peer-review of the allocation of benefits between communities, administration and council overheads by the wildlife agency and through CAMPFIRE Association meetings;

• • the development of open competitive marketing for hunting and tourism concessions (B. Child Reference Child1995; Child & Weaver Reference Child and Weaver2006);

• • experimentation with participatory democracy and revenue distribution (Child Reference Child2006)

• • a capacity-building process where district councils had a substantial involvement in community capacity-building and support, and provided general information while developing participatory technical management approaches (such as quota-setting) through a participatory technology development process (Goredema et al. Reference Goredema, Bond and Taylor2006); and

• • exceptionally strong support from government in its initial phases, but a commitment to the CAMPFIRE principles that is readily observable today (M. Mutsambiwa, C. Jonga, personal communication 2008).

CAMPFIRE was conceived and implemented by Zimbabweans, with the United States Agency for International Development (USAID), Norway and others providing some US$ 30–40m in support. Carefully crafted, CAMPFIRE immediately worked and, given its dependence on elephant hunting, influenced global debates about sustainable use and trade in endangered species (Duffy Reference Duffy2000; Oldfield Reference Oldfield2003). Despite Zimbabwe's later political instability and the withdrawal of most donor and technical support, CAMPFIRE was still performing strongly in the mid-2000s, including generating and devolving revenues and protecting wildlife and habitats (Child et al. Reference Child, Jones, Mazambani, Mlalazi and Moinuddin2003; Murphree Reference Murphree2009; Taylor Reference Taylor, Suich and Child2009). However, in a national culture of weakening public accountability and increasing impunity, centralization and elite capture at a local level became problematic (Mapedza & Bond Reference Mapedza and Bond2006). Nevertheless, cumulative data showed a recovery in devolved income from 20% in 2002 to 55% by 2008 (CAMPFIRE Association 2007), and some communities (with a participatory track record) partially reversed centralization and local elite capture (Rihoy et al. Reference Rihoy, Chirozva and Anstey2007; Taylor & Murphree Reference Taylor and Murphree2007). Both the wildlife agency and the CAMPFIRE Association still articulate the original CAMPFIRE principles.

CBNRM IN NAMIBIA

Since 2000, following Zimbabwe's problems and assisted by the World Wide Fund for Nature's ability to raise money, the Namibian national CBNRM programme has led the way in the region with a strong emphasis on natural resource management and monitoring, and on revenue creation. CBNRM in Namibia has its roots in the mid-1980s, when Garth Owen-Smith initiated a community game-guard scheme (Jones & Weaver Reference Jones, Weaver, Suich and Child2009) to protect desert rhinos and, through local participation, to recreate dignity lost during the apartheid era of top-down conservation and poaching by the South African military. CBNRM was then institutionalized by dedicated bureaucrats who, learning from personal contacts with CAMPFIRE, crafted the legislative foundation for CBNRM in Namibia to devolve wildlife use rights and all benefits to self-defined communities. This established community ‘Conservancies’ as independent structures, side-stepping local government to avoid Zimbabwe's experience where district councils absorbed a disproportionate share of wildlife revenues (Jones & Murphree Reference Jones, Murphree and Child2004).

Initially, Namibia's CBNRM programme sought social empowerment by using local men and women to monitor and manage natural resources (M. Jacobsen, personal communication 1993). Once legislation formalized ‘Conservancies’ in 1997, the adoption of open competitive marketing led to rapid income growth (NACSO [Nambian Association of CBNRM Support Organizations] 2008), and CBNRM in Namibia adopted similar economic language to CAMPFIRE (Ashley et al. Reference Ashley, Barnes, Brown and Jones1997). The Namibian programme is well organized through NACSO, with a rapid increase in registered and emerging Conservancies, and clearly demonstrates the recovery of wildlife populations including endangered species (Jones & Weaver Reference Jones, Weaver, Suich and Child2009). CBNRM is a moving target, where success generates new challenges. Rapid growth now places insatiable demands on a small group of technical experts. Self-generated income is quickly replacing external funding, but cannot as easily be supervised by NGOs, suggesting that emerging financial problems require a new system of governance based on bottom-up accountability and high levels of local participation and benefit.

CBNRM IN BOTSWANA

Unlike Zimbabwe and Namibia, CBNRM in Botswana was developed by USAID-funded external technical expertise. CBNRM communities were defined, sometimes arbitrarily, and received wildlife revenues through the mechanism of a 15-year land lease. The open and competitive allocation of hunting and tourism concessions to the private sector quickly generated income (Rozemeijer Reference Rozemeijer, Suich and Child2009). However, institutional design was neglected, and much of the current political threat to CBNRM in Botswana arises from problems spending this income: money is wasted or misused by committees, and levels of understanding, participation and benefit by local people are generally low.

Botswana never developed effective national civic structures like NACSO (which coordinates NGOs) or the CAMPFIRE Association (community political representation), weakening implementation and the ability of communities to advocate politically. Dedicated government officials play many of the capacity-building roles performed by NGOs in other countries, but are officially precluded from lobbying for greater community empowerment. Understandably, they are frustrated by proposals to recentralize wildlife management, believing that financial mismanagement by committees should be resolved through institutional design, not by removing rights and benefits from local people.

ZAMBIA

In weak states like Zambia, CBNRM tends to ‘bubble up’ (Wilson Reference Wilson, Brosius, Tsing and Zerner2005) in response to individual champions and donor initiatives, but outside of any national programme or policy. Two well known examples in Zambia are the Luangwa Integrated Rural Development Project (LIRDP) and Administration Management Design (ADMADE) (Gibson Reference Gibson1999). LIRDP was initiated by President Kaunda and the Norwegians, and attempted to control rampant elephant poaching in South Luangwa, partly through improved law enforcement, but also by pursuing rural development through a top down, but integrated planning process. Communities were supposed to benefit from considerable Norwegian investment and a 40% share of hunting and park fees, but most of the money was absorbed by transaction costs (meetings and allowances) or captured by local elites. In 1996, the CBNRM component of this project was radically altered. Eighty per cent of hunting revenues (US$ 220 000–240 000 annually) were devolved to some 45 participatory democracies (Village Action Groups), inculcating many of the CAMPFIRE principles. For a time these units performed extraordinarily well (Child & Dalal-Clayton Reference Child, Dalal-Clayton, McShane and Wells2004), but were based on project agreements not legislation. Key personnel in the project and the Norwegian Embassy ensured procedural conformance to CBNRM principles and, in the absence of legislated rights, protected the project at the national level. However, once these personnel left, over 70% of revenues were immediately captured by the Zambia Wildlife Authority. The remainder was paid to newly legislated and representative Community Resource Boards established through a poorly conceptualized and managed national CBNRM programme, where it was absorbed illegally or unproductively (Malenga Reference Malenga2004). Similarly, ADMADE had some successes as a community game-guard programme co-managed by chiefs and the wildlife agency, but it also succumbed to recentralization.

In the 1990s, donors funded a plethora of fashionable CBNRM projects, which produced little more than a couple of years of frenetic workshopping, micro-projects and planning documents. A possible exception was the USAID-funded Cooperative League of the United States of America (CLUSA) project in Zambia (P. Hoon, personal communication 2009), which organized local farmers to improve markets for products and inputs, and encouraged conservation farming and forest protection initiatives under the rubric of CBNRM. However, it is questionable whether individualized natural resource management (such as arable fields and domestic plants) should be labelled CBNRM.

TOWARDS A CBNRM ENABLING ENVIRONMENT

These thumbnail sketches do little justice to the years of effort by numerous individuals and organizations, but they do provide several valuable lessons on the enabling environment needed for CBNRM.

THE MULTI-SCALE CONFIGURATION OF CBNRM

Accepting that knowledge of how to craft multi-scale institutions is still evolving, we favour an approach that can be easily understood and can deal with this complexity in ways that can be operationalized. Implementing CBNRM appears onerous because it seeks simultaneously to: (1) get prices and economic processes right (to turn relatively low-ordered raw materials and information into more highly ordered environmental products and services), and (2) allocate power and resources to promote social justice and democratization. However, both economic liberalization and political democratization originate from a belief in individual discretion, so the unifying axiom of CBRNM lies in the rights of ordinary people to allocate, manage and benefit from resources within a framework of individual and collective accountability. The operational mechanism for this is the community sitting together to make decisions and review financial and technical performances, in other words the institution of participatory democracy. Achieving this is CBNRM's unifying purpose, but is not a panacea for development because of the underlying environmental constraint to provide goods and services to people.

This leads us to the principle that at the micro-level effective CBNRM depends on the devolution and delineation of the rights to benefit, manage and allocate natural resources at a scale which facilitates a process of participatory (face-to-face) democracy (roughly 150 decision-making members). To meet ecological or economic imperatives at larger scales, management institutions can be scaled upwards through a process of upward delegation, implying that these institutions remain forever accountable to the people in whom discretionary power originates (downward accountability).

This contention is strengthened because the so-called ‘crisis’ in CBNRM, including problems of low participation, inequitable benefit sharing and elite capture, are strongly associated with representational governance (committee-based management.) We contend that downward accountability, where power genuinely originates in the people, creates important benefits (like internal legitimacy, the building of a citizenry, resource proprietorship and accountability) that easily trump efficiency-gains associated with representational governance. Indeed, while the difference between participatory democracy and representational governance may be difficult to perceive (especially when viewed from above), it is so profound that the latter does not fulfil the conceptual requirements of CBNRM as set out in this paper. While common in practice, especially with donor funded projects, committee-based natural resource management exemplifies Murphree's (Reference Murphree2009) contention that robust jurisdictional devolution (namely CBNRM) has not been tried and found wanting, but has been found difficult and not tried.

The foundation of CBNRM, therefore, is the micro-scale institution of participatory democracy, which is scaled up through upward delegation, maintaining its conceptual integrity by complying to principles of participation, accountability, democracy and accountability. This will build capacity through experiential learning gained from the exercise of rights to manage, benefit from and allocate resources. However, the initiation and persistence of local institutional complexity in what North et al. (Reference North, Wallis and Weingast2009) term the ‘natural state’ of social order (with its characteristics of elite personalized power and privilege), depends on critical cross-scale linkages which are only beginning to be understood.

While the principles for micro-institutional design are general, the process of developing an enabling macro-environment for CBNRM is historically, politically and economically specific, even if the goals are the same. We suggest that successful CBNRM requires the achievement of three primary outcomes at the macro-level:

• • the external legitimization of local boundaries and the legislative (or, in the short term, administrative) devolution and protection of rights and principles;

• • monitoring (and enforcing) conformance to procedural principles that guarantee accountability, participation and equity to ordinary people, and protect them against marginalization and elite capture that is ‘natural’ (see North et al. Reference North, Wallis and Weingast2009) even at the very local level; and

• • linkages to macro-level intellectual resources that facilitate cross-scale, cross-site learning, support adaptive management processes and facilitate to processes like participatory technology development and extension.

The likelihood of achieving these goals, and their configuration of actors necessary to do so, depends on many macro-factors including the quality of the civil service, national educational levels, and the openness and vigour of the economy. A critical factor affecting implementation strategy is whether the state is characterized by neo-partimonialism and personalized exchange (for example as in Zambia and Tanzania) or is developing a neo-liberal character that allows for impersonal exchange and greater reliance on formal rules (such as in Botswana) (cf. Hyden Reference Hyden2005; North et al. Reference North, Wallis and Weingast2009).

Given the above, a small team of technical and political professionals may implement CBNRM at several pilot sites with a high probability of success. The difficult intellectual questions are whether meso institutions are necessary to add geographic spread and temporal persistence to CBNRM programmes; which combinations of local government, NGOs, the private sector and even traditional leadership work best; what their roles should be, and how these should change as CBNRM communities build their own capacity and begin to scale up?

THE FUTURE OF CBNRM

CBNRM, not coined in southern Africa, is used, somewhat uncomfortably, as shorthand for a set of principles that seek to:

• • internalize resource costs and benefits and remove market failures (to get prices right),

• • develop local proprietorship by devolving benefits and management to landholders, and

• • devolve choices and management to the people who live with these resources (subsidiarity) through sound institutions.

Thus CBNRM is an institutional approach that empowers communal people to allocate resources to their highest value; economic output and environmental sustainability are important ends. It is often associated with wildlife in southern Africa because wildlife authorities pioneered radical policies that translated wildlife's theoretical economic and ecological advantages into landholder incentives (including democratization and other non-financial values associated with local control). Not only did devolution transform the political economy of wildlife, but they actively and prominently fought trade bans and diversified uses by encouraging entrepreneurship (see G. Child Reference Child1995). Put simply, policy was designed to maximize benefits to landholders who were deterministic of land use. Weakened by pragmatic political compromises, the vision of small communities with title to their land and defendable property rights to all natural resources (not just wildlife) has not yet been fulfilled.

CBNRM has been slow to diversify into other environmental goods and services that may offer similar benefits. The quick answer is that product and market development is a slow process not suited to state agencies or diffuse or centralized ownership (such as state-owned forests or wild resources on communal lands), and that payments for environmental services like water, biodiversity and carbon remain theoretical. However, many of these resources are valuable to local communities, and the real reason is that the authorities responsible for forests, fish, grazing and other resources have seldom matched the rhetoric of participation with devolved legally-defendable use and management rights. Even where CBNRM is associated with high wildlife income, rights to the resources are at least as important as income; the real value of the few dollars that a villager gets is that this money symbolizes a new political economy in which people have the rights to choose how to manage and develop themselves.

If CBNRM is to address the needs of the poorest rural inhabitants, it will have to be extended to other wild resources (such as fibres, fuels and medicinal plants), some of which may not have a great market value, but which nevertheless contribute substantially to the livelihoods of the poor (Shackleton Reference Shackleton2001). With defendable rights, communities are more likely to manage a complexity of these low-income (but valuable) resources sustainably (Chhatre & Agrawal Reference Chhatre and Agrawal2008; Nelson & Chomitz Reference Nelson and Chomitz2009) provided local management is not undermined by the centralization of authority (Ostrom Reference Ostrom2000). Defendable collective property rights, for land and resource use, are central to CBNRM. In parts of Latin America, CBNRM is associated with the concept of private community property (with restrictions on alienation, prescription, subdivision and mortgaging), and titling provides a protective shell allowing internal systems of management and tenure to develop at their own pace. In Mexico, where more than half the country was titled to communities, collective management remains the norm even though many restrictions have been removed, and property can even be individualized if two-thirds of the community members agree (Barnes Reference Barnes2009). Formal communal land titling, including the genuine devolution of usufruct rights to natural resources, is the obvious next step in southern Africa, especially if widespread rural poverty is to be addressed.

In short, proprietorship (1) internalizes costs and benefits, (2) provides the protective shell mentioned above, (3) is valuable in its own right by empowering communities to manage locally important environmental services and low-income products and (4) as observed with wildlife on private land, is a precondition for the entrepreneurial development of products and markets for wild resources. CBNRM also requires a radical redefinition of the role and objectives of natural resource government agencies from controlling or preventing use towards making wild resources as economically competitive as possible by (1) proactively promoting sustainable uses of wild resources, (2) ‘light touch’ regulation similar to agricultural commodities to reduce disincentives to a minimum (for example through devolved peer-based regulation) and (3) removing all differential taxes, charges and license fees on wild resources. Whether the techno-bureaucratic elite that currently controls wild resources can make these radical mind-shifts remains an open question.

Despite equating to half of the world's economy (Constanza et al. Reference Costanza, d'Arge, de Groot, Farber, Grasso, Hannon, Limburg, Naeem, O'Neill, Paruelo, Raskin, Sutton and van den Belt1998), environmental services, many of which are collective, are seldom appreciated in the marketplace. CBNRM provides a working model to use payments for environmental services to simultaneously promote conservation, poverty reduction and democratization, with the crux being how costs, benefits and responsibilities are allocated across organizational scales.

A communal resource attracting a lot of attention in the climate change debate is carbon. Carbon's per hectare value is of the same order of magnitude as the income from wildlife in dry savannahs. Consequently, if processes such as REDD-Plus (see URL http://www.redd-monitor.org/redd-an-introduction/) could internalize the lessons of CBNRM, they might revolutionize both the natural and political environment in non-agricultural zones in Africa and elsewhere. Based on past CBNRM experience, this would require maximizing the benefits to communities from carbon payments, and ensuring compliance with the ideals of participatory democracy. Creating a direct link between global carbon markets and the production of carbon by local communities is essential to circumvent patronage in weak states. Conformance to the principles of CBNRM would combine incentives for carbon sequestration and environmental sustainability, as well as mechanisms for transforming rural economic and political institutions. Two interesting questions are whether a global insistence on conformance to the principles of participatory democracy in income-receiving communities could be reached without being watered down to ineffectiveness, and whether fiscal devolution and rural democratization could be combined to strengthen weak states?

The comparative advantage that wild resources and ecosystem services have in many places is undermined by highly subsidized alternatives. For example, many rural communities are heavily supported by transfer payments in the form of agricultural subsidies, subsidized infrastructure and services, aid and wage remittances, and these subsidies compound the economic dysfunction associated with open-access communal lands.

Despite these challenges, there are few alternatives to CBNRM for the conservation, development and democratization of non-agricultural environments in which people live. Granted, implementation requires rigorous adherence to aligned principles of tenure and resource allocation, proprietorship and participatory governance, but 20 years of CBNRM experience can provide valuable lessons.

Effective CBNRM is no small task. It needs, simultaneously, to reconcile increasing demographic and consumption demands with a smaller resource base, and to develop mechanisms for economic (markets and property rights) and political accountability (democracy, participation and equity). CBNRM is neither a conservation nor a development ‘quick-fix.’ It is a major reorientation of how society governs and protects the sustainability of the wild natural resources on which numerous rural communities depend, and in whose hands the future of these resources lie.